ava i lab le at www.sc iencedi rec t . com

journa l homepage : www.e l sev i er . com/ loca te /mycres

m y c o l o g i c a l r e s e a r c h 1 1 0 ( 2 0 0 6 ) 485 – 492

Impact of moisture on in vitro germinationof Metarhizium anisopliae and Beauveria bassianaand their activity on Triatoma infestans

Gustavo Maximiano Junqueira LAZZARINI, Luiz Fernando Nunes ROCHA,Christian LUZ*

DMIPP, Instituto de Patologia Tropical e Saude Publica, Universidade Federal de Goias, CP 131, 74001-970 Goiania (GO), Brasil

a r t i c l e i n f o

Article history:

Received 19 July 2005

Received in revised form

30 October 2005

Accepted 4 December 2005

Published online 20 March 2006

Corresponding Editor:

Richard A. Humber

Keywords:

Biocontrol

Chagas disease

Entomopathogens

Insect pathology

Triatomines

a b s t r a c t

The in vitro germination of 11 Metarhizium anisopliae and 11 Beauveria bassiana isolates orig-

inating from substrates collected in rural peridomestic areas in Central Brazil where triato-

mines are common was tested. Conidia completed germination up to 24 h after exposure to

water activity of >0.99 aw in all isolates tested. At lower 0.93 aw germination was delayed

but conidia of most isolates germinated at high rates (>98 %) within 216 h of incubation.

Activities of 2 M. anisopliae and 2 B. bassiana isolates with different patterns of germination

at 0.93 aw were tested in Triatoma infestans third instar nymphs. There was no relationship

between germination kinetics in vitro at 0.93 aw and their activity in vivo at 98, 75 and 43 %

relative humidity (rh). Isolates with accelerated germination at 0.93 aw were not more vir-

ulent at 75 and 43 % rh compared with isolates with retarded or no germination. Highest

mortalities were observed at 98 % rh, and they did not exceed 25 % after 25 d incubation

at lower 75 and 43 % rh. Isolates that originated from a region with an extensive annual

arid period showed no adaptation to lower humidity in their activity against T. infestans.

ª 2006 The British Mycological Society. Published by Elsevier Ltd. All rights reserved.

Introduction

Entomopathogenic fungi are important candidates for biocon-

trol of triatomines (Lecuona et al. 2001; Luz et al. 1998a, b).

These insects transmit Trypanosoma cruzi, the agent of Chagas

disease in Latin America. Recent studies in Central Brazil

showed that Metarhizium anisopliae and Beauveria bassiana oc-

cur naturally in peridomestic habitats and may act as antago-

nists of these vectors (Luz et al. 2004a). Another

entomopathogenic fungus, Evlachovaea sp., was isolated

from a dead Triatoma sordida collected in the same region

(Luz et al. 2003). When tested under laboratory conditions, ac-

tivity of these fungi against T. infestans and other species was

highest at moisture close to saturation, and a reduction and

high variability of mortality was observed at lower relative hu-

midity (rh) of 75 % (Luz et al. 2004a, c). Results obtained re-

cently in field tests with a B. bassiana isolate and T. sordida

underline the high potential of this fungus to control perido-

mestic triatomine populations (Luz et al. 2004b).

Infective conidia adhere to the cuticle, germinate at favour-

able conditions, and the fungus invades the host. After host

death fungi produce new conidia. Extracuticular development

on triatomines depends on the level and exposure time to

favourable moisture as shown for B. bassiana Bb 297 and Rhod-

nius prolixus (Fargues & Luz 1998, 2000; Luz & Fargues 1998,

1999). When tested in vitro the same isolate revealed a clear

* Corresponding author.E-mail address: wolf@iptsp.ufg.br

0953-7562/$ – see front matter ª 2006 The British Mycological Society. Published by Elsevier Ltd. All rights reserved.doi:10.1016/j.mycres.2005.12.001

486 G. M. J. Lazzarini et al.

threshold for germination at 0.96 aw (Luz & Fargues 1997).

However, other studies showed that T. infestans may be killed

by B. bassiana regardless to ambient humidity (Lecuona et al.

2001). The divergent behaviour of B. bassiana may be related

to genetic variability among strains. Both M. anisopliae and B.

bassiana are found worldwide, and in arid or semi arid regions

strains may occur with a better adaptation to unfavourable

moisture. A better understanding of the impact of moisture

on fungal activity during infection of triatomines is important,

as many species of these vectors are common in regions with

dry climate. We report on the impact of water activity on in

vitro germination of different M. anisopliae and B. bassiana iso-

lates collected in triatomine-infested areas in Central Brazil

and their activity against T. infestans at different humidities

under laboratory conditions.

Materials and methods

Fungus origin and culture

Eleven isolates of Metarhizium anisopliae (IP 220, IP 221, IP 222, IP

225, IP 226, IP 227, IP 230, IP 232, IP 233, IP 234, IP 235) and Beau-

veria bassiana (IP 155, IP 157, IP 161, IP 165, IP 170, IP 171, IP 180,

IP 224, IP 228, IP 229, IP 231) were tested. All isolates originated

from substrates collected in rural peridomestic areas in

Central Brazil where triatomines are common (Luz et al.

2004a). Fungi were obtained from the collection of entomopa-

thogenic fungi at the Institute of Tropical Pathology and Public

Health (IPTSP), Federal University of Goias, Goiania, Brazil.

