Transcriptional regulation by the Velvet complex during development in Neurospora crassa

Cover by Marta Florido Capilla UNIVERSIDAD DE SEVILLA FACULTAD DE BIOLOGIA DEPARTAMENTO DE GENETICA Transcriptional regulation by the velvet complex during development in Neurospora crassa Sara Cea Sánchez Tesis doctoral Universidad de Sevilla 2023 UNIVERSIDAD DE SEVILLA FACULTAD DE BIOLOGIA DEPARTAMENTO DE GENETICA Transcriptional regulation by the velvet complex during development in Neurospora crassa Memoria presentada por la licenciada Sara Cea Sánchez para optar al título de Doctora en Biología con Mencion Internacional por la Universidad de Sevilla Directores: Prof. Luis María Corrochano Peláez, Catedrático del departamento de Genética y Prof. David Cánovas López, profesor titular del departamento de Genética, ambos de la Universidad de Sevilla. La realización de esta Tesis Doctoral ha sido financiada gracias a las siguientes ayudas para proyectos estatales I+D+i del Ministerio de Ciencia e Innovación: • Regulación de la Degradación, Transporte al Núcleo y Formación de Complejos de VE-1 Durante el Desarrollo y la Biosíntesis de Carotenoides en Neurospora Crassa (RTI2018-098636-B-I00). • La Regulación del Regulador: Estabilidad y Localización de Velvet Durante la Conidiación y la Biosíntesis de Carotenoides en Neurospora (BIO2015- 67148-R) Durante este periodo, el doctorando también se ha beneficiado de dos ayudas para estancias de corta duración en centros de investigación extranjeros: • Convocatoria de ayudas, en el marco del Campus de Excelencia Internacional Andalucía Tech, para la movilidad e intercambio de investigadores y profesores de la Universidad de Sevilla. • Estancia de corta duración de la organización Europea de Biología Molecular (“EMBO short-term fellowship”) Parte de los resultados de esta Tesis han sido publicados en los siguientes medios: 1. Artículos científicos Cea-Sánchez S, Corrochano-Luque M, Gutiérrez G, Glass NL, Cánovas D, Corrochano LM (2022) Transcriptional Regulation by the Velvet Protein VE- 1 during Asexual Development in the Fungus Neurospora crassa. mBio, 13(4): e0150522. (Premio Publicación Científica del Mes de la Facultad de Biología). Gil-Sánchez MM, Cea-Sánchez S, Luque EM, Cánovas D, Corrochano LM (2022) Light regulates the degradation of the regulatory protein VE-1 in the fungus Neurospora crassa. BMC Biology, 20(1):149. Bayram ÖS, Dettmann A, Karahoda B, Moloney NM, Ormsby T, McGowan J, Cea- Sánchez S, Miralles-Durán A, Brancini GTP, Luque EM, Fitzpatrick DA, Cánovas D, Corrochano LM, Doyle S, Selker EU, Seiler S, Bayram Ö. (2019) Control of Development, Secondary Metabolism and Light-Dependent Carotenoid Biosynthesis by the Velvet Complex of Neurospora crassa. Genetics. 212(3):691-710. 2. Comunicaciones y ponencias en congresos Cea-Sánchez S, Martín-Villanueba S, Gutiérrez G, Cánovas D, Corrochano LM. Regulatory Role of VE-1 in transcription during sexual development in the fungus Neurospora crassa.31st Fungal Genetics Conference, March 15-20, 2022. Asilomar, Pacific Grove, California, USA (POSTER). Cea-Sánchez S, Martín-Villanueba S, Corrochano-Luque M, García-Marcelo MJ, Gutiérrez G, Cánovas D, Corrochano LM. Regulation of asexual and sexual development by the velvet complex in Neurospora crassa. Neurospora 2021. Camp Allen, Navasota, Texas, USA (POSTER). 4.3.1. Asexual development (conidiation) experiments ....................................................... 48 4.3.1.1. Culture conditions ............................................................................................. 48 4.3.1.2. cDNA libraries preparation and statistical analysis ........................................... 48 4.3.2. Sexual development experiments ............................................................................... 49 4.3.2.1. Culture conditions ............................................................................................. 49 4.3.2.2. cDNA libraries preparation and statistical analysis ........................................... 49 5. PROTEIN HANDLING ........................................................................................................... 51 5.1. Preparation of total protein extracts .......................................................................... 51 5.2. Western blot .............................................................................................................. 51 5.3. Coomassie staining of polyacrylamide gels .............................................................. 52 5.4. Cell fractionation and nuclei isolation ...................................................................... 53 5.5. Immunoprecipitation of tagged proteins ................................................................... 54 5.6. Identification of posttranslational modifications (PTM) ........................................... 55 5.6.1. Culture conditions and protein immunoprecipitation. ................................................ 55 5.6.2. Digestion of proteins with trypsin .............................................................................. 55 5.6.3. Peptide identification by MS/MS-based analysis ....................................................... 56 5.6.4. Data analysis ............................................................................................................... 57 6. PROTEIN-DNA INTERACTION ANALYSIS ............................................................................ 57 6.1. Heterologous expression of VE-1 and VE-2 in E. coli ............................................. 57 6.2. Chitin column affinity chromatography .................................................................... 58 6.3. Preparation of labeled probes .................................................................................... 58 6.4. Electrophoretic Mobility Shift Essay (EMSA) ......................................................... 59 7. MICROSCOPY ...................................................................................................................... 59 7.1. Optical microscopy ................................................................................................... 59 7.2. Stereomicroscopy ...................................................................................................... 60 7.3. Scanning Electron Microscopy (SEM) ..................................................................... 60 CHAPTER I: CHARACTERIZATION AND REGULATION OF THE VELVET COMPLEX FORMATION IN NEUROSPORA CRASSA ............................................................................... 63 1. INTRODUCTION ................................................................................................................... 63 2. RESULTS ............................................................................................................................. 64 2.1. The genome of N. crassa contains four members of the velvet family .................... 64 2.2. Light does not affect the expression of ve-2, vos-1, and lae-1 during vegetative growth 65 2.3. VE-2 and LAE-1 localize in nuclei and cytoplasm during vegetative growth in a light-independent manner ....................................................................................................... 67 2.4. Velvet proteins VE-1, VE-2, and LAE-1 methyltransferase form a heterotrimeric complex in N. crassa .............................................................................................................. 68 2.5. The velvet protein VE-2 and the LAE-1 methyltransferase are stable proteins during vegetative growth ................................................................................................................... 69 2.6. The components of the velvet complex in N. crassa undergo post-translational modifications in vegetative mycelia ....................................................................................... 70 .................................................................................................................................................... 72 2.7. Prediction of kinases responsible for the phosphorylation sites identified in VE-1 and VE-2 ................................................................................................................................. 72 3. DISCUSSION ........................................................................................................................ 74 4. SUPPLEMENTARY MATERIAL ............................................................................................. 77 CHAPTER II: TRANSCRIPTIONAL REGULATION BY THE VELVET PROTEIN VE-1 DURING ASEXUAL DEVELOPMENT IN THE FUNGUS NEUROSPORA CRASSA ........... 81 1. INTRODUCTION ................................................................................................................... 81 2. RESULTS ............................................................................................................................. 83 2.1. The ve-1 and ve-2 genes are required for aerial growth and conidiation .................. 83 2.2. VE-1 and VE-2 localize to nuclei and cytoplasm of aerial tissue ............................. 86 2.3. Transcriptomic comparisons identified genes regulated by VE-1 ............................ 88 2.4. Comparative transcriptomics revealed a role of VE-1 in transcriptional regulation during asexual development but not during vegetative growth ............................................. 90 2.5. VE-1 regulates the transcription of methyltransferase genes, including .................. 93 lae-1 93 2.6. Role of VE-1 in aerial tissue length: identification of VIB-1 ................................... 96 2.7. Role of VE-1 in conidiation: regulation of fl and microconidiation ......................... 97 3. DISCUSSION ....................................................................................................................... 99 4. SUPPLEMENTARY MATERIAL ........................................................................................... 103 CHAPTER III: VE-1 REGULATION OF MAP KINASE SIGNALING CONTROLS SEXUAL DEVELOPMENT IN NEUROSPORA CRASSA ....................................................................... 107 1. INTRODUCTION ................................................................................................................ 107 2. RESULTS .......................................................................................................................... 112 2.1. The N. crassa velvet proteins VE-1 and VE-2 play a role in sexual development . 112 2.2. The members of the velvet complex in N. crassa are not involved in male fertility 113 2.3. Proteins VE-1, VE-2, and LAE-1 are required for female fertility during sexual development in N. crassa ..................................................................................................... 114 2.4. N. crassa VE-1, VE-2, and LAE-1 proteins accumulate in early stages of sexual development ......................................................................................................................... 119 2.5. Comparative transcriptomic analysis reveals major transcriptional changes of the ∆ve-1 strain during sexual development .............................................................................. 119 2.6. VE-1 controls the expression of genes required for growth, asexual and sexual development during fruiting body formation ....................................................................... 122 2.7. VE-1 controls the expression of predicted G-protein coupled receptor genes ....... 124 2.8. VE-1 regulates the expression of serine/threonine kinases and serine/threonine phosphatases ........................................................................................................................ 126 2.9. The expression of cell-to-cell fusion genes involved in sexual development is affected in the mutant ∆ve-1 ................................................................................................ 129 2.10. VE-1 controls the expression of transcription factors that regulate sexual development ......................................................................................................................... 131 2.11. Coordinated regulation of genes involved in fruiting body formation by VE-1 protein and the MAP kinases MAK-1 and MAK-2 ............................................................. 133 2.12. Melanin synthesis genes are involved in perithecium development and are downregulated in the ∆ve-1 strain ....................................................................................... 134 2.13. The VE-1/VE-2 protein complex binds to the promoters of MAP Kinase genes of the PR, CWI and OS signaling pathways ............................................................................ 135 3. DISCUSSION ..................................................................................................................... 136 GENERAL DISCUSSION ........................................................................................................ 145 CONCLUSIONS ....................................................................................................................... 151 BIBLIOGRAPHY ..................................................................................................................... 155 Summary 1 SUMMARY Developmental transitions in fungi are key to their success in colonizing new habitats and adapting to stressful environmental conditions. Developmental transitions are regulated by environmental factors such as light. Neurospora crassa is an ascomycete heterothallic filamentous fungus and a model organism for research on several aspects of fungal biology, including the study of light sensing and morphogenesis during development. The life cycle of N. crassa includes asexual development with the formation of vegetative conidia that are easily dispersed, and sexual development, a complex process involving the formation of sexual reproductive structures (perithecia), were the ascospores (meiotic products) are formed. The velvet complex is a fungal-specific protein complex that participates in the regulation of gene expression in response to environmental signals such as light, as well as developmental processes, pathogenesis, and secondary metabolism. In this thesis, we have characterized the role of the velvet complex (the velvet proteins VE-1 and VE-2, and the methyltransferase LAE-1) during different stages of the life cycle of N. crassa. We first started studying the role of this complex during asexual development. Mutations in ve-1 or ve-2, but not in lae-1 led to shorter height of aerial tissue and an increased development of macroconidia. Additionally, when ve-1 or ve-2 mutations were combined with deletion of the transcription factor gene fl, which is an activator of macroconidiation, we observed an increased microconidiation. VE-2 and LAE-1 were detected during vegetative growth and conidiation, unlike VE-1 which was mostly observed in samples obtained from submerged vegetative hyphae. We propose that VE-1 is the limiting component of the velvet complex during conidiation and has a major role in the transcriptional regulation of conidiation. Characterization of the role of VE-1 by RNA seq experiments during mycelial growth and asexual development (conidiation) allowed the identification of a set of genes regulated by VE-1, most notably the transcription factor genes vib-1 and fl, that participate in the regulation of conidiation. We propose that VE-1 and VE-2 regulate the development of aerial tissue and the balance between macro- and microconidiation in coordination with FL and VIB-1. During sexual development, strains lacking VE-1 and/ or VE-2 display a markedly delayed and reduced sexual development with fewer fruiting bodies compared to the wildtype strain. Alterations in the development of female structures, protoperithecia, in the ve- Summary 2 1 and ve-2 mutants suggested that the VE-1/VE-2 complex should regulate transcription during sexual development. We have characterized the transcriptome of wild-type and Dve- 1 mutant strains over the time course of sexual development in both dark and light. Among the misregulated genes, we detected genes that are essential for sexual development, such as genes for mitogen-activated protein (MAP) kinase signaling pathways, cell-cell fusion genes (ham genes) and transcription factor genes. Electrophoretic mobility shift essays of the promoter regions of the four MAP kinase genes suggest that VE-1 could be regulating sexual development by binding directly to promoters of these key regulatory genes. Furthermore, we detected transcription of ve-1, ve-2, and lae-1 during all stages of sexual development, but the three proteins were not detected in the later stages of development (four and six days after fertilization), suggesting a major role for the velvet complex in the early stages of sexual development. Our results provide key insights into the control of multistage development processes by the regulatory velvet complex in the fungus N. crassa, and will help to understand how environmental signals are integrated in the fungal cell to regulate development. Resumen 3 RESUMEN Los cambios morfológicos que se producen durante el desarrollo de los hongos son clave para su éxito en la colonización de nuevos hábitats y para su adaptación a condiciones ambientales estresantes. Los cambios morfológicos están regulados por factores ambientales como la luz. Neurospora crassa es un hongo filamentoso heterotálico ascomiceto y un organismo modelo para la investigación de varios aspectos de la biología fúngica, incluido el estudio de la detección de luz y la morfogénesis durante el desarrollo. El ciclo de vida de N. crassa está compuesto por dos etapas, el desarrollo asexual con la formación de conidios vegetativos que se dispersan fácilmente y el desarrollo sexual, un proceso complejo que implica la formación de estructuras reproductivas sexuales (peritecios), en cuyo interior se forman las ascosporas (productos meióticos). El complejo velvet es un complejo proteico específico de hongos que participa en la regulación de la expresión génica en respuesta a señales ambientales como la luz, así como en los procesos de desarrollo, de patogénesis y en el metabolismo secundario. En esta tesis hemos caracterizado el papel del complejo velvet (formado por las proteínas velvet VE-1 y VE-2, y la metiltransferasa LAE-1) en diferentes etapas del ciclo de vida de N. crassa. Primero comenzamos a estudiar el papel de este complejo durante el desarrollo asexual. La deleción de los genes ve-1 o ve-2, pero no de lae-1, dieron lugar a una reducción en la longitud del tejido aéreo y a un incremento en la producción de macroconidios. Además, cuando las mutaciones ve-1 o ve-2 se combinaron con la eliminación del gen que codifica el factor de transcripción FL, que es un activador de la macroconidiación, se produjo un aumento de la microconidiación. Las proteínas VE-2 y LAE-1 se detectaron durante el crecimiento vegetativo y la conidiación, a diferencia de VE-1, que se observó principalmente en muestras obtenidas de hifas vegetativas sumergidas. Nuestros resultados nos llevaron a proponer que VE-1 es el componente limitante del complejo velvet durante la conidiación y tiene un papel importante en la regulación transcripcional de este proceso. La caracterización del papel de VE-1 mediante experimentos de secuenciación de ARNm durante el crecimiento micelial y el desarrollo asexual (conidiación) permitió la identificación de un conjunto de genes regulados por VE-1, de los cuales destacamos los genes vib-1 y fl, que participan en la regulación de la conidiación. Proponemos que VE-1 y VE-2 regulan el desarrollo del tejido aéreo y el equilibrio entre los procesos de macro y microconidiación en coordinación con los factores de transcripción FL y VIB-1. Resumen 4 Durante el desarrollo sexual, las cepas que carecen de las proteínas VE-1 y/o VE-2 muestran un desarrollo sexual notablemente retrasado y reducido con una menor producción de cuerpos fructíferos en comparación con la cepa de tipo silvestre. Las alteraciones en el desarrollo de estructuras femeninas, protoperitecios, en los mutantes ∆ve- 1 y ∆ve-2 sugirieren la participación del complejo VE-1/VE-2 en la regulación de la transcripción durante el desarrollo sexual. Hemos caracterizado el transcriptoma de las cepas mutantes ∆ve-1 y la de tipo silvestredurante el desarrollo sexual en condiciones de oscuridad y de luz continua. Entre los genes desregulados, detectamos genes que son esenciales para el desarrollo sexual, como los genes que codifican componentes de las vías de señalización de la proteína quinasa activada por mitógenos (MAPK), los genes de fusión celular (genes ham) y los genes que codifican factores de transcripción. Los resultados obtenidos en los ensayos de cambio de movilidad electroforética (EMSA) de unión a las regiones promotoras de los genes mak-1, mak-2, mek-2, and os-4 (genes que codifican componentes de las rutas MAP quinasa) sugieren que VE-1 podría estar regulando el desarrollo sexual al unirse directamente a los promotores de estos genes claves en la regulación del desarrollo sexual. Además, detectamos transcripción de los genes ve-1, ve- 2 y lae-1 en todas las etapas del desarrollo sexual, pero las tres proteínas no se detectaron en las etapas posteriores del desarrollo (cuatro and seis días después de la fertilización), lo cual sugiere que el complejo velvet tiene un papel regulador esencial en las primeras etapas del desarrollo sexual. Nuestros resultados brindan información clave sobre el papel del complejo velvet en el control de los cambios morfológicos que suceden durante el desarrollo en el hongo N. crassa y nos ayudarán a comprender cómo se integran las señales ambientales para regular el desarrollo. 