Full paper
Unguispora grylli, a new species of amphibious fungi associated with crickets (Gryllidae), transforms attachment structures of sporangiola in the host gut.
2025 Volume 66 Issue 3 Pages 162-170
Details
2025 Volume 66 Issue 3 Pages 162-170
Amphibious fungi exhibit both saprophytic and arthropod gut symbiotic lifestyles and are ideal materials for investigating how these two different lifestyles have evolved in Kickxellomycotina. We herein added a new species of amphibious fungi, Unguispora grylli, to a monotypic genus Unguispora. This species was found in the proventriculus (foregut) of several genera of Gryllidae and on their feces. It is distinguished from U. rhaphidophoridarum in that it has a different host, and the number and morphology of the claws of sporangiola also differ. After the host ingested sporangiola, we observed that the upper half of the sporangiola disappeared in the gut and that the claws of the sporangiola, which function as attachment structures in the proventriculus, underwent morphological changes. The claws transformed to bottle opener-like structures arranged in chains vertically with hollowed-out centers. Fine hairs on the inner surface of the proventriculus became entangled with these structures and were stuck in the gaps between the denticles of the claws, leading to attachment to the host gut. In this article, zygospores were reported for the first time in Unguispora, which represents the fourth zygospore-confirmed genus in Kickxellales.
Kickxellomycotina is one of the basal lineages of the fungi and constitutes Zoopagomycota along with Zoopagomycotina (parasites of fungi and small animals) and Entomophthoromycotina (insect parasites) (Hibbett et al., 2007; Spatafora et al., 2016). Phylogenomic analyses including these taxa found that Kickxellomycotina and Zoopagomycotina are both monophyletic groups, while Entomophthoromycotina is polyphyletic, with Basidiobolus Eidam and other taxa each forming distinct monophyletic clades (Davis et al., 2019; Li et al., 2021; Wang et al., 2023). In these studies, monophyly of Zoopagomycota excluding Basidiobolus was supported, but the phylogenetic position of Basidiobolus, whether inside or outside of Zoopagomycota, remains unresolved.
Kickxellomycotina contains eight orders, and their lifestyles vary: mycoparasites (Dimargaritales), saprobes (Kickxellales, Ramicandelaberales, Spiromycetales), and arthropod gut symbionts (Asellariales, Barbatosporales, Harpellales, and Orphellales) (Tretter et al., 2014; White et al., 2018). Even though phylogenetic analyses suggest that the origin of gut symbiosis could have stemmed from a single event or up to three independent occurrences (Reynolds et al., 2023; Tretter et al., 2014), how these lifestyles evolved remains controversial due to the lack of evidence aside from the phylogenetic relationships. Recently, we proposed a new ecological group, amphibious fungi, named so because they have growth stages both inside and outside the arthropod gut (Ri et al., 2022). This group is intermediate between gut symbionts and terrestrial saprobes of excrement or exuviae and are the key to understanding the transitions between gut symbionts and saprophytes (Ri et al., 2022).
When we proposed the concept, an amphibious fungus was newly described as Unguispora rhaphidophoridarum Ri & Degawa. The species has distinctive asexual reproductive structures consisting of sporocladia, pseudophialides, and unispored sporangiola and thereby belongs to Kickxellales, which was recognized as a saprophytic clade. The genus Unguispora Ri & Degawa is siter to Linderina Raper & Fennell., producing unicellular and dome-shaped sporocladia that are peculiar to Kickxellales (Chang, 1967; Raper & Fennell, 1952). In contrast, the asexual reproductive structures in Unguispora resemble those of Coemansia Tiegh. & G. Le Monn. and many other genera in Kickxellales in appearance except for sterile appendages on its sporocladia. However, Unguispora and Linderina are identical in that their sporangiola contain a space at the apex, and their sporangiospores germ at the base to elongate downward (Ri et al., 2022).
