www.nature.com/ismej
ARTICLE OPEN
Seasonality of parasitic and saprotrophic zoosporic fungi:
linking sequence data to ecological traits
Silke Van den Wyngaert 1,10✉, Lars Ganzert 1,2,11, Kensuke Seto3,4, Keilor Rojas-Jimenez 5, Ramsy Agha 6, Stella A. Berger 1,
Jason Woodhouse1, Judit Padisak 7, Christian Wurzbacher8, Maiko Kagami 3✉ and Hans-Peter Grossart 1,9✉
© The Author(s) 2022
Zoosporic fungi of the phylum Chytridiomycota (chytrids) regularly dominate pelagic fungal communities in freshwater and marine
environments. Their lifestyles range from obligate parasites to saprophytes. Yet, linking the scarce available sequence data to
specific ecological traits or their host ranges constitutes currently a major challenge. We combined 28 S rRNA gene amplicon
sequencing with targeted isolation and sequencing approaches, along with cross-infection assays and analysis of chytrid infection
prevalence to obtain new insights into chytrid diversity, ecology, and seasonal dynamics in a temperate lake. Parasitic
phytoplankton-chytrid and saprotrophic pollen-chytrid interactions made up the majority of zoosporic fungal reads. We explicitly
demonstrate the recurrent dominance of parasitic chytrids during frequent diatom blooms and saprotrophic chytrids during pollen
rains. Distinct temporal dynamics of diatom-specific parasitic clades suggest mechanisms of coexistence based on niche
differentiation and competitive strategies. The molecular and ecological information on chytrids generated in this study will aid
further exploration of their spatial and temporal distribution patterns worldwide. To fully exploit the power of environmental
sequencing for studies on chytrid ecology and evolution, we emphasize the need to intensify current isolation efforts of chytrids
and integrate taxonomic and autecological data into long-term studies and experiments.
The ISME Journal; https://doi.org/10.1038/s41396-022-01267-y
INTRODUCTION the winner” [18]), chytrid parasites maintain and promote genetic
Recent advances in sequencing technologies have revealed that diversity in phytoplankton populations [19, 20]. Furthermore,
fungi are ubiquitous and highly diverse in aquatic ecosystems chytrids efficiently siphon carbon and nitrogen from the photo-
[1, 2]. Yet, a substantial fraction of aquatic “dark matter” fungi, synthetic host, bypassing the microbial loop (i.e., fungal shunt
especially the early diverging lineages, has not been described [3]. [21]), which is further transferred to zooplankton through the
Zoosporic fungi of the phylum Chytridiomycota (chytrids), consumption of chytrid zoospores (i.e., mycoloop [22]). By this,
regularly dominate pelagic communities in freshwater and marine chytrids modify microbial interactions, enhance herbivory [23–25]
environments [4–6]. Chytrids encompass a wide range of taxa with and accelerate carbon transfer to higher trophic levels in pelagic
a continuum of consumer strategies spanning from strict food webs.
saprotrophs to obligate parasites [7, 8]. As such, chytrids are Despite recent advances, we are still far from comprehensively
decomposers of autochthonous and allochthonous organic matter characterizing the phylogenetic and ecological diversity of chytrids.
such as zooplankton exuviae and pollen grains [9] and lethal Although 18S and 28S rRNA gene sequencing approaches have
parasites of phytoplankton [8, 10, 11]. The integration of chytrids been applied to unearth chytrid diversity [26, 27], our current
in the PEG (plankton ecology group) model [12, 13] exemplifies knowledge on the diversity, especially of phytoplankton parasites, is
the emerging recognition of chytrids as ecological and evolu- almost exclusively based on >100 years of morphology-based
tionary drivers of phytoplankton bloom dynamics. Chytrids can identification [7]. The scarcity of reference chytrid sequences in
suppress the development of phytoplankton blooms [14–16], databases creates difficulties in linking chytrid sequences to specific
selective chytrid parasitism can alter interspecific competition, ecological traits or their host ranges.
affecting phytoplankton coexistence and succession [14, 17] and, To overcome these limitations, we aimed at improving the
by imposing negative frequency-dependent selection (e.g., “killing linkage between chytrid sequence diversity and consumer-
1Department of Plankton and Microbial Ecology, Leibniz Institute of Freshwater Ecology and Inland Fisheries (IGB), Zur alten Fischerhütte 2, 16775 Stechlin, Germany. 2GFZ
German Research Centre for Geosciences, Section Geomicrobiology, Telegrafenberg, 14473 Potsdam, Germany. 3Faculty of Environment and Information Sciences, Yokohama
National University, Tokiwadai 79-7, Hodogayaku, Yokohama, Kanagawa 240-8501, Japan. 4Department of Ecology and Evolutionary Biology, University of Michigan, Ann Arbor
48109 MI, USA. 5Escuela de Biología, Universidad de Costa Rica, 11501 San José, Costa Rica. 6Department of Ecosystem Research, Leibniz Institute of Freshwater Ecology and
Inland Fisheries (IGB), Müggelseedamm 301, 12587 Berlin, Germany. 7Research Group of Limnology, Centre of Natural Sciences, University of Pannonia, Egyetem u. 10, 8200
Veszprém, Hungary. 8Chair of Urban Water Systems Engineering, Technical University of Munich, Am Coulombwall 3, 85748 Garching, Germany. 9Institute of Biochemistry and
Biology, Potsdam University, Maulbeerallee 2, 14469 Potsdam, Germany. 10Present address: Department of Biology, University of Turku, Vesilinnantie 5, 20014 Turku, Finland.
11Present address: Marbio, UiT- The Arctic University of Norway, Sykehusveien 23, 9019 Tromsø, Norway. ✉email: silke.vandenwyngaert@utu.fi; kagami-maiko-bd@ynu.ac.jp;
hgrossart@igb-berlin.de
Received: 20 December 2021 Revised: 28 May 2022 Accepted: 7 June 2022
1234567890();,:
S. Van den Wyngaert et al.
2
resource interactions by studying their seasonal dynamics in a For establishing chytrid cultures, a similar procedure was used, where
well-studied lake ecosystem. We combined isolation approaches after washing, single phytoplankton cells with attached sporangia were
including 1) direct cultivation, 2) single-cell isolation, and 3) in situ transferred into wells of a 24-well plate containing each 1mL of CHU-10
baiting to target phytoplankton parasites and saprotrophic pollen- medium of a phytoplankton host culture or pollen suspension. After
degrading chytrids. Cultivation enables detailed morphological successful infection, phytoplankton-chytrid co-cultures were established
and molecular studies on all chytrid life stages, while experimental and maintained as previously described [34]. Saprotrophic chytrids isolated
from pollen were transferred and maintained in liquid mPmTG medium
cross-infection assays provide insights into their host range and [35] (for cultivation details see Supplementary Text S3).
specificity [28, 29]. Yet, cultivation is difficult and time-consuming,
which can arguably underestimate diversity because not all DNA extraction and sequencing: DNA of single infected cells was
chytrids can grow under the given laboratory conditions and extracted using the Hot-SHOT extraction method [30] or Illustra Single Cell
phytoplankton hosts available. This limitation can be partially GenomiPhi DNA amplification kit (GE-Healthcare). DNA of culture isolates
overcome by single-cell isolation, i.e., manual isolation and was extracted from zoospores (separated from host cells by filtration
subsequent sequencing of single infected phytoplankton colo- through a 10 µm nylon mesh) or from host chytrid co-cultures using the
nies/cells or pollen grains [28, 30]. For higher throughput and a peqGOLD Tissue DNA Mini Kit (Peqlab Biotechnology GmbH, Germany) or
greater coverage of diversity compared to manual cell picking, we Hot-SHOT extraction method [36]. The 28S and 18S rRNA genes of chytrids
applied an in situ baiting approach combined with amplicon were amplified with primers LROR-LR5 [37, 38] and NS1-NS4 [39] or EF4-
sequencing to target and amplify chytrids associated with pollen. EF3 [40], using MyTaq Red DNA Polymerase as previously described in [29],
and sequenced by Macrogen Europe. Sequences were quality-controlled
These targeted approaches allowed us to establish a taxonomic and assembled using BioEdit [41]. Additionally, the rRNA operon of single
and ecological annotated library compiling information on cells was sequenced using Oxford Nanopore sequencing with primer pair
sequence, morphology, and host/substrate ranges. We applied NS1short and RCA95m, as described in [42] and 18 S rRNA and 28 S rRNA
this reference library to an amplicon-based high-throughput genes of single-cell Dolichospermum-MDA2-akinete were retrieved from
sequencing (HTS) dataset from the freshwater Lake Stechlin with shotgun metagenome sequencing (Supplementary Text S4, Willis et al. in
the objective to i) estimate the contribution of Chytridiomycota revision). DNA extraction and sequencing methods for each isolate/single
phytoplankton parasites and pollen-degrading saprotrophs to the cell are listed in Supplementary Table S3.
total pelagic zoosporic fungal community, and ii) assess their
diversity and seasonal dynamics in relation to host association and Phylogenetic analysis of culture and single-cell isolates
inferred lifestyle. By synergizing state-of-the-art methods with For phylogenetic analysis, we created datasets of 18S and 28S rRNA gene
chytrid infection prevalence data, we provide new insights into sequences containing environmental sequences of uncultured chytrids
chytrid diversity, ecology, and seasonal dynamics in a related to culture and isolate sequences. Salpingoeca infusionum and
temperate lake. Monosiga brevicollis (Choanozoa) and Nuclearia simplex (Cristidiscoidea)
were selected as outgroup taxa. Sequences were automatically aligned
with MAFFT v. 7.475 [43], independently for each gene region.
Ambiguously aligned regions were excluded using trimAl v. 1.2 [44] with
MATERIALS AND METHODS a gappyout model. A concatenated alignment was generated and
A schematic overview of the workflow is presented in partitioned by genes for analysis with maximum likelihood (ML) methods.
figure S1 The ML tree was inferred using RAxML v. 8.2.12 [45] on Cipres Science
Lake sampling. The sampling period spanned 15 months from March Gateway [46]. For further details see Supplementary Table S8 and
2015 to June 2016 in temperate, dimictic, and mesotrophic Lake Stechlin, Supplementary text S5.