5 INTRODUCTION 6 7 GENERAL INTRODUCTION Figura 1. tesis A B GENERAL INTRODUCTION 1. Neurospora crassa Neurospora crassa is a filamentous fungus that belongs to the class of Ascomycetes and is a member of the Sordariaceae family. The first published account of N. crassa was in 1843, and it described the presence of a mold that produced crops of powdery orange spores (conidia) in infected bakeries in France (Payen, 1843). This mold, named Monilia sitophila, was frequently found in bread and other carbohydrate-rich food products, as well as colonizing the remains of burnt vegetation after forest fires (Figure 0. 1). It was a century later when the genus received its actual name after the discovery of the formation of sexual fruiting bodies (perithecia) by the mycologists Cornelius L. Shear and Bernard O. Dodge (Davis and Perkins, 2002). A B Figure 0. 1. N. crassa. (A) N. crassa growing in a Petri dish. (B) Neurospora sp. growing on the trunk of a burned tree (Luque et al., 2012). This fungus is commonly found in nature in tropical, subtropical, and temperate areas, and it is often observed in recently burned vegetation and trees (Turner et al., 2001; Jacobson et al., 2004; Jacobson et al., 2006). It has been proposed that the heat caused by fires stimulates the germination of ascospores in the soil, and provides the fungus with a sterile environment containing nutrients that stimulate its growth and facilitates colonization of these type of niches (Jacobson et al., 2006). Introduction 14 2002). The WC-2 protein also contains a GATA-type zinc finger, a PAS domain, and a nuclear localization signal (Linden and Macino, 1997). WC-1 and WC-2 dimerize through their PAS domains to form the WCC complex (Ballario et al., 1998; Cheng et al., 2002). Figure 0. 4. Light sensing signaling components in N. crassa. The figure shows the different domains of the components of the light sensing system in N. crassa. LOV-domain photoreceptors (WC-1 and VVD) contain the flavin chromophore-binding domain (LOV), and may also contain the protein-interaction domain (PAS) and the Zn finger domain (ZnF). Rhodopsins (NOP-1) contain the retinal-binding domain. CRY contains the FAD chromophore-binding domain and the antenna cofactor-binding domain. Phytochromes contain an amino-terminal sensory domain and a carboxy-terminal output domain. The sensory domain involved in binding the bilin chromophore is composed of three domains (PAS, GAF, and PHY). The output domain is composed of the histidine kinase domain (HK), the ATPase domain found in ATP binding proteins, and the response regulator domain (REC). Image modified from (Corrochano, 2007). Another blue-light photoreceptor is the protein encoded by the vvd gene, a small protein (186 amino acids) with a LOV domain that acts as the main regulator of the WCC, and regulates the transient light-dependent transcription (photoadaptation) (Schwerdtfeger and Linden, 2001; Heintzen et al., 2001; Shrode et al., 2001; Schwerdtfeger and Linden, 2003; Zoltowski et al., 2007) (Figure 0.4). Other photoreceptors genes are the opsin nop-1 the cryptochrome cry, and two phytochrome genes, phy-1 and phy-2 (Galagan et al., 2003). The presence of these photoreceptors encoding genes in addition to wc-1 was unexpected, as there is no evidence for N. crassa reacting to any other type of light, other than blue light, and only the activity of WCC is required for the blue light response, suggesting that these photoreceptors must N N N N N N C C C C C C 1167 aa 530 aa 186 aa 304 aa 745 aa 1242 aa 1166 aa WC-1 (NCU02356) WC-2 (NCU00902) VVD (NCU03967) NOP-1 (NCU10055) CRY (NCU00582) N PHY-1 (NCU04834) PHY-2 (NCU05790) 200 aa cofactor-binding C retinal-binding C 15 GENERAL INTRODUCTION have secondary roles in N. crassa (Corrochano, 2019). NOP-1 encodes a protein that binds to retinal in vitro, forming a rhodopsin that absorbs green light (Brown and Jung, 2006). CRY belongs to the family of cryptochromes, proteins capable of detecting blue light. The protein CRY is found in both the nucleus and the cytoplasm and is regulated by light and circadian rhythm. CRY binds to the FAD chromophore in vitro, and its sequence suggests that it belongs to the DASH subfamily of cryptochromes (Froehlich et al., 2010). Phytochromes are red light photoreceptors that use a tetrapyrrole molecule as a chromophore (Rockwell et al., 2006). PHY-1 and PHY-2 have a sensory domain responsible for binding to the chromophore. The phytochrome genes are not regulated by light, but phy-1 is regulated by the circadian clock (Froehlich et al., 2005). 3.3. Activation of transcription by WCC and adaptation to light The mechanisms of WCC-mediated transcription regulation by light have been well characterized (Figure 0. 5). It is proposed that the process begins with the formation of a flavin-cysteine adduct in a conserved cysteine from the LOV domain upon illumination (Salomon et al., 2000) that leads to a conformational change that affects protein-protein interaction (Crosson et al., 2003). The conserved cysteine in the WC-1 LOV domain is essential for the transcriptional regulation of light-responsive genes (Cheng et al., 2003). This formation of the flavin-cysteine adduct is completely reversible in darkness (Salomon et al., 2000). In cultures kept in the dark, the WCC binds to light-responsive elements (LREs) on the promoters of light-regulated genes (Froehlich et al., 2002; Froehlich et al., 2003; He and Liu, 2005; Smith et al., 2010) and interacts with the histone acetyltransferase NGF-1 (Brenna et al., 2012). Light triggers a conformational change in the WC-1 protein that results in the formation of WCC homodimers through the LOV domain of the WC-1 subunits (Malzahn et al., 2010). It has been shown that the formation of WCC homodimers leads to activation of the histone acetyltransferase activity of NGF-1, which results in a light-dependent acetylation of histone H3 lysine 14 (H3K14ac) in the nucleosomes associated with promoters of light-regulated genes (Grimaldi et al., 2006; Brenna et al., 2012). Histone acetylation results in a change in chromatin structure that likely allows a better access for the RNA polymerase to the promoters. This model is supported by observations that strains with a replacement of H3K14 with glutamine or with mutations in Introduction 16 the ngf-1 gene show a blind phenotype (Grimaldi et al., 2006). Light also plays a key role in controlling the abundance and phosphorylation of the WCC. Light activates transcription of the wc-1 gene (Ballario et al., 1996; Káldi et al., 2006) and phosphorylation of WC-1, leading to its degradation and replacement in the WCC by newly synthesized but inactive WC-1 (Talora et al., 1999; Schwerdtfeger and Linden, 2000; Schwerdtfeger and Linden, 2001; He and Liu, 2005). WC-2 is also phosphorylated by exposure to light (Schwerdtfeger and Linden, 2000). Regulation of light-activated transcription by the WCC is transient (He and Liu, 2005; Olmedo et al., 2010a). When mycelia grow in complete darkness and are exposed to light, the number of WCC-dependent mRNAs increases rapidly, and then, after extended illumination times, the amount of these mRNA decreases in a process called photoadaptation (Lauter and Yanofsky, 1993; Arpaia et al., 1999; Schwerdtfeger and Linden, 2001; Schwerdtfeger and Linden, 2003). In this state, increasing the intensity of light can provoke a second response and further photoadaptation (Schwerdtfeger and Linden, 2001; Malzahn et al., 2010). An essential protein in the photoadaptation process is the product of the gene vvd, which is also a light-responsive gene regulated by the WCC. Light-activated VVD forms dimers through its LOV domains (Zoltowski and Crane, 2008; Peter et al., 2012). It has been suggested that the similarity between the LOV domains in VVD and WC-1 could explain a possible interaction between the two photoreceptors (Malzahn et al., 2010). It has been shown that after exposure to light, the newly synthesized VDD protein binds to WC-1 resulting in the disruption of the transcriptionally active WCC dimers and leading to an attenuation of transcription activation. Although the amount of newly synthesized WC-1 increases in response to light-activated wc-1 transcription, this increase in the WC-1 protein would not be enough to compensate for its degradation. Instead, the WC-1 and VVD interaction prevents the degradation of WC-1 creating a pool of light-activated and transcriptionally inactive WC-1. As expected, VVD overexpression results in desensitization of the WCC (Chen et al., 2010a; Hunt et al., 2010; Malzahn et al., 2010). There is evidence showing that histone modification occurs during photoadaptation and that it may help to evict the WCC from the promoters. After long exposures to light, DIM-5 is responsible for an increase in the degree to which histone H3 lysine 9 is trimethylated (H3K9me3) in the frq promoter. In mutants lacking DIM-5, an increase in frq mRNA accumulation is detected after light exposure and 17 GENERAL INTRODUCTION more WC-2 binds to the promoter, supporting the role of H3K9me3 in WCC eviction (Ruesch et al., 2015). Figure 0. 5. Simplified model for light-regulated transcription in N. crassa. The activation of transcription by light is mediated by WCC and chromatin remodeling by the histone acetyltransferase NGF-1. The photoadaptation process is mediated by the interaction of LOV domains of WC-1 and VVD. Methylation of histones (H3K9me3) by DIM-5 further contributes to the eviction of WCC from promoters. Transient phosphorylation of WCC (black dots) results in partial degradation of WC-1 (Corrochano, 2019). Introduction 18 4. The velvet regulatory system The velvet family of regulatory proteins consists of fungal-specific proteins that participate in the regulation of several essential processes, such as the coordination of sexual and asexual development, the production of secondary metabolites, pathogenicity, and the responses to stress (Bayram and Braus, 2012). Most of the current knowledge regarding the role of the velvet proteins derives from A. nidulans, where the protein VeA, the founding member of the group, was discovered five and a half decades ago (Käfer, 1965). All the velvet proteins have a velvet domain that binds DNA at specific sequences and presumably regulates gene expression acting as transcription factors (Ahmed et al., 2013). The velvet domain comprises approximately 150 amino acids and does not show sequence similarity to other DNA binding domains. However, it has been described to be structurally similar to the DNA binding fold of the mammalian transcription factor NF-kB (Ahmed et al., 2013). In addition, the velvet domain is a protein-protein interaction domain, and mediates the interaction with other velvet proteins and also with non-velvet proteins driving the formation of protein complexes essential for the regulatory role of this family of proteins (Bayram et al., 2008a; Bayram et al., 2010; Sarikaya-Bayram et al., 2014; Sarikaya-Bayram et al., 2015; Park et al., 2017). 4.1. VeA in A. nidulans A. nidulans has two major reproductive programs during its life cycle, which are coordinated by light. A. nidulans produces asexual spores called conidia that can easily disperse in the environment to form a new mycelium. Furthermore, it undergoes sexual development by forming spherical fruiting structures called cleistothecia, in which sexual spores, called ascospores, are generated by meiosis (Adams et al., 1998; Dyer and O’Gorman, 2012) .These two reproductive programs involve several developmental events that require the spatiotemporal regulation of gene expression, allowing intercellular communication and the differentiation of specialized cells (Han et al., 2008), that ultimately leads to the formation of the reproductive structures. A key factor required for the coordination and regulation of sexual and asexual development in A. nidulans is the velvet gene veA (Kim et al., 2002b; Calvo, 2008). 19 GENERAL INTRODUCTION VeA has been described as a negative regulator of asexual development and a positive regulator of sexual development and, notably, its function is light-dependent (Kim et al., 2002b; Calvo, 2008). In strains carrying the wild-type allele veA, light delays and reduces cleistothecia formation and the fungus develops asexually, but sexual development is promoted in the dark leading to the formation of the female structures (Mooney and Yager, 1990; Adams et al., 1998; Purschwitz et al., 2008). It was found that veA1 strains can undergo asexual development in the dark (Timberlake, 1990; Mooney and Yager, 1990). The corresponding gene was identified by complementation revealing that veA1 mutants carry a single-point mutation in the initiation codon (G to T at the third position) that results in a truncated gene product lacking the first 36 amino acids (using the 37th codon as initiation codon) (Kim et al., 2002b). The veA1 mutant strain displays increased conidiation, as well as a markedly delayed and reduced sexual development with fewer fruiting bodies being formed compared to a wild-type strain carrying the intact veA allele. This mutation was very convenient for daily laboratory work, and was introduced into most laboratory strains, allowing researchers worldwide to grow A. nidulans in dark incubators to generate asexual spores (Adams et al., 1998). Mutants carrying deletion of the gene veA display a different phenotype than the original veA1 point mutation. Mutants with a deletion of veA cannot produce any fruiting bodies even under favorable dark conditions, while overexpression of the gene veA leads to a constitutive production of cleistothecia in a lightindependent manner (Kim et al., 2002b). These results suggested an essential role for VeA in activating sexual development and inhibiting asexual development. The veA gene encodes a 573 amino acid protein that contains a velvet domain and a bipartite nuclear localization signal (NLS) in the N-terminal half, a nuclear export signal (NES) located in its N-terminal part, and a PEST motif found in the C-terminal end of the protein (Figure 0. 6) (Kim et al., 2002b; Stinnett et al., 2007). PEST regions are motifs rich in proline (P), glutamic acid (E), serine (S), and threonine (T), commonly found in unstable proteins (Rechsteiner and Rogers, 1996; Roth and Davis, 2000). PEST domains are typical for proteins with a high turnover rate and often become phosphorylated (Rechsteiner, 1990; Rechsteiner and Rogers, 1996). Subcellular localization of VeA depends on illumination. VeA is mainly located in the nuclei under dark conditions, whereas in light, the amount of VeA in the cytoplasm increases. In VeA1, the bipartite NLS domain is not functional and fails to respond to light, leading to significant increase in the amount of VeA in the cytoplasm independently of Introduction 20 illumination. This result suggests an important role for the NLS motif located in the N- terminus of VeA in the regulation of nuclear entry (Stinnett et al., 2007). Cell differentiation or development has long been associated with the biosynthesis of natural products in fungi. These natural products, also known as secondary metabolites, can be noxious as mycotoxins or beneficial as antibiotics, and many of them are of great interest in medicine, agriculture, and industry (Demain and Fang, 2000). In 2003, a connection was made between fungal control of secondary metabolism and development with VeA as a key factor necessary for the expression of genes involved in the synthesis of the mycotoxin sterigmatocystin and the antibiotic penicillin (Kato et al., 2003; Calvo, 2008). 4.2. Other members of the velvet family Four members of the velvet family have been identified in A. nidulans. The three proteins in addition to VeA include VelB (velvet like B), VelC (velvet like C), and VosA (viability of spores A) (Bayram and Braus, 2012). VelB has also been defined as a global regulator of development and secondary metabolism in A. nidulans. Strains carrying deletion of the gene veA or velB display similar but not identical defects in development and secondary metabolism (Bayram et al., 2008a; Park et al., 2012b). VelB is the shorter member of the family with 360 amino acids, and the only member that carries two velvet domains located in the N-and C- terminal half, respectively. No NLS or NES motifs are found in VelB (Figure 0. 6) (Bayram and Braus, 2012).VelB can interact with other VelB proteins as homodimers, and with VeA and VosA, as heterodimers. The VeA/VelB interaction is required for an efficient nuclear entry of VelB (Bayram et al., 2008a). VosA has a length of 430 amino acids with a velvet domain in the N-terminal half like VeA. Indeed, VeA and VosA constitute a monophyletic clade, and it was suggested that vosA appeared from an early duplication of veA (Ojeda-López et al., 2018). VosA also has a NLS motif that mediates nuclear localization, but does not support VelB transfer into the nucleus (Bayram et al., 2010). Furthermore, VosA contains a potential transcription activation domain (TAD) in the C-terminal half (figure 0.6). VosA is localized in the nucleus of mature conidia and has been suggested that this protein is a key factor controlling the final stages of sporulation and the biogenesis of trehalose. Trehalose is a specialized 21 GENERAL INTRODUCTION sugar required for the long-term survival of both conidia and ascospores (Bayram et al., 2010; Park et al., 2012b; Park et al., 2017). The protein VelC contains a long velvet domain in the C-terminal region. This protein has been defined as an important activator of sexual development. Deletion of velC causes an increase in the number of conidia and reduced cleistothecia production, while overexpression of the gene leads to increased cleistothecia (Park et al., 2014). Figure 0. 6. Domain architecture of the velvet family proteins in A. nidulans. Proteins VeA, VelB, VosA, and VelC. SAM, S-adenosyl methionine-binding site; TAD, transcription activation domain; PEST, proline (P) glutamic acid (E) serine (S) and threonine (T) rich sequence. The size of the proteins and the position of the different domains are represented to scale. Adapted from (Bayram and Braus, 2012). 4.3. The methyltransferase protein LaeA LaeA was the first methyltransferase that was associated with VeA. The gene laeA of A. nidulans was identified by Book and Keller (Book and Keller, 2004). This gene encodes a polypeptide of 375 amino acids that contains a S-adenosylmethionine-dependent (SAM) methyltransferase domain that is highly conserved in filamentous fungi (Bayram and Braus, 2012). LaeA is considered a global regulator of secondary metabolism and development and does not contain any velvet domain in its sequence. The sequence of LaeA includes an NLS domain in the N-terminal half, and the protein is located primarily VeA (AN1052) 100 aa 537 aa N C VelB (AN0363) VosA (AN1959) VelC(AN2059) 369 aa 430 aa 524 aa N C N C NC Introduction 22 in the nucleus (Figure 0. 7). Deletion of laeA revealed that LaeA is required for the expression of metabolic gene clusters of biotechnological interest, including the sterigmatocystin (carcinogen), penicillin (antibiotic) and lovastatin (antihypercholesterolemic agent) biosynthetic clusters (Bok and Keller, 2004). After the first report of LaeA as a regulator of secondary metabolism in A. nidulans, several studies of LaeA homologs have supported this role in other fungi (Amaike and Keller, 2009; Hoff et al., 2010; Wiemann et al., 2010; Crespo-Sempere et al., 2013; Estiarte et al., 2016). Figure 0. 7. Domain architecture of the LaeA methyltransferase protein in A. nidulans. SAM, S-adenosyl methionine-binding site; MTD, methyltransferase domain. The size and the position of the different domains are represented to scale. Adapted from (Bayram and Braus, 2012). In 2008, it was found that LaeA binds to VeA and VelB proteins to form a heterotrimeric complex, the velvet complex, that coordinates development and secondary metabolism (Bayram et al., 2008a). The discovery that LaeA interacts with velvet proteins further extended the role of LaeA beyond secondary metabolism. In A. nidulans, in addition to its function in secondary metabolism, LaeA is required for the light-dependent formation of asexual spores. In the presence of light, LaeA promotes conidiation while reducing VeA and VelB levels within the nucleus (Bayram et al., 2010). The fungal responses of strains ∆veA and ∆laeA are antagonistic. Strains carrying deletion in laeA with an intact VeA protein show a strong reduction in asexual spore formation and are unable to suppress sexual development by light. In contrast, strains that carry deletion of veA with an intact LaeA protein constitutively perform the asexual program and do not produce sexual fruiting bodies (Bayram et al., 2010). There are several studies that indicate that LaeA together with velvet proteins play a role in the regulation of cell morphogenesis, development, and pathogenesis in several fungi. In Aspergillus flavus, strains lacking laeA show altered conidial development and are unable to produce sclerotia, while overexpression of laeA increases sclerotial formation (Kale et al., 2008). Penicillium chrysogenum strains with deletion of laeA show a severe impairment in conidiophore formation under both light and dark conditions (Hoff et al., LaeA (AN0807) 374 aa NC 23 GENERAL INTRODUCTION 2010). Deletion of the lae1 gene in Fusarium fujikuroi leads to a reduction in pigmentation and microconidia formation compared to the wild-type strain (Wiemann et al., 2010). The homologue of laeA in Trichoderma reesei, the gene lae1, is essential for conidiation among other processes such as cellulase gene expression (Seiboth et al., 2012). All these results support the idea that LaeA along with the velvet complex is a key global regulator of morphogenesis and development in fungi. Additionally, other methyltransferases have been identified as interacting partners of VeA. Three S-adenosyl L-methionine (SAM)-dependent methyltransferases have been shown to physically interact with VeA. One of them is the Lae-like methyltransferase LlmF, the product of one of the nine genes identified as Lae-like methyltransferase (llmA-llmJ) in A. nidulans (Palmer et al., 2013). LlmF was found to control ST production, as well as sexual development (Palmer et al., 2013). 