The rumen digesta samples were collected from goats (n=3) slaughtered at Kondhwa slaughterhouse, Pune (India). The samples were immediately brought to the laboratory (within 1 h), pooled and homogenised for 10 minutes under the gas phase of CO₂ before making dilutions (up to 10^-4) in the anaerobic diluent (McSweeney et al. 2005). The serum roll bottle method (Miller and Wolin 1974) was used to isolate pure cultures of anaerobic fungi from different dilutions. Briefly, the inoculum (0.5 ml) was added to 125 ml glass serum bottles containing 10 ml fungal culture medium (pH 6.8 ± 0.1), which comprised (per litre) of 3 g yeast extract, 5 g tryptone, 150 ml each of solution 1 (0.3% K₂HPO₄), solution 2 [0.3% KH₂PO₄, 0.6% (NH₄)₂SO₄, 0.6% NaCl, 0.06% MgSO₄.7H₂O and 0.06% CaCl₂.2H₂O] and clarified rumen fluid, 1 ml resazurin (0.1%), 1 ml hemin (0.05%), 6 g NaHCO₃, 1 g L-cysteine-HCl, 5 g cellobiose and 20 g agar. The antibiotics, benzylpenicillin and streptomycin sulphate (final concentration 200 µg/ml), were also added to inhibit bacterial growth.

All the roll bottles were incubated at 39±1 °C for 3 weeks and inspected regularly for the development of fungal colonies. The morphologically distinct colonies were picked under anaerobic conditions and inoculated into the fresh liquid culture medium. The serum roll bottle method was repeated two more times to get the axenic fungal cultures. All the cultures were cryopreserved at -80 °C and -196 °C for short-term and long-term storage, respectively using ethylene glycol (final concentration 0.64 M) as the cryoprotectant (Callaghan et al. 2015).

For molecular characterisation, the genomic DNA was extracted using the CTAB DNA extraction protocol (Sirohi et al. 2013). The complete internal transcribed spacer (ITS1-5.8S-ITS2; ITS) and D1/D2 domain of large-subunit (LSU) ribosomal DNA were amplified using ITS1/ITS4 and NL1/NL4 primer pairs (Edwards et al. 2017). The sequencing was outsourced to 1^st BASE (Singapore) and the obtained sequences were compiled and edited manually using BioEdit software (Hall 1999). Since most of the culture-independent studies have generated only ITS1 sequences, the ITS sequence was trimmed to obtain only the partial ITS1 region. Sequence similarity searches were performed using GenBank BLASTn. For phylogenetic analyses, the ITS1 and LSU sequences representing different anaerobic fungal genera and uncultured representatives, were downloaded from the NCBI GenBank database. All the sequences were aligned using the ClustalW programme (Thompson et al. 1994) with default settings in MEGA7 (Kumar et al. 2016). The aligned sequences were used to construct a phylogenetic tree in MEGA7 using the maximum-likelihood method based on the Tamura-Nei model (Tamura and Nei 1993) and tested by 500 bootstrap replications. The genus Gromochytrium mamkaevae was used as the outgroup for both ITS1 (accession number: KF586842) as well as LSU (accession number: NR_132054) based trees. The ITS1 and LSU alignments have been submitted to TreeBASE under submission ID 22988.

