Glutamate dehydrogenase 2 is required for virulence by facilitating fungal growth in the host hemocoel Authors: Yuzhen Lu, Denghui Wu, Jiawei Hu, Guojun Peng, Jielai Zhong, Yan Liu, Jing Li, Qiang Gao, Xiao-Qiang Yu Published: 2025 ABSTRACT ABSTRACTMetarhizium robertsii, a fungal pathogen employed in pest biocontrol, can alkalize the surrounding environment, although the biological implications remain unclear. Here, we found that M. robertsii glutamate dehydrogenase 2 (MrGDH2) was responsible for alkalization during fungal growth on insect wings or media containing cuticle powder. Loss of MrGDH2 function resulted in significantly reduced virulence during both cuticle-passing and cuticle-bypassing infections but did not affect appressorium formation or cuticle penetration. Deletion of MrGDH2 failed to alkalize amino acid-containing media under carbon deprivation, leading to impaired mycelia growth and conidiation. Hemolymph carbohydrates were decreased during M. robertsii infection, and the ΔMrgdh2 mutant exhibited delayed fungal growth and impaired alkalization in hemolymph cultures. Notably, expression of PacC, a pH-responsive transcription factor critical for virulence, was downregulated in hyphal bodies of the ΔMrgdh2 mutant. Contrary to the established models in plant and human fungal pathogens, we demonstrate that Gdh2 activity is dispensable for appressorium formation but essential for fungal colonization in insect hemocoel during M. robertsii infection. Introduction Environmental pH, one of the most crucial physiochemical variables, influences not only cell growth and development but also the pathogen’s ability to colonize, penetrate, and eventually kill the target host [1]. Phytopathogenic fungi are able to modify environmental pH, and the accumulation of ammonium has been linked to the pathogenicity in Colletotrichum species, such as C. coccodes and C. gloeosporioides [2]. This ambient alkalization mechanism is accomplished by the active release of ammonia or small regulatory peptides from fungi that mimic plant rapid alkalinizing factors (RALFs) [3,4]. The primary source of free ammonia in Saccharomyces cerevisiae is generated by the mitochondron-localized NAD+-dependent glutamate dehydrogenase (Gdh2), which catalyzes the deamination of glutamate to α-ketoglutarate, resulting in the production of NADH and NH3 [5]. Via α-ketoglutarate, an important intermediate for many biological processes, Gdh2 connects amino acid metabolism to the tricarboxylic acid cycle (TCA cycle) [6]. Deletion of GHD2 resulted in reduced generation of ammonia in C. gloeosporioide and also 85% suppression of appressorium formation, as well as a decrease in pathogenicity [7]. Exogenous addition of ammonia could fix the deficiency in appressoria formation [7]. Mgd1 (GDH2 homolog) was highly expressed in Magnaporthe oryzae appressoria, and deleting Mgd1 decreased the ability to form the appressoria and resulted in a weaker virulence [8]. Candida albicans strains lacking GDH2 (gdh2−/−) did not extrude ammonia on amino acid-based media, but gdh2−/− mutants were as virulent as the wild-type strain [9]. Entomopathogenic fungi, such as Metarhizium robertsii, are used as biological control agents against a wide range of agricultural pests [10]. The conidia that germinate on an insect’s cuticle develop into an infectious structure known as an appressorium. Then the appressorium penetrates the cuticle via a synergistic action of turgor pressure and cuticle-degrading enzymes [11]. When the fungal hyphae successfully invade insect hemolymph, they transform into yeast-like cells known as blastospores or hyphal bodies. After insects are killed, the fungus in the hemolymph re-breaches the insect cuticle and produces multilayered spores that cover the cadaver’s surface [10]. Previous studies showed that M. robertsii can generate ammonia to alkalize the media and increase the pH of insect cuticle during infection [12,13]. However, the role of ambient alkalinization during M. robertsii infection remains undefined. In this study, we showed that M. robertsii Gdh2 (MrGdh2) was responsible for alkalization by catalyzing glutamate deamination and was required for amino acid-dependent fungal growth and reproduction during carbon deprivation. In contrast to plant fungal pathogens, MrGdh2 activity was not required for appressorium formation and cuticle penetration in M. robertsii. Differing from the human fungal pathogen C. albicans, MrGdh2 was essential for virulence by facilitating colonization in insect hemocoel. We therefore uncovered the unique role of Ggh2-dependent alkalization during the infection of entomopathogenic fungus M. robertsii. Material and methods Fungal strains and culture conditions M. robertsii strain ARSEF2575 was obtained from the Agricultural