Functional Genomics Provide New Insights Into Regulation of Morphogenesis and Secondary Metabolism in the Industrial Penicillin Producer Penicillium Chrysogenum

Functional genomics provide new insights into regulation of morphogenesis and secondary metabolism in the industrial penicillin producer Penicillium chrysogenum

Dissertation to obtain the degree Doctor Rerum Naturalium (Dr. rer. nat.) Submitted to the International Graduate School of Biosciences, Faculty of Biology and Biotechnology Ruhr-University Bochum, Germany

this thesis was performed at the Department of General and Molecular Botany

submitted by Kordula Becker from Essen, Germany

Bochum April, 2015

1st supervisor: Prof. Dr. Ulrich Kück

2nd supervisor: Prof. Dr. Franz Narberhaus

Funktionelle Genomanalysen zur Regulation von Morphogenese und Sekundärmetabolismus in dem industriellen Penicillin-Produzenten Penicillium chrysogenum

Dissertation zur Erlangung des Grades eines Doktors der Naturwissenschaften der Fakultät für Biologie und Biotechnologie an der Internationalen Graduiertenschule Biowissenschaften der Ruhr-Universität Bochum

angefertigt am Lehrstuhl für Allgemeine und Molekulare Botanik

vorgelegt von Kordula Becker aus Essen

Bochum April, 2015

Referent: Prof. Dr. Ulrich Kück

Korreferent: Prof. Dr. Franz Narberhaus

DANKSAGUNG

VIELEN DANK:

Meinem Doktorvater Prof. Dr. Ulrich Kück möchte ich für das mir entgegengebrachte Vertrauen, die hervorragende Betreuung, sowie für unzählige Gespräche danken, die mich nicht nur fachlich sondern auch persönlich voran gebracht haben. Ich habe in den vergangenen Jahren nicht nur die Privilegien sondern auch die Verpflichtungen, die mit der Arbeit an Ihrem Lehrstuhl einhergehen, zu schätzen gelernt und bin froh, mich für die Promotion in der Allgemeinen und Molekularen Botanik entschieden zu haben.

Herrn Prof. Dr. Franz Narberhaus gilt mein besonderer Dank für die Übernahme des Korreferates.

Allen aktuellen und ehemaligen Mitarbeitern des Lehrstuhls für Allgemeine und Molekulare Botanik danke ich für die freundschaftliche Zusammenarbeit während der vergangenen Jahre. Es waren viele kleine und große Dinge, die dazu beigetragen haben, dass ich meine Zeit als Doktorandin in guter Erinnerung behalten werde.

Mein Dank gilt insbesondere Ingeborg Godehardt, die mich tatkräftig bei der Durchführung sämtlicher DNA-Bindungsstudien unterstützt hat. Besser als mit den Worten eines Gutachters kann auch ich deine Arbeit nicht beschreiben: “The binding site analysis is compelling”. Mein Dank für hervorragende technische Unterstützung gilt außerdem Kerstin Kalkreuter und Stefanie Mertens. Ihr danke ich außerdem für viele ermutigende Worte, gute Ratschläge und ihr immer offenes Ohr. PD Dr. Minou Nowrousian danke ich für die professionelle Unterstützung bei zahlreichen bioinformatischen Analysen. Darüber hinaus möchte ich ihr und Dr. Julia Böhm ganz herzlich für die gewissenhafte und kritische Korrektur dieser Arbeit danken.

Meinen Doktorschwestern und Doktorbrüdern danke ich für die schöne gemeinsame Zeit. Ohne euch wäre alles nur halb so schön gewesen! Insbesondere möchte ich mich bei Tim Dahlmann bedanken – dank dir bin ich nicht nur zu einer realistischen Selbsteinschätzung meiner mathematischen Fähigkeiten gelangt, sondern hatte einen Büro- und Laborpartner, auf den ich mich stets verlassen konnte. – „Keks? Heißes Wasser?!“

Prof. Dr. Michael Freitag und seinen Mitarbeitern danke ich für den schönen Forschungsaufenthalt an der Oregon State University im Herbst 2012, der die Etablierung der ChIP-seq Technologie für die Anwendung in P. chrysogenum überhaupt erst ermöglicht hat.

Der Studienstiftung des Deutschen Volkes, der Sandoz GmbH, der Christian Doppler Forschungsgesellschaft und der RUB Research School danke ich für die großzügige finanzielle Unterstützung. Darüber hinaus gilt mein Dank unseren Kooperationspartnern bei der Sandoz GmbH, insbesondere Dr. Ivo Zadra und Dr. Hubert Kürnsteiner, für ihr fortwährendes Interesse am Fortgang dieser Arbeit.

TABLE OF CONTENTS 1

TABLE OF CONTENTS

ABBREVIATIONS ...... 2 I. INTRODUCTION ...... 3 1. From Sanger sequencing to next-generation sequencing ...... 3 2. New directions in functional genomics...... 5 3. Location-based NGS approaches ...... 7 4. Bioinformatics analysis of ChIP-seq data ...... 11 5. Functional downstream analysis: from physical context to biological function ...... 13 6. ChIP-seq analyses in fungi ...... 15 7. Summary ...... 15 II. SCOPE OF THE THESIS ...... 18 1. Regulation of fungal secondary metabolism ...... 18 2. Aim of this thesis ...... 20 III. BECKER et al. 2015a ...... 23 IV. BECKER et al. 2015b ...... 24 V. DISCUSSION ...... 25 1. ChIP-seq analyses with MAT1-1-1 ...... 25 1.1 MAT1-1-1 regulates target genes beyond sexual development ...... 26 1.2 Rewiring of MAT-regulated transcriptional networks ...... 30 1.3 A new MAT1-1-1 working model ...... 33 2. ChIP-seq analyses with PcVelA ...... 36 2.1 PcVelA acts as a transcriptional regulator on DNA level ...... 37 2.2 The putative SAM-dependent methyltransferase PcLlmA is a direct interaction partner of PcVelA ..... 40 2.3 An expanded model of PcVelA regulatory functions ...... 42 3. Overall analysis of ChIP-seq data ...... 43 3.1 Genome-wide TF binding beyond direct target-gene control ...... 43 3.2 MAT1-1-1 and PcVelA bind DNA via specific DNA-consensus sequences ...... 45 3.3 Concluding remarks ...... 47 VI. SUMMARY ...... 49 VII. ZUSAMMENFASSUNG ...... 50 VIII. REFERENCES ...... 51

IX. EIGENANTEIL AN PUBLIKATIONEN ...... 76 X. CURRICULUM VITAE ...... 77 XI. ERKLÄRUNG ...... 79 ABBREVIATIONS 2

ABBREVIATIONS bp base pairs BiFC bimolecular fluorescence complementation ChIP chromatin immunoprecipitation ChIP-chip ChIP combined with microarray hybridization ChIP-DNA DNA obtained from ChIP ChIP-PCR ChIP combined with PCR ChIP-seq ChIP combined with NGS DNA deoxyribonucleic acid ENCODE Encyclopedia of DNA Elements GRN gene regulatory network HMG high-mobility group input-DNA DNA sample removed prior to ChIP MAT mating type NGS next-generation sequencing NHGRI National Human Genome Research Institute nt nucleotide PCR polymerase chain reaction qRT-PCR quantitative real time PCR RNA ribonucleic acid RNA-seq RNA sequencing SAM S-adenosyl-L-methionine SM secondary metabolite TF transcription factor TFBS transcription factor binding site TSS transcription start site WGS whole-genome sequencing Y2H yeast two-hybrid analysis αsg mating-type α-specific gene Δ deletion I. INTRODUCTION 3

I. INTRODUCTION

1. From Sanger sequencing to next-generation sequencing

When Frederick Sanger first introduced his method to sequence DNA by “dideoxy chain-termination” and fragmentation techniques in 1977 (Sanger et al. 1977), few might have envisioned the revolutionary character of his discovery, for which he was awarded the Nobel Prize in Chemistry less than ten years later. The technique, commonly referred to as Sanger sequencing, dominated the DNA-analysis field for the next 30 years and, ultimately, enabled the completion of the first human genome sequence in 2004 (The International Human Genome Sequencing Consortium(2004). However, the Human Genome Project required vast amounts of time and resources, and an increasing demand for faster, cheaper, and higher-throughput technologies emerged. As a consequence, the National Human Genome Research Institute (NHGRI) initiated a funding program to reduce the sequencing cost of a human genome to US$1,000 within the next 10 years (Schloss 2008). Shortly afterwards, a new generation of sequencing technologies, summarily termed next-generation sequencing (NGS) technologies, arrived on the scene (Table 1). Compared to the Sanger method, which is considered a first-generation technology, NGS technologies share three characteristic features: [1] they rely on the preparation of NGS libraries in a cell-free system in place of bacterial cloning of DNA fragments, [2] instead of hundreds, thousands to many millions of sequencing reactions are produced in parallel, and [3] the sequencing output is directly detected without need for electrophoresis (van Dijk et al. 2014). Nevertheless, error rates are high and far apart from those of the Sanger sequencings’ > 0.001%. Furthermore, with some exceptions, read lengths are restricted to a maximum of some hundred base pairs. While brief descriptions of the most important NGS platforms, namely 454, Illumina, and SOLiD, will be given in the next paragraph, please refer to Mardis (2008), Metzker (2010), and van Dijk (2014) for descriptions of less commonly used techniques and detailed information.

The 454 Genome Sequencer, the first commercial NGS system for individual laboratory use was introduced by 454 Life Sciences (today Roche) in 2005. The system is based on the principle of “pyrosequencing”, a sequencing-by-synthesis technique that measures the release of inorganic pyrophosphate by chemiluminescence (Margulies et al. 2005). The DNA library is amplified by emulsion PCR (Tawfik and Griffiths 1998, Nakano et al. 2003) on the surface I. INTRODUCTION 4

Table 1: Sequencing technologies at a glance

Max. read length Output Platform Mechanism Run time Error rate [bp] data/run Sanger 3730xl1) 1st generation; 400 – 900 1.9 – 84 Kb 20 min – > 0.001% Dideoxy chain termination 3 hours 454 FLX+2) 2nd generation (NGS); 1000 0.7 Gb 23 hours > 0.8% Pyrosequencing Illumina HiSeq 2nd generation (NGS); 2*150 95 Gb 10 days > 0.8% 25003) Sequencing-by-synthesis SOLiD 5500xl1) 2nd generation (NGS); 75 15 Gb 8 days > 0.5% Ligation-based sequencing MiSeq3) 2nd generation (NGS); 2*300 15 Gb 3 days > 0.8% Sequencing-by-synthesis PacBio RS II4) 3rd generation; up to > 40,000 0.3 – 1 Gb 4 hours > 10% Real-time sequencing 1) http://www.lifetechnologies.com; 2) http://www.454.com; 3) http://www.illumina.com; 4) http://www.pacificbiosciences.com of 28-µm agarose beads, covered with millions of oligomers, which are complementary to the adaptor sequences used for NGS-library construction. After amplification, several hundred thousand such agarose beads, which harbor up to 1,000,000 copies of the originally annealed DNA fragment, are added to 454 picotiter plates for sequencing. Nucleotides are added in a defined manner and imaging of light signals, produced by a chemiluminescent enzyme present in the reaction mix, is used to record their incorporation. In 2006, one year after the introduction of the 454 platform, Solexa (today Illumina) commercialized the Genome Analyzer, which is based on the concept of “sequencing-by-synthesis” (Bentley et al. 2008). Here, during the so-called cluster generation, single-stranded DNA fragments are hybridized and amplified on the surface of a glass flow cell prior to sequencing. After the amplification step, flow cells contain more than 40 million clusters, each composed of approximately 1,000 copies of a single template molecule. The templates are sequenced in a massively parallel fashion using differentially 3’-labeled fluorescent nucleotides. Each base incorporation cycle is followed by an imaging step, identifying the incorporated nucleotide, and the chemical removal of the fluorescent group, introducing the next incorporation cycle. In 2007, shortly after commercialization of the Illumina platform, Applied Biosystems (today Life Technologies) released the SOLiD (Sequencing by Oligo Ligation Detection) system (Shendure and Ji 2008). SOLiD is based on massive parallel sequencing by ligation, using a ligation technique referred to as polony sequencing (Shendure et al. 2005). Here, adaptor-linked DNA fragments are coupled to magnetic beads, covered with complementary oligonucleotides. Bead-DNA complexes are amplified using emulsion PCR and subsequently beads are covalently attached to glass slides for sequencing. The “ligation-based sequencing” process starts with the annealing of universal sequencing primers, complementary to the I. INTRODUCTION 5 adaptor sequences used for library preparation. Next, 8mer oligonucleotides and a DNA ligase are added. Matching 8mer oligonucleotides next to the universal sequencing primer 3’-end are linked by the DNA ligase and, depending on the cycle number, either the fifth or second position of the 8mer is identified using a fluorescent readout. The 8mer is chemically cleaved between positions five and six, removing the fluorescent group and enabling the next round of ligation. This way, the sequence of each DNA fragment is identified at five-nucleotide intervals upon completion of the sequencing cycle. Synthesized fragments are denaturized and removed, and a second round of sequencing starts, using either an universal primer that is set back one or more bases from the adaptor-insert junction or differentially labelled 8mers.

Meanwhile, benchtop-sequencing machines, like Ion Torrent Systems’ (today Life Technologies) Personal Genome Machine (Rothberg et al. 2011) and Illumina’s MiSeq, are available on the market, making large-scale sequencing affordable even for small laboratories. Driven by the competition between the main players on the NGS market, sequencing costs for a human genome have fallen well below US$5,000 until today and, finally, Illumina’s HiSeq X Ten platform was announced to burst the US$1,000 boundary set by the NHGRI in 2004 (McPherson 2014).

The next step in development of new DNA-sequencing platforms now meets the challenge of single-molecule sequencing without any amplification of the DNA template, ushering in the era of third-generation sequencing. Currently, this field of DNA analysis is dominated by Pacific Biosciences (PacBio), which released the PacBio RS, a system based on the detection of natural DNA synthesis by a single DNA polymerase (Eid et al. 2009), and Oxford Nanopore Technologies (ONT), which started a testing phase for the MinION sequencing device, reading the sequences of individual DNA strands while driving them through biological nanopores, in 2014 (Schneider and Dekker 2012, Jain et al. 2015).

2. New directions in functional genomics

The advent of NGS technologies marked a radical change in genomics research and provided a virtually inexhaustible basis for a variety of functional-genomics approaches, aiming at turning the huge amount of data obtained by genomic projects into a description of the interactions between the genome, gene products, and metabolites. Doing so, functional-genomics approaches characteristically focus on dynamic aspects, such as protein-DNA interactions, transcription, and translation (Werner 2010). I. INTRODUCTION 6

One of the first NGS-based applications in the field of functional genomics was RNA sequencing (RNA-seq), enabling the identification and quantification of transcripts, even without prior knowledge of particular genes, and providing insights into alternative splicing events and sequence variation (Wang et al. 2009). Early applications of RNA-seq comprised the generation of high-resolution transcriptome maps of Saccharomyces cerevisiae and Schizosaccharomyces pombe (Nagalakshmi et al. 2008, Wilhelm et al. 2008), transcriptome analyses in Arabidopsis thaliana (Lister et al. 2008) and human HeLa cells (Morin et al. 2008), as well as mapping and quantification of mouse transcriptomes from brain, liver, and skeletal muscle tissues (Mortazavi et al. 2008). Another early application of NGS was chromatin immunoprecipitation followed by high-throughput sequencing (ChIP-seq), allowing the genome-wide identification of protein-DNA interactions and epigenetic marks, e.g. DNA methylation and/or histone modifications (Park 2009). The first studies using ChIP-seq were published in 2007. They described the identification of DNA-binding sites of the human neuron-restrictive silencer factor (NRSF) (Johnson et al. 2007), mapping of target regions of the transcriptional regulator STAT1 in interferon-γ-stimulated and unstimulated human HeLa cells (Robertson et al. 2007), and high-resolution profiling of 20 histone methylations along with histone variant H2A.Z, RNA Pol II, and the DNA-binding protein CTCF in the human genome (Barski et al. 2007).

Besides RNA-seq and ChIP-seq, one other major field of NGS applications is whole-genome sequencing (WGS). Most importantly, NGS-based WGS approaches not only reveal the genome sequence of interest, but also provide valuable information about genomic deletions, rearrangements, copy number variations (CNVs), and short nucleotide polymorphisms (SNPs). The first eukaryotic organisms to be sequenced exclusively by NGS were the giant panda (Li et al. 2010) and the filamentous ascomycete Sordaria macrospora (Nowrousian et al. 2010). Since then, NGS-based WGS has been successfully used in a variety of experimental approaches, e.g. for analyzing the epidemiology of Staphylococcus aureus clinical isolates (Francois et al. 2007), generation of a draft genome sequence of the Neandertal (Green et al. 2010), and determination of the genome sequence of an 18.5 weeks unborn human fetus (Kitzman et al. 2012). In fungi, NGS-based WGS pipelines have been applied for sequencing of Pyronema confluens (Traeger et al. 2013), as well as strains of the biotechnologically relevant species Penicillium chrysogenum (Specht et al. 2014) and Acremonium chrysogenum (Terfehr et al. 2014). Moreover, NGS-based workflows have been developed for the discovery of fungal secondary metabolite (SM) gene clusters (Cacho et al. 2015). One of the earliest examples describes the identification of two Penicillium I. INTRODUCTION 7 aethiopicum SM gene clusters, encoding the tetracycline-like viridicatum toxin 1 and the antifungal agent griseofulvin 2, by 454 shotgun sequencing (Chooi et al. 2010).

Until today, the enormous power of NGS promoted its establishment in various new fields of applications, such as non-invasive prenatal diagnostics (Nepomnyashchaya et al. 2013), clinical diagnostics (Desai and Jere 2012), forensics (Yang et al. 2014), molecular barcoding (Smith et al. 2010), metagenomics (Kim et al. 2013, Burton et al. 2014), and drug development (Rodriguez and Miller 2014).

3. Location-based NGS approaches

Transcription factors (TFs) are the most important players in gene regulatory networks (GRNs). They orchestrate the gene expression control of the cell and thereby determine organismal complexity and diversity (Shelest 2008, Charoensawan et al. 2010). Hence, a precise map of binding sites of DNA-binding proteins and epigenetic marks is vital for understanding the regulatory mechanisms that underlie various biological processes (Farnham 2009, Park 2009). Since the ascent of NGS technologies, large-scale projects aiming at creating comprehensive maps of GRNs have been initiated. Prominent examples are The Encyclopedia of DNA Elements (ENCODE), an international collaboration of research groups funded by the NHGRI (The_ENCODE_Project_Consortium 2004, 2012), and the Human Epigenome Project (Esteller 2006), both aiming at the provision of a platform for the improvement of human biology and health.

For a long time, binding sites of DNA-binding proteins and histone modifications have been analyzed using chromatin immunoprecipitation (ChIP) (Hecht et al. 1996, Strahl-Bolsinger et al. 1997). ChIP provides a snapshot of all factors bound to specific chromatin regions in different functional states, and therefore provides an exquisite tool to investigate the interplay between structural or regulatory proteins and DNA (Won and Kim 2006). A typical DNA-binding protein ChIP experiment is based on the enrichment of DNA fragments associated with a protein of interest, such as TFs, components of the core transcriptional machinery, or histones. Therefore, interactions between proteins and DNA are crosslinked in vivo by treating the starting material with formaldehyde, a highly reactive compound, which reacts with the amino groups of proteins and amino acids (McGhee and Hippel 1975a, 1975b, Orlando 2000). In the following step, the chromatin is sheared, e.g. by sonication or endonuclease treatment, to small fragments of ~ 200-500 nt in length, and protein-specific antibodies are used to immunoprecipitate the protein-DNA complexes of interest. Finally, the I. INTRODUCTION 8 crosslinks are reversed and the released DNA fragments are applied to downstream analyses. Downstream strategies reach from quantitative PCR (ChIP-PCR) to microarray hybridization (ChIP-chip) and NGS (ChIP-seq) approaches (Figure 1).

In case of ChIP-PCR analysis, the immunoprecipitated DNA (ChIP-DNA) is analyzed by PCR, in order to verify the association of the protein of interest to selected DNA regions (Tanaka et al. 1997, Chen et al. 1999). However, each protein-DNA interaction must be validated on its own, strongly restricting the information content of ChIP-PCR analyses in comparison to genome-wide approaches. This drawback was - up to a certain point - overcome by ChIP-chip, combining ChIP with microarray hybridization. Here, single- stranded ChIP-DNA fragments, labeled with fluorescent tags, are hybridized to a DNA microarray, and probe-target hybridization is quantified using a fluorescent readout. The first ChIP-chip analysis was performed in 1999 and described the distribution of two cohesins along the yeast chromosome III (Blat and Kleckner 1999). Soon after, the first combinatorial approach, using location and expression profiles from ChIP-chip and microarray experiments, was used to identify direct target genes of S. cerevisiae TFs Gal4 and Ste12 in response to changes in carbon source and the mating pheromone, respectively (Ren et al. 2000). In 2002, the first high-throughput ChIP-chip analysis, focusing on determination of genome-wide positions of over 100 S. cerevisiae TFs followed (Lee et al. 2002). For long, ChIP-chip has been the method of choice for studying gene regulation and epigenetic mechanisms on an almost genome-wide scale. However, application of this method was limited by the fact that whole-genome microarrays are very expensive and not available for many organisms. As a consequence and fostered by a tremendous progress in NGS technology, ChIP-seq took rapidly over from its array-based predecessor. Today, ChIP-seq is the gold standard for determining binding sites of DNA-binding proteins in vivo (Hahn and Young 2011). As shown in Table 2, compared to ChIP-chip, ChIP-seq offers higher, up to nucleotide-level resolution, fewer artefacts, a larger dynamic range, less noise and greater coverage (Park 2009).

Figure 1: Chromatin immunoprecipitation - from quantitative PCR to ChIP-seq. DNA-fragments, isolated using chromatin immunoprecipitation (ChIP), can be applied to single-gene studies based on quantitative PCR (ChIP-PCR) or genome-wide approaches using ChIP coupled to microarray hybridization (ChIP-chip) and next-generation sequencing (ChIP-seq). (adapted from Thürmer (2014)) I. INTRODUCTION 9

Originally developed for the identification of in vivo protein-DNA interactions on a genome-wide scale (Johnson et al. 2007), ChIP-seq soon has been adapted for the investigation of a wide variety of biological processes. The experimental procedures of some of the most important adaptations, which have been described in detail in Furey (2012), Dekker et al. (2013), and de Wit and de Laat (2012), are depicted in Figure 2. They can be used for studying RNA-protein interactions (cross-linked immunoprecipitation followed by next-generation sequencing = CLIP-seq) (Sanford et al. 2009), RNA-DNA interactions (chromatin isolation through RNA purification = ChIRP-seq) (Chu et al. 2011, Simon et al. 2011), as well as DNA-DNA interactions (chromosome conformation capture = Hi-C, 5C; chromatin-interaction analysis by paired-end sequencing = ChIA-PET) (Dostie et al. 2006, Dostie and Dekker 2007, Fullwood et al. 2009, Lieberman-Aiden et al. 2009). Furthermore, nucleosome-depleted open chromatin can be mapped using DNase-seq (DNase I hypersensitive sites sequencing) (Crawford et al. 2006, Boyle et al. 2008) and FAIRE-seq (formaldehyde-assisted identification of regulatory elements) (Gaulton et al. 2010, Song et al. 2011).

Development and standardization of the above mentioned and new pipelines, as well as ongoing cost reduction will surely lead to the establishment of NGS and, in particular, ChIP-seq as key technologies in basic science. However, a number of technical considerations must be taken into account when setting up ChIP-seq pipelines, aiming at exploiting the full strength of this technology (Landt et al. 2012, Flensburg et al. 2014, Jung et al. 2014). Important issues in experimental design are selection and validation of high-quality antibodies, optimization of sample quantity (dependent on the abundance of the chromatin-

Table 2: Comparison of ChIP-chip and ChIP-seq

ChIP-chip ChIP-seq Max. resolution Array-specific, generally 30-100 bp Up to single nucleotide, depends on size of chromatin fragments and sequencing depth Coverage Limited by sequences on the array; Limited only by alignability of reads to repetitive regions are usually masked out the genome; increases with read length; many repetitive regions can be covered Source of platform noise Cross-hybridization between probes and Some GC bias can be present non-specific targets Cost-effective cases Profiling of selected regions; when a Large genomes; when a small fraction large fraction of the genome is enriched of the genome is enriched for the for the modification or protein of interest modification or protein of interest Required amount of ChIP-DNA High (a few micrograms) Low (10 – 50 ng) Dynamic range Lower detection limit; saturation at high Not limited signal Multiplexing Not possible Possible (adapted from Park (2009)) I. INTRODUCTION 10

I. INTRODUCTION 11

Figure 2: Comparison of selected experimental pipelines for location-based NGS approaches. Simplified schematics of the main steps are shown. (A) Chromatin immunoprecipitation followed by next-generation sequencing (ChIP-seq) for the genome-wide analysis of DNA-binding patterns of DNA-binding proteins, such as TFs. (B) ChIP-seq of histone modifications, such as H3K4me3 or H3K9me3, can be performed for the identification of heterochromatic and euchromatic genomic regions. (C) DNase-seq and (D) formaldehyde-assisted identification of regulatory elements (FAIRE-seq) can be used for the identification of nucleosome-depleted open chromatin. (E) Analogous to ChIP-seq, CLIP-seq can be used to identify binding sites of RNA-binding proteins and (F) ChIRP-seq can be applied for the identification of genomic regions which are bound by a specific RNA or a ribonucleoprotein containing the RNA of interest. (G) Interactions between distal genomic regions on the same or different chromosomes can be analyzed using chromatin conformation capture, providing information about possible targets for DNA-bound proteins. (H) Similarly, chromatin interaction analysis with paired-end tag sequencing (ChIA- PET) provides information about chromatin interactions mediated by a specific DNA-binding protein, such as RNA polymerase II. (adapted from Park (2009) and Furey (2012)) associated protein targets and/or histone modifications), sequencing depth, and implementation of control experiments. Currently, there is no consensus on which control experiment is the most appropriate, but the most commonly used strategies are: [1] sequencing of input-DNA (DNA sample removed prior to ChIP), [2] mock IP DNA (DNA obtained from ChIP without antibody), or [3] DNA from non-specific ChIP (using an antibody against a protein not involved in DNA binding or chromatin modification).

4. Bioinformatics analysis of ChIP-seq data

NGS technologies are well suited to provide all the primary data required for a variety of functional-genomics approaches, reaching from genomics to epigenetics, transcriptomics, and TF-binding studies. Nevertheless, a thorough down-stream analysis is necessary to bridge mere data, represented by sequence tags, and functional connections of biological relevance (Werner 2010).

Bioinformatics analysis of location-based NGS projects starts with a vast amount of raw data, which are converted to sequence reads during the so-called base-calling process (Figure 3). Sequence reads are then mapped to the corresponding reference genome using short read aligners, optimized for extremely fast mapping of short NGS reads. Next, genomic regions that show statistically significant enrichment in the ChIP sample relative to the control are identified during the so-called peak-calling process. As peaks can be generally divided into three categories, selection of the right peak-calling software is essential for effective data analysis. Peak categories are based on the architecture of peak regions, which in turn is caused by the DNA-binding properties of the associated proteins. Peak categories are: [1] point source (highly localized signals, e.g. for TFs), [2] broad source (signals spanning lager domains, e.g. for some histone modifications such as H3K36me3), and [3] mixed source (signals that have elements of both, e.g. RNA polymerase II binding) (Pepke et al. 2009). I. INTRODUCTION 12

Several “peak-callers”, such as the very popular open source peak-caller MACS (Model- based Analysis of ChIP-Seq) (Zhang et al. 2008), are currently available. They have been summarized in great detail by Bailey (2013), Furey (2012), and Wilbanks and Facciotti (2010).

Since bioinformatics analysis of NGS data can be challenging for users who are not skilled in advanced bioinformatics, several platform based analysis tools have been released, some of them even enabling data analysis via desktop applications. Prominent examples are HOMER, offering solid tools and methods for analyzing and interpreting ChIP-seq experiments (Heinz et al. 2010), Nebula, a web service provided by the Institute Curie (Boeva et al. 2012), and the Cistrome Project, an integrative and reproducible bioinformatics data-analysis platform featuring 29 ChIP-seq-specific tools from preliminary peak calling to downstream analysis (Liu et al. 2011). However, even if the growing number of bioinformatics tools simplifies data analysis in a great measure, it has to be regarded that varieties in sample preparation as well as data acquisition, processing, and interpretation can introduce bias, which results in divergent conclusions that do not necessarily reflect the biology of the factors of interest (Bailey et al. 2013, Park et al. 2013, Meyer and Liu 2014).

Figure 3: Overview of ChIP-seq data analysis. Raw sequencing data obtained from the sequencing platform is converted into sequence reads during the base-calling process. Subsequently, sequencing reads are mapped to a reference genome. Several quality-control analyses, such as examination of base-calling reliability, clonal-tag distribution, and check for nucleotide frequency relative to read positions and GC bias, can be performed in order to provide information about the quality of the experiment. Next, peak calling, using data from the ChIP-DNA profile and, if available, a control profile (usually input-DNA), is performed. Regions showing statistically significant enrichment in ChIP-DNA compared to the input profile are submitted to downstream analyses. (adapted from Park (2009)) I. INTRODUCTION 13

5. Functional downstream analysis: from physical context to biological function

Starting from a set of peak regions, follow-on bioinformatics analyses can be used to address the various biological implications of ChIP-seq data (Figure 3). For DNA-binding protein ChIP-seq approaches, the most common follow-up analysis is the identification of DNA-binding sequence motifs, either by de novo motif prediction or by comparison with pre-defined weight matrices of known DNA-binding factors. However, as de novo pattern detection algorithms are capable of defining novel patterns without prior experience, they are more in line with the genome-wide unbiased approach of ChIP-seq (Werner 2010). To date, hundreds of algorithms for the prediction of DNA motifs have been developed (Tompa et al. 2005, Weirauch et al. 2013), most of them using position weight matrices (PWMs) for representation of protein-DNA binding specificity (Stormo and Zhao 2010). One of the most widely used tool for de novo motif prediction, MEME (Multiple Em for Motif Elicitation), also allows the analysis of very large ChIP-seq datasets (Machanick and Bailey 2011). Based on sequences obtained from ChIP-seq peak regions, MEME performs all steps necessary for efficient DNA-binding motif analysis, ranging from ab initio motif discovery to motif enrichment analysis, motif visualization, binding-affinity analysis, and motif identification. Nevertheless, these and other models are restricted to a description of the DNA-base readout by a DNA-binding protein, and they rely on the assumption that positions within a TF binding site (TFBS) independently contribute to the binding affinity of the corresponding protein (Slattery et al. 2014). In order to overcome these restrictions, efforts have been made to develop new, more complex models of protein-DNA interactions. For example, algorithms based on Bayesian networks (Friedman 2004), Markov networks (Sharon et al. 2008), or thermodynamic/energy-based models (Zhao et al. 2012) have been introduced.

When aiming at the identification of a DNA-binding consensus sequence based on the comparison to known DNA-binding motifs, the easiest way is the use publicly available databases. One of the most popular open-access databases is JASPAR, representing a collection of annotated, high-quality, matrix-based TFBS profiles for multicellular eukaryotes (Sandelin et al. 2004). The latest release of the database contains a curated non-redundant set of 593 DNA-binding motifs derived from published collections of experimentally defined TFBSs. 205 of them were validated in vertebrates and 177 in fungi (Mathelier et al. 2014). Even in case of successful de novo prediction of a new DNA-binding motif, comparison to a I. INTRODUCTION 14

DNA-binding motif database can be rewarding, as similarities to known motifs might point to related regulatory functions or binding properties of the corresponding proteins.

In case of TF-binding studies, another important step in functional downstream analysis is the elucidation of the biological relevance of the identified TFBSs in terms of transcriptional regulation of neighboring genes. Based on the observation that transcriptional control is highly combinatorial, which means that cooperative binding of multiple TFs and/or cooperative recruitment of RNA polymerase II is often required for transcriptional activation, it has been postulated that TF binding only indicates the potential for a neighboring gene to be regulated (Gao et al. 2004). Furthermore, based on TF binding alone, it is not possible to determine whether the respective protein acts as a transcriptional activator or as a repressor. Hence, incorporation of other data types into the analysis is inevitable to establish functional TF–target gene relations. For example, TF ChIP-seq data can be complemented by expression data from RNA-seq or microarray analyses. Doing so, direct TF target genes can be identified based on the correlation between binding strength of the protein, as deduced from ChIP-seq data, and the expression levels of neighboring genes. Corresponding approaches were successfully used for the identification of genes under direct control of TF SrbA in Aspergillus fumigatus (Chung et al. 2014) and analysis of the binding properties of the core circadian TF WCC in Neurospora crassa (Hurley et al. 2014). Alternatively, data from DNA-binding protein ChIP-seq experiments can be complemented by the corresponding data from histone-modification ChIP-seq analyses, enabling the identification of transcriptional active genomic regions next to peak regions. While trimethylation of H3K4, H3K36, and H3K79 is generally accepted to be associated with actively transcribed euchromatic regions, trimethylation of H3K9, H3K27, and H4K20 is characteristically for heterochromatin formation and transcriptional silencing (Noma et al. 2001, Berger 2007). A corresponding strategy was used by Thurtle and Rine (2014), who used ChIP-seq of Sir proteins, histones, and histone modification H4K16-ac for mapping of silenced chromatin in S. cerevisiae.

Once a set of differentially expressed target genes has been identified, Gene Ontology (GO) analyses can be performed to identify over-representation of genes assigned to particular molecular functions or biological processes (Ashburner et al. 2000) and, finally, comparative analysis of multiple ChIP-seq data sets can be used for the generation of differential binding profiles, e.g. as a function of developmental processes (Li et al. 2015), in response to external stimuli (Chung et al. 2014), or for functional distinction of multiple factors involved in the same regulatory network (Fitzgerald et al. 2014). I. INTRODUCTION 15

6. ChIP-seq analyses in fungi

While ChIP-seq has been widely used for analyzing DNA-binding proteins and epigenetic mechanisms, especially in mice and human tissues, the number of published experiments performed in fungi is rather limited (Table 3). The very first publication dates back to 2007 and describes the generation of a comprehensive map of H2A.Z nucleosomes in S. cerevisiae by sequencing of DNA from 322,000 individual nucleosomes (Albert et al. 2007). In 2011, the first ChIP-seq analysis performed in a filamentous ascomycete, namely N. crassa, followed. Here, extensive co-localization of centromeric proteins CenH3, CEN-C, CEN-T, and histone H3K9me3 was demonstrated, and a model, in which centromere proteins nucleate at the core kinetochore but require additional factors for spreading, was proposed (Smith et al. 2011). Meanwhile, ChIP-seq experiments in various other fungi followed. For example, in the biotechnologically relevant species Fusarium fujikuroi and Trichoderma reesei, ChIP-seq was used for the analysis of epigenetic marks linked to SM gene-cluster expression (Seiboth et al. 2012, Karimi-Aghcheh et al. 2013, Niehaus et al. 2013, Studt et al. 2013, Wiemann et al. 2013). Furthermore, DNA-binding protein ChIP-seq analysis was used for the generation of a genome-wide binding profile of TF Tri6, which is involved in the regulation of trichothecene gene-cluster expression in Fusarium graminearum (Nasmith et al. 2011).

7. Summary

Driven by an increasing demand for faster, cheaper, and higher-throughput sequencing technologies, development of NGS technologies led to dramatic changes in genomics research during the past decade. Today, virtually all functional-genomics approaches can be addressed by NGS-based experimental pipelines. Specialized protocols have been published for use of NGS in WGS and transcriptomics (RNA-seq), as well as for the investigation of protein-DNA interactions (ChIP-seq, DNase-seq, FAIRE-seq), protein-RNA interactions (CLIP-seq), RNA-RNA interactions (ChIRP-seq), and chromatin conformation studies (Hi-C, 5C, ChIA-PET) (Furey 2012).

When focusing on location-based NGS approaches aiming at the generation of precise maps of epigenetic marks and genome-wide TF DNA-binding profiles, the advantages of ChIP-seq compared to its predecessor ChIP-chip are obvious. Compared to ChIP-chip, ChIP-seq offers higher resolution, fewer artefacts, a larger dynamic range, less noise, and greater coverage (Park, 2009). However, conscientious experimental design and thorough downstream analysis are necessary to exploit the full strength of this versatile technology (Landt et al. 2012). I. INTRODUCTION 16

Table 3: ChIP-seq analyses in fungi

Species Experimental approach Outcome Reference Aspergillus fumigatus ChIP-seq of Identification of genes under direct SrbA (Chung et al. 2014) transcription factor transcriptional regulation in hypoxia SrbA; RNA-seq Candida albicans ChIP-seq of histone Set3C acts as a transcriptional co-factor of (Hnisz et al. 2012) deacetylase Set3C; metabolic and morphogenesis-related genes RNA-seq Candida parapsilosis ChIP-seq of TF Efg1 Genome-wide binding profile of Efg1, a (Connolly et al. transcriptional regulator of morphogenesis, 2013a) biofilm formation, and virulence Cryptococcus ChIP-seq of H3K9ac; A homolog of the yeast protein Ada2, a (Haynes et al. neoformans RNA-seq member of the SAGA complex, is involved in 2011) the direct regulation of capsule and mating responses; it may also play a direct role in regulating capsule-independent anti-phagocytic virulence factors Fusarium fujikuroi ChIP-seq of H3K4me2, Identification of epigenetic marks linked to SM (Wiemann et al. H3K9me3, and H3K9ac gene-cluster expression 2013) Fusarium fujikuroi ChIP-seq of H3K9ac Identification of epigenetic marks linked to SM (Niehaus et al. gene-cluster expression 2013) Fusarium fujikuroi ChIP-seq of H3K9ac in Identification of genome-wide changes in (Studt et al. 2013) ffdah1 deletion strains histone acetylation-patterns Fusarium ChIP-seq of TF Tri6 Genome-wide binding profile of Tri6, involved (Nasmith et al. graminearum in trichothecene gene-cluster expression 2011) Neurospora crassa ChIP-seq of circadian TF Mapping of binding sites of the core circadian (Hurley et al. 2014) WCC; RNA-seq TF WCC Neurospora crassa ChIP-seq of histone Mapping of binding sites of yH2A, important (Sasaki et al. 2014) variant yH2A, for stabilization of stalled replication forks and H3K9me3, H3K4me2 promotion of DNA double-strand-break repair Neurospora crassa ChIP-seq of centromere Identification of centromeric DNA (Smith et al. 2011) proteins CenH3/CEN-C, kinetochore protein CEN-T, and H3K4me2, H3K4me3, H3K9me3 Saccharomyces ChIP-seq of histone H3 Genome-wide mapping of nucleosome (Wal and Pugh cerevisiae positions 2012) Saccharomyces ChIP-seq of Sir proteins, Analysis of the molecular topography of (Thurtle and Rine cerevisiae histone H3, H4K16ac silenced chromatin 2014) Saccharomyces ChIP-seq of linker Hho1 is required for efficient sporulation and (Bryant et al. 2012) cerevisiae histone Hho1 full compaction of the spore genome

Saccharomyces ChIP-seq of H2A.Z Generation of a H2A.Z nucleosome map (Albert et al. 2007) cerevisiae Saccharomyces ChIP-seq of Cse4, Ste12 Concurrent analysis of TFBSs (Lefrançois et al. cerevisiae and Pol II 2009) Schizosaccharomyces ChIP-seq of TF Pho7 Characterization of the phosphate starvation (Carter-O'Connell pombe response and hydrogen peroxide-mediated et al. 2012) oxidative stress response Trichoderma reesei ChIP-seq of H3K4me3, Identification of epigenetic marks linked to SM (Seiboth et al. H3K9me3, and gene-cluster expression 2012, Karimi- H3K4me2 Aghcheh et al. 2013) Zymoseptoria tritici ChIP-seq of H3K9me3 Mapping of euchromatic and heterochromatic (Soyer et al. 2015) and H3K4me2 genomic regions I. INTRODUCTION 17

While ChIP-seq has been widely used in mice and human tissues, the number of experiments performed in fungi has remained rather low until today. Nevertheless, further application of ChIP-seq in a broader variety of fungi harbors a great potential, especially in terms of functional analyses in biotechnologically relevant species.

II. SCOPE OF THE THESIS 18

II. SCOPE OF THE THESIS

1. Regulation of fungal secondary metabolism

Filamentous fungi are renowned for their ability to produce SMs, low-molecular-weight molecules, which are not essential for normal growth or survival of the producing organism (Keller et al. 2005). Generally, at least four major classes of fungal SMs, namely non-ribosomal peptides, polyketides, alkaloids, and terpenes, can be distinguished. As shown in Table 4, some of the most prominent fungal SMs include important pharmaceuticals, such as penicillins, cyclosporines and statins, as well as high-potency fungal toxins, e.g. aflatoxins and trichothecenes.

Fungal secondary metabolism is influenced by a number of genetic and environmental factors, which are ranging from light to the availability and type of carbon/nitrogen sources, as well as the pH of the surrounding medium and temperature (Calvo et al. 2002). Furthermore, a tight association between production of SMs and developmental processes has been described (Reiss 1982, Hicks et al. 1997, Guzmán-de-Peña et al. 1998).

Genes for fungal SMs are organized in clusters (Keller et al. 2005), which are mostly located in sub-telomeric regions (Palmer and Keller 2010). Transcriptional regulation of these SM

Table 4: Classes and examples of fungal SMs

Compound Application Species Reference Non-ribosomal peptides Cephalosporin Antibiotic Acremonium chrysogenum (Elander 2003) Cyclosporin Immunosuppressive drug Tolypocladium inflatum (Survase et al. 2011) Gliotoxin Mycotoxin, Aspergillus fumigatus (Sutton et al. 1994, Scharf Immunosuppressive drug et al. 2012) Penicillin Antibiotic Penicillium chrysogenum (Brakhage et al. 2004) Polyketides Aflatoxin B1 Mycotoxin Aspergillus flavus (Yu et al. 2004) Compactin Cholesterol-lowering drug Penicillium citrinum (Chakravarti and Sahai 2004) Fumonisin B Mycotoxin Fusarium verticillioides (Nelson et al. 1993) Lovastatin Cholesterol-lowering drug Aspergillus terreus (Manzoni and Rollini 2002) Mycophenolic acid Immunosuppressive drug Penicillium brevicompactum (Regueira et al. 2011) Alkaloids Ergotamin Acute treatment of migraine Claviceps purpurea (Tudzynski et al. 1999b, Tfelt-Hansen and Koehler 2008) Fumigaclavine C Anti-atherosclerotic agent Aspergillus fumigatus (Du et al. 2011) Terpenes Aflatrem Tremorgenic mycotoxin Aspergillus flavus (TePaske et al. 1992) Deoxynivalenol (DON) Mycotoxin Fusarium graminearum (Audenaert et al. 2014) Gibberellin GA3 Plant hormone Gibberella fujikuroi (Tudzynski 1999) Trichothecene T2 toxin Mycotoxin Fusarium sporotrichoides (Desjardins et al. 1993) II. SCOPE OF THE THESIS 19 gene clusters is mediated by a variety of transcriptional regulatory elements, ranging from pathway specific TFs to broad domain TFs and multiple-subunit protein complexes (Yin and

Keller 2011). Pathway specific TFs include AflR, a sequence-specific Zn(II)2Cys6 protein necessary for regulation of sterigmatocystin/aflatoxin biosynthesis in Aspergillus species (Woloshuk et al. 1994, Yu et al. 1996, Fernandes et al. 1998), as well as AcFKH1, a member of the forkhead family of TFs, and CPCR1, a eukaryotic regulatory X (RFX) family TF, both involved in regulation of cephalosporin C production in A. chrysogenum (Schmitt et al. 2004a, Schmitt et al. 2004b). Furthermore, PcRFX1 was shown to be a direct regulator of the expression of the penicillin-biosynthesis genes pcbAB, pcbC and penDE in P. chrysogenum (Domínguez-Santos et al. 2012). Besides pathway-specific transcriptional regulators, a number of broad domain TFs establishing a link between fungal SM production and environmental signals have been described. Important representatives of this category are CreA (Cre1), involved in carbon signaling (Dowzer and Kelly 1989), AreA, involved in nitrogen signaling (Hynes 1975), and PacC, involved in pH sensing (Tilburn et al. 1995). In A. chrysogenum, evidence was provided for Cre1 acting as a carbon catabolite-dependent repressor of the cephalosporin-gene cluster (Jekosch and Kück 2000), and a homolog of AreA was shown to directly influence the expression of the gibberellin-gene cluster in Gibberella fujikuroi (Tudzynski et al. 1999a, Mihlan et al. 2003). Furthermore, PacC was demonstrated to positively regulate penicillin production and negatively regulate production of sterigmatocystin in P. chrysogenum and Aspergillus species, respectively (Espeso et al. 1993, Suarez and Peñalva 1996, Keller et al. 1997). A totally new feature of regulation of SM biosynthesis in filamentous fungi was uncovered by recent work in P. chrysogenum, providing evidence for the mating-type (MAT) TF MAT1-1-1 to be involved in regulation of penicillin biosynthesis. It was shown that expression of the penicillin-biosynthesis genes is significantly down-regulated in a ΔMAT1-1-1 strain compared to wild type (Böhm et al. 2013). This finding was of exceptional importance, because for long MAT1-1-1 has been regarded as a regulatory protein restricted to the orchestration of sexual reproduction alone (Martin et al. 2010). The last group of transcriptional regulatory elements involved in regulation of fungal secondary metabolism involves multi-subunit protein complexes, such as the CCAAT-binding complex AnCF/PNR1 (Then Bergh et al. 1996, Brakhage et al. 1999) and the velvet complex (Bayram et al. 2008, Hoff et al. 2010, Wiemann et al. 2010, Kopke et al. 2013). Both are involved in regulation of SM biosynthesis in a number of fungal species. The founding member of the velvet complex, VeA (Velvet A), was shown to interact with the putative methyltransferase LaeA to positively regulate SM production and, together with II. SCOPE OF THE THESIS 20 another velvet protein, VelB, to induce sexual development in Aspergillus nidulans (Bayram and Braus 2012). Similar effects were also observed in P. chrysogenum, where PcVelA, a homolog of VeA, plays an important regulatory role during penicillin biosynthesis and conidiation (Hoff et al. 2010).

Although a broad variety of pathway-specific TFs, global transcriptional regulators, and multi-subunit complexes, which are involved in the regulation of fungal secondary metabolism, have been identified until today, the complexity of these regulatory networks, including multiple target sites and interconnections to other regulatory circuits, is far away from being understood (Calvo et al. 2002). This is mainly due to the fact that almost all works on the isolation and characterization of fungal TFs focused on the detailed analysis of particular genes or families until today. Although these studies played a large part in explaining the fundamental principles of gene regulation, they only presaged the whole dynamics of large GRNs. Today, NGS-based approaches provide excellent tools to analyze genome-wide TF-binding patterns and gene regulation at unprecedented depth. Using ChIP-seq combined with microarray/RNA-seq analysis, TFBSs and genome-wide expression profiles can be determined for various cell types, different developmental stages, or in different environmental conditions. This provides the information needed to uncover the regulatory dynamics associated with changes in cell physiology and development, and establishes a framework for a comprehensive understanding of multi-layer GRNs (Stormo and Zhao 2010).

2. Aim of this thesis

The aim of this thesis was the genome-wide analysis of GRNs controlling morphogenesis and secondary metabolism in the filamentous fungus P. chrysogenum by using ChIP-seq.

P. chrysogenum is the main industrial producer of the pharmaceutically relevant β-lactam antibiotic penicillin (Fleming 1929), the most commonly used drug in the treatment of bacterial infections. With yearly sales of about US$ 8 billion, penicillin is one of the most valued products in the global anti-infective market (Barber et al. 2004). Progressive optimization of P. chysogenum strains used for industrial penicillin production was started in 1943, after isolation of P. chrysogenum strain NRRL 1951 from a moldy cantaloupe in Peoria, IL, USA (Raper et al. 1944, Raper 1946). Using random-mutagenesis approaches, based on X-ray, ultraviolet irradiation, and nitrogen-mustard mutagenesis, penicillin-biosynthesis performance was increased from 60 µg/ml penicillin to more than 50 mg/ml in modern II. SCOPE OF THE THESIS 21 overproducer strains (Backus and Stauffer 1955, Peñalva et al. 1998, Barreiro et al. 2012). However, release of the P. chrysogenum genome sequence in 2008 (van den Berg et al. 2008) paved the way for the replacement of random mutagenesis by targeted genetic engineering. Today, detailed knowledge of cellular and developmental processes affecting penicillin biosynthesis and other traits of biotechnological relevance, such as growth rates, hyphal morphology and pellet formation, sporulation, and stress tolerance, is crucial for further optimization of this industrially highly relevant organism. As a consequence, the functional investigation of regulators of secondary metabolism and morphogenesis in P. chrysogenum has been in the limelight of numerous research projects during the past years (Kosalková et al. 2009, Hoff et al. 2010, Kamerewerd et al. 2011, Veiga et al. 2012, Böhm et al. 2013, Kopke et al. 2013, Böhm et al. 2015, Wolfers et al. 2015). However, as these studies were mainly built on phenotypic characterization of deletion and overexpression strains, protein-protein interaction studies, and microarray analyses, little is known about genome-wide GRNs and direct target genes of regulators of secondary metabolism and morphogenesis in P. chrysogenum until today.

This work describes the application of DNA-binding protein ChIP-seq analysis for genome-wide DNA-binding studies of two regulators of penicillin biosynthesis, development, and morphogenesis in P. chrysogenum, namely MAT1-1-1 and PcVelA. The α-box MAT TF MAT1-1-1 is one of the main regulators of the sexual life cycle in P. chrysogenum (Hoff et al. 2008, Böhm et al. 2013) and PcVelA acts as one of the core components of the multi-subunit velvet complex (Hoff et al. 2010, Kopke et al. 2013). Previous studies pointed to MAT1-1-1 and PcVelA regulatory functions that are not restricted to one cellular or developmental process alone. For example, MAT1-1-1 was shown to affect various traits of biotechnological relevance, including penicillin biosynthesis, hyphal morphology, and formation of asexual conidiospores (Böhm et al. 2013), extending its regulatory function far beyond its recognized role in orchestration of the sexual life cycle. Correspondingly, besides its well-known regulatory functions in terms of secondary metabolism and formation of asexual conidiospores, PcVelA was shown to influence pellet formation and hyphal morphology in P. chrysogenum (Hoff et al. 2010, Kopke et al. 2013). Additionally, recent work in related species described the ability of homologs of PcVelA to bind DNA in a sequence-dependent manner and to specifically interact with putative methyltransferases outside the velvet complex (Jiang et al. 2011, Palmer et al. 2013, Sarikaya-Bayram et al. 2014, Sarikaya-Bayram et al. 2015). These observations suggest that MAT1-1-1 and PcVelA might perform as global II. SCOPE OF THE THESIS 22 transcriptional regulators, marking them as extremely interesting candidates for further characterization on a genome-wide scale.