Compared to known gut-symbiotic fungi, Unguispora represents two novelties, namely attachment structures to host gut and host orders. The genus attaches to the host foregut by the claw-like ornamentation on the surface of sporangiola, whereas known gut-symbiotic fungi attach to the midgut or hindgut of their host by acellular or cellular holdfasts, which are typically made with adhesive substances secreted from their sporangiospores (Lichtwardt et al., 2001). The host Orthoptera of the genus represents also the first order that has been reported among known gut-symbiotic fungi. Additionally, terrestrial hosts are only known in Unguispora and Asellariales among Kickxellomycotina although most gut-symbiotic fungi of Kickxellomycotina live within aquatic hosts. Whether the habitat of fungal hosts is aquatic or terrestrial reflects the morphology of gut-symbiotic fungi. For example, trichospores of aquatic gut fungi (Barbatosporales, Harpellales, Orphellales) are generally entangled to the vegetation or other substrates with appendages or accompanying terminal cells to withstand the flow until being ingested with the substrates by the next host, but the spores of terrestrial gut fungi (most of Asellariales) lack them (Lichtwardt et al., 2001; Valle & Santamaria, 2005). The shape of zygospores is also distinct among aquatic and terrestrial gut-fungi; it is conical or biconical in Harpellales and helicoidal in Orphellales but spherical in Asellariales and Kickxellales, which is likely adaptive to their habitats (Lichtwardt et al., 2001; Valle & Cafaro, 2008; Valle & Santamaria, 2005). In Unguispora, the appendage and accessory cells of the sporangiola are absent, but zygospores are still unknown.
Unguispora rhaphidophoridarum was discovered on the feces and in the gut of camel crickets (Orthoptera; Rhaphidophoridae) (Ri et al., 2022). However, the arthropod feces have been given less attention as isolation sources in Kickxellomycotina. Most species of saprobes and mycoparasites in the subphylum have been reported from soil or mammalian dung. There are few reports of isolating Kickxellomycotina fungi from arthropod feces (Kurihara & Degawa, 2006; Takashima et al., 2022). Furthermore, gut-symbiotic fungi of Kickxellomycotina predominantly live in larval aquatic insects, and the terrestrial hosts were limited to Collembola and Isopoda (Lichtwardt et al., 2001). Hence, we anticipate that there is a wealth of hidden diversity of Kickxellomycotina fungi associated with the feces or digestive tract of terrestrial arthropods. While investigating insects closely related to camel crickets, we discovered an undescribed species of Unguispora from crickets of the family Gryllidae. In this article, we describe it as Unguispora grylli and report all the stages of its life cycle.
Adult or nymph crickets were collected from grassy areas in parks, along roadsides, or on footpaths between rice fields in Nagano and Niigata prefectures during summer and autumn 2019, 2020, and 2022. Crickets were collected by hand using plastic containers and kept individually in the laboratory to obtain their excrement. Identification of crickets was based on Machida (2016). Fungal isolation from the excrement of crickets followed Ri et al. (2022). We prepared cultures for observation and photography under a light microscope by incubating them on half-strength malt-extract yeast-extract agar (1/2 ME-YE agar; Kurihara et al., 2000) for 7 d under anaerobic conditions, or 1 to 21 d under aerobic conditions. Materials were mounted in lactic acid (LA) or lacto-fuchsin (LF; acid fuchsin 0.1 g, lactic acid 100 mL; Carmichael, 1955). Zygospore formation was observed by maintaining the cultures on 0.3% shrimp agar (ShA; Degawa & Tokumasu, 1997) until 60 d. All light microscopic observations were carried out by an Olympus BX53 (Olympus, Tokyo, Japan) with differential interference contrast equipment. The sporangiola in culture were also observed with an SM-300 scanning electron microscope (SEM) (TOPCON, Tokyo, Japan). For material preparation, a suspension of the sporangiola in distilled water was filtered with Whatman NucleporeTM Track-Etched Membrane, size 25 mm, pore size 0.2 µm (Cytiva, Tokyo, Japan) and the membrane filter was cut into 5 mm square pieces. The sequence of steps from fixation to SEM observation was the same as described in Ri et al. (2022). Observations of the yeast-like growth stage within the host gut were conducted with both light microscopy and SEM. Adult Gryllus bimaculatus Degeer were purchased from a local pet shop and raised so that we could use adult females of the second generation as the hosts. Artificial infection, dissection of host crickets, and SEM observation were prepared as described in Ri et al. (2022).
