Metarhizium robertsii Produces an Extracellular Invertase (MrINV) That Plays a Pivotal Role in Rhizospheric Interactions and Root Colonization

As well as killing pest insects, the rhizosphere competent insect-pathogenic fungus Metarhizium robertsii also boosts plant growth by providing nitrogenous nutrients and increasing resistance to plant pathogens. Plant roots secrete abundant nutrients but little is known about their utilization by Metarhizium spp. and the mechanistic basis of Metarhizium-plant associations. We report here that M. robertsii produces an extracellular invertase (MrInv) on plant roots. Deletion of MrInv (⊿MrInv) reduced M. robertsii growth on sucrose and rhizospheric exudates but increased colonization of Panicum virgatum and Arabidopsis thaliana roots. This could be accounted for by a reduction in carbon catabolite repression in ⊿MrInv increasing production of plant cell wall-degrading depolymerases. A non-rhizosphere competent scarab beetle specialist Metarhizium majus lacks invertase which suggests that rhizospheric competence may be related to the sugar metabolism of different Metarhizium species.


Introduction
The entomopathogenic fungus Metarhizium robertsii (formerly known as Metarhizium anisopliae var anisopliae [1]) is currently applied as a biological control agent against various insect pests [2][3][4].Recent investigations have revealed that species of Metarhizium which are widely distributed in soil also play other ecological roles, including establishing mutualistic interactions with plants as rhizospheric associates [5].Seed treatment with these Metarhizium spp.increases crop yields by killing soil insects [6], inhibiting plant pathogens [7], enhancing the uptake of micronutrients [8], and translocating nitrogen from killed insects to plants [9].However, Metarhizium has yet to be exploited in agriculture as a plant growth enhancer and the mechanistic basis for its many interactions are poorly understood.The mycoparasite Trichoderma is the best studied plant root colonizer and the most widely applied fungal plant growth promoter [10,11].Much effort and expense has been directed at finding better ways to apply both Metarhizium spp.(as insecticides) and Trichoderma spp.(as plant symbionts).However, as with Metarhizium, exploitation of Trichoderma has been handicapped by inconsistent field performance [12,13].
More consistent biopesticide products might be achieved based on combinations of microbes that economically achieve breadth of action.To this end, identifying commonalities and differences between Metarhizium and Trichoderma in their mode of action could be used to facilitate synergistic effects when applied together as a seed treatment.Comparative genomics reveals Metarhizium and Trichoderma are related [14], but studies to date have shown mechanistic differences in root colonization that suggest they have independently evolved rhizosphere competence.For instance, Trichoderma uses a hydrophobin to adhere to roots [15], whereas Metarhizium uses an adhesin [16].
Nevertheless, given their shared habitat there should also be similarities and some of these could potentially lead to competition for resources.In soils, both Metarhizium and Trichoderma exploit various products of plant photosynthesis.The ability to respond to and utilize these products will likely determine their success as plant root colonizers.Sucrose is one of the main carbohydrate products of photosynthesis and has been detected in high concentrations near root tips [17][18][19].Several fungi such as Thermomyces lanuginosus, Uromyces fabae, Aspergillus niger and Trichoderma virens produce invertases capable of hydrolyzing sucrose to monosaccharides [20][21][22][23].In U. fabae, invertase is highly expressed in haustoria during the process of infection in the leaf.Furthermore, plantderived sucrose is a key component in symbiotic associations between T. virens and maize [22].
In the present study, a gene (MrInv) with homology to the intracellular invertase of T. virens was identified and characterized in M. robertsii.MrInv encodes a single copy extracellular invertase that is principally responsible for sucrose hydrolysis by M. robertsii.Disrupting the gene resulted in poor fungal growth on sucrose, root exudates and rhizospheric soils.However, root colonization by the mutant was increased.We present evidence that this resulted from increased expression of cell wall-degrading enzymes because of reduced carbon catabolite repression in the mutant.Our data suggest that while MrInv plays an important role in providing the fungus with a carbon source, this limits the extent of root colonization during Metarhizium-plant associations.
Escherichia coli DH5α and Agrobacterium tumefaciens AGL-1 were used for DNA cloning and fungal transformation.Arabidopsis thaliana eco-type Col-0 seeds were purchased from LEHLE SEEDS (Round Rock, Texas, USA).Panicum virgatum (switchgrass) seeds were obtained from OSC seeds (Waterloo, Ontario, Canada). A. thaliana and switchgrass seeds were surface-disinfected according to Sauer and Burroughs [24] and Miché and Balandreau [25], respectively.The sterilized seeds were kept at 4 °C overnight to allow for synchronization of growth before fungal inoculation.

