Sex Pheromone Receptor Ste2 Orchestrates Chemotropic Growth towards Pine Root Extracts in the Pitch Canker Pathogen Fusarium circinatum

In ascomycetous fungi, sexual mate recognition requires interaction of the Ste2 receptor protein produced by one partner with the α-factor peptide pheromone produced by the other partner. In some fungi, Ste2 is further needed for chemotropism towards plant roots to allow for subsequent infection and colonization. Here, we investigated whether this is also true for the pine pitch canker fungus, Fusarium circinatum, which is a devastating pathogen of pine globally. Ste2 knockout mutants were generated for two opposite mating-type isolates, after which all strains were subjected to chemotropism assays involving exudates from pine seedling roots and synthetic α-factor pheromone, as well as a range of other compounds for comparison. Our data show that Ste2 is not required for chemotropism towards any of these other compounds, but, in both wild-type strains, Ste2 deletion resulted in the loss of chemotropism towards pine root exudate. Also, irrespective of mating type, both wild-type strains displayed positive chemotropism towards α-factor pheromone, which was substantially reduced in the deletion mutants and not the complementation mutants. Taken together, these findings suggest that Ste2 likely has a key role during the infection of pine roots in production nurseries. Our study also provides a strong foundation for exploring the role of self-produced and mate-produced α-factor pheromone in the growth and overall biology of the pitch canker pathogen.


Introduction
Sensing and responding to environmental cues are vital for all living organisms.This is particularly true for soil-dwelling microorganisms that use complex, coordinated processes to communicate with their surroundings [1][2][3].These organisms can sense a variety of chemical signals, and many can then respond by manipulating the surroundings for their benefit [4,5].For example, filamentous fungi can sense chemical stimuli and direct their growth toward or away from such compounds [6].The directed growth in response to chemical gradients is broadly referred to as chemotropism, which is a well-known process in many fungi, as it promotes access to nutrients and contact with mating partners [7,8].
Research conducted during the last decade has demonstrated that chemotropism plays a vital role in the pathogenesis of certain fungal pathogens of plants [9,10].This is because, in addition to mediating the directed growth of soil-inhabiting pathogens towards or away from certain chemicals [11,12], chemotropism also seems to afford fungi the ability to grow towards plant roots [4,13,14].Previous studies have shown that chemotropism, involving chemical signals in root exudates from tomato and wheat, is involved in the colonization of these plants by Fusarium oxysporum and Fusarium graminearum, respectively [15][16][17].
All fungi are maintained in the culture collection of the Forestry and Agricultural Biotechnology Institute (FABI), University of Pretoria, South Africa.For routine culturing, the fungi were grown for 7 days at 25 • C on potato dextrose agar (PDA; 20 g/L BD Difco TM Potato Dextrose Agar; Becton, Dickinson and Company, Franklin Lakes, NJ, USA).
DNA was extracted from the fungi using two methods.The first utilized PrepMan Ultra Sample Preparation Reagent (Thermo Fisher Scientific, Waltham, MA, USA) to obtain crude DNA extracts for subsequent PCR-based screening of the putative mutants.For the preparation of high-quality genomic DNA, mycelium was taken from the actively growing margins of 7-day-old cultures and used to inoculate 100 mL of potato dextrose broth (PDB; Becton).Following incubation for seven days on an orbital shaker (Orbishake 262, Labotec, Midrand, South Africa) at 200 revolutions per minute (rpm) and room temperature, mycelia were collected by filtration through 2-3 layers of sterile Miracloth (Merck KGaA, Darmstadt, Germany).These fungal tissues were then freeze-dried and subjected to DNA extraction using CTAB (N-cetyl-N, N, N-trimethyl-ammonium bromide) as described previously [35].Following spectrophotometric quantification (NanoDrop ND-2000, Thermo Fisher Scientific), the extracted CTAB-extracted DNA was stored at −20 • C until use.

Identification of ste2 Gene
The ste2 gene sequence in the genome of F. circinatum FSP34 was identified using the known ste2 gene of F. oxysporum, FOXG_10633 [15].For this purpose, the FOXG_10633 sequence was retrieved from the database of the National Centre for Biotechnology Information (NCBI; https://www.ncbi.nlm.nih.gov) using accession number XM_018390091, while the FSP34 genome was retrieved using accession number AYJV00000000.The FOXG_10633 sequence was then used as query in a BLASTn search of the FSP34 genome sequence using CLC Main Workbench version 8.1.2(Qiagen Bioinformatics, Aarhus, Denmark).We then confirmed that the ste2 gene identified in F. circinatum contained the expected transmembrane motifs using the SOSUI [36].We also compared it with the known ste2 gene from F. graminearum (FGSG 02655; accession number XM_011320304; [37,38]).

