i

MAMMALIAN TESTIS-DETERMINING FACTOR SRY HAS

EVOLVED TO THE EDGE OF AMBIGUITY

by

YEN-SHAN CHEN

Submitted in partial fulfillment of the requirements

For the degree of Doctor of Philosophy

Dissertation Advisor: Dr. Michael A. Weiss

Department of Biochemistry

CASE WESTERN RESERVE UNIVERSITY

August, 2013

CASE WESTERN RESERVE UNIVERSITY

SCHOOL OF GRADUATE STUDIES

We hereby approve the dissertation of

YEN-SHAN CHEN candidate for the Doctor of Philosophy degree.

(signer) Hung-Ying Kao, Ph.D.

(chair of the committee)

Michael A. Weiss, M.D. Ph.D.

David Samols, Ph.D.

Pieter deHaseth, Ph.D.

Shigemi Matsuyama, Ph.D.

(date) Jun 4th , 2013

i

Table of Contents

List of Tables iv

List of Figures v

Acknowledgements xi

List of Abbreviations xiii

Abstract xv

Chapter I 1

“Introduction”

Chapter II 39

“Inherited human sex reversal due to impaired nucleocytoplasmic trafficking of SRY defines a male transcriptional threshold”

Chapter III 112

“A binary fate decision in human development is rendered ambiguous by accelerated proteosomal degradation of a transcription factor”

Chapter IV 155

“A microsatellite-encoded domain in rodent Sry functions as a genetic capacitor to enable the rapid evolution of biological novelty SRY”

Chapter V 220

“Summary and discussion”

ii

Appendix I 235

“Summary of inherited mutations in human SRY and state of characterizations”

Appendix II 237

“Model: CH34 cell”

Appendix III 239

“Floppy Sox”

Appendix IV 243

“Summary of human XY intersex syndromes”

Appendix V 247

“Methods in detail”

Appendix VI 259

“Screening an appropriate cell models”

Appendix VII 264

“Improving the transient transfection protocol”

Bibliography 271

iii

List of Tables

Table 2.1 Properties of SRY variants 89

Table S2.1 Nucleocytoplasmic trafficking of SOX factors 108

Table S4.1 Construction of chimeric mSry/hSRY proteins 215

Table S4.2 Pairwise differences between HMG boxes 216

Table S4.3 Sox3 conservation 217

Table S4.4 The alignment of HMG boxes of the grass mice with selected

Muroidea rodents and non-rodent mammals 218

Table S4.5 ChIP primer sets for qPCR and ChIP (testis-specific enhancer

TESCO element) 219

iv

List of Figures

Figure 1.1 Sex determination mechanisms among selected bony

vertebrates 19

Figure 1.2 The ZW sex determination 20

Figure 1.3 The XY sex determination 21

Figure 1.4 Schematic of Y chromosome 22

Figure 1.5 Sox9 testis-specific enhancer 24

Figure 1.6 SRY-Sox9 regulatory axis and related chronological flow 25

Figure 1.7 Human SRY and mouse Sry 27

Figure 1.8 Structure of SRY, SRY-DNA complex, and role of minor wing 29

Figure 1.9 Alignment of selected rodents in Murinae subfamily 31

Figure 1.10 Phylogenetic tree of selected rodents from sister families

(Muroidea and Cricetidae) 32

Figure 1.11 Proposed human SRY (typical SRY) functional cascades 33

Figure 1.12 Cantilever substitution in hSRY and the related transcriptional

activity 34

Figure 1.13 Box-only transcriptional activity assays by Yeast-One-Hybrid

(Y1H) 36

Figure 1.14 A ‘medusa’ network architecture for the gene regulatory

v

network 37

Figure 1.15 Lorenz attractor model describes sex determination at the

edge of ambiguity. 38

Figure 2.1 Domain organization of hSRY and summary of human

genetics 76

Figure 2.2 Structure of SRY and outline of SRY-Sox9 regulatory axis 77

Figure 2.3 Transcriptional activity and nuclear localization 79

Figure 2.4 Chromatin immunoprecipitation and interactions with nuclear

import-export machinery 81

Figure 2.5 Calmodulin binding and Wnt signaling 83

Figure 2.6 Coupling between nucleocytoplasmic trafficking of SRY and

its phosphorylation 85

Figure 2.7 Potential N-terminal phosphorylation sites and NES of

primate SRY alleles 87

Figure S2.1 I90M (residues 35 in HMG box) does not perturb folding or

stability 91

Figure S2.2 FRET-based equilibrium binding and kinetics 92

Figure S2.3 Emission spectra of free and SRY-bound DNA after excitation

at 490 nm 94

vi

Figure S2.4 SRY-regulated testicular gene-regulatory network (GRN) and

transcriptional activation of Sox9 95

Figure S2.5 MG132 rescues expression of V60L and V60A to achieve

levels similar to wild-type SRY 96

Figure S2.6 Subcellular localization of wild-type SRY as analyzed by

immunostaining 97

Figure S2.7 SRY-Exportin-4 co-IP assays. Histogram provides a

quantitative summary of Western blots repeated in triplicate 98

Figure S2.8 Subcellular localization and CRM1 binding of mammalian

SRYs 99

Figure S2.9 Subcellular localization and CRM1 binding of hSRY variants

with NES modifications 101

Figure S2.10 V60L and V60A do not perturb binding of calmodulin (CaM) 103

Figure S2.11 Evidence that nucleocytoplasmic trafficking of hSRY does not

depend on phosphorylation 105

Figure S2.12 Transient transfection of hSRY in PC-3 cells activates

endogenous SOX9 106

Figure S2.13 Luciferase-based co-transfection assay of hSRY and variants 107

Figure 3.1 Domain organization of SRY and summary of clinical

vii

mutations 141

Figure 3.2 Design and SRY-dependent -galactosidase activity of

Y1H-screening system 143

Figure 3.3 Biophysical studies of the free and DNA bound F54S clinical

mutant 145

Figure 3.4 SRY-dependent transcriptional activation of Sox9 in

per-Sertoli cell model 147

Figure 3.5 Cellular turnover of SRY affects the regulation of SRY-Sox9

central axis 149

Figure 3.6 Truncated SRY mutations accelerated proteosomal

degradation, resulting in reduced Sox9 activation 151

Figure 3.7 SRY-regulated testicular gene-regulatory network (GRN) and

transcriptional activation of Sox9 153

Figure 4.1 Structures of hSRY and mSry 183

Figure 4.2 Gln-rich domain of mSry contributes to transcriptional

activation and TES occupancy 185

Figure 4.3 ChIP analysis of Sox9 testis-specific core enhancer

occupancy 187

Figure 4.4 Biochemical differences between HMG boxes of mSry and

viii

hSRY 188

Figure 4.5 Subcellular localization of mSry, hSRY, and chimeric proteins 190

Figure 4.6 Function of mSry Gln-rich domain in chimeric constructs 192

Figure 4.7 Correlation of CH34 model with sex reversal in transgenic

mice 194

Figure 4.8 Rodent Srys with CAG-encoded GRD contain attenuated 196

NES motif

Figure S4.1 The phylogenetic tree of selected rodents 198

Figure S4.2 Selected gene expression patterns in CH34 cell line 200

Figure S4.3 Schematic illustration of FRET-based DNA bending probe

and stopped-flow experimental design 202

Figure S4.4 A consensus NES motif restores CRM1-mediated nuclear

export of mSry 204

Figure S4.5 Studies of SRY phosphorylation and its effects on

gene-regulatory activity 206

Figure S4.6 Studies of SRY/Sry-directed sex reversal in XX transgenic

mice 208

Figure S4.7 Transfected hSRY/mSry up-regulates endogenous SOX9 in

human PC-3 cell line 210

ix

Figure S4.8 Transcriptional activity of hSRY/mSry variants in NT2-D1

cells 212

Figure S4.9 Predicted phosphorylation sites of SRY/Sry from selected

species 214

x

Acknowledgements

I would first like to thank my advisor, Michael Weiss, who is my mentor in personal and professional growth. He shapes the wonderful and key concepts presented in my work. He opened a new window for my spirit of independence and motivated my continual learning enthusiasm, which will be the most valuable treasure for my career.

I am grateful to thank my thesis committee for their support: Dr. Hung-Ying Kao as my chair of my committee, Dr. David Samols, Dr. Pieter deHaseth, and Dr.

Shigemi Matsuyama. This work would not have been completed without your encouragement and assistance. Within in the member of Dr. Weiss Lab, Dr. Nelson

Phillips gave me invaluable hands-on support and education, including experience and writing. He is always willing to improve my drafts and provide great personal and professional support, make me confident to reach my next stage of career. Dr.

Wan and Dr. Yang have been very kind in helping me operating computer programs.

I also want to thank Dr. Hua and Ms. Jia for their warm encouragement for my spirit and career. Last but not the least; I want to present my gratitude to my “team SRY” partner, Joe, thank you for being a wonderful friend.

Finally, I want to thank my family for their support. My parents always try their best to understand my difficulties and show me the positive side. Especially, I

xi

want to present my special gratitude to my fiancée, I-Ju Yeh, my lovely and smart partner for my life. She brought me these wonderful days and her constant love will keep me forward forever. Thank you!

xii

List of Abbreviations

CaM calmodulin

CD circular dichroism

ChIP chromatin immunoprecipitation

CTD C-terminal domain

DMEM Dulbecco's Modified Eagle's Medium

Exp4 exportin-4

FGF9 fibroblast growth factor 9

FRET fluorescence resonance energy transfer

GATA4 GATA binding protein 4

GRD glutamine rich domain

GRN gene regulatory network

HMG high mobility group

HPLC high-performance/pressure liquid chromatography

IgG immunoglobulin G

LHX9 LIM homeobox 9

LIM1 homeobox 1 LHX1

MIS/AMH Müllerian-inhibiting hormone/anti-Müllerian hormone

MYA million years ago

xiii

NES nuclear export signal

NLS nuclear localization signal

NRY non-recombining region

NTD N-terminal domain

PAR pseudoautosomal region

PTGDS prostaglandin-H2 D-isomerase

SF1 splicing factor 1

SOX Sry-related HMG box

SRY sex-determining region Y

TBP TATA-box binding protein

TDF testis determining factor

TES testis-specific enhancer of Sox9

TESCO core of testis-specific enhancer of Sox9

TSPY testis-specific protein on Y chromosome

WNT wingless-type

WT1 Wilms tumor protein 1

Y1H yeast-one-hybrid

ZFY Zinc finger Y-chromosomal protein

xiv

Mammalian Testis-Determining Factor SRY Has Evolved to the Edge of

Ambiguity

Abstract

by

YEN-SHAN CHEN

The male sex determination program in eutherian mammals is regulated by Sry, a transcription factor encoded by the Y chromosome. Specific DNA binding is mediated by the high-mobility-group (HMG) box domain. Expression of Sry in the gonadal ridge activates a Sox9-dependent gene-regulatory network leading to testis formation.

Mutations in human SRY cause gonadal dysgenesis and a female somatic phenotype.

A series of subtle variants define inherited alleles shared by an XY sterile daughter and fertile father. While their specific DNA binding and bending are unaffected, the variant proteins exhibit selective defects in nucleocytoplasmic shuttling (V60L; impairment of Exportin-4-mediated nuclear import, and I90M; impairment of

CRM1-regulated nuclear export) or protein stability (F109S and L163X; rapid degradation modulated by proteasomes). Decreased shuttling limits nuclear accumulation of phosphorylated (and hence activated) SRY, in turn reducing occupancy of DNA control sites regulating Sertoli-cell differentiation (the testis-specific Sox9 enhancer). Despite distinct patterns of biochemical and

xv

cell-biological perturbations, V60L and I90M each attenuate SRY- Sox9 activation in transient transfection assays by twofold. The twofold attenuation of SRY-Sox9 regulation was also observed in the hydrophobic core-destabilized mutation, F109S, and the C-terminal truncated SRY, L163X. Accelerated degradation in each case reduces nuclear accumulation of SRY and leads to deficient target enhancer occupancy. The threshold unmasked by these inherited mutations (twofold) is reminiscent of autosomal syndromes of transcription-factor haploinsufficiency, including XY sex reversal associated with mutations in SOX9 (campomelic dysplasia) and SF1 (hypospadias, anorchia). Our results demonstrate that sufficient nuclear accumulation of activated SRY regulated by nucleocytoplasmic shuttling or protein half-life is necessary for robust initiation of testicular development. Although shared by goat and deer Sry, such shuttling is not conserved among rodents wherein impaired nuclear export is compensated by a microsatellite-associated transcriptional activation domain. Human sex reversal due to subtle defects causing the twofold reduced

SRY-SOX9 regulation suggests that the transcriptional activity of wild-type SRY lies near the edge of developmental ambiguity.

A subset of Sry alleles in superfamily Muroidea (order Rodentia) is remarkable for insertion of DNA microsatellite repeat tracts, most commonly encoding (as in mice) a CAG-repeat-associated glutamine-rich domain (GRD). We xvi

provide evidence that this domain functions at a threshold length as a genetic capacitor to facilitate accumulation of variation elsewhere in the protein, including the

HMG box. The GRD compensates for otherwise deleterious substitutions in the box and the absence of non-box phosphorylation sites to ensure sufficient occupancy of the DNA target sites. Such compensation enables activation of a male transcriptional program despite perturbations to the HMG box. Whereas human SRY requires nucleocytoplasmic shuttling and coupled phosphorylation, mouse Sry contains a defective nuclear export signal analogous to a variant human SRY associated with inherited sex reversal. We propose that the rodent GRD has (i) fostered accumulation of cryptic intragenic variation and (ii) enabled unmasking of such variation due to

DNA replicative slippage. This model highlights genomic contingency as a source of protein novelty at the edge of developmental ambiguity and may underlie emergence of non-Sry-dependent sex determination in the radiation of Muroidea.

xvii

Chapter I

Introduction

1-1. Evolution of Y chromosome and SRY

Sex determination is a critical process in sexual reproduction. The underlying mechanism of sex determination is either environmental or genetic (Fig. 1.1) (Uller and Helantera 2011). Genetic sex determination factors are diverse and include ploidy of autosomes/sex chromosomes (Hodgkin 2005). Although it is believed that sex determination is modulated by a dichotomy of sex chromosome configurations, our work unveils surprising results which indicate a sophisticated underlying molecular mechanism, suggesting that the border between the two genders is ambiguous.

Sex-specific chromosomes in vertebrates can be identified by pairs of heteromorphic chromosomes bearing sex determining genes (Fig. 1.1) (Marin and

Baker 1998). The sex determining genes in birds and mammals serve as good models for related studies because the karyotypes of heteromorphic sex chromosomes are morphologically distinguishable (compared to those in turtles) and highly conserved (compared to those in frogs) (Graves and Shetty 2000). Birds have a

ZZ/ZW system determining sex which also functions in snakes, butterflies, and some fish (Fig. 1.2) (Ezaz et al. 2006). In this system the Z chromosome is large and 1

gene-rich but the partner chromosome, W, is small and gene-poor. Moreover, it is the ovum that determines the sex of the offspring. In mammals, chromosomal partners XY determine sex development (Vilain and McCabe 1998). Gene mapping has determined that this system is not homologous to the ZW system; however, this system has a similar heterogametic determination program for sex. In the mammalian system, heterogamety in XY determines maleness, as opposed to the homogametic male determination in the ZW system (Fig. 1.3) (Piprek 2009). The X is large (165

Mb; human), gene-rich (bearing approximately 1000 genes; human), and highly conserved whereas the Y is small (60 Mb; human), gene-poor (about 45 known genes; human), and repetitive (Charlesworth 2003). The large X and small Y share a short homologous region at each tip, termed the pseudoautosomal region (PAR) (Fig. 1.4)

(Charlesworth 2003). In normal situations, this region exhibits an autosomal-like phenomenon and is the site of X-Y chromosomal crossing over. Genes outside of the

PAR are recombination-inhibited since it is unnecessary or even harmful for females if the X chromosome gains male-specific genes from the Y chromosome

(Sachidanandam et al. 2001). The genes specific for male development have accumulated and are preserved on a male specific region in the non-recombining region of the Y chromosome (NRY) (Fig. 1.4). In humans 95% of genes on the Y chromosome are recombination-inhibited (Sachidanandam et al. 2001).

2

The inhibition of recombination prevents the Y chromosome from obtaining alleles from the X chromosome prohibiting recombination repair. As a result, many genes on the Y chromosome have been lost. A “shrinking theory” has been proposed to explain the loss of 97% of the genes on the human Y chromosome (1,393 lost of an original 1,438 genes) since it came into existence, and also predicts that, under the observed rate of gene loss, the human Y chromosome may be lost completely in the next 10 million years (a loss of 4.6 genes per million years) (Graves 2004).

Moreover, studies of rodents, which are believed to exhibit higher evolutionary rates

(Li et al. 1996), indicate that specific species in the muroidea super family, including the Transcaucasian mole vole (Elobius lutescens), the Zaisan mole vole (Elobius tancrei), and some Japanese spiny rats (Tokudaia osimensis and Tokudaia tokunoshimensis), have lost their Y chromosome entirely (Arakawa et al. 2002; Just et al. 2007). To compensate for the loss of the Y chromosome (including the crucial sex determining gene, SRY (Sex-determining region Y)), another sex-determination system has evolved in these species.

Sex determination is the primary function of the mammalian Y chromosome, and, although it is not the first candidate proposed for this function, SRY has been clearly-identified as the critical testis-determining gene. The zinc finger

Y-chromosomal protein (ZFY) was the first candidate for the testis-determining factor 3

(TDF) (Page et al. 1987), and other genes were also characterized that exhibited testis specific function or were suspected to play roles in spermatogenesis. However, except for SRY, all other candidates have been found to be either expressed generally or lack the “trigger role” for the initiation of male-sex differentiation. These results indicate that the SRY gene is the most critical factor in the male-specific development program. Moreover, the deployment of gene loci on the Y chromosome reflects the crucial role of SRY: SRY is located on the NRY (nonrecombining region of the Y chromosome) region and within the center of a cluster containing male-differentiation related genes, including TSPY (testis-specific protein) and ZFY (zinc finger protein)

(Fig. 1.4) (Sachidanandam et al. 2001). Also, the “leading role” of SRY in the Y chromosome was highlighted by the observation that the transposition of the SRY gene relocates the cluster of male-specific genes resulting in SRY redefining a new pseudoautosomal region (PAR) boundary. The SRY gene was regarded as a truncated version of its ancestor gene on the X, SOX3 (SRY-related HMG box 3), encoding a transcriptional regulator for embryonic brain development (Katoh and Miyata 1999).

Mutations in SOX3 are associated with X-linked hypopituitsarism (XH) and mental retardation, but are not associated with disorders in sex determination (Bylund et al.

2003). Recent studies indicate that the divergence between SRY and SOX3 may have occurred 166 million years ago (MYA), near the Therian/Monotreme split (Katoh and

4

Miyata 1999; Veyrunes et al. 2008). Although monotreme mammals (platypus and echidna) have XY sex-determining systems, they lack an SRY gene and use a completely different sex determination mechanism (Veyrunes et al. 2008). Thus,

SRY is unique to the therian mammals (marsupials and placentals) and serves as a switch in sex-determination.

1-2. Discovery of SRY’s role in sex development

The experimental assignment of SRY as a testis-determining factor is supported by transgenic mice studies (Sinclair et al. 1990). Koopman and colleagues found that ectopically expressed Sry initiated the testis-development pathway in XX transgenic mice (Koopman et al. 1991). Furthermore, if the Sry is not present in experimental XX individuals or impaired functionally in XY individuals, no testicular development is observed (Sinclair et al. 1990). A series of studies by Koopman and

Gubbay (Koopman et al. 1991; Gubbay et al. 1992) highlighted that Sry holds the key for initiation of male differentiation. After the initial functional characterization of

SRY using transgenic mice, further studies focused on the relationship between disorders of sex development (DSD) and SRY, especially in humans (Knower et al.

2003; Knower et al. 2011). Cases were identified with mismatches between the chromosomal sex and the congenital external phenotype or through the observation of 5

ambiguous anatomical sex. Loss of function or delayed expression of SRY leads to a failure of testicular differentiation in embryogenesis, and therefore exhibits a wide range of female-like somatic phenotypes, including total sex reversal or ovotestis.

Translocation of SRY may result in 46, XX female to male sex reversal, or an intermediate intersex phenotype (Haqq et al. 1994; Pontiggia et al. 1994; Werner et al.

1995; Murphy et al. 2001; Phillips et al. 2006). SRY functions as a transcriptional regulator through the interaction with a specific DNA sequence. Until 2008, when

Sekido and Lovell-Badge found the core testis-specific enhancer of Sox9 (TESCO) in mice (Fig. 1.5) (Sekido and Lovell-Badge 2008), it remained unclear what the direct

SRY-mediated downstream gene(s) were and the related mechanism. Before the discovery of TESCO, studies using the domestic mouse as a model showed that the expression of Sry is highly regulated in a short time window (10.5-12.5 days post coitum; d.p.c.). In mice, expression of Sry is initiated at 10.5 d.p.c., reaches a maximum expression peak at 11.5 d.p.c., and falls below the detectable limit of accumulation at 12.5 d.p.c. at which time the expression of Sox9 (Sry-related HMG box 9) is detected (Wright et al. 1995; Clarkson and Harley 2002; Kashimada and

Koopman 2010) (Fig. 1.6). Sox9, through activation by Sry, is a critical factor in male-specific development (Koopman 1999). Overexpression of Sox9 without activation of Sry led to the development of testes in 46, XX transgenic mice. Sox9

6

exhibits an auto-regulation loop and upregulates a group of genes (fibroblast grown factor 9; Fgf9, prostaglandin D2; Pgd2, and related factors) that directly modulate the differentiation of Sertoli cells (Wright et al. 1995; Koopman 1999; Clarkson and

Harley 2002) (Fig. 1.6). This laid the foundation for the central dogma that

SRY-SOX9 serves as a starter node for the complex male-specific gene regulatory network (GRN). Together, this central dogma and the direct interaction of Sry with the enhancer region of Sox9 validated a “SRY functional monitor system” using the extent of SRY-activation of endogenous Sox9 as a readout.

In contrast to the short time window of expression in the mouse embryo, human

SRY is broadly expressed in multiple tissues, including the adrenal gland and heart, and in the testis the expression of SRY is sustained through adulthood (Schafer and

Goodfellow 1996). This differential expression between human SRY and mouse Sry might be reflected in the differences in their molecular characteristics that are responsible for their functions; mouse Sry contains a specific glutamine-rich motif encoded by tri-nucleotide CAG microsatellite repeats on the C-terminus (Bowles et al.

1999) but lacks the N-terminal domain of human SRY and hence the function-mediating protein kinase A (PKA) acting sites (Desclozeaux et al. 1998) (Fig.

1.7). The detailed analysis of the DSD-related clinical SRY mutations and studies on the Sry-specific domains in the following section may offer insight into the functions

7

of various domains in SRY protein.

1-3. Molecular characteristics of SRY

SRY contains a high mobility group (HMG) box, a conserved motif for

DNA-binding and bending (Ner 1992) (Fig. 1.8A). This signature domain and its in-box C-terminal basic tail are conserved among the SOX (Sry-related HMG box) family throughout eukaryotic species (Phillips et al. 2006). SOX proteins are developmentally important and have no singular function. This family possesses the ability to function as a transcription factor that binds to the minor groove in DNA with the HMG box and regulates several different aspects of development (Kamachi et al.

2000; Weiss 2001) (for a detailed introduction of characteristics of SOX HMG boxes see section “Floppy Sox” in Appendix III). The HMG box is a domain with an

L-shaped structure comprised of a -strand at the N-terminus (consensus 1-11 in human SRY HMG box) (Fig. 1.8B), followed by three -helices (Weir et al. 1993;

Werner et al. 1995; Murphy et al. 2001). The first, second, and N-terminal portion of the third helices form the major wing containing a hydrophobic core for stabilizing the HMG box structure. Packing of the N-terminal -strand with the C-terminal portion of the third helix comprises the minor wing that may be involved in stabilizing the SRY-DNA interaction (van Houte et al. 1995; Weiss 2001). Unlike some other

8

HMG families such as HMG-1 and HMG-D that interact with DNA by an architecture-specific mechanism and actually form the ordered conformation in the absence of DNA (Read et al. 1993; Jones et al. 1994), the structure of the DNA-free

SRY HMG box is partially disordered in the N-terminal segment and exhibits an induced-fit mechanism during the SRY-DNA complex formation (van Houte et al.

1995; Weiss 2001). In such complexes, the major and minor wings bind to the DNA minor groove and DNA bending is modulated by the intercalation of the cantilever side-chain (an isoleucine at the consensus position 13 in human HMG box (68 in full-length of human SRY)), which inserts into DNA to perturb base-stacking but not pairing (King and Weiss 1993; Weiss et al. 1997). Moreover, although the function of the minor wing in the SRY HMG box still remains unclear, the formation process of the DNA-bound complex might be stabilized by the “aromatic box” motif in the minor wing, serving a clamp-like role (Fig. 1.8C) (Phillips et al. 2011). Therefore, the perturbance in the cantilever side chain causes obvious loss of DNA-interaction, and transcriptional regulation (King and Weiss 1993; Weiss et al. 1997) (this phenomenon validates the use of cantilever mutants as negative controls in this study).

As a transcriptional regulator, nuclear localization is critical in mediating SRY function. There are two nuclear localization signals (NLS) in the SRY HMG box: an

9

N-terminal bipartite NLS motif (N-NLS in human; KRP***********RRK), and a

C-terminal classical NLS domain (C-CLS in human; RPRRK) (Poulat et al. 1995).

Mutations located at any one of these positions are associated with clinical cases of

SRY-related DSD (Li et al. 2001; Sim et al. 2005; Hersmus et al. 2009), and it is notable that the indirect disturbance of N-NLS is associated with the attenuation of

SRY function as well (Phillips et al. 2011).

In the 1990s, studies reported two independent families having an isoleucine-methionine mutation at consensus position 35 in HMG box (90 in full-length human SRY) (Hawkins et al. 1992; Dork et al. 1998). This mutation was associated with pure gonadal dysgenesis (detail see Appendix IV) resulting in 46, XY sex reversal. Primary sequence analysis implies that this IleMet substitution impairs the potential nuclear export signal (NES), a short amino acid sequence with a conserved deployment of hydrophobic residues (la Cour et al. 2004) (typically, Ile or

Leu in common sequence: (I/L)xxxLxxxxL) that targets proteins for export from the nucleus to the cytoplasm by the nuclear transport machinery. Interestingly, Knower et al (Knower et al. 2011) showed that this mutation exhibited higher transcriptional activation of the Sox9 promoter-carrying reporter luciferase assay under overexpression experimental conditions. This observation appears to be at odds with the phenotype of the patients displaying pure gonadal dysgenesis. Although

10

previous studies provided a potential relationship between over-activation SOX9 and gonadal tumorigenesis, this cannot explain the I90M patients displaying uninhibited female differentiation (expression of SOX9 inhibits the Wnt/-catenin pathway and hence the female differentiation (Bernard et al. 2008)). Thus, by using a minimum dose transfection approach to maximally eliminate the effect of overexpression, we re-investigated the SRY-Sox9 regulation mediated by wild type and mutation I90M

SRY. Furthermore, we attempted to correlate the IleMet substitution to the related phenotype.

Previous biophysical studies investigating protein folding, Kd, and koff values of specific DNA binding indicated that there was no physiologically significant difference between the HMG boxes of I90M mutations and those of wild-type SRY.

These results therefore motivated us to think “out of the box”. In full-length human

SRY, there is a motif with three tandem serines (position 31-33 in full-length human

SRY sequence) in the out-box N-terminal domain (NTD) exhibiting phosphorylation modification mediated by PKA (Desclozeaux et al. 1998). Desclozeaux and colleagues (Desclozeaux et al. 1998) demonstrated that the phosphorylated SRY at these serines increased the SRY-DNA interaction approximately ten-fold. This result indicated that the out-box domain might play an important role in enhancing SRY function. This hypothesis is supported by another HMG box protein, SOX9, in

11

which the out-box modifications mediated by protein shuttling are also found to regulate transcriptional function (Sim et al. 2008). Taken together, the

PKA-mediating phosphorylation sites in the NTD of SRY might be affected if the nucleocytoplasmic trafficking lost efficiency due to the mutation in the NES. The loss of phosphorylation attenuates the extent of intergraded SRY-DNA interaction.

We have highlighted the role of the out-box NTD in SRY function; however, the

Sry protein from specific rodents including old world mice, rats, and Okinawa spiny rats contain no NTD but still exhibit normal functional Sry (Griffiths and Tiwari 1993;

Murata et al. 2010) (Fig. 1.9). Primary sequence analysis showed that those

NTD-lacking Sry proteins contain a unique glutamine-rich motif encoded by

microsatellite tri-nucleotide repeats, (CAG)n, at the C-terminal of the full-length Sry protein (Bowles et al. 1999). The repeating motif was uncharacterized although in vitro studies using a GAL4-responsive reporter and the transgenic XX mice sex reversal assays have indicated the glutamine-rich motif is critical for the transcriptional function of these rodent Srys (Dubin and Ostrer 1994).

Interestingly, even between the most closely related families (sister families muridae and cricetidae), rodents have widely divergent mechanisms of genetic sex determination (Fig. 1.10). For example, mice and rats (muridae) have Sry containing microsatellite repeats encoding a glutamine-rich tail and lack an NTD, but the

12

Transcaucasian mole vole and Zaisan mole vole (cricetidae) have lost their Sry and their entire Y-chromosome (Just et al. 2007). Moreover, two kinds of Japanese spiny rats, Tokudaia osimensis and Tokudaia tokunoshimensis, have lost their Y chromosome and Sry entirely but the Okinawa spiny rat (Tokudaia muenninki) has kept its Sry (Arakawa et al. 2002), which contains an open reading frame containing a glutamine-rich motif. The different evolutionary fates of Sry (loss of Sry or gain of a special repeating domain) in these closed families implies an evolutionary mechanism whereby specific amino-acid repeating motifs from microsatellite invasions favor conservation of Sry, and consequently, may protect against rapid changes on the Y chromosome.

