Electrophysiological identification of antennal pH receptors in the ground beetle Pterostichus...

ANTENNAL CONTACT CHEMORECEPTION IN GROUND BEETLES

(COLEOPTERA: CARABIDAE)

JOOKSIKLASTE (COLEOPTERA: CARABIDAE)ANTENNAALNE KONTAKTNE KEMORETSEPTSIOON

MARIT KOMENDANT

A Thesis

submitted for the degree of Doctor of Philosophy in Entomology

Väitekiri

filosoofiadoktori kraadi taotlemiseks entomoloogia erialal

Tartu 2011

EESTI MAAÜLIKOOLESTONIAN UNIVERSITY OF LIFE SCIENCES

ANTENNAL CONTACT CHEMORECEPTION

IN GROUND BEETLES

COLEOPTERA: CARABIDAE

JOOKSIKLASTE (COLEOPTERA: CARABIDAE)ANTENNAALNE KONTAKTNE KEMORETSEPTSIOON

MARIT KOMENDANT

A Th esis

submitted for the degree of Doctor of Philosophy in Entomology

Väitekiri

fi losoofi adoktori kraadi taotlemiseks entomoloogia erialal

Tartu 2011

Institute of Agricultural and Environmental Sciences

Estonian University of Life Sciences

According to the verdict No 80 of March 21, 2011, the Doctoral

Committee of the Agricultural and Natural Sciences of the Estonian

University of Life Sciences has accepted this thesis for the defence of the

degree of Doctor of Philosophy in Entomology.

Opponent: Prof. Habil. Dr. Vincas Būda

Institute of Ecology,

Vilnius University, Lithuania

Supervisors: Dr. Enno Merivee


Prof. Anne Luik


Defence of this thesis:

Estonian University of Life Sciences, Karl Ernst von Baer House, Veski

4, Tartu, on May 6, 2011, at 10.00.

Th e English language was edited by Ingrid H. Williams and the Estonian

by Tiina Halling.

Publication of this thesis is granted by the Estonian University of Life

Sciences and by the Doctoral School of Earth Sciences and Ecology

created under the auspices of European Social Fund.

© Marit Komendant 2011

ISBN 978-9949-484-00-3

5

CONTENTS

LIST OF ORIGINAL PUBLICATIONS .......................................... 7

1. INTRODUCTION ....................................................................... 8

2. AIMS OF THE STUDY ............................................................. 12

3. MATERIAL AND METHODS ................................................. 13

3.1. Test beetles ............................................................................. 13

3.2. Electrophysiology ................................................................... 13

3.3. Chemicals and preparation of stimulating solutions ................ 14

3.4. Feeding bioassays (V) ............................................................. 15

3.5. Data analysis ........................................................................... 15

4. RESULTS .................................................................................... 17

4.1. Taste sensilla on the antennal fl agellum of P. oblongopunctatus and

P. aethiops .............................................................................. 17

4.2. Spike shapes and amplitudes evoked by tested chemicals ........ 17

4.3. Eff ect of stimulus pH to the responses of the electrolyte-sensitive

chemoreceptor neurons innervating antennal taste sensilla of P.

aethiops and P. oblongopunctatus ............................................. 18

4.3.1. Response to 100 mmol l -1 acetate buff ers at pH 3–6 ..... 18

4.3.2. Response to 100 mmol l -1 phosphate buff ers at

pH 5.8–8.1 (11.1) ......................................................... 19

4.3.3. Response to 100 mmol l -1 Na+-salts with and without

addition of 10 mmol l -1 NaOH..................................... 19

4.4. Plant sugar sensitivity in ground beetles ................................. 20

4.4.1. Electrophysiological identifi cation of the sugar-sensitive

neuron in antennal taste sensilla of P. aethiops .............. 20

4.4.2. Electrophysiological responses of the chemoreceptor

neurons from antennal taste sensilla of P.

oblongopunctatus to plant sugars and amino acids ......... 21

4.4.2.1. Response of antennal taste neurons to sugars ... 21

4.4.2.2. Response of antennal taste neurons to amino

acids ................................................................ 22

4.5. Response of antennal taste neurons to plant alkaloids and

glucosides ............................................................................... 22

4.5.1. Electrophysiological identifi cation of the alkaloid-sensitive

neuron in antennal taste sensilla of P. oblongopunctatus .. 22

6

4.5.2. Stimulating eff ect of plant alkaloids and glucosides on

the responses of the antennal chemoreceptor neurons ... 23

4.5.3. Peripheral inhibition of the salt- and pH-sensitive

neurons by some plant secondary compounds .............. 23

4.5.4. Eff ect of quinine hydrochloride on the feeding response

of P. oblongopunctatus ................................................... 24

5. DISCUSSION .............................................................................. 25

5.1. pH sensitivity in ground beetles (I, II) .................................... 25

5.1.1. Th e occurrence of two chemoreceptor neurons in insect

contact chemoreceptors responding to salts ................... 25

5.1.2. Th e eff ect of stimulus pH to the salt- and pH-sensitive

neurons in antennal taste sensilla of P. aethiops and P.

oblongopunctatus ........................................................... 26

5.1.3. pH preference in ground beetles .................................... 27

5.2. Sensitivity of ground beetles to plant sugars and amino acids

(III, IV) ................................................................................. 28

5.2.1. Response of the chemoreceptor neurons from antennal

taste sensilla to sugars ................................................... 28

5.2.2. Response of the chemoreceptor neurons from antennal

taste sensilla to amino acids .......................................... 30

5.3. Response of the chemoreceptor neurons in the antennal taste sensilla

of P. oblongopunctatus to plant secondary compounds ............ 31

5.3.1. Stimulating eff ect of plant alkaloids and glucosides on the

responses of the antennal chemoreceptor neurons ......... 31

5.3.2. Peripheral inhibition of the salt- and pH-sensitive neurons

by some plant secondary compounds ............................ 33

CONCLUSIONS ............................................................................. 35

REFERENCES ................................................................................ 37

SUMMARY IN ESTONIAN ......................................................... 46

ACKNOWLEDGEMENTS ............................................................ 51

PUBLICATIONS ............................................................................ 53

CURRICULUM VITAE ............................................................... 129

ELULOOKIRJELDUS .................................................................. 130

LIST OF PUBLICATIONS ........................................................... 131

7

LIST OF ORIGINAL PUBLICATIONS

Th is thesis is a review of the following papers, which are referred to by

Roman numerals in the text. Th e papers are reproduced by kind permission

of the following journals: Physiological Entomology (I) and Journal

of Insect Physiology (II, III, IV).

I Merivee, E., P loomi, A., Milius, M., Luik, A., Heidemaa,

M., 2005. Electrophysiological identifi cation of antennal pH

receptors in the ground beetle Pterostichus oblongopunctatus. Physiological Entomology 30(2), 122–33.

II Milius, M., Merivee, E., Williams, I., Luik, A., Mänd, M., Must,

A., 2006. A new method for electrophysiological identifi cation

of antennal pH receptor cells in ground beetles: the example of

Pterostichus aethiops (Panzer, 1796) (Coleoptera, Carabidae). .

Journal of Insect Physiology 52, 960–967.

III Merivee, E., Must, A., Milius, M., Luik, A., 2007.

Electrophysiological identifi cation of the sugar cell in antennal

taste sensilla of the predatory ground beetle Pterostichus aethiops. Journal of Insect Physiology 53(4), 377–384.

IV Merivee, E., Märtmann, H., Must, A., Milius, M., Williams, I.,

Mänd, M., 2008. Electrophysiological responses from neurons

of antennal taste sensilla in the polyphagous predatory ground

beetle Pterostichus oblongopunctatus (Fabricius 1787) to plant

sugars and amino acids. Journal of Insect Physiology 54, 1213–

1219.

V Komendant, M., Merivee, E., Must, A., Tooming, E., Williams,

I., Luik, A. Electrophysiological response of the chemoreceptor

neurons in the antennal taste sensilla to plant alkaloids and

glucosides in the granivorous ground beetle Pterostichus

oblongopunctatus. (in review)

Table 1. Author’s contribution to each article (%)

I II III IV VIdea and design 10% 80% 25% 0% 80%Data acquisition 50% 100% 50% 40% 100%Data analysis, statistics 50% 80% 50% 40% 80%Writing 10% 80% 15% 0% 90%

8

1. INTRODUCTION

Th e ground beetles (Coleoptera, Carabidae) with 40,000 species belong to

the most numerous families of beetles, their distribution being cosmopoli-

tan. Approximately 275 species have been found in Estonia (Haberman,

1968; Silfverberg, 2004). Binding of ground beetles to certain habitat

types and microhabitats is determined by numerous biotic and abiotic

factors such as food conditions, vegetation type, presence and distribution

of competitors, life history and season, landscape characteristics, agricul-

tural cultivation impacts, temperature and humidity conditions, light

intensity, and others. Favourite wintering sites are well aerated. Habitat

choice in many ground beetles is so specifi c that they are often used to

characterize habitats (Th iele, 1977; Lövei and Sunderland, 1996; Ings

and Hartley, 1999; Cole et al., 2002; Blake et al., 2003; Eyre and Luff ,

2004; Purtauf et al., 2005).

Chemically mediated habitat selection has been shown to occur in several

species of ground beetle by Evans (1982, 1983, 1984, 1988). By behavioural

experiments, he demonstrated that olfactory neurons receptive to methyl

esters of palmitic and oleic acid emitted by the mat-forming blue-green

algae Oscillatoria animalis Agardh and Oscillatoria subbrevis Schmidle are

present in sensillae on the antennae of those Bembidiini associated with

the shores of saline lakes. Salt content and pH of the soil may also have

importance in determining the distribution of ground beetles (Th iele,

1977; Paje and Mossakowski, 1984; Hoback et al., 2000; Irmler, 2001,

2003; Magura, 2002). Behavioural ablation experiments have shown that

presumptive contact chemoreceptive sensilla are located on the ground

beetles’ antennae (Paje and Mossakowski, 1984; Hoback et al., 2000).

Scanning electron microscope studies show that the antennae of ground

beetles are well equipped with various chemoreceptors. In addition, to

numerous small basiconic olfactory sensilla, approximately 56 to 70 larger

bristle-like contact chemosensilla (35–200 μm in length) have been found

on the antennal fl agellum of adults in several genera (Kim and Yamasaki,

1996; Merivee et al., 2000, 2001, 2002). In the ground beetle Nebria

brevicollis these large sensilla are each innervated by one mechanoreceptive

and four chemoreceptive neurons (Daly and Ryan, 1979). A similar set

of neurons is typical for contact chemoreceptors in various insect groups

(Morita and Shiraishi, 1985; Chapman, 1998; Jørgensen et al., 2006).

9

By electrophysiological single sensillum stimulations the occurrence of

two chemoreceptor neurons responding to salts has been demonstrated

in the antennal taste bristles of the ground beetle Pterostichus aethiops. One of them, producing spikes with negative polarity was highly sensitive

to both ionic composition and concentration of the tested nine salt solu-

tions showing a phasic-tonic type of reaction with a pronounced phasic

component. Th e stimulating eff ect was dominated by the cations involved,

and in most cases, monovalent cations were more eff ective stimuli than

divalent cations. Th is neuron was classifi ed as the salt-sensitive neuron.

Th e specifi c function for the second neuron (the “A-cell”) generating two-

tipped spikes with positive polarity still remains to be explained. It was

observed, however, that alkaline salts usually evoked substantially more spikes in the second neuron than acid salts, but the ionic composition of the salts also aff ected the responses of the second neuron (Merivee et al., 2004). Th us, the “A-cell” with positive spikes seems to be the most probable candidate for pH sensitivity in ground beetles, but no electro-physiogical evidence is available. Th e special function of the third and fourth chemoreceptor neurons of the sensillum also needs to be clarifi ed in ground beetles.

Data on food items are available for approximately 1100 species and sub-species of ground beetles of the world. A total of 73% are carnivorous, 19% are omnivorous and 8% are exclusively herbivorous (Th iele, 1977; Larochelle, 1990; Toft and Bilde, 2002). Of these, many species predate on seeds (Honek et al., 2003). Ground beetles use a variety of external stimuli in order to locate food. Behavioural tests and observations have indicated that visual, tactile, olfactory and gustatory cues may be involved (Wheater, 1989; Kielty et al., 1996; Toft and Bilde, 2002). Various sugars and amino acids function as strong phagostimulants in most phytophagous insects equipped with specialized receptor neurons for these chemicals (Chapman, 1998; Hansen-Delkeskamp, 1998; Schoonhoven and van Loon, 2002; Chapman, 2003) suggesting that these chemicals may have a phagostimulatory role in herbivorous and omnivorous ground beetles,

too. Adults of the genus Pterostichus appear to be primarily carnivorous,

feeding on small prey animals, but in many species, plant matter also

forms part of their diet, and, in some the proportion of animal to plant

matter eaten varies with the season (Th iele, 1977; Hengeveld, 1980; Lövei

and Sunderland, 1996). Johnson and Cameron (1969) collected data on

the extent of herbivory in more than 150 species of ground beetle. Some

species in the genera Pterostichus, including Pterostichus oblongopunctatus,

10

Amara, Agonum and Harpalus consume conifer seeds and young seedlings

immediately following germination. In Sweden, P. oblongopunctatus and

Calathus micropterus are the most important predators of Pinus sylvestris seed. In the fi eld, seed predation typically resulted in >20% seed mortal-

ity, reaching even higher levels (up to 60%) on some occasions and most

predation (81% of all damaged seeds) was due to ground beetles (Heikkilä,

1977; Nystrand and Granström, 2000). Th erefore, as in other phytopha-

gous insects, ground beetles feeding exclusively or partly on plant material,

may also have plant sugar- and amino acid-sensitive neurons associated

with their feeding habits although no experimental data is available.

In addition to detecting phagostimulants, most animals, including her-bivorous insects, possess chemoreceptor neurons responding to a diverse range of feeding deterrent compounds. When stimulated, deterrent neurons reduce or fully stop feeding activity (Dethier, 1980; Schoonhoven and van Loon, 2002; Chapman, 2003; Wang et al., 2004). Natural deterrent stimuli recognized by chemoreceptor neurons of herbivorous insects con-stitute the largest and most structurally diverse class of gustatory stimuli. It has been shown that plant glucosides, for example, sinigrin and salicin, and various alkaloids like nicotine (pyrrolidine group), strychnine and quinine (quinoline group), caff eine (purine group) and several others stimulate the deterrent neuron in many herbivorous insects. Th reshold concentrations for compounds that stimulate a deterrent neuron vary from 10-7 to 10-2 mol l-1, but the majority of these are in the range of 10-4 to 10-2 mol l-1. Th erefore, deterrent neurons fulfi l a central role in host-plant recognition and have, since their discovery (Ishikawa 1966), attracted much interest (Glendinning, 2002; Ryan, 2002; Schoonhoven and van Loon, 2002; Chapman, 2003; Jørgensen et al., 2007).

Herbivory, including pre- and post-dispersal plant seed predation, is ex-

tremely common in virtually all ecosystems. Seeds with their very high

nutrient values per unit volume are extremely attractive to granivores as

food (Janzen, 1971). To counterbalance the eff ects of herbivory, plants

have developed a broad spectrum of secondary metabolites involved in

plant defense, which are collectively known as antiherbivory compounds

and can be classifi ed into three main sub-groups: nitrogen compounds

(including alkaloids, cyanogenic glycosides and glucosinolates), terpenoids,

and phenolics. Th ese compounds can act as feeding deterrents or toxins

with various modes of action to herbivores, or reduce plant digestibility

(Janzen, 1971; Jolivet, 1998; Ryan, 2002; Chapman, 2003; Hulme and

11

Benkman, 2004; Davis et al., 2008). Seeds are sources of some of the

most toxic natural products known to humans and the secondary chemi-

cals in seeds present formidable challenges to granivores. However, seed

chemical defenses can only be assessed with reference to specifi c target

organisms since a secondary chemical may not be equally toxic to all

granivores. Some seeds are so toxic that all seed-predators avoid them

(Harborne, 1993; Hulme and Benkman, 2004). Most toxins are usually

in small concentrations (<5%) in the seed and some alkaloids, for exam-

ple, can be lethal at concentrations as low as 0.1% (Hatzold et al., 1983;

Harborne, 1993). Unfortunately, there is no evidence that plant secondary

compounds cause feeding deterrency in ground beetles. However, due to

selective seed predation by ground beetles (Toft and Bilde, 2002; Tooley

and Brust, 2002), the presence of a deterrent neuron in their antennal

contact chemoreceptors should be expected.

12

2. AIMS OF THE STUDY

Th e aims of this study were:

1) by electrophysiological experiments, to test the response of the

chemoreceptor neurons in the antennal taste sensilla of the

ground beetle P. oblongopunctatus to acetate and phosphate

buff ers over a wide range of pH (3–11) in order to clarify the

eff ect of stimulus pH on the neurons, and to clarify the possible

existence of a pH-specifi c neuron in this species (I);

2) using various alkalized and non-alkalized salts as stimuli, to

test the electrophysiological response of the antennal contact

chemoreceptor neurons of the ground beetle P. aethiops to

stimulus pH in order to collect additional data on functioning

of a specifi c pH-sensitive neuron in ground beetles (II);

3) by electrophysiological experiments, to determine whether sugar-

and amino acid-sensitive receptor neurons are present in the

antennal taste sensilla of P. aethiops (III) and P. oblongopunctatus (IV) by testing for stimulatory eff ects of various sugars of live

and decaying plant origin, and amino acids on these sensilla;

4) to test the stimulatory eff ect of some toxic plant secondary

compounds (alkaloids and glucosides) on the chemoreceptor

neurons of the antennal taste sensilla of the granivorous P. oblongopunctatus in order to identify the specifi c feeding deterrent

neuron stimulated by these chemicals (V).

13

3. MATERIAL AND METHODS

3.1. Test beetles

Adult P. oblongopunctatus and P. aethiops were collected from a local

population in southern Estonia. Th ey were kept in 30×20×10 cm plastic

boxes fi lled with moistened moss and pieces of brown-rotted wood in a

refrigerator at 5–6 °C. Th ree to four days prior to the experiments, the bee-

tles were transferred to room temperature (20–23 °C), fed with moistened

cat food (Kitekat, Master Foods, Poland, I, II; Friskies Vitality+, Nestle´

Purina, Hungary, III, IV, V) and given clean water to drink every day.

3.2. Electrophysiology

Each beetle to be tested was restrained securely in a conical tube made

of thin sheet-aluminium of a size that allowed the head and antennae

to protrude from the narrower end. Th e wider rear end of the tube was

blocked with plasticine to prevent the beetle from retreating out of the

tube. Its antennae were fi xed horizontally to the edge of an aluminium

stand with special clamps and beeswax, so that the horizontally located

large contact chemoreceptors (chaetoid taste sensilla) were visible from

above under a light microscope, and easily accessible for micromanipula-

tion from the side. Care was taken not to contaminate the taste sensilla

by contact with instruments or beeswax. Each prepared test beetle was

placed in a Faraday cage (15×10×5 cm) for electrophysiology.

Spikes from taste neurons of the taste sensilla were recorded by the conven-tional single sensillum tip-recording technique (Hodgson et al., 1955). To achieve a good signal-to-noise ratio, the indiff erent tungsten microelectrode was inserted into the base of the fl agellum free of muscular tissue. It was connected to an AC amplifi er via a calibration source, and earthed. Th e recording glass micropipette fi lled with stimulating solution was brought into contact with the tip of the sensillum by means of a micromanipula-tor under visual control through a light microscope at a magnifi cation of 300×. Immediately before each stimulation, 2–3 drops of solution were squeezed out of the micropipette tip by a syringe connected to the rear end of the micropipette by means of a silicone tube, and adsorbed onto a piece of fi lter paper attached to another micromanipulator, in order to

14

avoid concentration changes in the capillary tip due to evaporation. Spikes picked up by the recording electrode were fi ltered with a bandwidth set at 100–2000 Hz, amplifi ed (input impedance 10 GV, amplifi cation factor 2000×), monitored on an oscilloscope screen, and relayed to a computer via an analogue-to-digital input board DAS-1401 (Keithley, Taunton, MA, USA) for data acquisition, storage, and analysis using TestPoint software (Capital Equipment Corp., Billerica, MA, USA) at a sampling rate of 10 kHz. Responses were recorded from neurons of one (II) to fi fteen (I, III, IV, V) randomly selected sensilla on the fl agellomeres 1–9 of each test beetle. Each sensillum was stimulated once only with one solution (IV). In most cases, however, each sensillum was stimulated twice, with control (10 mmol l-1 electrolyte) and test stimulus solution, respectively. In each subsequent test beetle, the order of stimulation was reversed (I, II, III, V). During each 1 s (IV) or 5 s (I, II, III, V) stimulation period, spikes picked up by the recording electrode were counted and analysed. Responses of the sensilla to mechanical stimulation were also tested, especially in those rare cases when it was not clear whether a certain class of recorded spikes originated from the chemosensory or mechanosensory neuron. Th e tip of the sensillum was rapidly moved from its resting position toward the antennal shaft by 15–20°, held in this position for 2–3 s, and then rapidly moved back to its initial position using the recording electrode.

3.3. Chemicals and preparation of stimulating solutions

To explain the eff ect of stimulus pH to the responses of antennal chemo-receptor neurons, 100 mmol l-1 acetate and phosphate buff ers in the pH range 3–8.1 were prepared. For some experiments, the pH of the phosphate buff er was raised to 11.1 with NaOH. To produce stimulating buff ers of approximately the same pH but of diff erent concentrations of ions, the 100 mmol l-1 phosphate buff er (pH 7.6) was diluted in distilled water to 10 mmol l-1 with pH 7.8 (I). To test for the presence of the pH neuron in the antennal taste bristles of the ground beetle P. aethiops more precisely, three solutions of 100 mmol l-1 Na+-salts, Na

2HPO

4, NaCl and NaH

2PO

4 were prepared, acid, neutral

and alkaline, respectively; these served as control stimuli. Test stimuli were prepared by adding small amounts of NaOH to the 100 mmol l-1 salts resulting in three electrolyte mixtures with 1 mmol l-1 NaOH and three mixtures with 10 mmol l-1 NaOH. Addition of NaOH to these salt solu-tions does not change their ionic composition and the Na+ concentration

15

increases only to a small extent (1% and 10%, respectively). At the same time, the pH level of these solutions rises substantially. Th e stimulating eff ect of these sodium salt solutions with and without NaOH was expected to be approximately similar for the salt-sensitive neuron but not for the pH-sensitive neuron if present (Table 1, II). Th e fi nal pH of the stimulat-ing solutions was measured with a pH meter E6115 (Evikon, Estonia).Sugars (III; Table 1, IV), amino acids (Table 1, IV), alkaloids and gluco-sides (Table 1, V) were dissolved in 10 mmol l-1 electrolyte (KCl, NaCl, choline chloride) to give good electrical conductivity. Stimulating chemicals were kept at 5 °C for less than a week. Chemicals were obtained from AppliChem (Germany) and Sigma-Aldrich company.

3.4. Feeding bioassays (V)

Two-choice feeding experiments were carried out in 110-mm diameter

glass Petri dishes. To ensure high air humidity inside the dishes, each

was lined with a moistened Whatman fi lter paper circle (Schleicher and

Schuell, England). Two pine seeds (Pinus sibirica), treated for 20 min with

distilled water (control) and 0.01 to 50 mmol l-1 quinine hydrochloride,

respectively, were placed in the centre of each Petri dish, 50 mm from each

other. Only seeds with seed coats removed were used in the experiments.

Test beetles, one in each Petri dish, were allowed to choose between, and

feed on the seeds for 24 h (20 °C, 14:10 light:dark). After 24 h, the beetles

were removed and the percentage of predation on control seeds and seeds

treated with alkaloid, was visually determined. At each concentration of

quinine hydrochloride, the feeding preference of the beetles was tested

in fi ve replications (10 beetles in each).

3.5. Data analysis

Automated classifi cation and counting of action potentials from taste

sensilla was not an appropriate method for data analysis because two ac-

tion potentials frequently coincided to produce an irregular waveform.

Instead, spikes from several chemosensory neurons innervating anten-

nal taste sensilla tested were visually distinguished by their polarity and

amplitude (Figs. 1, III, IV, V), and counted, using TestPoint software.

Firing rates (spikes/sec) were analysed by counting the number of spikes

within 5 s after stimulus onset. Mean fi ring rates and their standard errors

16

were calculated. t-test for paired samples (I, II, III,), t-test for independ-

ent samples (IV), Wilcoxon test (V) and ANOVA, Tukey test (IV) were

used to determine the signifi cance of diff erences between means. Spike

amplitude analysis was performed in order to explain which specifi c neuron

was stimulated by the chemical classes tested (IV, V).

17

4. RESULTS

4.1. Taste sensilla on the antennal fl agellum of P. oblongopunctatus and P. aethiops

Large blunt-tipped 100–200 μm long taste bristles are located in whorls

in the distal part of fl agellomeres. Th ere are six on the fi rst fl agellomere

and seven on the following fl agellomeres counting from the beetle’s head.

In addition to those, there are six bristles symmetrically at the tip of the

terminal fl agellomere. Hence, the total number of taste bristles on each

antennal fl agellum is 68. All these bristles seemed to respond in a similar

manner to the stimuli tested.

4.2. Spike shapes and amplitudes evoked by tested chemicals

Spike waveform analysis showed that spikes from four diff erent chemore-

ceptor neurons innervating antennal taste sensilla of the ground beetles P.

oblongopunctatus and P. aethiops could be distinguished in the recordings

obtained in response to tested chemicals. Spikes with negative polarity,

1–2.5 mV in amplitude, arose from the salt-sensitive neuron. Th is neuron

generated spikes in a phasic-tonic manner and had an extremely short

response latency lasting less than 10 ms (Fig. 1, III; Fig. 1, V). Two-tipped

spikes with positive polarity, 2–8 mV in amplitude, were generated by

the pH-sensitive neuron (Fig. 1, I, III; Figs. 1 and 2, IV). Compared to

the spikes from the salt-sensitive neuron, these positive spikes appeared

with a much longer delay (Fig. 1, II; Fig. 1A, I, V). Th e shape of these

impulses is complex because they are composed of two separate electrical

events in quick succession and most probably produced by two separate

spike generation sites of the same neuron. Minor diff erences in the time

interval between these two events occuring among antennal taste sensilla

may cause considerable diff erences in the shape of the resulting double-

spike. Usually, the double-spikes of the pH-sensitive neuron described are

two-tipped (Fig. 2A and B, IV). In some sensilla, however, only a small

characteristic notch or discontinuity in the rising phase of these impulses

indicates their complex nature (Fig. 2C and D, IV). Th e sugar-sensitive

neuron also produced spikes with negative polarity, but smaller in ampli-

tude than those of the salt-sensitive neuron (Fig. 1B, III, IV; Fig. 1J, V).

Th e fourth chemoreceptor neuron of the taste sensillum was stimulated

by some alkaloids. Th e negative spikes it produced were variable in size

18

(Fig.1 B–E, V) but diff erent from those originating from the salt- and

sugar-sensitive neurons (Fig. 2, V).

Impulse activity of the mechanosensory neuron belonging to these taste

sensilla and responding to bending of the bristle was recorded from some

intact bristles as well as from bristles with a broken tip. Spikes produced

by the mechanosensory neuron were small but variable in size, and nega-

tive in polarity (Fig. 1C and D, IV).

4.3. Eff ect of stimulus pH to the responses of the electrolyte-sensitive chemoreceptor neurons innervating antennal taste

sensilla of P. aethiops and P. oblongopunctatus

Two chemoreceptor neurons from fl agellar taste bristles of P. aethiops and

P. oblongopunctatus may respond to stimulation with various electrolyte

solutions in a phasic-tonic manner. According to their response charac-

teristics these neurons were classifi ed as salt- (cation-) and pH-sensitive

neuron distinguished by the amplitude, shape and polarity of spikes they

generated (Fig. 1, II). Th e phasic component of the response was more pro-

nounced in the salt-sensitive neuron compared to that of the pH-sensitive

neuron (Figs. 2 and 3, II). Due to large variations in responsiveness of

the pH-sensitive neuron from diff erent taste bristles and diff erent samples

of beetles to the same stimulus (Fig. 4, A-cell, I), stimulating solutions

were tested in pairs to allow a more precise comparison of the responses

to the tested stimuli.

4.3.1. Response to 100 mmol l-1 acetate buff ers at pH 3–6

Th e pH-sensitive neuron did not respond to acetate buff er at pH 3 and

only a very weak response was observed at pH 4; fi rst second response

below 1 spikes/s (Fig. 2A, A-cell, I). Th e number of spikes generated by

the salt-sensitive neuron at pH 3 and 4 was much higher, fi rst second

responses 5.7 and 9.5 spikes/s, respectively (Fig. 2A, B-cell, I). Accord-

ing to how much the pH level of the stimulating buff ers increased fur-

ther, more spikes by the pH- and salt-sensitive neurons were produced.

However, even at pH 6, the mean fi ring rate of the pH-sensitive neuron

remained below 5 spikes/s while that of the salt-sensitive neuron reached

23 spikes/s (Fig. 2B,C, I).

19

4.3.2. Response to 100 mmol l-1 phosphate buff ers at pH 5.8–8.1 (11.1)

In most cases, phosphate buff ers were remarkably more stimulating for

the pH-sensitive neuron compared to acetate buff ers; at pH 5.8, the neu-

ron produced a mean of 8.1–10.5 spikes/s (Fig. 4A,B, A-cell, I). When

buff er pH was increased from 5.8 to 7, the fi rst second fi ring rate of the

pH-sensitive neuron increased by 50%, from 10.5 to 15.8 spikes/s. Even

larger increases in the response, reaching up to 300–800%, were observed

during next four seconds of the response after stimulus onset (Fig. 4A, A-

cell, I). In contrast, the response of the salt-sensitive neuron to these two

phosphate buff ers was approximately equal (Fig. 4A, B-cell, I). However,

the salt-sensitive neuron responded to the larger jumps in stimulus pH.

For example, when buff er pH was increased by three pH units, from 8.1

to 11.1, both of the neurons responded with signifi cant fi ring rate increase

whereby relative change in the response of the pH-sensitive neuron was

considerably larger compared to that of the salt-sensitive neuron, the fi r-

ing rate increased during the fi rst second of the response by 1350% and

25%, respectively (Fig. 4C, A- and B-cell, respectively, I).

4.3.3. Response to 100 mmol l-1 Na+-salts with and without addition of 10 mmol l-1 NaOH

Th e limitation of the buff er series method used for testing the responses

of the chemoreceptor neurons to stimulus pH is that buff ers with diff erent

pH vary largely in both pH and ionic concentrations of the component

salts. Th e method of alkalized salts reported here allows the preparation

of stimulating salt solutions over a wide range of pH whereas changes in

ionic concentrations of the component salts are minimal.

Remarkable diff erences between the pH- and salt-sensitive neurons re-

garding their responsiveness to the pH of the stimulating salt solutions

were demonstrated by electrophysiological experiments with alkalized

and non-alkalized salts. In the pH-sensitive neuron, rates of fi ring evoked

by the mixture of 100 mmol l-1 Na+-salts and 10 mmol l-1 NaOH were

signifi cantly higher compared to those caused by the salts alone (Fig. 3,

II). Th e stimulating eff ect of NaOH to the pH neuron was highest in the

mixture with NaCl (pH 9.6), fi ring rate during the fi rst second of the

20

response was 31 spikes/s and it did not fall below 17 spikes/s also during

the next several seconds recorded. Th e diff erence compared to the response

caused by 100 mmol l-1 NaCl alone (pH 7.6) was 10–22 times depend-

ing on the response time. Th e mixtures of 10 mmol l-1 NaOH with 100

mmol l-1 NaH2PO

4 (pH 5.8) and Na

2HPO

4 (pH 10.6) were 2–7 times

more eff ective stimuli for the pH-sensitive neuron compared to these non-

alkalized salts (pH 4.6 and 8.6, respectively). In contrast, the presence or

absence of 10 mmol l-1 NaOH in the stimulating salt solution had only

little or no eff ect on the spike production of the salt-sensitive neuron.

4.4. Plant sugar sensitivity in ground beetles

4.4.1. Electrophysiological identifi cation of the sugar-sensitive neuron in antennal taste sensilla of P. aethiops

In the ground beetle P. aethiops, by single sensillum tip recording tech-

nique, in addition to the salt- and pH-sensitive neurons, the third chemo-

receptor neuron of the antennal taste sensilla was electrophysiologically

identifi ed as a sugar-sensitive neuron. Th is neuron generated spikes of

considerably smaller amplitude than those of the salt- and pH-sensitive

neurons (Fig. 1B, III), and phasic-tonically responded to sucrose and glu-

cose over the range of 1–1000 mmol l-1 tested. Th e responses of the sugar-

sensitive neuron to stimulating solutions containing glucose or sucrose

were signifi cantly stronger compared to the control solution (p<0.05, t-test

for paired samples; Figs. 2A and 3A, III). Responses were concentration

dependent, with sucrose evoking more spikes than glucose. During the

fi rst second of the response, mean rates of fi ring of the sugar-sensitive

neuron reached up to 19 and 37 spikes/s when stimulated with 1000 mmol

l-1 glucose and sucrose, respectively. Th ereafter, the rates of fi ring quickly

fell below 5 spikes/s (sucrose), or close to zero (glucose).

In contrast, the responses of the salt-sensitive neuron were not signifi cantly

aff ected by the glucose content of the stimulating solutions (p>0.05, t-test

for paired samples; Fig. 2B, III). It was observed, that sucrose had a little

stimulating eff ect to the fi ring rate of the neuron, however (p<0.05, t-test

for paired samples; Fig. 3B, III). Th e responses of the pH-sensitive neuron

were also aff ected to a small degree by the content of sucrose and glucose

in the stimulating solutions, but in a more complicated manner. Firing

rate of the neuron increased to some degree at low concentrations (1 and

21

10 mmol 11) of the sugars (p<0.05, t-test for paired samples). At higher

concentrations (100 and 1000 mmol l-1) of the sugars the opposite was

observed, however: fi ring rates of the pH-sensitive neuron were slightly

suppressed by both glucose and sucrose compared to the responses evoked

by 10 mmol l-1 KCl (control) alone (p<0.05, t-test for paired samples;

Figs. 2 and 3, III).

4.4.2. Electrophysiological responses of the chemoreceptor neurons from antennal taste sensilla of P. oblongopunctatus to

plant sugars and amino acids

Choline chloride at a concentration of 10 mmol l-1 served as the control

stimulus, and as the electrolyte in stimulating mixtures with sugars and

amino acids to ensure a good signal-to-noise ratio. Typically, choline

chloride evoked only the salt-sensitive neuron to discharge with a relatively

low rate of fi ring (approximately 15 spikes/s) during the fi rst second after

stimulus onset (Fig. 3A, IV). Only in rare cases were a few spikes from

other sensory neurons of the sensillum observed (Fig. 1A; Table 1, IV).

Th erefore, the stimulating properties of choline chloride were suitable for

classifying and counting action potentials of the responses evoked by the

mixtures of this electrolyte with sugars and amino acids.

4.4.2.1. Response of antennal taste neurons to sugars

Responses of antennal taste neurons to stimulation by various 100 mmol

l-1 sugars in mixtures with 10 mmol l-1 choline chloride, varied with

the sugar tested. Only three out of twelve sugars, maltose, sucrose and

glucose, evoked strong phasic-tonic responses from the sugar-sensitive

neuron (Fig. 3B–E; Table 1, IV). Th e mean rates of fi ring of the neuron

during the fi rst second after 100 mmol l-1 stimulation reached 49.9, 46.5

and 20.9 spikes/s, respectively. Dose/response curves of these three active

sugars are demonstrated in Fig. 4 (IV). In the range of 1–1000 mmol l-1,

the response to glucose was approximately two times lower than that of

maltose and sucrose (p<0.05, ANOVA, Tukey test). In contrast, no dif-

ferences were found between responses to maltose and sucrose (p>0.05,

ANOVA, Tukey test). Th e remaining nine 100 mmol l-1 sugars had only a

small or no stimulatory eff ect on the neuron. To test for possible seasonal

variation in the functioning of the sugar-sensitive neuron, its responses

22

to 100 mmol l-1 sucrose were compared in active and hibernating beetles

in July (46.5 spikes/s) and February (48.8 spikes/s), respectively but no

diff erences between their responses were observed (t = -0.57; d.f. = 18;

p= 0.57; t-test for independent samples).

4.4.2.2. Response of antennal taste neurons to amino acids

Responses of antennal taste neurons to the seven 10 and 100 mmol l-1

L-amino acids in a mixture with 10 mmol l-1 choline chloride varied with

the type and concentration of amino acid tested. At 10 mmol l-1, none of

the amino acids stimulated the neurons. At 100 mmol l-1, these chemicals

evoked only a few spikes (Table 1, IV) with amplitudes smaller than those

generated by the salt-sensitive neuron (Fig. 3G–I, IV). No initial phasic

component in the sequence of these spikes was observed. Frequently,

they appeared with some delay in the responses, sometimes 5–10 s after

stimulus onset (Fig. 3J, IV). Spike amplitude analysis showed that these

small impulses were generated by the 4th chemosensory neuron rather

than by the sugar-sensitive neuron. Th ough amplitudes of spikes induced

by sugars (sucrose) and amino acids varied and overlapped to a large extent

their histograms had diff erent maxima (Fig. 5, IV), indicating that two

neurons may be involved. Since the mean numbers of these small spikes

during the fi rst second after stimulus onset did not exceed 3 spikes/s, this

neuron has probably no specialized receptor sites for amino acids, at least

for those which were tested.

4.5. Response of antennal taste neurons to plant alkaloids and glucosides

4.5.1. Electrophysiological identifi cation of the alkaloid-sensitive neuron in antennal taste sensilla of P. oblongopunctatus

Some tested alkaloids (quinine, caff eine) and an alkaloid salt (quinine

hydrochloride) induced a neuron of the taste sensilla to produce spikes

with negative polarity, but with a considerably longer period of latency

compared to the spikes of the salt- and sugar-sensitive neurons (Fig. 1B–E,

V). Spike amplitude histograms show two maxima, indicating that spikes

evoked by salt (NaCl), sugar (sucrose) and alkaloid (quinine hydrochloride)

were generated by three diff erent neurons, however, the salt-, sugar- and

23

alkaloid-sensitive neuron, respectively (Fig. 2, V). Spike amplitudes of

the chemoreceptor neurons from diff erent taste sensilla varied to a some

extent, from 0.7 to 2 mV (Fig. 1C,E, V). As a result, in most cases, spike

amplitudes generated by the alkaloid-sensitive neuron were shorter than

those of the salt-sensitive neuron (Fig. 1B,C, V). In some cases, however,

opposite spike amplitude ratios of the neurons were observed (Fig. 1D, V).

4.5.2. Stimulating eff ect of plant alkaloids and glucosides on the responses of the antennal chemoreceptor neurons

Only two of the tested chemicals (Table 1, V), 1 mmol l-1 quinine and 10

mmol l-1quinine hydrochloride, were highly stimulating for the antennal

alkaloid-sensitive neuron, evoking 8.9 and 22.8 spikes/s, respectively. Th e

stimulatory eff ect of 10 mmol l-1 caff eine was very weak, 2.4 spikes/s. In

contrast, a slight inhibiting eff ect of strychnine and sinigrin to the activity

of the neuron was observed. Respective fi ring rates were very low, how-

ever, below 2 spikes/s (Table 1, V). Other plant chemicals tested had little

eff ect on spike production of the alkaloid-sensitive neuron. Phasic-tonic

response of the alkaloid-sensitive neuron to quinine hydrochloride was

concentration dependent over the range of 0.001 to 50 mmol l-1 with the

response threshold at 0.01 mmol l-1, and maximum rate of fi ring equal to

67 spikes/s at 50 mmol l-1 (Figs. 3 and 4, V). However, compared to the

spike trains of the salt-sensitive neuron (Figs. 1 and 5A–F, V), the phasic

component of the response produced by the alkaloid-sensitive neuron was

less pronounced (Fig. 1B–E; Fig. 4; Fig. 5C–F, V).

4.5.3. Peripheral inhibition of the salt- and pH-sensitive neurons by some plant secondary compounds

In addition to stimulation of the alkaloid-sensitive neuron, both quinine

and quinine hydrochloride strongly inhibited spike generation by the salt-

and pH-sensitive neurons when presented in mixtures with 10 mmol l-1

NaCl (Table 1, V). Increasing the quinine hydrochloride concentration

while keeping the concentration of NaCl constant at 10 mmol l-1 caused

a dose-dependent decrease in spike generation by the salt-sensitive neuron

(Figs. 3–5, V). In most tested sensilla, at concentrations above 10 mmol

l-1 quinine hydrochloride, only the alkaloid-sensitive neuron fi red while

the fi ring rate of the salt-sensitive neuron of the taste sensillum decreased

24

close to zero during the 5 s adaptation period (Fig. 4, V). In response to

10 mmol l-1 NaCl alone, the mean number of spikes produced by the pH-

sensitive neuron was relatively low compared to that of the salt-sensitive

neuron, and varied to some extent in diff erent sensilla (Table 1, V). At

quinine hydrochloride concentrations above 0.01 mmol l-1, the mean fi ring

rate of the pH-sensitive neuron dropped to zero (Fig. 3, V). It was also

observed that nicotine and salicin, which did not aff ect the fi ring rate of the

alkaloid-sensitive neuron, greatly reduced the response of the pH-sensitive

neuron (Table 1, V). Sinigrin and caff eine suppressed neural activity of

the salt- and pH-sensitive neuron, respectively. Th is suggested that the

antennal taste sensilla of P. oblongopunctatus may detect plant defensive

compounds not only through the activation of specifi c alkaloid-sensitive

neuron but also through inhibition of the salt- and pH-sensitive neurons.

4.5.4. Eff ect of quinine hydrochloride on the feeding response of P. oblongopunctatus

Feeding choice experiments showed that one of the most stimulating

chemicals for the alkaloid-sensitive neuron, quinine hydrochloride, had a

concentration dependent deterrent eff ect on seed predation in the ground

beetle P. oblongopunctatus, with the threshold concentration at 1 mmol l-1.

Th us, the threshold concentration of this alkaloid to evoke a behavioural

response was a hundred times higher than that for an electrophysiological

response of the alkaloid-sensitive neuron. At higher concentrations, the

deterrent eff ect of quinine hydrochloride was very strong. It was observed

that pine seeds treated with 50 mmol l-1 quinine hydrochloride were never

predated by the beetles (Fig. 6, V).

25

5. DISCUSSION

5.1. pH sensitivity in ground beetles (I, II)

5.1.1. Th e occurrence of two chemoreceptor neurons in insect contact chemoreceptors responding to salts

In a number of insects, including the ground beetles P. oblongopunctatus (I) and P. aethiops (Merivee et al., 2004; II) frequently two neurons of

the taste sensilla respond to salt solutions (Dethier and Hanson, 1968;

Haskell and Schoonhoven, 1969; Den Otter, 1972; Mitchell and Schoon-

hoven, 1973; Mitchell and Seabrook, 1973; Blaney, 1974; Hanson, 1987;

Bernays and Chapman, 2001a; Schoonhoven and van Loon, 2002). It is

suggested that they function as a cation and an anion neuron (Ishikawa,

1963; Elizarov, 1978; Hanson, 1987; Schoonhoven and van Loon, 2002).

Temporal patterns of fi ring and the concentration/response curves are

quite diff erent for the two neurons. In the cation neuron with a phasic-

tonic type of reaction, the magnitude of the response is proportional to

the logarithm of salt concentration over the dynamic range (Evans and

Mellon, 1962; Hanson, 1987). It is believed that the stimulating eff ect

of salts to taste receptors is dominated by the cations involved, and that

monovalent cations are more eff ective stimuli than divalent cations (Evans

and Mellon, 1962; Schoonhoven and van Loon, 2002). Th is neuron type

corresponds to the salt-sensitive neuron of the antennal taste bristles in

Pterostichus (Merivee et al., 2004).

Specifi c stimuli for the second salt neuron are yet to be clarifi ed: it has

been named the ‘anion neuron’ on the basis of a supposed sensitivity to

salt anions or the ‘fi fth neuron’ (Ishikawa, 1963; Dethier and Hanson,

1968; Hanson, 1987; Liscia et al., 1997; Liscia and Solari, 2000). Th is

neuron is characteristically less responsive to diff erences in salt concentra-

tion (Dethier, 1977; Liscia et al., 1997), similar to that observed in the

pH-sensitive neuron of P. oblongopunctatus. In the strict sense of the term,

it is not correct to give the name ‘salt neuron’ to a neuron that does not

respond to changes in salt concentration. On the other hand, a study in

Grammia geneura shows increasing activity in two neurons in response

to potassium and sodium chlorides at concentrations ranging from 10 to

1000 mmol l-1 (Bernays and Chapman, 2001a). Dethier (1977) was the fi rst

to realize that it is not at all certain that the best stimulus for the second

salt neuron is indeed salt. Electrophysiological evidence is provided that

26

the so-called ‘fi fth’ neuron in taste chemosensilla of blowfl ies responds

to deterrent compounds, such as quinine, amiloride, nicotine and caf-

feine (Liscia and Solari, 2000). Very little attention has been paid to this

neuron. Th is is unfortunate because this lack of information handicaps

any attempt to construct an overview of the control of food detection

and habitat selection in insects on the basis of their chemosensory input.

5.1.2. Th e eff ect of stimulus pH to the salt- and pH-sensitive neurons in antennal taste sensilla of P. aethiops and P.

oblongopunctatus

Results of the electrophysiological experiments conducted with acid, neu-

tral and alkaline salts (Merivee et al., 2004), acetate and phosphate buff ers

(I), and alkalized and non-alkalized salts (II) demonstrated that, in ad-

dition to the salt-sensitive neuron, the existence of a pH-sensitive neuron

in antennal chaetoid taste sensilla of the ground beetles P. aethiops and P.

oblongopunctatus is possible. A strong relationship between stimulus pH

and spike production by the antennal pH-sensitive neuron was found:

signifi cantly higher rates of fi ring were recorded at higher pH values com-

pared to those observed at lower pH of the electrolyte solutions tested at

pH ranged from 3 to 11. However, the methods of various salts and buff ers

used to test the eff ect of pH on contact chemoreceptor neurons in ground

beetles have limitations in that these stimuli with diff erent pH vary largely

in both pH and ionic concentration of the component salts leaving the

exact stimulating role of these two factors for the neurons open (Merivee

et al., 2004; I). Th erefore, a method of alkalized (with NaOH) Na+ salts

was developed allowing preparation of stimulating solutions over a wide

range of pH with no diff erences in their ionic composition and minimal

diff erences in their salt concentration. Using this method for stimulation

of antennal contact chemoreceptors in P. aethiops (II), 2.5–10-fold increase

in the fi rst second fi ring rate of the pH-sensitive neuron in response to the

mixture of various 100 mmol l-1 salts and 10 mmol l-1 NaOH (pH was

observed (4–30 spikes/s) compared to that of 100 mmol l-1 NaCl alone

(1–8 spikes/s). Th us, the high stimulating eff ect of alkalized salts to the

pH-sensitive neuron could be attributed to pH alone as it was increased

by 1.3 to 2.0 pH units depending on the salt. In comparison, fi ring rate

of the salt-sensitive neuron changed by only 16–27%.

27

If the response of the pH-sensitive neuron depends on solution pH level only, there should be equivalent responses to buff ers at diff erent con-centrations at the same pH. However, the results do not confi rm this: 10 mmol l-1 phosphate buff er had a greater stimulating eff ect on the pH-sensitive neuron of P. oblongopunctatus than 100 mmol l-1 buff er (I). Th is phenomenon can be explained by interaction of diff erent chemical stimuli to chemosensory neurons. Other chemicals, as well as the pH of the stimulating solution, may infl uence the responses of insect taste receptors to one or another stimulating compound (Hanson, 1987; Liscia et al., 1997; Bernays et al., 1998; Liscia and Solari, 2000; Bernays and Chapman, 2001b; Schoonhoven and van Loon, 2002). Presumably, salts of stimulating buff er solutions at high concentration suppressed the neural activity of the pH-sensitive neuron when stimulated with 100 mmol l-1 phosphate buff er in 10 and 100 mmol l-1 phosphate buff er experiments in P. oblongopunctatus.

5.1.3. pH preference in ground beetles

Preferences for particular soil pH, varying between 3 and 9, in ground

beetles were discovered using laboratory choice experiments some decades

ago (Krogerus, 1960; Paje and Mossakowski, 1984). More recently, in

ecological

fi eld experiments, a correlation between the abundance of some ground

beetles and soil pH has been found (Irmler, 2001, 2003). Our results

suggest that in P. aethiops and P. oblongopunctus in their preferred acid

forest habitats and overwintering sites in brown-rotted wood at pH 3–5,

the antennal pH-sensitive neuron does not discharge or discharges at

very low frequency with the fi rst s fi ring rate close to 1 spike/s or lower.

Areas with a higher pH seem to be unfavourable to this insect and when

contacted the pH-sensitive neuron signals with a stronger response. Th is

may occur, for example, in places with decaying plant material on the soil

surface. Many of the materials composted contain signifi cant amounts

of protein, which is converted to ammonia toxic to many organisms and

serving as the primary substrate in the nitrifi cation processes. Nitrifi cation

processes have been observed at pH levels ranging from 6.6 to 9.7 (Odell

et al., 1996). Obviously, these ground beetles avoid places with a high pH.

Adults of P. aethiops and P. oblongopunctus may spend seven to eight

months, from September to April, in their preferred overwintering sites in

28

brown-rotted wood in Estonia. Fungi often lower the pH of the extracel-

lular medium through secretion of organic acids and thus establish a pH

gradient around the mycelium. Th e organic acid oxalate is produced by

most, if not all, wood-degrading fungi. Usually, a pH of 3–6 is observed

in both white- and brown-rotted wood. Stable moisture content is quaran-

teed as a result of metabolic water production during the wood-degrading

process. For cultures of the white-rot fungus Merulius tremellosus incubated

in air, the optimum water content was 2 g/1 g of wood. Moister conditions

were less favourable to delignifi cation, possibly because the water impeded

aeration (Reid, 1985; Varela and Tien, 2003). Similar prolonged stability

of water content and aeration is probably also quaranteed in brown-rotting

wood which seems to be a favourable microenvironment of ground bee-

tles for their prolonged hibernation. Brown-rotted wood may off er good

protection for hibernating ground beetles against entomopathogenic fungi

and parasites which play an important role in their population dynamic.

Up to 41% parasitism by nematodes and ectoparasitic fungi was found

on 14 species of Bembidion in Norway depending on habitat selection

of the host (Andersen and Skorping, 1991). Eggs of P. oblongopunctatus

suff ered 83% mortality in fresh litter but only 7% in sterilized soil (Hees-

sen, 1981). Although predators, parasites, and pathogens aff ect all ground

beetle developmental stages, quantitative data remain scarce. So, antennal

pH receptors of the ground beetles seem to be a powerful tool in their

selection for favourable acid habitats and microhabitats crucial to survival.

5.2. Sensitivity of ground beetles to plant sugars and amino acids (III, IV)

5.2.1. Response of the chemoreceptor neurons from antennal taste sensilla to sugars

Th ree out of the twelve sugars tested in P. aethiops and P. oblongopunctatus, maltose, sucrose and glucose, evoked strong responses from the sugar-sensitive neurons while the others had little or no eff ect. Most stimulating were the two disaccharides, maltose and sucrose (III, IV), and this is in agreement with results for a number of other insect species (Mitchell and Gregory, 1979; Schoonhoven and van Loon, 2002; Albert, 2003; Sandoval and Albert, 2007). Sucrose is a disaccharide consisting of two monosac-charides, α-glucose and β-fructose, joined by a glycosidic bond between carbon atom 1 of the glucose unit and carbon atom 2 of the fructose unit.

29

Maltose is composed of two molecules of glucose joined together through α-1.4-glycoside linkage. Th us, the presence of a terminal α-glucose unit seems to be a common and important feature of the molecular structure determining the stimulatory properties of these disaccharides in Pteros-tichus. Other monosaccharide units in the molecular structure of sucrose and maltose were of no importance in this respect as the sugar-sensitive neuron responses to these two disaccharides were similar. In contrast, the trisaccharide raffi nose has a non-terminal α-glucose unit and does not evoke a response from the antennal sugar-sensitive neuron in this species (IV). In the red turnip beetle, Entomoscelis americana, maltose (α-glucose-1.4-glucose) was at least an order of magnitude more eff ective in activating the maxillary sugar-sensitive neuron than trehalose (α-glucose-1.1-glucose) illustrating the importance of the 1.4 linkage in these disac-charide molecules (Mitchell and Gregory, 1979). Compared to maltose and sucrose, the third active sugar glucose was approximately two times less eff ective at stimulating the antennal sugar-sensitive neuron of both P. oblongopunctatus and P. aethiops (III, IV). Th is can be partly explained by the fact that glucose exists in at least two anomeric forms, α and β. Th ese two forms interconvert over a timescale of hours in aqueous solu-tion, to a fi nal stable ratio of α:β 36:64, in a process called mutarotation. So the physiologically active α-anomere forms only 36% of the glucose solution which is refl ected in its relatively low dose/response curve. Th e order of stimulatory eff ectiveness of sugars may vary in diff erent insect species or even be diff erent for receptors on various parts of the body (Blaney, 1974; Rüth, 1976; Bernays and Chapman, 2001; Schoonhoven and van Loon, 2002).

No insect so far investigated responds to all sugars, and no single sugar

is a major phagostimulant for all insects. For a given insect species, only

some sugars elicit a response to a greater or lesser extent while other sug-

ars elicit no response. Plant sugars which were found to evoke responses

from the antennal contact chemoreceptors in P. oblongopunctatus, i.e.,

maltose, sucrose and glucose (IV), function as strong phagostimulants

for most phytophagous insects, fl ies, cockroaches and others, equipped

with specialized receptors to detect sugars (Dethier et al., 1956; Blaney,

1974; Rüth, 1976; Amakawa, 2001; Liscia et al., 2002; Schoonhoven and

van Loon, 2002; Chapman, 2003; Sandoval and Albert, 2007). Th is sug-

gests that these sugars may have the same function in P. oblongopunctatus, probably due to its partial herbivory. Although, in contrast to sucrose and

glucose, maltose is uncommon in nature, it can be found in considerable

30

amounts in germinating seeds as a product of the enzymatic degradation

of starch. Obviously, high sensitivity of P. oblongopunctatus to maltose

is related to predation by the species on conifer seeds (Heikkilä, 1977;

Nystrand and Granström, 2000).

Several carbohydrates, in addition to glucose, such as cellobiose, arabinose,

xylose, mannose, rhamnose and galactose are known as components of

cellulose and hemicelluloses. Th ey are released by brown-rot fungi dur-

ing enzymatic wood decay. All water-soluble carbohydrates are ultimately

consumed, so that brown-rotted wood consists almost entirely of modifi ed

lignin. Brown-rot fungi appear to depolymerize wood in the early stages

of decay much more rapidly than the decay products can be metabolized.

Th e occurrence of excess wood decomposition products may help explain

the frequent presence of other wood scavengers in brown-rotted wood

(Eriksson et al., 1990; Zabel and Morrell, 1992; Valášková and Baldrian,

2006). As wood decomposes, it passes through a range of decay stages,

each colonized by a succession of diverse saproxylic insect assemblages

(Grove, 2002). Wood-decay products may serve as chemical cues for these

saproxylic insects and be involved in this colonization process although

no electrophysiological or behavioural evidence is yet available. Some

ground beetles, including P. oblongopunctatus, prefer brown-rotted wood

for their prolonged hibernation (Haberman, 1968; Lindroth, 1986). Our

results showed, however, that P. oblongopunctatus was not able to sense

water-soluble wood sugars released by brown-rot fungi, except for glucose

(IV), indicating that these chemicals are not involved in the search for

hibernation sites by this species, although others may be.

5.2.2. Response of the chemoreceptor neurons from antennal taste sensilla to amino acids

Since amino acids, in addition to sugars, stimulate feeding behaviour in various phytophagous insects these animals have amino acid-specifi c neurons in their taste sensilla. Amino acids and sugars may stimulate the same neuron. Such a versatile receptor neuron is present in the polypha-gous caterpillar of Grammia geneura. Th ese insects have, in addition to an amino acid sensitive neuron in their lateral sensillum styloconicum, a neuron in another sensillum which responds to seven (out of twenty) amino acids. Th is latter neuron can also be stimulated by sucrose, glucose, and trehalose (Schoonhoven and van Loon, 2002; Chapman, 2003). Because

31

of its multiple specifi city this neuron was named a ‘‘phagostimulatory neuron’’, rather than a ‘‘sugar’’ or ‘‘amino acid’’ neuron (Bernays and Chapman, 2001). However, in P. oblongopunctatus, the seven amino acids tested (IV) had only little or no stimulatory eff ect on the neurons of the antennal taste sensilla suggesting that these neurons have no ability to perceive these amino acids. Th e weakly stimulatory eff ect (below 3 spikes/s) of some 100 mmol l-1 amino acids on the 4th chemosensory neuron of these sensilla can be characterized as non-specifi c, or modulating to the responses of non-target chemosensory neurons, the phenomenon well known in insect chemoreceptor functioning (Hanson, 1987; Liscia et al., 1997; Liscia and Solari, 2000; Bernays and Chapman, 2001; Chapman, 2003; Schoonhoven and van Loon, 2002; I; III).

5.3. Response of the chemoreceptor neurons in the antennal taste sensilla of P. oblongopunctatus to plant secondary compounds

5.3.1 Stimulating eff ect of plant alkaloids and glucosides on the responses of the antennal chemoreceptor neurons

Th e experimental data presented show that the “fourth” chemoreceptor

neuron in the antennal taste sensilla of the ground beetle P. oblongopunc-tatus (IV) may respond to some plant alkaloids and alkaloid salts such

as quinine and quinine hydrochloride (V). Th is fi nding seems to be in

agreement with literature data on the feeding habits of this species. P. oblongopunctatus is one of the most common and widespread ground

beetles in the forests of northern Europe (Lindroth, 1985; Niemelä et al.,

1994). In most stands of Pinus sylvestris, seed predation results in 20% seed

mortality, although occasionally it reaches 60%. Post-dispersal pine seed

predation in the Swedish boreal forests studied was due mainly to the two

ground beetle species P. oblongopunctatus and Calathus micropterus. Both

are able to consume conifer seeds. Th e larger species, P. oblongopunctatus, is considered to be the more important pine seed predator of the two

(Heikkilä, 1977; Nystrand and Granström, 2000).

Like many herbivores, numerous insect seed-predators have biochemical

adaptations to deal with toxic plant secondary compounds, including vari-

ous detoxifi cation and sequestration mechanisms (Jolivet, 1998; Hulme

and Benkman, 2004; Schoonhoven et al., 2005; Després et al., 2007).

However, the diffi culties of dealing with more than a few diff erent kinds

32

of defensive compounds has favoured specialization in seed-predators

and helps to explain why a large proportion of them are specialist feed-

ers (Hulme and Benkman, 2004). Th e omnivorous ground beetle P. oblongopunctatus predates on many small animals, most frequently on

Coleoptera, Diptera, Collembola, Acari etc. (Larochelle, 1990), as well

as on conifer seeds (Heikkilä, 1977; Nystrand and Granström, 2000).

Probably, the species feeds on seeds of various plants but respective data

are not available. Selective seed predation has been observed in many other

ground beetles, however (Toft and Bilde, 2002; Tooley and Brust, 2002).

To discriminate between toxic and nontoxic seeds, P. oblongopunctatus

possesses a specifi c feeding deterrent neuron innervating its antennal

chaetoid taste sensilla (V). A strong stimulatory eff ect of quinine (qui-

noline group) and quinine hydrochloride to this neuron was observed.

Th e threshold concentration for the latter at 0.01 mmol l-1 was close to

the upper limit of the threshold concentrations for the plant secondary

compounds that stimulate a deterrent neuron in other herbivorous insects

(Schoonhoven and van Loon, 2002; Chapman, 2003; Schoonhoven et al.,

2005), probably due to the fact that toxin concentration in seeds is usually

much higher than in other parts of the plant (Janzen, 1971; Hulme and

Benkman, 2004; Davis et al., 2008). Th e response of the neuron to qui-

nine hydrochloride was concentration dependent with maximum rates of

fi ring reaching up to 70 spikes/s at 50 mmol l-1. In two-choice behavioural

experiments, quinine hydrochloride had a strong concentration dependent

deterrent eff ect on feeding of P. oblongopunctatus beetles with the response

threshold at 1 mmol l-1. Th us, the threshold concentration of this alkaloid

to evoke a behavioural response was a hundred times higher than that

for an electrophysiological response of the alkaloid-sensitive neuron. In

contrast to quinine hydrochloride, the response of the alkaloid-sensitive

neuron to caff eine (purine group) was very weak, only 2.4 spikes/s at 10

mmol l-1 concentration. Th e eff ect of another quinoline alkaloid, strych-

nine, and sinigrin, to the neuron was rather inhibitory. No response of

the neuron to tested pyrrolidine alkaloid, nicotine, and salicin (glucoside)

was observed suggesting that the antennal alkaloid-sensitive neuron of P.

oblongopunctatus is not equally sensitive to all plant defensive chemicals.

Nevertheless, the list of toxic compounds capable of stimulating the an-

tennal alkaloid-sensitive neuron of P. oblongopunctatus is likely to grow

longer as more plant defence chemicals are tested (V).

33

Quinoline alkaloids are among the most potent insect feeding deter-

rents occurring mainly in Rutaceous and Rubiaceous plants. Th ey can be

present in virtually every part of the plant or just a specifi c part like the

bark, rhizome, leaf or seed (Openshaw, 1967; Michael, 2003; Aniszewski,

2007). Th ese alkaloids represent a large and structurally diverse group

of plant defensive compounds. When contacted, some of them may be

detected by stimulation of a specifi c antennal alkaloid-sensitive neuron of

the ground beetles predating on seeds, similarly to the quinine-sensitivity

of P. oblongopunctatus. Th e ability of granivores to perceive plant defensive

compounds allows them to avoid toxic seeds, and survive. High sensitiv-

ity of P. oblongopunctatus to plant sugars, commonly known as strong

phagostimulants for most phytophagous insects, seems to be also related

to its partially granivorous feeding habits. Th e disaccharide maltose was

one of the most stimulatory sugars for the sugar-sensitive neuron also

innervating antennal taste sensilla of P. oblongopunctatus (IV).

5.3.2. Peripheral inhibition of the salt- and pH-sensitive neurons by some plant secondary compounds

While stimulating the alkaloid-sensitive neuron, quinine hydrochloride strongly reduced spike production by the salt- and pH-sensitive neurons which also innervate these sensilla in Pterostichus spp. (Merivee et al., 2004; I, II). Increasing the quinine hydrochloride concentration, while keeping the concentration of NaCl constant, caused a dose-dependent decrease in spike generation by the salt-sensitive neuron. At 10 mmol l-1 and higher concentrations of quinine hydrochloride, predominant fi ring of the alkaloid-sensitive neuron was observed while the fi ring rate of other neu-rons dropped almost to zero. It was also observed that several tested plant secondary compounds, salicin, sinigrin, caff eine and nicotine, which had only little or no eff ect on the fi ring rate of the alkaloid-sensitive neuron, would greatly reduce the responses of the salt- and pH-sensitive neurons of the antennal taste sensillum of P. oblongopunctatus. Th ese results suggest that the antennal taste sensilla in P. oblongopunctatus may detect toxic and deterrent plant secondary compounds not only through activation of the specifi c alkaloid-sensitive neuron but also through inhibition of taste neurons activated by salts and stimulus pH; this is similar to deterrent inhibition of taste neurons activated by sugars and water in Drosophila (Meunier et al., 2003). Since foodplant acceptability to herbivores depends on the balance between stimulation and deterrency, similar interactions

34

between stimulating chemicals at peripheral taste neurons of herbivorous insects are of widespread occurrence. Plant toxins and deterrents may inhibit sugar neurons (Frazier, 1986; van Loon, 1990; Messchendorp et al., 1996; Bernays and Chapman, 2000; Schoonhoven and van Loon, 2002; Meunier et al., 2003; Schoonhoven et al., 2005), water neurons (Meunier et al., 2003), and sugars and salts may inhibit deterrent neurons (Simmonds and Blaney, 1983; Shields and Mitchell, 1995; Glendinning et al., 2000; Schoonhoven and van Loon, 2002; Schoonhoven et al., 2005). Deterrent compounds that on their own do not stimulate any neuron within a sensillum may also decrease the responsiveness of a neuron re-sponding to a nutrient, as exemplifi ed by sinigrin inhibiting the inositol neuron in Heliothis virescens and H. subfl exa (Bernays and Chapman, 2000; Schoonhoven and van Loon, 2002).

Survival of most terrestrial animals requires ability to detect environmental

NaCl, a nutrient essential for fl uid and electrolyte homeostasis and for a

multitude of other physiological processes (Lindemann, 1996; Contreras

and Lundy, 2000). Th e gustatory system allows insects to detect and ingest

salt, to discriminate between diff erent salts, and to avoid high salt concen-

trations. However, there are only few data that relate input from inorganic

salts to feeding behaviour in phytophagous insects. In some insects, salts

have a deterrent eff ect on feeding (Chapman, 2003; Schoonhoven and

van Loon, 2002). In contrast, larvae of Drosophila show dose-dependent

responses to NaCl: the appetitive responses to low concentration gradu-

ally turn into aversion as concentration is increased up to 200 mmol l-1

and higher (Miyakawa, 1982; Liu et al., 2003; Niewalda et al., 2008).

Some other insects also respond positively to dilute salts. Cockroaches

and housefl ies prefer dilute sodium and potassium chlorides and refuse

distilled water. In the two-choice situation, many animals give a bimodal

response to sodium chloride. Acceptance at low concentrations switches to

rejection at high concentrations (Dethier, 1977). In these cases, inorganic

salt at low concentration may act as a phagostimulant. Th erefore, NaCl

at 10 mmol l-1 may also have a phagostimulatory eff ect in P. oblongopunc-tatus although no behavioural data is available. Th is assumption could

explain the inhibition of spike production of the salt-sensitive neuron

by some toxic compounds in P. oblongopunctatus (V) as being similar to

how plant deterrents inhibit activity of other phagostimulatory neurons

in other phytophagous insects.

35

6. CONCLUSIONS

1. For the fi rst time, a specifi c pH-sensitive chemoreceptor neuron was

identifi ed in insect taste sensilla, most probably related to habitat and

microhabitat selection of the species (P. aethiops and P. oblongopunc-tatus; I, II);

2. For the fi rst time, a sugar-sensitive chemoreceptor neuron responding

to 1–1000 mmol l-1 sucrose and glucose was identifi ed in the taste

sensilla of ground beetles. Th e phasic-tonic response of the neuron

to these sugars was concentration dependent (P. aethiops, III);

3. Th e disacharides with an a-1.4-glycoside linkage, sucrose and maltose,

were the two most stimulatory sugars out of 12 tested for the antennal

sugar-sensitive neuron of P. oblongopunctatus evoking 70 spikes/s at

1000 mmol l-1. Th e stimulatory eff ect of glucose was approximately

two times lower. Due to the partial herbivory of P. oblongopunctatus

these plant sugars are probably involved in its search for food, for

example, for conifer seeds (IV);

4. None of the water-soluble sugars released by brown-rot fungi during

enzymatic wood decay (cellobiose, arabinose, xylose, mannose, rham-

nose and galactose) stimulated the antennal sugar-sensitive neuron in

P. oblongopunctatus (IV);

5. Th e weak stimulating eff ect (below 3 spikes/s) of some 100 mmol

l-1 amino acids (methionine, serine, alanine, glutamine) to the 4th

chemosensory neuron of the antennal taste sensilla of P. oblongopunc-tatus was characterized as non-specifi c, or modulating the responses

of non-target chemosensory neurons (IV);

6. For the fi rst time, the occurrence of an alkaloid-sensitive feeding

deterrent neuron in the antennal taste sensilla of ground beetles (P.

oblongopunctatus) was documented (V);

7. Some tested compounds (quinine and quinine hydrochloride) that

stimulate the alkaloid-sensitive neuron in P. oblongopunctatus may

strongly inhibit activity of other neurons (salt- and pH-sensitive neu-

ron) of the sensillum (V);

36

8. Several tested plant secondary compounds, salicin, sinigrin, caff eine

and nicotine, which had only little or no eff ect on the fi ring rate of

the alkaloid-sensitive neuron, through peripheral inhibition, would

greatly reduce the responses of other chemoreceptor neurons of the

sensillum in P. oblongopunctatus (V).

37

REFERENCES

Albert, P.J., 2003. Electrophysiological responses to sucrose from a gusta-

tory sensillum on the larval maxillary palp of the spruce budworm,

Choristoneura fumiferana (Clem.) (Lepidoptera: Tortricidae). Journal

of Insect Physiology 49, 733–738.

Amakawa, T., 2001. Eff ects of age and blood sugar levels on the proboscis

extension of the blow fl y Phormia regina. Journal of Insect Physiology

47, 195–203.

Andersen, J., Skorping, A., 1991. Parasites of carabid beetles: prevalence

depends on habitat selection of the host. Canadian Journal of Zool-

ogy 69, 1216–1220.

Anizewski, T., 2007. Alkaloids–secrets of life. Alkaloid chemistry, biologi-

cal signifi cance, applications and ecological role. Elsevier, Amsterdam,

Boston, Heidelberg, London, New York, Oxford, Paris, San Diego,

San Francisco, Singapore, Sydney, Tokyo, P 334.

Bernays, E.A., Chapman, R.F., 2000. A neurophysiological study of sen-

sitivity to a feeding deterrent in two sister species of Heliothis with

diff erent diet breadths. Journal of Insect Physiology 46, 905–912.

Bernays, E.A., Chapman, R.F., 2001a. Electrophysiological responses

of taste cells to nutrient mixtures in the polyphagous caterpillar of

Grammia geneura. Journal of Comparative Physiology A 187, 205–213.

Bernays, E.A., Chapman, R.F., 2001b. Taste cell responses in the poly-

phagous arctiid, Grammia geneura: towards a general pattern for cat-

erpillars. Journal of Insect Physiology 47, 1029–1043.

Bernays, E.A., Glendinning, J.I., Chapman, R.F., 1998. Plant acids modu-

late chemosensory responses in Manduca sexta larvae. Physiological

Entomology 23, 193–201

Blake, S., McCracken, D.I., Eyre, M.D., Garside, A., Foster, G.N., 2003.

Th e relationship between the classifi cation of Scottish ground bee-

tle assemblages (Coleoptera, Carabidae) and the national vegetation

classifi cation of British plant communities. Ecography 26, 602–616.

Blaney, W.M., 1974. Electrophysiological responses of the terminal sensilla

on the maxillary palps of Locusta migratoria L. to some electrolytes

and non-electrolytes. Journal of Experimental Biology 60, 275–293.

Chapman, R.F., 1998. Th e Insects. Structure and Function. 4th edn.

Cambridge University Press, Cambridge, P 770.

Chapman, R.F., 2003. Contact chemoreception in feeding by phytopha-

gous insects. Annual Review of Entomology 48, 455–484.

38

Cole, L.J., McCracken, D.I., Dennis, P., Downie, I.S., Griffi n, A.L., Foster,

G.N., Murphy, K.J., Waterhouse, T., 2002. Relationships between

agricultural management and ecological groups of ground beetles

(Coleoptera: Carabidae) on Scotish farmland. Agriculture, Ecosystems

and Environment 93, 323–336.

Contreras, R.J., Lundy, R.F., 2000. Gustatory neuron types in the pe-

riphery: a functional perspective. Physiology and Behavior 69, 41–52.

Daly, P.J., Ryan, M.F., 1979. Ultrastructure of antennal sensilla of Nebria brevicollis (Fab.) (Coleoptera: Carabidae). International Journal of

Insect Morphology and Embryology 8, 169–181.

Davis, A.S., Schutte, B.J., Iannuzzi, J., Renner, K. A., 2008. Chemical and

physical defense of weed seeds in relation to soil seedbank persistence.

Weed Science 56, 676–684.

Den Otter, C.J., 1972. Diff erential sensitivity of insect chemoreceptors

to alkali cations. Journal of Insect Physiology 18, 109–131.

Després, L., David, J.-P., Gallet, C., 2007. Th e evolutionary ecology of

insect resistance to plant chemicals. Trends in Ecology and Evolution

22, 298–307.

Dethier, V.G., 1977. Th e taste of salt. American Science 65, 744–751.

Dethier, V.G., 1980. Evolution of receptor sensitivity to secondary plant

substances with special reference to deterrents. American Naturalist

115, 45–66.

Dethier, V.G., Evans, D.R., Rhoades, M.V., 1956. Some factors control-

ling the ingestion of carbohydrates by the blowfl y. Biological Bulletin

111, 204–222.

Dethier, V.G., Hanson, F.E., 1968. Electrophysiological responses of the

blowfl y to sodium salts of fatty acids. Proceedings of the National

Academy of Sciences (U.S.A.) 60, 1269–1303.

Elizarov, Y.A., 1978. Insect Chemoreception. Moscow University Press,

Russia (in Russian).

Eriksson, K.-E.L., Blanchette, R.A., Ander, P., 1990. Microbial and Enzy-

matic Degradation of Wood Components. Springer-Verlag, New York.

Evans, D.R., Mellon, D., 1962. Stimulation of a primary taste receptor

by salts. Journal of General Physiology 45, 651–661.

Evans, W.G., 1982. Oscillatoria sp. (Cyanophyta) mat metabolites im-

plicated in habitat selection in Bembidion obtusidens (Coleopt.: Car-

abidae). Journal of Chemical Ecology 8, 671–678.

Evans, W.G., 1983. Habitat selection in the Carabidae. Th e Coleopterists’

Bulletin 37, 164–167.

39

Evans, W.G., 1984. Odour-mediated responses of Bembidion obtusidens (Coleoptera: Carabidae) in a wind tunnel. Canadian Journal of En-tomology 116, 1653–1658.

Evans, W.G., 1988. Chemically mediated habitat recognition in shore insects (Coleoptera: Carabidae, Hemiptera: Saldidae). Journal of Chemical Ecology 14, 1441–1454.

Eyre, M.D., Luff , M.L., 2004. Ground beetle species (Coleoptera, Car-abidae) associations with land cover variables in northern England and southern Scotland. Ecography 27, 417–426.

Frazier, J.L., 1986. Th e perception of plant allelochemicals that inhibit feeding. In: Brattsten, L.B., Ahmad, S. (Eds.), Molecular aspects of insect-plant associations. Plenum Press, New York, pp. 1–42

Glendinning, J.I., 2002. How do herbivorous insects cope with noxious secondary plant compounds in their diet? Entomologia Experimentalis et Applicata 104,15–25.

Glendinning, J.I., Nelson, N., Bernays, E.A., 2000. How do inositol and glucose modulate feeding in Manduca sexta caterpillars? Journal of Experimental Biology 203, 1299–1315.

Grove, S.J., 2002. Saproxylic insect ecology and the sustainable man-agement of forests. Th e Annual Review of Ecology, Evolution, and Systematics 33, 1–23.

Haberman, H., 1968. Ground beetles of Estonia. Valgus, Tallinn, P 589 (in Estonian).

Hansen-Delkeskamp, E., 1998. Development of specifi c responses in antennal taste hairs after ecdysis. An electrophysiological investigation of the cockroach, Periplaneta brunnea. Journal of Insect Physiology 44, 659–666.

Hanson, F.E., 1987. Chemoreception in the fl y: the search for the liver-wurst receptor. In: Chapman, R.F., Bernays, E.A., Stoff olano, J.G. (Eds.), Perspectives in Chemoreception and Behaviour. Springer, Germany, pp. 99–122.

Harborne, J.B., 1993. Introduction to Ecological Biochemistry. 4th edn. Academic Press, London, San Diego, P. 318.

Haskell, P.T., Schoonhoven, L.M., 1969. Th e function of certain mouth part receptors in relation to feeding in Schistocerca gregaria and Locusta migratoria migratorioides. Entomologia Experimentalis et Applicata 12, 429–440.

Hatzold, T., Elmadfa, I., Gross, R., Wink, M., Hartmann, T., Witte, L., 1983. Quinolizidine alkaloids in seeds of Lupinus mutabilis. Journal of Agricultural and Food Chemistry 31, 934–930.

40

Heessen, H.J.L., 1981. Egg mortality in P. oblongopunctatus (Coleoptera, Carabidae). Oecologia, 50, 233–235.

Heikkilä, R., 1977. Eläimet kylvetyn männyn ja kuusen siemenen tuhooji-na Pohjois-Suomessa (Summary: Destruction caused by animals to sown pine and spruce seed in northern Finland). Metsäntutkimus-laitoksen Julkaisuja 89, 1–35.

Hengeveld, R., 1980. Polyphagy, oligophagy and food specialization in ground beetles (Coleoptera, Carabidae). Netherlands Journal of Zool-ogy 30, 564–584.

Hoback, W.W., Golick, D.A., Svatos, T.M., Spomer, S.M., Higley, L.G., 2000. Salinity and shade preferences result in ovipositional diff er-ences between sympatric tiger beetle species. Ecological Entomology 25, 180–187.

Hodgson, E.S., Lettvin, J.Y., Roeder, K.D., 1955. Physiology of a primary chemoreceptor unit. Science 122, 417–418.

Honek, A., Martinkova, Z., Jarosik, V., 2003. Ground beetles (Carabidae) as seed predators. European Journal of Entomology 100, 531–544.

Hulme, P.E., Benkman, C.W., 2004. Granivory. In: Herrera, C.M., Pell-myr O. (Eds.), Plant-animal interactions. An evolutionary approach. Blackwell Science, Oxford, pp. 132–154

Ings, T.C., Hartley, S.E., 1999. Th e eff ect of habitat structure on carabid communities during the regeneration of a native Scotish forest. Forest Ecology and Management 119, 123–136.

Irmler, U., 2001. Charakterisierung der Laufkäfergemeinschaften schleswigholsteinischer Wälder und Möglichkeiten ihrer ökologischen Bewertung. Angewandte Carabidologie, Supplement 2. Laufkäfer im Wald, 21–32.

Irmler, U., 2003. Th e spatial and temporal pattern of carabid beetles on arable fi elds in northern Germany (Schleswig–Holstein) and their value as ecological indicators. Agriculture, Ecosystems and Environ-ment 98, 141–151.

Ishikawa, S., 1963. Responses of maxillary chemoreceptors in the larvae of the silkworm, Bombyx mori, to stimulation by carbohydrates. Journal of Cellular and Comparative Physiology 61, 99–107.

Ishikawa, S., 1966. Electrical response and function of bitter substance receptor associated with the maxillary sensilla of the larva of the silk-worm, Bombyx mori L. Journal of Cellular and Comparative Physiol-ogy 67, 1–12.

Janzen, D.H., 1971. Seed predation by animals. Annual Review of Ecol-

ogy, Evolution, and Systematics 2, 465–492.

41

Johnson, U.E., Cameron, R.S., 1969. Phytophagous ground beetles. An-nals of the Entomological Society of America 62, 909–914

Jolivet, P., 1998. Interrelationships between insects and plants. CRC Press, Boca Raton, Florida, P. 309.

Jørgensen, K., Almaas, T.J., Marion-Poll, F., Mustaparta, H., 2007. Elec-trophysiological characterization of responses from gustatory receptor neurons of sensilla chaetica in the moth Heliothis virescens. Chemical Senses 32, 863–879.

Jørgensen, K., Kvello, P., Almaas, T.J., Mustaparta, H., 2006. Two closely located areas in the suboesophageal ganglion and the tritocerebrum receive projections of gustatory receptor neurones located on the an-tennae and the proboscis in the moth Heliothis virescens. Th e Journal of Comparative Neurology 496, 121–134.

Kielty, J.P., Allen-Williams, L.J., Underwood, N., Eastwood, E.A., 1996. Behavioural responses of three species of ground beetle (Coleoptera: Carabidae) to olfactory cues associated with prey and habitat. Journal of Insect Physiology 9, 237–250.

Kim, J.L., Yamasaki, T., 1996. Sensilla of Carabus (Isiocarabus) fi duciarius saishutoicus Csiki (Coleoptera: Carabidae). International Journal of Insect Morphology and Embryology, 25, 153–172.

Krogerus, R., 1960. Ökologische Studien über nordische Moorarthro-poden. Societas Scientiarum. Fennica, Commentationes Biologicae, 21 (3), 1–238.

Larochelle, A., 1990. Th e food of the carabid beetles (Coleoptera: Carabi-dae, including Cicindelinae). Fabreries, Supplément 5, 1–132.

Lindemann, B., 1996. Taste reception. Physiological Reviews 76, 719–766.Lindroth, C.H., 1985. Th e Carabidae (Coleoptera) of Fennoscandia and

Denmark. Fauna Entomologica Scandinavica 15, 1–225.Lindroth, C.H., 1986. Th e Carabidae (Coleoptera) of Fennoscandia and

Denmark. Fauna Entomologica Scandinavica 15, 233–497.Liscia, A., Crnjar, R., Masala, C., Sollai, G., Solari, P., 2002. Sugar re-

ception in the blowfl y: a possible Ca++ involvement. Journal of Insect Physiology 48, 693–699.

Liscia, A., Solari, P., 2000. Bitter taste recognition in the blowfl y: elec-trophysiological and behavioral evidence. Physiology and Behaviour 70, 61–65.

Liscia, A., Solari, P., Majone, R., Tomassini Barbarossa, I., Crnjar, R., 1997. Taste receptor mechanisms in the blowfl y: evidence of amiloride-sensitive and insensitive receptor sites. Physiology and Behaviour 62, 875–879.

42

Liu, L., Leonard, A.S., Motto, D.G., Feller, M.A., Price, M.P., Johnson, W.A., Welsh, M.J. 2003. Contribution of Drosophila DEG/ENaC

Genes to Salt Taste. Neuron 39, 133–146.

Lövei, G.L., Sunderland, K.D., 1996. Ecology and behavior of ground

beetles (Coleoptera: Carabidae). Annual Review of Entomology 41,

231–256.

Magura, T., 2002. Carabids and forest edge: spatial pattern and edge

eff ect. Forest Ecology and Management 157, 23–37.

Merivee, E., Ploomi, A., Luik, A., Rahi, M., Sammelselg, V., 2001. An-

tennal sensilla of the ground beetle Platynus dorsalis (Pontoppidan,

1763) (Coleoptera, Carabidae). Microscopy Research and Technique

55, 339–349.

Merivee, E., Ploomi, A., Rahi, M., Bresciani, J., Ravn, H. P., Luik, A.,

Sammelselg, V., 2002. Antennal sensilla of the ground beetle Bembid-ion properans Steph. (Coleoptera, Carabidae). Micron 33, 429–440.

Merivee, E., Ploomi, A., Rahi, M., Luik, A., Sammelselg, V., 2000. Anten-

nal sensilla of the ground beetle Bembidion lampros Hbst (Coleoptera,

Carabidae). Acta Zoologica (Stockholm) 81, 339–350.

Merivee, E., Renou, M., Mänd, M., Luik, A., Heidemaa, M., Ploomi A.,

2004. Electrophysiological responses to salts from antennal chaetoid

taste sensilla of the ground beetle Pterostichus aethiops. Journal of

Insect Physiology 50, 1001–1013.

Messchendorp, L., van Loon, J.J.A., Gols, G.J.Z., 1996. Behavioural and

sensory responses to drimane antifeedants in Pieris brassicae larvae.

Entomologia Experimentalis et Applicata 79, 195–202.

Meunier, N., Marion-Poll, F., Rospars, J.-P., Tanimura, T., 2003. Periph-

eral coding of bitter taste in Drosophila. Journal of Neurobiology 56,

139–152.

Michael, J.P., 2003. Quinoline, quinazoline and acridone alkaloids. Natu-

ral Product Reports Articles 20, 476–493.

Mitchell, B.K., Gregory, P., 1979. Physiology of the maxillary sugar sensi-

tive cell in the red turnip beetle, Entomoscelis americana. Journal of

Comparative Physiology 132, 167–178.

Mitchell, B.K., Schoonhoven, L.M., 1973. Taste receptors in Colorado

beetle larvae. Journal of Insect Physiology 20, 1787–1793.

Mitchell, B.K., Seabrook, W.D., 1973. Electrophysiological investigations

on tarsal chemoreceptors of the spruce budworm, Choristoneura fumif-erana (Lepidoptera). Journal of Insect Physiology, 20, 1209–1218.

43

Miyakawa, Y., 1982. Behavioural evidence for the existence of sugar, salt

and amino acid taste receptor cells and some of their properties in

Drosophila larvae. Journal of Insect Physiology 28, 405–410.

Morita, H., Shiraishi, A., 1985. Chemoreception physiology. In: Kerkut,

G.A., Gilbert, L.I. (Eds.), Comprehensive Insect Physiology, Bio-

chemistry and Pharmacology. Pergamon Press, Oxford, pp. 133–170.

Niemelä, J., Spence, J. R., Langor, D., Haila, Y., Tukia, H., 1994. Logging

and boreal ground-beetle assemblages on two continents: implica-

tions for conservation. In: Gaston, K., Samways, M., New, T. (Eds.),

Perspectives in insect conservation. Intercept Publications, Andover,

pp. 29–50

Niewalda, T., Singhal, N., Fiala, A., Saumweber, T., Wegener S., Gerber,

B., 2008. Salt Processing in Larval Drosophila: Choice, Feeding,

and Learning Shift from Appetitive to Aversive in a Concentration-

Dependent Way. Chemical Senses 33, 685–692.

Nystrand, O., Granström, A., 2000. Predation on Pinus sylvestris seeds

and juvenile seedlings in Swedish boreal forests in relation to stand

disturbance by logging. Journal of Applied Ecology 37, 449–463.

Odell, L.H., Kirmeyer, G.J., Wilczak, A., Jasangelo, J.G., Marcinko, J.P.,

Wolfe, R.L., 1996. Controlling nitrifi cation in chloraminated systems.

Journal AWWA 88, 86–98.

Openshaw, H.T., 1967. Quinoline alkaloids other than those of Cinchona.

In Manske R.H.F. (Ed.), Th e Alkaloids. Chemistry and Physiology.

Vol. IX. Academic Press, New York, London, pp. 223–267.

Paje, F., Mossakowski, D., 1984. pH-preferences and habitat selection in

carabid beetles. Oecologia (Berlin) 64, 41–46.

Purtauf, T., Roschewitz, I., Dauber, J., Th ies, C., Tscharntke, T., Wolters,

V., 2005. Landscape context of organic and conventional farms: infl u-

ences on carabid beetle diversity. Agriculture, Ecosystems and Envi-

ronment 108, 165–174.

Reid, I.D., 1985. Biological delignifi cation of aspen wood by solid-state

fermentation with the white-rot fungus Merulius tremellosus. Applied

and Environmental Microbiology 50, 133–139.

Rüth, E., 1976. Elektrophysiologie der Sensilla Chaetica auf den Anten-

nen von Periplaneta americana. Journal of Comparative Physiology

105, 55–64.

Ryan, M.F., 2002. Insect Chemoreception. Fundamental and Applied.

Kluwer Academic Publishers, New York, Boston, Dordrecht, London,

Moscow, P. 330.

44

Sandoval, M., Albert, P.J., 2007. Chemoreception of sucrose and amino

acids in second and fourth instars of the spruce budworm Choristo-

neura fumiferana (Clem.) (Lepidoptera: Tortricidae). Journal of Insect

Physiology 53, 84–92.

Schoonhoven, L.M., van Loon, J.J.A., 2002. An inventory of taste in

caterpillars: each species its own key. Acta Zoologica Academiae Sci-

entiarum Hungaricae 48, 215–263.

Schoonhoven, L.M., van Loon, J.J.A., Dicke, M., 2005. Insect-Plant

Biology. 2nd edn.. Oxford University Press, Oxford, P. 421

Shields, V.D.C., Mitchell, B.K., 1995. Responses of maxillary stylocon-

ic receptors to stimulation by sinigrin, sucrose and inositol in two

crucifer-feeding, polyphagous lepidopterous species. Philosophical

Transactions of the Royal Society London 347, 447–457.

Silfverberg, H., 2004. Enumeratio nova Coleopterorum Fennoscandiae,

Daniae et Baltiae. Sahlbergia 9, 1–111.

Simmonds, M.S.J., Blaney, W.M., 1983. Some neurophysiological eff ects

of azadirachtin on lepidopterous larvae and their feeding responses.

In: Schmutterer, H., Ascher, K.R.S. (Eds.), Proceedings of the Second

International Neem Conference, G.T.Z., Eschborn, pp. 163–180

Zabel, R.A., Morrell, J.J., 1992. Wood Microbiology Decay and Its Pre-

vention. Academic Press, San Diego, pp. 195–224.

Th iele, H.-U., 1977. Carabid Beetles in Th eir Environment. Zoophysi-

ological Ecology 10.Springer, Berlin, P 369

Toft, S., Bilde, T. 2002. Carabid diets and food value. In: Holland, J.M.

(Ed.), Th e Agroecology of Carabid Beetles. Intercept, Andover, UK,

pp. 81–110.

Tooley, J., Brust, G. 2002. Weed seed predation by carabid beetles. In:

Holland, J.M. (Ed.), Th e Agroecology of Carabid Beetles. Intercept,

Andover, UK, pp. 215–229.

Valaškova, V., Baldrian, P., 2006. Degradation of cellulose and hemicel-

luloses by the brown rot fungus Piptoporus betulinus—production,

of extracellular enzymes and characterization of the major cellulases.

Microbiology 152, 3613–3622.

Van Loon, J.J.A., 1990. Chemoreception of phenolic acids and fl avonoids

in larvae of two species of Pieris. Journal of Comparative Physiology

A 166, 889–899.

Wang, Z., Singhvi, A., Kong, P. and Scott, K. 2004. Taste representations

in the Drosophila brain. Cell 117, 981–991.

45

Varela, E., Tien, M., 2003. Eff ect of pH and oxalate on hydroquinone-

derived hydroxyl radical formation during brown rot wood degrada-

tion. Applied and Environmental Microbiology 69, 6025–6031.

Wheater, C.P., 1989. Prey detection by some predatory Coleoptera (Car-

abidae and Staphylinidae). Journal of Zoology A 218, 171–185.

46

SUMMARY IN ESTONIAN

JOOKSIKLASTE (COLEOPTERA: CARABIDAE)

ANTENNAALNE KONTAKTNE KEMORETSEPTSIOON

Rööv- ja segatoidulised jooksiklased oma suure liigirikkuse ja arvukuse tõttu omavad arvestatavat majanduslikku tähtsust kui põllu- ja aiakahjurite arvukuse olulised looduslikud reguleerijad.

Mitme- ja segatoiduliste jooksiklaste jaoks leidub põldudel alati piisavalt toitu. Jooksiklaste peamisteks toiduobjektideks on lehetäid, hooghänna-lised, liblikate ja mardikate munad, vastsed ning valmikud, kahetiivaliste munad ja vastsed ning vaablaste ebaröövikud. Suuremate liikide toiduks sobivad ka vihmaussid ja teod. Suure osa sega- ja taimtoiduliste liikide toi-dumenüüsse kuuluvad taimede, sealhulgas umbrohtude seemned. Just selle tõttu võivad jooksiklased erinevalt kitsalt spetsialiseerunud röövtoidulistest putukatest ja parasitoididest migreeruda põldudele varakevadel, enne kui sinna ilmuvad kultuurtaimede kahjurid. Hävitades kahjureid arvukuse tõusu algfaasis hoitakse ära nende hilisem massesinemine ja majanduslik kahju. Seepärast on mahetootmise ja integreeritud kahjuritõrje tingimustes manipulatsioonid jooksiklaste liigirikkuse ja arvukusega tähtsal kohal.

Jooksiklaste seotus teatud elupaigatüüpide ning varje- ja talvituskohtadega on määratud mitmesuguste biootiliste ja abiootiliste teguritega nagu toitu-mistingimused, taimkattetüüp, konkurentide olemasolu ja levik, aastaaeg, maastiku iseärasused, maaharimise mõju, temperatuuri, valgus- ja niiskus-tingimused jne. Pinnase sooladesisaldus ja pH võivad olla samuti tähtsad tegurid jooksiklaste levikul. Käitumiskatsetega on näidatud, et vastavad kontaktsed kemoretseptoosed sensillid ehk maitseharjased paiknevad nende mardikate tundlatel. Paljude jooksiklaste elupaigavalik on nii spetsiifi line, et sageli kasutatakse neid elupaikade iseloomustamiseks. Välisstiimulid ja otsingulise käitumise mehhanismid, mille abil jooksiklased oma toidu asukoha kindlaks teevad ja saagirikastele aladele peatuma jäävad, pole täpselt teada. Samal ajal, kui paljud liigid leiavad oma toidu nähtavasti juhusliku otsingu teel, kütivad mitmed teised, päevase aktiivsusega liigid, saaki nägemise abil. Paljud liigid kasutavad aga toiduga seotud lõhna-signaale. Keemilise informatsiooni kasutamine on jooksiklastel ilmselt tunduvalt rohkem levinud kui väheste avaldatud andmete põhjal võiks arvata. Jooksiklaste käitumist reguleerivaid keemilisi välisstiimuleid ning neid vastuvõtvate kemoretseptorite talitlust on seni üllatavalt vähe uuritud.

47

Elektronmikroskoopilised uuringud on näidanud, et jooksiklaste tundlatel

esineb tuhandeid erinevaid kemoretseptoorseid sensille. Lisaks arvukatele

pisikestele pulkjatele haistesensillidele, paikneb nende fl agellumil 56 kuni

70 suurt 35–200 μm pikkust harjasjat kontaktset kemosensilli. Kivijook-

sikul Nebria brevicollis on need sensillid varustatud nelja kemoretseptoorse

ja ühe mehhanoretseptoorse neuroniga. See on tüüpiline neuronite komp-

lekt enamiku putukarühmade kontaktsetes kemosensillides. Nendest

on jooksiklastel seni elektrofüsioloogiliselt kindlaks tehtud ainult ühe,

soolaneuroni, funktsioon. Ülejäänud kolme kemoretseptoorse neuroni

täpsem funktsioon pole teada. Kontaktset kemoretseptsiooni on põhjali-

kumalt uuritud paljudel taimtoidulistel putukatel, kellel need neuronid

võivad reageerida väga mitmesugustele taimsetele keemilistele stiimulitele,

enamasti aga toitumise stimulantidele (taimsed suhkrud, aminohapped,

suhkuralkoholid jt.) ja deterrentsetele ühenditele (alkaloidid, glükosiidid

jt.). Kuna stimulantsetel ja deterrentsetel ühenditel on toidu- ja peremees-

taime äratundmisel keskne roll, on nad viimastel aastakümnetel pälvinud

üha suuremat tähelepanu.

Käesoleva töö eesmärgiks on elektrofüsioloogilise meetodi abil selgitada

antennaalseid kontaktseid kemoretseptoreid (maitseharjaseid) innerveeri-

vate neuronite funktsiooni ümarselg-süsijooksikul (Pterostichus aethiops) ja

metsa-süsijooksikul (P. oblongopunctatus). Erinevate meetoditega (puhver-

lahuste seeriad, leelistatud ja leelistamata soolad) testiti ärrituslahuste pH

mõju nende neuronite reaktsioonile ning spetsiifi lise pH-tundliku neuroni

võimalikku esinemist nende liikide antennaalsetes maitseharjastes. Kuna

mõlemad uuritavad liigid on osaliselt taimtoidulised (granivoorid), testiti

ka antennaalsete maitseneuronite reaktsiooni mitmesugustele taimsetele

suhkrutele ja aminohapetele, et selgitada nende ainete vastuvõtmisele

spetsialiseerunud neuronite olemasolu ja talitluse iseärasusi jooksiklaste

tundlatel. Olulise osa tööst moodustavad elektrofüsioloogilised ja etoloogi-

lised eksperimendid, mille käigus selgitati taimeseemnetes leiduvate toksi-

liste ja deterrentsete ainete (alkaloidid ja glükosiidid) stimuleerivat toimet

nimetatud jooksiklaste antennaalsete maitseharjaste kemosensoorsetele

neuronitele. Katsete eesmärgiks oli selgitada spetsiifi lise deterrendineuroni

olemasolu ja nende ainete võimalikku inhibeerivat toimet maitseharjase

teistele neuronitele. Nii ümarselg-süsijooksik kui ka metsa-süsijooksik

on videvikuaktiivsusega omnivoorsed metsaliigid, kelle toidumenüüs on

tähtsal kohal okaspuude seemned. Katsemardikad koguti Lõuna-Eesti

metsadest.

48

Elektrofüsioloogilistes katsetes mõõdeti spetsiaalse aparatuuri abil mardi-

kate antennaalseid maitseharjaseid innerveerivate neuronite reaktsioone

ärrituslahustele, kasutades sensilli tipust neuronite impulss-aktiivsuse

rakuvälise registreerimise meetodit. Selleks, et suurendada ärrituslahuste

elektrijuhtivust, lahustati suhkrud, aminohapped, alkaloidid ja glükosiidid

vajalikus kontsentratsioonis 10 mmol l-1 elektrolüüdi (NaCl, koliinklo-

riid) lahuses. Puhtad ained hangiti fi rmadelt AppliChem (Saksamaa) ja

Sigma-Aldrich. Lahuste pH taset mõõdeti portatiivse pH-meetri E6115

(Evikon, Estonia) abil. Neuronite reaktsiooni mõõdeti toimepotentsiaa-

lide (närviimpulsside) arvuna sekundis. Et testida elektrofüsioloogiliselt

aktiivsete alkaloidide deterrentset toimet mardikate toitumise aktiivsusele,

viidi läbi vajalikud käitumuslikud valikukatsed alkaloidiga töödeldud

ja töötlemata männiseemnetega. Katseandmete analüüsimisel kasutati

erinevaid statistilisi teste, kasutades arvutitarkvara Statistica (StatSoft,

Inc., USA).

Elektrofüsioloogilised katsed atsetaat- ja fosfaatpuhvrite seeriatega ning

leelistatud ja leelistamata sooladega pH vahemikus 3 kuni 11 näitasid, et

ümarselg-süsijooksiku ja metsa-süsijooksiku antennaalsetes maitseharjastes

paikneb kaks neuronit, soola- ja pH-tundlik neuron, mis reageerivad nende

elektrolüütide lahustele. pH-tundliku neuroni keskmine impulss-aktiiv-

sus tõusis alati koos ärrituslahuse pH-taseme tõusuga ulatudes aluselise

reaktsiooniga ärrituslahuste puhul (100 mmol l-1 NaCl + 10 mmol l-1

NaOH; pH 9,6) maksimaalselt kuni 30 imp/s, samal ajal kui soola-tund-

liku neuroni aktiivsus ei muutunud või muutus ainult vähesel määral.

Samas täheldati, et ka ärrituslahustes olevate soolade iooniline koostis

mõjutas märgatavalt pH-tundliku neuroni närviimpulsside sagedust.

Saadud tulemus on huvipakkuv, kuna varem pole putukate kontaktsetes

kemosensoorsetes sensillides spetsiifi lise pH-tundliku neuroni olemasolu

täheldatud. Uuritud jooksiklaste antennaalne pH-tundlik neuron näib

olevat seotud mardikate elupaiga ja talvituskohtade valikuga. pH-tundliku

neuroni impulssaktiivsus puudub või on madalseisus (alla 5 imp/s) pH

väärtustel 3 kuni 6. See happelisuse piirkond vastab pinnase pH tase-

mele uuritud jooksiklaste metsabiotoopides ja eelistatud talvituskohtades

pruunmädaniku poolt lagundatud puidus, kus mardikad veedavad 7–8

kuud, septembrist aprillini.

Käesoleva uurimuse raames identifi tseeriti esmakordselt granivoorsetel

lülijalgsetel (ümarselg-süsijooksikul ja metsa-süsijooksikul) elektrofüsio-

49

loogiliselt suhkrutundliku neuroni olemasolu. Katsed näitasid, et 12-st

testitud taimsest suhkrust 3 (maltoos, sahharoos ja glükoos) osutusid

füsioloogiliselt aktiivseteks, stimuleerides antennaalsete maitseharjaste

suhkrutundliku neuroni impulssaktiivsust. Tüübilt oli neuroni reakt-

sioon 1–1000 mmol l-1 suhkrulahusele faasilis-tooniline, sõltudes lahuse

kontsentratsioonist. Metsa-süsijooksikul osutusid 1,4-glükosiidsidemega

disahhariidid maltoos ja sahharoos füsioloogiliselt aktiivseimateks, tõstes

1000 mmol l-1 kontsentratsioonil suhkrutundliku neuroni impulss-aktiiv-

suse tasemele 70 imp/s. Glükoosi stimuleeriv toime oli ligi kaks korda

madalam. Kuna maltoosi leidub looduses peamiselt idanevates seemnetes

ja noortes idantites, võib oletada, et uuritud jooksiklaste suhkrutundlikkus

on seotud nende osalise taimtoidulisusega (mardikad toituvad okaspuude

seemnetest ja noortest idanditest). Ükski puidu ensümaatilise lagunemise

käigus tekkivast vees lahustuvast suhkrust (tsellobioos, arabinoos, ksüloos,

mannoos, ramnoos ja galaktoos) ei osutunud füsioloogiliselt aktiivseks,

mistõttu nad ei osale nähtavasti metsa-süsijooksiku otsingulises käitu-

mises. Mõned testitud seitsmest 100 mmol l-1 aminohappest (metioniin,

seriin, alaniin, glutamiin) avaldasid nõrka stimuleerivat toimet (alla 3

imp/s) metsa-süsijooksiku maitseharjase kemosensoorsetele neuronitele,

mida võib aga iseloomustada kui maitseharjase neuronite reaktsiooni

mittespetsiifi list modulatsiooni.

Antennaalsete maitseharjaste kemosensoorsete neuronite reaktsiooni taim-

setele alkaloididele ja glükosiididele testiti metsa-süsijooksikul. Katsete

tulemusena identifi tseeriti granivoorsetel lülijalgsetel esmakordselt alka-

loiditundlik neuron. Ilmnes, et testitud viiest alkaloidist avaldasid sellele

neuronile tugevat stimuleerivat toimet vaid kiniin ja selle sool kiniinhüd-

rokloriid. Alkaloiditundliku neuroni faasilis-toonilised reaktsioonid 0,001

kuni 50 mmol l-1 olid võrdelises sõltuvuses lahuse kontsentratsioonist.

Neuroni erutusläveks kiniinhüdrokloriidi suhtes osutus kontsentratsioon

0,01 mmol l-1, maksimaalseid reaktsioone (67 imp/s) täheldati kontsentrat-

sioonil 50 mmol l-1. Käitumiskatsetes osutus kiniinhüdrokloriid tugevaks

toitumise deterrendiks. Kofeiini stimuleeriv efekt sellele neuronile oli väga

nõrk (alla 2 imp/s), nikotiin aktiivsust ei mõjutanud, strühniin aga pärssis

nõrgalt selle neuroni impulssaktiivsust.

Lisaks kiniini ja kiniinhüdrokloriidi alkaloidineuronit stimuleerivale

toimele võivad nad tugevalt pärssida metsa-süsijooksiku maitseharja-

se soola- ja pH-tundliku neuroni impulssaktiivsust. Samuti ilmnes, et

50

glükosiidid sinigriin ja salitsiin ning alkaloidid kofeiin ja nikotiin, mille

mõju alkaloiditundlikule neuronile oli nõrk või puudus üldse, võivad

samuti tugevalt pärssida soola- ja pH-tundliku neuroni impulssaktiivsust.

Saadud tulemused lubavad oletada, et jooksiklase antennaalsed maitse-

harjased võivad taimseid toksilisi ja deterrentse toimega aineid avastada

nii alkaloiditundliku neuroni stimuleerimise kui ka maitseharjase teiste

kemosensoorsete neuronite periferaalse inhibitsiooni teel.

51

ACKNOWLEDGEMENTS

Th is study was carried out at the Institute of Agricultural and Environ-

mental Sciences of the Estonian University of Life Sciences. I would like

to express my gratitude to my supervisors Enno Merivee and Prof. Anne

Luik who has been supporting and helping me completing in this thesis.

I am grateful to Kaljo Voolma and Heino Õunap for their review of my

thesis and good comments on how to improve it.

I sincerely thank my colleagues in the Institute of Agricultural and En-

vironmental Sciences for their support.

Th e Estonian Science foundation (Grants No 5423 and 6958) fi nancially

supported this study.

I

PUBLICATIONS

Merivee, E., Ploomi, A., Milius, M., Luik, A., Heidemaa, M., 2005.

ELECTROPHYSIOLOGICAL IDENTIFICATION OF

ANTENNAL PH RECEPTORS IN THE GROUND BEETLE

PTEROSTICHUS OBLONGOPUNCTATUS.

Physiological Entomology 30(2), 122–33.

55

Electrophysiological identification of antennal pHreceptors in the ground beetle Pterostichusoblongopunctatus

ENNO MER I V E E 1 , A NGE LA P LOOM I 1 , MAR I T M I L I U S 1 , ANNE

LU IK 1 and M IKK HE I D EMAA 2

1Estonian Agricultural University, Institute of Plant Protection, Tartu and 2University of Tartu, Institute of Zoology and

Hydrobiology, Tartu, Estonia

Abstract. Electrophysiological responses of antennal taste bristles to 100mM

acetate and phosphate buffers were tested at pH 3–11 in the ground beetlePterostichus oblongopunctatus (F.) (Coleoptera, Carabidae). Additionally,responses of these sensilla to 10 and 100mM phosphate buffers were comparedwith each other. Generally, in response to these stimulating solutions, two sensorycells, classified as a salt cell (cation cell) and a pH cell, respectively, showed actionpotentials distinguished by differences in their amplitudes and polarity of spikes.The firing rate of the cation cell increased with increasing buffer concentration,and was influenced by buffer pH in a complicated way. The best stimulus for thesecond cell (pH cell) was pH of the stimulating buffer solution. As the pH of thestimulus solution increased, higher rates of firing were produced by the pH cell.For example, the number of action potentials elicited by 100mM phosphate bufferat pH11.1 was approximately 16-fold higher compared with that at pH 8.1, andfiring rates during the first second of the response were 27.9 and 1.7 imp/s,respectively. The pH cell did not fire or fired at very low frequency (first secondresponse below 5 imp/s) at pH3–6. This level of acidity probably represents thepH preferences of this ground beetle in its forest habitat and hibernating sites.By contrast to the cation cell, the pH cell responded to increases in bufferconcentration by decreasing its firing rate.

Key words. Action potentials, firing rate, taste sensilla, tip recording.

Introduction

The life of carabids from egg to death of the adult proceeds

in direct physical contact with the ground (from which the

name of the beetles is derived in English: ground beetles).

Due to scarce data, we can only assume that, in searching

and selection of their habitat, refugia and hibernating sites,

besides other stimuli, ground beetles perceive and analyse

external chemical information, such as soil pH, salt content

and concentration. However, some data obtained from

behavioural experiments concerning the ground beetles’

ability to perceive salt content and pH of the ground

are contradictory (Thiele, 1977). Nevertheless, Paje &

Mossakowski (1984) demonstrated convincingly that

behaviourally ground beetles have certain pH preferences

concerning their habitat; ablation experiments conducted

by these authors suggested that pH receptors are located

on the beetles’ antennae.

Scanning electron microscope studies demonstrate that

the most probable candidates for pH receptors (i.e. large

blunt-tipped sensilla) can be found abundantly on the

antennae of all ground beetles. These antennae consist of

a scapus, pedicel and nine flagellomeres. The taste receptors

stand in a whorl of six to seven bristles, in distal parts of the

flagellomeres. In addition, six bristles are located in pairs

Correspondence: Dr Enno Merivee, Estonian Agricultural

University, Institute of Plant Protection, 64 Kreutzwaldi Street,

51014 Tartu, Estonia. E-mail: [email protected]

Physiological Entomology (2005) 30, 122–133

122 # 2005 The Royal Entomological Society

symmetrically at the antennal tip (Kim & Yamasaki, 1996;

Merivee et al., 2000, 2001, 2002). According to Daly &

Ryan (1979), these taste bristles in the ground beetle

Nebria brevicollis are innerved with five neurones: four

chemoreceptor cells and one mechanoreceptor cell. In

electrophysiological experiments with acid, neutral and

alkaline salts, it was demonstrated that two chemoreceptor

cells of the antennal taste bristles respond to these salt

solutions in ground beetles. One of the cells, the salt cell,

was sensitive to ionic (cationic) content and concentration

of the stimulating solution. Firing rates of the second cell

were proportional to the pH level of the stimulating

salt solutions tested (Merivee et al., unpublished data).

Unfortunately, the tested salts were dissolved in distilled

water and not in buffer solutions. It is likely that the pH

level of the tested salt solutions was unstable and changed

during experiments. Thus, the function of the second salt

cell (pH cell) requires further testing using buffer-controlled

pH solutions. To date, electrophysiological evidence con-

cerning the occurrence of pH receptors in insects is lacking,

but it has been shown repeatedly that the pH of stimulating

solutions may affect neural responses of insect taste sensilla

(Shiraishi & Morita, 1969; Bernays et al., 1998).

The aim of the present study is to test responses of

antennal taste bristles to acetate and phosphate buffers in

a wide range of pH values (3–11) in the ground beetle

Pterostichus oblongopunctatus F. (Coleoptera, Carabidae)

to investigate the functioning of pH receptors in ground

beetles.

Materials and methods

Test beetles

Test beetles were collected in April 2003 from their

hibernation sites in forest margins in southern Estonia.

The beetles were kept in plastic boxes filled with moistened

moss in a refrigerator at þ5 �C. Three or four days before

the electrophysiological tests, the beetles were kept in

similar boxes at room temperature (þ 22 �C). The beetles

were given clean water to drink and fed with moistened cat

food (Kitekat, Master Foods, Poland) every day.

Test beetles were immobilized by placing them tightly

into a special conical tube made of thin sheet-aluminium

of a size that allowed their head and antennae to protrude

only a little from the narrower end. The wider rear end of

the conical tube was blocked with a piece of plasticine to

prevent the beetle from retreating out of the tube. The

antennae of immobilized intact beetles were fastened hori-

zontally on the edge of a special aluminium stand with tiny

amounts of beeswax, such that some horizontally located

antennal taste bristles were visible from above under a light

microscope, and easily accessible for micromanipulations

from the side. Contamination of tested taste bristles and

their contact with preparation instruments and beeswax was

avoided.

Chemicals and preparation of buffer solutions

Before experiments, 100mM acetate buffers in the pH

range 3–6 were prepared by dissolving glacial acetic acetate

and anhydrous sodium acetate in distilled water. Stimulat-

ing 100mM buffers with higher levels of pH (5.8–8.1) were

made with combinations of Na2HPO4 and NaH2PO4. For

some experiments, the pH of phosphate buffer was raised to

11.1 with NaOH. The pH of the 100mM borax used in some

test series with acetate buffers was 9.3. All chemicals

were obtained from AppliChem (Germany). To produce

stimulating buffers of approximately the same pH but of

different concentrations of ions, the 100mM phosphate

buffer (pH7.6) was diluted in distilled water to 10mM

solution (pH7.8). The final pH of the stimulating solutions

was measured with a pH meter E6115 (Evikon, Estonia).

Stimulation and recording of action potentials

Electrophysiological recordings were made from large

chaetoid taste sensilla on the antennal flagellum of the

beetle with a tip-recording technique (Hodgson et al.,

1955). To achieve a good signal-to-noise ratio and avoid

registration of muscular activities from two basal antenno-

meres, the scape and pedicel and from other parts of insect’s

body, an indifferent tungsten microelectrode was inserted

into the base of the flagellum. The recording microelec-

trode, a glass micropipette (diameter at the tip approxi-

mately 20 mm) filled with stimulating buffer solution was

placed over the sensillum tip by means of a micromanipu-

lator under visual control with a light microscope at

magnification � 300. Immediately before each stimulation,

two to three drops of the stimulating solution were squeezed

out of the micropipette tip, and absorbed onto a piece of

filter paper attached to an additional micromanipulator to

reduce concentration increases due to evaporation. A

copper wire made contact with high-impedance preampli-

fier input. The signal picked up by the recording electrode

was amplified and filtered with a band width set at

100–2000Hz and monitored on an oscilloscope. Recordings

of the first 5 s of the response were transferred directly to a

computer via an analogue-to-digital input board DAS-1401

(Keithley, Taunton, Massachusetts) for data acquisition,

storage and further analysis using TestPoint (Capital

Equipment Corp., Billerica, Massachusetts) software. The

sampling data rate was 10 kHz.