Gene Disruption and Complementation
For an efficient disruption of ste2 from the two wild-type strains, the split-marker homologous recombination approach was employed [39,40] using the methodology developed by Phasha and colleagues [35].The selectable markers used were obtained from plasmids SK2241 [41] and pGEN-Not-1 [42].The relevant PCRs (see below) utilized the CTAB-extracted DNA, while the primers used for generating the respective constructs were designed using CLC Main Workbench, (see Supplementary Tables S1 and S2 for primer sequences and PCR annealing temperatures).
Split-marker constructs were generated by obtaining fragments of the regions flanking ste2 in both wild-type strains, as well as overlapping fragments of the region flanking the GFP gene (ZsGreen) and the hygromycin B (hyg) resistance gene cassette in plasmid SK2241 (see Supplementary Figure S1).We used Phusion high-fidelity PCR Master Mix (Thermo Fisher Scientific) in reactions with primer sets RP8+RP7A and RP6A+RP5, respectively, to amplify the 5 ′ and 3 ′ flanking fragments of the ste2 gene in both wild-type strains.The Phusion high-fidelity PCR Master Mix was also used to amplify the two overlapping ZsGreen/hyg fragments from plasmid SK2241 using primer sets ZS-HYG-F1+ZS-HYG-HYR and ZS-HYG-YGF+ZS-HYG-R1.By making use of LongAmp Taq 2X Master Mix (New England Biolabs, Ipswich, MA, USA) and the manufacturer's guidelines, we then "fused" the 5' flanking fragment of ste2 with the first ZsGreen/hyg fragment using primer set RP8+ZS-HYG-HYR.Similarly, the 3' flanking fragment of ste2 was fused with the second ZsGreen/hyg fragment using primer set ZS-HYG-YGF+RP5.
The deletion constructs generated were then used to transform protoplasts of the respective wild-type strains as described previously [35,43].This involved incubation of the respective constructs with protoplasts of the wild-type strains, and recovery of mutants based on their capacity to grow in the presence of the aminoglycoside antibiotic hygromycin B (Sigma-Aldrich, Mannheim, Germany).Putative transformants were then transferred to PDA containing 150 µg/mL hygromycin B and grown as described above.
To complement the mutant strains, the ste2 gene was randomly integrated into the genomes of the deletion mutants using the geneticin resistance gene nptII from plasmid pGEN-Not-1.This was performed with the Phusion high-fidelity PCR Master Mix using primer set RP8+RP5A to amplify ste2 from each wild-type strain, and using primer set TY22+TY23 to amplify the nptII gene.As before, the LongAmp Taq 2X Master Mix was used to generate the fusion construct (ste2-nptII) using primer set RP8+TY23.The same procedure was used as described above for transformation with the complementation construct, except that the protoplasts obtained from the deletion mutants were used, and resistance to Geneticin G-418 (120 µg/mL, Thermo Fisher Scientific) was used for the recovery of transformants.
Initial screening for mutant transformants was conducted using the PrepMan-extracted DNAs, PCRs with Super-Therm Taq DNA Polymerase (Thermo Fisher Scientific) and the reaction conditions described previously [44].Primer set Pre2GeneF+Pre2GeneR was used to verify the absence of the ste2 gene, while primer set ZS-HYG-YGF+ZS-HYG-HYR was used to verify the presence of ZsGreen/hyg.Pure cultures for all the confirmed mutants were then prepared by overnight incubation of spore suspensions on water agar medium and then using single germinating conidia to inoculate fresh PDA.These fungal cultures were then examined for GFP expression using fluorescence microscopy.
To confirm that the ZsGreen/hyg construct correctly replaced ste2 in the genome, PCR reactions were conducted using CTAB-extracted DNA for the mutant and wild-type strains and the amplicons sequenced.This was performed with the LongAmp Taq 2X Master Mix in reactions with primers CFPRE2-1+CF-PRE2-2 (designed outside of the flanking gene constructs).We also sequenced each of the flanking regions and the ZsGreen/hyg construct, following PCR with LongAmp Taq and primer sets CF-PRE2-1+ZS-HYG-HYR and ZS-HYG-YGF+CF-PRE2-2.The two amplicons were cleaned using the Zymo DNA Clean and Concentrator Kit (Zymo Research, Orange, CA, USA) and sequenced as described previously [30] with the respective PCR primers and relevant internal primers (i.e., ZS-HYG-F1, COVER-PRE2, ZS-HYG-R1 and RP6).
Copy number of hyg in the genomes of the mutants was evaluated with Southern hybridization analysis [45,46] using Roche's DIG DNA Labelling and Detection Kit and DIG Easy Hyb™ (Merck).For this purpose, primer set ZS-HYG-YGF+ZS-HYG-HYR was used to generate the 562-base pair (bp) DIG-labelled hyg probe, while Southern blots were prepared using 20-25 µg/µL of FastDigest HindIII and BamHI (Thermo Fisher Scientific) digested CTAB-extracted DNA for each of the four mutants of the two wild-type strains.Similarly, the complementation mutants were screened by Southern blot analysis for the presence of the ste2 gene by using a probe made with primer set Pre2Gene-F+Pre2Gene-R.PCRs for making the respective probes utilized Super-Therm Taq DNA Polymerase.

