Mycelial compatibility, anastomosis, and nucleus numbers of eight Mexican Hirsutella citriformis strains isolated from Diaphorina citri

Background Among entomopathogenic fungi, H. citriformis has been recognized as potential biocontrol agent against the Asian citrus psyllid Diaphorina citri (Hemiptera: Liviidae). Nevertheless, this fungus is poorly characterized. Previous molecular studies have shown high sequence similarities among strains, but significant differences in Diaphorina citri virulence. Objective The aim of the present study was to determine mycelial compatibility and anastomosis, and nucleus numbers in mycelium and conidia of eight H. citriformis strains isolated from mycosed D. citri adults collected from several Mexican states. Methods Mycelial compatibility and anastomosis evaluation was performed after pairing strains, leading to 36 confrontations, and cultured in chlorate minimum medium to obtain mutants for vegetative compatibility group. Results Hypha or conidia nuclei were visualized with safranin-O and 3% KOH, and 0.05% trypan blue–lactophenol solution. H. citriformis strains showed compatibly and anastomosis events after confrontation. In addition, they showed one nucleus per conidium and mycelium section. It was not possible to obtain H. citriformis nit mutants from the chlorate concentrations tested. Conclusions To date, this is the first report demonstrating mycelial compatibility, anastomosis occurrence, and hyphae and conidia nuclei number among H. citriformis strains.

H. citriformis, as well as other entomopathogenic fungi, does not have a known sexual state. This group represents the imperfect state of Cordyceps, Ophiocordyceps or Torrubiella (Tanada & Kaya, 1993). Genetic recombination in asexual fungi (parasexual cycle) is a potential source for genotypic diversity among fungi populations, including H. citriformis, which occurs after compatible vegetative hyphae are fused (Paccola-Meirelles & Azevedo, 1991;Couteaudier & Viaud, 1997;Bello & Paccola-Meirelles, 1998;Dalzoto et al., 2003). No changes were observed when fungus mycelium is crosslinked during growth, thus vegetative compatibility was not tested since all strains showed similar growth in different nitrogen sources. Crosslinking during growth is a random event that might result in virulence alteration against common pathogens or change the common variety of the host strain (Burdon & Silk, 1997).
The aim of the present study was to determine mycelial compatibility and anastomosis, and nucleus numbers in mycelium and conidia of eight H. citriformis strains isolated from mycosed D. citri adults, collected from several Mexican states.

Mycelial compatibility
Strains cultured on PDA, were confronted among them, evaluating all possible combinations (36 confrontations) in order to determine strains mycelial compatibility. Two mycelial discs collected from 14-day-old colonies, were placed in 100× 15 mm Petri dishes with 25 mm to 35 mm separation distance between them. Plated were then incubated at 25 • C ± 2 • C for 3 wk to verify mycelium fusion. Confronted strains with potential to fuse their mycelia were considered compatible, whereas formation of a barrier between them was considered as incompatible strains. Experiments were performed three times with three replicate determinations per treatment.

Anastomosis
For anastomosis experiments, strains were combined and placed on 76× 26 mm sterile slides (Chance Propper Ltd., Smethwick, Warely, England) coated with a thin layer of WA, supplemented with 2% dextrose in a Petri dish at a distance of 2 ± 0.5 cm from each other. After 21-28 d incubation at 25 ± 1 • C, slide was taken from the plate and the original mycelium fragments were carefully removed, leaving only newly grown hyphae, which were stained with trypan blue-lactophenol [0.05 g of trypan blue solution in 100 mL of lactophenol (20 mL of 85% lactic acid and 20 g of phenol, 40 mL of glycerol, and 20 mL of distilled water)] (Burpee et al., 1978). A cover slip was placed on top of the developed hyphae and examined by light microscopy (Olympus CX41, Olympus, Mexico City). Anastomosis between strains was confirmed by visually tracing the anastomosing cells origin. Experiments were performed three times with three replicate determinations per treatment.

Vegetative compatibility
Vegetative compatibility between strains was determined by testing non-utilizing nitrate mutants (Brooker, Leslie & Dickman, 1991). To select mutants, strain mycelial discs (0.5 mm diameter) of the monoconidial culture were used to inoculate on chlorate plates. Chlorate medium was prepared by adding 1.5%, 2.0%, 2.5%, and 3% potassium chlorate to the MM agar (MM-C). After 3 wk to 4 wk, thin fast-growing sectors were observed.
To phenotypically characterize nit mutants, MM was prepared using the following nitrogen sources: MM supplemented with 1.6 g/L ammonium tartrate (MM-TA), 2 g/L of nitrate (MM-N), 0.5 g/L nitrite (MM-NI), 0.2 g/L hypoxanthine (MM-H), or 0.2 g uric acid (MM-UA) (Correll, Klittich & Leslie, 1987). Colonies growth in chlorate plates were transferred to 60 mm ×15 mm Petri dishes containing MM supplemented with each of the nitrogen sources to detect nit mutant type, and incubated at 25 • C for 2 wk to 3 wk.

