Review of Palaeobotany and Palynology 221 (2015) 171–178

Contents lists available at ScienceDirect

Review of Palaeobotany and Palynology

j ourna l homepage: www.e lsev ie r .com/ locate / revpa lbo

Research paper

Pollen as nutrient source in Holocene ombrotrophic bogs

Lyudmila S. Shumilovskikh a,b,c,⁎, Frank Schlütz d, Inke Achterberg a, Anna Kvitkina e,Andreas Bauerochse f, Hanns Hubert Leuschner a

a Department of Palynology and Climate Dynamics, Georg-August-University of Göttingen, Göttingen, Germanyb Laboratory of Biogeochemical and Remote Techniques of Environmental Monitoring, National Research Tomsk State University, Tomsk, Russiac Mediterranean Institute of Marine and Terrestrial Biodiversity and Ecology (IMBE, UMR CNRS 7263/IRD 237/Aix-Marseille University), Europôle Méditerranéen de l'Arbois, Pavillon Villemin,BP 80, 13545 Aix-en-Provence Cedex 04, Franced Lower Saxony Institute for Historical Coastal Research, Viktoriastrasse 26/28, 26382 Wilhelmshaven, Germanye Institute of Physicochemical and Biological Problems in Soil Science of the Russian Academy of Sciences, ul. Institutskaya 2, 142290 Pushchino, Moscow Oblast, Russiaf Lower Saxony State Service for Cultural Heritage, Scharnhorststrasse 1, 30175 Hannover, Germany

⁎ Corresponding author at: Mediterranean InstituBiodiversity and Ecology (IMBE, UMR CNRS 7263/IRD 2Europôle Méditerranéen de l'Arbois, Pavillon Villemin,Cedex 04, France. Tel.: +33 442908468.

E-mail addresses: shumilovskikh@yahoo.com, lyudmil(L.S. Shumilovskikh), schluetz@nihk.de (F. Schlütz),Inke.Achterberg@biologie.uni-goettingen.de (I. Achterbergandreas.Bauerochse@NLD.Niedersachsen.de (A. Baueroch(H.H. Leuschner).

http://dx.doi.org/10.1016/j.revpalbo.2015.07.0010034-6667/© 2015 Elsevier B.V. All rights reserved.

a b s t r a c t

a r t i c l e i n f o

Article history:Received 24 November 2014Received in revised form 23 June 2015Accepted 1 July 2015Available online 11 July 2015

Keywords:PaleoecologyFungusDecompositionAnemophilyNon-pollen palynomorph

Several laboratory experiments and field observations demonstrate that saprotrophic and ectomycorrhizal fungieffectively use pollen cytoplasm and suggest pollen produced by wind-pollinated trees as a crucial nutrientsource for fungi and their plant hosts in nutrient-limited environments. However, the role of such interactionsis still underestimated or neglected in ecology and paleoecology. Here, we consider pollen attacked by fungi inpalynological records fromHolocene raised peat bogs as a nutrient source for ecosystems in the past. The attacksoccurred through the pollen germination areas by a variety of saprotrophic, ericoidmycorrhizal, ectomycorrhizalor dark septate endophyte fungi. Large amounts of attacked pollen in phases rich in Calluna vulgaris highlight theimportance of Ericaceae shrubs, hosting ericoid mycorrhizal fungi and forming hotspots of decomposition innutrient-deficit bogs. Applying estimations of pollen rain from literature, and based on own observed pollen in-fection rates we estimate the annual release of nitrogen, phosphate and potassium from pollen, and highlighttheir significance in pushing the ecosystem nutrient cycle in early spring time, when several species releasetheir pollen. We highlight the significant role of anthropogenic changes in pollen deposition for pre-industrialbogs and hypothesize about the consequences of the pollen-based interrelation between wind-pollinated plantsand their mycorrhizal fungi in paleoecology and evolution.

© 2015 Elsevier B.V. All rights reserved.

1. Introduction

In northern terrestrial ecosystems and especially in organic-richsoils, nutrients are present mostly in organic forms that are not directlyavailable to plant roots, what makes the decaying activities in the soiland presence of easily degradable organic matter critical for the func-tioning of the entire ecosystem (e.g. Michelsen et al., 1998; Read et al.,2004; Vohník et al., 2011).

Pollen cytoplasm contains important nutrients such as N, P and K(Knight et al., 1972). The N content makes up about 2–4% of pollen dryweight and 60% of this N is water-soluble, making it a high quality andreadily degradable nutrient-rich material for organisms (Greenfield,

te of Marine and Terrestrial37/Aix-Marseille University),BP 80, 13545 Aix-en-Provence

a.shumilovskikh@imbe.fr

), aqvia@mail.ru (A. Kvitkina),se), hleusch@gwdg.de

1999). By entering terrestrial ecosystems, all the pollen that was notused for reproduction turns to a nutritional source. The total annual pol-len deposition estimations reported vary from 16–25 kg/ha in the borealforest in Canada (Lee et al., 1996) to 55–96 kg/ha in temperate forests inJapan (Saito and Takeoka, 1985). For boreal forests in Canada, this polleninput corresponds to nutrient inputs of 0.34–0.49 kgN/ha, 0.04–0.07kgP/ha and 0.11–0.2 kgK/ha (Lee et al., 1996). Although these values ap-pear relatively small compared with boreal forest nutrient uptake rates,pollen fallen to ground provides a rapid pulse of easily available nutri-ents and energy to soil microbes in the spring (Stark, 1972; Lee et al.,1996; Greenfield, 1999). Lee et al. (1996) concluded that pollen-mediated “nutrient cascades” operate in boreal forest ecosystems andare important in ombrotrophic bogs and oligotrophic lakes, where at-mospheric deposition is the principal form of nutrient input.