Quantitative Chemotropism Plate Assay
The chemotropic effects of six test compounds were analyzed in this study.Three of these were the nutrients glucose, glycerol, and glutamate, which typically provoke chemotropic responses in fungi and generally utilize GPCRs other than Ste2 [47].The fourth compound tested was the α-factor peptide pheromone of F. circinatum that was expected to interact with Ste2 [48], while the fifth compound was horseradish peroxidase (HRP), known to elicit chemotropic responses in fungi [10,[15][16][17].These five test compounds were prepared as described previously [15], which involved the use of sterile water to prepare 50 mM glucose, 15% (v/v) glycerol and 295 mM sodium glutamate (Sigma Aldrich), 4 µM HRP (Sigma-Aldrich).For comparison, peroxidase activity was abolished by adding 60 mM sodium L-ascorbate (Sigma Aldrich) directly to the HRP solution, followed by 5 min of incubation at 95 • C for 10 min.The lyophilized peptide (WCTWRGQPCW) of the synthetically produced α-factor pheromone of F. circinatum (GenScript, Piscataway, NJ, USA) was dissolved in 0.1% (v/v) of dimethyl sulfoxide and then assayed at a concentration of 378 µM [15].In all these cases, water was used as the negative control in the plate assays.
The sixth compound tested was root exudate obtained from the roots of 7-month-old Mexican weeping pine seedlings.The exudate was prepared by gently uprooting the plants from their growth substrate and carefully washing them under tap water.Following a final round of rinsing with sterilized water, the plants were placed in 500 mL Erlenmeyer flasks containing 400 mL of autoclaved distilled water, ensuring that only the roots were underwater.After covering flask openings with foil, plants were incubated for 5 days at 24 • C in a phytotron.After the removal of seedlings, the exudate-containing water was vacuum-filtered through Grade 3-HW filter membranes (Merck).Filtrates were stored in sterile 50 mL falcon tubes at −20 • C until use.
The chemotropic effects of the six test compounds were evaluated using a modified version of the plate assay (Supplementary Figure S2) reported by Turrà et al. [15].Briefly, 7-day-old PDA cultures were flooded with 15% glycerol, and 10 µL of the conidiacontaining mixture was suspended in 8 mL of 0.5% (w/v) agar (cooled to around 50 • C) to a final concentration of 2.5 × 10 6 conidia per mL.The agar suspension was poured into a 90 mm Petri plate and left to solidify.A central scoring line was then drawn on the bottom of the plate, and two parallel wells were cut on both sides into the medium at a 10 mm distance from the central line.Then, 50 µL of the test compound solution and the control were added to the wells on either side of the central line, respectively.Following the application of the test compounds, plates were incubated at 25 • C in the dark for 14 h.For each isolate, the assay was performed ten times with a particular compound.The growing direction of conidial germ tubes was recorded using a Nikon Eclipse Ni binocular microscope (100× magnification) with an affixed Nikon Ds-Ri2 camera (Nikon, Tokyo, Japan) and associated Nikon RIS Elements software (version BR 4.30.00).
A total of 200 germinating conidia were scored for each test compound, by considering only microconidia with single germ tubes.In all cases, the chemotropic responses towards compounds were compared with a negative control plate where water was added to both the test and control wells.For each strain, we quantified responses to the different compounds by determining a chemotropic index (CI).This was performed using the formula (H test − H solv )/H total × 100, where H test was the number of germ tubes pointing to the test solution, H s was the number of germ tubes pointing to the control and H total was the total number of germ tubes counted [15].The data were subjected to analyses of variance (ANOVA) using GraphPad Prism version 9.4.1 (https://www.graphpad.com).Significant differences between treatments were determined at a 5% significance level, while differences between means were evaluated with Tukey's post hoc tests.

Identification of ste2 Gene
Sequence comparisons with F. oxysporum FOXG_10633 allowed for the identification of a single ste2 homologue in the genome of strain FSP34 of F. circinatum.Unlike the ste2 genes of F. oxysporum and F. graminearum, the F. circinatum gene had one intron with a length of 61 bp.In terms of the coding region (i.e., the gene sequence without the intron), the F. circinatum gene shared 93% identity at an Expect-value (E) of 0.0 with the coding sequences of ste2 in F. oxysporum, and 76% identity with the one in F. graminearum (E = 7 × 10 −149 ).The predicted protein in both F. circinatum and F. oxysporum contained 380 amino acid residues, and the two polypeptides shared 94% identity.As expected, the predicted protein for the F. circinatum gene represented a membrane protein with seven transmembrane helices [19].We thus regarded the gene identified in F. circinatum as ste2 (NCBI accession number OR260338.1).