Germany [31], [Supplementary Text S1, Fig. S1, Supplementary Table S1].
Two integrated water samples of the upper mixed water layer (6–14m)
were taken weekly or bi-weekly (except in August 2015 and February 2016) Evaluation of chytrid host range and consumer strategy
with a hose (5 cm diameter) or an integrating water sampler (HYDRO-BIOS To examine the host range of parasitic chytrid strains, cross-infection
IWS III, Kiel). assays were performed as described in [29]. Briefly, 0.5 mL of zoospore
For environmental DNA extraction, volumes of 0.5–1 L of lake water were suspensions (after filtration of a 7 days old, infected culture through a 10
ltered onto 5 µm pore size polycarbonate lters (47mm diameter, Merck µm plankton mesh) were added to 1mL of exponentially growing
fi fi
Millipore, Germany) to enrich particle-associated fungi. All lters were stored phytoplankton host. The original chytrid host strain served as a reference.
fi
in cryotubes at−80 °C until further processing. One integrated water sample Each assay was performed in triplicates using 24-well plates. Visual
(6 L) was gently concentrated in situ by a 25 µm-plankton net underwater inspection of the infection was performed by inverted light microscopy
and subsequently pre- ltered over a 280 µm sized mesh to remove (Nikon Eclipse TS100). Cross-infection results of the following chytrid
fi
mesozooplankton. A subsample of 50mL served to screen for chytrid cultures have been published in previous studies: isolates STAU-CHY3 [34],
infections on phytoplankton, subsequent cultivation, and single-cell isola- SVdW-EUD1, SVdW-EUD2, SVdW-EUD3 [29], and SVdW-SYN-CHY1 [47]. In
tion. The rest (50mL) was xed with alkaline Lugol´s solution and stored at this study, two additional diatom parasite strains, Fragilaria-CHY1 and AST-
fi
4 °C for quantifying the the percentage of a host population infected by CHY1, were evaluated for their infection potential on eight different host
chytrids [32]. Chytrid sporangia were visualized using a dual staining species, including five diatoms (Fragilaria crotonensis, Ulnaria sp. (former
protocol with Calco uor White (CFW) and Wheat Germ Agglutinin, Synedra sp. [47]), Asterionella formosa, Aulacoseira ambigua, Aulacoseira
fl
conjugated to Alexa Fluor 488 (WGA) [33]. Whenever possible, 300 granulata) and three green algae (Yamagishiella unicocca, Eudorina elegans,
individuals of each phytoplankton species with visible chytrid infection Staurastrum sp.). Chytrid isolates Staurastrum-CHY4 and Staurastrum-CHY5
were counted by using an inverted microscope (Nikon eclipse Ti2, 400X, were tested on three desmid species (Staurastrum sp., Closterium sp.,
uorescence channels CFW: 387/11 nm excitation and 442/46 nm emission, Cosmarium sp.). All parasitic strains were tested for their saprotrophic
fl
WGA: 482/35 nm excitation and 536/40 nm emission). In cases of low growth capability on pine pollen grains and artificial mPmTG medium [35].
phytoplankton abundance, the whole Utermöhl counting chamber was All host-chytrid associations identified from single-cell data and cross
screened. Samples for phytoplankton biomass quanti cation were collected infection assays were represented in an association matrix using the vegan
fi
separately as part of a routine monitoring program with bi-weekly or package in R (Fig. 3). Based on the cross-infection results we categorized a
monthly intervals (Supplementary Text S2, Supplementary Table S2). chytrid species as 1) “specialist parasite” when it infected solely a single
phytoplankton species, 2) “generalist parasite” when it was associated with
more than one phytoplankton species and 3) “facultative parasite” when it
Single-cell isolation and cultivation of phytoplankton and pollen-associated was found in association with both phytoplankton and pollen, and/or was
chytrids. Individual, infected phytoplankton cells and pollen grains were capable of growth on mPmTG medium or senescent phytoplankton.
picked using a 0.5–10 µm micropipette under an inverted light microscope
(Nikon Eclipse TS100, 100X). Picked single cells were transferred and
washed thrice in 0.2 µm filtered MilliQ water before being transferred into Field experiment: in situ pollen baiting
0.5 mL PCR tubes (total volume: 1 µL of 0.2 µm filtered MilliQ water) and Pollen was collected on a dry and canopied surface close to Lake Stechlin
stored at −20 °C until further processing. in spring 2015. Most pollen were from Pinus sylvestris, but also birch and
The ISME Journal
S. Van den Wyngaert et al.
3
beech trees. A mixed pollen solution was prepared by adding 200mg distributed) for alpha diversity and PERMANOVA [55] for beta diversity.
pollen in 750mL sterile MilliQ water (0.27 g L−1). Thirty-five mL of this The correlation between chytrid ASV47 (parasite on diatom Fragilaria) and
solution was transferred to custom-made baiting chambers and incubated putative hyperparasite Rozellomycota ASV141 (Fig. S8) was determined by
just below the surface in Lake Stechlin for 1 week (15th to 22nd May 2015) calculating Pearson correlation coefficient using R.
at 4 locations: 1) littoral zone macrophyte area, 2) littoral zone reed stand,
3) littoral zone above sandy sediment, and 4) pelagic zone. Three replicates
were deployed at each littoral and six at the pelagic site, yielding RESULTS
15 samples in total. After incubation, pollen was rinsed to remove non- Diversity of cultured isolates and single cells/colonies
attached organisms and re-suspended in 40mL of 0.2 µm filtered lake
water. Twenty mL of pollen solution were ltered onto 5 µm pore size In total, 18 chytrid strains were isolated and 157 single-infected
fi
polycarbonate filters (47mm diameter, Merck Millipore), plunged into (host-chytrid) cells/colonies were collected between 2015–2017.
liquid nitrogen, and stored at −80 °C until further processing. More details Good quality sequences from single cells were obtained from
on the set-up and handling are given in Supplementary Text S6, Fig. S2). 31 samples. This resulted in a reference library of 22 unique partial
LSU sequences, of which 19 were associated with 14 phytoplank-
DNA extraction and sequence data analysis of lake and in situ ton host species and 3 with pollen (Table 1). All sequences
pollen baiting samples obtained by cultivation or single-cell isolation belonged to the
Genomic DNA was extracted using a CTAB-phenol-chloroform-isoamyl phylum Chytridiomycota, except two zoosporic incertae sedis
alcohol/bead-beating protocol (modified after [48], Supplementary fungi, which were associated with akinetes and vegetative cells of
Text S7). PCR, library preparation, and sequencing were performed by Dolichospermum spp. cyanobacteria (Fig. 1).
LGC Genomics (Berlin, Germany). Briefly, the D1 region of the LSU was Chytrid strains represented five species that have been
amplified using forward primer ITS4ngsF (5’-GCATATCAATAAGCGSAGGA- identified or newly described as Staurastromyces oculus (Rhizo-
3’) and reverse primer LF402R (5’-TTCCCTTTYARCAATTTCAC-3’) (modified phydiales) [34], Endocoenobium eudorinae (Polyphagales), Dangar-
after [49]), followed by library preparation and sequencing (2 × 300 bp) on dia mamillata (incertae sedis), Algomyces stechlinensis
a MiSeq (Illumina) platform. A total of 42 lake samples and 15 pollen-bait (Lobulomycetales) [29], Zygophlyctis planktonica (Zygophlycti-
samples were sequenced. Demultiplexed raw sequence data was quality
checked and analyzed using the DADA2 package [50] in R using default dales) [47]. Strains Staurastrum-CHY4 (Rhizophydiales) and
parameters (maxN= 0, maxEE= 2, truncQ= 2), generating sequences of Pollen-CHY1 (Rhizophydiales), were identified as known species
about 350 nt. Protrudomyces lateralis and Globomyces pollinis-pini, respectively
To analyze the fraction of zoosporic fungal diversity identified using the [56]. The remaining strains represent yet undescribed taxa. Strain
targeted cultivation-dependent and -independent approaches, all gener- Fragilaria-CHY1 (Lobulomycetales), parasitic on the diatom Fragi-
ated field ASVs (amplicon sequence variants) were compared against all laria crotonensis, together with single-cell sequences retrieved
sequences obtained from culture strains, single cells, and pollen-baiting from the diatom Fragilaria showed a close affiliation to
experiment. Additionally, sequences and ASVs were compared to the NCBI Zygorhizidium affluens, a known parasite of the diatom Asterionella
nt database release 246: October 15 2021. A Lake Stechlin ASV was formosa [57]. Strain AST-CHY1, parasitic on Asterionella formosa,
considered identical to a sequence generated in this study or from the
NCBI nt database when reaching a sequence similarity of 99% and a was placed within the novel clade RH-1 together with single-cell
≥
minimum sequence coverage of 85%. sequences from other diatom parasites, related to Alphamyceta-
ASVs from in situ pollen baiting were taxonomically assigned by ceae and Kappamycetaceae (Fig. 1). Strain Fragilaria-B6 was
manually searching the NCBI nt database using BLAST (BLAST+ v2.10.0) isolated from a single-infected diatom cell belonging to Stepha-
(Supplementary Table S4). Initial taxonomic assignment of Lake Stechlin nodiscus, but could be maintained in the lab on senescent
ASVs was done using the SILVA Online classifier with the LSU database Fragilaria and Ulnaria diatoms. Its partial LSU sequence was
v138 [51] (Supplementary Table S5). Fungal assignment followed the identical to that of the single-cell sample Stephanodiscus-MDA04,
criteria given by [49] for the LSU D1 barcode. When sequence similarity of forming a novel clade together with two uncultured clones from
fungal ASVs assigned to one of the zoosporic fungal lineages Chytridio- oxygen-depleted marine sediment and paddy field soil (RH-2),
mycota, Blastocladiomycota, Aphelidiomycota and Rozellomycota was
lower than 85% to a reference sequence, the ASV was manually veri ed by related to Halomycetaceae, within the Rhizophydiales. We
fi
searching the NCBI nt database using BLASTn. Only ASVs with an 80% identified another novel clade RH-3 within Rhizophydiales
sequence similarity and 85% query coverage of a zoosporic fungal including single-cell sequences of desmid parasites and sapro-
sequence in the NCBI nt database were treated as “zoosporic fungi”. Final trophs on pollen. Single cell isolate Staurastrum-MDAExp11 fell in
taxonomic verification and sequence affiliation of zoosporic fungal ASVs the clade CH-D sensu Kagami et al. (2020) [28]. Parasites of diatoms
was based on a phylogenetic approach (Supplementary Text S8, Figure S9). Cyclotella and Diatoma represented new species within the order
The extracted ASV abundance matrix of zoosporic fungi (including three Zygophlyctidales (Fig. 1).