1-4. Functional cascade of SRY-mediated developmental function.

In the last section, we discussed the known functional domains in or out of the box. Here, we will focus on our cascade model of SRY in cells reflected by molecular biological functions.

The SRY proteins are synthesized in the cytosol. The nuclear localization of de novo synthesized SRY is mediated by its NLS interacting with exportin-4 (here it functions as importin) (Gontan et al. 2009). Functioning as a transcriptional regulator, nuclear localized SRY binds to TESCO to activate the critical downstream

13

gene, SOX9. This process is proposed to be regulated by the cantilever side chain of

SRY and the clamp forming at the minor wing (King and Weiss 1993; Weiss 2001).

In the case of human sex development, previous studies have observed that the majority of SRY is distributed in the nucleus but it is also found in the cytosol of gonadal cells (Sim et al. 2008; Gontan et al. 2009; Malki et al. 2010). Furthermore, an NES-deficient SRY mutant is associated with pure gonadal dysgenesis (Dork et al.

1998; Knower et al. 2011). These results indicate that the shuttling of SRY between the nucleus and the cytosol is critical in humans. Although the detailed mechanism still needs further characterization, the phosphorylation hypothesis suggests that this shuttling is necessary for formation of a charged SRY.

We therefore can draw a brief life cycle of SRY in humans during development

(Fig. 1.11). The SRY is “born” in the cytosol, “goes to work” by localizing to the nucleus, “does its job” as a gene regulator in the nucleus, and then leaves for some

“rest and relaxation” or charge/recharge by export out of the nucleus. However, the murine Sry has a different story. The low efficiency of the NES means that murine

Sry is trapped in the nucleus after the NLS-mediating nuclear localization. This phenomenon was supported by the observation of Sekido and Wilhelm (Sekido et al.

2004; Wilhelm et al. 2005) who demonstrated that mSry was found exclusively in the nuclei of cells in the XY genital ridge. This exclusive phenomenon of nuclear

14

localization implies that phosphorylation in the cytosol is dispensable for mSry, reflected by the fact that murine Sry lacks an NTD, which contains phosphorylation sites. Despite mSry in our preliminary study possesses a more efficient cantilever

(methionine) than that of hSRY (isoleucine) (Fig. 1.12), the integrated transcriptional activity regulated by murine Sry-DNA binding is degenerated, which is demonstrated by our preliminary Yeast-One-Hybrid study comparing the “box-only function” between mouse and human SRY (Fig. 1.13). Moreover, murine Sry has a C-terminal glutamine-rich motif outside of the box. We hypothesize that the more efficient cantilever residue and the extra glutamine-rich motifs compensate for the effect of losing the NTD in murine Sry, ensuring its sufficient function since it is expressed in such a narrow window during embryogenesis (10.5-12.5 d.p.c.).

1-5. The characteristics of SRY in gene-regulatory network (GRN) of male sex

development.

A gene regulatory network (GRN) is a web of gene nodes (Fig. 1.14), including partial DNA segments or entire functional genes, which interact with each other by indirect programs (through their RNA or related expressed products), or with other substances in the cell (Levine and Davidson 2005). In general, some proteins serve to activate or repress other genes by binding to specific regulatory regions. These

15

transcriptional regulators are the main players in regulatory networks.

The gene-regulatory networks in unicellular organisms help the cell survive in various environments by responding to stimulation from external conditions. In the yeast cell, for example, a high sugar environment activates a gene network to make enzymes that process fermentation (Rossignol et al. 2003). In multicellular organisms, the similar but more complex networks serve as models of gene cascades that regulate sophisticated physiological activities. This control is delicate; even two cells containing identical genomes can still be regulated by this process and exhibit different gene expression patterns (Levine and Davidson 2005). A threshold level at every node decides the direction of the regulatory pathway in the GRN web

(Barabasi and Oltvai 2004; Kuo and Eliasmith 2005). These threshold levels and cell cross-talk determine cell type differentiation. Therefore, GRN signaling controls embryogenesis, the building and maintaining of a body plan, and guides the magnificent process of cell differentiation (Levine and Davidson 2005).

Metazoan development is under the control of GRNs and is ordinarily a robust system which allows persistence of certain characteristics under perturbations or variation conditions. Waddington canalization principle (Waddington 1959) suggests that developmental stability can overcome the minor genetic variation or environmental fluctuations in the genetic capacitors (for example, the heat shock

16

protein family (Rutherford and Lindquist 1998)) and the topology of GRNs (Siegal et al. 2007). An example of the developmental stability phenomenon is the Hox gene family, which is invariant even in species with varying body plans (Barmina and

Kopp 2007). However, metazoans contain divergent sex-determining protocols, including temperature, social cues, and genes (Rhen and Crews 1999; Kobayashi et al.

2009). Intersexual phenotypes are abundant in the wild, especially in the presence of hormone disruptors (Baker 2011). These phenomena motivated us to study the sex developmental GRN and the role of SRY in a critical transcriptional threshold.

For exploiting the transcriptional threshold of the SRY-mediated GRN, we used the cases of inherited human sex reversal as a model. Based on the regulation of the

SRY-SOX9 central dogma, which activates the male-specific GRN and leads to the differentiation of Sertoli cells, we measured the endogenous Sox9 activation in CH34 cells to monitor the function of the whole network. SRY mutations in inherited sex reversal cases exhibit alternative developmental outcomes: a fertile father with testicular differentiation and virilization, and a sterile daughter with gonadal dysgenesis (46, XY genotype) and nascent ovarian differentiation (Harley et al. 1992;

Jäger et al. 1992; Phillips et al. 2011). The reason behind this phenomenon was perplexing. Several groups have proposed that some dramatic attenuated SRY functions are responsible for inherited DSD (Knower et al. 2003; Knower et al. 2011).

17

However, while these hypotheses might be sufficient for explaining the sex reversal daughter, they fail to explain the father overcoming ostensibly defective SRY.

Therefore, we reinvestigated the SRY related to the inherited clinical cases. Our studies focused on clinical mutants exhibiting unrelated molecular characteristics, and located at different structural positions. It is surprising that all inherited mutants cause similar two-fold reduction in the transcriptional activity of SRY and this “factor of two” defining testicular differentiation or gonadal dysgenesis implies that the sex determination GRN program might not be a very robust system; a tiny difference

(two-fold) at the initiation could lead to widely divergent results.

Thus, the sex determination GRN walks the edge of ambiguity (Fig. 1.15) and the sensitive critical threshold of SRY-SOX9 regulation is a novel example of a violation of Waddington canalization principle. This phenomenon is not species-dependent because it is consistent with the sex reversal results of mice studies in the Y chromosome/autosome compatibility (Eicher et al. 1995; Eicher and

Washburn 2001; Washburn et al. 2001). Although further investigation is needed into the evolutionary advantage of a sex determination program undermining the robustness of gonadogenesis, mammalian multilevel selection evolution of

SRY-mediated GRN has been disposed at the edge of chaos.

18

Figure 1.1 Sex determination mechanisms among selected bony vertebrates

Boxes indicate known sex-determining mechanisms for each clade. Multi-box lineages possess multiple species with different sex determination mechanisms.

Birds and mammals possess greater stability of genetic sex determination mechanisms whereas species in other lineages, including ray-finned fish, amphibians, lizards and snakes, and turtles, have diverse mechanisms. ESD: environmental sex determination.

19

Figure 1.2 The ZW sex determination

The gene encodes a homodimer in ZZ males (blue) that stimulate a factor required for the differentiation of the testes. Whereas in ZW females, the factor of heterodimer

(from blue and red highlighted genes) may prevent the activation of that factor or stimulate directly the differentiation of ovaries.

20

Figure 1.3 The XY sex determination

In therian mammals, the females possess two homomorphic sex chromosomes (XX), and the males contain two heteromorphic sex chromosomes (XY). The Y chromosome is shorter and heterochromatic (made of heterochromatin). XY pair chromosomes separate during meiosis. The Y chromosome encodes 'maleness' so is restricted to the male lineage, which is different from the ZW system where W chromosome encodes 'femaleness' so is restricted to the female lineage. 21

Figure 1.4 Schematic of Y chromosome

The long arm (bottom arm) of the Y chromosome has a euchromatic proximal and a heterochromatic distal portion, and the upper arm of the Y chromosome is the short arm. PAR means the pseudoautosomal regions that permit pairing and recombination with the X chromosome during meiosis. Sex-determining Region Y gene encodes testis determining factor SRY; Zinc Finger Y-chromosomal protein is a protein that in humans is encoded by the ZFY gene, which encodes a zinc

22

finger-containing protein that function as a transcription factor; The Testis-Specific

Protein Y gene (TSPY) is one of the first genes identified on the human Y chromosome. It is a tandemly repeated and evolutionarily conserved gene on the mammalian Y chromosome. The recombination between X and Y chromosome is inhibited, except for the pseudoautosomal regions. The bulk of the Y chromosome which does not recombine is called the Non-recombining Region of the Y chromosome (NRY).

23

Figure 1.5 Sox9 testis-specific enhancer

The boxes in the TES region ((Sekido and Lovell-Badge 2008); Grey, with SRY binding sites (ATTGTT) and white, sites with no detectable binding).

24

Figure 1.6 SRY-Sox9 regulatory axis and related chronological flow

SRYSOX9 regulatory axis (red box) with genetic inputs (box at left) and outputs to a male-specific GRN () leading to inhibition of granulosa-cell fate (solid ), and

Müllerian regression (dashed ). Red curved arrow indicates SOX8/9-mediating feedback maintaining SOX9 expression. Bottom panel shows the expression timeline of Sry and Sox9 in the mouse sex determination process. The grey area indicates the initiation period of sex determination where bi-potential gonads arise from the genital ridges by 10.5 dpc. In XY genital ridges, Sry expression (blue) starts at 10.5 dpc, reaches highest at 11.5 dpc and then wanes by 12.5 dpc. Sox9 expression (green) is upregulated just few hours late after Sry expression, and its expression continues to be expressed postnatally supported by several 25

positive-feedback loops. The factor Sox9 induces differentiation of Sertoli cells.

Abbreviations: FGF9, fibroblast growth factor 9; GATA4, GATA binding protein 4;

LHX9, LIM homeobox 9; LIM1, homeobox protein Lhx 1; MIS (AMH), Müllerian

Inhibiting Substance (Anti-Müllerian Hormone); PTGDS, Prostaglandin D2 synthase;

SF1, steroidogenic factor 1; WT1, Wilm’s tumor 1; Wnt, wingless-type.

26

Figure 1.7 Human SRY and mouse Sry

Schematic diagram of full-length mouse Sry (395 residues) and human SRY (204 residues). Top: Human SRY (hSRY), which is shorter than mSry but contains a similar HMG box (black; 56-141). It lacks the C-terminal glutamine-rich region but contains a longer N-terminal domain (NTD; light purple; 1-55) that contains PKA phosphorylation sites (grey; 31-33). Other domains outside of the human HMG box:

PDZ-binding motif (dark purple) and the C-terminal domain (CTD; white; 142-204) including a bridge domain (orange, Br). Bottom: The domain of mouse Sry (mSry) is delineated: an N-terminal HMG box (green; 3-86) and a C-terminal glutamine-rich region encoded by CAG repeats (Gln-rich repeat; egg green; 144-367). They are connected by a bridge domain (orange, Br). Both of the HMG boxes contain a basic tail (blue; bt), which in human SRY has been shown to mediate specific SRY-DNA interactions. Blue numbers under the HMG box indicate the identity scores

27

compared to the amino acid sequence of the same domain of human SRY.

28

Figure 1.8 Structure of SRY, SRY-DNA complex, and role of minor wing

(A) Functional domains in SRY HMG box. The HMG box (blue) indicates residues

55-141, and the extreme C-terminus contains a PDZ-binding motif (grey), which is proposed to modulate SRY-SIP-1 interaction. HMG-box nuclear localization signals

N-NLS (amber) and C-NLS in the HMG box (yellow) are proposed to interact with

Exportin-4 and importin-. Red highlights the Valine at position 60. The side chain of V60 interacts with that of H120, Y124, and Y127 (green) and packs into the hydrophobic mini core in the SRY structure.

29

(B) Left: Structure of SRY HMG box domain. The domain contains a major wing

(consisting of -helix 1, -helix 2, the first two turns of -helix 3, and connecting loops) and a minor wing (the N-terminal -strand, the remainder of -helix 3, and

C-terminal tail). In a specific complex an interface forms between -helix 3 and the

N-terminal -strand to stabilize an L-shaped structure. Right: Ribbon model of SRY

HMG box (dark blue). The DNA is highlighted with red color. The core DNA target site has sequence 5’-TTGTGCA-3’ and complement.

(C) Overview of SRY minor wing in presence of DNA (DNA is hidden). The packing of V60 within an aromatic box defined by the side chains of H120, Y124, and

Y127 (red).

30

Figure 1.9 Alignment of selected rodents in Murinae subfamily.

Full-length SRY protein sequences of mouse, rat, and Okinawa spiny rat.

Spectroscopy diagram indicates the similarity of amino acid from less (blue) to greater (red).

31

Figure 1.10 Phylogenetic tree of selected rodents from sister families (Muroidea and Cricetidae)

Representative species in Muroidea superfamily. The color codes depict variations in the Sry frame: brown; follows mSry frame, such as HMG-bridge-“domain with repeating Gln-tracts encoded by CAG”, magenta; species with Sry containing poly-A repeating tracts encoded by GCA, and species with different evolutionary fates of Sry are framed in boxes. The relationships are derived by molecular information from

Huchon, D. et al. (Huchon et al. 2002) and Poux, C. et al (Poux et al. 2002). This phylogenetic tree was modified from previous review articles (Sitnikova et al. 2007;

Badenhorst et al. 2011).

32

Figure 1.11 Proposed human SRY (typical SRY) functional cascades.

33

Figure 1.12 Cantilever substitution in hSRY and the related transcriptional activity.

(A) ChIP assays probing SRY occupancy of target sites within Sox9 testis-specific enhancer core element (TESCO). Left: TESCO fragment boxes (black, with SRY binding sites and white, sites with no detectable binding). Right: Primer sets A and B were used to probe SRY occupancy. Negative controls using primer sets associated with white boxes (no detectable SRY-binding site) are not shown. 34

(B) Histogram showing qPCR results of Sox9 gene expression. Nature-occurring cantilever side chains are highlighted with black (wild type human SRY; Ile) and orange (Phe, Met, and Leu). Horizontal brackets in histogram designate statistical comparisons: (* or **) Wilcox p-values <0.05 or 0.01 whereas “ns” indicates p-values > 0.05.

(C) Gel showing results of ChIP with SRY variants.

35

Figure 1.13 Box-only transcriptional activity assays by Yeast-One-Hybrid (Y1H).

Integrated Y1H reporter containing three adjoining SRY target sites (5’-ATTGTT-3’ and complement) inserted upstream of reporter lacZ and downstream of selectable marker ura3. Expressed HMG box domain from human SRY (orange bar) and mouse Sry (green bar) exhibited two-fold difference in transcriptional activity. This data was obtained by Joe Racca as part of his collaboration on this project.

36

Figure 1.14 A ‘medusa’ network architecture for the gene regulatory network.

Individual circles represent individual gene nodes in the network (examples described inside the box). Arrows indicates the regulatory activities. The central area

(clustered with red nodes) outlines the core of network with strongly connections

(‘medusa’s head’). Genes outside the core area represent the peripheral, non-core genes (‘medusa’s arms’).

37

Figure 1.15 Lorenz attractor model describes sex determination at the edge of ambiguity.

Edward Lorenz proposed a Lorenz system model to describe a sensitive effect that exhibits dependence on initial conditions: a tiny variation at one place in a nonlinear system can result in wide differences to a later state. The Lorenz attractor is a set of solutions of the Lorenz system in chaos theory (Camargo et al. 2012). The figure set shows three panels of the time segment-3D differentiation of 2 trajectories (blue; male and yellow; female) in the Lorenz attractor starting at two very close initial points, which indicates the boundary between male and female development threshold.

Initially, the male and female trajectories seem coincident but, after some time, the divergence is obvious. This implies the widely divergent phenotypes in the inherited cases (virilization father and pure gonadal dysgenesis daughter).

38

Chapter II

Inherited human sex reversal due to impaired nucleocytoplasmic trafficking of SRY defines a male transcriptional threshold

Introduction

The male phenotype in eutherian mammals is determined by Sry (Sinclair et al.

1990), a Y-chromosomal gene encoding an architectural transcription factor (TF)

(Sekido 2010). SRY contains a central high-mobility-group (HMG) box, a conserved motif of specific DNA binding and bending (Fig. 2.1A) (Ner 1992; King and Weiss

1993). Assignment of Sry as the testis-determining factor is supported by transgenic mouse models (Koopman et al. 1991) and human mutations leading to gonadal dysgenesis with female somatic phenotype (Swyer’s Syndrome; detail see Appendix

IV) (Knower et al. 2003). Expressed in the pre-Sertoli cells of the differentiating gonadal ridge, Sry activates Sox9, an autosomal gene broadly conserved among vertebrate sex-determining pathways (Sekido and Lovell-Badge 2008). Direct binding of Sry to specific DNA sites within the testis-specific core enhancer of Sox9 (TESCO) activates a male-specific gene-regulatory network (GRN) in the fetal gonadal ridge

39

(Sekido and Lovell-Badge 2008). Sustained Sox9expression orchestrates programs of cell-cell communication, migration and differentiation leading to gonadogenesis

(Lovell-Badge et al. 2002), regression of the female anlagen (following Sertoli-cell secretion of anti-Müllerian Hormone/Müllerian Inhibiting Substance; AMH/MIS)

(Josso et al. 2006), and somatic virilization (through fetal Leydig-cell secretion of testosterone) (McClelland et al. 2012). Mutations in this pathway are associated with disorders of sexual development (DSD) (Wilhelm and Koopman 2006).

Structure-function relationships in human SRY (hSRY) have been investigated through comparative biochemical and biophysical studies of Swyer variants (Knower et al. 2003). Most often arising de novo in spermatogenesis and clustering in the HMG box (Fig. 2.1A), such mutations typically impair specific DNA binding by direct or indirect perturbation of an angular protein-DNA interface (Knower et al. 2003). The present study has focused on inherited mutations (V60L and I90M; asterisks in Fig.

2.1A) that by contrast allow near-native DNA binding and bending (Knower et al.

2011; Phillips et al. 2011). Such mutations are compatible with alternative developmental outcomes (family trees in Fig. 2.1B): testicular differentiation leading to virilization (fertile 46, XY father) or nascent ovarian differentiation leading to gonadal dysgenesis (sterile 46, XY daughter; (Knower et al. 2011). In the structure of the wild-type SRY-DNA complex V60 and I90 pack within the minor and major

40

wings (Fig. 2.2A) of the HMG box (consensus positions 5 and 35, respectively; Fig.

2.2B). The clinical substitutions each represent a subtle interchange of non-polar side chains. Known inherited mutations in SRY are summarized in Appendix I.

Inherited mutations in hSRY provide experiments of nature probing the threshold molecular properties of a developmental switch beyond DNA binding and bending. To this end, our studies exploited a model of the central SRY-Sox9 regulatory axis (Fig.

2.2C) in a rodent fetal pre-Sertoli cell line (Haqq et al. 1994; Haqq and Donahoe 1998;

Phillips et al. 2011). Our findings demonstrate that hSRY undergoes nucleocytoplasmic shuttling (NCS) and that such NCS is coupled to an activating

N-terminal serine phosphorylation (Desclozeaux et al. 1998; Malki et al. 2005).

Respective V60L and I90M variants exhibit selective impairment of nuclear import

(as mediated by Exportin-4) or nuclear export (mediated by CRM1) but otherwise retain substantial gene-regulatory activity in accordance with the phenotypes of the fathers. Mutational impairment of NCS in either direction reduces nuclear accumulation of phosphorylated hSRY, in turn attenuating SRY-dependent transcriptional activation of Sox9 in accordance with the phenotypes of the XY daughters. Although NCS-dependent transcriptional regulation is likely to be broadly conserved among Sox genes (Table S2.1; (Sim et al. 2008; Malki et al. 2010), its importance to the function of SRY has been obscure in studies of mouse models

41

(Poulat et al. 1995). In a companion study (Chen et al. 2013b) we demonstrate that murine Sry fails to undergo NCS as a consequence of an inactive NES motif. Whereas the HMG box of hSRY contains an active NES (IxxxLxxxxxML), microsatellite-associated rodent variants contain the variant sequence IxxxLxxxxxSL

(Chen et al. 2013b).Such variation is associated with clade-specific insertion of a

DNA microsatellite in the divergent evolution of the Y chromosome in superfamily

Muroidea (order Rodentia; (Miller et al. 1995; Bowles et al. 1999).

The contribution of NCS-linked phosphorylation of hSRY to its transcriptional activity was estimated at progressively lower levels of protein expression. In each case the limiting extent of attenuation was twofold, a quantitative threshold reminiscent of DSD-associated syndromes of autosomal transcription-factor (TF) haploinsufficiency (Wilhelm and Koopman 2006). An analogous threshold in XY mice has been inferred based on studies of intersexual phenotypes due to the strain-specific phenomenon of Y chromosome/autosome incompatibility (Albrecht et al. 2003; Bullejos and Koopman 2005; Polanco and Koopman 2007). This seeming paradox of mice and men highlights the tenuous beginning of the male program

(Wilhelm and Koopman 2006). Defining a critical boundary between testicular self-organization and gonadal dysgenesis, inherited human sex reversal associated with subtle perturbations of the NCS of hSRY thus violates the principle of

42

Waddington canalization (Waddington 1959). Whereas the latter emphasizes the importance of robustness in developmental pathways (Masel and Siegal 2009), we speculate that SRY has evolved to the edge of ambiguity as a developmental mechanism to ensure male phenotypic variation in social mammals.

43

Results

Our studies employed rat embryonic pre-Sertoli XY cell line CH34 (Haqq and

Donahoe 1998). We previously described a transient-transfection assay of hSRY-directed transcriptional activation of endogenous Sox9 (Phillips et al. 2011).

Employing quantitative PCR (real-time-Q-rtPCR; qPCR), this assay measures the time-dependent accumulation of mRNAs in a downstream GRN (Koopman et al.

2001). Sox9 expression was measured following transient transfection of wild-type or mutant SRY constructs. Key findings were replicated in two human male cell lines.

Inherited DSD mutations in hSRY (V60L and I90M; first and second red asterisks in Fig. 2.1A, respectively) were chosen based on their compatibility with native-like structure and DNA binding (Table 2.1; (Phillips et al. 2011). Structural environments are shown in Figure 2.2 (Murphy et al. 2001). Conserved as Val among mammalian SRY/Sry domains, V60 (consensus position 5 in the minor wing) adjoins a bipartite nuclear localization signal (NLS; (Phillips et al. 2011). A second mutation at this site (V60A; father uncharacterized) was associated with ovotestes (Hiort et al.

1995). V60L and V60A do not affect box stability and permit DNA-dependent folding of the minor wing with near-native DNA bending (Phillips et al. 2011). I90 projects from the second -helix (2) into the major-wing core as part of an aliphatic motif

44

(I90, L94, M100, and L101; consensus positions 35, 39, 45, and 46) proposed to function as a Sox nuclear export signal (NES;(Sim et al. 2008; Malki et al. 2010).

o Although I90M destabilizes the free box (ΔTm 4 C and ΔΔGu 0.5 kcal/mole; Table

2.1 and Fig. S2.1), near-native specific DNA affinity (Kd) and DNA bending are nonetheless maintained (Table 2.1, Fig. S2.2, and S2.3). Like V60L and V60A box-DNA complexes (Phillips et al. 2011), the I90M complex exhibits a reduced kinetic lifetime with compensating changes in on- rate constants similar to wild-type.

Transcriptional Regulation in a Pre-Sertoli Cell Model. CH34 cells were employed to probe hSRY-dependent activation of Sox9; transfection efficiency was 32.6(±1.2) percent as inferred from control co-transfection of a plasmid encoding green fluorescent protein (GFP). Transient transfection of wild-type hSRY under standard conditions (1g/well) activated expression of Sox9 by eightfold relative to an empty vector (black bar in Fig. 2.3A). Extent of transcriptional activation decreased to fivefold on successive dilution of the hSRY plasmid by the empty vector (maintaining total transfected DNA constant) to a final dilution of 0.02 g hSRY plasmid and

0.98g empty vector per well (50-fold dilution; white bar in Fig. 2.3A). Such dilution provided a control for over-expression artifacts.

Although Sox9 is the principal target of Sry (Sekido and Lovell-Badge 2008),

45

additional qPCR assays were undertaken to characterize a downstream GRN in relation to in situ transcriptional profiling of the murine XY gonadal ridge (Bullejos and Koopman 2005). Whereas transient expression of hSRY did not affect mRNA accumulation of non-sex-related Sox genes and housekeeping genes (boxes at bottom in Fig. 2.2C), specific up-regulation of Sox8, Sox9,fibroblast growth factor 9 (Fgf9) and prostaglandin D2 synthetase (Ptgd2)were observed in accord with their known roles in testicular differentiation (Moniot et al. 2009). No such changes in mRNA accumulation were observed on transient transfection of an empty plasmid or a control plasmid expressing a stable hSRY variant devoid of specific DNA-binding activity (I68A; (Weiss et al. 1997)). Further, following 24-h incubation in serum-rich medium, no significant changes were observed in the expression of rat Sry itself.

Activities and Stabilities of hSRY Variants.Comparative studies of variant hSRY constructs (V60L, V60A, and I90M) demonstrated that (i) the two minor-wing substitutions exhibited decreased transcriptional activity at each dilution tested whereas (ii) major-wing substitution I90M enhanced Sox9 expression when over-expressed but exhibited progressive loss of function with serial dilution (at right in Fig. 2.3A). Remarkably, at highest dilution (50-fold; white bars in Fig. 2.3A) the three substitutions were each associated with twofold loss of Sox9 activation relative

46

wild-type at the same dilution. These trends were also observed in the downstream

GRN modulated by the SRY-Sox9axis (Fig. S2.4). The anomalous dilution-related properties of I90M hSRY motivated assessment of proteolytic stability and cellular

NCS as factors that could lead to over-expression artifacts.

Cellular turnover of hemagglutinin (HA)-tagged hSRY constructs (transfected without dilution) was evaluated following cycloheximide inhibition of translation (Fig.

2.3B). Comparison of anti-HA Western-blot intensities demonstrated that V60L and

V60A variants are more susceptible to degradation than are wild-type or I90M hSRY

(graph in Fig. 2.3B). Such differential degradation could be circumvented through addition of proteosome-inhibitor MG132 (Fig. S2.5). Subcellular localization was investigated using immunofluorescence microscopy (Fig. 2.3C); control GFP co-transfection studies demonstrated that wild-type and variant constructs achieved similar transfection efficiencies; 900 cells were counted in each case (triplicate by blinded coworkers). Representative images are shown in Figure 2.3C (lower panel) in relation to corresponding nuclear staining of the same cells with

4',6-diamidino-2-phenylindole (DAPI; upper panel). The variant proteins exhibited contrasting perturbations relative to wild-type (Fig. 2.3D). Whereas V60L and V60A often exhibited pancellular distributions with significant reduction in exclusive nuclear staining (gray bars in Fig. 2.3D), I90M hSRY exhibited a reduced pancellular

47

fraction (white bars) with increase in exclusive nuclear staining. Partial restoration of nuclear localization of V60L and V60A hSRY was achieved by N-terminal fusion of an exogenous nuclear localization signal (NLS) derived from SV40 large T Antigen

(Kalderon et al. 1984) (“+NLS” at right in Fig. 2.3D).No significant changes in nuclear accumulation of wild-type hSRY were observed in control studies of

NLS-hSRY (Fig. S2.6); I90M NLS-hSRY was not tested.

Immunofluorescence studies were repeated on addition of MG132 for 24-h post-transfection (Fig. 2.3E). As expected, proteosomal inhibition did not affect nuclear localization of wild-type or I90M hSRY (Fig. 2.3D and E). In the absence of the fused NLS, MG132 treatment led to small decreases in extent of residual

GFP-positive cells lacking detectable hSRY expression (100-sum of gray and white bars in Fig. 2.3D and E); this trend did not achieve statistical significance. Strikingly, the combination of MG132 treatment and SV40 NLS fusion led to near-complete rescue of nuclear localization of V60L and V60A hSRY (right-hand bars in Fig. 2.3E; for representative cellular images see Fig. 2.3C).The efficiency of NLS/MG132

“double rescue” motivated re-investigation of the functional properties of V60L and

V60A hSRY under conditions wherein wild-type and variant proteins were expressed at similar levels and with similar patterns of subcellular localization (Fig. 2.3F).

Whereas NLS fusion and/or addition of MG132 had no significant effects on

48

transcriptional activation of Sox9 by wild-type or I90M hSRY at any dilution, defective activation of Sox9 by V60L or V60A hSRY was partially overcome by either maneuver, and double rescue led to native-like Sox9 expression (Fig. 2.3F, right).

Chromatin immunoprecipitation (ChIP) studies were undertaken of SRY/Sry binding sites in the TESCO element of Sox9 (Fig. 2.4A with primer sets defined in Fig.

2.4B; (Sekido and Lovell-Badge 2008). Occupancy (as probed by ChIP band intensities relative to input controls following transfection without plasmid dilution) of I90M hSRY was indistinguishable from wild-type (lanes 5 and 6 in Fig. 2.4C; see also bottom histogram). Control studies of inactive I68A hSRY (Weiss et al. 1997) demonstrated an absence of enhancer binding; further control studies employed de novo clinical mutants R62G and R75N in the N-terminal bipartite NLS of the HMG box (Gontan et al. 2009). R62G and R75N (which impair both NLS1 specific DNA binding) demonstrated weak ChIP band intensities (Fig. 2.4C; lanes 1 and 2 in gel panel and the bottom histogram). Whereas TESCO-specific ChIP band intensities for

V60L and V60A hSRY in the absence of NLS/MG132 were reduced to ca. half of the wild-type level (Fig. 2.4C), double rescue restored near-native enhancer occupancy

(Fig. 2.4D). As in the Sox9 qPCR assay, double-rescue of V60A hSRY was more complete than that of V60L hSRY (set c in Fig. 2.4D). These findings imply that the minor-wing variants, once bound to the Sox9 enhancer, retain native-like

49

gene-regulatory properties in a cellular milieu in accordance with their native-like biophysical properties (Phillips et al. 2011).

Analysis of hSRY Protein-Protein Interactions.The above findings motivated investigation of how the mutations affect binding of hSRY to the NCS machinery.

V60L and V60A adjoin the N-terminal bipartite basic NLS of the HMG box (Poulat et al. 1995); this motif (residues 61-77 of hSRY; KRPMNAFIVWSRDQRRK; basic residues in bold) binds to Exportin-4 (Exp4) to mediate nuclear import (Gontan et al.

2009) rather than importin- or (Sim et al. 2005). Co-immunoprecipitation (co-IP) studies were thus undertaken based on co-transfection of HA-tagged hSRY and

FLAG-tagged Exp4 (Fig. 2.4E). The studies revealed that whereas I90M mutation does not affect binding to Exp4 (lanes 1 and 6 in Fig. 2.4F), a graded series of perturbations were observed among the other variants (lanes 2-5). Measurement of relative band intensities (in triplicate; Fig. S2.7) defined the order V60L (mildest impairment; 60(±13)% relative to wild-type hSRY) and V60A (41(±13)%) relative to control mutations R75N (23(±6)%) and R62G (most severe; 17(±7)%) as previously characterized (Gontan et al. 2009).

Nuclear export of SOX proteins is mediated by CRM1 via a conserved NES in the HMG box (Sim et al. 2008). Binding of hSRY (NES consensus IxxxLxxxxxML;

50

residues 35-46 in the consensus HMG box) to endogenous CRM1 was demonstrated in CH34 cells by co-IP (lane 7 in Fig. 2.4G). Analogous CRM1-dependent NCS was observed in studies of goat Sry and deer Sry, whose HMG boxes bear active NES variant IxxxLxxxxxRL(Fig. S2.8). Binding of hSRY to CRM1 was unaffected by

V60L or V60A (lanes 9 and 10 in Fig. 2.4G), but markedly impaired by I90M (lane 8) in accordance with its enhanced nuclear accumulation. Binding of hSRY to CRM1 and resulting NCS were also impaired by mSry-related substitution IxxxLxxxxxSL and by multiple Ala substitutions (AxxxAxxxxxAL)(Fig. S2.9).Mutation-specific perturbations of hSRY-Exp4 or hSRY-CRM1 interactions stand in contrast to the absence of perturbations in assays of hSRY-CaM binding. Although such binding

(mediated by the N-terminal segment of the hSRY HMG box) has been proposed to direct nuclear entry (Sim et al. 2005), co-IP studies of endogenous CaM (outlined in

Fig. 2.5A) revealed similar levels of binding to diverse mutations in the HMG box, including V60L and V60A (Fig. 2.5B and C). Biochemical and biophysical studies of

V60L and V60A hSRY-CaM complexes likewise demonstrated similar affinities and structural features (Fig. S2.10).

Protein-protein interactions involving SRY have also been implicated in male-specific inhibition of Wnt/-catenin signaling (Fig. 2.5D; (Bernard et al. 2008;

Tamashiro et al. 2008). Such signaling is operative in ovarian development, and its

51

inappropriate activation in the XY gonadal ridge (via stabilization of -catenin) can be associated with human sex reversal (Maatouk et al. 2008). Although in pre-Sertoli cells this pathway is not well characterized, a functional SRY assay has been described in non-gonadal and germ-cell lines (Bernard et al. 2008; Tamashiro et al.

2008). This assay was adapted to CH34 cells stably transfected to express activated

-catenin variant S37A and transfected by a chimeric SRY construct (hSRY residues

1-155 and mouse Sry residues 128-396 as above); a functional read-out was provided by luciferase under the control of a -catenin-responsive promoter (TOPflash; Fig.

2.5E). Transient transfections were performed with 50-fold dilution of SRY plasmids to minimize over-expression and in the presence of MG132 to equalize protein levels.

Comparison of the CH34 cells expressing S37A -catenin with the parent cell line revealed a 40-fold increase in luciferase activity (lanes 1 and 2 in Fig. 2.5F). This activity was inhibited by 3.5-fold on transient transfection of an SRY chimera bearing the wild-type human HMG box (“wt” in Fig. 2.5F, lane 3). Inhibition required specific

DNA binding as demonstrated by maintenance of luciferase activity on transient transfection of SRY chimeras bearing inactive control mutations I68A or G95R (lanes

7 and 8 in Fig. 2.5F). Inherited mutations V60L and I90M as well as ovotestis-associated mutation V60A gave rise to intermediate levels of luciferase activity (lanes 4-6 in Fig. 2.5F). Whereas NLS fusion did not affect the inhibitory

52

activity of the I90M construct (lanes 6 and 11 in Fig. 2.5F), partial rescue of nuclear localization enabled the V60L and V60A constructs to achieve near-native inhibition of luciferase activity (lanes 9 and 10).

Nuclear Trafficking Affects hSRY Phosphorylation.SRY in primates contains potential phosphorylation sites N-terminal to the HMG box (hSRY residues 26-38;

PALRRSSSFLCTE) recognized by protein kinase A (PKA); phosphorylation augments specific DNA binding (Desclozeaux et al. 1998). To investigate whether altered NCS affects phosphorylation of hSRY (HA tagged) and hence enhancer binding, we evaluated by co-IP the extent of phosphorylation in cytosolic and nuclear fractions; an anti-phosphoserine antiserum was used to pull down phosphoproteins for

Western blot by anti-HA antiserum (Fig. 2.6A). Molecular markers for fidelity of fractionation were provided by glyceraldehyde-3-phosphate dehydrogenase (GAPDH; cytosol) and Ying-Yang 1 (YY1; nucleus) as shown in Figure 2.6B (bottom).

Epitope-tagged hSRY variants containing PKA-site substitutions RRAAAFL

(“phospho-dead”) or RRDDDFL (“phospho-mimic”) were not detected as further described in our companion study (Chen et al. 2013b); these findings indicate that in

CH34 cells this is the only site of serine phosphorylation (green box in Fig. 2.6C).

Studies were conducted following 50-fold dilution of the hSRY plasmid and in the

53

presence of MG132. Because of the low uniform levels of hSRY expression under these conditions, each assay required pooling extracts from fifteen 10-cm plates grown to confluence.

As expected, wild-type hSRY was detected predominantly but not exclusively in the nuclear fraction (lanes 1 and 5 in Fig. 2.6B). Extent of phosphorylation was similar in the two fractions as indicated by the ratio of band intensities in the top panel

(phospho-hSRY) to band intensities beneath (total HA-tagged hSRY). Also in accordance with the results of immunofluorescence microscopy (above), I90M hSRY exhibited increased nuclear accumulation (lanes 2 and 6 in Fig. 2.6B). Strikingly, however, the total phosphorylation and relative phosphorylation of I90M hSRY were decreased in the nucleus despite its enhanced accumulation in that fraction. The twofold reduction in Sox9 transcriptional activation observed under these conditions

(50-fold plasmid dilution) presumably reflects a defect in phosphorylation, which more than offsets any increase in transcriptional-regulatory activity that would otherwise be associated with its enhanced nuclear accumulation. V60L and V60A hSRY were predominantly detected in the cytosolic fraction in accordance with impairment of nuclear entry (Fig. 2.6B, comparison of lanes 3 and 4 with lanes 7 and

8, respectively). Whereas the relative extent of phosphorylation in the cytosolic fraction is similar to that of wild-type hSRY, a striking reduction was observed in each

54

case in the absolute amount of phosphorylated protein in the nuclear fraction (top panel of lanes 7 and 8 in Fig. 2.6B).

To investigate the relationship between N-terminal phosphorylation of hSRY and transcriptional activation of Sox9, phospho-dead and phospho-mimic variants (Fig.

2.6C) were employed in qPCR assays (Fig. 2.6D). The assays were conducted as above with 50-fold plasmid dilution and in the presence of MG132. The non-box Ala or Asp substitutions did not affect nuclear localization (Fig. S2.11) in accordance with past studies (Desclozeaux et al. 1998). Wild-type and variant PKA sites were introduced into three epitope-tagged hSRY constructs, bearing either a wild-type

HMG box, the I90M HMG box, or a variant HMG box in which three putative NES residues (consensus positions 35, 39, and 45) were each substituted by Ala. The latter construct (designated NES in Fig. 2.6D) thus yielded an hSRY variant containing six Ala substitutions, three in the non-box PKA site (S31A, S32A, and S33A) and three in the major wing (I90A, L94A, and M100A). In the presence of the native PKA motif, I90M caused the expected twofold reduction in Sox9mRNA accumulation

(relative to wild-type hSRY) as did the NES construct (left-hand set of bars in Fig.

2.6D). Strikingly, mimicry of PKA phosphorylation site by tandem Asp substitutions in each case led to enhanced and equal Sox9 expression (right-hand set of bars in Fig.

2.6D). This finding provides evidence that (i) serine phosphorylation enhances the

55

activity of hSRY in an appropriate cellular milieu and (ii) defective phosphorylation of I90M hSRY underlies its partial loss of activity. Conversely, in the phospho-dead context, wild-type hSRY exhibited the same twofold loss of function as was conferred by I90M or the NES Ala substitutions (middle group of bars in Fig. 2.6D). These findings suggest that unphosphorylated hSRY retains partial activity whether the defect in phosphorylation is due to mutation of the PKA site or to mutations in the

HMG box that impair nuclear exit.

Extension to Human Cell Lines. To confirm key findings in a human cellular milieu, transient transfection studies were conducted in male cell lines PC-3 and NT2-D1, respectively derived from prostate- and testicular cancer (Kaighn et al. 1979; Knower et al. 2007). Transient transfection of wt or variant hSRY in PC-3 cells yields a pattern of relative SOX9-regulatory activities that mirrors that observed above in CH34-based studies of rat Sox9 transcriptional activation (Fig. S2.12). Although transcriptional of the endogenous SOX9 gene in NT2-D1 cells is not robustly activated on transient transfection of hSRY (only 2.5 fold), co-transfection of wt or variant hSRY with a luciferase reporter under the transcriptional control of SRY- and SF1 DNA target sites likewise yields a similar pattern of relative luciferase activities (Fig. S2.13).

56

Discussion

Clinical identification of variant human SRY alleles associated with 46, XY pure gonadal dysgenesis (Swyer’s Syndrome) has provided an opportunity to investigate a sex-specific GRN and its function in gonadogenesis. The present study has focused on subtle variants inherited by sterile XY daughters from fertile fathers. We have uncovered perturbed NCS due to impaired binding of hSRY to Exp4 (V60L and V60A) or CRM1 (I90M). Impaired nuclear import or export is in each case associated with twofold reduction in male-pattern Sox9 transcription and twofold enhancement of female-pattern Wnt/β-catenin signaling. The inherited variants of SRY are thus poised at a critical threshold of activity (Nykter et al. 2008), highlighting the narrow margin of “decision making” in the differentiating gonadal ridge. The twofold perturbations uncovered in this study are functionally analogous to a syndrome of autosomal TF haploinsufficiency (Seidman and Seidman 2002).

V60L and V60A SRY retain near-native biochemical properties (Table 2.1).

Structural accommodation of the variant side chains reflects the flexibility of the minor wing of the HMG box. To probe their gene-regulatory functions, we investigated the intrinsic gene-regulatory activity of these variants following

“correction” of their altered trafficking (by NLS fusion) and accelerated degradation

57

(by proteosome inhibition). Under these conditions occupancy of the testis-specific

Sox9 enhancer by wild-type and variant proteins was similar and led to near-complete

“double rescue” of Sox9 mRNA accumulation; native-like inhibition of Wnt/β-catenin signaling was likewise restored. I90M impairs binding of SRY to CRM1, a mediator of nuclear export (Fornerod et al. 1997), leading to enhanced nuclear accumulation.

Although transcriptional activation of Sox9 is increased under conditions of over-expression (in accordance with the results of Knower, K.C., et al. (Knower et al.

2011)), successive plasmid dilution unmasked a twofold defect in Sox9 regulation.

Nucleocytoplasmic Shuttling of SRY. On plasmid dilution impaired nuclear export of I90M hSRY attenuates its gene-regulatory activity, suggesting that wild-type hSRY undergoes an NCS-coupled regulatory post-translational modification. Among primates (Fig. 2.7A) SRY alleles are highly conserved and share potential N-terminal phosphorylation sites (Fig. 2.7B). Their HMG boxes in each case contain consensus

NES sequences (IxxxLxxxxxML; Fig. 2.7C). The divergent boxes of rat and mouse

Sry (at bottom in Fig. 2.7C) by contrast contain an unfavorable NES substitution

(MS), shown in our companion study to impair nuclear export in association with gain of a microsatellite-associated transcriptional activation domain (Chen et al.

2013b). Whereas I90M (HMG position 35) occurs within -helix 2 and perturbs

58

packing of the major wing, the rodent substitution occurs in a loop following this

-helix and so may be better tolerated. Competence for CRM1-mediated nuclear export is maintained in goat and deer Sry, which bear NES variant IxxxLxxxxxRL.

It would be of future interest to investigate these and other mammalian SRYs in relation to NCS-coupled phosphorylation.

Like SOX9, hSRY contains a PKA site N-terminal to the HMG box

(Desclozeaux et al. 1998; Sim et al. 2008). In each case phosphorylation enhances specific DNA-binding affinity (Sim et al. 2008). We therefore hypothesized that

I90M-associated impaired nuclear export perturbs this phosphorylation, which in turn attenuates Sox9 activation—despite increased total nuclear hSRY abundance. We therefore evaluated hSRY phosphorylation following biochemical fractionation, yielding independent estimates of phospho-hSRY in the nucleus and cytoplasm.

Comparative studies of “phospho-dead” (RRAAAFL) and “phospho-mimic”

(RRDDDFL) constructs verified that negative charges enhance transcriptional potency.

In either context the extent of wild-type or I90M hSRY-dependent Sox9 expression is indistinguishable at a reduced (phospho-dead) or elevated (phospho-mimic) levels.

Because divergent rodent Sry proteins lack this N-terminal PKA site, we speculate that its C-terminal glutamine-rich expansion provides an alternative mechanism to enhance Sox9 transcription as explored in our companion study (Chen et al. 2013b).

59

Coupling between NCS and post-translational modification occurs in diverse systems (Sim et al. 2008; Malki et al. 2010) and may be a hallmark of mammalian

Sox proteins (Smith and Koopman 2004; Malki et al. 2010). A paradigm is provided by SOX9, which contains two PKA sites: S64 (N-terminal to the HMG box) and S181

(adjoining NLS2in the C-terminal tail of the HMG box). The latter phosphorylation enhances NLS2 binding to importin- and hence nuclear import (Sim et al. 2008).

Autocrine regulation of PKA activity by prostaglandin D2 in the differentiating pre-Sertoli cell is proposed to provide a mechanism to amplify and maintain Sox9 expression (Moniot et al. 2009). Evidence for the developmental importance of such mechanisms was obtained in engineered mice (Moniot et al. 2009). Although phosphorylation-regulated nuclear localization is not a feature of hSRY, lysine acetylation (K136 in NLS2) provides an alternative mechanism to enhance its binding to importin-β (Thevenet et al. 2004).

Mouse Sry Exhibits an Analogous Threshold. The inherited Swyer syndrome resembles Y-chromosome incompatibility among mouse strains wherein Y chromosomes bearing Sry alleles derived from diverse mouse strains can cause abnormal gonadal development in B6 strain C57BL/6J (Albrecht et al. 2003; Bullejos and Koopman 2005). Aberrant interaction between such alleles and B6-derived

60

autosomal genes leads to strain-dependent intersexual phenotypes (Nagamine et al.

1987). Such phenotypes could not be correlated with changes in Sry sequence

(Bullejos and Koopman 2005) but instead depend on the extent and timing of Sry expression (Albrecht et al. 2003; Bullejos and Koopman 2005; Wilhelm et al. 2009).

As in the present studies, changes of twofold or less in Sry expression in the differentiating gonadal ridge were associated with developmental abnormalities. Such genetic-background dependence strongly suggests that the rodent Sry-Sox9-directed transcriptional program lies close to a threshold of function.

The similar twofold biochemical thresholds of SRY function in mice and humans—mammals with otherwise divergent SRY/Sry genes (Bullejos and Koopman

2005)—demonstrates that lack of robustness in nascent Sertoli-cell specification has been independently maintained in lineages that separated 80 million years ago. Such conservation seems to violate Waddington’s Principle: that fundamental developmental pathways are canalized, at least in their upstream steps, and so robust to genotypic variation and environmental fluctuations (Waddington 1959; Masel and

Siegal 2009). It is possible that the apparent fragility of hSRY is illusory. The transcriptional set point of hSRY, for example, may be so tightly controlled by upstream factors (such as GATA4 and WT1; (Miyamoto et al. 2008) and transcriptional co-regulators (such as SF1 in sex-specific transcriptional pre-initiation

61

complexes (Sekido and Lovell-Badge 2008) that twofold reductions in hSRY-directed

SOX9 activation would be unlikely. Similarly, it is possible that variation in hSRY activity is buffered by feedback and feed-forward regulatory circuits in the

SOX9-directed GRN (such as mediated by prostaglandin D2 and FGF9; (Moniot et al.

2009).

Multi-level Selection in Mammalian Evolution? The similar transcriptional thresholds of murine and human SRY (despite their biochemical divergence; see companion study (Chen et al. 2013b)) suggests that its thin thread of function

(Polanco and Koopman 2007) is the product of selection.This poses an intriguing problem given the general robustness of developmental processes (Siegal and

Bergman 2002).It is possible that an hSRY of higher transcriptional potency could impair individual fitness, such as through induction of gonadal neoplasias (Knower et al. 2011). Alternatively, higher potency could lead to intra-genomic conflict with female-specific genes and so impair the fitness of daughters (Gibson et al.

2002).Genes that contribute to variation in male-specific straits, including hormone-dependent behaviors and social competencies†(Vom Saal 1981; Vom Saal et al. 1999), may also be subject to intra-sexual selection (such as in male dominance hierarchies) or inter-sexual selection (female mate choice) (Civetta and Singh 1998).

62

We speculate that genetic variation in fetal testosterone production may have influenced the evolution of eutherian mammals, especially species (like humans and mice) that evolved within social groups. Given epidemiological linkages between human fetal testosterone exposure (as measured in mid-trimester amniotic fluid) and behavioral styles in childhood (Baron-Cohen et al. 2004), it is possible that genetic or stochastic variation in fetal Leydig cell function could ultimately affect mate choice, reproductive success or social integration within the framework of multilevel selection (Wilson and Wilson 2007). Indeed, non-robustness is a hallmark of human genetics at successive stages of male differentiation (Huang et al. 1999).

Heterozygous nonsense and missense mutations in SF1 associated with 46, XY pure gonadal dysgenesis likewise suggest (in the absence of adrenal abnormalities) a syndrome of haploinsufficiency (Mallet et al. 2004; Lin et al. 2007). Mutations in

SOX9, moreover, result in a syndrome of TF haploinsufficiency, designated campomelic dysplasia, wherein abnormalities of bone may coincide (in XY patients) with male, intersex, or female somatic phenotypes (Baek and Kim 2011). Such phenotypic variability suggests that the twofold transcriptional threshold characteristic of hSRY extends to its immediate downstream target. Similarly, hemizygosity of chromosome 9p24.3 (which contains three DM genes related to the classical

Doublesex gene of the sex-determining hierarchy of Dipterans (Raymond et al. 1998))

63

is associated with 46, XY gonadal dysgenesis in the presence of wild-type SRY and

SOX9 alleles (Barbaro et al. 2009). This trend extends to the androgen receptor itself.

Studies of the X-linked androgen insensitivity syndrome have demonstrated that the same receptor mutation can be associated with complete feminization (“testicular feminization”), partial insensitivity, or minimal perturbations in virilization or fertility‡(Bennett et al. 2010). Together, these clinical entities highlight the anomalous non-robustness of sexual dimorphism at multiple steps in the male developmental program.

64

Concluding Remarks

Our studies have addressed an overarching issue inhuman development: biochemical properties of TFs that distinguish critical boundaries between organized and disorganized states of cellular differentiation or downstream pathways of pattern formation (Nykter et al. 2008). Inherited alleles of SRY provide probes of this boundary in gonadogenesis. Our results, demonstrating that NCS and NCS-coupled phosphorylation of hSRY contribute at the margin to its genetic function, highlight the tenuous transcriptional threshold of human Sertoli-cell specification. A similar threshold pertains to testicular differentiation in mice (Wilhelm and Koopman 2006) despite the marked biochemical divergence of murine Sry (Chen et al. 2013b).

Given general trends toward the evolution of developmental stability

(Waddington 1959), why have human and murine SRY evolved to the edge of ambiguity? We speculate that sex determination differs from canonical embryonic patterning through its coupling to variation in extent of testosterone secretion by fetal

Leydig cells, in turn enabling male neurodevelopmental diversity. This perspective highlights the complementary and potentially conflicting roles of within-group and between-group selection as a feature of multi-level selection in social mammals

(Wilson and Wilson 2007). Implicit in this view are connections between genotype, development, differentiation of the central nervous system, and complex behaviors,

65

including empathy and other social competencies as defined in longitudinal studies of human fetal testosterone exposure (Baron-Cohen et al. 2004). The potential connection between multi-level selection and social competencies was anticipated by

Darwin’s surmise that “although a high standard of morality gives but a slight or no advantage to each individual man and his children over the other men of the same tribe,… an increase in the number of well-endowed men and advancement in the standard of morality will certainly give an immense advantage to one tribe over another” (Darwin 1871). The anomalous non-robustness of male sex determination and sexual differentiation in social mammals, as evidenced by inherited alleles of human SRY at the edge of ambiguity, may relate Darwin’s insight to the tenuous biochemistry of a genetic switch.

66

Materials and Methods

Plasmids. Plasmids expressing hSRY or variants were constructed by polymerase chain reaction (PCR) (Phillips et al. 2011). Following the initiator methionine, the cloning site encoded an HA tag in triplicate. In selected constructs an element encoding a nuclear localization signal (NLS; (sequence PRRRKV as derived from the large T antigen of simian virus 40 (SV40)) was inserted after HA-related codons.

Mutations in SRY were introduced using QuikChangeTM (Stratagene).

Rodent Cell Culture. CH34 cells were cultured in Dulbecco’s modified Eagle

medium containing 5% heat-inactivated fetal bovine serum at 37ºC under 5% CO2.

For proteasome-inhibitor studies, transfected cells were maintained for 24 h in serum-free conditions and then treated with MG132 for 6 h followed by 18 h incubation in 5% serum-containing medium.

Human Cell Lines. NT2-D1 cells were grown in Dulbecco’s modified Eagle's

medium in an atmosphere of 5% CO2; the complete growth medium contained FBS to a final concentration of 10%. PC-3 cells were cultured in the F-12K medium (ATCC)

with 10% FBS in 5% CO2 atmosphere. SOX9- and TBP-specific PCR primers were in accordance with human genomic sequences.

67

Transient Transfection. Transfections were performed using Fugene6 (Hoffmann

LaRoche). After 24 h in serum-free medium, cells were recovered using fresh DMEM medium containing 5% heat-inactivated fetal bovine serum. Transfection efficiencies were determined by ratio of GFP positive cells to untransfected cells following co-transfection with pCMX-SRY and pCMX-GFP in equal amounts). Subcellular localization was visualized by immunostaining 24-h post transfection following treatment with 0.01% trypsin (Invitrogen) and plating on 12-mm cover slips. SRY expression was monitored by Western blot. SRY expression was monitored in triplicate by western blot in relation to -tubulin (see Appendix VII).

Cycloheximide-chase Assay and Western Blot. 24-h post transfection cells were split evenly into 6-well plates and treated with cycloheximide to a final concentration of 20 mg/mL in regular medium for indicated times; cells were then lysed by RIPA buffer (Hoffmann LaRoche). After protein normalization, cell lysates were subjected to 12% SDS-PAGE and Western blot using anti-HA antiserum (Sigma-Aldrich) at a dilution ratio of 1:5000 with -tubulin as a loading control. Quantification was performed by Image J. Experiments were performed in triplocate.

68

Sox9 Activation Assay. SRY-mediated transcriptional activation of Sox9 and other endogenous CH34 genes was measured in triplicate by quantitative real-time-Q-rtPCR (qPCR) as described (Phillips et al. 2011). Cellular RNA was extracted using RNeasy (Qiagen).

Luciferase-Based Co-Transfection Assay. A firefly luciferase reporter plasmid was constructed using the Dual-Luciferase® Reporter Assay System (Promega); a

68-base-pair DNA duplex containing consensus SRY- and SF1 target sites was synthesized based on mouse TESCO fragment 6 (Sekido and Lovell-Badge 2008).

NT2-D1 cells were seeded at a density of 106 cells per well in 6-well plates.

Following co-transfection, lysates were collected, and luciferase assays performed as described by the vendor. An internal control was provided by renilla luciferase

(Promega). Fold-expression of firefly luciferase in response to wt or variant hSRY was defined in relation to empty-vector controls. Each measurement was monitored in triplicate.

Immunocytochemistry. Transfected cells were plated evenly on 12-mm cover slips, fixed with 3% para-formaldehyde in phosphate-buffered saline (PBS) on ice for 30 min, treated with cold-permeability buffer solution (PBS containing 10% goat serum

69

and 1% Triton X-100; Sigma-Aldrich) for 10 min, blocked with 10% goat serum and

0.1% Tween-20 in cold PBS (Sigma-Aldrich), and incubated overnight at 4 ºC with

FITC-conjugated anti-HA antibody (diluted to 1:400 ratio; Santa Cruz). After washing and DAPI staining, cells were visualized by fluorescent microscopy. Nuclear localization was evaluated by the ratio of SRY detected in nucleus to the total number of GFP-positive cells; 800-1000 cells were countedin each case.

Chromatin Immunoprecipitation. Cells were transfected with SRY variants, exposed to MG132, and subjected to ChIP. In brief, recovered cells were cross-linked in wells by formaldehyde, collected, and lysed after quenching the cross-linking reaction. Chromatin lysates were sonicated to generate 300-400-bp fragments and immunoprecipitated with anti-HA antiserum (Sigma-Aldrich) coupled with Protein A slurry (Santa Cruz) after pre-clearing; a non-specific antiserum (Santa Cruz) served as control. After reversal of cross-linking at 65 oC overnight, fragments were treated with proteinase K and RNase (Hoffmann LaRoche), followed by extraction with 1:1 phenol-CIAA solution. A high-fidelity PCR protocol was provided by the vender

(Hoffmann LaRoche). Experiments were performed in triplicate.

Phosphorylation Assay. HA-tagged SRY variants in cytosolic or nuclear fractions

70

(below) were immunoprecipitated with rabbit polyclonal anti-phosphoserine antiserum (Abcam). Western blot following 12% SDS-PAGE employed

HRP-conjugated anti-HA antibody (Hoffmann LaRoche). Loading controls were provided by cytosolic enzyme GAPDH (Sigma-Aldrich) and nuclear proteins histone

H1 and YY1 (Santa Cruz).

Wnt/-catenin Luciferase Assay.CH34 cells were engineered to stably express S37A

-catenin as described (Bernard et al. 2008). Assays, adapted from previous studies

(Bernard et al. 2008; Tamashiro et al. 2008), employed a chimeric SRY containing residues 1-155 of hSRY followed by residues 128-396 of mouse Sry. Experiments were conducted in triplicate using Dual-Luciferase Reporter Assay System (Promega); lysates were simultaneously analyzed for firefly luciferase activity encoded by

TOPflash (Millipore) and Renilla luciferase activity encoded by phRL-TK (Promega).

A negative control for -catenin-directed luciferase expression was provided by a

TOPflash variant containing an inactive -catenin-responsive promoter (FOPflash;

Millipore). Experiments were monitored in triplicate.

Co-immunoprecipitation Assays. CH34 cells expressing HA-tagged SRY variants were treated with MG132 and lysed using complete Lysis-M buffer containing a

71

protease inhibitor cocktail as described by the vendor (Hoffmann LaRoche). In

SRY-calmodulin (CaM) studies lysates were precipitated with monoclonal anti-HA agarose beads (Sigma-Aldrich). Following 12% SDS-PAGE, Western blots employed an anti-CaM antiserum (Abcam). Equal CaM loading was verified by Western blot. In

SRY-CRM1 studies lysates were analyzed by immunoprecipitation using anti-CRM1 antibody with agarose-conjugated protein G (Santa Cruz). Pellets were subjected to

10% SDS-PAGE, and HRP-conjugated anti-HA antiserum (Hoffmann LaRoche) was used to investigate the CRM1-bound HA-SRY variants. In a reverse protocol lysates were treated with monoclonal agarose-conjugated anti-HA antiserum (Sigma-Aldrich), subjected to 10% SDS-PAGE electrophoresis, and analyzed by anti-CRM1 antibody

(Santa Cruz) to detect SRY-bounded CRM1. In SRY-Exp4 studies transfected cells were co-transfected with pCMX-FLAG-human Exp4. MG132-treated cell lysates were immunoprecipitated by monoclonal anti-FLAG agarose using vendor’s protocol

(Sigma-Aldrich). After analysis by 10% SDS-PAGE and electroblotting, hybridization solutions containing HRP-conjugated anti-HA antiserum (Hoffmann LaRoche) were used to investigate Exp4-bound SRY. Anti-FLAG antiserum was used to monitor

Exp4 expression; respective antisera against HA tag and -tubulin (Sigma-Aldrich) provided SRY input- and general loading controls.

72

Cellular Fractionation. Cells were pelleted and suspended in lysis buffer (10 mM

HEPES (pH 7.9), 20 mM KCl, 3mM MgCl2, 0.5% NP-40, 5% glycerol, 10 g/ml leupeptin, 10 g/ml aprotinin, and 1 mM phenylmethylsulfonyl fluoride;

Sigma-Aldrich). Lysates were kept on ice, sheared by 5 passages through 25-gauge needle, and centrifuged at 2500 g for 15 min at 4 ºC; supernatants provided cytosolic extract. Pellets were suspended in nuclear lysis buffer (20 mM HEPES (pH 7.9),

0.225 M NaCl, 1 mM EDTA, 3 mM MgCl2, 0.5% NP-40, 10% glycerol, 10 g/ml leupeptin, 10 g/ml aprotinin, and 1 mM phenylmethylsulfonyl fluoride), sheared by needle passage, kept on ice for 15 min, and subjected to 13,000 g centrifugation for 15 min at 4 ºC. Supernatants provided nuclear extracts.