Fungi were grown on complete medium (CM: 0.001 g FeSO4;

0.5 g KCl; 1.5 g KH2PO4; 0.5 g MgSO4$H2O; 6 g NaNO3; 0.001 g

ZnSO4; 1.5 g hydrolyzed casein; 0.5 g yeast extract; 10 g glu-

cose; 2 g peptone; 20 g agar; 1000 ml water) at 25 �C� 1 �C,

and a 12 �h photophase. All isolates were previously host-pas-

saged on third-instar nymphs of Triatoma infestans at the be-

ginning of the reported experiments in order to stimulate

fungal activity. Conidia were obtained directly from 10 d-old

sporulating cultures by scraping and then suspending in

10 ml of 0.1 % Tween 80. Suspensions were adjusted to defined

concentrations based on haemocytometer counts.

Insect origin and rearing

Triatoma infestans originated from Jacarezinho, Parana, Brazil

and were mass-reared at the IPTSP. Insects were blood-fed

on chickens (Gallus gallus domesticus) at two-weeks intervals,

held at 25� 1 �C, 75� 5 % rh, and 12 h photophase (Silva &

Silva 1988).

In vitro tests

Conidial germination of all isolates was screened in liquid

minimal medium (MM: 2 g glucose; 2 g yeast extract; 1000 ml

H2O) at two different water activities, 0.93 aw and >0.99 aw.

The water activity 0.93 aw was adjusted by equilibration

with sodium chloride (117 g 1000 ml�1 medium) and no salt

was added to MM to obtain>0.99 aw (Luz & Fargues 1997). Con-

idia were inoculated in 125 ml MM at final 1� 106 conidia ml�1

and then incubated at 25� 1 �C and darkness at 120 rev min�1

up to 216 h. Samples of 1 ml were taken 4, 8, 12, 24, 48, 72, 96,

120, 144, and 216 h post inoculation ( p. i.) and stored at�20 �C

until microscopic observation. Water was added daily to MM

in order to provide a permanent volume of 125 ml. Progress

of germination was monitored by scoring conidia as ungermi-

nated, swollen, germination initiated or germinated. Conidia

were considered germinated with an elongating germ tube

longer than the conidial diameter (Luz & Fargues 1997). Germi-

nation rates were determined by examination of 100 conidia

per replicate. Each treatment consisted of four independent

replicates.

In vivo tests

Two Metarhizium anisopliae isolates (IP 225 and IP 230) and two

Beauveria bassiana isolates (IP 170 and IP 229) which showed ac-

celerated (IP 170 and IP 230), retarded (IP 229) or no germination

(IP 225) at 0.93 aw, were assayed on insects. At least five up

to six concentrations of suspended conidia, 1� 106, 3� 106,

1� 107, 3� 107, 1� 108, and 3� 108 conidia ml�1 were tested

corresponding to 2.33� 103, 7� 103, 2.33� 104, 7� 104,

2.33� 105, and 7� 105 CFU (colony forming unit) cm�2 treated

surface, respectively. Ten recently molted and unfed third in-

star nymphs (N3) were treated by directly spraying 5 ml of

each concentration with a Potter spray tower (Burkard, Hert-

fordshire). Final deposit was determined by spraying 5 ml sus-

pended conidia at different concentrations, as mentioned

before, on sterile cover slips (18� 18 mm), transferring in

10 ml sterile 0.1 % Tween 80, and inoculating 100 ml on chlor-

amphenicol added (0.5 g 1000 ml�1) CM. The number of CFU

was checked for 5 d p. i. Control insects were treated with

0.1 % Tween 80 only. After drying for 1 h at ambient tempera-

ture and humidity nymphs were placed on filter paper in plas-

tic Petri dishes (90� 15 mm) and transferred to test chambers

at 43 % rh, 75 % rh and humidity close to saturation. Humidities

of 43, 75 and 98 % inside the chambers were regulated with sat-

urated solutions of potassium carbonate, sodium chloride, and

potassium sulphate, respectively (Winston & Bates 1960). Test

chambers consisted of an airtight plastic box (33� 37� 22 cm)

with the salt solution arranged at the bottom (1000 g salt

500 ml�1 water). Petri dishes were put on a plastic rack that

permitted free air exchange inside the chamber. Nymphs

were incubated at 25� 1 �C and a 12 h photophase and were

not fed during the assays. Four independent replicates of ten

N3 each were used per treatment. Mortality was monitored

daily for 25 d. Dead insects were transferred to a humid cham-

ber and fungal development on cadavers examined for 15 d.

Data analysis

Germination and mortality data were arcsine-square root

transformed and then analysed using either Kruskal–Wallis

test (H) or one- or two-way analysis of variance (ANOVA) (F)

and the Student–Newman–Keuls (SNK) multiple range test

for comparison of means. Means were considered signifi-

cantly different at P< 0.05. Times to obtain 50 and 90 % germi-

nation (GT50 and GT90), lethal times (LT50 and LT90), and

concentrations (LC50 and LC90) to obtain 50 and 90 % mortality

were calculated by non-linear regression analysis or probit

analysis, respectively. The survival curves of nymphs were

Impact of moisture on Beauveria bassiana and Metarhizium anisopliae 487

fixed with the Kaplan–Meier test and analysed using the log

rank test (SAS Institute Inc. 2000).