Several groups of soil microorganisms are able to use nutrients frompollen cytoplasm. Microscopic observations (Goldstein, 1960; Stark,1972; Greenfield, 1999) demonstrated that the contents of pollen grainsare susceptible to decomposition by bacteria, actinomycetes, chytrids,and hyphal fungi. Laboratory experiments with Pinus pollen as singlenutrient source for filamentous fungi (Hutchison and Barron, 1997a)and observations in the ectomycorrhizal association between Paxillus

172 L.S. Shumilovskikh et al. / Review of Palaeobotany and Palynology 221 (2015) 171–178

involutus and Betula pendula (Perez-Moreno and Read, 2001a) showthat the fungal hyphae attack pollen grains and utilize their cytoplasmfor nutrients. Moreover, mycorrhizal fungi pass a significant proportionof pollen nutrients to their autotrophic host, and therefore, to one of theproducers of pollen.

Experienced Quaternary palynologists are familiar with the occur-rence of fungal growth inside pollen grains. Although laboratory exper-iments and field observation highlight pollen as an important nutrientprovider for the recent boreal ecosystem, there are no systematic stud-ies available on the relationships between pollen deposition, fungal ac-tivities and nutrient cycling. This paper presents the records of pollenattacked by dark-colored fungal hyphae found in theHolocenepeat sed-iments from several raised bogs in northern Germany; thereafter, wediscuss in detail about the importance of pollen as a nutritional sourcefor ecosystems in the past, including mechanisms of fungal attack onpollen, taxonomic groups of fungi involved, and estimates of nutrientloading on the boreal bog ecosystem via pollen deposition.

2. Material and methods

As part of a dendroecological study, six peat profiles from fourSphagnum raised bogs in Northern Germany (Fig. 1) were investigatedfor palynological analyses, using a standard procedure including treat-ments with HCl, KOH, HF and acetolysis. All records prove the presenceof pollen attacked by fungi (further details to each site in Achterberg,

Fig. 1.Map of Lower Saxony (Germany) with locations of investigated pollen records: BorstelMoor (Bauerochse et al., in press), Totes Moor: M724 and M729 (Achterberg, 2010), M726 (Sc

2010; Eckstein et al., 2010; Bauerochse et al., in press; Shumilovskikhet al., 2015). In order to discuss the relation of such pollen to vegetationand environmental changes, we present themost relevant palynologicaltaxa as diagrams (Fig. 2). Changes in the pollen frequency of Pinusdiploxylon-type (further Pinus), Quercus robur-type (further Quercus),Corylus, Alnus, Betula, and Calluna vulgaris reflect vegetation changesin the surrounding forests and dwarf shrubs on the bog, Sphagnum to-gether with the thecamoebes Amphitrema, Arcella and Assulina indicatehydrological changes within the bog while charcoal fragments docu-ment fire influence. Taxa percentages in the pollen diagrams are basedon the total sum of identified pollen grains of 600–1200.

3. Results

3.1. Morphology

Fungal infection by dematiaceous (dark-colored) hyphae is docu-mented in our material for pollen of gymnosperms and angiospermsas well as for spores of ferns and of Sphagnum (Plate I). From 17 pollentypes attacked by fungi, 12 are wind-pollinated (i.e. Pinus, Alnus, Betula,Corylus) and 4 are (mostly) insect-pollinated (i.e. Tilia, Calluna vulgaris)(Table 1). The observed hyphae are concentrated in the aperture areas.In Angiosperm pollen, these are pores and/or colpi; thin regions in thepollen wall are supposed to be passed by the germination tube andtherefore not entirely covered by the very inert sporopollenin layer of

er Moor (Shumilovskikh et al., 2015), Venner Moor (Eckstein et al., 2010), Großes Uchterhlütz unpubl.).

Fig. 2. Selected curves from pollen diagrams (%) of Borsteler Moor, Venner Moor, Großes Uchter Moor and Totes Moor with percentages of attacked pollen calculated for common taxa.Yellow zones indicate phases with high amounts of attacked pollen.

173L.S. Shumilovskikh et al. / Review of Palaeobotany and Palynology 221 (2015) 171–178

the outer pollen wall (Punt et al., 2007). In the gymnosperm pollen ofPinus and Picea, aperturoid areas andmajor occurrence of hyphae are lo-cated at the distal side between the two air-sacs of the pollen grains(Plate I: 1, 2, 4). From there hyphae often grow into the air-sacs (PlateI: 2, 4, 7). In angiosperms, the main hyphal concentrations are in thepore areas of Alnus, Corylus, Tilia, Matricaria-type and C. vulgaris (PlateI: 5, 6, 8, 9, 11, 12) . In an initial stage, the presence of dark-brown hy-phae in the apertures of yellowish pollen grains is typical (Plate I: 5, 6,8, 12). Progressive infection is characterized by a developed dark-brown layer inside the pollen grains (Plate I: 1, 2, 4, 9, 11). Furthermore,pollen grains can befilled by objects of possibly sclerotious nature (PlateI: 3, 10, 13). The hyphal attack of Sphagnum spores occurs through thetrilete mark, the region of spore germination (Plate I: 14, 15).