unclassified ASVs that matched with the sequences from Dolichospermum
spp. attached chytrids obtained in this study) was imported into R for Host/substrate specificity. Five strains were classified as specialist
further analysis (Supplementary Table S6). All sequence reads are available
in the NCBI Sequence Read Archive (SRA) under BioProject PRJNA682007. parasites (i.e., infecting only one host), two strains as generalist
Sequences from strains and single cell isolates were deposited under parasites, and three strains as facultative parasites (Fig. 2). The
accession no. OL869010-OL869016; OM859415-OM859422 (18S Sanger), generalist parasite Algomyces stechlinensis had the most extensive
OL868971-OL869009 (28S Sanger), OL869133 (28S shotgun metagenome), host range, including two members of Chlorophyta and one
OL869110 (18S shotgun metagenome), OL869111-OL869121 (Nanopore). desmid. The desmid Staurastrum sp. displayed the highest
diversity of associated chytrids (four species).
Statistical analysis
Statistical analyses have been carried out using PASTv3.25 [52], unless Fungal community associated with pollen: pollen-baits
stated otherwise. We detected 51 fungal ASVs in the in situ pollen bait experiment,
Alpha diversity measures and principal coordinates analysis (PCoA) were the majority were assigned to Chytridiomycota (75%), followed by
calculated on a subsampled dataset including ASVs belonging to zoosporic Ascomycota (10%), Rozellomycota (6%), Blastocladiomycota (2%),
fungi. All environmental samples were subsampled to 1000 sequences and Mucoromycota (2%) (Supplementary Table S4). Within
because of the high variation in sequencing depth. Subsampling was done Chytridiomycota, members of Rhizophydiales were most abun-
using the ‘rrarefy’ function in the vegan package v2.5–7 [53] in R 3.6 [54], dant (95%). The ten most abundant ASVs represented 98% of the
and samples with fewer reads were removed. The rarefied ASV table was
Hellinger-transformed, and Bray-Curtis dissimilarities were used for PCoA sequences, nine of which belonged to Rhizophydiales and one to
analysis. Samples were sorted into seasons according to the meteorolo- Rozellomycota. The most abundant ASV matched with a single cell
gical calendar. Differences between seasons were analyzed with ANOVA sequence of Pollen-MDA36. None of the top ten most abundant
for normally distributed data (Kruskal-Wallis when non-normally ASVs matched with any described facultative or saprotrophic
The ISME Journal
S. Van den Wyngaert et al.
4
The ISME Journal
Table 1. Annotated chytrid reference sequences originating from cultivation and single-cell isolation, obtained from Lake Stechlin.
ID Chytrid ID Isolation date Original host species Chytrid morphology Chytrid phylogeny Accession no. 18 S/28 S/
isolated nanopore (18S-ITS-28S)
1 Cyclotella-MDA01 6 Apr 2016 Cyclotella sp. not determinable Zygophlyctidales sp.1 OL869011/OL868972/
OL869112
2 Asterionella-MDA57 22 Jun 2016 Asterionella formosa Zygophlyctis asterionellae1 Zygophlyctidales; Zygophlyctis OM859421/-/OL869111
asterionellae1
3 Diatoma-MDA07 20 Apr 2016 Diatoma tenuis not determinable Zygophlyctidales; OL869016/OL868990/
Diatoma-MDA15 27 Apr 2016 Zygophlyctidales sp.2 OL869117
Diatoma-MDA24 4 May 2016 OM859415/OL868991/-
Diatoma-MDA30 11 May 2016 OM859416/OL868992/-
OM859417/-/-
4 Stephanodiscus-MDA04 30 Mar 2016 Stephanodiscus sp. Podochytrium cornutum2 Rhizophydiales sp.1 -/OL868985/OL869115
Stephanodiscus-MDA05 30 Mar 2016 -/OL868986/-
Fragilaria-B6 13 Apr 2016 OL869014/OL868983/-
Synedra-A1 13 Apr 2016 -/OL868984/-
5 STAU-CHY33 25 Jul 2015 Staurastrum sp. Staurastromyces oculus Rhizophydiales; Staurastromyces KY350147/KY350145/-
Staurastrum-MDAExp2 28 Jun 2016 oculus OM859418/OL868999/
Staurastrum-CHYA2 28 Jun 2016 OL869119
Staurastrum-CHYB1 28 Jun 2016 -/KY555729/-
Staurastrum-CHYC1 28 Jun 2016 -/OL868997/-
-/OL868998/-
6 AST-CHY1 2 Dec 2016 Asterionella formosa distinct Rhizophydiales sp.2 OL869010/OL868971/-
7 Synedra-MDA20 4 May 2016 Ulnaria sp. (former Rhizophydium fragilariae4 Rhizophydiales sp.3a -/OL868988/-
Synedra-MDA23 4 May 2016 Synedra sp.) OL869015/OL868987/
OL869116
8 Fragilaria-MDA54 (LSU 2 bp 26 May 2016 Fragilaria crotonensis Rhizophydium fragilariae4 Rhizophydiales sp.3b -/OL868989/OL869121
difference)
9 Diatoma-MDA19 4 May 2016 Diatoma tenuis cannot be determined Rhizophydiales sp.4 -/OL868993/-
10 Staurastrum-Chy4 15 Sept 2016 Closterium sp. Protrudomyces lateralis5 Rhizophydiales; Protrudomyces -/OL869003/-
Staurastrum-Chy5 15 Sept 2016 Cosmarium sp. lateralis5 -/OL869004/-
11 Cosmarium-MDAExp14 28 Jun 2016 Cosmarium sp. cannot be determined Rhizophydiales sp.5a OM859419/OL869001/
cannot be determined OL869120
12 Cosmarium-MDAExp17 (LSU 28 Jun 2016 Cosmarium sp. Rhizophydiales sp.5b -/OL869002/-
3 bp difference)
13 SVdW-EUD26 15 Jul 2015 Yamagishiella unicocca Dangeardia mamillata Order incertae sedis; Dangeardia MG605054/MG605051/-
mamillata
14 Staurastrum-MDAExp11 28 Jun 2016 Staurastrum sp. cannot be determined Chytridiales sp.1 OM859422/OL869000/-
15 FRA-CHY1 5 Mar 2015 Fragilaria crotonensis Chytridium versatile/ “Species Lobulomycetales sp.1 OL869012/OL868973/-
Fragilaria-A1 27 Apr 2016 3 7” -/OL868975/-
Fragilaria-A3 27 Apr 2016 -/OL868977/-
Fragilaria-B1 20 Apr 2016 -/OL868974/-
Fragilaria-B7 6 Apr 2016 -/OL868976/-
Fragilaria-MDA06 13 Apr 2016 -/OL868978/-
Fragilaria-MDA08 20 Apr 2016 -/-/OL869114
Fragilaria-MDA25 4 May 2016 -/OL868981/OL869113
Fragilaria-MDA26 4 May 2016 OL869013/OL868982/-
Fragilaria-MDA39 11 May 2016 -/OL868979/-
Fragilaria-MDA42 11 May 2016 -/OL868980/-
S. Van den Wyngaert et al.
5
chytrid species, but they were highly similar (96–99.7%) to
sequences from other uncultivated pollen-associated chytrids
(Supplementary Table S4).
Community composition of lake fungi
We determined 1741 ASVs in 42 pelagic samples of Lake Stechlin
collected between March 2015 and June 2016. This period
included two diatom spring blooms, and two “pollen rain” events.
Among these ASVs, 1545 (89%) were classified within the fungal
kingdom. The highest proportion of fungal ASVs belonged to
Ascomycota (43%), followed by Basidiomycota (30%), Chytridio-
mycota (18%), Mucoromycota (3%), and Aphelidiomycota (1%).
Rozellomycota, Neocallimastigomycota, and Blastocladiomycota
made up together only 1% of the fungal community and 3% of
fungal ASVs belonged to Fungi incertae sedis (Supplementary
Table S5, Fig. S4).
Illuminating the “dark matter” zoosporic fungi
We identified the host-substrate association of 26% (83 ASVs) of all
zoosporic fungi in Lake Stechlin (319 ASVs), including Chytridio-
mycota, Blastocladiomycota, Aphelidiomycota, and Rozellomy-
cota. Almost two-thirds of assignments were derived from our
targeted approaches (cultivation/single cells 57%, in situ baiting
43%) and one-third stemmed from public reference databases. In
total, 11.3% of ASVs were associated with pollen, and 13.2% of
ASVs with phytoplankton (10.7% diatoms, 1.5% green algae, and
1% cyanobacteria). 1.5% of ASVs were associated with multiple
substrates, i.e., green algae/pollen and green algae/diatoms
(Fig. S5). When considering ASV sequence abundance instead of
number, we could identify 68.5% of total zoosporic fungal reads.
The majority of reads were associated with pollen (34%) and
diatoms (30%), and only 1.1, 0.1, and 3.3% were associated with
green algae, cyanobacteria, and multiple substrates, respectively
(Fig. S5).
Temporal dynamics of lake fungi abundance and prevalence
of infection
In early spring 2015, the fungal community was dominated by
Ascomycota associated with the diatom spring bloom, whereas
Chytridiomycota dominated the fungal community during both
“pollen rain” events in late spring 2015 and 2016, and during the
diatom spring bloom in 2016. In summer, the fungal community
was more diverse including the presence of Aphelidiomycota and
a higher proportion of unclassified fungi. Autumn and winter
periods were dominated by Ascomycota and Basidiomycota, with
Chytridiomycota increasing in relative abundance towards January
(Fig. 3A).