Results

Germination of conidia at different water activities

At >0.99 aw, the first swollen conidia of all Metarhizium aniso-

pliae and Beauveria bassiana isolates were observed 4 h p. i. For-

mation of germ tubes initiated from then and up to 8 h p. i.

Three M. anisopliae and eight B. bassiana isolates showed ger-

mination rates �10 % at this time (Tables 1 and 2). Except for

IP 235, germination of M. anisopliae isolates was >98 % at

12 h p. i. (Table 1). Germination of B. bassiana isolates varied

from<1 % (IP 180) to 75 % (IP 161) at 12 h p. i. and�98 % germi-

nation of all isolates was observed only at 24 h p. i. (Table 2). A

highly significant effect of the isolate on the germination rate

at 8 h and 12 h p. i. of both fungal species (P< 0.001) was noted.

At 0.93 aw the first swollen conidia of both fungal species

were observed also at 4 h p. i., but germination started only be-

tween 48 h and 72 h p. i. (Tables 1 and 2). Whereas no germina-

tion of M. anisopliae conidia could be detected at 48 h p. i.,

B. bassiana IP 170 and IP 180 had already initiated germination.

At 72 h p. i. germination of M. anisopliae was <10 %, except IP

230 with 47.5 % of conidia germinated. Germination of B. bassi-

ana isolates at 72 h p. i. was distinctly higher compared with

M. anisopliae and reached 83.7 % (IP 170). Nine isolates had

values of germination >10 %. No germination of M. anisopliae

IP 225 incubated at 0.93 aw could be observed at 216 h p. i. Ex-

cept for IP 225 and IP 235 (47.5 %), all other isolates of this spe-

cies showed germination rates >96 % at 216 h p. i. This was

also found for B. bassiana isolates, except for IP 171 and IP

229 where 92 % and 57.5 % of the conidia had germinated

216 h p. i., respectively. A highly significant effect (P< 0.001)

of the isolate on the germination was found for M. anisopliae

and B. bassiana between 72 and 216 h. Regression curves

were fitted for most isolates (F> 120, df¼ 2, P< 0.0001) and

values of GT50 of M. anisopliae isolates varied between 98.2 h

(IP 233) and 137 h (IP 227) and for B. bassiana isolates between

59.3 h (IP 170) and 102 h (IP 155; Table 3). Lowest values of GT90

were found at 104 h for M. anisopliae (IP 233) and 78.8 h for

B. bassiana (IP 170). The number of germinated conidia of

M. anisopliae IP 225 and IP 235 and B. bassiana IP 229 were insuf-

ficient or germination was completed too quickly (IP 230 and IP

234) to calculate GT50/90.

Fungal infection of Triatoma infestans at differenthumidities

The first mortalities of fungus-treated nymphs were found 4–

6 d p. i. at all humidities tested. There was a highly significant

effect of humidity on cumulative mortalities between 10 and

25 d p. i. (P< 0.001; Table 4). At rh close to saturation (98 %

rh) mortalities of both M. anisopliae and B. bassiana-treated

N3 increased with the conidial concentration and time of in-

cubation (Figs 1 and 2). Except B. bassiana IP 170, all isolates

had induced total mortality within a 10 d exposure to this hu-

midity. Only 30.9 % of N3 treated with IP 170 had been killed at

10 d p. i. (Table 4). Values of LC50/90 found for isolates after 10 d

Ta

ble

1–

Cu

mu

lati

ve

germ

ina

tio

n(%

;±

S.E

.)o

fM

eta

rhiz

ium

an

isop

lia

eco

nid

iain

liq

uid

min

ima

lm

ed

ium

ad

just

ed

at

wa

ter

act

ivit

ies>

0.9

9a

wa

nd

0.9

3a

w,

25� C

,1

20

rev

min

L1

aw

Tim

e(h

)Is

ola

tes

test

ed

IP220

IP221

IP222

IP225

IP226

IP227

IP230

IP232

IP233

IP234

IP235

0.9

94

**

**

**

**

**

*

8þ

16.8�

2.6

a18.0�

1.5

a6.3�

1.4

b4.8�

1.5

b4.0�

0.6

b*b

20.5�

3.0

a*

b*

b*

b*b

12þ

**a

**a

**a

**a

**a

**a

**a

**a

**a

**a

61.5�

2.2

b

24

****

****

****

****

****

**

0.9

348

**

**

**

**

**

*

72þ

8.8�

0.9

b*b

*b

*b

1.0�

0.4

b*b

47.5�

2.3

a6.5�

0.6

b4.3�

0.3

b*

b*b

96þ

17.8�

1.5

c9.8�

1.0

d6.0�

0.4

e*f

25.3�

0.9

b10.5�

1.3

d**

a10.2�

0.6

d22.0�

1.1

b*

g4.5�

0.6

ef

120þ

43.3�

2.5

b44.8�

1.9

b44.8�

0.7

b*d

42.3�

0.9

b35.8�

0.6

d**

a12.3�

1.0

c**

a10.5�

0.6

c13.0�

1.3

c

144þ

75.3�

1.0

d86.3�

1.5

b79.8�

1.1

c*g

**a

53.8�

1.3

e**

a**

a**

a**

a38.0�

0.9

f

216þ

**a

96.3�

0.5

b**

a*d

**a

**a

**a

**a

**a

**a

47.5�

1.3

c

*G

erm

ina

tio

n<

1%

.**

Germ

ina

tio

n>

98

%.