Gene Disruption
A MrInv disruption vector pPK2BargfpDMrInv was constructed to knock out the MrInv gene in M. robertsii.The 5′end and 3′-end of MrInv, cloned by PCR, were inserted into a modified master Ti vector pFBarGFP using the Xba I and Bgl II/ EcoR V sites, respectively.The disruption mutant (⊿MrInv) was obtained by A. tumefaciens-mediated transformation [26].To complement ⊿MrInv, a ~ 3.9 kb genomic fragment containing the MrInv open reading frame and its flanking sequence was cloned and inserted into the EcoR V site of pPK2SurGFP with the Sur selective marker under control of a constitutive glyceraldehyde-3-phosphate dehydrogenase (Pgpd) promoter [27,28].The ⊿MrInv revertant (⊿MrInv-rv) was obtained by transforming the construct into ⊿MrInv.

qPCR Analysis
To quantify gene expression in response to sugars, wild-type M. robertsii and M. robertsii ⊿MrInv were grown in Sabouraud Dextrose Broth (SDB, Difco) for 30 h, and mycelia were transferred to minimal medium (M100 medium minus glucose) plus 1% sucrose (sucrose cultures).To observe the time course of MrInv expression, RNA was harvested from mycelia for qPCR analysis as described previously [29].The primers used in this study are listed in Table S1.

Sugar Utilization Assay
The ability of ⊿MrInv and wild-type strains to utilize different sugars was analyzed according to a modification of the method of Fang and St. Leger [30].Sterile distilled water plus or minus a nitrogen source (0.1% NaNO 3 ) was combined with a carbohydrate (0.1%).The soybean root exudate was prepared as described by Pava-Ripoll et al. [31].To determine growth rates, mycelial inoculums (0.2 g wet mycelia, approximately 25 mg dry biomass) from SDB cultures were transferred into liquid minimal medium supplemented with 1% sucrose and/or 1% glucose.Mycelia harvested after 12 h growth was used for dry weight determinations.

Rhizosphere Competence and Root Colonization Assays
Sterile synchronized P. virgatum (switchgrass) seeds were inoculated by immersion for 1 h in 1×10 8 /mL -1 M. robertsii wildtype or ⊿MrInv spores as described by Wyrebek et al. [32].The seeds were planted in pots (0.5 g seeds per pot) filled with sterile soil (Scotts Turf Builder Seeding Soil, Scotts Company, USA).Each treatment was replicated three times.To determine fungal survival in bulk soil with or without switchgrass seeds, spore suspensions were spread evenly through soil producing approximately 5 × 10 3 spores/g -1 soil.Half the pots were planted with uninoculated seeds.All the pots were kept in a growth chamber at 25 °C with 14:10 h light:dark cycle, and within two weeks the grass had produced a lawn covering the soil surface.Sterile water was added regularly to avoid drying.The soil population of wild-type and ⊿MrInv strains was monitored at set intervals by a slight modification of Fang and St. Leger [30].Briefly, 0.5 g of soil from each pot was collected using a cork borer.This soil contained a high density of roots so the fungus would be existing in overlapping rhizospheres.Soil suspensions were prepared by adding 5 mL 0.05% Tween 80 solution and vortexing vigorously.Aliquots (100 μL) were spread on Rose-bengal selective agar plates and CFUs were determined after 10 days at 27 °C [5].
To assay root colonization, 10-d-old A. thaliana seedlings were inoculated in liquid Murashige and Skoog medium (Sigma, USA) containing 5 × 10 6 spores mL -1 M. robertsii suspensions.After 48 h incubation, A. thaliana roots were collected, and washed three times in 0.05% Tween 80 solution.To observe fungal colonization, individual A. thaliana root were stained with fungi lactophenol cotton blue (ENG Scientific, Inc., USA) and mounted on a slide for microscopy.M. robertsii root colonization was quantified by a modification of Viterbo et al. [33].To quantify wild-type and mutant growth, three replicates each containing nine A. thaliana roots were weighed and homogenized by vortexing vigorously for 2 min in 500 μL 0.05% Tween 80 solution.Serial dilutions were assayed for CFU on Rose-bengal selective agar plates [5].The same procedure was applied on switchgrass except that each replicate contained three switchgrass roots from each pot.
Roots were weighed and ground with mortar and pestle in 2 mL 0.05% Tween 80 solution.The CFUs were quantified as described above.