Within the scope of this thesis, ChIP-seq will be adapted for the application in P. chrysogenum and an experimental pipeline will be established for sample preparation, bioinformatics analysis of sequencing data, and further downstream analysis. A comprehensive ChIP-seq approach will be used in order to identify as many MAT1-1-1 and PcVelA DNA-binding sites as possible, independent of physiological culture conditions, developmental stages, or external stimuli. Downstream analyses will include the validation of data obtained from ChIP-seq analyses by ChIP-PCR as well as the identification of direct target genes of both, MAT1-1-1 and PcVelA, by integration of previous microarray data and qRT-PCR analyses. Based on peak regions identified in ChIP-seq analyses, de novo prediction of DNA-binding motifs specific for MAT1-1-1 as well as PcVelA will be performed. Subsequently, predicted DNA-binding consensus sequences will be tested for functionality and specificity in vitro, ex vitro, and in vivo by applying DNA-binding studies (electrophoretic mobility shift essays; EMSAs), yeast one-hybrid (Y1H), and DsRed reporter gene assays. Furthermore, functional characterization of new MAT1-1-1 and PcVelA downstream factors will complete this work.

III. BECKER et al. 2015a 23

III. BECKER et al. 2015a

Genome-wide identification of target genes of a mating-type α-domain transcription factor reveals functions beyond sexual development

Kordula Becker, Christina Beer, Michael Freitag, and Ulrich Kück (2015)

Molecular Microbiology doi:10.1111/mmi.12987

Molecular Microbiology (2015) ■ doi:10.1111/mmi.12987

Genome-wide identiﬁcation of target genes of a mating-type α-domain transcription factor reveals functions beyond sexual development

Kordula Becker,1 Christina Beer,1 Michael Freitag2 Introduction and Ulrich Kück1* 1Christian Doppler Laboratory for Fungal Biotechnology, Sexual propagation in euascomycetes is controlled by two Lehrstuhl für Allgemeine und Molekulare Botanik, alternative mating-type loci, namely MAT1-1 and MAT1-2, Ruhr-Universität Bochum, Universitätsstr. 150, D-44780 which consist of dissimilar sequences occupying the Bochum, Germany. same locus on the chromosome. These sequences are 2Department of Biochemistry and Biophysics, Oregon termed idiomorphs to indicate that they do not represent State University, Corvallis, Oregon 97331-7305, USA. the alleles of a single gene (Metzenberg and Glass, 1990). A common feature specific to mating types from euascomycetes is the presence of a MAT1-1-1 gene, Summary defining the MAT1-1 idiomorph and encoding an α-domain transcription factor (TF). The alternative idi- Penicillium chrysogenum is the main industrial pro- omorph, MAT1-2, is characterized by the presence of a ducer of the β-lactam antibiotic penicillin, the most MAT1-2-1 gene, encoding a TF carrying a high mobility commonly used drug in the treatment of bacterial group (HMG) domain (Turgeon and Yoder, 2000; Lee infections. Recently, a functional MAT1-1 locus et al., 2010). encoding the α-box transcription factor MAT1-1-1 was While DNA-binding HMG-domain proteins are ubiqui- discovered to control sexual development in P. chry- tous and well characterized, α-domain proteins have sogenum. As only little was known from any organ- limited distribution and their evolutionary origin is still ism about the regulatory functions mediated by obscure (Martin et al., 2010). In Saccharomyces cerevi- MAT1-1-1, we applied chromatin immunoprecipitation siae,MATα1, one of two proteins encoded by the α-type combined with next-generation sequencing (ChIP- mating locus, acts as a transcriptional co-activator and is seq) to gain new insights into the factors that influ- involved in the regulation of mating-type-specific gene ence MAT1-1-1 functions on a molecular level and its expression (Herskowitz, 1989). MATα1 binds coopera- role in genome-wide transcriptional regulatory net- tively with the MADS-box TF Mcm1 to 26-bp P′Q promoter works. Most importantly, our data provide evidence elements to activate the expression of α-specific genes for mating-type transcription factor functions that (αsgs) (Bender and Sprague, 1987). Surprisingly, only a reach far beyond their previously understood role in few direct target genes of mating-type TFs are known sexual development. These new roles include regula- until today. For example, chromatin immunoprecipitation tion of hyphal morphology, asexual development, as (ChIP)-chip analysis in S. cerevisiae identified five αsgs well as amino acid, iron, and secondary metabolism. and six a-specific genes (asgs), which, with the exception Furthermore, in vitro DNA–protein binding studies of one αsg, were all involved directly in some aspect of and downstream analysis in yeast and P. chrysoge- mating, e.g. those encoding the mating pheromone num enabled the identification of a MAT1-1-1 DNA- α-factor and the a-pheromone receptor Ste3 (Galgoczy binding motif, which is highly conserved among et al., 2004). Similarly, microarray analysis in Candida euascomycetes. Our studies pave the way to a more albicans identified two αsgs and at least two asgs general understanding of these master switches for (Tsong et al., 2003), and genome-wide ChIP analysis in development and metabolism in all fungi, and open Lachancea kluyveri identified a total of nine asgs, of which up new options for optimization of fungal high pro- six were orthologs of asgs in either C. albicans or S. cer- duction strains. evisiae (Baker et al., 2012). Against this background, it appears somehow contradictory that several micro- Accepted 26 February, 2015. *For correspondence. E-mail [email protected]; Tel.: (+49) 23 4322 6212; Fax (+49) 23 4321 array analyses demonstrated that MAT genes have a 4184. rather wide-ranging effect on fungal gene expression

© 2015 The Authors. Molecular Microbiology published by John Wiley & Sons Ltd This is an open access article under the terms of the Creative Commons Attribution-NonCommercial License, which permits use, distribution and reproduction in any medium, provided the original work is properly cited and is not used for commercial purposes. 2 K. Becker, C. Beer, M. Freitag and U. Kück ■

(Pöggeler et al., 2006; Bidard et al., 2011; Wada et al., encoded TFs and should thus open new avenues for the 2012; Böhm et al., 2013). Hence, further research is study of fungal sexual development. Finally, as we per- needed in order to distinguish between primary and sec- formed ChIP-seq experiments with a laboratory strain that ondary target genes of mating-type encoded TFs, and to has already undergone several rounds of mutagenesis to provide a comprehensive understanding of mating-type increase penicillin production (Nielsen, 1997), our results controlled regulatory circuits on a genome-wide level. are applicable to fungal strains used for today’s industrial As most research performed on characterizing mating- production of pharmaceutically relevant secondary type locus-encoded TFs used yeasts as a model organ- metabolites. ism, little is known about mating-type protein function in euascomycetes. Lack of research on many euascomycetes, especially those of major medical and industrial Results importance, was fostered by the fact that these fungi have Construction of MAT1-1-1 strains for ChIP-seq analysis been considered to be asexual since no sexual propagation had been observed under laboratory conditions for a APgpd::egfp::MAT1-1-1 fusion construct (pGFP-MAT1, very long time (Dyer and O’Gorman, 2012; Kück and Fig. S1A) was transformed into recipient P2niaD18 to Böhm, 2013). Recent description of a hetherothallic generate strain MAT1-ChIP. Pgpd was used to obtain an sexual cycle in P. chrysogenum now makes the fungus a elevated expression level of the MAT1-1-1 gene, since valuable object for the investigation of mating-type con- expression of mating-type genes under control of their trolled transcriptional regulatory networks and fungal native promoter is known to be low. For example, RMA- sexual reproduction in general. These mechanisms are of express (http://rmaexpress.bmbolstad.com) analysis of major importance, as the possibility to generate offspring normalized raw data obtained from microarray analysis with novel combinations of traits relevant to penicillin pro- using P. chrysogenum strain P2niaD18 revealed relative duction provides promising starting points for industrial MAT1-1-1 expression levels of about 13.6% and 3.1% strain development purposes (Böhm et al., 2013). referred to actin (Pc20g11630) and myosin (Pc21g00710) Chromatin immunoprecipitation combined with next- expression levels, respectively (Fig. S1B) (Böhm et al., generation sequencing analysis (ChIP-seq) is one of the 2013). Furthermore, transcripts of mating-type genes most powerful tools for genome-wide profiling of DNA- were reported to be barely detectable by Northern hybridi- binding proteins, which has greatly benefited from tre- zation in Podospora anserina as well as RNA-seq analy- mendous progress in next-generation sequencing sis in Neurospora crassa (Coppin and Debuchy, 2000; technology (Smith et al., 2010; Magnúsdóttir et al., 2013; Wang et al., 2014). Myers et al., 2013). Today, ChIP-seq is an indispensable Successful transformation was verified by polymerase tool for studying gene regulation and epigenetic mecha- chain reaction (PCR) and sodium dodecyl sulfate– nisms at the genomic level (Park, 2009). Here, we present polyacrylamide gel electrophoresis (SDS–PAGE)/ the first application of ChIP-seq for the functional charac- Western blot analysis, confirming the presence of the terization of a TF from P. chrysogenum, and, more impor- epitope-tagged protein EGFP-MAT1-1-1 in crude protein tantly, the first genome-wide analysis focusing on extract from recombinant strains (Fig. S1C and D). Using unraveling the transcriptional regulatory network con- fluorescence microscopy, the presence and nuclear locali- trolled by a mating-type locus-encoded TF. zation of the fusion protein were verified prior to each While MAT1-1-1 has been described as a regulatory ChIP experiment (Fig. S1E). Functionality of the fusion protein restricted to the orchestration of sexual reproduc- protein was further confirmed when pellet formation was tion (Debuchy et al., 2010), our data clearly expand this investigated in shaking cultures (Fig. S1F). Overexpres- current view of MAT1-1-1 function beyond transcriptional sion of MAT1-1-1 in the MAT1-ChIP strain resulted in the regulation of sexual development alone. We provide formation of significantly larger pellets (Ø 4–5 mm) when strong evidence of new and additional roles for MAT1-1-1 compared with P2niaD18 (Ø 1–2 mm), matching the phe- in regulating asexual development and morphogenesis, notypic characteristics of a previously described MAT1- as well as amino acid, iron, and secondary metabolism. 1-1 overexpression strain (OE MAT1-1-1) (Böhm et al., Furthermore, our analyses, using bioinformatics, electro- 2013). phoretic mobility shift assays (EMSAs), yeast one-hybrid (Y1H), and DsRed reporter gene assays in P. chrysoge- ChIP-seq analysis reveals a genome-wide binding num, led to the identification of a MAT1-1-1 DNA-binding profile of MAT1-1-1 motif that shows a high degree of conservation within euascomycetes. We performed ChIP-seq experiments on three independ- Taken together, our data extend the general under- ent biological samples, namely ‘shaking 1’, ‘shaking 2’, standing of the biological functions of mating-type- and ‘surface’ (Table 1). In an effort to identify as many

Table 1. ChIP-seq design and results.

# Peaks # Differential # Total Estimated Sample # Readsa # Mappedb % Mappedc FDR ≤ 0.001d peakse peaksf fragment lengthg shaking 1 44,608,426 27,190,663 60.9 % 7453 430 327 237 shaking 2 39,771,172 23,994,317 60.3 % 6523 379 276 226 surface 14,364,485 12,890,352 89.7 % 6324 218 102 212 shaking_input 16,952,199 15,380,186 90.7 % – – – – surface_input 12,879,889 11,422,613 88.7 % – – – – a. Total number of sequenced reads. b. Total number of reads mapped to P. chrysogenum P2niaD18 genome (Specht et al., 2014). c. Fraction of tags found in peaks versus genomic background determined by HOMER (Heinz et al., 2010). d. Number of peaks passing FDR ≤ 0.001 threshold. e. Number of peak regions showing at least fourfold enrichment in ChIP-sample compared to input. f. Total number of peak regions after local background ﬁltering and clonal ﬁltering. g. Estimated fragment length used for sequencing, determined from tag auto correlation analysis.

MAT1-1-1 binding sites as possible, independent of physi- Starting from ChIP-seq datasets, we classified peaks ological culture conditions, two samples were derived according to their genomic location with regard to neigh- from shaking cultures and one was obtained from a boring coding sequences. Seventy-nine percent (193/ surface-grown culture. Input-DNA from shaking (‘shaking- 243) of peaks were exclusively located within intergenic _input’) and surface cultures (‘surface_input’) was regions and 21% (50/243) showed intragenic localization sequenced as a control. Only regions meeting the follow- (Fig. 1B). Of 193 peaks showing intergenic localization, ing criteria were considered as specific peak regions: (1) 21 were positioned within the 3′ region of both neighbor- at least fourfold enrichment in ChIP-DNA versus input- ing open reading frames, and 90 showed 5′ localization to DNA, (2) a false discovery rate (FDR) threshold ≤ 0.001, only one adjacent gene. Eighty-two peak regions were, and (3) a Poisson p-value ≤ 1.00e–04. Intersection of our however, positioned within divergent promoters, resulting datasets identified 243 sites that were specifically bound in a total of 254 genes that may be directly controlled by by MAT1-1-1 in at least two independent biological repli- MAT1-1-1. Comparison to expression data obtained from cates, thus meeting the standards set by the ENCODE previous microarray analyses (Böhm et al., 2013) con- and modENCODE consortia (Landt et al., 2012) (Dataset firmed changes in expression profiles by at least twofold S1, Fig. 1A). in a ΔMAT1-1-1 strain compared with P2niaD18 for 29.9%

Fig. 1. Genome-wide distribution of MAT1-1-1 binding regions. A. Venn-diagram showing intersection between MAT1-1-1 ‘shaking 1’, ‘shaking 2’, and ‘surface’ datasets. Only peaks within a maximum distance of 100 nt were regarded as overlapping. B. Distribution of ChIP-enriched regions overlapping or positioned within intragenic regions vs. ChIP-enriched regions that were exclusively located within intergenic regions (based on peak regions present in at least two independent datasets). C. Distance between MAT1-1-1 ChIP-seq peak summits and ATG of neighboring genes positioned in 5′–3′ orientation with regard to the corresponding peak region (based on peak regions present in at least two independent datasets).

(76/254) of these genes. Analysis of the distance between relative to this ratio in the input-DNA sample. An additional peak summits and the predicted translation start sites control region (NC) is shown as a negative control. ChIP- (nearest ATG in good initiation context) revealed an PCR results showed significant overlap with the corre- average distance of 200–500 nt (Fig. 1C). Approximately sponding peak values obtained from bioinformatics 50% of all analyzed genes fit this pattern. analysis, confirming specific enrichment of all tested target regions in ChIP-DNA vs. input-DNA, and validating peak values as a convincing parameter for estimation of Categorization of putative MAT1-1-1 target genes MAT1-1-1 binding affinity to target regions identified in Gene ontology (GO) analysis of proteins encoded by the ChIP-seq analyses. 254 putative MAT1-1-1 target genes revealed a significant Next, quantitative real time (qRT)-PCR analyses were (p ≤ 0.05) overrepresentation of the following categories: performed to validate MAT1-1-1 target genes next to the (1) metabolism, including proteins related to amino acid peak regions mentioned earlier as well as a selection of and secondary metabolism, (2) energy, (20) cellular trans- additional, non-mating-related target genes, covering all port, transport facilities, and transport routes, (32) cell functional protein categories mentioned in Table 2. Com- rescue, defense, and virulence, (34) interaction with pared with wild type P2niaD18, expression levels of puta- the environment, including proteins involved in cellular tive target genes were examined in shaking cultures of a sensing and response to external stimuli (e.g. pheromone MAT1-1-1 deletion strain (ΔMAT1) or MAT1-1-1 overex- response) (Fig. S2). Besides expected putative MAT1-1-1 pression strain (MAT1-ChIP), grown under the same con- target genes that could be directly assigned to sexual ditions as for ChIP-seq sample preparation. A total of four development, e.g. ppg1 (Pc14g01160), the homolog of mating- and 13 non-mating-related genes were selected S. cerevisiae MFα1/2, encoding the α-factor pheromone, for our investigation. Compared with P2niaD18, overex- and pre1 (Pc22g15650), the homolog of the S. cerevisiae pression of MAT1-1-1 in the MAT1-ChIP strain led to a-factor receptor encoding gene STE3 (Galgoczy et al., significant changes in expression levels of 3 genes related 2004), ChIP-seq analysis identified many new putative to some aspect of sexual reproduction, namely pre1, kex1, MAT1-1-1 target genes that had never been linked to and ppg1, as well as seven non-mating-related genes, mating-type-encoded TFs before. Table 2 provides a namely Pc20g00090, dewA, atf21, Pc19g00140, sidD, detailed summary of selected MAT1-1-1 target genes Pc22g27040, and Pc22g22160 (Fig. 2B and Fig. S3A). arranged according to the description and proposed func- Similar results were obtained when ΔMAT1 expression tion of encoded proteins, as obtained from blastp analysis levels were measured. It is remarkable that all these genes and literature. All genes listed are positioned in 5′–3′ are located in 5′–3′ orientation relative to the adjacent orientation with regard to neighboring MAT1-1-1 peak peaks, thus, validating our criteria applied for identification regions. Corresponding peak values, expression profiles of putative MAT1-1-1 target genes based on data obtained of each gene in a MAT1-1-1 deletion strain compared with from ChIP-seq analysis. wild type P2niaD18 and occurrence of the MAT1.1 motif (to be described later) are given. For reasons of clarity De novo prediction of a MAT1-1-1 DNA-binding motif and comprehensibility, the categories mentioned here do not necessarily correspond directly to categories used in To gain further insight into MAT1-1-1 DNA-binding prop- GO analysis. erties, de novo motif prediction based on MAT1-1-1- binding regions, identified in our ChIP-seq analysis, was performed. We used MEME to identify conserved motifs, Validation of MAT1-1-1 targets and therefore the most likely binding site of MAT1-1-1 in To validate MAT1-1-1 DNA-binding regions identified by P. chrysogenum. MEME analysis, based on 62 MAT1-1-1 our ChIP-seq approach, we performed ChIP-PCR analy- binding regions, present in three independent ChIP-seq sis (Fig. 2A). Five representative MAT1-1-1 target regions experiments, identified one highly significant motif, desig- were analyzed for MAT1-1-1-specific enrichment in ChIP- nated MAT1.1, which showed a high degree of central DNA compared to input-DNA, obtained from shaking enrichment across MAT1-1-1 peak regions in CentriMo cultures. Target regions were selected according to the analysis (Fig. 3). Furthermore, FIMO analysis confirmed following two key criteria: (1) they either possessed a the presence of at least one copy of MAT1.1 within 202 of statistically highly significant peak value (Pc20g00090)or 243 (83.1%; p-value ≤ 0.01) MAT1-1-1 peak regions, indi- (2) proteins encoded by adjacent genes were known to be cating that the vast majority but not all MAT1-1-1 target involved in regulation of sexual reproduction in yeast sites are bound at this motif. (pre1, kex1, kex2, ppg1). Enrichment was calculated as Comparison of MAT1.1 to known binding motifs present the ratio of the region of interest to a control region in the JASPAR CORE (2014) databases for fungi and showing no MAT1-1-1-specific enrichment in ChIP-DNA vertebrates revealed strong similarity to the binding sites

Microarrayc MAT1.1d

Peak p ≤ p ≤ Identiﬁer Descriptiona Proposed function valueb 36 h 96 h 0.001 0.01

Sexual development Pc22g18600e serine carboxypeptidase Kex1 Homolog of S. cerevisiae α-pheromone processing endoproteases KEX1 (Dmochowska et al., 1987) 4540 −0.14 −1.07 3 6 Pc22g15650e a-pheromone receptor Pre1 Homolog of the S. cerevisiae a-factor receptor STE3 (Galgoczy et al., 2004) 3869 −0.11 −0.52 2 6 Pc22g02910e pheromone-processing endoprotease Kex2 Homolog of S. cerevisiae α-pheromone processing endoprotease KEX2 (Julius et al., 1984) 1396 0.42 0.02 2 3 oeua Microbiology Molecular Pc20g00540 cyclin-dependent protein kinase Bur1 Protein required for a G-α subunit-mediated adaptive pheromone-response in S. cerevisiae (Irie et al., 1991) 861 0.23 −0.32 2 6 Pc14g01160e mating α-pheromone Ppg1 Homolog of S. cerevisiae α-factor pheromone MFα1 (Galgoczy et al., 2004) 663 0.30 −0.46 2 5 Pc12g15890 cAMP-independent regulatory protein Pac2 cAMP-independent regulatory protein modulating onset of sexual development in Schizosaccharomyces pombe and 444 1.01 −0.04 0 0 Magnaporthe oryzae (Kunitomo et al., 1995; Chen et al., 2014) Morphogenesis and asexual development Pc21g04930 trehalose-6-phosphate synthase subunit 3 Involved in biosynthesis of trehalose, a compound necessary for long-term viability of fungal spores (Elbein et al., 2003) 747 −0.36 −0.18 0 1 Pc18g03940e 14-3-3 family protein ArtA 14-3-3 family protein involved in regulation of polarization of germinating conidiospores in Aspergillus nidulans (Kraus et al., 724 0.13 0.07 0 1 2002) Pc16g06690e spore wall fungal hydrophobin DewA Spore wall fungal hydrophobin responsible for hydrophobicity of conidiospores in A. nidulans (Stringer and Timberlake, 1995; 640 −3.87 0.05 0 4 Grünbacher et al., 2014) Pc06g01300 thioredoxin TrxA Involved in regulation of growth and formation of reproductive structures, e.g. conidiophores and cleistothecia, in A. nidulans 584 −3.91 0.41 1 4

published (Thön et al., 2007) Pc12g15180 chitin biosynthesis protein Involved in biosynthesis of chitin, an essential component of the cell walls and septa, necessary for polarized growth, septa 556 −1.37 0.11 0 1 formation during hyphal growth, and conidia development (Fukuda et al., 2009) Pc21g20900 cell morphogenesis protein PAG1 Cell morphogenesis protein, related to polarized morphogenesis and proliferation in S. cerevisiae (Du and Novick, 2002; Nelson 387 0.66 0.08 0 0 et al., 2003) e by Pc22g26820 bZIP TF Atf21 bZIP TF and repressor of sexual development in A. nidulans (Lara-Rojas et al., 2011); activating 324 2.00 0.74 1 2 TF/cAMP-response-element-binding protein, central role in maintaining cellular homeostasis and production of spores in onWly&Sn Ltd, Sons & Wiley John S. pombe (Morita et al., 2011) Pc21g09870 related to integral membrane protein Pth11 Functions at the cell cortex as an upstream effector of appressorium differentiation in response to surface cues in M. grisea 155 −1.29 0.56 0 4 (DeZwaan et al., 1999) Pc19g00140e trehalose-6-phosphate synthase subunit Involved in biosynthesis of trehalose, a compound necessary for long-term viability of fungal spores (Elbein et al., 2003) 106 1.70 0.17 0 3 Amino acid and secondary metabolism Pc18g02620 cyanide hydratase/nitrilase Likely to be involved in the cyanoamino acid metabolism 1712 −4.35 −0.40 0 2 Pc12g00820 MFS multidrug transporter Tpo1 Controls spermidine and spermine concentrations and mediates induction of antioxidant proteins, including Hsp70, Hsp90, 787 −0.85 0.37 2 6 Hsp104 and Sod1 in S. cerevisiae (Krüger et al., 2013) Pc16g11470 ABC multidrug transporter AtrF Overexpression correlates with itraconazole resistance in A. fumigatus (Slaven et al., 2002) 690 1.25 1.06 0 0 Pc22g18630e homocysteine S-methyltransferase Catalyzes the chemical reaction of L-homocysteine to L-methionine [KEGG database] 645 0.59 −0.08 1 4 Pc12g02630 carbon catabolite repression protein CreD 623 −1.39 0.13 0 1 Pc16g06630 MFS multidrug transporter 597 −2.44 0.97 0 4 oeua Microbiology Molecular Pc20g03900 MFS multidrug transporter 525 −0.88 0.22 0 4 Pc22g06500 amino acid transporter 367 0.54 1.04 0 2 Iron metabolism Pc22g20410 siderophore biosynthesis lipase 1138 0.50 1.38 0 4 Pc22g20400e non-ribosomal peptide synthetase SidD Non-ribosomal siderophore peptide synthetase important for biosynthesis of intracellular siderophore triacetylfusarinine C (TAFC) 1138 0.46 0.31 0 4 (Schrettl et al., 2007) Pc21g08020e iron transporter multicopper oxidase FetC Ferrooxidoreductase involved in reductive iron assimilation in A. fumigatus (Schrettl et al., 2004) 882 −1.04 0.95 0 1 Pc21g08030e high-affinity iron ion transporter FtrA High-affinity iron permease that mediates uptake of Fe2+ during reductive iron acquisition in A. fumigatus (Schrettl et al., 2004; 882 −1.18 1.57 0 1 factor transcription mating-type a of genes Target Schrettl and Haas, 2011) Pc21g13060 ferric reductase transmembrane component 834 −0.08 −0.15 1 2 Pc13g11520 siderophore biosynthesis family protein 624 −0.19 1.81 0 1 Pc22g02380 MFS siderophore iron transporter 461 0.40 −0.36 0 0 Transcription factors Pc18g00880e bZIP TF MeaB bZIP TF involved in regulation of expression of nitrogen-dependent genes (Wong et al., 2007; Schönig et al., 2008) 1064 0.84 −0.03 2 3 Pc22g22160e F-box domain protein 859 −2.35 0.16 2 4 Pc12g03120e transcription factor Sin3 Master transcriptional scaffold protein and co-repressor that regulates cellular proliferation, differentiation, apoptosis, and cell 616 0.33 −0.26 1 5 cycle regulation in yeast as well as higher eukaryotes (Grzenda et al., 2009) Pc20g05880 HLH TF 594 −1.95 −0.11 0 0 Pc18g01520 transcription initiation protein IIB 511 0.20 −0.11 1 3 Pc24g00540 C6 zinc ﬁnger domain protein 413 −2.89 1.16 1 2 Pc21g01450 TFIIIC transcription initiation factor complex subunit 393 0.47 −0.22 2 7 e Pc22g27040 C2H2 zinc ﬁnger domain protein 369 1.25 3.02 1 1

a. As obtained from blastp analysis (http://www.ncbi.nlm.nih.gov). b. Statistical peak value = average tag count found at peak normalized to 10 Mio. total mapped tags. c. Microarray data showing expressional changes in ΔMAT1 compared with wild type after 36 and 96 h of cultivation (Böhm et al., 2013). 5 d. Number of MAT1.1 occurrences within peak region, p-value ≤ 0.001/0.01. e. Veriﬁed using EMSA. 6 K. Becker, C. Beer, M. Freitag and U. Kück ■

Fig. 2. Verification of MAT1-1-1 ChIP-seq data. A. ChIP-PCR analysis was performed to verify enrichment of selected MAT1-1-1 binding regions in ChIP-DNA compared with input-DNA. Enrichment was calculated as the ratio of the region of interest to a control region showing no MAT1-1-1-specific enrichment in ChIP-DNA, relative to this ratio in the input-DNA sample. A region showing no MAT1-1-1-specific enrichment in ChIP-seq analysis is shown as a control (NC). Each qPCR ratio (gray bars) is shown in comparison to the corresponding peak value generated during bioinformatics analysis of ChIP-seq data (black bars). Values for qPCRs are the mean score of three biological replicates; average ± standard deviations are indicated. Tested peak regions are named according to neighboring genes (see Dataset S1).

B. Analysis of relative log2 fold gene expression ratios in a MAT1-1-1 overexpression strain (MAT1-ChIP; gray bars) or MAT1-1-1 deletion strain (ΔMAT1; black bars) compared with wild type strain P2niaD18 led to the identification of MAT1-1-1 specific target genes. Values are the mean score of three biological replicates. Tested genes represent pairs of genes positioned upstream and downstream of MAT1-1-1 target regions identified in ChIP-seq analysis (see Dataset S1). Directions of open reading frames are indicated by arrows.

of the S. cerevisiae mating-type protein MATa1 (Haber, to motifs known from vertebrates revealed strong similar- 2012) and Mcm1, a TF involved in cell-type-speciﬁc tran- ity to the binding sites of Sox9, a SRY-related HMG-box scription and pheromone response in yeast (Mead et al., protein, regulating the development of the skeleton and 2002) (Fig. S4). Furthermore, MAT1.1 showed similarity to the reproductive system (Mertin et al., 1999), Nkx2-5, a DNA-binding motifs for Yhp1, a homeobox transcriptional homeobox TF involved in the regulation of heart formation repressor known to bind Mcm1 (Pramila et al., 2002), and development (Chen and Schwartz, 1995), as well as and Hcm1, a forkhead TF regulating expression of genes Sox17 and Sox2, SRY-related HMG-box proteins involved involved in chromosome segregation, spindle pole in the regulation of embryonic development and cell fate dynamics and budding (Pramila et al., 2006). Comparison (Kanai et al., 1996; Maruyama et al., 2005).

Fig. 3. De novo prediction of a MAT1-1-1 DNA-binding motif. The central 100 nt region of 62 MAT1-1-1 specific peak regions identified in three independent ChIP-seq experiments was submitted to MEME for identification of enriched motifs. Only the most significant putative DNA-binding motif (‘MAT1.1’) is shown. The size of each letter is proportional to the frequency of each nucleotide at this position within the consensus sequence. CentriMo analysis, using MAT1.1 as an input, revealed central enrichment of the motif within a 500 nt range around MAT1-1-1 binding regions used for motif prediction.

the MAT_alpha1 domain (pfam04769), especially the region spanning the MATA_HMG-box (cd01389), of euascomycetes compared to hemiascomycetes, in particular, C. albicans (Fig. S5).

MAT1-1-1 binds in vitro to MAT1.1

Since our motif analysis suggested that MAT1-1-1 associates with DNA via the predicted DNA-binding consensus sequence CTATTGAG (MAT1.1), EMSAs were performed to test direct binding between DNA and protein. For this purpose, a GST-MAT1-1-1 fusion protein was purified from Escherichia coli BL21 (DE3) and the quality of the Fig. 4. Conservation of the MAT1.1 binding motif within isolated protein was verified by SDS–PAGE/Western blot ascomycetes. The 1000 nt upstream region of ppg1, pre1, kex2, analysis using an antibody to GST (Fig. S6). The promoter and kex1 from selected ascomycetes was screened for occurrences of MAT1.1 using FIMO. The 1000 nt upstream region regions of mating-related genes pre1, kex1, kex2, and of the actin gene act was used as a negative control. Total ppg1, as well as 13 non-mating-related genes (marked in numbers of detected MAT1.1 copies within input sequences Table 2), which were bound by MAT1-1-1 in ChIP-seq meeting a statistical threshold of p ≤ 0.001 and p ≤ 0.0001, respectively, are given. Locus tags according to NCBI database analysis, were used to design oligonucleotide probes cov- (http://www.ncbi.nlm.nih.gov/) are: ppg1: Pc14g01160, ering a region with at least one copy of MAT1.1 (Fig. 5A, AFUA_6G06360, AN5791.2, CaO19.11961*, FG05061.1, Table S3). NCU02500.1, YPL187W, TRIREDRAFT_104292; pre1: Pc22g15650, AFUA_5G07880, AN7743.2, CaO19.2492*, All oligonucleotides harboring a complete, central copy FG07270.1, NCU00138, YKL178C*, TRIREDRAFT_57526; kex2: of MAT1.1 were bound by GST-MAT1-1-1 (e.g. mating- Pc22g02910, AFUA_4G12970, AN3583.2, CaO19.12219, related Pre1-2, Kex1-2, Kex2-2, Ppg1-2, and non-mating- FG09156.1, NCU03219, YNL238W*, TRIREDRAFT_123561*; kex1: Pc22g18600, AFUA_1G08940, AN1384.2, CaO19.7020*, related Pc20g00090-2, Pc22g27040-1, ArtA-1, DewA- FG10145.1, NCU04316, YGL203C, TRIREDRAFT_74517; act: 1/-2, FetC/FtrA-1, and SidD-1), whereas probes lacking Pc20g11600, AFUA_6G04740, AN6542.2, CaO19.5007, MAT1.1 (e.g. Pc20g00090-3, Pre1-3, Kex1-1) showed no FG07335.1, NCU04173, YFL039C*, TRIREDRAFT_77541. Asterisks are sequences shorter than 1000 nt. binding (Fig. 5B and S3B). Only weak binding or no binding between DNA and protein was observed when MAT1.1 was positioned at the very end of the oligonucleo- The MAT1.1 binding motif shows conservation tide or contained obvious deviations from the predicted within euascomycetes consensus sequence (e.g. Kex1-3, Kex2-1, Ppg1-1, To address the question whether the predicted MAT1-1-1 TrxA-1, and Pc16g06630-1). GST alone showed no DNA-binding consensus sequence MAT1.1 is conserved binding to oligonucleotide Ppg1-2, confirming that the among ascomycetes, we performed FIMO analysis. For observed formation of protein–DNA complexes is medi- this purpose, the 1000 nt upstream region of ppg1, pre1, ated by MAT1-1-1, and not by the tag. kex2, and kex1 from P. chrysogenum and the correspond- Specificity of MAT1-1-1 binding to MAT1.1 was further ing homologs from A. fumigatus, A. nidulans, C. albicans, verified using mutated Ppg1-2 oligonucleotides. A single Fusarium graminearum, N. crassa, S. cerevisiae, and Tri- A → GorT→ C substitution at position three of one of choderma reesei were screened for occurrences of two copies of MAT1.1 present in oligonucleotide Ppg1- MAT1.1. The corresponding 1000 nt upstream sequences 2_m1 led to a drastic reduction of protein–DNA complex of the actin gene (act) were used as a negative control formation, whereas mutation of both motifs (Ppg1-2_m2) (Fig. 4). A high degree of conservation of MAT1.1 within totally abolished complex formation (Fig. 6A). Further- the tested promoter regions of euascomycetes became more, competition assays using Ppg1-2 as a probe and obvious, whereas significant deviations were recognized unlabeled Ppg1-2 oligonucleotide as a competitor showed when compared with hemiascomycetes. For example, that the level of MAT1-1-1 binding to the labeled probe is applying a statistical threshold of p ≤ 0.0001, occurrences diminished by addition of increasing amounts of the unla- of MAT1.1 were detected in seven out of eight ppg1 beled competitor. In the corresponding autoradiogram, an (no occurrence in C. albicans) and six out of eight pre1 attenuation of the shift band and accumulation of free (no occurrence in S. cerevisiae and C. albicans) labeled probe became visible (Fig. 6B; left panel). In con- upstream sequences. These observations were further trast, Western blotting of the shift gel and immunodetec- confirmed when sequence alignments of the protein tion using an antibody to GST clearly showed an increase sequences of MAT1-1-1 DNA-binding domains revealed in complex signal strength when competing with unla- a significantly higher degree of conservation within beled Ppg1-2 probe (Fig. 6B; right panel). Both, EMSA

Fig. 5. Electrophoretic mobility shift assays (EMSAs) confirm MAT1-1-1-binding to ChIP-enriched genomic regions. A. Zoomed ChIP-seq profiles of selected MAT1-1-1 ChIP-enriched regions. Positions and sequences of oligonucleotides used for shift analysis are indicated. Occurrences of the predicted MAT1-1-1 DNA-binding motif MAT1.1 are marked in red. Single nucleotides that do not fit the predicted consensus sequence are indicated in small letters. Maximum read counts at the summit of ChIP-seq peaks are indicated at the left. ORFs next to MAT1-1-1 ChIP-seq peak regions are marked by black boxes; arrowheads indicate 5′–3′ orientation. B. EMSAs were performed using radiolabeled double-stranded oligonucleotide probes covering the central region of selected MAT1-1-1 target regions, identified in ChIP-seq analysis. Addition of GST-MAT1-1-1 protein is marked by (+), samples without protein are marked by (−). Positions of free probe (*) and protein–DNA complexes (→) are indicated.

and Shift–Western analyses, confirmed the specificity of performed to validate binding ex vitro. Triple repeats of MAT1-1-1 binding to the Ppg1-2 oligonucleotide since oligonucleotides Kex1-2 and Ppg1-2, as well as Ppg1- addition of unlabeled DNA minimized binding of MAT1-1-1 2_m1 and Ppg1-2_m2 in the promoter of the lacZ or HIS3 to the radiolabeled probe, while overall complex formation reporter gene, were used as preys for MAT1-1-1. As bait, was maximized. As expected, unlabeled Ppg1-2_m2 did we used vector pMAT1-AD, containing the MAT1-1-1 not compete for binding to MAT1-1-1 with labeled Ppg1-2, cDNA sequence and the activation domain of yeast Gal4 leading to a steady protein–DNA complex signal in shift TF. Both prey and bait vectors were integrated into yeast analysis and a decrease in signal intensity in Shift– a- and α-strains. Diploid strains, generated by mating and Western analysis due to interference in overall complex carrying one of the prey and the bait vector, were identi- formation as a result of a great excess of unbound com- fied by growth on selective media lacking uracil and petitor DNA (Fig. 6C). leucine. Furthermore, HIS3 reporter gene activity, indicating MAT1-1-1 binding to the respective prey sequence, was analyzed on selective media lacking uracil, leucine, MAT1-1-1 binding to MAT1.1 activates reporter gene and histidine, but containing increasing amounts of 3-AT. expression in an ex vivo yeast one-hybrid (Y1H) assay In addition, qualitative and quantitative β-galactosidase As biochemical assays confirmed MAT1-1-1 binding to the assays were performed to measure lacZ reporter gene newly identified MAT1-1-1 DNA-binding motif MAT1.1 in activity, thereby enabling evaluation of protein–DNA inter- vitro, yeast one-hybrid (Y1H) reporter gene assays were actions based on two independent reporter gene systems.

Fig. 6. Single-bp substitutions and Shift–Western analyses conﬁrm speciﬁcity of MAT1-1-1 DNA-binding. A. GST-MAT1-1-1 shows strong binding to a 30 nt double-stranded oligonucleotide derived from the ppg1 promoter sequence (Ppg1-2), carrying two copies of the predicted MAT1-1-1 DNA-binding motif MAT1.1. A single A→G/T→C substitution at position 3 within one of two motif sequences (Ppg1-2_m1) results in a diminished formation of protein–DNA complexes. Complex formation is completely suppressed when both consensus sequences are mutated (Ppg1-2_m2). B. Competition with increasing amounts of unlabeled Ppg1-2 oligonucleotide decreased the level of MAT1-1-1 binding to the labeled probe, leading to an attenuation of the shift band and accumulation of free labeled probe in shift experiments (autoradiogram, left panel). Western blotting and immunodetection (Shift–Western analysis), using an antibody to GST, showed an increase in complex signal strength when competing with unlabeled Ppg1-2 probe (GST-immunoblot, right panel). C. Unlabeled Ppg1-2_m2 oligonucleotide did not compete for binding to MAT1-1-1 with labeled Ppg1-2, leading to a steady signal for protein–DNA complexes in shift experiments (autoradiogram, left panel) and a decrease in signal intensity in Shift–Western analysis (GST-immunoblot, right panel). The amount of protein used for shift analyses is indicated on top of each lane (1 μg of GST-MAT1-1-1 equals a molar concentration of 0.76 μM). Positions of free probe (*) and protein–DNA complexes (→) are indicated.

Fig. 7. Yeast one-hybrid analysis conﬁrms MAT1-1-1 binding to MAT1.1. Yeast strains were grown on SD-ura-leu in order to conﬁrm the presence of both, a bait and a prey vector, after mating. HIS3 reporter gene activity was analyzed on SD-ura-leu-his supplemented with 3-AT as indicated. lacZ reporter gene activity was analyzed using qualitative and quantitative β-galactosidase assays. A diploid strain harboring the mutated CPCR1 binding site BSIIm1 as a prey and the transcription factor CPCR1 as a bait construct was used as a negative control. A diploid strain carrying the native BSII binding site as a prey and CPCR1 as a bait is shown as a positive control (Schmitt et al., 2004).

Previously successfully employed Y1H plasmids were MAT1-ChIP and P2niaD18, and plasmid integration was used as a positive and negative control, and served as conﬁrmed using PCR analysis. A plasmid containing the standard for quantitative β-galactosidase assays (Schmitt DsRed gene without a promoter sequence (pDsRed) was and Kück, 2000) (Fig. 7). integrated into MAT1-ChIP as a control. As MAT1-ChIP

Y1H analysis confirmed binding of MAT1-1-1 to oligo- contained the Pgpd::egfp::MAT1-1-1 overexpression con- nucleotides Kex1-2, Ppg1-2, and Ppg1-2_m1 based on struct used for ChIP analysis, all derivatives of this strain both, HIS3 and lacZ, reporter gene activity. Moreover, showed clear nuclear EGFP signals, while no signals quantitative β-galactosidase assays confirmed our results were detectable in the P2niaD18 background. DsRed obtained from Shift–Western assays using oligonucleo- expression in MAT1-ChIP+Pppg1::DsRed and MAT1- tide Ppg1-2_m1. In both analyses, binding between ChIP+Pkex1::DsRed confirmed binding of the MAT1-1-1 MAT1-1-1 and the oligonucleotide was reduced due to a protein to the promoter regions of kex1 and ppg1, while no single point mutation within one copy of MAT1.1. As inte- fluorescence was recorded for the MAT1-ChIP+pDsRed gration of Ppg1-2_m2 into the prey vector pHISi led to control strain (Fig. 8). Because only weak DsRed transactivation with the empty bait vector pGADT7, Y1H fluorescence was detectable for Pkex1::DsRed in P2niaD18 analysis did not yield reliable results in this particular and no DsRed fluorescence was detectable for case. Most probably, this activation was mediated by a P2niaD18+Pppg1::DsRed, overall activation of reporter yeast protein that binds with high affinity to the mutated gene expression could be clearly attributed to high MAT1- binding sequence. Additional control experiments were 1-1 gene expression in the MAT1-ChIP background. performed to exclude transactivation between pGADT7 Thus, fluorescence microscopy confirmed the in vivo and the remaining prey vectors (Fig. S7). specificity of MAT1-1-1 binding to promoter regions of kex1 and ppg1.

DsRed reporter gene assays conﬁrm MAT1-1-1 binding to the kex1 and ppg1 promoter sequence in vivo Characterization of a MAT1-1-1 target gene that functions beyond sexual development To further verify binding between MAT1-1-1 and the promoter regions of kex1 and ppg1 in vivo, we performed To further validate functionality of a new MAT1-1-1 target DsRed reporter gene assays in P. chrysogenum. For this gene, identiﬁed in our ChIP-seq approach and unlikely purpose, reporter gene constructs carrying the DsRed to be involved in regulation of sexual development, we gene under control of the upstream sequence of kex1 and generated artA (Pc18g03940) deletion strains (ΔartA) by ppg1 were transformed into P. chrysogenum recipients homolog integration of a PtrpC-nat1 resistance cassette

Fig. 8. In vivo DsRed reporter gene analysis confirms MAT1-1-1 binding to selected target gene promoter regions. Reporter gene constructs carrying the DsRed gene under control of the upstream sequence of kex1 and ppg1 (Pkex1: 1445 nt; Pppg1: 843 nt) were transformed into P. chrysogenum strains MAT1-ChIP and P2niaD18. Fluorescence microscopy confirmed EGFP-MAT1-1-1 expression and nuclear localization in the MAT1-ChIP background. DsRed protein expression confirmed binding of MAT1-1-1 to the respective promoter regions. Scale bar = 20 μm. in Δku70FRT2 background. Correct integration of the reduction in conidiospore germination (∼ 30% germina- knockout construct was verified using PCR analysis. tion after 24 h), when compared with the recipient ArtA codes for a 14-3-3 family protein, which was pre- Δku70FRT2 and wild type P2niaD18 (∼ 90% germination viously shown to be involved in a pathway controlling after 24 h). This effect was further verified using micro- conidiospore germination in A. nidulans (Kraus et al., scopic analysis, confirming an impaired growth in ΔartA 2002). As shown in Fig. 9A, deletion of the correspond- compared with the reference strains after 24 h of culti- ing homolog in P. chrysogenum results in a severe vation (Fig. 9B).

Fig. 9. Characterization of artA deletion strains. A. Three independent artA deletion (ΔartA) mutants, recipient Δku70FRT2, and wild type P2niaD18 were grown on solid CCM. For each time point 400 conidiospores from each strain were investigated for determination of germination rates after 12, 15, 18, 21, and 24 h (given in %). B. Microscopic analysis of strains used in (A) after 24 h of cultivation. Scale bar = 50 μm.