2.2. Molecular phylogeny2.2.1. DNA extractionDNA was extracted from 1-2 wk old mycelia grown on 1/2 ME-YE agar by the Cetyltrimethylammonium bromide (CTAB) method of Ishida et al. (1999) with minor modifications: the CTAB extraction buffer did not include 2% mercaptoethanol; mycelia were suspended in 0.3 mL CTAB extraction buffer and incubated at 65 °C for 30 min; pellets after CTAB precipitation were dissolved in 200 µL of HTE (10 mM Tris-HCl [pH 8.0], 0.1 mM EDTA, 1 M NaCl); and DNA was precipitated by mixing well with 500 µL of 100% ethanol and centrifuging for 10 min. Preparation for DNA extraction was performed with a modified protocol of Gottlieb and Lichtwardt (2001). The mycelia suspended CTAB extraction buffer were frozen and thawed twice by putting them at –80 °C in a deep freezer for at least 1 h and then incubating at 65 °C in an aluminum block bath. After the final thaw, the mycelia were crushed with sterile pestles and treated according to the forementioned CTAB method. The DNA extracts were used in PCR amplifications either directly or after dilution in sterile, de-ionized water (DIW).
2.2.2. PCR and sequencingWe amplified the 18S rDNA, ITS1-5.8S-ITS2 (ITS), and 28S rDNA regions of the isolates. Primer combinations used were SR1R/NS8Z for 18S rDNA and ITS1F/LR5 for ITS and 28S rDNA. The PCR reactions were conducted on a Mastercycler Pro S thermal cycler (Eppendorf, Hamburg, Germany) and consisted of 0.2 µL of KOD FX Neo DNA polymerase (Toyobo, Osaka, Japan); 2 µL of 2 mM dNTP mix; 5 µL of 2× PCR buffer for KOD FX Neo; 0.15 µL of each 20 µM primer; 1 µL of DIW; and 1.5 µL of template DNA. The amplification protocol for SR1R/NS8Z was as follows: initial denaturation at 94 °C for 3 min; 10 cycles of 94 °C for 10 s, 55 °C for 30 s, decreasing 0.5 °C per cycle, and 68 °C for 1.5 min; followed by 25 cycles of 94 °C for 10 s, 50 °C for 30 s, and 68 °C for 1.5 min. For ITS1F/LR5, the same protocol as above was applied except for the extension condition (68 °C for 1 min). PCR products were diluted 10-fold with DIW, and 30 µL of the diluted PCR products were treated with 8 µL of ExoSAP-IT Express (Thermo Fisher Scientific, Waltham, MA, USA), which was diluted 20-fold with DIW, and incubated for 4 min at 37 °C followed by 1 min at 80 °C.
The sequence reaction was performed by the sSTeP method (Platt et al., 2007) on a BigDye Terminator 3.1 cycle sequencing kit (Thermo Fisher Scientific). The primers used were SR1R, NS4, NS6, and NS8Z for 18S rDNA; ITS1F and ITS4 for ITS; and LR0R, LR3, and LR5 for 28S rDNA. Nucleotide sequences were determined by using the Fasmac sequencing service (Kanagawa, Japan). The obtained sequences were deposited in GenBank under accession numbers PQ481183-PQ481186 for 18S rDNA, PQ481187-PQ481190 for 28SrDNA, and PQ481191-PQ481194 for ITS.
2.2.3. Sequence comparison and phylogenetic analysisThe sequences of Unguispora spp. for each region were aligned by MAFFT 7.511 using the default parameters (Katoh et al., 2019). Multiple alignments were edited further with AliView 1.28 (Larsson, 2014), and sequence identity matrices of each region were executed to assess sequence similarity by BioEdit 7.2.5 (Hall, 1999). We obtained the sequences of Kickxellomycotina and Zoopagomycotina from NCBI GenBank (Supplementary Table S1) and created 18S and 28S rDNA datasets for phylogenetic analysis. The ITS region was not used for the analysis because no sites with base substitutions remained when the ambiguous regions were excluded. The sequences for the analysis were automatically aligned with MAFFT version 7.511. Ambiguously aligned regions were excluded using trimAl 1.2 (Capella-Gutiérrez et al., 2009) with a gappyout model. Best-fit substitution models of each gene were selected by ModelFinder (Chernomor et al., 2016; Kalyaanamoorthy et al., 2017) with edge-proportional partition. The maximum likelihood phylogenetic tree was inferred by using IQ-TREE 2.1.3 (Minh et al., 2020), and support values for each node were calculated by 1000 nonparametric bootstraps. The multiple sequence alignments and the tree files are available from Zenodo online repository (DOI: 10.5281/zenodo.13683513).
