Influence of Different Light Regimes on the Mycoparasitic Activity and the Production of the Secondary Metabolite 6-Pentyl-α-Pyrone in Two Strains of Trichoderma Atroviride

The ascomycete Trichoderma atroviride is well known for its mycoparasitic lifestyle. Similar to other organisms, light is an important cue for T. atroviride. However, besides triggering of conidiation, little is known on the physiological responses of T. atroviride to light. In this study, we analyzed how cultivation under different light wavelengths and regimes impacted the behavior of two T. atroviride wild-type strains, IMI206040 and P1. While colony extension of both strains was slightly affected by light, massive differences in the photoconidation response between the two strains became evident. T. atroviride P1 colonies conidiated under all conditions tested including growth in complete darkness, while IMI206040 required white, blue or green light to trigger asexual reproduction. Interestingly, deletion of the stress-activated MAP kinase-encoding gene tmk3 abolished the ability of strain P1 to conidiate in red and yellow light as well as in darkness. Furthermore, light-dependent differences in the mycoparasitic activity of T. atroviride and in the biosynthesis of the secondary metabolite 6-pentyl--pyrone (6-PP) became evident. 6-PP production was highest upon dark incubation while light, especially exposure to white light as light/dark cycles, had an inhibitory effect on its biosynthesis. We conclude that the response of T. atroviride to light is strain-dependent and impacts differentiation, mycoparasitism and 6-PP production and hence should be considered in experiments testing the mycoparasitic activity of these fungi.


Introduction
The mycoparasitic fungus Trichoderma atroviride is applied in agriculture to protect plants against a wide range of fungal pathogens. The mycoparasitic attack comprises several processes such as the production of antifungal metabolites and hydrolytic enzymes and is largely affected by environmental conditions [1]. Much of our knowledge on fungal mycoparasitism comes from studies with T. atroviride, of which two strains are frequently used as model organisms. T. atroviride IMI206040, a strain isolated from a plum tree in an orchard in Southern Sweden, and T. atroviride P1 (ATCC 74058), a fungicide resistant isolate from the UK. Both T. atroviride strains are known as producers of 6-pentyl-α-pyrone (6-PP), a secondary metabolite with strong antifungal activity and concentration-dependent plant-growth promoting characteristics, that is responsible for the characteristic "coconut aroma" of certain Trichoderma species [2,3]. However, the biosynthetic route 3 of 16 radial growth, conidiation behavior, and mycoparasitic activities as well as the influence of the different light regimes on the production of the antifungal secondary metabolite 6-pentyl-α-pyrone . In addition, we assessed the role of the Tmk3 MAP kinase in these processes by employing respective Δtmk3 deletion mutants of both strains.

Strains and growth conditions
Trichoderma atroviride strains IMI206040 (ATCC 20476) and P1 (ATCC 74058) as well as the plant pathogens Botrytis cinerea B05.10, Rhizoctonia solani (pathogenic isolate obtained from the collection of the Institute of Plant Pathology, Università degli Studi di Napoli "Federico II", Naples, Italy), and Fusarium oxysporum f. sp. lycopersici strain 4287, were used in this study. Fungi were cultivated and maintained on potato dextrose agar (PDA; Becton, Dickinson and Company, Le Pont De Claix, France) at 25 °C in darkness or under different light regimes. The Δtmk3 mutant strain derived from IMI206040 [18] was maintained in the presence of 200 µ g/mL Hygromycin B (Calbiochem ® , Merck KGaA, Darmstadt, Germany).

Generation of tmk3 gene deletion mutants of T. atroviride P1
To knock-out the tmk3 gene, T. atroviride P1 was transformed with a deletion cassette containing 1 kb of the 5′ and 3′ noncoding regions of tmk3 flanking the hph (Hygromycin B-mediating resistance) cassette obtained from plasmid pGFP-XYR1 [48]. Primers used for DNA fragment amplification and assembly with the NEBuilder ® Hifi DNA Assembly Kit (New England Biolabs, Massachusetts, USA) are given in Supplementary Table 1. Resulting transformants were selected on PDA containing 200 µ g/mL Hygromycin B (Calbiochem ® , Merck KG, Darmstadt, Germany) and purified to mitotic stability by three rounds of single spore isolation. Deletion of tmk3 and locus specific integration of the hygromycin resistance cassette was verified by PCR-based genotyping using gene-and locusspecific primer pairs (Supplementary Table 1) as previously described [4].