The obtained strains were grown in a fungal culture medium without yeast extract or tryptone and cellobiose was replaced by different substrates (Table 1) as a carbon source (Hanafy et al. 2017; Hanafy et al. 2018). Following initial growth for 2 d on monosaccharides and disaccharides and 5 d on polysaccharides, the cultures were subcultured three times at 10% inoculum size to evaluate their substrate utilisation abilities. The growth was measured in terms of visible biomass accumulation and total gas production. The cultures were scored based on their ability to grow luxuriantly (++) or slowly (+) within the stipulated incubation periods or following extended incubation up to 10 days. The avicelase, CMCase, xylanase and β-glucosidase activities of type strain were also determined following its growth on cellulose, xylan, wheat straw and rice straw for 5 d as per the method described previously (Dagar et al. 2018). The fermentation gases (H₂, CO₂), volatile fatty acids (VFAs; acetate, propionate, butyrate etc.) and alcohols (ethanol and butanol) of growth positive substrates were determined using gas chromatographs equipped with flame ionisation or thermal conductivity detectors, similar to previous reports (Dighe et al. 1998; Kamalaskar et al. 2010; Singh et al. 2016). The organic acids (formate, lactate, succinate, malate etc.) were analysed using HPLC LC20A (Shimadzu, Japan) equipped with a refractive index detector (Thakker et al. 2006). An ordination of the obtained fermentation products was also generated to see if there is any pattern of substrate utilisation using Non-metric Multidimensional Scaling (NMDS) analysis based on the distance matrix in PAST 3.20 software (Hammer et al. 2001).

For the routine anaerobic fungal isolation, the inoculated roll tubes are usually incubated for 2–4 d after which the developed colonies are picked anaerobically and pure cultures are obtained. However, we chose to incubate the roll tubes for an extended period of time i.e. up to 3 weeks. The decision for extended incubation was taken in the wake of the fact that several genera of anaerobic fungi remain to be uncultured and incubation time may be one of the limiting factors. As shorter incubation times favours fast-growing cultures, some slow growing cultures might be omitted and which may grow after prolonged incubation. The prolonged incubation might also help some cultures to cope better in stressed conditions of different growth environment, the presence of antibiotics or some oxygen exposure during sample collection or processing. In a previous study, prolonged incubation was recommended for the isolation of anaerobic bacteria from clinical specimens and correct bacteriological diagnosis (Wren 1980). Janssen et al. (2002) also speculated that the extended incubation period may help isolation of previously uncultured groups of soil bacteria. In this case, the extended incubation of 15 d resulted in the development of two fungal colonies, but no further growth even after 21 d of incubation. Following isolation and establishment of pure cultures, the morphological, molecular and substrate utilisation characteristics of both strains, namely G1SC and G6SC were found to be identical. Therefore, one of them, i.e. strain G1SC, was denoted as the type strain and used for detailed characterisation. The type strain has been deposited to the MACS collection of microorganisms (MCM), Agharkar Research Institute, Pune, India under the accession number MCMB-1469.

The colony morphology of strain G1SC is shown in Fig. 1A–C, showing 1–2 mm sized colony, attached to rice straw (Fig. 1A) and at different stages of growth. A large number of newly formed sporangia at the periphery of the colony were observed, as shown by numerous dot-like structures of varying sizes (Fig. 1B–C). A similar observation was also made in broth culture, where numerous fungal thalli, attached to the bottom of glass bottles, were seen on the initial period of growth (Suppl. material 1: Figure S1), which later developed into a thin mat or biofilm-like structure (Fig. 1D) similar to Pecoramyces (Hanafy et al. 2017). Zoospores were abundantly produced, mostly uniflagellate (Fig. 1E–F) and rarely biflagellate (Fig. 1G), spherical to ovoid in size (5–6 µm in diameter) and the flagellum 15–20 µm in length. The zoospores were found to germinate either endogenously or exogenously (Fig. 1H–I) into different shapes of sporangia (Fig. 1J) like globose or ellipsoidal at the very early stages of development. Different forms of rhizoidal development, like long single rhizoid (Fig. 1K), two rhizoids (Fig. 1L) and even multiple rhizoids originating from sporangia (Fig. 1M), were also seen.

Figure 1. 