Enzymatic Assays
Enzymatic activities were assayed in filtrates from cultures grown with sucrose or A. thaliana seedlings as described previously.The Pr1 subtilisin protease activity was assayed with N-succinyl-Ala-Ala-Pro-Phe-p-nitroanilide [34].One unit of protease activity was defined as the amount of enzyme that produces 1 μM of para-nitro aniline per minute.Invertase and endochitinase activities were assayed using the EnzyChrom TM Invertase Assay Kit (BioAssay Systems, USA) and the Chitinase Assay Kit, Fluorimetric (Sigma, USA).Total protein content was measured with a Protein Assay Kit II (Bio-Rad, USA).Pectinase activity versus polygalacturonic acid was assayed according to the protocol from Worthington Biochemical Corporation (New Jersey, USA).One unit of pectinase was defined as the amount of enzyme that liberates 1 µmole of D-galacturonic acid from polygalacturonic acid per minute.

Plant Growth
To assess the effect of M. robertsii on switchgrass, stem lengths were measured and leaf chlorophyll content was read using a SPAD-502 Plus chlorophyll meter (Konica Minolta Sensing, Inc. Japan).The plants were harvested after three months, and rhizospheric soils washed off the roots, and root lengths measured as described previously [5].The dry weight of whole plants was determined as a measure of biomass.

Identification and Characterization of an Invertase MrINV in M. robertsii
Using an invertase TvINV from T. virens (accession no.EHK21605) as query, we identified a single homolog (MrInv) in the M. robertsii genome [14] with a maximum identity of 41% (5e -7 ).The Open Reading Frame (ORF) of MrInv is 1782-bp long coding for a putative 573 amino acid protein with a predicted molecular weight of 63.9 kDa, and is interrupted by one 60-bp intron.As deduced by SignalIP 4.1 [35], MrINV contains a predicted 24-aa signal peptide for secretion.The predicted cleavage site is between G24 and H25 and the calculated molecular weight of the mature protein is 61.3 kDa.The signal peptide resembled those on extracellular invertases from bacteria and some fungi including yeasts.Phylogenetic reconstruction grouped fungal invertases into two seperate clades (Figure 1).MrINV grouped with fungal invertases containing a NDPN box (conserved β-fructosidase motif) [36].However, no homologous sequence was found in the scarab beetle specialist M. majus.
To better understand the role of MrInv during M. robertsii development, regulation of MrInv was determined under different cultural conditions using quantitative reverse transcription PCR (qPCR).Total RNA was isolated from mycelia grown on different carbohydrates including monosaccharides, homologous disaccharides and heterologous dioligosaccharides (Table 1).MrInv expression was up-regulated 40-fold in the presence of sucrose (Figure 2A), but other carbohydrates had no effect on expression (data not shown).A time-course for MrInv expression in sucrose culture showed onset at 20 minutes, and a peak at 4 h which then plateaud (Figure 2B).

MrInv Disruption Suppressed M. robertsii Growth on Sucrose
To confirm that MrINV is a functional invertase involved in sucrose metabolism in M. robertsii, we produced a null mutant (⊿MrInv), deletion of the wild-type allele.To complement ⊿MrInv, the genomic fragment of MrInv was transformed into ⊿MrInv resulting in ⊿MrInv-rv.The deletion of MrInv in ⊿MrInv was confirmed by PCR using the genomic DNA as template (Figure S1).RT-PCR confirmed there was no MrInv expression in ⊿MrInv, and as expected, ⊿MrInv showed no invertase activity during 8 h growth on sucrose.Invertase activity was detected in the wild-type strain and ⊿MrInv revertant (⊿MrInvrv) (Figure 2C), confirming that MrInv encodes the invertase up-regulated in the presence of sucrose.
To determine the role of MrInv, the wild-type and ⊿MrInv strains were compared in colony morphology, germination rate and growth.Conidial germination of ⊿MrInv (8.0% ± 1.2%) after 24 h in sucrose medium was reduced by 61% compared to the wild-type strain (20.7% ± 1.2%) (Table 1, P < 0.01).Likewise, colony growth (Figure S2) and biomass in sucrose medium were significantly reduced in ⊿MrInv compared to the wild-type strain (Figure 3, P < 0.01).The addition of an inorganic nitrogen source (NaNO 3 ) did not affect the ability of MrInv to utilize sucrose, and ⊿MrInv and the wild-type strain germinated at similar rates in other carbohydrates, including glucose (Table 1).In addition,⊿MrInv grew at a similar rate as the wild-type strain using glucose as sole carbon source (Figure S3).Thus, the only impairment of⊿MrInv is in its ability to utilize sucrose which confirms MrINV is a functional invertase.