Gene Disruption and Complementation
Following the ste2 disruption experiments, we selected eight hygromycin-B-resistant colonies as putative transformants for each of the MAT1-1 and MAT1-2 wild-type strains for further characterization.PCR screening for the absence of ste2 and the presence of hyg identified five suitable transformants for each of the wild-type isolates.Southern blot analysis then allowed for the identification of mutants that had an integration of only one copy of hyg into the genome.From these single-copy hyg mutant strains, we chose two for the MAT1-1 wild type (designated ∆Fcste2-1 and ∆Fcste2-2) and two for the MAT1-2 wild type (designated ∆Cmste2-1 and ∆Cmste2-2) to use in subsequent experiments.Additionally, complementation of the selected deletion mutants with ste2 resulted in only a few colonies growing on PDA amended with Geneticin.Of these, PCR screening showed that only two of the MAT1-1 mutants contained the ste2 gene, and, of these, we selected one (designated ∆Fcste2C) for further experimentation.The one MAT1-2 mutant containing the ste2 gene (designated ∆Cmste2C) was selected for further use.

Deletion of ste2 Does Not Affect Chemotropism towards Glucose, Glycerol and Glutamate
Different chemotropic responses were elicited by the test compounds on the conidia of the two F. circinatum wild-type strains (Figure 1).The respective responses of the MAT1-1 wild type were not distinguishable from those evoked in the presence of water only.The chemotropic index values recorded for the glucose, glycerol and glutamate treatments were not significantly different from those observed for the control treatment with water.However, for the MAT1-2 wild-type strain, chemotropic index values recorded for the glucose and glutamate treatments differed significantly from those of the water control (p < 0.05).and ∆Cmste2 mutants did not differ significantly when exposed to glutamate and glycerol when compared to ∆Cmste2C and the wild type, respectively (Figure 1B).This indicated that ste2 deletion in the MAT1-2 wild type abolished chemotropism of its conidia to glucose in one of the deletion mutants (∆Cmste2-1) and not the other (∆Cmste2-2), while the presence of the ste2 gene did not appear to affect chemotropism in the presence of glycerol and glutamate in either of the deletion mutants.

Figure 1.
Chemotropic index values obtained for Fusarium circinatum in the presence of glucose (Gluc), glycerol (Glyc), and glutamate (Glut).Graphs indicate the directed hyphal growth of the MAT1-1 (A) and MAT1-2 (B) wild-type strains and their respective ste2 deletion mutant strains (ΔFc-ste2-1 and ΔFcste2-2, as well as ΔCmste2-1 and ΔCmste2-2) and complementation mutant strains (ΔFcste2C and ΔCmste2C) towards the different compounds, compared to the control (sterile water) after 14 h of exposure.Means that differed significantly (p < 0.05) from those obtained for the control treatment are indicated with asterisks (*).Data represent the average from at least two experiments.n = 250 germ tubes.Error bars represent standard deviation.

Ste2 Is Required for Chemotropism towards α-Factor Pheromone
The MAT1-1 and MAT1-2 wild-type strains did not differ from each other when exposed to synthetically produced α-factor pheromone of F. circinatum.This finding was confirmed by examining the responses of two additional opposite mating-type strains, CMWF1217 (MAT1-1) and CMWF1219 (MAT1-2).When subjected to synthetic The three test compounds also provoked different chemotropic responses from the mutant strains, as well as the complemented mutant strains of the two wild types (Figure 1A,B).For the MAT1-1 strains, neither of the mutants (∆Fcste2-1 and ∆Fcste2-2), nor the complemented mutant strain (∆Fcste2C) provoked responses that were significantly different from that of the wild type (Figure 1A).In other words, ste2 deletion did not affect the chemotropic responses of wild-type MAT1-1 conidia in the presence of glucose, glycerol, and glutamate.
Similarly, for the MAT1-2 strains, chemotropic index values significantly higher (p ≤ 0.05) than those for the wild type (Figure 1B) were recorded for the glutamate treatment of both deletion mutants and ∆Cmste2-2 in the glucose treatment.Mutants ∆Cmste2-1 and ∆Cmste2C generally lacked chemotropic response towards glucose.The MAT1-2 wild type and ∆Cmste2 mutants did not differ significantly when exposed to glutamate and glycerol when compared to ∆Cmste2C and the wild type, respectively (Figure 1B).This indicated that ste2 deletion in the MAT1-2 wild type abolished chemotropism of its conidia to glucose in one of the deletion mutants (∆Cmste2-1) and not the other (∆Cmste2-2), while the presence of the ste2 gene did not appear to affect chemotropism in the presence of glycerol and glutamate in either of the deletion mutants.