3.2. Records

In the pollen record Borsteler Moor, 41 of 45 samples contain up to22% pollen and spores infected by fungi. The pollen spectra are dominat-ed by Alnus, Corylus, Quercus and Ericaceae (Fig. 2). However, the mostattacked pollen types are Pinus (16%), followed by Alnus (9%), Corylus(8%), and by Ericaceae pollen like Calluna vulgaris (7%) and theVaccinium-type (8%). From Quercus pollen only 2% and from the abun-dant Sphagnum spores (73%) only 1.4% show signs of infection.

In the Venner Moor, in half of samples pollen and spores wereattacked (max. 10%). The trends are similar to the Borsteler Moor(Fig. 2) with Alnus, Corylus, Quercus and Calluna vulgaris dominatingthe pollen record, while Pinus (10%), Alnus (7%), Corylus (6%) andC. vulgaris (5%) are most infected. Quercus pollen was not attacked.

Three records from the Totes Moor are characterized by generallylow percentages of attacked pollen with maxima varying between 2%(core M726) and 8% (core M724). Similar to the Borsteler Moor andthe Venner Moor, the dominating pollen taxa are Alnus, Corylus andQuercus, in core M726 also Pinus and Betula. Ericaceae are rare. Themost attacked taxa are Betula (5%), Pinus (4%) and Corylus (4%) inM724 and Pinus (9%) and Alnus (3%) in core M729, and Alnus, Corylusand Betula in core M726.

In the pollen record from Großes Uchter Moor 8 of 24 samplescontained attacked pollen, mostly of Alnus, Calluna vulgaris and Corylus.The maxima (4–10%) occurred in samples with high amounts ofC. vulgaris pollen and charcoal fragments.

4. Discussion

Depending on the species, wind-pollinated trees spread out be-tween17 and 450 kg/ha of pollen per year (Saito, 2012). Only aminorityof the grains reach a stigma and germinate there. The major amount of

174 L.S. Shumilovskikh et al. / Review of Palaeobotany and Palynology 221 (2015) 171–178

Table 1Characteristics of plants producing pollen types attacked by fungi in the studiedpollen records: flowering months and pollinator following Jäger (2011), pollen volume(cm3 × 10−6) and protein content (%) from Roulston et al. (2000).

Type Months Pollination Pollen volume Protein content

Abies V–VI Wind – 24.1Alnus II–IV Wind 3.9 24.2Artemisia VII–X Wind – 19.5Betula IV–V Wind 3.2 28.8Calluna vulgaris VIII–X Insects, wind – –Corylus II–IV Wind 6.7 30.2Empetrum/Ledum IV–VII Insects – –Fraxinus excelsior-type IV–V Wind – –Matricaria-type V–VIII Insects – 15Picea IV–VI Wind – 20.9Pinus diploxylon-type V–VI Wind 20–80 15.7Poaceae V–VIII Wind 11–14 20–32Quercus robur-type V Wind 5.6 30.6Tilia VI–VII Insects – –Typha latifolia-type VII–VIII Wind 6.9 19.2Ulmus III–IV Wind – 26.4Vaccinium V–VIII Insects 9 43

175L.S. Shumilovskikh et al. / Review of Palaeobotany and Palynology 221 (2015) 171–178

the huge released pollen masses lands on the ground and isdecomposed. However, details of decomposition in general, and the de-composition of pollen in particular are still unknown (Wilkinson, 2008).Since raised bogs present an extreme of nutrient impoverishment andare characterized by excellent pollen preservation (e.g. Erdtman,1943), the six studied Holocene raised bog cores offer a unique insightinto the decay of pollen by fungi. Based on the paleo-records and a re-view of mycological research we discuss 1) how do fungi attack pollen,2) which fungi can attack pollen, 3) which pollen is the “tastiest”, and4) the significance of pollen as nutrient source for paleo-ecosystems.

4.1. How do fungi attack pollen?

High concentrations of fungal hyphae in pollen aperture areasstrongly suggest that penetration occurs through sites intended for pol-len tube emergence (Plate I). This is supported bymicroscopic observa-tions of Goldstein (1960), Fægri (1971), Stark (1972) and Hutchisonand Barron (1997a), showing that pollen cytoplasm supports microbialgrowth but the sporopollenin of the outer pollen wall is left intact. Thecontent of pollen grains is protected by the pollen wall, which consistsof two layers, the inner pectic–cellulosic intine and the outer exine,protected by degradation-resistant sporopollenin (Havinga, 1967).When a pollen grain falls on a receptive stigma, the stored RNA, protein,and bioactive small molecules allow rapid germination and outgrowthof a tube, an extension of the pectinaceous intine (Taylor and Hepler,1997). Thereby, the sporopollenin at the apertures, which is nearly al-ways fragmented and forms a series of loosely linked granules, isattacked by enzymes from the intine (Heslop-Harrison, 1971).

