The Two Cryptochrome/Photolyase Family Proteins Fulfill Distinct Roles in DNA Photorepair and Regulation of Conidiation in the Gray Mold Fungus Botrytis cinerea

ABSTRACT

The plant-pathogenic leotiomycete Botrytis cinerea is known for the strict regulation of its asexual differentiation programs by environmental light conditions. Sclerotia are formed in constant darkness; black/near-UV (NUV) light induces conidiation; and blue light represses both differentiation programs. Sensing of black/NUV light is attributed to proteins of the cryptochrome/photolyase family (CPF). To elucidate the molecular basis of the photoinduction of conidiation, we functionally characterized the two CPF proteins encoded in the genome of B. cinerea as putative positive-acting components. B. cinerea CRY1 (BcCRY1), a cyclobutane pyrimidine dimer (CPD) photolyase, acts as the major enzyme of light-driven DNA repair (photoreactivation) and has no obvious role in signaling. In contrast, BcCRY2, belonging to the cry-DASH proteins, is dispensable for photorepair but performs regulatory functions by repressing conidiation in white and especially black/NUV light. The transcription of bccry1 and bccry2 is induced by light in a White Collar complex (WCC)-dependent manner, but neither light nor the WCC is essential for the repression of conidiation through BcCRY2 when bccry2 is constitutively expressed. Further, BcCRY2 affects the transcript levels of both WCC-induced and WCC-repressed genes, suggesting a signaling function downstream of the WCC. Since both CPF proteins are dispensable for photoinduction by black/NUV light, the origin of this effect remains elusive and may be connected to a yet unknown UV-light-responsive system.

IMPORTANCE Botrytis cinerea is an economically important plant pathogen that causes gray mold diseases in a wide variety of plant species, including high-value crops and ornamental flowers. The spread of disease in the field relies on the formation of conidia, a process that is regulated by different light qualities. While this feature has been known for a long time, we are just starting to understand the underlying molecular mechanisms. Conidiation in B. cinerea is induced by black/near-UV light, whose sensing is attributed to the action of cryptochrome/photolyase family (CPF) proteins. Here we report on the distinct functions of two CPF proteins in the photoresponse of B. cinerea. While BcCRY1 acts as the major photolyase in photoprotection, BcCRY2 acts as a cryptochrome with a signaling function in regulating photomorphogenesis (repression of conidiation).

INTRODUCTION

Sensing light is an important ability in organisms from all kingdoms, allowing for adaptation to ongoing and upcoming environmental changes. Filamentous fungi use light as a signal to regulate, for instance, secondary metabolism, morphogenesis, tropism, and the entrainment of circadian clocks. The perception of light of specific wavelengths is mediated by photoreceptors, specialized proteins that bind light-absorbing molecules (chromophores). Fungi may comprise variable numbers of red/far-red-light-sensing phytochromes, green-light-sensing opsins, blue/near-UV (NUV) light-sensing proteins from the cryptochrome/photolyase family (CPF), and blue-light-sensing LOV (light, oxygen, voltage) domain proteins (1–3).

To date, the best-known LOV domain protein is the GATA-type transcription factor (TF) WC-1 from Neurospora crassa, which, together with the second GATA-type TF, WC-2, forms the White Collar complex (WCC). Since their discovery in N. crassa (4, 5), orthologs of the two TFs have been shown to play a major role in blue-light perception in many filamentous fungi by mediating a wide variety of photoresponses (6–13). For instance, the WCC is described as a mediator of UV tolerance through its regulation of genes encoding DNA repair enzymes (13–18).

Irradiation of DNA with UV light causes the formation of cyclobutane pyrimidine dimers (CPD) and (6-4)-pyrimidine-pyrimidone photoproducts (6-4PP), which are recognized by photolyases. Photolyases are DNA repair enzymes that utilize light as an energy source for the repair of UV-damaged DNA (photoreactivation). To achieve this, they noncovalently bind flavin adenine dinucleotide (FAD) and often an additional chromophore that acts as an antenna pigment, such as methylenetetrahydrofolate (MTHF). Upon photoactivation of the antenna molecule, the energy is transferred to the fully reduced FAD (FADH⁻), which leads to an electron transfer to the pyrimidine dimer and causes a splitting into monomers (19). Cryptochromes are photoreceptors that are closely related to photolyases, but they do not necessarily exhibit DNA repair functionality and may possess regulatory functions (20).

Three main classes of CPF proteins are known in fungi: CPD photolyases, 6-4PP photolyases, and Drosophila, Arabidopsis, Synechocystis, Homo cryptochromes (cry-DASH). Interestingly, regulatory functions have been reported for members of all three classes. Thus, the 6-4PP photolyase PHL1 from Cercospora zeae-maydis acts as a regulator of cercosporin biosynthesis and a repressor of conidiation in the dark (21). Members of the CPD photolyases, such as Trichoderma atroviride PHR1 and Aspergillus nidulans CryA, possess regulatory functions in the form of negative autoregulation (14) and repression of sexual development (22), respectively. N. crassa CRY, which belongs to the cry-DASH family, modulates the transcription of some light-induced genes (23) and is required for the entrainment of a WCC-independent circadian oscillator system (24). Another cry-DASH (CryD) regulates the formation of macroconidia as well as secondary metabolism in Fusarium fujikuroi (25). In contrast, T. atroviride cry-DASH does not have obvious regulatory functions (15), and Sclerotinia sclerotiorum CRY1 (SsCRY1) has only a minor effect on sclerotial and apothecial development (26). Overall, CPF proteins show great variability in their regulatory functions, making an experimental approach indispensable to the elucidation of their roles in different fungi.

The plant-pathogenic leotiomycete Botrytis cinerea, also known as gray mold fungus, shows a wide variety of photoresponses, including tropisms of germ tubes and reproductive structures (27–29), regulation of secondary metabolism (30, 31), entrainment of the circadian clock (32), and strict regulation of morphogenesis. When this fungus is exposed to light, conidia are produced for dispersal, whereas sclerotia, strongly melanized enduring structures, are formed in constant darkness (DD) (33). However, the formation of conidia is not dependent only on the presence of light per se but is differentially affected by at least four different light qualities: NUV, blue, red, and far-red light (34). In this context, the antagonistic functions of blue and NUV light are of special interest: exposure to blue light inhibits conidiation and can even cause the dedifferentiation of developing conidial or sclerotial initials, while subsequent exposure of blue-light-treated conidiophores or conidial initials to NUV light can repromote conidiation (35–37).

B. cinerea possesses four putative LOV domain-containing photoreceptors to sense blue light, including the small LOV protein B. cinerea VVD1 (BcVVD1) and the GATA-type TF BcWCL1 (38). BcWCL1 interacts with BcWCL2, resulting in the formation of the WCC (39), which acts as a negative regulator of conidiation. Loss of BcWCL1 causes light-independent formation of conidia (the “always conidia” phenotype) through the deregulation of bcltf2, encoding a C₂H₂ TF as a positive regulator of conidiation (40, 41). Further, the WCC regulates virulence on Arabidopsis thaliana by allowing a time-of-day adaptation of the infection process through entrainment of a BcFRQ1-based circadian clock (32).