Ste2 Is Required for Chemotropism towards α-Factor Pheromone
The MAT1-1 and MAT1-2 wild-type strains did not differ from each other when exposed to synthetically produced α-factor pheromone of F. circinatum.This finding was confirmed by examining the responses of two additional opposite mating-type strains, CMWF1217 (MAT1-1) and CMWF1219 (MAT1-2).When subjected to synthetic pheromone, these strains also displayed positive chemotropism that did not differ significantly from one another or from FSP34 and CMWF1218 (Figure 2A).
Deletion of ste2 from MAT1-1 strain FSP34 resulted in a significant (p < 0.05) loss of chemotropic response to α-pheromone for both mutant strains ∆Fcste2-1 and ∆Fcste2-2 (Figure 2B).Its complementation mutant (∆Fcste2C) displayed chemotropic index values comparable to that of the wild type.For the MAT1-2 strain CMWF1218, none of the responses of the various mutants differed significantly from that of the wild type.Deletion of ste2 from this strain did not have a significant impact on chemotropism towards the α-factor pheromone in either of the mutants, which was also the case for the relevant complementation mutant (Figure 2C).
pheromone, these strains also displayed positive chemotropism that did not differ icantly from one another or from FSP34 and CMWF1218 (Figure 2A).
Deletion of ste2 from MAT1-1 strain FSP34 resulted in a significant (p < 0.05) chemotropic response to α-pheromone for both mutant strains ∆Fcste2-1 and ∆F (Figure 2B).Its complementation mutant (∆Fcste2C) displayed chemotropic index comparable to that of the wild type.For the MAT1-2 strain CMWF1218, none of sponses of the various mutants differed significantly from that of the wild type.De of ste2 from this strain did not have a significant impact on chemotropism towards factor pheromone in either of the mutants, which was also the case for the relevan plementation mutant (Figure 2C).

Pine Root Exudate Provokes Ste2-Dependent Chemotropism
Both wild-type strains used in this study showed positive chemotropism to pine root exudate (Figure 3).However, deletion of ste2 from the MAT1-1 wild type y significantly (p < 0.05) reduced chemotropic index values, although the values ob for the corresponding complementation mutant did not differ significantly from th the wild type (Figure 3A).Likewise, the deletion of ste2 from the MAT1-2 wild type tively impacted the chemotropic response towards root exudate in the two mutant not in the complementation mutant ∆Cmste2C (Figure 3B).
To evaluate whether the chemotropic responses observed might be due to the ence of peroxidases in the pine root exudate, we subjected our wild-type and m strains to assays using active and inactivated HRP (Supplementary Figure S3A,B results show that untreated HRP stimulated a positive chemotropic response in bo MAT1-1 and MAT1-2 wild-type strains of F. circinatum, while treatment with inact HRP yielded much a reduced response in the MAT1-1 wild type and a negative c tropic response in the MAT1-2 wild type compared to when untreated HRP was However, these responses did not appear to be dependent on ste2, as responses rec for the deletion and complementation mutants generally followed the same pa those observed for the wild-type strains.

Pine Root Exudate Provokes Ste2-Dependent Chemotropism
Both wild-type strains used in this study showed positive chemotropism towards pine root exudate (Figure 3).However, deletion of ste2 from the MAT1-1 wild type yielded significantly (p < 0.05) reduced chemotropic index values, although the values obtained for the corresponding complementation mutant did not differ significantly from those of the wild type (Figure 3A).Likewise, the deletion of ste2 from the MAT1-2 wild type negatively impacted the chemotropic response towards root exudate in the two mutants, and not in the complementation mutant ∆Cmste2C (Figure 3B).
To evaluate whether the chemotropic responses observed might be due to the presence of peroxidases in the pine root exudate, we subjected our wild-type and mutant strains to assays using active and inactivated HRP (Supplementary Figure S3A,B).The results show that untreated HRP stimulated a positive chemotropic response in both the MAT1-1 and MAT1-2 wild-type strains of F. circinatum, while treatment with inactivated HRP yielded much a reduced response in the MAT1-1 wild type and a negative chemotropic response in the MAT1-2 wild type compared to when untreated HRP was used.However, these responses did not appear to be dependent on ste2, as responses recorded for the deletion and complementation mutants generally followed the same pattern as those observed for the wild-type strains.