Research Service Collection of Entomopathogenic Fungi. The wild-type (WT) and transformants were routinely cultivated on potato dextrose agar (PDA, pH 5.6) (BD Difco, New Jersey, USA) at 25 °C. For growth analysis with different pH, the PDA or 1% casein medium was supplemented with the following buffers and adjusted to the indicated pH: 77 mM Na2HPO4, 60 mM citric acid (pH4.2); 30 mM Na2HPO4, 62 mM NaH2PO4, 80 mM NaCl (pH 6.8); 50 mM Tris-HCl, 125 mM NaCl (pH 9.4) [14]. For nutrient-rich cultivation, fungi were cultured on the complete medium (CM: 10 g/L glucose, 6 g/L NaNO3, 0.52 g/L KCl, 0.52 g/L MgSO4 · 7 H2O, 0.25 g/L KH2PO4, 1 g/L yeast extract, 2 g/L peptone, 1 g/L casamino acid, 1 ml/L trace elements, 1 ml/L vitamin supplement, pH 6). The minimal medium without glucose (MM: 6 g/L NaNO3, 0.52 g/L KCl, 0.52 g/L MgSO4 · 7 H2O, 0.25 g/L KH2PO4, 1 ml/L trace elements, 10 mg/L vitamin B1, pH 6 or 6.5) was used as a basal medium for nutrient studies [15]. The cuticle was dissected from Tenebrio molitor larvae and ground into powder, and fungal strains were cultivated on the MM supplied with 2% (w/v) cuticle powder for 2 to 10 days. Gene deletion and complementation The M. robertsii GDH2 gene (MAA_02480, MrGDH2) was deleted by homologous recombination via Agrobacterium-mediated fungal transformation as described previously [16]. In brief, the upstream and downstream flanking sequences were amplified from genomic DNA with primer sets UF/UR and DF/DR, respectively (Table S1). The PCR products were digested with the restriction enzymes Bam HI or Bcu I, and then inserted into the corresponding sites of the binary vector pDHt-bar (conferring resistance against ammonium glufosinate) for fungal transformation. For null mutant complementation, the 1350 bp of upstream sequence of MrGDH2, MrGDH2 gene (3279 bp), and the 308 bp of downstream sequence were amplified with primers (Comp-F and Comp-R) containing Xba I. The PCR products were digested and inserted into the binary vector pDHt-ben (conferring resistance against benomyl) for fungal transformation with ΔMrgdh2 [17]. Total RNAs of mycelia collected from 3 days of cultivation in sabouraud dextrose broth (SDB, pH 5.6) were extracted for confirming the expression of MrGDH2 gene by RT-PCR. Induction of appressorium and examination of cuticle penetration The sterile cicada wings were immersed in conidial suspension containing 4–10 × 107 conidia/mL for 30 s before being placed on the 0.8% agar plate as described previously [17]. Appressoria were examined and counted using a microscope after incubation for 18 h. To conduct the insect cuticle penetration assay, a segment of mealworm (T. molitor) abdominal cuticle was dissected [18]. A drop of 0.2 μL conidial suspension (1 × 107 conidia/mL) was inoculated in the center of the sterilized cuticle. The cuticles were removed after 48 h of cultivation on the agar plate, and the penetrated mycelia were stained with lactophenol cotton blue to determine the diameter under a microscope (Nikon eclipse Ni-U, Tokyo, Japan). RNA extraction and quantitative real-time PCR (qRT-PCR) analysis The WT strain was cultivated on PDA (pH 5.6) for 2 weeks, and conidia were collected for RNA extraction. Mycelia were harvested from 3 days of cultivation in SDB (pH 5.6). Appressoria were induced on the cicada wings, collected, and ground with liquid nitrogen for RNA extraction. Hyphal bodies were harvested from the hemolymph of 6th instar Spodoptera litura larvae injected with conidia for 48 h. Total RNAs of conidia, mycelia, appressoria, and hyphal bodies were extracted, respectively, with TRIzol and reversely transcribed into cDNAs with HiScript III RT SuperMix reagent kit (Vazyme, Nanjing, China). Gene expressions were determined by qRT-PCR with specific primers (Table S1) and ChamQ Universal SYBR Color qPCR Master Mix (Vazyme, Nanjing, China) on an Applied QuantStudio 6 Flex Real-Time PCR System. Three biological replicates were performed, and quantification of transcript level of the target gene was conducted using the 2−ΔΔCt method [19]. Enzyme activity assays The proteinase activity was detected by the azocasein assay [20]. In brief, azocasein was dissolved at a concentration of 5 mg/mL in a buffer containing 50 mM Tris (pH 7.5), 0.2 M NaCl, 5 mM CaCl2, 0.05% Brij 35, and 0.01% sodium azide. Appressoria were induced on the cicada wing or cellophane membrane for 36 h, collected, and ground with 10 mM PBS (phosphate buffered saline, pH 7.4) to extract total proteins. The supernatant (50 μL) was mixed with the azocasein solution (100 μL), and the mixture was incubated in a 37 °C water bath for 90 min. The reactions were stopped by trichloroacetic acid, and 50 μL of 1 M NaOH was added. The absorbance at 436 nm of released azo dye was determined with a spectrophotometer (FlexStation 3 microplate reader, Molecular Devices, California, USA). One unit of protease activity was defined as the amount of enzyme required to cause an increase in absorbance of 0.01 at 436 nm/min. Chitinase hydrolysis assay was performed on colloidal chitin (1%, wt/vol) medium containing 7 mM KCl, 2 mM MgSO4, 50 mM Na2HPO4-NaH2PO4 (pH 5.3), 0.05% (wt/vol) Triton X-100, and trace elements [14]. Fungal growth in the hemolymph and fat body The surface of Galleria mellonella larvae at the last instar was disinfected with 75% alcohol in a clean bench. Hemolymph was collected, and phenylthiourea (PTU) was added to inhibit melanization caused by exposure to air. Additionally, 1% kanamycin and 1% ampicillin were supplemented to prevent bacterial infections. Plasma (cell-free hemolymph) was harvested by centrifuged hemolymph at 2000 rpm for 10 min to remove hemocytes. Conidia were inoculated into hemolymph or plasma and grown for 5 to 7 days. The pH of the hemolymph or plasma after 6 days of fungal growth was indicated by adding phenol red. Fat bodies were dissected from sterilized G. mellonella larvae and washed in sterile water before placing in PBS containing 1% kanamycin and 1% ampicillin. Conidia were inoculated into PBS or PBS containing a layer of fat bodies and cultivated for 7 days. Insect bioassays Insect bioassays were conducted using the last instar larvae of the mealworm beetle T. molitor. Conidia from PDA plates were used for topical infection by spraying 800 μL suspension of 107 conidia/mL. For injection assays, 10 μL of an aqueous suspension containing 5 × 105 conidia/mL were injected into the second proleg of each larva. Each treatment had three replicates with at least 15 larvae in each replicate, and the experiments were repeated three times. Larval mortality was recorded every 12 h, and the LT50 values were estimated by Kaplan–Meier analysis [21]. Detection of total carbohydrate content To determine the total sugars in the hemolymph of G. mellonella larvae, the anthrone method was performed as previously described [22]. Glucose was used for the standard curve. Briefly, hemolymph from different treated larvae groups was collected and centrifuged at 2000 rpm for 10 min, and plasma was collected. Anthrone solution was made by adding 0.1 g anthrone to 80% concentrated sulfuric acid. Aliquots (20 μL) of plasma samples were mixed with 1.98 mL water and 6 mL anthrone solution, and the mixtures were treated at 90 °C for 15 min. After cooling on ice, the absorbance of samples at 625 nm was measured in a spectrophotometer (FlexStation 3 microplate reader, Molecular Devices, California, USA). Statistical analysis Graphical representation and analysis of the data were performed using GraphPad Prism 8.0 (GraphPad Software, Inc.). Multiple comparisons across strains were performed by one-way or two-way ANOVA with Tukey’s honest significance test (Tukey’s HSD) (***p < 0.001, **p < 0.01, *p < 0.05, and ns for p > 0.05). For multiple comparisons involving more than three samples, different letters indicate statistically significant differences (p < 0.05). Results MrGdh2 is responsible for alkalization during fungal growth on insect wing or cuticle powder Consistent with a previous study demonstrating that M. robertsii was able to alkalinize the media [13], we also found that M. robertsii was able to alkalinize environmental pH when cultured on insect wing for 7 days (Figure 1(a)). Gdh2 has been proven to catalyze ammonia accumulation in Colletotrichum species during the alkalinization process [7]. We found that MrGDH2 gene was expressed at the highest level in hyphal bodies (fungal hyphae in the hemocoel), followed by in appressoria and mycelia (Figure 1(b)). To evaluate the function of Gdh2 in M. robertsii, a deletion mutant (ΔMrgdh2) and a complement strain (Comp) were created (Figure S1). We found that the ΔMrgdh2 mutant failed to catalyze glutamate deamination, leading to its incapacity to generate ammonia and alkalization in media containing only insect wings or cuticle powders as well as in the minimal medium (MM) with 10 mM glutamate (Figure 1(c)). Figure 1.Fungal growth on medium containing insect wing or cuticle powder. (a) M. robertsii 2575 strain (WT) was able to alkalize ambient pH during growth on cicada wing. The wing treated with 0.05% tween solution was used as the blank control. 0.001% phenol red was supplied as pH indicator. The colorimetric response of 0.001% phenol red to pH was present. (b) The expression pattern of MrGDH2 gene in conidia, mycelia, appressoria and hyphal bodies. Values are means ± SD of three replicates; different letters indicate significant differences (one-way ANOVA with Tukey’s test, p < 0.05). (c) Growth of the WT, ΔMrgdh2 and Comp strains on minimal medium containing 10 mM glutamate (left), cicada wing (middle) or 2% cuticle powder (right) for 2 to 10 days. Fungal growth on medium containing insect wing or cuticle powder. (a) M. robertsii 2575 strain (WT) was able to alkalize ambient pH during growth on cicada wing. The wing treated with 0.05% tween solution was used as the blank control. 