The zoosporic fungal community in Lake Stechlin exhibited
clear seasonal dynamics (Figs. 3B and 4, PERMANOVA Bray Curtis,
p < 0.001). In spring, 72% ± 0.17 (2015) and 62% ± 0.40 (2016) of
zoosporic fungal sequences matched diatom- or pollen-associated
chytrids. During spring diatom blooms, parasitic chytrids domi-
nated the community (Fig. 3B) and only a small proportion (<2%)
was attributed to pollen-associated chytrids. Two parasites
infecting the diatom Fragilaria, namely specialist Fragilaria-CHY1
(LO-1, ASV47) and generalist Fragilaria-MDA54/Synedra-MDA20
(RH-1, ASV23-ASV44), capable of infecting Fragilaria and Ulnaria,
equally dominated the zoosporic fungal community (17–82%)
during the spring bloom in 2015 (Fig. 5A). During this time,
Fragilaria represented 2–10% of the total phytoplankton biomass
(Supplementary table S2), and prevalence of infection was 5–44%
(Fig. 5A). In spring 2016, both parasites were present in much
lower relative abundance (1–12%), and during this time, Fragilaria
did not exceed 1% of the total phytoplankton biomass and
prevalence of infection reached 16%. Instead, Cyclotella sp. (a
small centric diatom) was highly impacted by chytrids (max.
prevalence 41%) with ASV6 corresponding to single-cell Cyclo-
tella-MDA01 (max. relative abundance 91%) dominating the
The ISME Journal
Table 1. continued
ID Chytrid ID Isolation date Original host species Chytrid morphology Chytrid phylogeny Accession no. 18 S/28 S/
isolated nanopore (18S-ITS-28S)
16 SVdW-EUD36 2 Dec 2015 Eudorina elegans Algomyces stechlinensis Lobulomycetales; Algomyces MG605055/MG605052/-
stechlinensis
17 SVdW-EUD16 9 Jun 2015 Yamagishiella unicocca Endocoenobium eudorinae Polyphagales; Endocoenobium MG605053/MG605050/-
Yamagishiella-MDA59 22 Jun 2016 eudorinae -/-/OL869118
Yamagishiella-MDAExp1 28 Jun 2016 -/OL868994/-
Yamagishiella-MDAExp5 28 Jun 2016 -/OL868995/-
Yamagishiella-MDAExp6 28 Jun 2016 -/OL868996/-
18 Dolichospermum-MDA2- 9 Aug 2017 Dolichospermum sp. Rhizosiphon akinetum8 Fungi Incertae sedis sp.1 OL869110/OL869133/-
akinet 9 Aug 2017 -/OL869005/-
Dolichospermum-MDA7-
akinet
19 Dolichospermum-MDA5- 9 Aug 2017 Dolichospermum sp. Rhizosiphon crassum4 Fungi Incertae sedis sp.2 -/OL869006/-
vegetative
20 Pollen-CHY1 22 May 2015 pollen pinus Globomyces pollinis-pini9 Rhizophydiales; Globomyces pollinis- OM859420/OL869009/-
pini9
21 Pollen-MDA28 11 May 2016 pollen other cannot be determined Rhizophydiales sp.6 -/OL869007/-
22 Pollen-MDA36 11 May 2016 pollen other cannot be determined Rhizophydiales sp.7 -/OL869008/-
References: 1 Seto et al. [47], 2Canter 1970, 3Van den Wyngaert et al. [34], 4Canter 1953, 5Letcher et al. [56], 6Van den Wyngaert et al. [29], 7Canter and Lund 1953, 8Canter [58]
S. Van den Wyngaert et al.
6
Fig. 1 Maximum-likelihood tree of Fungi using concatenated rRNA gene sequences (18S, 28S). The maximum likelihood bootstrap values
of 1000 repetitions are indicated at the nodes. Isolates and single cell sequences from this study are marked in bold and color coded
according to their host/substrate; brown (diatom host), dark green (chlorophyte host), light green (desmid host), blue (cyanobacteria host),
black (pollen substrate).
The ISME Journal
S. Van den Wyngaert et al.
7
Cyclotella-MDA1
Zygophlyctidiales SVdW-SYN-CHY1 Zygophlyctis planktonica
Asterionella-MDA57
Diatoma-MDA07
Pollen-CHY1
Staurastrum-CHY4/Staurastru-CHY5
Pollen-MDA36
Pollen-MDA28
Syendra-MDA23/Fragilaria-MDA54
Stephanodiscus-MDA04/Fragilaria-B6/Synedra-A6
AST-CHY1
Diatoma-MDA19
Cosmarium-MDAExp14
STAU-CHY3 Staurastromyces oculus
FRA-CHY1
Lobulomycetales
SVdW-EUD3 Algomyces stechlinensis
Polyphagales SVdW-EUD1 Endocoenobium eudorinae
Chytridiales StaurastrumMDAExp11
Chytridiomycota incertae sedis SVdW-EUD2 Dangeardia mamillata
Dolichospermum-MDA5-vegetative
Fungi incertae sedis
Dolichospermum-MDA2-akinet
Fig. 2 Host-chytrid association matrix based on experimental cross infection data and occurrence data. On the x-axis substrate (pollen)
and phytoplankton host species and on the y-axis chytrid strains and single cell isolates, clustered according to their taxonomic relatedness.
Rectangles indicate compatible host/substrate-chytrid pairs with the color code referring to host/substrate taxa; dark brown (pennate
diatoms), light brown (centric diatoms), dark green (Chlorophyta), light green (desmids), blue (cyanobacteria), black (pollen).
fungal community (Fig. 5A). Diatom bloom decay and the onset of to summer (Fig. S7), but 40% of the population was infected.
pollen rain were reflected by a shift towards saprotrophic pollen- Chytrids associated with green algae, e.g., Dangardia mamillata
degrading chytrids, reaching 97% (±0.09) and 92% (±0.07) of the (ASV94) and Endocoenobium eudorinae (ASV92), were mainly
total reads in May 2015 and 2016, respectively (Fig. 3B). present in summer and Algomyces stechlinensis (ASV585) in
Summer and autumn samples showed the highest zoosporic autumn. Patterns of prevalence of infection on the host species
fungal diversity being significantly higher compared to spring followed occurrence patterns of the respective parasites (Fig. 5B).
2016 (summer 2016-spring 2016: p < 0.001; autumn 2016-spring Staurastrum-MDAExp11 (ASVs 29, 1003) was present on different
2016: p < 0.001, Supplementary data, Fig. S6). These seasons occasions throughout the year. The highest relative abundance
displayed a mix of saprotrophic, parasitic, and facultative parasitic occurred during autumn when Staurastrum was present and
chytrids associated with major phytoplankton groups (green infected, but did not match with the prevalence of infection
algae, cyanobacteria, diatoms) and contained a higher percentage pattern (Fig. 5B). Parasites of cyanobacteria, infecting Dolichos-
of “unknown” sequences (Fig. 3B). Zygophlyctis asterionellae (ASVs permum solitaria and D. circinalis vegetative cells and akinetes only
4, 96, 98, 102, 125, 149, 220), a host-specific parasite on occurred in autumn and at low relative abundance (<2%).
Asterionella, reached a high relative abundance in summer, Prevalence of infection on D. solitaria and D. circinalis ranged
coinciding with two chytrid epidemics on the diatom Asterionella from 1–16% (note: the maximum value was only based on six
and infecting up to 68% (2015) and 78% (2016) of the population filaments) (Fig. 5C). Microscopic observations confirmed the
(Fig. 5A). Zygophlyctis planktonica (ASVs 105, 255, 416, 630, 668), a absence of chytrid infections on Dolichospermum spp. during
closely related but host-specific chytrid for Ulnaria sp. [47], showed summer blooms when total cyanobacteria biomass was highest
a similar temporal pattern as Zygophlyctis asterionellae (Fig. 5A). A (Fig. S7), but the relative proportion of D. solitaria in the
second parasite of Asterionella, AST-CHY1 (ASVs 34, 1302), Dolichospermum community (total biomass) was lower, i.e., max.
frequently occurred during the whole year, albeit in low relative 0.2% in summer vs. max. 24% in autumn (Supplementary
abundance (0.2–11%), even when the Asterionella biomass was table S2).
very low or non-detectable and no infected Asterionella cells were During winter, a mixed community of saprotrophic and parasitic
detected by microscopy (Fig. 5A). In autumn (16th November chytrids persisted with higher proportions of diatom parasites
2015), ASV 34 dominated the zoosporic fungal community (57%) (Fig. 3B). Prevalence of infection on diatom species was low (<2%)
when Asterionella was present at relatively low biomass compared in winter compared to other seasons (Fig. 5A).
The ISME Journal
Rhizophydiales
Pollen
Stephanodiscus sp.
Cyclotella sp.
Fragillaria sp.
Ulnaria sp.
Asterionella formosa
Diatoma sp.
Closterium sp.
Cosmarium sp.
Staurastrum sp.
Eudorina elegans
Yamagishiella unicocca
Dolichospermum sp. (vegetative)
Dolichospermum sp. (akinete)
S. Van den Wyngaert et al.
8
Fig. 3 Seasonal dynamics of the fungal and phytoplankton community in Lake Stechlin. Fungal phyla and their relative abundance (A),
identified zoosporic fungi substrate associations, including microscopy images illustrating the the succession of different phytoplankton/
substrate-chytrid pairs (B) and biomass and relative proportions of phytoplankton taxa (C). Note different dates on X-axis for the lower
phytoplankton plot.
