Sfwc-1 is essential for the phototropism of perithecial beak

It was previously shown that N. crassa wc-1 mutant stains showed the defect of phototrophic response of perithecial beak (Harding and Melles, 1983). To investigate the phenotypes of the Sfwc-1^(∆lov) mutant in sexual development, S. fimicola strains were grown on malt extract agar III under unilateral constant illumination and a 12-h light-dark photoperiod. Under the 12L:12D regimes, the wild-type but not Sfwc-1^(∆lov) mutant strain showed perithecial zonation and complementation of the Sfwc-1 gene into the Sfwc-1^(∆lov) strain restored the zonation pattern similar to that seen in the wild type (Fig.

6A-C). Furthermore, positive phototropism of perithecial beaks in the Sfwc-1^(∆lov) mutant was lost under unidirectional illumination (Fig. 6B). After 21 days under the same conditions, more than 76% of perithecial beaks in the wild-type showed a positive orientation toward the light source (Fig. 6A; Table 4). Perithecial beaks in Sfwc-1^(∆lov) were shorter than those of the wild type, with less than 5% positively phototropic orientation under same illumination conditions (Fig. 6B; Table 4). The beak length and phototropism of perithecia were phenotypically restored to the wild-type features in the complementation strains (Fig. 6C; Table 4). These results confirm that the Sfwc-1 gene affected the formation and phototropic behavior of the sexual fruiting body in S.

fimicola, confirming that peritheical zonation and phototropism of perithecial beak were governed by Sfwc-1 via light dependent regulation.

3.1.5 Disruption of the Sfwc-1gene reduces protoperithecial formation

To determine the effect of photoreceptor disruption on self-fertility in S. fimicola, strains were grown on 6-cm malt extract agar III plates under uniformly constant illumination and darkness. After 7 days, the wild type and the complementation strains

produced more perithecia under illumination than in darkness. In contrast, the Sfwc-1^(∆lov) mutant showed greatly decreased number of fruiting bodies under illumination condition and was a similar amount of perithecia in the darkness (Fig. 7A). To better evaluate the difference, we quantified the number of fruiting bodies formed on agar plates by manually counting under a dissection microscope. After a 7-day incubation under constant illumination, the number of fruiting bodies in the Sfwc-1^(∆lov) mutantwas reduced by up to 213- and 207-fold relative to the wild type and complementation strain, respectively. Under darkness conditions, the wild type and complementation strain produced substantially less perithecia, with a slight reduction (< 2-fold) observed in the Sfwc-1^(∆lov) mutant(Fig. 7B). These results indicate that fruiting-body formation in S.

fimicola was photo-induced and the Sfwc-1 gene played an important role in this process.

To more precisely determine the developmental defects of the Sfwc-1^(∆lov) mutant, we observed sexual structures of the strains over 14 days. Under the same conditions described previously, early-differentiated sexual structures including ascogonial coils were observed in all tested strains after 24-h incubation; however, the Sfwc-1^(∆lov) strain seemed to produce much fewer ascogonial coil structures as compared with the other two strains (Fig. 8A-C). The wild type and complementation strain showed incipient protoperithecia or protoperithecium–perithecium transition structures at about 72 h post-incubation and these structures further differentiated into small immature perithecia up to 200 µm in width after 5 days incubation. In contrast, the Sfwc-1^(∆lov) mutantshowed developmental defects at these stages. Protoperithecia were barely seen, and instead, clumps of hyphae or tiny protoperithecia were observed after 3 days incubation (Fig.

8B). Many of these structures were delayed or unable to further develop into perithecia, which resulted in a structure of 50 µm in width after 5 days incubation. After 7 days

incubation, the wild type and complementation strain formed black flask-shaped perithecia containing abundant asci with an eight-ascospore linear arrangement under illumination conditions; however, development in the darkness was delayed and immature perithecia were observed at the same period (Fig. 8A and C). In contrast, the Sfwc-1^(∆lov) mutant at the same stage showed only a few immature perithecia with asci in various stages of development under both illumination and darkness conditions.

Moreover, after 2 weeks incubation, the wild type and complementation strain showed elongated beaks and mature ascospore discharge, whereas the Sfwc-1^(∆lov) mutant mostly formed immature perithecia. The few mature perithecia produced by the mutant were further examined for the rosette of asci, and mature asci with black eight ascospores were observed in limited amounts (Fig. 8B). Thus, disruption of the Sfwc-1 gene significantly compromised the sexual and fruiting body development in S. fimicola, which indicates that Sfwc-1 played important roles in the light-regulated sexual development.