Macroscopic and microscopic features of Liebetanzomyces polymorphus. Colony morphology on agar roll tubes (A–C), showing the development of a colony attached (A) to a straw particle (arrowed), dense growth in the centre surrounded by numerous sporangia and zoospores (B–C) causing expansion of colony size. Growth in liquid medium showing a biofilm-like growth (D). Zoospores are spherical and Uniflagellate (E–F) or biflagellate (G). Germinating zoospore (H) showing a zoospore cyst (arrowed), presence and absence of sporangiophore indicating the endogenous and exogenous type of sporangial development (I) and different shapes of sporangia (I, J). Early stages of thallus development showing a single (K), bifurcated (L) and multifurcated (M) rhizoidal system. Scale bar: 1 mm (A–C); 10 µM (E–M).

We also noticed pleomorphism in sporangial and rhizoidal structures of strain G1SC on different substrates like rice straw (Fig. 2A), wheat straw (Fig. 2B), cellulose (Fig. 2C), xylan (Fig. 2D), starch (Fig. 2E), cellobiose (Fig. 2F), lactose (Fig. 2G), maltose (Fig. 2H), sucrose (Fig. 2I), glucose (Fig. 2J), xylose (Fig. 2K) and fructose (Fig. 2L). In the case of complex substrates like rice straw, wheat straw, cellulose and xylan, the rhizoidal growth was observed to be more extensively branched. Conversely, thicker and less branched rhizoidal growth was noticed on dimeric and monomeric substrates, probably due to the ready availability of fermentable sugars. It was also observed that, on rice and wheat straw, an exogenous type of sporangia development with short or long sporangiophores was more prominent in comparison to endogenous sporangia on all other substrates. The presence of sporangiophores on complex straw particles might be helpful for sporangia to come out of straw and release zoospores to further areas for faster colonisation. The pleomorphism, in morphological features with a change in culture conditions, particularly with carbon source, is also well described earlier (Brookman et al. 2000; Gruninger et al. 2014). However, the sporangial shapes varied, not only on different carbon sources, but also on the same carbon source (Fig. 3). Different sporangial shapes like globose, ellipsoid, clavate, ovoid, with or without a papilla were noticed on the same substrate i.e. rice straw as shown in Fig. 3A–F. Similar polymorphism was also observed on several other substrates, e.g. xylose (Fig. 3G–I) and cellobiose (Fig. 3J–L), where conspicuous irregular sporangial structures were seen. These findings, therefore, highlight the pleomorphic nature of strain G1SC, irrespective of the carbon sources used in the growth medium.

Figure 2. 

Microscopic features of Liebetanzomyces polymorphus showing pleomorphism in sporangial and rhizoidal structures, when grown on different carbon sources like rice straw (A), wheat straw (B), cellulose (C), xylan (D), starch (E), cellobiose (F), lactose (G), maltose (H), sucrose (I), glucose (J), xylose (K) and fructose (L). The exogenous sporangia appearing on a sporangiophore (A–B) and endogenous sporangia of different shapes (C–L) are shown. Scale bar: 20 µM.

Figure 3. 

Microscopic images of Liebetanzomyces polymorphus. On rice straw (A–F) sporangia of varying sizes and shapes, like elongate (A), ellipsoid (B), ovoid (C), clavate (E) and globose (F). The zoospore cyst (A–C arrowed) is also visible, highlighting bipolar germination. Sporangiophore of varying length and shape, from short to long (A), sometimes eggcup shaped (C) and constricted (E arrowed) is shown. The sporangia with a papilla (D arrowed) and septum (B, C, D, E starred) are also indicated. Sporangia and thalli of irregular morphology on xylose (G–I) and cellobiose (J–L), including some pseudo-intercalary sporangia (J–L) can be observed. Scale bar: 20 µM.

The sporangial size also varied in diameter (10–90 µm wide, 10–75 µm long), always borne at the terminal end of a variable length sporangiophore (15–80 µm), in the case of exogenous development. Several cyst-like structures were also visible on the sporangiophore (Fig. 3A–C) highlighting bipolar germination of zoospores. In some cases, an eggcup shaped sporangiophore at the bottom of sporangium (Fig. 3C) and a constricted sporangiophore were also noted (Fig. 3E). In the case of soluble substrates, few pseudo-intercalary sporangia, similar to Oontomyces and Feramyces (Dagar et al. 2015; Hanafy et al. 2018), were also noted (Fig. 3J–L). The thallus development was clearly monocentric as evident by the mature sporangia full of zoospores (Fig. 4A–D, I), anucleated rhizoids (Fig. 4E–I) and a single thallus having a single sporangium (Fig. 4J–K). All these morphological features were similar to various Piromyces spp., Buwchfawromyces, Oontomyces and Pecoramyces, which also have the monocentric thallus and extensive variations in sporangial shape and size and produce uniflagellate zoospores (Ho and Barr 1995). Similar to P. rhizinflatus, the sporangia of strain G1SC showed a constricted, isthmus-like neck (Fig. 2C, J), sporangiophore appearing like an eggcup (Fig. 3C) like in P. communis and sporangium with a papilla (Fig. 3D) similar to P. mae (Ho and Barr 1995; Li et al. 1990).

Figure 4. 

Phase contrast (A, C, E, G), fluorescence (B, D, F, H) and confocal (I) microscopic images of Liebetanzomyces polymorphus showing elongate (A–B) and triangular (C–D) sporangium filled with zoospores. Nuclei were seen in sporangium but not in sporangiophore (B) or rhizoidal system (F, H, I). No constrictions were observed in rhizoids. Scanning electron microscopy images of elongate (J) and globose (K) sporangium showing monocentric thallus. Scale bar: 20 µM (A–I); 10 µM (J–K).

The similarities in morphological features of monocentric and uniflagellate genera make it very difficult to identify and differentiate this group of anaerobic fungi. It is also interesting to note that, so far, the maximum number of species has been described in the Piromyces genus and most of the newly described genera of anaerobic fungi share morphological similarities with these different species, including strain G1SC. These observations point towards the possibility that some of these newly described genera might have been isolated previously as well but were identified as different species of Piromyces. It is also worth noting that all newly described genera have primarily been described using molecular tools, thus emphasising the need to use these modern tools.

The successful amplification of ITS and LSU regions yielded product sizes of ca. 700 bp and 750 bp, respectively. The obtained sequences were submitted to NCBI GenBank to obtain the accession numbers for strain G1SC (MH468765 and MH468763; ITS and LSU) and G6SC (MH468766 and MH468764; ITS and LSU). The ITS region based sequence similarity search results showed that strain G1SC was 88.14% and 87.03% similar to A. robustus (accession number: NR 148182) and A. contortus (accession number: MG605706), respectively. Likewise, the LSU region of strain G1SC was most closely similar (97.71%) to A. contortus (accession number: MG605690). Although the sequence similarity of ca. 97% usually relates to a novel culture at the species level with its nearest match, the stark morphological dissimilarities of strain G1SC with A. contortus led to its classification as a novel genus.

To better understand the ecological distribution of strain G1SC, only ITS1 based searches were performed, which indicated <96% sequence similarities with several uncultured representatives of phylum Neocallimastigomycota (Suppl. material 2: Figure S2) reported from the USA (Liggenstoffer et al. 2010). The nearest matches were with different clones obtained from cattle (93–96%) and other clones of cattle, sheep and llama (88%-92%), indicating the presence of members of the G1SC clade in cattle and close relatives in sheep and llama. The nearest matches of G1SC clade were also found in the SP4 clade designated by Paul et al. (2018), represented by accession number GQ767184, which consisted of 120 sequences of uncultured fungi, all from cattle faeces. The results, hence, indicate the presence of related fungi in both large as well as small ruminants. Surprisingly, no similar sequences were obtained from goat, which might be due to a lesser number of culture-independent studies conducted on these animals. The phylogenetic analysis of LSU and ITS1 regions showed distinct lineage for the G1SC clade, supported by high bootstrap values (Fig. 5 and Fig. 6), nearest to the Anaeromyces cluster. The results, therefore, underscore the importance of morphology-based identifications along with molecular tools, as none on its own can be relied on for accurate identification of anaerobic fungi.