Due to variations in responsiveness of different taste

bristles to the same stimuli, on the antennae of one and

the same as well as different beetles, stimulating buffers

were tested in pairs to allow a more precise comparison of

the responses to the tested stimuli. Thus, each taste sensil-

lum was tested twice. Up to 15 sensilla on one antenna of a

test beetle were in a suitable position for stimulations and

recordings. Primarily, these sensilla were tested with the first

stimulus. Then, after approximately 30min, the stimulat-

ing/recording micropipette was rinsed with distilled water

several times and filled with the second stimulus to test the

pH receptors in P. oblongopunctatus 123

# 2005 The Royal Entomological Society, Physiological Entomology, 30, 122–133

57

same sensilla again. The space of time between the two

successive stimulations ensured complete disadaptation of

the receptor cells. After the test series, test beetles remained

in good condition, and no differences were observed in the

responses of their taste bristles recorded at the beginning

and the end of the test series. Antennal sensilla from four to

six beetles were tested in each pair of stimuli.

Data management and analysis

Action potentials from several sensory cells of taste sen-

silla were distinguished by differences in their amplitude,

duration, shape and polarity of spikes. In some cases, the

regularity of firing as the principal parameter for separating

and identifying action potentials was used. Automatic

classifying and counting of action potentials from taste

sensilla was clearly not an appropriate method for data

analysis because spike amplitude of a certain cell often

changed in time or with concentration, and two action

potentials frequently interacted to produce an irregular

waveform. Instead, the analysis used was visual, employing

TestPoint software (CEC Capital Equipment, Bedford,

New Hampshire). Firing rates of receptor cells were

expressed as a number of spikes per second. All data

were analysed using Microsoft Excel (Microsoft Corp.,

Redmond, Washington) and Statistica (StatSoft, Inc.,

Tulsa, Oklohoma) software.

Results

Taste sensilla on the antennal flagellum in

P. oblongopunctatus

Large blunt-tipped 100–200-mm long flagellar taste

bristles of the ground beetle P. oblongopunctatus can be

quantified at low magnification under the light microscope.

The largest bristles occur on the first flagellomere counting

from the beetle’s head. The length of the bristles towards the

tip of the antenna decreases, and the shortest of the bristles

can be found at the tip of the terminal flagellomere. Bristles

are located in whorls in the distal part of flagellomeres.

There are six on the first flagellomere and seven on the

following flagellomeres. In addition to those, there are six

bristles symmetrically at the tip of the terminal flagellomere.

Hence, the total number of taste bristles on each antennal

flagellum is 68. All these bristles seemed to respond in a

similar manner to the stimuli tested.

Characterization of action potentials recorded from antennal

taste sensilla

Two sensory cells innervating flagellar taste bristles in

P. oblongopunctatus responded to stimulation with acetate

buffers, phosphate buffers and borax: A- and B-cell. Action

potentials with positive spikes generated by the A-cell are

extremely large, approximately 8mV peak-to-peak. By con-

trast, action potentials with negative spikes originating

from the B-cell are three- to four-fold smaller in amplitude

compared with the action potentials generated by the A-cell

(Fig. 1A). The responses to tested stimuli were phasic–tonic

in type, with a more pronounced phasic component in

B-cells. On the recordings, action potentials from two cells

frequently interacted to produce an irregular waveform. In

these cases, the regularity of firing as the principal para-

meter for visual separating and identifying action potentials

was used (Fig. 1A). In some cases, responses of the A-cells

were of a bursting nature (Figs 1C,D). Usually action

potentials generated by the same cell had similar amplitude.

However, in a number of cases, especially at the beginning

of the B-cell’s response, a few action potentials could be

observed whose amplitude was remarkably greater than the

others (Fig. 1B).

Responses to acetate buffers at pH 3 and 4

The A-cell did not respond to 100mM acetate buffer at

pH 3 and only a very weak response was observed at pH4:

the average number of impulses during the first and follow-

ing seconds of the response was less than 1. At the same

time, the A-cell of the same bristles strongly responded to

borax: on average, 23 impulses per first second of the

response were recorded (Fig. 2A, A-cell), demonstrating

that the tested cells were in good functional condition.

The B-cell responded to 100mM acetate buffers at pH 3

and 4 also relatively weakly: 5.7 and 9.5 impulses per first

second of the response, respectively (P< 0.05). The

response of the same B-cells to borax was 24 imp/s. Dur-

ing the following seconds of the response (in the tonic

part), the firing rate of the B-cell stabilized at the level of

1–3 imp/s; compared with the first second of the response,

the opposite trend was noted. The firing rate of the cell at

pH 4 appeared to be lower than that at pH3, although

the difference was not statistically significant (P> 0.05;

Fig. 2A, B-cell). Some typical 5-s responses of antennal

taste sensilla to stimulating solutions at pH 3 and 4 are

shown in Fig. 3.


In response to stimulation with acetate buffer at pH5, the

A-cells generated significantly more action potentials (first

second mean response 3.6 impulses) compared with the

response at pH 4 (0.9 impulses per first second of the

response) (P< 0.05; Fig. 2B, A-cell).

During the first two seconds of the response, B-cells

fired at a significantly higher frequency at pH 5 than at

pH 4 (P< 0.05; first second responses 15.9 and 9.7 imp/s,

respectively). The third second response was approxi-

mately equal (P> 0.05). As observed in the test series

with stimulating solutions at pH 3 and 4, during the last

124 E. Merivee et al.


58

2 s of stimulation, B-cells generated significantly less

action potentials at pH 5 compared with the response at

pH 4 (P< 0.05; Fig. 2B, B-cell).


As the pH of the 100mM acetate buffers tested

increased, higher rates of firing were generated by the

A-cells, although these remained at a relatively low level,

below 5 imp/s. When acetate buffers at pH 5 and pH 6

were used for stimulation, in every pair of stimuli tested,

the less acid buffer solutions caused A-cells to produce

more action potentials per second: first second firing

rates of 3.7 and 4.6 imp/s, respectively. Although differ-

ences in mean firing rates of A-cells were not statistically

significant during the first second of the response

(P> 0.05), these differences were significant during

the following seconds of the response (P< 0.05; Fig.

2C, A-cell).

First second mean firing rates of the B-cells in response

to stimulating solutions at pH 5 and 6 were 17 and

22.7 imp/s, respectively. The difference between the

means was statistically significant (P< 0.05). During the

next 4 s of the response, no differences were observed in

relatively low firing rates caused by the two buffer solu-

tions (P> 0.05; Fig. 2C, B-cell). Thus, comparing B-cell

responses to acetate buffers with pH3–6, it became evi-

dent that solutions of higher pH caused B-cells to fire at

higher frequencies during the first second of the response.

Fig. 1. Action potentials recorded from

two cells of antennal taste bristles in

response to 100mM phosphate buffer at

pH8.1 in the ground beetle Pterostichus

oblongopunctatus. (A) The beginning of

the electrophysiogical response. Arrow-

head indicates the beginning of stimula-

tion. AC and BC, action potentials from

A- and B-cells, respectively, distinguished

by differences in their amplitudes and

polarity of spikes. Action potentials from

two cells interacted frequently to produce

an irregular waveform (arrows). In the

phasic part of the response, the B-cell

fired in a highly regular manner (black

dots), which facilitated differentiation and

counting of impulses of two cells. Cali-

bration bar¼ 5mV. (B) The phasic begin-

ning of the response to phosphate buffer

at pH8.1. Only one cell responded. Often,

especially at the beginning of the B-cell’s

response, a few action potentials could be

observed whose amplitude was remark-

ably greater than others (indicated by

arrows). (C) The A-cell frequently fired in

bursts. Impulses of one of the bursts

(marked with black bar) are shown at a

shorter timescale (D).



59

Fig. 2. Responses of antennal taste bristles to 100mM acetate buffers at pH 3–6. Because the test insects and their taste sensilla varied

greatly in their responses to the same stimulus, stimuli were tested in pairs with one and the same bristle. (A) The A-cell did not fire,

or fired at very low frequency, when stimulated by acetate buffers at pH 3 and 4. However, strong responses to borax showed that the

same cells were in a good functional condition. (B and C) According to how the pH level of stimulus solution increased, higher rates

of firing were observed in A-cells. The increase in pH caused B-cells to fire more frequently, especially during the first second of the

response, but to a remarkably lesser extent than in A-cells. Vertical bars indicate SE of the means; asterisk (*) and n.s. indicate

whether the difference between the mean firing rates is significant or not, respectively (paired t-test, P< 0.05). n, Number of taste

bristles tested.



60

During the next 4 s of the response, in most cases, similar

firing rates were observed in response to the buffers tested

in pairs. However, in some cases (stimulations at pH 4

and 5), stimulating solution at lower pH elicited a signifi-

cantly higher firing rate in the tonic part of the response

of the B-cell (Fig. 2B).

Responses to phosphate buffers at pH 5.8 and 7.0

The frequency of action potentials, generated by 100mM

phosphate buffers both in A- and B-cells, was remarkably

higher than with acetate buffers. In A-cells, the number of

spikes per first second of the response was 10.5 and 15.8,

caused by stimulating solutions at pH5.8 and 7, respect-

ively. In addition, during the following seconds of the

response, firing rate (4–8.3 imp/s) generated by the neutral

buffer (pH7) was remarkably higher compared with that

(0.4–2.4 imp/s) generated by the slightly acid buffer (pH5.8)

(P< 0.05; Fig. 4A, A-cell). The responses of B-cells to these

two phosphate buffers were approximately equal (Fig. 4A,

B-cell).


Responses of the A-cell to stimulation with phosphate

buffers at pH5.8 and 8.1 were similar to these recorded in a

test series with buffers at pH5.8 and 7. However, differences

between respective firing rates were considerably greater in

Fig. 3. Some typical recordings of the

responses of an antennal taste sensillum

to 100mM acetate buffers and borax. (A

and B) Responses to acetate buffers at pH

3 and 4. Only one cell responded.

Stimulating the same taste bristle with

borax solution caused two cells to fire,

demonstrating that both cells of the

bristle were in a good condition and

responded strongly when an adequate

stimulus was provided (C). Some small

impulses from B-cells among large

impulses originating from A-cells are

indicated by arrows. (D) Beginning of

the same recording (C) at a shorter

timescale.



61

Fig. 4. Responses of antennal taste sensilla to 100mM phosphate buffers at pH5.8–11.1. Stimuli were tested in pairs with the same sensilla to

overcome the different sensitivity of different taste bristles. In A-cells, as a rule, stimulating solutions of high pH caused remarkably higher

rates of firing compared with those caused by solutions at lower pH (A–C). The pH of the stimulus solution also affected the responses of the

B-cells but in a much more complicated way. For example, the B-cell’s responses to buffers at pH5.8 and 7 did not differ very much (A). In a

test series with pH5.8 and 8.1, the firing rate of the B-cells was affected by an increase in pH in the opposite way at the beginning and end of

stimulation, positively and negatively, respectively (B). When the responses of the B-cells to phosphate buffers at pH8.1 and 11.1 were

compared with each other, it became evident that the solution with pH11.1 was a more effective stimulus for this cell. However, compared

with the responses of the A-cell, B-cell’s firing rate depended remarkably less on the pH value of the stimulating solution (C). Vertical bars

indicate standard errors (SE) of the means; asterisk (*) and n.s. indicate a significant and not significant difference between the mean firing

rates, respectively (paired t-test, P< 0.05). n, Number of bristles tested.



62

the test series with stimulating solutions at pH5.8 and 8.1

(first second responses 8.0 and 23.1 imp/s, respectively; Fig.

4B, A-cell).

In B-cells, differences between firing rates elicited by

stimulating buffers at pH5.8 and 8.1 were relatively small

but statistically significant (P< 0.05). During the first sec-

ond of the responses, mean firing rate at pH5.8 was a little

lower compared with that caused at pH8.1, 25 and

28.2 imp/s, respectively. The opposite was observed in the

tonic part of the responses; firing rates caused by buffer

solution at pH5.8 were slightly higher compared with those

at pH8.1 (Fig. 4B, B-cell). Some typical original recordings

of the responses to stimulating buffers at pH5.8 and 8.1 are

shown in Fig. 5.


Recordings obtained in test series with phosphate buffers

demonstrate that different sets of taste sensilla, belonging to

different beetles, may vary greatly in their responses to the

same stimulus. For example, 1.7 and 23.1 spikes per first

second of the response to buffer solution with pH8.1 were

recorded from A-cells in test series where responses to

buffers at pH5.8 and 8.1 and 8.1 and 11.1 were compared,

respectively (Fig. 4B,C). At the same time, the same A-cells,

which responded very weakly to buffer at pH8.1, responded

strongly to buffer at pH11.1 (P< 0.05), with first second

responses of 1.7 and 27.9 imp/s, respectively (Fig. 4C,

A-cell).

Fig. 5. Typical responses recorded from

one and the same antennal taste sensillum

stimulated by 100mM phosphate buffers

at pH5.8 (A) and 8.1 (C). (B) and (D)

show the same responses on a shorter time

scale, respectively. Two cells are firing.

(A,C) and (B,C) show action potentials

from A- and B-cells, respectively. Some

impulses generated by the B-cell are

marked with arrows. Compared with the

B-cell, the neural activity level of the

A-cell depended much more on the pH of

the stimulating solution.



63

Although the B-cells responded to buffer at pH11.1 more

strongly than at pH8.1 (P< 0.05), these cell’s responses to

those buffers during the first second differed to a much less

extent compared with those observed in A-cells, 34.4 and

27.9 imp/s, respectively. Furthermore, in the tonic part of

the response, the number of spikes at pH11.1 was remark-

ably higher than at pH8.1 (Fig. 4C, B-cell). Some examples

of the responses of the antennal taste sensilla to phosphate

buffers at pH8.1 and 11.1 are demonstrated in Fig. 6.

Responses to 10 and 100mM phosphate buffers (pH7.8 and

7.6, respectively)

When responses of A-cells to 10 and 100mM phosphate

buffers were compared, it became evident that remarkably

higher firing rates were elicited by the solution at lower

concentration. Differences were significant during the first

4 s of the response (P< 0.05), but not during the final fifth

second (P> 0.05; Fig. 7, A-cell). The opposite was noticed

when responses of the B-cell were compared with each

other; higher firing rates occurred when stimulating with

buffer solution at higher concentration, first second

responses 31.2 and 21.9 imp/s, elicited by 100 and 10mM

phosphate buffers, respectively (P< 0.05). In the tonic

part of the B-cell’s response, firing rates decreased rapidly,

and the cell responded equally to 10 and 100mM buffers

(Fig. 7, B-cell).

Discussion

In a number of insects, including the ground beetle

P. oblongopunctatus, frequently two cells of the taste sensilla

Fig. 6. Recordings of some typical

responses of an antennal chaetoid taste

sensillum to phosphate buffers at pH8.1

(A) and pH11.1 (C). (B) and (D) show the

same responses on a shorter time scale,

respectively. Some small neural impulses

generated by the B-cell among large

impulses originated by the A-cell are

marked with arrows. Responses of both

A-cell, and B-cells were of a phasic–tonic

type.



64

respond to salt solutions (Dethier & Hanson, 1968; Haskell

& Schoonhoven, 1969; Den Otter, 1972; Mitchell

& Schoonhoven, 1973; Mitchell & Seabrook, 1973; Blaney,

1974; Hanson, 1987; Bernays & Chapman, 2001a;

Schoonhoven & van Loon, 2002). It is suggested that they

function as an anion and a cation cell (Ishikawa, 1963;

Elizarov, 1978; Hanson, 1987; Schoonhoven & van Loon,

2002). Temporal patterns of firing and the concentration–

response curves are quite different for the two cells. In the

cation cell with a phasic–tonic type of reaction, themagnitude

of the response is proportional to the logarithm of salt

concentration over the dynamic range (Evans & Mellon,

1962; Hanson, 1987). It is believed that the stimulating effect

of salts to taste receptors is dominated by the cations

involved, and that monovalent cations are more effective

stimuli than divalent cations (Evans &Mellon, 1962; Schoon-

hoven & van Loon, 2002). This cell type corresponds to the

B-cell of the antennal taste bristles in P. oblongopunctatus.

Specific stimuli for the second salt cell are yet to be

identified: it has been named the ‘anion’ cell on the basis

of a supposed sensitivity to salt anions or the ‘fifth’ cell

(Ishikawa, 1963; Dethier & Hanson, 1968; Hanson, 1987;

Liscia et al., 1997; Liscia & Solari, 2000). This cell is char-

acteristically less responsive to differences in salt concentra-

tion and often responds in bursts (Dethier, 1977; Liscia

et al., 1997), similar to that observed in P. oblongopunctatus.

In the strict sense of the term, it is not correct to give the

name ‘salt cell’ to a cell that does not respond to changes in

salt concentration. On the other hand, a study in Grammia

geneura shows increasing activity in two salt cells in

response to KCl at concentrations ranging from 10 to

1000mM. NaCl also produced responses in the two

cells with slightly higher firing rates than KCl (Bernays &

Chapman, 2001a). Dethier (1977) was the first to realize

that it is not at all certain that the best stimulus for the

second salt cell is indeed salt. Electrophysiological evidence

is provided that the so-called ‘fifth’ cell in taste chemosen-

silla of blowflies responds to deterrent compounds, such as

quinine, amiloride, nicotine and caffeine (Liscia & Solari,

2000). Very little attention has been paid to this cell. This is

unfortunate because this lack of information handicaps any

attempt to construct an overview of the control of food

detection and habitat selection in insects on the basis of

their chemosensory input.

In the ground beetle P. oblongopunctatus, the firing rate

of the second cell, the A-cell, innervating antennal taste

sensilla, was correlated positively with pH values of the

100mM acetate and phosphate buffers tested at pH3–11

over the range of the 5-s stimulation period. Differences in

firing rates of the A-cell elicited by the different pH of the

stimulating solutions may be very large. For example, the

number of impulses elicited by 100mM phosphate buffer at

pH11.1 was approximately 16-fold higher compared with

that elicited by 100mM phosphate buffer at pH8.1. When

responses to 10 and 100mM phosphate buffers with nearly

the same pH (pH7.8 and 7.6, respectively) were compared

with each other, it became evident that a solution at lower

concentration had a stronger stimulating effect on the

A-cell than a solution at higher concentration. At the

same time, as expected, the response of B-cells to 10 and

100mM phosphate buffer was the opposite: a solution at a

higher concentration caused higher rates of firing.

Therefore, it can be concluded, that the best stimulus for

the antennal taste bristles’ A-cell in the ground beetle

P. oblongopunctatus is not salt but the pH level of the

stimulus solution. The present results confirm the

suggestion made by Paje & Mossakowski (1984) that, in

ground beetles, pH receptors are located on the antennae.

To date, pH receptors have not been described in other

insects, and it is possible that the pH of the stimulating

solution may prove to be an important stimulus for other

chemoreceptors.

If the response of the A-cell depends on solution pH level

only, there should be equivalent responses to buffers at

different concentrations at the same pH. However, the

results do not confirm this: 10mM buffer had a greater

Fig. 7. Responses of antennal taste sensilla to 100 and 10mM phosphate buffers (pH measured 7.6 and 7.8, respectively). A- and B-cells

responded to changes in buffer solution concentration in opposite manner. By contrast to A-cells, in B-cells, buffer solution at higher

concentration was a more effective stimulus, especially during the first 2 s of the response. Asterisk (*) and n.s. inidcate whether differences

between mean firing rates were statistically significant or not, respectively (paired t-test, P< 0.05).



65

stimulating effect on the A-cell than 100mM buffer. This

phenomenon can be explained by interaction of different

chemical stimuli to chemosensory cells. Other chemicals, as

well as the pH of stimulating solution, may influence the

responses of insect taste receptors to one or another stimu-

lating compound (Hanson, 1987; Liscia et al., 1997; Bernays

et al., 1998; Liscia & Solari, 2000; Bernays & Chapman,

2001b; Schoonhoven & van Loon, 2002). Presumably,

salts of stimulating buffer solutions at high concentration

suppressed the neural activity of pH cell when stimulated

with 100mM phosphate buffer in 10 and 100mM phosphate

buffer experiments in P. oblongopunctatus.

The pH cell of antennal taste sensilla in P. oblongopunc-

tatus does not fire, or fires at very low frequency (first

second response below 5 imp/s) at pH3–6. This acid range

of pH values most likely represents the pH values in the

forest habitat and hibernating sites of this ground beetle

(Thiele, 1977; Paje & Mossakowski, 1984).

It is noteworthy that the action potentials registered by

antennal taste bristles’ cation and pH cells in the ground

beetle P. oblongopuctatus are negative and positive in their

polarity, respectively. Furthermore, other electrophysiolo-

gists studying insects’ chemoreceptors have noticed that

spikes in sensillum tip-recording technique (Hodgson

et al., 1955) may have positive as well as negative polarity

(Hanson & Wolbarsht, 1962; Dethier & Hanson,

1965; Elizarov & Sinitzina, 1974; Elizarov, 1978; Hansen-

Delkeskamp, 2001; Albert, 2003). Because the sensillum is

an integral part of the recording circuit, any cellular

properties and changes may contribute to uncontrolled

variability in the shape of the recorded action potentials

(Frazier & Hanson, 1986). At present, there appears to be

no satisfactory explanation for this phenomenon.

Acknowledgements

Contract grant sponsor: Estonian Science Foundation;

Contract grant numbers: 5423 and 4105.

References

Albert, P.J. (2003) Electrophysiological responses to sucrose from a

gustatory sensillum on the larval maxillary palp of the spruce

budworm, Choristoneura fumiferana (Clem.) (Lepidoptera:

Tortricidae). Journal of Insect Physiology, 49, 733–738.

Bernays, E.A. & Chapman, R.F. (2001a) Taste cell responses in the

polyphagous arctiid,Grammia geneura: towards a general pattern

for caterpillars. Journal of Insect Physiology, 47, 1029–1043.

Bernays, E.A. & Chapman, R.F. (2001b) Electrophysiological

responses of taste cells to nutrient mixtures in the polyphagous

caterpillar of Grammia geneura. Journal of Comparative Physiol-

ogy A, 187, 205–213.

Bernays, E.A., Glendinning, J.I. & Chapman, R.F. (1998) Plant

acids modulate chemosensory responses in Manduca sexta

larvae. Physiological Entomology, 23, 193–201.

Blaney, W.M. (1974) Electrophysiological responses of the terminal

sensilla on the maxillary palps of Locusta migratoria (L.) to some

electrolytes and non-electrolytes. Journal of Experimental

Biology, 60, 275–293.

Daly, P.J. & Ryan, M.F. (1979) Ultrastructure of antennal sensilla

of Nebria brevicollis (Fab.) (Coleoptera: Carabidae). Interna-

tional Journal of Insect Morphology and Embryology, 8, 169–181.

den Otter, C.J. (1972) Differential sensitivity of insect chemo-

receptors to alkali cations. Journal of Insect Physiology, 18,

109–131.

Dethier, V.G. (1977) The taste of salts. American Scientist, 65,

744–751.

Dethier, V.G. & Hanson, F.E. (1965) Taste papillae of the blowfly.

Journal of Cellular and Comparative Physiology, 65, 93–100.

Dethier, V.G. & Hanson, F.E. (1968) Electrophysiological

responses of the blowfly to sodium salts of fatty acids.

Proceedings of the National Academy of Sciences, U.S.A., 60,

1269–1303.

Elizarov, Y.A. (1978) Insect Chemoreception. Moscow University

Press, Russia.

Elizarov, Y.A. & Sinitzina, E.E. (1974) Contact chemoreceptors in

Aedes aegypti L. Zoological Journal, 53, 577–584 (in Russian).

Evans, D.R. & Mellon, D. (1962) Stimulation of a primary taste

receptor by salts. Journal of General Physiology, 45, 651–661.

Frazier, J.L. & Hanson, F.E. (1986) Electrophysiological recording

and analysis of insect chemosensory responses. Insect–Plant

Interactions (ed. by T. A. Miller and J. Miller), pp. 285–330.

Springer, Germany.

Hansen-Delkeskamp, E. (2001) Responsiveness of antennal taste

hairs of the apterygotan insect, Thermobia domestica (Zygento-

ma); an electrophysiological investigation. Journal of Insect

Physiology, 47, 689–697.

Hanson, F.E. (1987) Chemoreception in the fly: the search for

the liverwurst receptor. Perspectives in Chemoreception

and Behaviour (ed. by R. F. Chapman, E. A. Bernays and

J. G. Stoffolano), pp. 99–122. Springer, Germany.

Hanson, F.E. & Wolbarsht, K.L. (1962) Dendritic action potentials

in insect chemoreceptors. American Zoologist, 2, 528–534.

Haskell, P.T. & Schoonhoven, L.M. (1969) The function of certain

mouth part receptors in relation to feeding in Schistocerca

gregaria and Locusta migratoria migratorioides. Entomologia

Experimentalis et Applicata, 12, 429–440.

Hodgson, E.S., Lettvin, J.Y. & Roeder, K.D. (1955) Physiology of

a primary chemoreceptor unit. Science, 122, 417–418.

Ishikawa, S. (1963) Responses of maxillary chemoreceptors in the

larvae of the silkworm, Bombyx mori, to stimulation by

carbohydrates. Journal of Cellular and Comparative Physiology,

61, 99–107.

Kim, J.L. & Yamasaki, T. (1996) Sensilla of Carabus (Isiocarabus)

fiduciarius saishutoicus Csiki (Coleoptera: Carabidae). Interna-

tional Journal of Insect Morphology and Embryology, 25,

153–172.

Liscia, A. & Solari, P. (2000) Bitter taste recognition in the blowfly:

Electrophysiological and behavioral evidence. Physiology and

Behaviour, 70, 61–65.

Liscia, A., Solari, P., Majone, R., Tomassini Barbarossa, I. &

Crnjar, R. (1997) Taste receptor mechanisms in the blowfly:

evidence of amiloride-sensitive and insensitive receptor sites.

Physiology and Behavior, 62, 875–879.7.

Merivee, E., Ploomi, A., Rahi, M., Luik, A. & Sammelselg, V.

(2000) Antennal sensilla of the ground beetle Bembidion lampros

Hbst (Coleoptera, Carabidae). Acta Zoologica (Stockholm), 81,

339–350.



66

Merivee, E., Ploomi, A., Luik, A., Rahi, M. & Sammelselg, V.

(2001) Antennal sensilla of the ground beetle Platynus dorsalis

(Pontoppidan, 1763) (Coleoptera, Carabidae). Microscopy

Research and Technique, 55, 339–349.

Merivee, E., Ploomi, A., Rahi, M., Bresciani, J., Ravn, H.P., Luik, A. &

Sammelselg,V. (2002)Antennal sensilla of the groundbeetleBembidion

properans Steph. (Coleoptera, Carabidae).Micron, 33, 429–440.

Mitchell, B.K. & Schoonhoven, L.M. (1973) Taste receptors

in Colorado beetle larvae. Journal of Insect Physiology, 20,

1787–1793.

Mitchell, B.K. & Seabrook, W.D. (1973) Electrophysiological

investigations on tarsal chemoreceptors of the spruce budworm,

Choristoneura fumiferana (Lepidoptera). Journal of Insect

Physiology, 20, 1209–1218.

Paje, F. & Mossakowski, D. (1984) pH-preferences and habitat

selection in carabid beetles. Oecologia (Berlin), 64, 41–46.

Schoonhoven, L.M. & van Loon, J.J.A. (2002) An inventory of

taste in caterpillars: each species its own key. Acta Zoologica

Academiae Scientiarum Hungarica, 48, 215–263.

Shiraishi, A. & Morita, H. (1969) The effects of pH on the labellar

sugar receptor of the fleshfly. Journal of General Physiology, 53,

450–470.

Thiele, H.-U. (1977) Carabid beetles in their environment.

Zoophysiology and Ecology 10. Springer, Germany.

Accepted 20 October 2004



Milius, M., Merivee, E., Williams, I., Luik, A., Mänd, M., Must, A.,

2006.

A NEW METHOD FOR ELECTROPHYSIOLOGICAL

IDENTIFICATION OF ANTENNAL PH RECEPTOR CELLS

IN GROUND BEETLES: THE EXAMPLE OF PTEROSTICHUS

AETHIOPS (PANZER, 1796) (COLEOPTERA, CARABIDAE).


69

Journal of Insect Physiology 52 (2006) 960–967

A new method for electrophysiological identification of antennal pHreceptor cells in ground beetles: The example of Pterostichus aethiops

(Panzer, 1796) (Coleoptera, Carabidae)

Marit Miliusa, Enno Meriveea,�, Ingrid Williamsb, Anne Luika, Marika Manda, Anne Musta

aEstonian University of Life Sciences, Institute of Agricultural and Environmental Sciences, 64 Kreutzwaldi Street, 51014 Tartu, EstoniabRothamsted Research, Harpenden, Hertfordshire, AL5 2JQ, UK

Received 20 March 2006; received in revised form 2 June 2006; accepted 7 June 2006

Abstract

The responses of antennal taste sensilla of the ground beetle Pterostichus aethiops to 100mM Na+-salts and their mixtures with 1 and

10mM NaOH were compared. An increase in pH by 0.3–0.6 units in 100mM Na+-salt solutions, caused by the content of 1mM NaOH,

was too small, except for alkaline Na2HPO4, to influence the firing rate of the cation cell and pH cell significantly. However, different

sensitivity of the two cells to increased pH was clearly demonstrated when the concentration of NaOH in 100mM stimulating salt

solutions was increased to 10mM. Increasing pH by 1.2–2 units caused the 1st s firing rate to increase by 140–1050% and 0–26% in the

pH cell and cation cell, respectively. Compared to the buffer series method used for identification of the pH receptors in ground beetles

earlier, considerably stronger responses of the pH cell to a similar increase in pH were observed when the NaOH method was used for

testing. At the same time, undesirable changes in salt ions concentration that occur when stimulating solutions differing by 1–2 pH units

are prepared were much smaller using the latter method. Behavioural and ecological relevance of the results is discussed.

r 2006 Elsevier Ltd. All rights reserved.

Keywords: Taste sensilla; Tip recording; Action potential rate; Predatory insects; Habitat preferences

1. Introduction

The ground beetles (Coleoptera, Carabidae) with 40,000species belong to the most numerous families of beetles,their distribution being cosmopolitan. Approximately 430species have been found in Fennoscandia, Denmark andBaltic countries, 530 species occur in Czech Republic(Lovei and Sunderland, 1996; Kula and Purchart, 2004;Silferberg, 2004). Their bio-indication importance isderived from their dependence on specific biotic andabiotic conditions of the site: food availability, competi-tors, vegetation type, humidity, light and temperatureconditions, soil pH and salt content etc. Habitat choice isso specific that carabids are often used to characterizehabitats (Thiele, 1977; Lovei and Sunderland, 1996).

External stimuli crucial in habitat choice of ground beetleshave been purely studied on the receptor level, however.Preferences for particular soil pH, varying between 3 and

9, in ground beetles were discovered using laboratorychoice experiments some decades ago (Krogerus, 1960;Paje and Mossakowski, 1984). More recently, in ecologicalfield experiments, a strong correlation between theabundance of some ground beetles and soil pH has beenfound (Irmler, 2001, 2003). Electrophysiological studieswith alkaline, neutral and acid salts have indicated that inaddition to the salt (cation) cell, the pH cell also probablyinnervates the antennal taste bristles in the ground beetlePterostichus aethiops (Panzer, 1796) (Merivee et al., 2004).Experiments with acetate and phosphate buffer series in thepH range of 3–11 confirmed this assumption in the groundbeetle Pterostichus oblongopunctatus (Fabricius, 1787)(Merivee et al., 2005). The pH cell discovered seems to bespecific to ground beetles, as it is not found in other insects.In order to explain the role of this cell in the beetles’

ARTICLE IN PRESS

www.elsevier.com/locate/jinsphys

0022-1910/$ - see front matter r 2006 Elsevier Ltd. All rights reserved.

doi:10.1016/j.jinsphys.2006.06.003

�Corresponding author.

E-mail address: [email protected] (E. Merivee).

70

behaviour more precisely, further electrophysiologicalstudies on the distribution and functioning of the pH-receptive cell in ground beetles are justified.

P. aethiops is a European species found from the BritishIsles east to Perm in Russia, south to Serbia and north intosouthern Fennoscandia and Finland. In mountains ofCentral and Eastern Europe their abundance considerablyincreases with increasing altitude up to subalpine zones.This spring-breeding, wingless, eurytopic ground beetlepopulates wet coniferous, deciduous as well as mixedforests with acid soil pH. Also, its preferred overwinteringsites in brown-rotted wood laying on the ground where, inautumn, the beetles concentrate in great numbers arestrongly acid (pH 3–5; Merivee, unpublished data).Frequently, hibernating beetles have been found in mossand litter. In its preferred habitats, it is one of the mostnumerous ground beetles (Haberman, 1968; Thiele, 1977;Lindroth, 1985, 1986; Wachmann et al., 1995; Ermakov,2004; Kula and Purchart, 2004; Silferberg, 2004; Sk"o-dowski, 2005).

In P. oblongopunctatus, it was demonstrated that the pHcell did not fire or fired at very low frequency at pH 3–6,first s response below 5 imp/s declining during the next fours of the stimulation close to zero. This level of acidityprobably represents the pH preferences of this groundbeetle in its forest habitat and hibernating sites (Meriveeet al., 2005). In this study we show that similar correlationsbetween electrophysiological responses of the pH cell andpH of the preferred habitats may occur also in the groundbeetle P. aethiops.

The limitation of the buffer series method used fortesting the responses of pH receptor cells to pH in groundbeetles earlier (Merivee et al., 2005) is that buffers withdifferent pH vary largely in both pH and electrolyteconcentration. The NaOH method reported here allows toprepare stimulating salt solutions over a wide range of pHwhereas changes in electrolyte concentrations are minimal.Over the dynamic range of the stimulus/response curve, theaction potential rate of a taste cell is a logarithmic functionof the stimulus concentration. The maximum firing rate ofa phasic-tonically responding taste cell evoked by strongstimuli does usually not exceed 40–50 imp/s (Evans andMellon, 1962; Hanson, 1987; Merivee et al., 2004).Consequently, a 1–10% increase in concentration ofthe stimulating salt solution causes a change of less than1 imp/s in the firing rate of the taste cell, which is hardlydetectable in electrophysiological experiments. These cal-culations were taken into account when designing thecurrent experiments.

To test for the presence of the pH cell in the antennaltaste bristles of the ground beetle P. aethiops moreprecisely, three solutions of 100mM Na+-salts, Na2HPO4,NaCl and NaH2PO4 were prepared, acid, neutral andalkaline, respectively; these served as control stimuli. Teststimuli were prepared by adding small amounts of NaOHto the 100mM salts resulting in three electrolyte mixtureswith 1mM NaOH and three mixtures with 10mM NaOH.

Addition of NaOH to these salt solutions does not changetheir ionic composition and the Na+ concentrationincreases only to a small extent (1% and 10%, respec-tively). At the same time, the pH level of these solutionsrises substantially. The stimulating effect of these sodiumsalt solutions with and without NaOH was expected to beapproximately similar for the salt cell but not for the pHcell if present. Results of these experiments are reported inthis paper. The size, number and location of antennal tastebristles in the ground beetle P. aethiops have beendescribed earlier (Merivee et al., 2004).

2. Material and methods

2.1. Test beetles

The specimens of P. aethiops used in the experimentswere collected in May 2005 in southern Estonia. They werestored in plastic boxes filled with moistened moss and sandin a refrigerator at +5 1C. Two or three days beforeelectrophysiological tests the beetles were transferred tosimilar boxes at room temperature (+22 1C). They weregiven clean water to drink and fed with moistened cat food(Kitekat, Master Foods, Poland) every day.For the electrophysiological tests, each test beetle was

restrained by placing it into a conical tube made of thinsheet of aluminium of a size that allowed its head andantennae to protrude a little from the narrower end. Thewider rear end of the conical tube was closed with a piece ofplasticine to prevent the beetle from retreating out of thetube. The antennae of the restrained beetle were fastenedhorizontally on the edge of a special aluminium stand withtiny amounts of beeswax, so that horizontally locatedantennal taste bristles were visible from above under a lightmicroscope, and easily accessible for micromanipulationsfrom the side. Contamination of tested taste bristles andtheir contact with preparation instruments and beeswaxwas carefully avoided.

2.2. Chemicals and preparation of stimulating solutions

Chemicals of analytical grade of purity used in theexperiments were obtained from AppliChem (Germany).Stimulating test solutions prepared are shown in Table 1.pH of the solutions ranging from 4.6 to 10.6 was measuredwith the pH meter E6115 (Evikon, Estonia).

