Ducellieriales ord. nov. and evidence for a novel clade of endobiotic pollen-infecting “lagenidiaceous” Peronosporomycetes

Abstract

The genus Ducellieria (Ducellieriaceae) contains three species (D. chodatii, D. tricuspidata, D. corcontica), and a single variety (D. chodatii var. armata) of obligate endobiotic pollen parasites. These organisms have been first assigned to the green alga genus Coelastrum, as they form very similar spherical structures, but the observation of heterokont zoospores has led to their reclassification to the phylum Oomycota. However, despite their widespread nature, these organisms are only known from their descriptive morphology, and life cycle traits of some species still remain incompletely known. Only the type species, D. chodatii, has been rediscovered several times, but the phylogeny of the genus remains unresolved, since none of its species has been studied for their molecular phylogeny. At present the genus is still included in some algal databases. To clarify the evolutionary affiliation of Ducellieria, efforts were undertaken to isolate D. chodatii from pollen grains, to infer its phylogenetic placement based on nrSSU sequences. By targeted isolation, the pollen endoparasitoid was rediscovered from three lakes in Germany (Mummelsee, Okertalsperre, Knappensee). Apart from the typical coelastrum-like spheroids, oomycetes sporulating directly from pollen grains in a lagenidium-like fashion were observed, and molecular sequences of both types of oomycetes were obtained. Phylogenetic reconstruction revealed that coelastrum-like and lagenidium-like forms are unrelated, with the former embedded within the deep branching early-diverging lineages, and the later stage forming a distinct clade in Peronosporales. Consequently, the life cycle of D. chodatii needs careful revision using single-spore isolates of the species, to infer if previous lifecycle reconstructions that involve various different thallus types are stages of a single species or potentially of several ones.

Citation: Buaya AT, Thines M (2023). Ducellieriales ord. nov. and evidence for a novel clade of endobiotic pollen-infecting “lagenidiaceous” Peronosporomycetes. Fungal Systematics and Evolution 12: 247–254. doi: 10.3114/fuse.2023.12.12

INTRODUCTION

Freshwater holocarpic oomycetes are diverse and widespread (Karling 1942, Sparrow 1960, Dick 2001, Buaya & Thines 2020). They are mostly obligate biotrophic endobiotic parasites, infecting various hosts, including diatoms (Sparrow 1960, Buaya et al. 2019a, Buaya & Thines 2019b, 2020, 2021, Thines & Buaya 2022), filamentous macroalgae (Zopf 1884, de Wildeman 1893, 1896, 1897, Scherffel 1925), cyanobacteria (Ingold 1949), plants (Zopf 1887, Karling 1981), oomycetes (Cornu 1872, Barrett 1912, McLarty 1941, Whiffen 1942, 1946, Sparrow 1950, Buaya et al. 2019d), and invertebrate animals (Sparrow 1939, Drechsler 1940, Karling 1944, Barron 1980, Glockling & Beakes 2000). Most of the species belong to the early-diverging lineages of oomycetes (e.g. Diatomophthora, Olpidiopsis), while others are in the crown clades of Saprolegniomycetes (e.g. Aphanomycopsis, Ectrogella) and Peronosporomycetes (e.g. Lagena, Lagenidium) (Karling 1942, Sparrow 1960, Dick 2001, Buaya & Thines 2000). But despite recent progress, the taxonomic affiliations of most freshwater holocarpic oomycetes remain obscure, since many species have not been investigated with respect to their molecular phylogeny (Beakes & Thines 2017, Buaya & Thines 2020).