Light conditions
T. atroviride was cultivated in triplicates on PDA plates at 25° C for up to 7 days. To this end, agar plugs of six millimeter diameter of the actively growing colony margins from pre-cultures were inoculated at the center of fresh PDA plates and incubated under: a) complete darkness, b) white light-dark cycles (12:12 h; Economic lux chamber, Snijders Labs, 30 W/m 2 max intensity); or c) varying light wavelengths. For the latter purpose, light chambers were designed using LED bulbs emitting blue (light wavelength ≈ 459 nm, luminous intensity 12,85 µ W/cm 2 ), green (light wavelength ≈ 517 nm, luminous intensity 6,22 µ W/cm 2 ), red (light wavelength ≈ 630 nm, luminous intensity 2,70 µ W/cm 2 ), or yellow (light wavelength ≈ 590 nm, luminous intensity 2,00 µW/cm 2 ) light. LEDs were positioned at 9 cm vertical height over the fungal colonies. For determination of the radial growth rate, colony radii were measured and the radial growth rate [cm/d] was calculated for each time point. Conidia were quantified with a haemocytometer after being harvested from a seven days old culture grown on PDA.

Mycoparasitic activity assay
Plate confrontation assays of T. atroviride against B. cinerea, R. solani, or F. graminearum were performed in triplicates on PDA plates. Fungi were inoculated at a distance of 6 cm and plates incubated at 25° C for a total of seven days under the different light treatments described above.

Quantification of 6-PP
For determination of secreted 6-PP, T. atroviride was cultivated in triplicates for 48 h (wild type) or 72 h (Δtmk3 mutants due to their slower growth) at 25° C on cellophane-covered PDA plates under the different defined light regimes. After removal of the mycelia-covered membrane, 1 g of agar situated below the colony was harvested and the triplicates pooled. Mycelial dry weight was determined from each membrane to assess biomass production. 5 mL of the extraction solvent, consisting of methanol (MeOH):H2O 3:1 [v/v] + 0.1% formic acid (FA), were added to 1 g of agar (MeOH: Merck, Darmstadt, Germany; water purified with ELGA Purelab Ultra-AN-MK2: Veolia Water, Vienna, Austria; FA: MS-grade, Sigma-Aldrich, Vienna, Austria) and sonicated for 15 min. 0.5 mL of acidified water containing 0.1% FA were added to 1 mL extract to obtain an organicsolvent:water ratio of 1:1 [v/v]. For liquid-chromatography coupled to high-resolution mass spectrometry (LC-HRMS) analysis, the samples were additionally diluted 1:10 [v/v] with MeOH:H2O 1:1 [v/v] + 1% FA. Samples were analyzed on a LC-HRMS system consisting of a Vanquish ultra-highperformance liquid chromatography (UHPLC) coupled to a QExactive Orbitrap HF mass spectrometer (Thermo Fisher Scientific, Bremen, Germany). 2 µ l of the sample were injected and chromatographed on a reversed phase C18 column XBridge 150 × 2.1 mm i.d., 3.5 µm (Waters, Milford, USA). H2O and MeOH, both acidified with 0.1% FA, were used as eluents A and B, respectively, to obtain a linear-gradient elution with increasing MeOH content. After an initial hold time of 1 min at 10% eluent B, the methanol content was increased to 100% within 9 min (3 min hold) before the system was re-equilibrated for 7 min at 10% eluent B (total run time 20 min). Flow rate was kept constant at 0.25 mL/min. Mass spectra were recorded from m/z 100 to 1,000 in positive ionization mode with a resolving-power setting of 120,000 at m/z 200. Quantification was carried out using the XCalibur software (Thermo Fisher Scientific, Bremen, Germany) after external standard calibration of 6-PP (purity > 96%; 1, 5, 10, 50, 100, 500, 1000 and 5000 µ g/L; Sigma-Aldrich, Vienna, Austria). 6-PP values were normalized to the mycelial dry weight of each triplicate.