Here we report on the distinct functions of two CPF members from B. cinerea. The CPD photolyase BcCRY1 is essential for light-driven repair of UV-damaged DNA (photoreactivation). In contrast, the cry-DASH protein BcCRY2 is dispensable for photorepair but acts as a negative regulator of photoinduced conidiation.

RESULTS

B. cinerea possesses two putative cryptochrome/photolyase family (CPF) proteins.

B. cinerea is known for its ability to regulate conidiation in dependence on at least four light qualities. Of special interest are the partially overlapping light qualities NUV light and black light blue (BLB), which act as the main inducers of conidiation (34, 42). As blue/NUV-light-sensing photoreceptors, CPF proteins were regarded as candidates for inducers of conidiation in B. cinerea. Two putative CPF proteins, BcCRY1 (EnsemblFungi B. cinerea B05.10 database protein Bcin05g08060) (607 amino acids [aa]) and BcCRY2 (Bcin09g01620) (644 aa), have been identified previously by BLASTP analyses (38). Putative orthologs of the CPF proteins in other fungi were identified using the BLASTP algorithm. Multiple-sequence alignments and phylogenetic analyses revealed that BcCRY1 clusters with CPD photolyases, whereas BcCRY2 groups with cry-DASH proteins (Fig. 1A). Prediction of conserved protein domains revealed similarities, but also differences, between the two CPF proteins (Fig. 1B). BcCRY1 comprises an N-terminal DNA photolyase domain (Pfam accession no. PF00875) (E value, 4.4e−37) and a C-terminal FAD-binding domain (PF03441) (3.7e−80). Further sequence analyses predicted a nuclear localization signal (¹⁹²KRKR¹⁹⁵) in the DNA photolyase domain. BcCRY2 also features DNA photolyase (PF00875) (1.6e−20) and FAD-binding (PF03441) (3.1e−43) domains but contains an additional C-terminal R/G-rich region that is often associated with the ability to interact with RNA (23, 43–45).

FIG 1.

B. cinerea possesses two putative cryptochrome/photolyases with distinct subcellular localizations. (A) Phylogenetic analyses identify them as members of CPD photolyase and cry-DASH families. The B. cinerea BcCRY1 (EnsemblFungi B. cinerea database ID Bcin05g08060) and BcCRY2 (Bcin09g01620) sequences were obtained from EnsemblFungi. The following sequences were retrieved from UniProt: Aspergillus nidulans (Anid) CryA (UniProt accession no. Q5BGE3), Cercospora zeae-maydis (Czea) PHL1 (B2CG58), Fusarium fujikuroi (Ffuj) CryD (H6SG34, S0DK59, S0E065), Neurospora crassa (Ncra) CRY (NCBI protein database accession no. Q7SI68) and PHR (P27526), Saccharomyces cerevisiae (Scer) PHR1 (NCBI protein database accession no. P05066), S. sclerotiorum CRY1 (SsCRY1) (UniProt accession no. A7EIM0) and SsCRY2 (A7F2C9), Trichoderma atroviride (Tatr) PHR1 (G9PA82), CRY1 (G9NYX9), and cry-DASH (G9P3D7), and Trichoderma reesei (Tree) CRY1 (G0RI60, G0RGE7, G0RKR6). Protein alignments were conducted using the ClustalOmega algorithm at EBI. CPD, cyclobutane pyrimidine dimer; (6-4)PP, (6-4)-pyrimidine-pyrimidone photoproduct; cry-DASH, Drosophila, Arabidopsis, Synechocystis, Homo cryptochromes. (B) BcCRY1 and BcCRY2 display similar domain organizations. Both proteins contain N-terminal DNA photolyase (NCBI protein database accession no. IPR006050) and C-terminal FAD-binding domains (IPR005101). BcCRY2 additionally possesses a C-terminal extension containing an R/G-rich region. A nuclear localization signal (¹⁹²KRKR¹⁹⁵) (NLS) was predicted in BcCRY1. (C) BcCRY1 and BcCRY2 are located in the nuclei and the cytosol, respectively. Subcellular localization was visualized by expression of GFP fusion proteins under the control of the constitutive oliC promoter from A. nidulans. The functionality of the GFP fusion proteins is shown in Fig. S3 in the supplemental material. Nuclei were visualized by staining with Hoechst 33342.

Transcription of both CPF protein-encoding genes is induced by light in a WCC-dependent manner.

The WCC mediates the transcription of CPF protein-encoding genes upon light exposure in several fungi, including N. crassa and T. atroviride (14, 23). In accordance, bccry1 and bccry2 have been described as light-induced genes (30, 40). Besides the WCC, BcVEL1 and BcLAE1, as members of the VELVET complex (46, 47), and the GATA-type TF BcLTF1 (30) are involved in the transcriptional responses of B. cinerea to light. To expand knowledge of the light-dependent transcriptional regulation of bccry1 and bccry2, reverse transcription-quantitative PCR (RT-qPCR) analyses were performed using bcwcl1, bcltf1, bcvel1, and bclae1 deletion mutants (Fig. 2). The strains were cultivated on solid complete medium (CM) in constant light (LL) or constant darkness (DD) for 2 days. Dark-grown cultures were then either harvested immediately or exposed to white-light pulses (LP) for 15, 60, or 180 min. Analyses of the transcript levels revealed that the light induction of both bccry1 (10.2-fold over DD levels with a 60-min LP) and bccry2 (52.2-fold over DD levels with a 60-min LP) occurs in a WCC-dependent manner, although low residual light-dependent transcription of bccry1 was observed in the Δbcwcl1 mutant (1.7-fold over DD levels with a 60-min LP). bcvel1 and bclae1 deletion mutants exhibited reduced transcript levels of bccry1 (3.6-fold reduction from the wild-type [WT] level with a 15-min LP in both mutants). In contrast, transcript levels of bccry2 in response to light were elevated in the Δbcvel1 mutant (1.9-fold over the WT level with a 60-min LP) and were not affected in the Δbclae1 mutant. Further, deletion of bcltf1 led to increased transcript levels of bccry2 (1.5-fold over the WT level with a 60-min LP) but had no effect on the transcription of bccry1. Thus, while the light responsiveness of both genes depends strictly on the WCC, they are not entirely coregulated, since their transcript levels are differently modulated by the key components of the light response system.

FIG 2.

Transcription of bccry1 and bccry2 is induced by light in a WCC-dependent manner. Light-dependent transcription of bccry1 and bccry2 was investigated in relation to components of the WCC (BcWCL1), the VELVET complex (BcVEL1, BcLAE1), and the light-responsive GATA-TF BcLTF1. The indicated strains were cultivated on solid CM with cellophane layers for 2 days in constant darkness (DD) or constant white light (LL). Cultures grown in DD were either harvested immediately (DD) or exposed to light pulses (LP) for 15 to 180 min. Total RNA from three biological replicates was used for cDNA synthesis and subsequent RT-qPCR analyses in two technical replicates. Shown are the transcript levels of the genes of interest (GOI) relative to those of bcact1 as a reference gene. Significant differences from the WT are indicated by asterisks (*, P < 0.05).

The two CPF proteins exhibit distinct subcellular localization patterns.

To elucidate the functions of BcCRY1 and BcCRY2, deletion mutants (Δbccry1, Δbccry2) and overexpression strains (OE::bccry1, OE::bccry2) for both genes were generated and characterized as described in Materials and Methods (Table 1; see also Fig. S1 and S2 in the supplemental material). For each approach, at least two independent mutants were obtained and tested. Because they displayed identical phenotypes, the data for arbitrarily chosen strains (one per construct) are shown. In addition, double mutants were generated in which both bccry1 and bccry2 were either deleted (ΔΔbccry1/2) or overexpressed (OE::bccry1/2). Their phenotypes matched those of the bccry1 or bccry2 single mutants, and no additive effects were observed (data not shown).

TABLE 1.

B. cinerea strains used in this study

B. cinerea strain	Genotype	Source or reference
B05.10 (WT)	Isolate from Vitis sp.; recipient strain	68
Δbcwcl1 (T11)	B05.10 Δbcwcl1::hph; homokaryon	40
Δbcltf1 (A6)	B05.10 Δbcltf1::hph; homokaryon	30
Δbcvel1 (T22)	B05.10 Δbcvel1::hph; homokaryon	46
Δbclae1 (T3)	B05.10 Δbclae1::nat1; homokaryon	47
OE::bccry1 (T2, T4)	B05.10 PoliC::bccry1::hph at bcniaD; heterokaryon	This study
OE::bccry2 (T23, T26)	B05.10 PoliC::bccry2::nat1 at bcniiA; heterokaryon	This study
OE::bccry1/2 (T2, T6, T9)	B05.10 PoliC::bccry1::hph at bcniaD, PoliC::bccry2::nat1 at bcniiA; heterokaryon	This study
Δbccry1 (T2, T3, T5)	B05.10 Δbccry1::nat1; homokaryon	This study
Δbccry2 (T1, T2, T5)	B05.10 Δbccry2::hph; homokaryon	This study
ΔΔbccry1/2 (T1, T2, T5)	B05.10 Δbccry1::nat1, Δbccry2::hph; homokaryon	This study
bccry1CiL (T3, T5, T7)	B05.10 Δbccry1::nat1, bccry1::hph in loco; heterokaryon	This study
bccry2CiL (T1, T4, T7)	B05.10 Δbccry2::hph, bccry2::nat1 in loco; heterokaryon	This study
bccry2Δ562-623aa (T7, T8, T11)	B05.10 Δbccry2::hph, bccry2Δ562-623aa::nat1 in loco; heterokaryon	This study
bccry1-gfp (T3, T4, T5)	B05.10 Δbccry1::nat1, PoliC::bccry1-gfp::hph at bcniiA; heterokaryon	This study
gfp-bccry2 (T2, T3, T4)	B05.10 PoliC::gfp-bccry2::nat1 at bcniiA; heterokaryon	This study
Δbcwcl1/OE::bccry2 (T2, T4, T9)	B05.10 Δbcwcl1::hph; homokaryon; PoliC::bccry2::nat1 at bcniiA; heterokaryon	This study

To determine the subcellular localization patterns of BcCRY1 and BcCRY2, strains constitutively expressing green fluorescent protein (GFP) fusion proteins were generated and analyzed as described in Materials and Methods. BcCRY1-GFP was found exclusively in the nuclei, as visualized by costaining with Hoechst stain. In contrast, the GFP-BcCRY2 fusion protein localized to the cytosol and was excluded from nuclei (Fig. 1C). The BcCRY1-GFP and GFP-BcCRY2 fusion proteins were functional, as evidenced by the fact that they restored the wild-type phenotypes (BcCRY1-GFP) or phenocopied overexpression strains (GFP-BcCRY2) (see Fig. S3 in the supplemental material). Taken together, these data show that B. cinerea possesses two CPF proteins, the putative CPD photolyase BcCRY1 and the cry-DASH BcCRY2, with distinct subcellular localizations.

BcCRY1 is crucial for photorepair.

The defining trait of photolyases is their ability to use light to repair UV-damaged DNA. Cryptochromes, on the other hand, are defined as proteins that are related to photolyases but show reduced or no DNA repair functionality and have gained regulatory functions (20). Thus, the roles of BcCRY1 and BcCRY2 in the photorepair of B. cinerea were investigated by testing the survivability of conidia of deletion and overexpression strains after exposure to UV light. The Δbcwcl1 mutant was included in this experiment, since the transcription of both CPF protein-encoding genes is reduced in this strain (Fig. 2). Droplets containing conidia from the different mutants were dripped onto solid CM supplemented with 0.01% Triton X-100 for reduction of hyphal growth, leading to defined colonies. Conidia were then exposed to UV light for various periods up to 8 min in order to induce DNA damage. Afterwards, the cultures were either incubated in DD to prevent photoreactivation or in a 12-h white-light/12-h dark shift (LD) to induce photoreactivation. Conidia of the wild type were able to form colonies after 6 min of UV exposure and showed greater tolerance after photoreactivation in LD than in DD. While Δbcwcl1 conidia were less tolerant than WT conidia and showed reduced photoreactivation capacities, the effect of the bccry1 deletion was even more severe. The Δbccry1 mutant was unable to grow after UV exposure for 6 min, and no survival was observed for strains cultivated in LD compared to those in DD, indicating the absence of photoreactivation (Fig. 3; also Fig. S3A in the supplemental material). Overexpression of bccry1, on the other hand, increases UV tolerance, as OE::bccry1 conidia still survived UV treatment of 8 min in LD, indicating more-efficient photoreactivation than that for the wild type (Fig. 3). Reintroduction of bccry1 into the Δbccry1 mutant restored the photorepair activity to wild-type levels (Fig. S3A). In contrast, no changes in UV tolerance and therefore in the effectiveness of photoreactivation were observed for bccry2 deletion or overexpression strains (Fig. 3). Taking these findings together, BcCRY1 plays an essential role in the photorepair in B. cinerea, whereas BcCRY2 is dispensable.

FIG 3.

BcCRY1, but not BcCRY2, mediates photoreactivation. Petri dishes containing CM plus 0.01% Triton X-100 were inoculated with 10-μl droplets of water containing 500 (left) or 100 (right) conidia of the indicated strains. Conidia were exposed to UV light for the periods indicated above the images and were subsequently incubated for 6 days in DD to prevent photorepair (not shown) and in LD to allow photorepair.

BcCRY2 functions in the regulation of vegetative growth.

To reveal possible regulatory functions of BcCRY1 and BcCRY2, the deletion and overexpression strains were characterized with regard to growth rates, virulence, and the light-dependent formation of reproductive structures. Radial growth of the strains was tested on solid CM, which was inoculated with plugs of vegetative mycelia or with droplets of conidial suspensions and was incubated in DD or LD (Fig. 4A). Both Δbccry1 and Δbccry2 mutants exhibited growth rates similar to those of the wild type under all conditions. However, OE::bccry2 mutants showed decreased radial growth rates when inoculated with vegetative mycelia (∼80% of the WT rate) or conidia (<50% of the WT rate) under both light conditions. Further, OE::bccry2 conidia exhibited a delay of the vegetative growth phase. Interestingly, this effect was observed at equal levels in DD and LD (Fig. 4A). To determine whether the retardation of radial growth observed for the OE::bccry2 strain affects virulence, the aggressiveness of the strains was tested on primary leaves of Phaseolus vulgaris by inoculation with plugs of vegetative mycelia or conidial suspensions. In this experiment, all strains infected the plants in a wild-type-like manner when vegetative mycelia were used as the inoculum, but slightly retarded lesion spreading (∼75% of WT) was noted for OE::bccry2 strains following inoculation with conidia (Fig. 4B). Thus, an overabundance of BcCRY2 caused reduced radial growth rates and delayed initiation of vegetative growth of germinating conidia in axenic culture, and this effect occurred independently of the light conditions.

FIG 4.

An overabundance of BcCRY2 affects radial growth rates and virulence. (A) Overexpression of bccry2 causes reduced radial growth in axenic culture in both DD and LD. Solid CM was inoculated either with 10-μl droplets containing 1,000 conidia in water or with plugs of vegetative mycelia from 3-day-old cultures. Pictures were taken from representative cultures deriving from conidial suspensions at 4 days postinoculation (dpi). Average colony diameters were determined from at least three colonies per condition with two measurements per colony. Statistical analyses revealed significant differences between mutants and the WT (*, P < 0.01). (B) Conidia of the OE::bccry2 strain exhibit delayed infection of P. vulgaris. The virulence of the indicated strains was investigated by inoculation of primary leaves either with 7.5-μl droplets containing 1,500 conidia in Gamborg B5 medium plus 2% glucose or with plugs of vegetative mycelia derived from 3-day-old cultures on solid CM. Average lesion diameters were calculated from 12 lesions per strain, with two measurements per lesion. Asterisks indicate statistically significant differences between mutants and the WT (*, P < 0.001).

BcCRY2 has a negative impact on conidiation.

The influence of BcCRY1 and BcCRY2 on the formation of reproductive structures was tested by cultivating the deletion and overexpression strains on solid CM for as long as 10 days under different light conditions (Fig. 5A). For the induction of sclerotial development, strains were incubated in DD, whereas conidial development was induced by cultivation in LD (12 h of white light/12 h of darkness). Since members of the CPF are generally regarded as blue/NUV photoreceptors, black light (12 h of black light/12 h of darkness [BLB-D]) was tested as well. Cultivation under BLB-D conditions was sufficient to induce conidiation in the wild type, although the conidial yield was significantly lower (∼50%) than that in LD after 5 days, whereas only slight differences between BLB-D and LD were observed after 10 days of incubation (∼80% of the LD yield). Neither deletion nor overexpression of bccry1 affected differentiation under the conditions tested; the strains showed wild-type-like conidiation (Fig. 5A). Furthermore, no impact of BcCRY1 or BcCRY2 on sclerotial development in DD was observed (see Fig. S4 in the supplemental material). In contrast, deletion of bccry2 resulted in an earlier onset of conidiation in white or black light, accompanied by higher conidial yields (∼150% of the WT yield in LD and ∼400% of the WT yield in BLB-D after 5 days) and stronger melanization (Fig. 5A; also Fig. S4). Overexpression of bccry2 had the opposite effect, resulting in fewer conidia (∼70% of the WT yield in LD and <50% of the WT yield in BLB-D after 10 days), decreased melanization, and an overabundance of aerial mycelia. The negative impact of BcCRY2 on conidiation was particularly visible in BLB-D, where OE::bccry2 mutants formed exclusively aerial mycelia after 5 days. Further, no significant differences in the conidial yields of the Δbccry2 mutant in LD and BLB-D were observed. Reintroduction of bccry2 into the Δbccry2 mutant restored wild-type-like conidiation in LD (Fig. S3B).

FIG 5.

BcCRY2 affects conidiation and the transcript levels of conidiation-related genes. (A) Deletion and overexpression of BcCRY2 promote and repress conidiation, respectively. Strains were grown on solid CM in 12-h-light/12-h-dark shifts with white light (LD) or black light (BLB-D). Shown are top views of representative cultures. Conidia from three cultures per strain and condition were harvested and counted. Statistical analyses revealed statistical differences between wild type and mutant strains (*, P < 0.05). (B) Transcript levels of conidiation-related genes are altered by deletion or overexpression of bccry2. Transcript levels in the WT and the indicated mutants were determined by RT-qPCR after cultivation for 2 or 3 days on solid CM overlaid with cellophane in DD or LL. Mean values and standard deviations were calculated from two technical replicates of three biological replicates. Asterisks indicate statistical differences between mutants and the WT under the respective condition (*, P < 0.05; **, P < 0.01).

To test whether the negative effect of BcCRY2 on conidiation can also be observed at the level of mRNA, the transcript levels of the conidiation-related genes were detected by RT-qPCR (Fig. 5B). The genes analyzed were bcltf2, encoding a C₂H₂ TF acting as a positive regulator of conidiation; bcltf12, encoding a Myb-TF whose transcription is regulated through BcLTF2 (41); bcpks13, encoding the key enzyme of conidial melanogenesis (31); and bcltf1, encoding a GATA-type TF acting as a repressor of conidiation (30). For this purpose, the strains were cultivated on solid CM overlaid with cellophane for 2 or 3 days in DD or LL. While the transcript levels of bcltf2, bcltf12, and bcpks13 increased between 2 and 3 days of incubation in LL in the wild type, this effect was not visible for bcltf1. Relative transcript levels of bcltf2, bcltf12, and bcpks13 were reduced in OE::bccry2 strains after 3 days in LL (2.4-, 1.7-, and 3.0-fold relative to WT levels, respectively) and were increased in Δbccry2 mutants after 2 days in LL (2.3-, 2.0-, and 2.1-fold relative to WT levels, respectively). No significant effect of BcCRY2 on bcltf1 was detected on the transcriptional level. Further, deletion or overexpression of BcCRY1 had no impact on the transcription of the genes investigated, confirming the absence of a regulatory function of BcCRY1 in differentiation processes (Fig. 5B).

Further, the transcriptional profiles of the conidiation-related TF-encoding genes mentioned above (bcltf1, bcltf2, bcltf12), as well as that of an opsin-encoding gene (bop1) (48), in response to light were investigated (Fig. 6). Two of the genes (bcltf1, bop1) are induced by light in a partially WCC dependent manner (40), whereas bcltf2 and bcltf12 are overexpressed in the absence of light (41). Strains were cultivated on solid CM with a cellophane overlay in DD for 2 days and were then either harvested immediately or exposed to light for different periods. As with cultivation in constant light, BcCRY2 also represses the transcription of bcltf2 and bcltf12 in response to light. Thus, the overall transcript levels of bcltf2 and its light responsiveness were increased in bccry2 deletion mutants (30-fold higher than DD levels and 2-fold higher than WT levels with a 60-min LP) and decreased in overexpression strains (4-fold higher than DD levels and 7-fold lower than WT levels with a 60-min LP) in comparison to the wild type (in which transcript levels with a 60-min LP were 20-fold higher than those in DD). However, while transcript levels of bcltf2 in the OE::bccry2 strain showed low light responsiveness, no induction of bcltf12 was detected in this mutant (Fig. 6). As was observed for bcltf2, transcript levels of bcltf1 and bop1 were reduced (2- and 4-fold lower than WT levels with a 60-min LP), but these genes were still responsive in the OE::bccry2 strain. Deletion of bccry2 had no significant impact on the transcription of bcltf1, bcltf12, or bop1. Taken together, overexpression of bccry2 negatively affected the transcript levels of light-responsive genes, but its deletion had only a minor effect on the transcriptional light responses (Fig. 6).

FIG 6.

The induction of light-responsive genes is attenuated by an overabundance of BcCRY2. The transcript levels of selected light-induced genes in the WT and the indicated mutants were determined by RT-qPCR. Cultures were cultivated on solid CM overlaid with cellophane for 2 days in DD and were then either harvested immediately or exposed to white light for 30 to 300 min. Mean values and standard deviations were calculated from three biological replicates with two technical replicates each. Asterisks indicate statistical differences between mutants and the WT under the respective conditions (*, P < 0.05; **, P < 0.01).

While no significant transcriptional changes for the genes tested were observed in the dark, overexpression of bccry2 caused decreased radial growth rates independently of the light condition (Fig. 4A). To test whether the regulation of conidiation by BcCRY2 is dependent on light, bccry2 was expressed under the control of a constitutive promoter (PoliC) in the “always conidia” Δbcwcl1 mutant (40). Transformation yielded three independent mutants (Δbcwcl1/OE::bccry2) showing identical phenotypes. Characterization of the mutants with regard to conidiation revealed that constitutive expression of bccry2 was sufficient to reduce conidiation in the light as well as in the dark from that of the Δbcwcl1 recipient strain. However, Δbcwcl1/OE::bccry2 strains still showed an “always conidia” phenotype (Fig. 7). These results show that the regulation of conidiation through BcCRY2 occurs independently of light when bccry2 transcription is decoupled from the WCC and that BcCRY2 cannot entirely compensate for the repression of conidiation by the WCC.

FIG 7.

BcCRY2 represses conidiation independently of light. Constitutive expression of bccry2 in the Δbcwcl1 mutant (Δbcwcl1/OE::bccry2) was achieved through integration of the PoliC::bccry2 construct at the bcniaD locus. Strains were cultivated on solid CM in LD, BLB-D, or DD. Top views of representative cultures are shown. Conidia from three cultures per strain and condition were harvested and counted. Significant differences in conidial yields between the Δbcwcl1 and Δbcwcl1/OE::bccry2 strains in the dark are indicated by asterisks (*, P < 0.05). The Δbcwcl1/OE::bccry2 strain showed a lower level of conidiation than the Δbcwcl1 strain under all light conditions tested.

The R/G-rich region of BcCRY2 is not essential for the regulation of conidiation.

The cytosolic localization of GFP-BcCRY2 (Fig. 1C) may indicate a posttranscriptional regulation of conidiation-related genes. In this context, the reported RNA-binding properties (23, 44) and the conserved putative RNA-interacting R/G-rich region of fungal cry-DASHs are of special interest. To test whether the C-terminal region is essential for the functionality of the protein, a variant lacking the R/G-rich region (BcCRY2^Δ562-623aa) (see Fig. S5 in the supplemental material) was expressed in the Δbccry2 background as described in Materials and Methods (Table 1). To test conidiation, wild type, the Δbccry2, bccry2^CiL, and bccry2^Δ562-623aa strains were cultivated on solid CM under LD or BLB-D for 5 days. The premature conidiation and lack of repression in black light observed for the Δbccry2 strain were complemented in both the bccry2^CiL and bccry2^Δ562-623aa strains (Fig. S5). Thus, the putative RNA-interacting R/G-rich domain in BcCRY2 is dispensable for the regulation of conidiation.

DISCUSSION

Studies of the light response system in B. cinerea have a long history. The first reports of wavelength-specific regulation of differentiation programs reach back to the late 19th century (49). The fungus senses at least four different light qualities (NUV, blue, red, and far-red) to regulate conidiation (34). One remarkable feature of the light response is the antagonistic functions of blue and NUV light in repressing and promoting conidiation and in the dedifferentiation and redifferentiation of developing conidia in later stages of conidiophore development (35, 37). This led to the theory of a two-receptor model, whereby reversible NUV/blue light and a far-red/red light receptor systems interact to regulate conidiation (35). Today's knowledge of fungal photoreceptors and their properties, combined with available molecular tools, now enables the study of the molecular mechanisms behind this phenomenon. The sensing of NUV and blue light is attributed to the action of CPF members (2, 3). Therefore, we investigated the functions of the two CPF proteins encoded in the genome of B. cinerea: the putative CPD photolyase BcCRY1 and the putative cry-DASH BcCRY2.

A common trait among fungal CPF protein-encoding genes is their transcriptional dependency on the blue-light-sensing WCC, leading to WCC-dependent UV tolerance (9, 14–17, 23, 25). Indeed, transcription of both bccry1 and bccry2 is induced by light in a WCC-dependent manner and is additionally modulated through different other components of the light response system. While BcVEL1 and the putative methyltransferase BcLAE1 (47) are required for full induction of bccry1 transcription by light, the GATA-type TF BcLTF1 (30) and BcVEL1 from the VELVET protein family (46) act as weak repressors of bccry2 transcription.