0.001% phenol red was supplied as pH indicator. The colorimetric response of 0.001% phenol red to pH was present. (b) The expression pattern of MrGDH2 gene in conidia, mycelia, appressoria and hyphal bodies. Values are means ± SD of three replicates; different letters indicate significant differences (one-way ANOVA with Tukey’s test, p < 0.05). (c) Growth of the WT, ΔMrgdh2 and Comp strains on minimal medium containing 10 mM glutamate (left), cicada wing (middle) or 2% cuticle powder (right) for 2 to 10 days. The full virulence of M. robertsii requires the activity of MrGdh2 To investigate the effect of MrGdh2 activity on pathogenicity, bioassays were performed by topically applying conidia to insect cuticle (cuticle-passing infection) or injecting conidia into hemocoel (cuticle-bypassing infection). The survival patterns of inoculated larvae showed that the lethal action of ΔMrgdh2 was delayed during cuticle-passing or cuticle-bypassing infections (Figure 2). As a topical application against T. molitor, ΔMrgdh2 had a substantially longer LT50 (4.52 ± 0.076 days) than the WT (4.0 ± 0.091 days, p < 0.001) or Comp (4.0 ± 0.061 days, p < 0.001) (Figure 2(a)). Topical infection against G. mellonella larvae also revealed a reduced virulence in the ΔMrgdh2 mutant when compared to the WT and Comp (Figure 2(c)). When T. molitor larvae were infected via direct injection, the ΔMrgdh2 mutant also had a longer LT50 (2.83 ± 0.036 days) than the WT (2.61 ± 0.034 days, p < 0.001) (Figure 2(b)), and similar results were observed when G. mellonella, Spodoptera litura, and Bombyx mori larvae were infected (Figures 2(d) and S1(c-d)). These results suggest that MrGdh2 activity is essential for the full virulence of M. robertsii against insect hosts. Figure 2.Insect bioassays. (a, c) survival of T. molitor (a) and G. mellonella (c) larvae following topical infection with spore suspensions. Control insects were treated with 0.05% tween 20. (b, d) survival of T. molitor (b) and G. mellonella (d) larvae following injection with conidia. Control insects were injected with 0.05% tween 20. Asterisk indicates significant differences (Kaplan−Meier analysis, ***p < 0.001, **p < 0.01, and *p < 0.05). Insect bioassays. (a, c) survival of T. molitor (a) and G. mellonella (c) larvae following topical infection with spore suspensions. Control insects were treated with 0.05% tween 20. (b, d) survival of T. molitor (b) and G. mellonella (d) larvae following injection with conidia. Control insects were injected with 0.05% tween 20. Asterisk indicates significant differences (Kaplan−Meier analysis, ***p < 0.001, **p < 0.01, and *p < 0.05). Deletion of MrGDH2 has no effect on appressorium formation and host cuticle penetration ex vivo Deletion of GHD2 decreased the formation of appressoria and resulted in reduced virulence in the plant pathogens [7,8,23]. To explore the reason for the reduced pathogenicity of ΔMrgdh2, we examined its capacity to generate appressorium. Interestingly, the morphology and formation ratio of appressoria on the cicada wings remained unchanged among the WT, ΔMrgdh2 and Comp strains (Figure 3(a)). Figure 3.Appressorium formation and penetration analyses. (a) Comparison of the morphology and formation rate of appressoria induced on the cicada wings for 18 h. Values are means ± SEM of replicates. Bar, 5 μm. (b) Fungal penetration on the cuticle of T. molitor larvae. The diameter of the penetrated mycelia was measured after inoculating conidia on the cuticle for 48 h. Values represent means ± SD of replicates. Bar, 500 μm. No significant difference (ns) was observed between groups based on one-way ANOVA with Tukey’s test. Appressorium formation and penetration analyses. (a) Comparison of the morphology and formation rate of appressoria induced on the cicada wings for 18 h. Values are means ± SEM of replicates. Bar, 5 μm. (b) Fungal penetration on the cuticle of T. molitor larvae. The diameter of the penetrated mycelia was measured after inoculating conidia on the cuticle for 48 h. Values represent means ± SD of replicates. Bar, 500 μm. No significant difference (ns) was observed between groups based on one-way ANOVA with Tukey’s test. Cuticle-degrading enzyme is a pathogenic factor for fungi to breakdown cuticle, and the expression and activity of enzymes are affected by changes in pH [12,24]. Compared to the WT and complement strains, the ΔMrgdh2 mutant showed no significant difference in the protease activity of