2.3. Stimulation and recording of action potentials

Electrophysiological recordings were made from thelarge chaetoid taste sensilla on the antennal flagellum of therestrained beetle using a tip-recording technique (Hodgsonet al., 1955). To achieve a good signal-to-noise ratio andavoid registration of muscular activities from two basalantennomeres, the scape and pedicel and from other partsof insect’s body, an indifferent tungsten microelectrode wasinserted into the base of the flagellum. The recording

ARTICLE IN PRESSM. Milius et al. / Journal of Insect Physiology 52 (2006) 960–967 961

71

microelectrode, a glass micropipette with a tip diameter ofapproximately 20 mm filled with stimulating solution, wasplaced over the sensillum tip by means of a micromanipu-lator under visual control with a light microscope atmagnification � 300. Immediately before each stimulation,two to three drops of the stimulating solution weresqueezed out of the micropipette tip, and adsorbed ontoa piece of filter paper attached to an additional micro-manipulator to reduce concentration increases due toevaporation. A copper wire made contact with a high-impedance preamplifier input. The signal picked up by therecording electrode was amplified and filtered with a bandwidth set at 100–2000Hz and monitored on an oscilloscopescreen. Recordings of the first 5 s of the response weretransferred directly to a computer hard disc via ananalogue-to-digital input board DAS-1401 (Keithley,Taunton, Massachusets) for data acquisition, storage andfurther analysis using TestPoint (Capital EquipmentCorporation, Bedford, New Hampshire) software. Therate of sampling data was 10 kHz. The first s firing rateincludes initial phasic burst of action potentials. Four nexts characterize the following low decline in the rate of firingof the sensory cells tested.

Due to variations in responsiveness of different tastebristles to the same stimuli, stimulating solutions weretested in pairs to allow more precise comparison of theresponses to salt solutions with and without NaOH. Thus,each taste bristle was tested twice. Responses of one tastebristle of each test beetle were recorded. To eliminate thepossible effect of stimulation order to the responses,100mM Na+-salt alone and that containing NaOHalternatively were chosen for the first stimulus. Nodeteriorating or disturbing effect on the receptor cellscaused by used low content of NaOH in stimulating saltsolutions was observed. Then, after approximately 10min,the stimulating/recording micropipette was rinsed withdistilled water several times and filled with the secondstimulus to test the same sensillum again. The time between

the two successive stimulations ensured complete disadap-tation of the receptor cells allowing similar responses in ourpreliminary experiments to several successive stimulationswith the same stimulus. Taste bristles from 30 to 69 beetleswere tested in each pair of stimuli.

2.4. Data management and analysis

Action potentials from several sensory cells of tastesensilla were distinguished by differences in their ampli-tude, shape and polarity of spikes. Automated classifica-tion and counting of action potentials was clearly not anappropriate method for data analysis because actionpotentials from two or three cells frequently interacted toproduce an irregular waveform. Instead, the analysis usedwas visual, employing TestPoint software. Firing rates ofreceptor cells were expressed as a number of spikes persecond. All data were analysed using Microsoft Excel(Microsoft Corp., Redmond, Washington) and Statistica(StatSoft, Inc., Tulsa, Oklahoma) software.

3. Results

Two sensory cells from the antennal taste bristlesresponded to the tested electrolytes in a phasic-tonicmanner. These two cells were the cation and the pH cellsdistinguished by the amplitude and polarity of action

ARTICLE IN PRESS

Table 1

Stimulating solutions prepared for the experiments

Stimulating solution Concentration (mM) pH

1. NaH2PO4 100 4.6

2. NaH2PO4 100 4.9

NaOH 1

3. NaH2PO4 100 5.8

NaOH 10

4. NaCl 100 7.6

5. NaCl 100 8.2

NaOH 1

6. NaCl 100 9.6

NaOH 10

7. Na2HPO4 100 8.6

8. Na2HPO4 100 9.0

NaOH 1

9. Na2HPO4 100 10.6

NaOH 10

Fig. 1. Example recordings from antennal taste sensilla of the ground

beetle Pterostichus aethiops in response to stimulating electrolyte solutions

with different pH. CC and pHC show action potentials from the cation

cell and pH cell, respectively. Arrows indicate the beginning of the

stimulation. Calibration bar: 5mV. (A) Response to 100mM NaCl

(pH 7.6). (B) Response to the mixture of 100mMNaCl and 10mMNaOH

(pH 9.6).

M. Milius et al. / Journal of Insect Physiology 52 (2006) 960–967962

72

potentials they generated. The phasic component of theresponse was more pronounced in the cation cell comparedto the pH cell. The action potentials with positive polarityfrom the pH cell were very large, 7–8mV peak-to-peak,whereas those with negative polarity generated by thecation cell were 3–4 times smaller (Fig. 1).

3.1. Responses to 100mM Na+-salts with and without

addition of 1mM NaOH

The stimulating effect of the 100mM salts tested on thepH cell, depended on their pH. Salt solutions with higherpH evoked stronger responses. Most test beetles did notrespond to the acid 100mM NaH2PO4 (pH 4.6) nor to themixture (pH 4.9) of this 100mM salt and 1mM NaOH;respective mean firing rates during the 1st s of the responsewere below 1 imp/s. The stimulating effects of 100mMNaCl (pH 7.6) and Na2HPO4 (pH 8.6) solutions wereconsiderably higher compared to NaH2PO4, with 1st sfiring rates of 9 and 13 imp/s, respectively. During thefollowing seconds these rates of firing fell 2–3 times. Nosignificant differences were observed between the firingrates of the pH cell caused by 100mM NaCl and themixture (pH 8.2) of 100mM NaCl and 1mM NaOH.Significantly more action potentials were generated by thepH cell in response to the solution of 100mM Na2HPO4

and 1mM NaOH (pH 9.0), however, compared to theresponse evoked by 100mM Na2HPO4 alone (Fig. 2).

By contrast, the relationship between the salt pH and thenumber of action potentials generated was not observed in

the cation cell. Usually NaCl was the most stimulating saltfor this cell. When the responses to the three Na+-saltswith and without NaOH were compared with each other, itbecame evident that the presence of 1mM NaOH in the100mM salts had no effect on the firing rate of the cationcell, except for the 1st s response in the test series withNaCl, where a few more action potentials were generatedby the mixture of this salt and NaOH compared to NaClalone (Fig. 2).

3.2. Responses to 100mM Na+-salts with and without of

10mM NaOH

Differences between the pH and cation cells regardingtheir responsiveness to the pH of the stimulating solutionsbecame much more obvious when the concentration ofNaOH in the alkalized test salt was raised from 1 to10mM. In the pH cell, rates of firing evoked by the mixtureof 100mM Na+-salts and 10mM NaOH were significantlyhigher compared to the responses caused by the salts alone(Fig. 3). The stimulating effect of NaOH to the pH cell wasgreatest in the mixture with NaCl (pH 9.6), firing rateduring the 1st s of the response was 31 imp/s and it did notfall below 17 imp/s also during the next several secondsrecorded. The difference compared to the response causedby 100mM NaCl alone was 10–22 times depending on theresponse time. The mixtures of 10mM NaOH with100mM NaH2PO4 (pH 5.8) and Na2HPO4 (pH 10.6) were2–7 times more effective stimuli for the pH cell comparedto these non-alkalized salts (pH 4.6 and 8.6, respectively).

ARTICLE IN PRESS

Fig. 2. Responses of the pH cell and cation cell of the antennal taste sensilla of the ground beetle P. aethiops to various 100mM Na+-salts and their

mixtures with 1mM NaOH. Vertical bars denote 7SE of the means. Asterisks indicate significantly different means: paired t-test, Po0.05.

M. Milius et al. / Journal of Insect Physiology 52 (2006) 960–967 963

73

The presence or absence of 10mM NaOH in thestimulating salt solution had little or no effect on the rateof firing of the cation cell. However, a 10–40% increase inthe firing rate was observed in response to the mixture of10mM NaOH with 100mM NaH2PO4 and NaCl at thebeginning of the 5 s stimulation period (Fig. 3).

4. Discussion

The results of the electrophysiological experimentsconducted with Na+-salts and Na+-alkali solutions in theground beetle P. aethiops agree well with earlier data oncontact chemoreception of ground beetles, obtained fromexperiments with various salts (Merivee et al., 2004) andbuffers (Merivee et al., 2005). Accordingly, two sensorycells of the taste bristles located on the antennae of groundbeetles respond to electrolytes: one of them is a salt (cation)

cell, the other a pH cell. The salt as well as the pH receptorsseem to be common in ground beetles and are probablyrelated to habitat choice by these ground-dwelling insects(Thiele, 1977; Paje and Mossakowski, 1984; Irmler, 2001,2003). Surprisingly, no pH receptors have been found inother insects so far. This is probably due to the fact thatinsect’ gustation has been less studied than their olfaction(Chapman, 1998; Hansson, 1999; Blomquist and Vogt,2003; Christensen, 2005). Instead, in some other insects, twosensory cells of a taste sensillum respond to stimulating saltsolutions (Dethier and Hanson, 1968; Haskell and Schoon-hoven, 1969; Den Otter, 1972; Mitchell and Schoonhoven,1973; Mitchell and Seabrook, 1973; Blaney, 1974; Hanson,1987; Bernays and Chapman, 2001a; Schoonhoven and vanLoon, 2002) but it is believed that they function as an anionand a cation cell (Ishikawa, 1963; Elizarov, 1978; Hanson,1987; Schoonhoven and van Loon, 2002).

ARTICLE IN PRESS

Fig. 3. Responses of the pH cell and cation cell of the antennal taste sensilla of the ground beetle P. aethiops to various 100mM Na+-salts and their

mixtures with 10mM NaOH. Vertical bars denote 7SE of the means. Asterisks indicate significantly different means: paired t-test, Po0.05. Note the

different y-scales.

M. Milius et al. / Journal of Insect Physiology 52 (2006) 960–967964

74

In addition to the specific stimulus, also frequentlynamed the best stimulus, other chemical factors such asionic composition and concentration of electrolytes, thepresence of non-electrolytes, and pH of the stimulatingsolution, may all influence the response of different kindsof insect contact chemoreceptive cells to some extent(Hanson, 1987; Liscia et al., 1997; Bernays et al., 1998;Liscia and Solari, 2000; Bernays and Chapman, 2001b;Schoonhoven and van Loon, 2002). In this context, it is notsurprising that the response of the antennal cation cell ofP. aethiops, in some cases, was also increased by 17–40%by pH increase up to two pH units of the stimulatingsolution, which is some thirty times smaller increase infiring rate than was observed in the pH cell responses underthe same stimulating conditions. Similar stimulatory effectof pH to the cation cell responses was observed inelectrophysiological experiments with acetate and phos-phate buffers also in the ground beetle P. oblongopunctatus(Merivee et al., 2005).

Our tests demonstrated that an increase in pH by 0.3–0.6units in 100mM Na+-salt solutions, caused by the 1mMNaOH content, was too small, except for Na2HPO4, toinfluence the firing rate of the pH cell or cation cellsignificantly. However, different sensitivity of the two cellsto increased pH was clearly demonstrated when theconcentration of NaOH in 100mM stimulating saltsolutions was increased to 10mM resulting in pH increaseby 1.2–2 units depending on the salt. Compared to theacetate and phosphate buffers method (Merivee et al.,2005), a comparable increase in pH by 1–2 units of thestimulatory salt solution using the method of adding smallamounts of NaOH to 100mM Na+-salts evoked aconsiderably larger increase in the rate of firing of the pHcell: respective rates of firing during the 1st s of theresponse increased by 30–400% and 140–1050%, respec-tively, depending on buffers and salts used.

In the stimulating solutions differing by 1.2–2 pH units,and using the NaOH method, the difference in Na+

concentrations is only 10% and the salt anion concentra-tion does not change. By contrast, using the method ofbuffers, respective concentrations of content ions maydiffer up to 30 times, depending on the pH range and saltsinvolved. Taking into consideration that both salt cationsand anions of the stimulating solution may affect the pHcell firing rate, large differences in the concentration of saltions are undesirable when testing pH cell responses to pH.Thus, it seems that the NaOH method is, besides thebuffers method, a reliable tool for studying pH receptorfunctioning in insects.

Our results suggest that in P. aethiops, in its preferredacid forest habitats and overwintering sites in brown-rottedwood at pH 3–5, the antennal pH sensitive cell does notdischarge or discharges at very low frequency with the firsts firing rate close to 1 imp/s or lower. Areas with a higherpH seem to be unfavourable to this insect and whencontacted the pH cell signals with a stronger response. Thismay occur, for example, in places with decaying plant

material on the soil surface. Many of the materialscomposted contain significant amounts of protein, whichis converted to ammonia toxic to many organisms andserving as the primary substrate in the nitrificationprocesses. Nitrification processes have been observed atpH levels ranging from 6.6 to 9.7 (Odell et al., 1996).Obviously, the ground beetle P. aethiops avoids theseplaces with a high pH.Six to seven months, from September to April, adults of

P. aethiops may spend in its preferred overwintering sites inbrown-rotted wood in Estonia. Fungi often lower the pHof the extracellular medium through secretion of organicacids and thus establish a pH gradient around themycelium. The organic acid oxalate is produced by most,if not all, wood-degrading fungi. Usually, pH 3–6 isobserved in both white- and brown-rotted wood. Stablemoisture content is quaranteed as a result of metabolicwater production during the wood-degrading process. Forcultures of white-rot fungus Merulius tremellosus incubatedin air, the optimum water content was 2 g/1 g of wood.Moister conditions were less favourable to delignification,possibly because the water impeded aeration (Reid, 1985;Varela and Tien, 2003). Similar prolonged stability ofwater content and aeration is probably also quaranteed inbrown-rotting wood which seems to be a favourablemicroenvironment to ground beetles for their prolongedhibernation.Brown-rotted wood may offer good protection for

hibernating ground beetles against entomopathogenicfungi and parasites which play an important role in theirpopulation dynamic. Up to 41% parasitism by nematodesand ectoparasitic fungi was found on 14 species ofBembidion in Norway depending on habitat selectionof the host (Andersen and Skorping, 1991). Eggs ofP. oblongopunctatus suffered 83% mortality in fresh litterbut only 7% in sterilized soil (Heessen, 1981). Althoughpredators, parasites, and pathogens affect all ground beetledevelopmental stages, quantitative data remain scarce. So,antennal pH receptors of the ground beetles P. aethiops

seem to be a powerful tool in their selection for favourableacid habitats and microhabitats crucial to survive.Chemically mediated habitat selection has been shown to

occur in several species of ground beetles by Evans (1982,1983, 1984, 1988). He demonstrated that olfactory cellsreceptive to methyl esters of palmitic and oleic acid emittedby the mat-forming blue-green algae Oscillatoria animalis

Agardh and Oscillatoria subbrevis Schmidle are present insensillae on the antennae of those Bembidiini associatedwith the shores of saline lakes. Low concentrations offormic acid, a defence secretion in the ground beetleHarpalus rufipes, was shown by Luff (1986) to causeaggregation in this species. Formic acid is also produced bysome wood-decaying fungi as a result of enzymaticdecarboxylation of oxalic acid accumulating in largequantities in the cell sap of various plants and as a majorproduct of carbohydrate metabolism by many molds(Shimazono and Hayaishi, 1957). Probably, formic acid


75

could be perceived and used for long-distance orientationand aggregation in brown-rotted wood by P. aethiops aswell.

Acknowledgements

Contract grant sponsor: Estonian Science Foundation;Contract Grant No. 5423 and 5736.

References


depends on habitat selection of the host. Canadian Journal of Zoology

69, 1216–1220.

Bernays, E.A., Chapman, R.F., 2001a. Taste cell responses in the

polyphagous arctiid, Grammia geneura: towards a general pattern for

caterpillars. Journal Insect Physiology 47, 1029–1043.

Bernays, E.A., Chapman, R.F., 2001b. Electrophysiological responses

of taste cells to nutrient mixtures in the polyphagous caterpillar

of Grammia geneura. Journal of Comparative Physiology A 187,

205–213.

Bernays, E.A., Glendinning, J.I., Chapman, R.F., 1998. Plant acids

modulate chemosensory responses in Manduca sexta larvae. Physio-

logical Entomology 23, 193–201.

Blaney, W.M., 1974. Electrophysiological responses of the terminal

sensilla on the maxillary palps of Locusta migratoria (L.) to some

electrolytes and non-electrolytes. Journal of Experimental Biology 60,

275–293.

Blomquist, G., Vogt, R. (Eds.), 2003. Insect Pheromone Biochemistry and

Molecular Biology. The Synthesis and Detection of Pheromones and

Plant Volatiles. Elsevier, Academic Press, Amsterdam, Boston,

Heidelberg, London, New York, Oxford, Paris, San Diego, San

Francisco, Singapore, Sidney, Tokyo, p. 745.

Chapman, R.F., 1998. The Insects Structure and Function. Cambridge

University Press, Cambridge, p. 770.

Christensen, T.A. (Ed.), 2005. Methods in Insect Sensory Neuroscience.

CRC Press, Boca Raton, London, New York, Washington, p. 435.

Den Otter, C.J., 1972. Differential sensitivity of insect chemoreceptors to

alkali cations. Journal of Insect Physiology 18, 109–131.

Dethier, V.G., Hanson, F.E., 1968. Electrophysiological responses of the

blowfly to sodium salts of fatty acids. Proceedings of the National

Academy of Sciences, USA 60, 1269–1303.

Elizarov, Yu.A., 1978. Insect Chemoreception. Moscow University Press,

Moscow, p. 232. (in Russian).

Ermakov, A.I., 2004. Structural changes in the carabid fauna of forest

ecosystems under a toxic impact. Russian Journal of Ecology 35,

403–408.

Evans, D.R., Mellon, D., 1962. Stimulation of a primary taste receptor by

salts. Journal of General Physiology 45, 651–661.

Evans, W.G., 1982. Oscillatoria sp. (Cyanophyta) mat metabolites

implicated in habitat selection in Bembidion obtusidens (Coleopt.:

Carabidae). Journal of Chemical Ecology 8, 671–678.

Evans, W.G., 1983. Habitat selection in the Carabidae. Coleopterists

Bulletin 37, 164–167.

Evans, W.G., 1984. Odour-mediated responses of Bembidion obtusidens

(Coleoptera: Carabidae) in a wind tunnel. Canadian Journal of

Entomology 116, 1653–1658.

Evans, W.G., 1988. Chemically mediated habitat recognition in shore

insects (Coleoptera: Carabidae, Hemiptera: Saldidae). Journal of

Chemical Ecology 14, 1441–1454.

Haberman, H., 1968. Ground Beetles of Estonia. Valgus, Tallinn. p. 598

(in Estonian).

Hanson, F.E., 1987. Chemoreception in the fly: the search for the

liverwurst receptor. In: Chapman, R.F., Bernays, E.A., Stoffolano,

J.G. (Eds.), Perspectives in Chemoreception and Behaviour. Springer,

Berlin, Heidelberg, New York, pp. 99–122.

Hansson, B.S. (Ed.), 1999. Insect Olfaction. Springer, Berlin, Heidelberg,

New York, Barcelona, Hong Kong, London, Milan, Paris, Singapore,

Tokyo, p. 457.

Haskell, P.T., Schoonhoven, L.M., 1969. The function of certain mouth

part receptors in relation to feeding in Schistocerca gregaria and

Locusta migratoria migratorioides. Entomologia Experimentalis et

Applicata 12, 429–440.

Heessen, H.J.L., 1981. Egg mortality in P. oblongopunctatus (Coleoptera,

Carabidae). Oecologia 50, 233–235.

Hodgson, E.S., Lettvin, J.Y., Roeder, K.D., 1955. Physiology of a primary

chemoreceptor unit. Science 122, 417–418.

Irmler, U., 2001. Characterisierung der Laufkafergemeinschaften schles-

wig-holsteinischer Walder und Moglichkeiten ihrer okologischen

Bewertung. Angewandte Carabidologie. Laufkafer im Wald

(Suppl. 2), 21–32.

Irmler, U., 2003. The spatial and temporal pattern of carabid beetles on

arable fields in northern Germany (Schleswig–Holstein) and their value

as ecological indicators. Agriculture, Ecosystems & Environment 98,

141–151.

Ishikawa, S., 1963. Responses of maxillary chemoreceptors in the

larvae of the silkworm, Bombyx mori, to stimulation by carbohy-

drates. Journal of Cellular and Comparative Physiology 61,

99–107.

Krogerus, R., 1960. Okologische Studien uber nordische Moorarthropo-

den. Societas Scientiarum. Fennica, Commentationes Biologicae 21

(3), 1–238.

Kula, E., Purchart, L., 2004. The ground beetles (Coleoptera: Carabidae)

of forest altitudinal zones of the eastern part of the Krusne hory Mts.

Journal of Forest Science 50, 456–463.

Lindroth, C.H., 1985. The Carabidae (Coleoptera) of Fennoscandia and

Denmark. Fauna Entomologica Scandinavica 15, 1–225.

Lindroth, C.H., 1986. The Carabidae (Coleoptera) of Fennoscandia and

Denmark. Fauna Entomologica Scandinavica 15, 233–497.

Liscia, A., Solari, P., 2000. Bitter taste recognition in the blowfly:

electrophysiological and behavioral evidence. Physiology & Behavior

70, 61–65.

Liscia, A., Solari, P., Majone, R., Tomassini Barbarossa, I., Crnjar, R.,

1997. Taste receptor mechanisms in the blowfly: evidence of amiloride-

sensitive and insensitive receptor sites. Physiology & Behavior 62,

875–879.

Lovei, G.L., Sunderland, K.D., 1996. Ecology and behavior of ground

beetles (Coleoptera:Carabidae). Annual Review of Entomology 41,

231–256.

Luff, M.L., 1986. Aggregation of some Carabidae in pitfall traps. In: den

Boer, P.J. (Ed.), Carabid Beetles. Gustav Fisher, Stuttgart,

pp. 385–397.

Merivee, E., Renou, M., Mand, M., Luik, A., Heidemaa, M., Ploomi, A.,





Electrophysiological identification of antennal pH-receptors in the

ground beetle Pterostichus oblongopunctatus (Coleoptera, Carabidae).

Physiological Entomology 30, 122–133.

Mitchell, B.K., Schoonhoven, L.M., 1973. Taste receptors in Colorado

beetle larvae. Journal of Insect Physiology 20, 1787–1793.

Mitchell, B.K., Seabrook, W.D., 1973. Electrophysiological investiga-

tions on tarsal chemoreceptors of the spruce budworm,

Choristoneura fumiferana (Lepidoptera). Journal of Insect Physiology

20, 1209–1218.

Odell, L.H., Kirmeyer, G.J., Wilczak, A., Jasangelo, J.G., Marcinko, J.P.,

Wolfe, R.L., 1996. Controlling nitrification in chloraminated systems.

Journal AWWA 88, 86–98.



Reid, I.D., 1985. Biological delignification of aspen wood by solid-state



ARTICLE IN PRESSM. Milius et al. / Journal of Insect Physiology 52 (2006) 960–967966

76


caterpillars: each species its own key. Acta Zoologica Academiae

Scientiarum Hungarica 48, 215–263.

Shimazono, H., Hayaishi, O., 1957. Enzymatic decarboxylation of oxalic

acid. Journal of Biological Chemistry 227, 151–159.

Silferberg, H., 2004. Enumeratio nova Coleopterorum Fennoscandiae,

Daniae et Baltiae. Sahlbergia 9, 1–111.

Sk"odowski, J.J.W., 2005. European Carabidology 2003. Proceedings of the

11th European Carabidologist Meeting. DIAS Report 114, 291–303.

Thiele, H.-U., 1977. Carabid Beetles in Their Environment. Zoophysiol-

ogy and Ecology, vol. 10. Springer, Berlin, p. 369.

Varela, E., Tien, M., 2003. Effect of pH and oxalate on hy-

droquinone-derived hydroxyl radical formation during brown rot

wood degradation. Applied and Environmental Microbiology 69,

6025–6031.

Wachmann, E., Platen, R., Barndt, D., 1995. Laufkafer. Beobachtung,

Lebensweise. Naturbuch Verlag, Leipzig, p. 295.


Merivee, E., Must, A., Milius, M., Luik, A., 2007.

ELECTROPHYSIOLOGICAL IDENTIFICATION OF THE

SUGAR CELL IN ANTENNAL TASTE SENSILLA OF THE

PREDATORY GROUND BEETLE PTEROSTICHUS AETHIOPS.

Journal of Insect Physiology 53(4), 377–384.

79


Electrophysiological identification of the sugar cell in antennal tastesensilla of the predatory ground beetle Pterostichus aethiops

Enno Merivee�, Anne Must, Marit Milius, Anne Luik

Estonian University of Life Sciences, Institute of Agricultural and Environmental Sciences, 64 Kreutzwaldi Street, 51014 Tartu, Estonia

Received 6 October 2006; received in revised form 14 December 2006; accepted 21 December 2006

Abstract

By single sensillum tip recording technique, in addition to the salt and pH cells found in antennal taste sensilla of some ground beetles

earlier, the third chemosensory cell of four innervating these large sensilla was electrophysiologically identified as a sugar cell in the

ground beetle Pterostichus aethiops. This cell generated action potentials of considerably smaller amplitude than those of the salt and pH

cells, and phasic-tonically responded to sucrose and glucose over the range of 1–1000mM tested. Responses were concentration

dependent, with sucrose generating more spikes than glucose. During the first second of the response, maximum rates of firing of the

sugar cell reached up to 19 and 37 imp/s when stimulated with 1000mM glucose and sucrose, respectively. Three to four seconds later,

the responses decreased close to zero. Both sugars are important in plant carbohydrate metabolism. These ground dwelling insects may

come into contact with live and decayed plant material everywhere in their habitat including their preferred overwintering sites in brown-

rot decayed wood. In conclusion, we hypothesize that high content of soluble sugars in their overwintering sites and refugia is

unfavourable for these ground beetles, most probably to avoid contact with dangerous fungi.

r 2007 Elsevier Ltd. All rights reserved.

Keywords: Ground beetles; Antennal taste sensilla; Sugar cell; Tip recording technique; Habitat selection; Brown-rotted wood; Hibernation

1. Introduction

Most electrophysiologists studying insect contact che-moreception have focused on taste sensilla of herbivores(Chapman, 1998,2003; Schoonhoven and van Loon, 2002)and some other insect groups (van der Starre, 1972; Ruth,1976; Hanson, 1987; Hansen-Delkeskamp, 1992; Murataet al., 2002), related to their feeding and oviposition.Usually, these sensilla are located on mouthparts, anten-nae, ovipositor and tarsi. Very little is known aboutfunctioning of taste sensilla in predatory insects such asground beetles (Carabidae). In the tribes Platynini,Pterostichini and Bembidiini, approximately 70 tastebristles are located on the antennae of adults (Meriveeet al., 2000, 2001, 2002). Four chemoreceptive and onemechanoreceptive cell innervate these sensilla in the groundbeetle Nebria brevicollis (Daly and Ryan, 1979). Similar setof sensory cells is widespread in contact chemoreceptors of

insects suggesting that it is characteristic for other groundbeetles too. In the ground beetles Pterostichus aethiops andPterostichus oblongopunctatus, two chemoreceptive cellshave been electrophysiologically identified: they are the salt(cation) cell and pH cell probably involved in choice ofhabitat and overwintering sites of these insects (Meriveeet al., 2004, 2005; Milius et al., 2006). The occurrence of thepH receptive cell in the sensillar apparatus seems to beunique for ground beetles, as it is not found in otherinsects. Stimulating chemical compounds, specific for othertwo chemoreceptive cells innervating antennal taste sensillaof ground beetles, are not known.Besides other chemosensory cells, phagostimulatory

sugar cells are present in all phytophagous insects that havebeen studied and are almost certainly present universally(Schoonhoven and van Loon, 2002; Chapman, 2003). Sugarcells have been also found in various flies (Shiraishi andMorita, 1969; van der Starre, 1972; Hanson, 1987; Liscia etal., 1998, 2002; Furuyama et al., 1999; Amakawa, 2001;Sadakata et al., 2002; Murata et al., 2004), cockroaches(Ruth, 1976; Becker and Peters, 1989; Hansen-Delkeskamp

ARTICLE IN PRESS

www.elsevier.com/locate/jinsphys

0022-1910/$ - see front matter r 2007 Elsevier Ltd. All rights reserved.

doi:10.1016/j.jinsphys.2006.12.012

�Corresponding author.


and Hansen, 1995; Hansen-Delkeskamp, 1998) andapterygotan insects (Zygentoma) (Hansen-Delkeskamp,2001). These results suggest that also in ground beetles,taste sensilla may house the sugar cell. The aim of this studyis to test stimulatory effect of glucose and sucrose, importantin plant carbohydrate metabolism (Kruger, 1993), toantennal taste bristles in the ground beetle P. aethiops

(Panzer, 1796). This ground dwelling forest insect(Thiele, 1977) may come into contact with live and decayedplant material everywhere in its habitat including itspreferred overwintering sites in brown-rot decayed wood.Results of these experiments are reported in this paper.

2. Material and methods

2.1. Test beetles

Test beetles were collected in their preferred overwinteringsites in brown-rot decayed wood at forest margins in southernEstonia in September and October 2005. Beetles were kept inplastic boxes filled with moistened moss and pieces of brown-rot decayed wood in refrigerator at 5–6 1C for a couple ofweeks. Three to four days before experiments, the beetleswere exposed to room temperature (20 1C), fed withmoistened cat food (Kitecat, Master Foods, Poland) andgiven clean water to drink every day.

Intact test beetles were restrained by placing them tightlyinto a special conical tube made of thin sheet aluminium ofa size that allowed their head and antennae to protrudeonly a little from the narrower end. The wider rear end ofthe conical tube was blocked with a piece of plasticine toprevent the beetle from retreating out of the tube.Thereafter, antennae of the beetles were fixed horizontallyon the edge of an aluminium stand with special clamps andtiny amounts of beeswax, so that horizontally located largechaetoid taste sensilla were well visible from above under alight microscope, and easily accessible for micromanipula-tions from the side. Contamination of tested taste bristlesand their contact with preparation instruments andbeeswax was carefully avoided.

2.2. Stimulation and recording of action potentials

Chemicals of analytical grade of purity used in theexperiments were obtained from AppliChem (Germany).Ten millimolar KCl was used as an electrolyte in the mixtureswith sugars, and as a control solution. pH of stimulatingsolutions varying between 6.1 and 6.2 was measured with thepH meter E6115 (Evikon, Estonia). To minimize the possiblemicrobial contamination, stimulating sugar solutions wereprepared no longer than 24h before experiments.

Action potentials from taste bristles were recorded byconventional single sensillum tip-recording technique. Toreach good signal-to-noise ratio, a grounded tungstenmicroelectrode was inserted into the base of the flagellum.The recording glass micropipette with tip diameter of 20mmand filled with stimulating solution was placed over the bristle

tip by means of a micromanipulator under visual controlthrough a light microscope at a magnification of 300� .Immediately before each stimulation, 2–3 drops of solutionwere squeezed out of the micropipette tip, and adsorbed ontoa piece of filter paper attached to an additional micromani-pulator, in order to avoid concentration changes in thecapillary tip due to evaporation. Due to response, variabilityof different taste bristles to the same stimulus stimulatingsolutions were tested in pairs to allow a more precisecomparison of the responses to the tested stimuli. Thus, eachtaste sensillum was tested twice. Primarily, three to five sensillaof each test beetle were tested with the control stimulus. Then,approximately after 30min, the stimulating/recording micro-pipette was rinsed with distilled water several times and filledwith the second stimulating solution containing sugar to testthe same sensilla again. In each next test beetle the order ofpresentation of stimulating solutions was alternated and thenfiltered with a band width set at 100–2000Hz and amplified(input impedance 10GO) signals were monitored on anoscilloscope screen, and relayed to a computer via ananalogue-to-digital input board DAS-1401 (Keithley, Taun-ton, MA, USA) for data acquisition, storage, and analysisusing TestPoint software (Capital Equipment Corporation,Billerica, MA, USA) at a sampling rate of 10kHz. Recordingswere made during 5-sec stimulation period in order tocharacterize both rapid phasic and relatively slow stabilizationperiod of the responses of the chemosensory cells involved.

2.3. Data analysis

Automatic classifying and counting of action potentialsfrom taste sensilla was clearly not an appropriate methodfor data analysis because two action potentials frequentlyinteracted to produce an irregular waveform. Instead, spikesfrom several chemosensory cells innervating antennal tastesensilla tested were visually distinguished by their amplitudeand polarity of spikes, and counted, using TestPointsoftware. In some cases, the regularity of firing as theprincipal parameter for separating and identifying actionpotentials was used, whereas contact artefact lasting up to25ms seen at onset of stimulus was not corrected by ourrecording amplifier, 3–4 spikes may have lost. So as thecontact artefact of the same duration occurs in both cases,when stimulated with control as well as with test solutioncontaining sugar, it does not essentially affect main resultsand conclusions of the work. Firing rate (imp/s) was used tomeasure the response. t-test for paired samples was used todetermine different means significantly. Responses fromthree to five taste sensilla of each test beetle were analysed.The number of beetles (N) in each test series was 7–16.

3. Results

3.1. Responses of antennal taste sensilla to 10mM KCl

Stimulation of antennal taste bristles of the groundbeetle P. aethiops by 10mM KCl only, serving as an

ARTICLE IN PRESSE. Merivee et al. / Journal of Insect Physiology 53 (2007) 377–384378

81

electrolyte and control solution in the experiments, causedtwo chemosensory cells to respond phasic-tonically. Thesecells were the salt (cation) cell and the pH cell. Comparedto the action potentials with negative polarity generated bythe salt cell those of the pH cell were positive andconsiderably larger in amplitude (Fig. 1A). In varioussamples of test insects, the first second response of the pHcell to this electrolyte varied to a great degree, from 4to15 imp/s. During the next 4 s of stimulation, however, therate of firing of the pH cell fell to the level of 0–2 imp/s(Figs. 2C and 3C). In contrast to the pH cell, firing rate ofthe salt cell was highly stable in different samples of testbeetles with first second responses varying between 11 and13 imp/s (Figs. 2B and 3B). In some recordings, a few smallaction potentials with negative polarity of the third cellwere observed. Their mean rate of firing usually did notexceed 1 imp/s at the beginning of stimulation. Thereafter,this low rate of firing fell to the level close to zero.

3.2. Responses of antennal taste sensilla to the mixtures of

10mM KCl and glucose

In response to the mixtures of 10mM KCl and sugars, inaddition to the action potentials from the salt and pH cells,

numerous spikes of the third cell were observed in therecordings (Fig. 1B). The responses of the third cell tostimulation solutions containing glucose were significantlystronger compared to control solution (po0.05), except forthe solutions with 1 and 1000mM glucose at the end of the5-s stimulation period, when the responses fell much below1 imp/s and they did not differ from control solution(p40.05). During the first second of stimulation, over therange of 1–1000mM glucose tested, firing rate of the thirdcell drastically increased with concentration increase from2 to 19 imp/s (po0.05). During the next 4 s of stimulation,firing rate of the third cell fell below 1–2 imp/s, and nodependence of the response on glucose concentration wasobserved any longer. This cell was classified as a phasic-tonic sugar cell (Fig. 2A).In contrast, the responses of the salt cell were not

significantly affected by the glucose content of thestimulating solutions (p40.05). Its first second responsesvaried between 11 and 14 imp/s. The responses fell below1 imp/s during the next s of stimulation, and close to zerothereafter (Fig. 2B). The responses of the pH cell were alsoaffected by the content of glucose in the stimulatingsolutions, but in a more complicated manner. Firing rate ofthis cell increased to some degree at low concentrations of

ARTICLE IN PRESS

Fig. 1. Typical responses from antennal taste bristles of the ground beetle Pterostichus aethiops: (A) 10mM KCl (control solution), (B) the mixture of 10

mMKCl and 100mM sucrose. Arrows indicate the beginning of stimulation. Action potentials of the sugar cell (SC), salt cell (CC) and pH cell (pHC) were

distinguished.

E. Merivee et al. / Journal of Insect Physiology 53 (2007) 377–384 379

82

glucose (1 and 10mM solutions). At the beginning of theresponses (first two seconds), stimulating effect of glucoseto the pH cell was statistically significant (po0.05). Athigher concentrations (100 and 1000mM solutions) theopposite was observed, however: firing rates of the pH cellwere slightly suppressed by glucose compared to theresponses evoked by 10mM KCl alone (Fig. 2C).

3.3. Responses of antennal taste sensilla to the mixtures of

10mM KCl and sucrose

Generally, responses of antennal taste bristles to sucrosewere similar to those obtained in tests with glucose (Fig. 3A).

Our results demonstrated that the presence of sucrosein stimulating solutions caused the sugar cell to respondstrongly being nearly silent when stimulated by 10mM KClalone. Difference was statistically significant (po0.05).Stimulatory effect of sucrose to the sugar cell was nearlytwo times stronger, however, compared to that of glucose.First second responses varied from 3 to 37 imp/s dependingon the sugar concentration. A strong dependence of thesugar cell firing rate on the sucrose concentration wasnoted throughout the 5-s period of stimulation.In contrast, 1–1000mM sucrose had only little or no

stimulatory effect to the salt cell (Fig. 3B). It was alsoobserved that 1 and 10mM solutions of sucrose were

ARTICLE IN PRESS

Fig. 2. Firing rates of antennal taste sensilla in response to glucose in the ground beetle P. aethiops. To compare stimulating effects of glucose more

precisely, the control and stimulus solutions were tested in pairs. Asterisks show significantly different means (po.05). Vertical bars indicate SE. Note the

different y-scales.

E. Merivee et al. / Journal of Insect Physiology 53 (2007) 377–384380

83

slightly stimulatory for the pH cell. At higher concentra-tions, however, this sugar caused the pH cell to fire at lowerfrequency of action potentials compared to that evoked by10mM control solution (Fig. 3C).