Nonetheless, the WCC plays an essential role in tolerance of UV light, as seen by the reduced photoreactivation efficiencies of the Δbcwcl1 mutant due to the altered transcript levels of the photorepair enzymes involved. In contrast, deletion of bccry1 led to a complete loss of photorepair, and overexpression of bccry1 conferred elevated light-dependent UV tolerance. The remaining photoreactivation capability in Δbcwcl1 mutants might be mediated by the low residual light-dependent transcription of bccry1 in this mutant. BcCRY2 does not play an obvious role in photorepair. Since BcCRY1 and BcCRY2 represent the only proteins containing photolyase domains, we consider BcCRY1 the main photolyase of B. cinerea, although a DNA repair activity of BcCRY2 cannot be excluded without biochemical analyses. For instance, the F. fujikuroi cry-DASH does possess DNA repair activity in vitro, but its disruption does not affect photoreactivation (44). In contrast to its ortholog A. nidulans CryA, which acts as a repressor of sexual development (22), BcCRY1 has no obvious regulatory function. The functional role of CPF proteins in B. cinerea might thus be more closely related to that in N. crassa, where a cry-DASH (CRY) fulfills signaling functions and a CPD photolyase (PHR) acts as a major photorepair enzyme (23, 24, 50, 51).

The function of CRY1 from S. sclerotiorum, a close relative of B. cinerea, is of special interest. Here, the deletion causes reduced sclerotial masses and higher numbers of hyphal projections on apothecial stipes in NUV light but does not affect the formation of apothecia per se (26). Like SsCRY1, BcCRY2 has regulatory functions, but these are most prominent in the formation of conidia, which are not formed by S. sclerotiorum. Thus, BcCRY2 affects the radial growth and virulence of mycelia originating from conidia and, most prominently conidiation, itself. Given the inducing properties of NUV/black light on the formation of conidia in B. cinerea (34, 52), we considered a cryptochrome to be a possible inducer of the process. On the contrary, BcCRY2 acts as a repressor of conidiation and is therefore not required for photoinduction by black light. Hence, deletion of bccry2 elevates conidiation in black light to levels observed in white light, indicating the full induction of conidiation by black light that is counteracted by BcCRY2 in the wild-type background. Given the transcriptional regulation of bccry2 by the WCC, BcCRY2 likely contributes to the inhibitory effect of blue light on conidiation. However, artificial constitutive expression of bccry2 in a Δbcwcl1 strain, exhibiting an “always conidia” phenotype, was sufficient to reduce conidiation in light and darkness, suggesting a light-independent signaling mechanism of BcCRY2, although it is considered a photoreceptor. Since the strongest effect of bccry2 overexpression was observed in black light, signaling might nonetheless be facilitated by the presence of light. Another reason for the observed effect of black light could be the absence of other light qualities (e.g., UV, red, or far-red) that might directly or indirectly counteract the signaling function of BcCRY2.

The origin of the induction of conidiation in black light, however, remains elusive, and one can speculate that an unknown photoreceptor is involved in this process. In A. thaliana, the response to NUV light is mediated by UVR8, a protein with similarities to human RCC1 (regulator of chromatin condensation 1). Here, the perception of light relies not on a prosthetic chromophore but on endogenous tryptophan residues (53). Accordingly, NUV-sensing filamentous fungi, such as B. cinerea, Phycomyces blakesleeanus, and Fusarium oxysporum (52) may possess so far uncharacterized proteins containing critical tryptophan residues for perceiving UV light. Indeed, the action spectrum of photoinduced conidiation in B. cinerea (∼230 nm and ∼280 nm) matches the peaks of absorbance by tryptophan residues (42). Furthermore, photoperception could be related to photoreceptor-independent light-sensing mechanisms such as those described recently for Saccharomyces cerevisiae. Here, a peroxisomal oxidase generates H₂O₂ in a light-dependent manner, and the H₂O₂ is sensed by a peroxiredoxin to induce signaling (54). Whereas the addition of exogenous H₂O₂ bypasses the requirement for the peroxisomal oxidase in yeast (54), it is not sufficient to induce conidiation in B. cinerea (J. Schumacher, unpublished data).

In accordance with its impact on conidiation, BcCRY2 affected the transcriptional responses of conidiation-related genes in the light. bcltf2 encodes a C₂H₂ TF that acts as a positive regulator of conidiation in B. cinerea. Transcription of bctlf2 is induced by light in the wild type, and its transcript levels correspond to conidiation (41). Thus, delayed and reduced conidiation in strains overexpressing bccry2 could be explained by the attenuated light induction of bcltf2 and lower transcript levels under continuous illumination. This is further reinforced by the reversed bcltf2 transcription patterns and phenotypes of the Δbccry2 strain. A similar role of BcCRY2 was observed for melanization of the cultures and transcript levels of bcpks13, encoding the key enzyme of conidial melanogenesis (31). Interestingly, BcCRY2 has a negative impact on the light-dependent transcription of both WCC-induced (i.e., bop1, bcltf1) and WCC-repressed (i.e., bcltf2, bcltf12) genes, indicating a function of BcCRY2 downstream of the WCC. Further, the WCC is not required for the signaling function of BcCRY2, as evidenced by the fact that constitutive expression of bccry2 in a Δbcwcl1 background was sufficient to reduce conidiation. Thus, even though it is transcriptionally regulated by the WCC, BcCRY2 modulates the transcript levels of light-responsive genes downstream of the WCC and independently of its activity.

The modes of action of several plant cryptochromes have been elucidated, but little is known about the signaling function of cry-DASH proteins (20). Considering the ability of BcCRY2 to regulate conidiation through bcltf2 transcript levels combined with its cytosolic localization, the assumption of a posttranscriptional mode of action is tempting. Such mechanisms have been suggested previously for F. fujikuroi CryD and N. crassa CRY (23, 25). In this context, it is noteworthy that biochemical analyses revealed the ability of fungal cry-DASH proteins to directly bind nucleic acids (23, 25, 44). For instance, F. fujikuroi CryD is capable of binding RNA at the photolyase domain independently of the putative RNA-interacting R/G-rich region. However, it was suggested that this region might nevertheless mediate the signaling function of the protein (44). To test whether the putative RNA-interacting R/G-rich region of BcCRY2 plays a role in its signaling function, a truncated variant of the protein lacking the C-terminal R/G-rich region was expressed in the Δbccry2 mutant. However, the truncated variant of BcCRY2 was able to restore WT-like conidiation, leaving the function of the R/G-rich region elusive (Fig. S5).

Taken together, we identified BcCRY2 as a negative regulator of conidiation that is transcriptionally regulated by the WCC. In addition to the previously described WCC-dependent transcriptional regulators of bcltf2, e.g., BcLTF1, BcLTF3, and BcREG1 (30, 55), BcCRY2 may contribute to posttranscriptional fine tuning of transcript levels, adding another layer to the tight regulation of conidiation through BcLTF2. A model of this process is shown in Fig. 8. The basis for the induction of conidiation by NUV light, however, remains elusive and may rely on unknown (N)UV light-responsive systems. Thus, so far undescribed tryptophan-based photoreceptors or indirect sensing of (N)UV light through damage or conversion of cellular components may be involved in this process.

FIG 8.