þK

rusk

al–

Wa

llis

test

:42.9>

H>

40.6

;P<

0.0

01.

Sig

nifi

can

td

iffe

ren

ces

wit

hin

on

eli

ne

are

ind

ica

ted

by

dif

fere

nt

lett

ers

(Stu

den

t–N

ew

ma

n–K

eu

lste

st,

P<

0.0

5).

488 G. M. J. Lazzarini et al.

Ta

ble

2–

Cu

mu

lati

ve

germ

ina

tio

n(%

;±S

.E.)

ofB

eau

ver

iaba

ssia

na

con

idia

inli

qu

idm

inim

alm

ed

ium

ad

just

ed

at

wa

ter

act

ivit

ies>

0.9

9a

wa

nd

0.9

3a

w,2

5� C

,12

0re

vm

inL

1

aw

Tim

e(h

)Is

ola

tes

test

ed

IP155

IP157

IP161

IP165

IP170

IP171

IP180

IP224

IP228

IP229

IP231

0.9

94

**

**

**

**

**

*

8þ

10.0�

0.8

cd10.2�

0.6

cd16.2�

0.5

b20.5�

0.3

a13.2�

1.2

bc

15.5�

0.3

b*e

9.5�

0.9

d10.8�

0.8

cd10.5�

0.6

cd7.7�

0.8

d

12þ

44.0�

2.5

d13.5�

0.6

e75.0�

1.3

a67.2�

0.6

b71.5�

0.6

a74.2�

0.5

a*f

59.5�

3.5

c57.2�

1.6

c60.0�

1.3

c69.5�

1.8

ab

24

****

****

****

****

****

**

0.9

336

**

**

**

**

**

*

48

**

**

10.0�

0.4

*3.0�

0.4

**

**

72þ

15.5�

2.1

e36.5�

2.3

d40.5�

0.6

d52.7�

0.6

c83.7�

0.5

a5.0�

0.4

f44.0�

0.4

cd36.7�

0.5

d68.5�

0.6

b8.0�

0.4

f74.2�

0.8

b

96þ

57.7�

0.5

e70.7�

0.5

d76.7�

0.5

cd80.5�

0.6

c88.5�

0.6

b48.7�

1.0

f61.2�

1.3

e94.7�

1.7

a88.2�

0.5

b21.0�

0.4

g91.2�

0.5

ab

120þ

63.2�

0.8

bc

**a

80.0�

0.8

ab

86.5�

0.6

ab

**a

79.5�

0.6

ab

**a

**a

**a

26.5�

0.6

c**

a

144þ

67.5�

0.9

b**

a**

a**

a**

a85.2�

3.6

b**

a**

a**

a41.0�

0.4

c**

a

216þ

**a

**a

**a

**a

**a

92.0�

0.9

b**

a**

a**

a57.5�

0.6

c**

a

*G

erm

ina

tio

n<

1%

.**

Germ

ina

tio

n>

98

%.

þK

rusk

al–

Wa

llis

test

;42.9>

H>

42.4

;P<

0.0

01.

Sig

nifi

can

td

iffe

ren

ces

wit

hin

on

eli

ne

are

ind

ica

ted

by

dif

fere

nt

lett

ers

(Stu

den

t–N

ew

ma

n–K

eu

lste

st,

P<

0.0

5).

Table 3 – Germination time (h; ±S.E.) to achieve 50 or 90 %germination (GT50/90) of Metarhizium anisopliae andBeauveria bassiana conidia incubated in liquid minimalmedium adjusted at water activities 0.93 aw, 25 �C and120 rev minL1

Species Isolate GT50 (h) GT90 (h)

M. anisopliae IP 220 123.0� 4.0 f 173.0� 7.5 g

IP 221 120.9� 19.0 ef 157.0� 33.1 e-g

IP 222 122.8� 5.1 f 158.0� 8.7 f

IP 225 * *

IP 226 115.9� 12.1 ef 150.0� 21 e-g

IP 227 137.7� 27.7 ef 191.0� 49.3 fg

IP 230 ** **

IP 232 123.6� 15.4 ef 130.0� 27.1 d-f

IP 233 98.2� 21.8 c-f 104.0� 35.6 a-e

IP 234 ** **

IP 235 * *

B. bassiana IP 155 102.0� 7.9 de 181.2� 13.9 g

IP 157 79.9� 16.0 a-d 112.1� 24.5 b-e

IP 161 78.5� 6.3 c 120.0� 9.9 de

IP 165 72.4� 7.5 a-c 108.1� 11.3 c

IP 170 59.3� 5.1 a 78.8� 7.3 a

IP 171 97.1� 15.8 b-e 150.7� 27.1 de

IP 180 79.6� 11.5 bc 118.1� 18.6 de

IP 224 75.0� 19.4 a-d 90.6� 28.7 a-d

IP 228 66.1� 4.6 ab 87.6� 6.5 a-c

IP 229 * *

IP 231 64.5� 4.9 ab 82.8� 6.9 a

* Number of germinated conidia insufficient to calculate GT50/90.** Number of values insufficient for nonlinear regression analysis.

Significant differences based on confidence intervals (CI¼ 95 %)

within one column are indicated by different letters.