Unguispora grylli Ri & Degawa, sp. nov. Figs. 1, 2, 3.
MycoBank no.: MB 856220.
Diagnosis: Distinguished from U. rhaphidophoridarum by host range, larger size of sporangiola, the number of lengthwise rows of claws on the sporangiola, presence of morphological change of claws within the host gut, and homothallic zygospore formation.
Type: JAPAN, Nagano Pref., Ueda City, Sanada, Soehi, on excrement of Modicogryllus siamensis Chopard, 26 Jun 2020, T. Ri (holotype, KPM-NC 30743; ex-type cultures, NBRC 116289).
Gene sequences ex-holotype: PQ481185 (18S), PQ481189 (28S), PQ481193 (ITS).
Etymology: Named after its host.
Description: Colonies on 1/2 ME-YE agar pale yellow, reaching 5-6 mm diam at 20 °C for 7 d under aerobic conditions following incubation at 20 °C for 7 d under anaerobic conditions. Vegetative hyphae hyaline, septate, irregularly branched, 2-27 µm wide, sometimes with gangliform swellings near the base of sporangiophores. Sporangiophores yellowish, erect, septate with median pores and plugs, simple or branched, asperulate, 14-20 µm wide, up to 7 mm high, producing one sporocladium at the top or fertile branches sympodially. Sporocladial stipes constricted near the apex, (54-)58.5-120.5 × 9.0-16.5 µm at the base, 4.5-8 µm wide at the most constricted part. Sporocladia tapered, slightly recurved, asperulate, 68-137.5 µm long except for the basal and terminal cells, 16-24 µm wide at the middle, composed of 14-23 cells (excluding the stipes), bearing 6-17 pairs of sterile filiform appendages laterally; basal cells sterile, 22-34.5 × 13.5-21 µm at the terminal, bearing three appendages at the middle; each appendage of the basal cell sterile, composed of 4-8 cells, corniform, elongate upward; terminal cells sterile, tapered, (7-)7.5-15 × 9.5-18.5 µm at the base, bearing one appendage at the distal end; the appendage of the terminal cell sterile, composed of 2-7 cells, corniform, elongate downward; each intercalary cell of sporocladia producing pseudophialides in one or two transverse rows on their lower surface. Pseudophialides lageniform, (11-)12-16.5(-17) × 4-6(-7) µm, bearing single sporangiola terminally. Sporangiola hyaline, cylindrical, (31.5-)32.5-38.5(-39) × 5-6 µm, containing a single sporangiospore at the basal end, bearing ornamentation terminally, immersed in a liquid droplet at maturity; ornamentation with five lengthwise repeated units of six minute elongate denticles in six transverse rows, like claws, ornamentation of lengthwise rows linked together to form a chain. Sporangiospores hyaline, cylindrical, 17-22.5 × 3.5-4.5 µm, producing secondary spores basally in yeast-like form under anaerobic conditions. Secondary spores at maturity hyaline, oblong, up to 32.5-52.5(-54.5) × 11.5-19.5 µm, germinating 1-4 hyphae from both ends under aerobic conditions. Zygospores, formed intercalary in and on the substrate, pale orange-brown by transmitted light, globoid, smooth, (46-)47.5-72.5(-73.5) µm diam, with a brownish wall 7-11.5(-12) µm thick, containing numerous globules, homothallic.
Sporangiola attaching to proventriculus (foregut) lining of Gryllidae with claws, producing secondary spores as well as in culture under anaerobic conditions. Claws morphologically changing with disappearance of the upper half of sporangiolar wall within the host gut: each claw, except the lowest one, shaped like a bottle opener with a hollow center, vertically connected in a chain.
Host: Gryllidae (Modicogryllus siamensis, Loxoblemmus campestris Matsuura, Loxoblemmus sp., and Teleogryllus emma Ohmachi & Matsuura).
Distribution: Japan (Nagano and Niigata prefectures).
Additional specimens/cultures examined: JAPAN, Nagano Pref., Ueda City, Sanada, Ohata, on excrement of Teleogryllus emma, 1 Sep 2019, T. Ri, NBRC 116287; on excrement of Loxoblemmus sp., 8 Sep 2019, T. Ri, NBRC 116288. JAPAN, Niigata Pref., Joetsu City, Kanayasan park, on excrement of Loxoblemmus sp., 15 Sep 2022, T. Ri, NBRC 116290.