The ISME Journal
S. Van den Wyngaert et al.
9
26.05.2016 Pollen rain
18.05.2016
0.23 11.05.2016
16.04.2016
08.03.2016
02.06.2016 20.05.2015 06.04.2016
02.06.2015 spring 2016 30.03.2016
0.15 20.04.2016 23.03.2016
05.06.2015
Diatom spring bloom
0.08 spring 2015
18.03.2015
24.03.2015 26.01.2016
04.05.2016 27.04.2016
0.00
summer 2015/2016 winter 2015/2016
01.01.2016
-0.08 16.06.2015
12.06.2015
-0.15 22.06.2016 02.12.2015
23.06.2015
16.11.2015
30.06.2015 07.07.2015
-0.23 15.09.2015 autumn 2015
14.07.2015 16.10.2015
05.11.2015
02.09.2015
-0.30
-0.30 -0.23 -0.15 -0.08 0.00 0.08 0.15 0.23 0.30
Axis 1 (14.3%)
Fig. 4 PCoA of zoosporic fungal composition over the sampling period across seasons (spring summer, autumn, winter) 2015 to 2016.
Transparent gray areas indicate sampling dates during diatom spring bloom and pollen rain periods.
DISCUSSION obtained from the pollen baits, the majority of pollen-associated
Illuminating “dark matter” zoosporic fungi fungal ASVs were not assigned to any known species. Although
Over the course of 15 months we identified zoosporic fungi that most reference sequences belong to saprotrophic chytrids, such
exhibited varying degrees of phytoplankton host specific parasit- low agreement reflects that many saprotrophic chytrids in the
ism, and saprotrophy on pollen. A high turnover of fungal databases have been isolated primarily from soil, ponds, and
diversity, driven by changes in autochthonous and allochthonous wetlands [56, 59] and that lake ecosystems harbor unique,
available carbon, is remarkable and has implications for both the uncharacterized pollen-degrading chytrids. Moreover, pure pine
diversity of fungi and associated phytoplankton. We show that or sweet gum pollen is commonly used for isolating saprophytic
chytrid epidemics on diatom species (including small edible chytrids [59, 60], whereas our study used natural pollen bait
species) occur throughout the year and are driven by multiple originating mainly from pine trees, but also including pollen from
parasite species that either co-occur or occupy different temporal other tree species (presumably birch and beech) collected from
niches. Revealing those dynamics was only made possible by the local environment. Our result suggests that the diversity of
linking targeted isolation approaches, laboratory infection assays, pollen-degrading chytrids is likely to be underestimated when
microscopy, and metabarcoding which greatly improved our only baiting with single pollen types and that saprophytic pollen
ability to assign ecological functions to environmental sequences. degrading chytrids display some degree of specificity for different
Of all zoosporic fungal ASVs, 26% could be assigned to parasitic pollen types. Importantly, ASVs assigned to either parasitic
phytoplankton-infecting or saprotrophic pollen-degrading life- phytoplankton-chytrid and saprotrophic pollen-chytrid interac-
styles. This assignment would be substantially lower (<10%) based tions made up almost 70% of all zoosporic fungal reads in Lake
on the current status of the NCBI sequence database. Moreover, Stechlin, suggesting that they are major components of the
this study obtained the first sequences of two parasitic chytrids zoosporic fungal community. We do point out that, in accordance
tentatively identified as Rhizosiphon akinetum and R. crassum with previous freshwater studies [61, 62], Ascomycota and
associated with the nuisance cyanobacterium Dolichospermum Basidiomycota presented the majority of fungal ASVs (see
[12, 58] and revealed their putative phylum-level phylogenetic Supplementary Text S9 for more details). The combination of
novelty. Phylogenomic analysis is necessary to clarify their precise targeted isolation with environmental sequencing, proven suc-
phylogenetic position. We further identified novel diatom-specific cessful in our study system, could be applied to any type of
parasites within Zygophlyctidales, Rhizophydiales, and Lobulomy- ecosystem and fungal group. Transferability of such an approach
cetelaes, emphasizing the large potential of phytoplankton- for higher fungi will depend on identification and isolation
associated fungal parasites to fill current research gaps concerning expertise of researchers. The heterogeneous morphologies and
aquatic fungal diversity and taxonomy. Besides two sequences often complex life cycles of Ascomycota and Basidiomycota, i.e.,
The ISME Journal
Axis 2 (11.3%)
S. Van den Wyngaert et al.
10
A diatoms B green algae C cyanobacteria
100 centric diatoms <10µm Fragilaria Synedra Asterionella 100 Eudorina/Yamagishiell Colonial green algae spp. Staurastrum 100 D. solitaria-vegetative D. circinales-vegetative D. circinales-akinete
90 90 90
80 80 80
70 70 70
60 60 60
50 50 50
40 40 40
30 30 30
20 20 20
10 10 10
0 0 0
ASV105 SVdW-Syn-Chy1-S ASV23 Synedra-MDA20/Fragillaria-MDA54-G
ASV4 Ast-MDA57-S ASV34 AST-Chy1-S 100 ASV94 SVdW-EUD2-G  ASV585 SVdW-EUD3-G
100 100 ASV337 Dolichospermum-akinete ASV1065 Dolichospermum-vegetative
ASV47 Fragilaria-Chy1-S ASV44 Fragillaria-MDA54/Synedra-MDA20-G ASV92 SVdW-EUD1-S ASV29 Staurastrum-Exp11
90 90 90
ASV6 Cyclotella-MDA01-S
80 80 80
70 70 70
60 60 60
50 50 50
40 40 40
30 30 30
20 20 20
10 10 10
0 0 0
Fig. 5 Temporal dynamics of the most abundant parasitic chytrid ASVs and prevalence of infection on the respective host species.
Parasitic chytrids associated with diatoms (A), with green algae (B), and cyanobacteria (C).
from small, free living single celled yeast to large substrate support its more generalist lifestyle (Fig. 5B). Whether this
associated filamentous hyphae, may provide additional challenges represents a rare case of a generalist chytrid with an inter-
compared to the rather simple life cycle and morphology of taxonomic host range would require additional isolation and cross
attached sporangial forms of chytrids. infection assays.
Specialist vs. generalist Seasonal dynamics of zoosporic fungi
Our cross-infection experiments showed that strain Fragilaria- The zoosporic fungal community in Lake Stechlin showed a clear
CHY1 was host-specific for Fragilaria. Its partial SSU, LSU, and ITS seasonality with distinct winter-spring, summer, and autumn
sequences were, however, almost identical to the recently communities. Our observations support the hypothesis that
rediscovered and sequenced species Zygorhizidium affluens, saprotrophic chytrids are related to the input of allochthonous
parasitic on the diatom Asterionella formosa in Lake Pavin, France organic matter (i.e., pollen) and parasitic chytrids to the seasonal
[57]. Nanopore sequencing of the rRNA operon from single cells dynamics of their phytoplankton hosts [4, 12, 32]. Chytrid infection
confirmed partial SSU, LSU, and ITS sequences being identical to on phytoplankton occurred throughout all seasons and years
Fragilaria-CHY1, but also showed several introns in the SSU region. examined. Whereas infected phytoplankton could not be
Host speci city of Z. af , ,
fi fluens has not been investigated, however, observed by microscopy on June 23rd 2015, and January 26th
differences between Z. affluens and Fragilaria-CHY1 in host 2016, metabarcoding revealed the presence of ASVs matching
specificity and introns may suggest genetic isolation with ongoing with diatom parasites that accounted for 10 to 80% of the
diversification and host specialization [63]. Or, as seen in other zoosporic fungal community (Fig. 3B, Supplementary tables S7).
host-parasite systems, both specialist and generalist strains likely Different seasonal patterns were detected between multiple
coexist [64]. In case Fragilaria-CHY1 would have a more generalist chytrid parasites sharing the same host. Whereas Rhizophydiales
lifestyle, we would expect it to occur also during times when sp. (AST-CHY1) and Zygophlyctis asterionellae parasites of Aster-
Fragilaria is absent or not infected, indicating its potential to ionella dominated in different seasons, parasitic generalist
reproduce on alternative host species. However, the correspond- (Fragilaria-MDA54/Synedra-MDA23) and specialist (Fragilaria-
ing prevalence of infection pattern on Fragilaria with the presence CHY1) of Fragilaria also coexisted, though specialists are expected
and relative abundance of ASV47 (Fragilaria-CHY1) and the to be superior competitors on a common diatom host [28].
absence of ASV47 during both epidemics on Asterionella, supports Species-specific environmental optima may drive such different
its host preference for Fragilaria (Fig. 5A). On the contrary, AST- seasonal dominance patterns [65–67] and the presence of host-
CHY1 was only infective on Asterionella in our cross-infection specific hyperparasites could provide another mechanism for the
assays, while the corresponding ASV34 frequently occurred even coexistence of multiple parasite species on the same host
when Asterionella was absent or not infected (Fig. 5A), pointing to population [68, 69]. Microscopic observation identified a putative
a generalist lifestyle for this parasite. It should be noted that our Rozellomycota hyperparasite encysted on a chytrid sporangia
cross-infection assays used a combination of a single clonal infecting Fragilaria (Supplementary Fig. S8). Additionally, a strong
chytrid strain with a single clonal host strain, thus, extrapolating to correlation (Pearson´s R= 0.97, p < 0.001) was found between
the population level advises caution. Another issue that needs specialist Fragilaria-CHY1 (ASV47) and the most abundant
consideration is phylogenetic resolution. The LSU D1 marker Rozellomycota ASV (ASV141), suggesting a putative Rozellomy-
showed limitations to resolve the closely related diatom parasites cota hyperparasite infecting chytrid Fragilaria-CHY1, as described
within clade RH2 (ASV23 matched 100% with Syn-MDA20-Fra- previously [70].