appressoria induced on the cicada wing and cellophane membrane (Figure S2a-b). Compared to the ΔMrPacC mutant which was disrupted in chitinase activity [14], the WT, ΔMrgdh2 and Comp strains showed no difference in chitinase activity based on the growth in 1% colloidal chitin medium (Figure S2c). To further explore the role of MrGdh2 in fungal penetration, we examined its capacity to penetrate the insect cuticle and wing. The ΔMrgdh2 mutant displayed the same capacity as the WT and Comp strains to penetrate the larval cuticle of T. molitor and cicada wing (Figures 3(b) and S3). These findings suggest that Ggh2-dependent alkalization is not required for M. robertsii to produce appressoria or penetrate insect cuticle ex vivo. MrGdh2 is vital for amino acid-dependent fungal growth and reproduction during carbon deprivation The WT, ΔMrgdh2, and Comp strains exhibited identical colony diameter and conidial production on rich media such as PDA and complete medium (CM) (Figure 4(a-b)). Furthermore, these strains displayed no difference in conidial yield and pH alteration when grown on PDA buffered to pH 4.2, 6.8, and 9.4 (Figure S4). However, a reduction in glucose content from 10 mg/mL to 1 mg/mL impaired the ability of the ΔMrgdh2 mutant to alkalize the medium. The media inoculated with the WT and Comp strains exhibited a pH around 7.4; however, following cultivation with the ΔMrgdh2 mutant, the medium pH only reached approximately 6.4 (Figure 4(b)). The colony of ΔMrgdh2 was just slightly smaller than the WT and Comp strains, but a significantly lower number of spores emerged from the ΔMrgdh2 mutant (Figures 4(b-c)). Figure 4.Growth and conidiation of M. robertsii strains on media with different nutrition. (a) Growth of the WT, ΔMrgdh2 and Comp strains for 10 days on PDA plates. (b) Fungal strains grown for 10 days on complete medium containing 10 mg/mL glucose (CM) or 1 mg/mL glucose (CM-C). 0.001% phenol red was supplied as pH indicator. (c) The diameter and spore production of the colonies were analyzed. Values represent means ± SD of three replicates, the comparisons were performed by on one-way ANOVA with Tukey’s test (***p < 0.001, **p < 0.01, *p < 0.05 and ns for p > 0.05). Growth and conidiation of M. robertsii strains on media with different nutrition. (a) Growth of the WT, ΔMrgdh2 and Comp strains for 10 days on PDA plates. (b) Fungal strains grown for 10 days on complete medium containing 10 mg/mL glucose (CM) or 1 mg/mL glucose (CM-C). 0.001% phenol red was supplied as pH indicator. (c) The diameter and spore production of the colonies were analyzed. Values represent means ± SD of three replicates, the comparisons were performed by on one-way ANOVA with Tukey’s test (***p < 0.001, **p < 0.01, *p < 0.05 and ns for p > 0.05). To further explore the effect of nutrients on MrGdh2 activity, a minimal medium without glucose (MM) was used as a basal medium. Only sparse vegetative mycelia of the WT, ΔMrgdh2, and Comp strains were maintained on the MM medium; however, the amounts of mycelia and conidia were increased with increasing concentrations of glucose in the medium (Figure 5(a-b)). There was no difference in colony morphology or conidial yield among the WT, ΔMrgdh2 and Comp strains during growth under these conditions (Figure 5(a-c)). Interestingly, ΔMrgdh2 produced fewer aerial mycelia and conidia, and failed to alkalize MM with 1% hydrolyzed casein as an amino acid resource (Figures 5(d-f) and S5). The ΔMrgdh2 strain also failed to alter the pH in casein medium with 1 or 5 mg/mL glucose (Figure 5(d)). Although aerial mycelia and conidia of the ΔMrgdh2 strain were more abundant when grown in media with 5 mg/mL glucose than 1 mg/mL (Figure 5(a,d)), conidial yield of the mutant strain was decreased in casein media with both 1 and 5 mg/mL glucose when compared to the WT and Comp strains (Figure 5(e,f)). No differences were observed in fungal growth and conidia yield among the three strains when grown in casein medium with 10 mg/mL glucose (Figure 5(d-f)), and none of these strains altered the pH of the medium (Figure 5(d)), suggesting that M. robertsii does not necessitate deamination of amino acids to produce energy when glucose is sufficient. Figure 5.M. robertsii Gdh2 is required for amino acid-dependent alkalization during glucose deprivation. Minimal medium without glucose (MM) was used as a basal medium. The WT, ΔMrgdh2 and Comp strains were cultivated for 10 days on MM or hydrolyzed casein media supplemented with different concentrations of glucose (a, d). 