Note: This species grew and sporulated well on 1/2 ME-YE agar and 0.3% ShA but produced sporangiophores and vegetative hyphae better on 1/2 ME-YE agar than on 0.3% ShA. The number of collected crickets (Gryllidae and Trigonidiidae) was 119, of which 15 (= 12.6%) were infected by U. grylli. The infection rates for each collection site ranged from 0% to 85.7% and averaged 18.5% (the minimum number of collections for each site was 7).
3.2. Sequence comparison and phylogenetic analysisOn the basis of aligned rDNA data between U. grylli (NBRC 116287-116290) and U. rhaphidophoridarum (NBRC 114905-114907), the length of the 18S rDNA sequences were 1630-1632 bp, the length of the partial 28S rDNA sequences were 855-859 bp, and the length of the ITS sequences were 561-647 bp. The intraspecific similarities of all regions in U. grylli were 100%. In U. rhaphidophoridarum, the intraspecific similarity was 99.8-99.9% for 18S rDNA and 94.2-99.7% for 28S rDNA. The ITS sequences could not be compared in U. rhaphidophoridarum because we could only obtain the ITS sequence from a single strain in the species. In contrast, the interspecific similarity between the two species was 98.3-98.5% for 18S rDNA, 89-89.9% for 28S rDNA, and 59.9% for ITS.
In the phylogenetic analysis, TIM3+F+R4 for 18S and 28S rDNA was selected as the best-fit substitution models according to corrected Akaike information criterion. In the phylogenetic tree of Kickxellomycotina (Fig. 4), U. grylli and U. rhaphidophoridarum clustered together (97% bootstrap), and Unguispora was sister to Linderina (92%). The tree topology of Kickxellales was similar to Ri et al. (2022) except for Dipsacomyces R.K. Benj., Martensiomyces J.A. Mey., Myconymphaea Kurihara, Degawa & Tokum., and Pinnaticoemansia Kurihara & Degawa.
Unguispora is the only genus of Kickxellales whose sporocladia produce two types of sterile appendages: corniform appendages and filiform appendages (Ri et al., 2022). The claws on the upper half of the sporangiola in Unguispora are also unique (Ri et al., 2022). Unguispora grylli possesses these characteristics and forms the monophyletic clade with U. rhaphidophoridarum (Fig. 4), while being distinguished from U. rhaphidophoridarum in terms of asexual and sexual structures and host specificity (Table 1). The colony size of U. grylli is much smaller than that of U. rhaphidophoridarum under the same culture conditions. Fertile branches of U. grylli are sympodial, but those of U. rhaphidophoridarum are monopodial in many cases. Compared to U. rhaphidophoridarum, sporocladia are generally smaller in size and number of cells. However, pseudophialides and sporangiola are larger in size. Unguispora grylli also differs in the shape of the sporangiospores and the number of lengthwise rows of claws. The size of mature secondary spores is smaller, and germ hyphae are also fewer in vitro.