4.4. The trimeric VelB/VeA/LaeA (velvet) complex and its mechanism of action The mechanism of action of velvet proteins in A. nidulans is summarized in Figure 0. 8). In the dark, VeA is primarily expressed and enriched in the nucleus. In the cytoplasm VeA interacts with VelB and supports the entry of VelB into the nucleus through the importin KapA (Bayram et al., 2008a). The formation of a VelB/VeA heterodimer is required for regulation of sexual development in the dark (Bayram et al., 2008a). The VeA/VelB heterodimer interacts with LaeA in the nucleus, and it has been proposed that this interaction helps to coordinate development and support the epigenetic role of LaeA regulating the expression of clusters of secondary metabolite genes (Bayram et al., 2008a; Bayram et al., 2010). VelB has a second role as part of the VosA/VelB heterodimer, repressing asexual differentiation during development in the absence of light. Furthermore, VosA/VelB plays a role in spore viability and trehalose biogenesis (Bayram et al., 2010; Park et al., 2012b; Park et al., 2017). After exposure to light, VeA is mostly retained in the cytoplasm, and thus, the amount of the trimeric velvet complex in the nucleus is reduced, leading to the suppression of fruiting body formation and the sterigmatocystin production (Bayram et al., 2008a). Light also drastically reduces the levels of VelB and VosA during development, which is consistent with increased conidiation. VelB is also able to form homodimers with other VelB proteins, but the function is currently unknown (Bayram et al., 2010; Bayram and Braus, 2012). 30 31 MATERIALS AND METHODS 32 33 MATERIALS AND METHODS MATERIALS AND METHODS 1. Strains and primers used in this thesis Table MM.1 list of strains used in this thesis Strain Genotype Source N. crassa FGSC#2489 4-OR23-1V mat A FGSC FGSC#2490 ORS-SL6a mat a FGSC FGSC#9719 ∆mus-52::bar mat a FGSC FGSC#11401 ve-1KO mat A FGSC FGSC#11400 ve-1KO mat a FGSC FGSC #13050 ve-2KO mat a FGSC FGSC #13536 vos-1KO mat A FGSC FGSC #22596 lae-1KO mat a FGSC FGSC #22596 lae-1KO mat a FGSC FGSC #11044 flKO mat a FGSC FGSC #11308 vib-1KO mat a FGSC ∆ve-1 ∆wc-1 ve-1KO wc-1KO Gil-Sánchez et al., 2022 Δve-1; Δmus-52; his-3 ve-1KO; delta mus-52::bar; his-3 mat A This study Δve-2; Δmus-52; his-3 ve-2KO; delta mus-52::bar; his-3 mat A This study SC-∆ve-2 ve-2KO mat A This study SC-∆vos-1 vos-1KO mat a This study SC-∆lae-1 lae-1KO mat A This study VE-1::3xFLAG ve-1::10xGly3xFLAG Gil-Sánchez et al., 2022 VE-2::3xFLAG ve-2::10xGly3xFLAG This study LAE-1::3xFLAG lae-1::10xGly3xFLAG This study SC- ∆ve-1 LAE- 1::3xFLAG ve-1KO lae-1::10xGly3xFLAG This study VE-1::3xFLAG; VE-2::3xHA ve-1::10xGly3xFLAG, ve- 2::10xGly3xHA This study VE-1::3xFLAG; LAE-1::3xHA ve-1::10xGly3xFLAG, lae- 1::10xGly3xHA This study VE2::3xFLAG ; LE1::3xHA ve-2::10xGly3xFLAG, lae- 1::10xGly3xHA This study SC- ∆ve-1∆ve-2 ve-1KO ve-2KO This study Materials and methods 34 SC- ∆ve-1∆vos-1 ve-1KO vos-1KO This study SC- ∆ve-2∆vos-1 ve-2KO vos-1KO This study SC- ∆ve-1∆lae-1 ve-1KO lae-1KO This study SC- ∆ve-2 ∆lae-1 ve-2KO lae-1KO This study SC- ∆ve-1∆vib-1 ve-1KO vib-1KO This study SC- ∆ve-2 ∆vib-1 ve-2KO vib-1KO This study SC- ∆ve-1∆fl ve-1KO flKO This study SC- ∆ve-2∆fl ve-2KO flKO This study S. cerevisiae FY834 MAT a his3∆200 ura3-52 leu2∆1 lys2∆202 trp1∆63 E. coli DH5 a huA2 lac(del)U169 phoA glnV44 Φ80' lacZ(del)M15 gyrA96 recA1 relA1 endA1 thi-1 hsdR17 (Taylor et al., 1993) ER2566 F−λ−fhuA2 ompT lacZ::T7gene1 gal1 sulA11 Δ(mcrC-mrr)114::IS10(R9mcr- 73::mini-Tn10-Tets) 2R(zgb-210::Tn10) (Tets) endA1 NEW ENGLAND BioLabs Table MM.2 list of primers used in this thesis. Primers Sequence 5'-3' Use con10-45F CAGCCACAGCGGAGGC RT-PCR con10-104R TTGGAGCAATTTCGCGC RT-PCR ve1-939F CGGAGGCCATCTGGGTTT RT-PCR ve1-998R GGTGGTTGAGCGGGATACTG RT-PCR ve-2-86F CGCGTCATCACCAATTGC RT-PCR ve-2-187R ACATGGTGGGATCTTGGTATGG RT-PCR vos-1-460F GCACAGCCCTCAGTCTAACCA RT-PCR vos-1-560F CACTCCGGTACCATGCAACA RT-PCR lae-1-33 F CCAGGACAGTTATCCGTCTTACAGA RT-PCR lae-1-133 R CCCCATAGAAGCGTCCGTTT RT-PCR RTcsp-2F AGGTTCAACTCGAGCGATGCT RT-PCR RTcsp-2R GACTCAATCCAGCCCGTCAA RT-PCR acon3-771F GCAACGTGCCCCTAACTTTG RT-PCR acon3-832R TGGTACCCGCGCAATTTC RT-PCR acon-2-76F TCCGAGGACCGCTGCAT RT-PCR acon-2-145R AGTTTGGTTCCGTCTTTGCAA RT-PCR fl-811F GGCGATTCCCGCTACTGTT RT-PCR fl-869R TTGCAGGCCTTTCCCAAA RT-PCR fld-1275F CGCCAACGCCCAGAAA RT-PCR 35 MATERIALS AND METHODS fld-1350R GCAGCGCATGACAACGAA RT-PCR vib-740F CCACCCAACTCACCAGAAGT RT-PCR vib-842R GTGATCTCCGCCATTGTCTT RT-PCR tub-1073F CCCGCGGTCTCAAGATGT RT-PCR tub-1139R CGCTTGAAGAGCTCCTGGAT RT-PCR 5’UTR-ve1-sc TCATACCCGTCGCTCTGGTTTCTGG PCR 3’ve1int-sc GTTGGCTCGTGGGGTAGACGG PCR 5’UTR-ve1-sc AACCTGATTATGCCAGGCCGAT PCR ve-2-5'F-ORF-sc ATGCAACCGTACACTTCAAACC PCR ve-2-3'R-ORF-sc CCG CCT TAATAATCATCATCATCC PCR 3’UTRve-2R-sc CCGCCTGCGCCTTACCTCC PCR 5’ve-2intF-sc. TGGTCAGGGCGGACAGCA PCR vos-1-5'F-ORF-sc ATGGCAACAGCATCGCCGCC PCR vos-1-3'R-ORF-sc GGC CTT ACT TTG GGT CGG C PCR 5’UTRlaeA-sc CGA CCG CTA TCT AGC ACG TT PCR 3’UTRlaeA-sc TCG GAC AGA ACC CAC CAA AG PCR 3’lae1int-sc ACCCAATCGATTTCCACCTGCTC PCR hygB-R CGTGCACGCGGATTTCGGCTCC PCR hygB-1-sc GGTCAATACACTACATGGCGTG PCR hygB-2-sc CACGCCATGTAGTGTATTGACCG PCR F-ve-1-Inteína TAG-sc GGTCATATGATGGGTGCTCAGGTTATCG PCR R-ve-1-Inteína TAG-sc CGC TGCTCTTCCGCA ATACCCGCCAATATCTGC PCR F-ve-2-Inteína TAG-sc GGTCATATGATGCAACCGTACACTTCAA PCR R-ve-2-Inteína TAG-sc GGC TGCTCTTCCGCA ATAATCATCATCATCCTCATTA PCR 5'F mak-1Region1- 6FAM CATCATTCGTCAGGAACCGACAAG PCR 3'R mak-1Region1 TGGCGAATGTGTCTGTGTGGTGT PCR 5'F mak-2Region1- 6FAM CAGTCTTGCTGCTGAGCCGG PCR 3'R mak-2Region1 TTCGACGTGTCCCGAAAAGTTGC PCR 5'F mek-2Region1- 6FAM GCTGCCAAGATACGTTAAGACAAAGG PCR 3'R mek-2Region1 GATGTCTGTTTGTTTAAGCGTAGTCG PCR 5'F os-4 Region2- 6FAM GACCGGGACTGGCAGTGAC PCR 3'R os-4 Region2 TGTCGCCGACAAGTCAAAGG PCR Materials and methods 36 2. Media and culture conditions 2.1. Media and culture conditions for Neurospora crassa The medium used for the growth of Neurospora was Vogel minimal medium (Vogel H, 1956; Vogel, 1964) with different carbon sources depending on the experiment. The stocks required for this media are listed below: • Vogel Salts 50X: 125 g Na3C6H5O7 · 3H2O, 250g KH2PO4 anhydrous, MgSO4 3 7H2O, 100 g NH4NO3 anhydrous, 5 g CaCL2 · 2H2O, 5 ml of biotin solution and 5ml of trace elements solution were sequentially dissolved in 755 ml of distilled water. Then it was filled up to 1 liter and 1-2 ml of trichloromethane were added as preservative. CaCl2 was dissolved previously in distilled water and was added slowly to salts solution to avoid its precipitation. • Biotin Solution: 5mg biotin was dissolved in 100 ml ethanol 50%. It was stored at -20ºC. • Trace elements: 5g citric acid· 2H2O, 5g Zn SO4 ·7 H2O, 1 Fe (NH4)2(SO4)2· 6H2O, 0.25 g CuSO4 · 5H2O, 0.05 g H3BO3 and 0.05 Na2MoO4 · 2 H2O were sequentially dissolved in 95 ml of distilled water. It was stored at 4ºC protected from light. 2.1.1. Liquid media cultures For cultures in liquid media, we used petri dishes or flasks. This type of culture was used for DNA, RNA, and protein extractions from vegetative mycelia. Tween 80 was used to avoid the formation of aerial hyphae and conidia, allowing mycelia to develop vegetatively. Cultures inoculated with a conidial suspension or with mycelia pads (when the strain is unable to develop asexually) were incubated at 22ºC, 30ºC or 34ºC for 24 / 48 h depending on the requirements of the experiment. This growth media consists of: - Minimal medium sucrose: 1X Vogel salts, 2%[w/v] sucrose, 0.2%[v/v] Tween 80. 37 MATERIALS AND METHODS 2.1.1.1. Liquid cultures with cycloheximide To assay protein stability, strains were cultured in a Petri dish with 20 ml of liquid Vogel without shaking for 24 h at 34°C. Once the mycelium was grown, discs of mycelium were clipped with a sterile 15 ml falcon and transferred to a 500 ml flask containing 200 ml of Vogel´s liquid medium that was incubated under agitation at 30ºC for another 24 h in dark, light, or exposed to 30 min of light. Then, cycloheximide was added to the cultures, and samples (mycelium discs) were removed at different times. Cycloheximide was used at a concentration of 0.1 mg/ml and thiolutin at 50μM (Figure MM. 1). Figure MM. 1. Culture condition for the analysis of protein stability in vegetative mycelia. Cycloheximide (CHX) was added to inhibit protein synthesis. Modified from (Maria del Mar Gil thesis, 2017). 2.1.2. Solid media cultures Solid cultures were used for strains propagation, production and collection of conidia, colonial growth, to plate transformations, crossing strains and induction of development (conidiation and sexual development). Materials and methods 38 2.1.2.1. Strains propagation Strains were grown and maintained on slants tubes containing 3-5 ml of Vogel´s solid medium. Slants tubes were made by pouring melted medium and tilting the tubes before they solidified to get a thin layer of medium in the highest area of the tube to favor conidiation. Inoculated slants tubes were incubated at 34º C for two days and then allowed to develop asexually, being exposed to dark/light cycles on the bench for a minimum of five days. Medium supplemented with hygromycin was used for strains carrying a copy of the gene hph conferring hygromycin resistance through its inactivation by phosphorylation. This growth media consists of: - Minimal medium sucrose: 1X Vogel salts, 2%[w/v] sucrose, 1.5%[w/v] agar. 2.1.2.2. Collection of conidia Approximately 50 ml of melted medium was added to 250 ml flask before agar solidification. The medium was spread on the flask walls to get a thin layer of medium to favor conidiation. Once inoculated, the cultures were incubated at 34º C for two days and then allowed to develop asexually, being exposed to dark/light cycles on the bench at room temperature. Conidia were recovered adding 50ml of 1M sorbitol and shaking flasks for 5 min. Then, conidial suspension was filtered through cheesecloth. Conidia were quantified in a Burker chamber and conidial stocks were stored at -20ºC before use. For long-term maintenance, conidia were stored at -80ºC. This growth media consists of: - Minimal medium sucrose: 1X Vogel salts, 2%[w/v] sucrose, 1.5%[w/v] agar. - Minimal medium sucrose with hygromycin B: 1X Vogel salts, 2%[w/v] sucrose, 1.5%[w/v] agar, 200 µg/ml Hygromycin B (Roche). 2.1.2.3. Colonial growth Colonial growth in N. crassa is achieved by adding sorbose as the main carbon source on the medium. Neurospora is unable to use sorbose as a source of nutrient for growth, so the mycelium has a limited spread, and the production of conidia is reduced. This medium was used for the germination of ascospores (meiotic products from sexual crosses). The 39 MATERIALS AND METHODS medium was supplemented with hygromycin for the selection of resistant transformants with a copy of the gene hph. - FGS solution 10X: 10% [w/v] sorbose, 0.5%[w/v] glucose, 0.5%[w/v] fructose. - Minimal medium with sorbose: 1X Vogel salts, 1X FGS solution, 1.5%[w/v] agar. - Minimal medium with sorbose and hygromycin B: 1X Vogel salts, 1X FGS solution, 1.5%[w/v] agar, 300 µg/ml Hygromycin B (Roche). 2.1.2.4. Synthetic cross medium (SCM) Strains were inoculated by spreading a suspension of conidia into nitrogen-poor Synthetic Crossing Medium (SCM) agar´s plates (Westergaard and Mitchell, 1947) to stimulate sexual development and mating between strains of opposing mating type (mat A and mat a). Plates were incubated at 22°C or room temperature for a minimum of two weeks until ascospores were released. Ascospores were recovered from the petri dish lid with water and heat activated (60ºC, 45 min) prior to being cultured in FGS medium plates. - Westgard Salts 2X: 1.4 g K2HPO4, 1 g KH2PO4 anhydrous, 1 g MgSO4 · 7H2O, 0.2 g NaCl, 2 g KNO3, 0.2 g CaCl2 · 2H2O, 0.2 ml of biotin solution and 0.2 ml of trace elements solution were sequentially dissolved in 950 ml of distilled water and then, filled up ap to 1 liter. 2 ml of trichloromethane were used as a preservative. It was stored at room temperature. - SCM: 1X Westergaard Salts, 1% [w/v] sucrose, 1.5% [w/v] agar. 2.1.2.5. Induction of asexual development Induction of asexual development was carried out using the method described by Baley-Shrode and Ebbole (Bailey-Shrode and Ebbole, 2004) with some minor modifications. A total of 107 conidia were inoculated in 200 ml of Vogel’s liquid medium in 500 ml flasks and incubated for 24 hr at 30ºC under agitation. Vegetative mycelia were collected by filtration onto filter paper, covered by an additional layer of filter paper, and placed on top of 25 ml Vogel’s solid medium in 90-mm plates to induce conidiation. Cultures were incubated at 30ºC for 15 h or 24 h to allow the aerial hyphae to develop and grow through the filter paper. Samples of the aerial tissue (aerial hyphae and developing conidia) located on top of the filter papers, and the supporting vegetative mycelial mat Materials and methods 46 h at 37 ºC with moderate shaking. After this time, 100 µl of cell suspension were plated on a Petri dish with 25 ml of LB solid medium with the appropriate antibiotic and the rest was centrifuged and the pellet was spread on a different plate. Plates were grown for 12 h at 37 ºC. This method was normally used to amplify vectors. 3.7.2. Electroporation method The electroporation method was used for plasmids isolated from yeasts minipreps. A single colony of E. coli DH5a was used to inoculate 25 ml of LB medium that was cultured for 12 h at 37 ºC with shaking. The culture was diluted to 0.1 OD600 in 25 ml of LB medium and incubated at 37 ºC until it reached 0.4 OD600. Cells were then harvested and washed twice by centrifugation and resuspended on 500 µl of bi-distilled water. Around 40 µl of the cell suspension were mixed with the DNA and placed on ice for 5 min before being transferred to an electroporation cuvette. Electroporation was performed with a BTX Electro Cell Manipulator (Model ECM 630) instrument at 2500V, 200Ω, and 25μF. After the electro-pulse 1 ml of cold 1 M sorbitol was immediately added to the cuvette and 100 µl of this cell suspension were spread on al solid LB plate with antibiotic. The rest of the cells were centrifuged, and the pellet was placed on another LB plate. Plates were grown for 12 h at 37 ºC. 4. Isolation and manipulation of RNA 4.1. Isolation of RNA 4.1.1. RNA purification from vegetative and aerial mycelia Mycelia (vegetative and aerial tissue) were disrupted by two 0.5 min pulses in a cell homogenizer (FastPrep-24, MP Biomedicals) with 1.5 g of zirconium beads (0.5 mm diameter) in 1.9 ml screw-cap tubes by using the RNeasy Plant Mini Kit (Qiagen) with the RLC buffer following the manufacturer’s procedure. The extracts in screw-cap tubes were clarified by centrifugation (13,000 rpm) for 5 min prior to RNA purification. The RNA samples were treated with DNase I (USB) prior to use. 47 MATERIALS AND METHODS In an alternative method, mycelia were ground into a fine powder with liquid nitrogen in a mortar with a pestle. RNA was isolated from the powder using the RNeasy Plant Mini Kit (Qiagen) with the RLC buffer following the manufacturer procedure. RNA quantity and quality was analyzed using Nano-drop ND-1000 spectrophotometer and agarose gel electrophoresis. The RNA samples were stored at -80 ° C until use. 4.1.2. Total RNA purification from samples collected during sexual development Tissue samples from biological replicates were pooled for RNA extraction at each sample point. Mycelia was grinded using a precooled mortar and pestle, homogenized with TRI REAGENT (Invitrogen), and then phase separated by adding chloroform. The nucleic acid phase was precipitated with isopropanol, washed with 75% ethanol, and solubilized in RNase-free water. RNA clean-up and DNase treatment was performed using Nucleo-Spin RNA (MACHEREY-NAGEL) and RNase-free DNase Set (QIAGEN), following the manufacturer procedures. RNA quantity and quality were analyzed using Nano-drop ND- 1000 spectrophotometer and agarose gel electrophoresis. The RNA isolated using this procedure is suitable for applications such as Northern blotting, poly(A)+ mRNA selection, and RT-qPCR. The RNA samples were stored at -80 ° C before use. 4.2. Quantitative RT-PCR The primers used for RT-qPCR are detailed in Table S2. RT-qPCR experiments were performed using one-step RT-PCR, in a LightCycler 480 II instrument (Roche) by using the One Step SYBR® PrimeScript™ RT-PCR Kit (Takara Bio Inc.), 0.2 µM of each primer and 12.5 ng of RNA in a 10 µl reaction. The reaction consisted of 5 min at 42º C, followed by 10 s at 95º C, and then 40 cycles of DNA amplification (5 s at 95º C and 20 s at 60ºC). After each PCR we performed a melting curve analysis to show the specific amplification of single DNA segments and the absence of nonspecific amplified DNA. The fluorescent signal obtained for each gene was normalized to the corresponding fluorescent signal obtained with tub-2 to correct for sampling errors. Materials and methods 48 4.3. RNA seq analysis 4.3.1. Asexual development (conidiation) experiments 4.3.1.1. Culture conditions For the light induction experiments in vegetative mycelia, 106 conidia were inoculated on 25 ml Vogel’s liquid medium in 90-mm plates, which were incubated in complete darkness at 22ºC for 48 h. The cultures were then exposed to light for 30 min. A control treatment was always kept in the dark. Vegetative mycelia were collected, dried on paper, frozen in liquid nitrogen, and stored at -80ºC. For the induction of conidiation and total RNA isolation from vegetative and aerial mycelia we followed the method previously explained in sections 3.1.2.5. and 5.1.1 of material and methods. 4.3.1.2. cDNA libraries preparation and statistical analysis Total mRNA purification by Poly-A selection and cDNA libraries preparation was done according to the Illumina mRNA Sample Preparation Guide. Sequencing was performed on Illumina Genome Analyzer (Illumina HiSeqV4) by the Next Generation Sequencing Facility at Vienna BioCenter Core Facilities (VBCF), Austria. Single endreads of 100 nucleotides each were mapped against the N. crassa OR74A genome (NC12.37) obtained from JGI mycocosm web portal https://mycocosm.jgi.doe.gov/Neucr2/Neucr2.home.html (Grigoriev et al., 2014) by using Tophat 2.11 (Kim et al., 2013). The read quality was checked using FastQC 0.11.8 (http://www.bioinformatics.babraham.ac.uk/projects/fastqc/) and automatic preprocessing of the reads was carried out with AFTERQC 0.9.6 (Chen et al., 2017). SeqMonk version 1.45.4 (https://www.bioinformatics.babraham.ac.uk/projects/seqmonk/) was used for visualization and quantification of BAM files. Transcript features were generated over genes. Quantification was performed using the RNAseq quantification pipeline on merged transcripts counting reads over exons, assuming opposing strand specificity. Differential gene expression analysis was done by DESeq2 (Love et al., 2014). Genes with Log2FC ≥ 2 or ≤ -2 and adjusted P values below 0.001 after Benjamin and Hochberg correction were labelled as differentially expressed genes (DEGs) for comparison between conditions. Three different conditions were analyzed for each strain (Figure II.3A): genes regulated by 49 MATERIALS AND METHODS light during vegetative growth (Light regulation experiment), and genes regulated by conidiation (Developmental regulation experiment), either in complete darkness (A15 Dark), or developed in the dark and then followed by 30 min of light exposure (A15 Light). In addition, differential gene expression analysis was carried out to examine pairwise differences between wild type and ∆ve-1 in different developmental conditions. Genes showing significant differences in relative expression levels (DEGs) were classified according to their known or putative gene information in the integrated functional genomic database Fungi DB (Stajich et al., 2012). All the genes selected were used to create VENN diagrams using http://www.venndiagrams.net/, followed by GO analysis using FungiFun2 database (Priebe et al., 2014) for enrichment of GO terms into biological processes, cellular components, and molecular functions. The significantly associated gene ontology terms (adjusted p-value <0.05) were imported to REVIGO where they were clustered based on their relatedness and any redundancy was removed (Supek et al., 2011). Heat maps were generated with log2 TPM (Transcripts Per Kilobase Million) values using ClustVis (Metsalu and Vilo, 2015). 4.3.2. Sexual development experiments 4.3.2.1. Culture conditions For this experiment, strains of complementary mating type, ∆ve-1 mat A and wild type mat a (Table MM.1) were used. Induction of formation of protoperithecia by ∆ve-1 and then maturation of perithecia after fertilization were performed as described in section 3.1.2.6 of materials and methods. Samples were collected 2 and 1 day prior to fertilization, immediately before fertilization (day 0) and 1, 2, 4, and 6 days after fertilization (Figure III. 9A). For transcriptomic analysis all tissues were immediately frozen in liquid nitrogen and stored at -80°C. ARN purification method is described in section 5.1.2. of materials and methods. 