4. Discussion

Binding of ground beetles to certain habitat types andmicrohabitats is determined by numerous biotic andabiotic factors such as food conditions, vegetation type,presence and distribution of competitors, life history andseason, landscape characteristics, agricultural cultivationimpacts, temperature and humidity conditions, lightintensity, and others. Favourite wintering sites are wellaerated. Habitat choice in ground beetles is so specific that

they are often used to characterize habitats (Thiele, 1977;Lovei and Sunderland, 1996; Ings and Hartley, 1999; Coleet al., 2002; Blake et al., 2003; Eyre and Luff, 2004; Purtaufet al., 2005). Chemically mediated habitat selection hasbeen shown to occur in several species of ground beetles byEvans (1982, 1983, 1984, 1988). He demonstrated thatolfactory cells receptive to methyl esters of palmitic andoleic acid emitted by the mat-forming blue-green algaeOscillatoria animalis Agardh and Oscillatoria subbrevis

Schmidle are present in sensillae on the antennae of thoseBembidiini associated with the shores of saline lakes. Saltcontent and pH of the soil are also important factors indistribution of ground beetles (Thiele, 1977; Paje andMossakowski, 1984; Hoback et al., 2000; Irmler,2001,2003; Magura, 2002). Behavioural experiments have

ARTICLE IN PRESS

Fig. 3. Responses of antennal taste sensilla to sucrose in the ground beetle P. aethiops. Details as in Fig. 2.


84

shown that presumptive contact chemoreceptive sensillaeare located on the ground beetles’ antennae (Paje andMossakowski, 1984; Hoback et al., 2000).

By electrophysiological experiments it is convincinglydemonstrated that the salt and pH cells innervatenumerous antennal taste bristles in ground beetles indeed.The used conventional single sensillum tip-recordingtechnique (Hodgson et al., 1955), and the location of thegrounded microelectrode at the base of antennal flagellumfree of muscle tissues guaranteed that action potentialsrecorded originated from the sensory cells of the tastesensillum contacted and stimulated by the recordingmicroelectrode. The unusual positive polarity and extre-mely long duration of the potentials generated by the pHcell have been repeatedly pointed out and discussed earlierin various ground beetles. Unfortunately, we have nosatisfactory explanation for this phenomenon today(Merivee et al., 2004, 2005). Detailed transmission electronmicroscope studies are needed to get new detailedinformation on the inner structure of these sensilla.Intracellular recordings of action potentials from the pHcell may probably finally solve the problem of the realpolarity of their unusual positive potentials recorded byextracellular single sensillum tip-recording technique. Dueto the technical complicacy of intracellular recording ofaction potentials from insect antennal taste sensilla coveredwith a thick and hard cuticular wall, no such attempts havebeen made so far, however. With regard to the pHreceptive cell, it was suggested that in P. aethiops andP. oblongopunctatus, in their preferred acid forest habitatsand overwintering sites in brown-rotted wood at pH 3–5the antennal pH sensitive cell does not discharge ordischarges at very low frequency with the first secondfiring rate close to 1 imp/s or lower. Areas with higher pHseems to be unfavourable for these insects and whencontacted the pH cell signals with a stronger response(Merivee et al., 2004, 2005; Milius et al., 2006). Therefore,we hypothesize that the function of the third chemorecep-tive cell of the taste bristles identified as a sugar cell in theforest ground beetle P. aethiops, is also related to habitatand/or microhabitat choice. This cell generated actionpotentials of considerably smaller amplitude than those ofthe salt and pH cells, and responded phasic-tonically tosucrose and glucose over the all range of 1–1000mMstimulating solutions tested, a span that covers the range ofsoluble sugar levels generally present in green plants, i.e.1–50mM/l (Schoonhoven and van Loon, 2002; Chapman,2003). Significantly, higher rates of firing were recorded inresponse to sucrose compared to that evoked by glucose.During the first second of the response, maximum rates offiring of the sugar cell reached up to 19 and 37 imp/s whenstimulated with glucose and sucrose, respectively. Bothsugars are important in plant carbohydrate metabolism.These results are in a good consistency with literature dataobtained by similar experiments with phytophagousinsects. Various mono-, di-, and trisaccharides maystimulate the sugar cell with different effectiveness in some

insects, whereas in other species this receptor respondsexclusively to sucrose or glucose or fructose or othercarbohydrate. These compounds function as strong pha-gostimulants and important stimuli for host recognition tomost herbivorous insects (Mitchell and Gregory, 1979;Schoonhoven, 1987; Bernays and Chapman, 2001; Bernayset al., 2002; Schoonhoven and van Loon, 2002; Chapman,2003) suggesting that the sugar cell found in antennal tastebristles of the predatory ground beetle P. aethiops may alsobe related to detection for food.On the other hand, gustatory apparatuses are usually

located in the labella, legs, wings, palpi and less frequently inthe antennae. Further, there exist strong evidences that thepH cell at least, and probably also the salt cell innervatingthese antennal taste bristles are responsible for habitatselection in some ground beetles (Thiele, 1977; Paje andMossakowski, 1984; Hoback et al., 2000). So, participationof the sugar cell in searching for suitable habitat is alsopossible in ground beetles. The predatory ground beetleP. aethiops may come into contact with live and decayedplant material everywhere in its habitat including itspreferred overwintering sites in brown-rot decayed woodwhich does not contain soluble sugars and cellulose, andwhich is composed of pure lignin. Stable moisture contentand aeration are quaranteed as a result of metabolic waterproduction during the wood-degrading process. The contentof sugars and pH depend on the assembly of wood-decayingfungi involved, and change during this process (Reid, 1985;Enoki et al., 1988; Zabel and Morrell, 1992; Varela and Tien,2003) reflecting the quality of decaying wood as a favourablemicroenvironment to ground beetles for their prolongedhibernation. In addition, brown-rotted wood may offer goodprotection for hibernating ground beetles against entomo-pathogenic fungi and parasites which play an important rolein their population dynamic. Up to 41% parasitism bynematodes and ectoparasitic fungi was found on 14 speciesof Bembidion in Norway depending on habitat selectionof the host (Andersen and Skorping, 1991). Eggs ofP. oblongopunctatus suffered 83% mortality in fresh litterbut only 7% in sterilized soil (Heessen, 1981). Althoughpredators, parasites, and pathogens affect all developmentalstages of ground beetles, quantitative data remain scarce.Hence, antennal pH, salt and sugar receptors of the groundbeetle P. aethiops seem to be a powerful tool in their selectionof suitable habitats and microhabitats crucial to survive.

Acknowledgements

Contract grant sponsor: Estonian Science Foundation;Contract grant numbers 5423 and 5736.

References

Amakawa, T., 2001. Effects of age and blood sugar levels on the proboscis

extension of the blow fly Phormia regina. Journal of Insect Physiology

47, 195–203.


85


depends on habitat selection of the host. Canadian Journal of Zoology

69, 1216–1220.

Becker, A., Peters, W., 1989. Localization of sugar-binding sites in contact

chemosensilla of Periplaneta americana. Journal of Insect Physiology

35, 239–250.

Bernays, E.A., Chapman, R.F., 2001. Taste cell responses in the

polyphagous arctiid, Grammia geneura: towards a general pattern

for caterpillars. Journal of Insect Physiology 47, 1029–1043.

Bernays, E.A., Chapman, R.F., Hartmann, T., 2002. A highly sensitive

taste receptor for pyrrolizidine alkaloids in the lateral galeal sensillum

of a polyphagous caterpillar, Estigmene acraea. Journal of Compara-

tive Physiology A 188, 715–723.

Blake, S., McCracken, D.I., Eyre, M.D., Garside, A., Foster, G.N., 2003.

The relationship between the classification of Scottish ground

beetle assemblages (Coleoptera, Carabidae) and the national vegeta-

tion classification of British plant communities. Ecography 26,

602–616.

Chapman, R.F., 1998. The Insects. Structure and Function. Cambridge

University Press, Cambridge, p. 770.

Chapman, R.F., 2003. Contact chemoreception in feeding by phytopha-

gous insects. Annual Review of Entomology 48, 455–484.

Cole, L.J., McCracken, D.I., Dennis, P., Downie, I.S., Griffin, A.L.,

Foster, G.N., Murphy, K.J., Waterhouse, T., 2002. Relationships

between agricultural management and ecological groups of ground

beetles (Coleoptera: Carabidae) on Scotish farmland. Agriculture,

Ecosystems and Environment 93, 323–336.

Daly, P.J., Ryan, M.F., 1979. Ultrastructure of antennal sensilla of Nebria

brevicollis (Fab.) (Coleoptera: Carabidae). International Journal of

Insect Morphology and Embryology 8, 169–181.

Enoki, A., Tanaka, H., Fuse, G., 1988. Degradation of lignin-related

compounds, pure cellulose, and wood components by white-rot and

brown-rot fungi. Holzforschung 42, 85–93.

Evans, W.G., 1982. Oscillatoria sp. (Cyanophyta) mat metabolites

implicated in habitat selection in Bembidion obtusidens (Coleopt.:

Carabidae). Journal of Chemical Ecology 8, 671–678.

Evans, W.G., 1983. Habitat selection in the Carabidae. The Coleopterists’

Bulletin 37, 164–167.

Evans, W.G., 1984. Odour-mediated responses of Bembidion obtusidens

(Coleoptera: Carabidae) in a wind tunnel. Canadian Journal of

Entomology 116, 1653–1658.

Evans, W.G., 1988. Chemically mediated habitat recognition in shore

insects (Coleoptera: Carabidae, Hemiptera: Saldidae). Journal of

Chemical Ecology 14, 1441–1454.

Eyre, M.D., Luff, M.L., 2004. Ground beetle species (Coleoptera,

Carabidae) associations with land cover variables in northern England

and southern Scotland. Ecography 27, 417–426.

Furuyama, A., Koganezawa, M., Shimada, I., 1999. Multiple receptor

sites for nucleotide reception in the labellar taste receptor cells of the

fleshfly Boettcherisca peregrina. Journal of Insect Physiology 45,

249–255.

Hansen-Delkeskamp, E., 1992. Functional characterization of antennal

contact chemoreceptors in the cockroach Periplaneta americana.


Hansen-Delkeskamp, E., 1998. Development of specific responses in

antennal taste hairs after ecdysis. An electrophysiological investigation

of the cockroach, Periplaneta brunnea. Journal of Insect Physiology 44,

659–666.

Hansen-Delkeskamp, E., 2001. Responsiveness of antennal taste hairs of

the apterygotan insect, Thermobia domestica (Zygentoma); an electro-

physiological investigation. Journal of Insect Physiology 47, 689–697.

Hansen-Delkeskamp, E., Hansen, K., 1995. Responses and spike

generation in the largest antennal taste hairs of Periplaneta brunnea

Burm. Journal of Insect Physiology 41, 773–781.

Hanson, F.E., 1987. Chemoreception in the fly: the search for the

liverwurst receptor. In: Chapman, R.F., Bernays, E.A., Stoffolano,

J.G. (Eds.), Perspectives in Chemoreception and Behaviour. Springer,

Germany, pp. 99–122.

Heessen, H.J.L., 1981. Egg mortality in P. oblongopunctatus (Coleoptera,

Carabidae). Oecologia 50, 233–235.

Hoback, W.W., Golick, D.A., Svatos, T.M., Spomer, S.M., Higley, L.G.,

2000. Salinity and shade preferences result in ovipositional differences

between sympatric tiger beetle species. Ecological Entomology 25,

180–187.

Hodgson, E.S., Lettvin, J.Y., Roeder, K.D., 1955. Physiology of a primary

chemoreceptor unit. Science 122, 417–418.

Ings, T.C., Hartley, S.E., 1999. The effect of habitat structure on carabid

communities during the regeneration of a native Scotish forest. Forest

Ecology and Management 119, 123–136.

Irmler, U., 2001. Charakterisierung der Laufkafergemeinschaften schles-

wig-holsteinischer Walder und Moglichkeiten ihrer okologischen

Bewertung. Angewandte Carabidologie, Supplement 2. Laufkafer im

Wald, 21–32.

Irmler, U., 2003. The spatial and temporal pattern of carabid beetles on

arable fields in northern Germany (Schleswig–Holstein) and their value

as ecological indicators. Agriculture, Ecosystems and Environment 98,

141–151.

Kruger, N.J., 1993. Carbohydrate synthesis and degradation. In: Dennis,

D.T., Turpin, D.H. (Eds.), Plant Physiology, Biochemistry and

Molecular Biology. Longman Scientific and Technical, Singapore,

pp. 59–76.

Liscia, A., Majone, R., Solari, P., Tomassini Barbarossa, I., Crnjar, R.,

1998. Sugar response differences related to sensillum type and location

on the labella of Protophormia terraenovae: a contribution to spatial

representation of the stimulus. Journal of Insect Physiology 44, 471–481.

Liscia, A., Crnjar, R., Masala, C., Sollai, G., Solari, P., 2002. Sugar

reception in the blowfly: a possible Ca++ involvement. Journal of


Lovei, G.L., Sunderland, K.D., 1996. Ecology and behavior of ground

beetles (Coleoptera: Carabidae). Annual Review of Entomology 41,

231–256.

Magura, T., 2002. Carabids and forest edge: spatial pattern and edge

effect. Forest Ecology and Management 157, 23–37.

Merivee, E., Ploomi, A., Rahi, M., Luik, A., Sammelselg, V., 2000.

Antennal sensilla of the ground beetle Bembidion lampros Hbst

(Coleoptera, Carabidae). Acta Zoologica (Stockholm) 81, 339–350.

Merivee, E., Ploomi, A., Luik, A., Rahi, M., Sammelselg, V., 2001.

Antennal sensilla of the ground beetle Platynus dorsalis (Pontoppidan,

1763) (Coleoptera, Carabidae). Microscopy Research and Techique 55,

339–349.

Merivee, E., Ploomi, A., Rahi, M., Bresciani, J., Ravn, H.P., Luik, A.,

Sammelselg, V., 2002. Antennal sensilla of the ground beetle

Bembidion properans Steph. (Coleoptera, Carabidae). Micron 33,

429–440.

Merivee, E., Renou, M., Mand, M., Luik, A., Heidemaa, M., Ploomi, A.,





Electrophysiological identification of antennal pH-receptors in the

ground beetle Pterostichus oblongopunctatus (Coleoptera, Carabidae).

Physiological Entomology 30, 122–133.

Milius, M., Merivee, E., Williams, I., Luik, A., Mand, M., Must, A., 2006.

A new method for Electrophysiological identification of antennal pH

receptor cells in ground beetles: the example of Pterostichus aethiops

(Panzer, 1796) (Coleoptera, Carabidae). Journal of Insect Physiology

52, 960–967.

Mitchell, B.K., Gregory, P., 1979. Physiology of the maxillary sugar

sensitive cell in the red turnip beetle, Entomoscelis americana. Journal

of Comparative Physiology 132, 167–178.

Murata, Y., Kataoka–Shirasugi, N., Amakawa, T., 2002. Electrophysio-

logical studies of salty taste modification by organic acids in the

labellar taste cell of the blowfly. Chemical Senses 27, 57–65.

Murata, Y., Mashiko, M., Ozaki, M., Amakawa, T., Nakamura, T., 2004.

Intrinsic nitric oxide regulates the taste response of the sugar receptor

cell in the blowfly, Phormia regina. Chemical Senses 29, 75–81.

ARTICLE IN PRESSE. Merivee et al. / Journal of Insect Physiology 53 (2007) 377–384 383

86



Purtauf, T., Roschewitz, I., Dauber, J., Thies, C., Tscharntke, T., Wolters,

V., 2005. Landscape context of organic and conventional farms:

influences on carabid beetle diversity. Agriculture, Ecosystems and

Environment 108, 165–174.

Reid, I.D., 1985. Biological delignification of aspen wood by solid-state



Ruth, E., 1976. Elektrophysiologie der Sensilla Chaetica auf den Antennen

von Periplaneta americana. Journal of comparative Physiology 105,

55–64.

Sadakata, T., Hatano, H., Koseki, T., Koganezawa, M., Shimada, I.,

2002. The effect of amiloride on the labellar taste receptor cells of the

fleshfly Boettcherisca peregrina. Journal of Insect Physiology 48,

565–570.

Schoonhoven, L.M., 1987. What makes a caterpillar eat? The sensory

code underlying feeding behavior. In: Chapman, R.F.et al. (Ed.),

Perspectives in Chemoreception and Behavior. Springer, New York,

pp. 69–97.


caterpillars: each species its own key. Acta Zoologica Academiae

Scientiarum Hungaricae 48, 215–263.

Shiraishi, A., Morita, H., 1969. The effects of pH on the labellar sugar

receptor of the fleshfly. The Journal of General Physiology 53, 450–470.

Thiele, H.-U., 1977. Carabid Beetles in Their Environment. Zoophysiology

and Ecology, vol.10. Springer, Berlin, p. 369.

Van der Starre, H., 1972. Tarsal taste discrimination in the blowfly,

Calliphora vicina Robineau-Desvoidy. Netherlands Journal of Zool-

ogy 22, 227–282.

Varela, E., Tien, M., 2003. Effect of pH and oxalate on hydroquinone-

derived hydroxyl radical formation during brown rot wood degrada-

tion. Applied and Environmental Microbiology 69, 6025–6031.

Zabel, R.A., Morrell, J.J., 1992. Chemical changes in wood caused by

decay fungi. In: Wood Microbiology Decay and its Prevention.

Academic, San Diego, pp. 195–224.


Merivee, E., Märtmann, H., Must, A., Milius, M., Williams, I.,

Mänd, M., 2008.

ELECTROPHYSIOLOGICAL RESPONSES FROM NEURONS

OF ANTENNAL TASTE SENSILLA IN THE POLYPHAGOUS

PREDATORY GROUND BEETLE PTEROSTICHUS

OBLONGOPUNCTATUS (fABRICIUS 1787) TO PLANT SUGARS

AND AMINO ACIDS.


89

Electrophysiological responses from neurons of antennal taste sensilla inthe polyphagous predatory ground beetle Pterostichus oblongopunctatus(Fabricius 1787) to plant sugars and amino acids

Enno Merivee *, Helina Martmann, Anne Must, Marit Milius, Ingrid Williams, Marika Mand

Estonian University of Life Sciences, 1 Kreutzwaldi Street, 51014 Tartu, Estonia

1. Introduction

Many species of ground beetles (Coleoptera, Carabidae) areimportant as natural enemies of phytophagous pest insects(Kromp, 1999; Symondson et al., 2002), but the chemical cuesthat they use in habitat selection and food recognition have beenlittle studied. More knowledge of these cues and their sensorydetection may aid the development of strategies to manipulatecarabid assemblages for plant protection in the future. Thecommon Euro-Siberian ground beetle, Pterostichus oblongopuncta-tus, the subject of this study, is a eurytopic woodland species,occuring in both deciduous and coniferous forests. It reproduces

in the spring; newly emerged adults appear in the autumn andhibernate in decayingwood, under the bark of tree trunks, in forestlitter and under stones (Haberman, 1968; Lindroth, 1986).

Scanning electron microscope studies show that the antennaeof ground beetles are well equipped with various chemoreceptors.In addition to numerous small basiconic olfactory sensilla, largerbristle-like contact chemosensilla have been found on the antennalflagellum of adults in several genera: approximately 70 suchsensilla in Bembidion, Platynus and Pterostichus (Merivee et al.,2000, 2001, 2002, 2005) and 56 in Carabus (Kim and Yamasaki,1996). In Nebria brevicollis these large sensilla are each innervatedby one mechanoreceptive and four chemoreceptive neurons (Dalyand Ryan, 1979). In Pterostichus aethiops and P. oblongopunctatus,specific chemical stimuli have been identified by electrophysiologyfor three of the four chemosensory neurons. One responds tovarious salts and another to the pH of the stimulating solution(Merivee et al., 2004, 2005; Milius et al., 2006). In P. aethiops, thethird neuron is sensitive to sucrose and glucose, important in plant


A R T I C L E I N F O

Article history:

Received 10 March 2008

Received in revised form 13 May 2008

Accepted 19 May 2008

Keywords:

Contact chemoreceptors

Electrophysiology

Firing rate

Feeding

Phytophagy

Polyphagy

A B S T R A C T

The responses of antennal contact chemoreceptors, in the polyphagous predatory ground beetle

Pterostichus oblongopunctatus, to twelve 1–1000 mmol l�1 plant sugars and seven 10–100 mmol l�1

amino acids were tested. The disaccharides with an a-1.4-glycoside linkage, sucrose and maltose, were

the twomost stimulatory sugars for the sugar-sensitive neuron innervating these contact chemosensilla.

The firing rates they evoked were concentration dependent and reached up to 70 impulses/s at

1000 mmol l�1. The stimulatory effect of glucose on this neuronwas approximately two times lower. This

can be partly explained by the fact that glucose exists in at least two anomeric forms, a and b. These two

forms interconvert over a timescale of hours in aqueous solution, to a final stable ratio of a:b 36:64, in a

process called mutarotation. So the physiologically active a-anomere forms only 36% of the glucose

solution which was reflected in its relatively low dose/response curve. Due to the partial herbivory of P.

oblongopunctatus these plant sugars are probably involved in its search for food, for example, for conifer

seeds. Several carbohydrates, in addition to glucose, such as cellobiose, arabinose, xylose, mannose,

rhamnose and galactose are known as components of cellulose and hemicelluloses. They are released by

brown-rot fungi during enzymatic wood decay. None of them stimulated the antennal sugar-sensitive

neuron. They are therefore not implicated in the search for hibernation sites, which include rottingwood,

by this beetle. The weak stimulating effect (below 3 impulses/s) of some 100 mmol l�1 amino acids

(methionine, serine, alanine, glutamine) to the 4th chemosensory neuron of these sensilla was

characterized as non-specific, or modulating the responses of non-target chemosensory neurons.

� 2008 Elsevier Ltd. All rights reserved.

* Corresponding author at: Estonian University of Life Sciences, Institute of

Agricultural and Environmental Sciences, 1 Kreutzwaldi Street, 51014 Tartu,

Estonia.


Contents lists available at ScienceDirect

Journal of Insect Physiology

journa l homepage: www.e lsev ier .com/ locate / j insphys

0022-1910/$ – see front matter � 2008 Elsevier Ltd. All rights reserved.

doi:10.1016/j.jinsphys.2008.05.002

90

carbohydrate metabolism (Merivee et al., 2007). As habitatselection in ground beetles appears to be influenced by soil pHand salinity (Krogerus, 1960; Thiele, 1977; Paje and Mossakowski,1984; Hoback et al., 2000; Irmler, 2001), the antennal pH- and salt-sensitive neuron may be involved in this process. Electrophysio-logical experiments with the pH-sensitive neuron in P. oblongo-

punctatus and P. aethiops demonstrated that this neuron did not fireor fired at very low frequency (first second response below5 impulses/s) when stimulated at pH 3–6. This level of acidityreflects the pH preferences of these species in their forest habitatand hibernation sites in brown-rotted wood. Firing rates ofthe neuron increased with increase in pH, up to 20–30 impulses/sat pH 8–11 (Merivee et al., 2005; Milius et al., 2006). Theseelectrophysiological results are in agreement with the acid soilpH preferences of P. oblongopunctatus obtained in behaviouraland ecological experiments (Paje and Mossakowski, 1984; Irmler,2001).

In addition to the salt- and pH-sensitive neuron, the sugar-sensitive neuron present in the same antennal taste sensillummayalso be involved in habitat/microhabitat selection by these species.Various sugars and amino acids function as strong phagostimu-lants in most phytophagous insects equipped with specializedreceptor neurons for these chemicals (Schoonhoven and van Loon,2002; Chapman, 2003). Adults of the genus Pterostichus appear tobe primarily carnivorous, feeding on small prey animals, but inmany species, plant matter also forms part of their diet, and, insome the proportion of animal to plant matter eaten varies withseason (Thiele, 1977; Hengeveld, 1980; Lovei and Sunderland,1996). Johnson and Cameron (1969) collected data on the extent ofherbivory in more than 150 species of ground beetles. Somespecies in the genera Pterostichus, including P. oblongopunctatus,Amara, Agonum and Harpalus consume conifer seeds and youngseedlings immediately following germination. In Sweden, P.

oblongopunctatus and Calathus micropterus are the most importantpredators of Pinus sylvestris seed. In the field, seed predationtypically resulted in >20% seed mortality, reaching even higherlevels (up to 60%) on some occasions andmost predation (81% of alldamaged seeds) was due to ground beetles (Heikkila, 1977;Nystrand and Granstrom, 2000). Therefore, as in phytophagousinsects, polyphagous predatory ground beetles, including P.

oblongopunctatus, may also have plant sugar- and amino acid-sensitive neurons associated with their feeding habits.

The aim of this study was to determine whether sugar- andamino acid-sensitive receptor neurons are present in the antennaltaste sensilla of adult P. oblongopunctatus by testing for stimulatoryeffects of various sugars, mostly of plant origin, and amino acids onthese sensilla by electrophysiology. As the composition and contentof water-soluble sugars differ substantially in live and decayingplant matter (Eriksson et al., 1990; Zabel and Morrell, 1992;Valaskova and Baldrian, 2006) the reaction spectrum of any sugar-sensitive antennal neurons, if present, should give some informationon the functioning of antennal contact chemoreceptors and howthey might be involved in habitat selection and food recognition bythis species.

2. Materials and methods

2.1. Insects

Adult P. oblongopunctatus were collected from a local popula-tion in southern Estonia. Theywere kept in plastic boxes filledwithmoistened moss and pieces of brown-rotted wood in a refrigeratorat 5–6 8C. Three to four days prior to the experiments, the beetleswere transferred to room temperature (23 8C), fed with moistenedcat food (Friskies Vitality+, Nestle Purina, Hungary) and given clean

water to drink every day. Electrophysiological experiments withhibernating beetles were conducted in February and March 2006and with active beetles from June to August 2007.

2.2. Electrophysiology

Each beetle to be testedwas restrained securely in a conical tubemade of thin sheet-aluminium of a size that allowed the head andantennae to protrude from the narrower end. The wider rear end ofthe tube was blocked with plasticine to prevent the beetle fromretreating outof the tube. Its antennaewerefixedhorizontally to theedge of an aluminium stand with special clamps and beeswax, sothat the horizontally located large chaetoid taste sensilla werevisible fromaboveundera lightmicroscope, andeasily accessible formicromanipulation from the side. Care was taken not to contam-inate the taste sensilla by contact with instruments or beeswax.Each prepared test beetle was placed in a Faraday cage (15 cm�10 cm� 5 cm) for electrophysiology.

Action potentials from taste neurons of the sensilla wererecorded by the conventional single sensillum tip-recordingtechnique. To achieve a good signal-to-noise ratio, the indifferenttungsten microelectrode was inserted into the base of the flage-llum free of muscular tissue. It was connected to an AC amplifiervia a calibration source, and earthed. The recording glass micro-pipette filled with stimulating solution was brought into contactwith the tip of the sensillum by means of a micromanipulatorunder visual control through a lightmicroscope at amagnificationof 300�. Immediately before each stimulation, 2–3 drops ofsolution were squeezed out of the micropipette tip, and adsorbedonto a piece of filter paper attached to anothermicromanipulator,in order to avoid concentration changes in the capillary tip dueto evaporation. Action potentials picked up by the recordingelectrode were filtered with a bandwidth set at 100–2000 Hz,amplified (input impedance 10 GV), monitored on an oscillo-scope screen, and relayed to a computer via an analogue-to-digitalinput board DAS-1401 (Keithley, Taunton, MA, USA) for dataacquisition, storage, and analysis using TestPoint software(Capital Equipment Corp., Billerica, MA, USA) at a sampling rateof 10 kHz. Responses were recorded from neurons of three to fiverandomly selected taste sensilla on the flagellomeres 1–9 of eachtest beetle. Each sensillum was stimulated once only with onesolution. Action potentials generated during the first second afterthe beginning of stimulus applicationwere counted and analysed.Responses of the sensilla to mechanical stimulation were alsotested, especially in those rare cases when it was not clearwhether a certain class of recorded action potentials originatedfrom the chemosensory ormechanosensory neuron. The tip of thesensillumwas rapidly moved from its resting position toward theantennal shaft by 15–208, held in this position for 2–3 s, and thenrapidly moved back to its initial position using the recordingelectrode.

2.3. Chemical stimuli

Themain tests were carried out using twelve 1–1000 mmol l�1

sugars and seven 10–100 mmol l�1 amino acids (Table 1) dis-solved in 10 mmol l�1 choline chloride to give good conductivity.Compared to KCl and NaCl, which are frequently used aselectrolytes in solutions of stimulating non-electrolytes, cholinechloride elicits only a few action potentials of large amplitudegenerated by the salt- and pH-sensitive neuron innervating theantennal taste sensilla in ground beetles. As a result, discrimina-tion and counting of small spikes generated by the sugar-sensitiveneuron is easier, especially at the beginning of the response.Chemicals of analytical grade purity were obtained from


91

AppliChem (Germany). To minimize microbial contamination,stimulating solutions were prepared within 48 h of electrophy-siology.

2.4. Data analysis

Automated classification and counting of action potentials fromtaste sensilla was not an appropriate method for data analysisbecause two action potentials frequently interacted to produce anirregular waveform. Instead, spikes from several chemosensoryneurons innervating antennal taste sensilla tested were visuallydistinguished by their polarity and amplitude (Fig. 1), and counted,using TestPoint software. In some cases, the regularity of firingwasused as the principal parameter for separating and identifyingaction potentials. Firing rates (impulses/s) were analysed bycounting the number of action potentials within 1 s after stimulusonset. Mean responses and their standard errors were calculated.

3. Results

3.1. Spike waveforms generated by neurons of the

antennal chaetoid taste sensilla

Spike waveform analysis of electrophysiological recordingsobtained in chemical and mechanical stimulation experimentsshowed that at least four chemoreceptive and one mechanor-eceptive neurons innervate each antennal chaetoid taste sensillumin P. oblongopunctatus. In response to 10 mmol l�1 choline chloride,

usually only the salt-sensitive neuron fired, producing spikes withnegative polarity, 1–2 mV in amplitude. However, in somerecordings, a few spikes from other sensory neurons were alsoobserved. The pH-sensitive neuron generated very large actionpotentials, 2–4 mV peak-to-peak, with initial positive polarity(Fig. 1A). The shape of these impulses is complex because they arecomposed of two separate electrical events in quick succession andmost probably produced by two separate spike generation sites ofthe same neuron. Minor differences in the time interval betweenthese two events occuring among antennal taste sensilla maycause considerable differences in the shape of the resulting double-impulse. Usually, the double-impulses of the pH-sensitive neurondescribed are two-tipped (Fig. 2A and B). In some sensilla,however, only a small characteristic notch or discontinuity inthe rising phase of these impulses indicates their complex nature(Fig. 2C and D). Typically the mixtures of 10 mmol l�1 cholinechloride and some sugars evoked a sugar-sensitive neuron, inaddition to the salt-sensitive neuron, to discharge. The sugar-sensitive neuron generated negative spikes smaller than those ofthe salt-sensitive neuron. In a few recordings, irrespective of thestimulating solution, rare, very small action potentials withnegative polarity from the 4th chemosensory neuron from thesesensilla were also observed (Fig. 1B). So far, chemical stimulieliciting a response specific to the 4th chemosensory neuron havenot been identified. Impulse activity of the mechanosensoryneuron belonging to these taste sensilla and responding to bendingof the bristlewas recorded from some intact bristles aswell as frombristles with a broken tip. Action potentials produced by the

Table 1Responses of neurons of antennal chaetoid taste sensilla of P. oblongopunctatus for 1 s to stimulation with various carbohydrates and amino acids presented in 10 mmol l�1

choline chloride

Stimulating solution Concentration

(mmol l�1)

Mean impulses/s

(�S.E.)

Neuron responding Beetles/bristles Monosaccharide units

and glycosidic bonds

Choline chloride (control) 10 0.15 � 0.04 SNa and/or 4thN 26/130

D(+)-Pentoses

Arabinose 100 0.23 � 0.83 SN and/or 4thN 6/30

Xylose 100 0.62 � 0.15** SN and/or 4thN 7/34

D(+)-Hexoses

Glucose 100 20.87 � 1.41** SC 15/72 a-glu, 36%; b-glu, 64%Galactose 100 0.16 � 0.06 SN and/or 4thN 6/30

Mannose 100 0.46 � 0.19* SN and/or 4thN 9/45

D(�)-Hexoses

Fructose 100 0.00* 6/30

L-Hexoses

Rhamnose 100 0.26 � 0.18 SN and/or 4thN 6/30

D(+)-Disaccharides

Cellobiose 100 0.18 � 0.11* SN and/or 4thN 9/43 b-glu, glu; 1! 4

Lactose 100 0.51 � 0.15* SC and/or 4thC 8/40 b-gal, glu; 1! 4

Maltose 100 49.9 � 2.65** SN 10/48 a-glu, glu; 1! 4

Sucrose 100 46.5 � 2.01** SN 23/111 a-glu, b-fru; 1! 2

D(+)-Trisaccharides

Raffinose 100 0.13 � 0.13 SN and/or 4thN 5/25 a-gal, a-glu, b-fru; 1! 6, 1! 2

L-Amino acids

Methionine 100 2.32 � 0.42** SN and/or 4thN 12/60

Proline 100 0.68 � 0.23* SN and/or 4thN 7/35

Serine 100 2.76 � 0.51** SN and/or 4thN 6/30

Alanine 100 2.78 � 0.46** SN and/or 4thN 9/42

Asparagine 100 0.91 � 0.2** SN and/or 4thN 10/50

Glutamine 2.49 � 0.44** SN and/or 4thN 12/60

g-Aminobutyric acid 100 0.35 � 0.09 SN and/or 4thN 10/47

The number of spikes from the salt-sensitive neuron is not indicated.a At very low rate of firing firm classification of the spiking neuron was not possible.* Responses significantly different from choline chloride alone at P < 0.01.** Responses significantly different from choline chloride alone at P < 0.001 (t-test for independent samples).


92

mechanosensory neuron were small but variable in size, andnegative in polarity (Fig. 1C and D).

3.2. Responses of neurons from antennal taste sensilla to 10 mmol l�1

choline chloride alone

Choline chloride at a concentration of 10 mmol l�1 served as thecontrol stimulus, and as the electrolyte in stimulating mixtureswith sugars and amino acids to ensure a good signal-to-noise ratio.Typically, choline chloride evoked only the salt-sensitive neuronto discharge with a relatively low rate of firing (approximately15 impulses/s) during the first second after stimulus onset(Fig. 3A). Only in rare cases were a few action potentials fromother sensory neurons of the sensillum observed (Fig. 1A; Table 1).Therefore, the stimulating properties of choline chloride weresuitable for classifying and counting action potentials of the

responses evoked by the mixtures of this electrolyte with sugarsand amino acids.

3.3. Responses of antennal taste neurons to sugars

Responses of antennal taste neurons to stimulation by various100 mmol l�1 sugars in mixtures with 10 mmol l�1 cholinechloride, varied with the sugar tested. Only three out of twelvesugars, maltose, sucrose and glucose, evoked strong phasic-tonicresponses from the sugar-sensitive neuron (Table 1; Fig. 3B–E). Themean rates of firing of the neuron during the first second after100 mmol l�1 stimulation reached 49.9, 46.5 and 20.9 impulses/s,respectively. Dose/response curves of these three active sugars aredemonstrated in Fig. 4. In the range of 1–1000 mmol l�1, theresponse to glucose was approximately two times lower than thatof maltose and sucrose. In contrast, no differences were foundbetween responses to maltose and sucrose. The remaining nine100 mmol l�1 sugars had only a small or no stimulatory effect onthe neuron. Action potentials generated by the sugar-sensitive andthe 4th chemosensory neuron were small, varied in size, and theiramplitudes appeared to overlap. Therefore, at a very low rate offiring, firm classification of the spiking neuron was not alwayspossible with these non-stimulatory or slightly stimulatory sugars.Themean response they evokedwasmuch lower than 1 impulses/s

Fig. 1. Action potential waveforms recorded from the tip of antennal chaetoid

taste sensilla in the ground beetle P. oblongopunctatus. CN, SN, pHN, 4thN, andMN

represent action potentials produced by the salt-, sugar-, pH-, 4th chemo- and

mechano-sensitive neuron, respectively. Downward deflection from the baseline

represents negative polarity. (A) Response of a taste sensillum to 10 mmol l�1

choline chloride (control) 0.2–0.4 s after the beginning of stimulation. Arrowhead

shows a characteristic notch in the rising phase of the spike from the pH-sensitive

neuron indicating the presence of two spike generation sites in the neuron

membrane. (B) A recording of an atypical response from a taste sensillum to the

mixture of 10 mmol l�1 choline chloride and 100 mmol l�1 sucrose 0.5 s after the

beginning of stimulation demonstrating negative action potentials with different

amplitudes generated by three different chemosensory neurons. (C) Response of

the taste bristle to bending. Arrows show 1 mV calibration pulses from a

calibration source. Horizontal doubled black line indicates period of the rapid

movement of the sensillum tip towards antennal shaft and holding it in this

position (single black line) approximately 1.4 s which evoked strong firing of the

MN. The movement of the sensillum back to its initial resting position (dotted

line) caused only a weak response. (D) The same recording as (C) but in a shorter

timescale.