The areas that are less, or even unprotected by sporopollenin, suchas apertures and pollen tubes, can be easily attacked by fungal hyphaeproducing corresponding enzymes. Laboratory experiments with fila-mentous fungi by Hutchison and Barron (1997a) demonstrate that fun-gal hyphae grow towards pollen grains, contact with them, wraparound the pollen tube or pollen grain and gain access to the contentvia physical penetration through germ pores or lysing the pollen tube.The attraction of hyphae to the pollen grains is presumed to bechemotrophic in nature (Hutchison and Barron, 1997a) and could be

Plate I. Pollen attacked by fungi: 1 Pinus pollen with hypha invasion at distant pole area (Borste26 cm); 3 Pinus pollen filled by cells of possibly pseudosclerotious nature (BorstelerMoor, 6 cminvasion at germination pore areas (BorstelerMoor, 8 cm); 6Matricaria-type pollenwith hypha8 cm); 7 Abies air-sac coveredbyhyphae (Bortsel, 26 cm); 8 Alnuspollenwith hyphal invasion atand entire pollen grain brown colored (BorstelerMoor, 28 cm); 10 Alnus pollen filled by cells ofgermination pores and entire pollen grain (TotesMoor:M726, 173 cm); 12Calluna vulgaris tetraby cells of possibly pseudosclerotious nature (Borsteler Moor, 6 cm); 14, 15 Sphagnum spore w10 μm.

caused by germination of grains or their leakage, provoked by exposureof desiccated grains to bulkwater (Taylor andHepler, 1997). In addition,compounds of the wall adhering pollenkitt like waxes, lipids, and smallaromatic molecules and proteins (Taylor and Hepler, 1997), may alsoattract hyphae.

4.2. Which fungi can attack pollen?

Even though bogs, especially Sphagnum bogs, are very well knownfor their good preservation of organic material (Erdtmann 1943), theycontain large microbial populations of wide metabolic diversity suchas fungi, bacteria and archaea (Andersen et al., 2013). Under the aerobicconditions of the peat surface mainly saprotrophic fungi, differentmycorrhizal fungi and yeasts perform the decay of organic litter(Andersen et al., 2013).

Saprotrophic fungi are a major group with capacity to use nutrientsfrompollen. Hutchison and Barron (1997a) demonstrated that 41 out of157 filamentous fungal species positively responded to pollen grains ofPinus nigra. Especially lignicolous Basidiomycota comprising litter- andwood-decaying species were able to utilize the pollen content. The abil-ity of lignicolous fungi to parasitize is also documented for algae(Hutchison and Barron, 1997b), bacteria (Barron, 1988), yeasts(Hutchison and Barron, 1996), filamentous fungi (Griffith and Barnett,1967), basidiospores (Fries and Swedjemark, 1985), testate amoebae(Vohník et al., 2011) and nematodes (Barron and Dierkes, 1977).Barron (1992) suggested these sources allow fungi to obtain extra nitro-gen for the degradation of nitrogen-poor litter and wood.

Ectomycorrhizal (ECM) fungi (Hymenomycetes, Basidiomycetes,rarely Ascomycetes) form mutualistic relations with the roots of trees,improving plant nutrition, especially of nitrogen, in exchange for carbo-hydrates (Klironomos and Hart, 2001). ECM fungi are characterized byhigh proteolytic capabilities and rapid absorption of the products of pro-teolysis such as amino-acids and small peptides (Read et al., 2004) andhave a critical role in nutrient acquisition in soils of low fertility. Besidespollen (Perez-Moreno and Read, 2001a), they use collembolans(Klironomos and Hart, 2001), other ECM fungal mycelium (Anderssonet al., 1997), and nematodes (Perez-Moreno and Read, 2001b) as nutri-ent source. ECM fungi such as Suillus tomentosus, Cenococcumgeophilum,Russula sp. indeed colonize the roots of Pinus, Picea and Betula(Thormann et al., 1999) and even Vaccinium (Vohník et al., 2007) inpeatlands, albeit successful colonizationof trees inwetlands is restrictedby limited oxygen availability and low tree productivity (Wurzburgeret al., 2004).

The dark colored slow growing sterile fungi (Read, 1996) of ericoidmycorrhiza (ERM) on roots of for instance Calluna vulgaris andVaccinium is a potential pollen attacker too. ERM fungi are abundantin soils with low N availability and taxonomically belong mostly to theAscomycetes (e.g. Rhizoscyphus ericae, Hymenoscyphus ericae), but canbe Hymenomycetes (Basidiomycetes) as well (Andersen et al., 2013).ERM fungi assimilate nitrate, ammonium and phosphate ions as wellas contribute to nutrient-releasing decomposer activities by enzymesable to attack proteins (Langdale and Read, 1989), pectin (Cervoneet al., 1988), polyphenols and lignin (Haselwandter et al., 1990), cellu-lose, cellobiose, hemicellulose, and chitin (review by Read et al.,2004). For instance, in the laboratory, the mycorrhizal fungus H. ericaemobilizes nitrogen presented in sterile root and shoot material ofVaccinium macrocarpon and transfers a sufficient amount to the host,enhancing its growth (Kerley and Read, 1998). In vitro experiments

ler Moor, 8 cm); 2 Pinus pollen body and both air-sacs covered inside by hyphae (Bortsel,); 4 Pinus pollen body covered by hypha (BorstelerMoor, 26 cm); 5 Tilia pollenwith hyphainvasion at germination pore areas and several fungal cells inside the grain (BorstelerMoor,germination pores (BorstelerMoor, 4 cm); 9Alnuspollenwith hypha at germination porespossibly pseudosclerotious nature (BorstelerMoor, 4 cm); 11 Corylus pollenwith hypha atdewith hypha at pori (BorstelerMoor, 8 cm); 13 single cells of Calluna vugaris tetradefilledith fungal cells at trilete mark (Borsteler Moor, 26 and 30 cm, respectively). Scale bar is

176 L.S. Shumilovskikh et al. / Review of Palaeobotany and Palynology 221 (2015) 171–178

demonstrate that R. ericae colonize testate amoebae as Trigonopyxis andCyclopyxis and utilize their nutrients (Vohník et al., 2009). Despite thelack of laboratory experiments on pollen, the ability of ERM fungi to uti-lize such a variety of biotic sources suggest their potential ability to di-gest the contents of pollen as well.