4.3.2.2. cDNA libraries preparation and statistical analysis The mRNA quality control and the preparation of the strand-specific cDNA library was performed according to the Illumina mRNA sample preparation guide. Sequencing Materials and methods 50 was performed on Illumina Genome Analyzer (Illumina NovaSeq 6000) by the Next Generation Sequencing Facility Novogene. Pair-end reads of 150 nucleotides each (firststrand PE150) were mapped against the N. crassa OR74A genome (NC12.37) obtained from JGI mycocosm web portal https://mycocosm.jgi.doe.gov/Neucr2/Neucr2.home.html (Grigoriev et al., 2014) using STAR 2.7 (Dobin et al., 2013). The quality of the read was checked using FastQC 0.11.9 (http://www.bioinformatics.babraham.ac.uk/projects/fastqc/). SeqMonk version 1.47.2 (https://www.bioinformatics.babraham.ac.uk/projects/seqmonk/) was used for visualisation and quantification of BAM files. Transcript features were generated over genes. Quantification was performed using the RNA seq quantification pipeline on merged transcripts counting reads over exons, assuming opposing strand specificity. The identification of genes showing differences in gene expression levels in the time course of sexual development was done using maSigPro (Conesa et al., 2006; Ia Jos E Nueda et al., 2014). This method first adjusts a global regression model with all defined variables to identify differentially expressed genes, and then a variable selection strategy is applied to study differences between groups. Genes with differences in mRNA accumulation in at least one time point during sexual development (FDR=0.001) were labelled as differentially expressed genes (DEGs). Two different experimental conditions (Dark and Light) were analyzed for each strain (∆ve- 1 and wild-type strains). Then, a second strategy was applied to find statistically significant different profiles between dark and light in each strain. In addition, a third analysis was done to study differences in expression profiles between the wild-type and the ∆ve-1 strains during sexual development either in complete darkness (WT vs ∆ve-1 dark) or in continuous light (WT vs ∆ve-1 light). Genes showing significant differences in their expression profile (DEGs) were classified according to their known or putative gene information in the integrated functional genomic database Fungi DB (Stajich et al., 2012). All the genes selected were used to create VENN diagrams using http://www.venndiagrams.net/, followed by GO analysis using Fungi DB database for enrichment of GO terms into biological processes. The significantly associated gene ontology terms (adjusted p-value <0.05) were imported to REVIGO where they were clustered based on their relatedness and any redundancy was removed (Supek et al., 2011). Heat maps were generated with CPM (Counts Per Million) values using 51 MATERIALS AND METHODS GraphPad Prism 8.2.1 (Macintosh version) (GraphPad Software, San Diego, California USA, www.graphpad.com). 5. Protein handling 5.1. Preparation of total protein extracts Mycelial pads were harvested, frozen in liquid nitrogen, and ground to a fine powder using a mortar and pestle. Powdered tissue was mixed in a microcentrifuge tube with different volumes of PEB buffer containing protease and phosphatases inhibitors (when indicated) and dissolved by vortexing. Vortexing pulses were followed by incubations of 5 to 10 min on ice. After mixing, crude extracts were centrifuged at 13,000 rpm and 4 ºC. Supernatants were taken, frozen in liquid nitrogen and stored at -80 ºC until used. Protein concentrations were measured by spectrometry using a Nanodrop instrument (Abs 280 nm) or a Bradford assay (Bradford, 1976). - PEB: 50 mM HEPES pH 7.4, 10% [v/v] glycerol, 137 mM NaCl, 5 mM EDTA. - Protease inhibitors: leupeptin (1 µM), pepstatin (1 µM), and PMSF (50 µM). - Phosphatase inhibitors: PhosStop (Roche). 5.2. Western blot Proteins from total cell lysates, cytoplasmic, nuclear or immunoprecipitated fractions were subjected to SDS-PAGE on 7.5% or 10% acrylamide:bisacrylamide (29:1) gels. Samples were resuspended in Laemmli buffer 1X (Laemmli, 1970) and boiled for 5 min before loading into gels. Proteins were semidry transferred to nitrocellulose blotting membranes (GE Healthcare). As molecular weight marker the Precision PlusTM Standards All Blue from Bio-Rad was used. Equal loading was confirmed by staining membranes with Ponceau S solution. Membranes were washed out for 15 min prior to blocking using TBS 1X. Blocking was carried out using blocking solution for at least 1 h with moderate shaking at room temperature prior to incubation with primary antibodies (Table MM.3). Secondary antibodies used were conjugated with the horseradish peroxidase. Peroxidase activity was detected using chemiluminescence (ECL Prime Reagent GE Healthcare). Materials and methods 52 - Laemmli Buffer (4X): 240 mM Tris-HCl pH 6.8, 8% [p/v] SDS, 40% [v/v] glycerol, 4% [v/v] β-mercaptoethanol, 0.02% [p/v] bromophenol blue. - Stacking gel: 4% [p/v] acrylamide:bisacrylamide (29:1), 125 mM Tris-HCl pH 6.8, 0.1% [p/v] SDS, 0.5% [v/v] APS 10% [p/v], 0.2% [v/v] TEMED. - Running gel: 7.5%, 10% [p/v] acrylamide:bisacrylamide (29:1 or 30:02), 375 mM Tris HCl pH 8.8, 0.1% [p/v] SDS, 0.5% [v/v] APS 10% [p/v], 0.2% [v/v] TEMED. - SDS-running buffer (10X): 0.4 M Tris base, 1.2 M Glycin, 1% [p/v] SDS. - Semi-dry transfer buffer: 20 mM Tris base, 150 mM Glycin, 20% [v/v] methanol, 0.05% [p/v] SDS. - Ponceau S solution: 0.1% [p/v] Ponceaus S, 3% [p/v] TCA. - TBS (10X): 1.5 M NaCl, 33 mM Tris base. Adjust to pH 7.5. - Blocking solution: 5% [p/v] Skim Milk (Difco), TBS 1X. Table MM.3. Antibodies used in this thesis 5.3. Coomassie staining of polyacrylamide gels Acrylamide gels were stained using Coomassie solution for 1 h at room temperature. To visualize the proteins, gels were then submerged in destaining solution while shaking for at least 1 h at room temperature. Antibody Catalog number Type Origin Use FLAG F3165(Sigma) Monoclonal Mouse Western blot, IP HA 901513 (Biolegend) Polyclonal Mouse Western blot hH3 ab1791 (abcam) Polyclonal Mouse Western blot NOP-1 MA1-10025 (TermoFisher) Monoclonal Mouse Western blot TUB sc-32293 (Santa Cruz) Monoclonal Mouse Western blot Ubiquitine sc-8017(Santacruz) Monoclonal Rabbit Western blot a MOUSE #170-6515 (BioRad) Monoclonal Goat Western blot a RABBIT #172-1011 (BioRad) Monoclonal Goat Western blot - Coomassie Staining Solution: 45% [v/v] methanol, 10% [v/v] acetic acid, 0.1% [p/v] Coomassie Blue. - Destaining Solution: 5% [v/v] methanol, 7.5% [v/v] acetic acid. 53 MATERIALS AND METHODS 5.4. Cell fractionation and nuclei isolation The total cell lysates and nuclei were prepared according to a method described by Baum and Giles with some minor modifications (Baum and Giles, 1985; Schwerdtfeger and Linden, 2000; Froehlich et al., 2002). Culture conditions depend on the experiment. Mycelia were fixed by adding 0.1% formaldehyde for 15 min, and then the reaction was stopped by adding glycine (125mM) prior to collection. Mycelial pads (4-5 g wet weight) were harvested, vacuum dried, frozen in liquid nitrogen, and ground to a fine powder with liquid nitrogen using a mortar and pestle. Powdered tissue was mixed with 8 ml of buffer A. The crude extract was filtered through a cheesecloth and 2 vol of buffer B were added to the filtrate while stirring. The homogenate was then layered onto 10.4 ml of buffer C in a Ultra-Clear tube (25x89 mm, Beckman), and centrifuged at 3,000 g for 10 min at 4°C in a SW32Ti rotor (Beckman) to remove cell debris. Aliquots from the supernatant were taken (1 ml, total cell lysate), frozen in liquid nitrogen and stored at -80 ºC. The remain supernatant was then layered onto a 5 ml step sucrose gradient and centrifuged at 9,400 g for 30 min at 4 ºC in a SW32Ti rotor. Aliquots from the supernatant were taken (cytoplasmic fraction), frozen in liquid nitrogen and stored at -80 ºC. The resulting nuclear pellet was resuspended in 0.5 ml of buffer D or nuclei lysis buffer (depending on the experiment), frozen in liquid nitrogen and stored at -80 ºC. All buffers contained protease inhibitors. Total, cytosolic, and nuclear fractions were used for western blot. Protein concentrations were determined by Bradford assay (Bradford, 1976). - Buffer A: 1 M sorbitol, 7% [w/v] Ficoll®, 20% [v/v] glycerol, 5 mM magnesium acetate, 5 mM EGTA, 3 mM CaCl2, 3 mM DTT, 50 mM Tris-HCl, pH 7.5. - Buffer B: 10% [v/v] glycerol, 5 mM magnesium acetate, 5 mM EGTA, 25 mM Tris-HCl, pH 7.5. - Buffer C: Composed of a 1:1.7 mix of buffers A and B. - Buffer D: 25% [v/v] glycerol, 5 mM magnesium acetate, 3 mM DTT, 0.1 mM EDTA, 25 mM Tris-HCl, pH 7.5. - Nuclei lysis buffer: 50 mM Tris-HCl pH 7.5, 300 mM NaCl, 1.5 mM MgCl2, 0.5% [v/v] NP-40. - Protease inhibitors: 1 µM leupeptin, 1 µM pepstatin and 50 µM PMSF. Materials and methods 54 - Step sucrose gradient: 1 M sucrose, 10% [v/v] glycerol, 5 mM magnesium acetate, 1 mM dithiothreitol, 25 mM Tris-HCl, pH 7.5. - Protease inhibitors: 1 µM leupeptin, 1 µM pepstatin and 50 µM PMSF. 5.5. Immunoprecipitation of tagged proteins Mycelial pads (8-10 g wet weight) were harvested, frozen in liquid nitrogen, and ground to a fine powder under liquid nitrogen using a mortar and pestle. Powdered tissue was mixed in a 50 ml tube with 20 ml of IPB buffer with protease inhibitors and dissolved by vortexing. Vortexing pulses were followed by incubations of 5 to 10 min on ice. After mixing, crude extracts were centrifuged at 14,300 g and 4 ºC for 20 min (rotor JA-10 or JA-20 Beckman). Supernatants were collected and transferred to centrifuge tubes. Then they were centrifuged again at 169,000 g and 4 ºC for 45 min (rotor Sorvall T-647.5 or Beckman 70Ti). A fraction (1 ml) of each supernatant was transferred to a new tube, marked as each “input” sample and frozen in liquid nitrogen. Previously, 300 µl of M2-FLAG agarose beads (Sigma) were equilibrated by washing them three times with 1 ml of IPB buffer followed by centrifugation at 1,000 g and 4 ºC. After equilibration they were added to each remaining supernatant and incubated in a rotator wheel at 4 ºC for 3 h. Beads were then collected by centrifugation at 1,500 g at 4 ºC for 10 min (rotor SX4250). A fraction (1 ml) of each supernatant was transferred to a new tube, marked as each “flowthrough” (FT) sample and frozen in liquid nitrogen. Beads pellet was washed first with 30 ml of IPB buffer by gentle rotation at 4 ºC for 3 min and finally collected with 500 µl of IPB buffer and transferred to new 2 ml tubes. Beads were washed twice with 1 ml IPB buffer by centrifugation at 1,000 g and 4 ºC. At the end, 120 µl of LDS NuPAGE buffer (Invitrogen) were added to the beads and incubated for 10 min at 72 ºC with moderate shaking. Eluted proteins were collected by centrifugation at 1,000 g at room temperature and transferred to new tubes marked as “IP”. Each sample was treated with DTT (50 mM final concentration) and incubated again for 10 min at 72 ºC with moderate shaking. Protein concentrations from the INPUT and FT samples were measured by spectrometry using a Nanodrop instrument (Abs 280 nm). INPUT, FT, and IP samples were used for western blot and IP samples were additionally subjected to Coomassie or silver staining. 55 MATERIALS AND METHODS In the immunoprecipitation experiments performed to prepare samples for mass spectrometry analysis, protein elution was carried out by adding 100 μl of arginine pH 3.5 buffer to the beads and incubated for 15 min at room temperature with moderate shaking. Eluted proteins were collected by centrifugation at 5,000 g at room temperature and transferred to new tubes marked “IP” with 10 μl of neutralization solution. For co-immunoprecipitation experiments, vegetative mycelia were fixed by adding 0.1% formaldehyde for 15 min and then the reaction was stopped by adding glycine (125mM) prior to collection and then the immunoprecipitation experiment followed as described in this section. - IPB Buffer: 50 mM Tris-HCl pH 7.5, 150 mM NaCl, 1.5 mM MgCl2, 0.1% [v/v] NP-40. - Protease inhibitors: 1 µM leupeptin, 1 µM pepstatin and 50 µM PMSF. - Neutralization solution: 0,5M Tris HCl pH 7.4, 1,5M NaCl. 5.6. Identification of posttranslational modifications (PTM) 5.6.1. Culture conditions and protein immunoprecipitation. N. crassa strains expressing gene fusions with the FLAG epitope: VE-1FLAG, VE- 2FLAG and LAEFLAG were used for immunoprecipitation using a FLAG antibody (ANTIFLAG M2 Affinity gel A2220 MERCK). For the analysis of PTMs, a total of 107 conidia were inoculated in 500 ml flask containing 200 ml of Vogel´s liquid medium with shaking and allowed to develop for 48 h in dark or under continuous light. Mycelia were fixed prior being collected by adding 0.1% formaldehyde for 15 min and then the reaction was stopped by adding glycine (125 mM). Mycelia were then washed and collected by filtration and frozen in liquid nitrogen. The immunoprecipitation procedure was performed as described in section 6.5 of this material and methods. 5.6.2. Digestion of proteins with trypsin For analysis of PTMs, immunoprecipitated samples were subjected to onedimensional SDS-polyacrylamide gel electrophoresis using 1 mm-thick 10% acrylamide gels. Protein spots of interest were excised form the gel and subjected to protein digestion 62 63 CHAPTER I Chapter I: Characterization and regulation of the velvet complex formation in Neurospora crassa 1. Introduction In N. crassa light regulates several essential processes such as the formation of asexual spores (conidia), required for its dispersion and colonization of new environments (Ruger-Herreros and Corrochano, 2019), and the production of photoprotective pigments (carotenoids and melanin), which are secondary metabolites with antioxidant properties that protect fungi from stress and UV radiation (Avalos and Carmen Limón, 2015; Cordero and Casadevall, 2017). Regulation of development and metabolism occurs when environmental signals, such as light, and internal cues are perceived and transmitted to the nucleus to rewire gene expression. This requires a complex regulatory network comprising several components, such as sensing systems, cytoplasmic signal transduction cascades, and downstream regulatory protein complexes that work in coordination to integrate different signals into a biological response ( Corrochano, 2019; Yu and Fischer, 2019). A central protein complex that coordinates fungal development, secondary metabolism, and light responses is the velvet complex. This complex was first identified in A. nidulans, the VelB/VeA/LaeA velvet complex (Bayram et al., 2008a). It has been proposed that VeA interacts with the red light phytochrome receptor protein FphA, which is further associated with the blue light receptors, LreA and LreB, to modulate gene expression in response to light, resulting in the regulation of different morphogenetic programs and metabolite production (Purschwitz et al., 2008). The molecular function of this complex is determined by its abundance, subcellular localization, and interaction with other regulatory proteins, in a process that is coordinated by light (Reviewed in Bayram et al. 2012). Initial studies of the ortholog of veA in N. crassa, the gene ve-1, revealed that it is required for the repression of asexual development as deletion of ve-1 lead to an increase in conidia production and defects in the length of aerial hyphae (Bayram et al. 2008b). Subsequently, a link between light and the mechanism of action of VE-1 was made in a study carried out in our research group (Gil- Sánchez et al., 2022). This work revealed that VE-1 is required for carotenoids accumulation and light-dependent transcription of genes involved in carotenogenesis (al-1, al-2, al-3, and cao-2) as well as genes involved in the light response and photoadaptation Chapter I. Velvet complex 64 (wc-1 and vvd). Furthermore, VE-1 showed light-dependent accumulation and subcellular localization in the nuclei during early stages of asexual development, presumably to regulate gene expression (Gil-Sánchez et al., 2022). VE-1 is an unstable protein and undergoes degradation via the proteosome pathway in a process regulated by the COP9 signalosome (Gil-Sánchez et al. 2022). It has been proposed that in the early stages of asexual development, light prevents the degradation of VE-1, resulting in a major accumulation of the protein within the nucleus. The presence of PEST domains is typical for proteins with a high turnover rate, and they often become phosphorylated. Like VE-1, A. nidulans VeA also contains a PEST domain and has been described as an unstable protein. In addition, the degradation of VeA requires the ubiquitinase UspA (Meister et al., 2019). VeA is modified by post-translational modifications (PTMs). In particular, VeA is phosphorylated, and reversible phosphorylation of VeA has been proposed to control its localization and interaction with VelB and the phytochrome FphA, affecting its function in development and SM production (Rauscher et al., 2016). The molecular function of other components of the velvet complex in N. crassa and the interaction between them remain largely unknown. Here, we characterize the accumulation and subcellular localization of the proteins VE-2 and LAE-1, homologs of VelB and the methyltransferase LaeA of A. nidulans, and show that they interact with VE- 1 to form a heterotrimeric velvet complex, as in A. nidulans. Furthermore, we performed MS- spectrometry to identify PTMs since they could play a role in VE-1 stability, subcellular localization, and interactions between the members of the N. crassa velvet complex. 2. Results 2.1. The genome of N. crassa contains four members of the velvet family In addition to VE-1, the genome of N. crassa contains three other genes that encode proteins of the velvet family, as well as an ortholog of A. nidulans Lae-A (Galagan et al., 2003). VE-2 (470 aa; 47% identity with VelB of A. nidulans) contains a velvet domain in its C-terminal half and no obvious NLS motif in its sequence, similar to VelB in A. nidulans. 65 CHAPTER I The sequence of VOS-1 (447 aa; 44% identity with VosA of A. nidulans) is predicted to contain an NLS and a velvet domain in its C-terminal half. VE-3 shows a domain structure that resembles that of VelC of A. nidulans (40% identical), but with a truncated velvet domain (Ojeda-López et al. 2018). There is also an ortholog of the LaeA protein in N. crassa, the protein LAE-1(326 aa; 42% identity with A. nidulans LaeA), which has an S- adenosyl methionine (SAM) binding site and a methyltransferase domain in the central part of the protein (Figure I. 1). Figure I. 1. Predicted domain architecture of the velvet proteins and LAE-1 of N. crassa. The size of the proteins and the position of the different domains are represented to scale. Domains were identified using the databases Pfam, Prosite, SMART, EMBOSS explorer, and NLS mapper. 2.2. Light does not affect the expression of ve-2, vos-1, and lae-1 during vegetative growth In N. crassa, illumination of the mycelium activates the transcription of lightresponsive genes, e.g., the con genes, genes involved in the carotenogenesis pathway such as the albino genes (al-1, al-2, and al-3) and the cao-2 gene, as well as the genes responsible for photoinduction and photoadaptation, wc-1 and vvd, respectively (Shrode et al., 2001; He and Liu, 2005; Chen et al., 2009). Light does not change the accumulation of ve-1 mRNA or protein, but VE-1 is required for the correct accumulation of light-regulated mRNAs (Gil-Sánchez et al., 2022). Therefore, we decided to characterize the effect of light 554 aa NC 100 aa 470 aa 447 aa 348 aa 326 aa VE-1 (NCU01731) VE-2 (NCU02775) VOS-1 (NCU05964) VE-3 (NCU07553) LAE-1 (NCU00646) N C NC NC NC Chapter I. Velvet complex 66 on the transcription of ve-2, vos-1, and the methyltransferase gene lae-1. The gene encoding for VE-3 was not included in the analysis due to their truncated velvet domain and the lack of phenotype in the corresponding deletion mutant (data not shown). To characterize the accumulation of mRNAs by light, we collected samples from vegetative mycelium cultured in liquid medium in the dark and after exposure to light. The gene ve-1 and the light-responsive gene con-10 were included as controls (Figure I. 2). As expected, photoinduction and photoadaptation of con-10 were observed after 30 minutes and 300 min of light exposure, respectively. The expression of ve-1, ve-2, vos-1, and lae-1 was detected at all experimental points (0, 30, and 300 min of light exposure) and did not appear to be regulated by light. Figure I. 2. Relative expression of con-10, ve-1, ve-2, vos-1, and lae-1 after different exposures to light in the wild-type strain. The wild-type strain was grown in liquid media for 48 hours at 22°C in the dark and subsequently exposed to light during different times prior to RNA extraction and quantification by quantitative RT-PCR. The mRNA levels from each gene were normalized with tubulin (tub-2) mRNA to correct for sampling errors and then normalized with the mRNA accumulation obtained with samples exposed to 30 min of light. The graph shows the mean and standard error of at least three independent experiments. The plot shows the mean and standard error of four independent experiments (* p< 0.05, ** p<0.01, ***p< 0.001, ****p<0.0001). According to the previous characterization of the VE-1 protein (Gil-Sánchez et al., 2022), its accumulation during vegetative growth was not altered by light, confirming that light does not have a regulatory role in the accumulation of VE-1 in the vegetative mycelium. In the present work, we have expanded the characterization of protein accumulation to VE-2, VOS-1 and LAE-1, using strains with alleles of the corresponding genes fused to the coding sequence of the FLAG epitope (ve-2FLAG, vos-1FLAG and lae-1FLAG) 67 CHAPTER I to allow immunodetection by Western blot. Each new allele replaced the corresponding wild-type allele under the control of the original promoter. Both proteins VE-2 and LAE-1 were detected at all experimental times of light exposure (Figure I. 3), but no differences in protein accumulation were observed. Western blots revealed more than one band for proteins VE-2 and LAE-1. Neither of these bands was the result of unspecific antibody binding as a total protein extract of the wild-type strain did not show any bands (data not shown). The protein VOS-1 could not be detected in any of the samples. The results indicate that the accumulation of VE-2 and LAE-1 in vegetative mycelia is not regulated by light, similar to VE-1. Figure I. 3. Detection of VE-2 and LAE-1 proteins in the vegetative mycelium at different times of light exposure. ve-2FLAG and lae-1FLAG strains were grown in vogel liquid media for 48 hours at 22°C in the dark and subsequently exposed to 0, 30, 60, 120 and 300 min of light. Proteins were separated by SDS-PAGE and hybridized with antibodies specific for FLAG and Tubulin. Each lane was loaded with 70 µg of protein. The protein tubulin was used as a loading control. 2.3. VE-2 and LAE-1 localize in nuclei and cytoplasm during vegetative growth in a light-independent manner In A. nidulans, VeA shows a light-dependent nuclear localization (Stinnett et al., 2007; Bayram et al., 2008a). According to previous results in our group, VE-1 localizes to the nuclei and cytoplasm during vegetative growth and that localization is independent of the light condition. In this thesis, VE-2 and LAE-1 were examined in cytoplasmic and nuclear fractions. Both proteins were detected in the nuclear and cytoplasmic fractions under dark and light conditions, but VE-2 was less abundant in nuclei compared to LAE-1 (Figure I. 4). 75 kDa 50 50 37 kDa 50 !-FLAG !-Tub !-FLAG !-Tub ve-2FLAG lae-1FLAG 0 30 60 120 300 0 30 60 120 300 (Time, min)(Time, min) Chapter I. Velvet complex 68 Figure I. 4. Subcellular localization of VE-2 and LAE-1. Nuclear and cytoplasmic enrichment of VE-2 and LAE-1. The cultures were grown in the dark for 48 hours and then exposed to light for 30 minutes, or under continuous dark, or in continuous white light for 48 hours. T, C, and N represent total, cytoplasmic, and nuclear extracts, respectively. Proteins were separated by SDSPAGE and hybridized with antibodies specific for FLAG, Tubulin, and histone 3 (H3). Each lane was loaded with 70 µg of protein. Tubulin and histone 3 (H3) were used as loading controls for cytoplasmic and nuclear proteins, respectively. The nuclear subpopulation of LAE-1 showed a higher molecular weight than the cytoplasmic LAE-1, suggesting that this protein may undergo post-translational modifications depending on its subcellular location. The presence of the velvet proteins and the LAE-1 methyltransferase in nuclei suggests that these proteins may play key transcriptional regulatory functions that control gene expression in N. crassa. 2.4. Velvet proteins VE-1, VE-2, and LAE-1 methyltransferase form a heterotrimeric complex in N. crassa Next, we investigated whether the N. crassa velvet proteins VE-1 and VE-2 and the methyltransferase LAE-1 interact to form a regulatory protein complex in the vegetative mycelium. To do that, co-immunoprecipitation experiments (co-IP) of strains ve-1FLAG ve- 2HA, ve-1FLAG lae-1HA and ve-2FLAG lae-1HA were carried out during vegetative growth (Figure I. 5). VE-1 co-precipitated with both VE-2 and LAE-1. Furthermore, VE-2 also coprecipitated with LAE-1, consistent with the formation of a heterotrimeric velvet complex in vivo, as shown in A. nidulans (Bayram et al., 2008a). These experiments were carried out under three growth conditions: continuous darkness, continuous light, and illumination of the mycelium for 30 minutes, and we obtained similar results (not shown). 69 CHAPTER I Figure I. 5. Detection of a velvet complex during vegetative growth in N. crassa. (A) Coimmunoprecipitations (co-IPs) of the velvet complex using triple (3x) FLAG and HA-tagged versions of VE-1, VE-2 and LAE-1 expressed under their native promoters during vegetative growth. INPUT, FT, and IP represent the input, flow-through, and immunoprecipitated fractions. VE-2 and LAE-1 were co-immunoprecipitated with VE-1. LAE-1 was precipitated with either VE-1 or VE-2. Proteins were separated by SDS-PAGE and hybridized with antibodies specific for FLAG and HA. We used Ponceau as loading control. 2.5. The velvet protein VE-2 and the LAE-1 methyltransferase are stable proteins during vegetative growth Previous results obtained in our group suggest that light regulates the degradation of VE-1 via the proteasome pathway. This regulation of VE-1 stability by light may represent a new mechanism that allows N. crassa to respond quickly to changes in illumination that may not be obtained by modulating the transcriptional response by light (Gil-Sánchez et al., 2022). Therefore, we decided to assay the stability of the other two members of the velvet complex, VE-2 and LAE-1, using strains with the tagged alleles ve-1FLAG, ve-2HA, and lae-1FLAG. Cultures were grown for 24 h in liquid media. Then protein synthesis was inhibited with cycloheximide for up to 480 min, samples were collected at different times Chapter I. Velvet complex 70 and the proteins were detected by western blot using antibodies against FLAG or HA epitope. Cycloheximide disrupts ribosome activity and prevents translation elongation (Schneider-Poetsch et al., 2010). Unlike VE-1, we observed the accumulation of VE-2 and LAE-1 after long incubations in cycloheximide (480 min), suggesting that VE-1 was the only component of the velvet complex subjected to degradation when protein synthesis was inhibited (Figure I. 6). Figure I. 6. Stability and degradation of the components of the velvet complex. Vegetative mycelia cultures from strains ve-1FLAGve-2HA or lae-1FLAG strains were grown at 30° C in liquid media for 24h in light, then cycloheximide was added to the cultures, and the samples were collected at different times. The proteins were separated by SDS-PAGE and hybridized with an antibody against FLAG, HA, or NOP1P. 30 μg of proteins were loaded per lane. We used Ponceau as loading control. 2.6. The components of the velvet complex in N. crassa undergo posttranslational modifications in vegetative mycelia VE-1 is an unstable protein, and its degradation by the proteasome is mediated by light and requires the coordinated action of FWD-1 and the COP9 signalosome (Gil- Sánchez et al., 2022). This implies that VE-1 must undergo post-translational modifications (PTMs), such as ubiquitylation, prior to its degradation. It is also possible that PTMs play a role in modulating the interactions and subcellular localization of the components of the velvet complex. We have detected several hybridization bands of VE-2 and LAE-1 by western blot, which suggests that these proteins are modified by PTMs. We characterized the presence of PTMs in VE-1, VE-2, and LAE-1 by mass spectrometry with protein samples isolated from vegetative cultures grown under two experimental conditions, in the 0 30 60 120 300 480 min in CHX !-NOP1P lae-1FLAG !-FLAG Ponceau 37 kDa 37 kDa 0 30 60 120 300 480 min in CHX ve-1FLAG ve-2HA !-NOP1P !-FLAG Ponceau !-HA 75 50 75 50 37 71 CHAPTER I dark or exposed to continuous light (Figure I. 7). The mass spectrometry analysis was carried out in collaboration with the laboratory of Prof. G. Braus at the University of Göttingen (Germany). Figure I. 7. Schematic diagram showing post-translational modifications (PTMs) identified in proteins of the velvet complex in N. crassa. Major functional domains and possible PTMs identified by mass spectrometry are shown. The size and the position of the different domains are represented to scale in each protein. Our results show that VE-1 is also a multiphosphorylated protein with five different phosphosites identified by MS/MS corresponding to positions T114, T264, S291, S414, and S455 (Figure I. 7). Phosphorylation of amino acids T114 and T264 could be essential for VE-1 regulation, as they are located within the velvet domain and close to the NLS, respectively. We detected the ubiquitylation of K170 in the PEST area of the velvet domain of VE-1. This modification was confirmed by western blot under both dark and light conditions (Figure I. 8), supporting the proposal that VE-1 is subjected to ubiquitinmediated proteasome degradation (Gil-Sánchez et al., 2022). We detected several methylation sites in VE-1, and most of them were located at the N-terminal half of the protein. Seven residues of the velvet domain were methylated, and three additional methylated residues were located close to this domain, suggesting that they may have a role in regulating its activity. Two other methylated amino acids were located in the C-terminal region. We did not observe any light-regulation in the distribution of PTMs. Phosphorylation Ubiquitination Methylation VE-1 VE-2 LAE-1 K 170 NLS S455 NLS NLS N N N C C C Chapter I. Velvet complex 78 Table I. S2. Prediction of kinases that could phosphorylate VE-2. Position code kinases peptide score cutoff 158 T Other/CDC7 DKDRRPITPPPCVKL 75,481 69,788 158 T Other/ULK DKDRRPITPPPCVKL 32,358 30,737 158 T STE/STE7 DKDRRPITPPPCVKL 26,895 25,765 158 T CMGC/DYRK DKDRRPITPPPCVKL 21,375 14,351 158 T CMGC/MAPK DKDRRPITPPPCVKL 12,327 8,865 158 T Atypical/PIKK DKDRRPITPPPCVKL 11,216 10.89 158 T CMGC/CDK DKDRRPITPPPCVKL 10,886 6,537 158 T STE DKDRRPITPPPCVKL 7,407 5,264 158 T CMGC/RCK DKDRRPITPPPCVKL 6,382 3,201 158 T TKL DKDRRPITPPPCVKL 6,327 4,132 158 T STE/STE20 DKDRRPITPPPCVKL 5,487 4,709 158 T AGC/NDR DKDRRPITPPPCVKL 0.003 0.001 327 S Other/NEK GPLQHRMSVSSTSGQ 110,671 106,827 327 S AGC/PKA GPLQHRMSVSSTSGQ 22,683 21,268 327 S CMGC/DYRK GPLQHRMSVSSTSGQ 15.15 14,351 327 S Other/Aur GPLQHRMSVSSTSGQ 9,458 7,351 327 S TKL/RIPK GPLQHRMSVSSTSGQ 9,166 8.42 327 S CAMK/CAMKL GPLQHRMSVSSTSGQ 6,768 6,392 327 S CK1 GPLQHRMSVSSTSGQ 5,606 5,148 327 S AGC/PDK1 GPLQHRMSVSSTSGQ 4,997 4,554 327 S CAMK GPLQHRMSVSSTSGQ 2.93 1,956 79 CHAPTER II 80 81 CHAPTER II Chapter II: Transcriptional regulation by the velvet protein VE-1 during asexual development in the fungus Neurospora crassa 1. Introduction Asexual reproduction in fungi facilitates dispersal and colonization of new substrates and is a key step in the infection of plants and animals by pathogenic fungi. Although the formation of vegetative spores or conidia is a conserved fungal trait, the molecular mechanisms and the regulatory genes that participate in asexual development show differences among fungi (Park and Yu, 2012; Park and Yu, 2016; Huang and Hull, 2017; Ojeda-López et al., 2018; Ruger-Herreros and Corrochano, 2019; Etxebeste et al., 2019). Two types of vegetative spores, conidia, have been described in the ascomycete Neurospora crassa: macroconidia and microconidia. Macroconidia are large and multinucleate, while microconidia are smaller and uninucleate. Macroconidia are produced abundantly in most growth conditions, but microconidia are only observed under specific growth conditions including high humidity, low temperature, and limited nutrients (Springer, 1993; Maheshwari, 1999). We will use the terms macroconidia and conidia interchangeably, as they are the most easily observed and most frequently used type of N. crassa conidia. Macroconidiation is usually induced by desiccation, the transfer of cultures from liquid to an air interface, and a reduction of nutrients. In addition, macroconidiation is regulated by light, the levels of CO2, and the circadian clock (Springer, 1993; Ruger- Herreros and Corrochano, 2019). During macroconidiation, vegetative hyphae lift up from the surface substrate and develop a mass of aerial hyphae that leads to the development of a string of conidia held together by connective threads until they are dispersed by air currents (Springer and Yanofsky, 1989). Single microconidia, on the contrary, are produced by budding of vegetative hyphae and have been proposed to serve as male fertilization cells during the sexual cycle (Maheshwari, 1999). Mutations that block macroconidiation do not modify microconidiation, although conidiation (con) genes isolated by their expression during macroconidiation are also detected in microconidia (Springer and Yanofsky, 1992). Several genes that participate in the regulation of macroconidiation have been identified by mutations that block this developmental pathway. A key regulatory gene is fl, a gene that encodes a Zn2Cys6 binuclear zinc cluster domain transcription factor (Bailey and Ebbole, Chapter II. Conidiation 82 1998). Overexpression of fl in vegetative mycelia leads to macroconidiation in submerged cultures, and FL binding sites have been identified in genes regulated by FL, such as the gene eas encoding the hydrophobin that coats macroconidia (Bailey-Shrode and Ebbole, 2004; Rerngsamran et al., 2005). The gene fl is induced by light, in a process regulated by binding of the light-dependent transcription factor complex, White Collar Complex (WCC), to the promoter. It has been proposed that the activation of fl by light is a key element in the regulation by light of macroconidiation (Olmedo et al., 2010a). The velvet complex is a fungal-specific protein complex that participates in the regulation of conidiation in many fungi (Calvo, 2008; Bayram and Braus, 2012). The activity and regulation of the velvet complex has been described in detail in Aspergillus nidulans and it is composed of three proteins: VeA, VelB and LaeA (Bayram et al., 2008a). VeA and VelB contain a 150-amino acid velvet domain that participates in protein-protein interactions and DNA binding. The structure of the velvet domain resembles the DNA- binding fold of the mammalian transcription factor NF-kB (Ahmed et al., 2013). It has been proposed that the velvet complex regulates transcription by DNA binding to promoters and chromatin modification (Sarikaya-Bayram et al., 2014). In A. nidulans, the components of the velvet complex are observed in the cytoplasm and nuclei. VeA and VelB interact with each other, and VeA also interacts with the importin KapA to enter the nucleus in a process that is regulated by light (Stinnett et al., 2007). The third component of the velvet complex is LaeA, a methyltransferase that is required for the regulation of secondary metabolism (Bok and Keller, 2004; Bayram et al., 2008a). Other methyltransferases, kinases, and photoreceptors interact with components of the velvet complex, presumably to regulate its activity (Purschwitz et al., 2008; Sarikaya-Bayram et al., 2014; Rauscher et al., 2016). In addition, VosA, which also contains a velvet domain, interacts with VelB to regulate spore maturation and viability in A. nidulans (Ni and Yu, 2007; Wu et al., 2021). The velvet family also includes a fourth member, ve-3/velC, which belongs to a nonmonophyletic group of velvet genes (Ojeda-López et al., 2018). In A. nidulans, VelC is a negative regulator of conidiation and a positive regulator of sexual development (Park et al., 2014), while in N. crassa VE-3 appears to have a truncated velvet domain and is not a component of the velvet complex (Sarikaya Bayram et al., 2019). The velvet complex in N. crassa is composed of VE-1, VE-2, and LAE-1. They interact in vegetative hyphae to form a complex that participates in the regulation of sexual and asexual development, secondary metabolism, and the accumulation of carotenoid 83 CHAPTER II pigments (chapter I; Sarikaya Bayram et al.,2019). Strains with mutations in ve-1, ve-2 or lae-1 showed reduced sexual development, but alterations in asexual development (shortened aerial hyphae and increased conidiation) were observed in ve-1 or ve-2 mutant strains only (Sarikaya Bayram et al., 2019). Mutations in the genes encoding components of the velvet complex resulted in changes in the accumulation of secondary metabolites, including the siderophore coprogen, and carotenoids. RNA seq analysis performed in liquid cultures under growth conditions that promoted the accumulation of secondary metabolites identified a set of about 1500 genes regulated by VE-1 (Sarikaya Bayram et al., 2019). Heterologous expression of ve-1 and ve-2 complemented the phenotypes of A. nidulans mutants of veA or velB, respectively. Similarly, expression of A. nidulans veA in the N. crassa ve-1 mutant restored the wild-type phenotype (Sarikaya Bayram et al., 2019). VE-1 has been observed in the cytoplasm and nuclei of vegetative hyphae, and the nuclear accumulation of VE-1 increases during conidiation. In addition, VE-1 is degraded by the proteasome during conidiation, but the degradation of VE-1 is regulated by light, presumably to allow a quick transcriptional response to changes in light intensity during conidiation (Gil-Sánchez et al., 2022). It is not known how VE-1 and the other components of the velvet complex interact and regulate transcription during conidiation. The other velvet protein, VOS-1, was found to interact with VE-1 and VE-2, but no phenotype in conidial development was found in N. crassa (Sarikaya Bayram et al., 2019). Here, we characterize the role of VE-1 in transcriptional regulation during mycelial growth and during conidiation by RNA seq experiments performed with cultures kept in dark or exposed to light. We propose that VE-1 is the limiting component of the velvet complex during conidiation and that it has a major role in the transcriptional regulation of the developmental process of conidiation. 2. Results 2.1. The ve-1 and ve-2 genes are required for aerial growth and conidiation Initial studies have shown that the absence of VE-1 led to a light-dependent shortening of aerial tissue, a mixture of aerial hyphae and developing macroconidia, that required the activity of the photoreceptor WC-1 (Gil-Sánchez et al., 2022). This observation together with the proposed role of the velvet complex in the regulation of Chapter II. Conidiation 84 asexual development in N. crassa (Sarikaya Bayram et al., 2019) prompted us to further characterize the role of VE-1 and VE-2 in the growth of aerial tissue and the amount of conidia in single and double deletion mutants, and their possible regulation by light. Aerial tissue is defined by the growth of conidiating hyphae away from the supporting mycelia in glass tubes with solid agar media. We observed a 25% reduction in the length of aerial hyphae in the wild-type strain by illumination. Shorter aerial hyphae were observed in ∆ve- 1 and ∆ve-2 mutant strains compared to the wild type both in dark or light (Figure II. 1A). A reduction in the growth of aerial tissue has been reported for a strain with reduced adenylyl cyclase activity, cr-1. The growth of aerial tissue in these mutants was restored after addition of cAMP to the growth media (Kays et al., 2000), but supplementation with cAMP did not restore the growth of aerial tissue in ∆ve-1 or ∆ve-2 mutant strains (Figure II. S1A). In the wild-type strain, conidiation was increased three-fold by light exposure. In addition, ∆ve-1 and ∆ve-2 mutant strains displayed 3.2 and 3.3-fold higher conidiation levels, respectively, but only in cultures kept in the dark (Figure II. 1B). The ∆vos-1 mutant did not show any alteration in the growth of aerial tissue or conidial production and behaved like the wild-type strain under all conditions. Deletion of vos-1 did not alter the phenotypes of either ∆ve-1 or ∆ve-2 mutant strains, and the double ∆ve-1 ∆ve-2 mutant displayed the same short aerial tissue phenotype, and increased conidiation in the dark (3.5-fold higher as the wild type) as the corresponding single mutants (Figure II. 1). Deletion of ve-2, vos- 1 or both did not affect the viability of conidia in N. crassa, unlike in A. nidulans (Ni and Yu, 2007; Wu et al., 2021) (Figure II. S1B). Our results support the proposal that absence of either VE-1 or VE-2 disrupts the activity of the velvet complex resulting in alterations in the growth of aerial tissue and during conidiation in the dark. 85 CHAPTER II Figure II. 1. Mutants in ve-1 and ve-2 have defects in aerial tissue and increased conidiation. (A) Photographs of growth tube phenotypes of wild type, ∆ve-1, ∆ve-2, ∆vos-1, and double deletion mutants after three days of growth at 30°C in the dark or the light. Graphs show the length of aerial hyphae over time in the wild type and the mutants in the light or the dark. (B) Relative amount of conidia produced by wild type, single, and double deletion mutants. Conidia were collected from cultures grown in dark and light at 30°C after three days and plotted relative to the amount produced by the wild type in dark or light. The wild-type strain produced 1.5x108 conidia/tube in dark compared to 4.7x108 conidia/tube in light. The plots show the mean and standard error of three independent experiments. Two-way ANOVA comparisons were performed between all strains. Only statistical significances between the wild type and single mutant strains, and between every combination of a single mutant with the corresponding double mutant strains (∆ve-1 vs ∆ve-1 ∆ve-2, ∆ve-1 vs ∆ve-1 ∆vos-1 etc.) are shown. Significant comparisons are indicated by asterisks (*p<0.05, **p<0.01, ***p<0.001, ****p<0.0001). Chapter II. Conidiation 86 2.2. VE-1 and VE-2 localize to nuclei and cytoplasm of aerial tissue The role of the velvet complex in asexual development suggested that it could be present in the nuclei of aerial tissue to regulate transcription. We have previously reported the presence of the velvet complex in nuclei and cytoplasm of vegetative hyphae (chapter I), and the accumulation of VE-1 in nuclei and cytoplasm of aerial tissue in a light-regulated manner (Gil-Sánchez et al., 2022). In order to characterize the accumulation of mRNAs and proteins during conidiation, we collected samples from vegetative mycelia cultured in liquid media (submerged vegetative mycelia) prior to induction of conidiation. To induce conidiation, we transferred vegetative mycelia to the surface of a solid media plate. Fifteen hours after the induction of conidiation, we collected samples from the supporting vegetative mycelia (M15) and from aerial tissue (A15), for RNA and protein quantification. In some cases, we further separated mature conidia (C) from aerial hyphae (AH) to characterize the protein content from each developmental structure (Figure II. 2A). For RNA quantification we used the wild-type strain, and for protein quantification we employed strains with alleles of ve-1 or ve-2 fused to the coding sequence of the FLAG epitope to allow immunodetection by Western blots. Each new allele, ve-1FLAG and ve-2FLAG, replaced each wild-type allele and was under the control of their original promoters. The growth of strains with the ve-1FLAG and ve-2FLAG alleles was similar to that of the wild-type strain (Supplementary Figure II. S1C). The ve-1 and ve-2 mRNAs were observed in vegetative mycelia (both submerged vegetative mycelia, S, and supporting vegetative mycelia, M15) and aerial tissue (A15). We did not detect any major difference between these developmental stages, regardless of light conditions, except for an increase in the amount of ve-1 mRNA in supporting vegetative mycelia in the dark (Figure II. 2B). We confirmed the detection of VE-1 and VE-2 in submerged vegetative mycelial samples, but the amount of VE-1 in aerial hyphae and conidia was reduced, unlike VE-2 that accumulated in similar amounts in submerged vegetative mycelia, aerial hyphae, and conidia (Figure II. 2C). Light exposure did not modify the pattern of VE-2 accumulation (Supplementary Figure II. S2). 87 CHAPTER II Figure II. 2. Accumulation and subcellular localization of VE-1 and VE-2 during asexual development. (A) Samples from cultures kept in dark or light were obtained from mycelia growing vegetatively in submerged liquid media for 24h (submerged vegetative mycelia, S), from supporting vegetative mycelia after 15h of induction of conidiation (M15), and aerial tissue after 15h of the induction of conidiation (A15). Aerial hyphae (AH) and conidia (C) were separated from aerial tissue (A15) by centrifugation. (B) Relative accumulation of ve-1 and ve-2 mRNAs during asexual development. Quantitative RT-PCR experiments were performed to measure relative accumulation of mRNA in three different stages of development. The results were normalized to the corresponding tub-2 mRNA accumulation to correct for sampling errors. Then, the results were normalized to those obtained with submerged vegetative mycelia growing in liquid media for 24h (S). (C) Accumulation of VE-1 and VE-2 proteins during conidiation in the light. We used the ve-1FLAG and the ve-2FLAG strains. Proteins were separated by SDS-PAGE and hybridized with an antibody specific for FLAG. 70 µg of proteins were loaded per lane. As loading controls, we used Ponceau staining of each protein sample. (D) Subcellular localization of VE-1 during conidiation in the dark. We used the ve-1FLAG strain. Total protein samples (T), or samples enriched in cytoplasmic (C) or nuclear (N) proteins were separated by SDS-PAGE, and hybridized with antibodies specific for FLAG or histone H3. 80-120 µg of proteins were loaded per lane. As loading controls, we used Ponceau staining of each protein sample. (E) Subcellular localization of VE-2 during conidiation in the light. We used the ve-2FLAG strain. Total protein samples (T), or samples enriched in cytoplasmic (C) or nuclear (N) proteins were separated by SDS-PAGE, and hybridized with antibodies specific for FLAG or histone H3. 70 µg of proteins were loaded per lane. As loading controls, we used Ponceau staining of each protein sample. Plots show the mean and standard error (* p< 0.05, ** p<0.01, ***p<0.001, **** p<0.0001, two-way ANOVA) of three independent experiments. Chapter II. Conidiation 94 Figure II. 5. VE-1 regulates transcription of methyltransferase genes. (A) Heat map of relative transcript levels of DEGs encoding methyltransferases in submerged vegetative mycelia (V) or in aerial tissue at 15 h (A15) after induction of conidiation in the wild-type and ∆ve-1 mutant strains. (B) Photographs of culture tube phenotypes of wild type, ∆ve-1, ∆ve-2, ∆lae-1, and double deletion mutants after three days of growth at 30°C in the dark or the light. (C) Length of aerial tissue of the wild-type and mutant strains after three days of growth at 30°C in dark or light. (D) Relative amount of conidia produced in the wild type, single, and double deletion mutants. Conidia were collected from cultures grown in dark and light at 30°C after three days of incubation and plotted relative to the amount produced by the wild type in dark or light. Plots show the mean and standard error of three independent experiments. Two-way ANOVA comparisons were performed between all strains. Only statistical significances between the wild type and single mutant strains, and between every combination of a single mutant with the corresponding double mutant strains (∆lae-1 vs ∆ve-1 ∆lae-1, ∆lae-1 vs ∆ve-2 ∆lae-1 etc.) are shown. Significant comparisons are indicated by asterisks (*p<0.05, **p<0.01, ***p<0.001, ****p<0.0001) 95 CHAPTER II Then, we characterized the accumulation of the protein LAE-1 during conidiation by using a flag-tagged lae-1 strain. We observed accumulation of LAE-1 during asexual development that followed a similar pattern to that of VE-2. Levels of LAE-1 in both vegetative mycelia (submerged and supporting mycelia) and aerial tissue were similar, with a small increase in the amount of LAE-1 in aerial hyphae and conidia 15 and 24h after induction of asexual development; LAE-1 localized in the cytoplasmic and nuclear fractions, but it was less abundant in nuclei of aerial tissue (Figure II. 6E-F, Figure II. S2). By transcriptome analysis, developmental regulation of lae-1 was not observed (Data Set S3), but the increase in lae-1 mRNA levels in the ∆ve-1 strain suggested that the absence of VE-1 altered the accumulation of LAE-1. Indeed, we observed a two-fold increase in LAE-1 protein levels in the absence of VE-1 in submerged vegetative mycelia, and a slight increase in aerial tissue in the ∆ve-1 mutant (Figure II. 6F). Our results suggest that VE-1 plays a regulatory role not only in the transcription of lae-1, but also in the accumulation of LAE-1. Figure II. 6. Characterization of LAE-1 during conidiation. (A) Subcellular localization of LAE-1 during conidiation. We collected samples from submerged vegetative mycelia (S), supporting vegetative mycelia (M15), and aerial tissue (A15) from the lae-1FLAG strain growing in light. Total protein samples (T), or samples enriched in cytoplasmic (C) or nuclear (N) proteins were separated by SDS-PAGE, and hybridized with antibodies specific for FLAG or histone H3. 70 µg of proteins were loaded per lane. As loading control, we used Ponceau staining of each protein sample. (B) Accumulation of LAE-1 during conidiation. Protein samples from cultures kept in the light were obtained from submerged vegetative mycelia (S). Aerial tissue (A15) were collected after 15 h for induction of conidiation, and separated by filtration and cleaned by centrifugation to obtain separate samples of aerial hyphae (AH) and conidia (C). We used the lae- 1FLAG and ∆ve-1 lae-1FLAG strains. Proteins were separated by SDS-PAGE and hybridized with an antibody specific for FLAG. 70 µg of proteins were loaded per lane. As loading, control we used Ponceau staining of each protein sample. Plots show the mean and standard error of three independent experiments. Chapter II. Conidiation 96 2.6. Role of VE-1 in aerial tissue length: identification of VIB-1 As shown above, the deletion of ve-1 produced two major phenotypes during growth on solid media: a drastic reduction in the length of aerial tissue and increased conidiation in the dark. It is possible that some of the genes misregulated in the ∆ve-1 mutant could play a role in the developmental alterations produced by a non-functional velvet complex. A number of genes were selected that were identified as DEGs in our transcriptomic results based on the following criteria: (1) genes up- and downregulated by conidiation exclusively in the wild type with differences in FC log2 > 3, and (2) genes showing differences in mRNA accumulation (up and down) after pairwise comparisons between the wild type and the ∆ve-1 mutant in aerial tissue in dark and light with differences in FC log2 > 3. We then used the collection of N. crassa deletion mutants (Colot et al., 2006) with the most promising candidate genes to characterize their developmental phenotypes with the aim to identify mutants with a phenotype similar to that of the ∆ve-1 mutant. We identified 268 genes, whose deletion strains were available in the collection of N. crassa deletion mutants (Data Set S4); the phenotype of each strain was assayed by growth in solid medium in slants and in 55 mm petri dishes. Mutants with defects in the length of aerial tissue and in the pattern of conidiation across the petri dish were further selected for phenotypic confirmation. Eight mutants showed alterations in their growth pattern, out of which three strains displayed a growth phenotype similar to the ∆ve-1 strain. One of these three mutants had a growth phenotype that did not segregate with the hygromycin resistance marker used for deleting the gene, while the gene deleted in the second mutant was expressed weakly under all conditions. The third candidate gene was NCU03725 and had lower expression in the ∆ve-1 mutant than in the wild type during conidiation (FC log2 = -2.37 in dark, FC log2 = -3.20 in light; Data Set S1). Mutation in this gene resulted in a defect in the length of aerial tissue and a conidiation pattern intermediate between the ∆ve-1 mutant and the wild-type strain (Figure II. 7). The gene NCU03725 has been characterized previously as vib-1 (vegetative incompatibility blocked- 1) (Hutchison and Glass, 2010), and encodes a homolog of Saccharomyces cerevisiae NDT80, a transcription factor involved in the regulation of genes during meiosis (Winter, 2012). VIB-1 does not play a role in meiosis in N. crassa (Hutchison and Glass, 2010) but was identified as a major regulator of responses to nitrogen and carbon starvation (Wu et al., 2020) and is essential for the expression of genes involved in non-self-recognition and cell death in N. crassa (Xiang and Louise Glass, 2002; Dementhon et al., 2006). In addition, 97 CHAPTER II VIB-1 has been described as a negative regulator of conidiation (Xiang and Glass, 2004). We hypothesize that VE-1 could regulate conidiation by acting as a positive regulator of the expression of vib-1. To test this hypothesis, we created double mutants of ve-1 and ve- 2 with vib-1. All single and double mutants (∆ve-1 ∆vib-1 and ∆ve-2 ∆vib-1) showed reduced length of aerial tissue both in light and dark, and increased conidiation in the dark, as each of the single ∆ve-1 and ∆ve-2 mutants (Figure II. 7 A-B). The two double mutants (∆ve-1 ∆vib-1 and ∆ve-2 ∆vib-1) showed a decrease in the length of aerial tissue compared to each single mutant alone (with the exception of ∆ve-2 ∆vib-1 in light), suggesting that VIB-1 may operate, at least partially, in a different route for the regulation of asexual development, and that control of aerial hyphal growth requires both VE-1 and VIB-1 in N. crassa. Previously, Wu et al identified 1,742 VIB-1 binding sites by DAP-seq (Wu et al., 2020); VIB-1 binds the promoters of 94 (dark) and 86 (light) DEGs out of 237 and 343 genes that required VE-1 for proper expression during conidiation in the dark and light, respectively (Figure II. 7C). This accounts for 39% (dark) and 25% (light) of the DEGs, which supports the proposal that a significant proportion of the VE-1 regulon requires VIB- 1. 2.7. Role of VE-1 in conidiation: regulation of fl and microconidiation Genetic analysis has led to the proposal that FL and VIB-1 participate in the same pathway for the regulation of conidiation and that FL partially activates conidiation by repressing VIB-1 (Xiang and Glass, 2004). We have shown that VE-1 is required for the transcription of vib-1 and VE-1 is a repressor of fl and other conidiation genes (Figure II. 8A, Figure II. 4D, Data Set S3). FL is a key transcription factor in the regulation of conidiation since fl mutants are blocked in the transition from minor to major constrictions during the development of conidia (Bailey and Ebbole, 1998; Bailey-Shrode and Ebbole, 2004). To characterize a possible interaction between FL and VE-1 in the regulation of conidiation, we created double mutant strains with mutations in ve-1 or ve-2 with fl. Chapter II. Conidiation 98 Figure II. 7. VE-1 regulates vib-1 during the development of aerial tissue and conidiation. (A) Photographs of culture tube phenotypes of the wild type, ∆ve-1, ∆ve-2, ∆vib-1, and double deletion mutants after three days of growth at 30°C in dark or light. (B) Length of aerial tissue of the wild-type and the mutant strains after three days of growth at 30°C in dark or light. (C) Relative amount of conidia produced in the wild type, single, and double deletion mutants. Conidia were collected from cultures grown in dark and light at 30°C after three days of incubation and plotted relative to the amount produced by the wild type in dark or light. Plots show the mean and standard error of three independent experiments. Two-way ANOVA comparisons were performed between all strains. Only statistical significances between the wild type and single mutant strains, and between every combination of a single mutant with the corresponding double mutant strains (∆vib-1 vs ∆ve-1 ∆vib-1, ∆vib-1 vs ∆ve-2 ∆vib-1 etc.) are shown. Significant comparisons are indicated by asterisks (*p<0.05, **p<0.01, ***p<0.001, ****p<0.0001). 99 CHAPTER II The double mutant strains ∆ve-1 ∆fl and ∆ve-2 ∆fl displayed the short aerial phenotype that is characteristic of each single mutant ∆ve-1, ∆ve-2, and ∆fl (Figure II. 8B- C), but the double mutants showed shorter aerial tissue than the ∆fl strain (with the exception of ∆ve-2 ∆fl in light) suggesting that VE-1, VE-2 and FL act, at least partially, in different pathways to regulate the growth of aerial tissue. The ∆ve-1 ∆fl and ∆ve-2 ∆fl strains lacked macroconidial production as the single ∆fl mutant, but the double mutants produced large amounts of microconidia, unlike each single mutant (Figure II. 8D-F). The increase in microconidiation in the ∆ve-1 ∆fl and ∆ve-2 ∆fl mutants suggests that the velvet complex acts as a repressor of microconidiation when the macroconidiation pathway is blocked by a deletion of fl. Our results support the proposal that the velvet complex plays a key role in the transcriptional regulation of conidiation and the balance between macro- and microconidiation in N. crassa. 3. Discussion Velvet proteins, and the velvet complexes, regulate development, secondary metabolism, and pathogenicity in fungi but the specific role of the velvet proteins may differ from one species to another, and is particularly evident in the regulation of fungal development. For example, in A. nidulans, VeA plays a crucial role in conidiation in response to red light (Mooney and Yager, 1990), and its role depends on its subcellular localization, which is regulated by light (Stinnett et al., 2007; Bayram et al., 2008a). The VeA homolog is required for conidiation in Penicillium chrysogenum (Hoff et al., 2010) and Trichoderma reesei (Liu et al., 2016). However, deletion of the veA homolog results in a hyperconidiating phenotype in A. fumigatus (Park et al., 2012a). In several species of Fusarium, mutations in veA homologs results in alterations in conidiation that are specific for each species. For example, deletion of the veA homolog in Fusarium verticilloides and F. fujikoroi resulted in an increase in macroconidiation and decreased microconidiation (Li et al., 2006; Wiemann et al., 2010), but the same deletion in F. oxysporum results in a reduction in aerial hyphal formation and conidial production (Li et al., 2015). VeA, and homologous proteins, act as global regulators of fungal development (Calvo, 2008; Bayram and Braus, 2012). Here we show that in N. crassa, mutations in ve-1 or ve-2 result in a reduced length of aerial tissue, and increased conidiation in the dark by modulating the transcriptional responses. Chapter II. Conidiation 100 Figure II. 8. The velvet complex regulates microconidiation. (A). Heat map of relative transcript levels of genes encoding conidiation-related genes in submerged vegetative mycelia (S) or in aerial tissue 15 h (A15) after induction of conidiation in the dark or after exposure to light for 30 min prior to collection. We used the wild-type and the ∆ve-1 mutant strains. (B) Photographs of culture tube phenotypes of the wild type, ∆ve-1, ∆ve-2, ∆fl, and double deletion mutants after three days of growth at 30°C in the dark or the light. (C) Length of aerial tissue of the wild-type and mutant strains after three days of growth at 30°C in the dark or with light. (D) Vegetative hyphae and conidiation (macro- and microconidiation) in samples of each single and double mutant. Macro- and microconidia from ∆ve-1, and ∆ve-1, ∆fl strains were obtained after filtration with cheese cloth. (E) Relative amounts of macroconidia in the wild type, single, and double deletion mutants. Results are plotted relative to the amount produced by the wild type in dark or light. (F) Amounts of microconidia in the wild type, single, and double deletion mutants. Conidia (macro- and microconidia) were collected from cultures grown in dark and light at 30°C after three days of incubation. Plots show the mean and standard error of three independent experiments (*p<0.05, **p<0.01, ***p<0.001, ****p<0.0001, two-way ANOVA). 101 CHAPTER II The N. crassa velvet complex is composed of VE-1, VE-2 and LAE-1 (chapter I; Sarikaya Bayram et al, 2019). In addition, the velvet protein VOS-1 interacts with VE-1 and VE-2 (Sarikaya Bayram et al., 2019). Our observation that mutations in ve-1 or ve-2 results in alterations of asexual development, while mutations in vos-1 or lae-1 had a wildtype phenotype suggests a role in the regulation of asexual development for a regulatory complex composed of VE-1 and VE-2. The stable levels of VE-2 during conidiation, unlike VE-1, suggests that VE-1 acts as the limiting component in the formation of complexes with VE-2 and LAE-1, and that the activity of these complexes is regulated by the availability of VE-1. Given the key role that VE-1 should have on the formation and activity of the velvet complex, we characterized the transcriptome of a ∆ve-1 mutant strain during vegetative growth and in the early stages of conidiation. Our results show that the absence of VE-1 led to major changes in the conidiation transcriptome, suggesting that the developmental phenotypes we observed are due to changes in transcription. Given the DNA-binding capability of the velvet domain (Ahmed et al., 2013) it is conceivable that VE-1 regulates transcription directly by binding to target promoters as has been shown for VeA in A. nidulans (Rauscher et al., 2016). VE-1 may also indirectly regulate conidiation-dependent transcription factors as we identified a set of transcription factor genes with altered transcriptional profiles in the absence of VE-1 that could participate in the regulation of conidiation in coordination with VE-1. In addition, the absence of VE-1 led to changes in the accumulation of mRNAs for LAE-1 (and genes encoding other methyltransferases) and VOS-1, suggesting that VE-1 also regulates the composition of its interacting partners through transcriptional regulation. We have identified the transcription factor VIB-1 as a candidate protein that participates with VE-1 in the regulation of development in N. crassa. vib-1 was initially identified in N. crassa as a gene required for the regulation of vegetative incompatibility and encodes a transcription factor (Xiang and Louise Glass, 2002). More recently, it was shown that VIB-1 directly binds and regulates genes associated with nutrient assimilation and regulation of genes associated with plant cell wall deconstruction (Wu et al., 2020). In particular, two genes encoding LaeA-like methyltransferase domains (NCU05832, and NCU05501) that were down regulated in the ∆ve-1 mutant are in the core VIB-1 regulon, while four additional LaeA-like genes (NCU04909, NCU04717, NCU04707, and NCU01148) are direct targets of VIB-1. Genetic interactions between vib-1 and fl, a key Chapter II. Conidiation 102 regulator of conidiation that encodes a conidiation-dependent transcription factor, led to the proposal that FL acts as negative regulator of VIB-1, and that VIB-1 is a repressor of conidiation (Xiang and Glass, 2004). In this hypothesis, the aconidial phenotype of fl is due to an excessive repression of conidiation by VIB-1. Our transcriptomic analysis showed that vib-1 is downregulated in the absence of VE-1 while fl and other conidiation genes are upregulated, suggesting that VE-1 participates in the regulation of conidiation by repressing the transcription of fl, among other TF genes, and activating the transcription of vib-1. When the aconidial strain ∆fl was combined with deletions in either ve-1 or ve-2, the resulting double mutant strains produced large amounts of conidia. However, instead of producing large multinucleate macroconidia, the double ∆ve-1 ∆fl and ∆ve-2 ∆fl strains produced large amounts of microconidia. These are small in size, are uninucleate (unlike macroconidia), and are only observed under specific environmental conditions with high humidity, low temperature, and limiting nutrients (Maheshwari, 1999). In our growth conditions we did not detect microconidia in the wild-type or the fl mutant strain. In addition, we could not detect microconidia in the single ∆ve-1 or ∆ve-2 strains suggesting that absence of VE-1 or VE-2 alone was not sufficient to induce microconidiation. Instead, the suppression of macroconidiation by the absence of FL and the absence of VE-1 or VE- 2 led to a misregulation of microconidiation and an abundant production of microconidia in these double mutant strains. How could VE-1 and VE-2 regulate microconidiation? It is possible that the VE- 1/VE-2 complex may regulate microconidiation by controlling the transcription of key regulatory genes, as observed in the regulation of macroconidiation. Two strains that overproduce microconidia have been reported in the past, and they carry mutations in fl together with mutations in other genes (pe;fl and dn;fl) (Maheshwari, 1999). The presence of a mutation in fl in several microconidiating strains suggest that in order to overproduce microconidia, N. crassa requires a blockage of macroconidiation and mutations in other regulatory genes. Genetic mapping and mutant sequencing identified pe as an allele of ve- 1 (Scott Baker, Devin Wright, and Michael Freitag, pers. comm.), and it is possible that dn is an allele of ve-2 or a novel gene involved in the regulation of microconidiation. In addition, several conidiation (con) genes are expressed during macro- and microconidiation (Springer and Yanofsky, 1992) suggesting common regulatory elements, such as VE-1 and VE-2. It is tempting to hypothesize that the VE-1/VE-2 complex participates in the 103 CHAPTER II repression of microconidiation by regulating, in addition, the transcription of microconidiation-specific regulators yet to be identified. 