BcCRY2 as a blue-light/WCC target negatively regulates conidiation in B. cinerea. Both CPF protein-encoding genes and the indicated TF-encoding genes are induced by (blue) light in a WCC-dependent manner (WCL1/WCL2). BcLTF1 and BcLTF3 are considered negative regulators of bcltf2 transcription, while BcCRY2 negatively affects the accumulation of transcripts of bcltf2 and other light-induced genes (lig). The near-UV-light-responsive system triggering the expression of bcltf2 and conidiation as an antagonist of the blue-light-sensing WCC remains to be identified.

MATERIALS AND METHODS

Bioinformatics.

Protein and DNA sequences of B. cinerea were exported from EnsemblFungi (http://fungi.ensembl.org/Botrytis_cinerea) (56–58). The UniProt database (http://www.uniprot.org/) was utilized for gathering protein sequences of other fungi. BLASTP algorithms (59) at UniProt (http://www.uniprot.org/blast) and NCBI (http://blast.ncbi.nlm.nih.gov/) were used for the identification of putative orthologs. Multiple-sequence alignments were carried out using the Clustal Omega algorithm at the European Bioinformatics Institute (EBI) (http://www.ebi.ac.uk/Tools/msa/clustalo/), and the phylogeny data generated were processed using Dendroscope 3 (60). Conserved protein domains were identified using InterPro (https://www.ebi.ac.uk/interpro/) (61), and putative nuclear localization signals were predicted using WoLF PSORT (http://www.genscript.com/wolf-psort.html). Statistical significance in the different experiments was determined by t tests.

Standard molecular methods.

Genomic DNA from B. cinerea was extracted according to the 1992 method of Cenis (62). For PCR applications, 1:10-diluted DNA was used as the template. For Southern blot analyses, the genomic DNA was digested with appropriate restriction enzymes (Thermo Fisher Scientific) before separation on 1% agarose gels. After transfer to Whatman Nytran SuPerCharge (SPC) blotting membranes (GE Healthcare), the DNA was hybridized with [α-³²P]dCTP-labeled probes generated by random primers according to the 1989 work of Sambrook et al. (63). Diagnostic PCR was carried out using BioTherm Taq DNA polymerase (GeneCraft), and Phusion High-Fidelity DNA polymerase (Thermo Fisher Scientific) was used for cloning and sequencing purposes. Primers are listed in Tables S1 and S2 in the supplemental material. Replacement fragments and expression constructs were assembled by cotransformation of Saccharomyces cerevisiae FY834 with overlapping DNA fragments and pRS426-derived shuttle vectors (39, 64, 65), followed by amplification of the assembled plasmids in Escherichia coli TOP10F′. Plasmid DNAs from S. cerevisiae and E. coli were isolated using the Easy Yeast plasmid isolation kit (Clontech) or the NucleoSpin Plasmid EasyPure kit (Macherey-Nagel), respectively. Sequencing reactions were performed using the BigDye Terminator sequencing kit (version 3.1; Thermo Fisher Scientific) and were analyzed in an ABI Prism capillary sequencer (Applied Biosystems). The sequences obtained were analyzed with the Lasergene software package (version 11; DNAStar). For analyses of transcript levels, total RNA was isolated with an RNA isolation reagent (TRI reagent; Sigma-Aldrich), and 25-μg samples were separated in 1% agarose gels containing formaldehyde (63). Hybridization was done with [α-³²P]dCTP-labeled probes after transfer of the RNA to Whatman Nytran SPC blotting membranes (GE Healthcare). For RT-qPCR, residual DNA in total-RNA samples (with A₂₆₀/A₂₈₀ and A₂₆₀/A₂₃₀ ratios between 2.0 and 2.3) was removed by treatment with RNase-free DNase (Macherey-Nagel GmbH & Co. KG). Then 1 μg of total RNA was used for cDNA synthesis using SuperScript II reverse transcriptase (Life Technologies) in combination with an oligo(dT)₂₀ primer. RT-qPCR was carried out with the iQ SYBR green supermix in a CFX96 real-time system (Bio-Rad Laboratories Inc.) on 1:7.5-diluted cDNA. The primers used are listed in Table S1 in the supplemental material. The RT-qPCR program consisted of an initial denaturation step (95°C) followed by 40 cycles of denaturation for 10 s (95°C), annealing for 10 s (58°C), and extension for 20 s (72°C). The data generated were analyzed using CFX Manager software (Bio-Rad Laboratories Inc.). Transcript levels were calculated relative to those of bcact1 (actin), which is constitutively expressed in actively growing colonies (vegetative mycelia), according to the 2^−ΔΔCT method (66).

Cultivation of B. cinerea.

Solid complete medium (CM) (67) was used. To induce conidiation or sclerotial development, strains were cultivated at 20°C in a 12-h-light/12-h-dark shift (LD) or complete darkness (DD), respectively. White light (WL; 2,000 μW/cm²) was generated by equal numbers of Sylvania Standard F18W/29-530 “warm-white” and F36W/33-640 “cool-white”' fluorescent tubes, and black light (BLB; 2.7 μW/cm²) was provided by Osram L36/73 neon tubes. B. cinerea cultures for protoplast formation and transformation were handled as described previously (39). Selective pressure on resistant colonies was generated by the addition of 70 μg/ml hygromycin B (Invitrogen) and/or 70 μg/ml nourseothricin (Werner BioAgents) to solid CM. Transformants were genetically purified by spreading conidia on selective medium and selecting resistant colonies. For RNA and DNA isolation, cultures were grown on solid CM overlaid with cellophane sheets. The virulence of B. cinerea strains was assayed on French bean (Phaseolus vulgaris) plants. Primary leaves were inoculated either with plugs of 3-day-old vegetative mycelia from CM or with 7.5-μl droplets of conidial suspensions (2 × 10⁵ conidia/ml of Gamborg B5 medium [Duchefa Biochemie] supplemented with 2% glucose; conidia derived from 7- to 10-day-old CM cultures). The infection process was monitored at 22°C under natural light (daylight/night) and humid conditions. UV tolerance of conidia was assayed by inoculation of solid CM (supplemented with 0.01% Triton X-100) with 10-μl droplets containing 5 × 10⁴ or 1 × 10⁴ conidia/ml in H₂O. The conidia were exposed to UV light (λ = 254 nm) at 0.3 μW/cm² and were then incubated in DD or LD in order to prevent or allow for photoreactivation, respectively.

Generation of B. cinerea mutants.