Discussion
This study demonstrated that the Ste2 mating pheromone receptor of F. circinatum mediates positive hyphal chemotropism towards root exudate obtained from pine seedlings.In both MAT1-1 and MAT1-2 individuals, significantly (p < 0.05) reduced chemotropic index values were obtained when ste2 deletion mutants were exposed to root exudates.For complementation mutants harboring a randomly integrated functional copy of ste2, we observed chemotropic indices comparable to those of the wild types.A similar elimination of positive chemotropism towards the roots of host plants was also observed for ste2 deletion mutants of F. oxysporum, F. graminearum, Verticillium dahliae and Trichoderma reesei [9,10,17,24].These results thus confirmed that Ste2 is needed for targeted hyphal growth of F. circinatum towards exudates released by the roots of its pine host, suggesting a primary role for the protein in this pathosystem.Accordingly, our future research will explore whether inoculation with ste2 deletion mutants is indeed associated with reduced infection of pine seedling roots.
The precise compounds responsible for Ste2-dependent chemotropism of F. circinatum toward pine root exudate remains unknown, as plant-encoded class III peroxidases appeared not to be involved in the process [10,15,17,24].These enzymes are ubiquitous in land plants [49], where they provide defense against biotic stresses by creating conditions that are toxic to potential pathogens [50].In the current study, higher chemotropic indices were observed in the presence of HRP than in inactivated HRP and in the water control treatments for both of the F. circinatum wild-type strains examined, which is similar to what was seen in F. oxysporum, F. graminearum, T. reesei and V. dahliae [10,15,17,24].However, the various ste2 deletion mutants of F. circinatum yielded chemotropic index values that were not significantly different from the wild types.By contrast, those of F. oxysporum, F. graminearum, and V. dahliae lacked chemotropism towards HRP [10,15,17,24].Because such a dependence on Ste2 was not seen in the F. circinatum, the compound(s) in pine root exudate that interacts with Ste2 thus remains to be determined.Indeed, the roots of plants, including those of pine, are known to release a diversity of compounds into the rhizosphere [4,51].Our future research will explore the composition of the exudates for pine roots and investigate the ligand(s) potentially mediating Ste2-dependent chemotropism.(B) Chemotropic responses of the MAT1-2 wild-type strain and its two ste2 deletion mutants (∆Cmste2-1 and ∆Cmste2-2) and complementation mutant (∆Cmste2C).Means that differed significantly (p < 0.05) from those obtained from the wild-type strains are indicated with asterisks (*).Data are presented as the mean from two experiments.n = 250 germ tubes.Error bars show standard deviation.