MrInv Disruption Impaired M. robertsii Rhizosphere Competency but Enhanced Root Colonization
To analyze the involvement of MrInv in the rhizosphere competency of M. robertsii, we investigated fungal survivorship in the soil and measured root colonization.Firstly, to mimic rhizospheric nutrient sources we used soybean root exudate, which contains plant-derived sucrose as a major component [31].No difference in the germination rate was observed between ⊿MrInv and wild-type strains at high concentration of root exudate (1 mg mL -1 ).But at low concentrations (0.1 mg mL -1 and 0.01 mg mL -1 ), germination of ⊿MrInv was reduced by16% and 56%, respectively (Table 1, P< 0.01).
The poor growth of ⊿MrInv in root exudate implied that ⊿MrInv may be impaired in rhizosphere competency.To investigate rhizospheric interactions and fungal root colonization, spores of ⊿MrInv and wild-type strains were inoculated into soil microcosms containing switchgrass.Switchgrass is easy to culture in lab conditions and is a well characterized host for M. robertsii [32].The rhizosphere competency of ⊿MrInv and wild-type strains was determined by counting colony-forming units (CFUs) in soil samples [30].Initial densities were determined by counting CFUs immediately after adding fungal spores.Two weeks postinoculation the number of ⊿MrInv and wild-type CFUs had dropped by half.Two months post-inoculation, ⊿MrInv levels were still 37% less than the initial density, but the number of wild-type CFUs had increased 2.4-fold.At three months,⊿MrInv and wild-type rhizospheric populations were 2.5-fold and 8-fold higher, respctively.Thus, compared to the wild-type, the number of ⊿MrInv CFUs were 72%, 75% and 71% less in the first, second and third month post-inoculation, respectively (Figure 5, P < 0.01).CFU counts of ⊿MrInv and wild-type strains in bulk soil (soil containing fungi but no seeds) declined at the same rate over 3 months.The significant reduction of ⊿MrInv in the rhizosphere microcosms relative to the wild-type suggests MrInv is important for rhizosphere competency through utilization of plant-derived sucrose as carbon source.
As well as colonizing the rhizosphere, M. robertsii also grows on root surfaces [30].Since deletion of MrInv reduced rhizosphere competency, we also determined if MrInv facilitates colonization of roots by counting CFUs extracted from switchgrass roots.Deleting MrInv produced a 2.7-fold increase in root colonization relative to the wild-type strain (Figure 6A, P < 0.05).To determine if the increased root colonization by ⊿MrInv on switchgrass is applicable to other plants, A. thaliana roots were inoculated with ⊿MrInv or wildtype strain in a hydroponic system.Root colonization was determined after 48 h incubation by counting CFUs.Colonization of A. thaliana roots by ⊿MrInv was increased 2.1fold relative to the wild-type strain (Figure 6B, P < 0.01).Microscopic observation confirmed that ⊿MrInv hyphae proliferated more than wild-type hyphae on A. thaliana roots (Figure 6C).
Production of many hydrolytic enzymes in fungi is regulated by carbon catabolite repression (CCR) [41].We hypothesized that hydrolytic enzymes are up-regulated in ⊿MrInv due to less glucose uptake and thereby reduced CCR.To test this we measured the expression of CCR-related genes.Consistent with greater CCR, sucrose induced higher levels of the CCRrelated genes tps1 (8.3-fold) and hxk1 (2.6-fold) in the wild-type than in ⊿MrInv (Figure 8, P < 0.01).The current study suggests that hydrolytic enzymes from M. robertsii are involved in the root colonization process and the hydrolytic enzyme activities in MrInv deficiency strain are up-regulated possibly through the release of CCR.

Plant growth
The growth of switchgrass inoculated with ⊿MrInv and wildtype strains was monitored for 3 months at one month intervals to determine whether MrInv expression had an impact on plant growth.No differences were found between plants inoculated with ⊿MrInv and wild-type strain in stem length, leaf chlorophyll content, root length and plant biomass (Figure S4).In addition, the growth of switchgrass uninoculated with either fungi was similar with the plants inoculated with M. robertsii.