Table 4 – Cumulative mortality (%) (±S.E.) of Triatomainfestans third instar nymphs, 10 and 25 d after treatmentwith Metarhizium anisopliae and Beauveria bassianaisolates (5 ml of 1 3 108 conidia mlL1) and incubation atrh [ 43, 75 and 98 %, 12 h photoperiod, and 25 �C

Day Species Isolate * Relative humidity: ***(rh) 43 %

75 % 98 %

10 * M. anisopliae IP 225 7.5� 4.1 bc 7.5� 4.1 bc 100.0 a

IP 230 0.6� 2.5 c 6.2� 9.5 bc 100.0 a

B. bassiana IP 170 11.2� 6.5 bc 7.5� 4.1 bc 30.9� 11.1 b

IP 229 2.6� 2.9 bc 14.9� 9.1 bc 100.0 a

25 ** M. anisopliae IP 225 7.5� 4.1 b 7.5� 4.1 b 100.0 a

IP 230 0.6� 2.5 b 10.6� 8.7 b 100.0 a

B. bassiana IP 170 15.6� 7.1 b 9.4� 2.5 b 98.7� 7.1 a

IP 229 23.7� 2.9 b 19.2� 8.5 b 100.0 a

* Effect of isolate, df¼ 3;6, F¼ 6.06, P¼ 0.002.** Effect of isolate, df¼ 3;6, F¼ 1.59, P¼ 0.2.*** Effect of relative humidity, df¼ 2;6, F> 146, P< 0.001.

Values followed by different letters within the same day indicate

different mean mortality (two-way analysis of variance and

Student–Newman–Keuls test, P< 0.05).

Impact of moisture on Beauveria bassiana and Metarhizium anisopliae 489

0

20

40

60

80

100

05

1015

2025

CFU cm-2

IP 225

2.323

233 707

Day

s 0

20

40

60

80

100

05

1015

2025

CFU cm-2

IP 230

2.3

2370

7

Day

s

0

20

40

60

80

100

05

1015

2025

CFU cm-2 2.323

233 707

Day

s 0

20

40

60

80

100

05

1015

2025

CFU cm-2 2.3

2370

7

Day

s

0

20

40

60

80

100

05

1015

2025M

orta

lity

(

)

CFU cm-2

Mor

talit

y (

)

Mor

talit

y (

)

Mor

talit

y (

)

Mor

talit

y (

)

Mor

talit

y (

)

2.323

233 707

Day

s 0

20

40

60

80

100

05

1015

2025

CFU cm-2 2.3

2370

7

Day

s

Fig 1 – Cumulative mortality (%) of Triatoma infestans third instar nymphs after treatment with Metarhizium anisopliae iso-

lates [2.33 3 103 up to 7 3 105, corresponding to 3.37 to 5.85 log values of CFU (colony forming unit) cmL2] and

exposure to 43 %, 75 % and 98 % relative humidity, 25 �C and 12 h photophase during 25 d. Number of CFU presented

as linear values (multiplied by 103) on log transformed axis.

incubation at 98 % rh and values of LT50/90 of N3 treated with

1� 108 conidia ml�1 and exposed to the same humidity are

shown in Table 5. The highest LC50/90 data were found for

B. bassiana IP 170, which differed significantly from values

found for the other isolates. Regression curves were fitted

for all isolates (F> 108, df¼ 2, P< 0.0001), and there was

a highly significant effect of the isolate on survival of nymphs

(X2¼ 94.6, df¼ 3, P< 0.0001). LT50 (13.6 d) and LT90 (20.1 d) of IP

170 were highest and differed from the other isolates tested.

Their LT50 values varied between 5.9 d (IP 230) and 6.6 d (IP

225) and LT90 values between 6.7 d (IP 229) and 7.9 d (IP 230; Ta-

ble 5). No distinct relation between conidial concentration and

quantitative mortality could be observed at 43 % or 75 % rh. At

these conditions cumulative mortality within 25 d did not ex-

ceed 40 %, independently of the fungus species or isolates

tested (Figs 1 and 2, Table 4). Mortality of control insects dur-

ing 25 d p. i. did not exceed 10 %, regardless of humidity tested.

Development of inoculated fungi was observed on all

cadavers.

Discussion

Results showed a clear impact of moisture on initiation and

progress of conidial germination in vitro. At optimal water ac-

tivity (>0.99 aw) conidia of both fungi began germinating

490 G. M. J. Lazzarini et al.

0

20

40

60

80

100

05

1015

2025

CFU cm-2

IP 170

23233

70

7

Day

s 0

20

40

60

80

100

05

1015

2025

CFU cm-2

IP 229

23

23370

7

Day

s

0

20

40

60

80

100

05

1015

2025

CFU cm-2

23

23370

7

Day

s 0

20

40

60

80

100

05

1015

2025

CFU cm-2

23233

70

7

Day

s

0

20

40

60

80

100

05

1015

2025

CFU cm-2

23

23370

7

Day

s 0

20

40

60

80

100

0

510

1520

25

CFU cm-2

232.3

23370

7

Day

s

Mor

talit

y (

)

Mor

talit

y (

)

Mor

talit

y (

)

Mor

talit

y (

)

Mor

talit

y (

)

Mor

talit

y (

)