Stress assays
To analyze stress resistance, conidia collected from 7 days old Trichoderma cultures grown at 25° C under different light conditions were exposed to stress agents. 10 6 conidia were inoculated on PDA plates containing 108 µ M congo red or 31 µ M calcofluor white to study cell wall stress resistance, 0.5 M NaCl to study osmotic stress resistance, and 2.5 mM H2O2 for oxidative stress resistance analysis. At least three biological replicates were incubated for two days at 25° C under light-dark conditions. Mycelial stress resistance was analyzed by inoculating mycelia-covered plugs from pre-cultures grown under different light or dark conditions on PDA plates supplemented with respective stress agents.

Microscopic analysis
Fungal hyphae were imaged with an inverted Nikon optiphot-2 microscope or a Nikon SMZ1500 stereomicroscope and images were captured with a digital camera.

Statistical analysis
Data were subject to one-way analysis of variation (ANOVA), and treatment means were separated using least significance difference (LSD) at P = 0.05 [49]. All analyses were performed using the package IBM SPSS Statistics 24.

Effect of different light regimes on colony extension and asexual development of T. atroviride
Light is a factor that influences many cellular processes. In T. atroviride IMI206040, blue light has been shown to regulate asexual reproduction through the Tmk3 MAPK pathway [18]. Based on this, we intended to assess the influence of not only blue light but light of different defined wavelengths on radial growth and conidiation of the T. atroviride wild type strains IMI206040 and P1 as well as of Δtmk3 mutants derived thereof.
When grown under the different light regimes (complete darkness, white light-dark cycle, or in the presence of either blue, green, yellow or red light), both wild type strains developed similar colony diameters irrespective of the applied light conditions tested. 48 h after inoculation, however, strain IMI206040 had developed slightly larger colonies upon incubation in the dark or when grown in the presence of yellow or red light, while light of shorter wavelengths, such as blue light, negatively impacted colony extension ( Figure 1). This effect was not visible with strain P1. Irrespective of the applied light or dark conditions, strain IMI206040 exhibited a slightly enhanced colony extension rate compared to strain P1 reaching the border of the plates already after 72 h of growth ( Figure 1). Compared to the wild type strains, the Δtmk3 mutants showed a clear growth reduction. Colonies of the IMI206040-derived Δtmk3 mutant covered the plates not until 144 h of cultivation in darkness, while light exposure, especially with blue light, led to reduced growth. The P1-derived Δtmk3 mutant reached the border of the plates after 144 h only when grown in the presence of green, yellow or red light, while the colony remained significantly smaller upon growth in complete darkness, white lightdark cycles, and in the presence of blue light ( Figure 1). The experiment was repeated three times with three replicates each. Asterisks denote significance level: *p <0,05, **p <0,01, ***p <0,001.
Although colony extension of both T. atroviride wild type strains was only slightly affected by light, massive light-dependent differences in the colony phenotypes between the two strains became evident. T. atroviride P1 colonies became green-colored, indicating sporulation, under all conditions tested including growth in complete darkness. In contrast, IMI206040 colonies stayed unpigmented upon exposure to yellow and red light and upon cultivation in darkness, indicating that this strain is only able to conidiate in the presence of white, blue and green light (Figure 2 A). The conidiation behavior of the IMI206040-derived Δtmk3 mutant resembled the parental strain, while interestingly deletion of the tmk3 gene prevented P1-derived mutants to produce green-pigmented conidia in darkness and upon growth in the presence of yellow and red light, a behavior resembling that of the IMI206040 wild type strain. produced by the two T. atroviride wild type strains (A) and their tmk3 mutants (B) upon growth under the different light regimes. The experiment was repeated three times with three replicates each. Results shown are means ± SD. p values: *p <0,05, **p <0,01, ***p <0,001.
To assess whether the observed effects of light on T. atroviride conidiation were actually due to alterations in conidia production or only in conidial pigmentation, the number of conidia produced by the different strains under the various light conditions was determined. Upon growth under white light-dark cycles, blue, and green light, T. atroviride IMI206040 produced similar numbers of conidia as strain P1, while no conidia could be obtained from IMI206040 colonies grown in the presence of red and yellow light, and in complete darkness. T. atroviride strain P1 in contrast produced similar numbers of conidia upon growth in darkness, exposure to blue and green light, and under white light-dark cycles. Interestingly, increased spore densities were observed under yellow and red light in T. atroviride P1 suggesting that these light wavelengths additionally trigger conidia production in this strain (Figure 2 B). Conidia numbers in the IMI206040-and P1-derived Δtmk3 mutants were similar to each other upon growth under white light-dark cycles and in the presence of blue or green light. However, the numbers of conidia from both mutants were massively reduced compared to their respective wild types. In addition, cultures of both Δtmk3 mutants were devoid of conidia upon cultivation in darkness and in the presence of yellow or red light, which is similar to the phenotype of the IMI206040 wild type strain but contrasts the behavior of strain P1.