Discussion
Enormous numbers of microbes inhabit the rhizosphere using various products of plant photosynthesis, and in return some microbes boost plant growth and health through several mechanisms including phytostimulation, biofertilisation, bioremediation and biological control [42].The rhizosphere competence of M. robertsii was discovered comparatively recently in a field plot of cabbages [5].M. robertsii is the only Metarhizium species found to associate with grass roots in the field [32], but in this study, we found that both switchgrass and the dicotyledonous A. thaliana supported extensive root colonization by M. robertsii.
Compared to interactions between fungal pathogens and their hosts, little is known about the molecular mechanisms underlying the complex interactions between roots, root exudate and any rhizospheric fungus, including the best studied examples in the genus Trichoderma.It is particularly important to comprehensively understand the natural ecological role of Metarhizium and Trichoderma in order to optimize their use as biological control agents.Table 2 summarizes currently understood similarities and differences between Metarhizium In vitro growth assays have shown that M. robertsii grows well in root exudates from soybeans and switchgrass [30,32].In the current study, we show that M. robertsii uses MrINV to specifically hydrolyze sucrose to monosaccharides, and while M. robertsii utilizes a wide variety of carbohydrates the highest germination rates are on monosaccharides.M. robertsii can also take up sucrose using its unique oligosaccharide transporter (MRT) [30,32].Disruption of Mrt resulted in reduced rhizosphere competency confirming a role for root-derived oligosaccharides in the symbiotic relationship between M. robertsii and plants [30].However, germination and growth utilizing sucrose are reduced in the absence of the extracellular sucrose-induced MrINV.Given the complexity of carbon sources in root exudate, diverse strategies for utilizing sugars seems well attuned to M. robertsii's opportunistic lifestyle and wide range of plant associations.
A phylogenetic reconstruction classifies MrINV as a fungal secreted acidic invertase, and it clusters separately from nonsecreted enzymes including the query sequence TvINV from T. virens.Blastp and genomic blast searches confirm that MrInv is a single copy gene in M. robertsii.The specialist beetle pathogen M. majus lacks a homolog of MrINV and does not grow on sucrose as sole carbon source.M. majus has not to date been identified as a root colonizer [32], suggesting that rhizospheric competence may be related to the sugar metabolism of different Metarhizium species.
Metarhizium is attracted to and associates intimately with root surfaces in the soil [32,43].However, few studies have focused on the characteristics of Metarhizium interactions with roots as they appear under the microscope.In this study, the microscopic observation of the colonization of A. thaliana root by M. robertsii suggests the root invasion occurs by penetration of the epidermis and further ingress into the outer cortex, which share a similar pattern of colonization with Trichoderma [15].However, unlike Trichoderma, we observed no specific adhesive structures during root colonization by M. robertsii.Sasan and Bidochka (2012) have reported that M. robertsii promotes plant root growth [43], and we have observed the same in field conditions when plants are growing under suboptimum feeding and watering regimes (Unpubl.Data).The lack of impact on plant growth in the current study is probably because the plants were growing in a nutrient rich soil with adequate water.
The most significant finding of the current study is that disruption of MrInv reduced M. robertsii's survival in the overlapping rhizospheres surrounding switchgrass roots, but significantly enhanced root colonization.Trichoderma lacking the intracellular TvInv also shows increased colonization of maize roots [22].There is no information about TvInv's effects on competency in soil but Trichoderma lacking TvInv showed increased expression of hydrolytic enzymes that weaken epidermal cell walls [44].This could result from carbon catabolite derepression because of a reduction in glucose and related sugars in the mutant [45,46].The tps1 (encoding trehalose-6-phosphate synthase) and hxk1 (encoding hexose kinase) are important mediators of CCR in fungi.They are induced by glucose and mediate glycolysis and carbon catabolite repression [47].In our study, sucrose induced greater expression of tps1 and hxk1 in the wild-type than in ⊿MrInv consistent with derepression in the mutant.M. robertsii invades insects by direct penetration of the cuticle facilitated by the production of a battery of extracellular enzymes, including proteinases (Pr1), chitinases and esterases [38].The higher levels of Pr1 and endochitinase produced by ⊿MrInv on Arabidopsis root probably result from carbon catabolite derepression in the mutant.In the field these enzymes could be involved in scavenging nutrients from fungi and insects in the soil since they do not target the plant cell wall.⊿MrInv's upregulation of pectinase, which hydrolyzes a major polysaccharide substrate in plant cell walls [37], provides a direct connection between Metarhizium-produced hydrolytic enzymes and plant cell wall-degradation.Furthermore, increased colonization of monocotyledon (switchgrass) and dicotyledon (A.thaliana) roots suggests that the effect is irrespective of the plant species.In spite of increased production of cell wall-degrading enzymes ⊿MrInv's competency in the soil is sharply reduced suggesting that at a small distance from the roots sucrose is a more important source of nutrients than polymers.