One of the major fungal groups in peatlands are dark septate endo-phytes (DSEs) (Andersen et al., 2013). DSEs are characterized by mela-nized hyphae of conidial or sterile fungi (Deuteromycotina and FungiImperfecti) colonizing plant roots (Jumpponen and Trappe, 1998;Jumpponen, 2000). DSEs are globally distributed from the tropics to arc-tic soils. They form a heterogeneous group that functionally and ecolog-ically overlaps with saprotrophic, pathogenic and mycorrhizal fungi(Jumpponen and Trappe, 1998). Although the taxonomic position andecological role of DSE is poorly understood, studies already demonstratetheir access to e.g. cellulose, starch, xylan, protein, and ribonucleic acidsby producing corresponding enzymes (Caldwell et al., 2000), whichshould allow the use of pollen as nutrient source as well. As DSE fungiproduce microsclerotia in plant cells for their distribution (Jumpponenand Trappe, 1998), they are possibly the cause for some of the docu-mented sclerotious fillings of pollen (Plate I: 3 10, 13).

Beside ECM and ERM also arbuscular mycorrhiza (AM) colonizesplant roots in peatlands. The involved fungi areGlomales (Zygomycota).They are mainly associatedwith herbaceous species and are more com-mon in fens than in Sphagnum bogs, where potential host plants are rare(Cornwell et al., 2001). AM fungi are characterized by a superior abilityto acquire mineral P but a restricted saprotrophic potential (Andersenet al., 2013). Thus, they can here be excluded from further consider-ation. The same goes for yeasts using only simple carbohydrates andproducing no preservable hyphae (Andersen et al., 2013) andChytridiomycetes, a major fungal group able to parasitize pollen, butcharacterized by hyaline structures distinctly different from the ob-served dematiaceous hyphae (e.g. Hesmer, 1929; Shumilovskikh et al.,2015).

Therefore, saprotrophic fungi, and fungi forming DSE, ECM and ERMare identified asmajor groups verifiable ormost probably capable to at-tack pollen by dematiaceous hyphae. These fungal groups are mutuallyexclusive neither in terms of taxonomy nor in terms of ecology.

4.3. Which pollen is the “tastiest”?

Pollen spectra of studied profiles are dominated by anemophiloustaxa, which are also mostly infected by fungi (Fig. 2). Moreover, fungalhyphae also attack fern and Sphagnum spores (Plate I). Although pollenspectra of the majority of records are dominated by Alnus, Corylus andQuercus (Fig. 2), fungal attacks do not follow the pollen dominance pat-tern. In contrast, Pinus is the most and Quercus the last attacked pollentype in almost all analyzed spectra. Goldstein (1960) already reportedsuch a selective behavior for chytrids, invading the conifer pollenmore readily.

One possible reason for the different pollen infection rates could bethe seasonality of pollen input and fungal activity. For example, Stark(1972) observed high activity of DSE fungi in winter and their disap-pearance in summer, when mycorrhizal and saprotrophic fungi domi-nate. Furthermore, Alnus and Corylus release pollen in early spring(Table 1), which can be immediately attacked by active fungi, bringingnew nutrients into the ecosystems. Later on, when different nutrientsources are accessible, released pollen of, e.g. Quercus becomes less at-tractive. However, Pinus is an exception. Pinus pollen has a much highervolume (20–80 × 10−6 cm3) in comparison to Alnus (3.9), Corylus (6.7),Betula (3.2), Quercus (5.6), Ericaceae (9.0) (Roulston et al., 2000,Table 1). Although the protein content in Pinus pollen is rather low(15.7%), the pollen volume correlates positively with the mass of pro-tein and particularly of nitrogen (Roulston et al., 2000). In terms of en-ergy and nutrient balance, the attack of Pinus pollen is therefore veryvaluable. Moreover, species-specific interactions between fungal hy-phae and Pinus pollen grains cannot be excluded.

4.4. Significance of pollen as nutrient source for paleoecosystems

The majority of attacked pollen occurs in layers characterized byhigh percentages of Calluna vulgaris/Ericaceae (Fig. 2), emphasising aspecial role of ericaceous shrubs in initialing decay processes in thepeat ecosystem. On the one hand, Ericaceae are a host of ERM fungi,which might transfer nutrients from pollen rain to the host plants(Kerley and Read, 1998). On the other hand, ericaceous shrubs and par-ticularly C. vulgaris are well known as ecosystem engineers in bogs(Read et al., 2004). Heath plants form dry hummocks with high air–oxygen availability between water saturated Sphagnum hollows, con-tributing to the formation of the horizontalmosaic. The conditions in er-icaceous hummocks can boost trees and ectomycorrhiza, saprotrophsand DSE to colonize here, effecting increased decomposition and nutri-ent availability, particularly frompollen. Previous paleoecological inves-tigations in ombrotrophic bogs already proved that in phases ofSphagnum hollows the number of fungal remains is low. In contrast,peat of ericaceous hummock always shows signs of highly increasedfungal activities and decomposition (e.g. van Geel, 1978; Middeldorp,1986). Additionally, it can be explained by the fact that it takes moretime for hummock-peat (compared with hollow-peat) to become wa-terlogged and therefore for decomposition.