Unless otherwise indicated, B. cinerea B05.10 (WT) (68) was used as the recipient strain for genetic modifications. For the generation of gene replacement fragments for bccry1 and bccry2, the 5′ and 3′ noncoding regions were amplified from genomic DNA using primer pairs bccry1-5F/bccry1-5R or bccry2-5F/bccry2-5R and bccry1-3F/bccry1-3R or bccry2-3F/bccry2-3R (Fig. S1 in the supplemental material). To generate pΔbccry1, S. cerevisiae was cotransformed with the flanks of bccry1, the nourseothricin resistance cassette PtrpC::nat1 (amplified from pZPnat1 using primer pair hph-F/nat1-R), and EcoRI/XhoI-digested pRS426. The replacement fragment from pΔbccry1 was amplified using the primer pair bccry1-5F/bccry1-3R. Transformation of the WT with the fragment yielded three independent transformants (Δbccry1; T2, T3, T5) that had undergone homologous recombination at the 5′ (primer pair bccry1-hi5F/nat1-hiF) and 3′ (primer pair bccry1-hi3R2/PtrpC-P2) ends. The absence of the wild-type allele after genetic purification by single-spore isolation was confirmed using primer pair bccry1-WT-F/bccry1-WT-R. Single integrations of the replacement fragment were shown by Southern blot analyses. Complementation of the Δbccry1 mutant (T2) was achieved by replacement of the previously introduced PtrpC::nat1 cassette with bccry1 fused to a terminator (Tgluc). For this purpose, pbccry1_CiL was generated by cotransformation of S. cerevisiae with pRS426 (EcoRI/XhoI digested), the 5′ noncoding region plus the open reading frame (ORF) (primer pair bccry1-5F/bccry1-Tgluc-R), Tgluc (primer pair Tgluc-F2/Tgluc-nat1-R), PtrpC::hph (primer pair hph-F/hph-R), and the 3′ noncoding region (primer pair bccry1-3F/bccry1-3R). Transformation of the Δbccry1 mutant (T2) yielded three independent strains (bccry1^CiL; T3, T5, T7) with homologous integration, detected by diagnostic PCR using primer pairs bccry1-hi5F/bccry1-WT-R and bccry1-hi3R2/hph-hiR, as indicated in Fig. S1A. The construction of pΔbccry2 was facilitated by the cotransformation of S. cerevisiae, with its 5′ and 3′ noncoding regions, with the hygromycin resistance cassette PtrpC::hph (amplified from pCSN44 using primer pair hph-F/hph-R) and pRS426 (EcoRI/XhoI digested). Transformation of the WT or the Δbccry1 mutant (T2) with the replacement fragment yielded three independent transformants (Δbccry2 [T1, T2, T5] and ΔΔbccry1/2 [T1, T2, T5]) that had undergone homologous recombination at the 5′ (primer pair bccry2-hi5F/hph-hiF) and 3′ (primer pair bccry2-hi3R/PtrpC-P2) ends. The absence of the wild-type allele in single conidial isolates was confirmed using primer pair bccry2-WT-F/bccry2-WT-R. Single integration events were detected by Southern blot analyses. Complementation of the Δbccry2 mutant (T1) was achieved by replacement of the PtrpC::hph cassette with bccry2 or the mutated allele bccry2^Δ562-623aa, lacking the part encoding the R/G-rich region. pbccry2_CiL was generated by cotransformation of S. cerevisiae with pRS426 (EcoRI/XhoI digested), the 5′ noncoding region plus the ORF (primer pair bccry2-5F/bccry2-Tgluc-R), Tgluc (see above), PtrpC::nat1 (see above), and the 3′ flank (primer pair bccry2-3F/bccry2-3R). pbccry2^Δ562-623aa was generated by cotransformation of pRS426 (EcoRI/XhoI digested) with the 5′ noncoding region plus the ORF (primers bccry2-5F/bccry2^Δ562-623aa-R) and the 3′ noncoding region plus PtrpC::nat1::Tgluc (primer pair bccry2^Δ562-623aa-F/bccry2-3R; template, pbccry2_CiL). Transformation of Δbccry2 (T1) yielded three independent mutants per construct (bccry2^CiL [T1, T4, T7] and bccry2^Δ562-623aa [T7, T8, T11]) with homologous integration, verified by diagnostic PCR using primer pairs bccry2-hi5F/bccry2-hi5R and bccry2-hi3R/nat1-hiR. The construct for overexpressing (Fig. S2A in the supplemental material) bccry1 (pNDH_PoliC::bccry1) was generated by cotransformation of S. cerevisiae with bccry1 (primer pair bccry1-PoliC-F/-Tgluc-R) and pNDH-OGG (NcoI/NotI digested) (39). Transformation of the WT yielded three independent transformants (OE::bccry1; T2, T4) with targeted integration of the expression construct at bcniaD (primer pairs bcniaD-hi5F/Tgluc-hiF and bcniaD-hi3R/hph-hiF). pNAN_PoliC::bccry2 was cloned by cotransformation of S. cerevisiae with bccry2 (primer pair bccry2-PoliC-F/bccry2-Tgluc-R) and pNAN-OGG (NcoI/NotI digested) (39). Transformation of the WT, OE::bccry1 (T4), and Δbcwcl1 strains (Fig. S2B) (40) yielded two (OE::bccry2; T23, T26), three (OE::bccry1/2; T2, T6, T9), and three (Δbcwcl1/OE::bccry2; T2, T4, T9) independent transformants, respectively. Targeted integration at bcniiA was verified by diagnostic PCR using primer pairs bcniiA-hi5F/Tgluc-hiF and bcniiA-hi3R/nat1-hiF. Increased transcript levels of bccry1 and bccry2 in the OE::bccry1 and OE::bccry2 strains were confirmed by Northern blot analyses (Fig. S2C). GFP fusion proteins were expressed by integration of the respective constructs under the control of the constitutive promoter PoliC at bcniaD (bccry1-gfp) and bcniiA (gfp-bccry2). pNAH_PoliC::bccry1-gfp or pNAN_PoliC::gfp-bccry2 were generated by cotransformation of yeast with pNAH-OGG (NcoI digested) and bccry1 (primer pair bccry1-GFP-F/bccry1-Tgluc-R) or pNAN-OGG (NotI digested) and bccry2 (primer pair bccry2-PoliC-F/bccry2-GFP-R) (39). Transformation of the Δbccry1 mutant (T2) with NAH_bccry1-gfp and of the WT with NAN_gfp-bccry2 yielded three independent transformants per construct (gfp-bccry2 [T2, T3, T4] and bccry1-gfp [T3, T4, T5]).

Microscopy.

Suspensions of 1 × 10⁵ conidia/ml in Gamborg B5 medium (Duchefa Biochemie) supplemented with 2% glucose and 0.02% (NH₄)₂PO₄ were incubated on object slides for 16 h to obtain germlings. Nuclei were stained with Hoechst 33342 (Sigma-Aldrich) as described in reference 39. An AxioImager M1 microscope equipped with an ApoTome.2 module was used for microscopy, and an AxioCam MRc camera was used for capturing images. Filter sets 38 (excitation, BP 470/40; beam splitter, FT 495; emission, BP 525/50) and 49 DAPI shift free (excitation, G 365; beam splitter, FT 395; emission, BP 445/50) were used to monitor GFP fluorescence and Hoechst staining, respectively. Data were processed using the AxioVision software package (release 4.8; Zeiss).