Fig. 2.Variation of the complex action potential shape produced by the pH-sensitive

neuron from different antennal taste sensilla. A, B: Two-tipped impulses with

equal and unequal positive peaks, respectively. In other cases, a notch (C) or a

discontinuity (D, arrowhead) in the rising phase of these impulses indicates their

complex nature. These double-impulses are caused by two separate action potential

generation sites of the same neuron. Note the shortening of time interval betveen

the two electrical events from A to D. Lower traces show upper traces in a shorter

timescale.


93

(Table 1; Fig. 3F). To test for possible seasonal variation in thefunctioning of the sugar-sensitive neuron, its responses to100 mmol l�1 sucrose were compared in active and hibernatingbeetles in July (46.5 impulses/s) and February (48.8 impulses/s),respectively but no differences between their responses wereobserved (t = �0.57; d.f. = 18; P = 0.57; t-test for independentsamples).

3.4. Responses of antennal taste neurons to amino acids

Responses of antennal taste neurons to the seven 10 and100 mmol l�1

L-amino acids in a mixture with 10 mmol l�1 cholinechloride varied with the type and concentration of amino acidtested. At 10 mmol l�1, none of the amino acids stimulated theneurons. At 100 mmol l�1, these chemicals evoked only a fewaction potentials (Table 1) with amplitudes smaller than thosegenerated by the salt-sensitive neuron (Fig. 3G–I). No initial phasiccomponent in the sequence of these spikes was observed.Frequently, they appeared with some delay in the responses,sometimes 5–10 s after stimulus onset (Fig. 3J). Spike amplitudeanalysis showed that these small impulses were generated by the4th chemosensory neuron rather than by the sugar-sensitive

Fig. 3. Example recordings from antennal taste sensilla to some carbohydrates and amino acids in P. oblongopunctatus. CN and SN represent action potentials from the salt-

and sugar-sensitive neuron, respectively. Arrowheads show small action potentials most probably generated by the 4th chemosensory neuron.

Fig. 4. Concentration/response curves of the sugar-sensitive neuron from antennal

taste sensilla obtained in experimentswith three physiologically active plant sugars

in P. oblongopunctatus. Action potentials were counted for 1 s after the onset of the

stimulus. Vertical bars show S.E. of themeans; means denotedwith different letters

grouped by the horizontal bar are significantly different at P < 0.05 (ANOVA, Tukey

test, 6 � N � 23).


94

neuron. Though amplitudes of spikes induced by sugars (sucrose)and amino acids varied and overlapped to a large extent theirhistograms had different maxima (Fig. 5). Since the mean numbersof these small spikes during the first second after stimulusonset did not exceed 3 impulses/s, this neuron has probably nospecialized receptor sites for amino acids, at least for those whichwere tested.

4. Discussion

Three out of the twelve sugars tested, maltose, sucrose andglucose, evoked strong responses from the sugar-sensitive neuronswhile other had little or no effect. Most stimulating were the twodisaccharides, maltose and sucrose, and this is in agreement withresults for a number of other insect species (Mitchell and Gregory,1979; Schoonhoven and van Loon, 2002; Albert, 2003; Meriveeet al., 2007; Sandoval and Albert, 2007). Sucrose is a disaccharideconsisting of two monosaccharides, a-glucose and b-fructose,joined by a glycosidic bond between carbon atom 1 of the glucoseunit and carbon atom2 of the fructose unit. Maltose is composed oftwo molecules of glucose joined together through a-1.4-glycosidelinkage. Thus, the presence of a terminal a-glucose unit seems tobe a common and important feature of the molecular structuredetermining the stimulatory properties of these disaccharides in P.

oblongopunctatus. Other monosaccharide units in the molecularstructure of sucrose and maltose were of no importance in thisrespect as the sugar-sensitive neuron responses to these twodisaccharides were similar. In contrast, the trisaccharide raffinosehas a non-terminal a-glucose unit and does not evoke a responsefrom the antennal sugar-sensitive neuron in this species. In the redturnip beetle, Entomoscelis americana, maltose (a-glucose-1.4-glucose) was at least an order of magnitude more effective inactivating the maxillary sugar-sensitive neuron than trehalose (a-glucose-1.1-glucose) illustrating the importance of the 1.4 linkagein these disaccharide molecules (Mitchell and Gregory, 1979).Compared to maltose and sucrose, the third active sugar glucosewas approximately two times less effective at stimulating theantennal sugar-sensitive neuron of P. oblongopunctatus. This can bepartly explained by the fact that glucose exists in at least twoanomeric forms, a and b. These two forms interconvert over atimescale of hours in aqueous solution, to a final stable ratio ofa:b

36:64, in a process called mutarotation. So the physiologicallyactive a-anomere forms only 36% of the glucose solution which isreflected in its relatively low dose/response curve. The order ofstimulatory effectiveness of sugars may vary in different insectspecies or even be different for receptors on various parts of thebody (Blaney, 1974; Ruth, 1976; Bernays and Chapman, 2001;Schoonhoven and van Loon, 2002).

Since amino acids, in addition to sugars, stimulate feedingbehaviour in various phytophagous insects these animals haveamino acid-specific neurons in their taste sensilla. Amino acids andsugars may stimulate the same neuron. Such a versatile receptorneuron is present in the polyphagous caterpillar of Grammia

geneura. These insects have, in addition to an amino acid sensitiveneuron in their lateral sensillum styloconicum, a neuron in anothersensillum which responds to seven (out of twenty) amino acids.This latter neuron can also be stimulated by sucrose, glucose, andtrehalose (Schoonhoven and van Loon, 2002; Chapman, 2003).Because of its multiple specificity this neuron was named a‘‘phagostimulatory cell’’, rather than a ‘‘sugar’’ or ‘‘amino acid’’ cell(Bernays and Chapman, 2001). However, in P. oblongopunctatus,the seven amino acids tested had only little or no stimulatory effecton the neurons of the antennal taste sensilla suggesting that theseneurons have no ability to perceive these amino acids. The weaklystimulatory effect (below 3 impulses/s) of some 100 mmol l�1

amino acids on the 4th chemosensory neuron of these sensilla canbe characterized as non-specific, or modulating to the responses ofnon-target chemosensory neurons, the phenomenon well knownin insect chemoreceptor functioning (Hanson, 1987; Liscia et al.,1997; Liscia and Solari, 2000; Bernays and Chapman, 2001;Chapman, 2003; Schoonhoven and van Loon, 2002; Merivee et al.,2005; Merivee et al., 2007).

No insect so far investigated responds to all sugars, and nosingle sugar is a major phagostimulant for all insects. For a giveninsect species, only some sugars elicit a response to a greater orlesser extent while other sugars elicit no response. Plant sugarswhich were found to evoke responses from the antennal contactchemoreceptors in P. oblongopunctatus, i.e., maltose, sucrose andglucose, function as strong phagostimulants for most phytopha-gous insects, flies, cockroaches and others, equipped withspecialized receptors to detect sugars (Dethier et al., 1956; Blaney,1974; Ruth, 1976; Amakawa, 2001; Liscia et al., 2002; Schoon-hoven and van Loon, 2002; Chapman, 2003; Sandoval and Albert,2007). This suggests that these sugars may have the same functionin P. oblongopunctatus, probably due to its partial herbivory.Although, in contrast to sucrose and glucose,maltose is uncommonin nature, it can be found in considerable amounts in germinatingseeds as a product of the enzymatic degradation of starch.

Several carbohydrates, in addition to glucose, such as cello-biose, arabinose, xylose, mannose, rhamnose and galactose areknown as components of cellulose and hemicelluloses. They arereleased by brown-rot fungi during enzymatic wood decay. Allwater-soluble carbohydrates are ultimately consumed, so thatextensively brown-rotted wood consists almost entirely ofmodified lignin. Brown-rot fungi appear to depolymerize woodin the early stages of decay much more rapidly than the decayproducts can be metabolized. The occurrence of excess wood-decomposition productsmay help explain the frequent presence ofother wood scavengers in brown-rotted wood (Eriksson et al.,1990; Zabel and Morrell, 1992; Valaskova and Baldrian, 2006). Aswood decomposes, it passes through a range of decay stages, eachcolonized by a succession of diverse saproxylic insect assemblages(Grove, 2002). Wood-decay products may serve as chemical cuesfor these saproxylic insects and be involved in this colonizationprocess although no electrophysiological or behavioural evidenceis yet available. Some ground beetles, including P. oblongopuncta-

Fig. 5. Histograms of relative spike amplitudes from neurons stimulated by sugars

and amino acids in antennal taste sensilla of P. oblongopunctatus. aSN, a4thN and

aCN represent spike amplitudes of the sugar-, 4th chemo-, and salt-sensitive

neuron, respectively. Normal distribution curves of respective samples are

indicated.


95

tus, prefer brown-rotted wood for their prolonged hibernation(Haberman, 1968; Lindroth, 1986). Our results showed, however,that P. oblongopunctatuswas not able to sense water-soluble woodsugars released by brown-rot fungi, except for glucose, indicatingthat these chemicals are not involved in the search for hibernationsites by this species, although others may be.

Acknowledgements

This work was supported by contract grant sponsor EstonianScience Foundation (contract grant numbers: 6958 and 7391).

References

Albert, P.J., 2003. Electrophysiological responses to sucrose from a gustatorysensillum on the larval maxillary palp of the spruce budworm, Choristoneurafumiferana (Clem.) (Lepidoptera: Tortricidae). J. Insect Physiol. 49, 733–738.

Amakawa, T., 2001. Effects of age and blood sugar levels on the proboscis extensionof the blow fly Phormia regina. J. Insect Physiol. 47, 195–203.

Bernays, E.A., Chapman, R.F., 2001. Gustatory neuron responses in the polyphagousarctiid, Grammia geneura: towards a general pattern for caterpillars. J. InsectPhysiol. 47, 1029–1043.

Blaney, W.M., 1974. Electrophysiological responses of the terminal sensilla on themaxillary palps of Locusta migratoria (L.) to some electrolytes and non-electro-lytes. J. Exp. Biol. 60, 275–293.

Chapman, R.F., 2003. Contact chemoreception in feeding by phytophagous insects.Ann. Rev. Entomol. 48, 455–484.

Daly, P.J., Ryan, M.F., 1979. Ultrastructure of antennal sensilla of Nebria brevicollis(Fab) (Coleoptera: Carabidae). Int. J. Insect Morphol. Embryol. 8, 169–181.

Dethier, V.G., Evans, D.R., Rhoades, M.V., 1956. Some factors controlling the inges-tion of carbohydrates by the blowfly. Biol. Bull. 111, 204–222.

Eriksson, K.-E.L., Blanchette, R.A., Ander, P., 1990. Microbial and Enzymatic Degra-dation of Wood Components. Springer-Verlag, New York.

Grove, S.J., 2002. Saproxylic insect ecology and the sustainable management offorests. Ann. Rev. Ecol. Syst. 33, 1–23.

Haberman, H., 1968. Ground Beetles of Estonia. Valgus, Tallinn, p. 598 [in Estonian]Hanson, F.E., 1987. In: Chapman, R.F., Bernays, E.A., Stoffolano, J.G. (Eds.), Chem-

oreception in the Fly: The Search for the Liverwurst Receptor. Perspectives inChemoreception and Behaviour. Springer, Germany, pp. 99–122.

Heikkila, R., 1977. Elaimet kylvetynmannyn ja kuusen siemenen tuhoojina Pohjois-Suomessa [Summary: Destruction caused by animals to sown pine and spruceseed in northern Finland]. Metsantutkimuslaitoksen Julkaisuja 89, 1–35.

Hengeveld, R., 1980. Polyphagy, oligophagy and food specialization in groundbeetles (Coleoptera Carabidae). Netherlands J. Zool. 30, 564–584.

Hoback, W.W., Golick, D.A., Svatos, T.M., Spomer, S.M., Higley, L.G., 2000. Salinityand shade preferences result in ovipositional differences between sympatrictiger beetle species. Ecol. Entomol. 25, 180–187.

Irmler, U., 2001. Charakterisierung der Laufkafergemeinschaften schleswig-holstei-nischer Walder undMoglichkeiten ihrer okologischen Bewertung. AngewandteCarabidologie, Suppl 2. Laufkafer im Wald, pp. 21–32.

Johnson, U.E., Cameron, R.S., 1969. Phytophagous ground beetles. Ann. Entomol.Soc. Am. 62, 909–914.

Kim, J.L., Yamasaki, T., 1996. Sensilla of Carabus (Isiocarabus) fiduciarius saishutoicusCsiki (Coleoptera: Carabidae). Int. J. Insect Morphol. Embryol. 25, 153–172.

Krogerus, R., 1960. Okologische Studien uber nordische Moorarthropoden. SocietasScientiarum. Fennica Commentationes Biologicae 21 (3), 1–238.

Kromp, B., 1999. Carabid beetles in sustainable agriculture: a review on pest controlefficacy, cultivation impacts and enhancement. Agric. Ecosyst. Environ. 74,187–228.

Lindroth, C.H., 1986. Fauna Entomologica Scandinavica, Vol. 15: The Carabidae(Coleoptera) of Fennoscandia and Denmark, II. E.J. Brill, Leiden, pp. 230–497.With an appendix on the family Rhysodidae.

Liscia, A., Solari, P., 2000. Bitter taste recognition in the blowfly: electrophysiolo-gical and behavioral evidence. Physiol. Behav. 70, 61–65.

Liscia, A., Solari, P., Majone, R., Tomassini Barbarossa, I., Crnjar, R., 1997. Tastereceptor mechanisms in the blowfly: evidence of amiloride-sensitive andinsensitive receptor sites. Physiol. Behav. 62, 875–879.

Liscia, A., Crnjar, R., Masala, C., Sollai, G., Solari, P., 2002. Sugar reception in theblowfly: a possible Ca2+ involvement. J. Insect Physiol. 48, 693–699.

Lovei, G.L., Sunderland, K.D., 1996. Ecology and behavior of ground beetles (Coleop-tera: Carabidae). Ann. Rev. Entomol. 41, 231–256.

Merivee, E., Ploomi, A., Rahi, M., Luik, A., Sammelselg, V., 2000. Antennal sensilla ofthe ground beetle Bembidion lampros Hbst (Coleoptera Carabidae). Acta Zool.(Stockholm) 81, 339–350.

Merivee, E., Ploomi, A., Luik, A., Rahi, M., Sammelselg, V., 2001. Antennal sensilla ofthe ground beetle Platynus dorsalis (Pontoppidan, 1763) (Coleoptera Carabidae).Microsc. Res. Tech. 55, 339–349.

Merivee, E., Ploomi, A., Rahi, M., Bresciani, J., Ravn, H.P., Luik, A., Sammelselg, V.,2002. Antennal sensilla of the ground beetle Bembidion properans Steph.(Coleoptera Carabidae). Micron 33, 429–440.

Merivee, E., Renou, M., Mand, M., Luik, A., Heidemaa, M., Ploomi, A., 2004. Electro-physiological responses to salts from antennal chaetoid taste sensilla of theground beetle Pterostichus aethiops. J. Insect Physiol. 50, 1001–1013.

Merivee, E., Ploomi, A., Milius, M., Luik, A., Heidemaa, M., 2005. Electrophysiologicalidentification of antennal pH-receptors in the ground beetle Pterostichus oblon-gopunctatus (Coleoptera Carabidae). Physiol. Entomol. 30, 122–133.

Merivee, E., Must, A., Milius, M., Luik, A., 2007. Electrophysiological identification ofthe sugar cell in antennal taste sensilla of the predatory ground beetle Pter-ostichus aethiops. J. Insect Physiol. 53, 377–384.

Milius, M., Merivee, E., Williams, I., Luik, A., Mand, M., Must, A., 2006. A newmethodfor electrophysiological identification of antennal pH receptor cells in groundbeetles: the example of Pterostichus aethiops (Panzer, 1796) (Coleoptera, Car-abidae). J. Insect Physiol. 52, 960–967.

Mitchell, B.K., Gregory, P., 1979. Physiology of the maxillary sugar sensitive cell inthe red turnip beetle, Entomoscelis americana. J. Comp. Physiol. 132, 167–178.

Nystrand, O., Granstrom, A., 2000. Predation on Pinus sylvestris seeds and juvenileseedlings in Swedish boreal forests in relation to stand disturbance by logging. J.Appl. Ecol. 37, 449–463.

Paje, F., Mossakowski, D., 1984. pH-preferences and habitat selection in carabidbeetles. Oecologia (Berlin) 64, 41–46.

Ruth, E., 1976. Elektrophysiologie der Sensilla Chaetica auf den Antennen vonPeriplaneta americana. J. Comp. Physiol. 105, 55–64.

Sandoval, M., Albert, P.J., 2007. Chemoreception of sucrose and amino acids insecond and fourth instars of the spruce budworm Choristoneura fumiferana(Clem.) (Lepidoptera: Tortricidae). J. Insect Physiol. 53, 84–92.

Schoonhoven, L.M., van Loon, J.J.A., 2002. An inventory of taste in caterpillars: eachspecies its own key. Acta Zoologica Academiae Scientiarum Hungaricae 48,215–263.

Symondson, W.O.C., Sunderland, K.D., Greenstone, M.H., 2002. Can generalistpredators be effective biocontrol agents. Ann. Rev. Entomol. 47, 561–594.

Thiele, H.-U., 1977. Carabid Beetles in Their Environment. Zoophysiology andEcology 10. Springer, Berlin, p. 369.

Valaskova, V., Baldrian, P., 2006. Degradation of cellulose and hemicelluloses by thebrown rot fungus Piptoporus betulinus—production, of extracellular enzymesand characterization of the major cellulases. Microbiology 152, 3613–3622.

Zabel, R.A., Morrell, J.J., 1992. Wood Microbiology Decay and Its Prevention. Aca-demic Press, San Diego, pp. 195–224.


Komendant, M., Merivee, E., Must, A., Tooming, E., Williams, I.,

Luik, A.

ELECTROPHYSIOLOGICAL RESPONSE OF THE CHEMO-

RECEPTOR NEURONS IN THE ANTENNAL TASTE SEN-

SILLA TO PLANT ALKALOIDS AND GLUCOSIDES IN THE

GRANIVOROUS GROUND BEETLE PTEROSTICHUS OBLON-GOPUNCTATUS.

Journal of Chemical Ecology (in review)

ELECTROPHYSIOLOGICAL RESPONSE OF

THE CHEMORECEPTOR NEURONS IN THE

ANTENNAL TASTE SENSILLA TO PLANT

ALKALOIDS AND GLUCOSIDES

IN THE GRANIVOROUS GROUND BEETLE

Pterostichus oblongopunctatus

MARIT KOMENDANT, ENNO MERIVEE*, ANNE MUST, ENE

TOOMING, INGRID WILLIAMS, ANNE LUIK


1 Kreutzwaldi Street, 51014 Tartu, Estonia

Abstract Electrophysiological responses of chemoreceptor neurons from

the antennal chaetoid taste sensilla to several plant alkaloids and glucosides

were tested in the ground beetle Pterostichus oblongopunctatus. For the

fi rst time, a specifi c alkaloid-sensitive neuron phasic-tonically responding

to quinine and quinine hydrochloride was found in these insects, most

probably related to their granivorous feeding habit. Th e response to qui-

nine hydrochloride was concentration-dependent at 0.001 to 50 mmol

l-1 with the response threshold at 0.01 mmol l-1, and a maximum rate of

fi ring of 67 spikes/sec at 50 mmol l-1. Th e stimulatory eff ect of caff eine

was very weak. In addition, both quinine and quinine hydrochloride

strongly inhibited spike production by the salt- and pH-sensitive neurons

when presented in mixtures with 10 mmol l-1 NaCl. It was also observed

that several tested plant secondary compounds, salicin, sinigrin, caff eine

and nicotine, which had only little or no eff ect on the fi ring rate of the

alkaloid-sensitive neuron, would greatly reduce the responses of the salt-

and pH-sensitive neurons, suggesting that the antennal taste sensilla of P. oblongopunctatus may detect plant defensive compounds both through the

activation of a specifi c alkaloid-sensitive neuron and through peripheral

inhibition of other chemoreceptor neurons of the taste sensillum.

Key Words Granivory, Contact chemoreception, Plant secondary com-

pounds, Alkaloid-sensitive neuron, Peripheral inhibition, Feeding deterrent

101

INTRODUCTION

Ground beetles (Coleoptera, Carabidae) are the dominant ground-living

invertebrate predators in many ecosystems. Because of their potential role

in foodweb dynamics and pest control, the food and feeding habits of

ground beetles have been investigated in more detail than of any other

group of predatory arthropods. Data on food items are available for ap-

proximately 1100 species and subspecies of ground beetles of the world. A

total of 73% are carnivorous, 19% are omnivorous and 8% are exclusively

herbivorous (Th iele 1977; Larochelle 1990; Toft and Bilde 2002). Of these,

many species predate on seeds (Honek et al. 2003). Ground beetles use a

variety of external stimuli in order to locate food. Behavioural tests and

observations have indicated that visual, tactile, olfactory and gustatory

cues may be involved (Wheater 1989; Kielty et al. 1996; Toft and Bilde

2002). Scanning electron microscopic studies have demonstrated that

thousands of various olfactory and gustatory sensilla are located on the

ground beetles’ antennae (Daly and Ryan 1979; Merivee et al. 2000, 2001,

2002). Very little is known, however, about the chemical compounds to

which they respond.

Electrophysiological investigations of the chemoreception capabilities of

ground beetles started only recently with experiments on antennal chaetoid

contact chemoreceptors. Approximately 70 of these large sensilla (35-

200 μm in length) have been found on the antennal fl agellum of various

ground beetles including the ground beetle Pterostichus oblongopunctatus (Merivee et al. 2000, 2001, 2002, 2005). Typically, four chemoreceptive

and one mechanoreceptive neuron innervate these sensilla in various in-

sect groups (Morita and Shiraishi 1985; Chapman 1998; Jørgensen et al.

2006). A similar set of sensory neurons has been found for the antennal

chaetoid contact chemoreceptors of the ground beetle Nebria brevicollis (Daly and Ryan 1979) suggesting that it may be characteristic for these

sensilla in P. oblongopunctatus, too, though, histological data for this

species are not available. In the ground beetles Pterostichus aethiops and P.

oblongopunctatus, specifi c chemical stimuli have been identifi ed by elect-

rophysiology for three of the four chemosensory neurons. Two respond

to various electrolyte solutions. By their specifi c response characteristics

these neurons were classifi ed as salt- (cation-) and pH-sensitive neurons

(Merivee et al. 2004, 2005; Milius et al. 2006). Th e third chemoreceptor

neuron is sensitive to sucrose, maltose and glucose, important in plant

102

carbohydrate metabolism (Merivee et al. 2007, 2008), most probably

related to granivory of these two forest species (Heikkilä 1977; Nystrand

and Granström 2000). Th e action potentials (spikes) produced by the

salt-, pH-, and sugar-sensitive neurons can be distinguished by their wave-

form, polarity and amplitude. Th e pH-sensitive neuron generates large

two-tipped spikes with positive polarity. In contrast, the polarity of the

one-tipped spikes from the salt- and sugar-sensitive neurons is negative.

It has been also observed that the spikes generated by the sugar-sensitive

neuron are shorter in amplitude compared to those generated by the

salt-sensitive neuron (Merivee et al. 2004, 2005, 2007). Carbohydrates

function as strong phagostimulants to most herbivorous insects, equipped

with specialized receptors to detect these compounds (Chapman 1998;

Schoonhoven and van Loon 2002; Chapman 2003) suggesting that these

chemicals may have phagostimulatory role in herbivorous and omnivorous

ground beetles, too. Unfortunately, no data exist on the function of the

4th chemoreceptor neuron probably also located in these antennal sensilla

of ground beetles, including P. oblongopunctatus (Daly and Ryan 1979;

Merivee et al. 2008).

In addition to detecting phagostimulants, most animals, including her-

bivorous insects, possess chemoreceptor neurons responding to a diverse

range of feeding deterrent compounds. When stimulated, deterrent neurons

reduce or fully stop feeding activity (Dethier 1980; Schoonhoven and van

Loon 2002; Chapman 2003; Wang et al. 2004). Natural deterrent stimuli

recognized by chemoreceptor neurons of herbivorous insects constitute

the largest and most structurally diverse class of gustatory stimuli. It has

been shown that plant glucosides, for example, sinigrin and salicin, and

various alkaloids like nicotine (pyrrolidine group), strychnine and quinine

(quinoline group), caff eine (purine group) and several others stimulate the

deterrent neuron in many herbivorous insects. Th reshold concentrations

for compounds that stimulate a deterrent neuron vary from 10-7 to 10-2

mol l-1, but the majority of these are in the range of 10-4 to 10-2 mol l-1.

Th erefore, deterrent neurons fulfi l a central role in host-plant recognition

and have, since their discovery (Ishikawa 1966), attracted much inter-

est (Glendinning 2002; Ryan 2002; Schoonhoven and van Loon 2002;

Chapman 2003; Jørgensen et al. 2007).

Herbivory, including pre- and post-dispersal plant seed predation, is ex-

tremely common in virtually all ecosystems. Seeds with their very high

103

nutrient values per unit volume are extremely attractive to granivores as

food (Janzen 1971). To counterbalance the eff ects of herbivory, plants

have developed a broad spectrum of secondary metabolites involved in

plant defense, which are collectively known as antiherbivory compounds

and can be classifi ed into three main sub-groups: nitrogen compounds

(including alkaloids, cyanogenic glycosides and glucosinolates), terpenoids,

and phenolics. Th ese compounds can act as feeding deterrents or toxins

with various modes of action to herbivores, or reduce plant digestibility

(Janzen 1971; Jolivet 1998; Ryan 2002; Chapman 2003; Hulme and

Benkman 2004; Davis et al. 2008). Seeds are sources of some of the most

toxic natural products known to humans and the secondary chemicals in

seeds present formidable challenges to granivores. However, seed chemical

defenses can only be assessed with reference to specifi c target organisms

since a secondary chemical may not be equally toxic to all granivores.

Some seeds are so toxic that all seed-predators avoid them (Harborne 1993;

Hulme and Benkman 2004). Most toxins are usually in small concentra-

tions (<5%) in the seed and some alkaloids, for example, can be lethal

at concentrations as low as 0.1% (Hatzold et al. 1983; Harborne 1993).

Unfortunately, there is no evidence that plant secondary compounds cause

feeding deterrency in ground beetles.

However, due to selective seed predation by ground beetles (Toft and

Bilde, 2002; Tooley and Brust, 2002), the presence of a deterrent neuron

in their antennal contact chemoreceptors should be expected. We tested

the stimulatory eff ect of some plant alkaloids and glucosides on the che-

moreceptor neurons innervating the antennal chaetoid taste sensilla in

the omnivorous P. oblongopunctatus, known as an important predator

on conifer seeds (Heikkilä 1977; Nystrand and Granström 2000). Th ese

experiments are reported in this paper.

104

METHODS AND MATERIALS

Insects Adult test beetles of P. oblongopunctatus were collected from a local

population in southern Estonia in September and October 2008. Th ey

were kept in 30×20×10 cm plastic boxes fi lled with moistened moss and

pieces of brown-rotted wood in a refrigerator at 5-6 ºC. A week prior

to the experiments, the beetles were transferred to room temperature

(22 ºC), fed with moistened cat food (Friskies Vitality+, Nestlé Purina,

Hungary) and given clean water to drink every day. Electrophysiological

experiments were conducted from November 2009 to February 2010.

Electrophysiology Th e beetle to be tested was intact and restrained securely

in a conical tube made of thin sheet-aluminium of a size that allowed the

head and antennae to protrude from the narrower end. Th e wider rear

end of the tube was blocked with plasticine to prevent the beetle from

retreating out of the tube. Its antennae were fi xed horizontally to the

edge of an aluminium stand with special clamps and beeswax, so that the

horizontally-located large chaetoid taste sensilla were visible from above

under a light microscope, and easily accessible for micromanipulation from

the side. Care was taken not to contaminate the taste sensilla by contact

with instruments or beeswax. Each prepared test beetle was placed in a

Faraday cage (15x10x5 cm) for electrophysiology.

Spikes from chemoreceptor neurons of the sensilla were recorded by the

conventional single sensillum tip-recording technique (Hodgson et al.

1955). To achieve a good signal to noise ratio, the indiff erent tungsten

microelectrode was inserted into the base of the fl agellum free of muscular

tissue. It was connected to an AC amplifi er via a calibration source, and

earthed. Th e recording glass micropipette fi lled with stimulating solu-

tion was brought into contact with the tip of the sensillum by means of

a micromanipulator under visual control through a light microscope at

a magnifi cation of 300X. Approximately, two sec before each stimula-

tion, 2-3 drops of solution were squeezed out of the micropipette tip by

a syringe connected to the rear end of the micropipette by means of a

silicone tube, and adsorbed onto a piece of fi lter paper attached to another

micromanipulator, in order to avoid concentration changes in the capil-

lary tip due to evaporation. During each 5 sec stimulation period, spikes

picked up by the recording electrode were fi ltered with a bandwidth set at

100–2000 Hz, amplifi ed (input impedance 10 GΩ; amplifi cation factor

105

2000X), monitored on an oscilloscope screen, and relayed to a computer

via an analogue-to-digital input board DAS-1401 (Keithley, Taunton,

MA, USA) for data acquisition, storage, and analysis using TestPoint

software (Capital Equipment Corp., Billerica, MA, USA) at a sampling

rate of 10 kHz. Responses were recorded from neurons of two to three

randomly-selected taste sensilla on the fl agellomeres 1 to 9 of each test

beetle. Each sensillum was stimulated twice, with control (10 mmol l-1

NaCl) and test stimulus solution, respectively. In each subsequent test

beetle, the order of stimulation was reversed.

Chemicals Sodium chloride, sucrose, sinigrin, salicin and caff eine were

purchased from AppliChem (Germany). Quinine, quinine hydrochloride

dihydrate, nicotine and L-strychnine were obtained from Sigma-Aldrich

company. Except for quinine and strychnine, stimulating chemicals, at

concentration of 10 mmol l-1, and were diluted in 10 mmol l-1 sodium

chloride. Due to their poor solubility in water, 1 mmol l-1 quinine and

0.1 mmol l-1 strychnine in 10 mmol l-1 sodium chloride were used in

the experiments. Additionally, to assess the threshold concentration and

concentration/response curve for one of the most active compounds, qui-

nine hydrochloride dihydrate, a water soluble salt of quinine, was tested

at concentrations ranging from 0.001 to 50 mmol l-1. All solutions were

kept at 5°C for less than one week.

Feeding bioassays Two-choice feeding experiments were carried out in

110-mm diameter glass Petri dishes. To ensure high air humidity inside

the dishes, each was lined with a moistened Whatman fi lter paper circle

(Schleicher & Schuell, England). Two pine seeds (Pinus sibirica), treated

for 20 min with distilled water (control) and 0.01 to 50 mmol l-1 quinine

hydrochloride, respectively, were placed in the centre of each Petri dish,

50 mm from each other. Only seeds with seed coats removed were used

in the experiments. Test beetles, one in each Petri dish, were allowed to

choose between, and feed on the seeds for 24 h (20 °C, 14:10 light:dark).

After 24 h, the beetles were removed and the percentage of predation on

control seeds and seeds treated with alkaloid, was visually determined.

At each concentration of quinine hydrochloride, the feeding preference

of the beetles was tested in fi ve replications (10 beetles in each).

Data analysis Automated classifi cation and counting of action potentials

from taste sensilla was not an appropriate method for data analysis be-

106

cause two action potentials frequently coincided to produce an irregular

waveform. Instead, spikes from several chemosensory neurons innervating

antennal taste sensilla tested were visually distinguished by their polarity

and amplitude (Fig. 1), and counted, using TestPoint software. Firing rates

(spikes/sec) were analysed by counting the number of action potentials

within 5 sec after stimulus onset. Mean fi ring rates and their standard

errors were calculated. Paired sample Wilcoxon signed rank was used to

determine the signifi cance of diff erences between means. Spike amplitude

analysis was performed in order to explain which specifi c neuron was

stimulated by the chemical classes tested.

107

RESULTS

Spike shapes and amplitudes evoked by tested chemicals Spike waveform

analysis showed that spikes from four diff erent chemoreceptor neurons

innervating antennal taste sensilla of the ground beetle Pterostichus oblon-gopunctatus could be distinguished in the recordings obtained in response

to tested chemicals. Earlier, three of the neurons, the salt- (Merivee et al.

2004), pH- (Merivee et al. 2005; Milius et al. 2006) and sugar-sensitive

neurons (Merivee et al. 2007, 2008), were identifi ed in this species. Spikes

with negative polarity, 1–2 mV in amplitude, arose from the salt-sensi-

tive neuron. Th is neuron generated spikes in a phasic-tonic manner and

had an extremely short response latency lasting less than 10 msec (Fig.

1A–J). Two-tipped spikes with positive polarity, 2–4 mV in amplitude,

were generated by the pH-sensitive neuron (Merivee et al. 2005; Milius

et al. 2006). Compared to the spikes from the salt-sensitive neuron, these

positive spikes appeared with a much longer delay (Fig. 1A,F). Th e sugar-

sensitive neuron also produced spikes with negative polarity, but smaller

in amplitude than those of the salt-sensitive neuron (Fig. 1J). Some tested

alkaloids (quinine, caff eine) and an alkaloid salt (quinine hydrochloride)

induced a neuron of the taste sensilla to produce spikes with negative

polarity, but with a considerably longer period of latency compared to

the spikes of the salt- and sugar-sensitive neurons (Fig. 1B–E). Spike

amplitude histograms show two maxima, indicating that spikes evoked

by salt (NaCl), sugar (sucrose) and quinine hydrochloride were generated

by three diff erent neurons, the salt-, sugar- and alkaloid-sensitive neuron,

respectively (Fig. 2). Spike amplitudes of the chemoreceptor neurons from

diff erent taste sensilla varied to a some extent, from 0.7 to 2 mV (Fig.

1C,E). As a result, in most cases, spike amplitudes generated by the alka-

loid-sensitive neuron were shorter than those of the salt-sensitive neuron

(Fig. 1B,C). In some cases, however, opposite spike amplitude ratios of

the neurons were observed (Fig. 1D).

Eff ect of plant alkaloids and glucosides on the responses of the antennal che-

moreceptor neurons Only two of the tested chemicals (Table 1), 1 mmol l-1

quinine and 10 mmol l-1quinine hydrochloride, were highly stimulating for

the antennal alkaloid-sensitive neuron, evoking 8.9 and 22.8 spikes/sec,

respectively. Th e stimulatory eff ect of 10 mmol l-1 caff eine was very weak,

2.4 spikes/sec. In contrast, some inhibiting eff ect of strychnine and sinigrin

to the activity of the neuron was observed. Respective fi ring rates were

108

very low, however (Table 1). Other plant chemicals did not considerably

aff ect the spike production of the alkaloid-sensitive neuron. Phasic-tonic

response of the alkaloid-sensitive neuron to quinine hydrochloride was

concentration dependent over the range of 0.001 to 50 mmol l-1 with the

response threshold at 0.01 mmol l-1, and maximum rate of fi ring equal

to 67 spikes/sec at 50 mmol l-1 (Figs. 3 and 4). However, compared to

the spike trains of the salt-sensitive neuron (Fig. 1; Fig. 5A–F), the phasic

component of the response produced by the alkaloid-sensitive neuron was

less pronounced (Fig. 1B–E; Fig. 4; Fig. 5C–F).

In addition to stimulation of the alkaloid-sensitive neuron, both quinine

and quinine hydrochloride strongly inhibited spike generation by the salt-

and pH-sensitive neurons when presented in mixtures with 10 mmol l-1

NaCl (Table 1). Increasing the quinine hydrochloride concentration while

keeping the concentration of NaCl constant at 10 mmol l-1 caused a dose-

dependent decrease in spike generation by the salt-sensitive neuron (Figs.

3–5). In most tested sensilla, at concentrations above 10 mmol l-1 quinine

hydrochloride, only the alkaloid-sensitive neuron fi red while the fi ring

rate of the salt-sensitive neuron of the taste sensillum decreased close to

zero during the 5 sec adaptation period (Fig. 4). In response to 10 mmol

l-1 NaCl alone, the mean number of spikes produced by the pH-sensitive

neuron was relatively low compared to that of the salt-sensitive neuron,

and varied to some extent in diff erent sensilla (Table 1). At quinine hy-

drochloride concentrations above 0.01 mmol l-1, the mean fi ring rate of

the pH-sensitive neuron dropped to zero (Fig. 3). It was also observed that

nicotine and salicin, which did not aff ect the fi ring rate of the alkaloid-

sensitive neuron, greatly reduced the response of the pH-sensitive neuron

(Table 1). Sinigrin and caff eine suppressed neural activity of the salt- and

pH-sensitive neuron, respectively. Th is suggested that the antennal taste

sensilla of P. oblongopunctatus may detect plant defensive compounds not

only through the activation of specifi c alkaloid-sensitive neuron but also

through inhibition of the salt- and pH-sensitive neurons.

Eff ect of quinine hydrochloride on the feeding response of P. oblongopunctatus

Feeding choice experiments showed that one of the most stimulating

chemicals for the alkaloid-sensitive neuron, quinine hydrochloride, had a

concentration dependent deterrent eff ect on seed predation in the ground

beetle P. oblongopunctatus, with the threshold concentration at 1 mmol l-1.

Th us, the threshold concentration of this alkaloid to evoke a behavioural

109

response was a hundred times higher than that for an electrophysiological

response of the alkaloid-sensitive neuron. At higher concentrations, the

deterrent eff ect of quinine hydrochloride was very strong. It was observed

that pine seeds treated with 50 mmol l-1 quinine hydrochloride were never

predated by the beetles (Fig. 6).