In raised bogs, sources of nitrogen are limited to atmospheric depo-sitions (nitrate, nitrite, ammonium and organic particles like pollen), bi-ological fixation, and animals (insectivory, droppings). Differentestimations reveal atmospheric nitrogen deposition between 0.1 and4.6 g/m2 per year equally to biological fixation (0.07–3.2 g/m2 peryear) (Middeldorp, 1986). Estimations for subarctic mires suggest thatorganic nitrogen (including pollen and spores) makes up half of thetotal atmospheric deposition (Rosswall and Granhall, 1980). Takeninto account low pre-industrial atmospheric pollution, nitrogen inputby wind-transported pollen can have played a considerable role in thepast. Applying estimations of pollen rain (Lee et al., 1996) and an ob-served pollen infection rate of up to 22% (Borsteler Moor) results in anuptake of up to 0.01 gN/m2, 0.001 gP/m2 and 0.003 gK/m2 by fungal hy-phae. Even the amounts are not enough to cover all needs of the in-volved fungi and associated hosts, deposition of easily degradableorganicmaterial such as pollen cytoplasm in a nutrient-deficit bog envi-ronment during spring could be a crucially surplus of essential elementsand energy (Stark, 1972; Lee et al., 1996; Greenfield, 1999).

In sense of ecosystem nutrient cycling, the amount of nutrient inputand therefore the competition situation in the bog depends not only onprecipitation but also verymuchon thepollination strategy and the spe-cies composition of the surrounding vegetation. Going further, onemayask how strong the mid- to late Holocene deforestation by humansaffected the nutrient budget in neighbored bogs? For example,McClymont et al. (2008) suggest the anthropogenic effect of fertilizationby increased dust from disturbed soils as a main trigger of the disap-pearance of Sphagnum imbricatum from peat bogs during the late Holo-cene. It seems plausible that in earlier times a decrease of tree polleninput due to deforestation caused traceable ecological alterations aswell. This can be tested in future paleoecological studies by taking fun-gal attacked pollen as a proxy.

Additionally, Holocene records of attacked pollen provide argu-ments for a hypothesis about reasons for wind-pollination in nutrient-limited environments like boreal forests. Anemophily is traditionallyconsidered as an inefficient process due to low pollen to ovule ratiosof N106:1 (Ackerman, 2000). Saito (2012) demonstrates that pollen isa major component of annual investment for sexual reproduction bytrees. Moreover, proportion of pollen yield from total investment bywind-pollinated trees e.g. Alnus sieboldiana (36%), Quercus serrata(28%), Pinus densiflora (24%) is much higher than the same parameterfor insect-pollinated trees e.g. Aesculus turbinata (3.8%) or Cinnamomumcamphora (5%) (Saito, 2012). From our point of view, energy and nutri-ents of the last year stored over winter in the wood rays and primaryinvested into pollen production effectively support symbiotic partners

177L.S. Shumilovskikh et al. / Review of Palaeobotany and Palynology 221 (2015) 171–178

during the nutrient-deficit period of the next spring. In terms ofecosystem nutrient balance, pollen rain is not simple waste but translo-cation of nutrients within the ecosystem. Such interactions might haveinfluenced the course of evolution in wind-pollinated trees (Perez-Moreno and Read, 2001a) and the establishment of modern forestsystems.

5. Conclusions and perspectives

Palynological records from Holocene raised peat bogs document thegeneral presence of pollen attacked by fungal hyphae. Fungal hyphaepenetrate pollen grains through the least protected apertures and cancover the inner side of the grains and even develop microsclerotia. Wepropose that attacks on pollen occur at the time of pollen depositionand can be performed by saprotrophic, ERM, ECM and DSE fungi. Thefact that large amounts of attacked pollen are found during Callunavulgaris phases suggest a special role of Ericaceae shrubs, which hostERM fungi and form relative dry, upraised hummocks in the bog withincreased fungal activity and nutrient release.

In the paleoecological context, pollen as nutrient source has beensuggested for the first time here, to our best knowledge. However, ourown observations on recent surface samples and discussions with col-leagues at the Palynological School in Moscow (April 2014) and theNPP-Workshop in Tallin (June 2014) reveal that pollen grains penetrat-ed by fungal hyphae are a common feature of palynologicalmaterial, butmostly neglected to its unknown meaning until now. Therefore, one ofthemain purposes of thiswork is to attain specialists of different ecolog-ical and paleoecological disciplines to this phenomenon. For example,there is an urgent need in further mycological experiments testing abil-ity of ERM and DSE fungi for the pollen attack suggested here, and eval-uation of the role of these fungi in decomposition processes in peat bogs.Proper documentation of attacked pollen by palynologists will allowtime–space correlations to determine the role of pollen nutrients onbog and forest developments and therefore of important carbon sinks.

Acknowledgments

We thank Shinya Sugita for fruitful discussions, Bas van Geel, ananonymous reviewer and the editor for improving the manuscript andLaura Sutcliffe for polishing the English language. The palynologicalwork was supported by DFG projects LE 1805 and HA 4439 as well asthe grant in accordance with Resolution of the Government ofthe Russian Federation № 220 dated April 09, 2010, under Agreement№ 14.В25.31.0001 with the Ministry of Education and Science of theRussian Federation dated June 24, 2013 (BIO-GEO-CLIM).

References

Achterberg, I., 2010. Dendroökologisch-vegetationsgeschictliche Untersuchung anEichen- und Kiefernhorizonten aus dem Toten Moor bei Hannover (Diplom thesis,Göttingen, Göttingen in German).