73

Footnotes

Acknowledgements.We thank Prof. P. K. Donahoe for cell line CH34 and encouragement and H.-Y. Kao for the pCMX plasmids. S. Jeong and P. Janki for assistance with nuclear localization studies; T. Feng, B. Li, and R. Singh for participation in early stages of this work; and P. DeHaseth, H.-Y. Kao, D. Samols, and

P. Sequeira for advice. MAW thanks B. Baker, F.A. Jenkins, Jr., P. Koopman, R.

Lovell-Badge, R. Sekido, and D. Wilhelm for discussion. This work, a contribution from the Cleveland Center for Membrane & Structural Biology, was supported in part by a grant to MAW from the National Institutes of Health (GM080505).

The present finding that I90M attenuates SRY function under conditions of plasmid dilution stands in disagreement with the results of co-transfection studies employing a reporter gene (luciferase) under the control of multimerized TESCO elements

(Knower et al. 2011). The latter studies demonstrated enhanced transcriptional activity of I90M SRY under conditions of its over-expression in Chinese hamster ovarian cells (presumably due to nuclear accumulation) in accordance with our results in the absence of plasmid dilution.

†Respective positioning of XX and XY fetuses in litter-bearing mammals influences local testosterone concentrations and hence postnatal reproductive traits including

74

timing of puberty, sexual behaviors and aggressiveness (Vom Saal 1989).

‡Instructive examples are provided by mutations R840C and R840H in -helix 9 of the ligand-binding domain of the androgen receptor (Beitel et al. 1994; Imasaki et al.

1994; Marcelli et al. 1994; Evans et al. 1997; Chu et al. 2002). These mutations are associated with a broad spectrum of partial phenotypes ranging from micropenis

(Reifenstein syndrome with patient raised as male) to external female genitalia

(incomplete testicular feminization with patient raised as female). Relative to the wild-type domain, biochemical affinities of the variant domains for testosterone are reduced by less than fourfold.

75

Figure 2.1 Domain organization of hSRY and summary of human genetics.

(A) hSRY and Swyer mutations. Residues 56-141 comprise HMG box (black) and basic tail (bt; light green).Other domains: PKA phosphorylation sites (dark green), bridge domain (purple, Br), and PDZ-binding motif (gray). De novo, unclassified mutations (amber and gray triangles, respectively), inherited mutations (filled red triangles), and mosaic fathers (open red triangles) are shown. Asterisks indicate V60L

(red), V60A (gray), and I90M (red). (B) Family trees pertaining to V60A, V60L, and

I90M. Arrows indicate proband; symbols are defined at Right. I90M was identified in two unrelated families.

76

Figure 2.2 Structure of SRY and outline of SRY-Sox9 regulatory axis.

(A) Ribbon model of SRY HMG box (front and side views; color code at right). V60 and I90 (balls and sticks) are highlighted in red and green.

(B) SRY (ribbon)-DNA (CPK model) complex. HMG color code is as in A; DNA atoms are light blue (bases), medium blue (deoxyribose), or dark blue (phosphodiester linkages). The minor wing comprises N-terminal -strand and C-terminal segment of 3; the major wing contains 1, 2, and N-terminal portion of 3. Asterisk indicates minor wing mini-core containing V60 (red balls and sticks).

(C) SRYSOX9 regulatory axis (red box) with genetic inputs (box at left) and outputs to a male-specific GRN () leading to inhibition of granulosa-cell fate (solid 77

), and Müllerian regression (dashed ). Red curved arrow indicates

SOX8/9-mediating feedback maintaining SOX9 expression. Abbreviations: FGF9, fibroblast growth factor 9; GATA4, GATA binding protein 4; LHX9, LIM homeobox

9; LIM1, homeobox protein Lhx 1; MIS (AMH), Müllerian Inhibiting Substance

(Anti-Müllerian Hormone); PTGDS, Prostaglandin D2 synthase; SF1, steroidogenic factor 1; WT1, Wilm’s tumor 1; Wnt, wingless-type. Boxes at bottom contain control qPCR primers: sex-independent SOX family genes (left) and housekeeping genes

(right).

78

Figure 2.3 Transcriptional activity and nuclear localization.

(A) Dependence of Sox9 expression on dose of transfected SRY plasmid: 1 g (black),

0.2 g (dark gray), 0.1 g (medium gray), 0.02 g (white); total transfected DNA was the same. Brackets designate p-values (*) <0.05 or (**) <0.01; “ns” indicates p-values > 0.05. 79

(B) Cycloheximide assay. Proteolysis is enhanced by V60L and V60A but unaffected by I90M. Gels are at left. Graph at right depicts SRY degradation following inhibition of protein synthesis.

(C) Subcellular localization of epitope-tagged SRY: DAPI nuclear staining (upper row blue) and SRY (lower row green). V60L and V60A SRY are pancellular; nuclear accumulation was rescued by SV40 NLS (right). I90M augments nuclear localization. Images were obtained following MG132 treatment.

(D and E) Histograms describing nuclear (gray) and pancellular (white) patterns in absence (D) or presence (E) of MG132. Gray and white bars sum to <100 due to occasional GFP-positive cells lacking SRY. SRY was at highest dose in panel A.

Brackets indicate statistical comparisons. “+NLS” indicates rescue of by SV40 NLS.

I90M augments nuclear localization under all conditions tested.

(F) Relationship between Sox9 expression and SRY dose. qPCR data were obtained in absence (right) or presence (left) of MG132. Within each set additional data were obtained with fused SV40 signal (+NLS). At lowest dose of V60L or V60A, rescue of

Sox9 expression required only fused NLS. I90M SRY at higher doses was super-active; plasmid dilution unmasked a twofold defect unaffected by fused NLS or

MG132.

80

Figure 2.4 Chromatin immunoprecipitation and interactions with nuclear import-export machinery.

(A-D) ChIP assays probing SRY occupancy of target sites within Sox9 testis-specific enhancer core element (TESCO). (A) TESCO fragment boxes (black, with SRY binding sites and white, sites with no detectable binding). (B) Primer sets a and c 81

probed for SRY occupancy; primer set b provided negative control. (C) Histogram and representative gel show results of ChIP of variants. Gel showing results of ChIP of variants without NLS rescue. R62G and R75N are de novo clinical mutations in the

N-terminal bipartite NLS of the HMG box and in part impair specific DNA binding.

(D) Histogram and representative gel showing ChIP studies of variants with

NLS/MG132 rescue: lanes 8 and 9, without rescue; lanes 10 and 11, NLS fused; lanes

12 and 13, NLS/MG132 “double rescue”; lane 14 and 15, positive and negative hSRY controls, respectively. At Right are non-specific immunoglobulin G (IgG) controls. (E)

Schematic outline of CRM1 and Exp4 co-IP assays. (F and G) Biochemical studies of the binding of epitope-tagged hSRY variants to nuclear import (Exp4) or nuclear export machinery (CRM1). Internal loading controls are provided by -tubulin

(bottom panels). (F) Exp4 co-IP assay. Whereas the de novo mutations in the

N-terminal NLS exhibit marked impairment, V60L exhibits only mild impairment, and V60A intermediate; no defect was observed for I90M. (G) CH34 CRM1 co-IP assay. Mutations at position 60 did not impair binding to CRM1; I90M was associated with almost complete lack of binding.

82

Figure 2.5 Calmodulin binding and Wnt signaling.

(A-C) hSRY binding to CaM. (A) Design of co-IP assay employing transfected

HA-tagged SRY variants. Complexes were analyzed by SDS-PAGE immunoblotted with anti-CaM antiserum. (B and C) Western blots documenting similar CaM binding in wild-type and variant complexes; differences were not statistically significant.

Results were normalized by CaM immunoblotting (bottom panels). Control studies employed other de novo mutations at positions 62, 64, 68, 75, and 76 lie within the 83

CaM-binding motif; position 109 lies on the back side of the major wing; and 133 lies within the C-terminal NLS. (D) Male gonadal ridge (Left) requires positive and negative regulatory steps ( and , respectively) to activate the testicular program

(SOX9-dependent) and repress the ovarian program (Wnt/-catenin pathway). Dotted interactions designate cross-talk; “X”s indicate absence of ovaries and female somatic differentiation; and “t” designates testosterone-dependent virilization. (E)

Wnt/-catenin reporter-gene assay. Cells were split, co-transfected by wild-type or variant SRY plasmids to enable luciferase read-out. (F) Histogram summarizing

Wnt/-catenin repression assay performed at the lowest plasmid dose (50-fold dilution as defined in Fig. 2.3). MG132 was added to ensure equal levels of protein expression. Repression relative to wild-type SRY was in part relieved by inherited mutations V60L, V60A, and I90M. The repression-related activity of V60L and V60A was almost fully restored by fused NLS. Full relief was observed in control studies employing substitution I68A and de novo mutation G95R (which block specific DNA binding(King and Weiss 1993; Tamashiro et al. 2008)). Brackets designate statistical comparisons as described in Figure 2.3.

84

Figure 2.6 Coupling between nucleocytoplasmic trafficking of SRY and its phosphorylation.

(A) SRY phosphorylation assay exploited fractionation of SRY into nuclear and cytosolic fractions with subsequent co-IP analysis of total and phosphorylated subsets.

85

(B) Western blots of cytosolic and nuclear SRY fractions (left and right). Mutations are as labeled; controls were provided by transcriptional repressor YY1 (nuclear protein) and GAPDH (cytosolic enzyme).

(C) Schematic depiction of SRY analogs containing substitution within its N-terminal three contiguous PKA phosphorylation sites (PALRRSSSFLCTE; putative phosphorylation sites underlined): top, wild-type motif; middle, SerAla substitutions (AAA) to render motif non-phosphorylatable (“phospho-dead”); bottom,

SerAsp substitutions (DDD) to mimic the negative charges (“phospho-mimic”).

Domains are defined as in Fig. 2.1A.

(D) Effects of AAA or DDD substitutions on transcriptional regulatory activity of wild-type SRY, inherited I90M variant, or SRY NES in which putative NES was impaired by Ala substitutions I90A, L94A, and M100A. Studies were performed at the lowest dose employed in Figure 2.3. Horizontal brackets designate statistical comparisons.

86

Fig. 2.7 Potential N-terminal phosphorylation sites and NES of primate SRY alleles.

(A) Phylogenetic tree of representative primates; Mya: million years ago (panelis modified from ref (Ferguson-Smith and Trifonov 2007). (B) Known or predicted phosphorylation sites (PKA); serines inmagenta (residues 31-33 in hSRY) depict observed phosphorylation sites (Desclozeaux et al. 1998). Serines in bold (non-human primates) represent potential phosphorylation sites as predicted by NetPhosK 1.0

(website http://www.cbs.dtu.dk/services/NetPhosK). R30, a site of clinical mutation in

87

hSRY adjoining the PKA site, is underlined. (C) Alignment of representative primate

SRY HMG box sequences in relation to rodent Sry proteins (rat and mouse). Boxed are

NLS-1, NES, and NLS-2. Critical NES residues are in bold. Sites of mutations impairing NCS of hSRY (as investigated in previous studies) are underlined (from Left with full-length residue numbers: R62(G), R75(N), and R133(W/G)). V60 and I90

(focus of the present study) are highlighted in yellow boxes. Highlighted in red at bottom is a non-conservative substitution in the HMG box of rat and mouse Sry (MS) proposed to disable its NES.

88

Table 2.1. Properties of SRY Variantsa

e SRY ∆∆Gu relative Sox9 mRNA accumulation 9 b o c Kdx 10  Tm ( C) variant (kcal/mol)d assay A assay B assay C wild type 14±3 74o 40 - 100± 4 100± 4 100± 6

V60A 27±5 72o 40 0.1± 0.1 42± 9 93± 8 98± 6

V60L 32±2 71o 40 0.1± 0.1 47± 9 89± 7 92± 5

I90M 15±1 74o 36 0.5± 0.1 43± 8 42± 9 46± 5

aBiophysical studies employed the isolated SRY HMG box; qPCR studies employed

full-length HA-tagged SRY. FRET-based estimates of dissociation constants (Kd)

revealed a small decrease in affinity of the V60A and V60L HMG box relative to

wild-type, whereas that of I90M was unchanged. Stopped-flow FRET studies of

protein-DNA dissociation demonstrated that respective dissociation-rate constants (koff)

-1 of V60L, V60A, and I90M domains are increased (koff 0.24(±0.008) s , 0.17(±0.005)

s-1; and 0.14(±0.001) s-1, respectively) relative to wild-type (0.033(±0.001) s-1; see ref

(Phillips et al. 2011) and Fig. S2.2).

bDNA bend angles were investigated by permutation gel electrophoresis; uncertainties

were ±1o(Phillips et al. 2006). V60A and V60L values are from Phillips, N.B., et al.

(Phillips et al. 2011); I90M value is from Knower, K.C., et al. (Knower et al. 2011).

cMidpoint unfolding temperatures were based on CD (King and Weiss 1993).

d Differences of free energies of unfolding (Gu) were estimated based on a two-state

model (Phillips et al. 2011); wild-type Gu was determined to be 3.8(±0.1) kcal/mole.

eValues of Sox9 mRNA accumulation were at lowest plasmid dose (50-fold dilution;

89

Fig. 2.3) to avoid potential artifacts of over-expression. Assay A provides baseline values in absence of SV40 NLS or proteosome inhibition; assay B, effects of fused

SV40 NLS; and assay C, effects of NLS-MG132 double rescue. This data was obtained by Joe Racca as part of his collaboration on this project.

90

Fig. S2.1 I90M (residue 35 in HMG box) does not perturb folding or stability.

(A) Far-UV CD spectra of the native- (●) and mutant domain (I90M, □) are similar at

4 C. (B) Guanidine-induced unfolding of native (●) and I90M domains (□) as

monitored by CD at 222 nm. Estimates of (ΔGu) were extracted by a two-state model.

(C) Thermal unfolding of wild-type(●) and I90M (□) SRY domains (CD 222 nm).

Midpoint temperatures were obtained from the derivative of the curve. This data was obtained by Joe Racca as part of his collaboration on this project.

91

Fig. S2.2 FRET-based equilibrium binding and kinetics.

(A) Stopped-flow experiment coupled to a fluorimeter, enabling measurement of

FRET-based dissociation rate. One syringe contained a preformed protein-DNA complex with the 15-bp DNA probe containing 5’-donor (fluorescein 6-FAM; green circle) on one strand and 5’-acceptor (TAMRA; red circle) on the other; the other syringe contained a 20-fold excess of unmodified DNA site. (B)FRET-based

stopped-flow measurement of dissociation rates (koff). Time-dependent increase in donor fluorescence was due to dissociation of FRET-labeled complex. Data and fitted solid lines at 15 C are shown for wild type- (blue) and I90M-complexes (red).

Dissociation reactions were monitored for 90 sec. Dissociation rate constants, koff, were determined by fitting traces to a single exponential equation and averaging the

-1 results of three replicates. Values of koff are 0.14(±0.001)sec for I90M and

0.041(±0.001)sec-1 for wild-type. (C) FRET-based equilibrium-binding studies of

wild-type and I90M domains enabled determination of Kd. The Kd of the variant

92

domain (15.0(±1.4) nM) is indistinguishable from wild-type (14.2(±2.6) nM. This data was obtained by Joe Racca as part of his collaboration on this project.

93

Fig. S2.3 Emission spectra of free and SRY-bound DNA after excitation at 490 nm.

(A) DNA singly labeled with donor 6-FAM in complex with wild type SRY (green) and in complex with I90M (light blue); DNA doubly labeled with 6-FAM and

TAMRA (acceptor), free DNA (black), in complex with wild type SRY (dark blue) and in complex with I90M (orange). Emission spectra of DNA singly labeled with

TAMRA (free, complexed with wild-type domain and complexed with the I90M domain) are shown in yellow (beneath overlapping spectra). (B) Emission of normalized rhodamine (TAMRA) spectra in double labeled DNA due to FRET indicating similar degree of bending by wildtype (dark blue), and I90M (orange). The free DNA spectrum is shown in black. This data was obtained by Nelson Phillips as part of his collaboration on this project.

94

Fig.S2.4 SRY-regulated testicular gene-regulatory network (GRN) and transcriptional activation of Sox9.

Transcriptional assays of selected genes activated by SRY variants in rat embroyonic gonadal cell line. RT-Q-rt-PCR was employed to probe mRNA abundances of (A) Sox family members, (B) candidate male GRN-related genes, and (C) housekeeping genes. qPCR was analyzed following transfection of SRY variant expression plasmids, empty vector, or control plasmid expression a stable but inactive SRY variant (I68A).

(A) Fold mRNA accumulation of Sox family, including endogenous Sry and transfected SRY-activated Sox9. (B) Fold mRNA accumulation of sex-related factors in the program of SRY-mediated GRN. Results show that CH34 cells express low endogenous levels of Fgf9, Ptgds, Sox8, and Wnt5 gene expression. (C) Fold mRNA accumulation of sex-unrelated housekeeping genes; these gene were not affected by expression of transfected SRY; statistical analyses: *Wilcox test, p<0.05.

95

Fig. S2.5 MG132 rescues expression of V60L and V60A to achieve levels similar to wild-type SRY.

Top, Western blots probing extent of total-cellular expression of wild-type or variant

SRY constructs following 24 hours in serum-rich medium in the absence of proteosome inhibitor MG132. In the absence of the SV40 NLS, V60L and V60A bands are ca. twofold less intense than that of wild-type; addition of the SV40 NLS enhances total-cellular expression; loading controls are as described in Figure 2.3.

Bottom, addition of MG132 leads to similar band intensities in each lane. Note that these data do not probe nucleus-specific accumulation of SRY. NLS-tagged constructs are 0.7 kDa larger than non-NLS-tagged constructs. It is not known why V60A constructs exhibit slightly faster mobilities. This figure was taken from the left-hand

(lanes 1-3) and right-hand (lanes 4 and 5) portions of a single gel and blot; the middle lanes (not pertinent to this study) were omitted.

96

Fig. S2.6 Subcellular localization of wild-type SRY as analyzed by immunostaining.

(A) Nuclear staining by DAPI (blue) and anti-HA immunofluorescence, which specifically detects the epitope-tagged transfected SRY (green). Wild-type SRY containing in-frame N-terminalSV40 NLS exhibits native-like extent of nuclear localization without further increase in the nuclear fraction. Images were obtained following MG132 treatment to equalize levels of total SRY protein expression. (B)

Histogram showing quantitative summary of extent of nuclear localization (gray) or pancellular distribution (white) in the without/with MG132. The dose of the

SRY-expression plasmid was undiluted (1 g/well).Horizontal brackets designate statistical comparisons as in Figure 2.3.

97

Fig. S2.7 SRY-Exportin-4 co-IP assays. Histogram provides a quantitative summary of Western blots repeated in triplicate.

Mutations in the SRY HMG box lead to a range of perturbations to the interaction between SRY and Exportin-4, which is mediated by the N-terminal bipartite NLS of the HMG box (see Fig. 3F in the main text). Error bars represent 1 standard deviation:

*, p-values less than 0.05, and “ns,” not significant.

98

Fig. S2.8 Subcellular localization and CRM1 binding of mammalian SRYs.

(A) Histogram indicating fractions of transfected CH34 cells exhibiting exclusive nuclear localization of SRY (filled bars) versus pancellular distribution (open bars); the transfected plasmid dose was in each case 1g. Whereas the localization patterns of human, goat, and deer SRY share indistinguishable (gray bars), mouse and rat Sry, exhibit near-exclusive nuclear localization (red bars; see companion paper in this issue; Chen Y.-S., et al. PNAS (2013)). At bottom is summarized CRM1-binding activity as observed in panel B(for experimental design, see Fig. 4E). Horizontal brackets indicate statistical comparisons (asterisk, p<0.05; ns, p>0.05 and in this case, the ns p-values were all >0.15). (B) CRM1 co-IP assay in transfected CH34 cells.

Whereas human SRY (hSRY), goat Sry (gSry; Capra hircus) and deer Sry (dSry;

Hydropotesinermis) exhibit similar CRM1-co-IP signals (lanes 1, 2, and 3, respectively), mouse Sry (mSry; Mus musculus) rat Sry (rSry; Rattus norvegicus) lack such binding in association with a variant NES motif (IxxxLxxxxxSL). Top boxes,

99

background bands (X) and SRY-CRM1 co-IP signal (arrow); middle boxes, species-specific Sry/SRY input bands; and bottom boxes, -tubulin control for protein loading. Images were taken from a single gel.

100

Fig. S2.9 Subcellular localization and CRM1 binding of hSRY variants with NES modifications.

(A) Histogram showing fractions of transfected CH34 cells exhibiting exclusive nuclear localization of SRY (filled bars) versus pancellular distribution (open bars).

Wild-type and variant NES motifs are as labeled; the transfected plasmid dose was in each case 1g. Variant NES motifs (IxxxLxxxSL and AxxxAxxxxxAL) in hSRY confer near-exclusive nuclear localization as is characteristic of mouse Sry (mSry; see companion paper in this issue). At bottom is summarized observed (+) or impaired (-) binding to CRM1. Horizontal brackets indicate statistical comparisons (asterisk, p<0.05; ns, p>0.05 and in this case, the ns p-values were all > 0.15). (B) Binding of

SRY variants to CRM1 as probed by co-IP assay (for experimental design, see Fig.

4E). Top boxes, background band (X) and SRY-CRM1 co-IP signal (arrow); middle boxes, mSry input band (upper band in lane 4) and hSRY input band (lower bands in lanes 1-3); and bottom boxes, -tubulin control for protein loading. Lanes 1-4 101

provide: (1) wild-type human SRY (wild-type NES motif designated at top –ML); (2) variant hSRY bearing mouse-specific NES substitution IxxxLxxxxxSL (designated at top –SL); (3) variant hSRY bearing Ala-substituted NES substitution AxxxAxxxxxAL

(designated –AL); and (4) control studies of mouse Sry (mSry) bearing impaired NES sequence IxxxLxxxxxSL (designated at top –SL). Images were taken from a single gel.

102

Fig. S2.10 V60L and V60A do not perturb binding of calmodulin (CaM).

(A) Model of the N-terminal segment of SRY (red) bound to CaM (gray)(Harley, V. R., et al. FEBS Lett. 391, 24-8 (1996)). Calcium ions in CaM are shown as silver balls.

Arrow indicates V60 (position 5 in HMG-box consensus). (B) CD spectra of wild-type, V60L, V60ASRY domains and their CaM complexes. Top group, spectra of free domains at 20 C: wild-type (), V60L (□), and V60A (). Middle group, spectrum of CaM alone ().Bottom group, spectra of SRY-CaM complexes: wild-type

(+), V60L (), or V60A SRY (); the latter were obtained at 20 C under the same conditions as above after incubating at 4 C for 30 min. SRY-CaM complexes exhibit

103

indistinguishable patterns. (C) Trp fluorescencespectra of SRY HMG box and variants either free or as CaM-bound complexes. Bottom group, free SRY domains: wild-type

(), V60L (), and V60A (). Top group, SRY-CaM complexes: wild-type (),

V60L (), and V60A (). In presence of CaM, similar enhancement and shift in emission wavelength maxima (representing reversal of the quenching and blue shift characteristic of the folded free domain) are observed. (D) Titration of SRY domains

with increasing [CaM] as monitored by Trp fluorescence. F0 is the tryptophan fluorescence at 350nm of SRY HMG box alone (2 M); F is the observed fluorescence upon CaM addition. CaM titration curves with the variants are similar, indicating similar CaM-binding affinities: wild-type (), V60L (), and V60A ().

(E and F) DNA gel-mobility shift assays (GMSA) depicting competition between

CaM and 33P-labeled DNA (36 bp) for wild-type (E)and V60A SRY (F)domain.

Similar extent of competition was observed. Domains were either pre-complexed with the specific DNA site prior to the addition of [CaM] (lanes 1-9 in panels E and F; 0,

0.08, 0.3, 0.6, 1.3, 2.5, 5, 10, and 20 M) or pre-complexed (lanes 2’-9’ in panels E and F) prior to the addition of DNA. Bands representing free DNA and shifted 1:1 complex C1 are indicated by arrows at right; higher-order complexes C2 and C3 are also indicated. This data was obtained by Nelson Phillips as part of his collaboration on this project.

104

Fig. S2.11. Evidence that nucleocytoplasmic trafficking of hSRY does not depend on phosphorylation. (A) Schematic depiction of hSRY analogs containing triplicate substitutions within its N-terminal PKA phosphorylation site (PALRRSSSFLCTE; putative phosphorylation sites underlined): Top, wild-type motif; Middle, SerAla substitutions

(AAA) to render motif non-phosphorylatable ( “phospho-dead ” ); Bottom, SerAsp substitutions (DDD) to mimic the negative charges of phospho-serine (“phospho-mimic”).

Domains are defined as in Fig. 7C in main text. (B) Non-box AAA or DDD substitutions do not affect nuclear localization. Histogram provides fractions of transfected CH34 cells exhibiting exclusive nuclear localization of SRY (gray filled bars) versus pancellular distribution (open bars). The transfected plasmid dose was in each case 1 g. The hSRY AAA or DDD variants thus exhibit indistinguishable subcellular localization. Horizontal brackets indicate statistical comparisons as in Figure 2.3 (ns, p>0.05 and in this case, the ns p-values were all > 0.2. 105

Fig. S2.12. Transient transfection of hSRY in PC-3 cells activates endogenous

SOX9. (A) Histogram showing results of wild-type and variant rt-Q-PCR assays

(plasmid dose 1 g) with an empty vector and inactive hSRY variant I68A as negative controls. V60L and V60A attenuate SOX9 activation by twofold. At this high plasmid dose I90M has 30% higher activity than wild-type hSRY. Statistical comparisons: p-value (*) < 0.05; “ns” indicates p-value > 0.05. (B) Western blots probing extent of total cellular expression of hSRY and variants. Top: similar band intensities are shown in each lane. Bottom: -tubulin loading controls. We note that PC-3 cells, derived from a human prostate-cancer cell line, has a Wnt/-catenin-responsive SOX9 gene

(Wang, H., et al. (2007) Cancer Res. 67, 528-536) but lacks endogenous SRY (Dasari,

V., et al. (2002) J. Urology 167, 335-338). 106

Fig. S2.13. Luciferase-based co-transfection assay of hSRY and variants. (A)

Design of luciferase reporter construct. Respective SRY- and SF1 consensus DNA binding sites are shown as red and green boxes. (B) Transfection of NT2-D1 cells with hSRY or variants (100 ng plasmid) stimulates expression of firefly luciferase.

NT2-D1 cells, derived from a human testis carcinoma, express FGF9, SF1, and

SOX-9 and so may in part resemble gonadal cell types (Knower, K.C., et al. (2007)

Sex Dev. 1, 114-126). As in CH34 cells, mutations V60L and V60A impair luciferase activity by ca. twofold. NES mutation I90M enhanced activity by 60% relative to wild type hSRY. Statistical comparisons: p-value (*) < 0.05, (**) < 0.01; “ns” indicates p-value > 0.05.

107

Table S2.1 Nucleocytoplasmic trafficking of SOX factorsa

aFor each member of the Sox family, the availability of prior studies of trafficking is indicated by + (yes) or – (no). Columns indicate known components of the nuclear import or export machinery implicated in trafficking of that factor.

108

Table References

1. Wegner M (1999) From head to toes: the multiple facets of Sox protein. Nucleic Acids

Res 6:1409-1420.

2. Bowles J, Schepers G, & Koopman P (2000) Phylogeny of the SOX family of developmental transcription factors based on sequence and structural indicators. Dev Biol

227:239-255.

3. Pevny LH, Sockanathan S, Placzek M, & Lovell-Badge R (1998) A role for SOX1 in neural determination. Development 125:1967-1978.

4. Malki S, Boizet-Bonhoure B, & Poulat F (2010) Shuttling of SOX proteins. Int J

Biochem Cell Biol 42:411-416.

5. Baltus GA, et al. (2009) Acetylation of Sox2 induces its nuclear export in embryonic stem cells. Stem Cells 27:2175-2184.

6. Gontan C, et al. (2009) Exportin 4 mediates a novel nuclear import pathway for Sox family transcription factors. J Cell Biol 185:27-34.

7. Woods KS, et al. (2005) Over- and underdosage of SOX 3 is associated with infundibular hypoplasia and hypopituitarism. Am J Hum Genet 76:833-849.

8. Collignon J, et al. (1996) A comparison of the properties of Sox-3 with Sry and two related genes, Sox-1 and Sox-2. Development 122:509-520.

9. Futaki S, Hayashi Y, Emoto T, Weber CN, & Sekiguchi K (2004) Sox7 plays crucial

109

roles in parietal endoderm differentiation in F9 embryonal carcinoma cells through regulating

Gata-4 and Gata-6 expression. Mol Cell Biol 24:10492-10503.

10. Kennedy CL, Koopman P, Mishina Y, & O'Bryan MK (2007) Sox8 and Sertoli-cell function. Ann N Y Acad Sci 1120:104-113.

11. Barrionuevo F & Scherer G (2010) SOX E genes: SOX9 and SOX8 in mammalian testis development. Int J Biochem Cell Biol 42:433-436.

12. Sim H, Argentaro A, & Harley VR (2008) Boys, girls and shuttling of SRY and SOX9.

Trends Endocrinol Metab19:213-222.

13. Rehberg S, et al. (2002) Sox10 is an active nucleocytoplasmic shuttle protein, and shuttling is crucial for Sox10-mediated transactivation. Mol Cell Biol 22:5826-5834.

14. Chen YH, Gao J, Fan G, & Peterson LC (2010) Nuclear expression of sox11 is highly associated with mantle cell lymphoma but is independent of t(11;14)(q13;q32) in non-mantle cell B-cell neoplasms. Mod Pathol 23:105-112.