Discussion (Pppg1,Pkex1), can be used for controlled expression of downstream reporter genes in a MAT1-1-1-dependent ChIP-seq analysis identifies MAT1-1-1 target genes that manner. Furthermore, DNA-binding assays and qRT-PCR have functions other than for sexual development analyses confirmed functionality of at least 13 non- Although concerted research efforts have been made to mating-related MAT1-1-1 target genes, identified in our analyze regulatory circuits controlled by mating-type- ChIP-seq approach and covering the functional catego- encoded TFs, little is still known about their specific target ries morphogenesis and development, amino acid and genes. Our work expands the current understanding of secondary metabolism, iron metabolism as well as TFs. mating-type protein functions far beyond the regulation of Important examples are the spore wall fungal hydro- sexual development alone, and provides unambiguous phobin encoding dewA, the non-ribosomal peptide syn- evidence for a participation of the mating-type α-domain thetase encoding sidD, the bZIP TF encoding atf21 TF MAT1-1-1 in regulation of asexual development and the F-box domain protein encoding Pc22g22160. and morphogenesis, as well as amino acid, iron and sec- Moreover, functional characterization of artA, a further ondary metabolism in P. chrysogenum. Furthermore, we MAT1-1-1 target gene, confirmed its role in regulation of present the first genome-wide analysis focusing on conidiospore germination in P. chrysogenum. A compara- unraveling the transcriptional regulatory network con- ble function was previously shown for A. nidulans (Kraus trolled by a mating-type locus-encoded TF and the com- et al., 2002). This observation supports our hypothesis prehensive characterization of a MAT1-1-1 DNA-binding that MAT1-1-1 functions on a genome-wide level are more motif in euascomycetes. far-ranging than expected, and that the number of primary Identification of Pc14g01160 and Pc22g15650 as MAT1-1-1 target genes might be significantly higher than MAT1-1-1 specific target genes confirms the biological previously assumed. To improve clarity, all MAT1-1-1 significance of our ChIP-seq analyses, as they are target genes, verified by EMSAs and/or qRT-PCR analy- homologs of S. cerevisiae MFα1 and STE3, respectively, sis, are labeled with an asterisk (*) throughout this discus- both αsgs (Galgoczy et al., 2004). Since their correspond- sion (see Table 2 for further information). ing peak regions revealed significantly high statistical Interestingly, none of the identified MAT1-1-1 target peak values, these observations are consistent with the genes, assigned to sexual development in P. chrysoge- general acceptance that the most highly bound regions num, showed MAT1-1-1-dependent changes in expres- in ChIP experiments occur near generally known func- sion profiles in microarray analysis comparing ΔMAT1 to tional targets, while many of the regions bound at much wild type P2niaD18, except for kex1* and the homolog of lower levels may represent ‘non-functional’ binding sites pac2 (Pc12g15890). Pac2 encodes a cAMP-independent (Todeschini et al., 2014). Nevertheless, low-affinity TF regulatory protein, modulating onset of sexual develop- binding may have a functional role in chromatin remod- ment in S. pombe and M. oryzae, and regulation of sporu- eling (Cao et al., 2010) or nucleosome positioning (Zaret lation in Ashbya gossypii (Kunitomo et al., 1995; and Carroll, 2011), which can influence gene expression Wasserstrom et al., 2013; Chen et al., 2014). On the con- at later developmental stages or have an additional non- trary, qRT-PCR analysis revealed a significant upregula- transcriptional function (Spitz and Furlong, 2012). tion of pre1*, kex1*, and ppg1* expression in a MAT1-1-1 We used previous microarray data (Böhm et al., 2013), overexpression strain (MAT1-ChIP) and significant down- which identified a total of 2421 genes as MAT1-1-1- regulation of pre1* in ΔMAT1 compared with P2niaD18 dependent in a MAT1-1-1 deletion strain (ΔMAT1) com- after 48 h of cultivation in shaking cultures. Accordingly, pared with wild type P2niaD18, to align ChIP-seq results pre1* and ppg1* expression was shown to be significantly and expression profiles of putative MAT1-1-1 target downregulated in ΔMAT1 compared with the parental genes, identified in our ChIP-seq analysis. This compari- strain after 72 h of cultivation in liquid shaking cultures son revealed an overlap of 29.9% (76/254), which is con- (Böhm et al., 2013). These findings are consistent with sistent to comparisons between TF binding events and reports, demonstrating that expression of pheromone pre- expression profiling data in yeast and higher eukaryotes, cursor genes, and most probably receptor genes, is con- showing a relatively small overlap of ∼ 50% and 10–25% trolled by mating-type gene expression in heterothallic between TF occupancy and expression of neighboring species, e.g. N. crassa (Kim and Borkovich, 2006). The genes (Spitz and Furlong, 2012). Nevertheless, our overexpression of MAT1-1-1 thus has an impact on the assumption that most of the 243 MAT1-1-1 binding sites expression of genes involved in regulation and onset of identified in ChIP-seq experiments affect the expression sexual reproduction. Similar observations were made in of neighboring genes at some point during development A. nidulans and N. crassa, in which sexual reproduction was strengthened when DsRed reporter gene assays correlates significantly with an increased expression of showed that MAT1-1-1 binding to promoter regions, iden- mating-type genes and key genes of a pheromone- tified as specific target regions in ChIP-seq analyses response MAP-kinase signaling pathway (Paoletti et al.,

2007; Wang et al., 2014). However, deletion of MAT1-1-1 of nitrogen limitation as a key feature of induction of in P. chrysogenum does not lead to significant changes in sexual reproduction in P. chrysogenum. expression levels of key genes from the pheromone- We found a variety of MAT1-1-1 target genes linked to response signaling pathway at early developmental amino acid and secondary metabolism, most of them stages, suggesting that these are, to a certain degree, showing downregulation in ΔMAT1 compared with wild independent of MAT1-1-1. type. Pc18g02620, encoding a cyanide hydratase/ A significant number of MAT1-1-1 target genes, identi- nitrilase, and Pc22g18630*, encoding an enzyme catalyz- fied in our ChIP-seq analyses, might be involved in the ing the chemical reaction of L-homocysteine to manifestation of the phenotypic characteristics of MAT1- L-methionine, are important candidates, as they might 1-1 overexpression and deletion strains, showing altered have a direct impact on penicillin biosynthesis, which polarity of germinating hyphae, unusual branching behav- starts with the formation of a tripeptide based on ior, and impaired hyphal growth and pellet formation L-cysteine, L-valine and L-α-aminoadipic acid. Several (Böhm et al., 2013). Examples are dewA* (Pc16g06690), multidrug (Pc12g00820, Pc16g06630, Pc16g11470, encoding a spore wall fungal hydrophobin responsible Pc20g03900) and amino acid transporter encoding genes for hydrophobicity of conidiospores in A. nidulans (Pc06g01080, Pc22g06500) complete this selection. As (Stringer and Timberlake, 1995; Grünbacher et al., 2014), deletion of MAT1-1-1 was shown to lead to a significant artA* (Pc18g03940), coding for a 14-3-3 family protein reduction in penicillin production (Böhm et al., 2013), involved in regulation of polarization of germinating con- these observations strengthen our idea of MAT1-1-1 idiospores in A. nidulans (Kraus et al., 2002), and PAG1 being a positive regulator of secondary metabolism in (Pc21g20900), encoding a cell morphogenesis protein P. chrysogenum. related to polarized morphogenesis and proliferation Furthermore, integration of ChIP-seq and microarray in S. cerevisiae (Du and Novick, 2002; Nelson et al., data led to the identification of MAT1-1-1 target genes 2003). Furthermore, genes assigned to the formation of involved in iron transport and iron acquisition, e.g. sidD* conidiospores were identified and showed significant (Pc22g20400), encoding a non-ribosomal siderophore upregulation in ΔMAT1, e.g. atf21* (Pc22g26820) and peptide synthetase important for biosynthesis of the intra- Pc19g00140*. Atf21* codes for a basic leucine zipper cellular siderophore triacetylfusarinine C (TAFC) (Schrettl (bZIP) TF and repressor of sexual development in A. nidu- et al., 2007), fetC* (Pc21g08020), encoding for a ferroxi- lans (Lara-Rojas et al., 2011), while Pc19g00140* shows dase, and ftrA* (Pc21g08030), encoding for a high affinity high similarity to trehalose-6-phosphate synthase subunit iron permease that mediates uptake of Fe2+ during reduc- encoding genes from Aspergilli, involved in the biosynthe- tive iron acquisition (Schrettl and Haas, 2011). It is known sis of trehalose, a compound necessary for long-term from Aspergillus species that imbalance in iron homeo- viability of fungal spores (Elbein et al., 2003). As formation stasis affects a variety of cellular functions, e.g. growth of conidiospores is generally accepted to be restricted to rates, germination, sensitivity of conidia to oxidative asexual development, this observation fits the notion of stress and formation of cleistothecia (Eisendle et al., MAT1-1-1 being a positive regulator of sexual reproduc- 2006a). Furthermore, deletion of sidD* in A. fumigatus tion and a negative regulator of asexual development. was shown to lead to decreased conidiation during iron- Consistent with this hypothesis is our recent finding that depleted conditions (Schrettl et al., 2007), whereas dele- sporulation was increased by about 25% in a ΔMAT1-1-1 tion of ftrA* displayed an eightfold increase in TAFC strain compared with wild type (Böhm et al., 2013). siderophore production under iron-depleted conditions, It is known that the developmental decision between demonstrating that lack of FtrA brings forward the onset of sexual and asexual reproduction in A. nidulans is siderophore production (Schrettl et al., 2004). dependent on environmental factors, such as nutritional status and culture conditions (Han et al., 2003). Conse- Identification of a new MAT1-1-1 DNA-binding motif quently, in most out-crossing ascomycetes, such as N. crassa and S. cerevisiae, nitrogen limitation is a key Using EMSAs, Shift–Western and Y1H analysis, we inducing condition for mating or sexual sporulation showed that MAT1-1-1 binds with high specificity to (Glass and Lorimer, 1991). As we identified meaB* the newly identified MAT1.1 DNA-binding consensus (Pc18g00880), a bZIP TF involved in regulation of sequence ‘CTATTGAG’. The motif was further shown to expression of nitrogen-dependent genes in A. nidulans be conserved among euascomycetes and showed simi- (Wong et al., 2007), as a target gene of MAT1-1-1 in larities to known DNA-binding motifs of proteins known to ChIP-seq analyses and microarray analysis indicated be involved in regulation of sexual reproduction in yeast, upregulation in ΔMAT1 compared with wild type, MAT1- e.g. MATa1, Mcm1, and Hcm1, and embryonic develop- 1-1 seems to act as a negative regulator of meaB* ment in vertebrates, e.g. Sox9, Nkx2-5, and Sox17. Even expression in P. chrysogenum, thus, supporting the idea though DNA-sequence recognition by TFs can be con-

© 2015 The Authors. Molecular Microbiology published by John Wiley & Sons Ltd, Molecular Microbiology 14 K. Becker, C. Beer, M. Freitag and U. Kück ■ served across large evolutionary distances, binding speci- et al., 2004; Baker et al., 2011), dimerization might also be ficity of MATα1 has been shown to have changed a regulatory feature of MAT1-1-1 in P. chrysogenum. substantially over small evolutionary distances (Tuch Interestingly, the most prominent MAT1-1-1 target et al., 2008). As our analysis pointed to obvious differ- genes, characterized by high statistical peak values com- ences between MAT1-1-1 binding sites in euascomycetes bined with an accumulation of MAT1.1 (p ≤ 0.001), were and hemiascomycetes, especially C. albicans, these find- assigned to sexual reproduction. This finding might indi- ings are consistent with the hypothesis that hemiascomy- cate a regulatory feature ensuring high-affinity binding of cetes and euascomycetes share a common ancestor, but MAT1-1-1 to the corresponding promoter regions, even that binding specificity of modern MATα1 proteins from under conditions where only a low amount of MAT1-1-1 C. albicans and euascomycetes might have changed sub- protein is available. Moreover, the occurrence of MAT1.1 stantially during evolution (Baker et al., 2011). within the upstream regions of new direct MAT1-1-1 target The strongest protein–DNA interaction was observed genes presented within this work points to an evolutionary between MAT1-1-1 and an oligonucleotide probe harbor- link between mating and other cellular functions which ing two copies of the MAT1.1 binding motif, forming the were believed to be independent of MAT1-1-1 protein imperfect palindrome 5′-TCAATA-N7-TATTGA-3′. Corre- functions until now. This hypothesis was further strength- spondingly, the strongest interaction between DNA ened by EMSAs and qRT-PCR analyses, verifying func- and MAT1-1-1, as deduced from ChIP-seq data, was tionality of selected MAT1-1-1 target genes identified in observed for those peak regions characterized by a our ChIP-seq approach. Further research is needed to noticeable high frequency of MAT1.1 with close matches identify interaction partners of MAT1-1-1 on protein level to the consensus sequence, whereas weak interactions and to understand interactions between the TF, enhancer were characterized by a relatively low abundance of the elements and other cis-regulatory elements. Furthermore, motif (compare with Table 2 and Dataset S1). Since as our analysis was designed to identify as many MAT1- eukaryotic TFs tend to recognize shorter DNA sequence 1-1 target genes as possible, further studies, however, will motifs compared with bacterial TFs, clustering of sites is be needed in order to decipher MAT1-1-1 mediated tran- often required to achieve specific recognition (Wunderlich scriptional regulation under control of its native promoter and Mirny, 2009). sequence, e.g. as a function of developmental stages or Although a large number of MAT1-1-1 peak regions physiological culture conditions. contained at least one copy of MAT1.1, some peaks com- Taken together, our discoveries concerning the sexual pletely lacked it. However, this might be due to statistical biology of P. chrysogenum presented within this work thresholds applied during motif prediction and motif greatly advance the current understanding of sexual detection procedures. On the other hand, this is a reproduction within ascomycetes, and open up new common observation: even if ChIP-seq peaks are typi- avenues for the study of fungal development as a whole. cally enriched in the consensus motif for the TF in ques- Based on our finding that the mating-type encoded TF tion, a significant proportion of peaks lacks clearly MAT1-1-1 not only regulates expression of αsgs related to identifiable motifs (Robertson et al., 2007; Valouev et al., sexual reproduction but also other key biological pro- 2008). For example, the consensus sequence for E2F cesses, it appears that mating-type regulated transcrip- family proteins that control various cellular and organismal tional networks have undergone drastic reorganization, functions in higher eukaryotes is present in less than 20% resulting in the presence of DNA binding sites in the of the regions recognized in ChIP-chip experiments in promoters of – at first glance – unrelated target genes that human and mouse cells (Rabinovich et al., 2008). This are bound and controlled by highly conserved transcrip- observation might be ascribed to the fact that most TFs tional regulators in different fungi. This hypothesis is sup- not only interact with DNA through a consensus site but ported by a recent discovery showing that targets of the also recognize divergent sequences. For example, a mating-type TF heterodimer Sxi2a-Sxiα1 from Cryptococ- study of approximately 100 mouse TF revealed that cus neoformans not only include genes known to be almost half of these proteins can recognize several differ- involved in sexual reproduction but also several well ent sequences in addition to the known DNA-binding con- studied virulence genes (Mead et al., 2015). Microarray sensus sequences (Badis et al., 2009). Furthermore, analyses in other euascomycetes also pointed to an unex- specific recognition of regulatory elements by a TF is pectedly large number of genes that are expressed in strongly influenced by its ability to interact with other pro- a mating-type dependent manner (Lee et al., 2006; teins that bind to neighboring DNA sites. The simplest Pöggeler et al., 2006; Keszthelyi et al., 2007; Bidard example of this mechanism is the formation of TF dimers et al., 2011). In combination, these data suggest that or higher order structures (Amoutzias et al., 2008). Since mating-type protein regulatory functions might reach far cooperative binding was described for the mating-type α1 beyond sexual development in these species as well. HMG domain TF and Mcm1 from S. cerevisiae (Carr Future research will be necessary in order to determine

© 2015 The Authors. Molecular Microbiology published by John Wiley & Sons Ltd, Molecular Microbiology Target genes of a mating-type transcription factor 15 exactly which changes in MAT1-1-1 and its corresponding Construction of recombinant P. chrysogenum strains DNA binding site were necessary to allow for the expan- For generation of strains used for ChIP-seq analysis, DsRed sion in MAT1-1-1 regulatory functions during evolution. reporter gene assays and deletion mutants (Table S1), the The observation that MAT1-1-1 is involved in regulation corresponding plasmids (Table S2) were transformed into of development, morphogenesis and metabolism in P. chrysogenum strain P2niaD18 and Δku70FRT2, respec- P. chrysogenum supports the idea that MAT genes are tively. Transformation was performed as described previously functionally retained even during the asexual part of the (Hoff et al., 2010; Kamerewerd et al., 2011) with some modi- life cycle and the apparent absence of a sexual phase, fications. Cultures were grown for 72 h and protoplasts were presumably because of the impact of positive selection on transformed with circular plasmid DNA for ectopic, and linear plasmid DNA for homologous integration. Transformants important processes unrelated to sexual development in − were selected on CCM media containing 150 μgmL1 nour- asexual fungal populations (Ádám et al., 2011). Since we seothricin (Werner BioAgents, Jena, Germany) and demonstrated that MAT1-1-1 regulates expression of a 40 μgml−1 phleomycin (Invivogen, CA, USA) as necessary. number of genes related to various traits of morphology Resistant colonies were isolated and tested for correct inte- and development, it is conceivable that the mating-type gration of plasmid DNA as previously described (Hoff et al., protein mediated regulation is necessary for efficient 2010). balance between morphologic features characteristic to the sexual and asexual parts of the life cycle. This might Sample preparation for ChIP-seq also be true for an involvement of MAT1-1-1 in regulation Chromatin immunoprecipitation (ChIP) was carried out of secondary, amino acid and iron metabolism. It is known essentially as described previously (Tamaru et al., 2003; from various euascomycetes that there is a concerted Smith et al., 2011) with the following modifications. P. chrys- balance between sexual development and secondary ogenum strains were grown in 100 mL CCM cultures inocu- × 7 metabolism (Bayram et al., 2008; Hoff et al., 2010; lated with 0.5 10 spores for 48 h at 120 rpm and 27°C. For chromatin fixation, freshly prepared formaldehyde (in NaOH) Wiemann et al., 2010; Kopke et al., 2012). Another impor- was added to a final concentration of 1%, and cultures were tant example is fungal iron metabolism, which was shown incubated at 27°C and 100 rpm for 30 min. Five milliliters of to affect both asexual and sexual development (Eisendle 2.5 M glycine was added to quench formaldehyde, and cul- et al., 2003; 2006b; Schrettl et al., 2007; Johnson, 2008). tures were incubated at room temperature with gentle Since these traits are also crucial in terms of applied shaking for 5 min. Approximately 250 mg mycelium were microbiology, our work will further not only contribute to resuspended in 750 μL lysis buffer (50 mM HEPES–KOH pH the advanced improvement of P. chrysogenum strains 7.5, 90 mM NaCl, 1 mM ethylenediaminetetraacetic acid (EDTA), 1% Triton X-100, 0.1% sodium deoxycholate (DOC) used for industrial production of β-lactam antibiotics supplemented with fresh protease inhibitors) and chromatin but also to other filamentous fungi with biotechnological was sheared using a Branson 250 sonifier (output 2, duty relevance. cycle 0.8, 6 × 20 impulses). After pre-clearing with protein A agarose beads (Invitrogen, Darmstadt, Germany) the soluble Experimental procedures chromatin fraction was immunoprecipitated using anti-GFP antibody (ab290; Abcam, Cambridge, UK). Fresh protein A Strains and culture conditions agarose beads were added to bind antibody–protein–DNA complexes. The supernatant was discarded and beads were Penicillium chrysogenum strains (Table S1) were grown in washed several times (1 × TE buffer: 10 mM Tris–HCl pH 8.8, shaking or surface cultures in complete culture medium (CCM; 1 mM EDTA; 2 × lysis buffer without protease inhibitors; 0.3% (w/v) sucrose, 0.05% (w/v) NaCl, 0.05% (w/v) K2HPO4, 1 × lysis buffer without protease inhibitors + 0.5 M NaCl; 0.05% (w/v) MgSO4, 0.001% (w/v) FeSO4, 0.5% (w/v) tryptic 1 × LiCl wash buffer: 0.25 M LiCl, 1 mM EDTA, 10 mM Tris– soy broth, 0.1% (w/v) yeast extract, 0.1% (w/v) meat extract, HCl pH 8.0, 0.5% NP-40, 0.5% DOC). Beads were incubated 0.15% (w/v) dextrin, pH 7.0) at 27°C. For inoculation, 0.5 × 107 two times in TE(S) (50 mM Tris–HCl pH 8.0, 10 mM EDTA, spores derived from cultures grown on M322 solid medium 1% SDS) at 65°C for 10 min with gentle agitation to elute (0.35% (w/v) (NH4)2SO4, 0.2% (w/v) KSO4, 0.02% (w/v) protein–DNA complexes. To reverse the crosslinking, KHSO4, 1 g N/l soy flour, 0.5% (w/v) lime stone powder, 5% samples were incubated at 65°C for 6–16 h. After RNaseA (w/v) lactose, pH 6.3) for 4–5 days were used. Escherichia coli and ProteinaseK digestion, DNA from immunoprecipitated strain XL1 blue was used for cloning and plasmid propagation chromatin (ChIP-DNA) and input samples (input-DNA) was purposes, while BL21 (DE3) served as a host for heterologous isolated. Construction of ChIP-libraries and sequencing of 50 overexpression of MAT1-1-1 (Bullock et al., 1987; Miroux and nt single-end reads on a Illumina HiSeq 2000 were performed Walker, 1996). Saccharomyces cerevisiae strains PJ69-4a by GATC Biotech AG (Konstanz, Germany) or at the OSU and PJ69-4α were used for yeast one-hybrid analysis (James CGRB core facility. et al., 1996). Strains were grown at 30°C on synthetic defined (SD) medium lacking selected amino acids used for auxotro- Data analysis and visualization phy marker selection. Mating of PJ69-4a and -4α strains was performed in liquid yeast peptone dextrose adenine (YPDA) Sequences corresponding to adaptors were removed from medium at 30°C and 50 rpm. reads, and remaining sequences were subsequently mapped

© 2015 The Authors. Molecular Microbiology published by John Wiley & Sons Ltd, Molecular Microbiology 16 K. Becker, C. Beer, M. Freitag and U. Kück ■ to the latest version of the P. chrysogenum P2niaD18 Quantification of protein levels and immunodetection genome (Specht et al., 2014) using Bowtie version 1.0.1 (Langmead et al., 2009) with the following settings: ‘–S –q – The concentration of purified GST-MAT1-1-1 and GST alone m 1’, which only retains unique alignments. Binary Alignment/ was determined by using Bradford reagent (BioRad, Map (BAM) files were sorted and indexed using SAMtools München, Germany). Western blotting and immunodetection (Li et al., 2009), and visualized using the Integrative Genom- of GST-tagged proteins were performed using RPN1236 anti- ics Viewer (IGV) (Thorvaldsdóttir et al., 2012). A genome- GST HRP conjugate (GE Healthcare, Freiburg, Germany). wide distribution figure of MAT1-1-1 binding sites is provided Detection of GFP-MAT1-1-1 from P. chrysogenum total in Fig. S8. Further data analysis was performed using the protein isolates was performed using JL-8 antibody to GFP HOMER software for motif discovery and next-generation (Clontech, Saint-Germain-en-Laye, France) and HRP- sequencing analysis (Heinz et al., 2010). Quality control coupled secondary antibody #7076 (Cell Signaling Technol- analysis included examination of clonal tag counts in order ogy, Leiden, The Netherlands). to determine the non-redundant fraction of mapped reads, autocorrelation analysis to enable sequencing fragment length estimation, nucleotide frequency analysis and frag- Electrophoretic mobility shift assays (EMSAs) and ment GC % distribution to rule out sequence biases and Shift–Western analysis analysis of ChIP-fragment density near MAT1-1-1-specific peak regions. Peaks were called using findPeaks.pl using Gel shift assays were performed using oligonucleotides the -style factor option, a FDR ≤ 0.001, and a p-value over derived from ChIP-enriched regions and purified GST-MAT1- local background cutoff of 1.00e–04. Peak regions for each 1-1. Double-stranded oligonucleotides were 5′-end-labeled individual experiment were intersected using mergePeak- using polynucleotide kinase (Roche, Basel, Switzerland) and s.pl –d 100, reporting peaks within a maximum distance of [γ-32P]-ATP (Hartmann Analytic, Braunschweig, Germany). 100 nt as overlapping. Peaks were assigned to neighboring For shift experiments, 3.5–7.0 fmol (∼ 50–100 cps) of radi- and overlapping genes using a custom-made Perl script olabeled oligonucleotides was incubated with varying protein based on BioPerl modules and blast2go analysis (Conesa concentrations in the presence of 2 μL binding buffer et al., 2005). Functional category enrichment analysis (250 mM Tris/HCl pH 8.0, 1 M KCl, 50 % glycerol) and 1 μg of genes associated with peaks was performed using the poly(dI-dC)-poly(dI-dC) (Affymetrix USB, CA, USA) in a total MIPS functional catalogue database (FunCat) (Ruepp et al., volume of 20 μL for 20 min at room temperature. Samples 2004). Raw sequencing data from ChIP experiments were run on 5% polyacrylamide gels at 4°C in 190 mM are available from the NCBI SRA database (http:// glycine, 27 mM Tris/HCl pH 8.5. Competition experiments www.ncbi.nlm.nih.gov/sra), study ID PRJNA257456, Acces- were performed by adding unlabeled oligonucleotide. Prepa- sion # SRP045261. ration of gels used for Shift–Western analysis (Demczuk et al., 1993) was performed as described earlier. Denatura- tion of proteins and blotting to a PVDF membrane (Perki- Sequence motif analysis nElmer, MA, USA) was performed as described previously (Granger-Schnarr et al., 1988) with a transfer time of The central 100 nt region of selected MAT1-1-1 peak regions 180 min at 1.3 A and a transfer buffer containing 25 mM Tris, was submitted to MEME (Multiple Em for Motif Elicitation; 192 mM glycine and 10 % methanol. The sequences of all http://meme.nbcr.net/meme/) (Bailey and Elkan, 1994) for oligonucleotides used for shift analyses are provided in Table de novo motif prediction. Further analysis was performed S3. using CentriMo (Bailey and Machanick, 2012) and FIMO (Grant et al., 2011). For comparison of the newly identified MAT1-1-1 DNA-binding consensus sequence against the Nucleic acids isolation, cDNA synthesis, quantitative JASPAR CORE (2014) fungi and vertebrates databases, RT-PCR and ChIP-PCR results were submitted to TOMTOM (Gupta et al., 2007) using default parameters. Isolation of nucleic acids, cDNA synthesis and qRT-PCR analysis were carried out as described earlier (Hoff et al., 2009; Böhm et al., 2013). ChIP-PCR analysis was performed Expression and purification of recombinant as described for qRT-PCR analysis, using ChIP- and input- GST-MAT1-1-1 protein DNA from independent ChIP experiments as a template. The sequences of all oligonucleotides used for PCR analyses are The MAT1-1-1 cDNA sequence was integrated into the given in Table S3. expression vector pGEX-4T3 (Amersham Bioscience, Frei- burg, Germany) to generate plasmid pGEX-MAT1 (see Table S2). GST and GST-MAT1-1-1 were purified from Microarray data analysis E. coli BL21 (DE3) cells. Purification of recombinant protein and GST alone was performed as described Analysis of microarray data was performed as described earlier using an elution buffer containing 50 mM Tris/HCl, previously using the affylmGUI R package (Wettenhall 30 mM reduced glutathione, 100 mM NaCl, pH 8.0 (Janus et al., 2006; Wolfers et al., 2014). p-Values for single time et al., 2007). Purified protein was supplemented with points were generated by treating datasets from light-grown 87% glycerol and stored at −70°C until used for further ΔMAT1 (48 h, 60 h, 96 h) as independent biological applications. replicates.

© 2015 The Authors. Molecular Microbiology published by John Wiley & Sons Ltd, Molecular Microbiology Target genes of a mating-type transcription factor 17 Yeast one-hybrid analysis Author contributions Complementary oligonucleotides harboring three copies of K.B., U.K., M.F. designed experiments; K.B., C.B. performed the corresponding oligonucleotide sequence used for EMSAs experiments; K.B. analyzed data; K.B., U.K., M.F. wrote the were cloned into plasmids pHISi and pLacZi to generate prey manuscript. All authors discussed results and commented on vectors for yeast one-hybrid analysis (see Table S3), as the manuscript. described previously (Schmitt and Kück, 2000). As a bait, the MAT1-1-1 cDNA sequence was integrated into plasmid pGADT7 to generate plasmid pMAT1-AD (see Table S2). Bait References and prey vectors were transferred into S. cerevisiae strains PJ69-4α and PJ69-4a, respectively. Diploid reporter strains Ádám, A.L., García-Martínez, J., Szücs, E.P., Avalos, J., and harboring both, the bait and a prey vector, were generated by Hornok, L. (2011) The MAT1-2-1 mating-type gene upregu- mating. For analyzing DNA–protein interactions between lates photo-inducible carotenoid biosynthesis in Fusarium MAT1-1-1 and putative DNA-binding sites, reporter strains verticillioides. FEMS Microbiol Lett 318: 76–83. were tested for growth on -his/-leu/-ura selective media sup- Amoutzias, G.D., Robertson, D.L., Van de Peer, Y., and plemented with 3-amino-1,2,4-triazole (3-AT) (Merck, Darm- Oliver, S.G. (2008) Choose your partners: dimerization in stadt, Germany) as indicated. Further, β-galactosidase activity eukaryotic transcription factors. Trends Biochem Sci 33: of reporter strains was analyzed by qualitative and quantitative 220–229. determination of 5-bromo-4-chloro-3-indolyl-beta-D-galacto- Badis, G., Berger, M.F., Philippakis, A.A., Talukder, S., pyranoside and O-nitrophenyl β-d-galactopyranoside turno- Gehrke, A.R., Jaeger, S.A., et al. (2009) Diversity and ver, respectively. complexity in DNA recognition by transcription factors. Science 324: 1720–1723. Bailey, T.L., and Elkan, C. (1994) Fitting a mixture model by Microscopy expectation maximization to discover motifs in biopolymers. Proc Int Conf Intell Syst Mol Biol 2: 28–36. P. chrysogenum strains were grown on glass slides with a Bailey, T.L., and Machanick, P. (2012) Inferring direct thin layer of CCM at 27°C. Fluorescence and light microscopy DNA binding from ChIP-seq. Nucleic Acids Res 40: was carried out with an AxioImager M1 fluorescence micro- e128. scope (Zeiss, Jena, Germany) using a SPECTRA Light Baker, C.R., Tuch, B.B., and Johnson, A.D. (2011) Extensive Engine® LED lamp (Lumencor, OR, USA) as described pre- DNA-binding specificity divergence of a conserved tran- viously (Engh et al., 2007). Images were captured with a scription regulator. Proc Natl Acad Sci USA 108: 7493– Photometrix Cool SnapHQ camera (Roper Scientific, AZ, 7498. USA) and MetaMorph software version 6.3.1. Recorded Baker, C.R., Booth, L.N., Sorrells, T.R., and Johnson, A.D. images were edited with MetaMorph and Adobe Photoshop (2012) Protein modularity, cooperative binding, and hybrid CS4. Counter staining of nuclei was performed using regulatory states underlie transcriptional network diversifi- NucBlue® Live Cell Stain (Life Technologies GmbH, Darm- cation. Cell 151: 80–95. stadt, Germany) as specified by the manufacturer. Pellet Bayram, Ö., Krappmann, S., Ni, M., Bok, J.W., Helmstaedt, quantification assays were conducted as described earlier K., Valerius, O., et al. (2008) VelB/VeA/LaeA complex coor- (Böhm et al., 2013). dinates light signal with fungal development and secondary metabolism. Science 320: 1504–1506. Multiple sequence alignments Bender, A., and Sprague, G.F., Jr (1987) MAT alpha 1 protein, a yeast transcription activator, binds synergistically Multiple sequence alignments were performed using the with a second protein to a set of cell-type-specific genes. Guidance server (http://guidance.tau.ac.il/) and MAFFT Cell 50: 681–691. default settings (Penn et al., 2010). Alignments were visual- Bidard, F., Ait Benkhali, J., Coppin, E., Imbeaud, S., Grognet, ized using Jalview according to the Clustalx color scheme P., Delacroix, H., and Debuchy, R. (2011) Genome-wide (http://www.jalview.org/) (Waterhouse et al., 2009). gene expression profiling of fertilization competent mycelium in opposite mating types in the heterothallic fungus Podospora anserina. PLoS ONE 6: e21476. Acknowledgements Böhm, J., Hoff, B., O’Gorman, C.M., Wolfers, S., Klix, V., Binger, D., et al. (2013) Sexual reproduction and mating- We thank L. Connolly, Dr. J. Galazka, Dr. E. Bredeweg, S. type-mediated strain development in the penicillin- Friedman and M. Dasenko for help with ChIP-seq analyses, producing fungus Penicillium chrysogenum. Proc Natl PD Dr. M. Nowrousian, M. Sc. T. A. Dahlmann and M. Sc. D. Acad Sci USA 110: 1476–1481. Terfehr for help with bioinformatics, and I. Godehardt for Bullock, W.O., Fernandez, J.M., and Short, J.M. (1987) Xl1- technical assistance. We thank Dr. I. Zadra, Dr. H. Kürnsteiner, Blue – a high-efficiency plasmid transforming reca Escheri- Dr. E. Friedlin, and Dr. T. Specht for their ongoing interest and chia coli strain with beta-galactosidase selection. support, and Dr. I. Teichert for critical reading of the manu- Biotechniques 5: 376–378. script. This work was funded by Sandoz GmbH, the Christian Cao, Y., Yao, Z., Sarkar, D., Lawrence, M., Sanchez, G.J., Doppler Society, the German National Academic Foundation, Parker, M.H., et al. (2010) Genome-wide MyoD binding in and the Ruhr-University Bochum Research School. skeletal muscle cells: a potential for broad cellular repro- The authors declare no conflict of interest. gramming. Dev Cell 18: 662–674.

Carr, E.A., Mead, J., and Vershon, A.K. (2004) Alpha1- Elbein, A.D., Pan, Y.T., Pastuszak, I., and Carroll, D. (2003) induced DNA bending is required for transcriptional activa- New insights on trehalose: a multifunctional molecule. Gly- tion by the Mcm1-alpha1 complex. Nucleic Acids Res 32: cobiology 13: 17R–27R. 2298–2305. Engh, I., Würtz, C., Witzel-Schlomp, K., Zhang, H.Y., Hoff, B., Chen, C.Y., and Schwartz, R.J. (1995) Identification of novel Nowrousian, M., et al. (2007) The WW domain protein DNA binding targets and regulatory domains of a murine PRO40 is required for fungal fertility and associates with tinman homeodomain factor, Nkx-2.5. J Biol Chem 270: woronin bodies. Eukaryot Cell 6: 831–843. 15628–15633. Fukuda, K., Yamada, K., Deoka, K., Yamashita, S., Ohta, A., Chen, Y., Zhai, S., Zhang, H., Zuo, R., Wang, J., Guo, M., and Horiuchi, H. (2009) Class III chitin synthase ChsB of et al. (2014) Shared and distinct functions of two Gti1/ Aspergillus nidulans localizes at the sites of polarized cell Pac2 family proteins in growth, morphogenesis and patho- wall synthesis and is required for conidial development. genicity of Magnaporthe oryzae. Environ Microbiol 16: Eukaryot Cell 8: 945–956. 788–801. Galgoczy, D.J., Cassidy-Stone, A., Llinas, M., O’Rourke, Conesa, A., Götz, S., García-Gómez, J.M., Terol, J., Talón, S.M., Herskowitz, I., DeRisi, J.L., and Johnson, A.D. (2004) M., and Robles, M. (2005) Blast2GO: a universal tool for Genomic dissection of the cell-type-specification circuit in annotation, visualization and analysis in functional genom- Saccharomyces cerevisiae. Proc Natl Acad Sci USA 101: ics research. Bioinformatics 21: 3674–3676. 18069–18074. Coppin, E., and Debuchy, R. (2000) Co-expression of the Glass, N.L., and Lorimer, I. (1991) More gene manipulations mating-type genes involved in internuclear recognition is in fungi. In Ascomycete Mating Types. Bennett, J.W., and lethal in Podospora anserina. Genetics 155: 657–669. Lasure, L.S. (eds). San Diego: CA: Academic Press, pp. Debuchy, R., Berteaux-Lecellier, V., and Silar, P. (2010) 193–216. Mating systems and sexual morphogenesis in ascomy- Granger-Schnarr, M., Lloubes, R., de Murcia, G., and cetes. In Cellular and Molecular Biology of Filamentous Schnarr, M. (1988) Specific protein-DNA complexes: Fungi. Washington, DC: ASM Press, pp. 501–535. immunodetection of the protein component after gel elec- Demczuk, S., Harbers, M., and Vennstrom, B. (1993) Identi- trophoresis and Western-blotting. Anal Biochem 174: 235– fication and analysis of all components of a gel retardation 238. assay by combination with immunoblotting. Proc Natl Acad Grant, C.E., Bailey, T.L., and Noble, W.S. (2011) FIMO: scan- Sci USA 90: 2574–2578. ning for occurrences of a given motif. Bioinformatics 27: DeZwaan, T.M., Carroll, A.M., Valent, B., and Sweigard, J.A. 1017–1018. (1999) Magnaporthe grisea Pth11p is a novel plasma mem- Grünbacher, A., Throm, T., Seidel, C., Gutt, B., Rohrig, J., brane protein that mediates appressorium differentiation in Strunk, T., et al. (2014) Six hydrophobins are involved in response to inductive substrate cues. Plant Cell 11: 2013– hydrophobin rodlet formation in Aspergillus nidulans and 2030. contribute to hydrophobicity of the spore surface. PLoS Dmochowska, A., Dignard, D., Henning, D., Thomas, D.Y., ONE 9: e94546. and Bussey, H. (1987) Yeast KEX1 gene encodes a puta- Grzenda, A., Lomberk, G., Zhang, J.S., and Urrutia, R. (2009) tive protease with a carboxypeptidase B-like function Sin3: master scaffold and transcriptional corepressor. involved in killer toxin and alpha-factor precursor process- Biochim Biophys Acta 1789: 443–450. ing. Cell 50: 573–584. Gupta, S., Stamatoyannopoulos, J.A., Bailey, T.L., and Du, L.L., and Novick, P. (2002) Pag1p, a novel protein asso- Noble, W.S. (2007) Quantifying similarity between motifs. ciated with protein kinase Cbk1p, is required for cell mor- Genome Biol 8: R24. phogenesis and proliferation in Saccharomyces cerevisiae. Haber, J.E. (2012) Mating-type genes and MAT switching in Mol Biol Cell 13: 503–514. Saccharomyces cerevisiae. Genetics 191: 33–64. Dyer, P.S., and O’Gorman, C.M. (2012) Sexual development Han, K.H., Lee, D.B., Kim, J.H., Kim, M.S., Han, K.Y., and cryptic sexuality in fungi: insights from Aspergillus Kim, W.S., et al. (2003) Environmental factors affecting species. FEMS Microbiol Rev 36: 165–192. development of Aspergillus nidulans. J Microbiol 41: Eisendle, M., Oberegger, H., Zadra, I., and Haas, H. (2003) 34–40. The siderophore system is essential for viability of Asper- Heinz, S., Benner, C., Spann, N., Bertolino, E., Lin, Y.C., gillus nidulans: functional analysis of two genes encoding Laslo, P., et al. (2010) Simple combinations of lineage- l-ornithine N 5-monooxygenase (sidA) and a non- determining transcription factors prime cis-regulatory ele- ribosomal peptide synthetase (sidC). Mol Microbiol 49: ments required for macrophage and B cell identities. Mol 359–375. Cell 38: 576–589. Eisendle, M., Schrettl, M., Kragl, C., Müller, D., Illmer, P., and Herskowitz, I. (1989) A regulatory hierarchy for cell speciali- Haas, H. (2006a) The intracellular siderophore ferricrocin zation in yeast. Nature 342: 749–757. is involved in iron storage, oxidative-stress resistance, ger- Hoff, B., Kamerewerd, J., Sigl, C., Zadra, I., and Kück, U. mination, and sexual development in Aspergillus nidulans. (2009) Homologous recombination in the antibiotic pro- Eukaryot Cell 5: 1596–1603. ducer Penicillium chrysogenum: strain ΔPcku70 shows Eisendle, M., Schrettl, M., Kragl, C., Müller, D., Illmer, P., and up-regulation of genes from the HOG pathway. Appl Micro- Haas, H. (2006b) The intracellular siderophore ferricrocin biol Biotechnol 85: 1081–1094. is involved in iron storage, oxidative-stress resistance, ger- Hoff, B., Kamerewerd, J., Sigl, C., Mitterbauer, R., Zadra, I., mination, and sexual development in Aspergillus nidulans. Kürnsteiner, H., and Kück, U. (2010) Two components Eukaryot Cell 5: 1596–1603. of a velvet-like complex control hyphal morphogenesis,

conidiophore development, and penicillin biosynthesis in Langmead, B., Trapnell, C., Pop, M., and Salzberg, S.L. Penicillium chrysogenum. Eukaryot Cell 9: 1236–1250. (2009) Ultrafast and memory-efficient alignment of short Irie, K., Nomoto, S., Miyajima, I., and Matsumoto, K. (1991) DNA sequences to the human genome. Genome Biol 10: SGV1 encodes a CDC28/cdc2-related kinase required for R25. a G alpha subunit-mediated adaptive response to phero- Lara-Rojas, F., Sanchez, O., Kawasaki, L., and Aguirre, J. mone in S. cerevisiae. Cell 65: 785–795. (2011) Aspergillus nidulans transcription factor AtfA inter- James, P., Halladay, J., and Craig, E.A. (1996) Genomic acts with the MAPK SakA to regulate general stress libraries and a host strain designed for highly efficient two- responses, development and spore functions. Mol Micro- hybrid selection in yeast. Genetics 144: 1425–1436. biol 80: 436–454. Janus, D., Hortschansky, P., and Kück, U. (2007) Identifica- Lee, S.C., Ni, M., Li, W., Shertz, C., and Heitman, J. (2010) tion of a minimal cre1 promoter sequence promoting The evolution of sex: a perspective from the fungal glucose-dependent gene expression in the β-lactam pro- kingdom. Microbiol Mol Biol Rev 74: 298–340. ducer Acremonium chrysogenum. Curr Genet 53: 35–48. Lee, S.H., Lee, S., Choi, D., Lee, Y.W., and Yun, S.H. (2006) Johnson, L. (2008) Iron and siderophores in fungal-host inter- Identification of the down-regulated genes in a mat1-2- actions. Mycol Res 112: 170–183. deleted strain of Gibberella zeae, using cDNA subtraction Julius, D., Brake, A., Blair, L., Kunisawa, R., and Thorner, J. and microarray analysis. Fungal Genet Biol 43: 295–310. (1984) Isolation of the putative structural gene for the Li, H., Handsaker, B., Wysoker, A., Fennell, T., Ruan, J., lysine-arginine-cleaving endopeptidase required for pro- Homer, N., et al. (2009) The sequence alignment/map cessing of yeast prepro-alpha-factor. Cell 37: 1075–1089. format and SAMtools. Bioinformatics 25: 2078–2079. Kamerewerd, J., Zadra, I., Kürnsteiner, H., and Kück, U. Magnúsdóttir, E., Dietmann, S., Murakami, K., Günesdogan, (2011) PcchiB1, encoding a class V chitinase, is affected by U., Tang, F.C., Bao, S.Q., et al. (2013) A tripartite transcrip- PcVelA and PcLaeA, and is responsible for cell wall integrity tion factor network regulates primordial germ cell specifi- in Penicillium chrysogenum. Microbiology 157: 3036–3048. cation in mice. Nat Cell Biol 15: 905–U322. Kanai, Y., Kanai-Azuma, M., Noce, T., Saido, T.C., Shiroishi, Martin, T., Lu, S.W., van Tilbeurgh, H., Ripoll, D.R., Dixelius, T., Hayashi, Y., and Yazaki, K. (1996) Identification of two C., Turgeon, B.G., and Debuchy, R. (2010) Tracing the Sox17 messenger RNA isoforms, with and without the high origin of the fungal alpha 1 domain places its ancestor in mobility group box region, and their differential expression the HMG-box superfamily: implication for fungal mating- in mouse spermatogenesis. J Cell Biol 133: 667–681. type evolution. PLoS ONE 5: e15199. Keszthelyi, A., Jeney, A., Kerényi, Z., Mendes, O., Waalwijk, Maruyama, M., Ichisaka, T., Nakagawa, M., and Yamanaka, C., and Hornok, L. (2007) Tagging target genes of the S. (2005) Differential roles for Sox15 and Sox2 in transcrip- MAT1-2-1 transcription factor in Fusarium verticillioides tional control in mouse embryonic stem cells. J Biol Chem (Gibberella fujikuroi MP-A). Antonie Van Leeuwenhoek 91: 280: 24371–24379. 373–391. Mead, J., Bruning, A.R., Gill, M.K., Steiner, A.M., Acton, T.B., Kim, H., and Borkovich, K.A. (2006) Pheromones are essen- and Vershon, A.K. (2002) Interactions of the Mcm1 MADS tial for male fertility and sufficient to direct chemotropic box protein with cofactors that regulate mating in yeast. polarized growth of trichogynes during mating in Neuros- Mol Cell Biol 22: 4607–4621. pora crassa. Eukaryot Cell 5: 544–554. Mead, M.E., Stanton, B.C., Kruzel, E.K., and Hull, C.M. Kopke, K., Hoff, B., Bloemendal, S., Katschorowski, A., (2015) Targets of the Sex Inducer homeodomain proteins Kamerewerd, J., and Kück, U. (2012) Members of the are required for fungal development and virulence in Cryp- Penicillium chrysogenum velvet complex play functionally tococcus neoformans. Mol Microbiol 95: 804–818. opposing roles in the regulation of penicillin biosynthesis Mertin, S., McDowall, S.G., and Harley, V.R. (1999) The and conidiation. Eukaryot Cell 12: 299–310. DNA-binding specificity of SOX9 and other SOX proteins. Kraus, P.R., Hofmann, A.F., and Harris, S.D. (2002) Charac- Nucleic Acids Res 27: 1359–1364. terization of the Aspergillus nidulans 14-3-3 homologue, Metzenberg, R.L., and Glass, N.L. (1990) Mating type and ArtA. FEMS Microbiol Lett 210: 61–66. mating strategies in Neurospora. Bioessays 12: 53–59. Krüger, A., Vowinckel, J., Mulleder, M., Grote, P., Capuano, Miroux, B., and Walker, J.E. (1996) Over-production of pro- F., Bluemlein, K., and Ralser, M. (2013) Tpo1-mediated teins in Escherichia coli: mutant hosts that allow synthesis spermine and spermidine export controls cell cycle delay of some membrane proteins and globular proteins at high and times antioxidant protein expression during the oxida- levels. J Mol Biol 260: 289–298. tive stress response. EMBO Rep 14: 1113–1119. Morita, T., Yamada, T., Yamada, S., Matsumoto, K., and Ohta, Kunitomo, H., Sugimoto, A., Wilkinson, C.R., and Yamamoto, K. (2011) Fission yeast ATF/CREB family protein Atf21 M. (1995) Schizosaccharomyces pombe pac2+ controls plays important roles in production of normal spores. the onset of sexual development via a pathway independ- Genes Cells 16: 217–230. ent of the cAMP cascade. Curr Genet 28: 32–38. Myers, K.S., Yan, H., Ong, I.M., Chung, D., Liang, K., Tran, F., Kück, U., and Böhm, J. (2013) Mating type genes and cryptic et al. (2013) Genome-scale analysis of Escherichia coli sexuality as tools for genetically manipulating industrial FNR reveals complex features of transcription factor molds. Appl Microbiol Biotechnol 97: 9609–9620. binding. PLoS Genet 9: e1003565. Landt, S.G., Marinov, G.K., Kundaje, A., Kheradpour, P., Nelson, B., Kurischko, C., Horecka, J., Mody, M., Nair, P., Pauli, F., Batzoglou, S., et al. (2012) ChIP-seq guidelines Pratt, L., et al. (2003) RAM: a conserved signaling network and practices of the ENCODE and modENCODE consor- that regulates Ace2p transcription factor and polarized tia. Genome Res 22: 1813–1831. morphogenesis. Mol Biol Cell 14: 3782–3803.

Nielsen, J. (1997) Physiological Engineering Aspects of Peni- Schrettl, M., Bignell, E., Kragl, C., Sabiha, Y., Loss, O., cillium chrysogenum. Singapore: World Scientific Publish- Eisendle, M., et al. (2007) Distinct roles for intra- and extra- ing Co. Pte. Ltd. cellular siderophores during Aspergillus fumigatus infec- Paoletti, M., Seymour, F.A., Alcocer, M.J., Kaur, N., Calvo, tion. PLoS Pathog 3: 1195–1207. A.M., Archer, D.B., and Dyer, P.S. (2007) Mating type and Slaven, J.W., Anderson, M.J., Sanglard, D., Dixon, G.K., the genetic basis of self-fertility in the model fungus Asper- Bille, J., Roberts, I.S., and Denning, D.W. (2002) Increased gillus nidulans. Curr Biol 17: 1384–1389. expression of a novel Aspergillus fumigatus ABC trans- Park, P.J. (2009) ChIP-seq: advantages and challenges of a porter gene, atrF, in the presence of itraconazole in an maturing technology. Nat Rev Genet 10: 669–680. itraconazole resistant clinical isolate. Fungal Genet Biol 36: Penn, O., Privman, E., Ashkenazy, H., Landan, G., Graur, D., 199–206. and Pupko, T. (2010) GUIDANCE: a web server for assess- Smith, K.M., Sancar, G., Dekhang, R., Sullivan, C.M., Li, S.J., ing alignment confidence scores. Nucleic Acids Res 38: Tag, A.G., et al. (2010) Transcription factors in light and W23–W28. circadian clock signaling networks revealed by genom- Pöggeler, S., Nowrousian, M., Ringelberg, C., Loros, J.J., ewide mapping of direct targets for Neurospora white collar Dunlap, J.C., and Kück, U. (2006) Microarray and real-time complex. Eukaryot Cell 9: 1549–1556. PCR analyses reveal mating type-dependent gene expres- Smith, K.M., Phatale, P.A., Sullivan, C.M., Pomraning, K.R., sion in a homothallic fungus. Mol Genet Genomics 275: and Freitag, M. (2011) Heterochromatin is required for 492–503. normal distribution of Neurospora crassa CenH3. Mol Cell Pramila, T., Miles, S., GuhaThakurta, D., Jemiolo, D., and Biol 31: 2528–2542. Breeden, L.L. (2002) Conserved homeodomain proteins Specht, T., Dahlmann, T.A., Zadra, I., Kürnsteiner, H., and interact with MADS box protein Mcm1 to restrict ECB- Kück, U. (2014) Complete sequencing and chromosome- dependent transcription to the M/G1 phase of the cell scale genome assembly of the industrial progenitor strain cycle. Genes Dev 16: 3034–3045. P2niaD18 from the penicillin producer Penicillium chrys- Pramila, T., Wu, W., Miles, S., Noble, W.S., and Breeden, L.L. ogenum. Genome Announc 2: e00577-14. (2006) The forkhead transcription factor Hcm1 regulates Spitz, F., and Furlong, E.E. (2012) Transcription factors: from chromosome segregation genes and fills the S-phase gap enhancer binding to developmental control. Nat Rev Genet in the transcriptional circuitry of the cell cycle. Genes Dev 13: 613–626. 20: 2266–2278. Stringer, M.A., and Timberlake, W.E. (1995) DewA encodes a Rabinovich, A., Jin, V.X., Rabinovich, R., Xu, X., and fungal hydrophobin component of the Aspergillus spore Farnham, P.J. (2008) E2F in vivo binding specificity: com- wall. Mol Microbiol 16: 33–44. parison of consensus versus nonconsensus binding sites. Tamaru, H., Zhang, X., McMillen, D., Singh, P.B., Nakayama, Genome Res 18: 1763–1777. J., Grewal, S.I., et al. (2003) Trimethylated lysine 9 of Robertson, G., Hirst, M., Bainbridge, M., Bilenky, M., Zhao, histone H3 is a mark for DNA methylation in Neurospora Y., Zeng, T., et al. (2007) Genome-wide profiles of STAT1 crassa. Nat Genet 34: 75–79. DNA association using chromatin immunoprecipitation and Thorvaldsdóttir, H., Robinson, J.T., and Mesirov, J.P. (2012) massively parallel sequencing. Nat Methods 4: 651–657. Integrative Genomics Viewer (IGV): high-performance Ruepp, A., Zollner, A., Maier, D., Albermann, K., Hani, J., genomics data visualization and exploration. Brief Bioin- Mokrejs, M., et al. (2004) The FunCat, a functional anno- form 14: 178–192. tation scheme for systematic classification of proteins from Thön, M., Al-Abdallah, Q., Hortschansky, P., and Brakhage, whole genomes. Nucleic Acids Res 32: 5539–5545. A.A. (2007) The thioredoxin system of the filamentous Schmitt, E.K., and Kück, U. (2000) The fungal CPCR1 fungus Aspergillus nidulans: impact on development protein, which binds specifically to beta-lactam biosynthe- and oxidative stress response. J Biol Chem 282: 27259– sis genes, is related to human regulatory factor X transcrip- 27269. tion factors. J Biol Chem 275: 9348–9357. Todeschini, A.L., Georges, A., and Veitia, R.A. (2014) Tran- Schmitt, E.K., Bunse, A., Janus, D., Hoff, B., Friedlin, E., scription factors: specific DNA binding and specific gene Kürnsteiner, H., and Kück, U. (2004) Winged helix regulation. Trends Genet 30: 211–219. transcription factor CPCR1 is involved in regulation of beta- Tsong, A.E., Miller, M.G., Raisner, R.M., and Johnson, A.D. lactam biosynthesis in the fungus Acremonium chrysoge- (2003) Evolution of a combinatorial transcriptional circuit: a num. Eukaryot Cell 3: 121–134. case study in yeasts. Cell 115: 389–399. Schönig, B., Brown, D.W., Oeser, B., and Tudzynski, B. Tuch, B.B., Galgoczy, D.J., Hernday, A.D., Li, H., and (2008) Cross-species hybridization with Fusarium verticil- Johnson, A.D. (2008) The evolution of combinatorial gene lioides microarrays reveals new insights into Fusarium fuji- regulation in fungi. PLoS Biol 6: e38. kuroi nitrogen regulation and the role of AreA and NMR. Turgeon, B.G., and Yoder, O.C. (2000) Proposed nomencla- Eukaryot Cell 7: 1831–1846. ture for mating type genes of filamentous ascomycetes. Schrettl, M., and Haas, H. (2011) Iron homeostasis – Achilles’ Fungal Genet Biol 31: 1–5. heel of Aspergillus fumigatus? Curr Opin Microbiol 14: Valouev, A., Johnson, D.S., Sundquist, A., Medina, C., Anton, 400–405. E., Batzoglou, S., et al. (2008) Genome-wide analysis of Schrettl, M., Bignell, E., Kragl, C., Joechl, C., Rogers, T., Arst, transcription factor binding sites based on ChIP-Seq data. H., et al. (2004) Siderophore biosynthesis but not reductive Nat Methods 5: 829–834. iron assimilation is essential for Aspergillus fumigatus viru- Wada, R., Maruyama, J., Yamaguchi, H., Yamamoto, N., lence. J Exp Med 200: 1213–1219. Wagu, Y., Paoletti, M., et al. (2012) Presence and

functionality of mating type genes in the supposedly fujikuroi, affect differentiation, secondary metabolism and asexual ﬁlamentous fungus Aspergillus oryzae. Appl virulence. Mol Microbiol 77: 972–994. Environ Microbiol 78: 2819–2829. Wolfers, S., Kamerewerd, J., Nowrousian, M., Sigl, C., Zadra, Wang, Z., Lopez-Giraldez, F., Lehr, N., Farré, M., Common, I., Kürnsteiner, H., et al. (2014) Microarray hybridization R., Trail, F., and Townsend, J.P. (2014) Global gene analysis of light-dependent gene expression in Penicillium expression and focused knockout analysis reveals genes chrysogenum identiﬁes bZIP transcription factor PcAtfA. J associated with fungal fruiting body development in Neu- Basic Microbiol 54: 1–10. rospora crassa. Eukaryot Cell 13: 154–169. Wong, K.H., Hynes, M.J., Todd, R.B., and Davis, M.A. (2007) Wasserstrom, L., Lengeler, K.B., Walther, A., and Wendland, Transcriptional control of nmrA by the bZIP transcription J. (2013) Molecular determinants of sporulation in Ashbya factor MeaB reveals a new level of nitrogen regulation in gossypii. Genetics 195: 87–99. Aspergillus nidulans. Mol Microbiol 66: 534–551. Waterhouse, A.M., Procter, J.B., Martin, D.M., Clamp, M., Wunderlich, Z., and Mirny, L.A. (2009) Different gene regu- and Barton, G.J. (2009) Jalview Version 2-a multiple lation strategies revealed by analysis of binding motifs. sequence alignment editor and analysis workbench. Bioin- Trends Genet 25: 434–440. formatics 25: 1189–1191. Zaret, K.S., and Carroll, J.S. (2011) Pioneer transcription Wettenhall, J.M., Simpson, K.M., Satterley, K., and Smyth, factors: establishing competence for gene expression. G.K. (2006) affylmGUI: a graphical user interface for linear Genes Dev 25: 2227–2241. modeling of single channel microarray data. Bioinformatics 22: 897–899. Supporting information Wiemann, P., Brown, D.W., Kleigrewe, K., Bok, J.W., Keller, N.P., Humpf, H.U., and Tudzynski, B. (2010) FfVel1 and Additional supporting information may be found in the online FfLae1, components of a velvet-like complex in Fusarium version of this article at the publisher’s web-site.