Ackerman, J.D., 2000. Abiotic pollen and pollination: ecological, functional, and evolution-ary perspectives. Plant Syst. Evol. 222, 167–185.

Andersen, R., Chapman, S.J., Artz, R.R.E., 2013. Microbial communities in natural and dis-turbed peatlands: a review. Soil Biol. Biochem. 57, 979–994.

Andersson, S., Ek, H., Söderström, B., 1997. Effects of liming on the uptake of organic andinorganic nitrogen by mycorrhizal (Paxillus involutus) and non-mycorrhizal Pinussylvestris plants. New Phytol. 135, 763–771.

Barron, G.L., 1988. Microcolonies of bacteria as a nutrient source for lignicolous and otherfungi. Can. J. Bot. 66, 2505–2510.

Barron, G.L., 1992. Lignolytic and cellulolytic fungi as predators and parasites. In: Carrol,G.C., Wicklow, D.T. (Eds.), The Fungal Community: Its Organization and Role in theEcosystem. Marcel Dekker, New York, pp. 311–326.

Barron, G.L., Dierkes, Y., 1977. Nematofagous fungi: Hohenbuehelia, the perfect state ofNematoctonus. Can. J. Bot. 55, 3054–3062.

Bauerochse, A., Niemuth, A., Jantz, N., Shumilovskikh, L., Metzler, A., 2015. Archäologischeund paläobotanische Untersuchungen im Darlaten Moor. In: Bauerochse, A.,Hassmann, H., Püschel, K., Schultz, M. (Eds.), “Moora”, das Mädchen aus demUchter Moor. Eine Moorleiche der Eisenzeit aus Niedersachsen II. – Materialheftezur Ur- und Frühgeschichte Niedersachsens (in press, in German).

Caldwell, B.A., Jumpponen, A., Trappe, J.M., 2000. Utilization of major detrital substratesby dark-septate, root endophytes. Mycologia 92, 230–232.

Cervone, F., Castoria, R., Spanu, P., Bonfante-Fasolo, P., 1988. Pectinolytic activity in someericoid mycorrhizal fungi. Trans. Br. Mycol. Soc. 91, 537–539.

Cornwell, W.K., Bedford, B.L., Chapin, C.T., 2001. Occurrence of arbuscular mycorrhizalfungi in a phosphorus-poor wetland and mycorrhizal response to phosphorusfertilisation. Am. J. Bot. 88, 1824–1829.

Eckstein, J., Leuschner, H.H., Giesecke, T., Shumilovskikh, L., Bauerochse, A., 2010.Dendroecological investigations at Venner Moor (northwest Germany) documentclimate-driven woodland dynamics and mire development in the period 2450–2050 BC. The Holocene 20, 231–244.

Erdtman, G., 1943. An Introduction to Pollen Analysis. Chronica Botanica, Waltham.Fægri, K., 1971. The preservation of sporopollenin membranes under natural conditions.

In: Brooks, J., Grant, P.R., Muir, M., van Gijzel, P., Shaw, G. (Eds.), Sporopollenin. Aca-demic Press, London, pp. 256–272.

Fries, N., Swedjemark, G., 1985. Sporophagy in hymenomycetes. Exp. Mycol. 9, 74–79.Goldstein, S., 1960. Degradation of pollen by phycomycetes. Ecology 41, 543–545.Greenfield, L.G., 1999.Weight loss and release ofmineral nitrogen from decomposing pol-

len. Soil Biol. Biochem. 31, 353–361.Griffith, N.T., Barnett, H.L., 1967. Mycoparasitism by basidiomycetes in culture. Mycologia

59, 149–154.Haselwandter, K., Bobleter, O., Read, D.J., 1990. Degradation of 14C-labelled lignin and

dehygropolymer of coniferyl alcohol by ericoid and ectomycorrhizal fungi. Arch.Microbiol. 153, 352–354.

Havinga, A.J., 1967. Palynology and pollen preservation. Rev. Palaeobot. Palynol. 2, 81–98.Heslop-Harrison, J., 1971. Sporopollenin in the biological context. In: Brooks, J., Grant, P.R.,

Muir, M., van Gijzel, P., Shaw, G. (Eds.), Sporopollenin. Academic Press, London,pp. 1–30.

Hesmer, H., 1929. Mikrofossilien in Torfen. Paläontol. Z. 11, 245–257 (in German).Hutchison, L.J., Barron, G.L., 1996. Parasitism of yeasts by lignicolous Basidiomycota and

other fungi. Can. J. Bot. 74, 735–742.Hutchison, L.J., Barron, G.L., 1997a. Parasitism of pollen as a nutritional source for

lignicolous Basidiomycota and other fungi. Mycol. Res. 101, 191–194.Hutchison, L.J., Barron, G.L., 1997b. Parasitism of algae by lignicolous Basidiomycota and

other fungi. Can. J. Bot. 75, 1006–1011.Jäger, E.J., 2011. Exkursionsflora von Deutschland. Gefäßpflanzen: Grundband. Springer

Spektrum Akademischer Verlag, Heidelberg (in German).Jumpponen, A., 2000. Dark septate endophytes – are they mycorrhizal? Mycorrhiza 11,

207–211.Jumpponen, A., Trappe, J.M., 1998. Dark septate endophytes: a review of facultative

biotrophic root-colonizing fungi. New Phytol. 140, 295–310.Kerley, S.J., Read, D.J., 1998. The biology of mycorrhiza in the Ericaceae. New Phytol. 139,

353–360.Klironomos, J.N., Hart, M.N., 2001. Food-web dynamics: animal nitrogen swap for plant

carbon. Nature 410, 651–652.Knight, A.H., Crooke, W.M., Shepherd, H., 1972. Chemical composition of pollen with par-

ticular reference to cation exchange capacity and uronic acid content. J. Sci. FoodAgric. 23, 263–274.