Among early light-induced transcription factors (TFs) in Neurospora, sub-1, a GATA type family TF gene, is essential for WCC-dependent light induction of numerous late light-inducible genes (Chen et al., 2009; Sancar et al., 2015; Wu et al., 2014). The sub-1 mutants exhibited defects in protoperithecia formation and altered expression of numerous genes, which indicates that SUB1 acts as a global regulator of light-induced transcription (Chen et al., 2009; Colot et al., 2006; Sancar et al., 2015). To correlate the phenotypic data with the gene potentially involved in the sexual development cascade, we determined S. fimicola sub-1 (Sfsub-1) mRNA levels in the wild type and Sfwc-1^(∆lov) mutant under different conditions of illumination (Fig. 9). The mRNA level of Sfsub-1 was photo-induced in the wild type and showed the highest expression with

60-min illumination. However, Sfsub-1 mRNA level in light-treated samples was similar to that under the darkness condition in the Sfwc-1^(∆lov) mutant (Fig. 9). Therefore, Sfsub-1 was light-induced in a Sfwc-1 dependent manner. These results indicate that SfWC-1 was required for the light-activated sexual developmental process by activating other regulatory genes such as Sfsub-1, which suggests that a SfWC-1–mediated light-responsive network was important for proper development.

3.1.6 Light-induced carotenoid pigmentation via Sfwc-1 gene regulation

Wild-type mycelia of S. fimicola exhibited an orange pigment after light induction as previously described (Ingold and Hadland, 1959). Phenotypically, carotenoid pigment was light-induced in the wild type and complementation strain but was unaltered in the Sfwc-1^(∆lov) mutant, even with continued illumination for 6 h (Fig.

2B). The Sfwc-1 photoreceptor gene might be required to regulate genes involved in the carotenoid biosynthesis pathway. In Neurospora, albino genes (al-1, al-2, and al-3) encoding enzymes responsible for carotenoid biosynthesis are light-induced in hyphae and regulated by WCC (Belozerskaya et al., 2012; Iigusa et al., 2005). To better understand the role of Sfwc-1 in the carotenoid biosynthesis pathway, we analyzed the mRNA expression profiles of putative genes involved in carotenogenesis, Sfal-1 (SF_NCU00552), Sfal-2 (SF_NCU00585), Sfal-3 (SF_NCU01427), and Sfcao-2 (SF_NCU11424) in the wild type and Sfwc-1^(∆lov) mutant under light induction. The mRNA levels of Sfal-1 and Sfal-2 peaked 30 min after exposure to blue light and decreased after longer exposure to blue light (Fig. 10). Moreover, Sfal-3 encodes for the enzyme responsible for the synthesis of geranylgeranyl diphosphate, a precursor molecule of several biochemical pathways involving the carotenoid biosynthesis pathway (Iigusa et al., 2005; Sandmann et al., 1993). The mRNA level of Sfal-3 peaked

at 15 min and decreased thereafter. However, the expression of Sfcao-2, probably required for a final step of the carotenoid biosynthesis pathway, gradually increased and peaked at 60 min, and decreased after long-term exposure to blue light for 120 min (Fig.

10). In contrast, the mRNA levels of these genes in the Sfwc-1^(∆lov) mutant showed no significant changes with light induction and were much lower in the mutant than the wild type and similar to that in constant darkness. These findings agreed with the light-induced carotenoid pigmentation of mycelia found in the wild type and undetectable accumulation of carotenoid pigment in Sfwc-1^(∆lov) mycelia (Fig. 2). Hence, carotenoid biosynthesis genes in S. fimicola were light-induced via Sfwc-1 regulation, similar to that found in N. crassa (Belozerskaya et al., 2012; Lee et al., 2003).

Besides the effect on carotenoid pigmentation, light also affects pigment production in sexual reproductive structures via melanin biosynthesis pathways. Several genes and enzymes required for melanization of sexual structure have been reported (Chinnici et al., 2014; Wang et al., 2014). Based on initial screening of transcriptome results, we identified the genes encoding enzymes involved in the melanin biosynthesis, including scytalone dehydratase (SF_NCU07823), Sfper-1 (per-1) (polyketide synthase;

SF_NCU03584), tetrahydroxynaphthalene reductase (SF_NCU06905/ SF_NCU09390), and T gene (Tyrosinase; SF_NCU00776) (Krobanan et al., 2019). Their transcript levels were downregulated under illumination relative to darkness in the wild type whereas subtle changes were observed in the Sfwc-1^(∆lov) mutant, indicating that genes involved in melanin production may be regulated by light via Sfwc-1 (Fig. 11, Krobanan et al., 2019). Moreover, a putative polyketide synthase-like gene was found in S. fimicola and it was homologous to N. crassa polyketide synthase-4 (pks-4; NCU08399). We also examined the transcript level of this Sfpks gene in response to illumination. Sfpks

transcript remained at a basal level in constant darkness for both wild type and mutant strains. Interestingly, Sfpks expression was down-regulated under illumination conditions both in the wild type and mutant strains, but the Sfwc-1^(∆lov) mutant displayed a different expression pattern. The role and regulation of this Sfpks gene in S. fimicola requires further investigation (Fig. 9).