© 2015 The Authors. Molecular Microbiology published by John Wiley & Sons Ltd, Molecular Microbiology Genome-wide identification of target genes of a mating-type α-domain transcription factor reveals functions beyond sexual development

- SUPPLEMENTARY MATERIAL -

Authors: Kordula Beckera, Christina Beera, Michael Freitagb, and Ulrich Kücka,1

Affiliations: aChristian Doppler Laboratory for „Fungal Biotechnology“, Lehrstuhl für Allgemeine und Molekulare Botanik, Ruhr-Universität Bochum, Universitätsstr. 150, 44780 Bochum, Germany bDepartment of Biochemistry and Biophysics, Oregon State University, Corvallis, Oregon 97331- 7305, USA

1To whom correspondence should be addressed: Ulrich Kück, Christian Doppler Laboratory for „Fungal Biotechnology“, Lehrstuhl für Allgemeine und Molekulare Botanik, Ruhr-Universität Bochum, Universitätsstr. 150, 44780 Bochum, Germany Tel.: +49 234 - 32 - 26212, [email protected]

Table S1: P. chrysogenum strains used in this work. Table S2: Plasmids used in this work. Table S3: Oligonucleotides used in this work.

Figure S1: Construction of P. chrysogenum strains used for ChIP analysis. Figure S2: Functional categorization of putative MAT1-1-1 target genes. Figure S3: Validation of non-mating related MAT1-1-1 target genes. Figure S4: Comparison of the predicted MAT1-1-1 DNA-binding motif MAT1.1 to known DNA- binding motifs. Figure S5: Alignment of MAT1-1-1 α-domain region amino acid sequences. Figure S6: Purification of recombinant GST-MAT1-1-1 from E. coli BL21 (DE3). Figure S7: Control experiments for yeast one-hybrid analyses. Figure S8: Genome-wide distribution of MAT1-1-1 binding sites.

Table S1: P. chrysogenum strains used in this work.

strain characteristics and genotype source P2niaD18 niaD- (Hoff et al., 2008) Δku70FRT2 ΔPcku70::FRT2; niaD− (Kopke et al., 2010)

OE MAT1-1-1 (T2) Pgpd::MAT1-1–1; ptrA; niaD− (Böhm et al., 2013) ΔMAT1 (EK6) MAT1-1-1Δ::ble; Pcku70Δ::nat1; niaD− (Böhm et al., 2013) − MAT1-ChIP (T28.8) Pgpd::EGFP::MAT1-1-1::TtrpC; nat1; niaD this study − T4.11 Pgpd::EGFP::MAT1-1-1::TtrpC; DsRed::TtrpC; nat1; ble; niaD this study − T2.7.1 Pppg1::DsRed::TtrpC; Pgpd::EGFP::MAT1-1-1::TtrpC; nat1; ble; niaD this study - T11.7 Pppg1::DsRed::TtrpC; ble; niaD this study − T3.15 Pkex1::DsRed::TtrpC; Pgpd::EGFP::MAT1-1-1::TtrpC; nat1; ble; niaD this study - T10.6 Pkex1::DsRed::TtrpC; ble; niaD this study T34.2 artAΔ::nat1; ΔPcku70::FRT2; niaD− this study T34.3 artAΔ::nat1; ΔPcku70::FRT2; niaD− this study T34.4 artAΔ::nat1; ΔPcku70::FRT2; niaD− this study

Table S2: Plasmids used in this work. name characteristics source pGEX-MAT1 MAT1-1-1 cDNA sequence of P. chrysogenum; used for heterologous this study expression of a GST-MAT1-1-1 fusion construct in E. coli BL21 (DE3)

pGFP-MAT1 Pgpd of A. nidulans, egfp, MAT1-1-1 gene of P. chrysogenum, TtrpC of this study A. nidulans, nat resistance gene of Streptomyces noursei; used for construction of P. chrysogenum ChIP-strains pLacZi prey vector for yeast one-hybrid analyses, lacZ reporter gene Clontech, (Luo et al., 1996)

pLacZ-kex1-2 triple repeat of Pkex1 fragment (chr1, 2536477-2536502) of P. chrysogenum this study integrated into MCS of pLacZi

pLacZ-ppg1-2 triple repeat of Pppg1 fragment (chr1: 278174-278202) of P. chrysogenum this study integrated into MCS of pLacZi

pLacZ-ppg1-2_m1 triple repeat of Pppg1 fragment (chr1: 278174-278202, AÆG at pos. 278184) this study of P. chrysogenum integrated into MCS of pLacZi

pLacZ-ppg1-2_m2 triple repeat of Pppg1 fragment (chr1: 278174-278202, AÆG at pos. 278184 this study and 278194) of P. chrysogenum integrated into MCS of pLacZi pHISi prey vector for yeast one-hybrid analyses, his3 reporter gene Clontech, (Alexandre et al., 1993)

pHIS-kex1-2 triple repeat of Pkex1 fragment (chr1, 2536477-2536502) of P. chrysogenum this study integrated into MCS of pHISi

pHIS-ppg1-2 triple repeat of Pppg1 fragment (chr1: 278174-278202) of P. chrysogenum this study integrated into MCS of pHISi

pHIS-ppg1-2_m1 triple repeat of Pppg1 fragment (chr1: 278174-278202, AÆG at pos. 278184) this study of P. chrysogenum integrated into MCS of pHISi

pHIS-ppg1-2_m2 triple repeat of Pppg1 fragment (chr1: 278174-278202, AÆG at pos. 278184 this study and 278194) of P. chrysogenum integrated into MCS of pHISi pGADT7 bait vector for yeast one-hybrid analyses, GAL4 transcription factor Clontech, (Chien et activation domain al., 1991) pMAT1-AD MAT1-1-1 cDNA sequence of P. chrysogenum integrated into MCS of this study pGADT7

pDsRed dsRed gene of Discosoma sp., TtrpC of A. nidulans this study pPkex1- Pkex1 (chr1:2536272-2537717) of P. chrysogenum, dsRed gene of Discosoma this study dsRed_ble sp., TtrpC of A. nidulans, ble resistance gene of Streptoalluteichus hindustanus pPppg1- Pppg1 (chr1:277853-278696) of P. chrysogenum, dsRed gene of Discosoma this study dsRed_ble sp., TtrpC of A. nidulans, ble resistance gene of Streptoalluteichus hindustanus pKO_artA 5’-flank Pc18g03940, nat resistance gene of Streptomyces noursei, 3’-flank this study Pc18g03940; used for construction of artA deletion strains

Table S3: Oligonucleotides used in this work. name sequence (5’ Æ 3’) specificity plasmid and MAT1-1-1_f TCTCGGCATGGACGAGCTGTACAAGATGTCTA Pc20g07800 gene; chr2:1892222-1892241 strain CCTCTCTTGATGC construction, MAT1-1-1_r CGTTAAGTGGATCCACTAGTTCTAGCTAGTTG Pc20g07800 gene; chr2:1891165-1891188 PCR analysis TGCCCAAAGATCCGGTC MAT1_SmaI_f CCCGGGATGTCTACCTCTCTTGATGC Pc20g07800 gene; chr2:1892222-1892241 MAT1_SmaI_r CCCGGGCTAGTTGTGCCCAAAGATCC Pc20g07800 gene; chr2:1891165-1891184 egfp_f GGTGAACTTCAAGATCCG egfp gene MAT1_r CTAGTTGTGCCCAAAGATCCGGTC Pc20g07800 gene; chr2:1891165-1891188 Pkex1_f ATTCTCGAGCGCTAATCACGGTATATCTG chr1:2537698-2537717 Pkex1_r GACTGAGCTCGATTGGAACTCGCGCTTTG chr1:2536272-2536290 Pppg1_f ATTGGGGCCCTGTGCCGTTCTGGAGAAC chr1:278679-278697 Pppg1_r GATTAAGCTTCTTGATGGTTGGAGGAGAAAC chr1:277852-277873 5’artA_f* GTAACGCCAGGGTTTTCCCAGTCACGACGGAA chr1:6422911-6422928 TTCAGGCGCAAGTAAGGAACC 5’artA_r* ATGCTCCTTCAATATCAGTTGAATTGAGGTGA chr1:6423803-6423821 GGGTTGAAAATC 3’artA_f* TGAGCATGCCCTGCCCCTGAGGGCCGACCTTG chr1:6424895-6424913 TTGATGGACACG 3’artA_r* GCGGATAACAATTTCACACAGGAAACAGCGA chr1:6425858-6425876 ATTCTACGCAGTATACCTTTCGC ChIP-PCR qPCR_Pex20_f CCCCATGAACAATAGCAATCACGT chr2:3627752-3627775 qPCR_Pex20_r AGACAGCCAATCAGAGACCGT chr2:3627840-3627860 qPCR_Pre1_f TCCATCGAAAATATGAGGCGGATG chr1:3250668-3250691 qPCR_Pre1_r TGAGCGAATCGTGTGAGGCA chr1:3250591-3250610 qPCR_Kex1_f CCAACTCCACTGGCCATGTCT chr1:2536505-2536525 qPCR_Kex1_r GACATGGTGGCGTTATTGAGCT chr1:2536418-2536439 qPCR_Kex2_f GAGGATCGTCAGAGGCCACC chr3:4731876-4731895 qPCR_Kex2_r TGGAGTAGAAACCGTGGCCTT chr3:4731747-4731767 qPCR_Ppg1_f GTGTTCTTTGTCGACATCAGCCT chr1:278126-278148 qPCR_Ppg1_r CGACAGGATGGCTTGCCCTA chr1:278253-278272 qPCR_NC1_f TTCTTCCGCAATCAAGCTCA chr1:5375234-5375253 qPCR_NC1_r GAAAAATTGCCGCTGGACTC chr1:5375364-5375383 qPCR_NC2_f GGTCGTTGATTCCCTTGAGC chr2:7621179-7621198 qPCR_NC2_r GGATCGGATTATTCGGGTGA chr2:7621294-7621313 qRT-PCR Pc20g00090_f CTGTCATCATCGCTGCGCTG Pc20g00090 gene; chr2:3628372-3628391 Pc20g00090_r GCTTGCGACCGTTGCTTTCT Pc20g00090 gene; chr2:3628547-3628566 Pex20_f TGGCTCAACAACAAGGGCCT Pc20g00100 gene; chr2:3626658-3626677 Pex20_r GGATTCGTCGAACTGCTGCG Pc20g00100 gene; chr2:3626811-3626830 Pc22g15640_f TTGTCGTTCATCAAGCCCGC Pc22g15640 gene; chr1:3251329-3251348 Pc22g15640_r AGACGGACCTGGCGTTCAAT Pc22g15640 gene; chr1:3251116-3251135 Pre1_f TGGGACACTGCTGGATGATCT Pc22g15650 gene; chr1:3248820-3248840 Pre1_r GCTAATAACCTGCCGCACATG Pc22g15650 gene; chr1:3248690-3248710 Sok1_f ACAAAAGAAGCCCGGACCCA Pc22g18590 gene; chr1:2537733-2537752 Sok1_r AGGCGAAGGGTTGTTGACGA Pc22g18590 gene; chr1:2537882-2537901 Kex1_f GGCTTCAACGACGTGCTAGC Pc22g18600 gene; chr1:2535442-2535461 Kex1_r CTTCTTCCGGGGTTGTGGGT Pc22g18600 gene; chr1:2535308-2535327 Kex2_f CAATACGTTGCAGCCCGGAC Pc22g02910 gene; chr3:4732652-4732671 Kex2_r TCCATGTCCAGCCCATCGTC Pc22g02910 gene; chr3:4732740-4732759 Pc22g02930_f GCACGTGAAGCACCAACACA Pc22g02930 gene; chr3:4729905-4729924 Pc22g02930_r CCGTGCCTTCTGATTGTCGC Pc22g02930 gene; chr3:4729771-4729790 Ppg1_f GCTTGCCCCTTGTCCTTCAGA Pc14g01160 gene; chr1:277648-277668 Ppg1_r CGCTGGTACGCTTGACCTCA Pc14g01160 gene; chr1:277567-277586 Pc14g01170_f CGCCAGAACCTTTGCCAGTG Pc14g01170 gene; chr1:278866-278885 Pc14g01170_r AGCACCGATACCGTCACCAG Pc14g01170 gene; chr1:279084-279104 DewA_f GGAGGCCTTCTGAACGGTGT chr4:861331-861350 DewA_r CGGTGCAAGTTGTGGTTCCC chr4:861596-861615 SidD_f GCTGGAAGTTCTGGATGCGC chr1:2128333-2128352 SidD_r TCCTTGCGGCCAACAAACAC chr1:2128417-2128436 Sin3_f GAACAGGCCGAGAAGTACGG chr3:712273-712292 Sin3_r GCAGTTCAGCCACGTCAAAG chr3:712353-712372

Pc22g27040_f GAATTCACACTGGCCAGCGG chr1:586115-586134 Pc22g27040_r GGAGGGTGAGAGCGGTGTTT chr1:586259-586278 Pc22g22160_f CGTGGTGAGAGTCTGGTGCA chr1:1697134-1697153 Pc22g22160_r TTCCCGAAAGCCCAAACCCT chr1:1696969-1696988 ArtA_f TACCACCGCTACCTTGCTGA chr1:6424345-6424364 ArtA_r CAGTGGAGGCGATCTCAGTG chr1:6424428-6424447 Atf21_f CCAGCGAATCAACCAGCAGC chr1:631651-631670 Atf21_r CGCACCATCTGAGACCGACT chr1:631523-631542 Pc19g00140_f GCCTGCGGTTCTCACATTGG chr2:4622762-4622781 Pc19g00140_r TCTCTGGCAGTCAATCCCCG chr2:4622615-4622634 Pc22g18630_f TGTGATCCAGGTCGACGAGC chr1:2530647-2530666 Pc22g18630_r GCGGTGGAGAGCTTGAAGGA chr1:2530738-2530757 FetC_f CCGAAGCAGATCCAGGAGCA chr2:6607858-6607877 FetC_r AAGCTGCTTGTTTTGGCCGG chr2:6607740-6607759 FtrA_f ACAACACCTGGAACCACGCT chr2:6610998-6611017 FtrA_r GAGCGCGTTGAAGATTCCCC chr2:6611127-6611146 MeaB_f CCCAACCCCGGACTTTCAGT chr1:5714759-5714778 MeaB_r CGAGGACCCATAGCTCCACC chr1:5714673-5714692 EMSAs and Pc20g00090-1 CATGAACAATAGCAATCACGTGATCTCTA chr2:3627755-3627783 Y1H** Pc20g00090-2 TCACGTGATCTCTATTGAGAACAATAGAA chr2:3627770-3627798 Pc20g00090-3 TGAGAACAATAGAAGTCCATTCAAGGATC chr2:3627785-3627813 Pre1-1 TCAATAGACTAGAAAGTCTAGATCAATAA chr1, 3250705-3250733 Pre1-2 AGATCAATAATAAACTCATTCATTCCATC chr1, 3250686-3250714 Pre1-3 ACTCATTCATTCCATCGAAAATATGAGGC chr1, 3250673-3250701 Kex1-1 ACTCCACTGGCCATGTCTTTGGCCACAAT chr1, 2536494-2536522 Kex1-2 GGCCACAATAACCCCACCGGCCTTATTGA chr1, 2536474-2536502 Kex1-3 CCTTATTGACACCCAAATCTGGCTCAACA chr1, 2536454-2536482 Kex2-1 TCCTATTGAGTCTCCTAAGAGGTCTATTG chr3, 4731883-4731866 Kex2-2 GGTCTATTGAGCTCAACTTAGCTTATTCA chr3, 4731818-4731846 Kex2-3 GCTTATTCAAACAGAAGCTAATTCCTTTG chr3, 4731798-4731826 Ppg1-1 CTCGAGATCGGCAGTTCTCAATAGGAATC chr1,278163-278191 Ppg1-2 CAGTTCTCAATAGGAATCTTATTGACCGA chr1, 278174-278202 Ppg1-3 ATAGGAATCTTATTGACCGACGTCAGTGT chr1,278183-278211 Ppg1-2_m1 CAGTTCTCAACAGGAATCTTATTGACCGA chr1, 278174-278202, AÆG at 278184 Ppg1-2_m2 CAGTTCTCAACAGGAATCTTGTTGACCGA chr1, 278174-278202, AÆG at 278184 and 278194 DewA-1 CAGTAGGCATTCTCAATAATCAAAGCGTC chr4:860854-860882 DewA-2 ATTGTCTGATTGATTCTCATTGACTTGAG chr4:859279-859307 SidD-1 GGAAATACCGAGTATTGATACCACGGTAT chr1:2122707-2122735 Sin3-1 TCGCCTGTCCACTCAATGCCAGTCATGTT chr3:708154-708182 Pc22g27040-1 CCAAGATGAGGTTATTGAGGCATTTCTTT chr1:590547-590575 Pc22g22160-1 TGCAAAGCTTGCAATTGAGTGACGCAAAG chr1:1699087-1699115 ArtA-1 GCAGGACGGGCAAATTGACGAAGCATGAT chr1:6422421-6422449 Atf21-1 CATACCGGAGGATTCATTGATCCTATGCA chr1:632283-632311 Pc19g00140-1 TATCCTGACTGTTCTCAATTGCAACCTTG chr2:4623848-4623876 Pc22g18630-1 GCCTTATCGGTGTCTCAATGGCGTGATCA chr1:2528324-2528352 FetC/FtrA-1 CAAAAGGCTAAGCTCAATACGAGTGGGTC chr2:6609927-6609955 MeaB-1 TAGGTATTCAAGTATTGACTCGGTATTGG chr1:5716368-5716396 *5‘-overhang for homologous recombination in yeast is marked by underlining **in case of double-stranded oligonucleotides only the sense sequences are given.

Figure S1: Construction of P. chrysogenum strains used for ChIP analysis. (A) A Pgpd::egfp::MAT1-1-1 fusion construct was used for the construction of MAT1-1-1 ChIP-strains. Plasmid pGFP-MAT1 was integrated ectopically into P. chrysogenum P2niaD18. (B) RMAexpress (http://rmaexpress.bmbolstad.com/) analysis of normalized raw data from microarray analysis using P. chrysogenum P2niaD18 after 48 h of cultivation was performed to obtain relative MAT1-1-1 expression levels. Actin (Pc20g11600) and myosin (Pc21g00710) relative expression levels are shown as a reference. (C) Integration of Pgpd::egfp::MAT1-1-1 was confirmed using PCR. Binding positions of primers egfp_f and MAT1_r are indicated as arrows in (A). NC = water control. (D) Presence of the epitope-tagged protein EGFP-MAT1-1-1 in crude protein extract from recombinant P. chrysogenum strains was confirmed using SDS-Page/Western blot analysis prior to ChIP-experiments. (E) Fluorescence microscopy confirmed nuclear localization of EGFP-MAT1-1-1 in the MAT1-ChIP strain. Strains were grown on solid medium for 48 h. Scale bar = 10 μm. (F) Pellet formation in MAT1-ChIP was analyzed in comparison to a previously described MAT1-1-1 overexpression strain (OE MAT1-1-1) (Böhm et al., 2013) and the parental strain P2niaD18 in shaking cultures after 72 h of cultivation. Scale bar = 5 mm.

Figure S2: Functional categorization of putative MAT1-1-1 target genes. (A) Illustration of functional categories assigned to proteins encoded by putative MAT1-1-1 target genes. For each functional category, the fold representation among all predicted proteins is given. Statistically overrepresented functional groups (p-value ≤ 0.05) are indicated by an asterisk. (B) For each functional category, the absolute and relative number of proteins from our dataset (abs SET, rel SET) and the reference genome (abs GENOME, rel GENOME), the fold representation among all predicted proteins (rel SET/rel GENOME), and the corresponding p-value are indicated.

Figure S3: Validation of non-mating related MAT1-1-1 target genes. (A) Analysis of relative log2fold gene expression ratios of selected MAT1-1-1 target genes, identified in ChIP-seq analyses, in a MAT1-1-1 overexpression strain (MAT1-ChIP; grey bars) or MAT1-1-1 deletion strain (ΔMAT1; black bars) compared to wild type strain P2niaD18. Values are the mean score of two biological replicates (B) EMSAs were performed using radiolabeled double stranded oligonucleotide probes covering the central region of selected MAT1-1-1 target regions, identified in ChIP-seq analysis. Addition of GST-MAT1-1-1 protein is marked by (+), samples without protein are marked by (-). Positions of free probe (*) and protein-DNA complexes (Æ) are indicated.

Figure S4: Comparison of the predicted MAT1-1-1 DNA-binding motif MAT1.1 to known DNA- binding motifs. MAT1.1 was submitted to TOMTOM for comparison with the JASPAR CORE (2014) databases for fungi and vertebrates. For each database, the top four hits, showing similarities to the predicted MAT1-1-1 DNA-binding sequence, are presented. The associated proteins, IDs from the JASPAR CORE database, p-values and E-values are indicated. The size of each letter is proportional to the frequency of each nucleotide at this position within the consensus sequence. Motifs are centered on common central nucleotide.

Figure S5: Alignment of MAT1-1-1 α-domain region amino acid sequences. Accession numbers for MAT1-1-1 proteins according to UniProt database (http://www.uniprot.org/): A. fumigatus (Q4G285), A. nidulans (G5EAT5), C. albicans (Q9UW19), F. graminearum (I1RX43), N. crassa (P19392), P. chrysogenum (B6HEL2), S. cerevisiae (P0CY06), T. reesei (C7SQ97. Localization of the MAT_alpha1 (pfam04769), MATA_HMG-box (cd01389), and Y-[LM]-x(3)-G-[WL] (Martin et al., 2010) domains are indicated by red, green, and blue labels. Alignments were visualized using Jalview according to the Clustalx colour scheme (http://www.jalview.org/).

Figure S6: Purification of recombinant GST-MAT1-1-1 from E. coli BL21 (DE3). Purified protein was analyzed using SDS-PAGE/Western blot analysis using an antibody to GST prior to DNA-binding experiments.

Figure S7: Control experiments for yeast one-hybrid analyses. Control experiments were performed to analyze trans-activation between prey constructs and the empty bait vector and vice versa. All strains used for reporter gene assays were grown on SD-ura-leu to confirm presence of both, a bait and a prey vector, in diploid yeast strains. HIS3 reporter gene activity was analyzed using growth tests on SD-ura-leu-his supplemented with 3-AT as indicated. lacZ reporter gene activity was analyzed using qualitative and quantitative ß-galactosidase assays.

Figure S8: Genome-wide distribution of MAT1-1-1 binding sites. The tracks are (from top): “shaking 1”, “shaking 2”, “surface”, “shaking IP (input)”, and “surface IP (input)” for each of the four P. chrysogenum P2niaD18 chromosomes. Selected peaks upstream of genes mentioned within this study are labeled with the name of the corresponding gene.

REFERENCES

Alexandre, C., Grueneberg, D. A., Gilman, M. Z. (1993) Studying heterologous transcription factors in yeast. METHODS: A Companion to Methods in Enzymology 5: 147–155. Böhm, J., Hoff, B., O'Gorman, C. M., Wolfers, S., Klix, V., Binger, D., et al. (2013) Sexual reproduction and mating-type-mediated strain development in the penicillin-producing fungus Penicillium chrysogenum. Proc Natl Acad Sci U S A 110: 1476-1481. Chien, C. T., Bartel, P. L., Sternglanz, R., Fields, S. (1991) The two-hybrid system: a method to identify and clone genes for proteins that interact with a protein of interest. Proc Natl Acad Sci U S A 88: 9578-9582. Hoff, B., Pöggeler, S., Kück, U. (2008) Eighty years after its discovery, Fleming's Penicillium strain discloses the secret of its sex. Eukaryot Cell 7: 465-470. Kopke, K., Hoff, B., Kück, U. (2010) Application of the Saccharomyces cerevisiae FLP/FRT recombination system in filamentous fungi for marker recycling and construction of knockout strains devoid of heterologous genes. Appl Environ Microbiol 76: 4664-4674. Luo, Y., Vijaychander, S., Stile, J., Zhu, L. (1996) Cloning and analysis of DNA-binding proteins by yeast one-hybrid and one-two-hybrid systems. Biotechniques 20: 564-568. Martin, T., Lu, S. W., van Tilbeurgh, H., Ripoll, D. R., Dixelius, C., Turgeon, B. G., Debuchy, R. (2010) Tracing the origin of the fungal alpha 1 domain places its ancestor in the HMG-box superfamily: implication for fungal mating-type evolution. PLoS One 5: e15199.

IV. BECKER et al. 2015b 24

IV. BECKER et al. 2015b

New insights into PcVelA regulatory functions on a genome-wide scale reveal evidence for methyltransferase PcLlmA acting as a downstream factor and direct interaction partner of PcVelA in Penicillium chrysogenum

Kordula Becker, Sandra Bloemendal, and Ulrich Kück (2015)

– prepared for submission –

Authors: Kordula Beckera, Sandra Bloemendala, and Ulrich Kücka,1

Affiliations: aLehrstuhl für Allgemeine und Molekulare Botanik, Ruhr-Universität Bochum, Universitätsstr. 150, D–44780 Bochum, Germany

1To whom correspondence should be addressed:

Ulrich Kück, Lehrstuhl für Allgemeine und Molekulare Botanik, Ruhr-Universität Bochum, Universitätsstr. 150, 44780 Bochum, Germany Tel.: +49 234 - 32 - 26212, [email protected]

Short title: Target genes of PcVelA

Key words: Penicillium chrysogenum, velvet complex, PcVelA, PcLaeA, PcLlmA, ChIP-sequencing, protein-DNA interaction, methyltransferase

ABSTRACT

Penicillium chrysogenum is the only industrial producer of the β-lactam antibiotic penicillin, the most commonly used drug in the treatment of bacterial infections. In P. chrysogenum and other fungi, secondary metabolism and morphogenesis were shown to be controlled via velvet, a highly conserved multi-subunit protein complex. However, only little is known about how velvet complex-mediated regulation is exerted on a molecular level. To address this question, we performed chromatin immunoprecipitation combined with next-generation sequencing (ChIP-seq) analysis of PcVelA, the founding member and one of the core components of the velvet complex. We present a genome-wide DNA binding profile and DNA-binding motif of PcVelA, providing the first experimental evidence for PcVelA acting as a direct transcriptional regulator on DNA level, possibly even as a transcription factor. Based on ChIP-seq data, we identified a total of 592 PcVelA-specific DNA-binding regions and 631 putative direct target genes. Besides a remarkable number of genes related to known PcVelA regulatory functions, e.g. in terms of conidiation, we also identified at least seven PcVelA-specific target genes coding for putative methyltransferases. Furthermore, by using yeast two-hybrid analysis and bimolecular fluorescence complementation (BiFC), one of the encoded putative methyltransferases, PcLlmA, was shown to interact with PcVelA.

INTRODUCTION

The discovery of β-lactam antibiotics has been described as one of the most significant milestones of the human history, and entailed a revolution in modern chemotherapy (Barreiro et al. 2012). Penicillin, the most commonly used drug within this group of antibiotics, is produced by the filamentous ascomycete P. chrysogenum, which was firstly described in 1928 (Fleming 1929). Since then, huge efforts have been made in order to maximize penicillin yields in large-scale industrial production. For long, strain improvement programs were based on random mutagenesis approaches, such as treatment with X-ray, ultraviolet irradiation, and nitrogen mustard mutagenesis (Backus and Stauffer 1955, Peñalva et al. 1998, Barreiro et al. 2012). One of the main drawbacks of random mutagenesis is that it introduces both intended and unintended mutations, and large-scale screening processes are necessary to identify those mutant strains displaying the desired characteristics. As a consequence, one main goal of modern strain-improvement programs is to replace random mutagenesis by targeted genetic engineering approaches in order to fasten and simplify the generation of new strains with optimized production properties. Hence, detailed knowledge of determinants regulating morphology and secondary metabolism in P. chrysogenum is crucial for further optimization of this industrially highly relevant organism.

Secondary metabolism and differentiation processes in various filamentous fungi are orchestrated by the multi-subunit velvet complex (Bayram et al. 2008, Hoff et al. 2010, Wiemann et al. 2010). The founding member of the complex, VeA (Velvet A), was firstly described as a light-dependent regulator in Aspergillus nidulans (Käfer 1965). Since then, characterization of veA deletion and overexpression mutants confirmed its regulatory functions in terms of sexual and asexual development, morphogenesis, virulence, and secondary metabolism in numerous species (Kim et al. 2002, Kato et al. 2003, Dreyer et al. 2007, Hoff et al. 2010, Wiemann et al. 2010, Merhej et al. 2012). In P. chrysogenum, deletion of PcvelA was shown to result in reduced production of penicillin, together with light- independent formation of conidiospores, dichotomous branching of hyphae, and increased pellet formation in shaking cultures (Hoff et al. 2010). Besides PcVelA, members of the velvet family of proteins in P. chrysogenum include PcVelB, PcVelC, and PcVosA (Kopke et al. 2013). Furthermore, the putative S-adenosyl-L-methionine (SAM)-dependent methyltransferase LaeA (loss of aflR expression A), which functions as a global regulator of secondary metabolism and development in a huge number of euascomycetes (Bok and Keller 2004, Hoff et al. 2010, Sarikaya Bayram et al. 2010, Wiemann et al. 2010), has been shown to

3 be part of the fungal velvet complex. According to our current working model, all velvet subunits, including PcLaeA, are able to interact with one or more other subunits (Hoff et al. 2010, Kopke et al. 2013). Using a comprehensive set of single and double deletion mutants, it has been shown that PcVelA, together with PcLaeA and PcVelC, is an activator of penicillin biosynthesis, whereas PcVelB represses this process. Moreover, PcVelB and PcVosA were identified as promoters of conidiation, while PcVelC has an inhibitory effect (Hoff et al. 2010, Kopke et al. 2013). In contrast to A. nidulans, it has not been solved if velvet sub-complexes are formed at distinct time points or as a function of developmental stages in P. chrysogenum.

Velvet proteins are characterized by the so-called velvet domain, a conserved protein domain that can be found in most fungi (Kopke et al. 2013, Gerke and Braus 2014). While velvet regulatory functions were for long thought to be solely mediated by the putative methyltransferase LaeA, Ni and Yu hypothesized that the velvet proteins might be acting as global transcriptional regulators, representing a new fungus-specific class of transcription factors (TFs) (Ni and Yu 2007). Further evidence for this assumption was provided by microarray analyses, revealing that the expression of 13.6 % of all nuclear genes in P. chrysogenum is influenced by PcVelA (Hoff et al. 2010). Accordingly, RNA-seq analyses in A. fumigatus and A. nidulans revealed that a total of 32 % and 26 % of all protein coding genes are differentially regulated in a ∆veA strain compared to wild type (Lind et al. 2015). The first experimental evidence for velvet proteins acting as regulators on DNA level was provided in 2013. Ahmed et al. (2013) not only showed that the velvet domain is involved in the dimerization of different velvet proteins, leading to the formation of different homo- and heterodimers, but also demonstrated that the velvet domains of A. nidulans VosA and the VosA-VelB heterodimer are acting as DNA-binding domains. Besides the finding that velvet proteins are able to bind DNA in a sequence-specific manner, another interesting new feature of velvet-protein properties deals with the observation that several putative methyltransferases, others than LaeA, are able to directly interact with VeA on the protein level. For example, a reverse genetics screen in A. nidulans identified LlmF (LaeA-like methyltransferase F) as a direct interaction partner of VeA and a negative regulator of sterigmatocystin production and sexual development. While over-expression of llmF was shown to decrease the nuclear to cytoplasmic ratio of VeA, deletion of llmF resulted in an increased nuclear accumulation of the protein (Palmer et al. 2013). Furthermore, A. nidulans VipC (velvet interacting protein C, also referred to as LlmB) and VapB (VipC associated protein B), both putative methyltransferases and interaction partners of VeA, were shown to act in the nucleus to promote asexual development or, together with the membrane protein 4

VapA, at the plasma membrane to support sexual development (Sarikaya-Bayram et al. 2014). Finally, using a yeast two-hybrid approach (Y2H), Fusarium graminearum FgVeA was shown to directly interact with a total of six putative methyltransferases that show sequence homologies to FgLaeA1 (Jiang et al. 2011).

A general understanding of PcVelA regulatory functions on a genome-wide scale as well as the identification and characterization of additional interaction partners and downstream factors is crucial for further optimization of P. chrysogenum high production strains. Although huge efforts have been made in order to decipher the molecular mechanisms that control velvet protein-mediated regulation in various fungi, they are still poorly understood. Here, we present the first application of chromatin immunoprecipitation combined with next- generation sequencing (ChIP-seq) for the functional characterization of PcVelA, one of the core components of the velvet complex from P. chrysogenum. Most importantly, we provide evidence for PcVelA acting as a direct regulator of transcription on DNA-level and introduce the putative methyltransferase PcLlmA as a new downstream factor and interaction partner of PcVelA.

RESULTS

Generation of a genome-wide PcVelA DNA-binding profile. For use in ChIP-seq experiments, a Pgpd::PcvelA::egfp fusion construct was ectopically integrated into P. chrysogenum ΔPcvelA, a marker-free PcvelA deletion strain (Fig. 1A). Successful transformation and expression of PcvelA::egfp in strain PcVelA-ChIP was verified using PCR and SDS-PAGE/Western blot analysis (Fig. 1B-C). Furthermore, fluorescence microscopy confirmed presence and nuclear localization of PcVelA-EGFP prior to ChIP-experiments (Fig. 1D). ChIP-seq experiments were performed on two independent biological samples obtained from shaking cultures, designated “PcVelA_shaking_1” and “PcVelA_shaking_2”. As a control, input-DNA (“PcVelA_shaking_input”; DNA sample removed prior to ChIP) was sequenced in parallel. During bioinformatics analysis, only those regions accomplishing the following criteria were regarded as specific PcVelA binding regions: [1] at least fourfold enrichment in ChIP-DNA versus input-DNA, [2] a false discovery rate (FDR) threshold ≤ 0.001, and [3] a Poisson p-value ≤ 1.00e-04. In total, we identified 764 and 1,001 regions to be specifically bound to PcVelA in the “PcVelA_shaking1” and “PcVelA_shaking2” dataset, respectively (see Table 1). Intersection of both datasets, regarding peaks within a maximum distance of 100 nt as overlapping, identified 592 sites that were specifically bound by PcVelA in both biological replicates (Fig. 2A and Dataset S1), thus meeting the standards set by the ENCODE and modENCODE consortia (Landt et al. 2012).

As part of our initial analysis, peak regions were classified according to their genomic location with regard to neighboring open reading frames (ORFs). 78.9 % (467/592) of peaks showed intergenic localization and 21.1 % (125/592) were positioned within intragenic regions (Fig. 2B). Of 467 peaks showing intergenic localization, 39 were positioned within the 3’-region of both adjacent ORFs, 225 showed 5’-localization to one neighboring gene, and 203 peak regions were positioned within divergent promoter regions, resulting in a total of 631 genes that may be directly controlled by PcVelA. Previous microarray analyses (Hoff et al. 2010) confirmed PcVelA-dependent changes in expression levels by at least twofold for 18.9 % (119/631) of these genes. Furthermore, analysis of the distance between peak summits and the predicted translation start site (ATG) of those genes showing 5’-3’ orientation with regard to neighboring peaks revealed an average distance within a range of 100-600 nt (Fig. 2C). This observation is in accordance with the general understanding that most regulatory DNA sequences for a given gene fall within a few hundred bp from its transcription start site (TSS) in S. cerevisiae (Nguyen and D'Haeseleer 2006, Lin et al. 2010).

Validation of PcVelA ChIP-seq data. In order to validate the biological significance of our dataset and to rule out bias from bioinformatics analysis, PcVelA-specific enrichment of four selected target regions identified in ChIP-seq analyses was confirmed using quantitative ChIP-PCR. Target regions were selected to cover a range from high-affinity to mid-affinity PcVelA binding sites, as deduced from ChIP data. Based on data obtained from ChIP-PCR analysis, the ratio of the region of interest to a control region showing no PcVelA-specific enrichment in ChIP-DNA relative to this ratio in input-DNA was calculated. Another region, showing no enrichment in ChIP-seq data, was analyzed as a negative control (NC). Data from ChIP-PCR analysis was compared to peak values obtained from bioinformatics analysis, representing the average number of sequence tags found within a peak region after normalization to a total of 10 million mapped tags. As shown in Fig. 3, ChIP-PCR results were consistent with peak values, thus confirming specific enrichment of all tested PcVelA target regions, and validating peak values as a convincing parameter for estimation of PcVelA binding affinity to target regions identified in ChIP-seq analyses.

Categorization of putative PcVelA target genes. Screening of our ChIP-seq dataset identified a remarkable number of putative PcVelA target genes, directly related to cellular and developmental processes known to be under velvet-mediated control (Table 2). Most of these genes exhibited PcVelA-dependent expression profiles in previous microarray analyses, when expression levels in a PcvelA deletion strain were compared to the corresponding wild type strain ΔPcku70 (Hoff et al. 2010). Prominent examples for direct PcVelA target genes include con-6 (Pc16g03240), flbC (Pc12g12190), flbD (Pc13g03170), artA (Pc18g03940), and brlA (Pc23g00400), all related to different aspects of conidiation (Adams et al. 1988, Kraus et al. 2002, Kwon et al. 2010, Olmedo et al. 2010, Arratia-Quijada et al. 2012). Interestingly, a PcVelA DNA-binding region was also identified within the upstream region of PcvelB, encoding another component of the velvet complex that acts as an activator of conidiospore formation in various filamentous fungi (Bayram et al. 2008, Wiemann et al. 2010, Kopke et al. 2013). Putative targets assigned to functions related to spore viability and protection included treA/ath1 (Pc16g11870), coding for an α,α-trehalose glucohydrolase, Pc16g06690, encoding a precursor of the spore-wall fungal hydrophobin DewA, and Pc13g09910, coding for a late embryogenesis abundant (LEA) domain protein, known to protect other proteins from aggregation due to osmotic stresses associated with low temperatures in plants and animals (Stringer and Timberlake 1995, d'Enfert and Fontaine 1997, Goyal et al. 2005). Furthermore, several genes encoding proteins related to various aspects of secondary metabolism have been identified, such as Pc21g08920, coding for a norsolorinic acid 7 reductase, Pc21g12630, coding for a non-ribosomal peptide synthetase, and stuA (Pc13g04920), encoding a basic helix-loop-helix (bHLH) domain TF. Remarkably, StuA was not only shown to be involved in regulation of penicillin biosynthesis in P. chrysogenum (Sigl et al. 2011), but also in regulation of asexual reproduction, especially conidiophore development in A. nidulans (Miller et al. 1992). Besides this selection of somehow obvious PcVelA target genes, ChIP-seq analysis also identified a remarkable number of target genes never related to PcVelA or any other component of the velvet complex before. Among these genes, we identified five genes encoding proteins with Acetyl-CoA/Acyl-CoA-related functions, numerous genes coding for uncharacterized TFs (only those showing PcVelA-dependent expression in microarray analysis are shown in Table 2), as well as seven genes encoding putative methyltransferases. As recent reports indicated a close link between VeA and various putative methyltransferases in A. nidulans and F. graminearum (Jiang et al. 2011, Palmer et al. 2013, Sarikaya-Bayram et al. 2014), we decided to dedicate follow-on analyses to this group of new PcVelA target genes.

Further characterization of putative methyltransferases. Interestingly, five out of seven genes coding for putative methyltransferases showed highly significant PcVelA-dependent expression profiles in previous microarray analyses. To further validate PcVelA-dependent expression of these genes, qRT-PCR analysis was performed with RNA from strains that were grown under exactly the same conditions as for ChIP-seq sample preparation. We compared expression levels from a PcvelA overexpression strain, PcVelA-ChIP, and PcvelA deletion strain, ΔPcvelA, with those from wild type P2niaD18. As shown in Fig. 4, PcVelA-dependent expression profiles were confirmed for four out of seven tested genes, namely PcllmA (Pc21g02240), PcvipC (Pc18g01840), Pc21g12700, and Pc22g01170. Moreover, Pc18g06010 showed PcVelA-dependent expression in qRT-PCR but not in microarray analysis (Fig. 4, Table 2).

Next, amino acid sequences of putative methyltransferases were compared to that of the velvet-complex component PcLaeA, in order to draw inferences to possible similarities in terms of their functional properties. In general, a set of three conserved sequence motifs (motif I-III), essential for catalytic activity, can be identified in most methyltransferases (Kagan and Clarke 1994, Hacker et al. 2000). The most prominent one, “motif I”, is characterized by three consecutive glycine (G) residues, which are conserved in fungi, plants, and humans (Kozbial and Mushegian 2005, Sarikaya-Bayram et al. 2014). As shown in

Fig. 5, comparison of amino acid sequences revealed strong accordance for PcLlmA, PcVipC, the one encoded by Pc21g12700, and PcLaeA. Interestingly, all of these putative methyltransferases are regarded as S-adenosyl-methionine (SAM)-dependent methyltransferases, and conservation seemed to be restricted to the regions spanning the methyltransferase sequence motifs. This finding fits the observation that although SAM-dependent methyltransferases share a highly conserved structural fold and - in most cases - carry a set of three conserved methyltransferase sequence motifs, they share little sequence similarity (Kagan and Clarke 1994, Hacker et al. 2000, Martin and McMillan 2002).

De novo prediction and validation of a PcVelA DNA-binding motif. We used MEME to perform a de novo prediction of a PcVelA DNA-binding motif, based on peak regions present in two independent ChIP-experiments. Our analysis revealed one highly significant motif sequence, designated PcVelA.M1, which was found to be present in 275 (46.5 %) out of 592 peak regions when applying a statistical threshold of p ≤ 0.001. While comparison of PcVelA.M1 to known binding motifs within the JASPAR CORE (2014) fungi database did not reveal any significant matches, comparison to the JASPER CORE (2014) vertebrates’ database revealed some interesting similarities. PcVelA.M1 most closely resembled DNA-binding motifs of NR2E3, a photoreceptor nuclear receptor TF involved in human photoreceptor development (Kobayashi et al. 1999, Milam et al. 2002), and NR2F1, a nuclear hormone receptor and transcriptional regulator playing an important role in the neurodevelopment of the visual system in humans (Bosch et al. 2014). Independently to bioinformatics analysis, we observed weak similarity between PcVelA.M1 and a DNA-binding consensus sequence that has recently been described for A. nidulans VosA (Ahmed et al. 2013).

To further verify biological significance of PcVelA.M1, we performed electrophoretic mobility shift assays (EMSAs). As shown in Fig. 7, a GST-tagged version of the PcVelA

N-terminal region (PcVelA1-256), purified from E. coli, showed specific binding to 50 nt oligonucleotides (PcLlmA_2 and PcLlmA_4) derived from a region within the PcllmA upstream sequence and harboring exactly one copy of PcVelA.M1. Specific binding was also documented for full-length PcVelA (data not shown), however, the PcVelA N-terminus seems to be sufficient to mediate effective DNA binding. As PcVelA1-256 harbors the complete velvet domain, this might indicate that the velvet domain mediates PcVelA DNA-binding, like it has been previously demonstrated for A. nidulans VosA and VelB (Ahmed et al. 2013). Binding specificity between PcVelA and the DNA-binding consensus sequence PcVelA.M1 was

9 further verified, when mutated versions of the aforementioned oligonucleotides

(PcLlmA_2_m and PcLlmA_4_m) were tested for binding to PcVelA1-256 (Fig. 7). Complex formation was diminished drastically due to four single nucleotide mutations within the motif sequence, thus, confirming PcVelA.M1 as a specific PcVelA DNA-binding motif.

PcVelA directly interacts with PcLlmA, a putative SAM-dependent methyltransferase. Starting from the observation that we identified PcllmA as a direct target gene of PcVelA, and comparison of amino acid sequences revealed high conservation within the methyltransferase motifs I-III of PcLaeA and PcLlmA (Fig. 5), we decided to submit PcLlmA to further functional characterization. As shown in Fig. 8A, we were able to confirm direct interaction between PcVelA and PcLlmA by using an ex vivo yeast two–hybrid (Y2H) approach. Here, diploid strains synthesizing both the bait and the prey protein were spotted on selective media, lacking adenine and histidine and supplemented with X-α-Gal, to demonstrate ADE2 and HIS3, as well as lacZ reporter gene activity. Interestingly, PcLlmA did interact with PcVelA but not with other velvet components (data not shown). This is also true for the interaction between PcLaeA and the velvet complex, which is solely mediated by PcVelA (Hoff et al. 2010, Kopke et al. 2013). To confirm the observed interaction between PcVelA and PcLlmA in vivo in the homologous system, we used bimolecular fluorescence complementation analysis (BiFC) (Hoff and Kück 2005). Genes encoding PcVelA and PcLlmA were fused to eyfp fragments encoding either the N- or the C-terminus of the yellow fluorescent protein, and strains harboring both constructs were analyzed using fluorescence microscopy. As a control, we investigated strains producing only split EYFPs and strains producing one split EYFP together with either EYFP-PcVelA or PcLlmA-EYFP. As shown in Fig. 8B, strains carrying both PcVelA and PcLlmA eyfp-fusion constructs showed clear EYFP signals while no fluorescence was detectable in control strains. Additional DAPI staining demonstrated that interaction between PcVelA and PcLlmA takes place in the nucleus, as it has previously been shown for the interaction between PcVelA and PcLaeA, PcVelB, PcVelC, and PcVosA as well as the interaction of PcVelA with itself (Hoff et al. 2010, Kopke et al. 2013).

DISCUSSION

Although functional characterization of the velvet-complex components in P. chrysogenum and related species has been in the limelight of research during the past years (Bayram et al. 2008, Hoff et al. 2010, Wiemann et al. 2010), the output mechanisms of genome-wide velvet protein-mediated regulatory functions on a molecular level remained enigmatic. Within this work, we present the first ChIP-seq analysis of one of the core components of the velvet complex and provide unambiguous evidence for an involvement of PcVelA in genome-wide transcriptional regulation on DNA-level, possibly even as a TF.

PcVelA acts as a global regulator of transcriptional regulation. Based on data obtained from ChIP-seq analysis, we identified a total of 592 highly specific PcVelA DNA-binding sites and 631 putative direct PcVelA target genes, of which 18.9 % showed PcVelA-dependent changes in expression levels by at least twofold in previous microarray analyses (Hoff et al. 2010). This apparently small overlap is in accordance with previous ChIP-seq analyses in P. chrysogenum and comparable analyses in yeast and higher eukaryotes, which revealed an overlap of ~50 % and 10-25 % between TF occupancy and expression of neighboring genes (Gao et al. 2004, Sandmann et al. 2006, Jakobsen et al. 2007, Vokes et al. 2008, Becker et al. 2015). A remarkable number of PcVelA target genes, identified in our analysis, could be clearly related to known velvet-regulated cellular and developmental processes. For example, we identified at least 14 genes involved in the regulation of conidiation and development, as well as 16 genes, which could be assigned to secondary metabolism (see Table 2). Genes related to these categories were localized next to some of the most significant PcVelA DNA- binding sites, which is in great accordance with the general acceptance that the most likely bound regions identified in DNA-binding protein ChIP-seq analysis are positioned next to generally known functional target genes (Todeschini et al. 2014). However, it was shown that even low-affinity TF binding may have a functional role in chromatin remodeling (Cao et al. 2010) or nucleosome positioning (Zaret and Carroll 2011), which can influence gene expression at later developmental stages or have an additional non-transcriptional function (Spitz and Furlong 2012). Based on our ChIP-seq data, this might also be true for a PcVelA peak region within the upstream sequence of PcvelB, encoding another component of the velvet complex that acts as an activator of conidiospores formation in various fungi (Bayram et al. 2008, Wiemann et al. 2010, Kopke et al. 2013). Although PcvelB did not show PcVelA- dependent expression in previous microarray analyses, and the peak value of the corresponding ChIP-seq peak region pointed to a rather low-affinity target region, further

11 research will be needed in order to analyze this remarkable observation and to draw inferences about its biological relevance.

Besides somehow expected PcVelA target genes involved in regulation of spore formation and secondary metabolism, our dataset also included a large number of new putative PcVelA target genes, e.g. five genes encoding proteins with functions related to Acyl-CoA/Acetyl-CoA synthesis and utilization and several genes encoding TFs with so far unknown functions. Acetyl-CoA is one of the key biochemical precursors used in fundamental cellular metabolism, such as fatty acid metabolism, and secondary metabolite synthesis (Hutchinson and Fujii 1995, Brown et al. 1996, Kistler and Broz 2015). For example, Acetyl-CoA is needed during the final enzymatic steps of penicillin biosynthesis in P. chrysogenum, where the enzyme acyl-coenzyme A:isopenicillin N acyltransferase converts isopenicillin N (IPN) to penicillin G by exchange of the α-amino adipyl side chain of IPN with CoA-activated phenylacetic acid (Brakhage et al. 2004). Among genes encoding proteins with functions related to Acyl-CoA/Acetyl-CoA, we identified a homolog of AnfacA, coding for a cytoplasmic Acetyl-CoA synthetase in A. nidulans. It has been shown that loss-of-function mutations in the AnfacA gene result in resistance to fluoroacetate in the absence of a repressing carbon source, which otherwise inhibits development, conidiation, and conidial pigmentation (Hynes and Murray 2010). Based on the diversity of Acetyl-CoA/Acyl-CoA functions, it will be highly interesting to characterize the functions of Acetyl-CoA- and Acyl-CoA-utilizing enzymes, encoded by specific target genes of PcVelA, in more detail. Another promising starting point for future experiments includes the characterization of TFs encoded by PcVelA target genes. As these proteins might be acting as important downstream factors of the velvet complex, knowledge of their specific regulatory functions will play a crucial part in improving our general understanding of the regulatory circuits governed by velvet.

PcVelA binds DNA in a sequence-dependent manner. Besides the identification of direct target genes, ChIP-seq data were also used for the de novo prediction of a PcVelA DNA- binding motif, PcVelA.M1, which was further verified by DNA-binding studies (EMSAs). Interestingly, with the exception of the DNA-binding motif described for A. nidulans VosA, we were not able to identify any fungal DNA-binding consensus sequences similar to PcVelA.M1. This finding is in accordance to a theory advanced by Ni and Yu in 2007, which implies that A. nidulans VeA, VelB, and VosA might be acting as global transcriptional regulators, representing a new fungus-specific class of TFs. Furthermore, when comparing

PcVelA.M1 to known DNA-binding motifs from vertebrates, close similarities to motif sequences of transcriptional regulators involved in development of the visual system became obvious. As PcVelA is known as a regulator of light-dependent formation of conidiospores in P. chrysogenum and related species, this finding points to a possible evolutionary interconnection between light sensing systems in filamentous fungi and vertebrates. Most importantly, PcVelA.M1 resembled the DNA-binding motif of NR2E3, a photoreceptor nuclear receptor TF involved in the regulation of human photoreceptor development (Kobayashi et al. 1999, Milam et al. 2002). It was shown that mutations in NR2E3 are associated with the enhanced S-cone syndrome (ESCS), characterized by night blindness, varying degrees of cone vision, and retinal degeneration in humans (Haider et al. 2000). Remarkably, this phenotype somehow resembles those of PcvelA/veA deletion mutants in P. chrysogenum and A. nidulans, which are characterized by an impaired light-sensing ability, leading to the formation of conidiospores in the absence of light, whereas conidiogenesis in the wild type is light-dependent (Käfer 1965, Mooney and Yager 1990, Hoff et al. 2010).