4. Supplementary material Figure. II S1. Growth tube phenotypes and viability of conidia. (A) Photographs of culture tube phenotypes of wild type and mutants with cAMP supplementation. The graph on the right shows aerial tissue length in the wild-type, ∆ve-1, ∆ve-2, and ∆ve-1 ∆ve-2 strains after five days of growth at 30°C in light with 2.5mM cAMP with the mean and standard error of three independent experiments. (B) Conidial samples from the wild type, and the ∆ve-2, ∆vos-1, and ∆ve-2 ∆vos-1 mutant strains were collected and stored at 4ºC from 0 to 21 days. Conidial viability was assayed by plating diluted samples and assessing the number of colonies after two days of growth at 34ºC from each sample associated with period storage at 4ºC. The plot shows the average viability of conidia stored at 4ºC relative to the viability observed at the time of collection from each strain, and the standard error of three experiments. (C) Photographs of culture tube phenotype of wild type, ∆ve-1, ∆ve-2, ∆lae-1, ve-1FLAG, ve-2FLAG, and lae-1FLAG strains after three days of growth at 30°C in the light. The graph on the right shows the length of aerial tissue of the wild-type and other strains with the mean and standard error of three independent experiments (*p<0.05, **p<0.01, ***p<0.001, ****p<0.0001, two-way ANOVA). Chapter III. Sexual development 110 cascades are required for sexual development, but each cascade displayed distinct functions during this process (Lichius et al., 2012). Functional crosstalk between the N. crassa MAP Kinase pathways was observed during regulation of cell-to-cell fusion and communication, and during the development of perithecia (Lichius et al., 2012; Leeder et al., 2013; Kamei et al., 2016; Fischer et al., 2018; Serrano et al., 2018; Gonçalves et al., 2020; Lan et al., 2021). The MAK-2 MAP Kinase pathway is predicted to act downstream of the G-protein signaling in the pheromone response, based on the observed similarities in the female fertility phenotype of their mutants and the analogy with the yeast S. cerevisiae (Kim and Borkovich, 2004). Additionally, many other signal transduction pathways have been reported to play a role in sexual development in N. crassa and other fungi, such as the STRIPAK pathway, the NOX pathway, the COP-9 and the PACC signal transduction pathways (Cano-Domínguez et al., 2008; Nahlik et al., 2010; Chinnici et al., 2014). The final step of the signal transduction cascades is the activation of target effectors such as transcription factors that are involved in the transcriptional regulation of cell differentiation genes among other processes. In recent years, many genes that encode transcription factors with a role in fruiting body formation have been functionally characterized (adv-1, fsd-1, rco-1, tcf-9, ada-3, ff-7, and pp-1). ADV-1 (arrested development-1) and PP-1 (protoperithecium-1) are conserved transcription factors in N. crassa required for cell fusion and communication during colony establishment, as well as for normal conidiation and female fertility (Li et al., 2005; Fu et al., 2011; Fischer et al., 2018). PP-1 is a zinc finger transcription factor homologous to Ste12, which is a downstream target of the Fus3/Kss1 (MAK-2 in N. crassa) pheromone response pathway in S. cerevisiae (Leeder et al., 2013). In N. crassa, PP-1 also acts downstream of the MAK- 2 pathway. Additionally, ADV-1, which is orthologous to the early fruiting body regulator PRO-1 of S. macrospora, functions downstream of the MAK-1 (CWI) pathway (Fischer et al., 2018). PP-1 also directly regulates the expression of adv-1, and there are several points of crosstalk between the MAK-1/ADV-1 and MAK-2/PP-1 pathways, which form a regulatory network that coordinate the transcriptional regulation of many genes involved in growth and development (Fischer et al., 2018). The transcriptomic landscape of N. crassa during the formation of the fruiting body has been examined by RNA sequencing experiments (Wang et al., 2012; Lehr et al., 2014; Wang et al., 2014), revealing major changes in the transcriptome of developing fruiting bodies during the sexual cycle, as well as a stage-specific expression pattern for many key 111 CHAPTER III genes during sexual differentiation. Similar transcriptional changes have also been found by many transcriptome analyses in other fungal species (Lütkenhaus et al., 2019; Kim et al., 2022; Nowrousian, 2022). In addition, comparative transcriptome analysis between different fungi has allowed the identification of putative regulators of sexual development by selecting those genes that display a conserved and a stage-specific expression profile during sexual development in different lineages (Nowrousian, 2022). The fungal-specific velvet protein complex is a central regulatory complex for the regulation of sexual development, among other developmental programs, in response to specific environmental conditions such as light (Sarikaya-Bayram et al., 2015; Sarikaya Bayram et al., 2019). In Aspergillus nidulans VeA and VelB physically interact and activate sexual development in the dark (Bayram et al., 2008a). In Aspergillus flavus, VeA, VelB and LaeA are required for the formation of sclerotia, structures where fruiting bodies are produced inside (Amaike and Keller, 2009; Chang et al., 2013; Smith et al., 2015). Trichoderma reesei VeA protein, VEL1, acts as a positive regulator of sexual development and is involved in the response to pheromones and communication between mating partners (Karimi Aghcheh et al., 2014; Bazafkan et al., 2015; Bazafkan et al., 2017). In N. crassa, a velvet complex composed of VE-1, VE-2, and LAE-1 has previously been characterized (Chapter I). Evidence for the role of the velvet complex VE-1/VE-2/LAE-1 in sexual development has been previously observed (Sarikaya Bayram et al., 2019). However, the role of the velvet complex in light-mediated regulation of gene expression during sexual development have not yet been established. The present study aims to investigate the regulatory role of VE-1 in sexual development, and to provide insight into the regulatory relationship between light and the velvet complex in female fertility. We performed a comparative genome-wide transcriptional analysis between the wild-type strain and the deletion mutant ∆ve-1 kept in the dark or exposed to light during the sexual cycle. We propose that VE-1, together with VE-2, plays a major role as a light-dependent positive regulator of female fertility and that deletion of VE-1 controls the expression of many genes involved in sexual development, including components of the N. crassa MAP Kinase signaling transduction pathways. Chapter III. Sexual development 112 2. Results 2.1. The N. crassa velvet proteins VE-1 and VE-2 play a role in sexual development During the daily work with N. crassa and the velvet complex mutants in the laboratory, we observed certain light-dependent alterations in different aspects of sexual development in crosses involving velvet mutants. In addition, the observations that the velvet complex is involved in the formation of protoperithecia in N. crassa (Sarikaya Bayram et al., 2019), prompted us to characterize in detail the role of VE-1 and VE-2 in the sexual cycle and, in particular, the connection with light. Strains with opposite matingtype were confronted in pairs, inoculated on opposite sides of low-nutrient solid medium plates (SCM plates). Crosses were made between the wild-type strain and single mutants, ∆ve-1 and ∆ve-2, in all possible mating-type combinations. The experiment was carried out under complete darkness or continuous light (Figure III. 2). Figure III. 2. Barrage formation in mating experiments of the wild-type strain (WT) and velvet mutants ∆ve-1 and ∆ve-2. Strains of opposite mating type were confronted by inoculating 3 µl of a conidia suspension (104 conidia/ml) on opposing sides on SCM plates followed by incubation at 22°C for 14 days. The plates were washed with ethanol to remove the aerial mycelium covering the plates before taking the images. A barrage has been previously defined as any form of demarcation near the point of contact between two genetically different individuals (Esser and Meinhardt, 1984). At the interface between wild-type strains (WT mat a x WT mat A), a well-defined double dark line of perithecia separated by a clear zone containing a low density of hyphae was observed 113 CHAPTER III in the meeting zone. This barrage type is close to that of a 'standard mating reaction,' previously defined as the barrage that occurs along the contact border between aconidial (fluffy) mutants of opposing mating types (Griffiths and Rieck, 1981). However, the barrage generated from the cross between the wild-type strain and the mutants ∆ve-1 or ∆ve-2 showed differences compared to the standard mating reaction described previously, and these differences were dependent on light. In darkness, both dark lines were observed adjacent to the zone of contact, but the line corresponding to the mutant appeared to be more diffuse than the line corresponding to the wild type, with fewer mature perithecia and most of them embedded in the agar. In light, the alterations were more evident with only one well-defined line of perithecia, no clear zone (low density of hyphae) formed in the zone of interaction, and an even distribution of the perithecia in the mutant side. Most of the female structures that developed in the mutant side showed a lower level of maturation and were mainly embedded in the agar. Confrontations between single mutants in light resulted in impairments in the formation of a well-defined barrage in the interaction area with a large amount of perithecia embedded in the agar. In the crosses with either ∆ve-1 and ∆ve-2 mutant strains, ascospore release was rarely observed but manually collected ascospores from squashed perithecia were viable. These results suggest that the velvet proteins VE-1 and VE-2 may play a light-mediated role in sexual development during protoperithecia formation and maturation, while they do not appear to be required to produce viable ascospores. 2.2. The members of the velvet complex in N. crassa are not involved in male fertility To further characterize the defects of the velvet mutants ∆ve-1 and ∆ve-2 in sexual development, and to determine the possible causes of these alterations, we decided to study the role of VE-1 and VE-2 during mating when they act as males. In the experiments, ∆vos- 1 and ∆lae-1 mutants were included, as the corresponding proteins are part of the N. crassa velvet complex. We inoculated equal amounts of conidia from the wild-type strain in the center of SCM plates and allowed the development of protoperithecia for seven days. Then, the wild-type strain was fertilized by applying conidia from each mutant strain. The plates were further incubated for seven days to allow perithecia maturation and production of ascospores. After seven days of incubation, normal fertilization and morphological Chapter III. Sexual development 114 development were observed for all mutants, as indicated by the normal size and color of mature perithecia, as well as the development and release of ascospores (Figure III. 3). These results were found to be independent of the light condition and the mating-type (results not shown). Our results show that the velvet proteins, VE-1, VE-2 and VOS-1, and the methyltransferase LAE-1 do not play a role in male fertility during sexual development in N. crassa. Figure III. 3. VE-1, VE-2, VOS-1, and LAE-1 are not required for male fertility during sexual development. Plates inoculated with 5 µl of a wild-type conidia suspension (104 conidia/ml) were incubated for seven days at 22°C to allow protoperithecia formation and were covered with a suspension of 106 conidia from mutant strains. The plates were incubated for a minimum of seven additional days under the same conditions until perithecia matured and ascospores were released. The plates were washed with ethanol to remove the aerial hyphae covering the plates prior to imaging. 2.3. Proteins VE-1, VE-2, and LAE-1 are required for female fertility during sexual development in N. crassa Since deletion of any of the members of the velvet complex did not show any impairment in male fertility during sexual reproduction, we decided to characterize the role of these proteins during the formation of protoperithecia and the maturation of perithecia. Conidia of a wild-type and mutant strains of the genes ve-1, ve-2, vos-1 and lae-1 were spread on SCM plates and incubated at 22°C under two experimental conditions, continuous darkness and continuous light. The formation of protoperithecia was observed in the wild-type strain as expected after 5-7 days of incubation. The crosses were then carried out as previously explained by spraying a suspension of conidia of the wild type of the opposite mating type and incubating the plates for up to seven additional days to allow maturation of perithecia and formation of ascospores. In the wild-type strain, no obvious tissue differentiation was observed within 24 h after fertilization, except for a slight increase in size and a slight darkening in color. As expected, the fertilized perithecia 115 CHAPTER III underwent maturation, i.e., they enlarged and darkened in the following five days after fertilization, which is indicative of the formation of melanin during perithecial development. In the 5- to 6-days interval, we observed the formation of a protuberance that led to the perithecial beak for the dispersal of ascospores. Comparisons between the wild-type strain and the single velvet mutants revealed defects in sexual development in mutants ∆ve-1 and ∆ve-2 under both dark and light conditions. Our analysis revealed that the ∆ve-1 and ∆ve- 2 mutants showed a reduction of 35% in the number of perithecia after seven days of incubation in the dark, and this reduction increased to 98% (in ∆ve-1) and 90% (in ∆ve-2) when plates were incubated under continuous light (Figure III. 4). A B Figure III. 4. VE-1, VE-2, VOS-1, and LAE-1 participate in the development of female structures during sexual development. (A) Formation of perithecia of wild-type, ∆ve-1, ∆wc-1 ∆ve-1, ∆ve-2, ∆ve-1 ∆ve-2, ∆vos-1, and ∆lae-1 strains in SCM plate seven days after fertilization in light (upper panel) and dark (lower panel). (B) Quantification of the total number of perithecia in plates seven days after crossing. The plot shows the mean and standard error of four independent experiments (* p< 0.05, ** p<0.01, ***p< 0.001, ****p<0.0001). We also assayed the development of female structures in mutants of the vos-1 and lae-1 genes. Deletion of vos-1 did not appear to have any effect on the number of perithecia. For the mutant ∆lae-1, a significant reduction was observed in the number of perithecia compared to the wild-type strain. However, in this case, the reduction was more evident when the experiment was performed in the dark (about 60% reduction in the number of perithecia) than in the light (about 35% reduction) (Figure III. 4). Chapter III. Sexual development 116 The observation that light led to a larger reduction in the amount of perithecia in the ∆ve-1 strain suggested that light, through the WC-1 photoreceptor, was inhibiting the development of perithecia in the absence of VE-1. To test this hypothesis, we analyzed the phenotype of a double mutant carrying deletions in the genes ve-1 and wc-1. The double mutant ∆wc-1 ∆ve-1 in light behaved as the single mutant ∆ve-1 in dark, confirming that the light-dependent inhibition of perithecial development in the absence of VE-1 required the WC-1 photoreceptor (Figure III. 4). To explore whether the phenotypes observed in ∆ve-1 and ∆ve-2 mutants were due to a reduction in the formation of protoperithecia, in the maturation of perithecia, or in both, we monitored protoperithecia development in the interval between five and seven days before fertilization. When cultures were incubated under light, a reduction in the formation of protoperithecia was observed for each single mutant (Figure III. 5). The effect of light in protoperithecia development in the ∆ve-1 mutant required WC-1 as shown by the phenotype of the double mutant ∆wc-1 ∆ve-1. Our results suggest a role for VE-1 and VE-2 during the formation of protoperithecia in the dark and a more relevant role during sexual development in light Figure III. 5. Formation of protoperithecia in the wild-type, ∆ve-1, ∆ve-2, and ∆wc-1 ∆ve-1 strains. A total of 2x104 conidia were spread on SCM plates and incubated for seven days at 22°C under both dark and light conditions prior to observing the development of protoperithecia. Scale bar = 1mm. Most of the perithecia observed in the wild-type strain seven days after fertilization were fully developed, black, and with a beak at the top of the perithecium. However, some variability in the size and maturation of the perithecia was found in the ∆ve-1 and ∆ve-2 strains under both dark and light conditions. Thus, we characterized the size distribution of the perithecia produced by the velvet deletion mutants by measuring and analyzing female 117 CHAPTER III structures with a size greater than 120 µm, and comparing them with those formed by the wild-type strain. The strains carrying a deletion in the genes ve-1 and ve-2 showed a reduction in the size of perithecia in dark and light. When sexual development was performed in the light, the number of perithecia was drastically reduced, as previously described, and most of the female structures found on the plate were smaller than 120 µm. Under dark conditions, we also observed a significant reduction in the median of the perithecia size. This size reduction is more significant in the double mutant ∆ve-1 ∆ve-2 under both light and dark conditions (Figure III. 6A-B). A B Figure III. 6. Characterization of the morphology and size of the perithecia in velvet mutants. (A) Microscope images showing perithecia developed after seven days postfertilization in the wild-type strain and single and double ∆ve-1 and ∆ve-2 mutants. Scale bar = 2 mm. (B) Measurement of the size of the perithecia population in the wild-type strain and single and double ∆ve-1 and ∆ve-2 mutants. The asterisks show significance with a two-way ANOVA test (* p< 0.05, ** p<0.01). Chapter III. Sexual development 118 In addition, we observed differences in the formation of the beak between the wild-type and the mutant strains. Most of the perithecia produced by the ∆ve-1 strain were not fully developed, as mature beaks were hardly observed in the top of the perithecia (Figure III. 7). As expected, the ∆ve-2 mutant displayed similar but milder alteration in the morphology of the beak. The alterations in the morphology of the perithecial beak may explain the reduction in the release of ascospores in strains lacking velvet genes. Figure III. 7. Characterization of perithecia in velvet mutants. SEM images of the perithecia of the wild-type strain and the ∆ve-1, ∆ve-2, and ∆ve-1 ∆ve-2 strains cultured in complete darkness for seven days after fertilization. Scale bar = 100 µm. Samples were visualized with a JEOL 6460LV instrument. 119 CHAPTER III 2.4. N. crassa VE-1, VE-2, and LAE-1 proteins accumulate in early stages of sexual development The amount of the VE-1 protein during asexual development was previously reported to be regulated by light (Gil-Sánchez et al., 2022). To better understand the role of the velvet complex in the regulation of female fertility, we further analyzed the expression of ve-1, ve-2, and lae-1 during sexual development. The transcripts of the ve-1, ve-2, and lae- 1 were detected in all stages of sexual development (from two days before fertilization to six days after fertilization). Maximum ve-1 mRNA accumulation was observed two days after fertilization. However, the VE-1 and VE-2 proteins were not detected after fertilization, which correspond to the maturation stages of perithecia and the formation and maturation of ascospores (three and six days after fertilization) (Figure III. 8). The accumulation of VE-1 and VE-2 was higher in dark than in light. The accumulation of LAE-1 was reduced during later stages of sexual development regardless of exposure to light. The effect of light on VE-1 and VE-2 accumulation and the overall reduction in accumulation of the components of the velvet complex during sexual development did not correlate with mRNA accumulation suggesting the degradation of these proteins as perithecia mature. 