3.1.7 Genome-wide transcriptional responses to light in S. fimicola

To elucidate the transcriptional regulatory networks in response to light in S.

fimicola, we analyzed blue-light-induced genome-wide transcriptional responses by high-throughput RNA-seq to reveal the transcriptional profiles between the wild type and Sfwc-1^(∆lov) mutant. We aimed to use the data from this approach to identify differentially expressed genes for RT-qPCR verification and future studies. Untreated mycelia (constant darkness sample) and mycelia treated with blue light for 15 and 45 min were labeled WTDD, WT15, and WT45, respectively, for the wild type, and ADD, A15, and A45 for the Sfwc-1^(∆lov) mutant. The N. crassa database was used as reference for predicting putative genes in S. fimicola.

Genome-wide expression profiling of Sordaria wild type and the Sfwc-1^(∆lov) mutant was performed with the Illumina sequencing platform and analyzed with the Trinity program. A total of 62,274 transcript isoforms were de novo assembled and annotated via Blast search. The estimated abundance of these transcripts were quantified at each time point and these was categorized as “expressed genes” under these growth conditions. Using fold change ≥2, FDR p-value <0.005 as the cut-offs for significance, we identified 874 genes differentially expressed in at least one pairwise comparison under illumination or darkness conditions in either S. fimicola wild type or Sfwc-1^(∆lov) mutant. The putative genes were described as SF_ (N. crassa accession number).

Among photoreceptor genes, cytochrome (Sfcryp; SF_NCU00582) and vivid (Sfvivid:

SF_NCU03967) genes were differentially expressed in the wild type upon light induction, and the expression was unaltered in the Sfwc-1^(∆lov) mutant under the same condition (Krobanan et al., 2019). The expression levels of Sfcryp and Sfvivid were confirmed by RT-qPCR analysis and both genes were induced by light via SfWC-1 regulation (Fig. 9). From Venn diagram analysis, 466 putative genes showed differential expression in the wild type, but not in Sfwc-1^(∆lov) mutant, after exposure to light for 15 and 45 min relative to constant darkness (DD) (Fig. 12A; Krobanan et al., 2019).

Moreover, 590 putative genes showed differential regulation between the wild type and Sfwc-1^(∆lov) mutant at each time point (constant darkness and after illumination by blue light for 15 and 45 min) (Fig. 12B; Table S6). In addition to 466 putative genes, we also found that some genes showed differential expression after 15 min relative to 45 min blue light induction. Therefore, a total of 478 putative genes showed significantly differential expression at least two-fold for at least one time point in response to light in the wild type, but not the Sfwc-1^(∆lov) mutant (Krobanan et al., 2019). The differential expression seen in the Sfwc-1^(∆lov) mutant after blue light induction indicates that additional blue light photoreceptor may exist and function. Genes differentially expressed in at least one time point in response to blue light in the wild type but not Sfwc-1^(∆lov) were also subjected to GO analysis for functional annotation to reveal their potential functions. Significantly, enriched GO terms (p ≤ 0.05) were oxidation-reduction processes, oxidoreductase activity, heme binding and single-organism metabolic processes (Fig. 13A, (Krobanan et al., 2019). To better understand the potential pathways regulated by Sfwc-1, the same set of putative genes were further assessed by catagorizing functions using FunCat. The enrichment of specific metabolic

pathways influenced by Sfwc-1 included metabolism of melanin, metabolism involving polyketides, ABC transporters, rhythm (e.g., circadian), metabolism of vitamins, cofactors, and prosthetic groups, lipid, fatty acid and isoprenoid metabolism, detoxification, secondary metabolism and other pathways (Fig 13B; (Krobanan et al., 2019).

Based on the changes in expression (estimated abundance) of 478 putative genes, 306 genes were upregulated and 172 genes were downregulated in the wild type, but not in Sfwc-1^(∆lov) strain, relative to darkness and illumination conditions. Genes which showed differential expression in response to light were further grouped into 4 clusters (Krobanan et al., 2019). Patterns of changes in gene expression in each group were upslope, hill, downslope and valley for cluster 1 to 4 respectively.