MDA54 but had only a 2 bp mismatch with Ast-Chy1, whereas Contrary to obligate parasites, ASVs matching with facultative
ASV34 matched 100% with Ast-Chy1 and had a 2 bp mismatch parasites only occurred in a few samples and never reached high
with Syn-MDA20-Fra-MDA54 and only 1 bp mismatch with relative abundances, i.e., strain Fragilaria B6/single-cell
Diatom-MDA19). Culture isolates of Syn-MDA20-Fra-MDA54 and Stephanodiscus-MDA04 (1 sample, 0.4%), Staurastrum-CHY4
Diatom-MDA19 are needed to resolve better inter- vs. intraspecific (5 samples, max. 1.7%), Aquamyces chlorogonii (1 sample, 0.5%).
variability within this clade. Interestingly, Staurastrum-MDAExp11, Whereas obligate parasites are likely to be superior competitors in
associated with desmid Staurastrum sp. in Lake Stechlin had the upper pelagic zone associated with active phytoplankton
almost identical LSU (99.77%) sequences as single-cell chytrids growth, the importance of facultative parasites may increase with
(819o12Aa and 704k6Ag) associated with two species of the depth, i.e., with increasingly senescent or dead cells of sinking
diatom genus Aulacoseira in Lake Inba (Japan) [28]. The phytoplankton in the hypolimnion.
discrepancy between the patterns of prevalence of infection on Chytrid epidemics have been mostly reported from large (e.g.
Staurastrum and the presence of Staurastrum-MDAExp11 (ASV 29) inedible) bloom forming diatom species [12, 71, 72], but we
The ISME Journal
Relative abundance (%) Prevalence of infection (%)
2015.03.05
2015.03.11
2015.03.18
2015.03.24
2015.04.08
2015.04.15
2015.04.21
2015.05.07
2015.05.20
2015.06.02
2015.06.05
2015.06.12
2015.06.16
2015.06.23
2015.06.30
2015.07.07
2015.07.14
2015.09.02
2015.09.15
2015.09.22
2015.09.29
2015.10.16
2015.10.27
2015.11.05
2015.11.16
2015.12.02
2015.12.17
2016.01.01
2016.01.26
2016.03.08
2016.03.23
2016.03.30
2016.04.06
2016.04.16
2016.04.20
2016.04.27
2016.05.04
2016.05.11
2016.05.18
2016.05.26
2016.06.02
2016.06.22
Relative abundance (%) Prevalence of infection (%)
2015.03.05
2015.03.11
2015.03.18
2015.03.24
2015.04.08
2015.04.15
2015.04.21
2015.05.07
2015.05.20
2015.06.02
2015.06.05
2015.06.12
2015.06.16
2015.06.23
2015.06.30
2015.07.07
2015.07.14
2015.09.02
2015.09.15
2015.09.22
2015.09.29
2015.10.16
2015.10.27
2015.11.05
2015.11.16
2015.12.02
2015.12.17
2016.01.01
2016.01.26
2016.03.08
2016.03.23
2016.03.30
2016.04.06
2016.04.16
2016.04.20
2016.04.27
2016.05.04
2016.05.11
2016.05.18
2016.05.26
2016.06.02
2016.06.22
Relative abundance (%) Prevalence of infection (%)
2015.03.05
2015.03.11
2015.03.18
2015.03.24
2015.04.08
2015.04.15
2015.04.21
2015.05.07
2015.05.20
2015.06.02
2015.06.05
2015.06.12
2015.06.16
2015.06.23
2015.06.30
2015.07.07
2015.07.14
2015.09.02
2015.09.15
2015.09.22
2015.09.29
2015.10.16
2015.10.27
2015.11.05
2015.11.16
2015.12.02
2015.12.17
2016.01.01
2016.01.26
2016.03.08
2016.03.23
2016.03.30
2016.04.06
2016.04.16
2016.04.20
2016.04.27
2016.05.04
2016.05.11
2016.05.18
2016.05.26
2016.06.02
2016.06.22
S. Van den Wyngaert et al.
11
observed that also small-sized diatoms (e.g. Cyclotella spp.) are mixed, pelagic zone of Lake Stechlin. We demonstrate a high
highly impacted by chytrids. Single-cell Cyclotella-MDA01 con- turnover of zoosporic fungal diversity, driven by changes in
stituted the third most abundant zoosporic fungal ASV consider- autochthonous and allochthonous available carbon and provide
ing the entire sampling period. During the 2016 spring bloom, this evidence that phytoplankton-parasites and saprotrophic pollen
chytrid dominated the overall fungal community, highlighting the degraders are key components of the zoosporic fungal commu-
importance of chytridiomycosis also for smaller and thus nity. Chytrid epidemics on diatoms (including small edible species)
potentially more edible diatoms. This is of great relevance as lake occur throughout the year and are driven by multiple parasite
warming mainly favors small-sized planktonic diatom species, species that either co-occur or occupy different temporal niches.
particularly within the genus Cyclotella [73]. Yet, the resulting Revealing those dynamics was only made possible by linking
effects for higher trophic levels, e.g. via the mycoloop [22] require targeted isolation approaches, laboratory infection assays, micro-
further investigations. scopy, and metabarcoding which greatly improved our ability to
In addition to parasites, we show that also saprotrophic chytrids assign ecological functions to environmental sequences. We
affect the seasonal succession of plankton communities. Over two highlight that successful identification of the most abundant
consecutive years, the transition from the spring diatom bloom to zoosporic fungal ASVs in Lake Stechlin was largely accomplished
a clear water phase with massive pollen input was consistently by single cell and culture isolate sequencing. As long read
reflected by a shift from parasitic- to saprotrophic-dominated metabarcoding and (meta)genomics are improving rapidly by
chytrid communities. Pollen input often occurs during the clear- getting more cost-efficient, they will ultimately solve single marker
water phase when phytoplankton biomass and nutrient concen- choices for complex environmental samples, providing increased
trations are low [74] and thus represents an important nutrient resolution and reduced taxonomic biases. Coupling these third-
input in spring-summer in many temperate and boreal lakes [75]. generation sequencing technologies to high quality reference
For example, in Lake Stechlin, pollen rain accounts for nearly half sequences with rich metadata, as generated in this study, will
of the yearly atmospheric phosphorus input [76]. Whereas pollen enable a better exploration of spatial and temporal distribution of
grains are hardly ingested by zooplankton, saprotrophic chytrids chytrids in temperate lakes worldwide.
render this otherwise inaccessible food source available to grazers
in the form of readily edible chytrid zoospores [77, 78]. Such a
mycoloop [22] effectively channels allochthonous organic matter DATA AVAILABILITY
to higher trophic levels such as zooplankton, which in particular is Raw sequence data is available in the NCBI Sequence Read Archive (SRA) under
important during the clear water phase when phytoplankton prey BioProject PRJNA682007. Sequences from strains and single cell isolates have been
abundance is low. deposited in GenBank under accession no. OL869010-OL869016; OM859415-
Summer represented a transitional period leading to a more OM859422 (18 S Sanger), OL868971-OL869009 (28 S Sanger), OL869133 (28 S
diverse saprotrophic and parasitic chytrid community which shotgun metagenome), OL869110 (18 S shotgun metagenome), OL869111-
OL869121 (Nanopore). All other data generated or analyzed during this study are
culminated in autumn when zoosporic fungal diversity was included in this published article and its supplementary information files.
highest. Microscopy confirmed widespread chytrid infections on
various phytoplankton taxa (highest number of infected species
recorded in autumn; Supplementary Table S7). A similar pattern
has been observed in other temperate lakes [6, 12] where a high REFERENCES
phytoplankton diversity in autumn but at lower abundance 1. Grossart H-P, Van den Wyngaert S, Kagami M, Wurzbacher C, Cunliffe M, Rojas-
compared to spring is suggested to favor the co-existence of a Jimenez K. Fungi in aquatic ecosystems. Nat Rev Microbiol. 2019;17:339–54.
diverse community of host-specific parasites. However, due to low 2. Panzer K, Yilmaz P, Weiß M, Reich L, Richter M, Wiese J, et al. Identification of
cell abundances, we were not able to capture this phenomenon habitat-specific biomes of aquatic fungal communities using a comprehensive
with our isolation approach. Moreover, Aphelidiomycota, present nearly full-length 18S rRNA dataset enriched with contextual data. PLOS ONE.
in summer, are known parasites of green algae, yellow-green 2015;7:e0134377.
algae, and diatoms [79, 80] that may have possibly been 3. Grossart H-P, Wurzbacher C, James TY, Kagami M. Discovery of dark matter fungi
overlooked in microscopic studies due to their intracellular in aquatic ecosystems demands a reappraisal of the phylogeny and ecology of
infection stages. However, the relative abundance of Aphelidio- zoosporic fungi. Fungal Ecol. 2016;19:28–38.
4. Wurzbacher C, Warthmann N, Bourne E, Attermeyer K, Allgaier M, Powell JR, et al.
mycota remained low (Fig. 5B and Fig. S4) indicating that chytrids High habitat-specificity in fungal communities in oligo-mesotrophic, temperate
represent the major zoosporic fungal phytoplankton parasites. Lake Stechlin (North-East Germany). MycoKeys. 2016;16:17–44.
During winter, despite low phytoplankton biomass, parasitic 5. Comeau AM, Vincent WF, Bernier L, Lovejoy C. Novel chytrid lineages dominate
chytrids on diatoms made up a substantial portion of the zoosporic fungal sequences in diverse marine and freshwater habitats. Sci Rep.
fungal community. As water temperatures below 3°C may inhibit 2016;6:30120.
chytrid infection [65, 81], cold winters with ice coverage and no or 6. Lefèvre E, Letcher PM, Powell MJ. Temporal variation of the small eukaryotic
little snow provide a disease-free window of opportunity for diatom community in two freshwater lakes: emphasis on zoosporic fungi. Aquat Micro
growth [15]. In Lake Stechlin, the observed low levels of infected Ecol. 2012;67:91–105.
diatoms during winter re ect the importance of cold winters for 7. Sparrow F. Aquatic phycomycetes. 2nd ed. Ann Arbor: University of Michigan
fl
Press; 1960.
diatom spring bloom development. Gradual loss of this environ- 8. Frenken T, Alacid E, Berger SA, Bourne EC, Gerphagnon M, Grossart H-P, et al.
mental refuge, e.g. by increasingly warmer winters, may contribute Integrating chytrid fungal parasites into plankton ecology: research gaps and
to earlier and less intense diatom spring blooms and thus may lead needs: Research needs in plankton chytridiomycosis. Environ Microbiol.
to possible shifts in phytoplankton community composition during 2017;19:3802–22.
the following season [82–84]. As the bloom inoculum affects the 9. Wurzbacher C, Bärlocher F, Grossart H. Fungi in lake ecosystems. Aquat Micro
subsequent phytoplankton dynamics in spring, the loss of the Ecol. 2010;59:125–49.
winter refuge from chytrids infection exemplifies the far-reaching 10. Ibelings BW, De Bruin A, Kagami M, Rijkeboer M, Brehm M, van Donk E. Host
consequences of gradual lake warming for plankton community parasite interactions between freshwater phytoplankton and chytrid fungi
dynamics, trophic interactions, and consequently ecosystem (Chytridiomycota). J Phycol. 2004;40:437–53.