The response to light and the corresponding molecular mechanisms that regulate many different aspects of the biology of organisms have been extensively studied in plants and many fungi, in particular Neurospora crassa (Chen et al. 2004, Chen et al. 2010). However, only little is known about the mechanisms underlying light perception in P. chrysogenum and closely related species. In A. nidulans, direct interaction between VeA and FphA, a phytochrome acting as a red-light sensor, as well as FphA-mediated interaction between VeA and LreA/LreB, orthologs of the respective N. crassa blue-light responsive WC-1 and WC-2, have been demonstrated (Blumenstein et al. 2005, Purschwitz et al. 2009). Accordingly, it has been hypothesized that development in A. nidulans is regulated through an interplay of two light-sensing systems, FphA and LreA/LreB, with VeA, interacting at the protein level (Calvo 2008). In A. nidulans, one of the major responses to light is the regulation of asexual development, a pathway that is controlled by the master regulator BrlA (Adams et al. 1988, Ruger-Herreros et al. 2011). It was demonstrated that the fluffy genes fluG and flbA-E encode regulators of light-dependent brlA expression. Their deletion reduces expression of brlA, resulting in aconidial, fluffy phenotypes (Ruger-Herreros et al. 2011). As deletion of lreB results in a complete loss of brlA expression but not of the fluffy genes, it was hypothesized that the photoreceptor complex interacts directly with brlA (Ruger-Herreros et al. 2011). Identification and validation of a PcVelA DNA-binding motif with high similarity to those of regulators of development of the visual system in vertebrates as well as the identification of specific PcVelA DNA-binding sites within the upstream region of the fluffy genes flbC and 13 flbD, as well as brlA, clearly indicates a need for reconsideration of the regulatory functions of PcVelA in terms of light perception in P. chrysogenum and related species. Most importantly, our findings might help to finally answer the question how photoreceptors induce the expression of the regulators of conidiation and whether it is a direct or indirect event through other components. Based on our data, a regulatory function of PcVelA on DNA-level instead or simultaneously to a regulatory function on protein level has to be taken into consideration. Further research will be needed in order to clarify the exact involvement of PcVelA in transcriptional regulation of light-dependent gene expression and to elucidate the possible evolutionary interconnection between velvet proteins and regulators of the visual system in vertebrates.

PcVelA directly interacts with the putative SAM-dependent methyltransferase PcLlmA. Recently, a number of putative methyltransferases other than LaeA have been described to directly interact with VeA in A. nidulans and F. graminearum. In A. nidulans, methyltransferase LlmF was shown to be involved in determination of VeA localization and methyltransferases VipC and VapB were demonstrated to be involved in regulation of sexual and asexual development (Palmer et al. 2013, Sarikaya-Bayram et al. 2014). In F. graminearum, a total of six putative methyltransferases were shown to directly interact with FgVeA using a Y2H screen (Jiang et al. 2011). Moreover, a number of putative methyltransferases have been demonstrated to exert LaeA-similar functions in various fungi. For example, in the maize pathogen Cochliobolus heterostrophus, Lae1-like methyltransferase Llm1, a homolog of A. nidulans LlmF, was shown to act as a negative regulator of T-toxin production and asexual sporulation (Bi et al. 2013), and in F. graminearum, methyltransferase KMT6 was shown to be involved in regulation of development as well as expression of genes for mycotoxins, pigments, and other secondary metabolites (Connolly et al. 2013). Against this background, it appeared highly interesting that we identified a total of seven putative methyltransferase-encoding genes as PcVelA targets. To the best of our knowledge, no putative methyltransferase-encoding gene has been identified as a specific downstream factor of VeA in P. chrysogenum or related species until today. As the methyltransferase-specific sequence motifs (motif I-III) of one of the encoded putative methyltransferases, PcLlmA, exhibited noticeably high consistency to those of PcLaeA, and the corresponding gene was located next to a highly significant PcVelA DNA-binding region obtained from ChIP-seq analysis, we decided to submit PcLlmA to further functional characterization. Using Y2H and BiFC analysis we were able to verify direct interaction between PcVelA and PcLlmA on protein level. Interestingly, this interaction 14 seemed to be restricted to PcVelA, as no other interaction between PcLlmA and components of the velvet complex could be demonstrated. This observation is in great accordance with previous Y2H analyses, revealing that interaction of the putative methyltransferase PcLaeA is restricted to PcVelA (Kopke et al. 2013). Furthermore, DAPI staining confirmed nuclear localization of PcVelA-PcLlmA in vivo, as it was previously described for PcVelA-PcLaeA and PcVelA-PcVelB (Hoff et al. 2010), PcVelA-PcVelC and PcVelA-PcVosA (Kopke et al. 2013) as well as VeA-LlmF (Palmer et al. 2013), VeA-VipC, and VipC-VapB in A. nidulans (Sarikaya-Bayram et al. 2014).

PcLaeA as well as PcLlmA belong to the family of SAM-dependent methyltransferases, which catalyze the transfer of methyl groups from SAM to a large variety of acceptor substrates. These substrates are ranging from small metabolites to bio-macromolecules, including DNA, proteins and small-molecule secondary metabolites (Martin and McMillan 2002, Jiang et al. 2011, Struck et al. 2012). As the biological functions of methylation are versatile, reaching from biosynthesis, metabolism, detoxification and signal transduction, to protein sorting and repair as well as nucleic acid processing (Martin and McMillan 2002), they have a significant potential for application in biotechnology (Struck et al. 2012). Accordingly, the putative SAM-dependent methyltransferase LaeA has been shown to influence regulation of secondary metabolite gene clusters in various fungi (Bok and Keller 2004, Kale et al. 2008, Kosalková et al. 2009, Sarikaya-Bayram et al. 2010, Wiemann et al. 2010, Karimi-Aghcheh et al. 2013). For example, deletion of P. chrysogenum PclaeA resulted in a significant reduction of penicillin biosynthesis (Hoff et al. 2010). It remains a task of the future to elucidate the regulatory functions of VeA-interacting methyltransferases on a molecular level and to analyze how the interaction between VeA and the growing number of methyltransferases described as direct interaction partners in P. chrysogenum and other fungi is mediated on a structural level. Based on data from A. nidulans, it was suggested that VeA should have an affinity domain for methyltransferases or a tertiary domain providing interaction between VeA and methyltransferases (Bayram et al. 2008, Sarikaya-Bayram et al. 2015). However, experimental evidence is needed to verify sustainability of these hypotheses and to elucidate the functional output of this interaction in more detail.

Taken together, our data provide unambiguous evidence for PcVelA acting as a global transcriptional regulator, involved in a variety of cellular and developmental processes, ranging from conidiation and development to secondary metabolism. Furthermore, at least seven genes encoding putative methyltransferases were identified as specific PcVelA target

15 genes, and PcLlmA, a putative SAM-domain methyltransferase was introduced as a new direct interaction partner of PcVelA on protein level. However, even if this work provided unprecedented deep insight into PcVelA regulatory functions on a genome-wide scale, much work remains to be done in order to fully understand PcVelA’s ambiguous nature as a transcriptional regulator on the one hand and as one of the core components of the multi-subunit velvet complex on the other hand.

ACKNOWLEDGEMENTS

We thank PD Dr. M. Nowrousian and M. Sc. T. A. Dahlmann for help with bioinformatics, and S. Mertens, I. Godehardt, K. Kalkreuter and I. Schelberg for excellent technical assistance. We further thank Drs. I. Zadra, H. Kürnsteiner, E. Friedlin, and T. Specht for their ongoing interest and support. This work was funded by Sandoz GmbH, the Christian Doppler Society, the German National Academic Foundation, and the Ruhr-University Bochum Research School.

REFERENCES

Adams TH, Boylan MT, Timberlake WE (1988) BrlA is necessary and sufficient to direct conidiophore development in Aspergillus nidulans. Cell 54: 353-62

Ahmed YL, Gerke J, Park HS, Bayram Ö, Neumann P, Ni M, Dickmanns A, Kim SC, Yu JH, Braus GH, Ficner R (2013) The velvet family of fungal regulators contains a DNA-binding domain structurally similar to NF-κB. PLoS Biol 11: e1001750

Arratia-Quijada J, Sanchez O, Scazzocchio C, Aguirre J (2012) FlbD, a Myb transcription factor of Aspergillus nidulans, is uniquely involved in both asexual and sexual differentiation. Eukaryot Cell 11: 1132-42

Backus MP, Stauffer JF (1955) The production and selection of a family of strains in Penicillium chrysogenum. Mycologia 47: 429-63

Bailey TL, Elkan C (1994) Fitting a mixture model by expectation maximization to discover motifs in biopolymers. Proc Int Conf Intell Syst Mol Biol 2: 28-36

Balazs A, Pocsi I, Hamari Z, Leiter E, Emri T, Miskei M, Olah J, Toth V, Hegedus N, Prade RA, Molnar M (2010) AtfA bZIP-type transcription factor regulates oxidative and osmotic stress responses in Aspergillus nidulans. Mol Genet Genomics 283: 289-303

Barreiro C, Martín JF, García-Estrada C (2012) Proteomics shows new faces for the old penicillin producer Penicillium chrysogenum. J Biomed Biotechnol 2012: 1-15

Bayram Ö, Krappmann S, Ni M, Bok JW, Helmstaedt K, Valerius O, Braus-Stromeyer S, Kwon NJ, Keller NP, Yu JH, Braus GH (2008) VelB/VeA/LaeA complex coordinates light signal with fungal development and secondary metabolism. Science 320: 1504-6

Becker K, Beer C, Freitag M, Kück U (2015) Genome-wide identification of target genes of a mating-type α-domain transcription factor reveals functions beyond sexual development. Mol Microbiol (Epub ahead of print)

Bi Q, Wu D, Zhu X, Gillian Turgeon B (2013) Cochliobolus heterostrophus Llm1 - a Lae1-like methyltransferase regulates T-toxin production, virulence, and development. Fungal Genet Biol 51: 21-33

Blumenstein A, Vienken K, Tasler R, Purschwitz J, Veith D, Frankenberg-Dinkel N, Fischer R (2005) The Aspergillus nidulans phytochrome FphA represses sexual development in red light. Curr Biol 15: 1833-8

Böhm J, Hoff B, O'Gorman CM, Wolfers S, Klix V, Binger D, Zadra I, Kürnsteiner H, Pöggeler S, Dyer PS, Kück U (2013) Sexual reproduction and mating-type-mediated strain development in the penicillin-producing fungus Penicillium chrysogenum. Proc Natl Acad Sci U S A 110: 1476-81

Bok JW, Keller NP (2004) LaeA, a regulator of secondary metabolism in Aspergillus spp. Eukaryot Cell 3: 527-35

Bosch DG, Boonstra FN, Gonzaga-Jauregui C, Xu M, de Ligt J, Jhangiani S, Wiszniewski W, Muzny DM, Yntema HG, Pfundt R, Vissers LE, Spruijt L, Blokland EA, Chen CA, Lewis RA, Tsai SY, Gibbs RA, Tsai MJ, Lupski JR, Zoghbi HY, Cremers FP, de Vries BB, Schaaf CP (2014) NR2F1 mutations cause optic atrophy with intellectual disability. Am J Hum Genet 94: 303-9 18

Brakhage AA, Spröte P, Al-Abdallah Q, Gehrke A, Plattner H, Tüncher A (2004) Regulation of penicillin biosynthesis in filamentous fungi. Adv Biochem Eng Biotechnol 88: 45-90

Brown DW, Adams TH, Keller NP (1996) Aspergillus has distinct fatty acid synthases for primary and secondary metabolism. Proc Natl Acad Sci U S A 93: 14873-7

Bullock WO, Fernandez JM, Short JM (1987) Xl1-Blue - a high-efficiency plasmid transforming recA Escherichia coli strain with β-galactosidase selection. Biotechniques 5: 376-8

Calvo AM (2008) The VeA regulatory system and its role in morphological and chemical development in fungi. Fungal Genet Biol 45: 1053-61

Cao Y, Yao Z, Sarkar D, Lawrence M, Sanchez GJ, Parker MH, MacQuarrie KL, Davison J, Morgan MT, Ruzzo WL, Gentleman RC, Tapscott SJ (2010) Genome-wide MyoD binding in skeletal muscle cells: a potential for broad cellular reprogramming. Dev Cell 18: 662-74

Chakrabortee S, Tripathi R, Watson M, Schierle GSK, Kurniawan DP, Kaminski CF, Wise MJ, Tunnacliffe A (2012) Intrinsically disordered proteins as molecular shields. Mol Biosyst 8: 210-9

Chen CH, Dunlap JC, Loros JJ (2010) Neurospora illuminates fungal photoreception. Fungal Genet Biol 47: 922-9

Chen M, Chory J, Fankhauser C (2004) Light signal transduction in higher plants. Annu Rev Genet 38: 87-117

Connolly LR, Smith KM, Freitag M (2013) The Fusarium graminearum histone H3 K27 methyltransferase KMT6 regulates development and expression of secondary metabolite gene clusters. PLoS Genet 9: e1003916 d'Enfert C, Fontaine T (1997) Molecular characterization of the Aspergillus nidulans treA gene encoding an acid trehalase required for growth on trehalose. Mol Microbiol 24: 203-16

DeZwaan TM, Carroll AM, Valent B, Sweigard JA (1999) Magnaporthe grisea Pth11p is a novel plasma membrane protein that mediates appressorium differentiation in response to inductive substrate cues. Plant Cell 11: 2013-30

Dreyer J, Eichhorn H, Friedlin E, Kürnsteiner H, Kück U (2007) A homologue of the Aspergillus velvet gene regulates both cephalosporin C biosynthesis and hyphal fragmentation in Acremonium chrysogenum. Appl Environ Microbiol 73: 3412-22

Engh I, Würtz C, Witzel-Schlomp K, Zhang HY, Hoff B, Nowrousian M, Rottensteiner H, Kück U (2007) The WW domain protein PRO40 is required for fungal fertility and associates with woronin bodies. Eukaryot Cell 6: 831-43

Fleming A (1929) On the antibacterial action of cultures of a Penicillium, with special reference to their use in the isolation of B. influenzae. British Journal of Experimental Pathology 10: 226- 36

Gao F, Foat BC, Bussemaker HJ (2004) Defining transcriptional networks through integrative modeling of mRNA expression and transcription factor binding data. BMC Bioinformatics 5: 31

Gerke J, Braus GH (2014) Manipulation of fungal development as source of novel secondary metabolites for biotechnology. Appl Microbiol Biotechnol 98: 8443-55

Goyal K, Walton LJ, Tunnacliffe A (2005) LEA proteins prevent protein aggregation due to water stress. Biochem J 388: 151-7

Gupta S, Stamatoyannopoulos JA, Bailey TL, Noble WS (2007) Quantifying similarity between motifs. Genome Biol 8: R24

Hacker C, Glinski M, Hornbogen T, Doller A, Zocher R (2000) Mutational analysis of the N- methyltransferase domain of the multifunctional enzyme enniatin synthetase. J Biol Chem 275: 30826-32

Hagiwara D, Suzuki S, Kamei K, Gonoi T, Kawamoto S (2014) The role of AtfA and HOG MAPK pathway in stress tolerance in conidia of Aspergillus fumigatus. Fungal Genet Biol 73: 138-49

Haider NB, Jacobson SG, Cideciyan AV, Swiderski R, Streb LM, Searby C, Beck G, Hockey R, Hanna DB, Gorman S, Duhl D, Carmi R, Bennett J, Weleber RG, Fishman GA, Wright AF, Stone EM, Sheffield VC (2000) Mutation of a nuclear receptor gene, NR2E3, causes enhanced S cone syndrome, a disorder of retinal cell fate. Nat Genet 24: 127-31

Heinz S, Benner C, Spann N, Bertolino E, Lin YC, Laslo P, Cheng JX, Murre C, Singh H, Glass CK (2010) Simple combinations of lineage-determining transcription factors prime cis- regulatory elements required for macrophage and B cell identities. Mol Cell 38: 576-89

Hoff B, Kück U (2005) Use of bimolecular fluorescence complementation to demonstrate transcription factor interaction in nuclei of living cells from the filamentous fungus Acremonium chrysogenum. Curr Genet 47: 132-8

Hoff B, Pöggeler S, Kück U (2008) Eighty years after its discovery, Fleming's Penicillium strain discloses the secret of its sex. Eukaryot Cell 7: 465-70

Hoff B, Kamerewerd J, Sigl C, Zadra I, Kück U (2009) Homologous recombination in the antibiotic producer Penicillium chrysogenum: strain ΔPcku70 shows up-regulation of genes from the HOG pathway. Appl Microbiol Biotechnol 85: 1081-94

Hoff B, Kamerewerd J, Sigl C, Mitterbauer R, Zadra I, Kürnsteiner H, Kück U (2010) Two components of a velvet-like complex control hyphal morphogenesis, conidiophore development, and penicillin biosynthesis in Penicillium chrysogenum. Eukaryot Cell 9: 1236- 50

Hutchinson CR, Fujii I (1995) Polyketide synthase gene manipulation: a structure-function approach in engineering novel antibiotics. Annu Rev Microbiol 49: 201-38

Hynes MJ, Murray SL (2010) ATP-citrate lyase is required for production of cytosolic acetyl coenzyme A and development in Aspergillus nidulans. Eukaryot Cell 9: 1039-48

Jakobsen JS, Braun M, Astorga J, Gustafson EH, Sandmann T, Karzynski M, Carlsson P, Furlong EE (2007) Temporal ChIP-on-chip reveals Biniou as a universal regulator of the visceral muscle transcriptional network. Genes Dev 21: 2448-60

James P, Halladay J, Craig EA (1996) Genomic libraries and a host strain designed for highly efficient two-hybrid selection in yeast. Genetics 144: 1425-36

Janus D, Hortschansky P, Kück U (2007) Identification of a minimal cre1 promoter sequence promoting glucose-dependent gene expression in the β-lactam producer Acremonium chrysogenum. Curr Genet 53: 35-48

Jiang J, Liu X, Yin Y, Ma Z (2011) Involvement of a velvet protein FgVeA in the regulation of asexual development, lipid and secondary metabolisms and virulence in Fusarium graminearum. PLoS One 6: e28291

Käfer E (1965) Origins of translocations in Aspergillus nidulans. Genetics 52: 217-32

Kagan RM, Clarke S (1994) Widespread occurrence of three sequence motifs in diverse S- adenosylmethionine-dependent methyltransferases suggests a common structure for these enzymes. Arch Biochem Biophys 310: 417-27

Kale SP, Milde L, Trapp MK, Frisvad JC, Keller NP, Bok JW (2008) Requirement of LaeA for secondary metabolism and sclerotial production in Aspergillus flavus. Fungal Genet Biol 45: 1422-9

Kamerewerd J, Zadra I, Kürnsteiner H, Kück U (2011) PcchiB1, encoding a class V chitinase, is affected by PcVelA and PcLaeA, and is responsible for cell wall integrity in Penicillium chrysogenum. Microbiology 157: 3036-48

Karimi-Aghcheh R, Bok JW, Phatale PA, Smith KM, Baker SE, Lichius A, Omann M, Zeilinger S, Seiboth B, Rhee C, Keller NP, Freitag M, Kubicek CP (2013) Functional analyses of Trichoderma reesei LAE1 reveal conserved and contrasting roles of this regulator. G3 (Bethesda) 3: 369-78

Kato N, Brooks W, Calvo AM (2003) The expression of sterigmatocystin and penicillin genes in Aspergillus nidulans is controlled by veA, a gene required for sexual development. Eukaryot Cell 2: 1178-86

Kim H, Han K, Kim K, Han D, Jahng K, Chae K (2002) The veA gene activates sexual development in Aspergillus nidulans. Fungal Genet Biol 37: 72-80

Kistler HC, Broz K (2015) Cellular compartmentalization of secondary metabolism. Front Microbiol 6: 68

Kobayashi M, Takezawa S, Hara K, Yu RT, Umesono Y, Agata K, Taniwaki M, Yasuda K, Umesono K (1999) Identification of a photoreceptor cell-specific nuclear receptor. Proc Natl Acad Sci U S A 96: 4814-9

Kopke K, Hoff B, Kück U (2010) Application of the Saccharomyces cerevisiae FLP/FRT recombination system in filamentous fungi for marker recycling and construction of knockout strains devoid of heterologous genes. Appl Environ Microbiol 76: 4664-74

Kopke K, Hoff B, Bloemendal S, Katschorowski A, Kamerewerd J, Kück U (2013) Members of the Penicillium chrysogenum velvet complex play functionally opposing roles in the regulation of penicillin biosynthesis and conidiation. Eukaryot Cell 12: 299-310

Kosalková K, García-Estrada C, Ullán RV, Godio RP, Feltrer R, Teijeira F, Mauriz E, Martín JF (2009) The global regulator LaeA controls penicillin biosynthesis, pigmentation and sporulation, but not roquefortine C synthesis in Penicillium chrysogenum. Biochimie 91: 214- 25

Kozbial PZ, Mushegian AR (2005) Natural history of S-adenosylmethionine-binding proteins. BMC Struct Biol 5: 19

Kraus PR, Hofmann AF, Harris SD (2002) Characterization of the Aspergillus nidulans 14-3-3 homologue, ArtA. FEMS Microbiol Lett 210: 61-6

Kwon NJ, Garzia A, Espeso EA, Ugalde U, Yu JH (2010) FlbC is a putative nuclear C2H2 transcription factor regulating development in Aspergillus nidulans. Mol Microbiol 77: 1203- 19

Landt SG, Marinov GK, Kundaje A, Kheradpour P, Pauli F, Batzoglou S, Bernstein BE, Bickel P, Brown JB, Cayting P, Chen Y, DeSalvo G, Epstein C, Fisher-Aylor KI, Euskirchen G, Gerstein M, Gertz J, Hartemink AJ, Hoffman MM, Iyer VR, Jung YL, Karmakar S, Kellis M, Kharchenko PV, Li Q, Liu T, Liu XS, Ma L, Milosavljevic A, Myers RM, Park PJ, Pazin MJ, Perry MD, Raha D, Reddy TE, Rozowsky J, Shoresh N, Sidow A, Slattery M, Stamatoyannopoulos JA, Tolstorukov MY, White KP, Xi S, Farnham PJ, Lieb JD, Wold BJ, Snyder M (2012) ChIP-seq guidelines and practices of the ENCODE and modENCODE consortia. Genome Res 22: 1813-31

Langmead B, Trapnell C, Pop M, Salzberg SL (2009) Ultrafast and memory-efficient alignment of short DNA sequences to the human genome. Genome Biol 10: R25

Li H, Handsaker B, Wysoker A, Fennell T, Ruan J, Homer N, Marth G, Abecasis G, Durbin R (2009) The sequence alignment/map format and SAMtools. Bioinformatics 25: 2078-9

Lin Z, Wu WS, Liang H, Woo Y, Li WH (2010) The spatial distribution of cis regulatory elements in yeast promoters and its implications for transcriptional regulation. BMC Genomics 11: 581

Lind AL, Wisecaver JH, Smith TD, Feng X, Calvo AM, Rokas A (2015) Examining the evolution of the regulatory circuit controlling secondary metabolism and development in the fungal genus Aspergillus. PLoS Genet 11: e1005096

Martin JL, McMillan FM (2002) SAM (dependent) I AM: the S-adenosylmethionine-dependent methyltransferase fold. Curr Opin Struct Biol 12: 783-93

Merhej J, Urban M, Dufresne M, Hammond-Kosack KE, Richard-Forget F, Barreau C (2012) The velvet gene, Fgve1, affects fungal development and positively regulates trichothecene biosynthesis and pathogenicity in Fusarium graminearum. Mol Plant Pathol 13: 363-74

Milam AH, Rose L, Cideciyan AV, Barakat MR, Tang WX, Gupta N, Aleman TS, Wright AF, Stone EM, Sheffield VC, Jacobson SG (2002) The nuclear receptor NR2E3 plays a role in human retinal photoreceptor differentiation and degeneration. Proc Natl Acad Sci U S A 99: 473-8

Miller KY, Wu JG, Miller BL (1992) StuA is required for cell pattern-formation in Aspergillus. Gene Dev 6: 1770-82

Miroux B, Walker JE (1996) Over-production of proteins in Escherichia coli: mutant hosts that allow synthesis of some membrane proteins and globular proteins at high levels. J Mol Biol 260: 289-98

Mooney JL, Yager LN (1990) Light is required for conidiation in Aspergillus nidulans. Genes Dev 4: 1473-82

Nguyen DH, D'Haeseleer P (2006) Deciphering principles of transcription regulation in eukaryotic genomes. Mol Syst Biol 2: 2006.0012

Ni M, Yu JH (2007) A novel regulator couples sporogenesis and trehalose biogenesis in Aspergillus nidulans. PLoS One 2: e970

Olmedo M, Ruger-Herreros C, Luque EM, Corrochano LM (2010) A complex photoreceptor system mediates the regulation by light of the conidiation genes con-10 and con-6 in Neurospora crassa. Fungal Genet Biol 47: 352-63

Palmer JM, Theisen JM, Duran RM, Grayburn WS, Calvo AM, Keller NP (2013) Secondary metabolism and development is mediated by LlmF control of VeA subcellular localization in Aspergillus nidulans. PLoS Genet 9: e1003193

Peñalva MA, Rowlands RT, Turner G (1998) The optimization of penicillin biosynthesis in fungi. Trends Biotechnol 16: 483-9

Purschwitz J, Müller S, Fischer R (2009) Mapping the interaction sites of Aspergillus nidulans phytochrome FphA with the global regulator VeA and the White Collar protein LreB. Mol Genet Genomics 281: 35-42

Ruger-Herreros C, Rodríguez-Romero J, Fernández-Barranco R, Olmedo M, Fischer R, Corrochano LM, Canovas D (2011) Regulation of conidiation by light in Aspergillus nidulans. Genetics 188: 809-22

Sandmann T, Jensen LJ, Jakobsen JS, Karzynski MM, Eichenlaub MP, Bork P, Furlong EEM (2006) A temporal map of transcription factor activity: Mef2 directly regulates at all stages of muscle target genes development. Dev Cell 10: 797-807

Sarikaya-Bayram Ö, Bayram Ö, Valerius O, Park HS, Irniger S, Gerke J, Ni M, Han KH, Yu JH, Braus GH (2010) LaeA control of velvet family regulatory proteins for light-dependent development and fungal cell-type specificity. PLoS Genet 6: e1001226

Sarikaya-Bayram Ö, Bayram Ö, Feussner K, Kim JH, Kim HS, Kaever A, Feussner I, Chae KS, Han DM, Han KH, Braus GH (2014) Membrane-bound methyltransferase complex VapA- VipC-VapB guides epigenetic control of fungal development. Dev Cell 29: 406-20

Sarikaya-Bayram Ö, Palmer JM, Keller N, Braus GH, Bayram Ö (2015) One Juliet and four Romeos: VeA and its methyltransferases. Front Microbiol 6: 1

Sarikaya Bayram Ö, Bayram Ö, Valerius O, Park HS, Irniger S, Gerke J, Ni M, Han KH, Yu JH, Braus GH (2010) LaeA control of velvet family regulatory proteins for light-dependent development and fungal cell-type specificity. Plos Genet 6: e1001226

Sigl C, Haas H, Specht T, Pfaller K, Kürnsteiner H, Zadra I (2011) Among developmental regulators, StuA but not BrlA is essential for penicillin V production in Penicillium chrysogenum. Appl Environ Microbiol 77: 972-82

Spitz F, Furlong EE (2012) Transcription factors: from enhancer binding to developmental control. Nat Rev Genet 13: 613-26

Stringer MA, Timberlake WE (1995) dewA encodes a fungal hydrophobin component of the Aspergillus spore wall. Mol Microbiol 16: 33-44

Struck AW, Thompson ML, Wong LS, Micklefield J (2012) S-adenosyl-methionine-dependent methyltransferases: highly versatile enzymes in biocatalysis, biosynthesis and other biotechnological applications. Chembiochem 13: 2642-55

Thorvaldsdóttir H, Robinson JT, Mesirov JP (2012) Integrative Genomics Viewer (IGV): high- performance genomics data visualization and exploration. Brief Bioinform 14: 178-92

Todeschini AL, Georges A, Veitia RA (2014) Transcription factors: specific DNA binding and specific gene regulation. Trends Genet 30: 211-9

Vokes SA, Ji H, Wong WH, McMahon AP (2008) A genome-scale analysis of the cis-regulatory circuitry underlying sonic hedgehog-mediated patterning of the mammalian limb. Genes Dev 22: 2651-63

Wiemann P, Brown DW, Kleigrewe K, Bok JW, Keller NP, Humpf HU, Tudzynski B (2010) FfVel1 and FfLae1, components of a velvet-like complex in Fusarium fujikuroi, affect differentiation, secondary metabolism and virulence. Mol Microbiol 77: 972-94

Wolfers S, Kamerewerd J, Nowrousian M, Sigl C, Zadra I, Kürnsteiner H, Kück U, Bloemendal S (2014) Microarray hybridization analysis of light-dependent gene expression in Penicillium chrysogenum identifies bZIP transcription factor PcAtfA. J Basic Microbiol 54: 1-10

Zaret KS, Carroll JS (2011) Pioneer transcription factors: establishing competence for gene expression. Genes Dev 25: 2227-41

Figure 1: Construction of PcVelA-ChIP strains. (A) Plasmid pVelA-EGFP, harboring a

Pgpd::PcvelA::egfp fusion construct, was used for ectopic integration into a marker-free PcvelA deletion strain, ΔPcvelA. (B) PCR analysis confirmed integration of Pgpd::PcvelA::egfp. Binding positions of primers PcvelA_f and egfp_r are indicated as arrows in (A). (C) Presence of the epitope-tagged protein PcVelA-EGFP in crude protein extract from recombinant P. chrysogenum strains was confirmed using SDS-PAGE/Western blot analysis. (D) Fluorescence microscopy confirmed nuclear localization of PcVelA-EGFP in the PcVelA-ChIP strain. Strains were grown on solid medium for 48 h.

Figure 2: Initial analysis of PcVelA ChIP-seq data. (A) Venn diagram showing intersection between peak regions identified within PcVelA_shaking_1 and PcVelA_shaking_2 datasets. Only peaks within a maximum distance of 100 nt were regarded as overlapping. (B) Distribution of ChIP-enriched regions overlapping with or positioned within intragenic regions vs. ChIP-enriched regions that were exclusively located within intergenic regions (based on peak regions identified in both biological replicates). (C) Distance between peak summits and ATG of neighboring genes positioned in 5’-3’ orientation with regard to the corresponding peak region (based on peak regions present in both biological replicates).

Figure 3: Validation of PcVelA ChIP-seq data. (A) ChIP-PCR analysis was performed to analyze enrichment of selected PcVelA target regions in ChIP-DNA compared to input-DNA. Enrichment was calculated as the ratio of the region of interest to a control region, showing no PcVelA-specific enrichment in ChIP-seq experiments. Another region showing no PcVelA- specific enrichment in ChIP-seq analysis is shown as a control (NC). ChIP-PCR ratios (grey bars) are shown in comparison to the corresponding peak values, as obtained from bioinformatics analysis (black bars). Values for ChIP-PCR are the mean score of three biological replicates; average ± standard deviations are indicated. Peak regions are named according to neighboring genes (see Dataset S1).

Figure 4: qRT-PCR analysis confirms PcVelA-dependent expression of putative methyltransferase-encoding genes. Analysis of relative log2fold gene expression ratios of putative methyltransferase-encoding genes, identified as specific PcVelA target genes in ChIP-seq analysis, confirmed PcVelA dependency. Expression ratios in PcVelA-ChIP (grey bars) and ΔPcvelA (black bars) compared to wild type P2niaD18 are shown. Values are the mean score of three biological replicates.

Figure 5: Multiple sequence alignments of methyltransferase sequence motifs I-III. Alignment of amino acid sequences revealed a high degree of conservation between methyltransferase sequence motifs I-III from putative SAM-dependent methyltransferases PcLaeA and PcLlmA, PcVipC, and the one encoded by Pc21g12700. Alignments were visualized using Jalview according to the ClustalX color scheme (http://www.jalview.org/). Numbers indicate amino acids that separate conserved motifs.

Figure 6: ChIP-seq reveals a PcVelA consensus-binding site. PcVelA-specific peak regions were submitted to MEME (Bailey and Elkan 1994) for de novo motif prediction. Only the most significant putative DNA-binding motif, PcVelA.M1, is shown. For comparison against the JASPAR CORE (2014) database, results were submitted to TOMTOM (Gupta et al. 2007), using default parameters. The top two matches to known DNA-binding motifs from vertebrates are given. The associated proteins, IDs from the JASPAR CORE database, p-values, and E-values are indicated. The size of each letter is proportional to the frequency of each nucleotide at this position within the consensus sequence. Motifs are centered on common central nucleotides.

Figure 7: Electrophoretic mobility shift assays (EMSAs) confirm PcVelA-binding to the predicted DNA-binding consensus sequence PcVelA.M1. (A) Zoomed ChIP-seq profiles from PcVelA_shaking1, PcVelA_shaking2, and input control next to Pc21g02240, encoding the putative SAM-domain methyltransferase PcLlmA are shown. Positions of oligonucleotides used for shift analysis (black bars) and occurrences of PcVelA.M1 (red arrows) are indicated. Orientation of ORFs (black boxes) next to PcVelA ChIP-seq peak regions are indicated by arrowheads. (B) EMSAs were performed using 50 nt radiolabeled double stranded oligonucleotide probes (PcLlmA_2, PcLlmA_4) derived from the PcllmA promoter region and rising amounts of purified GST-PcVelA1-256 protein. Positions of free probe (*) and protein-DNA complexes (Æ) are indicated. Single-bp substitutions within PcVelA.M1 in oligonucleotides PcLlmA_2_m and PcLlmA_4_m resulted in a diminished formation of protein-DNA complexes.

Figure 8: PcVelA directly interacts with the putative SAM-dependent methyltransferase PcLlmA. (A) For yeast two-hybrid analysis, diploid strains were spotted on selective media lacking adenine and histidine and supplemented with X-α-Gal to demonstrate ADE2 and HIS3 as well as lacZ reporter gene activity. (B) For BiFC analysis, genes encoding PcVelA and PcLlmA were fused to eyfp fragments encoding either the N- or the C-terminus of the yellow fluorescent protein, and strains harboring both constructs were analyzed using fluorescence microscopy. DAPI straining confirmed nuclear localization of the PcVelA-PcLlmA interaction. As a control, strains producing either both split EYFPs or one split EYFP together with EYFP-PcVelA/PcLlmA-EYFP are shown.

Table 1: ChIP-seq design and results Estimated % # Peaks FDR # Differential # Total Sample # Readsa # Mappedb fragment Mappedc ≤ 0.001d peakse peaksf lengthg PcVelA_shaking1 34.074.601 20.835.894 61.15% 6088 1937 764 235 PcVelA_shaking2 29.736.045 17.177.895 57.77% 6090 1362 1001 231 PcVelA_shaking_input 20.383.512 18.540.910 90.96% - - - - a total number of sequenced reads b total number of reads mapped to P. chrysogenum P2niaD18 genome c fraction of tags found in peaks versus genomic background determined by HOMER d number of peaks passing FDR ≤ 0.001 threshold e number of peak regions showing at least fourfold enrichment in ChIP-sample compared to input-DNA f total number of peak regions after local background filtering and clonal filtering g estimated fragment length used for sequencing determined from tag auto correlation analysis

Table 2: Selected PcVelA target genes identified using ChIP-seq analysis Microarray Peak Identifier Descriptiona Proposed function ΔPcVelAc valueb 48 h 60 h 96 h Development and conidiation over-expression causes a severe delay in the Pc18g03940 14-3-3 family protein ArtA polarization of conidiospores in A. nidulans (Kraus et al. 1090 -0.1 0 -0.1 2002) inactivation reduces expression of the penicillin gene cluster in P. chrysogenum (Sigl et al. 2011); cell pattern formation- Pc13g04920 required for differentiation and spatial organization of 1034 -0.1 0.4 -0.1 associated protein StuA cell types of the A. nidulans conidiophore (Miller et al. 1992) deletion mutant conidia show significant sensitivity to high temperature and oxidative stress in A. fumigatus bZIP transcription factor Pc13g09580 (Hagiwara et al. 2014); 1027 -0.9 -0.7 -1.1 AtfA regulates different types of stress responses in A. nidulans (Balazs et al. 2010) expressed during the formation of asexual spores or Pc16g03240 conidiation protein Con-6 after illumination of vegetative mycelia in N. crassa 801 0.9 0.8 0.3 (Olmedo et al. 2010) gene product is localized in the conidiospore wall; α,α-trehalose Pc16g11870 required for growth on trehalose as a carbon source in 761 -0.1 -0.2 -0.3 glucohydrolase TreA/Ath1 A. nidulans (d'Enfert and Fontaine 1997) functions at the cell cortex as an upstream effector of related to integral Pc21g09870 appressorium differentiation in M. grisea (DeZwaan et 746 0 1.5 1.6 membrane protein Pth11 al. 1999) spore-wall fungal encodes a fungal hydrophobin component of the Pc16g06690 hydrophobin DewA 730 -0.1 0.2 -0.2 conidial wall (Stringer and Timberlake 1995) precursor putative nuclear TF necessary for proper activation of C H conidiation Pc12g12190 2 2 conidiation, growth and development in A. nidulans 613 -0.5 0.3 -0.2 transcription factor FlbC (Kwon et al. 2010) associated with tolerance to water stress resulting LEA (late embryogenesis Pc13g09910 from desiccation and cold shock in plants and animals 560 -0.6 0 -0.8 abundant) domain protein (Goyal et al. 2005, Chakrabortee et al. 2012) MYB family conidiophore regulates both asexual and sexual differentiation in A. Pc13g03170 553 0.9 0.5 0.2 development protein FlbD nidulans (Arratia-Quijada et al. 2012) acts as an activator of conidiospore formation in developmental regulator Pc22g22320 various filamentous fungi (Bayram et al. 2008, Kopke et 506 0.2 0.6 0.4 VelB al. 2013) mediates developmental switch from apical growth of C H type conidiation Pc23g00400 2 2 vegetative cells to budding growth pattern of 396 -0.1 1.2 0.2 transcription factor BrlA conidiophores (Adams et al. 1988) Secondary metabolism Pc22g06500 amino acid transporter 2544 0.9 1.0 1.2 Pc22g17530 ABC multidrug transporter aa5 2154 0 -0.3 1.5 Pc22g06610 neutral amino acid permease 1585 0.2 0.6 1.4 Pc20g05090 ABC multidrug transporter 1262 1.0 0.4 0.2 Pc16g11480 PKS, putative 870 -2.7 -3.7 -4.3 Pc16g11470 ABC multidrug transporter 870 0.3 -1.2 0.5 Pc20g03900 MFS multidrug transporter 716 0.4 1.2 1.3 Pc18g00380 hybrid NRPS PKS 644 0.1 0.5 0.1 Pc21g12630 NRPS 621 -2.6 -1.5 -0.1 Pc20g12260 ABC drug exporter AtrF 538 0.3 0.7 1.1 Pc12g14890 MFS multidrug transporter 538 0.3 0.8 3.4 Pc22g22420 MFS transporter 528 1.3 1.5 4 Pc18g03610 ABC multidrug transporter 481 -0.1 0 2.8 Pc18g03610 ABC multidrug transporter 481 -0.1 0 2.8 Pc21g08920 norsolorinic acid reductase 343 -0.2 0.5 1.9 Acyl-CoA-related processes acetyl-CoA synthetase-like Pc22g24780 1291 -0.1 0.3 1.3 protein loss-of-function mutations result in resistance to fluoroacetate in the absence of a repressing carbon acetyl-coenzyme A Pc22g06680 source, which otherwise inhibits development, 1150 0.2 0 0.3 synthetase FacA conidiation, and conidial pigmentation in A. nidulans (Hynes and Murray 2010) acetyl-CoA- Pc22g00420 923 0.5 0 0.6 acetyltransferase 34

acyl-CoA N- Pc16g03600 606 0 1.4 1.9 acyltransferase acyl-CoA N- Pc21g08470 467 -1.5 -1.2 -2.1 acyltransferase Transcription factors

Pc20g05960 C2H2 transcription factor 1705 -0.7 -1.0 -1.7 Pc21g15330 bZIP TF 1040 -1.2 -1.1 -1.9

Pc06g02030 C2H2 finger domain protein 986 -3.1 -3.2 -2.3 Pc15g00130 F-box domain protein 885 -1.0 -0.2 -0.9 Pc21g15330 bZIP TF 725 -1.2 -1.1 -1.9 Pc12g10080 C6 finger domain protein 576 -1.4 -0.6 0 Methyltransferases LaeA-like SAM-dependent Pc21g02240 2756 2.6 1.6 0.8 methyltransferase PcLlmA part of a membrane-associated trimeric complex that LaeA-like SAM- controls a signal transduction pathway for fungal 2139 Pc18g01840 dependent -2.6 -2.1 -1.7 differentiation in A. nidulans (Sarikaya-Bayram et al. 1430 methyltransferase PcVipC 2014) SAM-dependent Pc21g12700 1101 -1.4 -0.4 5 methyltransferase SAM-dependent Pc18g04780 660 0.1 0.6 0.6 methyltransferase Pc18g06010 O-methyltransferase 470 0.1 0 0.1 nicotinamide Pc13g15570 334 0.8 0.7 1.0 N-methyltransferase Pc22g01170 O-methyltransferase 328 4.5 4.6 4.5 a as obtained from blastp analysis (http://blast.ncbi.nlm.nih.gov/Blast.cgi) b statistical peak value = average tag count found at peak normalized to 10 million total mapped tags c microarray data showing expressional changes in ΔPcvelA compared to wild type ΔPcku70 after 48, 60, and 96 h of cultivation (Hoff et al. 2010)

MATERIALS AND METHODS Strains and culture conditions. Penicillium chrysogenum strains (Table S1) were grown in CCM (Wolfers et al. 2014) shaking or surface cultures at 27°C. For inoculation, 0.5x107 spores derived from cultures grown on M322 solid medium (Wolfers et al. 2014) for 4-5 days were used. Escherichia coli strain XL1blue was used for cloning and plasmid propagation purposes, while BL21 (DE3) served as a host for heterologous overexpression of PcVelA- GST (Bullock et al. 1987, Miroux and Walker 1996). Saccharomyces cerevisiae strains PJ69-4a and PJ69-4α were used for yeast two-hybrid analysis (James et al. 1996). Strains were grown at 30°C on SD medium lacking selected amino acids used for auxotrophy marker selection. Mating of PJ69-4a and -α strains was performed in liquid YPDA medium at 30°C and 50 rpm.

Construction of P. chrysogenum strains. Strains were constructed by ectopic or homologous integration of plasmid DNA (Table S2) as described previously (Hoff et al. 2010, Kamerewerd et al. 2011) with some modifications. Recipient strains were grown for 72 h in shaking cultures and protoplasts were transformed with either circular (for ectopic integration) or linear (for homologous recombination) plasmid DNA. Transformants were selected on CCM media containing 150 μg/ml nourseothricin (Werner BioAgents, Germany). Resistant colonies were isolated and tested for integration of plasmid DNA. PCR analysis and SDS- PAGE/Western blot analysis was performed as described previously (Hoff et al. 2010).

Nucleic acids isolation, cDNA synthesis, qRT-PCR, and ChIP-PCR. Isolation of nucleic acids, cDNA synthesis, qRT-PCR, and ChIP-PCR analysis was carried out as described earlier (Hoff et al. 2009, Böhm et al. 2013, Becker et al. 2015). Oligonucleotides are listed in Table S3.

Sample preparation for ChIP-seq, data analysis and visualization. ChIP and analysis of sequencing data was carried out as previously described (Becker et al. 2015), using Bowtie version 1.0.1 (Langmead et al. 2009), SAMtools (Li et al. 2009), the Integrative Genomics Viewer (IGV) (Thorvaldsdóttir et al. 2012), MEME (Multiple Em for Motif Elicitation; http://meme.nbcr.net/meme/) (Bailey and Elkan 1994), TOMTOM (Gupta et al. 2007), and the HOMER software for motif discovery and next-generation sequencing analysis (Heinz et al. 2010). Raw sequencing data is available from the NCBI SRA database; study ID n.a., Accession # n.a..

Electrophoretic mobility shift assays (EMSAs). Gel shift assays were performed using oligonucleotides derived from ChIP-enriched regions and purified GST-PcVelA1-256. 50 nt double-stranded oligonucleotides (Table S3) were 5’-end-labeled using polynucleotide kinase (Roche, Basel, Switzerland) and [ɣ-32P]-ATP (Hartmann Analytic, Braunschweig, Germany). For shift experiments, 3.5-7.0 fmol (~50 – 100 cps) of radiolabeled oligonucleotides were incubated with varying protein concentrations in the presence of 2 μl binding buffer (250 mM Tris/HCl pH 8.0, 1 M KCl, 50 % glycerol) and 1 μg poly(dI-dC)-poly(dI-dC) (Affymetrix USB, CA, USA) in a total volume of 20 μl for 20 min at room temperature. Samples were run on 5 % polyacrylamide gels at 4°C in 190 mM glycine, 27 mM Tris/HCl pH 8.5. 36

Expression, purification and immunodetection of recombinant PcVelA-GST protein. Purification of recombinant PcVelA-GST protein from E. coli was performed as described earlier (Janus et al. 2007) using an elution buffer containing 50 mM Tris/HCl, 30 mM reduced glutathione, 100 mM NaCl, pH 8.0. Western blotting and immunodetection was performed using RPN1236 anti-GST HRP conjugate (GE Healthcare, Germany).

Yeast two-hybrid analysis. Yeast two-hybrid analysis was carried out as described by Kopke et al. (2013) using yeast strain PJ694a for Gal4 activation domain (AD) fusion derivatives and strain PJ69-4α for Gal4 DNA-binding domain (BD) fusion constructs.

Microscopy. Fluorescence and light microscopy was carried out as described previously (Engh et al. 2007, Hoff et al. 2010) with minor modifications. Images were captured with a Photometrix Cool SnapHQ camera (Roper Scientific, USA) and Metamorph (version 7.7.5.0; Universal Imaging). Recorded images were processed with MetaMorph and Adobe Photoshop CS4. Staining of nuclei was performed using DAPI (Sigma Aldrich, Germany).

SUPPLEMENTS

Table S1: P. chrysogenum strains used in this work

Strain Characteristics and genotype Source P2niaD18 niaD- (Hoff et al. 2008) − PcVelA-ChIP (T16.2) Pgpd::PcVelA::EGFP::TtrpC; nat1; niaD This study ΔPcvelAab Pcku70::FRT; PcVelA::FRT; niaD− (Kopke et al. 2013) ∆Pcku70b Pcku70::FRT; niaD- (Kopke et al. 2010)

BiFC-N/N Pgpd::eyfpC::TtrpC; Pgpd::eyfpN::TtrpC; PtrpC::nat1; niaD− This study BiFC-N/PcLlmA Pgpd::eyfpC::TtrpC; Pgpd::PcllmA::eyfpC::TtrpC; PtrpC::nat1; niaD− This study BiFC-PcVelA/N Pgpd::PcvelA::eyfpN::TtrpC; Pgpd::eyfpN::TtrpC; PtrpC::nat1; niaD− This study BiFC-PcVelA/PcLlmA Pgpd::PcvelA::eyfpN::TtrpC; Pgpd::PcllmA::eyfpC::TtrpC; PtrpC::nat1; niaD− This study a strains that still carry a resistance marker from the flipper knockout construct b strains without any resistance markers due to FLP/FRT marker recycling.

Table S2: Plasmids used in this work

Name Characteristics Source

pPcVelA-EGFP Pgpd of A. nidulans, egfp, PcvelA gene of P. chrysogenum, TtrpC of A. nidulans, nat This study resistance gene of Streptomyces noursei; used for construction of P. chrysogenum ChIP-strains via ectopic integration into strain ΔPcVelA

pEYFPC-nat gpd promoter of A. nidulans, eyfpC-fragment (aa 155-238), trpC terminator of (Hoff et al. 2010) A. nidulans, nat1 gene

pEYFPN-nat gpd promoter of A. nidulans, eyfpN-fragment (aa 1-154), trpC terminator of (Hoff et al. 2010) A. nidulans, nat1 gene pYNVELA PcvelA ORF in NotI site of pEYFPN-nat (Hoff et al. 2010) pYCLLMA PcllmA ORF in NcoI and NotI site of pEYFPC-nat (Hoff et al. 2010) pGADT7 ADH1(p)::gal4 AD::LEU2 Clontech pGBKT7 ADH1(p)::gal4 BD::TRP1 Clontech pAD-PcvelA PcvelA cDNA in SmaI and SacI site in pGADT7 (Kopke et al. 2013) pAD-PcllmA PcllmA cDNA in EcoRI and BamHI site in pGADT7 This study pAD-PclaeA PclaeA cDNA in EcoRI and XhoI site in pGADT7 (Kopke et al. 2013) pBD-PcvelA PcvelA cDNA in SmaI and SacI site in pGBKT7 (Kopke et al. 2013) pBD-PcllmA PcllmA cDNA in EcoRI and BamHI site in pGADT7 This study pBD-PclaeA PclaeA cDNA in EcoRI and PstI site in pGBKT7 (Kopke et al. 2013)

Table S3: Oligonucleotides used in this work

Name Sequence (5’ to 3’) Specificity Plasmid and strain egfp_r ACTTCAGGGTCAGCTTGC egfp gene construction PcvelA_f TCGGTCGACATGGCCAACAGACCATCTC PcvelA gene ChIP-PCR qPCR_NC1_f TTCTTCCGCAATCAAGCTCA chr1:5375234-5375253 qPCR_NC1_r GAAAAATTGCCGCTGGACTC chr1:5375364-5375383 qPCR_NC2_f GGTCGTTGATTCCCTTGAGC chr2:7621179-7621198 qPCR_NC2_r GGATCGGATTATTCGGGTGA chr2:7621294-7621313 qPCR_Pc21g02240_f CGAGAGAGAGGAACCCGGGA chr2:5277047-5277066 qPCR_Pc21g02240_r TTTCCCGTACCAGGCTGTCG chr2:5277176-5277195 qPCR_Pc22g17530_f AGGCACCGAAACCGTGAAGA chr1:2788884-2788903 qPCR_Pc22g17530_r ACGCCAGGCCAGAGTTCAAT chr1:2788799-2788818 qPCR_Pc20g02880_f CGTGAAATTCGAAGGTTCCCGA chr2:2995008-2995029 qPCR_Pc20g02880_r AGAAATTAAGCCGCAAAACCCAGA chr2:2994880-2994903 qPCR_Pc20g14090_f GTGGAAATTTCGGATGGGGTAGC chr2:378180-378202 qPCR_Pc20g14090_r GATGCCCTGGTATCGGCAAAA chr2:378053-378073 qRT-PCR qRT_Pc21g02240_f ACAAGGAAATCGGTCGCATC chr2:5275851-5275870 qRT_Pc21g02240_r GCCCTCTCCATATGCTCCTG chr2:5275747-5275766 qRT_Pc18g01840_f TTCGGCAAGGACATGACATC chr1:5930185-5930204 qRT_Pc18g01840_r TGGTACCGACCAAGCTCCTT chr1:5930361-5930380 qRT_Pc21g12700_f GGGTTTGTCGATACCCAGGA chr2:7743588-7743607

qRT_Pc21g12700_r CCGCAGTCCACTGATGGTAA chr2:7743677-7743696 qRT_Pc18g04780_f TGATGATGCGAAGACCATCC chr1:6636202-6636221 qRT_Pc18g04780_r TGAACCAATGCACCAGCTCT chr1:6636094-6636113 qRT_Pc18g06010_f CCCCGATACCAACGCATACT chr1:6920825-6920844 qRT_Pc18g06010_r CGTGATCTTCAACCCAGCAG chr1:6920718-6920737 qRT_Pc13g15570_f GGACCCGAACTCTGTTGCTC chr4:2949445-2949464 qRT_Pc13g15570_r GCATCCACCACCTTCTCAAA chr4:2949330-2949349 qRT_Pc22g01170_f TCGCTCGCTTCCTTGTATGA chr3:5122377-5122396 qRT_Pc22g01170_r CAGGACTCGCAGACCAACAG chr3:5122469-5122488 EMSAsa PcLlmA_2 TAGCGTCATTTATTTTTTTCTTCCAAGGTTTTTCCCTCTTCTT chr2:5277074-5277122 CGGAGT PcLlmA_2_m TAGCGTCATTTATTTTTTTCcTtCAgaGTTTTTCCCTCTTCTTC chr2:5277074-5277122 GGAGT (with mutations) PcLlmA_4 TCCGACAGCCTGGTACGGGAAACCTTGGAACCCATTCCAA chr2:5277174-5277222 ATCGGTCTG PcLlmA_4_m TCCGACAGCCTGGTACGGGAAACtcTGaAgCCCATTCCAAA chr2:5277174-5277222 TCGGTCTG (with mutations) a in case of double-stranded oligonucleotides used for EMSAs, only the sense sequences are given.

V. DISCUSSION 25

V. DISCUSSION

For long, functional analysis of transcriptional regulators in P. chrysogenum and many other species has been restricted to the characterization of single genes or gene families, e.g. by using deletion and overexpression mutants. Even though these studies contributed in many ways to the investigation of fundamental principles of gene regulation, they only presaged the whole dynamics of genome-wide GRNs, which control all kinds of cellular and developmental processes. Accordingly, in order to improve our current knowledge of transcriptional regulation of secondary metabolism and morphogenesis in the industrially highly relevant filamentous fungus P. chrysogenum, a ChIP-seq approach on two regulatory proteins, namely MAT1-1-1 and PcVelA, has been performed within the scope of this work.

1. ChIP-seq analyses with MAT1-1-1

For the most part, our general understanding of mating-related processes in ascomycetes is based on knowledge obtained from studies using the baker’s yeast S. cerevisiae and the filamentous fungus N. crassa (Ni et al. 2011). In S. cerevisiae, sex is determined by two alternative MAT loci, namely MATα and MATa, which consist of dissimilar sequences occupying the same locus on the chromosome (Astell et al. 1981). These sequences are termed idiomorphs to indicate that they do not represent the alleles of a single gene (Metzenberg and Glass 1990). In N. crassa, MAT loci are designated mata and matA, whereas in most other euascomycetes the terms MAT1-1 and MAT1-2 are used (Coppin et al. 1997, Pöggeler 2001). MAT loci from ascomycetes harbor one or more open reading frames (ORFs), of which at least one codes for a MAT TF (Kück and Böhm 2013). As a rule, the MAT1-1 locus encodes an α-domain TF and the alternative idiomorph, MAT1-2, is characterized by a gene coding for a TF carrying a high-mobility group (HMG)-domain. The corresponding genes are generally referred to as MAT1-1-1 and MAT1-2-1 (Turgeon and Yoder 2000, Lee et al. 2010).

While MAT genes were shown to control mating in all sexually reproducing ascomycetes (Coppin et al. 1997, Merlini et al. 2013), a total of 64 % and 73 % of all described Aspergillus and Penicillium species, respectively, are still lacking known sexual states (Dyer and O'Gorman 2011). Nevertheless, the presence of fully functional, constitutively transcribed MAT genes suggests that many of these supposed 'asexual' species do indeed have the potential to undergo sexual reproduction under appropriate environmental conditions. Until V. DISCUSSION 26 now, cryptic sexuality has been described for a growing number of fungi, which had been considered to be asexual since no sexual propagation was observed under laboratory conditions for a very long time. Prominent examples include the fungal pathogens Candida albicans, Aspergillus flavus, Aspergillus parasiticus, and A. fumigatus (Magee and Magee 2000, Horn et al. 2009a, Horn et al. 2009b, O'Gorman et al. 2009), as well as the biotechnologically relevant species P. chrysogenum, Penicillium roqueforti, and T. reesei (Seidl et al. 2009, Böhm et al. 2013, Ropars et al. 2014).

Functional MAT genes and homologs of pheromone and pheromone-receptor genes have been first described in P. chrysogenum in 2008 (Hoff et al. 2008). Since then, huge efforts have been made in order to elucidate the ability of P. chrysogenum to undergo sexual mating and, in 2013, the sexual life cycle of P. chrysogenum, leading to the production of recombinant ascospores, has been described (Böhm et al. 2013, Böhm et al. 2015). Furthermore, phenotypic characterization of MAT1-1-1 and MAT1-2-1 deletion and overexpression strains provided evidence for an involvement of MAT proteins in regulation of cellular and developmental processes other than mating. For example, deletion of MAT1-1-1 was shown to lead to a drastic reduction in penicillin biosynthesis and an increased formation of asexual conidiospores, whereas both deletion and overexpression of the gene resulted in the formation of significantly larger pellets compared to wild type (Böhm et al. 2013). On the contrary, MAT1-2-1 was shown to be involved in regulation of conidiospore germination, light-dependent asexual sporulation, and determination of surface properties of conidiospores (Böhm et al. 2015).

1.1 MAT1-1-1 regulates target genes beyond sexual development

A comprehensive ChIP-seq approach, enabling the generation of a genome-wide MAT1-1-1 DNA-binding profile and identification of 254 putative direct MAT1-1-1 target genes, was described within this work (see section III). Interestingly, some of the most highly bound regions in ChIP experiments occurred near homologs of known functional targets of the S. cerevisiae MATα1 protein, such as Pcppg1, a homolog of MFα1, encoding the α-pheromone, and Pcpre1, a homolog of STE3, coding for the a-factor receptor (Ammerer et al. 1985, Galgoczy et al. 2004). In combination with data obtained from qRT-PCR analyses, this observation not only demonstrated MAT1-1-1-mediated regulation of these highly conserved target genes in P. chrysogenum, but also confirmed biological significance of the presented ChIP-seq dataset. It is consistent with reports demonstrating that expression of pheromone-precursor genes and most probably receptor genes is controlled by expression of V. DISCUSSION 27

MAT genes in heterothallic species (Kim and Borkovich 2006), and that sexual reproduction correlates significantly with an increased expression of MAT genes and key genes of a pheromone-response MAP-kinase signaling pathway in A. nidulans and N. crassa (Paoletti et al. 2007, Wang et al. 2014). All in all, findings presented within this work fit the notion of MAT1-1-1 being a positive regulator of sexual reproduction and a negative regulator of asexual development in P. chrysogenum. This is consistent with the recent discovery that sporulation was increased by about 25 % in a ΔMAT1-1-1 strain compared to wild type, and that ΔMAT1-1-1 strains were sterile when crossed to a fertile MAT1-2 isolate (Böhm et al. 2013).

Besides known target genes of MAT TFs, ChIP-seq and downstream analyses identified a large number of new MAT1-1-1 target genes, which have never been related to any MAT TF before. A large part of these genes was assigned to the functional categories asexual development, morphogenesis, amino acid and secondary metabolism, as well as iron metabolism. It is noteworthy that for most of these processes indications for a relation to fungal sexual development can be found in the literature, but no experimental evidence for a direct regulatory impact of MAT encoded TFs has been reported until today. Two of the most diverse groups, morphogenesis and asexual development, contained genes related to the formation of conidiospores, hyphal growth, polarization of germinating conidiospores and hyphae, as well as surface hydrophobicity. These genes can be used to explain the phenotypic characteristics of MAT1-1-1 mutant strains that have previously been described by Böhm et al. (2013). Within this context, one of the most promising new MAT1-1-1 target genes is dewA, which codes for a fungal hydrophobin, likely to be involved in the formation of extraordinarily large pellets in MAT1-1-1 overexpression and deletion mutants.

It is known that the developmental decision between sexual and asexual reproduction as well as coordination of secondary metabolism is dependent on environmental factors, such as nitrogen sources, iron supply, pH, and culture conditions (Han et al. 2003, Bayram and Braus 2012). Accordingly, within the scope of this work, MAT1-1-1 was shown to directly regulate a number of genes assigned to iron transport and iron acquisition. Most importantly, sidD, encoding a non-ribosomal siderophore-peptide synthetase important for biosynthesis of the intracellular siderophore triacetylfusarinine C (TAFC) (Schrettl et al. 2007), was identified as a specific MAT1-1-1 target gene. It has previously been described that fungal intracellular siderophores are essential for asexual and sexual reproduction (Johnson 2008). For example, deletion of sidC, a gene involved in biosynthesis of the intracellular siderophore ferricrocin, V. DISCUSSION 28 resulted in a delayed germination of conidia, reduced production of asexual spores, as well as blocked sexual development in A. nidulans (Eisendle et al. 2003, Eisendle et al. 2006). Furthermore, in A. fumigatus, deletion of sidC revealed that intracellular siderophores are required for conidial germ-tube formation, asexual sporulation, and conidial catalase A activity (Schrettl et al. 2007), and in the heterothallic ascomycete Cochliobolus heterostrophus, deletion of the siderophore-encoding nps2 resulted in defective ascus and ascospore development (Oide et al. 2007). Besides the observation that MAT1-1-1 is involved in regulation of iron metabolism, identification of meaB, encoding a TF involved in the regulation of nitrogen-dependent gene expression in A. nidulans (Wong et al. 2007), as a specific MAT1-1-1 target gene, established a connection between MAT protein regulatory functions and nitrogen utilization in P. chrysogenum.

It has been hypothesized by Ádám et al. (2011) that MAT genes are functionally retained even during the asexual part of the life cycle and the apparent absence of a sexual phase, presumably because of their positive selective impact on important processes unrelated to sexual development. Accordingly, a direct connection between MAT1-1-1 and regulation of penicillin biosynthesis, which was shown to be significantly down-regulated in a ∆MAT1-1-1 strain compared to wild type, has been demonstrated (Böhm et al. 2013). Within this work, a possible explanation for this observation was provided by the identification of MAT1-1-1 target genes encoding a cyanide hydratase/nitrilase as well as an enzyme catalyzing the chemical reaction of L-homocysteine to L-methionine. As both encoded enzymes are likely to influence biosynthesis of L-cysteine, they might have a direct impact on penicillin biosynthesis, which starts with the formation of a tripeptide based on L-cysteine, L-valine, and L-α-aminoadipic acid (Figure 4).

Although validation of the overall biological significance of large ChIP-seq datasets is generally a challenging task, comprehensive qRT-PCR analyses and DNA-binding studies provided experimental evidence for – at least selected – MAT1-1-1 target genes, covering the functional categories mentioned above. Furthermore, functional characterization of the new MAT1-1-1 target gene artA, encoding a protein that was shown to be necessary for conidiospores germination, provided experimental evidence for MAT1-1-1 downstream factors being involved in processes others than mating. All in all, findings presented within this work are strongly supporting the hypothesis that MAT1-1-1 functions on a genome-wide level are more far ranging than expected. This hypothesis is in accordance with previous genome-wide gene-expression analyses in various euascomycetes, revealing that the number V. DISCUSSION 29

Figure 4: Schematic overview of penicillin biosynthesis. During the first step of penicillin biosynthesis L-α-aminoadipic acid, L-cysteine, and L-valine are condensed into a tripeptide. The condensation reaction is catalyzed by the enzyme δ-(L-α- aminoadipyl)-L-cysteinyl-D-valine synthetase (ACVS), a non-ribosomal peptide synthetase. The second step involves the oxidative conversion of linear ACV into the bicyclic intermediate isopenicillin N by isopenicillin N synthase (IPNS). Finally, the α-aminoadipyl side-chain of isopenicillin N is removed and exchanged by a phenylacetyl side-chain through the enzyme acyl-coenzyme A:isopenicillin N acyltransferase (Brakhage et al. 2004). (kindly provided by Dr. S. Bloemendal) of MAT-regulated genes is rather high (Table 5). For example, a total of 2,421 genes are expressed in a MAT1-1-1-dependent manner in P. chrysogenum (Böhm et al. 2013). Nevertheless, when including corresponding data from hemiascomycetes, a striking discrepancy between MAT-dependent gene expression in euascomycetes and hemiascomycetes can be observed. For example, ChIP-chip analysis in S. cerevisiae haploid a–cells and α–cells, as well as diploid a/α–cells identified a total of six a-specific genes (asgs), five α-specific genes (αsgs), and 19 a/α-specific genes (Galgoczy et al. 2004). Remarkably, with the exception of one αsg, all of these genes were shown to be related to some aspect of mating, such as pheromone signaling, mating-cassette recombination, pheromone-induced cell-cycle arrest, and agglutination. Based on this observation, one can assume that MAT TF target genes in hemiascomycetes are restricted to genes relevant for mating, whereas MAT-mediated regulation beyond sexual development might be a common feature in euascomycetes. Accordingly, deletion of the MAT1-2-1 gene in Fusarium verticillioides was shown to lead to a drastic reduction in carotenoid production, paralleled with a severe decrease in photo-induced expression of genes encoding key enzymes of the carotenoid biosynthesis pathway (Ádám et al. 2011). Moreover, MAT1-1-1 and MAT1-2-1 deletion mutants were shown to be reduced in virulence although none of the MAT locus genes was important for plant infection, indicating that MAT1-1-1 and MAT1-2-1 genes may play a host-specific role in colonization of corn stalks (Zheng et al. 2013). Another interesting example for MAT TFs acting outside the sexual life cycle can be found in the basidiomycete and human fungal pathogen Cryptococcus neoformans. Here, the heterodimer of MAT TFs Sxi2a and Sxi1α (Sex inducer 2a / Sex inducer 1α) was shown to be involved in regulation of several well-studied virulence genes (Mead et al. 2014). V. DISCUSSION 30

Table 5: MAT-dependent gene expression in euascomycetes and hemiascomycetes

Species Experimental approach Number of regulated genes Reference Euascomycetes S. Krappmann, RNA-seq ∆MAT1-1/∆MAT1-2 214 (MAT1-1) Aspergillus fumigatus personal vs WT 729 (MAT1-2) communication 596 (MAT1-1) Microarray after idiomorph Aspergillus oryzae 559 (MAT1-2) (Wada et al. 2012) replacement (only downregulated) Reverse Northern and 171 Fusarium graminearum (Lee et al. 2006) Microarray ∆MAT1-2 vs WT (only downregulated) Fusarium verticillioides Microarray ∆MAT1-2-1 vs WT 248 (Keszthelyi et al. 2007) 44, 61, and 469 (MatA)a 233, 159, and 744 (Mata)a Microarray during asexual Neurospora crassa (only upregulated in (Wang et al. 2012) development comparison to opposing mating type) Penicillium chrysogenum Microarray ∆MAT1-1-1 vs. WT 2,421 (Böhm et al. 2013) Podospora anserina Microarray mat+ vs mat− 167 (Bidard et al. 2011) Sordaria macrospora Microarray ∆Smta-1 vs WT 107 (Pöggeler et al. 2006) Cross-species microarray 978 (SmtA-1) Sordaria macrospora analysis of ΔSmtA-1 and (Klix et al. 2010) 854 (SmtA-2) ΔSmtA-2 Hemiascomycetes ChIP-chip MATα2 6 Saccharomyces cerevisiae ChIP-chip MATα1 5 (Galgoczy et al. 2004) ChIP-chip MATa1-MATα2 19 Microarray analysis with 2 (MATα1) Candida albicans strains of different MAT loci (Tsong et al. 2003) 2 (MATa2) configuration Lachancea kluyveri ChIP-chip MATa2 9 (Baker et al. 2012) Microarray on genes whose induction by Ste11p 12 (MAT1-M) (Mata and Bahler Schizosaccharomyces pombe overexpression is cell 4 (MAT1-P) 2006) type-dependent a) after 36, 60, and 96h of cultivation

1.2 Rewiring of MAT-regulated transcriptional networks

Within this work, the MAT1-1-1 DNA-bindi ng motif “CTATTGAG”, designated MAT1.1, was identified. Interestingly, MAT1.1 shows close similarity to the cis-regulatory sequence “TCATTGAT”, which has previously been described for αsgs in S. cerevisiae (Hagen et al. 1993, Baker et al. 2011). Furthermore, a high degree of conservation of MAT1.1 within promoter sequences of αsgs in the euascomycetes A. fumigatus, A. nidulans, F. graminearum, T. reesei, and N. crassa was demonstrated, whereas only moderate conservation was observed in promoter regions from S. cerevisiae, and as good as no conservation was documented within promoter regions from C. albicans. Taken together, these findings are in accordance with recent studies demonstrating the conservation of the αsg cis-regulatory sequence from S. cerevisiae and the filamentous fungus Uncinocarpus reesei. Moving of an αsg cis-regulatory sequence from U. reesei into S. cerevisiae resulted in efficient activation of V. DISCUSSION 31 reporter-gene expression by the S. cerevisiae Matα1 protein, whereas only weak activation was documented in C. albicans (Baker et al. 2011). Based on this observation, it has been hypothesized that S. cerevisiae, C. albicans, and filamentous ascomycetes may share a common ancestor with an α-domain MAT protein DNA-binding specificity similar to that of the modern S. cerevisiae Matα1 protein. Accordingly, DNA-binding specificity of the protein would have changed so extensively that its cis-regulatory sequence appears different even in related species, such as S. cerevisiae and C. albicans. However, as Matα1 still controls the same core set of mating-related genes in various fungi, Matα1 and its DNA-recognition site seem to have evolved together, preserving the protein-DNA interaction but significantly changing its molecular details (Baker et al. 2011). Comparable results were described in studies focusing on the evolution of gene regulation by the highly conserved transcriptional regulator Mcm1 (Tuch et al. 2008). Mcm1 is a founding member of the MADS-box family of TFs and an essential protein involved in regulation of diverse cellular and developmental processes, such as the cell cycle, osmotic regulation, and arginine metabolism in yeast (Shore and Sharrocks 1995, Carr et al. 2004). Furthermore, S. cerevisiae Mcm1 is necessary for cell-type-specific transcription and pheromone response, and plays a central role in the formation of repressor and activator complexes in cooperation with the MAT proteins MATα1 and MATα2 (Mead et al. 2002, Pachkov et al. 2007). A direct interaction between Mcm1 and Matα1 was shown to be essential for activation of the expression of yeast αsgs (Bender and Sprague 1987, Jarvis et al. 1989, Carr et al. 2004). Surprisingly, comparison of genes regulated by Mcm1 in the yeasts S. cerevisiae, C. albicans, and Kluyveromyces lactis revealed substantial differences. Most importantly, new Mcm1-cofactor interactions were shown to have evolved along different branches of the yeast lineage, whereas the core Mcm1-cofactor interactions associated with cell cycle and mating remained the same. For example, C. albicans Mcm1 was shown to bind to a new DNA-binding motif within the promoter region of genes involved in the white-opaque phenotypic switch, necessary for adaption within a human host (Kvaal et al. 1997). As the phenotypic switch has been only described in C. albicans and two related pathogenic species, Candida tropicalis and Candida dubliniensis (Pujol et al. 2004, Xie et al. 2012), one can assume that the adaption of Mcm1 DNA-binding properties was crucial to this development.

A comparable scenario, where a MAT-encoded TF acquired additional regulatory features, is also conceivable for P. chrysogenum MAT1-1-1. Here, expansion of MAT1-1-1 regulatory functions during the asexual part of the life cycle, resulting in an involvement in various cellular and developmental processes, such as formation of asexual conidiospores, amino V. DISCUSSION 32 acid, iron, and secondary metabolism, might have been promoted by a slow decline in sexual fertility within the species as a whole (Dyer and Paoletti 2005). This “slow decline” hypothesis is in accordance with the observation that not all P. chrysogenum crosses produce cleistothecia with ascospores and the same efficiency (Böhm et al. 2013). Furthermore, as no information is available about the extent of sexual fertility within natural P. chrysogenum populations, it is imaginable that the adaption of MAT1-1-1 regulatory functions to non-mating-related processes might be an evolutionary step towards preference of an asexual life style. A comparable evolutionary progress has recently been documented for S. cerevisiae, where the elimination of the expression of 23 mating-related genes resulted in a 2 % growth-rate advantage in sterile mutants compared to wild type (Lang et al. 2009). Evidently, further research will be necessary to elucidate the mechanisms leading to rewiring of the GRN governed by MAT1-1-1. In this context, various mechanisms, ranging from mutation or recombination of promoter sequences that might have brought additional genes under control of MAT1-1-1, to changes in MAT1-1-1 itself, which may have altered its binding specificity, activity, or interaction with other factors, have to be taken into account.

Although specificity and functionality of the DNA-binding motif MAT1.1 has been verified within the scope of this work, the exact mechanisms underlying MAT1-1-1-DNA interactions remained unclear. However, comparison of MAT1.1 to known DNA-binding motif sequences present in the JASPAR core (2014) databases revealed noticeable similarities to known DNA-binding consensus sequences from fungi and vertebrates. Within this context, a significant overrepresentation of DNA-binding motifs specific for homeobox- and HMG-box domain proteins from both fungi and vertebrates was observed. While homeobox-domain TFs are renowned for their involvement in regulation of development in higher eukaryotes (Lewis 1978, Svingen and Koopman 2007, Mukherjee et al. 2009, Hay and Tsiantis 2010, Mallo et al. 2010), they have also been shown to fulfill vital regulatory functions during fungal development and differentiation, e.g. in the form of MAT proteins from basidiomycetes (Casselton and Olesnicky 1998, Yan et al. 2007, Haber 2012), during philaide development and conidiogenesis in Fusarium species (Zheng et al. 2012), and as regulators of hyphal morphology and microconidogenesis in Podospora anserina (Arnaise et al. 2001). HMG-domain proteins are eukaryotic DNA-binding proteins, which are characterized by a functional HMG-box, a conserved motif containing approximately 80 amino acids arranged in a distinctive L-shaped three-α-helical fold (Read et al. 1993). Prominent examples for HMG-domain proteins are MAT TFs from various fungi (Glass et al. 1990, Idnurm et al. 2008, Martin et al. 2010, Böhm et al. 2015), as well as the SOX (SRY-type HMG-box) V. DISCUSSION 33 proteins, including SRY (sex-determining region Y), a crucial factor involved in mammalian male sex determination (Gubbay et al. 1990, Giese et al. 1994, Werner et al. 1995). It has for long been recognized that fungal MAT loci share structural and functional features of the mammalian X and Y sex determination system (Kronstad and Staben 1997, Fraser et al. 2004, Idnurm et al. 2008) and evidence for this assumption has recently been provided by Czaja et al. (2014). It was demonstrated that the human SRY protein is able to functionally replace the MAT protein MatA and drive sexual development in A. nidulans (Czaja et al. 2014). Although little is known about fungal TFs and their evolutionary relatedness to TFs in other eukaryotes (Shelest 2008), the observed similarities in DNA-binding consensus sequences of MAT1-1-1 and various homeobox- and HMG-domain proteins from vertebrates and fungi can be taken as a hint to similar DNA-binding properties. This is in accordance with recent work providing evidence for the hypothesis that extant α-box genes originated from an ancestral HMG gene, and that the α-domain should be able to bind DNA in a manner similar to canonical HMG domains (Martin et al. 2010). Further support for this theory comes from the observation that MATα1 is able to bend DNA (Carr et al. 2004), which is an important feature of members of the HMG family of regulatory proteins (Grosschedl et al. 1994, Bianchi and Agresti 2005, Malarkey and Churchill 2012).

Taken together, it appears that MAT-regulated GRNs have undergone drastic reorganization, resulting in the presence of TFBSs in the promoters of – at a first glance – unrelated target genes that are bound and controlled by highly conserved transcriptional regulators in different fungi. Overall, this hypothesis fits the general assumption that transcriptional network rewiring, allowing new regulatory patterns of existing gene products, is a key mechanism by which organismal complexity arises in evolution (Carroll 2000, Levine and Tjian 2003, Tsong et al. 2003, Lavoie et al. 2009, Booth et al. 2010, Li and Johnson 2010). Future research will be necessary in order to determine exactly which changes in MAT1-1-1 and its corresponding DNA-binding site were necessary to allow the enormous expansion in MAT1-1-1 regulatory functions described within this work and when these facets of MAT1-1-1 functions were acquired during evolution. Moreover, additional experiments will be needed to elucidate the exact mechanism of MAT1-1-1 DNA-binding and its transcriptional regulatory activity.

1.3 A new MAT1-1-1 working model

Based on data presented within this work, a new model of MAT1-1-1 action can be proposed. As shown in Figure 5, two levels of MAT1-1-1 regulatory functions can be distinguished: [a] V. DISCUSSION 34 highly conserved, mating-related functions and [b] so far undescribed, non-mating-related functions.

a) Highly conserved, mating-related MAT1-1-1 functions: The expression of genes necessary for sexual development in P. chrysogenum, such as those coding for the α-pheromone precursor PcPpg1, the α-pheromone processing endoprotease Kex1, and the a-pheromone receptor PcPre1, is induced by MAT1-1-1. However, none of the identified MAT1-1-1 target genes assigned to sexual development, except for kex1, shows MAT1-1-1-dependent changes in expression profiles at early developmental stages in a ΔMAT1-1-1 strain, suggesting that these are, up to a certain point, independent of MAT1-1-1. It is conceivable that binding of a so far uncharacterized a-pheromone to the respective a-pheromone receptor, encoded by Pcpre1, initiates a positive feedback loop resulting in an increased expression of MAT1-1-1-dependent genes.

b) New, non-mating-related MAT1-1-1 functions: Transcriptional control mediated by MAT1-1-1 is not restricted to genes encoding highly conserved key elements of sexual reproduction but also affects a considerable number of non-mating genes. These include genes related to asexual development, morphogenesis, secondary, amino acid, and iron metabolism, which can be used to explain the previously described phenotypic characteristics of MAT1-1-1 overexpression and deletion strains (Böhm et al. 2013). In contradiction to mating-related genes, expression levels of many non-mating-related genes were shown to be affected in both MAT1-1-1-deletion and overexpression background. V. DISCUSSION 35

Figure 5: Schematic representation of MAT1-1-1 regulatory functions. Two levels of MAT1-1-1 regulatory functions can be distinguished. One the one hand, MAT1-1-1 acts as the main regulator of sexual development in P. chrysogenum MAT1-1/α-cells (blue). Here, MAT1-1-1 induces the expression of key elements of sexual reproduction, such as Pcpre1, encoding the a-pheromone receptor, and Pcppg1, encoding the α-pheromone. On the other hand, MAT1-1-1 regulates genes with functions beyond the sexual part of the life cycle, e.g. those involved in asexual development and morphogenesis, as well as secondary, iron, and amino acid metabolism. Presumably, binding of the (so far uncharacterized) a-pheromone, produced by MAT1-2/a-cells (red), to the a-pheromone receptor of α-cells initiates a positive feedback loop (marked by +), which leads to an increased expression of MAT1-1-1-regulated genes.

V. DISCUSSION 36

2. ChIP-seq analyses with PcVelA

Members of the velvet family of proteins act as key regulators of secondary metabolism and differentiation processes. They are characterized by the presence of the so-called velvet domain, which is widely distributed within the fungal kingdom (Gerke and Braus 2014). Remarkably, velvet proteins have not been verified in fungi lacking SM gene clusters, such as the yeasts S. cerevisiae and C. albicans (Bayram and Braus 2012).

The founding member of the velvet family, VeA (Velvet A), was firstly described as a light-dependent regulator in A. nidulans (Käfer 1965). Since then, functional characterization of VeA and its homologs in various filamentous fungi confirmed its involvement in terms of sexual and asexual development, morphogenesis, virulence, and secondary metabolism. For example, deletion of veA in A. nidulans leads to defects in the formation of sexual fruiting bodies and abolishes sterigmatocystin production, whereas overexpression results in constitutive formation of cleistothecia, independent of light conditions (Kim et al. 2002, Kato et al. 2003, Calvo 2008). In F. fujikuroi, FfVel1 acts simultaneously as a positive and negative regulator of secondary metabolism, and deletion mutants are characterized by an aberrant formation of conidiospores and reduced virulence (Wiemann et al. 2010). Similarly, Fgve1 deletion mutants in F. graminearum are characterized by hyperbranching of the mycelium, suppression of aerial hyphae formation, reduced hydrophobicity of the mycelium, as well as reduced sporulation and virulence (Merhej et al. 2012). In P. chrysogenum, deletion of PcvelA leads to reduced production of penicillin, together with light-independent formation of conidiospores, dichotomous branching of hyphae, and increased pellet formation in shaking cultures (Hoff et al. 2010). Similarly, deletion of AcveA in A. chrysogenum leads to reduced production of cephalosporin, early hyphal fragmentation, and hyperbranching of hyphal tips (Dreyer et al. 2007).

Besides VeA, three other members of the velvet family, namely VelB, VelC, and VosA, have been identified, which are able to form multi-subunit protein complexes together with VeA and the putative S-adenosyl-L-methionine (SAM)-dependent methyltransferase LaeA (Hoff et al. 2010, Sarikaya-Bayram et al. 2010, Bayram and Braus 2012, Kopke et al. 2013). One of the best studied functions of LaeA is its involvement in regulation of SM gene-cluster expression (Bok and Keller 2004, Kale et al. 2008, Kosalková et al. 2009, Sarikaya-Bayram et al. 2010, Wiemann et al. 2010, Karimi-Aghcheh et al. 2013). For example, transcriptional profiling in A. fumigatus revealed that expression of 13 out of 22 SM gene clusters is significantly reduced in the ΔlaeA background (Perrin et al. 2007). Correspondingly, in V. DISCUSSION 37

P. chrysogenum, deletion of PclaeA resulted in a significant reduction in penicillin biosynthesis and conidiospore formation (Hoff et al. 2010). Besides its involvement in regulation of secondary metabolism, LaeA was shown to affect virulence in numerous pathogenic fungi (Bok et al. 2005, Sugui et al. 2007, Kale et al. 2008, Amaike and Keller 2009, Wiemann et al. 2010, Wu et al. 2012, López-Berges et al. 2013).

Comprehensive characterization of the velvet proteins and LaeA in A. nidulans and P. chrysogenum enabled the establishment of working models for velvet complex-mediated regulation, which are described in detail in Bayram et al. (2012) and Kopke et al. (2013). In brief, in A. nidulans, VeA and VelB are transported into the nucleus under dark conditions, where interaction with LaeA leads to the formation of a heterotrimeric VelB-VeA-LaeA complex that controls sexual development and secondary metabolism (Bayram et al. 2008). Moreover, LaeA-dependent shuffling of VelB between VelB-VeA-LaeA and a second complex, VelB-VosA, has been described. In the absence of light, VelB-VosA was shown to repress asexual differentiation and regulate biogenesis of trehalose, a compound necessary for long-term viability of fungal spores (d'Enfert and Fontaine 1997, Elbein et al. 2003, Sarikaya- Bayram et al. 2010). In P. chrysogenum, all velvet subunits, including PcLaeA, have been shown to interact with one or more other subunits (Hoff et al. 2010, Kopke et al. 2013). However, it has not been solved if sub-complexes are formed at distinct time points or as a function of developmental stages. Phenotypic characterization of a set of single- and double-deletion mutants revealed that PcVelA, together with PcLaeA and PcVelC activates penicillin biosynthesis, whereas PcVelB represses this process. Moreover, PcVelB and PcVosA were shown to promote conidiation, while PcVelC has an inhibitory effect (Hoff et al. 2010, Kopke et al. 2013).

2.1 PcVelA acts as a transcriptional regulator on DNA level

In order to advance our general understanding of velvet complex-mediated regulatory functions in P. chrysogenum, PcVelA ChIP-seq analyses were performed (see section IV). 467 statistically significant PcVelA DNA-binding sites, corresponding to 631 putative direct PcVelA target genes, were identified. Based on the fact that PcVelA regulatory functions were generally thought to be restricted to protein level, this number of direct PcVelA DNA-binding sites appeared surprisingly high. However, previous studies already indicated that VeA greatly influences overall gene expression levels in various fungi. For example, microarray analyses in P. chrysogenum revealed that a total of 13.6 % of all nuclear genes is expressed in a PcVelA-dependent manner (Hoff et al. 2010). Moreover, RNA-seq analyses in V. DISCUSSION 38

A. fumigatus and A. nidulans demonstrated that a total of 32 % and 26 % of all protein-coding genes are differentially regulated in a ∆veA strain compared to wild type (Lind et al. 2015).

A large number of high-affinity PcVelA target regions were found to be located next to genes known to be involved in processes that are affected by velvet proteins. For example, a total of at least 14 genes related to conidiation and development as well as 16 genes that could be assigned to secondary metabolism were identified. Interestingly, data from PcVelA ChIP-seq analyses showed overlap to previous ChIP-chip analyses of other velvet components in A. nidulans. Here, specific binding of VosA to the promoter sequences of brlA, coding for a master regulator of conidiogenesis (Adams et al. 1988), and treA, associated with trehalose biosynthesis, was demonstrated (Ahmed et al. 2013). Taken together, these observations not only suggest that data obtained from ChIP-seq analysis are indeed of high biological relevance, but also provide the first experimental evidence for PcVelA acting as a direct transcriptional regulator on DNA level, possibly even as a TF. All in all, data presented within this work provide evidence for a hypothesis made by Ni and Yu almost ten years ago, which implies that the velvet proteins might be acting as global transcriptional regulators, representing a new fungus-specific class of TFs (Ni and Yu 2007).

One of the best described features of VeA is its involvement in regulation of conidiation in various fungi (Kim et al. 2002, Kato et al. 2003, Calvo 2008, Tuch et al. 2008, Hoff et al. 2010, Wiemann et al. 2010, Merhej et al. 2012, Kopke et al. 2013). While almost nothing is known about the molecular details of light-dependent formation of asexual conidiospores in P. chrysogenum, our current understanding of these mechanisms in A. nidulans is comparatively profound. In general, conidiation is thought to be regulated by the master regulator BrlA, whose expression is dependent on a number of genes, including the fluffy genes fluG and flbA-E (Adams et al. 1988, Lee and Adams 1994, Etxebeste et al. 2008, Garzia et al. 2010, Arratia-Quijada et al. 2012, Oiartzabal-Arano et al. 2015). It was hypothesized that the light-sensing FphA-LreA-LreB photoreceptor complex might be signaling FlbB and FlbC, which in turn bind to the brlA promoter region to activate its expression. Alternatively, the photoreceptor complex itself might bind to the promoter of brlA in a mechanism that involves FlbB and FlbC (Ruger-Herreros et al. 2011). Another working model assumes direct involvement of VeA. Based on the fact that a direct interaction between FphA and VeA as well as light-dependent shuttling of VeA between nucleus and cytoplasm can be observed (Stinnett et al. 2007, Purschwitz et al. 2009), it was hypothesized that light signals are transmitted to photoreceptors, which in turn control VeA activity through direct V. DISCUSSION 39 protein-protein interaction (Bayram et al. 2008). VeA in turn would interact with additional downstream factors, such as LaeA, to orchestrate light-dependent development and biosynthesis of SMs (Bayram et al. 2008, Sarikaya-Bayram et al. 2010). Interestingly, not only brlA but also flbC, encoding a C2H2 zinc-finger protein, and flbD, encoding a Myb-like DNA-binding protein (Wieser and Adams 1995, Arratia-Quijada et al. 2012), have been identified as specific PcVelA target genes in ChIP-seq analyses presented within this work. In contradiction to the working models supposed by Ruger-Herreros et al. (2011) and Bayram et al. (2008), this finding points to a direct involvement of PcVelA in regulation of conidiation on DNA level. Nevertheless, it remains unclear whether PcVelA binding to promoter regions of its target genes is mediated by PcVelA alone, possibly acting as a downstream factor of the photoreceptor complex, or if DNA-binding of PcVelA is dependent on the interaction with additional proteins, such as the photoreceptors FphA, LreA/LreB and/or the velvet proteins.

Besides target genes that are related to regulatory pathways known to be affected by the velvet complex, ChIP-seq identified a significant number of direct PcVelA target genes that have never been related to PcVelA or any other component of the velvet complex before. Among these targets, a total of at least six TF-encoding genes and seven genes coding for putative methyltransferases were identified. This observation might indicate that overall PcVelA regulatory functions are dependent on additional downstream factors with direct impact on transcriptional regulation, either as TFs or as modifiers of epigenetic marks. Moreover, this observation is in line with current research, which demonstrated a close link between VeA and various putative methyltransferases in a number of filamentous ascomycetes (Jiang et al. 2011, Bi et al. 2013, Connolly et al. 2013b, Palmer et al. 2013, Sarikaya-Bayram et al. 2014). Based on this finding, as well as on data obtained from ChIP-seq and downstream analyses, PcLlmA (LaeA-like methyltransferase A), a putative SAM-dependent methyltransferase, encoded by a new PcVelA target gene, was identified as the most promising candidate for further characterization (see section V.2.2).

Besides identification of putative direct PcVelA target genes, ChIP-seq data were also used for the de novo prediction of the PcVelA DNA-binding consensus sequence “AACCTTGGAA” (PcVelA.M1), which was shown to specifically mediate protein-DNA binding in vitro. This observation is consistent with the recent finding that two other velvet proteins from A. nidulans, namely VosA and the VosA-VelB heterodimer, are able to bind DNA in a sequence-specific manner (Ahmed et al. 2013). However, despite the fact that PcVelA and A. nidulans VosA share a number of specific target genes, PcVelA.M1 displayed V. DISCUSSION 40 only moderate similarity to the described VosA binding site “CTGGCCAAGGC”. Hence, it is likely that even though both proteins bind to different DNA-consensus elements, they share some fundamental regulatory features. Interestingly, comparison of PcVelA.M1 to DNA-binding motifs present in the JASPAR core (2014) databases revealed significant overlap to DNA-binding consensus sequences of NR2F1 and NR2E3, both involved in the regulation of the development of the visual system in humans (Kobayashi et al. 1999, Milam et al. 2002, Bosch et al. 2014). It was shown, that mutations in NR2E3 are associated with the enhanced S-cone syndrome, characterized by night blindness, varying degrees of color vision, and retinal degeneration in humans (Haider et al. 2000). Remarkably, this phenotype somehow resembles those of PcvelA/veA deletion mutants in P. chrysogenum and A. nidulans. Here, deletion of veA/PcvelA results in impaired light-sensing abilities, leading to the formation of conidiospores in the absence of light, whereas conidiogenesis in the wild type is light-dependent (Käfer 1965, Mooney and Yager 1990, Hoff et al. 2010).

Taken together, a regulatory function of PcVelA on DNA level instead or simultaneously to a regulatory function on protein level has to be considered. Moreover, a direct involvement of PcVelA in light-dependent gene regulation has to be assumed. Further experiments will be necessary in order to fully elucidate DNA-binding properties of PcVelA and to understand the dynamics underlying PcVelA-dependent regulation of target gene expression in cooperation with other proteins.

2.2 The putative SAM-dependent methyltransferase PcLlmA is a direct interaction partner of PcVelA

ChIP-seq, qRT-PCR and microarray analyses, as well as DNA-binding studies unambiguously identified PcllmA as a direct PcVelA-target gene, coding for a putative SAM-dependent methyltransferase with noticeable similarity to PcLaeA. Furthermore, yeast two-hybrid (Y2H) and bimolecular fluorescence complementation (BiFC) analyses confirmed direct interaction between PcLlmA and PcVelA, as well as nuclear localization of the PcVelA-PcLlmA heterodimer. Interestingly, interaction between PcLlmA and the velvet complex appeared to be restricted to PcVelA, as it has previously been shown for the interaction between the velvet complex and the putative SAM-dependent methyltransferase PcLaeA (Kopke et al. 2013). It is a recent observation that interaction with various putative methyltransferases, others than LaeA, seems to be a characteristic feature of VeA. For example, a reverse genetics screen in A. nidulans identified LlmF (LaeA-like methyltransferase F), an interaction partner of VeA and a negative regulator of sexual V. DISCUSSION 41 development and secondary metabolism (Palmer et al. 2013). Furthermore, methyltransferases VipC (velvet interacting protein C) and VapB (VipC associated protein B were shown to directly interact with VeA in the nucleus to promote asexual development or, together with the membrane protein VapA, at the plasma membrane to support sexual development (Sarikaya-Bayram et al. 2014). Moreover, using a Y2H approach, F. graminearum FgVeA was shown to interact with a total of six putative methyltransferases that show sequence homologies to FgLaeA1 (Jiang et al. 2011). As it has been demonstrated for PcVelA-PcLlmA, interactions between PcVelA/VeA and other methyltransferases, such as PcLaeA (Hoff et al. 2010), as well as A. nidulans LlmF, VipC and VapB (Palmer et al. 2013, Sarikaya-Bayram et al. 2014) are restricted to the nucleus. However, it remains unclear how the interaction between VeA and the growing number of methyltransferases is mediated on a structural level. It was suggested that VeA should have an affinity domain for methyltransferases or a tertiary domain providing interaction with methyltransferases (Bayram et al. 2008, Sarikaya-Bayram et al. 2015) but experimental evidence is needed to verify these hypotheses and to elucidate the functional output of these interactions in more detail.

Besides its direct interaction with PcVelA, PcLlmA was shown to carry the methyltransferase-specific sequence motifs I-III (Kagan and Clarke 1994), and therefore to fulfill the requirements for epigenetic methyltransferase activity. Similar functions have already been hypothesized for LaeA and VapB but experimental evidence for an involvement in genome-wide chromatin remodeling is still lacking (Reyes-Dominguez et al. 2010, Sarikaya-Bayram et al. 2014). Against this background it will be highly interesting to see if PcLlmA is able to exert epigenetic methyltransferase activity, which would make it one of the most promising PcVelA interaction partners identified so far. However, when speculating about PcLaeA, VapB, and even PcLlmA functions on a molecular level, it has to be mentioned that biological functions of SAM-dependent methyltransferases are versatile. They catalyze the transfer of methyl groups from SAM to a large variety of acceptor substrates, ranging from small metabolites to bio-macromolecules, including DNA, proteins and SMs (Martin and McMillan 2002, Jiang et al. 2011, Struck et al. 2012). On the one hand, this marks them as interesting candidates for application in biotechnology (Struck et al. 2012), but, on the other hand, this emphasizes why numerous other functions besides those involved in epigenetic modification of chromatin have to be taken into account. Above that, occasional examples for proteins, which incorporate the core SAM-dependent methyltransferase fold but do not exert any quantifiable methyltransferase activity, can be found in the literature (Dong et al. 2001). V. DISCUSSION 42

2.3 An expanded model of PcVelA regulatory functions

Starting from the aforementioned observations, an expanded version of the current model of PcVelA regulatory functions can be hypothesized (Figure 6). On the one hand, PcVelA acts on the protein level as one of the core components of the velvet complex and, most likely, as an interaction partner of the light sensing FphA-LreA-LreB complex. On the other hand, PcVelA operates as a genome-wide transcriptional regulator on DNA level, possibly in cooperation with other proteins, such as the velvet components or the FphA-LreA-LreB photoreceptor complex. Examples from the literature demonstrate that these functions must not be mutually exclusive. For instance, the metabolic enzyme IMPDH, which controls the cellular guanidine nucleotide pool was shown to be also a DNA-binding transcriptional repressor involved in regulation of histone genes and E2f, a key driver of cell proliferation in Drosophila (Kozhevnikova et al. 2012). Furthermore, as the putative SAM-dependent methyltransferase PcLlmA was identified as a direct interaction partner of PcVelA, a third level of PcVelA regulatory functions, dependent on interacting methyltransferases, others than PcLaeA, has to be assumed.

Figure 6: Three levels of PcVelA regulatory functions. (1) PcVelA acts as one of the core components of the velvet complex, which is likely to form different sub-complexes in order to mediate control of development, morphology, and secondary metabolism. (2) Based on data obtained from ChIP-seq and follow-on analyses, PcVelA acts as a regulatory protein on DNA-level, probably even as a TF. It is conceivable that DNA-binding is dependent on interaction with other proteins, such as the velvet components or the FphA-LreA-LreB photoreceptor complex. (3) PcVelA directly interacts with putative methyltransferases, others than PcLaeA. For example, the putative SAM-dependent methyltransferase PcLlmA was identified as a direct interaction partner and downstream factor of PcVelA. (modified from a model provided by Dr. S. Bloemendal) V. DISCUSSION 43

3. Overall analysis of ChIP-seq data

3.1 Genome-wide TF binding beyond direct target-gene control

Using ChIP-seq analysis, a total of 243 and 467 specific DNA-binding sites have been identified for MAT1-1-1 and PcVelA, respectively. All of these binding sites passed a high statistical threshold and were present in at least two independent biological replicates. Overall, the observed DNA-binding patterns matched the characteristic features of TF DNA-binding in most eukaryotes. For example, a total of 79.4 % of MAT1-1-1 and 78.9 % of PcVelA DNA-binding regions was found to be located within intergenic regions, matching the general observation that regulatory genomic regions targeted by TFs are primarily found within intergenic or intronic DNA (Stergachis et al. 2013). Furthermore, the distance of peak summits and transcription start sites (TSSs) of neighboring genes was found to be within a range of 200-500 nt for MAT1-1-1 and 100-600 nt for PcVelA. This finding is consistent with the observation that, depending on TF identity, expression of target genes in S. cerevisiae reaches maximal values when the TFBS is either within 150 bp (short-range regulation), 150-300 bp (mid-range regulation), or 300-500 bp (long-range regulation) from the start codon, and that most regulatory DNA sequences for a given gene fall within a few hundred bp from its TSS (Nguyen and D'Haeseleer 2006, Lin et al. 2010).

Biological significance of ChIP-seq datasets was verified using ChIP-PCR, in order to rule out bias from bioinformatics analysis, as well as microarray and qRT-PCR analyses for the identification of specific MAT1-1-1 and PcVelA target genes. Comparison of ChIP-seq data to expression values from previous microarray analysis in ΔMAT1-1-1 and ΔPcvelA strains revealed that 29.9 % and 18.9 % of genes showing 5’-3’ orientation with regard to neighboring peak regions are expressed in a MAT1-1-1- and PcVelA-dependent manner, respectively. This observation is in line with previous works, analyzing TF binding in relation to changes in expression profiles of neighboring genes and demonstrating that 58 % of genes whose promoter region is bound by a TF are true regulatory targets of the respective factor in S. cerevisiae, whereas this is true for only 10-25 % of putative target genes in Drosophila and mammalian systems (Gao et al. 2004, Sandmann et al. 2006, Jakobsen et al. 2007, Vokes et al. 2008). Correspondingly, a comprehensive TF knock-down analysis of 59 TFs and chromatin modifiers in a human lymphoblastoid cell line, revealed that 46.4 % to 99.1 % of all analyzed TF binding events are likely to be non-functional, as no association between knock down of the respective factor and changes in expression levels could be documented (Cusanovich et al. 2014). V. DISCUSSION 44

Based on the growing amount of data from genome-wide TF DNA-binding studies (Table 6), it is a common observation that TFs vary greatly in their number of genomic binding sites. This suggests that TF binding events can significantly exceed the number of conceivable or even possible direct target genes. Accordingly, starting from the observation that genome-wide TF binding is not necessarily equivalent to expressional regulation of adjacent genes, three types of TF binding events can be distinguished: [1] specific, functional binding to cis-regulatory regions with a direct impact on gene regulation, [2] specific but non-functional binding, and [3] non-specific, non-functional binding (when functionality is defined as transcriptional regulation) (Todeschini et al. 2014). It is generally accepted, and might also be true for MAT1-1-1 and PcVelA TFBSs identified within this work, that although the most highly bound regions in ChIP-seq analyses are known functional targets, many of the thousands of regions bound at much lower levels may represent specific or non-specific, non-functional interactions. Support for this hypothesis has been provided by recent studies in Drosophila, analyzing DNA-binding patterns of 21 TFs involved in embryo development (MacArthur et al. 2009, Fisher et al. 2012). Here, it was demonstrated that high-affinity binding of TFs is more likely to occur within close proximity of genes showing transcriptional regulation, whereas low-affinity binding generally occurs in regions not regulated by the respective factor. Consistently, analyses focusing on the conservation of TF

Table 6: Numbers of TFBSs from selected ChIP-seq experiments

Species Transcription factor Reported number of TFBSs Reference Anabaena sp. All3953 142 (Picossi et al. 2015) Arabidopsis thaliana KAN1 4,183 (Merelo et al. 2013) Aspergillus fumigatus SrbA 111 (Chung et al. 2014) Caenorhabditis elegans PHA-4 4,350/4,808b (Zhong et al. 2010) Caenorhabditis elegans Tra-1 184 (Berkseth et al. 2013) Candida parapsilosis Efg1 931 (Connolly et al. 2013a) Fusarium graminearum Tri6 198 (Nasmith et al. 2011) Human NRSF 1,946 (Johnson et al. 2007) Human STAT1 11,004/41,582c (Robertson et al. 2007) Human TdIF1 1,274 (Koiwai et al. 2015) Mouse MyoD 25,956/59,267a (Cao et al. 2010) Neurospora crassa WCC 287 (Hurley et al. 2014) Penicillium chrysogenum MAT1-1-1 243 This work (section III) Penicillium chrysogenum PcVelA 467 This work (section IV) Saccharomyces cerevisiae Pho7 1,676 (Carter-O'Connell et al. 2012) Solanum lycopersicum ASR1 225 (Ricardi et al. 2014) Sordaria macrospora PRO1 215 E. Steffens, personal communication Zebrafish Gli2a 93/122e (Wang et al. 2013) Zebrafish Zic3 3,209/2,088d (Winata et al. 2013) a)binding sites at two different statistical cutoffs b)binding sites in embryos and L1 larvae c)binding sites in un-stimulated and interferon-γ-stimulated cells d)binding sites after 8 and 24 hours post-fertilization e)binding sites at the 5 and 15 somitic stage V. DISCUSSION 45 binding and gene expression patterns within different Drosophila species demonstrated that TFBSs producing strong peaks are more likely to be conserved across species than those characterized by weak signals in ChIP-seq analyses (He et al. 2011, Paris et al. 2013). However, as signal strengths obtained from ChIP assays usually represent mean values of the corresponding signals across millions of cells, caution should be exercised when trying to draw conclusions about the biological relevance, functionality, or binding-affinity of a DNA-binding protein from ChIP-seq signal strength alone (Slattery et al. 2014).

In contradiction to the assumption that the majority of low-affinity TFBSs identified in large-scale TF-binding studies is likely to serve no apparent biological purpose, Tanay (2006) hypothesized that TF binding to these sites might contribute to gene expression at levels that are low but sufficient enough to allow evolutionary conservation. Another model aiming at explaining the apparently contradictory DNA-binding patterns of many TFs proceeds on the assumption that genome-wide TF binding at non-regulatory sites might serve as a reservoir for TFs, sequestering them in a manner comparable to other biological buffering systems, and ensuring an optimal amount of available TF in the nucleus (MacQuarrie et al. 2011). Other explanations for the existence of apparently non-functional TFBSs include the possibility of TFs exerting their regulatory influence on genes over large genomic distances by distal elements like enhancers or silencers (most possibly mediated by chromatin looping), as well as TF-mediated induction of changes in chromatin and nuclear structure, and the evolution of new GRNs (Cao et al. 2010, MacQuarrie et al. 2011, Zaret and Carroll 2011, Weingarten- Gabbay and Segal 2014).

3.2 MAT1-1-1 and PcVelA bind DNA via specific DNA-consensus sequences

Starting from peak regions identified in ChIP-seq analyses, DNA-binding motif consensus sequences were predicted and verified for both, MAT1-1-1 and PcVelA. Remarkably, no evidence was found for the necessity of both motifs to be orientated in a definite direction in order to drive expression of neighboring target genes. This finding is in agreement with previous work in S. cerevisiae, revealing that activity of only 6 out of 75 (8 %) analyzed TFs is dependent on the orientation of the corresponding TFBS (Sharon et al. 2012).

When focusing on the overall distribution of DNA-binding motifs MAT1.1 and PcVelA.M1, it was striking that some of the most significant DNA-binding regions were found to carry noticeably high numbers of the corresponding DNA-binding consensus sequences. Interestingly, this observation is fits the recent discovery that functional binding of human V. DISCUSSION 46

TFs is enriched in genomic regions that carry large numbers of specific TFBSs, at sites with predicted higher binding affinity, and at sites that are clustered within genomic regions annotated as enhancers (Cusanovich et al. 2014). A possible explanation for this observation has been provided by a large-scale bioinformatics analysis of more than 950 TF-binding motifs, leading to the conclusion that clustering of DNA-binding motifs is necessary to target TFs to their specific binding sites in eukaryotic genomes (Wunderlich and Mirny 2009). Fundamental for this hypothesis was the observation that eukaryotic TFs typically recognize short sequences of 10-12 nt, whereas bacterial TFs tend to recognize extended DNA sites of ~ 23 nt, which is enough to ensure specificity in small genomes (Wunderlich and Mirny 2009, Stewart et al. 2012).

Although DNA-binding motifs were present in the majority of high-affinity MAT1-1-1 and PcVelA binding-sites, a significant proportion of peaks lacked clearly identifiable copies of MAT1.1 and PcVelA.M1, respectively. Accordingly, at least one copy of the respective consensus sequences was identified in 83.1 % of MAT1-1-1 peak regions (p ≤ 0.01) and 46.5 % of PcVelA peak regions (p ≤ 0.001). This is a common observation: although TF DNA-binding regions identified in ChIP-seq analyses are typically enriched in the consensus motif for the TF in question, a significant proportion of peaks lack it (Robertson et al. 2007, Li et al. 2008, Rabinovich et al. 2008, Valouev et al. 2008, Berkseth et al. 2013). Possible explanations for this observation can be found in recent studies providing evidence for the highly combinatorial nature of TF binding in eukaryotes. For example, it has been demonstrated that TFs not only interact with DNA via a consensus sequence but also recognize divergent sequences, and that TF regulatory functions involve interactions with other TFs binding to neighboring DNA and chromatin sites, as well as binding to chromatin without directly contacting DNA (Li et al. 2007, Brent et al. 2008, Badis et al. 2009, Georges et al. 2010, Todeschini et al. 2014). For example, expression of amdS, encoding an acedamidase enzyme that is vital for A. nidulans carbon and nitrogen metabolism, was shown to be regulated by at least six different TFs, which directly bind within a ~ 200 nt upstream region of the gene (Hynes 1994). Combinatorial binding can be mediated by direct protein-protein interactions, leading to the formation of homodimers, heterodimers, or large transcriptional complexes, as well as indirectly by co-binding of the same DNA-sequence (Walhout 2006, Amoutzias et al. 2008, Ravasi et al. 2010). Dimerization of TFs has been shown to be linked to several benefits, such as an increase in the binding affinity and effective length of the recognized DNA sites, which in turn increases binding specificity and decreases the probability of random occurrence of the DNA-binding motif within the genome (Georges V. DISCUSSION 47 et al. 2010). Furthermore, for some TFs it has been shown that they are able to assume divergent functions when interacting with different partners. For example, TF SREBP1, an important regulator of cholesterol and fatty acid metabolism in humans, has been shown to act in distinct functional pathways when interacting with its binding partners NFY and SP1 (Reed et al. 2008).

Based on findings presented within this work and in the literature, formation of homo- and heterodimers might indeed be a regulatory feature of P. chrysogenum MAT1-1-1 and PcVelA. For example, various direct interaction partners of PcVelA, including the velvet proteins PcVelB, PcVelC, PcVosA and PcVelA itself, along with the putative methyltransferases PcLaeA and PcLlmA, have been identified using Y2H and BiFC approaches in P. chrysogenum (Hoff et al. 2010, Kopke et al. 2013). Moreover, recent crystal structure analysis in A. nidulans provided evidence for an involvement of the velvet domain in the dimerization of different velvet-domain proteins (Ahmed et al. 2013), and, based on the fact that different combinations of dimerization exist in velvet proteins, their dimerization properties have been described to resemble those of the bZIP family of TFs (Sarikaya-Bayram et al. 2015). For MAT1-1-1 the situation turns out to be more vague, as dimerization was shown to be a common feature in MAT proteins (Dranginis 1990, Bruhn et al. 1992, Ho et al. 1994, Nolting and Pöggeler 2006), but no experimental evidence has been provided for interactions between MAT1-1-1 and any other proteins in P. chrysogenum until now. It will be a challenging task to further analyze the molecular details and dynamics of MAT1-1-1 and PcVelA DNA-binding properties in terms of putative interactions with other proteins, acting as cooperative binding partners on DNA-level or as direct interaction partners on protein level.

3.3 Concluding remarks

Taken together, data obtained from MAT1-1-1 and PcVelA ChIP-seq analyses in P. chrysogenum share both the strengths and weaknesses of many other datasets from ChIP-seq experiments, performed in a variety of organisms and tissues so far. While on the one hand, ChIP-seq represents the most powerful tool for genome-wide binding profiling of DNA-binding proteins and epigenetic marks, it is, on the other hand, a challenging task to fully elucidate the biological meaning of raw sequencing data. This difficulty becomes especially obvious, when it comes to the discrimination between functional and apparently non-functional DNA-binding sites. Although integration of additional information obtained from expression profiling or mapping of euchromatic and/or heterochromatic genomic regions V. DISCUSSION 48 can help to shed light onto the genome-wide regulatory functions of DNA-binding proteins, it is still not sufficient to fully elucidate the whole complexity of GRNs. It will be a major task for the future to reconsider the traditional relationship between TF binding and gene regulation, as well as to analyze new aspects of transcriptional regulation, such as the role of widespread TF binding outside direct target-gene control and dependencies between TF binding and the three-dimensional structure of chromatin, in more detail.

When focusing on ChIP-seq analyses performed within the scope of this work, it has to be regarded that the presented experimental workflow was designed to depict a preferably complete picture of as much MAT1-1-1- and PcVelA-specific DNA-binding regions as possible. Therefore, DNA-binding profiles were analyzed independent of culture conditions, developmental stages, or external stimuli and in MAT1-1-1 and PcvelA overexpression background, respectively. Although this approach turned out to be well suited to get an overview of the genome-wide regulatory functions of both proteins, it does not consider the fact that some TFs can occupy diverse sets of binding sites, depending on the developmental stage or condition considered. An example for this was provided by ChIP-seq analysis of the erythroid Kruppel-like factor (EKLF) in mice, which revealed highly divergent binding patterns in erythroid progenitor cells and more differentiated erythroblasts (Pilon et al. 2011). Furthermore, it has to be taken into account that some binding events may require co-factors that are only present after specific stimuli in order to become functional. For example, the non-DNA-binding transcriptional co-activator Met4 enables the regulatory centromere-binding factor Cbf1–Met4 heterodimer to recognize an extended DNA-binding motif, mediating effective expression of the sulfur metabolism genes in S. cerevisiae (Siggers et al. 2011). Hence, further studies will be needed in order to fully elucidate the GRNs governed by MAT1-1-1 and PcVelA as a function of developmental stages, physiological culture conditions, or environmental factors. VI. SUMMARY 49

VI. SUMMARY

P. chrysogenum is the only industrial producer of the β-lactam antibiotic penicillin, the most commonly used drug in the treatment of bacterial infections. In order to identify new starting points for further optimization of high-production strains, it is necessary to obtain a more comprehensive knowledge of GRNs controlling morphogenesis and secondary metabolism in P. chrysogenum. Within this context, functional characterization of TFs, which orchestrate gene expression control on the molecular level, is of major importance.

Within the scope of this work, the ChIP-seq technology, which is regarded as the most powerful tool for the analysis of protein-DNA interactions on a genome-wide scale, was successfully adapted for application in P. chrysogenum. A well-established experimental pipeline for sample preparation, ChIP-seq data analysis, and follow-on experiments, including validation of ChIP-seq data, identification of specific target genes, as well as prediction and validation of DNA-binding consensus sequences, is now available. Furthermore, comprehensive ChIP-seq analyses and follow-on experiments revealed important new insights into the regulatory properties of the MAT α-domain TF MAT1-1-1 and the velvet protein PcVelA. Most importantly, data presented within this work provide the first experimental evidence for a direct involvement of MAT1-1-1 in transcriptional regulation of numerous target genes beyond sexual development. Moreover, it was shown that extensive rewiring of MAT controlled transcriptional networks must have occurred in euascomycetes compared to hemiascomycetes. With regard to PcVelA, the most important finding presented within this work is that its regulatory properties are not restricted to protein-protein interactions with other components of the velvet complex, but instead involve regulatory functions on DNA level, probably even as a TF. Moreover, a new downstream factor and direct interaction partner of PcVelA, the putative SAM-dependent methyltransferase PcLlmA, was identified. This finding points to a third level of PcVelA regulatory functions, which involves interactions with putative methyltransferases others than the velvet-interacting putative methyltransferase PcLaeA.

Taken together, data from ChIP-seq and follow-on analyses not only enabled new insights into MAT1-1-1 and PcVelA regulatory functions, but also provide a versatile basis for further analysis of GRNs controlling morphogenesis and secondary metabolism in the biotechnologically highly relevant ascomycete P. chrysogenum. VII. ZUSAMMENFASSUNG 50

VII. ZUSAMMENFASSUNG

P. chrysogenum ist der einzige industriell genutzte Produzent des β-Laktam-Antibiotikums Penicillin, welches weltweit am häufigsten zur Behandlung bakterieller Infektionen eingesetzt wird. Für die weitere Optimierung industrieller Hochleistungs-Produktionsstämme sind fundierte Kenntnisse der transkriptionellen Regulation der Morphogenese und Sekundärmetabolit-Biosynthese in P. chrysogenum unerlässlich. Insbesondere die funktionelle Charakterisierung von Transkriptionsfaktoren, welche die Genregulation auf molekularer Ebene steuern, ist hierbei von besonderer Bedeutung.

Im Rahmen der vorliegenden Arbeit wurde die ChIP-seq Technologie, welche zurzeit als die vielversprechendste Methode zur Genom-weiten Analyse von Protein-DNA-Interaktionen angesehen wird, erfolgreich für die Anwendung in P. chrysogenum adaptiert. Es steht nun ein etabliertes Protokoll zur Verfügung, welches sämtliche Schritte einer ChIP-seq-Analyse und nachfolgender Kontrollexperimente abdeckt. Des Weiteren wurden umfassende ChIP-seq-Analysen durchgeführt, welche neue Einblicke in die regulatorischen Eigenschaften des Kreuzungstyp-Transkriptionsfaktors MAT1-1-1 und des Velvet-Proteins PcVelA lieferten. Es konnte gezeigt werden, dass MAT1-1-1 an der Regulation zahlreicher Gene beteiligt ist, welche nicht unmittelbar mit der sexuellen Entwicklung des Pilzes in Verbindung stehen. Darüber hinaus ergaben sich wichtige Hinweise auf grundlegende Umstrukturierungen von Kreuzungstyp-regulierten Genregulationsnetzwerken in Euascomyzeten im Vergleich mit Hemiascomyzeten. Für PcVelA konnte des Weiteren gezeigt werden, dass die regulatorischen Eigenschaften des Proteins nicht allein auf Protein-Protein-Interaktionen mit Komponenten des Velvet-Komplexes beschränkt sind, sondern auch regulatorische Funktionen auf DNA-Ebene, möglicherweise sogar als Transkriptionsfaktor, umfassen. Außerdem konnte ein neuer Downstream-Faktor und direkter Interaktionspartner von PcVelA, die putative SAM-abhängige Methyltransferase PcLlmA, identifiziert werden. Diese Beobachtung deutet auf eine weitere Ebene von PcVelA-vermittelter Regulation hin, welche auf der Interaktion mit anderen Methyltransferasen neben der als Komponente des Velvet-Komplexes beschriebenen putativen Methyltransferase PcLaeA beruht.

Zusammenfassend liefert die vorliegende Arbeit nicht nur wichtige neue Erkenntnisse über die regulatorischen Funktionen von MAT1-1-1 und PcVelA, sondern stellt auch eine fundierte Basis für die weitere Analyse genregulatorischer Netzwerke bereit, welche die Morphogenese und den Sekundärmetabolismus in P. chrysogenum kontrollieren. VIII. REFERENCES 51

VIII. REFERENCES

Ádám AL, García-Martínez J, Szücs EP, Avalos J, Hornok L (2011) The MAT1-2-1 mating-type gene upregulates photo-inducible carotenoid biosynthesis in Fusarium verticillioides. FEMS Microbiol Lett 318: 76-83

Adams TH, Boylan MT, Timberlake WE (1988) BrlA is necessary and sufficient to direct conidiophore development in Aspergillus nidulans. Cell 54: 353-62

Albert I, Mavrich TN, Tomsho LP, Qi J, Zanton SJ, Schuster SC, Pugh BF (2007) Translational and rotational settings of H2A.Z nucleosomes across the Saccharomyces cerevisiae genome. Nature 446: 572-6

Amaike S, Keller NP (2009) Distinct roles for VeA and LaeA in development and pathogenesis of Aspergillus flavus. Eukaryot Cell 8: 1051-60

Ammerer G, Sprague GF, Jr., Bender A (1985) Control of yeast α-specific genes: evidence for two blocks to expression in MATa/MATα diploids. Proc Natl Acad Sci U S A 82: 5855-9

Amoutzias GD, Robertson DL, Van de Peer Y, Oliver SG (2008) Choose your partners: dimerization in eukaryotic transcription factors. Trends Biochem Sci 33: 220-9

Arnaise S, Zickler D, Poisier C, Debuchy R (2001) pah1: a homeobox gene involved in hyphal morphology and microconidiogenesis in the filamentous ascomycete Podospora anserina. Mol Microbiol 39: 54-64

Ashburner M, Ball CA, Blake JA, Botstein D, Butler H, Cherry JM, Davis AP, Dolinski K, Dwight SS, Eppig JT, Harris MA, Hill DP, Issel-Tarver L, Kasarskis A, Lewis S, Matese JC, Richardson JE, Ringwald M, Rubin GM, Sherlock G (2000) Gene ontology: tool for the unification of biology. The Gene Ontology Consortium. Nat Genet 25: 25-9

Astell CR, Ahlstrom-Jonasson L, Smith M, Tatchell K, Nasmyth KA, Hall BD (1981) The sequence of the DNAs coding for the mating-type loci of Saccharomyces cerevisiae. Cell 27: 15-23

Audenaert K, Vanheule A, Hofte M, Haesaert G (2014) Deoxynivalenol: a major player in the multifaceted response of Fusarium to its environment. Toxins 6: 1-19

Backus MP, Stauffer JF (1955) The production and selection of a family of strains in Penicillium chrysogenum. Mycologia 47: 429-63

Badis G, Berger MF, Philippakis AA, Talukder S, Gehrke AR, Jaeger SA, Chan ET, Metzler G, Vedenko A, Chen X, Kuznetsov H, Wang CF, Coburn D, Newburger DE, Morris Q, Hughes TR, Bulyk ML (2009) Diversity and complexity in DNA recognition by transcription factors. Science 324: 1720-3

VIII. REFERENCES 52

Bailey T, Krajewski P, Ladunga I, Lefebvre C, Li Q, Liu T, Madrigal P, Taslim C, Zhang J (2013) Practical guidelines for the comprehensive analysis of ChIP-seq data. PLoS Comput Biol 9: e1003326

Baker CR, Tuch BB, Johnson AD (2011) Extensive DNA-binding specificity divergence of a conserved transcription regulator. Proc Natl Acad Sci U S A 108: 7493-8

Barber MS, Giesecke U, Reichert A, Minas W (2004) Industrial enzymatic production of cephalosporin-based β-lactams. Adv Biochem Eng Biotechnol 88: 179-215

Barreiro C, Martín JF, García-Estrada C (2012) Proteomics shows new faces for the old penicillin producer Penicillium chrysogenum. J Biomed Biotechnol 2012: 1-15

Barski A, Cuddapah S, Cui K, Roh TY, Schones DE, Wang Z, Wei G, Chepelev I, Zhao K (2007) High-resolution profiling of histone methylations in the human genome. Cell 129: 823-37

Bayram Ö, Braus GH (2012) Coordination of secondary metabolism and development in fungi: the velvet family of regulatory proteins. FEMS Microbiol Rev 36: 1-24

Bender A, Sprague GF, Jr. (1987) MAT α1 protein, a yeast transcription activator, binds synergistically with a second protein to a set of cell-type-specific genes. Cell 50: 681-91

Bentley DR, Balasubramanian S, Swerdlow HP, Smith GP, Milton J, Brown CG, Hall KP, Evers DJ, Barnes CL, Bignell HR, Boutell JM, Bryant J, Carter RJ, Keira Cheetham R, Cox AJ, Ellis DJ, Flatbush MR, Gormley NA, Humphray SJ, Irving LJ, Karbelashvili MS, Kirk SM, Li H, Liu X, Maisinger KS, Murray LJ, Obradovic B, Ost T, Parkinson ML, Pratt MR, Rasolonjatovo IM, Reed MT, Rigatti R, Rodighiero C, Ross MT, Sabot A, Sankar SV, Scally A, Schroth GP, Smith ME, Smith VP, Spiridou A, Torrance PE, Tzonev SS, Vermaas EH, Walter K, Wu X, Zhang L, Alam MD, Anastasi C, Aniebo IC, Bailey DM, Bancarz IR, Banerjee S, Barbour SG, Baybayan PA, Benoit VA, Benson KF, Bevis C, Black PJ, Boodhun A, Brennan JS, Bridgham JA, Brown RC, Brown AA, Buermann DH, Bundu AA, Burrows JC, Carter NP, Castillo N, Chiara ECM, Chang S, Neil Cooley R, Crake NR, Dada OO, Diakoumakos KD, Dominguez-Fernandez B, Earnshaw DJ, Egbujor UC, Elmore DW, Etchin SS, Ewan MR, Fedurco M, Fraser LJ, Fuentes Fajardo KV, Scott Furey W, George D, Gietzen KJ, Goddard CP, Golda GS, Granieri PA, Green DE, Gustafson DL, Hansen NF, Harnish K, Haudenschild CD, Heyer NI, Hims MM, Ho JT, Horgan AM, Hoschler K, Hurwitz S, Ivanov DV, Johnson MQ, James T, Huw Jones TA, Kang GD, Kerelska TH, Kersey AD, Khrebtukova I, Kindwall AP, Kingsbury Z, Kokko-Gonzales PI, Kumar A, Laurent MA, Lawley CT, Lee SE, Lee X, Liao AK, Loch JA, Lok M, Luo S, Mammen RM, Martin JW, McCauley PG, McNitt P, Mehta P, Moon KW, Mullens JW, Newington T, Ning Z, Ling Ng B, Novo SM, O'Neill MJ, Osborne MA, Osnowski A, Ostadan O, Paraschos LL, Pickering L, Pike AC, Chris Pinkard D, Pliskin DP, Podhasky J, Quijano VJ, Raczy C, Rae VH, Rawlings SR, Chiva Rodriguez A, Roe PM, Rogers J, Rogert Bacigalupo MC, Romanov N, Romieu A, Roth RK, Rourke NJ, Ruediger ST, Rusman E, Sanches-Kuiper RM, Schenker MR, Seoane JM, Shaw RJ, Shiver MK, Short SW, Sizto NL, Sluis JP, Smith MA, Ernest Sohna Sohna J, Spence EJ, Stevens K, Sutton N, Szajkowski L, Tregidgo CL, Turcatti G, Vandevondele S, Verhovsky Y, Virk SM, Wakelin S, Walcott GC, Wang J, Worsley GJ, Yan J, Yau L, Zuerlein M, Mullikin JC, Hurles ME, McCooke NJ, West JS, Oaks FL, Lundberg PL, Klenerman D, Durbin R, Smith AJ (2008) Accurate whole human genome sequencing using reversible terminator chemistry. Nature 456: 53-9

VIII. REFERENCES 53

Berger SL (2007) The complex language of chromatin regulation during transcription. Nature 447: 407-12

Berkseth M, Ikegami K, Arur S, Lieb JD, Zarkower D (2013) TRA-1 ChIP-seq reveals regulators of sexual differentiation and multilevel feedback in nematode sex determination. Proc Natl Acad Sci U S A 110: 16033-8

Bi Q, Wu D, Zhu X, Gillian Turgeon B (2013) Cochliobolus heterostrophus Llm1 - a Lae1-like methyltransferase regulates T-toxin production, virulence, and development. Fungal Genet Biol 51: 21-33

Bianchi ME, Agresti A (2005) HMG proteins: dynamic players in gene regulation and differentiation. Curr Opin Genet Dev 15: 496-506

Bidard F, Ait Benkhali J, Coppin E, Imbeaud S, Grognet P, Delacroix H, Debuchy R (2011) Genome-wide gene expression profiling of fertilization competent mycelium in opposite mating types in the heterothallic fungus Podospora anserina. PLoS One 6: e21476

Blat Y, Kleckner N (1999) Cohesins bind to preferential sites along yeast chromosome III, with differential regulation along arms versus the centric region. Cell 98: 249-59

Boeva V, Lermine A, Barette C, Guillouf C, Barillot E (2012) Nebula - a web-server for advanced ChIP-seq data analysis. Bioinformatics 28: 2517-9

Böhm J, Dahlmann TA, Gümüşer H, Kück U (2015) A MAT1-2 wild-type strain from Penicillium chrysogenum: functional mating-type locus characterization, genome sequencing and mating with an industrial penicillin-producing strain. Mol Microbiol 95: 859-74

Bok JW, Keller NP (2004) LaeA, a regulator of secondary metabolism in Aspergillus spp. Eukaryot Cell 3: 527-35

Bok JW, Balajee SA, Marr KA, Andes D, Nielsen KF, Frisvad JC, Keller NP (2005) LaeA, a regulator of morphogenetic fungal virulence factors. Eukaryot Cell 4: 1574-82

Booth LN, Tuch BB, Johnson AD (2010) Intercalation of a new tier of transcription regulation into an ancient circuit. Nature 468: 959-63

Boyle AP, Davis S, Shulha HP, Meltzer P, Margulies EH, Weng Z, Furey TS, Crawford GE (2008) High-resolution mapping and characterization of open chromatin across the genome. Cell 132: 311-22

Brakhage AA, Andrianopoulos A, Kato M, Steidl S, Davis MA, Tsukagoshi N, Hynes MJ (1999) HAP-Like CCAAT-binding complexes in filamentous fungi: implications for biotechnology. Fungal Genet Biol 27: 243-52

VIII. REFERENCES 54

Brakhage AA, Spröte P, Al-Abdallah Q, Gehrke A, Plattner H, Tüncher A (2004) Regulation of penicillin biosynthesis in filamentous fungi. Adv Biochem Eng Biotechnol 88: 45-90

Brent MM, Anand R, Marmorstein R (2008) Structural basis for DNA recognition by FoxO1 and its regulation by posttranslational modification. Structure 16: 1407-16

Bruhn L, Hwang-Shum JJ, Sprague GF, Jr. (1992) The N-terminal 96 residues of MCM1, a regulator of cell type-specific genes in Saccharomyces cerevisiae, are sufficient for DNA binding, transcription activation, and interaction with α1. Mol Cell Biol 12: 3563-72

Bryant JM, Govin J, Zhang L, Donahue G, Pugh BF, Berger SL (2012) The linker histone plays a dual role during gametogenesis in Saccharomyces cerevisiae. Mol Cell Biol 32: 2771-83

Burton JN, Liachko I, Dunham MJ, Shendure J (2014) Species-level deconvolution of metagenome assemblies with Hi-C-based contact probability maps. G3 (Bethesda) 4: 1339-46

Cacho RA, Tang Y, Chooi YH (2015) Next-generation sequencing approach for connecting secondary metabolites to biosynthetic gene clusters in fungi. Front Microbiol 5: 774

Calvo AM, Wilson RA, Bok JW, Keller NP (2002) Relationship between secondary metabolism and fungal development. Microbiol Mol Biol Rev 66: 447-59

Calvo AM (2008) The VeA regulatory system and its role in morphological and chemical development in fungi. Fungal Genet Biol 45: 1053-61

Carr EA, Mead J, Vershon AK (2004) α1-induced DNA bending is required for transcriptional activation by the Mcm1-α1 complex. Nucleic Acids Res 32: 2298-305

Carroll SB (2000) Endless forms: the evolution of gene regulation and morphological diversity. Cell 101: 577-80

Carter-O'Connell I, Peel MT, Wykoff DD, O'Shea EK (2012) Genome-wide characterization of the phosphate starvation response in Schizosaccharomyces pombe. BMC Genomics 13: 697

Casselton LA, Olesnicky NS (1998) Molecular genetics of mating recognition in basidiomycete fungi. Microbiol Mol Biol Rev 62: 55-70

Chakravarti R, Sahai V (2004) Compactin - a review. Appl Microbiol Biotechnol 64: 618-24

Charoensawan V, Wilson D, Teichmann SA (2010) Genomic repertoires of DNA-binding transcription factors across the tree of life. Nucleic Acids Res 38: 7364-77

Chen H, Lin RJ, Xie W, Wilpitz D, Evans RM (1999) Regulation of hormone-induced histone hyperacetylation and gene activation via acetylation of an acetylase. Cell 98: 675-86

Chooi YH, Cacho R, Tang Y (2010) Identification of the viridicatumtoxin and griseofulvin gene clusters from Penicillium aethiopicum. Chem Biol 17: 483-94

Chu C, Qu K, Zhong FL, Artandi SE, Chang HY (2011) Genomic maps of long non-coding RNA occupancy reveal principles of RNA-chromatin interactions. Mol Cell 44: 667-78

VIII. REFERENCES 55

Chung D, Barker BM, Carey CC, Merriman B, Werner ER, Lechner BE, Dhingra S, Cheng C, Xu W, Blosser SJ, Morohashi K, Mazurie A, Mitchell TK, Haas H, Mitchell AP, Cramer RA (2014) ChIP-seq and in vivo transcriptome analyses of the Aspergillus fumigatus SREBP SrbA reveals a new regulator of the fungal hypoxia response and virulence. PLoS Pathog 10: e1004487

Connolly LA, Riccombeni A, Grozer Z, Holland LM, Lynch DB, Andes DR, Gacser A, Butler G (2013a) The APSES transcription factor Efg1 is a global regulator that controls morphogenesis and biofilm formation in Candida parapsilosis. Mol Microbiol 90: 36-53

Connolly LR, Smith KM, Freitag M (2013b) The Fusarium graminearum histone H3 K27 methyltransferase KMT6 regulates development and expression of secondary metabolite gene clusters. PLoS Genet 9: e1003916

Coppin E, Debuchy R, Arnaise S, Picard M (1997) Mating types and sexual development in filamentous ascomycetes. Microbiol Mol Biol Rev 61: 411-28

Crawford GE, Holt IE, Whittle J, Webb BD, Tai D, Davis S, Margulies EH, Chen YD, Bernat JA, Ginsburg D, Zhou DX, Luo SJ, Vasicek TJ, Daly MJ, Wolfsberg TG, Collins FS (2006) Genome-wide mapping of DNase hypersensitive sites using massively parallel signature sequencing (MPSS). Genome Res 16: 123-31

Cusanovich DA, Pavlovic B, Pritchard JK, Gilad Y (2014) The functional consequences of variation in transcription factor binding. PLoS Genet 10: e1004226

Czaja W, Miller KY, Skinner MK, Miller BL (2014) Structural and functional conservation of fungal MatA and human SRY sex-determining proteins. Nat Commun 5: 5434 d'Enfert C, Fontaine T (1997) Molecular characterization of the Aspergillus nidulans treA gene encoding an acid trehalase required for growth on trehalose. Mol Microbiol 24: 203-16 de Wit E, de Laat W (2012) A decade of 3C technologies: insights into nuclear organization. Genes Dev 26: 11-24

Dekker J, Marti-Renom MA, Mirny LA (2013) Exploring the three-dimensional organization of genomes: interpreting chromatin interaction data. Nat Rev Genet 14: 390-403

Desai AN, Jere A (2012) Next-generation sequencing: ready for the clinics? Clin Genet 81: 503-10

Desjardins AE, Hohn TM, McCormick SP (1993) Trichothecene biosynthesis in fusarium species - chemistry, genetics, and significance. Microbiol Rev 57: 595-604

Domínguez-Santos R, Martín JF, Kosalková K, Prieto C, Ullán RV, García-Estrada C (2012) The regulatory factor PcRFX1 controls the expression of the three genes of β-lactam biosynthesis in Penicillium chrysogenum. Fungal Genet Biol 49: 866-81

Dong AP, Yoder JA, Zhang X, Zhou L, Bestor TH, Cheng XD (2001) Structure of human DNMT2, an enigmatic DNA methyltransferase homolog that displays denaturant-resistant binding to DNA. Nucleic Acids Res 29: 439-48

Dostie J, Richmond TA, Arnaout RA, Selzer RR, Lee WL, Honan TA, Rubio ED, Krumm A, Lamb J, Nusbaum C, Green RD, Dekker J (2006) Chromosome conformation capture carbon copy (5C): A massively parallel solution for mapping interactions between genomic elements. Genome Res 16: 1299-309

VIII. REFERENCES 56

Dostie J, Dekker J (2007) Mapping networks of physical interactions between genomic elements using 5C technology. Nat Protoc 2: 988-1002

Dowzer CE, Kelly JM (1989) Cloning of the creA gene from Aspergillus nidulans: a gene involved in carbon catabolite repression. Curr Genet 15: 457-9

Dranginis AM (1990) Binding of yeast a1 and α-2 as a heterodimer to the operator DNA of a haploid- specific gene. Nature 347: 682-5

Du RH, Li EG, Cao Y, Song YC, Tan RX (2011) Fumigaclavine C inhibits tumor necrosis factor α production via suppression of toll-like receptor 4 and nuclear factor κB activation in macrophages. Life Sci 89: 235-40

Dyer PS, Paoletti M (2005) Reproduction in Aspergillus fumigatus: sexuality in a supposedly asexual species? Med Mycol 43 Suppl 1: S7-14

Dyer PS, O'Gorman CM (2011) A fungal sexual revolution: Aspergillus and Penicillium show the way. Curr Opin Microbiol 14: 649-54

Eid J, Fehr A, Gray J, Luong K, Lyle J, Otto G, Peluso P, Rank D, Baybayan P, Bettman B, Bibillo A, Bjornson K, Chaudhuri B, Christians F, Cicero R, Clark S, Dalal R, Dewinter A, Dixon J, Foquet M, Gaertner A, Hardenbol P, Heiner C, Hester K, Holden D, Kearns G, Kong X, Kuse R, Lacroix Y, Lin S, Lundquist P, Ma C, Marks P, Maxham M, Murphy D, Park I, Pham T, Phillips M, Roy J, Sebra R, Shen G, Sorenson J, Tomaney A, Travers K, Trulson M, Vieceli J, Wegener J, Wu D, Yang A, Zaccarin D, Zhao P, Zhong F, Korlach J, Turner S (2009) Real-time DNA sequencing from single polymerase molecules. Science 323: 133-8

Eisendle M, Oberegger H, Zadra I, Haas H (2003) The siderophore system is essential for viability of Aspergillus nidulans: functional analysis of two genes encoding L-ornithine N 5- monooxygenase (sidA) and a non-ribosomal peptide synthetase (sidC). Mol Microbiol 49: 359-75

Eisendle M, Schrettl M, Kragl C, Müller D, Illmer P, Haas H (2006) The intracellular siderophore ferricrocin is involved in iron storage, oxidative-stress resistance, germination, and sexual development in Aspergillus nidulans. Eukaryot Cell 5: 1596-603

Elander RP (2003) Industrial production of β-lactam antibiotics. Appl Microbiol Biotechnol 61: 385- 92

Elbein AD, Pan YT, Pastuszak I, Carroll D (2003) New insights on trehalose: a multifunctional molecule. Glycobiology 13: 17R-27R

Espeso EA, Tilburn J, Arst HN, Jr., Peñalva MA (1993) pH regulation is a major determinant in expression of a fungal penicillin biosynthetic gene. EMBO J 12: 3947-56

Esteller M (2006) The necessity of a human epigenome project. Carcinogenesis 27: 1121-5

Etxebeste O, Ni M, Garzia A, Kwon NJ, Fischer R, Yu JH, Espeso EA, Ugalde U (2008) Basic- zipper-type transcription factor FlbB controls asexual development in Aspergillus nidulans. Eukaryot Cell 7: 38-48

VIII. REFERENCES 57

Farnham PJ (2009) Insights from genomic profiling of transcription factors. Nat Rev Genet 10: 605- 16

Fernandes M, Keller NP, Adams TH (1998) Sequence-specific binding by Aspergillus nidulans AflR, a C6 zinc cluster protein regulating mycotoxin biosynthesis. Mol Microbiol 28: 1355-65

Fisher WW, Li JJ, Hammonds AS, Brown JB, Pfeiffer BD, Weiszmann R, MacArthur S, Thomas S, Stamatoyannopoulos JA, Eisen MB, Bickel PJ, Biggin MD, Celniker SE (2012) DNA regions bound at low occupancy by transcription factors do not drive patterned reporter gene expression in Drosophila. Proc Natl Acad Sci U S A 109: 21330-5

Fitzgerald DM, Bonocora RP, Wade JT (2014) Comprehensive mapping of the Escherichia coli flagellar regulatory network. PLoS Genet 10: e1004649

Fleming A (1929) On the antibacterial action of cultures of a Penicillium, with special reference to their use in the isolation of B. influenzae. British Journal of Experimental Pathology 10: 226- 36

Flensburg C, Kinkel SA, Keniry A, Blewitt ME, Oshlack A (2014) A comparison of control samples for ChIP-seq of histone modifications. Front Genet 5: 329

Francois P, Hernandez D, Schrenzel J (2007) Genome content determination in methicillin-resistant Staphylococcus aureus. Future Microbiol 2: 187-98

Fraser JA, Diezmann S, Subaran RL, Allen A, Lengeler KB, Dietrich FS, Heitman J (2004) Convergent evolution of chromosomal sex-determining regions in the animal and fungal kingdoms. PLoS Biol 2: e384

Friedman N (2004) Inferring cellular networks using probabilistic graphical models. Science 303: 799-805

Fullwood MJ, Liu MH, Pan YF, Liu J, Xu H, Mohamed YB, Orlov YL, Velkov S, Ho A, Mei PH, Chew EG, Huang PY, Welboren WJ, Han Y, Ooi HS, Ariyaratne PN, Vega VB, Luo Y, Tan PY, Choy PY, Wansa KD, Zhao B, Lim KS, Leow SC, Yow JS, Joseph R, Li H, Desai KV, Thomsen JS, Lee YK, Karuturi RK, Herve T, Bourque G, Stunnenberg HG, Ruan X, Cacheux-Rataboul V, Sung WK, Liu ET, Wei CL, Cheung E, Ruan Y (2009) An oestrogen-receptor-α-bound human chromatin interactome. Nature 462: 58-64

Furey TS (2012) ChIP-seq and beyond: new and improved methodologies to detect and characterize protein-DNA interactions. Nat Rev Genet 13: 840-52

Galgoczy DJ, Cassidy-Stone A, Llinas M, O'Rourke SM, Herskowitz I, DeRisi JL, Johnson AD (2004) Genomic dissection of the cell-type-specification circuit in Saccharomyces cerevisiae. Proc Natl Acad Sci U S A 101: 18069-74

Gao F, Foat BC, Bussemaker HJ (2004) Defining transcriptional networks through integrative modeling of mRNA expression and transcription factor binding data. BMC Bioinformatics 5: 31

Garzia A, Etxebeste O, Herrero-García E, Ugalde U, Espeso EA (2010) The concerted action of bZip and cMyb transcription factors FlbB and FlbD induces brlA expression and asexual development in Aspergillus nidulans. Mol Microbiol 75: 1314-24

Gaulton KJ, Nammo T, Pasquali L, Simon JM, Giresi PG, Fogarty MP, Panhuis TM, Mieczkowski P, Secchi A, Bosco D, Berney T, Montanya E, Mohlke KL, Lieb JD, Ferrer J (2010) A map of open chromatin in human pancreatic islets. Nat Genet 42: 255-9 VIII. REFERENCES 58

Georges AB, Benayoun BA, Caburet S, Veitia RA (2010) Generic binding sites, generic DNA- binding domains: where does specific promoter recognition come from? Faseb Journal 24: 346-56

Gerke J, Braus GH (2014) Manipulation of fungal development as source of novel secondary metabolites for biotechnology. Appl Microbiol Biotechnol 98: 8443-55

Giese K, Pagel J, Grosschedl R (1994) Distinct DNA-binding properties of the high mobility group domain of murine and human SRY sex-determining factors. Proc Natl Acad Sci U S A 91: 3368-72

Glass NL, Grotelueschen J, Metzenberg RL (1990) Neurospora crassa A mating-type region. Proc Natl Acad Sci U S A 87: 4912-6

Green RE, Krause J, Briggs AW, Maricic T, Stenzel U, Kircher M, Patterson N, Li H, Zhai W, Fritz MH, Hansen NF, Durand EY, Malaspinas AS, Jensen JD, Marques-Bonet T, Alkan C, Prufer K, Meyer M, Burbano HA, Good JM, Schultz R, Aximu-Petri A, Butthof A, Hober B, Hoffner B, Siegemund M, Weihmann A, Nusbaum C, Lander ES, Russ C, Novod N, Affourtit J, Egholm M, Verna C, Rudan P, Brajkovic D, Kucan Z, Gusic I, Doronichev VB, Golovanova LV, Lalueza-Fox C, de la Rasilla M, Fortea J, Rosas A, Schmitz RW, Johnson PL, Eichler EE, Falush D, Birney E, Mullikin JC, Slatkin M, Nielsen R, Kelso J, Lachmann M, Reich D, Paabo S (2010) A draft sequence of the Neandertal genome. Science 328: 710-22

Grosschedl R, Giese K, Pagel J (1994) HMG domain proteins: architectural elements in the assembly of nucleoprotein structures. Trends Genet 10: 94-100

Gubbay J, Collignon J, Koopman P, Capel B, Economou A, Münsterberg A, Vivian N, Goodfellow P, Lovellbadge R (1990) A gene mapping to the sex-determining region of the mouse Y-chromosome is a member of a novel family of embryonically expressed genes. Nature 346: 245-50

Guzmán-de-Peña D, Aguirre J, Ruiz-Herrera J (1998) Correlation between the regulation of sterigmatocystin biosynthesis and asexual and sexual sporulation in Emericella nidulans. Antonie Van Leeuwenhoek 73: 199-205

Haber JE (2012) Mating-type genes and MAT switching in Saccharomyces cerevisiae. Genetics 191: 33-64

Hagen DC, Bruhn L, Westby CA, Sprague GF, Jr. (1993) Transcription of α-specific genes in Saccharomyces cerevisiae: DNA sequence requirements for activity of the coregulator α1. Mol Cell Biol 13: 6866-75

Hahn S, Young ET (2011) Transcriptional regulation in Saccharomyces cerevisiae: transcription factor regulation and function, mechanisms of initiation, and roles of activators and coactivators. Genetics 189: 705-36

Han KH, Lee DB, Kim JH, Kim MS, Han KY, Kim WS, Park YS, Kim HB, Han DM (2003) Environmental factors affecting development of Aspergillus nidulans. J Microbiol 41: 34-40

VIII. REFERENCES 59

Hay A, Tsiantis M (2010) KNOX genes: versatile regulators of plant development and diversity. Development 137: 3153-65

Haynes BC, Skowyra ML, Spencer SJ, Gish SR, Williams M, Held EP, Brent MR, Doering TL (2011) Toward an integrated model of capsule regulation in Cryptococcus neoformans. PLoS Pathog 7: e1002411

He Q, Bardet AF, Patton B, Purvis J, Johnston J, Paulson A, Gogol M, Stark A, Zeitlinger J (2011) High conservation of transcription factor binding and evidence for combinatorial regulation across six Drosophila species. Nat Genet 43: 414-20

Hecht A, Strahl-Bolsinger S, Grunstein M (1996) Spreading of transcriptional repressor SIR3 from telomeric heterochromatin. Nature 383: 92-6

Hicks JK, Yu JH, Keller NP, Adams TH (1997) Aspergillus sporulation and mycotoxin production both require inactivation of the FadA G-α protein-dependent signaling pathway. Embo Journal 16: 4916-23

Hnisz D, Bardet AF, Nobile CJ, Petryshyn A, Glaser W, Schöck U, Stark A, Kuchler K (2012) A histone deacetylase adjusts transcription kinetics at coding sequences during Candida albicans morphogenesis. PLoS Genet 8: e1003118

Ho CY, Adamson JG, Hodges RS, Smith M (1994) Heterodimerization of the yeast MATa1 and MATα2 proteins is mediated by two leucine zipper-like coiled-coil motifs. EMBO J 13: 1403- 13

Hoff B, Pöggeler S, Kück U (2008) Eighty years after its discovery, Fleming's Penicillium strain discloses the secret of its sex. Eukaryot Cell 7: 465-70

Horn BW, Moore GG, Carbone I (2009a) Sexual reproduction in Aspergillus flavus. Mycologia 101: 423-9

Horn BW, Ramirez-Prado JH, Carbone I (2009b) Sexual reproduction and recombination in the aflatoxin-producing fungus Aspergillus parasiticus. Fungal Genet Biol 46: 169-75

Hurley JM, Dasgupta A, Emerson JM, Zhou X, Ringelberg CS, Knabe N, Lipzen AM, Lindquist EA, Daum CG, Barry KW, Grigoriev IV, Smith KM, Galagan JE, Bell-Pedersen D, Freitag M, Cheng C, Loros JJ, Dunlap JC (2014) Analysis of clock-regulated genes in Neurospora reveals widespread posttranscriptional control of metabolic potential. Proc Natl Acad Sci U S A 111: 16995-7002

Hynes MJ (1975) Studies on the role of the areA gene in the regulation of nitrogen catabolism in Aspergillus nidulans. Aust J Biol Sci 28: 301-13

Hynes MJ (1994) Regulatory circuits of the amdS gene of Aspergillus nidulans. Antonie Van Leeuwenhoek 65: 179-82

VIII. REFERENCES 60

Idnurm A, Walton FJ, Floyd A, Heitman J (2008) Identification of the sex genes in an early diverged fungus. Nature 451: 193-6

Jain M, Fiddes IT, Miga KH, Olsen HE, Paten B, Akeson M (2015) Improved data analysis for the MinION nanopore sequencer. Nat Methods 12: 351-6

Jakobsen JS, Braun M, Astorga J, Gustafson EH, Sandmann T, Karzynski M, Carlsson P, Furlong EEM (2007) Temporal ChIP-on-chip reveals Biniou as a universal regulator of the visceral muscle transcriptional network. Gene Dev 21: 2448-60

Jarvis EE, Clark KL, Sprague GF, Jr. (1989) The yeast transcription activator PRTF, a homolog of the mammalian serum response factor, is encoded by the MCM1 gene. Genes Dev 3: 936-45

Jekosch K, Kück U (2000) Glucose dependent transcriptional expression of the cre1 gene in Acremonium chrysogenum strains showing different levels of cephalosporin C production. Curr Genet 37: 388-95

Johnson DS, Mortazavi A, Myers RM, Wold B (2007) Genome-wide mapping of in vivo protein- DNA interactions. Science 316: 1497-502

Johnson L (2008) Iron and siderophores in fungal-host interactions. Mycol Res 112: 170-83

Jung YL, Luquette LJ, Ho JW, Ferrari F, Tolstorukov M, Minoda A, Issner R, Epstein CB, Karpen GH, Kuroda MI, Park PJ (2014) Impact of sequencing depth in ChIP-seq experiments. Nucleic Acids Res 42: e74

Käfer E (1965) Origins of translocations in Aspergillus nidulans. Genetics 52: 217-32

Kale SP, Milde L, Trapp MK, Frisvad JC, Keller NP, Bok JW (2008) Requirement of LaeA for secondary metabolism and sclerotial production in Aspergillus flavus. Fungal Genet Biol 45: 1422-9

Keller NP, Nesbitt C, Sarr B, Phillips TD, Burow GB (1997) pH regulation of sterigmatocystin and aflatoxin biosynthesis in Aspergillus spp. Phytopathology 87: 643-8

VIII. REFERENCES 61

Keller NP, Turner G, Bennett JW (2005) Fungal secondary metabolism - from biochemistry to genomics. Nat Rev Microbiol 3: 937-47

Keszthelyi A, Jeney A, Kerényi Z, Mendes O, Waalwijk C, Hornok L (2007) Tagging target genes of the MAT1-2-1 transcription factor in Fusarium verticillioides (Gibberella fujikuroi MP-A). Antonie Van Leeuwenhoek 91: 373-91

Kim H, Han K, Kim K, Han D, Jahng K, Chae K (2002) The veA gene activates sexual development in Aspergillus nidulans. Fungal Genet Biol 37: 72-80

Kim H, Borkovich KA (2006) Pheromones are essential for male fertility and sufficient to direct chemotropic polarized growth of trichogynes during mating in Neurospora crassa. Eukaryot Cell 5: 544-54

Kim M, Lee KH, Yoon SW, Kim BS, Chun J, Yi H (2013) Analytical tools and databases for metagenomics in the next-generation sequencing era. Genomics Inform 11: 102-13

Kitzman JO, Snyder MW, Ventura M, Lewis AP, Qiu R, Simmons LE, Gammill HS, Rubens CE, Santillan DA, Murray JC, Tabor HK, Bamshad MJ, Eichler EE, Shendure J (2012) Non-invasive whole-genome sequencing of a human fetus. Sci Transl Med 4: 137ra76

Klix V, Nowrousian M, Ringelberg C, Loros JJ, Dunlap JC, Pöggeler S (2010) Functional characterization of MAT1-1-specific mating-type genes in the homothallic ascomycete Sordaria macrospora provides new insights into essential and nonessential sexual regulators. Eukaryot Cell 9: 894-905

Koiwai K, Kubota T, Watanabe N, Hori K, Koiwai O, Masai H (2015) Definition of the transcription factor TdIF1 consensus-binding sequence through genome-wide mapping of its binding sites. Genes Cells 20: 242-54

Kozhevnikova EN, van der Knaap JA, Pindyurin AV, Ozgur Z, van Ijcken WFJ, Moshkin YM, Verrijzer CP (2012) Metabolic enzyme IMPDH is also a transcription factor regulated by cellular state. Mol Cell 47: 133-9

Kronstad JW, Staben C (1997) Mating type in filamentous fungi. Annu Rev Genet 31: 245-76

Kück U, Böhm J (2013) Mating-type genes and cryptic sexuality as tools for genetically manipulating industrial molds. Appl Microbiol Biotechnol 97: 9609-20

Kvaal CA, Srikantha T, Soll DR (1997) Mis-expression of the white-phase-specific gene WH11 in the opaque phase of Candida albicans affects switching and virulence. Infection and Immunity 65: 4468-75

VIII. REFERENCES 62

Lang GI, Murray AW, Botstein D (2009) The cost of gene expression underlies a fitness trade-off in yeast. Proc Natl Acad Sci U S A 106: 5755-60

Lavoie H, Hogues H, Whiteway M (2009) Rearrangements of the transcriptional regulatory networks of metabolic pathways in fungi. Curr Opin Microbiol 12: 655-63

Lee BN, Adams TH (1994) Overexpression of flbA, an early regulator of Aspergillus asexual sporulation, leads to activation of brlA and premature initiation of development. Mol Microbiol 14: 323-34

Lee SC, Ni M, Li W, Shertz C, Heitman J (2010) The evolution of sex: a perspective from the fungal kingdom. Microbiol Mol Biol Rev 74: 298-340

Lee SH, Lee S, Choi D, Lee YW, Yun SH (2006) Identification of the down-regulated genes in a mat1-2-deleted strain of Gibberella zeae, using cDNA subtraction and microarray analysis. Fungal Genet Biol 43: 295-310

Lee TI, Rinaldi NJ, Robert F, Odom DT, Bar-Joseph Z, Gerber GK, Hannett NM, Harbison CT, Thompson CM, Simon I, Zeitlinger J, Jennings EG, Murray HL, Gordon DB, Ren B, Wyrick JJ, Tagne JB, Volkert TL, Fraenkel E, Gifford DK, Young RA (2002) Transcriptional regulatory networks in Saccharomyces cerevisiae. Science 298: 799-804

Lefrançois P, Euskirchen GM, Auerbach RK, Rozowsky J, Gibson T, Yellman CM, Gerstein M, Snyder M (2009) Efficient yeast ChIP-Seq using multiplex short-read DNA sequencing. BMC Genomics 10: 37