Cluster 1 included genes that were typically induced by light in a Sfwc-1 dependent manner and were further subdivided into cluster 1a and 1b based on different levels of estimated abundance. Each sub-cluster showing higher level of RNA at 45 min than at 15 min was considered as main cluster 1, which implied that this cluster may be involved in attenuating light responses in S. fimicola. Cluster 1a included the genes involved in the development of sexual spores such as the constitutively photomorphogenic 9 (COP9) signalosome (SF_NCU08342) and bromodomain protein-1 (SF_NCU08809). The cluster protein-1b was enriched for genes involved in biosynthesis of vitamins, cofactors, and prosthetic groups, secondary metabolism and C-compound and carbohydrate metabolism. This cluster also included genes required for photoperception and response, the well-studied photoadaptation gene Sfvvd (SF_NCU03967) and genes involved in homeostasis and metal ions transporters, such as ferrochelatase (SF_NCU08291), iron-sulfur clusters transporter atm-1 (SF_ NCU05029), oxidative

stress resistance (SF_NCU03145) and peptidyl-prolyl cis-trans isomerase, fkr-3 (SF_NCU04371). In addition, some reported genes directly targeted by WCC including light responsive transcription factors, such as Sffl (fluffy SF_NCU08726), Sfmig-12 TF (SF_NCU09830), and hypothetical protein (SF_NCU01871) and other light responsive genes such as Sfcryp (SF_NCU00582) and UV-endonuclease UVE-1 (SF_NCU08850) were also found in this cluster (Smith et al., 2010; Verma and Idnurm, 2013).

Cluster 2 represented genes that were rapidly induced by light and peaked in expression between 15 min and 45 min in the light, so called early light-induced genes.

Cluster 2 was significantly enriched for genes encoding enzymes for carotenoid biosynthesis including phytoene desaturase (Sfal-1), phytoene synthase (Sfal-2), geranylgeranyl pyrophosphate synthetase (Sfal-3). Some light-responsive transcriptional factor genes including Sfbeak-1 (SF_NCU00097), Sfwc-1 (SF_NCU02356), Sfcsp-1 (SF_NCU02713), hypothetical protein (SF_NCU00275) and other light responsive genes such as Sfcon-6 (conidiation-specific protein 6; SF_NCU08769), Sffrq gene (frq;

SF_NCU02265), Sfbli-4 (SF_NCU08699) were also included in this cluster.

Genes in cluster 3 were repressed by light. Cluster 3 was mainly enriched for genes involved in secondary metabolism, lipid, fatty acid and isoprenoid metabolism and C-compound and carbohydrate metabolism. Interestingly, this cluster also included polyketide synthase-like family genes, SF_NCU04865, SF_NCU02918 and SF_NCU08399, and the genes encoding enzymes involved in the melanin metabolism such as tetrahydroxynaphthalene reductase (SF_NCU06905/SF_ NCU09390), and T gene (Tyrosinase; SF_NCU00776) (Krobanan et al., 2019).

Genes in cluster 4 were repressed in the light by 15 min but their expression was returned to approximately darkness level by 45 min. Although cluster 4 was highly

enriched for genes involved in secondary metabolism and C-compound and carbohydrate metabolism similar to cluster 3, genes related to lipid, fatty acid and isoprenoid metabolism were not found in this cluster. Genes involved in melanin metabolism such as Aspergillus yellowish-green 1 (SF_NCU05821) and scytalone dehydratase (SF_NCU07823), Sfper-1 (per-1) (polyketide synthase; SF_NCU03584) and clock-controlled protein 9 (SF_NCU09559) were also included in this cluster. To investigate and verify light-responsive expression profiles, the wild type and Sfwc-1^(∆lov) mutant were grown for 2 days in constant darkness (DD) and white light (LL), and an additional set of DD samples were exposed to blue light for 15, 30, 45, 60, 120 and 180 min (Fig. 9). The transcript levels of light-responsive genes including Sfbeak-1, Sffrq, and Sfcryp peaked at 15 min and decreased thereafter. However, the light-induced mRNA level of light-responsive genes showed subtle expression changes in the Sfwc-1^(∆lov) mutant (Fig. 9). Because of the known role of wc-1 in the phototropism of perithecial beaks in N. crassa, Neurospora beak-1 is photo-regulated via a WCC mechanism. Lack of bek-1 confers defective perithecial beak development in Neurospora (Colot et al., 2006). Similarly, Sfbeak-1 transcripts were photo-induced in the wild type, whereas sustained low mRNA levels were found in the Sfwc-1^(∆lov) mutant.