15. Hoser M, et al. (2008) Sox12 deletion in the mouse reveals nonreciprocal redundancy with the related Sox4 and Sox11 transcription factors. Mol Cell Biol 28:4675-4687.

16. Kasimiotis H, et al. (2000) Sex-determining region Y-related protein SOX13 is a diabetes autoantigen expressed in pancreatic islets. Diabetes 49:555-561.

17. Hargrave M, et al. (2000) Fine mapping of the neurally expressed gene SOX14 to human 3q23, relative to three congenital diseases. Hum Genet 106:432-439.

110

18. Lee HJ, et al. (2004) Sox15 is required for skeletal muscle regeneration. Mol Cell Biol

24:8428-8436.

19. Gimelli S, et al. (2010) Mutations in SOX17 are associated with congenital anomalies of the kidney and the urinary tract. Hum Mutat 31:1352-1359.

20. Irrthum A, et al. (2003) Mutations in the transcription factor gene SOX18 underlie recessive and dominant forms of hypotrichosis-lymphedema-telangiectasia. Am J Hum Genet

72:1470-1478.

21. Dunn TL, et al. (1995) Sequence and expression of Sox-18 encoding a new HMG-box transcription factor. Gene 161:223-225.

22. Chakravarthy H, Ormsbee BD, Mallanna SK, & Rizzino A (2011) Rapid activation of the bivalent gene Sox21 requires displacement of multiple layers of gene-silencing machinery.

FASEB J 25:206-218.

23. Hosoya M, et al. (2011) Expression and function of Sox21 during mouse cochlea development. Neurochem Res 36:1261-1269.

111

Chapter III

A binary fate decision in human development is rendered ambiguous by accelerated proteosomal degradation of a transcription factor

Introduction

Developmental stability of ordinary metazoan development in the presence of cryptic genetic variation, known as Waddington canalization (Waddington 1959), has stimulated interest in the biochemical mechanisms of genetic capacitors (such as heat shock proteins (Rutherford and Lindquist 1998)) and the topological properties of gene-regulatory networks (Siegal et al. 2007). A common framework for understanding developmental stability is provided by the Hox gene cluster, invariant even among unrelated body plans (Barmina and Kopp 2007). Sex-determination is different. Metazoans exhibit a remarkable diversity of genetic sex-determining systems and signals, including temperature and social cues (Rhen and Crews 1999;

Kobayashi et al. 2009). Intersexual phenotypes are readily obtained in model organisms (Baker 2011) and abound in the wild (Storrs-Mendez and Semlitsch 2010).

This raises the question of a broad tolerance exhibited by the male developmental system. To investigate the extent and degree of this threshold, in this study we have

112

exploited the enigma of inherited human sex reversal (Harley et al. 1992; Jäger et al.

1992; Phillips et al. 2011) to measure the transcriptional threshold of testis determination.

The male phenotype in mammals with the exception of monotremes is determined by SRY, a gene on the Y chromosome encoding an architectural transcription factor (Fig. 3.1A) (Sinclair et al. 1990). The role of SRY as a testis-determining factor is supported by transgenic murine models (Koopman et al.

1991) and studies of human clinical mutations leading to gonadal dysgenesis (Swyer’s

Syndrome; 46, XY gonadal dysgenesis) (Knower et al. 2003; Knower et al. 2011).

Human SRY is encoded by an intronless gene (Su and Lau 1993) containing a high mobility-group (HMG) box with an N-terminal -strand (HMG box consensus residues 1-11, Fig. 3.1A) and three -helices (1, 2, and 3, Fig. 1C) (Murphy et al.

2001). The HMG box is a conserved motif mediating DNA binding and DNA bending (Ner 1992) and serves as a critical domain for the transcriptional regulatory role of SRY. This signature domain (dark blue in Fig. 3.1A) and its basic tail (bt, light blue) are conserved among an extensive family of SRY-related transcription factors

(designed SOX, Sry-related box) broadly involved in development. Direct binding of

SRY to specific regulatory DNA sites within the testis-specific enhancer (TES) of

SOX9 activates a male-specific gene-regulatory network in the fetal gonadal ridge

113

leading to Sertoli cell differentiation (Sekido and Lovell-Badge 2008). Swyer mutations in SRY are categorized into genetic classes (Fig. 3.1A). Most often arising de novo in spermatogenesis and clustering in the HMG box (green triangles Fig.

3.1A). Such mutations markedly impair specific DNA binding and/or DNA bending by direct or indirect perturbation of an angular protein-DNA interface (Weiss 2005;

Sekido and Lovell-Badge 2009; Kashimada and Koopman 2010; Knower et al. 2011).

Variant SRY alleles may also be inherited. The present study has focused on inherited mutations (F109S and L163X; asterisks in Fig. 3.1A). Whereas similar molecular mechanisms underlie inheritance of dysfunctional alleles from mosaic fathers (open red triangles in Fig. 3.1A), an enigma is posed by the inherited mutations (red triangles in Fig. 3.1A) with compatibility of the same SRY allele leading to alternative developmental outcomes (family trees in Fig. 3.1B): testicular differentiation leading to virilization (fertile 46, XY father) or nascent ovarian differentiation leading to gonadal dysgenesis (sterile 46, XY daughter) (Knower et al. 2011).

In the structure of the SRY-DNA complex, F109 (HMG box consensus position

54; Fig. 3.1A) flanks the major wing of the HMG box (Packing of the 1, 2, and the

N-terminal portion of 3 comprise the major wing, Fig. 3.1C). However, mutation at this position was not expected to lead to direct perturbation of the DNA-binding surface, and indeed, F109S was originally reported to exhibit normal specific

114

DNA-binding activity as probed by gel-mobility-shift assays (GMSA) (Jäger et al.

1992). Another familial mutant used in the present study is nonsense mutation SRY

L163X, which contains the native HMG box but lacks the last 41 amino acids of the protein. A cellular surveillance mechanism usually functioning to “search and kill” transcripts with nonsense mutations, the program of mRNA nonsense-mediated decay

(NMD), might prevent the expression of truncated SRY proteins, which could be associated with the L163X Swyer outcome. However, this surveillance system is triggered by exon junction complexes during the pre-mRNA processing. Thus, an

SRY transcript with a nonsense mutation could still survive because of the intronless open reading frame of the SRY gene (Nott et al. 2004; Chang et al. 2007).

Our investigation of these SRY mutations was motivated by a seeming paradox: its inheritance implies that these variant alleles are directed toward two alternative developmental fates. Whereas the reported incomplete penetrance of these inherited mutations could explain the phenotype of the father, what of the daughter displaying a

Swyer phenotype?

Inherited mutations in SRY provide experiments of nature that help to define the threshold molecular properties of a developmental switch beyond DNA binding and bending. To this end, our studies have exploited a model of the central SRY-Sox9 regulatory axis in a rodent fetal pre-Sertoli cell line (Haqq et al. 1994; Phillips et al.

115

2011). Our findings demonstrate that F109S and L163X attenuate SRY-dependent transcriptional activation of Sox9 and hence the representative factors in the downstream gene-regulatory network (Fgf9 and Ptgds) in accordance with the phenotypes of the XY daughters. The extent of attenuation is twofold. This factor of two defines a critical boundary between testicular self-organization and gonadal dysgenesis, and this subtle effect of SRY-SOX9 central regulation in these Swyer families challenges the principle of Waddington canalization, which suggests developmental processes would overcome variations in conditions and lead to a definite outcome (Waddington 1959; Masel and Siegal 2009) in the developmental program, including testicular development (Sekido and Lovell-Badge 2009).

116

Results

The investigations of inherited F109S and L163X SRY mutations were motivated by the seeming paradoxical result using HMG box domains of these clinical mutants whereby their transcriptional activation function was found to be indistinguishable from that of the wild-type SRY (F109S and L163X, L163X contains a wild-type box and hence as expected displayed native-like binding) in study of Jäger et al (Jäger et al. 1992), and the Y1H screening (Fig. 3.2). However, using a rat embryonic pre-Sertoli cell line, CH34 (Haqq et al. 1994), the F109S and L163X substitutions exhibited reduced SRY-Sox9 transcriptional activation. Although

DNA-binding function of F109S HMG box in the yeast model is tolerant and displayed native-like results (Fig. 3.2C), spectrometry-based biophysical investigations unmasked a less stable property of the PheSer substituted HMG box.

Furthermore, mutation L163X containing a native HMG box shared a similar extent of reduced Sox9 activation and a trend of accelerated degradation as was observed with the F109S substitution in the pre-Sertoli cell model.

Quantitative Yeast-1-Hybrid screening demonstrated mutant variants exhibiting native-like DNA-binding. An Y1H system with an integrated reporter constructed to express β-galactosidase under the control of triplicate consensus SRY-binding sites

117

was established (5’-ATTGTT-3’ and complement; Fig. 3.2A). This strain was employed in the screening of transcriptional responses of SRY variants. The test plasmids encoded a fusion of either wild-type or variant SRY HMG box domain attached to the transcriptional activation domain of Gal4 (Fig. 3.2B (Phillips et al.

2011)). Whereas the wild-type and F54S (numbered using the HMG box-only consensus, motif position 109 in the full-length sequence) fusion proteins gave rise to

β-galactosidase expression, reflected in the increased miller units (Fig. 3.2C, black and red, respectively), deletion of the SRY HMG box (“empty control”; Fig. 3.2C, white) led to a loss of β-galactosidase expression. Thus, the Y1H results strongly suggest that specific DNA-binding activity was indistinguishable between wild-type and F54S SRY HMG box variant.

The mutation F54S affects the stability of the free HMG box domain. As demonstrated by Clore and co-workers, the DNA-bound structure of the SRY HMG box is similar to that of isolated nonsequence-specific HMG boxes (Phillips et al.

2011). Conserved residues in both are seen to form a cluster of nonpolar and aromatic side chains. This cluster occurs at the confluence of helices α1–α3 (major wing); F54

(F109 in full length SRY) is located in this region of the HMG box and interacts in concert with W70 and W98 in stabilizing the major wing through aromatic-aromatic

118

contacts that exist between the tryptophan side chains (W70/W98) and the phenylalanine ring in the wild type structure (Fig. 3.1C). These residues are highly conserved among SRY and SOX HMG box domains. The Phe to Ser mutation results in a marked change from a large aromatic ring with a propensity to be buried in the interior of a hydrophobic core, to a much smaller and polar residue. Thus, the serine mutation at position 54 is predicted to have a destabilizing effect on the major wing.

To test this prediction, we investigated the structure and stability of the free SRY

HMG domain using intrinsic tryptophan fluorescence (a probe of quenching because of side-chain desolvation in the major wing (Phillips et al. 2011)). Stabilities of the native and variant domains were determined by guanidine-induced protein denaturation as probed at 4°C by intrinsic TRP fluorescence (at 390 nm emission after excitation at 270 nm) to measure the exposure of internal Trp residues and thus stability of the major wing. The clinical mutation F54S was found to have a two-fold destabilizing effect on the HMG domain. Application of a two-state model validated estimates of a reduced ∆Gu of 2.5 ± 0.09 kcal/mol compared to 4.3 ± 0.14 kcal/mol for the wild-type HMG domain (Fig. 3.3A).

Intrinsic tryptophan fluorescence spectra (monitored from 300 nm-450 nm after excitation at 270 nm) of the HMG box domain under native conditions, probed for the environment of the tryptophan residues. An increase in the emission signal and the red

119

shift of the emission maxima (broken line in Fig. 3.3B) for the F109S variant indicates that the tryptophan residues are more exposed and that the microenvironment around these residues are structurally different compared to that of the wild-type SRY.

Intrinsic TRP fluorescence probes for the structure and stability of the microenvironment in the major wing of the HMG domain of SRY. However, the

F54S mutation also displayed perturbation in the overall global structure compared to the wild type domain as determined by circular dichroism (CD) at a the helix sensitive wavelength of 222 nm (downward arrow in Fig. 3.3C).

The serine mutation in HMG box (F54S) does not affect specific DNA binding and bending. SRY-DNA binding was investigated by steady-state FRET studies of a

15-bp DNA duplex (5′-TCGGTGATTGTTCAG-3′; target site in boldface) as described previously (Phillips et al. 2006). Steady-state FRET efficiency was measured by changes between FAM (donor) and TAMRA (acceptor) flexibly linked to respective 5′-ends of the DNA. Changes in the distance between respective 5′-ends of the upper and lower strands upon protein binding are reflected in the FRET efficiencies. Equimolar solutions (1 µM) of the DNA and SRY domain were prepared as described (Phillips et al. 2006). Binding of the wild-type SRY HMG box to this

120

donor/acceptor-labeled DNA probe led to an increase in FRET efficiency at 15°C.

Similar increases in FRET efficiency were observed with binding of the F54S SRY

HMG box (Fig. 3.3D), indicating that under these conditions the variant is capable of binding and bending a consensus SRY target site with native-like properties.

Fluorescence emission spectra from 500-650 nm were recorded after excitation at 490 nm for both the steady state FRET measurements (Fig 3.3D) and the determinations of equilibrium dissociation constants. Dissociation constants were determined by measuring FRET changes in the donor/acceptor-labeled DNA probe at a fixed concentration (25 nm) with increasing SRY concentrations, from 1 nM to 2 μM at 15

°C and 37 oC. Data were fit to a single-site ligand-binding equation as described

o previously and yielded Kd values of 14±3 nm and 15±2 nm (wild-type domain 15 C and 37 oC respectively); 12±2 nM and 23±6 nM for variant F109S at 15 oC and 37 oC respectively. Although at physiological temperature there was an apparent 2-fold

decreased affinity for specific DNA binding (Kd 23±6 nM for F54S vs. 15±2 nM for wild-type), these values are not statistically significant. These results suggest that

DNA binding and bending are unaffected by the F54S mutation at physiological temperature.

DNA binding prevents destabilizing of the HMG box caused by F54S mutation.

121

The structure and stability of the free and DNA bound F54S domain was investigated by far-ultraviolet Circular dichroism (CD). The CD spectra of the DNA-bound domain shows a transition to a more alpha-helical content and is restored to that of the wild type spectra in Fig 3.3C as indicated by the downward arrow in Fig. 3.3E .

Circular dichroism (CD) was used to evaluate the thermal stabilities of the

DNA-bound native and mutant HMG domains by monitoring perturbations in

α-helical content at 222 nm. Although the DNA-complex F54S domain displayed decreased thermal stability at higher temperatures (> 45 oC ) compared to that of wild-type domain, they exhibited similar thermal stabilities at the physiological temperature of 37oC (Fig. 3.3F). It is not surprising that PheSer mutation in a hydrophobic core destabilizes the HMG box domain causing local and global structural changes. Interestingly, this mutation can lead to two development outcomes as an inherited mutation, meaning that its DNA binding and bending capabilities are unlikely to be altered drastically. Here, we show that indeed, at both low temperature (which can stabilize proteins) and at physiological temperature (37 oC) the F54S isolated domain is destabilized. However, the DNA interacting activities are not drastically altered in all other assays for specific DNA binding and bending under physiological conditions. Therefore the underlying mechanism for the cause of abnormal development does not reside in the inherit DNA interacting

122

functions of the mutant SRY.

These studies demonstrate that inherited mutation F109S (F54S in HMG box) and wt share similar DNA interacting parameters; affinity, relative bend angle, and complex thermo-stability at experimental low and physiological temperatures.

F109S and L163X SRY variants reduced the occupancies of the testis-specific enhancer core region of Sox9 gene. Chromatin immunoprecipitation (ChIP) studies were undertaken to evaluate SRY binding sites in the testis-specific enhancer core region (TESCO) of Sox9 (Fig. 3.4 with primer sets defined in Appendix V; (Sekido and Lovell-Badge 2008)). Relative enrichment of occupancy (Fig. 3.4A) was determined by measuring ChIP band intensities (Fig. 3.4B). TESCO occupancy of

F109S and L163X SRY variants were significantly reduced compared to that of wild-type (Fig. 3.4A histogram and lane 2, 3 in gel panel). Control studies of inactive

I68A SRY variant demonstrated an absence of binding. The enhancer-specific ChIP band intensities for F109S and L163X SRY in the different transfection conditions with various SRY-encoding plasmid amounts (none, (1X) and 50X dilutions; see material and methods) were both reduced to ca. half of the wild-type level (lanes 2 and 3 in both panels of Fig. 3.4B), which were verified quantitatively (histogram in

3.4A).

123

F109 and L163X Sry variants displayed attenuated gene regulatory activities.

Gene regulatory activity of SRY and variants were evaluated by SRY-directed transcriptional activation of the endogenous Sox9 gene in a male rat embryonic gonadal cell line following transient transfection of SRY or variant SRY alleles

(Phillips et al. 2011). CH34 cells were tranfected with various SRY-encoded plasmids at a transfection efficiency of 40.1(±2.1) percent as monitored by co-transfection of a control plasmid encoding green fluorescent protein (GFP). Transient transfection of wild-type SRY under standard conditions (1g of plasmid DNA/per million cells) exhibited an eight-fold increased accumulation of Sox9 mRNA by relative to the control empty vector or the inactive variant I68A (black bar in Fig. 3.4C). The extent of transcriptional activation decreased when the SRY-encoded plasmids were successively diluted with the empty vector with same backbone in order to maintain constant amount transfected DNA. The highest dilution was to a final dilution of 0.02

g SRY plasmid and 0.98 g empty vector per million cells (i.e. 50-fold dilution; white bar in Fig. 3.4C). Such dilutions demonstrated biological trends and provided a control for potential artifacts caused by over-expression.

The investigation of the variant SRY constructs (F109S and L163X) demonstrated that (i) the substitution F109S contains a native-like functional HMG

124

box in Y1H assays but exhibited half the transcriptional activity at each dilution tested

(Fig. 3.4C; black, gray, and white), and (ii) the substitution L163X contains a wild-type HMG box but shared similar Sox9 mRNA accumulation with F109S, (half) at all dilution conditions (Fig. 3.4D). The possibility of accelerated breakdown of the

SRY variants in CH34 cell model as an underlying mechanism for the observed attenuated transcriptional activities was evaluated as described below.

Clinical SRY variants displayed accelerated cellular protein turnover. The relative protein accumulation of HA-tagged SRY variant constructs (transfected without dilution) was evaluated following treatment with the translation inhibitor, cycloheximide. The Western-blot intensities measured after five hours using anti-HA antiserum demonstrated that F109S and L163X variants were more susceptible to degradation than the wild-type (gel graph in Fig. 3.5A). The quantitative plot of degradation (Fig. 3.5B) verified that the two SRY variants exhibited reduced protein accumulation in the pre-sertoli cell model. These findings suggest that the faster degradation of SRY could be the underlying cause for the reduced SRY-Sox9 central regulation. This possibility was evaluated by a “rescue approach” with the inclusion of the proteasome inhibitor, MG132 in the transcriptional activation assays.

125

Inhibition of proteosomal degradation of the SRY variants restored regulation activities. Based on the results observed in Fig 3.5C, MG132 treatment led to complete restoration of Sox9 mRNA accumulation of F109S and L163X SRY to the levels of wild type SRY (right-hand bars in Fig. 3.5C); that of untreated SRY variants are provided in the left of Figure 3.5C. At all concentrations of the transfected

SRY-encoded plasmid, the same biological trends of restoration were observed (black, gray, and white bars in Fig. 3.5C). Taken together, the ability of MG132 to restore native-like activities to the two inherited mutations in SRY (L163X, containing wild type HMG box domain with native specific DNA-binding properties, and F109S) demonstrate that the reduced SRY-Sox9 central regulations of the variants are associated with their faster proteosomal degradation.

The attenuation of SRY-Sox9 regulation by the truncated SRY variants is not location-specific. The reduced SRY-Sox9 regulation seen in the variant mutation outside the HMG box, L163X, has been attributed to its accelerated proteosomal degradation (Fig. 3.5A and B). This result implies that there is a potential relationship between the C-terminal truncation of the SRY protein and the protein stability, which mediates transcriptional function. To test this hypothesis, a series of truncated SRY variants outside of the HMG-box were constructed and employed to evaluate

126

proteosomal protein turnover and their potential ability to modulate SRY-Sox9 regulation (Fig. 3.6A; fusion of the PDZ domain serves as a control). Cycloheximide studies indicated that truncated SRY variants, independent of the presence or absence of the PDZ domain, exhibited accelerated degradation compared to wild-type SRY in the presertoli CH34 cell model (Fig. 3.6B). The presence of the PDZ sequence

(YSHWTKL) in the construct had no significant effect on the degradation rate (Fig.

3.6B; filled and empty symbols) nor did the length of the truncated variants. The

SRY-Sox9 regulation of these variants monitored by qPCR was partially reduced independent of the fused PDZ domain (Fig. 3.6C; black and white bars in the left cluster). Treatment with the proteosome inhibitor, MG132, completely restored the activation function of all of the variants (Fig. 3.6C; black and white bars in the right cluster). The “rescue” by MG132 indicates that the attenuation of Sox9 activation is associated with the proteosome-mediating degradation without location specificity.

Gene regulatory network in pre-Sertoli cell follows the same biological trend as the Swyer variants SRY-Sox9 regulation. The present studies focused on the central regulation of SRY-Sox9. However, besides the principal SRY target, Sox9 (Sekido and

Lovell-Badge 2008), qPCR was further undertaken to profile the downstream GRN

(Fig. 3.7A) of the XY gonadal ridge. Whereas transient expression of SRY did not

127

affect the mRNA accumulation associated with non-sex-related Sox genes and housekeeping genes (Fig. 3.7B), specific up-regulation of Sox9, fibroblast growth factor 9 (Fgf9) and prostaglandin D2 synthetase (Ptgd2) were observed in accord with their known roles in testicular differentiation ((Moniot et al. 2009); Fig. 3.7B).

No significant changes in mRNA accumulation were observed after transient transfection of an empty plasmid or a control plasmid expressing a stable SRY variant devoid of specific DNA-binding activity (I68A; (Weiss et al. 1997). These results suggest that SRY-dependent activation of Sox9 in turn activates a ramifying GRN, and provides evidence that SRY functions as a transcriptional activator (Sekido 2010).

128

Discussion

Organogenesis is an internal organs developmental process in animals, and testicular differentiation represents a model for this genetic program. The expression of SRY, the critical Y chromosome gene sufficient for male sex determination in eutherian mammals (Koopman et al. 1991), initiates a gene regulatory network that transforms the undifferentiated gonad into a testis. Absence or a defective SRY leads to the gonad differentiating into activity of intersexual phenotypes. Although the direct targets of

SRY in this gene regulatory network are incompletely characterized, an expression profile of male-specific differentiation gene pattern from rodent pre-Sertoli gonadal cells have been defined, and more significantly, quantified in the present study. Such regulation is the central dogma to testicular differentiation (Sekido et al. 2004; Sekido and Lovell-Badge 2008)), and involves an extensive network of candidate downstream genes, including Sox9, Fgf9, and Ptgds (Beverdam and Koopman 2006).

The direct binding of SRY in a Sox9 enhancer element have been identified by ChIP and shown to be critical for its tissue- and stage-specific transcriptional regulation

(Koopman 2001), which was validated and correlated with the Sry occupancy of the testis-specific enhancer core region (TESCO) and Sox9 gene activation. The crucial regulatory linkage between SRY and SOX9 provides a unified framework for understanding the mechanism behind Swyer outcomes and the diverse human genetic

129

syndromes (Knower et al. 2011; Phillips et al. 2011).

Naturally occurring variant SRY alleles associated with 46, XY pure gonadal dysgenesis (Swyer’s Syndrome, born externally as females but with nonfunctional streak gonads instead of ovaries or testes (Berta et al. 1990; Knower et al. 2003)), provide an important opportunity for investigating the sex-specific GRN and its function in gonadogenesis. The present study has focused on variants F109S and

L163X, containing HMG boxes that exhibit native-like transcriptional function in

Y1H screening, indicating that DNA binding is unaffected. These variants are inherited as sterile XY daughters from their fertile fathers. A possible insight into the cause of the Swyer’s Syndrome manifested by these two mutations is provided by the observation of a twofold reduction in male-pattern Sox9 transcription and near twofold reductions in signature testis developmental genes in male-specific GRN. The inherited variants of SRY are poised at a critical threshold of activity in regulation

(Nykter et al. 2008), highlighting the narrow margin of “decision making” by a sex-specific GRN in the differentiating gonadal ridge. The twofold perturbations in

SRY-SOX9 regulation uncovered in this study are functionally analogous to a syndrome of autosomal transcription factor (TF) haploinsufficiency (Seidman and

Seidman 2002). Specific haploinsufficiency phenomena have been observed resulting in disorders of gonadogenesis. For example, heterozygous nonsense and missense

130

mutations in SF1 display a phenotype of 46, XY pure gonadal dysgenesis, and the mutations in SOX9 result in campomelic dysplasia, wherein abnormalities of bone coincide (in XY patients) with male, intersex, or female somatic phenotypes (Baek and Kim 2011).

Surprisingly, the HMG box of clinical mutations F109S and L163X SRY retain their native transcriptional activation properties in yeast model (Fig. 3.2).

Serving as the signature domain critically mediating the DNA-interaction functions in

SRY, native-like SRY-Sox9 regulation in per-sertoli cell was expected with the native-like functional HMG boxes. Thus, what is the mechanism mediating the reduced regulatory function of these mutations? Biophysical investigations including

Trp-titration and CD spectroscopy demonstrated that the protein stability in the HMG box of F109S is perturbed. A protein surveillance system, the ubiquitin-proteasome system (Wong and Cuervo 2010), might detect and degrade this less stable F109S mutation. The reduced protein accumulation of mutation F109S associated with the accelerated degradation is supported by control immunoblotting analysis (data not shown) and cycloheximide chase assays, and hence affects the F109S SRY-Sox9 regulation. Similarly, mutation L163X exhibited reduced protein accumulation in the

CH34 pre-sertoli cell model. Because SRY is an intronless gene, nonsense mediated mRNA decay might not be triggered (Nott et al. 2004; Chang et al. 2007). The

131

cycloheximide chase study unmasked the L163X protein is rapidly degraded. This phenomenon might be the mechanism behind the similar twofold reduction Sox9 activation that was observed with F109S.

To probe their gene-regulatory functions, which are associated with proteasome-dependent early degradation, we investigated the intrinsic gene-regulatory activity of these variants following “correction” of their accelerated degradation by proteosome inhibition, MG132. Under this condition the wild-type and variant proteins led to complete rescue of Sox9 mRNA accumulation. Although transcriptional activation of Sox9 is increased under conditions of over-expression, successive plasmid dilution showed a constant twofold defect in Sox9 regulation in both mutations. This result implies the threshold for SRY-SOX9 regulation to determine the testis differentiation and gonadal dysgenesis.

Degradation, not position in SRY affects the regulation in male-specific GRN.

The gene regulatory studies monitored by Sox9 expression from mutations of nature

(F109S and L163X) and series of designed mutations (Q179X and S151X with and without PDZ domain) demonstrated that the rapid degradation of all cases (position independent) is associated with the attenuation of the SRY-Sox9 central regulation in testis differentiation regulatory network. The control constructs with PDZ domains are

132

functionally indistinguishable from related PDZ-absent constructs. Their Sox9 expression functions were able to be “rescued” by the treatment with proteasome inhibitor MG132, which unmasked that the degradation is crucial. Taken this result together with the clinical cases partially (50%) retaining transcriptional activation function (F109S and L163X SRY), associated with the observation of related clinical

Swyer phenotype, suggests that there is a threshold for gonadogenesis outcomes regulated by SRY-SOX9 regulatory process regulated by the protein level of SRY.

Rescue of this proteosomal degradation restores SRY-mediating transcriptional function completely supports our hypothesis that it is degradation and not a loss of

HMG box function (specifically, the DNA-binding) that is associated with the attenuation in Sox9 activation, and hence affects the downstream GRN. A model is provided by intersex mouse phenotypes associated with the incompatibility of the divergent Y chromosomes of Mus musculus domesticus and Mus domesticus poschiavinus (YDOM and YPOS) in C57BL/6J strains (Nagamine et al. 1987). The inherited Swyer syndrome resembles Y-chromosome incompatibility among mouse strains in which Y chromosomes bearing Sry alleles derived from diverse mouse strains can cause abnormal gonad development (Albrecht et al. 2003). Such phenotypes could not be correlated with changes in Sry sequence (Albrecht et al. 2003) but instead depend on the extent and timing of Sry expression (Bullejos and Koopman

133

2005; Wilhelm et al. 2009). Taken together, changes in the abundance of SRY protein, in mice and humans, in the differentiating gonadal ridge were associated with developmental abnormalities. Such genetic-background dependence strongly suggests that the SRY-SOX9-directed transcriptional program lies close to a threshold of function.

A threshold of SRY-initiated regulation in gonadogenesis and the enigma of inherited sex reversal. This study on the inherited mutations showing a twofold reduction in SRY-Sox9 regulation challenges the paradigm (Sekido and Lovell-Badge

2009): clinical mutations mostly were reported to block detectable specific

DNA-binding activity (Harley et al. 1992). Inherited F109S and L163X SRY mutations poised the father-daughter paradox in a reverse way: although the active

SRY HMG boxes of them could– within a DNA-centered transcriptional paradigm

(Sekido and Lovell-Badge 2009) –explain for the sex of the fertile father, what of the daughter with the Swyer phenotype?

The present study uncovers the answer clearly: twofold attenuation in the central regulation of male-specific GRN. Remarkably, the extent of SRY-dependent transcriptional activation of its major downstream target gene (Sox9) were reduced by ca. twofold in all three independent inherited cases (F109S, L163X, and V60L,

134

described in ref (Phillips et al. 2011) and chapter 2). Unlike de novo mutations I68T

(bearing a non-functional cantilever that affects DNA binding, (Haqq et al. 1994;

Peters et al. 1995)) and R133W (defective in nuclear localization (Li et al. 2001)) giving rise to negligible activation of SRY-Sox9 regulation are consistent with the absolute Swyer phenotypes, the twofold reduction of biological activity at the inherited variants rationalize the widely divergent phenotypes of both father and daughter because the genetic switch (the SRY-SOX9 regulation) is poised at the threshold of function.