In particular, the pollen-parasitic oomycetes, of which less than a dozen species have been reported, remain unresolved, despite their widespread occurrence and potential ecological role for carbon cycling in freshwater environments (Kagami et al. 2014, Masigol et al. 2019). Records of pollen inhabiting oomycetes are known from both gymnosperm and angiosperm plants, and almost none of them have been rediscovered, to date (Sparrow 1960). These include two varieties classified in the genus Lagenidium, (L. pygmaeum var. pygmaeum, L. pygmaeum var. pygmaeoides) (Zopf 1887, Karling 1981), one each from Aphanomycopsis (A. saprophytica) (Karling 1968), and Anisolpidium (A. saprobium) (Karling 1968), a handful of enigmatic species classified into genus Ducellieria (D. chodatii, D. corcontica, D. tricuspidata) (Teiling 1957, Kusel Fetzmann & Nouak 1981, Hesse et al. 1989, Matula 1980), and the provisional genus Lagenidicopsis (L. arctica) (Artemchuk 1972). Dick (2001) subsequently reclassified some of these parasites, and reassigned L. pygmaeum to Cornumyces, L. pygmaeum var. pygmaeoides to Pleocystidium, A. saprophytica to Hyphochytrium, and introduced the family Ducellieriaceae to accommodate the pine pollen pathogen D. chodatii within Oomycota. However, there are no molecular data supporting the placement of these species, including also the type species of the genus Ducellieria, D. chodatii, which is still sometimes listed as a member of the yellow-green algae family Botryochloridaceae of the class Xanthophyceae (Teiling 1957, Coute 1984) in algae databases (e.g. AlgaeBase) and papers in phycology (Wehr & Sheath 2003, Ettl 2009, Godínez Ortega 2017).

The genus Ducellieria was described by Teiling in 1957. It currently comprises three species, its types species, Ducellieria chodatii, and with two additional species, D. corcontica and D. tricuspidata (Matula 1980), along with a variety, D. chodatii var. armatum. In addition, one species, D. bicuspidata, has not been validly described, as no type specimen had been indicated (Oblinger 2005). In any case, the description provided does not contain enough detail to infer if it can be considered a member of the genus. Ducellieria is primarily characterized by forming colourless, multicellular, coelastrum-like spherical aggregates, connected by hollow spines, and outward-directed spines (Kusel Fetzmann & Nouak 1981, Hesse et al. 1989). The spherical aggregate is formed at the tip of a discharge tube from an endobiotic non-septate holocarpic thallus (Kusel Fetzmann & Nouak 1981, Hesse et al. 1989, Stoyneva et al. 2013). Ducellieria chodatii recurs seasonally in parallel to the peak of the pollen bloom of Pinus and Picea species, but has only rarely been reported (Kusel Fetzmann & Nouak 1981, Hesse, Kusel Fetzmann & Carniel 1989, Gorbulin 2012, Stoyneva et al. 2013, Bancsó 2023).

Ducellieria with its type species, D. chodatii, has been reclassified multiple times into different taxonomic lineages. Originally described as Coelastrum chodatii by Ducellier (1915) in the green algae family Chlorophyceae (Ducellier 1915, Bourelly 1968, Komarek & Fott 1983), the parasite has been reclassified into Ducellieria chodatii by Teiling (1957) with its own genus in the yellow-green algae family Xanthophyceae containing three species (D. chodatii, D. corcontica, D. tricuspidata) (Teiling 1957, Matula 1980), and one variety, D. chodatii var. armata (Teiling 1957), of which only D. chodatii has been proven to be an oomycete by electron microscopy (Hesse et al. 1989). Subsequently, a detailed re-examination of its life cycle and cellular ultrastructure confirmed its identification as an oomycete, and it was initially placed in the Saprolegniales (Hesse et al. 1989). Dick (2001) reclassified the species into its own family (Ducellieriaceae), provisionally assigning it to the Leptomitales, but excluding D. tricuspidata and D. chodatii var. armata.

In order to resolve the phylogeny of the type species of Ducellieria, attempts were made to isolate this oomycete from different freshwater environments and geographic locations in Germany. While screening for holocarpic oomycete parasites from pollen grains from the lakes Mummelsee, Okertalsperre, and Knappensee, the species was rediscovered, and it was the aim of this study to clarify its phylogenetic placement.