| Species | U. grylli | U. rhaphidophoridarum |
| Host | Gryllidae | Rhaphidophoridae |
| Colony diam. | 5-6 mm | 21-22 mm |
| Sporangiophores diam. | 14-20 µm | 14-25 µm |
| Fertile branch type | sympodial | monopodial or sympodial |
| Stipes | (54-)58.5-120.5 × 9.0-16.5 µm | 22-43(-49.5) × 11-20(-21.5) µm |
| Constriction part of stipes | 4.5-8 µm wide | 7-10(-10.5) µm wide |
| Sporocladia except for basal and terminal cells | 68-137.5 × 16-24 µm. 14-23 cells | 87-155.5(-161) × 17-20 µm. 24-44 cells |
| Number of corniform appendages | 4 | 4 |
| Number of filiform appendages | 14-27 pairs | 14-28 pairs |
| Basal cells | 22-34.5 × 13.5-21 µm | 36.5-53 × (14-)15.5-19 µm |
| Terminal cells | (7-)7.5-15 × 9.5-18.5 µm | (8-)10-18.5(-19) × (8.5-)9-11.5 µm |
| Pseudophialides | (11-)12-16.5(-17) × 4-6(-7) µm | 11-14.5 × 3-4(-4.5) µm |
| Sporangiola | (31.5-)32.5-38.5(-39) × 5-6 µm. Cylindrical | (27.5-)28-32.5 × 3.5-4 µm. Cylindrical |
| Claws | 5 lengthwise rows | 9-10 lengthwise rows |
| Number of denticles | 6 | 6 |
| Sporangiospores | 17-22.5 × 3.5-4.5 µm. Cylindrical | (17-)17.5-21 × 3-3.5 µm. Cylindrical, abruptly tapering toward the acute apex, round and truncate at the base |
| Secondary spores | 32.5-52.5(-54.5) × 11.5-19.5 µm | 43.5-68(-80) × (14-)14.5-27(-31) µm |
| Germ hyphae from secondary spores | 1-4 | 2-8 |
| Zygospores | Present | Unknown |
Surprisingly, the new species changed the morphology of the sporangiola and claws when in the host proventriculus. The upper half of the sporangiolar wall disappeared, and the claws were connected with loops in vertical columns (Fig. 2A-D). The bottle opener-like structures could be observed by using differential interference microscopy before the morphological change (Fig. 1D). The chains of claws got tangled in fine hair on the surface of the host's proventriculus (Fig. 2A, B). There were gaps between the base of the denticles of the claws, where fine hairs became trapped (Fig. 2F-H). This method of attachment for the claws is essentially identical to U. rhaphidophoridarum (Ri et al., 2022). However, the shape of the gaps is different between the two species: U. grylli has circular gaps on the first to forth columns of the claws counting from the top (Fig. 2C, D, F, G), whereas U. rhaphidophoridarum has rectangle gaps (Ri et al., 2022). In both species, the denticles and gaps in the lowest column are smaller and less conspicuous than those in the other columns. The circular gaps were presumed to emerge with the morphological change (Figs. 1H, 2C, D).
Although the trigger and mechanism of the morphological change are unclear, it seems to have been caused by the physical stimuli of peristalsis and/or the chemical stimuli from the intestinal environment. However, it is unlikely that the sporangiolar wall was broken down by the host's digestive enzymes since the lower half of the sporangiolar wall appears to be continuous with the upper half but remained even after the morphological change (Fig. 2C, E). Alternatively, the specific regions of the sporangiolar wall may have disappeared through autolysis. The morphological change did not occur under anaerobic conditions where the species shows yeast-like growth. Thus, it does not necessarily correlate with the germination and formation of secondary spores. As a similar case, sporangiospores in Zancudomyces culisetae Yan Wang, Tretter, Lichtw. & M.M. White and some species of Smittium R.A. Poiss. (Harpellales) change their morphology in the gut of hosts (mosquito larvae) (Williams & Lichtwardt, 1972). Their sporangiospores: 1) extrude from the sporangiola (trichospores); 2) rapidly elongate to three times its length; 3) secrete a holdfast substance; 4) shrink to the initial length; and 5) attach to the lining of the hindgut (Horn, 1989a, 1989b; Sato, 1993; Takeshita & Sato, 2023; Williams, 1983). All these steps are completed within 90 s, allowing sufficient adhesion even during short passage times through digestive tracts (Takeshita & Sato, 2023). In these species, the reaction can be induced in vitro by potassium with a pH shift that mimicks passage through the host gut (Horn, 1989a). Due to the difference in adhesion sites, the stimuli required for morphological changes most likely differ between these species and U. grylli. Further study is needed to identify the stimulus for the morphological change in U. grylli.