0.001% phenol red was supplied as pH indicator. Conidial production (b, e) and diameter (c, f) of colonies were analyzed. Values are means ± SD of three replicates, the differences across strains within each medium were analyzed by one-way ANOVA with Tukey’s test (***p < 0.001, **p < 0.01, *p < 0.05 and ns for p > 0.05). M. robertsii Gdh2 is required for amino acid-dependent alkalization during glucose deprivation. Minimal medium without glucose (MM) was used as a basal medium. The WT, ΔMrgdh2 and Comp strains were cultivated for 10 days on MM or hydrolyzed casein media supplemented with different concentrations of glucose (a, d). 0.001% phenol red was supplied as pH indicator. Conidial production (b, e) and diameter (c, f) of colonies were analyzed. Values are means ± SD of three replicates, the differences across strains within each medium were analyzed by one-way ANOVA with Tukey’s test (***p < 0.001, **p < 0.01, *p < 0.05 and ns for p > 0.05). Trehalose is a major circulatory sugar in insect hemolymph [25]. Similar outcomes to those observed with glucose were achieved when trehalose served as the carbon source (Figure 6). When the WT, ΔMrgdh2 and Comp strains were supplied with 1% gelatin as a nitrogen source [26], the colony diameter and conidial yield of ΔMrgdh2 strain were also decreased in gelatin media containing 0, 1, or 5 mg/mL glucose or trehalose (Figure S6). In addition, the growth defects of ΔMrgdh2 in casein media were independent of the environmental pH (Figure S7). These results suggest that during carbon deprivation, fungal development requires the MrGdh2 activity to convert amino acids to produce α-ketoglutarate and NADH, both are vital intermediates for energy metabolism [27]. Figure 6.M. robertsii Gdh2 is required for amino acid-dependent alkalization during trehalose deprivation. The WT, ΔMrgdh2 and Comp strains were cultivated for 10 days on MM or hydrolyzed casein media supplemented with different concentrations of trehalose (a, d). 0.001% phenol red was supplied as pH indicator. Conidial production (b, e) and diameter (c, f) of colonies were analyzed. Values are means ± SD of three replicates, the differences across strains within each medium were analyzed by one-way ANOVA with Tukey’s test (***p < 0.001, **p < 0.01, *p < 0.05 and ns for p > 0.05). M. robertsii Gdh2 is required for amino acid-dependent alkalization during trehalose deprivation. The WT, ΔMrgdh2 and Comp strains were cultivated for 10 days on MM or hydrolyzed casein media supplemented with different concentrations of trehalose (a, d). 0.001% phenol red was supplied as pH indicator. Conidial production (b, e) and diameter (c, f) of colonies were analyzed. Values are means ± SD of three replicates, the differences across strains within each medium were analyzed by one-way ANOVA with Tukey’s test (***p < 0.001, **p < 0.01, *p < 0.05 and ns for p > 0.05). The activity of MrGdh2 is required for efficient fungal colonization in insect hemolymph Both immune defense and pathogen replication within the host involve substantial energy demands [28]. We proposed that M. robertsii infection may cause carbon deprivation in insect hemolymph. We found that total carbohydrate content in the hemolymph of uninfected G. mellonella larvae was around 13 mg/mL (Figure 7(a)). However, the carbohydrate content in the hemolymph of larvae infected with M. robertsii was significantly reduced, dropping to 10 mg/mL at 1 d (1 day), 8.8 mg/mL at 1.5 d, 3.5 mg/mL at 2 d, and 2.5 mg/mL at 2.5 d after injection of M. robertsii conidia (Figure 7(a)). The carbohydrate contents in the hemolymph of ΔMrgdh2-infected larvae were also decreased but were slightly higher at 2 d and 2.5 d after infection compared to the WT and Comp strains (Figure 7(a), Table S2). Figure 7.Hemolymph carbohydrate contents during M. robertsii infection and fungal growth in the hemolymph. (a) Carbohydrate contents in the hemolymph of G. mellonella larvae infected by the WT, ΔMrgdh2 or Comp strain were quantified. Control larvae were treated with 0.05% tween 20 (tw, control). Values are means ± SD of three replicates. The comparisons across infections by different strains were performed by two-way ANOVA with Tukey’s test (also see Table S2), with different letters indicating statistically significant differences (p < 0.05). (b) The length of hyphal bodies in the hemolymph. Hyphal bodies emerged in the hemolymph of S. litura larvae were observed by light microscope after injection of the WT, ΔMrgdh2 or Comp strain for 2 days. (c) The expression level of MrPacc in hyphal bodies of the WT, ΔMrgdh2 or Comp strain was analyzed by qRT-PCR with EF-1 as a reference gene. Values are means ± SD, the comparisons were performed by one-way ANOVA with Tukey’s test (***p < 0.001, **p < 0.01, *p < 0.05 and ns for p > 0.05). (d) The growth of the WT, ΔMrgdh2 and Comp strains in the hemolymph or plasma for 5 and 7 days. (e) The pH of the hemolymph or plasma was indicated by adding phenol red after inoculating the WT, ΔMrgdh2 or Comp strain for 6 days. Hemolymph carbohydrate