Assessment of light-induced stress resistance
Photoperception allows fungi to trigger stress resistance in response to light. Blue light exposure of dark-grown Aspergillus fumigatus for example resulted in enhanced resistance to UV or hydrogen peroxide mediated stress [10] and growth of Metarhizium robertsii under visible light led to conidia that were more tolerant to UV than conidia from dark-grown cultures [50]. As the stress-activated MAP kinase pathway involving Tmk3 has been reported to integrate stress and light signals in T. atroviride [18], we evaluated the effect of different light regimes on cellular stress management in the two T. atroviride wild-type and ∆tmk3 mutant strains. All strains tested were able to develop colonies in the presence of 31 µ M calcofluor white that were of similar sizes as those grown without stressor ( Figure 3). Congo red mediated cell wall stress resulted in moderately reduced colony sizes in both P1 and IMI206040, which however were not affected by loss of tmk3. In contrast, both Δtmk3 mutants were completely unable to cope with NaCl-mediated osmotic stress, while colony development of the respective wild types was severely inhibited but still possible under this condition. Wild type strain P1 was more sensitive to H2O2-mediated oxidative stress than IMI206040. In addition, both Δtmk3 mutants could hardly grow in the presence of this stressor indicated by the fact that development of a colony could only start after 72 h of cultivation. For all strains tested, however, colony development and growth under the tested conditions were completely independent of the previous light exposure. This was even the case with wild type strain P1 whose conidia derived from light-exposed cultures did not show enhanced resistance to any of the stressors tested compared to conidia produced by dark-grown colonies.

Effect of different light regimes on the mycoparasitic activity of T. atroviride wild types and ∆tmk3 mutants
To study the effect of the different light treatments on the mycoparasitic activity of T. atroviride, the two wild type strains and their ∆tmk3 mutants were co-cultivated with the host fungi R. solani and F. oxysporum in plate confrontation assays. Similar to axenic cultures, wild type strain P1 conidiated in co-cultures with the two tested fungal hosts irrespective of the applied light regime, while conidiation of IMI206040 in the presence of the fungal hosts remained dependent on white, blue and green light (Figure 4). The ability to antagonize and overgrow R. solani only slightly differed between the two T. atroviride wild type strains and turned out to be only marginally influenced by the applied light regime. In all cases, both, IMI206040 and P1, were able to fully overgrow the host fungus within seven days. In the interaction with F. oxysporum, the mycoparasitic overgrowth ability of both T. atroviride wild type strains was better in darkness and upon yellow or red light exposure, while white light-dark cycle conditions hampered the mycoparasitic attack. Compared to the wild type strains, both ∆tmk3 mutants showed a reduced ability to overgrow the fungal hosts. The P1derived ∆tmk3 mutant showed partial overgrowth of both hosts under all light conditions tested while the IMI206040-derived mutant already stopped its growth at the interaction border. This behavior was most evident under blue, green, yellow and red light conditions where the mutant's colonies remained smaller than upon cultivation in darkness or under white light-dark cycle conditions ( Figure 4).

Effect of different light regimes on the production of the secondary metabolite 6-PP
To evaluate the influence of light on 6-PP production by T. atroviride, both wild type strains and their ∆tmk3 mutants were grown on PDA plates under different light conditions. 6-PP production was highest upon cultivation in complete darkness in all strains tested and lowest upon growth in the presence of white light-dark cycles ( Figure 5). This inhibitory effect of white light was most Preprints (www.preprints.org) | NOT PEER-REVIEWED | Posted: 11 September 2020 doi:10.20944/preprints202009.0251.v1 evident in wild type strain IMI206040, which was able to secrete ~ 6.5 mg 6-PP per gram mycelial dry weight in darkness but below 0.5 mg 6-PP per gram mycelial dry weight upon growth under white light-dark cycles (Figure 5 A). Blue, green, yellow and even red light inhibited 6-PP biosynthesis in both wild type strains compared to growth in darkness, although to a lower extent than white light cycles. The negative impact of light from across the whole spectrum on the amount of 6-PP secreted into the agar was less evident in the Δtmk3 mutants than in their respective wild types. Especially the P1-derived mutant produced similar levels of 6-PP upon growth under blue, green, yellow and red light than in darkness and also white light-dark cycles only had a minor negative effect on 6-PP biosynthesis in this mutant ( Figure 5). In the IMI-derived ∆tmk3 mutant, the repressive effect of white light treatment was similar than in its wild type.