110

DISCUSSION

Th e experimental data presented show that the “fourth” chemoreceptor

neuron in the antennal taste sensilla of the ground beetle Pterostichus oblongopunctatus may respond to some plant alkaloids such as quinine.

Th is fi nding is in agreement with literature data on the feeding habits of

this species. P. oblongopunctatus is one of the most common and wide-

spread ground beetles in the forests of northern Europe (Lindroth 1985;

Niemelä et al. 1994). In most stands of Pinus sylvestris, seed predation

results in 20% seed mortality, although occasionally it reaches 60%.

Post-dispersal pine seed predation in the Swedish boreal forests studied

was due mainly to the two ground beetle species P. oblongopunctatus and

Calathus micropterus. Both are able to consume conifer seeds. Th e larger

species, P. oblongopunctatus, is considered to be the more important pine

seed predator of the two (Heikkilä 1977; Nystrand and Granström 2000).

Like many herbivores, numerous insect seed-predators have biochemical

adaptations to deal with toxic plant secondary compounds, including vari-

ous detoxifi cation and sequestration mechanisms (Jolivet 1998; Hulme and

Benkman 2004; Schoonhoven et al. 2005; Després et al. 2007). However,

the diffi culties of dealing with more than a few diff erent kinds of defen-

sive compounds has favoured specialization in seed-predators and helps

to explain why a large proportion of them are specialist feeders (Hulme

and Benkman 2004). Th e omnivorous ground beetle P. oblongopunctatus predates on many small animals, most frequently on Coleoptera, Diptera,

Collembola, Acari etc. (Larochelle 1990), as well as on conifer seeds

(Heikkilä 1977; Nystrand and Granström 2000). Probably, the species

feeds on seeds of various plants but respective data are not available. Se-

lective seed predation has been observed in many other ground beetles,

however (Toft and Bilde, 2002; Tooley and Brust, 2002).

To discriminate between toxic and nontoxic seeds, P. oblongopunctatus

possesses a specifi c feeding deterrent neuron innervating its antennal chae-

toid taste sensilla. A strong stimulatory eff ect of quinine (quinoline group)

and quinine hydrochloride to this neuron was observed. Th e threshold

concentration for the latter at 0.01 mmol l-1 was close to the upper limit

of the threshold concentrations for the plant secondary compounds that

stimulate a deterrent neuron in other herbivorous insects (Schoonhoven

and van Loon 2002; Chapman 2003; Schoonhoven et al. 2005), probably

111

due to the fact that toxin concentration in seeds is usually much higher

than in other parts of the plant (Janzen 1971; Hulme and Benkman 2004;

Davis et al. 2008). Th e response of the neuron to quinine hydrochloride

was concentration dependent with maximum rates of fi ring reaching up

to 70 spikes/secat 50 mmol l-1. In two-choice behavioural experiments,

quinine hydrochloride had a strong concentration dependent deterrent

eff ect on feeding of P. oblongopunctatus beetles with the response thresh-

old at 1 mmol l-1. Th us, the threshold concentration of this alkaloid to

evoke a behavioural response was a hundred times higher than that for an

electrophysiological response of the alkaloid-sensitive neuron. In contrast

to quinine hydrochloride, the response of the alkaloid-sensitive neuron

to caff eine (purine group) was very weak, only 2.4 spikes/sec at 10 mmol

l-1 concentration. Th e eff ect of another quinoline alkaloid, strychnine,

and sinigrin, to the neuron was rather inhibitory. No response of the

neuron to tested pyrrolidine alkaloid, nicotine, and salicin (glucoside)

was observed suggesting that the antennal alkaloid-sensitive neuron of P. oblongopunctatus is not equally sensitive to all plant defensive chemicals.

Nevertheless, the list of toxic compounds capable of stimulating the an-

tennal alkaloid-sensitive neuron of P. oblongopunctatus is likely to grow

longer as more plant defence chemicals are tested.

Quinoline alkaloids are among the most potent insect feeding deterrents

occurring mainly in Rutaceous and Rubiaceous plants. Th ey can be pres-

ent in virtually every part of the plant or just a specifi c part like the bark,

rhizome, leaf or seed (Openshaw 1967; Michael 2003; Aniszewski 2007).

Th ese alkaloids represent a large and structurally diverse group of plant

defensive compounds. When contacted, some of them may be detected by

stimulation of a specifi c antennal alkaloid-sensitive neuron of the ground

beetles predating on seeds, similarly to quinoline alkaloid-sensitivity of

P. oblongopunctatus. Th e ability of granivores to perceive plant defensive

compounds allows them to avoid toxic seeds, and survive. High sensitiv-

ity of P. oblongopunctatus to plant sugars, commonly known as strong

phagostimulants for most phytophagous insects, seems to be also related

to its partially granivorous feeding habits. Th e disaccharide maltose was

one of the most stimulatory sugars for the sugar-sensitive neuron also

innervating antennal taste sensilla of P. oblongopunctatus (Merivee et al.

2008). Maltose is uncommon in nature, it can be found in considerable

amounts in germinating seeds as a product of the enzymatic degradation

of starch.

112

While stimulating the alkaloid-sensitive neuron, quinine hydrochloride

strongly reduced spike production by the salt- and pH-sensitive neurons

which also innervate these sensilla in Pterostichus spp. (Merivee et al.

2004, 2005; Milius et al. 2006). Increasing the quinine hydrochloride

concentration, while keeping the concentration of NaCl constant, caused

a dose-dependent decrease in spike generation by the salt-sensitive neu-

ron. At 10 mmol l-1 and higher concentrations of quinine hydrochloride,

predominant fi ring of the alkaloid-sensitive neuron was observed while

the fi ring rate of other neurons dropped almost to zero. It was also ob-

served that several tested plant secondary compounds, salicin, sinigrin,

caff eine and nicotine, which had only little or no eff ect on the fi ring rate

of the alkaloid-sensitive neuron, would greatly reduce the responses of

the salt- and pH-sensitive neurons of the antennal taste sensillum of P. oblongopunctatus. Th ese results suggest that the antennal taste sensilla in

P. oblongopunctatus may detect toxic and deterrent plant secondary com-

pounds not only through activation of the specifi c alkaloid-sensitive neuron

but also through inhibition of taste neurons activated by salts and stimulus

pH; this is similar to deterrent inhibition of taste neurons activated by

sugars and water in Drosophila (Meunier et al. 2003). Since foodplant

acceptability to herbivores depends on the balance between stimulation

and deterrency, similar interactions between stimulating chemicals at

peripheral taste neurons of herbivorous insects are of widespread occur-

rence. Plant toxins and deterrents may inhibit sugar neurons (Frazier 1986;

van Loon 1990; Messchendorp et al. 1996; Bernays and Chapman 2000;

Schoonhoven and van Loon 2002; Meunier et al. 2003; Schoonhoven

et al. 2005), water neurons (Meunier et al. 2003), and sugars and salts

may inhibit deterrent neurons (Simmonds and Blaney 1983; Shields and

Mitchell 1995; Glendinning et al. 2000; Schoonhoven and van Loon

2002; Schoonhoven et al. 2005). Deterrent compounds that on their

own do not stimulate any neuron within a sensillum may also decrease

the responsiveness of a neuron responding to a nutrient, as exemplifi ed

by sinigrin inhibiting the inositol neuron in Heliothis virescens and H. subfl exa (Bernays and Chapman 2000; Schoonhoven and van Loon 2002).

Survival of most terrestrial animals requires ability to detect environmental

NaCl, a nutrient essential for fl uid and electrolyte homeostasis and for a

multitude of other physiological processes (Lindemann 1996; Contreras

and Lundy 2000). Th e gustatory system allows insects to detect and ingest

salt, to discriminate between diff erent salts, and to avoid high salt concen-

113

trations. However, there are only few data that relate input from inorganic

salts to feeding behaviour in phytophagous insects. In some insects, salts

have a deterrent eff ect on feeding (Chapman 2002; Schoonhoven and

van Loon 2003). In contrast, larvae of Drosophila show dose-dependent

responses to NaCl: the appetitive responses to low concentration gradually

turn into aversion as concentration is increased up to 200 mmol l-1 and

higher (Miyakawa 1982; Liu et al. 2003; Niewalda et al. 2008). Some other

insects also respond positively to dilute salts. Cockroaches and housefl ies

prefer dilute sodium and potassium chlorides and refuse distilled water.

In the two-choice situation, many animals give a bimodal response to

sodium chloride. Acceptance at low concentrations switches to rejection

at high concentrations (Dethier 1977). In these cases, inorganic salt at

low concentration may act as a phagostimulant. Th erefore, NaCl at 10

mmol l-1 may also have a phagostimulatory eff ect in P. oblongopunctatus although no behavioural data is available. Th is assumption could explain

the inhibition of spike production of the salt-sensitive neuron by some

toxic compounds in P. oblongopunctatus as being similar to how plant

deterrents inhibit activity of other phagostimulatory neurons in other

phytophagous insects.

In conclusion we state that

• the occurrence of an alkaloid-sensitive neuron in the antennal taste

sensilla of the ground beetle P. oblongopunctatus was documented;

• the alkaloid-sensitive neuron found is most probably related to the

granivorous feeding habit of the species;

• some tested compounds (quinine and quinine hydrochloride) that

stimulate the alkaloid-sensitive neuron may inhibit activity of other

neurons (salt- and pH-sensitive neuron) of the sensillum;

• several tested plant secondary compounds, salicin, sinigrin, caff eine

and nicotine, which had only little or no eff ect on the fi ring rate of

the alkaloid-sensitive neuron, through peripheral inhibition, would

greatly reduce the responses of other chemoreceptor neurons of the

sensillum.

114

Acknowledgements Th e study was supported by Estonian target fi nancing

project SF0170057s09 and Estonian Science Foundation grant no. 6958.

115

REFERENCES

ANISZEWSKI, T. 2007. Alkaloids – secrets of life. Alkaloid chemistry,

biological signifi cance, applications and ecological role. P 334. Else-

vier, Amsterdam, Boston, Heidelberg, London, New York, Oxford,

Paris, San Diego, San Francisco, Singapore, Sydney, Tokyo.

BERNAYS, E. A. and CHAPMAN, R. F. 2000. A neurophysiological

study of sensitivity to a feeding deterrent in two sister species of

Heliothis with diff erent diet breadths. J. Insect Physiol. 46:905-912.

CHAPMAN, R. F. 1998. Th e Insects. Structure and Function. 4th edn.

P. 770. Cambridge University Press, Cambridge.

CHAPMAN, R. F., 2003. Contact chemoreception in feeding by phy-

tophagous insects. Annu. Rev. Entomol. 48:455-484.

CONTRERAS, R. J. and LUNDY, R. F. 2000. Gustatory neuron types

in the periphery: a functional perspective. Physiol. Behav. 69:41-52.

DALY, P. J. and RYAN, M.F. 1979. Ultrastructure of antennal sensilla of

Nebria brevicollis (Fab.) (Coleoptera: Carabidae). Int. J. Insect Morphol.

Embryol. 8:169-181.

DAVIS, A. S., SCHUTTE, B. J., IANNUZZI, J. and RENNER, K.

A. 2008. Chemical and physical defense of weed seeds in relation to

soil seedbank persistence. Weed Sci. 56: 676-684.

DESPRÉS, L., DAVID, J. P. and GALLET, C. 2007. Th e evolution-

ary ecology of insect resistance to plant chemicals. Trends Ecol. Evol. 22:298-307.

DETHIER, V. G. 1977. Th e taste of salt. Amer. Sci. 65, 744-751.

DETHIER, V. G. 1980. Evolution of receptor sensitivity to second-

ary plant substances with special reference to deterrents. Am. Nat. 115:45–66.

FRAZIER, J. L. 1986. Th e perception of plant allelochemicals that in-

hibit feeding. Pp. 1-42 in L. B. BRATTSTEN and S. AHMAD

(eds.). Molecular aspects of insect-plant associations. Plenum Press,

New York.

GLENDINNING, J. I. 2002. How do herbivorous insects cope with

noxious secondary plant compounds in their diet? Entomol. Exp. Appl. 104:15-25.

GLENDINNING, J. I., NELSON, N. and BERNAYS, E. A. 2000.

How do inositol and glucose modulate feeding in Manduca sexta

caterpillars? J. Exp. Biol. 203: 1299-1315.

116

HARBORNE, J. B. 1993. Introduction to Ecological Biochemistry. 4th

edn. P. 318. Academic Press, London, San Diego.

HATZOLD, T., ELMADFA, I., GROSS, R., WINK, M., HART

MANN, T. and WITTE, L. 1983. Quinolizidine alkaloids in seeds

of Lupinus mutabilis. J. Agric. Food Chem. 31:934-930.

HEIKKILÄ, R. 1977. Eläimet kylvetyn männyn ja kuusen siemenen

tuhoojina Pohjois-Suomessa (Summary: Destruction caused by ani-

mals to sown pine and spruce seed in northern Finland). Comm. Inst. Forest. Fenniae 89:1-35.

HODGSON, E. S., LETTVIN, J. Y. and ROEDER, K. D. 1955. Physi-

ology of a primary chemoreceptor unit. Science 122:417-418.

HONEK, A., MARTINKOVA, Z. and JAROSIK, V. 2003. Ground

beetles (Carabidae) as seed predators. Eur. J. Entomol. 100:531-544.

HULME, P. E. and BENKMAN C. W. 2004. Granivory. Pp. 132-

154 in C. M. HERRERA and O. PELLMYR (eds.). Plant-animal

interactions. An evolutionary approach. Blackwell Science, Oxford.

ISHIKAWA, S. 1966. Electrical response and function of bitter substance

receptor associated with the maxillary sensilla of the larva of the

silkworm, Bombyx mori L. J. Cell Comp. Physiol. 67:1-12.

JANZEN, D. H. 1971. Seed predation by animals. Annu. Rev. Ecol.

Syst. 2:465-492.

JOLIVET, P. 1998. Interrelationships between insects and plants. P. 309.

CRC Press, Boca Raton, Florida.

JØRGENSEN, K., ALMAAS, T. J., MARIONPOLL, F. and MU

STAPARTA, H. 2007. Electrophysiological characterization of re-

sponses from gustatory receptor neurons of sensilla chaetica in the

moth Heliothis virescens. Chem. Senses 32:863-879.

JØRGENSEN, K., KVELLO, P., ALMAAS, T. J. and MUSTAPARTA,

H. 2006. Two closely located areas in the suboesophageal ganglion

and the tritocerebrum receive projections of gustatory receptor neu-

rones located on the antennae and the proboscis in the moth Heliothis

virescens. J. Comp. Neurol. 496:121-134.

KIELTY, J. P., ALLENWILLIAMS, L. J., UNDERWOOD, N. and

EASTWOOD, E. A. 1996. Behavioural responses of three species

of ground beetle (Coleoptera: Carabidae) to olfactory cues associated

with prey and habitat. J. Insect Physiol. 9:237-250.

LAROCHELLE, A. 1990. Th e food of the carabid beetles (Coleoptera:

Carabidae, including Cicindelinae). Fabreries, Supplément 5:1-132.

LINDEMANN, B. 1996. Taste reception. Physiol. Rev.76:719-766.

117

LINDROTH, C. H. 1985. Th e Carabidae (Coleoptera) of Fennoscandia

and Denmark. Fauna Entomol. Scand. 15:1-225.

LIU, L., LEONARD, A. S., MOTTO, D. G., FELLER, M. A., PRICE, M. P., JOHNSON, W. A. and WELSH, M. J. 2003. Contribution

of Drosophila DEG/ENaC Genes to Salt Taste. Neuron 39:133-146.

MERIVEE, E., MUST, A., MILIUS, M. and LUIK, A. 2007. Electro-

physiological identifi cation of the sugar cell in antennal taste sensilla

of the predatory ground beetle Pterostichus aethiops. J. Insect Physiol. 53:377-384.

MERIVEE, E., MÄRTMANN, H., MUST, A., MILIUS, M., WIL

LIAMS, I. and MÄND, M. 2008. Electrophysiological responses

from neurons of antennal taste sensilla in the polyphagous predatory

ground beetle Pterostichus oblongopunctatus (Fabritius 1787) to plant

sugars and amino acids. J. Insect Physiol. 54:1213-1219.

MERIVEE, E., PLOOMI, A., LUIK, A., RAHI, M. and SAM

MELSELG, V. 2001. Antennal sensilla of the ground beetle Platynus dorsalis (Pontoppidan, 1763) (Coleoptera, Carabidae). Microsc. Res.

Tech. 55:339-349.

MERIVEE, E., PLOOMI, A., MILIUS, M., LUIK, A. and HEIDE

MAA, M. 2005. Electrophysiological identifi cation of antennal pH-

receptors in the ground beetle Pterostichus oblongopunctatus (Coleop-

tera, Carabidae). Physiol. Entomol. 30: 122-133.

MERIVEE, E., PLOOMI, A., RAHI, M., BRESCIANI, J., RAVN, H.

P., LUIK, A. and SAMMELSELG, V. 2002. Antennal sensilla of the

ground beetle Bembidion properans Steph. (Coleoptera, Carabidae).

Micron 33:429-440.

MERIVEE, E., PLOOMI, A., RAHI, M., LUIK, A. and SAMMEL

SELG, V. 2000. Antennal sensilla of the ground beetle Bembidion lam-

pros Hbst (Coleoptera, Carabidae). Acta Zool. (Stockholm) 81:339-350.

MERIVEE, E., RENOU, M., MÄND, M., LUIK, A., HEIDEMAA,

M. and PLOOMI A., 2004. Electrophysiological responses to salts

from antennal chaetoid taste sensilla of the ground beetle Pterostichus aethiops. J. Insect Physiol. 50:1001-1013.

MESSCHENDORP, L., VAN LOON, J. J. A. and GOLS, G. J. Z.,

1996. Behavioural and sensory responses to drimane antifeedants in

Pieris brassicae larvae. Entomol. Exp. Appl. 79:195-202.

MEUNIER, N., MARIONPOLL, F., ROSPARS, J. P. and TANI

MURA, T. 2003. Peripheral coding of bitter taste in Drosophila. J.

Neurobiol. 56:139-152.

118

MICHAEL, J. P. 2003. Quinoline, quinazoline and acridone alkaloids.

Nat. Prod. Rep., 20:476-493.

MILIUS, M., MERIVEE, E., WILLIAMS, I., LUIK, A., MÄND, M.

and MUST, A. 2006. A new method for electrophysiological identifi -

cation of antennal pH receptor cells in ground beetles: the example of

Pterostichus aethiops (Panzer, 1796) (Coleoptera, Carabidae). J. Insect

Physiol. 52:960-967.

MIYAKAWA, Y. 1982. Behavioural evidence for the existence of sugar,

salt and amino acid taste receptor cells and some of their properties

in Drosophila larvae. J. Insect Physiol. 28:405-410.

MORITA, H. and SHIRAISHI, A. 1985. Chemoreception physiol-

ogy. Pp. 133-170 in G. A. KERKUT and L. I. GILBERT (eds.).

Comprehensive Insect Physiology, Biochemistry and Pharmacology.

Pergamon Press, Oxford,.

NIEMELÄ, J., SPENCE, J. R., LANGOR, D., HAILA, Y. and TUKIA,

H. 1994. Logging and boreal ground-beetle assemblages on two con-

tinents: implications for conservation. Pp. 29-50 in K. GASTON, M.

SAMWAYS and T. NEW (eds.). Perspectives in insect conservation.

Intercept Publications, Andover.

NIEWALDA, T., SINGHAL, N., FIALA, A., SAUMWEBER, T.,

WEGENER S. and GERBER, B. 2008. Salt Processing in Larval

Drosophila: Choice, Feeding, and Learning Shift from Appetitive to

Aversive in a Concentration-Dependent Way. Chem. Senses 33:685-

692.

NYSTRAND, O. and GRANSTRÖM, A. 2000. Predation on Pinus sylvestris seeds and juvenile seedlings in Swedish boreal forests in

relation to stand disturbance by logging. J. Appl. Ecol. 37:449-463.

OPENSHAW, H. T. 1967. Quinoline alkaloids other than those of Cinc-

hona. Pp. 223-267 in R. H. F. MANSKE (ed.). Th e Alkaloids. Che-

mistry and Physiology. Vol. IX. Academic Press, New York, London.

RYAN, M. F. 2002. Insect Chemoreception. Fundamental and Applied.

P. 330. Kluwer Academic Publishers, New York, Boston, Dordrecht,

London, Moscow.

SCHOONHOVEN, L. M. and VAN LOON, J. J. A. 2002. An inven-

tory of taste in caterpillars: each species its own key. Acta Zool. Hung. 48:215-263.

SCHOONHOVEN, L. M., VAN LOON, J. J. A. and DICKE, M.,

2005. Insect-Plant Biology. 2nd edn. P. 421. Oxford University Press,

Oxford.

119

SHIELDS, V. D. C. and MITCHELL, B. K. 1995. Responses of max-

illary styloconic receptors to stimulation by sinigrin, sucrose and

inositol in two crucifer-feeding, polyphagous lepidopterous species.

Philos. Trans. R. Soc. London 347:447-457.

SIMMONDS, M. S. J. and BLANEY, W. M. 1983. Some neurophysi-

ological eff ects of azadirachtin on lepidopterous larvae and their feed-

ing responses. Pp. 163-180 in H. SCHMUTTERER and K. R.

S. ASCHER (eds.). Proceedings of the Second International Neem

Conference, G.T.Z., Eschborn.

THIELE, H. U. 1977. Carabid Beetles in Th eir Environment. Zoo-

physiol. Ecol. 10. P 369. Springer, Berlin.

TOFT, S. and BILDE, T. 2002. Carabid diets and food value. Pp. 81-

110 in J. M. Holland (ed.). Th e Agroecology of Carabid Beetles.

Intercept, Andover, UK.

TOOLEY, J. and BRUST, G. 2002. Weed seed predation by carabid

beetles. Pp. 215-229 in J. M. Holland (ed.). Th e Agroecology of Ca-

rabid Beetles. Intercept, Andover, UK.

VAN LOON, J. J. A. 1990. Chemoreception of phenolic acids and fl avo-

noids in larvae of two species of Pieris. J. Comp. Physiol., A 166:889-

899.

WANG, Z., SINGHVI, A., KONG, P. and SCOTT, K. 2004. Taste

representations in the Drosophila brain. Cell 117:981-991.

WHEATER, C. P. 1989. Prey detection by some predatory Coleoptera

(Carabidae and Staphylinidae). J. Zool., A 218:171-185.

120

Legends for fi gures

Figure 1. Spike waveforms and response types of the chemoreceptor

neurons from the antennal chaetoid taste sensilla in P. oblongopunctatus. CN, AN, pHN and SN represent spikes produced by the salt-, alkaloid-,

pH- and sugar-sensitive neuron, respectively. Downward defl ection from

the baseline represents negative polarity. Recordings A–C, D, and E were

obtained from three diff erent sensilla from the same animal, respectively.

F–J, recordings were made from the antennal taste sensilla of diff erent

beetles. B, C, E, F, I, note the long response latency of the AN; fi rst spike

in the spike train of the AN is shown by an arrow. Usually, spike ampli-

tude of the AN was smaller compared to that of CN (see B and C), but in

some sensilla they were larger than CN spikes (demonstrated in D). Note

the pronounced phasic component of the response from CN (see A–J)

and SN (see J), and response with no clear phasic component of the AN

(see B, C, E) in these 0.5 sec fragments of the original 5 sec recordings.

Figure 2. (A) Histogram of relative spike amplitudes from neurons stimu-

lated by sucrose and quinine hydrochloride in antennal taste sensillum

of P. oblongopunctatus. aAN, aCN and aSN represent spike amplitudes

of the alkaloid-, salt- and sugar-sensitive neuron, respectively. Normal

distribution curves of respective samples are indicated. Two successive

stimulations of the same sensillum with 10 mmol l-1 sucrose and 10 mmol

l-1 quinine hydrochloride in the mixture with 10 mmol l-1 NaCl, respec-

tively, were performed. Spikes amplitudes from the two 30 sec record-

ings were measured and analysed. (B,C) Sample 1 sec fragments of the

respective original recordings.

Figure 3. Spike frequencies of the salt- and alkaloid-sensitive neurons in

the antennal chaetoid taste sensilla of P. oblongopunctatus upon stimulation

with 10 mmol l-1 NaCl mixed with diff erent concentrations of quinine

hydrochloride (QHC). CN, AN, pHN, salt-, alkaloid and pH-sensitive

neuron, respectively. N, the number of tested beetles. Responses from

two to three taste sensilla were tested from each test beetle. Vertical bars

denote ± SE of the means.

Figure 4. Responses of the salt- and alkaloid-sensitive neurons from

antennal taste sensilla of P. oblongopunctatus during 5 sec stimulation

with 10 mmol l-1 NaCl mixed with diff erent concentrations of quinine

121

hydrochloride. CN and AN show spike numbers produced by the salt- and

alkaloid-sensitive neuron, respectively; N, the number of tested beetles.

Responses from two to three taste sensilla were tested from each test

beetle. Vertical bars denote ± SE of the means.

Figure 5. Sample responses of an antennal taste sensillum of P. oblon-gopunctatus to diff erent concentrations of quinine hydrochloride. Note

that spike production of the alkaloid-sensitive neuron increased and that

of the salt-sensitive neuron decreased with quinine hydrochloride (QHC)

concentration increase. CN and AN show spikes from the salt- and al-

kaloid-sensitive neurons, respectively.

Figure 6. Deterrent eff ect of quinine hydrochloride (QHC) to the feeding

response of P. oblongopunctatus in two-choice experiments. At each variant

of seed treatment the feeding preference of the beetles between treated

(QHC) and control (distilled water) seeds was tested in fi ve replications

(10 beetles in each).

Table 1. Eff ect of tested plant alkaloids and glucosides to the spike pro-

duction of the chemoreceptor neurons from antennal chaetoid taste sen-

silla of P. oblongopunctatus. All test chemicals were diluted in 10 mmol

l-1 sodium chloride. Due to low solubility of quinine and strychnine in

water these chemicals were tested at a lower concentration than others.

pHN, CN, AN indicate spikes from pH-, salt- and alkaloid-sensitive

neurons, respectively. Asterisks indicate signifi cantly diff erent means,

P<0.05 (paired t-test).

122

A

F

B

I

C

H

D

G

0 0.1 0.2 0.3 0.4 0.5

CN

CN

CN

CN

pHN

pHN

1 mV

CN

CN

CN

CN

CN

CN

AN

AN

AN

SN

quinine hydrochloride dihydrate 1 mmol -1l

caffeine 10 mmol -1l

NaCl 10 mmol -1l

NaCl 10 mmol -1l

NaCl 10 mmol -1l

NaCl 10 mmol -1l

NaCl 10 mmol -1l

NaCl 10 mmol -1l

NaCl 10 mmol -1l

NaCl 10 mmol -1l

NaCl 10 mmol -1l

NaCl 10 mmol -1l

nicotine 10 mmol -1l

strychnine 0.1 mmol -1l

salicin 10 mmol -1l

sinigrin 10 mmol -1l

sucrose 10 mmol -1l



AN

AN

AN

AN

AN

AN

Time (sec)

E

J

Figure 1.

123

0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1.0

Time (sec)

10 mmol quinine hydrochloride 10 mmol NaCl -1 -1l in l

10 mmol sucrose 10 mmol NaCl -1 -1l in l

salt-sensitive neuron

salt-sensitive neuron

alkaloid-sensitive neuron

sugar-sensitive neuron

B

A

C

0,45 0,50 0,55 0,60 0,65 0,70 0,75 0,80 0,85 0,90 0,95 1,000

5

10

15

20

25

30

35N

um

be

r o

f o

bse

rva

tio

ns

Ratio of spike amplitudes

aAN/aCN

aSN/aCN

Figure 2.

124

10

0

0 0.010.001

20

30

40

50

60

70

80

0.1 1 10 50-1

Concentration of QHC (mmol l )

First se

co

nd

re

sp

on

se

(sp

ike

s/s

ec)

ANpHN

CN

N=7

Figure 3.

125

0

0

0

0

10

10

10

10

20

20

20

20

30

30

30

30

40

40

40

40

50

50

50

50

60

60

60

60

70

70

70

70

Time (sec)1 2 3 4 5

Me

an

firin

g r

ate

(sp

ike

s/s

ec)

Me

an

firin

g r

ate

(sp

ike

s/s

ec)

Me

an

firin

g r

ate

(sp

ike

s/s

ec)

Me

an

firin

g r

ate

(sp

ike

s/s

ec)

AN

AN

AN

AN

CN

pHN

pHN

pHN

pHN

CN

CN

CN

A

B

C

D

N=7

N=23

N=23

N=26

QHC 1 mmol l ¹

QHC 10 mmol l ¹

QHC 50 mmol l ¹

QHC 0.1 mmol l ¹

Figure 4.

126

A

B

C

D

E

CN

CN

CN

CN

CN

AN

AN

AN

8

F

CNAN

AN

AN

-1 -1 0.01 mmol l QHC in 10 mmol l NaCl

-1 -1 1 mmol l QHC in 10 mmol l NaCl



-1 -1 0.1 mmol l QHC in 10 mmol l NaCl

-1 10 mmol l NaCl (control)

0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1.00

Time (sec)

Figure 5.

127

0

20

40

60

80

100

120

Fe

ed

ing

pre

fere

nce

(%

)

1-

QHC

1m

0

ml-1

QHC

50 m

ml

-1

HC1

m

Q

m

l-1

0

QHC

.1 m

ml

1-

HC0

Q

.0

1 mm

l

C

l

ontro

C

rl

ont o

Contrlo

Contrlo

C

l

ontro

*

*

*

n.s n.s

Figure 6.

128

Tab

le 1

. Com

pou

nd

Con

cen

trat

ion

(mm

ol l-1

)R

esp

on

din

gn

euro

n

Mea

n 5

s r

esp

on

se

to N

aCl 10 m

mol l-1

(s

pik

es/s

±SE

)

Mea

n 5

s r

esp

on

se

to t

est

solu

tion

(s

pik

es/s

±SE

)T

este

d b

eetl

es N

(bri

stle

s)P

uri

ne

alk

alo

ids

Caff

ein

e10

pH

N3.2

± 0

.88

1.7

± 0

.73*

11(3

4)

CN

12.4

± 0

.94

11.7

± 0

.85

AN

1.0

± 0

.20

2.4

± 0

.44**

Pyr

roli

din

e al

kal

oid

sN

icoti

ne

10

pH

N12.3

± 2

.97

0.4

± 0

.17*

25(5

0)

CN

14.2

± 0

.67

10.9

± 2

.5A

N 0

.1 ±

0.1

10.1

± 0

.11

Qu

ino

lin

e al

kal

oid

sQ

uin

ine

1p

HN

0.4

± 0

.21

0.0

± 0

.00*

9(2

7)

CN

11.7

± 0

.82

10.1

± 0

.81*

AN

1.1

± 0

.37

8.9

± 1

.19**

Stry

chn

ine

0.1

pH

N0.2

± 0

.08

1.6

± 0

.67*

8(2

4)

CN

11.1

± 0

.86

10.2

± 0

.84

AN

0.7

± 0

.21

0.3

± 0

.08*

Qu

inin

e h

ydro

chlo

rid

e10

pH

N8.2

± 2

.30.0

± 0

.00**

15(4

5)

(sal

t of

qu

inin

e)C

N16.2

± 1

.88

1.6

± 0

.60**

AN

1.0

± 0

.27

22.8

± 2

.91**

Glu

cosi

des

Sali

cin

10

pH

N1.2

± 0

.64

0.7

± 0

.10**

15(3

0)

CN

10.9

± 0

.63

12.1

± 0

.82

AN

0.8

± 0

.20

0.9

± 0

.24

Sin

igri

n10

pH

N0.6

± 0

.22

0.5

± 0

.19

16(4

8)

CN

11.8

± 0

.61

5.2

± 0

.35**

AN

1.7

± 0

.46

0.8

± 0

.22**

129

CURRICULUM VITAE

First name: Marit

Last name: Komendant

Date birth: May 10th 1980

Address: Institute of Agricultural and Environmental

Sciences, Estonian University of Life Sciences,

Kreutzwaldi 1, Tartu 51014, Estonia

E-mail: [email protected]

Institution and position held: 2006–... Selteret OÜ offi ce manager

Education:2006–2010 PhD studies in Entomology, Estonian University

of Life Sciences

2004–2006 MSc studies in Plant Protection, Estonian

University of Life Sciences

2000–2004 BSc studies in Horticulture, Estonian Agricultural

University

1998–2000 Räpina Higher Gardening School

1986–1998 Gymnasium of Forseliuse

Academic degree: Master’s Degree (2006), Th esis: Antennal pH

receptors in the ground beetles Pterostichus aethiops and P. oblongopunctatus (Coleoptera,

Carabidae), Estonian University of Life Sciences.

Research interests: Biosciences and Environment, Ecology,

Biosystematics and -physiology (Insect behaviour

and sensor physiology)

Languages spoken:Estonian, English, German, Russian

130

ELULOOKIRJELDUS

Eesnimi: Marit

Perekonnanimi: Milius

Sünniaeg: 10. mai 1980

Aadress: Põllumajandus- ja keskkonnainstituut, Eesti

Maaülikool, Kreutzwaldi 1, Tartu 51014

E-mail: [email protected]

Töökoht ja amet: 2006–... Selteret OÜ büroojuht

Haridus:2006–2010 doktoriõpe entomoloogia erialal, Eesti

Maaülikool

2004–2006 magistriõpe taimekaitse erialal, Eesti Maaülikool

2000–2004 bakalaureuseõpe aianduse erialal, Eesti

Põllumajandusülikool

1998–2000 Räpina Kõrgem Aianduskool

1986–1998 Forseliuse Gümnaasium

Teaduskraad: Teadusmagister 2006, väitekiri: “Antennaalsed

pH-retseptorid süsijooksikutel Pterostichus aethiops ja P. oblongopunctatus (Coleoptera,

Carabidae)”, Eesti Maaülikool

Teadustöö põhisuunad: Bio- ja keskkonnateadused, Ökoloogia,

biosüstemaatika ja -füsioloogia (Putukate

käitumine ja sensoorne füsioloogia)

131

LIST OF PUBLICATIONS

1.1. Publications indexed in the ISI Web of Science database:

Merivee, E., Märtmann, H., Must, A., Milius, M., Williams, I., Mänd,

M., 2008. Electrophysiological responses from neurons of antennal

taste sensilla in the polyphagous predatory ground beetle Pterostichus

oblongopunctatus (Fabricius 1787) to plant sugars and amino acids.


Merivee, E., Must, A., Milius, M., Luik, A., 2007. Electrophysiological

identifi cation of the sugar cell in antennal taste sensilla of the predatory

ground beetle Pterostichus aethiops. Journal of Insect Physiology 53,

377–384.

Milius, M., Merivee, E., Williams, I., Luik, A., Mänd, M., Must, A.,

2006. A new method for Electrophysiological identifi cation of antennal

pH receptor cells in ground beetles: the example of Pterostichus aethiops (Panzer, 1796) (Coleoptera, Carabidae). Journal of Insect Physiology 52,

960–967.


Electrophysiological identifi cation of antennal pH receptors in the

ground beetle Pterostichus oblongopunctatus. Physiological Entomology

30(2), 122–33.

1.2. Papers published in other peer-reviewed international journals with a registered code:

Merivee, E., Must, A., Milius, M., Luik, A., 2006. External stimuli

in searching for favourable habitat, overwintering sites and refugia of

ground beetles: a short review. Agronomy Research 4, 299–302.

3.5. Papers in Estonian and in other peer-reviewed research journals with a local editorial board:

Milius, M., Merivee, E., Mänd, M., Ploomi, A., 2005. Metsa-

süsijooksiku antennaalsete maitseharjaste reaktsioonid 10 ja 100 mM

fosfaatpuhvritele. EPMÜ Teadustööde kogumik 220, 228–230.

ISBN 978-9949-484-00-3

Trükitud taastoodetud paberile looduslike trükivärvidega © Ecoprint

VIIS VIIMAST KAITSMIST

THEA KULL

REPRODUCTION ECOLOGY AND GENETIC DIVERSITY OF

DECLINING SEDGE (CAREX) SPECIES

Prof. Tiiu Kull Dr. Tatjana Oja

November 23, 2011

MARGIT HEINLAAN

ECOTOXICOLOGICAL EVALUATION OF SYNTHETIC NANOPARTICLES AND

PARTICULATE ENVIRONMENTAL SAMPLES.

Prof. Henri-Charles Dubourguier

Prof. Kalev Sepp

Dr. Anne Kahru

December 14, 2011

REIN DRENKHAN

EPIDEMIOLOGICAL INVESTIGATION OF PINE FOLIAGE DISEASES

BY THE USE OF THE NEEDLE TRACE METHOD

Dr. Märt Hanso

January 13, 2011

MIHKEL KIVISTE

CONDITION AND RESIDUAL BEARING CAPACITY OF

EXISTING REINFORCED CONCRETE STRUCTURES

Prof. Jaan Miljan

January 21, 2011

TATJANA KUZNETSOVA

PLANTATIONS OF NATIVE AND INTRODUCED TREE SPECIES IN THE

RECLAMATION OF OIL SHALE POST-MINING AREAS

Dr. Malle Mandre

Dr. Henn Pärn January 25, 2011

Electrophysiological identification of antennal pH receptors in the ground beetle Pterostichus...

Documents

Transcript of Electrophysiological identification of antennal pH receptors in the ground beetle Pterostichus...