MATERIALS AND METHODS

Isolation, characterisation, and host and parasite dual culture attempts

Dense pollen films (predominantly from Pinus sylvestris and Picea abies) were collected from shores of the lake Mummelsee, in the state of Baden-Württemberg, Southwest Germany (48°35’53.1”N, 8°12’03.9”E), the reservoir Okertalsperre in the state of Sachsen-Anhalt, Central Germany (51°51’02.9”N, 10°27’33.0”E), the lake Knappensee (50°26’27.3”N, 8°53’55.2”E), and the pond Forbachteich (50°22’47.2”N, 8°36’45.1”E) in the state of Hessen, Central Germany. Pollen clusters with living aggregates of Ducellieria chodatii were collected in June 2020 from Mummelsee, Knappensee, and Forbachteich, and in June 2021 from Mummelsee and Okertalsperre, by directly collecting masses of floating pollen grains using multiple 50 mL sterile plastic tubes (Sarstedt, Nümbrecht, Germany). About 10 mL of pollen concentrates were poured onto several Petri dishes (150 × 20 mm) (Sarstedt, Nümbrecht, Germany) on the day of collection, and 100 mL of autoclaved distilled water were added to dilute the samples. Subsequently, samples were screened for D. chodatii using an inverted compound light microscope (AE31, Motic, Xiamen, China) and were individually picked using a 10 µL pipette (Brandt, Wertheim, Germany), rinsed multiple times in autoclaved deionized water, and transferred to 2 mL tubes containing 0.5 mL RNALater solution (Invitrogen, Thermo Fisher, Lithuania) or 70 % ethanol (VWR, France) for subsequent DNA extraction. Approximately 100 spherical coelastrum-like aggregates, and pine pollen grains with endobiotic oomycete thalli were collected in this manner for DNA extraction. Samples preserved in 70 % ethanol were deposited in the herbarium collection of the Senckenberg Museum of Natural History (Herbarium Senckenbergianum, FR), Cryptogams Section, Frankfurt am Main (accession numbers (Forbachteich 2020 isolate FR-0046136, Knappensee 2020 isolate FR-0046137) for pollen grains with endobiotic thalli, and (Mummelsee 2021 isolate FR-0046135, Okertalsperre 2021 isolate FR-0046156) for coelastrum-like spheroids). Morphological characterisation of the parasite was done as described earlier (Buaya et al. 2022) using a compound light microscope (Imager2, Carl Zeiss Göttingen, Germany) with DIC, and photographs were taken using a Zeiss Axiocam MRc5 (Carl Zeiss, Göttingen, Germany), or an SLR digital camera (EOS 500D, Canon, Tokyo Japan) for life cycle observations, mounted on the inverted compound light microscope. To inhibit motility of freshly released zoospores for microscopic photography, small amounts of Sorbitol (Carl Roth GmbH, Karlsruhe, Germany) solution (2 mg/mL) were added to the slides.

Parasite cultures with pollen grains were also attempted using living spherical aggregates of D. chodatii and pine pollen directly collected from pine trees (P. sylvestris) in 15 mL Petri dishes (60 × 15mm), with the addition of 50 µg/mL ampicillin or 100 µg/mL rifampicin (Carl Roth GmbH, Karlsruhe, Germany) and 50 mg/L Benomyl (Edgington et al. 1971) (Sigma-Aldrich Chemie GmbH, Steinheim, Germany) to control unwanted microbial contaminants. Culture replicates were incubated in a climate chamber (CMP 6010, Conviron, Canada) at 16 °C and 12 °C, 14 h in light and in 10 h darkness (1 000 lux, Narva, bio-vital, Germany), respectively, per day cycle. To these cultures, new pine pollen and autoclaved water were added regularly. However, long term culture trials to obtain single spore cultures from living spherical aggregates of D. chodatii were unsuccessful.