In contrast, Sfbeak-1 expression in Sfwc-1^(∆lov) was not affected by light (Fig. 9). This result was consistent with the phenotype of the mutant, with the Sfwc-1^(∆lov) mutant showing a shorter perithecial neck than the wild type. These findings suggest that Sfbeak-1 was regulated by light via Sfwc-1 and may be involved in perithecial beak initiation and elongation in S. fimicola.

3.1.8 Generation of SfWC-1-GFP and nuclear localization Generation of the nuclear-labeled strains

To examine nuclear dynamics in S. fimicola, we aimed to generate fluorescent, nuclear-labeled strains to record live images of the nuclear distribution and behavior during different stages of the life cycle. Since histones are essential structural proteins for chromatin assembly and play functional roles in all living cell types, we used the fluorescent mCherry protein to tag H2B histone to monitor nuclear distribution in the ascospore germlings, vegetative hyphae, and initial fusing and other sexual structures.

As shown in Fig. 14 we successfully generated the histone H2B-labeled strains using Agrobacterium-mediated protoplast transformation. Observation of the hygromycin B-resistant transformants under the fluorescent microscope revealed that five transformants clearly displayed strong mCherry signals. These transformants were further verified by PCR to detect the presence of the PR27p-H2B-mCherry fragment.

Indeed, 974-bp and 1.1-kb fragments were amplified from these transformants with the primer pairs WC1218/WC2904 and WC2903/WC602, respectively, whereas no PCR product was obtained from the wild type (Fig. 14B, left and middle). Amplification of gDNA with the universal ITS1/ITS4 primer pair as a control resulted in a 584-bp product in both the wild-type and transformed strains (Fig 14B right). Based on these results, the transformant WTH2B/4 was chosen for further investigation of nuclear dynamics.

To examine the nuclear dynamics in different sexual structures, ascospores of S.

fimicola strains were first collected and inoculated onto malt extract agar III.

Approximately 1–2 h post-inoculation, ascospores of the H2B-mCherry-labeled strain formed a germination pore at one end of the spore, followed by the formation of the

germination vesicle, which allowed the nuclei to enter the vesicle (Fig. 15). At 6 h post-inoculation, various inclusions were visible in this germling, including a single large or multiple medium-sized lipid droplet vacuoles in the center of the germination vesicle.

These structures likely reflect vacuole inclusions that might help push nuclei toward the germination vesicle and the peripheral growth zone of a fungal colony and/or serve as nutrient reservoirs during establishment on solid agar. However, at 15 h post-inoculation, the hyphal tips were devoid of nuclei (Figs. 15, 16). As sexual development further proceeded, ascogonial coils, likely formed as side branches of vegetative hyphae ca. 8–10 μm wide, were observed at 27 h and 48 h post-inoculation (Fig. 15). At 48 h post-inoculation, in addition to ascogonial coil formation, multinucleate cellular compartments resulting from differentiated ascogenous hyphae were observed in this culture and defined as early protoperithecia (Fig. 15). These structures were ca. 30 μm in width and showed the packing of various hyphae. In fact, the protoperithecia developed from the ascogonium were formed by the aggregation of enveloping hyphae that emerged from either the ascogonial coil or neighboring hyphae. Through the growth of the protoperithecium accompanied by septation and tissue differentiation, the dark-pigmented mature fruiting body (perithecium) with its neck was observed at approximately 192 h post-inoculation, with a size of ca. 250–300 μm wide. Moreover, an ascus rosette containing asci of various stages of maturity derived from the perithecium of the mCherry-labeled strain was observed. The black, mature ascospore contained multiple nuclei derived from repetitive mitotic division, as shown in Figure 15 (192 h post-inoculation).

SfWC-1-GFP and nuclear localization

N. crassa WC-1 plays an important role for activating frq in the darkness, whereas light-activated WCC binds to two light-response elements in the promoter region of light-responsive genes (Cheng et al., 2003b). Previously, the localization of WC-1 in the nucleus does not require light stimuli in N. crassa (Schwerdtfeger and Linden, 2000). Thus, we generated the SfWC-1-GFP-labeled strains to investigate cellular localization of SfWC-1 under both darkness and illumination conditions in S.

N. crassa WC-1 plays an important role for activating frq in the darkness, whereas light-activated WCC binds to two light-response elements in the promoter region of light-responsive genes (Cheng et al., 2003b). Previously, the localization of WC-1 in the nucleus does not require light stimuli in N. crassa (Schwerdtfeger and Linden, 2000). Thus, we generated the SfWC-1-GFP-labeled strains to investigate cellular localization of SfWC-1 under both darkness and illumination conditions in S.