Fig 2 – Cumulative mortality (%) of Triatoma infestans third instar nymphs after treatment with Beauveria bassiana iso-

lates [2.33 3 103 up to 7 3 105 CFU (colony forming unit) cmL2] and exposure to 43 %, 75 % and 98 % relative humidity, 25 �C

and 12 h photophase during 25 d. Number of CFU presented as linear values (multiplied by 103) on log transformed axis.

between 4 and 8 h of incubation. Most Metarhizium anisopliae

isolates completed germination up to 12 h and B. bassiana up

to 24 h p. i. Water activity below 0.93 had a considerable neg-

ative effect on the germination kinetics. Conidia initiated

spherical growth within 4 h of incubation but germinated

only between 48 and 72 h p. i. Interestingly, Beauveria bassiana

conidia germinated faster at 0.93 aw than did M. anisopliae con-

idia. Even at the unfavourable 0.93 aw, most isolates of both

fungal species completed germination within 216 h, except

M. anisopliae IP 225 where no germination was detected. Differ-

ences of quantitative germination between isolates of the

same species were more apparent at unfavourable 0.93 aw,

than at >0.99 aw.

An increasing delay of germination and growth of M. aniso-

pliae, B. bassiana and other entomopathogenic fungi at de-

creasing water activities has been reported by other authors

(Florido et al. 2002; Hallsworth & Magan 1999; Humphreys

et al. 1989; Inch & Trinci 1987; Luz & Fargues 1997; Milner

et al. 1997). A distinct species- and strain-related variability

of germination was reported by Hywel-Jones and Gillespie

(1990) for B. bassiana and M. anisopliae conidia on Sabouraud

dextrose agar (SDA) at 100 % rh. These authors found, as did

we in this study, that conidia of M. anisopliae strains germi-

nated generally more quickly than those of B. bassiana strains.

Various other studies showed that water activities of 0.90–0.96

aw are necessary for in vitro development of these fungi. In

Impact of moisture on Beauveria bassiana and Metarhizium anisopliae 491

Table 5 – Lethal concentration [LC50/90: CFU (colony forming unit) cmL2] of Metarhizium anisopliae and Beauveria bassiana, 10and 25 d after treatment, and lethal time (LT50/90: days) to kill 50 or 90 % of Triatoma infestans third instar nymphs atrh [ 98 % and 25 �C

Days aftertreatment

Species Isolate LC50/90 LT50/90 (1� 108 conidia ml�1)

LC50 (CI) LC90 (CI) LT50 (CI) LT90 (CI)

10 M. anisopliae IP 225 2.8� 103 a (5.8� 102–4.9� 103) 1.8� 104 a (9.3� 103–1.8� 105) 6.6� 1.1 a 7.5� 1.1 a

IP 230 2.8� 103 a (2.2� 102–6.3� 103) 2.1� 104 a (9.1� 103–3.3� 105) 5.9� 1.2 a 7.9� 1.2 a

B. bassiana IP 170 1.2� 106 b (4.4� 105–1.4� 107) * 13.6� 1.8 b 20.1� 1.8 b

IP 229 2.8� 103 a (9.1� 100–5.1� 103) 8.6� 103 a (4.9� 103–1.9� 1010) 6.0� 1.5 a 6.7� 1.5 a

25 M. anisopliae IP 225 ** ** d d

IP 230 ** ** d d

B. bassiana IP 170 5.6� 102 (3.3� 101–2� 103) 7.7� 104 (4.4� 104–1.8� 105) d d

IP 229 ** ** d d

* Number of dead nymphs insufficient to calculate LC50/90.** Number of dead nymphs too high to calculate LC50/90.

Significant differences within one column and different days (LC50/90) based on confidence intervals (CI 95 %, probit analysis) and within one

column (LT50/90) based on nonlinear regression and T test are indicated by different letters.

addition to inter- and intraspecific variability, techniques of

submerged and solid cultivation, the solute used to adjust wa-

ter activity and nutrients added to the medium interfered with

test conditions and determined fungal development at re-

duced water availability. B. bassiana conidia germinated either

on nutrient agar adjusted to 0.90 aw and exposed to a corre-

sponding relative humidity or in liquid medium at 0.92 aw

(Luz & Fargues 1997). Hallsworth and Magan (1999) found

that M. anisopliae and B. bassiana inoculated on SDA medium

developed at 0.93 aw. However, germination of M. anisopliae

conidia in liquid medium was reported to be completely

inhibited in another study at <0.96 aw by Milner et al. (1997)

and was not observed, even after as much as 240 h of incuba-

tion on SDA, for either M. anisopliae and B. bassiana at �0.94 aw

(Hallsworth & Magan 1995). The same authors showed that

manipulation of the polyol content in conidia extends the

range of water availability as conidia of B. bassiana and M. ani-

sopliae with higher intracellular concentrations of glycerol and

erythritol germinated at water activities as low as 0.89 aw

(Hallsworth & Magan 1995).

In this study, daily compensated liquid medium with

added minimal nutrients and salt provided constant and pre-

cise conditions of water activity. However, in vitro conditions

did not reflect conditions in the microhabitats on the insect

and complex interactions of conidia and cuticle. Results

obtained in vitro should be extrapolated carefully. Previous

studies showed that B. bassiana, M. anisopliae and other fungi

were able to infect insects at lower humidities (Ferron 1977;

Hsiao et al. 1992; Lecuona et al. 2001; Marcandier & Khacha-

tourians 1987; Milner et al. 1997; Ramoska 1984).