11. Kagami M, de Bruin A, Ibelings BW, Donk EV. Parasitic chytrids: their effects on
functioning [85]. phytoplankton communities and food-web dynamics. Hydrobiologa.
2007;578:113–29.
Summary 12. Rasconi S, Niquil N, Sime-Ngando T. Phytoplankton chytridiomycosis: community
This study represents a unique effort to link zoosporic fungal structure and infectivity of fungal parasites in aquatic ecosystems. Environ
sequence diversity and consumer-resource interactions in the Microbiol. 2012;14:2151–70.
The ISME Journal
S. Van den Wyngaert et al.
12
13. Sommer U, Adrian R, De Senerpont Domis L, Elser JJ, Gaedke U, Ibelings B, et al. 39. White T, Bruns T, Lee S, Taylor J. Amplification and direct sequencing of fungal
Beyond the Plankton Ecology Group (PEG) Model: Mechanisms driving plankton ribosomal RNA genes for phylogenetics. In: Innis M, Gelfand D, Shinsky J, White T,
succession. Ann Rev Ecol Evol Syst. 2012;43:429–48. (ed.). PCR Protocols: A Guide to Methods and Applications. New York, NY: Aca-
14. Canter HM, Lund JWG. Studies on plankton parasites: III. Examples of the inter- demic Press Inc; 1990. p. 315–22.
action between parasitism and other factors determining the growth of diatoms. 40. Smit E, Leeflang P, Glandorf B, Dirk van Elsas J, Wernars K. Analysis of fungal
Ann Bot. 1951;15:359–71. diversity in the wheat rhizosphere by sequencing of cloned PCR-Amplified genes
15. Ibelings BW, Gsell AS, Mooij WM, van Donk E, Van den Wyngaert S, De Senerpont encoding 18S rRNA and temperature gradient gel electrophoresis. Appl Environ
Domis L. Chytrid infections and diatom spring blooms: paradoxical effects of Microbiol. 1999;65:2614–21.
climate warming on fungal epidemics in lakes. Freshw Biol. 2011;56:754–66. 41. Hall TA. BioEdit: a user-friendly biological sequence alignment editor and analysis
16. Gsell AS, Domis LNDS, Naus‐Wiezer SMH, Helmsing NR, Donk EV, Ibelings BW. program for Windows 95/98/NT. Nucleic Acids Symp Ser. 1999;41:95–98.
Spatiotemporal variation in the distribution of chytrid parasites in diatom host 42. Wurzbacher C, Larsson E, Bengtsson‐Palme J, Wyngaert SV, den, Svantesson S,
populations. Freshw Biol. 2013;58:523–37. Kristiansson E, et al. Introducing ribosomal tandem repeat barcoding for fungi.
17. van Donk E, Ringelberg J. The effect of fungal parasitism on the succession of Mol Ecol Resour. 2019;19:118–27.
diatoms in Lake Maarsseveen I (The Netherlands). Freshw Biol. 1983;13:241–51. 43. Katoh K, Standley DM. MAFFT multiple sequence alignment software version 7:
18. Thingstad TF. Elements of a theory for the mechanisms controlling abundance, improvements in performance and usability. Mol Biol Evol. 2013;30:772–80.
diversity, and biogeochemical role of lytic bacterial viruses in aquatic systems. 44. Capella-Gutiérrez S, Silla-Martínez JM, Gabaldón T. trimAl: a tool for automated
Limnol Oceanogr. 2000;45:1320–28. alignment trimming in large-scale phylogenetic analyses. Bioinforma Oxf Engl.
19. Agha R, Gross A, Rohrlack T, Wolinska J. Adaptation of a chytrid parasite to its 2009;25:1972–73.
cyanobacterial host is hampered by host intraspecific diversity. Front Microbiol. 45. Stamatakis A. RAxML version 8: a tool for phylogenetic analysis and post-analysis
2018;9:921. of large phylogenies. Bioinformatics. 2014;30:1312–13.
20. Gsell AS, de Senerpont Domis LN, Verhoeven KJ, van Donk E, Ibelings BW. Chytrid 46. Miller MA, Schwartz T, Pickett BE, He S, Klem EB, Scheuermann RH, et al. A RESTful
epidemics may increase genetic diversity of a diatom spring-bloom. ISME J. API for access to phylogenetic tools via the CIPRES Science Gateway. Evol
2013;7:2057–59. Bioinforma Online. 2015;11:43–48.
21. Klawonn I, Wyngaert SV, den, Parada AE, Arandia-Gorostidi N, Whitehouse MJ, 47. Seto K, Van Den Wyngaert S, Degawa Y, Kagami M. Taxonomic revision of the
Grossart H-P, et al. Characterizing the “fungal shunt”: Parasitic fungi on diatoms genus Zygorhizidium: Zygorhizidiales and Zygophlyctidales ord. nov. (Chy-
affect carbon flow and bacterial communities in aquatic microbial food webs. tridiomycetes, Chytridiomycota). Fungal Syst Evol. 2020;5:17–38.
Proc Natl Acad Sci. 2021;118:e2102225118. 48. Nercessian O, Noyes E, Kalyuzhnaya MG, Lidstrom ME, Chistoserdova L. Bacterial
22. Kagami M, Miki T, Takimoto G. Mycoloop: chytrids in aquatic food webs. Front populations active in metabolism of C1 compounds in the sediment of Lake
Microbiol. 2014;5:166. Washington, a freshwater lake. Appl Env MicrobioL. 2005;71:6885–99.
23. Gerphagnon M, Agha R, Martin-Creuzburg D, Bec A, Perriere F, Rad-Menéndez C, 49. Tedersoo L, Anslan S, Bahram M, Põlme S, Riit T, Liiv I, et al. Shotgun metagen-
et al. Comparison of sterol and fatty acid profiles of chytrids and their hosts omes and multiple primer pair-barcode combinations of amplicons reveal biases
reveals trophic upgrading of nutritionally inadequate phytoplankton by fungal in metabarcoding analyses of fungi. MycoKeys. 2015;10:1–43.
parasites. Environ Microbiol. 2019;21:949–58. 50. Callahan BJ, McMurdie PJ, Holmes SP. Exact sequence variants should replace
24. Rasconi S, Ptacnik R, Danner S, Van den Wyngaert S, Rohrlack T, Pilecky M, et al. operational taxonomic units in marker-gene data analysis. ISME J. 2017;11:2639–43.
Parasitic chytrids upgrade and convey primary produced carbon during inedible 51. Quast C, Pruesse E, Yilmaz P, Gerken J, Schweer T, Yarza P, et al. The SILVA
algae proliferation. Protist. 2020;171:125768. ribosomal RNA gene database project: improved data processing and web-based
25. Frenken T, Wolinska J, Tao Y, Rohrlack T, Agha R. Infection of filamentous phy- tools. Nucleic Acids Res. 2013;41:D590–96.
toplankton by fungal parasites enhances herbivory in pelagic food webs. Limnol 52. Electronica P, Hammer Ø, Harper DAT, Ryan PD. Past: Paleontological Statistics
Oceanogr. 2020;65:2618–26. Software Package for Education and Data Analysis. 2001.
26. James TY, Kauff F, Schoch CL, Matheny PB, Hofstetter V, Cox CJ, et al. Recon- 53. R Core Team. R: A language and environment for statistical computing. R
structing the early evolution of Fungi using a six-gene phylogeny. Nature. Foundation for Statistical Computing, Vienna, Austria. Vienna, Austria: R Foun-
2006;443:818–22. dation for Statistical Computing; 2020. https://www.R-project.org/
27. Letcher PM, Powell MJ, Churchill PF, Chambers JG. Ultrastructural and molecular 54. Oksanen J, Blanchet FG, Friendly M, Kindt R, Legendre P, McGlinn D, et al. vegan
phylogenetic delineation of a new order, the Rhizophydiales (Chytridiomycota). community ecology package version 2.5-7 November 2020.
Mycol Res. 2006;110:898–15. 55. Anderson MJ. A new method for non-parametric multivariate analysis of variance.
28. Kagami M, Seto K, Nozaki D, Nakamura T, Wakana H, Wurzbacher C. Single Austral Ecol. 2001;26:32–46.
dominant diatom can host diverse parasitic fungi with different degree of host 56. Letcher PM, Vélez CG, Barrantes ME, Powell MJ, Churchill PF, Wakefield WS.
specificity. Limnol Oceanogr. 2021;66:667–77. Ultrastructural and molecular analyses of Rhizophydiales (Chytridiomycota) iso-
29. Van den Wyngaert S, Rojas-Jimenez K, Seto K, Kagami M, Grossart H-P. Diversity lates from North America and Argentina. Mycol Res. 2008;112:759–82.
and hidden host specificity of chytrids infecting colonial volvocacean algae. J 57. Rad-Menéndez C, Gerphagnon M, Garvetto A, Arce P, Badis Y, Sime-Ngando T,
Eukaryot Microbiol. 2018;65:870–81. et al. Rediscovering Zygorhizidium affluens Canter: molecular taxonomy, infec-
30. Ishida S, Nozaki D, Grossart H-P, Kagami M. Novel basal, fungal lineages from tious cycle, and cryopreservation of a chytrid infecting the bloom-forming dia-
freshwater phytoplankton and lake samples. Environ Microbiol Rep. tom Asterionella formosa. Appl Environ Microbiol. 2018;84:e01826–18.
2015;7:435–41. 58. Canter HM. Fungal parasites of the phytoplankton. III. Trans Br Mycol Soc.