Discussion
In this study, we have shown that light exposure affects the phenotype and behavior of T. atroviride in many ways. We have found significant differences in the response to light between the two tested T. atroviride wild type strains IMI206040 and P1, and have evaluated the effect of loss of the Tmk3 MAP kinase. Reduction of growth by continuous white light and enhanced radial colony growth in darkness has been previously described for T. atroviride IMI206040 [53]. In our experiments, however, cultivation in the presence of white light-dark cycles and light of different wavelengths only had a minor effect on colony extension but clearly impacted differentiation. The latter is in accordance with previous reports showing that asexual reproduction of T. atroviride IMI206040 is tightly regulated by light [22,51,52,45,53,54]. Accordingly, we found that IMI206040 required light, either white, blue or green, to trigger conidiation and the strain produced similar numbers of conidia under these light conditions. In contrast, conidiation of strain P1 occurred in a light-independent way and also in complete darkness. Interestingly, however, yellow and red light seemed to additionally trigger conidia formation in strain P1 as the fungus produced enhanced numbers of conidia compared to the other light conditions tested upon illumination with these wavelengths. Taken together, asexual development is differently affected by light in the two T. atroviride strains IMI206040 and P1, suggesting differences in light sensing and/or activation of conidiation-related lightresponsive genes.
Besides a general growth reduction, deletion of the gene encoding the Tmk3 MAP kinase led to significantly less conidia upon growth under white light-dark cycles, blue and green light in both T. atroviride strains, which is in accordance with a previous study on strain IMI206040 showing that Tmk3 regulates photoconidiation in T. atroviride [18]. Interestingly, however, strain P1 lost the ability to conidiate in the dark or in the presence of yellow and red light upon tmk3 gene deletion, while conidiation in response to white, blue and green light was still possible in both P1-as well as IMI206040-derived Δtmk3 mutants. In IMI206040, the two blue light regulators Blr1 and Blr2 have been found to be essential for photoconidiation and control of light responsive genes. In addition, Tmk3 has been demonstrated to corroborate with the Blr photoreceptor complex in activation of gene expression [18]. Tmk3 gene transcription as well as phosphorylation of Tmk3 are triggered upon light exposure and light regulates asexual reproduction through the Tmk3 pathway implying that this MAP kinase is a key player in the light sensing pathway [18]. The role of Tmk3 has hitherto only been studied in strain IMI206040 in response to white and blue light. Our results additionally suggest a connection between Tmk3 and red light signaling, while, interestingly, tmk3 gene deletion still allowed conidiation in response to blue light.
Similar to other fungi, the HOG pathway and its Tmk3 MAPK has previously been demonstrated to participate in high osmolarity and oxidative stress resistance as well as cell wall integrity in T. reesei and T. atroviride IMI206040 [18,39]. In our study, we found a similar role of Tmk3 in T. atroviride P1. Respective tmk3 gene deletion mutants were unable to cope with NaCl-mediated osmotic stress and could hardly develop colonies in the presence of the oxidative stress-triggering agent H2O2. However, in the Δtmk3 mutants as well as their respective wild type strains, the previous light exposure or absence of light had no significant impact on the resistance to the stressors tested including NaCl (osmotic stress), H2O2 (oxidative stress), as well as congo red and calcofluor white (cell wall stress). These results are somehow contrasting previous findings with other fungi suggesting that visible light acts as a signal for stress [55]. In A. fumigatus, for example, the response to visible light included enhanced resistance to UV and oxidative stress and an increased susceptibility to cell wall perturbation [10]. Secondary metabolite production is as well among the various processes impacted by light in fungi [56]. The impact of light, however, seems to be dependent on the fungal species as well as the specific secondary metabolite. While for example in Aspergillus flavus aflatoxin biosynthesis was negatively affected by light, production of