Fungal activity is directly related to the dose. Quantitative

contamination of the cuticle with conidia depends on applica-

tion techniques in standard laboratory assays. James et al.

(1998) found five times greater doses per insect after immer-

sion than after spray application. Moreover, a boundary layer

of elevated moisture on the cuticle may permit germination of

a higher number of conidia even at lower ambient humidities.

Obviously there was no relationship between germination

kinetics at the unfavourable 0.93 aw and their activity against

T. infestans nymphs at different humidities. Isolates such as

M. anisopliae IP 230 and B. bassiana IP 170 that showed a faster

germination and higher number of germinated conidia at 0.93

aw, did not induce higher mortalities at the unfavourable rel-

ative humidities of 75 % or 43 %. On the contrary, IP 170 was

the least virulent isolate at all humidities and among isolates

tested. In addition, isolates with reduced rates of germination,

(B. bassiana IP 229) or without germination (M. anisopliae IP 225)

at 0.93 aw were not less virulent against nymphs, regardless of

the humidity tested. In vivo tests showed no difference in ac-

tivity between both fungal species in comparison with the in

vitro tests. The reduced virulence at lower 43 % and 75 % rh

may be related to a low number of germinating conidia and in-

vading propagules or to a loss of conidial adherence after pro-

longed exposure on the cuticle. In the present study mortality

did not increase at 43 % rh as observed by other authors

(Romana 1992; Luz 1990).

There are only few studies correlating in vitro and in vivo be-

haviour of entomopathogenic fungi. Drummond et al. (1987)

found no relationship between speed of germination in vitro

of Verticillium lecanii isolates and their activity against white-

flies. In contrast, Matawele et al. (1994) observed an elevated

virulence against green leafhopper (Nephotettix virescens)

among mutants of M. anisopliae and Paecilomyces farinosus

that germinated and grew better in vitro at reduced water ac-

tivity or humidity than the parental strains.

It is important to mention that our results regarding the

correlation between in vitro and in vivo behaviour of B. bassiana

and M. anisopliae are based on only few fungal isolates and

may be different for other strains or insect models. Isolates

of M. anisopliae and B. bassiana collected in an area with an ex-

tremely dry climate during several months of the year germi-

nated under unfavourable conditions of water activity after

prolonged incubation but were not more virulent against

T. infestans at unfavourable humidities. No relation between

behaviour in vitro and activity in vivo was found for the tested

isolates. More studies on the impact of moisture and other fac-

tors during host invasion will be useful to develop a bioinsec-

ticide based on these fungi.

492 G. M. J. Lazzarini et al.

Acknowledgements

The authors thank Ionizete G. da Silva for providing Triatoma

infestans, Alexandre S. G. Coelho for advice on statistical anal-

ysis, Mark S. Goettel for the constructive review of the manu-

script and the National Council of Scientific and Technological

Development (CNPq) for financial support.

r e f e r e n c e s

Drummond J, Heale JB, Gillespie AT, 1987. Germination and effectof reduced humidity on expression of pathogenicity in Verti-cillium lecanii against the glasshouse whitefly Trialeurodesvaporariorum. Annals of Applied Biology 111: 193–201.

Fargues J, Luz C, 1998. Effects of fluctuating moisture and temper-ature regimes on sporulation of Beauveria bassiana on cadaversof Rhodnius prolixus. Biocontrol Science and Technology 8: 323–334.

Fargues J, Luz C, 2000. Effects of fluctuating moisture and tem-perature regimes on the infection potential of Beauveriabassiana for Rhodnius prolixus. Journal of Invertebrate Pathology75: 202–211.

Ferron P, 1977. Influence of relative humidity on the developmentof fungal infection caused by Beauveria bassiana (Fungi Imper-fecti, Moniliales) in imagines of Acanthoscelides obtectus (Col.:Bruchidae). Entomophaga 22: 393–396.

Florido JEB, Rosas RA, Rojas MG, Gonzalez GV, Castaneda GS,2002. Criteria for the selection of strains of entomopathogenicfungi Verticillium lecanii for solid state cultivation. Enzyme andMicrobial Technology 30: 910–915.

Hallsworth JE, Magan N, 1995. Manipulation of intracellularglycerol and erythritol enhances germination of conidia at lowwater availability. Microbiology 141: 1109–1115.

Hallsworth JE, Magan N, 1999. Water and temperature relations ofgrowth of the entomogenous fungi Beauveria bassiana, Meta-rhizium anisopliae, and Paecilomyces farinosus. Journal of AppliedEntomology 74: 261–266.

Hsiao WF, Bidochka MJ, Khachatourians GG, 1992. Effect of tem-perature and relative humidity on the virulence of the ento-mopathogenic fungus, Verticillium lecanii, toward the oat-berryaphid, Rhopalosiphum padi (Hom., Aphidae). Journal of AppliedEntomology 114: 484–490.

Humphreys AM, Matawele P, Trinci APJ, Gillespie AT, 1989. Effectsof water activity on morphology, growth and blastosporeproduction of Metarhizium anisopliae, Beauveria bassiana andPaecilomyces farinosus in batch and fed-batch culture. Mycolog-ical Research 92: 257–264.