DNA extraction, PCR amplification, and phylogenetic analyses

DNA extraction was performed using an innuPREP Plant DNA extraction Kit (analytikjena, Jena, Germany), as previously described (Buaya et al. 2017) on isolated living spherical coelastrum-like aggregates and endobiotic oomycete thalli. Initially, the collected thalli were centrifuged at maximum speed (19 000 g) for 2 min at 22 °C to concentrate the cells. Subsequently, RNALater or 70 % ethanol was carefully removed using 1 000 µL pipette tips, and 400 µL SLS buffer from the extraction kit was added. About 100 mg of sterile 0.1 mm Silica Glass Beads (Carl Roth GmbH, Karlsruhe, Germany) were added into each 2 mL tube (Sarstedt, Nümbrecht, Germany) and the samples were homogenized at 25 Hz for 25 min in a Retsch Mixer Mill MM 200 (Retsch GmbH, Haan, Germany). DNA extraction of all samples was conducted following the manufacturer’s instructions of the innuPREP Plant DNA extraction Kit. The PCR amplification of the partial nuclear encoding small subunit (18S; nrSSU) was performed as described in Buaya et al. (2019a) using MangoTaq™ DNA Polymerase (Bioline, London, UK) with the 18S primer pair EUK422-445 and EUK1422-1440_R (Wang et al. 2014) on all samples. Subsequently, all positive amplicons were sent for sequencing to the laboratory centre of the Senckenberg Biodiversity and Climate Research Centre, Frankfurt am Main (SBiK-F, Frankfurt, Germany) using the 18S primers used in the PCR. In addition, direct PCRs (putting cells of the oomycete directly into PCR tubes without prior DNA extraction) using 18S primers (Wang et al. 2014) were also done as described in Buaya et al. (2019a). To obtain high quality nrSSU sequence data for D. chodatii coelastrum-like aggregates, PCR amplicons were cloned into competent Escherichia coli (Agilent Technologies, Santa Clara, United States) using a StrataClone TA cloning kit (Agilent Technologies, Santa Clara, United States) following instructions of the manufacturer. Single bacterial colonies were picked into 20 µL molecular grade water (Life Technologies, USA) and colony PCR was carried out with the MangoTaq™ DNA Polymerase using M13-F and M13-R plasmid primers with amplification conditions set to an initial denaturation at 96 °C for 10 min, 36 cycles at 96 °C for 20 s, 56 °C for 20 s and 72 °C for 60 s, and concluding with a final elongation at 72 °C for 10 min. Amplicons were sent for sequencing to the laboratory centre of the Senckenberg Biodiversity and Climate Research Centre (Frankfurt am Main, Germany) using M13 (M13-F, M13-R), T7 and T3 plasmid primers.

The resulting sequences were edited using Geneious v. 5.6, and the assembled sequences of D. chodatii were aligned together with sequences of various members of the Pythiales, Saprolegniales, and early-diverging oomycete lineages, using MAFFT v. 7 (Katoh & Stanley 2013), employing the Q-INS-i algorithm. The final alignment can be found on FigShare (doi: 10.6084/m9.figshare.24185379). Minimum Evolution phylogenetic inference with 1 000 bootstrap replicates was computed using MEGAX (Kumar et al. 2018) with pairwise deletion, the Tamura-Nei substitution model, and all other parameters set to default. Phylogenetic analyses using the GTR model were done on the TrEase webserver (http://thines-lab.senckenberg.de/trease/, Mishra et al. 2023) using RAxML v. 8 (Stamatakis 2014) for Maximum Likelihood inference with 1 000 bootstrap replicates. Bayesian inference was done on the same server using MrBayes v. 3.2 (Ronquist et al. 2012) and 5 M generations, while other parameters were set to default. Partial nrSSU sequences obtained in this study were deposited in GenBank under the accession numbers given in the phylogenetic tree.