2.5. Comparative transcriptomic analysis reveals major transcriptional changes of the ∆ve-1 strain during sexual development Our discovery that VE-1 was required for gene expression during conidiation, and the alterations observed in sexual development in the ∆ve-1 mutant, prompted us to characterize the role of VE-1 in transcriptional regulation during sexual development. To identify changes in the transcriptome due to the lack of VE-1 we first compared the transcriptomes of the wild-type and ∆ve-1 strains during sexual development in darkness and continuous light (Figure III. 9A). Wild-type and ∆ve-1 cultures from seven stages during the sexual cycle, including the last stages of protoperithecia formation (two days and one day prior to the fertilization point) and four stages of maturation of perithecia (the fertilization day (day 0), and one, four and six days after fertilization) were analyzed by RNA sequencing. Transcripts identified in our RNAseq libraries were mapped to 10,285 genes in the reference genome of N. crassa (10,590 genes in total) indicating that most genes are expressed during sexual development (Data Set S5). Identification of Chapter III. Sexual development 126 2.8. VE-1 regulates the expression of serine/threonine kinases and serine/threonine phosphatases Serine/threonine (S/T) kinases and phosphatases are essential components of signaling pathways in eukaryotes (Dickman and Yarden, 1999). The genome of N. crassa encodes 86 genes for serine/threonine protein kinases (Galagan et al., 2003; Park et al., 2011; Carrillo et al., 2020) that can be further classified into different classes based on their catalytic domain (AGC, CAMK, CK1, CMGC, STE, and others) (Park et al., 2011). Mutants in 77 kinase genes have been previously characterized, and 32 mutants exhibited defects in sexual development (Park et al., 2011). 20 kinases have been identified in our set of VE-1-regulated genes, of which 11 were required for sexual development (Figure III. 13A; Table III.1). We identified genes for components of the three MAP kinase signaling pathways, including components of the PR pathway (nrc-1, mek-2 and mak-2) and the CWI pathway, (mak-1) that showed lower mRNA accumulation in the ∆ve-1 mutant, unlike os-4 that showed higher mRNA accumulation in the mutant (Figure III. 13A). Most of the misregulated genes were identified in cultures kept in light (20 genes) than in dark (3 genes). Phosphatases participate in signal transduction pathways in coordination with kinases (Dickman and Yarden, 1999; Ghosh et al., 2014). There are 24 serine/threonine and tyrosine protein phosphatases in the N. crassa genome that have been functionally characterized and are required for growth and development (Ghosh et al., 2014; Carrillo et al., 2020). We identified six in our list of VE-1-regulated genes that are required during sexual development (Figure III. 13B; Table III.1). Our transcriptomic results suggest major changes in mRNA accumulation of genes that participate in MAP Kinase signaling pathways (kinases and phosphatases) that may result in altered signal transduction during sexual development, particularly in cultures kept in light 127 CHAPTER III Figure III. 13. Transcriptomic profiles of genes encoding S/T kinases (A) and S/T phosphatases (B) during sexual development. Heatmap showing relative transcription levels of genes misregulated by absence of VE-1 during sexual development in the dark (upper panel) or in the light (lower panel). Kinases stk-9, ime-2 and stk-14 (A) and phosphatases pty-2, dsp-3 and ppt-1 (B) were misregulated in dark and light. The table on the right side of the figure is a summary of the phenotype for the mutants lacking each kinase or phosphatase. The gray circles denote that the indicated structure shape or number of structures formed is abnormal compared to the WT, while black circles show that the indicated structure is not formed. The absence of a circle indicates there was no defect observed in sexual development. PP, Protoperithecia; P, Perithecia; A, Ascospores. Chapter III. Sexual development 128 Table III. 1. Signal transduction pathway genes required for sexual development that are misregulated in the ∆ve-1 mutant. 1. Alteration in sexual development. PP, Protoperithecia; P, Perithecia; A, Ascospores. Text in grey indicates that the structure shape or number of structures formed is abnormal compared to the WT; text in black shows that the indicated structure is not formed. Dark Light Dark vs Light Dark Light NCU00138 pre-1 STE3 P / A pheromone receptor pre-1 - - - - YES Kim and Borkovich 2004 NCU02500 ccg4 P / A clock-controlled pheromone ccg-4 YES YES YES YES - Kim and Borkovich 2006 NCU06182 nrc-1* STE11 PP / P / A MAPKK kinase - - - - YES Lichius et al. 2012 NCU04612 mek-2* STE7 PP / P / A Mitogen activated protein kinase kinase 2 YES YES YES -YES Lichius et al. 2012 NCU02393 mak-2* FUS3/KSS1 PP / P / A Mitogen-activated protein kinase 2 YES YES YES -YES Li et al. 2005 NCU00340 pp-1* STE12 PP / P / A transcription factor steA YES YES YES -YES Li et al. 2005 NCU09842 mak-1* MPK1 PP / P / A mitogen-activated protein kinase MKC1 - - YES -YES Park, Pan, and Borkovich 2008 NCU07392 adv-1* PP / P / A transcriptional regulatory protein pro-1 - - - - YES Carrillo et al. 2020; Fu et al. 2011 NCU03071 os-4 SSK22 PP / P / A MAP kinase kinase kinase SskB - - - - YES Lichius et al. 2012 NCU02794 so/ham-1* PP / P / A Fso1 YES YES - - YES Fleißner et al. 2005; Carrillo et al. 2020 NCU03727 ham-2* PP / P / A Hyphal anastamosis-2 protein - - - - YES Simonin et al. 2010 NCU01789 ham-5* PP / P / A Hyphal anastamosis-5 protein - - - - YES Dettmann et al. 2014 NCU02767 ham-6* PP / P / A Hyphal anastamosis-6 protein - - - YES YES Fu et al. 2014 NCU00881 ham-7* PP / P / A Hyphal anastamosis-7 protein YES - - YES YES Maddi et al. 2012 NCU02811 ham-8* PP / P / A hyphal anastamosis-8 protein - - - YES YES Fu et al. 2014 NCU10518 whi-2* WHI2 PP / P / A Fso1 YES YES YES YES YES Fu et al. 2014; Gonçalves et al. 2020 NCU08380 csp-6 PSR1 PP / P / A plasma membrane phosphatase required for sodium stress response YES YES - - YES Ghosh et al. 2014 NCU06205 rco-1* TUP1 PP / P / A transcriptional repressor rco-1 YES - - YES YES Yamashiro et al. 1996 NCU09123 camk-1 CMK1/CMK2 PP / P / A Ca/CaM-dependent kinase-1, variant 2 YES -YES -YES Park et al. 2011 NCU09071 dbf-2* DBF2/ DBF20 PP / P / A AGC/NDR protein kinase - - - - YES Park et al. 2011 NCU00786 gpr-1 PG-protein coupled receptor - - - - YES Cabrera et al. 2015 NCU04626 gpr-2 PG-protein coupled receptor YES YES - - YES Krystofova and Borkovich 2006 NCU07399 stk-9 ENV7 PP / P / A serine/threonine protein kinase-9 - YES YES YES YES Park et al. 2011 NCU04426 div-4 CAK1 PP/ P / A cell division protein kinase - YES -YES YES Park et al. 2011 NCU03187 nim-1 KIN3 AG2-specific protein kinase nimA - - - - YES Park et al. 2011 NCU05658 stk-36 PP / P / A serine/threonine protein kinase-36 - - - - YES Park et al. 2011 NCU09064 stk-53 KCC4/ SNF1/ GIN4/ HSL1 Aserine/threonine protein kinase-53 - YES - - YES Park et al. 2011 NCU00914 stk-16 KIN4/ FRK1 PP / P / A serine/threonine protein kinase - YES YES -YES Park et al. 2011 NCU00406 vel STE20/ CLA4/SKM1 Aprotein kinase CHM1 - YES - - YES Park et al. 2011 NCU06685 stk-47 CTK1 PP / P / A protein kinase - - - - YES Park et al. 2011 NCU06330 dsp-3 PP / P / A hypothetical protein - YES -YES YES Ghosh et al. 2014 NCU08158 dsp-4 YVH1 Adual specificity phosphatase YES YES - - YES Ghosh et al. 2014 NCU08948 pph-11 NEM1 PP / P / A NIF domain-containing protein, variant 1 - YES YES -YES Ghosh et al. 2014 NCU04600 pph-8 PTC2/ PTC3 PP / P / A protein phosphatase-9 YES YES YES -YES Ghosh et al. 2014 NCU08301 pp4 PP / A calcineurin-like phosphoesterase - YES YES -YES Ghosh et al. 2014 Components of CWI and cell fusion MAP kinase pathways Other signaling pathways DEGs WT vs ∆ve-1 Reference information Pheromone response (PR) / female fertility and Cell fusion pathway Cell wall integrity (CWI) Osmosensing pathway (OS) Gene ID Name orthologous in S. cerevisiae Sexual alterations Fungi DB annotation DEGs WT 129 CHAPTER III 2.9. The expression of cell-to-cell fusion genes involved in sexual development is affected in the mutant ∆ve-1 Cell fusion events and cell communication play a key role in the life cycle of N. crassa and other filamentous fungi (Fleißner et al., 2008; Read et al., 2010). In the vegetative phase, hyphal fusion is essential for the initiation and establishment of the colony allowing communication between hyphae, and the regulation of its homeostasis by redistribution of water, nutrients, and signaling molecules within the colony (Fleißner et al., 2008). Cell fusion is also essential for fertilization during mating, and there is an open discussion regarding the role of fusion events for the formation of protoperithecia (Glass et al., 2004; Fleißner et al., 2008; Lichius et al., 2012). Mutants defective in hyphal fusion show alterations in the formation of aerial hyphae and display a 'flat' conidiation phenotype due to their short aerial hyphae (Aldabbous et al., 2010). Most cell fusion mutants are also affected in female fertility and cannot develop protoperithecia (Xiang et al., 2002; Fleißner et al., 2005; Aldabbous et al., 2010; Simonin et al., 2010; Fu et al., 2014). We have observed that VE-1 was required for the accumulation of mRNAs of genes encoding proteins that participate in the regulation of the Cell Wall Integrity (CWI) pathway (Table III.1). The proteins HAM-6, HAM-7, and HAM-8 form a sensory complex on the cell wall/plasma membrane that is involved in activation of the CWI pathway by activation of the kinase MAK-1 (Fu et al., 2014). The levels of ham-6, ham-7, and ham-8 mRNAs were lower in the ∆ve-1 mutant than in the wild-type strain in both dark and light (Figure III.14). We also observed a down-regulation of the cell fusion genes ham-9, and whi-2, which are involved in the coordination of MAK-1 and MAK-2 during cell fusion, and are also required for vegetative growth and development (Fu et al., 2014; Gonçalves et al., 2020). Only when the cultures were exposed to light, we did observe lower levels of mRNA in the ∆ve-1 mutant in the cell-to-cell fusion genes, so/ham-1, ham-2, ham-5 and mob-3. Mutants lacking the so/ham-1 gene show a pleiotropic phenotype, including shortened aerial hyphae, altered conidiation pattern, and female sterility (Fleißner et al., 2005). HAM-5 functions as a MAP kinase scaffold during fusion and deletion of ham-5 results in the production of low number of infertile protoperithecia (Dettmann et al., 2014). HAM-2 and MOB-3 are components of the STRIPAK complex, and this complex is essential for the nuclear accumulation of MAK-1 in a MAK-2 dependent manner by phosphorylation of MOB-3 (Dettmann et al., 2013). In addition to the genes shown in this Chapter III 130 section, genes encoding for central components of the MAP Kinase signaling cascades (mak-1, nrc-1, mek-2, and mak-2), as well as for transcription factors acting downstream of these cascades (pp-1 and adv-1) have been previously identified as cell fusion genes and are misregulated in the ∆ve-1 strain (Table III.1). Down-regulation of cell fusion genes is consistent with our observation that the defects observed in protoperithecia formation and the shortened aerial hyphae of the ∆ve-1 strain are reminiscent of cell fusion mutants (Fleißner et al., 2005; Maddi et al., 2012; Fu et al., 2014). Figure III. 14. Transcriptomic profiles of cell-to cell fusion genes during sexual development. Heatmap showing relative transcription levels of genes misregulated by absence of VE-1 during sexual development in the dark (upper panel) or in the light (lower panel). The table on the right side of the figure is a summary of phenotype for the mutants lacking each cell-fusion gene. The severity of the alterations is represented with a gray circle (reduced /increased number or abnormal shape) and a black circle (structure not formed). The absence of a circle indicates that not phenotype has been described. PP, Protoperithecia; P, Perithecia; A, Ascospores. 131 CHAPTER III 2.10. VE-1 controls the expression of transcription factors that regulate sexual development Regulation of gene expression by transcription factors (TFs) is vital for the proper coordination of growth and development in all organisms, and they represent the last step in signal transduction cascades. The genome of N. crassa contains 312 TF genes (Galagan et al., 2003; Carrillo et al., 2017), and we identified 75 TFs genes that changed their mRNA accumulation in the ∆ve-1 mutant in the dark (24% of total TFs in the N. crassa genome). Exposure to light increased the number of misregulated TFs to 137 (46% of total TFs in N. crassa genome) (Figure III. 15A). Most of the misregulated TF genes in the dark were also misregulated in the light (60 out of 75), and approximately 60% of the genes misregulated in the light were specific for this condition. This suggests a major role for VE-1 in the regulation of TFs during sexual development in light, consistent with the more drastic phenotype of ∆ve-1 in light. A GO term analysis of the set of genes that were misregulated in each condition revealed an enrichment of TF genes involved in nitrogen metabolism, development, reproduction, growth, and cell differentiation (Figure III. 15B). Deletion of VE-1-regulated TF genes produced alterations in asexual development, growth rate and sexual development. Eight TFs that were misregulated in the dark and 23 in the light displayed different defects in sexual development (alterations in protoperithecia, perithecia or ascospore formation), and seven of them were misregulated under both conditions (Figure III. 15C). Of the 22 TF genes with at least one sexual alteration in light, the mutation of ten genes resulted in defects in two aspects of the sexual cycle, and ten were defective in the three phenotypes tested. 13 genes lead to alteration in protoperithecia production, with mutants in seven genes leading to a complete block in the formation of premature female organs. This group includes fsd-1, rco-1, tcf-9, ada-3, ff-7, and the genes that encode downstream effectors of the signal transduction pathways MAK-2/PR and MAK-1/CWI, pp-1 and adv-1, respectively (Table III.1). There are several crosstalks between the MAK-1/ADV-1 and MAK-2/PP-1 pathways during the regulation of growth and development (Fischer et al., 2018). For example, pp-1 is the homolog of the S. cerevisiae STE-12, and in N. crassa, it also directly regulates the expression of adv- 1(Leeder et al., 2013; Fischer et al., 2018). The repressor RCO-1 is the homolog of S. cerevisiae Tup1, and it has been characterized as a pleiotropic transcription factor involved in cell fusion, growth, and development (Yamashiro et al., 1996; Aldabbous et al., 2010; Chapter III 132 Ruger-Herreros et al., 2014). Our results suggest that VE-1 has an effect on gene regulation during sexual development by altering mRNA accumulation of key TF genes. Figure III. 15. Transcriptional profile of genes involved in transcriptional regulation. (A) Venn diagram showing the number of TFs displaying differential expression in the ∆ve-1 mutant during sexual development in dark or light. (B) GO term analysis showing enrichment of categories in each treatment. (C) Heat map representing relative expression of misregulated TFs genes with a mutant phenotype in sexual development (PP for protoperithecia; P for Perithecia, and A for ascospores). Phenotype data were obtained from (Carrillo et al., 2017). The severity of the alterations is represented with a gray circle (reduced /increased number or abnormal shape) and a black circle (structure not formed). The absence of a circle indicates that not phenotype has been described. 133 CHAPTER III 2.11. Coordinated regulation of genes involved in fruiting body formation by VE-1 protein and the MAP kinases MAK-1 and MAK-2 Our RNA-seq data suggests that VE-1 regulates the mRNA accumulation of genes for proteins that participate in the CWI pathway and the PR pathway. These two pathways are not independent, and previous studies have found crosstalks between the two key kinases in each of these pathways, MAK-1 and MAK-2 (Maerz et al., 2008; Maddi et al., 2012; Kamei et al., 2016; Lan et al., 2021). Since the expression of the velvet genes ve-1 and ve-2 during protoperithecium formation is not affected in mutants that carry deletion of these MAP Kinase genes (mak-1 and mak-2) (Lan et al., 2021), this led us to hypothesize that VE-1 could act on the PR and CWI cascades, regulating directly or indirectly, the expression of genes in these pathways. To further explore this hypothesis, we compared the list of genes co-regulated by MAK-1 and MAK-2 during the early formation of the protoperithecium (Lan et al., 2021) with our list of genes regulated by VE-1 to identify downstream effectors that could be co-regulated by VE-1 and these two MAP Kinases. Our analysis revealed a large regulatory overlap between the VE-1 and MAK-1/MAK-2 kinases. Among the 438 genes previously identified to be co-regulated by MAK-1/MAK-2 (Lan et al., 2021) the expression of 265 genes was also VE-1-dependent during sexual development in light (corresponding to 60% of the data set), 142 in the dark, and 117 genes overlapped under both conditions (Data Set S9). In the set of co-regulated genes between the MAP Kinase pathways and the VE-1, deletion mutants in 24 genes displayed a phenotype in sexual development with diverse defects in protoperithecium or perithecium formation (Carrillo et al., 2020; Lan et al., 2021). Specifically, ten single-gene deletion mutants (pfd- 1, oli, NCU05948, tcf-27, cea-6, mol-1, fpo-1, acw-5, gld-1, and eas) have been characterized as female sterile, as no perithecium formation was observed after mating (Lan et al., 2021). The deletion mutants of three of them (pfd-1, oli and pfd-2) shared a similar phenotype with the ∆mak-1 and ∆mak-2 strains, failing to form the protoperithecium (Lan et al., 2021), while mutants with deletion of tcf-27, cea-6 and mol-1 showed a reduction in protoperithecia formation (Lan et al., 2021). The other four genes were not required for protoperithecia formation, but the corresponding mutants showed defects in protoperithecia maturation, mating, or perithecia development (Lan et al., 2021). These results suggest a coordinated regulatory role between VE-1, MAK-1, and MAK-2 in the transcription of several genes required for perithecium development. Whether VE-1 directly regulates the Chapter III 134 expression of all these genes or acts indirectly upstream and/or at the MAP Kinase cascades remains to be investigated. 2.12. Melanin synthesis genes are involved in perithecium development and are downregulated in the ∆ve-1 strain During sexual development, the melanin synthesis pathway is responsible for producing the dark pigment (melanin) in ascospores and perithecia. The genes involved in the melanin synthesis pathway were downregulated in the ∆ve-1 mutant, which is consistent with the decrease in melanization of the mutant perithecia. Four of the genes involved in the melanin pathway (per-1, pkh-1, pkh-2 and lacm-1) were less expressed in the ∆ve-1 mutant compared to the wild-type strain throughout sexual development under both experimental conditions (dark and light) (Figure III. 16). The polyketide synthase gene (per-1) is required for the first step of the pathway. Deletion of per-1 leads to unmelanized perithecia in strains that act as the perithecial parent (female) (Howe and Johnson1, 1975). The genes pkh-1 and pkh-2 encode two polyketide hydrolase proteins required for the second step of the pathway. The pkh-1 deletion mutants produce melanized and unmelanized ascospores in a 1:1 ratio when used as a female in a mating with the wild type (Ao et al., 2019). The laccase gene for melanization-1 (lacm-1) is one of the five laccases responsible for the final step in melanin formation. Melanization of the perithecium and ascospores has been found to be delayed in a lacm-1 deletion mutant (Ao et al., 2019). Furthermore, two additional genes of the melanin pathway (pkr-2 and scy-1) and the transcriptional regulator mld-1 (Carrillo et al., 2017) were downregulated in the mutant only in light. Five of the genes involved in the synthesis of melanin (per-1, pkh-1, pkh-2, mld-1, and scy-1) are also included in the set of genes co-regulated by MAK-1 and MAK- 2 (Lan et al., 2021) suggesting that melanin biosynthesis is a key aspect in the development of perithecia, presumably due to their photoprotective capacities (Cordero and Casadevall, 2017). 135 CHAPTER III Figure III. 16. Expression of genes involved in the melanization pathway during the sexual development. Expression (CPMs) of genes involved in the synthesis of melanin that showed differential expression in the ∆ve-1 mutant throughout sexual development. The melanin biosynthetic pathway and the genes encoding the enzymes that participate in the pathway are shown on the right. 2.13. The VE-1/VE-2 protein complex binds to the promoters of MAP Kinase genes of the PR, CWI and OS signaling pathways The velvet domain of the VosA protein of A. nidulans is an RHD-like domain related to NF-kB TF and recognizes an 11-nucleotide sequence present in the promoter of many regulatory and structural genes (Ahmed et al., 2013). Since VE-1 contains a velvet domain it is tempting to speculate that VE-1 could regulate protoperithecia formation and maturation by binding directly to the promoters of key regulatory genes during sexual development. Therefore, we performed electrophoretic mobility shift assays (EMSA) to test whether the proteins VE-1, VE-2, and the heterodimer VE-1/VE-2 produced in E. coli cells in culture could bind to the promoters of some of the genes regulated by VE-1. We