Zygospore formation in the genus Unguispora was observed for the first time in this study. Unguispora grylli consistently formed zygospores across all isolates and on any media. More zygospores were produced in the mycelium differentiated from sporangiophores when transferred by sporangiophores rather than when transferred by vegetative mycelia. The species was determined to be homothallic, as zygospores formed in the culture derived from a single secondary spore. Among Kickxellales, zygospores have been reported in three genera (Coemansia, Kickxella Coem., and Spirodactylon R.K. Benj.) and are usually smooth, globoid, and thick-walled (Benjamin, 1958, 1959; Chien, 1971; Kwaśna et al., 1999; Linder, 1943). While U. grylli shares these characteristics, the zygospores vary in color: those of U. grylli are pale orange-brown, whereas those of the other genera except C. braziliensis Thaxt. ex Linder are nearly hyaline (Fig. 3T, U). Zygospores of Spiromyces R.K. Benj. and Mycoëmilia Kurihara, Degawa & Tokum., which were transferred from Kickxellales to Spiromycetales, are verrucose and distinguished from those of Kickxellales including U. grylli (Benjamin, 1963, 1979; Kurihara et al., 2004). However, the shape of zygospores in Kickxellales is simple like to that in Asellariales despite phylogenetic distance (Valle & Cafaro, 2008). Therefore, the morphology of zygospores in Kickxellomycotina appears to reflect not only their phylogenetic relationship but also their habitats. Furthermore, there are two methods of zygospore formation in Kickxellales. Zygospores arise from a cell formed at the junction of two differentiated hyphae, as in C. aciculifera Linder and Kickxella alabastrina Coem., or from a cell delimited by an undifferentiated hypha near its point of anastomosis with another hypha, as in C. mojavensis R.K. Benj. and C. braziliensis (Benjamin, 1958). Unguispora grylli produces zygospores in the latter way, and incipient zygotes enlarge as more or less globose thin-walled swelllings (Fig. 3L, M). Subsequently, granular cytoplasm fuses together, resulting in larger and fewer granules, while zygote walls become thicker and develop pigmentation (Fig. 3N-S). Even when the development ceases, there is considerable variation in the size and number of the granules.
In this study, U. grylli was found in several genera of Gryllidae but not in any genera of Trigonidiidae, which co-occurred with Gryllidae. The host range of this species remains unclear due to the limited number of collected crickets. Unguispora grylli and U. rhaphidophoridarum have thus far been discovered exclusively in Gryllidae and Rhaphidophoridae, respectively. Although they share the common characteristics of having host Orthoptera and attaching to the proventriculus, the internal surface structure of the proventriculus differs between Gryllidae and Rhaphidophoridae (Judd, 1948). Thus, it seems that the Unguispora species have morphologically distinct attachment structures that confer their host specificity. Moreover, the infection rate of U. rhaphidophoridarum was obviously higher than that of U. grylli (Ri et al., 2022). We believe that this could be attributed to the behavior of the host. For example, camel crickets are primarily active for foraging in surface habitats at night and then return to roost in caves or houses, digest food, and defecate (Hill, 2003; Peck, 1976; Poulson et al., 1995). Furthermore, they cluster in the roost to enhance water conservation (Yoder et al., 2002, 2010). The high relative humidity and stability of environments such as caves and indoors are advantageous for extremely dry-sensitive mycelia. In addition, there is a consistently high frequency of contact among individuals when aggregated. This may contribute to the spread of infection if, when grooming, individuals take up wet sporangiola adhered to the surface of their bodies. The infection rate of U. grylli is highly variable and generally low because the host crickets lack the aforementioned behaviors. Additionally, zygospore formation of U. grylli seems to reflect the life history of the host. Most crickets in temperate Honshu are univoltine and overwinter as eggs (Machida, 2016). Unguispora grylli may utilize the zygospore as a dormant structure where they can endure periods without nymphs or adults. On the contrary, camel crickets in Japan are semivoltine or univoltine, and either nymphs or adults are present throughout the year (Machida, 2016; Sugimoto & Ichikawa, 2003). Hence U. rhaphidophoridarum appears not to require dormant structures, and indeed, zygospores of this species have not been found.
The current diversity of Kickxellomycotina is evidently underestimated, as suggested by Reynolds et al. (2023), and identifying the undiscovered taxa is an urgent task in taxonomy. The data presented herein has expanded our knowledge of the potential hosts of the amphibious fungi and given us a glimpse of the morphological diversity of the host gut attachment structure. To develop a better understanding of the phylogeny of Kickxellomycotina and basal lineages of the fungi, the search for undescribed taxa should be conducted in a wider range of host taxa and regions.
The authors declare no conflicts of interest. All the experiments undertaken in this study comply with the current laws of the country where they were performed.
This study was supported, in part, by a Grant-in-Aid for Scientific Research (no. 24KJ0485) from the Japan Society for the Promotion of Sciences. We thank Morikazu Yanagisawa, a naturalist of Sugadaira, for cooperation in collecting crickets.