It is possible that stochastic changes in gene expression or the influence of autosomal polymorphisms influence the divergent developmental fates of the father and daughter. For example, our previous study, V60A, was found to differentiate to ovotestes and not pure gonadal dysgenesis, which is different from the phenotype of

V60L (Hiort et al. 1995; Phillips et al. 2011). Twofold changes in the biochemical properties of proteins may seem subtle in vitro but can have profound biological consequences. This phenomenon demonstrates that lack of robustness in nascent

Sertoli-cell specification seems to violate Waddington’s Principle: that fundamental developmental pathways are canalized, at least in their upstream steps, and so robust to genotypic variation and environmental fluctuations (Waddington 1959; Masel and

Siegal 2009). This phenotypic variability suggests that in humans the twofold

135

transcriptional threshold characteristic of SRY extends to its immediate downstream target, which is supported by the signature downstream factors of gonadogenesis

GRN monitored by q-PCR. Together, these clinical entities and the inherited Swyer phenotypical incompatibility among mouse strains highlight the anomalous non-robustness of sexual dimorphism at multiple steps in the developmental program.

136

Materials and Methods

Plasmids. Bacterial plasmids expressing human SRY HMG boxes were constructed as described (Phillips et al. 2011). For the yeast-1-hybrid system, DNA encoding the

HMG box of human SRY was subcloned into plasmid pGAD-T7 encoding a fusion protein containing an N-terminal nuclear localization signal (NLS), central activation domain (AD) derived from GAL4, and an SRY HMG box domain (Phillips et al.

2011). Mammalian CMV promoter vector pCMX, kindly provided by Dr. H.-Y. Kao

(Kao et al. 2001), expressing SRY variants were constructed as previously described

(Phillips et al. 2011). Mutations in SRY were introduced using QuikChangeTM

(Stratagene). The encoded sequences contain a triplicate hemagglutinin (HA) tag following the initiator methionine. Constructions were verified by DNA sequencing.

Y1H Reporter screenings. The SRY triplicate consensus binding sites within the plasmid pLacZi (bold; 5’-AATTCGCAATTGTTATTGTTATTGTT-3’) were constructed as described (Phillips et al. 2011), and this site lies upstream of lacZ encoding -galactosidase. Reporter strains bearing integrated SRY target or non-target sites were tested (Phillips et al. 2011). Colonies were grown in minimal medium under the same selection. Extent of SRY-dependent expression of -galactosidase was evaluated by quantitative assay of enzyme activity in liquid culture using

137

ortho-nitrophenyl-β-galactoside (ONPG) as described by the vendor (Clontech

Laboratories, Inc). Experiments were performed in triplicate.

Mammalian Cell Culture. Rodent cell line CH34 (Haqq et al. 1994), employed for studies of the SRY-regulating gene regulatory network, was kindly provided by T. R.

Clarke and P. K. Donahoe (Massachusetts General Hospital, Boston, MA). Cells were cultured in Dulbecco’s modified Eagle medium containing 5% heat-inactivated fetal

o bovine serum at 37 C under 5% CO2. In the specific cases of proteasome-inhibitor studies, transfected cells were maintained for 24 h in serum-free conditions and then treated with MG132 for 6 h followed by 18 h incubation in 5% serum-containing medium.

Transient Transfection. Transfections were performed using Fugene HD (Hoffmann

LaRoche) with protocols from the vendor: After 24 h in serum-free medium, cells were recovered using fresh DMEM medium containing 5% heat-inactivated fetal bovine serum. Transfection efficiencies were determined by the ratio of GFP positive cells to untransfected cells following co-transfection with pCMX-SRY expression was monitored in triplicate by Western blotting in relation to -tubulin.

138

Cycloheximide-chase assay. 24-h post transfection cells were split evenly into 6-well plates and treated with the translation inhibitor cycloheximide (final concentration is

20 mg/mL) in the regular medium for the indicated times; cells were then lysed by mammalian lysis buffer (Hoffmann LaRoche). Normalized cell lysates with total protein were subjected to 12% SDS-PAGE detected by Western blot using anti-HA antiserum (Sigma-Aldrich), and anti--tubulin as a loading control. Quantification was performed by Image J program. Experiments were performed by triplicate.

Sox9 activation assay. SRY-mediated transcriptional activation of Sox9 and other endogenous CH34 genes was measured in triplicate by quantitative real-time-Q-rtPCR (qPCR) as described (Phillips et al. 2011). Cellular RNA was extracted using RNeasy (Qiagen).

139

Footnotes

Acknowledgements.We thank Prof. P. K. Donahoe for cell line CH34 and encouragement and H.-Y. Kao for the pCMX plasmids. S. Jeong and P. Janki for assistance with nuclear localization studies; T. Feng, B. Li, and R. Singh for participation in early stages of this work; and P. DeHaseth, H.-Y. Kao, D. Samols, and

P. Sequeira for advice. MAW thanks B. Baker, F.A. Jenkins, Jr., P. Koopman, R.

Lovell-Badge, R. Sekido, and D. Wilhelm for discussion. This work, a contribution from the Cleveland Center for Membrane & Structural Biology, was supported in part by a grant to MAW from the National Institutes of Health (GM080505).

140

Fig. 3.1 Domain organization of SRY and summary of clinical mutations.

(A) Human SRY sequence depicting clinical mutations. Residues 56-141 (HMG–box consensus number 1-86 in parentheses) (dark blue) and basic tail (bt; light blue).

Other domains: PKA phosphorylation sites (brown) and proposed PDZ-binding motif

(yellow). De novo mutations (green triangles), inherited mutations (filled red triangles), and mosaic fathers (open red triangles) are shown. Red asterisks indicate

F109S (HMG-box consensus residue 54) and L163X (outside of the HMG box). (B)

Family trees pertaining to F109S and L163X SRY. Symbols are defined at right. (C)

Stereo pair showing ribbon model of SRY HMG box with DNA contacts; the domain 141

docks within minor groove of bent DNA site (space-filling representation).

Highlighted in red in the ribbon is the site of clinical mutation F109S which is not

DNA-contacting (position 54 in the HMG box; asterisk). F109 packs within the major wing core depicted with the side chains of W98 and L101 (green). The basic tail (bt) containing additional de novo mutations is highlighted in light blue. The DNA atoms are shown in dark gray (phosphodiester linkages), medium gray (deoxyribose moieties) and light gray (base pairs).

142

Fig. 3.2 Design and SRY-dependent -galactosidase activity of Y1H-screening system.

(A) Triplet adjoining SRY target sites were integrated within the Y1H reporter

(5’-ATTGTT-3’ and complement; red, black, and blue) upstream of lacZ gene (black rectangle) and downstream of selectable marker ura3 (gray rectangle). (B) Schematic of plasmid pGAD-T7 encoding a fusion protein under the control of the ADH1 promoter (red), consisting of an nuclear localization signal (NLS; blue), the transcriptional activation domain (AD; green) of yeast regulator Gal4, and the SRY

HMG box (black). (C) Quantitative analysis of -galactosidase activity in liquid culture in Miller units using the Yeast-1-hybrid system as described.

Ortho-nitrophenyl--galactosidase values observed with the F54S variant (residue

109 in full-length SRY number) fusion protein (red) were statistically indistinguishable from those with the wild-type fusion protein (black). Mutation I13A

143

(position 68 in full-length SRY) serves as negative control due to the loss of specific

DNA-binding property. Assays were performed in triplicate; error bars represent standard deviations. Values were normalized according to the level of protein expression. This data was obtained by Joe Racca as part of his collaboration on this project.

144

Fig. 3.3 Biophysical studies of the free and DNA bound F54S clinical mutant.

(A) A Serine mutation in the major wing of the HMG box causes destabilization of the protein domain by approximately 2 fold; ∆G for wt (black) is 4.3±0.14 kcal/mol compared to mutant (red) 2.5±0.09 kcal/mol. (B) Under native conditions the serine mutation causes structural alteration of the major wing, as determined by fluorescence probing for intrinsic tryptophan. Specifically the microenvironment around W70 and W98 is altered as indicated by the increase in intensity (red) and a shift to the right (red shift) compared to wt (black). (C) The defect in the major wing as probed by intrinsic tryptophan fluorescence in (B) affects the global structural integrity of the HMG box monitored by circular dichroism for α-helical content. At physiological temperature there is a distinct loss of a-helical signal (broken arrow indicating 222nm signal for α-helix) for the mutant (red) compared to wt (black) for unbound domain. However, DNA binding and bending appears to be unaffected for 145

the mutant domain. (D) Steady-state FRET experiments show that specific DNA bending is retained (closed red circles compared to open red circles for a double fluorescently-labeled DNA substrate) indicated by the decrease in the emission of the donor (FAM) fluorophore in the presence of the mutant domain. Whereas for each of the single-labeled DNA substrates, donor only green circles free (open) and in the presence of protein (closed) and acceptor only in purple circles free (open) and complex (closed), display minimal decrease in FRET transfer. (E) In the presence of

DNA the F54S mutant domain regains α-helical structure at physiological temperature. DNA bound (red closed circles) F54S mutant has different spectral characteristics determined by CD compared to the free mutant domain (open red circles). The solid arrow indicates a restoration of α-helical signal monitored by CD.

(F) Mutant and wt HMG box domains have similar thermo-stability at physiological temperature. The observed melting temperature (Tm) for wt (black circles) is 60oC

(solid black line) and mutant (red circles) is 50 oC (broken red line). The thermo-stability of the mutant complex is only affected above physiological temperature. This data was obtained by Joe Racca as part of his collaboration on this project.

146

Fig. 3.4 SRY-dependent transcriptional activation of Sox9 in pre-Sertoli cell model.

(A-B) Histogram and gel showing results of ChIP assays of variants without any rescue treatment. Clinical mutations F109S and L163X exhibit partial impaired

TESCO region occupancy under all conditions (none (1X) or 50X dilution in transfected SRY-encoded plasmids). Mutation I68A which lacks specific DNA binding (King and Weiss 1993) served as negative control. Lanes 1-4 in the gel panel represent the SRY wt and variant from left to right in the histogram. (C) Sox9 gene activation (monitored by Sox9 mRNA accumulation) as a function of dose of transfected plasmid SRY DNA: 1 g (black), 0.2 g (dark gray), and 0.02 g (white).

The total concentration of transfected DNA was held constant by filling in empty pCMX plasmids. Horizontal brackets designate statistical comparisons: (* or **)

147

Wilcox p-values <0.05 or 0.01 whereas “ns” indicates p-values > 0.05. (D) Relative

Sox9 activation regulated by SRY variants compared to wild-type SRY with 1X, 5X, and 50X-dilutions of transfected F109S (blue) and L163X (green) SRY plasmid conditions. Dash and dot lines indicate 100% and 50% of Sox9 mRNA accumulation of wild-type SRY, respectively. Statistical comparisons are defined in panel A.

148

Fig. 3.5 Cellular turnover of SRY affects the regulation of SRY-Sox9 central axis.

(A and B) Cycloheximide assay. Proteolysis is enhanced by mutation F109S and

L163X compared to wild-type SRY. Immunoblotting is shown on (A). The

HA-tagged SRY variants were blotted using anti-HA antiserum. Tubulin served as loading control. (B) Protein accumulation plot with time depicts the more rapid degradation of SRY variants following inhibition of continued protein synthesis by cycloheximide. (C) SRY-dependent Sox9 gene activation with varying doses of transfected wt SRY and variants. qPCR data were obtained in absence (left) or 149

presence (right) of MG132. A twofold deficit in transcriptional activation is displayed by the two SRY variants at different transfected plasmid dilutions. The deficit could be completely restored by MG132 rescue. Although it is not shown, the Sox9 mRNA accumulation exhibited no significant difference in wild type SRY in the presence or absence of MG132. The SRY mutant, I68A, lacking specific DNA binding, served as a negative control. The statistical comparisons: (**) Wilcox p-values <0.01 whereas

“ns” indicates p-values > 0.05.

150

Fig. 3.6 Truncated SRY mutations accelerated proteosomal degradation, resulting in reduced Sox9 activation.

(A) Schematic diagram of wild-type and truncated SRY constructs including the clinical mutation, L163X and two designed truncated mutations: Q179X and S151X where X represents the differentially truncated C-terminal residues of SRY Light gray boxes indicate the PDZ domains. Mutations with or without PDZ fusion domains are highlighted by the solid and dashed-lined boxes respectively. (B) Proteosomal 151

degradation assay following cycloheximide treatment. The SRY variants are more susceptible to proteolysis compared to wild-type SRY following treatment with the protein synthesis inhibitor, cycloheximide. The protein degradation data for SRY variants except wt SRY show statistics insignificantly (p-values between 0.15-0.3).

(C) Relative SRY dependent Sox9 gene activation by the various truncated SRY mutations. qPCR data were obtained in the absence or presence of the proteosome inhibitor, MG132. Dash and dot lines represent 100% and 50% of Sox9 mRNA accumulation in comparison to that of wild-type SRY. Results in the absence or presence of fused PDZ to the SRY variants are highlighted with black and white, respectively. The condition used was a 50X-dilution of transfected SRY variant-encoded plasmids.

152

Fig. 3.7 SRY-regulated testicular gene-regulatory network (GRN) and transcriptional activation of Sox9.

(A) The central regulatory axis of SRYSOX9 (red box) with genetic inputs (box at left) and outputs to a male-specific GRN (box at right) functions to inhibit granulosa-cell fate (solid ; Wnt pathway), and Müllerian regression (dashed ;

MIS(AMH)). Red curved arrow indicates SOX8/9-mediated feedback maintenance of

SOX9 expression once the SRYSOX9 regulation is initiated. Abbreviations: FGF9, fibroblast growth factor 9; GATA4, GATA binding protein 4; LHX9, LIM homeobox

9; LIM1, homeobox protein Lhx 1; MIS (AMH), Müllerian Inhibiting Substance

(Anti-Müllerian Hormone); PTGDS, Prostaglandin D2 synthase; WT1, Wilm’s tumor

153

1; Wnt, wingless-type. The blue color highlights the selected genes that are the subject of investigation as described in (B). (B) Transcriptional assays of selected genes activated by SRY variants in rat embroyonic gonadal cell line. RT-Q-rt-PCR was employed to measure mRNA abundances of Sox family members (left), candidate male GRN-related genes (middle), and sex-unrelated housekeeping genes (right). qPCR was analyzed following transient transfection of SRY variant expression plasmids, empty vector, or a control plasmid with an inactive SRY variant (I68A).

Left: Fold mRNA accumulation of Sox family genes, including Sox8 and Sox9 previously implicated in the program of SRY-mediated GRN. Middle: Fold mRNA accumulation of non-Sox sex-related factors. Results show that CH34 cells express low endogenous levels of Sry, Fgf9, Ptgds, and Wnt5 gene expression. Right: Fold mRNA accumulation of sex-unrelated genes; these genes were not affected by expression of transfected SRY. Abbreviations: as in panel A; statistical analyses:

*Wilcoxon test, p<0.05, indicating the significant comparison between wt and F109S

SRY transfections.

154

Chapter IV

A microsatellite-encoded domain in rodent Sry functions as a genetic capacitor to enable the rapid evolution of biological novelty SRY

Introduction

Protein innovation can emerge through gradual accumulation of mutations

(Soskine and Tawfik 2010), rearrangement of DNA segments (Bornberg-Bauer et al.

2005), alternative RNA splicing (Keren et al. 2010), and RNA editing (Reenan 2005).

Exon shuffling among eukaryotic genes and pseudogenes, for example, has provided combinatorial opportunities for protein diversity within a given taxonomy of folds

(Bogarad and Deem 1999). Exploiting similar principles, somatic recombination (in conjunction with the mutational spectra of hypervariable regions) underlies the extraordinary repertoire of the adaptive immune system (Cooper and Alder 2006). The present study focuses on clade-specific divergence of a transcription factor(TF)

(Koopman 1995) in association with insertion of a CAG triplet repeat (Coward et al.

1994; Dubin and Ostrer 1994). Can microsatellite dynamics (Schlötterer 2000) in itself influence the pace and direction of protein evolution? A model is provided by

Sry, an architectural TF in eutherian mammals encoded by the sex-determining region

155

of the Y chromosome (Koopman et al. 1991; Dubin and Ostrer 1994). Our results rationalize rapid changes in the mechanism and fate of a developmental switch in the radiation of rodent superfamily Muroidea (Fig. S4.1).

Sry is a sequence-specific DNA-binding protein containing a high-mobility-group (HMG) box, a conserved motif of DNA bending (Lovell-Badge et al. 2002). In the differentiating gonadal ridge Sry activates Sox9, an autosomal gene that in turn regulates male gonadogenesis (Sekido and Lovell-Badge 2008). Binding of murine Sry (mSry) to the testis-specific core enhancer of Sox9 (TESCO; (Sekido and Lovell-Badge 2008) thus activates a Sertoli-cell-specific gene-regulatory network

(GRN) that mediates programs of cell-cell communication, migration and differentiation leading to formation of the fetal testis (Lovell-Badge et al. 2002). The

Sry HMG box provides the signature motif of an extensive family of cognate TFs

(designated Sox; Sry-related HMG box) with broad functions in metazoan development and tissue-specific gene regulation (Guth and Wegner 2008). Sry itself is thought to have arisen by duplication of Sox3, an X-linked member of this family

(Katoh and Miyata 1999). Whereas Sox3 is highly conserved among mammals, the evolution of Sry has been rapid (Whitfield et al. 1993), particularly within Rodentia

(Pamilo and O'Neill 1997). As a seeming paradox, some members of Muroidea lack

Sry (such as spiny rats Tokudaia osimensis and T. tokunoshimensis and vole Ellobius

156

lutescens), leading to new (and uncharacterized)mechanisms of sex determination

(Soullier et al. 1998; Just et al. 2007). We thus sought to investigate variation in the biochemical properties of Sry as a model Y-encoded protein undergoing rapid change.

Our studies focused on mSry (derived from Mus musculus domesticus) and human

SRY (hSRY); their respective domain organizations are shown in Figure 4.1 in relation to the structure of the HMG box (Murphy et al. 2001). Whereas hSRY (like many non-rodent Sry alleles) contains an HMG box embedded between N- and

C-terminal domains (NTD/CTD), murine and rat Sry lack an NTD and contain a CTD extended by a glutamine-rich domain (GRD; Fig. 4.1A) containing 3-20 poly-Gln blocks separated by His-rich spacers (consensus FHDHH). Encoded by a CAG microsatellite unique to the Y chromosomes of Muroidea, the GRD of mSry is required for its function as a transgene in XX mice (Bowles et al. 1999).

Our investigation of mSry builds on companion studies of mutations in hSRY associated with inherited sex reversal (preceding article in this issue; (Chen et al.

2013a). Whereas GRDs in other TFs flank conserved DNA-binding motifs without change in mutational clocks (Dunah et al. 2002), the HMG boxes of mSry and its orthologs in Muroidea exhibit greater sequence variation (with respect to both synonymous and non-synonymous base substitutions) than do Sry boxes in other mammalian orders (Tucker and Lundrigan 1993; Pontiggia et al. 1995). Our results

157

demonstrate that such variation is associated with (i) impaired nuclear export by a mechanism analogous to a clinical mutation in hSRY, (ii) biophysical perturbations of the mSry HMG box and (iii) impaired occupancy of TESCO sites (in the absence of the GRD) required in the gonadal ridge for Sox9 transcriptional activation.

Biochemical compensation is provided by the GRD functioning at a threshold number of poly-Gln blocks. We envisage that variation in rodent Sry—suppressed or unmasked at the protein level by an unstable CAG-encoded GRD (Liu and Wilson

2012)—has been a source of evolutionary innovation: an historical contingency of genomic dynamics leading to divergence of a master switch and even to its anomalous disappearance (Graves 2002; Just et al. 2007). The Sry GRD, functioning as a genetic capacitor (Rutherford and Lindquist 1998; Lindquist 2009), has fostered the rapid generation of biological novelty in the radiation of alleles of Muroidea.

158

Results

Rat embryonic pre-Sertoli cell line CH34 (Haqq et al. 1994; Haqq and Donahoe

1998) was employed as a platform to monitor the gene-regulatory activities of

N-terminal hemagglutinin-tagged (HA) Sry constructs following transient transfection

(Phillips et al. 2011). Transcriptional activation of endogenous Sox9 was probed by quantitative PCR (real-time-Q-rtPCR; qPCR) and chromatin immunoprecipitation

(ChIP) (Chen et al. 2013a). Despite their structural differences (Fig. 4.1), mSry and hSRY exhibit similar activities in these assays in accordance with the ability of either protein to induce testicular differentiation in transgenic XX mice (Koopman et al.

1991; Lovell-Badge et al. 2002). Consistent with transcriptional profiling of the differentiating XY gonadal ridge (Moniot et al. 2009; Barrionuevo and Scherer

2010),Sry-dependent activation ofSox9 is associated with the selective activation of downstream genes Sox8 and fibroblast growth factor 9 (Fgf9) (Fig. S4.2).Transient transfection of Sry constructs does not lead to changes in abundance of control Sox mRNAs uninvolved in testis determination or mRNAs encoding housekeeping genes.

Our companion studies of clinical hSRY variants highlighted the technical importance of dilution of the expression plasmid by an empty vector to avoid TF over-expression (Chen et al. 2013a). Whereas at high plasmid dose without dilution

(1g per well) mSry and hSRY exhibit indistinguishable activities (black bars in Fig.

159

4.2A), on 50-fold plasmid dilution the activity of mSry is ca. 20% higher than that of hSRY (white bars in Fig. 4.2A). Negative controls were provided in this assay by the empty vector and a variant hSRY (I68A; consensus position 13 of the human HMG box) unable to bind specific DNA sites (King and Weiss 1993; Weiss et al. 1997).

Equal expression of the mSry/hSRY constructs was routinely verified by anti-HA

Western blot; loading controls were provided by housekeeping protein -tubulin as described previously (Phillips et al. 2011).

Deletion Analysis. The mSry GRD in M. musculus domesticus contains 20 Gln-rich blocks separated by His-rich spacers (Fig. 4.2B). Stepwise C-terminal deletion

(constructs 1-6 in Fig. 4.2C) unmasked a threshold requirement for at least 3 blocks to maintain mSry-dependent Sox9 expression (Fig. 4.2D); 3-20Gln-rich blocks conferred similar activities and ChIP-based estimates of TESCO occupancy (Fig.

4.3A,B,D). The dependence of TESCO occupancy on a threshold GRD length in mSry stands in contrast to the robust occupancy of hSRY in the absence of a GRD (see companion study (Chen et al. 2013a)) but is in accordance with its necessary inclusion in transgenes able to induce testicular differentiation in XX mice (Bowles et al. 1999).

Biophysical Degeneration of Murine HMG Box. The mSry HMG box has diverged

160

relative to non-rodent Sry domains (Table S4.2) and is associated with less precise

DNA bending (Phillips et al. 2004). The murine domain also exhibits an anomalous sensitivity to chemical denaturation by guanidine-HCl (Fig. 4.4A). Similarly, its thermal stability is reduced by 3-5 ºC relative to the hSRY domain (inset in Fig. 4.4B).

In each case partial unfolding occurs at physiological temperatures as indicated by attenuated -helical CD features (spectra Fig. 4.4B). -Helical structure was in each case enhanced on specific DNA binding but to a more marked extent in the more stable hSRY complex (Fig. 4.4B-D).

Respective affinities of murine and human boxes (Kd) for a consensus DNA target site (5’-TCGGTGATTGTTCAG-3’ and complement; bold), as determined by equilibrium FRET-based titration (Phillips et al. 2011), are similar at 15 ºC (11.2(±3) nM (murine) and 14.5(±2) nM (human)) but different at 37 ºC (22(±7) nM (murine) and 14.2(±2) nM (human)). The similar affinities at 15 ºC mask compensating changes in rates of protein-DNA dissociation and (by inference) protein-DNA association (Fig. 4.4E; for experimental design see Fig S4.3).At this temperature the

3 lifetime of the mSry domain-DNA complex (6.6 x 10 ms, corresponding to koff

0.15(±0.002) s-1) is foreshortened relative to the hSRY domain-DNA complex (31.3 x

3 -1 10 ms; koff 0.032(±0.001) s ). At 37 ºC the lifetime of the murine complex is markedly reduced (Fig. 4.4F); only an upper limit could be estimated (< 0.7 x 103 ms)

161

3 -1 relative to hSRY (12.5 x 10 ms; koff = 0.08(±0.003) s ). The twofold reduction in affinity of mSry (relative to hSRY) at 37 ºC thus reflects insufficient compensation in rate of association. Previous studies of hSRY variants have suggested that the lifetime of the box-DNA complex correlates with transcriptional potency (Ukiyama et al. 2001).

mSry/hSRY Chimeric Constructs Probe for Non-HMG Box Function. “Swap” of murine and human boxes within hSRY (chimera 1; Fig. 4.5A) resulted in reduced

TESCO occupancy relative to hSRY or mSry (Fig. 4.3C). Impaired occupancy was not due to inefficient nuclear import as chimera 1 exhibits enhanced nuclear localization relative to hSRY with decreased frequency of cells with pancellular distribution (Fig. 4.5 B and C), a pattern similar to mSry and a clinical hSRY variant bearing a defective NES (I90M at consensus position 35 (Chen et al. 2013a)) (Fig.

4.5B and C).Inspection of the mSry box sequence revealed a non-conservative substitution (bold) in its putative NES (humanmouse:

IxxxLxxxxxMLIxxxLxxxxxSL). Substitution of the rodent-specific Ser by Met restored CRM1-mediated nuclear export (Fig. S4.4).Impaired nuclear export of mSry and I90M hSRY stands in contrast to the impaired import of a control hSRY variant bearing an NLS mutation (R62G; consensus position 7) as previously characterized

162

(Gontan et al. 2009). The distinct distributions of mSry and hSRY in CH34 cells are in accordance with (a) the exclusive nuclear localization of mSry in the differentiating murine XY gonadal ridge (Dubin and Ostrer 1994; Sekido and Lovell-Badge 2008) and (b) the partial pancellular distribution of hSRY in an aborted human XY specimen

(Poulat et al. 1995; Malki et al. 2010).

Like I90M hSRY(Chen et al. 2013a), chimera 1 exhibits twofold attenuation of

Sox9 activation (“SSS” bars in Fig. 5D), which is rescued by acidic substitutions within the NTD PKA site (hSRY residues 26-38; PALRRSSSFLCTE) in accordance with an NCS-dependent activating phosphorylation (“DDD” bars in Fig.

4.5D;(Desclozeaux et al. 1998; Chen et al. 2013a); such phosphorylation is documented in Fig. S4.5. Elimination of this site in hSRY and chimera 1 led to equal residual transcriptional-regulatory activities (“AAA” bars in Fig. 4.5D). Design of chimeras 2, 3, and 4is depicted in Figure 4.6 A, B, and C. The functional dependence of hSRY on NTD phosphorylation state (as probed by AAA and DDD substitutions) was eliminated by swap of CTDs, including the murine GRD (chimera 3; Fig. 4.6B and E). Native TESCO occupancy and Sox9 activation by a TF bearing the murine box (mSry) was likewise conferred by the mSry CTD (chimeras 2 and 3; Fig. 4.6A,

D-E and Fig. 3C). In the context of chimera 3 the murine CTD with GRD also compensates for an inherited mutation in the human box (Y127F in Fig. 4.6C;

163

consensus position 72) associated with sex reversal and partial reduction of specific

DNA binding (Jordan et al. 2002). Sox9 activation is impaired by this substitution in the context of hSRY but not chimera 4 (Fig. 4.6F).

Transgene-inspired Chimera Probe for Non-box Sex-reversal Mechanism.

Chimeric transgenes expressing hSRY or goat Sry (also lacking a GRD; (Pannetier et al. 2006) under the transcriptional control of mSry regulatory DNA sequences are able to direct testicular differentiation in XX mice (Koopman et al. 1991; Lovell-Badge et al. 2002). Analogous biological activity was observed on swap of the mSry HMG box by its proposed X-encoded ancestor Sox3or homolog Sox9 (Bergstrom et al. 2000)

(see Fig. S4.6). Chimeras 5-8 exploited these findings to demonstrate that, with the exception of swap of the murine box with hSRY (chimera 1 above), the homologous boxes function in the context of either hSRY (Fig. 4.7A) or human NTD-extended mSry (Fig. 4.7B). Whereas at high- or low plasmid dose the Sox9-related transcriptional activities of the NTD-extended chimeras were indistinguishable from wild-type mSry (i.e., irrespective of box sequence; Fig 4.7D), the hSRY-based chimeras exhibited inequivalent activities on plasmid dilution (Fig. 4. 7C) in rank order mSry box < hSRY box, Sox3 box < goat Sry box.

Control Cell Lines. To extend key findings to a human cellular milieu, additional 164

studies were conducted in male cell lines PC-3 (Kaighn et al. 1979) and NT2-D1

(Knower et al. 2007) (respectively derived from prostate- and testicular cancers).

Although endogenous SOX9 in these lines is less amenable to transcriptional activation by hSRY or mSry, the two factors in each case exhibit similar relative activities (Supplemental Fig. S4.7 and S4.8). To enable comparative studies of variants despite reduced assay sensitivity, relative activities were further evaluated in

NT2-D1 cells on co-transfection of an mSry/hSRY-responsive luciferase reporter (Fig.

S4.8C). The results confirm key findings of the above CH34-based studies with respect to deletion analysis of the mSry GRD and the transcriptional regulatory properties of chimera 1-based constructs (Fig. S4.8D), in particular effects of NES

“repair” in the murine box (SerMet at box position 45) and DDD-based phosphor-mimicry of an activated hSRY N-terminal PKA site.

165

Discussion

Divergence of biochemical mechanisms underlying cognate GRNs (Booth et al.

2010) highlights the complementary roles of chance and necessity in the evolution of biological novelty (Monod 1972). The present study has investigated the relationship between a contingent genomic event—insertion of a DNA microsatellite—and its consequences for protein evolution in the adaptive radiation of a clade. A model was provided by a Y-encoded TF under strong selection (Sry).

Interplay between microsatellite instability and protein divergence in Muroidea may underlie the emergence of three-component populations (XX females, XY females, and XY males) and non-Y-dependent mechanisms of male sex determination (Soullier et al. 1998; Graves 2002; Just et al. 2007).

Sry Domain Organization and Drift of the HMG Box. Lacking an N-terminal non-box domain (NTD), the divergent HMG box of mSry is extended by a C-terminal

GRD unique to Muroidea (Fig. 4.1 and S4.9; (Bowles et al. 1999). We employed chimeric and deletion constructs, corresponding in part to transgenes previously characterized in XX mice (Fig. S4.6), to investigate the inter-relation of these domains in a pre-Sertoli cell line (Haqq and Donahoe 1998).Our companion study exploited this line to dissect coupling between NCS and NTD phosphorylation in hSRY (Chen

166

et al. 2013a). Impaired coupling is associated with an inherited form of Swyer’s syndrome (46, XY pure gonadal dysgenesis; (Knower et al. 2011) due to a twofold reduction of hSRY-directed Sox9 expression (Chen et al. 2013a).

GRDs are well known among eukaryotic transcriptional activation domains

(TADs; (Atanesyan et al. 2012). Such low-complexity sequences are found in diverse TFs, including Sox proteins (Kasimiotis et al. 2000), Sp1, Kruppel-related factors, and the cyclic-AMP-responsive factor CREB family (Persengiev et al. 1995).

GRDs can form oligomers (Tobaben et al. 2003) and/or contact the basal transcriptional machinery (Atanesyan et al. 2012). The CAG-encoded domain of mSry was first identified as a potential TAD in a yeast model (Dubin and Ostrer 1994).

Its deletion within an mSry transgene blocks the ability of the construct to induce testicular differentiation in XX mice (Bowles et al. 1999). A survey of mammalian Sry alleles indicates that the CAG-microsatellite in Muroidea is associated with loss of (i) an NTD bearing potential phosphorylation sites (Desclozeaux et al. 1998; Chen et al.

2013a) and (ii) a consensus NES within the HMG box as otherwise observed among mammalian Sry and Sox family members (Malki et al. 2010). The inactive NES of

(IxxxLxxxxxSL; Fig. 8C) is selectively found in the subset of Sry alleles of Muroidea rodents that also contain a CAG repeat. In mSry this variant NES blocks NCS as characterized in hSRY (Chen et al. 2013a). Competency for CRM1-mediated nuclear

167

export, conserved in deer and goat Sry (Chen et al. 2013a), was regained on reversion to the Sry consensus NES motif (IxxxLxxxxxML) (Fig. S4.4). The contribution of the murine GRD to testicular differentiation in vivo(Bowles et al. 1999) and to the gene-regulatory activity of mSry in cell culture (present results) thus resolves an apparent paradox posed by the clinical association between Swyer’s mutations in hSRY that selectively impair its NCS in a fashion similar to wild-type mSry (Chen et al. 2013a).

We speculate that the biochemical activity of the mSry GRD has attenuated selective pressure on its Sry HMG box, leading to genetic drift. Whereas the HMG box of Sox3 (the proposed X-encoded ancestor of Sry; (Katoh and Miyata 1999) is broadly conserved among eutherian mammals, including within Rodentia, the mSry domain differs from the boxes of primates, ungulates, and other mammalian orders at more sites (and at these sites by less conservative substitutions) than do the latter from the Sox3 box (Tables S4.2 and S4.3). Such variation in mSry was associated with attenuated thermodynamic stability and foreshortened residence time of a specific

DNA complex (Fig. 4.4). Although native-like -helical structure is largely regained on specific DNA binding (induced fit), the reduced lifetime of the mSry domain-DNA complex may underlie its impaired transcriptional-regulatory activity in the absence of the GRD (Bowles et al. 1999). These biophysical findings suggest that the

168

contribution of the GRD to Sox9 transcriptional activation (Phillips et al. 2004) compensates for biophysical instability, impaired NCS, and absence of NTD phosphorylation site.

Block Dissection of the GRD. A CAG microsatellite occurs in Sry is several lineages within Muroidea, most dramatically in Muridae (old world rats, mice and gerbils).

Repeat lengths are variable, ranging from 20 poly-Gln blocks (as in mSry in M. musculus domesticus; see Fig. 2B (Denny et al. 1992)) to 3 poly-Gln blocks (Rattus norvegicus (Griffiths and Tiwari 1993)). Even among laboratory strains of M. musculus, domesticus-derived Y chromosomes encode Sry proteins of different lengths than those encoded by molossinus-derived Y chromosomes (alleles SryB6 and

Sry129; (Coward et al. 1994)). Although block numbers vary, the downstream

Sox9-dependent GRN is presumably similar as indicated by heterogametic male development. Tolerance to variation in poly-Gln block number is in accordance with deletion analysis in CH34 cells wherein constructs containing 3 or more blocks activated Sox9 transcription, to an extent similar to that of canonical mSry (20 blocks;

Fig. 4.2D). Further, similar activities were observed in CH34 assays of wild-type Sry alleles derived from Rattus norvegicus and Tokudaia muenninki (Muennink's spiny rat;

Okinawa), which each contain 3 poly-Gln blocks (accession number in pubmed:

169

Rattus norvegicus, NP_036904; Tokudaia muenninki, BAJ08420). ChIP studies focused on Sry-binding sites in TESCO (Sekido and Lovell-Badge 2008) indicated

CTDs containing < 3 blocks are associated with loss of Sox9enhancer occupancy (Fig.

4.3B and D).

Microsatellite-based Biochemical Complementation. Chimeric mSry/hSRY constructs were prepared to test whether a CAG-associated TAD could relax biochemical constraints on the function of the HMG box. Chimera 1 is a variant of hSRY containing the murine box (Fig. 4.5). Its properties are analogous to those of

I90M hSRY (an inherited allele; see companion study (Chen et al. 2013a)) bearing a dysfunctional NES, leading in each case to reduced activity despite increased nuclear accumulation (Fig. 4.5B and C). Comparison of human NTD variants indicated that phospho-mimicry through acidic substitutions in a putative PKA site rescued the activity of chimera 1. Such rescue also implies that, on NTD phosphorylation and on enhanced nuclear accumulation due to impaired nuclear export, the function of hSRY

(at least in a rodent cell line) tolerates the many substitutions that otherwise distinguish between human and murine boxes. To test whether NTD phosphorylation could modulate the function of mSry, chimera 2 was fused to the human NTD (Fig.

4.6A). Its gene-regulatory properties were found to be robust to AAA or SSS

170

substitutions (Fig. 4.6D), implying that the GRD renders such regulation superfluous.

Chimera 3 contained both the human NTD and box fused to the C-terminal non-box sequences of mSry, including its GRD (Fig. 4.6B). Occupancy of TESCO sites was similar to its wild-type parents (Fig. 4.3C). The function of chimera is likewise robust to PKA site substitutions (Fig. 4.6E).

Biochemical complementation by the mSry CTD was further investigated in relation to an inherited human variant near the protein-DNA interface (Y127F in Fig.

4.6C; consensus position 72 in the HMG box), which partially impairs specific DNA binding (Jordan et al. 2002). The aromatic ring adjoins V60 (consensus position 5), also a site of inherited mutation (Chen et al. 2013a). Whereas in the context of hSRY

Y127F impairs Sox9 expression by ca. twofold (as observed in companion studies of

V60L and I90M (Chen et al. 2013a)), the mutation has no effect in the context of chimera 3 (Fig. 4.6F). Such intragenic complementation supports an evolutionary scenario wherein insertion of a CAG microsatellite in a founding lineage of Muroidea enabled drift of HMG box sequences. To further explore GRD complementation, chimeric constructs 5-8 employed the HMG boxes of mSox3 and goat Sry as inspired by studies of chimeric transgenes (Bergstrom et al. 2000; Pannetier et al. 2006).

Whereas in the context of hSRY respective “box swap” variants exhibited relative activities in the order mouse < human = Sox3 < goat (Fig. 4.7C), such functional

171

differences were abolished in the presence of the mSry CTD (Fig. 4.7D).

Evolution of Male Sex Determination. Whereas Sry is generally conserved among eutherian mammals as the testis-determining locus (Berta et al. 1990), an enigma is posed in Muriodea(Fig. S4.1).One member of the family Cricetidae, the vole Ellobius lutescens, has no Sry gene or Y chromosome (Fig. 4.8A); its mechanism of sex determination mechanism is unknown (Just et al. 2007). Evidence for the rapid evolution of non-Sry-dependent male-determining mechanisms has likewise been obtained within Muridae. Like genus Apodemus (which contains the common mouse and rat), the related genus Tokudaia contains species with GRD-associated Sry alleles as exemplified by T. muenninki (above). Despite the implication of a common ancestor whose Y chromosome contained the original CAG-associated microsatellite,

Tokudaiaalso contains species lacking a Y chromosome(Fig. 4.8B; (Arakawa et al.

2002) as exemplified by T. tokunoshimensis (Tokunoshima spiny rat) and T. osimensis

(Armani spiny rat).We propose an evolutionary scenario wherein(i) microsatellite invasion of Sry within a Muroidea common ancestor enabled drift of HMG box sequences with biophysical perturbation and loss of non-box phosphorylation sites and (ii) subsequent GRD repeat-number instability led (below the threshold of 3 poly-Gln blocks) to attenuated Sry-directed Sox9 activation in some lineages, leading

172

in turn to reproductive isolation and recruitment of non-Sry-dependent mechanisms of

Sox9 transcriptional activation in the bipotential gonadal ridge. Redundant or non-functional Sry alleles were a likely precondition for the rare anomalous loss of the Y chromosomes in this clade.†

The plausibility of this evolutionary scenario is strengthened by intermediate cases found within Cricetidae (grass mice Akodon boliviensis and Akodon azarae).

Although males represent the heterogametic sex, these species exhibit high percentages of XY females (Bianchi 2002). Their variant Sry genes encode a foreshortened NTD (lacking potential phosphorylation sites), a divergent box with non-consensus NES (Table S4.4) followed by a single-block Gln-rich motif (Bianchi

2002). We speculate that this remnant GRD is insufficient to rescue the function of the divergent NTD and box, providing partial activation of Sox9 at the threshold of testis determination: testis determination is thereby non-robust with respect to autosomal variation, environmental fluctuations, or stochastic gene expression. Such grass mice may thus stand at the crossroads of Sry loss and Y-chromosome degeneration in

Rodentia.

173

Concluding Remarks

Poly-Gln repeats encoded by CAG repeats were first observed in neurological disorders (Pearson et al. 2005; Walker 2007; van Eyk et al. 2011) in which length-dependent alterations of protein structure, function and toxicity can correlate with clinical severity or age of onset (Taylor et al. 2002). In Huntington’s disease, for example, aberrant gain-of-function by the variant huntingtin perturbs patterns of neuronal gene expression, in part through competitive binding of the GRD to transcriptional co-activators and the basal transcriptional machinery (Dunah et al.

2002). Whereas microsatellite instability within TFs is not generally associated with divergence of respective DNA-binding motifs,‡ the evolution of Sry in Muroidea is remarkable for both variation in GRD length and divergence of HMG-box sequences

(Tucker and Lundrigan 1993; Miller et al. 1995). We speculate that GRD-associated gain of function in a TESCO-directed Sox9 transcriptional regulatory complex circumvents biochemical requirements for NCS and NCS-coupled phosphorylation as defined in human SRY (Chen et al. 2013a).

We envisage that the CAG triplet repeat of rodent Sry alleles has functioned in the radiation of Muroidea as an intragenic "capacitor" to suppress phenotypic consequences of variation elsewhere in the protein, which (in the case of mSry) includes destabilizing substitutions in the HMG box, loss of NCS, and deletion of the

174

NTD. Such variation could then have been unmasked by microsatellite instability leading to truncation of the GRD below its critical threshold (repeat length 3). This model extends the paradigm of a genetic capacitor as defined by heat shock protein 90

(Rutherford and Lindquist 1998; Lindquist 2009). Because Hsp90 buffers the misfolding of proteins regulating metazoan development (thereby conferring interim stability to GRNs) discharge of the Hsp90 capacitor may underlie rapid morphological evolution as documented in the fossil record (Eldredge and Gould

1972). Similarly, the microsatellite capacitor of Sry in Muroidea may enable, via replicative DNA slippage (Liu and Wilson 2012), sudden shifts in molecular mechanisms of male sex determination. Operating through the biochemical properties of a GRD in a TESCO complex, this Sry capacitor may discharge to create reproductive barriers between nascent species.

We thus envisage that microsatellite instability within Sry has promoted the emergence of biological novelty in Muroidea. Such innovation reflects a combination of genomic and biochemical mechanisms distinct from general processes leading to Y degeneration (Marchal et al. 2003; Wilson and Makova 2009). Although biochemical properties of mSry and hSRY differ, each has evolved to regulate Sox9 expression just above the threshold of Sertoli-cell specification. Gonadogenesis at the edge of ambiguity is shared by rare human families (Knower et al. 2011; Phillips et al. 2011;

175

Chen et al. 2013a) and routine XY sex reversal among grass mice (Bianchi 2002). The thin thread of testis determination (Polanco and Koopman 2007), first glimpsed in studies of murine Y chromosome-autosome incompatibility (Albrecht et al. 2003), represents an apparent violation of the Waddington principle of developmental canalization (Waddington 1959). Addressing why sex is different will require deciphering a seeming paradox: multi-level selection (Wilson and Wilson 2007) against the robustness of male gonadogenesis. A fundamental problem at the intersection of biochemistry and evolutionary biology is posed by the developmental, neuroendocrine, behavioral, and social origins of such selection.

176

Materials and Methods

Plasmids. Mammalian plasmids expressing hSRY, mSry, and variants (Table S4.1) were constructed by PCR and cloned into pCMX (containing CMV promoter) between EcoRI and XhoI restriction sites (Phillips et al. 2011). Following the initiator methionine, the cloning site encoded a hemagglutinin (HA) tag in triplicate.

Mutations were introduced using the QuikChange™ kit (Stratagene).

Mammalian Cell Culture. CH34 cells (kindly provided by T. R. Clarke and P. K.

Donahoe (Haqq and Donahoe 1998)) were cultured in Dulbecco’s modified Eagle medium (DMEM) containing 5% heat-inactivated fetal bovine serum at 37ºC under

5% CO2.

Human Cell Lines. NT2-D1 cells (Knower et al. 2007) were grown in Dulbecco’s modified Eagle's medium in an atmosphere of 5% CO2; the complete growth medium contained FBS to a final concentration of 10%. PC-3 cells (Kaighn et al. 1979) were cultured in the F-12K medium (ATCC) with 10% FBS in 5% CO2 atmosphere.

Transient transfections were in effected by the Fugene HD protocol (Hoffmann

LaRoche). PCR primers were in accordance with human genomic sequences.

177

Transient Transfection. Transfections were performed using Fugene6

(Hoffmann-LaRoche). After 24 h in serum-free medium, cells recovered in

Dulbecco’s Modified Eagle’s Medium (Gibco) containing 5% heat-inactivated fetal bovine serum. Efficiencies were determined by ratio of GFP positive cells to untransfected cells following co-transfection with pCMX-SRY and pCMX-GFP.

Cellular localization was probed by immunostaining 24-h post transfection following treatment with 0.01% trypsin (Invitrogen) and plating on 12-mm cover slips (Fisher

Scientific). SRY expression was monitored in triplicate by Western blot in relation to

-tubulin.

Western Blot. Expression of mSry/hSRY and variants was monitored by Western blot using monoclonal anti-HA antiserum (Sigma-Aldrich) following lysis in Lysis-M buffer (Hoffmann-LaRoche). Protein concentrations were measured by BCA protein assay (Thermo Scientific). Samples were subjected to SDS PAGE, and electroblotted onto apolyvinylidene-difluoride membrane (Hoffmann-LaRoche). Solutions containing anti-HA antiserum in 5% non-fat milk Tris-buffered saline and Tween 20at a ratio of 1:5000 (v/v) were hybridized onto the membrane using the vendor’s protocol.

178

Real-time-Q-rtPCR Assay. Accumulation of Sox9 mRNA in transfected CH34 cells was probed by qPCR as described (Phillips et al. 2011). Cellular total RNA was extracted using the RNeasy kit (Qiagen). Primer sequences are provided in Table S4.5.

TFIID was used as internal control; measurements were made in triplicate with blind coded samples.

Immunocytochemistry. Transfected cells were evenly plated on 12-mm cover slips and fixed with 3% para-formaldehyde in phosphate-buffered saline (PBS) on ice for

30min. Fixed cells were treated with cold-permeability buffer solution (PBS containing 10% goat serum and 1% triton X-100; Sigma-Aldrich) for 10 min, followed by blocking using cold PBS containing 10% goat serum and 0.1% Tween-20.

Hybridization was performed by incubating FITC-conjugated anti-HA antibody

(diluted to 1:400) overnight at 4ºC. After washing and DAPI staining, cells were visualized by fluorescent microscopy in relation to the total number of GFP-positive cells. 800-1000 cells were counted in each case.

Chromatin Immunoprecipitation. Transfected cells were probed by ChIP using an anti-HA antiserum as described in our companion study (Chen et al. 2013a). An

179

expanded high-fidelity PCR protocol was provided by a vender (Hoffmann LaRoche).

The enrichment on each fragment was performed in triplicate.

Biophysical Assays.CD, fluorescent spectroscopy and FRET based Kd determinations were performed as described (Phillips et al. 2011). Stopped-flow FRET-based analysis of protein-DNA dissociation rates employed 15-bp specific FRET-labeled DNA site

(5’-TCGGTGATTGTTCAG-3’ and complement) in 1:1 SRY in one syringe.

Competitive displacement was achieved by rapidly mixing the complex with 20-fold excess of unlabeled 12-bp DNA with the same target site from a second syringe (Fig.

S4.3) (Phillips et al. 2011).

180

Footnotes

Acknowledgements. We thank Prof. P. K. Donahoe for cell line CH34 and encouragement and H.-Y. Kao for the pCMX plasmids. S. Jeong and P. Janki for assistance with nuclear localization studies; T. Feng, B. Li, and R. Singh for participation in early stages of this work; and P. DeHaseth, H.-Y. Kao, D. Samols, and

P. Sequeira for advice. MAW thanks B. Baker, F.A. Jenkins, Jr., P. Koopman, R.

Lovell-Badge, R. Sekido, and D. Wilhelm for discussion. This work, a contribution from the Cleveland Center for Membrane & Structural Biology, was supported in part by a grant to MAW from the National Institutes of Health (GM080505).

Although in past studies the mSry domain has been described as exhibiting more stringent sequence specificity than the hSRY domain (a seeming biochemical improvement) (Giese et al. 1994), it is possible that such findings represented kinetic artifacts of gel mobility-shift assays (i.e., even more rapid dissociation of variant mSry complexes) as discussed by Li and coworkers (Li et al. 2006). Similar sensitivity of GMSA to changes in protein-DNA dissociation rates may account for the murine domain’s seeming enhancement of discrimination against AT IC transitions

(Giese et al. 1994).

†At a time scale similar to that of the Rodentia radiation, the Y chromosome of primates has been stable (Hughes et al. 2012). It is not known whether selective

181

pressure to maintain SRY (in the absence of microsatellite instability) may have dampened the pace of Y-chromosome degeneration.

‡Microsatellite instability within the coding region of a basal TF may lead to disease as demonstrated by the association between variant TATA-binding proteins (TBP) and spinocerebellar ataxia (type 17) (Tomiuk et al. 2007). Such instability within a specific TF may also underlie the rapid divergence of a morphological program as inferred from studies of Runx2 in Carnivora (Sears et al. 2007).

182

Fig. 4.1 Structures of hSRY and mSry.

(A) Human protein (204 residues; upper bar) comprises: N-terminal domain (violet;

NTD; residues 1-55) with PKA sites (gray; residues 31-33); HMG box (black; residues 56-141) containing the basic tail (dark gray; bt; residues 129-141); and

C-terminal domain (white; CTD; residues 142-204) containing bridge- (Br) and

PDZ-binding motifs (orange and dark purple, respectively). Murine protein (395 residues; lower bar) comprises: HMG box (green; residues 3-86) with basic tail (dark gray; 74-86); Br motif (orange); and C-terminalnon-conserved domain (light gray) directly linked to glutamine-rich domain (chartreuse; residues 144-367). (B) Ribbon model of human HMG box/DNA complex (Murphy et al. 2001). Left, front view of bent DNA site (blue ribbon) overlying box with basic tail (black and gray). Side 183

chains at the protein-DNA interface are shown in red (R7, F12, I13, Y74, and P76; consensus HMG box numbering scheme), brown (R4, K6, Q62, R66, K73, K79, and

K81), or auburn (R20, N32, S33, and S36). A 90º rotation about vertical axis is shown at right. Coordinates were obtained from PDB entry 1J46. (C) Corresponding space-filling model of hSRY HMG box (front view). Color code of DNA contacts as in panel B; non-contact surfaces are gray. (D) Homology model of mSry HMG box.

Amino-acid substitutions are indicated by darker shades of respective colors

(DNA-binding surface) or darker gray (non-DNA-binding surface).

184

Fig. 4.2 Gln-rich domain of mSry contributes to transcriptional activation and

TES occupancy.

(A) Histogram showing baseline extent ofSox9mRNA accumulation on transfection by wild-type hSRY or mSry at high dose (1 g; black bars) or low dose (0.02 g; white bars). Inactive hSRY variant I68A served as negative control (Right). (B) Schematic diagram and amino-acid sequence of mSry glutamine-rich domain, comprising 20 poly-Gln (Q) repeats (chartreuse) separated by spacer with conserved FHDHH 185

element (black). (C) C-terminal deletion constructs of mSry: wild-type, 20 Q-repeats

(top); 1, 10 Q-repeats; 2, 8 Q-repeating tracts; 3, 4 Q-repeating tracts; 4, 3

Q-repeating tracts; 5, 2 Q-repeating tracts; 6, one Q-repeating tract (Bottom). (D)

Histogram showing qPCR results of Sox9 expression by the successive C-terminal deletion constructs with low-dose transfection (0.02 g plasmid with 50X empty-vector dilution as in panel A). In panels A and D horizontal brackets designate statistical comparisons: (* or “ns”), Wilcox p-values <0.05 or >0.05, respectively.

186

Fig. 4.3 ChIP analysis of Sox9 testis-specific core enhancer occupancy.

(A) Schematic model of the murine Sox9 gene with testis-specific enhancer elements, including TES/TESCO. (B-D) ChIP assays probing SRY occupancy of TESCO. (B)

ChIP assays of representative mSry C-terminal deletions (see Fig. 4.2C). (C) ChIP assays of representative chimeric proteins. Chimera 1 containing a mouse HMG box within hSRY framework (see Fig. 4.5A); chimera 3 a human HMG box within mSry framework (see Fig. 4.6B). A negative control is provided by inactive hSRY variant

I68A. In ChIP assays primer sets a and c probed for SRY occupancy and primer set b served as a negative control (see Table S4.5). At Right in panels B and D are shown non-specific immunoglobulin G (IgG) controls; equal loading was verified by primer set b. (D) Histogram showing relative TESCO occupancies by the C-terminal deletion series (see Fig. 4.2C) with the wild-type mSry signal defined as 100%.Horizontal brackets designate statistical comparisons as in Figure 4.2. 187

Fig. 4.4 Biochemical differences between HMG boxes of mSry and hSRY.

(A) The murine domain (green circles) exhibits increased sensitivity to chemical denaturation by guanidine-HCl relative to the human (black line) as probed by intrinsic tryptophan fluorescence. (B) Far-ultraviolet CD spectra at 37ºC demonstrates greater attenuation of -helical content of the murine domain (green circles) relative to human domain (black line). Inset, thermal unfolding mid-point of murine domain

188

(green circles; monitored by CD at 222 nm) is decreased by ca. 5 ºC relative to human

(black line). (C) Thermal denaturation of mSry and hSRY box-DNA complexes (green and back circles, respectively) as monitored by CD at 222 nm. Apparent midpoint temperatures (vertical lines) are 53ºC (mSry) and 59ºC (hSRY). The HMG boxes were complexed with a 12-bp DNA site, 5’-GTGATTGTTCAG-3’ and its complement. (D) Far-ultraviolet CD spectra at 37ºC of free mSry HMG box (open green circles), DNA complexes of mSry and hSRY (green closed circles and solid black line, respectively) showing the regain of -helical structure (downward arrow).

The spectrum of free DNA is shown as a red line. (E and F) Stopped-flow

FRET-based dissociation kinetic assay of HMG-DNA complexes at 15 ºC (E) and 37

ºC (F); Representative data and solid fitted lines showing time-dependent increase in donor fluorescence of FRET-labeled DNA due to dissociation from the SRY complex.

Dissociation rate constants (koff) were determined by fitting 3-4 individual traces to a single exponential equation (see Fig. S4.3 for schematic experimental design (Phillips et al. 2011)). At both temperatures the dissociation of the murine complex (green) is more rapid than that of the human complex (black). This data was obtained by Joe

Racca as part of his collaboration on this project.

189

Fig. 4.5 Subcellular localization of mSry, hSRY, and chimeric proteins.

(A) Design of chimeric mSry/hSRY chimera 1 (bottom) relative towild-type hSRY

(top) and mSry (middle). Domain color code is as in Figure 4.1A. NTDs of hSRY and chimera 1(violet) contained either native PKA site (PALRRSSSFLCTE; residues

26-38phosphorylation site underlined), AAA variant (designated “phospho-dead”), or

DDD variant (“phospho-mimic”). (B) Subcellular localization of epitope-tagged hSRY/mSry constructs as analyzed by immunostaining: DAPI nuclear staining

(Upper row; blue), and SRY immunofluorescence (Lower row; green). In most cells wild-type hSRY localizes in nucleus with a minor fraction exhibiting a pancellular distribution (see panel C). In contrast mSry, chimera 1, and hSRY variant I90M (with defective NES; see preceding study (Chen et al. 2013a)) exhibited augmented nuclear localization; human mutation R62G (impairing an NLS (Gontan et al. 2009)) led to consistent pancellular distribution of SRY. (C) Histogram indicating fractions of 190

transfected CH34 cells exhibiting exclusive nuclear localization of hSRY/mSry (gray bars) versus pancellular distribution (white bars). Lengths of gray and white bars do not sum to 100 due to occasional GFP-positive cells lacking SRY expression. The transfected plasmid dose was in each case 1 g.(D) Results of qPCR assays of Sox9 gene expression following low-dose transfection (0.02 g with 50X empty-vector dilution as in Fig. 4.2A). Respective right- and left-hand sets of data pertain to hSRY

NTD variants (SSS, AAA, and DDD as in panel A) or corresponding variants of chimera 1. Inactive hSRY variant I68A (Far Right) served as a negative control.

Horizontal brackets in panels C and D indicate statistical comparisons as defined in

Figure 4.2.

191

Fig. 4.6 Function of mSry Gln-rich domain in chimeric constructs.

(A-C) Design of chimeric proteins 2-4 in relation to parent proteins. The domain color code is as defined in Figure 4.1A. (A) Chimera 2: mSry containing an N-terminal extension provided by the hSRY NTD. (B) Chimera 3: NTD and HMG box of hSRY with basic tail (human residues 1-141) are linked to CTD of mSry (mouse residues

87-395). (C) Chimera 4: variant of chimera 3 containing human clinical mutation

Y127F (a substitution at HMG consensus position 72 that partially impairs specific

DNA binding (Jordan et al. 2002)). Variants SSS, AAA, and DDD of chimeras 2 and 3 were constructed as probes of the human NTD PKA site as defined in Figure 4.5A.

(D-F) Results of qPCR assays of Sox9 gene expression following low-dose

192

transfection (0.02 g with 50X empty-vector dilution as in Fig. 4.2A). A positive control was in each case provided by wild-type mSry; negative controls were provided by an empty vector, inactive hSRY variant I68A or impaired mSry variant

M13A (homolog of mutant hSRY described in companion study (Chen et al. 2013a)).

(D) Sox9 activation by wild-type mSry or chimera 2 (SSS, AAA, or DDD variants).

(E) Sox9 activation by wild-type mSry, hSRY (SSS, AAA, or DDD variants), or chimera 3 (SSS, AAA, or DDD variants). (F) Sox9 activation by wild-type mSry, chimera 3 (wild-type SSS NTD), hSRY bearing Y127F, or chimera 4 (wild-type SSS

NTD). Horizontal brackets in panels D-F indicate statistical comparisons as defined in

Figure 4.2; **, p-value < 0.01.

193

Fig. 4.7 Correlation of CH34 model with sex reversal in transgenic mice.

(A) Design of chimeric proteins 1, 5, and 6 in relation to wild-type hSRY and (B) chimeric proteins 3, 7, and 8 in relation to native mSry bearing the human NTD.

Transgenes encoding chimeric proteins 5 and 7 (containing the HMG boxes of Sox3; blue) are able to induce XX sex reversal in mice (Bergstrom et al. 2000). The goat Sry

HMG box was employed in chimeric protein 6 and 8 (aquamarine) due to the observation that the transgenic goat Sry gene is also able to induce testis development in XX transgenic mice(Pannetier et al. 2006). (C) Results of qPCR assays of Sox9 gene expression activated by wild-type hSRY and hSRY-type chimeric proteins (1, 5, and 6) following high-dose transfection (1g expression plasmid; black bars) or low-dose transfection (0.02 g with 50X empty-vector dilution; white bars as in Fig.

4.2A). (D) Results of qPCR assays of native mSry and mSry-type chimeric proteins (3,

7, and 8). The transfection was as described in C. In each case the function of the 194

chimeric proteins was indistinguishable from that of wild-type mSry. A negative control was provided by the inactive hSRY variant I68A (Right). Horizontal brackets indicate statistical comparisons; n.s., p-value >0.05.

195

Fig. 4.8 Rodent Srys with CAG-encoded GRD contain attenuated NES motif.

(A) Representative species in Muroidea superfamily. Color codes depicting variations in the Sry frame: brown; follows mSry frame, such as HMG-bridge-“domain with repeating Gln-tracts encoded by CAG”, magenta; species with Sry containing poly-A repeating tracts encoded by GCA, and species with different evolutionary fates of Sry are framed in boxes. (B) Phylogenetic relationships of three Tokudaia species. The phylogenetic tree in this panel was modified from Murata et al. (Murata et al. 2010).