RESULTS

Screening and observation

Floating pollen collected during spring of 2020 and 2021 from Mummelsee, Knappensee, and Okertalsperre in Germany were found to contain abundant Pinus sylvestris and Picea abies pollen grains, living spherical coelastrum-like aggregates of Ducellieria chodatii, and endobiotic oomycete thalli in the pollen grains. Spherical aggregates of D. chodatii were highly abundant on pollen samples from Mummelsee and Okertalsperre, but very rare in samples from Knappensee, while pollen from all sites contained dense clusters of endobiotic oomycete thalli and epibiotic thalli of monocentric fungi (e.g. Chytridiomycota), as well as filamentous oomycetes (e.g. Aphanomyces, Lagenidium, Pythium). All samples were further incubated for 2–4 wk under controlled conditions, with the addition of freshly collected pollen of Pinus sylvestris. Every week, about 50 mL of autoclaved deionised water and pine pollen were added to prevent desiccation and to provide continuous substrate for D. chodatii. These culture plates were screened daily, and after more than a week, spherical coelastrum-like aggregates of D. chodatii became more abundant, especially on Mummelsee samples. Two weeks after this, the amount of coelastrum-like aggregates declined rapidly and finally they disappeared from all culture plates. In all samples both colonial and single-celled varieties of algae were common. During zoospores release, some protists become attracted to the newly released mass of zoospores, centrally located within the cage of empty aggregates of pentagonal or triangular shaped cells (as seen from above or the side, respectively). Often, the cage provided an effective barrier against the external predators, protecting the newly released zoospores, before they attained full motility. However, sometimes other unicellular eukaryotes penetrated into the cage after multiple attempts, but frequently remained trapped, and sometimes died within the cage.

Axenic host and pathogen dual cultures of D. chodatii were attempted using freshly collected Pinus sylvestris pollen. For this living spherical coelastrum-like aggregates were isolated and inoculated into freshly collected and autoclaved pollen grains of Pinus sylvestris. Subsequently, cultures were inspected daily for 1 wk for the presence of the pathogen and production of coelastrum-like spherical aggregates. However, diverse microbial contaminants, especially filamentous oomycetes e.g. Aphanomyces, Lagenidium, Pythium and fungal organisms e.g. chytrids, filamentous ascomycetes, and yeasts, quickly colonised all the substrates, despite application of antibiotics, and no coelastrum-like aggregates of D. chodatii were observed after more than 2 wk of incubation in multiple attempts. In addition, growth of the pathogen was neither observed on autoclaved pollen grains nor on pollen grains treated with freezing at -80 °C after more than 2 wk of incubation.

While the development of the endobiotic thalli (Fig. 1A) seemed to show features of several genera, including Aphanomyces, Ducellieria, and Lagenidium, the development of the coelastrum-like form (Fig. 1A, B) was always in line with previous descriptions of D. chodatii (Kusel-Fetzmann & Nouak 1981, Kusel-Fetzmann & Carniel 1984, Hesse et al. 1989, Stoyneva et al. 2013). As we assume that multiple species of oomycetes were causing the endobiotic thalli in this study, only the coelastrum-like form is described here. The number of cells in a spherical aggregate varied, likely reflecting the nutritional state of the thallus from which they were formed, and, thus, probably depending on the size of the pollen grains consumed. Pinus pollen grains, which are much smaller than those of Picea, usually produced smaller aggregates. The hollow coelastrum-like aggregates from Picea pollen grains usually consisted of more than 20 cells, with a diameter of about 50–80 µm (Fig. 1B, C). The warty to papillate individual cells usually measured 10–15 µm diam, as seen from above, each cell forming 4–6 connections to neighbouring cells. Cells in aggregates were hyaline, at maturity contained fine granules with mixtures of larger droplets (Fig. 1B). In conventional light microscopy, the spore ball sometimes appears light yellow or faint yellow green. After some time, each cell of the aggregate developed a single zoospore almost simultaneously, which escaped into the inner side of the sphere. The newly released spores usually remained in the hollow centre of the sphere for a while, initially weakly moving inside the sphere. Subsequently, zoospore motility became more intense, and after 10–15 min zoospores swam away in an irregular motion, escaping through from the cage formed by the cell connections, leaving an empty spherical aggregate behind (Fig. 1C). The zoospores were about 8–13 µm long and 7–10 µm diam, broadly limoniform to pyriform or reniform, biflagellate with two subapically inserted flagella, of which one was forward-directed (Fig. 1D, red arrow). The heterokont zoospores usually swam for several minutes before they settled down, becoming aflagellate and spherical in shape, surrounded by thin cell wall (Fig. 1E). Sometimes, the spherical resting cysts fragmented into minute spherical globules (Fig. 1F, G) after several minutes leaving behind empty cysts (Fig. 1F, red arrows).