ochratoxin in Aspergillus ochraceus was enhanced [57,58]. Biosynthesis of the mycotoxin citrinin by Penicillium verrucosum was increased by blue light [59], toxin production in Alternaria alternata was reduced upon blue light irradiation [60,61]. In Aspergillus niger, fumonisin production was increased under blue and red light, while ochratoxin levels were reduced compared to dark incubation [62]. In addition, the wavelength of light also impacts fungal secondary metabolite production. In A. nidulans, the red light receptor FphA has been reported to suppress mycotoxin biosynthesis, whereas the blue light sensors LreA and LreB had a stimulatory effect [20]. Our results on T. atroviride revealed the highest 6-PP levels upon growth under dark conditions in both strains tested, while light, with only a minor influence of the wavelength, negatively affected 6-PP production. In A. nidulans and other filamentous ascomycetes, the velvet protein complex, consisting of VeA, VelB and LaeA, acts as a light-dependent key regulator of development and secondary metabolism. VeA is mainly cytoplasmic in the presence of light while it is imported into the nucleus under dark conditions, where the fully functional velvet complex acts as an activator of secondary metabolism-related genes Preprints (www.preprints.org) | NOT PEER-REVIEWED | Posted: 11 September 2020 doi:10.20944/preprints202009.0251.v1 [21]. Our findings that 6-PP production by T. atroviride mainly occurs in the dark and is repressed by light suggests regulation by LaeA/Lae1, the global regulator of fungal secondary metabolite gene clusters [63], and the velvet complex.
In Trichoderma virens, another potent mycoparasitic Trichoderma species (which, however, is unable to produce 6-PP), the velvet protein Vel1 has been identified as a key regulator of biocontrol. Vel1 mutants were defective in secondary metabolism, mycoparasitism and biocontrol efficacy [64]. In our study, mycoparasitic overgrowth of F. oxysporum by both T. atroviride wild type strains, IMI206040 and P1, was enhanced under dark conditions as well as in the presence of yellow and red light compared to white light-dark cycles and blue light illumination. In contrast, the mycoparasitic overgrowth of R. solani was largely light-independent. 6-PP has previously been shown to exhibit antifungal activity [2]. Our results from the plate confrontation assays however suggest that 6-PP only plays a minor role in the mycoparasitic interaction. On the other hand, the difficulties of T. atroviride to overgrow F. oxysporum in the presence of white, blue and green light might also be due to enhanced production of secondary metabolites by the host fungus under these conditions that inhibit Trichoderma growth or protect F. oxysporum against T. atroviride attack. Accordingly, it has been shown that light increases fumonisin biosynthesis in Fusarium spp. [65] and that blue light triggers red pigment content of F. oxysporum [66].
In conclusion, this study has shown that T. atroviride is able to sense and respond to different light regimes which impacted differentiation, mycoparasitic activity and production of the secondary metabolite 6-pentyl-α-pyrone. In addition, the two T. atroviride strains tested, IMI206040 and P1, differed in their light responses, which, however, is not due to differences in their major photosensory proteins. Alignment of Blr-1 and Blr-2, phytochrome, and opsin protein sequences revealed a 100% identity between the respective proteins encoded in the two T. atroviride strains (Supplementary Figures. S1 -S3). The same light sensory proteins hence seem to govern distinct, strain-specific responses.
Supplementary Materials: The following are available online at www.mdpi.com/xxx/s1, Table S1: Primers used for generation of tmk3 gene deletion cassette and genotyping, Figure S1: Percentage of sequence identity of fungal light sensors, Figure S2: Sequence alignment of fungal phytochromes and White Collar-like proteins, Figure S3: Sequence alignment of fungal opsins.
Author Contributions: DM and SZ conceived and directed this study and drafted the manuscript. DM and AF performed the experiments. KM and RS contributed to 6-PP analysis. All authors read, revised and approved the manuscript. All authors have read and agreed to the published version of the manuscript.
Funding: This research was supported by the Austrian Science Fund (FWF; grant P32179-B) and the doctoral program BioApp from the University of Innsbruck.