contents during M. robertsii infection and fungal growth in the hemolymph. (a) Carbohydrate contents in the hemolymph of G. mellonella larvae infected by the WT, ΔMrgdh2 or Comp strain were quantified. Control larvae were treated with 0.05% tween 20 (tw, control). Values are means ± SD of three replicates. The comparisons across infections by different strains were performed by two-way ANOVA with Tukey’s test (also see Table S2), with different letters indicating statistically significant differences (p < 0.05). (b) The length of hyphal bodies in the hemolymph. Hyphal bodies emerged in the hemolymph of S. litura larvae were observed by light microscope after injection of the WT, ΔMrgdh2 or Comp strain for 2 days. (c) The expression level of MrPacc in hyphal bodies of the WT, ΔMrgdh2 or Comp strain was analyzed by qRT-PCR with EF-1 as a reference gene. Values are means ± SD, the comparisons were performed by one-way ANOVA with Tukey’s test (***p < 0.001, **p < 0.01, *p < 0.05 and ns for p > 0.05). (d) The growth of the WT, ΔMrgdh2 and Comp strains in the hemolymph or plasma for 5 and 7 days. (e) The pH of the hemolymph or plasma was indicated by adding phenol red after inoculating the WT, ΔMrgdh2 or Comp strain for 6 days. After penetration into insects, fungal hyphae secrete enzymes and metabolites to inhibit host immunity or take in host nutrients, which facilitate the successful colonization of fungi in the hemocoel [29]. The number of hyphal bodies was similar between the WT and ΔMrgdh2 mutant in the hemolymph of S. litura larvae (Figure S8a); however, the length of hyphal bodies of the ΔMrgdh2 mutant was shorter than the WT and Comp strains in the hemolymph (Figures 7(b), S8(b)), indicating that MrGdh2 participated in fungal growth in insect hemocoel. The pH-responsive PacC transcription factor contributes to virulence by regulating blastosphere budding in M. rileyi [30]. We found that the expression of MrPacC in hyphal bodies of the ΔMrgdh2 mutant was lower compared to the WT and Comp strains (Figure 7(c)). To further investigate the impact of MrGdh2 activity during fungal colonization in host hemocoel, hemolymph was used as nutrition source for fungal growth. The hemolymph or plasma (cell-free hemolymph) was entirely covered by the white mycelia of the WT and Comp strains after 5 days of cultivation, whereas the growth of ΔMrgdh2 was delayed (Figure 7(d)). The conidia yield of ΔMrgdh2 was also notably reduced after cultivating for 7 days (Figure 7(d)). Correspondingly, ΔMrgdh2 failed to alkalize hemolymph or plasma (Figure 7(e)). In addition, there was no notable difference in hyphal growth between the WT and ΔMrgdh2 after 7 days of cultivation in PBS supplemented with fat bodies from G. mellonella larvae (Figure S8c). These results imply that MrGdh2 is crucial for fungal colonization in insect hemocoel. Discussion Fungal infections are often associated with pH changes in the surrounding host tissues [31]. Plant fungal pathogens induce alkalinization through regulated release of ammonia, and in certain phytopathogens by secreting small regulatory peptides that mimic plant rapid alkalinizing factors (RALFs) [3,4]. M. robertsii can alkalinize artificial medium by production of ammonia [13]. Given that yeast Gdh2 mediates glutamate deamination to α-ketoglutarate and ammonia [5], MrGdh2 also participated in deamination of glutamate (Figure 1(c)). A previous study reported that M. robertsii was able to alkalinize insect cuticle during infection [12]. We showed that M. robertsii alkalinized the medium containing wing or cuticle powder as a nutrition source after growth for 7 days, but this ability was impaired upon MrGDH2 deletion (Figure 1(c)). In plant pathogens M. oryzae and Colletotrichum, GDH2 deletion disrupts appressorium formation and virulence [7,8,23]. Although MrGdh2 activity is required for virulence during cuticle-passing bioassay (Figure 2(a)), the abilities to form appressorium and penetrate cuticle were not affected in ΔMrgdh2 (Figures 3, S3). Ambient pH is an important determinant for cuticle-degrading enzymes to penetrate insect cuticle by M. robertsii [11,12]. Deleting MrGDH2 resulted in no changes in protease and chitinase activities (Figures 3(b), S2, S3). The pH around the wing remained unchanged during the first 48 h post-inoculation (hpi), a period coinciding with appressorium formation (Figure 1(c)), suggesting that MrGdh2 appears to be dispensable for appressorium formation and cuticle penetration during early infection but essential for subsequent fungal growth under nutrient-limited conditions. Metarhizium was likely originated from plant endophytes or pathogens [32]. However, Gdh2 contributes differently to appressorium formation between Metarhizium and plant pathogens, suggesting distinct