Figure 5. 

The LSU based maximum-likelihood tree is showing the phylogenetic position of Liebetanzomyces polymorphus with other members of phylum Neocallimastigomycota. Bootstrap values (>50%) based on 500 replicates are indicated at branching points. The GenBank accession number of each strain is listed in parentheses. Scale bar: 0.1 substitutions per site.

Figure 6. 

The ITS1 based maximum-likelihood tree is showing the phylogenetic position of Liebetanzomyces polymorphus with nearest uncultured clones and other members of phylum Neocallimastigomycota. Bootstrap values (>50%) based on 500 replicates are indicated at branching points. The GenBank accession number of each strain is listed in parentheses. Scale bar: 0.5 substitutions per site.

The substrate utilisation profiles of strains G1SC and G6SC exhibited exactly identical patterns (Table 1). Both these strains could utilise various polymeric and monomeric substrates. Our results were also compared to substrate utilisation profiles of other genera of monocentric and uniflagellate filamentous anaerobic fungi. All these genera were found to utilise cellulose, xylan, starch, cellobiose, sucrose, maltose, glucose and fructose, while none utilised chitin, ribose, peptone and tryptone as the sole carbon and energy source. For inulin, raffinose, alginate, pectin, trehalose, mannose, arabinose, glucuronic acid and galactose, varying results were obtained. Interestingly, only strains G1SC and G6SC were found positive for pectin utilisation amongst compared genera.

The enzyme activities (µmol/ml/h) of type strain G1SC showed maximum enzyme activities on crude substrates like rice straw and wheat straw, instead of pure cellulose and xylan (Table 2). The highest avicelase (1.05 ± 0.19) and xylanase (88.33 ± 1.00) activities were obtained on rice straw, while highest CMCase (7.73 ± 0.31) and β-glucosidase (1.64 ± 0.08) activities were measured in wheat straw. When compared to the enzyme activities of Neocallimastix, Piromyces, Orpinomyces and Anaeromyces, grown on wheat straw (Dagar et al. 2018), the strain G1SC displayed poor β-glucosidase activities. The avicelase activities of strain G1SC were comparable to Orpinomyces and Anaeromyces, but lower than most strains of Neocallimastix and Piromyces. The CMCase and xylanase activities of strain G1SC were comparable to Orpinomyces, Neocallimastix and most strains of Piromyces, but higher than Anaeromyces. These results suggest that strain G1SC is a moderate lignocellulose degrader and can be of use in applications involving lignocellulose degradation.

The substrate utilisation results were also confirmed by measuring the fermentation products on each substrate (Table 3). The strain G1SC produced hydrogen, carbon dioxide, formate, acetate, lactate, succinate and ethanol on all substrates, but did not produce propionate, butyrate, butanol or malate which is in line with the previous reports (Trinci et al. 1994; Edwards et al. 2017). The absence of propionate and butyrate also implies that there was no bacterial contamination and all products are solely of fungal origin. The fermentation product quantities also indicated the efficiency of substrate utilisation. The substrates like sucrose and lactose, which recorded poor growth after 2 d of incubation, also showed lesser production of formate, acetate and lactate and the absence of succinate. Likewise, the polymeric substrate like rice straw, wheat straw, cellulose and xylan showed lesser production of hydrogen, formate, acetate, lactate, succinate and ethanol at 2 d of incubation (Table 3), which increased substantially after 5 d (Suppl. material 4: Table S1). The NMDS plot showed clear clustering of various carbon sources based on fermentation products, an observation duly supported by a low stress value (Suppl. material 3: Figure S3). These results, therefore, signify the usefulness of metabolite profiling for measuring the substrate utilisation abilities of anaerobic fungi.