31. Padisák J, Scheffler W, Kasprzak P, Koschel R, Krienitz L. Interannual variability in 1954;37:111–33.
the phytoplankton composition of Lake Stechlin (1994-2000). Arch für Hydrobiol 59. Davis WJ, Picard KT, Antonetti J, Edmonds J, Fults J, Letcher PM, et al. Inventory of
Beih Adv Limnol. 2003;58:135–55. chytrid diversity in two temporary forest ponds using a multiphasic approach.
32. Holfeld H. Fungal infections of the phytoplankton: seasonality, minimal host Mycologia. 2018;110:811–21.
density, and specificity in a mesotrophic lake. N. Phytol. 1998;138:507–17. 60. Vélez CG, Letcher PM, Schultz S, Mataloni G, Lefèvre E, Powell MJ. Three new
33. Klawonn I, Dunker S, Kagami M, Grossart H-P, Van den Wyngaert S. Inter- genera in Chytridiales from aquatic habitats in Argentina. Mycologia. 2013;105:
comparison of two fluorescent dyes to visualize parasitic fungi (Chytridiomycota) 1251–65.
on phytoplankton. Microb Ecol 2021 https://doi.org/10.1007/s00248-021-01893-7 61. Lepère C, Domaizon I, Humbert J-F, Jardillier L, Hugoni M, Debroas D. Diversity,
34. Van den Wyngaert S, Seto K, Rojas-Jimenez K, Kagami M, Grossart H-P. A New spatial distribution and activity of fungi in freshwater ecosystems. 2019;7:e6247.
parasitic chytrid, Staurastromyces oculus (Rhizophydiales, Staurastromycetaceae fam. 62. Khomich M, Davey ML, Kauserud H, Rasconi S, Andersen T. Fungal communities
nov.), infecting the freshwater desmid Staurastrum sp. Protist. 2017;168:392–407. in Scandinavian lakes along a longitudinal gradient. Fungal Ecol. 2017;27:36–46.
35. Longcore JE. Rhizophydium brooksianum sp. nov., a multipored chytrid from soil. 63. Karpov SA, Mamanazarova KS, Popova OV, Aleoshin VV, James TY, Mamkaeva MA,
Mycologia. 2004;96:162–71. et al. Monoblepharidomycetes diversity includes new parasitic and saprotrophic
36. Montero-Pau J, Gómez A, Muñoz J. Application of an inexpensive and high- species with highly intronized rDNA. Fungal Biol. 2017;121:729–41.
throughput genomic DNA extraction method for the molecular ecology of zoo- 64. Koehler AV. S. Genetic and phenotypic influences on clone-level success and host
planktonic diapausing eggs. Limnol Oceanogr Methods. 2008;6:218–22. specialization in a generalist parasite. J Evol Biol. 2012;25:66–79.
37. Rehner SA, Samuels GJ. Taxonomy and phylogeny of Gliocladium analysed from 65. Bruning K. Effects of temperature and light on the population dynamics of the
nuclear large subunit ribosomal DNA sequences. Mycol Res. 1994;98:625–34. Asterionella-Rhizophydium association. J Plankton Res. 1991;13:707–19.
38. Vilgalys R, Hester M. Rapid genetic identification and mapping of enzymatically 66. Agha R, Gross A, Gerphagnon M, Rohrlack T, Wolinska J. Fitness and eco-
amplified ribosomal DNA from several Cryptococcus species. J Bacteriol. physiological response of a chytrid fungal parasite infecting planktonic cyano-
1990;172:4238–46. bacteria to thermal and host genotype variation. Parasitology. 2018;145:1279–86.
The ISME Journal
S. Van den Wyngaert et al.
13
67. Rohrlack T, Haande S, Molversmyr Å, Kyle M. Environmental conditions determine Maria Belyaeva, Erik Sperfeld, Maren Lentz, and Susanne Stephan for their assistance
the course and outcome of phytoplankton chytridiomycosis. PLOS ONE. 2015;10: during field sampling. Anusuya Willis is acknowledged for her assistance with the
e0145559. metagenome sequencing of single-cell Dolichospermum-MDA2-akinete.
68. Parratt SR, Laine A-L. The role of hyperparasitism in microbial pathogen ecology
and evolution. ISME J. 2016;10:1815–22.
69. Holt RD, Hochberg ME. The coexistence of competing parasites. Part II—Hyper- AUTHOR CONTRIBUTIONS
parasitism and food chain dynamics. J Theor Biol. 1998;193:485–95. SVdW, LG, MK and HPG conceptualized and designed the study. Lake sampling was
70. Canter HM. Studies on British Chytrids. XXIX. A taxonomic revision of certain fungi coordinated and performed by SVdW, MK and SAB. Pollen-bait experiment was
found on the diatom Asterionella. Bot J Linn Soc. 1969;62:267–78. performed by SVdW, RA and LG. DNA extractions of lake and pollen-bait samples
71. Canter-Lund HL. Freshwater algae: their microscopic world explored. Bristol, were performed by LG. JP analyzed phytoplankton. SVdW performed single cell
England: Biopress Ltd; 1995. isolation and cultivation and conducted growth assays and cross infection assays. KJR
72. Song P, Yi R, Tanabe S, Goto N, Seto K, Kagami M, et al. Temporal variation in and SVdW generated Sanger sequences, CW generated nanopore sequences and JW
community structure of zoosporic fungi in Lake Biwa, Japan. Aquat Micro Ecol. generated 18 S rRNA and 28 S rRNA sequences of single-cell Dolichospermum-MDA2-
2021;87:17–28. akinete from shotgun metagenome sequencing. Data was analyzed by SVdW, LG,
73. Winder M, Reuter JE, Schladow SG. Lake warming favours small-sized planktonic and KS. SVdW, LG, RA, MK and HPG drafted the manuscript. All authors provided
diatom species. Proc R Soc B Biol Sci. 2009;276:427–35. input for data interpretation, revised the manuscript, and approved the final version.
74. Lampert W, Sommer U. Limnoecology: The Ecology of Lakes and Streams. 2nd
edition. Oxford: Oxford University Press; 2007.
75. Graham MD, Vinebrooke RD, Turner M. Coupling of boreal forests and lakes:
Effects of conifer pollen on littoral communities. Limnol Oceanogr. FUNDING
2006;51:1524–29. Open Access funding enabled and organized by Projekt DEAL.
76. Rösel S, Rychła A, Wurzbacher C, Grossart H-P. Effects of pollen leaching and
microbial degradation on organic carbon and nutrient availability in lake water.
Aquat Sci. 2012;74:87–99. COMPETING INTERESTS
77. Kagami M, Motoki Y, Masclaux H, Bec A. Carbon and nutrients of indigestible The authors declare no competing interests.
pollen are transferred to zooplankton by chytrid fungi. Freshw Biol.
2017;62:954–64.
78. Masclaux H, Bec A, Kagami M, Perga M-E, Sime-Ngando T, Desvilettes C, et al. ADDITIONAL INFORMATION
Food quality of anemophilous plant pollen for zooplankton. Limnol Oceanogr. Supplementary information The online version contains supplementary material
2011;56:939–46. available at https://doi.org/10.1038/s41396-022-01267-y.
79. Karpov SA, Mamkaeva MA, Aleoshin VV, Nassonova E, Lilje O, Gleason FH. Mor-
phology, phylogeny, and ecology of the aphelids (Aphelidea, Opisthokonta) and Correspondence and requests for materials should be addressed to Silke Van den
proposal for the new superphylum Opisthosporidia. Front Microbiol. 2014;5:112. Wyngaert, Maiko Kagami or Hans-Peter Grossart.
80. Seto K, Matsuzawa T, Kuno H, Kagami M. Morphology, Ultrastructure, and
molecular phylogeny of Aphelidium collabens sp. nov. (Aphelida), a parasitoid of a Reprints and permission information is available at http://www.nature.com/
green alga Coccomyxa sp. Protist. 2020;171:125728. reprints
81. Gsell AS, de Senerpont Domis LN, van Donk E, Ibelings BW. Temperature alters
host genotype-specific susceptibility to chytrid infection. PLoS ONE. 2013;8: Publisher’s note Springer Nature remains neutral with regard to jurisdictional claims
e71737. in published maps and institutional affiliations.
82. Winder M, Schindler DE. Climatic effects on the phenology of lake processes.
Glob Change Biol. 2004;10:1844–56.
83. Shatwell T, Köhler J, Nicklisch A. Warming promotes cold-adapted phytoplankton
in temperate lakes and opens a loophole for Oscillatoriales in spring. Glob
Change Biol. 2008;14:2194–200. Open Access This article is licensed under a Creative Commons
84. Yankova Y, Neuenschwander S, Köster O, Posch T. Abrupt stop of deep water Attribution 4.0 International License, which permits use, sharing,
turnover with lake warming: Drastic consequences for algal primary producers. adaptation, distribution and reproduction in anymedium or format, as long as you give
Sci Rep. 2017;7:13770. appropriate credit to the original author(s) and the source, provide a link to the Creative
85. Cavaliere E, Fournier IB, Hazuková V, Rue GP, Sadro S, Berger SA, et al. The Lake Commons license, and indicate if changes were made. The images or other third party
Ice Continuum Concept: Influence of winter conditions on energy and ecosystem material in this article are included in the article’s Creative Commons license, unless
dynamics. J Geophys Res Biogeosciences. 2021;126:e2020JG006165. indicated otherwise in a credit line to the material. If material is not included in the
article’s Creative Commons license and your intended use is not permitted by statutory
regulation or exceeds the permitted use, you will need to obtain permission directly
ACKNOWLEDGEMENTS from the copyright holder. To view a copy of this license, visit http://creativecommons.
This work was supported by IGB Postdoc Fellowship and the German Research org/licenses/by/4.0/.
Foundation project WY175/1-1 to SVdW, WU 890/2-1 to CW, AG 284/1-1 to RA and
GR1540/30-1 to HPG and Leibniz SAW project “MycoLink” (SAW-2014-IGB). MK was
supported by JSPS KAKENHI (15KK0026, 16H02943, 19H05667). We are grateful to © The Author(s) 2022
The ISME Journal