Langdale, A.R., Read, D.J., 1989. Substrate decomposition and product release by ericoidand ectomycorrhizal fungi grown on protein. Agric. Ecosyst. Environ. 28, 285–291.

Lee, E.L., Kenkel, N., Booth, T., 1996. Atmospheric deposition of macronutrients by pollenin the boreal forest. Ecoscience 3, 304–309.

McClymont, E.L., Mauquoy, D., Yeloff, D., Broekens, P., van Geel, B., Pancost, R.D.,Chambers, F.M., Evershd, R.P., 2008. The disappearance of Sphagnum imbricatumfrom Butterburn Flow, U.K. The Holocene 18, 991–1002.

Michelsen, A., Quarmby, C., Sleep, D., Jonasson, S., 1998. Vascular plant 15N natural abun-dance in heath and forest tundra ecosystems is closely correlated with presence andtype of mycorrhizal fungi in roots. Oecologia 115, 406–418.

Middeldorp, A.A., 1986. Functional palaeoecology of the Hahnenmoor raised bog ecosys-tem— a study of vegetation history, production and decomposition by means of pol-len density dating. Rev. Palaeobot. Palynol. 49, 1–73.

Perez-Moreno, J., Read, D.J., 2001a. Exploitation of pollen by mycorrhizal mycelia systemswith special reference to nutrient recycling in boreal forests. Proc. R. Soc. B 268,1329–1335.

Perez-Moreno, J., Read, D.J., 2001b. Nutrient transfer from soil nematodes to plants: a di-rect pathway provided by the mycorrhizal mycelial network. Plant Cell Environ. 24,1219–1226.

Punt, W., Hoen, P.P., Blackmore, S., Nilsson, S., le Thomas, A., 2007. Glossary of pollen andspore terminology. Rev. Palaeobot. Palynol. 143, 1–81.

Read, D.J., 1996. The structure and function of the ericoid mycorrhizal root. Ann. Bot. 77,365–374.

Read, D.J., Leake, J.R., Perez-Moreno, J., 2004. Mycorrhizal fungi as drivers of ecosystemprocesses in heathland and boreal forest biomes. Can. J. Bot. 82, 1243–1263.

Rosswall, T., Granhall, U., 1980. Nitrogen cycling in a subarctic ombrotrophic mire. Ecol.Bull. 30, 209–234.

Roulston, T.H., Cane, J.H., Buchmann, S.L., 2000. What governs protein content of pollen:pollinator preferences, pollen–pistil interactions, or phylogeny? Ecol. Monogr. 70,617–643.

Saito, H., 2012. Biomass Allocation in Forest to Pollen and Seed Production. Research andData on Biomass Production and Sexual Reproduction of Trees. Ito Press, Japan(in Japanese).

Saito, H., Takeoka, M., 1985. Pollen production rates in a young Japanese red pine forest.Jpn. J. Ecol. 35, 67–76.

Shumilovskikh, L.S., Schlütz, F., Achterberg, I., Bauerochse, A., Leuschner, H.H., 2015. Non-pollen palynomorphs from mid-Holocene peat of the raised bog Borsteler Moor(Lower Saxony, Germany). Stud. Q. 32, 5–18.

178 L.S. Shumilovskikh et al. / Review of Palaeobotany and Palynology 221 (2015) 171–178

Stark, N., 1972. Nutrient cycling pathways and litter fungi. Bioscience 22, 355–360.Taylor, L.P., Hepler, P.K., 1997. Pollen germination and tube growth. Annu. Rev. Plant

Physiol. Plant Mol. Biol. 48, 461–491.Thormann, M.N., Currah, R.S., Bayley, S.E., 1999. The mycorrhizal status of the dominant

vegetation along a peatland gradient in southern boreal Alberta, Canada. Wetlands19, 438–450.

Van Geel, B., 1978. A palaeoecological study of Holocene peat bog sections in Germanyand the Netherlands. Rev. Palaeobot. Palynol. 25, 1–120.

Vohník, M., Fendrych, M., Albrechtová, J., Vosátka, M., 2007. Intracellular colonization ofRhododendron and Vaccinium roots by Cenococcum geophilum. Geomyces pannorumand Meliniomyces variabilis 52. Folia Microbiologica, pp. 407–414.

Vohník, M., Burdíková, Z., Albrechtová, J., Vosátka, M., 2009. Testate amoebae (Arcellinidaand Euglyphida) vs. ericoidmycorrhizal and DSE fungi: a possible novel interaction inthe mycorrhizosphere of ericaceous plants? Microb. Ecol. 57, 203–214.

Vohník, M., Burdíková, Z., Vyhnal, A., Koukol, O., 2011. Interactions between testate amoe-bae and saprotrophic microfungi in a Scots Pine litter microcosm. Microb. Ecol. 61,660–668.

Wilkinson, D.M., 2008. Testate amoebae and nutrient cycling: peering into the black boxof soil ecology. Trends Ecol. Evol. 23, 596–599.

Wurzburger, N., Hartshorn, A.S., Hendrick, R.L., 2004. Ectomycorrhizal fungal communitystructure across a bog-forest ecotone in southeastern Alaska. Mycorrhiza 14,383–389.