Discussion
This study demonstrated that the Ste2 mating pheromone receptor of F. circinatum mediates positive hyphal chemotropism towards root exudate obtained from pine seedlings.In both MAT1-1 and MAT1-2 individuals, significantly (p < 0.05) reduced chemotropic index values were obtained when ste2 deletion mutants were exposed to root exudates.For complementation mutants harboring a randomly integrated functional copy of ste2, we observed chemotropic indices comparable to those of the wild types.A similar elimination of positive chemotropism towards the roots of host plants was also observed for ste2 deletion mutants of F. oxysporum, F. graminearum, Verticillium dahliae and Trichoderma reesei [9,10,17,24].These results thus confirmed that Ste2 is needed for targeted hyphal growth of F. circinatum towards exudates released by the roots of its pine host, suggesting a primary role for the protein in this pathosystem.Accordingly, our future research will explore whether inoculation with ste2 deletion mutants is indeed associated with reduced infection of pine seedling roots.
The precise compounds responsible for Ste2-dependent chemotropism of F. circinatum toward pine root exudate remains unknown, as plant-encoded class III peroxidases appeared not to be involved in the process [10,15,17,24].These enzymes are ubiquitous in land plants [49], where they provide defense against biotic stresses by creating conditions that are toxic to potential pathogens [50].In the current study, higher chemotropic indices were observed in the presence of HRP than in inactivated HRP and in the water control treatments for both of the F. circinatum wild-type strains examined, which is similar to what was seen in F. oxysporum, F. graminearum, T. reesei and V. dahliae [10,15,17,24].However, the various ste2 deletion mutants of F. circinatum yielded chemotropic index values that were not significantly different from the wild types.By contrast, those of F. oxysporum, F. graminearum, and V. dahliae lacked chemotropism towards HRP [10,15,17,24].Because such a dependence on Ste2 was not seen in the F. circinatum, the compound(s) in pine root exudate that interacts with Ste2 thus remains to be determined.Indeed, the roots of plants, including those of pine, are known to release a diversity of compounds into the rhizosphere [4,51].Our future research will explore the composition of the exudates for pine roots and investigate the ligand(s) potentially mediating Ste2-dependent chemotropism.
As reported for F. oxysporum and V. dahliae, the Ste2 receptor of F. circinatum also does not mediate chemotropic responses towards glutamate, glycerol, and glucose [10,15].Positive chemotropic responses towards the three compounds were observed for the wild-type F. circinatum strains, but these mostly did not differ significantly from the control treatment nor those observed for the ste2-deletion mutants.This indicates that chemotropism by F. circinatum towards these compounds requires the activity of one or more receptor proteins other than Ste2.In T. reesei and F. graminearum GPCRs unrelated to Ste2, respectively, mediate chemotropism towards glucose [24] and nutrients extracted from wheat spikelets [52].Investigation of the repertoire of GPCRs in F. circinatum might also lead to the identification of receptors mediating chemotropic responses to specific nutrients.For the latter, a suitable candidate might be strain ∆Cmste2-1 obtained in the current study, as the generation of this mutant apparently affected target(s) in addition to the ste2 gene, causing the fungus to behave unusually in the chemotropic assays involving glucose.This suggests that a receptor (or other relevant protein) required for chemotropism towards glucose was compromised in the mutant while retaining its chemotropic responsiveness to pine root exudates.Further examination of mutant ∆Cmste2-1 might thus contribute to our understanding of nutrient acquisition in F. circinatum.
By focusing on F. circinatum, this study is the first to investigate Ste2-mediated chemotropism to one of the canonical pheromones in a heterothallic member of the Ascomycota.Both F. oxysporum and V. dahliae are thought to be asexual [53,54], although high levels of recent meiotic recombination were reported in a lineage of F. oxysporum [55].In the case of V. dahliae, sexual reproduction is apparently prevented by karyotype variation that interferes with homologous chromosome pairing during meiosis [56].Nevertheless, strains of F. oxysporum and V. dahliae harbor either the MAT1-1 or MAT1-2 versions of the MAT1 locus [57,58] and not both as is the case for homothallic species such as F. graminearum.The F. oxysporum strain in which chemotropism was studied carries the MAT1-1 version of the locus, while the V. dahliae strain studied carries the MAT1-2 version [15,59].In the current study, we examined both MAT1-1 and MAT1-2 isolates in the sexual fungus F. circinatum and found that strains of either mating specificity responded chemotropically to synthetically produced α-peptide pheromone.Previously, synthetically produced αpheromone for, respectively, F. oxysporum and F. graminearum, provoked high levels of positive chemotropism in these fungi [15,17], while the synthetic α-peptide from F. oxysporum elicited a similar response in V. dahliae [10].In F. circinatum, deletion of the ste2 gene also abolished the chemotropism towards the peptide, while it was restored in the complementation mutants.Therefore, under the conditions tested in the current study, our findings suggest that Ste2 of F. circinatum mediates positive chemotropism towards self-produced α-peptide pheromone in MAT1-1 isolates, as well as MAT1-2 isolates.
The chemotropic response to α-peptide pheromone by the F. circinatum MAT1-1 individual and the MAT1-2 individual contrasted with expectations for a fungus with a typical heterothallic system.We expected to observe chemotropic responses consistent with paracrine pheromone signaling (sensu Vitale et al. [23]) where the signals from one mating-type act on the hyphae of the opposite mating-type.This is because activation of the dedicated MAPK signaling cascade needed for sexual development requires the interaction of the α-pheromone produced by the MAT1-1 strain with Ste2 from the MAT1-2 strain, and interaction between the MAT1-2 produced a-pheromone and the MAT1-1 produced Ste3 [60].Most of what we currently know about the process was gathered from gene deletion and complementation studies of model Ascomycota such as Saccharomyces cerevisiae, Neurospora crassa and Podospora anserina, and a small number of non-model fungi (for a comprehensive review, see [19]).Moreover, to the best of our knowledge, only the study on T. reesei directly examined the impact of synthetic pheromone peptides on the chemotropic responses of opposite mating-type isolates of a true heterothallic fungus [24].However, the set of pheromones mediating mating in T. reesei is unusual in that its a-pheromone represents a hybrid between the canonical αand a-pheromones.Also, T. reesei experiments using the synthetic a-peptide derived from F. oxysporum elicited responses resembling paracrine pheromone signaling, although assays using the F. oxysporum α-pheromone failed due to solubility issues [24].Nevertheless, the collective findings of these studies highlight that there are still gaps in our understanding of sexual reproduction in Ascomycota and emphasize the importance of including non-model fungi in research enterprises.
The chemotropic responses of F. circinatum conidia to α-pheromone might point to one or more processes other than sexual reproduction being regulated by pheromone signaling [15,44].The latter could include so-called autocrine pheromone signaling where the peptide signal effects changes in the cells producing it.For example, Vitale et al. [23] showed that this form of signaling in F. oxysporum regulates cell density-dependent germination of conidia via the CWI pathway.In this fungus, the self-produced αand a-pheromones interact with the self-produced Ste2 and Ste3 GPCRs, respectively, and the corresponding α/Ste2 and a/Ste3 signals either promote or inhibit conidial germination via a quorumsensing mechanism.In the MAT1-1 strain of F. oxysporum examined, the Bar1 aspartyl protease degrades α-pheromone at low cell densities (i.e., low α-pheromone concentrations), thus allowing for the competitive recruitment of a/Ste3 to the CWI MAPK cascade and the promotion of conidial germination.At higher cell densities (i.e., high α-pheromone concentrations), conidial germination is repressed because Bar1's degradative activity is not sufficient for preventing the recruitment of α/Ste2 to the MAPK cascade.This is consistent with anecdotal observations in F. circinatum, where high concentrations of synthetic α-pheromone also inhibited conidial germination (unpublished).Even so, it is unclear whether a similar autocrine pheromone signaling system regulates conidial germination or other phenotypes in F. circinatum, but it would be interesting to explore the role(s) of αand a-pheromone as quorum-sensing molecules in this fungus.
The notion that different concentrations of the various components of pheromone signaling might mediate different biological outcomes in F. circinatum aligns well with the limited information available for this fungus.For example, RNA-seq data on F. circinatum conidial germination in a MAT1-1 wild-type strain revealed nearly four times more ste2 than ste3 transcripts in libraries produced under similar conditions [35,61].Additionally, the α-peptide pheromone gene of F. circinatum also differs widely between isolates in the number of peptide motifs included in its preprotein, as well as the exact amino acid composition of the motifs, suggesting that individuals of the fungus produce different amounts of mature peptides that vary at their N-termini [44].The latter region in the α-peptide's overall structure has been shown to determine growth inhibition in F. oxysporum [62].Therefore, any one of these or other unknown factors could have impacted the outcome of our chemotropic assays with α-peptide pheromone and the two wild-types included, particularly in terms of the functional levels of Ste2 and α-peptide pheromone (endogenously produced and added in vitro during the assays).