mechanistic adaptations in host surface recognition and infection strategies. Carbon limitation triggers extracellular accumulation of ammonia and alkalinization in several plant and human fungal pathogens [9,23]. For example, Colletotrichum gloeosporioides alkalinizes tryptone-containing media under carbon deprivation [23]. Unlike the WT, ΔMrgdh2 failed to alkalize hydrolyzed casein media supplemented with low glucose/trehalose (0–5 mg/mL; Figures 5–6) or CM with 1 mg/mL glucose (Figure 4(b)). However, neither strain alkalinized PDA, CM, MM, or 1% casein medium with 1% carbon source (Figures 4–6), suggesting that MrGdh2 mediates glutamate deamination specifically during carbon shortage when amino acids are available, with alkalization occurring as a metabolic byproduct of amino acid utilization. C. albicans Gdh2 expression is repressed by glucose [9]. We also found that alkalization of the casein media disappeared after increasing glucose content to 10 mg/mL (Figure 5). Intriguingly, both the WT and ΔMrgdh2 strains alkalized amino acid-free MM containing 10 mg/mL glucose/trehalose (Figure 5(a), 6(a)), but this ability was lost upon casein addition (Figure 5(d), 6(d)). These findings suggest that glucose metabolites may undergo enhanced catalysis to compensate for the nitrogen deficiency in the absence of amino acids, resulting in alkalinization. In C. albicans, Gdh2 is required for growth using amino acids as sole carbon and nitrogen sources [9]. Similarly, in M. robertsii, MrGdh2 is vital for mycelia growth (particularly aerial mycelia) under such conditions. Moreover, MrGDH2 deletion significantly impaired conidiation (Figure 5(d), 6(d)). Aerial hyphae growth depends on vegetative mycelia for nutrients, and serves as the foundation of conidiation [32,33]. Accordingly, increasing the carbon source enhanced both aerial mycelia growth and conidia yield in M. robertsii (Figure 5,6). Gdh2 catalyzes the deamination of glutamate to α-ketoglutarate, a key metabolic intermediate required for many biological processes and linking amino acid metabolism to the TCA cycle [5,6]. In ΔMrgdh2, loss of MrGdh2 function abolished α-ketoglutarate production from amino acids, and carbon limitation halts the TCA cycle, resulting in defects in both aerial mycelia growth and conidiation. C. albicans gdh2−/− mutants were as virulent as the wild-type strain [9]. However, loss of MrGDH2 led to reduced virulence upon hemocoel injection (Figure 2(b-c)), indicating that MrGdh2 is involved in host colonization within insect hemocoel. Both immune defense and pathogen replication require substantial energy during infection [28], and hemolymph-derived nutrients vary across host-parasite systems [34,35]. For instance, parasitism of Heliothis virescens larvae by Microplitis croceipes but not Cardiochiles nigriceps resulted in an increase in host hemolymph trehalose concentration [36], whereas parasitization of Pimpla turionellae in G. mellonella pupae reduced the pupal hemolymph carbohydrate at 24 h postoviposition [37]. During M. robertsii infection, G. mellonella hemolymph carbohydrates dropped from 13 to 2.5 mg/mL (Figure 7(a)), suggesting carbon limitation. Amino acids are particularly important nutrients that are central to the host and bacterial pathogen interaction [37]. MrGDH2 expression peaks in hyphal bodies (Figure 1(b)); its deletion shortened hyphal bodies and delayed growth in hemolymph/plasma (Figures 7(b), S8(c)). As hemolymph is rich in amino acids [38,39], we propose that amino acids serve as critical nutrients upon carbon source depletion during M. robertsii infection, and MrGdh2 facilitates host colonization via amino acid catabolism. The pH-responsive PacC transcription factor is required for environmental fitness, development, and virulence of different fungal pathogens [30,40]. In Metarhizium, PacC contributes to virulence by regulating cuticle penetration and immune evasion in M. acridum and M. robertsii, as well as by regulating blastosphere budding in M. rileyi [14,30,41]. MrPacC expression in appressoria of ΔMrgdh2 was similar to that in the WT and Comp strains (Figure S9), while it was downregulated in hyphal bodies of ΔMrgdh2 (Figure 7(c)), implying a potential link between Gdh2 and PacC in blastosphere growth. Further studies are needed to elucidate this interaction. Our study revealed that Gdh2 activity is required for virulence by facilitating fungal colonization in insect hemolymph in the entomopathogenic fungus M. robertsii, which is distinct from the established models of plant and human fungal infections. These findings advance our understanding of the diverse mechanisms of fungal infection across various hosts. Supplementary Material Table S2.xlsx Supplementary Material revise2.docx Figure S6.tif Figure S7.tif Figure S4.tif