Fig. 1.

DIC light micrographs of parasites of coniferous pollen. A. Representative of an endobiotic parasite thallus within a Pinus sp. pollen grain. B–G. Development of the coelastrum-like state of Ducellieria chodatii, at different stages B. Fully developed spherical aggregate. C. Spherical aggregate after most zoospore have been released. D. Biflagellate zoospores (red arrow pointing to the forward-directed tinsel flagellum). E. Zoospores resting after encystation. F. Empty cysts with scattered granular globules (F, G) after disintegration. Scale bars: A–C = 50 µm; D–G = 20 µm.

Molecular phylogeny

In the phylogenetic reconstructions inferred from partial nrSSU sequences (Fig. 2), the coelastrum-like form of Ducellieria chodatii and the endobiotic thalli belonged to different oomycete lineages, in line with the observation that in the samples obtained most endobiotic thalli showed a lagenidium-like spore discharge. Ducellieria chodatii coelastrum-like isolates had identical sequences and formed a monophyletic group with maximum support. This group was resolved as a member of the early-diverging oomycete lineages, forming separate, deep-branching clade. The clade was grouped with Olpidiopsidales and Miraculales, but without support. The lagenidium-like thalli were imbedded within the crown oomycete order Peronosporales with lacking to weak support and no clear affinities to any of the genera of the Peronosporales sequenced so far.

Fig. 2.

Molecular phylogeny using Minimum Evolution inferred from partial nrSSU sequences. The numbers at the branches denote bootstrap support values from Maximum Likelihood and Minimum Evolution analyses, in respective order. A hyphen (-) indicates less than 60 % bootstrap support or a conflicting topology. The third number refers to posterior probabilities equal to or greater than 0.95 from the Bayesian phylogenetic inference. Also in this case, a hyphen (-) sign denotes lack of support for the presented or an alternate topology.

TAXONOMY

Based on the life-cycle traits and phylogenetic placement, Ducellieria chodatii is a bona fide member of the early-diverging oomycete lineages. Thus, the provisional placement of the parasitoid in the Leptomitales by Dick (2001) cannot be upheld, and the assignment is revised in the present study by reclassifying D. chodatii to an order of its own.

Ducellieriales A.T. Buaya & Thines, ord. nov. MycoBank MB 849453.

Obligate biotrophic parasite of Pinus and Picea pollen grains, thallus holocarpic, forming hollow sphere at discharge, which at maturity forms a multicellular hollow spherical aggregate, individual cells with a thin, colourless wall, ornamented with warts and spines directed outwards; zoospores single per individual cell, biflagellate and heterokont.

Type genus: Ducellieria Teiling

Type species: Ducellieria chodatii (F. Ducell.) Teiling, Svensk bot. Tidskr. 51: 209. 1957.

The typification of Dick (2001) was made by choice of an iconotype, the figures in Hesse et al. (1989). Thus, no molecular phylogenetic investigation of the type can be done to clarify if morphologically similar and divergent species are conspecific. Therefore, FR-0046135 (Germany, Baden-Württemberg, Black Forest, Mummelsee, 14 Jun. 2021, M. Thines, ex type partial nrSSU sequence deposited in GenBank under thre accession number OR282458) deposited in the Herbarium Senckenbergianum (FR), is here designated as an epitype (MBT 10014079).

DISCUSSION

Holocarpic parasites of plant pollen are widespread in freshwater aquatic environments and ecologically important for carbon cycling (Karling 1942, Sparrow 1960, Dick 2001). Despite their ubiquitous nature, none of these organisms have been investigated for molecular phylogeny, leaving their taxonomic placement largely unresolved (Beakes & Thines 2017, Buaya & Thines 2020). Arguably the most enigmatic of these organisms belong to the genus Ducellieria (D. chodatii, D. chodatii var. armata, D. corcontica, D. tricuspidata), which forms a unique coelastrum-like sporangial state that has led to a classification of the pathogen in green and later golden-green algae (Ducellier 1915, Teiling 1957, Karling 1968, Artemchuk 1972, Matula 1980).

Originally, the type species of the genus was classified as Coelastrum chodatii in the Chlorophyceae by Ducellier (1915). It was later transferred to a new genus in the Xanthophyceae, Ducellieria, by Teiling (1957). However, the placement of D. chodatii among algae was questioned by subsequent investigators (Kusel-Fetzmann & Nouak 1981, Kusel-Fetzmann & Carniel 1984) due to the lack of plastids, absence of solid connection between cells (unlike Coelastrum), and other traits of its complex life cycle evidently divergent from algae (e.g. the formation of a holocarpic thallus). After detailed re-examination of its life cycle and cellular ultrastructure, its affiliation to the Oomycota was confirmed by Hesse et al. (1989). The pathogen was initially thought to be a member of the Saprolegniales (Kusel-Fetzmann & Nouak 1981). However, Dick (2001) reclassified the species to the order Leptomitales in separate family (Ducellieriaceae), excluding other related species originally included by Teiling (1957), due to the difference in the structure of connections between cells (D. chodatii var. armata), or incomplete life cycle observations by the original author (D. tricuspidata).

The isolate investigated in the present study agrees well with the description of previous investigators (Kusel Fetzmann & Nouak 1981, Hesse et al. 1989, Stoyneva et al. 2013), in terms of morphology and parasitism to coniferous plant pollen. However, it is noteworthy that the coelastrum-like forms co-occurred with a widespread lagendium-like form and several other pollen parasites of both chytrids and oomycetes (e.g. an aphanomyces-like pathogen). The possibility of confusing D. chodatii with other species has already been pointed out before (Kusel Fetzmann & Nouak 1981, Kusel Fetzmann & Carniel 1984, Hesse et al. 1989, Stoyneva et al. 2013), but it also seems possible, that forms attributed to the life cycle of D. chodatii might actually represent other oomycetes or fungi, since drifting gymnosperm pollen harbours a great diversity of fungi and oomycetes (Sparrow 1960, Sparrow 1968, Czeczuga & Muszyńska 2004, Wurzbacher et al. 2014, Van den Wyngaert et al. 2022), as we have also observed in all our samples. Due to the cage-like nature of the aggregates it seems possible that even isolations of single spheroids might contain additional organisms able to infect pollen grains. Thus, until dual cultures using single zoospore isolates confirm the various stages previously reported, only cycle “A” and “B” from Hesse et al. (1989) featuring the coelastrum-like phase and the fact that thalli can develop within pollen grains can be considered doubtless, while variations of the endobiotic stage and the additional stages (A1, B1, D1) found by Stoyneva (2013) should be confirmed with single spore isolates. This might not always be easy, as it is conceivable, that the formation of these stages depends on environmental conditions. Also, the resting stage reported (e.g. Hesse et al. 1989) should be considered as not conclusively belonging to D. chodatii, as Hesse et al. (1989) did not succeed to trigger the formation of coelastrum-like forms from these. The exclusion of D. tricuspidata (basionym: Coelastrum tricuspidatum), and D. chodatii var. armata (basionym: Coelastrum augustae var. armata) from Ducellieria by Dick (2001) also requires re-evaluation. Considering the similarities with respect to the formation of a coelastrum-like stage, these species are probably better retained in the genus until a re-evaluation of their life cycle and molecular phylogenetic reconstructions can clarify their evolutionary affinities.

Conflict of interest

M. Thines is a senior editor of the journal. However, he was not involved in the handling or review process of this manuscript.