Of insects
Of fungi (some strains)

Figure 1 .
Figure 1.Neighbour-joining (NJ) tree of invertase from M. robertsii (MrINV) and other 23 invertases from bacteria, plants, yeasts and other fungi.Amino acid sequences were aligned using the ClustalX algorithm and the NJ tree was constructed with the MEGA4 software.Bootstrap values are adjacent to each internal node, representing the percertage of 1,000 bootsrap replicates.doi: 10.1371/journal.pone.0078118.g001

Figure 2 .
Figure 2. Characteristics of MrInv in M. robertsii.A, Expression assay with qPCR after mycelia transferred from SDB to minimal medium plus 1% sucrose, grown at 27 °C for 8 h.B, Relative transcript levels of MrInv gene versus the housekeeping gpd gene.gpd, glyceraldehyde-3-phosphate dehydrogenase gene.Values are means of three replicates with corresponding standard deviation.C, The invertase activity of M. robertsii 2575 wild-type strain, ⊿MrInv and ⊿MrInv revertant.Mycelia were transferred from SDB to minimal medium plus 1% sucrose, grown at 27 °C for 8 h.The filtrates from cultures were collected for enzymatic activity assay.Means are calculated from three replicates and bars represent standard error.doi: 10.1371/journal.pone.0078118.g002

Figure 3 .
Figure 3. Biomass production of M. robertsii 2575 wild-type strain and ⊿MrInv in the presence of glucose and/or sucrose.Mycelia were transferred from SDB to minimal liquid medium supplemented with 1% glucose (Glu) and (Glu+Suc)/or 1% sucrose (Suc), grown at 27 °C for 12 h.Means were calculated from 3 replicates with corresponding standard errors.Experiments were repeated twice.** indicates means statistically different at the 0.01 level.doi: 10.1371/journal.pone.0078118.g003

Figure 6 .
Figure 6.Root colonization of switchgrass (Pv) and A. thaliana (At) by M. robertsii wild-type strain (wt) and MrInv disruption mutant (⊿MrInv).A, 3-month-old switchgrass roots from the soils containing M. robertsii strains were collected for root colonization assay.B, A. thaliana seedlings were inoculated with M. robertsii spores from each strain in the hydroponic system.After 48 h incubation, A. thaliana roots were collected for root colonization assay.Means are calculated from 9 (Pv) or 27 (At) replicates of each treatment and bars represent the standard error.** and * indicate means statistically different at the 0.01 and 0.05 level.C, Photographs of A. thaliana root colonization by M. robertsii wild-type strain (top) and ⊿MrInv (bottom) after 48 h incubation.C, conidium; H, hypha.Scale bar = 5 μm.doi: 10.1371/journal.pone.0078118.g006

Figure 7 .
Figure 7.The activity assay of hydrolytic enzymes produced by M. robertsii 2575 wild-type strain and ⊿MrInv.The filtrates from cultures of each strain grown with A. thaliana seedlings for 48 h were collected, and the enzymatic activity of pectinase (A), subtilisin Pr1 (B) and endochitinase (C) were measured.Means are calculated from three replicates and bars represent standard deviation.Similar results were obtained in two independent experiments.** and * indicate means statistically different at the 0.01 and 0.05 level.doi: 10.1371/journal.pone.0078118.g007

Figure 8 .
Figure 8. Expression of CCR-related genes in M. robertsii 2575 wild-type strain and ΔMrInv in the presence of sucrose.A, Semi-quantitative RT-PCR analysis of CCR-related genes.Mycelia were transferred from SDB to minimal medium plus 1% sucrose, grown at 27 °C for 30 min.Total RNA was extracted from fungal biomass and qPCR was performed.B, Relative transcript levels of CCR-related genes versus the housekeeping gpd gene.tps1, trehalose-6-phosphate synthase gene; hxk1, hexose kinase gene; gpd, glyceraldehyde-3-phosphate dehydrogenase gene.Values are means of three replicates with corresponding standard deviation.** indicates means statistically different at the 0.01 level.doi: 10.1371/journal.pone.0078118.g008 photosynthesis, particularly sugars .1371/journal.pone.0078118.t002

Table 1 .
In vitro growth in different carbohydrates and root exudates.

Table 2 .
Commonalities and differences betweenMetarhizium and Trichoderma in root colonization.