Hywel-Jones NL, Gillespie AT, 1990. Effect of temperature onspore germination in Metarhizium anisopliae and Beauveriabassiana. Mycological Research 94: 389–392.

Inch JMM, Trinci APJ, 1987. Effects of water activity on growth andsporulation of Paecilomyces farinosus in liquid and solid media.Journal of General Microbiology 133: 247–252.

James RR, Croft BA, Shaffer BT, Lighthart B, 1998. Impact oftemperature and humidity on host–pathogen interactionbetween Beauveria bassiana and a coccinelid. EnvironmentalEntomology 27: 1506–1513.

Lecuona RE, Edelstein JD, Berretta MF, Rossa FR, Argas JA, 2001.Evaluation of Beauveria bassiana (Hyphomycetes) strains aspotential agents for control of Triatoma infestans (Hemiptera:Reduviidae). Journal of Medical Entomology 38: 172–179.

Luz C, 1990. Zur Pathogenitat von Beauveria bassiana (Fungi, Im-perfecti) gegenuber mehreren Raubwanzenarten (Reduviidae,Triatominae) und Einfluss der relativen Luftfeuchtigkeit auf dieInfektion von Rhodnius prolixus. Mitteilungen der DeutschenGesellschaft fur Allgemeine und Angewandte Entomologie 7: 510–511.

Luz C, Fargues J, 1997. Temperature and moisture requirementsfor conidial germination of an isolate of Beauveria bassiana,pathogenic to Rhodnius prolixus. Mycopathologia 138: 117–125.

Luz C, Fargues J, 1998. Factors affecting conidial production ofBeauveria bassiana from fungus-killed cadavers of Rhodniusprolixus. Journal of Invertebrate Pathology 72: 97–103.

Luz C, Fargues J, 1999. Dependence of the entomopathogenicfungus, Beauveria bassiana, on high humidity for infection ofRhodnius prolixus. Mycopathologia 146: 33–41.

Luz C, Rocha LFN, Humber RA, 2003. Record of Evlachovaea sp.(Hyphomycetes) on Triatoma sordida in the State of Goias, Braziland its activity against Triatoma infestans (Reduviidae, Triato-minae). Journal of Medical Entomology 40: 451–454.

Luz C, Rocha LFN, Nery GV, 2004a. Detection of entomopatho-genic fungi in peridomestic triatomine-infested areas in Cen-tral Brazil and fungal activity against Triatoma infestans (Klug)(Hemiptera: Reduviidae). Neotropical Entomology 33: 783–791.

Luz C, Rocha LFN, Nery GV, Magalhaes BP, Tigano MS, 2004b.Activity of oil-formulated Beauveria bassiana against Triatomasordida in peridomestic areas in Central Brazil. Memorias doInstituto Oswaldo Cruz 99: 211–218.

Luz C, Rocha LFN, Silva IG, 2004c. Pathogenicity of Evlachovaea sp.(Hyphomycetes), a new species isolated from Triatoma sordida,in Chagas disease vectors under laboratory conditions. Revistada Sociedade Brasileira de Medicina Tropical 37: 189–191.

Luz C, Silva IG, Cordeiro CMT, Tigano MS, 1998a. Beauveria bassi-ana (Hyphomycetes) as a possible control agent for the vectorsof Chagas disease. Journal of Medical Entomology 35: 977–979.

Luz C, Tigano MS, Silva IG, Cordeiro CMT, Aljanabi SM, 1998b.Selection of Beauveria bassiana and Metarhizium anisopliae tocontrol Triatoma infestans. Memorias do Instituto Oswaldo Cruz93: 839–846.

Marcandier S, Khachatourians GG, 1987. Susceptibility of themigratory grasshopper, Melanoplus sanguinipes (Fab.) (Orthop-tera: Acrididae), to Beauveria bassiana (Bals.) Vuillemin (Hypho-mycetes): influence of relative humdity. Canadian Entomologist119: 901–907.

Matawele P, Trinci APJ, Gillespie AT, 1994. Mutants of entomo-pathogenic fungi that germinate and grow at reduced wateractivities and reduced relative humidities are more virulent toNeophotellix virescens (green leafhopper) than the parentalstrains. Mycological Research 98: 1329–1333.

Milner RJ, Staples JA, Lutton GG, 1997. The effect of humidity ongermination and infection of termites by the hyphomycete,Metarhizium anisopliae. Journal of Invertebrate Pathology 69: 64–69.

Ramoska WA, 1984. The influence of relative humidity on Beau-veria bassiana infectivity and replication in the chinch bug,Blissus leucopterus. Journal of Invertebrate Pathology 43: 389–394.

Romana CA, 1992. Recherches sur les potentialites des hypho-mycetes entomopathogenes (Fungi Imperfecti) dans la luttecontre les Triatominae (Heteroptera). These de Doctorat,Montpellier, 129 pp.

SAS Institute Inc., 2000. What’s New in SAS� Software for Release8.1. SAS Institute Inc., Cary, N.C.

Silva IG, Silva HHG, 1988. Influencia da temperatura na biologia detriatomıneos. IV. Triatoma infestans Klug, 1834 (Hemiptera, Re-duviidae). Anais da Sociedade Entomologica do Brasil 17: 443–454.

Winston PW, Bates DH, 1960. Saturated solutions for the controlof humidity in biological research. Ecology 41: 232–237.