Levine M, Tjian R (2003) Transcription regulation and animal diversity. Nature 424: 147-51

Lewis EB (1978) A gene complex controlling segmentation in Drosophila. Nature 276: 565-70

Li B, Carey M, Workman JL (2007) The role of chromatin during transcription. Cell 128: 707-19

Li C, Chen C, Gao L, Yang S, Nguyen V, Shi X, Siminovitch K, Kohalmi SE, Huang S, Wu K, Chen X, Cui Y (2015) The Arabidopsis SWI2/SNF2 chromatin remodeler BRAHMA regulates polycomb function during vegetative development and directly activates the flowering repressor gene SVP. PLoS Genet 11: e1004944

Li H, Johnson AD (2010) Evolution of transcription networks - lessons from yeasts. Curr Biol 20: R746-53

Li R, Fan W, Tian G, Zhu H, He L, Cai J, Huang Q, Cai Q, Li B, Bai Y, Zhang Z, Zhang Y, Wang W, Li J, Wei F, Li H, Jian M, Nielsen R, Li D, Gu W, Yang Z, Xuan Z, Ryder OA, Leung FC, Zhou Y, Cao J, Sun X, Fu Y, Fang X, Guo X, Wang B, Hou R, Shen F, Mu B, Ni P, Lin R, Qian W, Wang G, Yu C, Nie W, Wang J, Wu Z, Liang H, Min J, Wu Q, Cheng S, Ruan J, Wang M, Shi Z, Wen M, Liu B, Ren X, Zheng H, Dong D, Cook K, Shan G, Zhang H, Kosiol C, Xie X, Lu Z, Li Y, Steiner CC, Lam TT, Lin S, Zhang Q, Li G, Tian J, Gong T, Liu H, Zhang D, Fang L, Ye C, Zhang J, Hu W, Xu A, Ren Y, Zhang VIII. REFERENCES 63

G, Bruford MW, Li Q, Ma L, Guo Y, An N, Hu Y, Zheng Y, Shi Y, Li Z, Liu Q, Chen Y, Zhao J, Qu N, Zhao S, Tian F, Wang X, Wang H, Xu L, Liu X, Vinar T, Wang Y, Lam TW, Yiu SM, Liu S, Huang Y, Yang G, Jiang Z, Qin N, Li L, Bolund L, Kristiansen K, Wong GK, Olson M, Zhang X, Li S, Yang H (2010) The sequence and de novo assembly of the giant panda genome. Nature 463: 311-7

Li XY, MacArthur S, Bourgon R, Nix D, Pollard DA, Iyer VN, Hechmer A, Simirenko L, Stapleton M, Luengo Hendriks CL, Chu HC, Ogawa N, Inwood W, Sementchenko V, Beaton A, Weiszmann R, Celniker SE, Knowles DW, Gingeras T, Speed TP, Eisen MB, Biggin MD (2008) Transcription factors bind thousands of active and inactive regions in the Drosophila blastoderm. PLoS Biol 6: e27

Lieberman-Aiden E, van Berkum NL, Williams L, Imakaev M, Ragoczy T, Telling A, Amit I, Lajoie BR, Sabo PJ, Dorschner MO, Sandstrom R, Bernstein B, Bender MA, Groudine M, Gnirke A, Stamatoyannopoulos J, Mirny LA, Lander ES, Dekker J (2009) Comprehensive mapping of long-range interactions reveals folding principles of the human genome. Science 326: 289-93

Lin Z, Wu WS, Liang H, Woo Y, Li WH (2010) The spatial distribution of cis regulatory elements in yeast promoters and its implications for transcriptional regulation. BMC Genomics 11: 581

Lister R, O'Malley RC, Tonti-Filippini J, Gregory BD, Berry CC, Millar AH, Ecker JR (2008) Highly integrated single-base resolution maps of the epigenome in Arabidopsis. Cell 133: 523-36

Liu T, Ortiz JA, Taing L, Meyer CA, Lee B, Zhang Y, Shin H, Wong SS, Ma J, Lei Y, Pape UJ, Poidinger M, Chen Y, Yeung K, Brown M, Turpaz Y, Liu XS (2011) Cistrome: an integrative platform for transcriptional regulation studies. Genome Biol 12: R83

López-Berges MS, Hera C, Sulyok M, Schäfer K, Capilla J, Guarro J, Di Pietro A (2013) The velvet complex governs mycotoxin production and virulence of Fusarium oxysporum on plant and mammalian hosts. Mol Microbiol 87: 49-65

MacArthur S, Li XY, Li J, Brown JB, Chu HC, Zeng L, Grondona BP, Hechmer A, Simirenko L, Keranen SV, Knowles DW, Stapleton M, Bickel P, Biggin MD, Eisen MB (2009) Developmental roles of 21 Drosophila transcription factors are determined by quantitative differences in binding to an overlapping set of thousands of genomic regions. Genome Biol 10: R80

Machanick P, Bailey TL (2011) MEME-ChIP: motif analysis of large DNA datasets. Bioinformatics 27: 1696-7

MacQuarrie KL, Fong AP, Morse RH, Tapscott SJ (2011) Genome-wide transcription factor binding: beyond direct target regulation. Trends Genet 27: 141-8

Magee BB, Magee PT (2000) Induction of mating in Candida albicans by construction of MTLa and MTLα strains. Science 289: 310-3

Malarkey CS, Churchill MEA (2012) The high mobility group box: the ultimate utility player of a cell. Trends Biochem Sci 37: 553-62

VIII. REFERENCES 64

Mallo M, Wellik DM, Deschamps J (2010) Hox genes and regional patterning of the vertebrate body plan. Dev Biol 344: 7-15

Manzoni M, Rollini M (2002) Biosynthesis and biotechnological production of statins by filamentous fungi and application of these cholesterol-lowering drugs. Appl Microbiol Biotechnol 58: 555- 64

Mardis ER (2008) The impact of next-generation sequencing technology on genetics. Trends Genet 24: 133-41

Margulies M, Egholm M, Altman WE, Attiya S, Bader JS, Bemben LA, Berka J, Braverman MS, Chen YJ, Chen Z, Dewell SB, Du L, Fierro JM, Gomes XV, Godwin BC, He W, Helgesen S, Ho CH, Irzyk GP, Jando SC, Alenquer ML, Jarvie TP, Jirage KB, Kim JB, Knight JR, Lanza JR, Leamon JH, Lefkowitz SM, Lei M, Li J, Lohman KL, Lu H, Makhijani VB, McDade KE, McKenna MP, Myers EW, Nickerson E, Nobile JR, Plant R, Puc BP, Ronan MT, Roth GT, Sarkis GJ, Simons JF, Simpson JW, Srinivasan M, Tartaro KR, Tomasz A, Vogt KA, Volkmer GA, Wang SH, Wang Y, Weiner MP, Yu P, Begley RF, Rothberg JM (2005) Genome sequencing in microfabricated high-density picolitre reactors. Nature 437: 376-80

Martin JL, McMillan FM (2002) SAM (dependent) I AM: the S-adenosylmethionine-dependent methyltransferase fold. Curr Opin Struct Biol 12: 783-93

Martin T, Lu SW, van Tilbeurgh H, Ripoll DR, Dixelius C, Turgeon BG, Debuchy R (2010) Tracing the origin of the fungal α1 domain places its ancestor in the HMG-box superfamily: implication for fungal mating-type evolution. PLoS One 5: e15199

Mata J, Bahler J (2006) Global roles of Ste11p, cell type, and pheromone in the control of gene expression during early sexual differentiation in fission yeast. Proc Natl Acad Sci U S A 103: 15517-22

Mathelier A, Zhao X, Zhang AW, Parcy F, Worsley-Hunt R, Arenillas DJ, Buchman S, Chen CY, Chou A, Ienasescu H, Lim J, Shyr C, Tan G, Zhou M, Lenhard B, Sandelin A, Wasserman WW (2014) JASPAR 2014: an extensively expanded and updated open-access database of transcription factor binding profiles. Nucleic Acids Res 42: D142-7

McGhee JD, Hippel PHV (1975a) Formaldehyde as a probe of DNA structure I. Reaction with exocyclic amino-groups of DNA bases. Biochemistry 14: 1281-96

McGhee JD, Hippel PHV (1975b) Formaldehyde as a probe of DNA structure II. Reaction with endocyclic imino groups of DNA bases. Biochemistry 14: 1297-303

McPherson JD (2014) A defining decade in DNA sequencing. Nat Methods 11: 1003-5

Mead J, Bruning AR, Gill MK, Steiner AM, Acton TB, Vershon AK (2002) Interactions of the Mcm1 MADS box protein with cofactors that regulate mating in yeast. Mol Cell Biol 22: 4607-21

Mead ME, Stanton BC, Kruzel EK, Hull CM (2014) Targets of the Sex Inducer homeodomain proteins are required for fungal development and virulence in Cryptococcus neoformans. Mol Microbiol 95: 804-18

Merelo P, Xie Y, Brand L, Ott F, Weigel D, Bowman JL, Heisler MG, Wenkel S (2013) Genome- wide identification of KANADI1 target genes. PLoS One 8: e77341

VIII. REFERENCES 65

Merlini L, Dudin O, Martin SG (2013) Mate and fuse: how yeast cells do it. Open Biology 3: 130008

Metzenberg RL, Glass NL (1990) Mating type and mating strategies in Neurospora. Bioessays 12: 53-9

Metzker ML (2010) Sequencing technologies - the next generation. Nat Rev Genet 11: 31-46

Meyer CA, Liu XS (2014) Identifying and mitigating bias in next-generation sequencing methods for chromatin biology. Nat Rev Genet 15: 709-21

Mihlan M, Homann V, Liu TWD, Tudzynski B (2003) AREA directly mediates nitrogen regulation of gibberellin biosynthesis in Gibberella fujikuroi, but its activity is not affected by NMR. Mol Microbiol 47: 975-91

Mooney JL, Yager LN (1990) Light is required for conidiation in Aspergillus nidulans. Genes Dev 4: 1473-82

Morin RD, Bainbridge M, Fejes A, Hirst M, Krzywinski M, Pugh TJ, McDonald H, Varhol R, Jones SJM, Marra MA (2008) Profiling the HeLa S3 transcriptome using randomly primed cDNA and massively parallel short-read sequencing. Biotechniques 45: 81-94

Mortazavi A, Williams BA, McCue K, Schaeffer L, Wold B (2008) Mapping and quantifying mammalian transcriptomes by RNA-Seq. Nat Methods 5: 621-8

Mukherjee K, Brocchieri L, Burglin TR (2009) A comprehensive classification and evolutionary analysis of plant homeobox genes. Mol Biol Evol 26: 2775-94

Nagalakshmi U, Wang Z, Waern K, Shou C, Raha D, Gerstein M, Snyder M (2008) The transcriptional landscape of the yeast genome defined by RNA sequencing. Science 320: 1344-9

Nakano M, Komatsu J, Matsuura S, Takashima K, Katsura S, Mizuno A (2003) Single-molecule PCR using water-in-oil emulsion. J Biotechnol 102: 117-24

Nasmith CG, Walkowiak S, Wang L, Leung WW, Gong Y, Johnston A, Harris LJ, Guttman DS, Subramaniam R (2011) Tri6 is a global transcription regulator in the phytopathogen Fusarium graminearum. PLoS Pathog 7: e1002266

Nelson PE, Desjardins AE, Plattner RD (1993) Fumonisins, mycotoxins produced by fusarium species: biology, chemistry, and significance. Annu Rev Phytopathol 31: 233-52

Nepomnyashchaya YN, Artemov AV, Roumiantsev SA, Roumyantsev AG, Zhavoronkov A (2013) Non-invasive prenatal diagnostics of aneuploidy using next-generation DNA sequencing technologies, and clinical considerations. Clin Chem Lab Med 51: 1141-54

VIII. REFERENCES 66

Nguyen DH, D'Haeseleer P (2006) Deciphering principles of transcription regulation in eukaryotic genomes. Mol Syst Biol 2: 2006.0012

Ni M, Yu JH (2007) A novel regulator couples sporogenesis and trehalose biogenesis in Aspergillus nidulans. PLoS One 2: e970

Ni M, Feretzaki M, Sun S, Wang X, Heitman J (2011) Sex in fungi. Annu Rev Genet 45: 405-30

Niehaus EM, Kleigrewe K, Wiemann P, Studt L, Sieber CMK, Connolly LR, Freitag M, Güldener U, Tudzynski B, Humpf HU (2013) Genetic manipulation of the Fusarium fujikuroi fusarin gene cluster yields insight into the complex regulation and fusarin biosynthetic pathway. Chem Biol 20: 1055-66

Nolting N, Pöggeler S (2006) A MADS box protein interacts with a mating-type protein and is required for fruiting body development in the homothallic ascomycete Sordaria macrospora. Eukaryot Cell 5: 1043-56

Noma K, Allis CD, Grewal SIS (2001) Transitions in distinct histone H3 methylation patterns at the heterochromatin domain boundaries. Science 293: 1150-5

Nowrousian M, Stajich JE, Chu M, Engh I, Espagne E, Halliday K, Kamerewerd J, Kempken F, Knab B, Kuo HC, Osiewacz HD, Pöggeler S, Read ND, Seiler S, Smith KM, Zickler D, Kück U, Freitag M (2010) De novo assembly of a 40 Mb eukaryotic genome from short sequence reads: Sordaria macrospora, a model organism for fungal morphogenesis. PLoS Genet 6: e1000891

O'Gorman CM, Fuller H, Dyer PS (2009) Discovery of a sexual cycle in the opportunistic fungal pathogen Aspergillus fumigatus. Nature 457: 471-4

Oiartzabal-Arano E, Garzia A, Gorostidi A, Ugalde U, Espeso EA, Etxebeste O (2015) Beyond asexual development: modifications in the gene expression profile caused by the absence of the Aspergillus nidulans transcription factor FlbB. Genetics (doi: 10.1534/genetics.155.174342; Epub ahead of print)

Oide S, Krasnoff SB, Gibson DM, Turgeon BG (2007) Intracellular siderophores are essential for ascomycete sexual development in heterothallic Cochliobolus heterostrophus and homothallic Gibberella zeae. Eukaryot Cell 6: 1339-53

Orlando V (2000) Mapping chromosomal proteins in vivo by formaldehyde-crosslinked-chromatin immunoprecipitation. Trends Biochem Sci 25: 99-104

Pachkov M, Erb I, Molina N, van Nimwegen E (2007) SwissRegulon: a database of genome-wide annotations of regulatory sites. Nucleic Acids Res 35: D127-31

Palmer JM, Keller NP (2010) Secondary metabolism in fungi: does chromosomal location matter? Curr Opin Microbiol 13: 431-6

Paoletti M, Seymour FA, Alcocer MJ, Kaur N, Calvo AM, Archer DB, Dyer PS (2007) Mating type and the genetic basis of self-fertility in the model fungus Aspergillus nidulans. Curr Biol 17: 1384-9

VIII. REFERENCES 67

Paris M, Kaplan T, Li XY, Villalta JE, Lott SE, Eisen MB (2013) Extensive divergence of transcription factor binding in Drosophila embryos with highly conserved gene expression. PLoS Genet 9: e1003748

Park D, Lee Y, Bhupindersingh G, Iyer VR (2013) Widespread mis-interpretable ChIP-seq bias in yeast. PLoS One 8: e83506

Park PJ (2009) ChIP-seq: advantages and challenges of a maturing technology. Nat Rev Genet 10: 669-80

Peñalva MA, Rowlands RT, Turner G (1998) The optimization of penicillin biosynthesis in fungi. Trends Biotechnol 16: 483-9

Pepke S, Wold B, Mortazavi A (2009) Computation for ChIP-seq and RNA-seq studies. Nat Methods 6: S22-32

Perrin RM, Fedorova ND, Bok JW, Cramer RA, Wortman JR, Kim HS, Nierman WC, Keller NP (2007) Transcriptional regulation of chemical diversity in Aspergillus fumigatus by LaeA. PLoS Pathog 3: e50

Picossi S, Flores E, Herrero A (2015) The LysR-type transcription factor PacR is a global regulator of photosynthetic carbon assimilation in Anabaena. Environ Microbiol (doi: 10.1111/1462- 2920.12800; Epub ahead of print)

Pilon AM, Ajay SS, Kumar SA, Steiner LA, Cherukuri PF, Wincovitch S, Anderson SM, Mullikin JC, Gallagher PG, Hardison RC, Margulies EH, Bodine DM (2011) Genome- wide ChIP-seq reveals a dramatic shift in the binding of the transcription factor erythroid Kruppel-like factor during erythrocyte differentiation. Blood 118: e139-48

Pöggeler S (2001) Mating-type genes for classical strain improvements of ascomycetes. Appl Microbiol Biotechnol 56: 589-601

Pöggeler S, Nowrousian M, Ringelberg C, Loros JJ, Dunlap JC, Kück U (2006) Microarray and real-time PCR analyses reveal mating type-dependent gene expression in a homothallic fungus. Mol Genet Genomics 275: 492-503

Pujol C, Daniels KJ, Lockhart SR, Srikantha T, Radke JB, Geiger J, Soll DR (2004) The closely related species Candida albicans and Candida dubliniensis can mate. Eukaryot Cell 3: 1015- 27

Rabinovich A, Jin VX, Rabinovich R, Xu X, Farnham PJ (2008) E2F in vivo binding specificity: comparison of consensus versus nonconsensus binding sites. Genome Res 18: 1763-77

Raper KB, Alexander DF, Coghill RD (1944) Penicillin: II. Natural variation and penicillin production in Penicillium notatum and allied species. J Bacteriol 48: 639-59

Raper KB (1946) The development of improved penicillin-producing molds. Annals of the New York Academy of Sciences 48: 41-56

Ravasi T, Suzuki H, Cannistraci CV, Katayama S, Bajic VB, Tan K, Akalin A, Schmeier S, Kanamori-Katayama M, Bertin N, Carninci P, Daub CO, Forrest AR, Gough J, Grimmond S, Han JH, Hashimoto T, Hide W, Hofmann O, Kamburov A, Kaur M, VIII. REFERENCES 68

Kawaji H, Kubosaki A, Lassmann T, van Nimwegen E, MacPherson CR, Ogawa C, Radovanovic A, Schwartz A, Teasdale RD, Tegner J, Lenhard B, Teichmann SA, Arakawa T, Ninomiya N, Murakami K, Tagami M, Fukuda S, Imamura K, Kai C, Ishihara R, Kitazume Y, Kawai J, Hume DA, Ideker T, Hayashizaki Y (2010) An atlas of combinatorial transcriptional regulation in mouse and man. Cell 140: 744-52

Read CM, Cary PD, Cranerobinson C, Driscoll PC, Norman DG (1993) Solution structure of a DNA-binding domain from Hmg1. Nucleic Acids Res 21: 3427-36

Reed BD, Charos AE, Szekely AM, Weissman SM, Snyder M (2008) Genome-wide occupancy of SREBP1 and its partners NFY and SP1 reveals novel functional roles and combinatorial regulation of distinct classes of genes. PLoS Genet 4: e1000133

Regueira TB, Kildegaard KR, Hansen BG, Mortensen UH, Hertweck C, Nielsen J (2011) Molecular basis for mycophenolic acid biosynthesis in Penicillium brevicompactum. Appl Environ Microbiol 77: 3035-43

Reiss J (1982) Development of Aspergillus parasiticus and formation of aflatoxin B1 under the influence of conidiogenesis affecting compounds. Arch Microbiol 133: 236-8

Ren B, Robert F, Wyrick JJ, Aparicio O, Jennings EG, Simon I, Zeitlinger J, Schreiber J, Hannett N, Kanin E, Volkert TL, Wilson CJ, Bell SP, Young RA (2000) Genome-wide location and function of DNA binding proteins. Science 290: 2306-9

Reyes-Dominguez Y, Bok JW, Berger H, Shwab EK, Basheer A, Gallmetzer A, Scazzocchio C, Keller N, Strauss J (2010) Heterochromatic marks are associated with the repression of secondary metabolism clusters in Aspergillus nidulans. Mol Microbiol 76: 1376-86

Ricardi MM, González RM, Zhong SL, Domínguez PG, Duffy T, Turjanski PG, Salter JDS, Alleva K, Carrari F, Giovannoni JJ, Estévez JM, Iusem ND (2014) Genome-wide data (ChIP-seq) enabled identification of cell wall-related and aquaporin genes as targets of tomato ASR1, a drought stress-responsive transcription factor. BMC Plant Biol 14: 29

Robertson G, Hirst M, Bainbridge M, Bilenky M, Zhao Y, Zeng T, Euskirchen G, Bernier B, Varhol R, Delaney A, Thiessen N, Griffith OL, He A, Marra M, Snyder M, Jones S (2007) Genome-wide profiles of STAT1 DNA association using chromatin immunoprecipitation and massively parallel sequencing. Nat Methods 4: 651-7

Rodriguez R, Miller KM (2014) Unravelling the genomic targets of small molecules using high- throughput sequencing. Nat Rev Genet 15: 783-96

Ropars J, López-Villavicencio M, Dupont J, Snirc A, Gillot G, Coton M, Jany JL, Coton E, Giraud T (2014) Induction of sexual reproduction and genetic diversity in the cheese fungus Penicillium roqueforti. Evol Appl 7: 433-41

Rothberg JM, Hinz W, Rearick TM, Schultz J, Mileski W, Davey M, Leamon JH, Johnson K, Milgrew MJ, Edwards M, Hoon J, Simons JF, Marran D, Myers JW, Davidson JF, Branting A, Nobile JR, Puc BP, Light D, Clark TA, Huber M, Branciforte JT, Stoner IB, Cawley SE, Lyons M, Fu Y, Homer N, Sedova M, Miao X, Reed B, Sabina J, Feierstein E, Schorn M, Alanjary M, Dimalanta E, Dressman D, Kasinskas R, Sokolsky T, Fidanza JA, Namsaraev E, McKernan KJ, Williams A, Roth GT, Bustillo J (2011) An integrated semiconductor device enabling non-optical genome sequencing. Nature 475: 348-52

Ruger-Herreros C, Rodríguez-Romero J, Fernández-Barranco R, Olmedo M, Fischer R, Corrochano LM, Canovas D (2011) Regulation of conidiation by light in Aspergillus nidulans. Genetics 188: 809-22 VIII. REFERENCES 69

Sandelin A, Alkema W, Engstrom P, Wasserman WW, Lenhard B (2004) JASPAR: an open- access database for eukaryotic transcription factor binding profiles. Nucleic Acids Res 32: D91-4

Sanford JR, Wang X, Mort M, Vanduyn N, Cooper DN, Mooney SD, Edenberg HJ, Liu Y (2009) Splicing factor SFRS1 recognizes a functionally diverse landscape of RNA transcripts. Genome Res 19: 381-94

Sanger F, Nicklen S, Coulson AR (1977) DNA sequencing with chain-terminating inhibitors. Proc Natl Acad Sci U S A 74: 5463-7

Sarikaya-Bayram Ö, Palmer JM, Keller N, Braus GH, Bayram Ö (2015) One Juliet and four Romeos: VeA and its methyltransferases. Front Microbiol 6: 1

Sasaki T, Lynch KL, Mueller CV, Friedman S, Freitag M, Lewis ZA (2014) Heterochromatin controls γH2A localization in Neurospora crassa. Eukaryot Cell 13: 990-1000

Scharf DH, Heinekamp T, Remme N, Hortschansky P, Brakhage AA, Hertweck C (2012) Biosynthesis and function of gliotoxin in Aspergillus fumigatus. Appl Microbiol Biotechnol 93: 467-72

Schloss JA (2008) How to get genomes at one ten-thousandth the cost. Nat Biotechnol 26: 1113-5

Schmitt EK, Bunse A, Janus D, Hoff B, Friedlin E, Kürnsteiner H, Kück U (2004a) Winged helix transcription factor CPCR1 is involved in regulation of β-lactam biosynthesis in the fungus Acremonium chrysogenum. Eukaryot Cell 3: 121-34

Schmitt EK, Hoff B, Kück U (2004b) AcFKH1, a novel member of the forkhead family, associates with the RFX transcription factor CPCR1 in the cephalosporin C-producing fungus Acremonium chrysogenum. Gene 342: 269-81

Schneider GF, Dekker C (2012) DNA sequencing with nanopores. Nat Biotechnol 30: 326-8

Schrettl M, Bignell E, Kragl C, Sabiha Y, Loss O, Eisendle M, Wallner A, Arst HN, Haynes K, Haas H (2007) Distinct roles for intra- and extracellular siderophores during Aspergillus fumigatus infection. PLoS Pathog 3: 1195-207

Seiboth B, Karimi Aghcheh R, Phatale PA, Linke R, Hartl L, Sauer DG, Smith KM, Baker SE, Freitag M, Kubicek CP (2012) The putative protein methyltransferase LAE1 controls cellulase gene expression in Trichoderma reesei. Mol Microbiol 84: 1150-64

Seidl V, Seibel C, Kubicek CP, Schmoll M (2009) Sexual development in the industrial workhorse Trichoderma reesei. Proc Natl Acad Sci U S A 106: 13909-14

VIII. REFERENCES 70

Sharon E, Lubliner S, Segal E (2008) A feature-based approach to modeling protein-DNA interactions. PLoS Comput Biol 4: e1000154

Sharon E, Kalma Y, Sharp A, Raveh-Sadka T, Levo M, Zeevi D, Keren L, Yakhini Z, Weinberger A, Segal E (2012) Inferring gene regulatory logic from high-throughput measurements of thousands of systematically designed promoters. Nat Biotechnol 30: 521-30

Shelest E (2008) Transcription factors in fungi. FEMS Microbiol Lett 286: 145-51

Shendure J, Porreca GJ, Reppas NB, Lin X, McCutcheon JP, Rosenbaum AM, Wang MD, Zhang K, Mitra RD, Church GM (2005) Accurate multiplex polony sequencing of an evolved bacterial genome. Science 309: 1728-32

Shendure J, Ji H (2008) Next-generation DNA sequencing. Nat Biotechnol 26: 1135-45

Shore P, Sharrocks AD (1995) The MADS-box family of transcription factors. Eur J Biochem 229: 1-13

Siggers T, Duyzend MH, Reddy J, Khan S, Bulyk ML (2011) Non-DNA-binding cofactors enhance DNA-binding specificity of a transcriptional regulatory complex. Mol Syst Biol 7: 555

Simon MD, Wang CI, Kharchenko PV, West JA, Chapman BA, Alekseyenko AA, Borowsky ML, Kuroda MI, Kingston RE (2011) The genomic binding sites of a non-coding RNA. Proc Natl Acad Sci U S A 108: 20497-502

Slattery M, Zhou T, Yang L, Dantas Machado AC, Gordan R, Rohs R (2014) Absence of a simple code: how transcription factors read the genome. Trends Biochem Sci 39: 381-99

Smith AM, Heisler LE, St Onge RP, Farias-Hesson E, Wallace IM, Bodeau J, Harris AN, Perry KM, Giaever G, Pourmand N, Nislow C (2010) Highly-multiplexed barcode sequencing: an efficient method for parallel analysis of pooled samples. Nucleic Acids Res 38: e142

Smith KM, Phatale PA, Sullivan CM, Pomraning KR, Freitag M (2011) Heterochromatin is required for normal distribution of Neurospora crassa CenH3. Mol Cell Biol 31: 2528-42

Song L, Zhang Z, Grasfeder LL, Boyle AP, Giresi PG, Lee BK, Sheffield NC, Graf S, Huss M, Keefe D, Liu Z, London D, McDaniell RM, Shibata Y, Showers KA, Simon JM, Vales T, Wang T, Winter D, Clarke ND, Birney E, Iyer VR, Crawford GE, Lieb JD, Furey TS (2011) Open chromatin defined by DNaseI and FAIRE identifies regulatory elements that shape cell-type identity. Genome Res 21: 1757-67

Soyer JL, Moller M, Schotanus K, Connolly LR, Galazka JM, Freitag M, Stukenbrock EH (2015) Chromatin analyses of Zymoseptoria tritici: methods for chromatin immunoprecipitation followed by high-throughput sequencing (ChIP-seq). Fungal Genet Biol (doi: doi:10.1016/j.fgb.2015.03.006; Epub ahead of print)

Specht T, Dahlmann TA, Zadra I, Kürnsteiner H, Kück U (2014) Complete sequencing and chromosome-scale genome assembly of the industrial progenitor strain P2niaD18 from the penicillin producer Penicillium chrysogenum. Genome Announc 2: e00577-14

Stergachis AB, Haugen E, Shafer A, Fu W, Vernot B, Reynolds A, Raubitschek A, Ziegler S, LeProust EM, Akey JM, Stamatoyannopoulos JA (2013) Exonic transcription factor binding directs codon choice and affects protein evolution. Science 342: 1367-72

Stewart AJ, Hannenhalli S, Plotkin JB (2012) Why transcription factor binding sites are ten nucleotides long. Genetics 192: 973-85 VIII. REFERENCES 71

Stinnett SM, Espeso EA, Cobeno L, Araujo-Bazan L, Calvo AM (2007) Aspergillus nidulans VeA subcellular localization is dependent on the importin α-carrier and on light. Mol Microbiol 63: 242-55

Stormo GD, Zhao Y (2010) Determining the specificity of protein-DNA interactions. Nat Rev Genet 11: 751-60

Strahl-Bolsinger S, Hecht A, Luo K, Grunstein M (1997) SIR2 and SIR4 interactions differ in core and extended telomeric heterochromatin in yeast. Genes Dev 11: 83-93

Studt L, Schmidt FJ, Jahn L, Sieber CM, Connolly LR, Niehaus EM, Freitag M, Humpf HU, Tudzynski B (2013) Two histone deacetylases, FfHda1 and FfHda2, are important for Fusarium fujikuroi secondary metabolism and virulence. Appl Environ Microbiol 79: 7719-34

Suarez T, Peñalva MA (1996) Characterization of a Penicillium chrysogenum gene encoding a PacC transcription factor and its binding sites in the divergent pcbAB-pcbC promoter of the penicillin biosynthetic cluster. Mol Microbiol 20: 529-40

Sugui JA, Pardo J, Chang YC, Müllbacher A, Zarember KA, Galvez EM, Brinster L, Zerfas P, Gallin JI, Simon MM, Kwon-Chung KJ (2007) Role of laeA in the regulation of alb1, gliP, conidial morphology, and virulence in Aspergillus fumigatus. Eukaryot Cell 6: 1552-61

Survase SA, Kagliwal LD, Annapure US, Singhal RS (2011) Cyclosporin A - a review on fermentative production, downstream processing and pharmacological applications. Biotechnol Adv 29: 418-35

Sutton P, Newcombe NR, Waring P, Müllbacher A (1994) In vivo immunosuppressive activity of gliotoxin, a metabolite produced by human pathogenic fungi. Infect Immun 62: 1192-8

Svingen T, Koopman P (2007) Involvement of homeobox genes in mammalian sexual development. Sex Dev 1: 12-23

Tanaka TU, Knapp D, Nasmyth K (1997) Loading of an Mcm protein onto DNA replication origins is regulated by Cdc6p and CDKs. Cell 90: 649-60

Tanay A (2006) Extensive low-affinity transcriptional interactions in the yeast genome. Genome Res 16: 962-72

Tawfik DS, Griffiths AD (1998) Man-made cell-like compartments for molecular evolution. Nat Biotechnol 16: 652-6

TePaske MR, Gloer JB, Wicklow DT, Dowd PF (1992) Aflavarin and β-Aflatrem - New anti- insectan metabolites from the sclerotia of Aspergillus flavus. J Nat Prod 55: 1080-6

Terfehr D, Dahlmann TA, Specht T, Zadra I, Kürnsteiner H, Kück U (2014) Genome sequence and annotation of Acremonium chrysogenum, producer of the β-lactam antibiotic Cephalosporin C. Genome Announc 2: e00948-14

Tfelt-Hansen PC, Koehler PJ (2008) History of the use of ergotamine and dihydroergotamine in migraine from 1906 and onward. Cephalalgia 28: 877-86

VIII. REFERENCES 72

The ENCODE Project Consortium (2004) The ENCODE (ENCyclopedia Of DNA Elements) Project. Science 306: 636-40

The ENCODE Project Consortium (2012) An integrated encyclopedia of DNA elements in the human genome. Nature 489: 57-74

The International Human Genome Sequencing Consortium (2004) Finishing the euchromatic sequence of the human genome. Nature 431: 931-45

Then Bergh K, Litzka O, Brakhage AA (1996) Identification of a major cis-acting DNA element controlling the bidirectionally transcribed penicillin biosynthesis genes acvA (pcbAB) and ipnA (pcbC) of Aspergillus nidulans. J Bacteriol 178: 3908-16

Thürmer A (2014) Next Generation Sequencing in der mikrobiellen (Meta-)Genomforschung. BIOSpektrum 20: 168-71

Thurtle DM, Rine J (2014) The molecular topography of silenced chromatin in Saccharomyces cerevisiae. Genes Dev 28: 245-58

Tilburn J, Sarkar S, Widdick DA, Espeso EA, Orejas M, Mungroo J, Peñalva MA, Arst HN, Jr. (1995) The Aspergillus PacC zinc finger transcription factor mediates regulation of both acid- and alkaline-expressed genes by ambient pH. EMBO J 14: 779-90

Todeschini AL, Georges A, Veitia RA (2014) Transcription factors: specific DNA binding and specific gene regulation. Trends Genet 30: 211-9

Tompa M, Li N, Bailey TL, Church GM, De Moor B, Eskin E, Favorov AV, Frith MC, Fu Y, Kent WJ, Makeev VJ, Mironov AA, Noble WS, Pavesi G, Pesole G, Regnier M, Simonis N, Sinha S, Thijs G, van Helden J, Vandenbogaert M, Weng Z, Workman C, Ye C, Zhu Z (2005) Assessing computational tools for the discovery of transcription factor binding sites. Nat Biotechnol 23: 137-44

Traeger S, Altegoer F, Freitag M, Gabaldon T, Kempken F, Kumar A, Marcet-Houben M, Pöggeler S, Stajich JE, Nowrousian M (2013) The genome and development-dependent transcriptomes of Pyronema confluens: a window into fungal evolution. PLoS Genet 9: e1003820

Tsong AE, Miller MG, Raisner RM, Johnson AD (2003) Evolution of a combinatorial transcriptional circuit: a case study in yeasts. Cell 115: 389-99

Tuch BB, Galgoczy DJ, Hernday AD, Li H, Johnson AD (2008) The evolution of combinatorial gene regulation in fungi. PLoS Biol 6: e38

Tudzynski B (1999) Biosynthesis of gibberellins in Gibberella fujikuroi: biomolecular aspects. Appl Microbiol Biotechnol 52: 298-310

Tudzynski B, Homann V, Feng B, Marzluf GA (1999a) Isolation, characterization and disruption of the areA nitrogen regulatory gene of Gibberella fujikuroi. Mol Gen Genet 261: 106-14

Tudzynski P, Holter K, Correia T, Arntz C, Grammel N, Keller U (1999b) Evidence for an ergot alkaloid gene cluster in Claviceps purpurea. Mol Gen Genet 261: 133-41

Turgeon BG, Yoder OC (2000) Proposed nomenclature for mating type genes of filamentous ascomycetes. Fungal Genet Biol 31: 1-5

VIII. REFERENCES 73

Valouev A, Johnson DS, Sundquist A, Medina C, Anton E, Batzoglou S, Myers RM, Sidow A (2008) Genome-wide analysis of transcription factor binding sites based on ChIP-Seq data. Nat Methods 5: 829-34 van den Berg MA, Albang R, Albermann K, Badger JH, Daran JM, Driessen AJM, Garcia- Estrada C, Fedorova ND, Harris DM, Heijne WHM, Joardar V, Kiel JAKW, Kovalchuk A, Martín JF, Nierman WC, Nijland JG, Pronk JT, Roubos JA, van der Klei IJ, van Peij NNME, Veenhuis M, von Döhren H, Wagner C, Wortman J, Bovenberg RAL (2008) Genome sequencing and analysis of the filamentous fungus Penicillium chrysogenum. Nat Biotechnol 26: 1161-8 van Dijk EL, Auger H, Jaszczyszyn Y, Thermes C (2014) Ten years of next-generation sequencing technology. Trends Genet 30: 418-26

Veiga T, Nijland JG, Driessen AJ, Bovenberg RA, Touw H, van den Berg MA, Pronk JT, Daran JM (2012) Impact of velvet complex on transcriptome and penicillin G production in glucose- limited chemostat cultures of a β-lactam high-producing Penicillium chrysogenum strain. OMICS 16: 320-33

Vokes SA, Ji H, Wong WH, McMahon AP (2008) A genome-scale analysis of the cis-regulatory circuitry underlying sonic hedgehog-mediated patterning of the mammalian limb. Genes Dev 22: 2651-63

Wada R, Maruyama J, Yamaguchi H, Yamamoto N, Wagu Y, Paoletti M, Archer DB, Dyer PS, Kitamoto K (2012) Presence and functionality of mating type genes in the supposedly asexual filamentous fungus Aspergillus oryzae. Appl Environ Microbiol 78: 2819-29

Wal M, Pugh BF (2012) Genome-wide mapping of nucleosome positions in yeast using high- resolution MNase ChIP-Seq. Methods Enzymol 513: 233-50

Walhout AJM (2006) Unraveling transcription regulatory networks by protein-DNA and protein- protein interaction mapping. Genome Res 16: 1445-54

Wang X, Zhao Z, Muller J, Iyu A, Khng AJ, Guccione E, Ruan Y, Ingham PW (2013) Targeted inactivation and identification of targets of the Gli2a transcription factor in the zebrafish. Biol Open 2: 1203-13

Wang Z, Gerstein M, Snyder M (2009) RNA-Seq: a revolutionary tool for transcriptomics. Nat Rev Genet 10: 57-63

Wang Z, Kin K, Lopez-Giraldez F, Johannesson H, Townsend JP (2012) Sex-specific gene expression during asexual development of Neurospora crassa. Fungal Genet Biol 49: 533-43

Wang Z, Lopez-Giraldez F, Lehr N, Farré M, Common R, Trail F, Townsend JP (2014) Global gene expression and focused knockout analysis reveals genes associated with fungal fruiting body development in Neurospora crassa. Eukaryot Cell 13: 154-69

Weingarten-Gabbay S, Segal E (2014) The grammar of transcriptional regulation. Hum Genet 133: 701-11

Weirauch MT, Cote A, Norel R, Annala M, Zhao Y, Riley TR, Saez-Rodriguez J, Cokelaer T, Vedenko A, Talukder S, Bussemaker HJ, Morris QD, Bulyk ML, Stolovitzky G, Hughes TR (2013) Evaluation of methods for modeling transcription factor sequence specificity. Nat Biotechnol 31: 126-34

VIII. REFERENCES 74

Werner MH, Huth JR, Gronenborn AM, Clore GM (1995) Molecular basis of human 46X,Y sex reversal revealed from the three-dimensional solution structure of the human SRY-DNA complex. Cell 81: 705-14

Werner T (2010) Next generation sequencing in functional genomics. Brief Bioinform 11: 499-511

Wiemann P, Sieber CM, von Bargen KW, Studt L, Niehaus EM, Espino JJ, Huss K, Michielse CB, Albermann S, Wagner D, Bergner SV, Connolly LR, Fischer A, Reuter G, Kleigrewe K, Bald T, Wingfield BD, Ophir R, Freeman S, Hippler M, Smith KM, Brown DW, Proctor RH, Münsterkötter M, Freitag M, Humpf HU, Güldener U, Tudzynski B (2013) Deciphering the cryptic genome: genome-wide analyses of the rice pathogen Fusarium fujikuroi reveal complex regulation of secondary metabolism and novel metabolites. PLoS Pathog 9: e1003475

Wieser J, Adams TH (1995) flbD encodes a Myb-like DNA-binding protein that coordinates initiation of Aspergillus nidulans conidiophore development. Genes Dev 9: 491-502

Wilbanks EG, Facciotti MT (2010) Evaluation of algorithm performance in ChIP-seq peak detection. PLoS One 5: e11471

Wilhelm BT, Marguerat S, Watt S, Schubert F, Wood V, Goodhead I, Penkett CJ, Rogers J, Bahler J (2008) Dynamic repertoire of a eukaryotic transcriptome surveyed at single- nucleotide resolution. Nature 453: 1239-43

Winata CL, Kondrychyn I, Kumar V, Srinivasan KG, Orlov Y, Ravishankar A, Prabhakar S, Stanton LW, Korzh V, Mathavan S (2013) Genome wide analysis reveals Zic3 interaction with distal regulatory elements of stage specific developmental genes in zebrafish. PLoS Genet 9: e1003852

Wolfers S, Kamerewerd J, Nowrousian M, Sigl C, Zadra I, Kürnsteiner H, Kück U, Bloemendal S (2015) Microarray hybridization analysis of light-dependent gene expression in Penicillium chrysogenum identifies bZIP transcription factor PcAtfA. J Basic Microbiol 55: 480-9

Woloshuk CP, Foutz KR, Brewer JF, Bhatnagar D, Cleveland TE, Payne GA (1994) Molecular characterization of aflR, a regulatory locus for aflatoxin biosynthesis. Appl Environ Microbiol 60: 2408-14

Won J, Kim TK (2006) Histone modifications and transcription factor binding on chromatin ChIP- PCR assays. Methods Mol Biol 325: 273-83

Wong KH, Hynes MJ, Todd RB, Davis MA (2007) Transcriptional control of nmrA by the bZIP transcription factor MeaB reveals a new level of nitrogen regulation in Aspergillus nidulans. Mol Microbiol 66: 534-51

Wu D, Oide S, Zhang N, Choi MY, Turgeon BG (2012) ChLae1 and ChVel1 regulate T-toxin production, virulence, oxidative stress response, and development of the maize pathogen Cochliobolus heterostrophus. PLoS Pathog 8: e1002542

Wunderlich Z, Mirny LA (2009) Different gene regulation strategies revealed by analysis of binding motifs. Trends Genet 25: 434-40

VIII. REFERENCES 75

Xie J, Du H, Guan GB, Tong YJ, Kourkoumpetis TK, Zhang LX, Bai FY, Huang GH (2012) N- acetylglucosamine induces white-to-opaque switching and mating in Candida tropicalis, poviding new Insights into adaptation and fungal sexual evolution. Eukaryot Cell 11: 773-82

Yan Z, Hull CM, Sun S, Heitman J, Xu J (2007) The mating type-specific homeodomain genes SXI1α and SXI2a coordinately control uniparental mitochondrial inheritance in Cryptococcus neoformans. Curr Genet 51: 187-95 Yang Y, Xie B, Yan J (2014) Application of next-generation sequencing technology in forensic science. Genomics Proteomics Bioinformatics 12: 190-7

Yin W, Keller NP (2011) Transcriptional regulatory elements in fungal secondary metabolism. J Microbiol 49: 329-39

Yu J, Chang PK, Ehrlich KC, Cary JW, Bhatnagar D, Cleveland TE, Payne GA, Linz JE, Woloshuk CP, Bennett JW (2004) Clustered pathway genes in aflatoxin biosynthesis. Appl Environ Microbiol 70: 1253-62

Yu JH, Butchko RA, Fernandes M, Keller NP, Leonard TJ, Adams TH (1996) Conservation of structure and function of the aflatoxin regulatory gene aflR from Aspergillus nidulans and A. flavus. Curr Genet 29: 549-55

Zaret KS, Carroll JS (2011) Pioneer transcription factors: establishing competence for gene expression. Genes Dev 25: 2227-41

Zhang Y, Liu T, Meyer CA, Eeckhoute J, Johnson DS, Bernstein BE, Nusbaum C, Myers RM, Brown M, Li W, Liu XS (2008) Model-based analysis of ChIP-Seq (MACS). Genome Biol 9: R137

Zhao Y, Ruan S, Pandey M, Stormo GD (2012) Improved models for transcription factor binding site identification using non-independent interactions. Genetics 191: 781-90

Zheng Q, Hou R, Juanyu, Zhang, Ma J, Wu Z, Wang G, Wang C, Xu JR (2013) The MAT locus genes play different roles in sexual reproduction and pathogenesis in Fusarium graminearum. PLoS One 8: e66980

Zheng WH, Zhao X, Xie QR, Huang QP, Zhang CK, Zhai HC, Xu LP, Lu GD, Shim WB, Wang ZH (2012) A conserved homeobox transcription factor Htf1 is required for phialide development and conidiogenesis in Fusarium species. PLoS One 7: e45432

Zhong M, Niu W, Lu ZJ, Sarov M, Murray JI, Janette J, Raha D, Sheaffer KL, Lam HY, Preston E, Slightham C, Hillier LW, Brock T, Agarwal A, Auerbach R, Hyman AA, Gerstein M, Mango SE, Kim SK, Waterston RH, Reinke V, Snyder M (2010) Genome- wide identification of binding sites defines distinct functions for Caenorhabditis elegans PHA-4/FOXA in development and environmental response. PLoS Genet 6: e1000848

IX. EIGENANTEIL AN PUBLIKATIONEN 76

IX. EIGENANTEIL AN PUBLIKATIONEN

Genome-wide identification of target genes of a mating-type α-domain transcription factor reveals functions beyond sexual development

Kordula Becker, Christina Beer, Michael Freitag, and Ulrich Kück (2015) Molecular Microbiology doi:10.1111/mmi.12987

Planung (P): 50 %

Experimentelle Durchführung (E): 90 %

Verfassen des Manuskripts (M): 60 %

Kordula Becker, Sandra Bloemendal, and Ulrich Kück (2015) – prepared for submission –

Planung (P): 50 %

Experimentelle Durchführung (E): 60 %

Verfassen des Manuskripts (M): 70 %

X. CURRICULUM VITAE 77

X. CURRICULUM VITAE

Kordula Becker geb. 31. Mai 1987, Essen

Witteringstr. 1 45130 Essen [email protected]

AUSBILDUNG seit 07/2011 Promotionsstudium, Ruhr-Universität Bochum „Functional genomics provide new insights into regulation of morphogenesis and secondary metabolism in the industrial penicillin producer Penicillium chrysogenum“ angefertigt am Lehrstuhl für Allgemeine und Molekulare Botanik, Christian Doppler Labor für Biotechnologie der Pilze; Betreuer: Prof. Dr. U. Kück 04/2010 - 06/2011 Vorbereitung der Fast-Track Promotion, Ruhr-Universität Bochum Lehrstuhl für Allgemeine und Molekulare Botanik, Christian Doppler Labor für Biotechnologie der Pilze; Betreuer: Prof. Dr. U. Kück 10/2006 - 09/2009 Bachelorstudium der Biologie, Ruhr-Universität Bochum „Entwicklungsbiologie bei dem Ascomyceten Sordaria macrospora: Bioinformatorische und biochemische Charakterisierung von Interaktionspartnern des Entwicklungsproteins PRO22“ angefertigt am Lehrstuhl für Allgemeine und Molekulare Botanik; Betreuer: Prof. Dr. U. Kück

06/2006 Erwerb der Allgemeinen Hochschulreife, Maria-Wächtler Gymnasium, Essen

AUSLANDSAUFENTHALTE 09/2012 - 11/2012 Forschungsaufenthalt an der Oregon State University, Corvallis, USA Etablierung und Anwendung der ChIP-seq Technologie in P. chrysogenum Department for Biochemistry and Biophysics; Betreuer: Prof. Dr. M. Freitag 10/2009 - 02/2010 ERASMUS-Studiensemester Karl-Franzens Universität Graz und Technische Universität Graz (Österreich)

STIPENDIEN seit 12/2012 Promotionsstipendium der Studienstiftung des Deutschen Volkes 10/2009 - 09/2011 Stipendium des Bildungsfonds der Ruhr-Universität Bochum X. CURRICULUM VITAE 78

AUSZEICHNUNGEN 03/2014 Novozymes Poster Award th im Rahmen des „11 International Aspergillus Satellite Meeting“, Sevilla, Spanien

09/2013 1. Posterpreis der Deutschen Gesellschaft für Genetik anlässlich der Jahrestagung der Deutschen Gesellschaft für Genetik (GfG), Braunschweig

PUBLIKATIONEN Becker K, Beer C, Freitag M, Kück U (2015) Genome-wide identification of target genes of a mating-type α-domain transcription factor reveals functions beyond sexual development. Mol Microbiol (in press; doi: 10.1111/mmi.12987)

Becker K, Bloemendal S, Kück U (2015) New insights into PcVelA regulatory functions on a genome-wide scale reveal evidence for methyltransferase PcLlmA acting as a downstream factor and direct interaction partner of PcVelA in Penicillium chrysogenum (prepared for submission)

Becker K, Böhm J, Dahlmann T, Kück U (2015) Sex und Penicillin-Biosynthese in Schimmelpilzen. Biospektrum (in press)

Marlinghaus L, Becker K, Korte M, Neumann S, Gatermann SG, Szabados F (2011) Construction and characterization of three knockout mutants of the fbl gene of Staphylococcus lugdunensis. APMIS 120: 108-116

KONGRESSBEITRÄGE

K. Becker, C. Beer, M. Freitag, U. Kück (2015) Genome-wide identification of target genes of a mating- type α-domain transcription factor reveals functions beyond sexual development. Abstracts, 28th Fungal Genetics Conference, Asilomar, CA, USA, Poster #201

K. Becker, M. Freitag, U. Kück (2014) Use of ChIP-seq technology for the functional characterization of the mating-type protein MAT1-1-1 from the industrial penicillin producer Penicillium chrysogenum. Abstracts, 12th European Conference on Fungal Genetics, ECFG12, Sevilla, Spain, Poster #250

K. Becker, M. Freitag, U. Kück (2013) Use of ChIP-seq technology for the functional characterization of two transcription factors from the industrial penicillin producer Penicillium chrysogenum. Abstracts, Annual Meeting of the German Genetics Society, Braunschweig, Poster #41

K. Becker, S. Bloemendal, U. Kück (2012) Genetic and molecular characterization of the Penicillium chrysogenum PcrsmA gene, encoding a homologue of the Aspergillus nidulans bZIP transcription factor RsmA. Abstracts, 11th European Conference on Fungal Genetics, ECFG11, Marburg, PR8.21

K. Becker, K. Kopke, B. Hoff, A. Katschorowski, S. Milbredt, J. Kamerewerd, S. Bloemendal, U. Kück (2011) The velvet-like complex in Penicillium chrysogenum participates in pathways controlling morphogenesis and penicillin production. Abstracts, Molecular Biology of Fungi, 10th VAAM- Symposium, Marburg, Poster #3 XI. ERKLÄRUNG 79

XI. ERKLÄRUNG

Hiermit erkläre ich, dass ich die Arbeit selbständig verfasst und bei keiner anderen Fakultät eingereicht und dass ich keine anderen als die angegebenen Hilfsmittel verwendet habe. Es handelt sich bei der heute von mir eingereichten Dissertation um sechs in Wort und Bild völlig übereinstimmende Exemplare.

Weiterhin erkläre ich, dass digitale Abbildungen nur die originalen Daten enthalten und in keinem Fall inhaltsverändernde Bildbearbeitung vorgenommen wurde.

Bochum, den

______(Kordula Becker)