Conidia and mycelium nuclear characterization
Hypha nuclei were stained with 3% KOH-safranin O (Bandoni, 1979) and 0.05% trypan blue-lactophenol solution. The culture technique was similar to that used by the anastomosis test. Conidia nucleus preparations were visualized with the same stain used for mycelium nuclei. Slides were then examined by light microscopy (Olympus, Mexico City).

Mycelial compatibility
Thirty-six pairings were performed with the eight evaluated stains. All combinations resulted in a compatible reaction, where strains mycelia intermingled at the interaction zone. Both the compatibility between the same strain and the compatibility among strains, showed what appears to be a separation line in the colony border among them; nevertheless, this was only observed by the aerial mycelia since in the agar internal growth, mycelia combination among strains was observed. Cell death related to incompatibility or growth inhibitions due to antagonism were not detected among strains (Fig. 1).

Anastomosis
Anastomosis experiments revealed all strain combinations performed anastomosis (Fig. 2). Anastomosis events occurred between hyphal tip with other hyphae lateral portion (first type) or by parallel bridges between hyphae cells (second type). The second type was the most frequent anastomosis event observed.

Vegetative compatibility
Vegetative compatibility groups were observed, but all the colonies showing the strain typical growth of nit mutant were reported. All strains grew and produced non-dense mycelium in MM. In contrast, colonies developing on chlorate plates showed dense fluffy grown mycelium (Fig. 3). Seventeen colonies grew on 2.5% chlorate MM-C plates, whereas only one colony (INIFAP-Hir2 strain) grew on 3% chlorate MM-C. INIFAP-Hir-1, INIFAP-Hir-5, INIFAP-Hir-6, and IB-Hir-1 strains did not grow at these chlorate concentrations. After mycelium samples of these strains were transferred from the MM-C to MM, all developed non-dense mycelia.
H. citriformis strain developed colonies were characterized upon different nitrogen sources utilization (Fig. 4). All strains produced dense mycelium on MM-TA, MM-H, and MM-UA, whereas non-dense mycelium was developed on MM-N and MM-NI. These wild-type strains produced similar macroscopic colonies than those isolated from MM-C when testing different nitrogen sources. It was then not possible to obtain nit mutants since the isolated colonies similarly grew as the wild strains on the different nitrogen sources media.

Conidia and mycelium nuclear characterization
Nuclei from 10 conidia and 10 cells of randomly selected H. citriformis strains were counted, from a location in the stain gradient, where nuclei and septa are distinctly observed. All strains showed a single nucleus in both conidia and mycelium (Fig. 5).
Colonies developed from H. citriformis strains on MM-C transferred to MM with nitrate as the sole source of nitrogen, showed thin, expansive, and appressed growth. This type of growth is similar than that reported by nit mutants. Nevertheless, wild strains showed similar growth on MM with nitrate. Furthermore, mycelium developed by the strains cultured on MM-C (2.5% and 3% chlorate) showed that Hirsutella strains were stimulated to develop dense mycelium colonies. All colonies from MM-C as well as the wild strains, presented similar growth on different nitrogen sources (dense mycelium on organic nitrogen sources and nondense mycelia on inorganic nitrogen sources). Results provided evidence of the expected compatibility among tested isolates, thus indicating that selected strains are closely related.
Enzymes production studies developed on this fungus did not show significant correlation with its pathogenicity (Cortez-Madrigal et al., 2014;Quesada-Sojo & Rivera-Méndez, 2016;Pérez-González & Valdes-Gonzalez, 2017). However, it has not been determined whether the pathogenic metabolism relies on transposon or mycovirus elements that may play a role as observed by B. bassiana (Kotta-Loizou & Coutts, 2017;Pitaluga et al., 2020). Results from our study may help to better understand this fungus metabolism. In this regard, compatibility results might be a factor to determine whether each strain characteristics are related to growth and development on culture media, as well as pathogenicity in the laboratory and in the field.
Future research on H. citriformis is underway to better understand if nuclei exchange between strains or extrachromosomal material exchange affects its pathogenicity. In addition, dissemination, population richness, and infection mechanism field studies on this fungus are required, since limited research related to the presence of the fungus parasitizing the insect has been performed.

CONCLUSIONS
H. citriformis strains tested in this study presented compatible mycelium and anastomosis among them. Evaluated strains showed one nucleus per conidium and mycelium section. All tested strains developed dense mycelium in inorganic nitrogen sources. Results also indicated that genetic information may be exchanged by asexual reproduction (hyphal anastomosis) between different H. citriformis strains isolated from D. citri. It was not possible to obtain H. citriformis nit mutants at the chlorate concentrations tested. To date, this is the first report describing the mycelial compatibility, anastomosis occurrence, and number of nuclei present in hyphae and conidia among H. citriformis strains.

ADDITIONAL INFORMATION AND DECLARATIONS Funding
This work was supported by CONACYT-MEXICO (No. A1-S-33091) to OPG. Authors were supported by Sistema Nacional de Investigadores (SNI-Conacyt, Mexico) SNI-73211 to Orquídea Pérez-González, SNI-9942 to Ricardo Gomez-Flores, and SNI-16614 to Patricia Tamez-Guerra. The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript.