G-protein subunit Ga i in mitochondria, MrGPA1, affects conidiation, stress resistance, and virulence of entomopathogenic fungus Metarhizium robertsii

G proteins are critical modulators or transducers in various transmembrane signaling systems. They play key roles in numerous biological processes in fungi, including vegetative growth, development of infection-related structures, asexual sporulation, and virulence. However, their function in entomopathogenic fungi remains unclear. Here, we characterized the roles of MrGPA1, a G-protein subunit Gα i , in conidiation, stress resistance, and virulence in Metarhizium robertsii . MrGPA1 was localized in the mitochondria. MrGpa1 deletion resulted in a signi�cant reduction (47%) in the conidiation capacity, and reduced expression of several key conidiation-related genes, including �uG , �bD , brlA , wetA , phiA , and stuA . Further, MrGpa1 disruption resulted in decreased fungal sensitivity to UV irradiation and thermal stress, as determined based on conidial germination of ∆ MrGpa1 and wild-type strains. Chemical stress analysis indicated that MrGpa1 contributes to fungal antioxidant capacity and cell wall integrity, but is not involved in antifungal ability and osmotic stress. Importantly, insect bioassays involving (topical inoculation and injection) of Galleria mellonella larvae revealed decreased virulence of ∆ MrGpa1 strain after cuticle infection. This was accompanied by decreased rates of appressorium formation and reduced expression of several cuticle penetration-related genes. These observations suggest that MrGpa1 contributes to the regulation of conidiation, UV irradiation, thermal stress response, antioxidant capacity and cell wall integrity in M. robertsii . This gene is also involved in insect cuticle penetration during infection. These �ndings raise the possibility of designing powerful strategies for genetic improvement of M. robertsii conidiation capacity and virulence for killing pests.


Introduction
G protein with GTP-hydrolase activity is a type of signaling protein that binds to guanine nucleotides (Robishaw and Berlot 2004).It participates in signal transduction pathways linking activated cell-surface receptors with intracellular effectors, including adenylate cyclase and phospholipase through a series of signaling cascades involved in the regulation of physiological and biochemical processes (Chakravorty and Assmann 2018;Ortiz-Urquiza and Keyhani 2015).In fungi, G protein is associated with sexual and asexual reproduction, virulence and response to external signal stimuli (Guo et al. 2016b;Ivey et al. 1996;Liu and Dean 1997).
The heterotrimeric G protein is composed of three subunits α, β, and γ, wherein α subunit binds to GDP, and β and γ subunits from a heterodimers (Birnbaumer 2007;Lambert 2008;McIntire 2009).When the heterotrimeric G protein is stimulated by a G-protein-coupled receptor (GPCR) that senses external signals, GDP is exchanged for GTP, and Gα and Gβγ complexes dissociate (Wedegaertner 2012).Then, Gα-GTP and Gβγ act on the respective downstream effectors (Barren and Artemyev 2007).The cycle is reset by the hydrolysis of GTP to GDP, and Gα recombining with Gβγ and GPCR (Gilchrist et al. 1999;Slessareva and Dohlman 2006).
In mammals, G-protein α subunits are divided into four classes, Gα s , Gα i , Gα q , and Gα 12 , based on the amino acid sequence identity (Neer 1995).Further, the Gα i family is composed of four subfamilies, Gα i , Gα o , Gα t and Gα z .The functions of Gα i family proteins are diverse, and include regulation of adenylyl cyclase, K + and Ca + channels, and cGMP phosphodiesterase activities (Simon et al. 1991).The conserved functional motif of Gα i protein is characterized by possession of N-myristoylation and ADP-ribosylation (Buss et al. 1987).
The functions of the G-protein α subunit have been characterized in some fungi.In Saccharomyces cerevisiae, two kinds of Gα proteins have been identi ed (Gpa1 and Gpa2).Gpa1 is involved in pheromone regulation (Jahng et al. 1988), while Gpa2 regulates pseudohyphal development via cyclic AMP (cAMP)-dependent pathways and heat resistance (Kubler et al. 1997).However, these proteins have different functions in other lamentous fungi.For example, Fusarium oxysporum f. sp.cubense possesses three Gα proteins (Gα-fga1, -fga2, and -fga3), and the deletion of encoding genes leads to phenotypic defects in colony morphology, reduced conidiation, increased heat tolerance, reduced virulence, and decreased intracellular cAMP levels (Guo et al. 2016a, b;Jain et al. 2002).Further, three different Gα proteins control unique signal transduction pathways in Magnaporthe grisea, in uencing fungal vegetative growth, conidiation, conidium attachment, appressorium formation, mating, and pathogenicity (Liu and Dean 1997;Zhang et al. 2012).However, it is unclear whether the Gα proteins are involving in vegetative growth, conidiation, stress resistance, and virulence in entomopathogenic fungi, such as Metarhizium robertsii.
M. robertsii, an important entomopathogenic fungus, has been developed as an environmentally friendly alternative to chemical insecticides (Frazzon et al. 2000;Lord 2005;Wang and Wang 2017).
Unfortunately, commercialized broad application of M. robertsii formulations is limited by the low conidiation rate, failure of conidia germination under high-temperature and UV stress, slow killing speed, and inconsistent eld performance (Fang et al. 2012;Faria and Wraight 2007;Muniz-Paredes et al. 2017).
Genetic improvements of this mycoinsecticide require extensive understanding of the molecular mechanisms and M. robertsii genes involved in stress tolerance and virulence (Zhang and Feng 2018).
In the current study, we aimed to investigate the role of Gα proteins in M. robertsii.BLASTP search of against the assembled draft genome sequence of M. robertsii identi ed four putative Gα proteins.Among these Gα proteins, MrGPA1 (EFZ00892) shared the highest identity (96.32%) with GNA-1 protein of Neurospora crassa, which is required for the extension of basal hypha, growth, conidiation, and formation of female reproductive structures (Ivey et al. 1996;Yang and Borkovich 1999).We then, characterized the biological function of MrGPA1 by constructing and analyzing MrGpa1 gene deletion mutant.We show that MrGpa1 in uences conidiation, stress resistance, and virulence in M. robertsii.

Sequence analysis
To construct the phylogenetic tree of GPA proteins and analyze the structural domains of guanine nucleotide-binding site, amino acid sequences of the G-protein a subunits were downloaded from the National Center for Biotechnology Information (https://www.ncbi.nlm.nih.gov/), and phylogenetic analysis was performed using MEGAX software (https://www.megasoftware.net/).

Construction of MrGPA1-GFP fusion vector and analysis of subcellular localization of MrGPA1
To monitor subcellular localization of MrGpa1, gfp and MrGpa1 gene fragments were ampli ed by polymerase chain reaction (PCR), using gfp-F/gfp-R, and gfpMrGpa1-F/gfpMrGpa1-R primers (Supplementary Table S1), high-delity Taq DNA polymerase (KOD Plus Neo, Toyobo, Osaka, Japan) and M. robertsii genomic DNA as a template.The ampli cation products were inserted into the EcoRI restriction site in pDHt-SK-bar vector (kindly provided by Dr. ChengshuWang; the vector conferred resistance against glufosinate-ammonium) (Fang et al. 2006) containing a strong promoter and terminator to generate vector pDHt-MrGpa1-gfp for Agrobacterium tumefaciens transformation.The corresponding transformants resistant to glufosinate ammonium were obtained, and veri ed by PCR using the primers gfp-F and gfp-R (Supplementary Table S1).
The MrGPA1-GFP strain was cultured on sabouraud dextrose agar medium containing yeast extract (SDAY 4% glucose, 1% peptone, 2% agar, and 1% yeast extract powder, w/v) at 25 °C for 2 days.The hyphae were then washed off the plate with sterile water and mixed with 500 nM MitoTracker Red CMXRos (Invitrogen, Shanghai, China), a dye speci c to mitochondria.Subcellular localization of MrGPA1 was evaluated using a laser scanning confocal microscopy (LSCM, Zeiss LSM880).Before using laser scanning confocal microscopy, Wolf PSORT software (https://wolfpsort.hgc.jp) was used for prediction subcellular localization by analysing protein sequence of MrGPA1.
For gene complementation, the entire MrGpa1 gene and the 1000-bp upstream sequence and 600-bp downstream sequence were inserted into vector pDHt-SK-ben (containing benomyl resistance gene) digested with the SpeI restriction enzyme for fungal transformation.The 3050-bp fragment was ectopically integrated into ∆MrGpa1 strain by the same method as that used for gene deletion.
Complemented strains (cp∆MrGpa1) were obtained by selection for benomyl resistance, and veri ed by PCR using primer pairs MrGpa1-F/MrGpa1-R and ben-F/ben-R (Supplementary Table S1).

Phenotype assays
For phenotype assays, these experiments were performed with 3 technical and biological replicates per strain (WT, ∆MrGpa1, and cp∆MrGpa1).
Fungal conidiation ability was evaluated as previously described (Meng et al. 2017).Brie y, 30 ml of conidial suspension( 1 ´ 10 6 conidia /ml) was spread on PDA plate (35-mm diameter).After culturing at 25 °C for 14 days, the conidia on each plate were collected into 30 mL of 0.05% Tween-80 by vortexmixing, and conidial density was determined using a hemocytometer and converted to the number of conidia per square centimeter of colony.
For conidial germination assay, 10 ml of conidial suspension (5 ´ 10 6 conidia /ml) were spread on PDA medium.The conidial germination was observed by microscope (Olympus BX 51, Tokyo, Japan) at 2,4,6,8,10,12,14,16,18,20,22,24 hours after incubated at 25 °C.Conidia are considered to be germinated when the length of the germ tube reaches or longer than the length of the conidia (Wang et al. 2014).Three hundred conidia were counted at least by per plate and the germination rates were calculated by comparing the number of germinated conidia with the 300 counted conidia, and the median germinate time (GT 50 ) was calculated using the SPSS software For heat stress tolerance assays, 1 ml of conidial suspensions ( 5´ 10 6 conidia /ml) of WT, ∆MrGpa1, and cp∆MrGpa1 strains were placed in 1.5-ml Eppendorf tubes, and then incubated in a water bath at 42 °C or 28 °C (as control) for 1 h.Then 10 ml of the suspension were spread on PDA medium, incubated at 25 °C.
Conidial germination was observed under a microscope (Olympus BX 51, Tokyo, Japan) after 16 h and 24 h.Three hundred conidia were counted at least by per plate and the relative germination rates were calculated by comparing the number of germinated conidia with had not been heat stressed (Wang et al. 2019).
To determine fungal tolerance to ultraviolet B (UV-B) light, 10 ml of conidial suspensions ( 5 ´ 10 6 conidia /ml) of WT, ∆MrGpa1, and cp∆MrGpa1 was taken to PDA medium.The plates were then exposed to UV-B irradiation (312-nm wavelength at 100 mJ cm -2 ) using HL-2000 Hybrilinker (UVP, CA, USA) (Yao et al. 2010) or exposed to sunlight (as control).Relative UV-B tolerance was assessed and calculated by aforementioned methods to assess relative germination rate of tolerance to UV-B.
To assess the effects of MrGpa1 on virulence, bioassays with Galleria mellonella larvae (RuiQing Bait, Shanghai, China) were performed as described previously (Zhou et al. 2018).The larvae were immersed in conidial suspension (1 ´ 10 6 conidia /ml) for 90 s or injected (into the hemocoel) with 10 ml of conidial suspensions (1 ´ 10 5 conidia /ml) and incubated at 25 °C, Each treatment was performed in triplicate, with 18 larvae in each group.The experiment was repeated three times.Larva mortality was evaluated every 24 h, and the median lethal time (LT 50 ) was calculated using the SPSS software.
The appressorium formation assay was performed as described previously (Gao et al. 2013).Brie y, to test the appressorium formation on a hydrophobic surface, 1 ml of conidial suspension (1 ´ 10 6 conidia /ml) in MMGly (minimal medium amended with 1% glycerol) was spread on a sterile plastic Petri dishes (3.5-cm diameter), followed by 24 h incubation at 25 °C.At least 300 conidia of each strain were evaluated microscopically, and the induction rates of appressorium formation were quanti ed by observing different microscopic elds (inverted microscope, Olympus IX 71, Tokyo, Japan).

Quantitative RT-PCR (RT-qPCR)
To analyze the expression of conidiation-related genes, 200 ml of conidial suspensions (1 ´ 10 7 conidia/ml) of WT, ∆MrGpa1, and cp∆MrGpa1 strains were plated on PDA medium, and cultured in the dark at 25 °C for 2.5 days.The samples were collected and milled in liquid nitrogen to extract total RNA.
To analyze the expression of virulence genes related to cuticle infection, G. mellonella larvae were dipped in conidial suspensions (5 ´ 10 7 conidia/ml) of WT, ∆MrGpa1, and cp∆MrGpa1 strains for 1.5 min, transferred to 25 °C for 48 h, and then placed in liquid nitrogen for total RNA extraction.Total RNA was extracted by using Trizol reagent (Invitrogen, Foster City, CA, USA).cDNA was obtained by using the PrimeScript™ RT reagent kit with gDNA Eraser (TaKaRa, Dalian, China), and used as a template for RT-qPCR.The gene expression analysis was performed by using the CFBR96 TM Real-Time PCR System (Bio-Rad, Hercules, CA, USA) and SYBR® PremixEx TaqTM II (TaKaRa).Three biological repeats of each treatment were analyzed.The qPCR primers are listed in Supplementary Table S2.The expression of the gpd gene (MAA_07675, encoding glyceraldehyde 3-phosphate dehydrogenase) was used as an internal control (Fang and Bidochka 2006).The relative gene expression was calculated by using the 2 -ΔΔCt method (Livak and Schmittgen 2001).

Statistical analysis
All data are presented using GraphPad Prism version 6.0.Data are expressed as the mean ± standard error (SE) of the mean, from three biological replicates.Statistical analysis was performed by one-way analysis of variance (ANOVA).For multiple comparisons, Tukey's multiple comparison test was used to analyze statistical the signi cance.p < 0.05 was considered to be signi cant, and p < 0.01 was considered to be extremely signi cant.
Further bioinformatics analysis indicated that MrGpa1 (MAA-03488) is a single copy gene encoding Gprotein a subunit (353-aa protein) in M. robertsii.A BLASTP search of MrGPA1 homologs in NCBI revealed that the protein shares 100% amino acid similarity with GPA from M. acridum (XP_007811324), M. anisopliae (KFG82129), M. brunneum ((XP_014542423) and M. rileyi (OAA44756).Phylogenetic tree of GPA proteins from Metarhizium spp, and related fungal species was constructed with Saccharomyces cerevisiaeas an outgroup (Fig. 1a).All GPA proteins from the genus Metarhizium formed an independent branch (100% support value).These Metarhizium GPA proteins are closely related to GPA proteins from Pochonia chlamydosporia, Moelleriella libera and Purpureocillium lilacinum.
A conserved domain database search demonstrated that MrGPA1 contains a highly conserved guanine nucleotide-binding site (34-347-aa protein), which is the key identi cation domain of the G-protein a subunit (Fig. 1b).Furthermore, homologous alignment revealed the presence of N-myristoylation and ADP-ribosylation sites (two conserved positions in the G-protein a i subunit) in the conserved functional motif of MrGPA1.Hence, MrGPA1 is a member of the Ga i family.

MrGPA1 is a mitochondria protein
To investigate the subcellular localization of MrGPA1, the Wolf PSORT software was rst used.The analysis predicted that the protein is localized in the mitochondrial inner membrane.To verify this prediction, we generated MrGpa1-gfp strain (Supplementary Fig. S1a, b).As shown in LSCM images in Fig. 2, the mitochondria in hyphal is stained with a mitochondrial dye (red), in a punctate patterns, while punctate green uorescence was also observed in vegetative hyphae.Red and green uorescence was detected and overlapped (Fig. 2), suggesting that MrGPA1 is a mitochondria protein.

Construction of MrGpa1 knockout and complementation
To investigate the function of MrGPA1 in detail, we next generated MrGpa1 gene replacement and complemented strains (Supplementary Fig. S1a).The nature of the generated mutant strains was con rmed by using genomic DNA PCR.The analysis indicated the presence of a 1, 222-bp fragment corresponding to the partial MrGpa1 gene sequence in the WT and cp∆MrGpa1 stains, but not in the ∆MrGpa1 strain.In addition, a partial 806-bp bar gene fragment was detected in ∆MrGpa1 and cp∆MrGpa1 strains, and a partial 785-bp ben gene fragment was detected in the cp∆MrGpa1 strain.Furthermore, PCR analysis indicated the presence of a fragment containing upstream sequence of MrGpa1 and a partial bar gene (2689-bp ) and a fragment containing downstream sequence of MrGpa1 and a partial bar gene (1975-bp), and detected by using the primer sets up MrGpa1-F/upMrGpa1-R and dnMrGpa1-F/dnMrGpa1-R, respectively, in the ∆MrGpa1 strain (Supplementary Fig. S1c).Finally, RT-PCR analysis veri ed the loss or regain of the MrGpa1 gene expression in ∆MrGpa1 and cp∆MrGpa1 strains, accordingly (Supplementary Fig. S1d).These observations indicated a successful construction of the MrGpa1 knockout and complementation strains.

MrGpa1 contributes to fungal conidiation but is not involved in vegetative growth
To examine the effect of MrGpa1 on the growth and development of M. robertsii, we evaluated mycelial growth and conidial yield of WT and mutant strains on PDA and 1/4 SDAY medium, respectively.The 14day-old colonies of WT, ΔMrGpa1, and cp∆MrGpa1 strains formed 6.18 ´ 10 7 , 3.28 ´ 10 7 , and 6.55 ´ 10 7 conidia /cm −2 , respectively (Fig. 3a).The loss of MrGpa1 caused a signi cant, 47% reduction in conidiation, but little difference in the growth rate of WT, ΔMrGpa1, and cp∆MrGpa1 strains on PDA and 1/4 SDAY media was apparent (Fig. 3a, b).We also examined the expression of genes involved in conidiation in M. robertsii by RT-qPCR.The expression of uG, bD, brlA, wetA, phiA and stuA genes in the MrGpa1 strain was signi cantly reduced compared with that in the WT and cp∆MrGpa1 strains (Fig. 3c).
Collectively, these observations indicate that while MrGPA1 plays an important role in the conidiation of M. robertsii, it is not involved in vegetative growth.
MrGpa1 is important for heat and UV stresses tolerance, and is involved in antioxidant capacity and cell wall integrity of M. robertsii We observed that conidial germination rate of ΔMrGpa1 strain on PDA medium was signi cantly higher than that of WT and cp∆MrGpa1 strains.The GT 50 values for WT, ∆MrGpa1, and cp∆MrGpa1 were 11.99 h, 6.16 h (p < 0.01, compared with WT strain), and 12.16 h, respectively (Fig. 4a).
To investigate the effects of MrGpa1 deletion on UV irradiation and thermal stress, the relative germination rate of conidia exposed to these stresses was determined 16-h or 24-h after stress exposure.We found that the sensitivity of ∆MrGpa1 strain to 42 °C heat-stress was reduced.For example, compared with the WT, ∆MrGpa1 germination rates at 16 h and 24 h increased by 51% (p < 0.01) and 24% (p < 0.05), respectively (Fig. 4b).Similar results were obtained for conidial tolerance of UV irradiation; compared with the WT, ∆MrGpa1 germination rates at 16 h increased by 83% (p < 0.01), but only by 8% (p < 0.05) at 24 h (Fig. 4b).Hence, it appears that MrGPA1 plays an important role in conidial tolerance of both UV irradiation and thermal stress.
To evaluate the role of MrGpa1 in fungal growth under different chemical stress conditions, we investigated the mycelial growth of the WT and mutant strains on PDA containing carbendazim, NaCl, H 2 O 2 , or Congo red.The antioxidant capacity and cell wall integrity of ΔMrGpa1 strain were signi cantly different than those of the WT and cp∆MrGpa1 strains.For instance, compared with the WT, the relative inhibition of ∆MrGpa1 growth was decreased by 68.6% (p < 0.01) on PDA containing H 2 O 2 , while the sensitivity to Congo red was increased by 47.4%(p < 0.01).However, the relative inhibition of ∆MrGpa1 growth in the presence of carbendazim and NaCl was not markedly different from that of the control strains (Fig. 4c).These observations indicate that MrGpa1 contributes to fungal antioxidant capacity and cell wall integrity, but is not involved in antifungal ability and osmotic stress.

MrGpa1 plays an important role in insect cuticle penetration via appressorium formation
We next used G. mellonella bioassays to assess the consequences of MrGpa1 deletion on fungal virulence.In topical infection bioassays, the mean lethal times to death (LT 50 ) in insects infected with ∆MrGpa1, WT and cp∆MrGpa1 strains were 7.2 ± 0.45, 11.8±0.54 and 8.3 ± 0.71 days, respectively, with a signi cant (p < 0.05) attenuation of virulence in G. mellonella (Fig. 5a, b).The treatment with ∆MrGpa1 also resulted in a increased survival rate of the larvae compared with the WT and cp∆MrGpa1 strains treatment.By contrast, in the injection bioassays, we did not observe any differences in LT 50 values between larvae infected with ∆MrGpa1 (3.69 ± 0.15 days), and WT (3.53 ± 0.14 days) or cp∆MrGpa1 strains (3.57± 0.12 days) (Fig. 5c, d).
We then determined the expression of insect virulence-related genes during cuticle penetration by RT-qPCR.Indeed, the expression of several genes involved in the adhesion (mad1, 45% expression in ∆MrGpa1 strain compared with the WT strain), appressorium formation (mpl1, 72%, and gpa, 52%) and cuticle penetration (pr1A, 93%, and pr1C, 96%) was signi cantly decreased in the ∆MrGpa1 strain compared with their expression in the WT strains (Fig. 6a).
To determine the mechanism of the virulence defect of ∆MrGpa1 strain, we then assayed appressorium formation on a hydrophobic surface.We observed that the loss of MrGpa1 impaired appressorium differentiation, compared with the control strains.Speci cally, 24 h after induction, ∆MrGpa1 strain did not form appressoria, while, the appressorium formation rate of the WT strain was 80% (Fig. 6b).Further, 48 h after induction, the appressorium formation rate of ∆MrGpa1 was only approximately 20% and were signi cantly reduced (by 76.5%) compared with WT strains (Fig. 6c).Therefore, MrGpa1 plays an important role in cuticle penetration by impacting appressorium formation.

Discussion
G-proteins are key components of various signal transduction pathways and play show important biological roles function in the control of cell proliferation, behavior, and development in higher organisms (Harashima and Heitman 2005;Ivey et al. 1996;Regenfelder et al. 1997).In the current study, we described the identi cation of MrGpa1 gene encoding the Ga i subunit in M. robertsii, and showed that this Ga i protein plays an important role in conidiation, stress resistance, and virulence in the host fungus.
In lamentous fungi, genes encoding three G-protein a subunits were reported in N. crassa, F. oxysporum f. sp.cubense, Aspergillus nidulans and M. grisea (Guo et al. 2016b;Hicks et al. 1997;Ivey et al. 1996;Liu and Dean 1997).However, while genes for four G protein a subunits were identi ed in M. robertsii, two of the open reading frames (ORF) are highly similar (MrGpa2 and MrGpa4).Further, the existence of Ga i has been demonstrated in lamentous fungi including the M. robertsii in this study, but not in the yeasts S. cerevisiae or Schizosaccharomyces pombe (Kallal and Fishel 2000;Kubler et al. 1997).
Previously, to be activated by GPCR that sense external signals in plasma membrane (PM), G protein was considered to locate at the PM (Marrari et al. 2007).However, recent studies revealed G-protein localization beyond the plasma membrane, e.g., in the mitochondrion, endoplasmic reticulum, and Golgi apparatus (Michaelson et al. 2002;Nair et al. 2017).Wolf PSORT prediction in the current study indicated a possible location of MrGPA1 in the mitochondrion.Indeed, by using a uorescent protein fusion, we showed here that MrGPA1 is located in the mitochondria, which is consistent with location of Gα i in HEK293T cells (Lyssand and Bajjalieh 2007).
According to multiple studies, G-protein a subunits play a role in the conidial yield in some fungi.For example, the conidiation of magB deletion mutant of M. grisea and ∆fga1 mutant of F. oxysporum f. sp.cubense is impaired (Guo et al. 2016b;Liu and Dean 1997).A similar phenomenon was also observed in the current study.Further, some key genes involved in conidiation were down-regulated in the MrGpa1 deletion mutant, suggesting that the MrGpa1 gene is involved in the conidiation of M. robertsii by regulating the expression of conidiation-related genes.
We also observed that ∆MrGpa1 strain is more tolerant to UV irradiation and thermal stress than the WT and cp∆MrGpa1 strains.This observation was consistent with ndings for F. oxysporum f. sp.cubense and N.crassa (Guo et al. 2016b;Yang and Borkovich 1999).According to the two cited studies, intracellular cAMP levels are reduced in the Ga i deletion mutants, suggesting that the cAMP pathway might be involved in the response to thermal stress and tolerance of UV irradiation in some fungi.
The virulence of plant pathogenic fungi is regulated by multiple pathways, such as the mitogen activated protein kinase (MAPK) cascades and the cAMP-PKA pathway (Dean 1997).Further, the Ga i family contributes to pathogenicity in many plant pathogenic fungi.For example, the pathogenicity of the deletion strains M. grisea ∆magB, Botryils cinerea ∆bcg1, and F. oxysporum f. sp.cubense ∆fga1 is markedly reduced (Gronover et al. 2001;Guo et al. 2016b;Liu and Dean 1997).In the entomopathogenic fungus M. robertsii, MrGpa1 deletion resulted in a marked attenuation of virulence in the G. mellonella model, with an increased LT 50 value (by 63.9%) compared with that of the WT strain.Further analysis indicated that the reduced virulence is associated with an impaired appressorium differentiation in the mutant.This was similar to the effect of magB deletion in M. grisea, namely, blocked appressorium formation in a deletion strain (Liu and Dean 1997).To the best of our knowledge, the current study constitutes the rst report of virulence decrease resulting from a deletion of a G protein a subunit gene in an entomopathogenic fungus.
In conclusion, MrGPA1 is located in the mitochondria of M. robertsii cell and is a member of Ga i family.
MrGPA1 controls unique signal transduction pathways, and thus plays important role in conidiation, stress resistance, and virulence in that fungus.These ndings raise the possibility of designing powerful strategies for genetic improvement of M. robertsii conidiation capacity and virulence for killing pests.

Declarations
Ethics approval and consent to participate article does not contain any studies with human participants or animals performed by any of the authors.

Figures Figure 1
Figures