Conclusions
The findings of this study shed light on our knowledge of both the pathogenesis and mating system of F. circinatum.Given the importance of conidia as an inoculum source for root disease outbreaks in pine seedling nurseries [63], aspects related to GPCRs such as Ste2 and the pathogen's responses to synthetic α-peptide pheromone provide potential avenues for developing disease management strategies.In fact, Brown et al. [47] argued that the targeting of pheromone signaling for this purpose might have the dual effect of reducing a pathogen's dispersal and its genetic diversity because of interference in its interactions with the host and with mating partners.In fact, research into how this type of information is translated into applied products is well-developed in the medical field, where pheromone signaling and GPCRs form the basis of various therapeutics for human diseases [47].Adaptation of these proven procedures and the application of our findings might ultimately allow for the successful manipulation of the plant-pathogen interaction, thereby preventing losses due to F. circinatum-associated disease by pine-based industries worldwide.

Figure 1 .
Figure1.Chemotropic index values obtained for Fusarium circinatum in the presence of glucose (Gluc), glycerol (Glyc), and glutamate (Glut).Graphs indicate the directed hyphal growth of the MAT1-1 (A) and MAT1-2 (B) wild-type strains and their respective ste2 deletion mutant strains (∆Fcste2-1 and ∆Fcste2-2, as well as ∆Cmste2-1 and ∆Cmste2-2) and complementation mutant strains (∆Fcste2C and ∆Cmste2C) towards the different compounds, compared to the control (sterile water) after 14 h of exposure.Means that differed significantly (p < 0.05) from those obtained for the control treatment are indicated with asterisks (*).Data represent the average from at least two experiments.n = 250 germ tubes.Error bars represent standard deviation.
1 Δ F c s t e 2 -1 Δ F c s t e 2 -2 Δ F c s t e 2

Figure 3 .
Figure 3. Chemotropic index values obtained for F. circinatum in the presence of root exudates obtained from Mexican weeping pine seedlings.(A) Chemotropic responses of the MAT1-1 wild type and its ste2 deletion mutants (ΔFcste2-1 and ΔFcste2-2) and complementation mutant ΔFcste2C.(B) Chemotropic responses of the MAT1-2 wild-type strain and its two ste2 deletion mutants (ΔCmste2-1 and ΔCmste2-2) and complementation mutant (ΔCmste2C).Means that differed significantly (p < 0.05) from those obtained from the wild-type strains are indicated with asterisks (*).Data are presented as the mean from two experiments.n = 250 germ tubes.Error bars show standard deviation.

Figure 3 .
Figure 3. Chemotropic index values obtained for F. circinatum in the presence of root exudates obtained from Mexican weeping pine seedlings.(A) Chemotropic responses of the MAT1-1 wild type and its ste2 deletion mutants (∆Fcste2-1 and ∆Fcste2-2) and complementation mutant ∆Fcste2C.(B) Chemotropic responses of the MAT1-2 wild-type strain and its two ste2 deletion mutants (∆Cmste2-1 and ∆Cmste2-2) and complementation mutant (∆Cmste2C).Means that differed significantly (p < 0.05) from those obtained from the wild-type strains are indicated with asterisks (*).Data are presented as the mean from two experiments.n = 250 germ tubes.Error bars show standard deviation.