Color codes: brown; as in panel A and red; species that have lost their Sry. (C)

Alignments of the published SRY sequences without CAG-encoded repeating domain

(Upper bracket) and that with CAG-encoded GRD (bottom bracket). The NES motifs are highlighted in bold. The schematic figure (top) shows the secondary structural environment of NES motif. Residue numbers correspond to the consensus HMG box.

The second and third -helices in hSRY HMG box are depicted by 2 and 3. The 196

conserved serines proposed to attenuate the NES functional efficiency are in red

(bottom bracket).

197

Fig. S4.1 The phylogenetic tree of selected rodents.

(A) Phylogenetic tree of representative species within the Rodentiaorder. The evolutionary relationships are based on molecular information from Huchon et al.

(Huchon D, et al. (2002) Mol Biol Evol 19:1053-65) and Poux et al. (Poux C, et al.(2002) Mol Biol Evol 19:2035-7). The phylogenetic tree was modified from a review 198

article by Ferguson-Smith and Trifonov (Ferguson-Smith MA, et al. (2007) Nat

RevGenet 8:950-62). The Muroidea superfamily containing Muridae and Cricetidae families is highlighted by a box. Color codes for SRY frame: Green; follows the

NTD-HMG-CTD framework, Gray; insufficient information in published primary sequences, Black; absence of published SRY sequences. (B) The phylogenetic tree of

Muroidea superfamily. Families lacking published SRY sequence information are depicted with asterisks. Sister families Muridae and Cricetidae with CAG-encoded glutamine-rich domain Sry proteins are in brown letters.

199

Fig. S4.2 Selected gene expression patterns in CH34 cell line.

RT-Q-rt-PCR was employed to probe mRNA abundances of Sox family members (A), candidate male GRN-related genes (B), and housekeeping genes (C) following transfection of an SRY expression plasmid, empty vector, or control plasmid expressing a stable, but inactive, SRY variant (I68A). (A) Fold mRNA accumulation of Sox family 200

genes, including Sox8 and Sox9 previously implicated in the program of Sertoli-cell differentiation. (B) Fold mRNA accumulation of non-Sox sex-related factors. Results show that CH34 cells express low endogenous levels of Sry, AMH, Ptgds, and Capsulin but exhibit higher Wt1 gene expression. Only the transcription of Fgf9 was significantly activated when probed 24 hours post SRY transient transfection (asterisks).

(C) Fold mRNA accumulation of housekeeping genes; these genes were not affected by expression of transfected SRY.Abbreviations:AMH, Anti-Müllerian Hormone; Ptgds,

Prostaglandin D2 synthase; Fgf9, fibroblast growth factor 9; Sf1, steroidogenic factor 1;

Wt1, Wilms tumor 1; Wnt, wingless-type MMTV integration site family. Statistical analyses: *Wilcox test, p<0.05.

201

Fig. S4.3 Schematic illustration of FRET-based DNA bending probe and stopped-flow experimental design.

(A) A central bend reduces distance between respective 5’-ends. One 5’ end is labeled with fluorescein (FAM; donor), and the bottom showed 5’ end with tetramethylrhodamine (TAMRA; acceptor). The fluorophores are flexibly linked to the DNA site via hexanyl linkers. The bent DNA site depicted is based on the co-crystal structure of the Sox17-DNA complex. The photophysics of this model complex has been characterized in detail in both murine and human complexes

(Phillips NB, et al. (2004) Biochemistry43:7066-8). (B) Stopped-flow experimental design. The stopped-flow apparatus coupled to the fluorimeter allowed the measurement of FRET-based dissociation of SRY-DNA complex. One syringe 202

contained a preformed protein-DNA complex containing a 15-bp DNA probe containing 5’-donor (FAM; green circle) on one strand and 5’-acceptor (TAMRA; red circle) on the other; the other syringe contained a 20-fold excess of unmodified 12-bp

DNA site. (Fig. adapted from Phillips NB, et al. (2011) J Biol Chem 86:36787-807).

203

Fig. S4.4 A consensus NES motif restores CRM1-mediated nuclear export of mSry.

Histogram provides fractions of transfected CH34 cells exhibiting exclusive nuclear localization of SRY (filled bars) versus pancellular distribution (open bars). The transfected plasmid dose was in each case 1g. Substitution of the rodent-specific Ser at residue 45 by Met (as in human SRY) resulted in a quantitative pattern of subcellular localization indistinguishable from that of hSRY. Near-exclusive nuclear localization of wild-type mSry is highlighted in red (at Right). At bottom is summarized status of

CRM1-binding activity: (+) co-IP positive and (-) co-IP negative (for experimental design, see companion paper in this issue). Horizontal brackets indicate statistical comparisons as in Figure 2 (asterisk, p<0.05; ns, p>0.05 and in this case, the ns p-values were all > 0.15). (B) CH34 CRM1 co-IP assay. Lanes 1-4 provide: (1) wild-type human SRY (wild-type NES motif designated at top –ML); (2) variant mSry bearing consensus primate NES substitution IxxxLxxxxxML (designated at

204

top –ML); and (3) control studies of mouse Sry (mSry; red) bearing impaired NES sequence IxxxLxxxxxSL (designated at top –SL). Top boxes, background bands (X) and SRY-CRM1 co-IP signal (arrow); middle boxes, species-specific Sry/SRY input bands; and bottom boxes, -tubulin control for protein loading. Images were taken from a single gel.

205

Fig. S4.5 Studies of SRY phosphorylation and its effects on gene-regulatory activity.

A representative Western blot is provided as a control for the phosphorylation studies in the main text. Top blot, Western blot following precipitation with anti-phosphoserine antiserum and blotted using an HRP-conjugated anti-HA antiserum, thus showing the extent of SRY phosphorylation at 24 kD (bck indicates non-specific background band); middle blot, anti-HA antiserum documented total

SRY (phosphorylated or unphosphorylated); and bottom blot, anti-α-tubulin antiserum provided loading controls. Lane 1, HA-tagged “phospho-dead” SRY construct; lane 2,

HA-tagged wild-type SRY; and lane 3, HA-tagged “phospho-mimic” SRY. Dashed boxes in lanes 1 and 3 highlight diminished or absent SRY phosphorylation.

Nomenclature: The terms “phospho-dead” and “phospo-mimic” indicate substitutions at key serines within the putative PKA site of human SRY

(PALRRSSSFLCTE; residues 26-38 in human SRY). The phospho-dead construct 206

contains three SerAla substitutions (PALRRAAAFLCTE; underlined) whereas the phospho-mimic construct contains three SerAsp substitutions (PALRRDDDFLCTE; underlined) and so recapitulates the negative charges of phosphoserine modification.

207

Fig. S4.6 Studies of SRY/Sry-directed sex reversal in XX transgenic mice.

The male somatic phenotype is induced by a SRY/Sry open reading frame

(orf)-contained fragment from the Y chromosome (Koopman P, et al. (1991) Nature

351:117-21). The domain structures of SRY/Sry constructs are shown: (A) full-length mSry including mouse HMG box and C-terminal glutamine-rich region due to CAG repeat. (B and C) chimeric mSry constructions encoding the domain swap of the mSry

HMGbox with those of mSOX3 (B) or mSOX9 (C) (Bergstrom DE, et al. (2000)

208

Genesis 28:111-24). (D) The N-terminal portion of the mSry HMG box is replaced with that of hSry, including coding regions for N-terminal nonbox sequences

(Lovell-Badge R, et al. (2002) in The genetics and biology of sex determination pp

4-22, John Wiley, West Sussex, U.K.). (E) Construct contains intact the hSry orf, including the coding region for the human C-terminal nonbox sequences and the 3’ stop codon (arrowhead), instead of the N-terminal segment of the mSry orf. The glutamine-rich domain of mSry is thus not expressed (Lovell-Badge R, et al. (2002) in

The genetics and biology of sex determination pp 4-22, John Wiley, West Sussex,

U.K.). (F) Full-length goat SRY (Pannetier M, et al. (2006) FEBS Lett 580:3715-20).

(G and H) Constructs unable to direct sex reversal in XX mice are highlighted by red color. mSryStop1 and mSryStop2 contain premature stop codons after HMG box

(mSryStop1) or immediately before the glutamine-rich domain (mSryStop2) (Bowles

J, et al. (1999) Nat Genet 22:405-8).

209

Fig. S4.7. Transfected hSRY/mSry up-regulates endogenous SOX9 in human

PC-3 cell line. (A) Histogram showing rt-Q-PCR assays of expression of SOX9 on transient transfection of mSry or hSRY in PC-3 (1g plasmid DNA; blue bars) versus

CH34 cells (0.02 g; black bars). Vertical axes at Left or Right provide absolute fold-inductions in the respective cell lines. Negative controls were provided by an empty vector and an inactive hSRY variant (I68A). Statistical comparisons: p-value (*)

< 0.05; “ns” indicates p-value > 0.05. (B) Western blots probing extent of total cellular expression of mSry, hSRY or variants. Top: similar band intensities are shown in each lane. Bottom: -tubulin loading control. (C) Histogram showing qPCR results of

SOX9 expression by a partial C-terminal GRD deletion construct (5; for details of construction see main text) and chimera 1, which each exhibit reduced SOX9 activation relative to mSry. Color code and symbols as in panel A. (D) Western blots documenting extent of total cellular expression of Sry variants. Top and Bottom panels 210

are as in panel B. This figure was taken from a single gel and blot; middle lanes not pertinent to this study were omitted. We note that PC-3 cells are a human prostate cancer cell line with Wnt/-catenin-regulated SOX9 gene (Wang, H., et al. (2007)

Cancer Res. 67, 528-536) but lacking expression of endogenous SRY (Dasari, V., et al.

(2002) J. Urology 167, 335-338).

211

Fig. S4.8. Transcriptional activity of hSRY/mSry variants in NT2-D1 cells. (A)

Endogenous SOX9 is up-regulated on transient transfection of mSry or hSRY (1g) in

NT2-D1cells as probed by rt-Q-PCR; negative controls were provided by an empty vector and an inactive hSRY variant (I68A). The activities of mSry and hSRY were indistinguishable. Statistical comparisons: p-value (*) < 0.05; “ns” indicates p-value >

0.05. (B) Western blots probing extent of total cellular expression of mSry, hSRY and

I68A hSRY. Top: similar band intensities are shown in each lane.

Bottom-tubulinloading control. This image was taken from a single gel and blot; middle lanes not pertinent to this study were omitted. (C) Design of luciferase-based transient co-transfection assay. Respective SRY- and SF1 consensus DNA target site are shown as by red and green boxes. (D) Histogram providing results of luciferase co-transfection assay in NT2-D1 cells in response to expression of the following:

212

hSRY, mSry, mSry deletion constructs 2 and 5 (for details see main text) and chimera 1 and its variants (NES repair; PKA site variant DDD as described in main text); negative controls were provided by an empty vector and an inactive hSRY variant (I68A). Statistical comparisons: p-value (*) < 0.05, (**) < 0.01; “ns” indicates p-value > 0.05. We note that the NT2-D1 cell line, derived from a human testis carcinoma, is formally unrelated to an embryonic pre-Sertoli cell but may partially reflect gene-expression patterns of gonadal cell types, such as FGF9, SOX9 and SF1 (Knower, K.C., et al. (2007) Sex Dev. 1:114-126).

213

Fig. S4.9 Predicted phosphorylation sites of SRY/Sry from selected species.

Dark red indicates the predicted phosphorylation sites in (A) Primates, (B) Artiodactyla, and (C) rodents. Red color highlights the known phosphorylation site in human SRY

(Desclozeaux M, et al. (1998) J Biol Chem 273:7988-95). Black and green boxes indicate the predicted kinases sites (black; PKA and green; PKC, respectively). The 84 a.a. HMG boxes are aligned and highlighted using thick line with kinase-related colors.

In (C) the glutamine-rich domain encoded by CAG is depicted in chartreuse. The potential phosphorylation sites and related kinases were predicted using NetPhosK 1.0; website: http://www.cbs.dtu.dk/services/NetPhosK/.

214

Table S4.1 Construction of Chimeric mSry/hSRY Proteinsa

aA consensus numbering scheme has been developed as a common designation of residues within the SRY/SOX-related HMG box and its basic C-terminal tail despite differences in positioning of the box within intact proteins (Murphy EC, et al. (2001)

J Mol Biol 312:481-99). Whereas the mSry HMG box begins at residue 1 of mSry, for example, the box of hSRY begins at residue 56 of intact human SRY. The cantilever side chain of the SRY/SOX box is thus at consensus position 13 (M13 in mSry) but occurs at residue 68 of human SRY (I68). Inherited Swyer’s mutation in hSRY Y127F thus affects consensus residue 72 at the beginning of the basic tail.

215

Table S4.2 Pairwise differences between HMG boxes

aFormat W(X/Y/Z) indicates: the total amino-acid difference (W); conservative change

(X), related substitution (Y), and non-conservative change (Z)

216

Table S4.3 Sox3 Conservation

The representative SOX3 HMG boxes exhibit identical protein sequences. Due to the insufficient sequence information of Artiodactyla animals including goat, sheep, cattle, and pig; the dog SOX3 was included for the comparison.

217

Table S4.4 The alignment of HMG boxes of the grass mice (Akodon) with selected Muroidea rodents (upper) and non-rodent mammals (bottom)

NES motifs are depicted in bold. The conservation symbols of alignment are shown: identical residues (*), conservative changes (:), related substitution (.), and non-conservative changes( ).

218

Table S4.5 ChIP Primer Sets for qPCR and ChIP (Testis-Specific Enhancer TESCO Element)a

aThe ChIP primer sets were derived from Sekido and Lovell-Badge (Sekido R, et al. (2008) Nature 453:930-34).

219

Chapter V

Summary and discussion

5.1 Nucleocytoplasmic shuttling in SRY function cascade

Clinical identification of SRY variants associated with the Swyer’s syndrome (46,

XY pure gonadal dysgenesis; (Michala et al. 2008)) provides a model to study the relationship between a gene-regulatory network (GRN) and the molecular functions of

SRY. We have focused on the subtle variants inherited by sterile XY sex reversal daughters from fertile fathers. The inherited mutations (V60L and I90M) in SRY were chosen based on family trees (Fig. 2.1B). Biophysical studies indicated that these mutations exhibited native-like structure and specific DNA-binding properties

(Table 2.1) but cell biological investigations uncovered associations with disturbances in nucleocytoplasmic shuttling (Fig. 2.3C and D). The clinical mutation V60L

(V60L in full length SRY; in consensus sequence it is position 5) exhibited partial impairment in nuclear localization due to the affected nuclear localization signal

(NLS). In another clinical mutation, I90M (full length SRY; in consensus position

35), we observed impaired nuclear export caused by reduced efficiency of the nuclear

220

export signal (NES), and hence exhibited exclusive nuclear localization (Fig. 2.4G).

It is conceivable that the impairment in nuclear localization reduces the

SRY-regulatory activity but it is a seeming paradox that the disturbance of nuclear export associated with the enhanced nuclear accumulation also exhibits the phenotype of gonadal dysgenesis.

The I90M SRY-driven activation function of endogenous Sox9 by the luciferase reporter assays from Knower and colleagues (Knower et al. 2011) shows increased activity after I90M overexpression. However, our successive decreases in the extent of I90M expression by dilution of transfected plasmid unmasked a twofold defect in

Sox9 regulation (Fig. 2.3F). The attenuated gene-regulatory activity associated with the impaired NES in mutation I90M suggests that nucleocytoplasmic trafficking mediates the function of SRY, and the post-translational modification is also involved.

A paradigm, developmental factor, SOX9, is modified by phosphorylation and this modification mediates SOX9 function as a transcriptional activator by enhancing its nuclear import function (Sim et al. 2008). Although the phosphorylation shows no regulatory function in SRY nuclear localization, it enhances the formation of the

SRY-DNA complex by phosphorylating the tandem triplet serines in the N-terminal domain (Desclozeaux et al. 1998). Thus, we proposed that SRY with impaired nucleocytoplasmic trafficking, I90M, is associated with the perturbed phosphorylation

221

modification, and hence attenuates the function of activating Sox9 even though the total nuclear SRY is increased. Fractionation assays support this model by the finding that nuclear phospho-I90M is dramatically decreased (Fig. 2.6B). Moreover, a phospho-mimic construct using Asp to replace the N-terminal PKA sites

(RRSSSRRDDD) verified that a negatively charged N-terminal domain (NTD) can rescue the NES-deficient variant (I90M)-mediated function of transcriptional activation (it also rescues the designed NES-disabled SRY (NES;

IxxxLxxxxxMLAxxxAxxxxxAL)). Together, these results of NLS or

NES-impairing clinical variants indicate a potential cascade model for describing how

SRY functions in promoting male differentiation in gonadal cells:

1. Newly synthesized SRY without phosphorylation is not “activated”. However, because phosphorylation has no effect on the nuclear localization of SRY, SRY is still transported into the nucleus mediated by both NLS domains, an atypical bipartite

N-NLS and a classical C-NLS.

2. PKA regulates the phosphorylation at the tandem triplet serines in the SRY NTD to

“activate” SRY. SRY, with or without phosphorylation, shares the same nuclear localization process modulated by the mechanism described above.

3. Serving as a transcriptional activator, SRY directly recognizes and interacts with the enhancer region of SOX9 (TESCO) (Sekido and Lovell-Badge 2008). In the

222

previous chapter, it was shown that this step is highly mediated by the process of

SRY-DNA complex formation, which involves a joint operation of DNA-intercalation by the cantilever side chain in SRY and the clamp function of the minor-wing structure of SRY to strengthen the complex (further discussion below).

4. Nuclear SRY travels back to cytoplasm. This nucleocytoplasmic shuttling is necessary for SRY to gain or re-gain phosphorylation in order to be activated. Our results regarding fractionation phosphorylation-status and related real time PCR assays using mutants and wild type SRY suggest that impairment of in-out trafficking of SRY significantly affects phosphorylation and reduces the SRY function of downstream gene activation.

5. Activated SRY by phosphorylation goes back to the nucleus, and activates downstream SOX9 through the formation of a high affinity SRY-DNA complex.

This cycle of activation properly describes how human SRY, a typical SRY, functions in cells. However, this model does not fit the behavior of mouse Sry

(Dubin and Ostrer 1994), which exhibits nuclear localization exclusively due to the disabled NES. The related discussions are mentioned in section 5.5.

5.2 Resolution of the enigma of inherited human sex reversal

There is a seeming paradox posed by inherited alleles of SRY associated with

223

human XY sex reversal. In such families, the daughter (46, XY with variant SRY allele) exhibits pure gonadal dysgenesis but has otherwise female external genitalia and internal reproductive structures (Swyer’s Syndrome). However, the father and this sterile daughter share the same SRY. Previous studies explained these inherited clinical cases by complete loss of SRY function or a gonadal tumor potentially associated with the mutant exhibiting super activity (Knower et al. 2011). These hypotheses may be able to explain the sex reversal daughter but are insufficient to explain the normally developed father.

Using the Sox9 quantitative platform, we found that two inherited clinical mutations, V60L and I90M, with opposite molecular characteristics surprisingly exhibited the same twofold decrease compared to native SRY in Sox9 responsiveness.

It is noticeable that the reduction is not a factor of 100 or 1000 which might be expected in the upstream control of a fundamental gene-regulatory network, or causing null-like transcriptional activities represented by some de novo clinical mutations, but instead is a factor of 2. With this result, we hypothesize that the threshold associated with the widely divergent fates of differentiation is at the edge of ambiguity. This hypothesis is supported by an experiment carried out by Eicher et al

(Eicher et al. 1996; Eicher and Washburn 2001; Washburn et al. 2001) demonstrating that Y chromosomes bearing Sry alleles derived from diverse mouse strains cause

224

divergent sexual phenotypes. Such phenotypes in mice are independent of changes in Sry sequence but depend on the extent and timing of Sry-Sox9 regulation.

Moreover, changes of twofold or less in Sry gene expression were highly associated with different levels of gonadal dysgenesis (normal testis differentiation versus Swyer phenotype). This observation implies a violation of Waddinton’s principle

(Waddington 1959) that fundamental development pathways should be canalized and robust to ensure that minor variations in development would not affect the outcomes.

The similar twofold biochemical threshold of SRY/Sry-Sox9 regulation in human and mouse indicates a thin boundary separates the differentiation fates of maleness or femaleness. Unlike de novo clinical SRY cases with null-like transcriptional functions, which are far away from the critical threshold for maleness, these inherited cases still retain partial function as transcriptional activators near the threshold.

Walking this close to the edge results in ambiguous fates in sex development.

5.3 The dynamic box in SRY functions as a transcriptional regulator

A key step for SRY functioning as a transcriptional activator is the formation of an SRY-DNA complex. The SRY-DNA complex is constructed by the major and minor wings of the HMG box binding in a widened minor groove of DNA (King and

Weiss 1993; Phillips et al. 2006; Phillips et al. 2011). Sharp DNA bending

225

accompanies partial side-chain intercalation at the crux of the angular domain. A cantilever side chain (isoleucine in position 68 of full length human SRY; in consensus position 13 of human HMG box) in the HMG box plays a critical role in this interaction, and is stabilized by the C-terminal basic tail of the HMG box (Weiss et al. 1997). Isolated SRY HMG box domains are partially disordered in the free

N-terminal segment of the minor wing. Unlike the minor wing which is very dynamic, the major wing exhibits higher structural order even in its free form (Weiss

2001).

However, a seeming paradox is posed by the mouse Sry (mSry) protein. It shares a similar threshold with human SRY (hSRY) in the cell-based model (CH34) but biophysical results suggest that the HMG box of mSry is significantly less structured than that of hSRY at physiological temperatures. Moreover, the CD spectrum of the mouse HMG box indicates increased induced-fit mechanism during

HMG-DNA complex formation (Fig. 4.4).

Although the induced-fit mechanism describes that the SRY-DNA interaction could function in a dynamic structure-independent manner, there are some limitations to this model. First, the cantilever side chain exhibits minimum tolerance: 1) cantilever residues are hyper-conserved. The only naturally-occurring residues are

Ile(I), Leu(L), Phe(F), and Met(M). Any non naturally-occurring replacement causes

226

null-like results. 2) Box-only in vitro biophysical studies using valine as a cantilever

showed a similar dissociation rate (koff) as mouse Sry. However, while the mSry box retains activity, the transcriptional activation of the valine cantilever in SRY is null-like (Fig. 1.12B).

Second, a clinical mutation, Y127F, indicates the importance of the minor wing in the induced-fit model. This TyrPhe substitution significantly increases the off rate in SRY-DNA binding by destabilizing a hydrogen bond that mediates formation of the HMG-DNA complex. Therefore, although the minor wing is “born to be dynamic”, it is still important for the “clamp” structure of the complex. This limitation is also highlighted by mutations in the minor wing, V60L and V60A (Fig.

2.2A). In the SRY-DNA complex the side chain of V60 (number in the full length

SRY) packs within an aromatic box formed by the side chains of H120, Y124, and

Y127 (number in the full length SRY), and this aromatic box serves as a clamp to strengthen the complex. Leu-Val substitution creates a mild spatial conflict and weakens the function of the clamp but Leu-Ala substitution still favors the structure of the aromatic clamp. Thus, the impaired DNA binding revealed by chromatin IP (Fig.

2.4) and reduced transcriptional activity (Fig. 2.3F) results in the HMG box-only assays reflect the slight disturbance of the V60L clamp structure whereas V60A SRY retains native-like function mediated by SRY-DNA complex formation (Fig. 2.4 and

227

Table 2.1).

Taken together, a model is proposed for describing how SRY functions as a transcriptional activator. Given that non-SOX-related HMG boxes have been characterized with greater structural and thermodynamic stability, it would be of future interest to identify if stabilizing substitutions would “improve” formation of the

SRY-DNA complex and affect related transcriptional activity by attenuating the flexibility.

5.4 The SRY-initiating male regulatory network is at the edge of ambiguity

Subtle clinical mutations V60L and I90M raised a father-daughter paradox.

Classical investigations for testing DNA-interaction of these mutations did not solve this paradox because their related results that are either subtly attenuated (V60L) or native-like (I90M) cannot explain such widely divergent phenotypes. Overall cell-based studies exploiting these inherited mutations exhibited a twofold reduction in downstream gene activity compared to that of wild type SRY (Fig. 2.3F). This result implies that this factor of 2, not a factor of 100 or 1000, defines a critical ambiguous boundary between testicular self-organization and gonadal dysgenesis, which means the GRN of male sex development, starting from the initiation central axis SRY-SOX9 regulation, is near the edge of ambiguity.

228

The idea that a GRN may function at the edge of ambiguity has been proposed as an evolutionary principle by Nykter and colleagues (Nykter et al. 2008) and the low transcriptional threshold of SRY provides a novel example of this principle. Such a tenuous program is nonetheless consistent with studies of strain-dependent sex reversal in mice due to Y chromosome/autosome compatibility (Albrecht et al. 2003).

We speculate that in the evolutionary history of mammals multi-level selection for male behavioral diversity—mediated in part by variation in the secretion of fetal testosterone—has undermined the robustness of gonadogenesis. Supportively, several key factors involved in the testicular differentiation regulatory network show the non-robustness phenomenon. Heterozygous nonsense and missense mutations in

SF1 associated with 46, XY pure gonadal dysgenesis suggest (in the absence of adrenal abnormalities) a syndrome of haploinsufficiency (Mallet et al. 2004).

Mutations in SOX9, moreover, result in a syndrome of transcription factor haploinsufficiency, designated campomelic dysplasia, wherein abnormalities of bone coincide (in XY patients) with male, intersex, or female somatic phenotypes

(Seymour et al. 2008; Dubois et al. 2011; Pritchett et al. 2011). These phenotypic variations suggest that in sex development-related factors of humans, the twofold transcriptional threshold characteristic of SRY extends to its immediate downstream targets. Together, these clinical entities highlight the anomalous non-robustness of

229

sexual dimorphism at multiple steps in the developmental program. Non-robustness of testicular differentiation may have offered selective advantages in the evolution of at least social therian species (human, mice, rat…etc) (De Vries et al. 2002; Gatewood et al. 2006; McPhie-Lalmansingh et al. 2008).

Our studies addressed a global issue of evolution and development: what biochemical properties of regulatory factors distinguish critical boundaries between organized and disorganized states of cellular differentiation or downstream pathways of pattern formation? We tried to answer this question using the experiments of nature-providing inherited alleles of SRY, which serve as probes for this boundary in gonadogenesis. Beyond the immediacy of cellular biochemistry, these probes establish a tenuous transcriptional threshold of Sertoli-cell specification in the bipotential gonadal ridge. We suggest that sex determination differs from canonical mechanisms of embryonic patterning and is a violation of Waddington’s principle. Sry has evolved to the edge of ambiguity.

5.5 Ways toward the evolutionary fate of SRY

Hundreds of SRY sequences from different species have been identified. Most of them encode typical domain deployments, N-terminal domain (NTD)-HMG box-C-terminal domain (CTD), whereas some rare Sry genes (recently found in

230

rodents) have a specific microsatellite CAG-repeating domain (Coward et al. 1994;

Bowles et al. 1999) at their 3’ end, and may also contain a shortened or deleted NTD.

As the previous chapter described, we found the NTD in human SRY, a typical SRY, enhances SRY-driven transcriptional activity by the phosphorylation of tandem serines, and mutations blocking phosphorylation (replace triple serines with triple alanines) significantly reduces related transcriptional activity (Fig. 2.6D). However, it is surprising that the gain of microsatellite repeat-encoding glutamine-rich motifs in typical SRY restores and normalizes SRY-mediating transcriptional activity (the central dogma regulation: SRY-Sox9 activation) even in different phosphorylation statuses (Fig. 4.6). Interestingly, the same phenomena are shown in the reverse experiment: adding the NTD from typical SRY to full-length mouse Sry protein which contains the CAG repeats-encoding glutamine-rich motifs on the C-terminus.

Moreover, these chimeras, with or without the NTD in various phosphorylation statuses, exhibit approximately the same extent of Sry-Sox9 activation (Fig. 4.6).

The compensatory role of the glutamine-rich motif highlighted by these results implies that this specific motif relaxes the requirement of phosphorylation in typical

SRY. Also, this motif normalizes the functions of transcriptional activity in SRY/Sry variants with different HMG boxes coupled with various phosphorylation statuses.

Thus, we believe that this glutamine-repeating region containing SRY variants is

231

independent of phosphorylation and hence the presence of the NTD, but also indicates that the specificity between the HMG boxes of various species is relaxed.

Together with the results described above, the HMG box only-transcriptional activity of human and mouse HMG boxes highlight the dramatic “normalizer” role of the glutamine-rich motif. The box-only results show that the activation function of the mouse HMG box is attenuated significantly (Fig. 4.5), which may be due to the higher dynamic box structure of mouse and the faster off-rate of the mouse box-DNA complex. However, native full-length mouse Sry shares a similar extent of transcriptional activation with native human SRY, and both exhibit indistinguishable functions in a phosphorylation independent manner with the glutamine-rich motif.

Thus, the adjoining domain (HMG box and phosphorylation region) in SRY with the glutamine-rich motif has a relaxed complex formation requirement and hence might raise the divergence of box function or even the entire Sry-driven mechanism.

To further study this hypothesis, we reinvestigated the evolutionary tree which contains these unique microsatellite repeats containing Sry genes, i.e. the muridae and cricetidae families, which are closely related (sister families). There are different

Sry evolutionary fates in muridae, including typical Sry genes (following the

NTD-HMG-CTD deployment), Sry genes with CAG microsatellite repeats (encoded protein known with shortened or deleted NTDs but containing various lengths or open

232

reading frames of CAG-repeating tracts), and strikingly, extinguished Sry genes along with extinguished Y chromosomes. We believe the invasion of microsatellite repeats within the Sry exon compensates and tolerates the degeneration of the joint function of the NTD-HMG box. Thus, the invaded microsatellite repeats in Sry might make the function of the NTD-HMG domain “hyper-divergent”. Nature supported this hypothesis with the Sry genes from Arvicola rodents in cricetidae (Megias-Nogales et al. 2003), and Hylomyscus in muridae (Lundrigan and Tucker 1994): these repeating tails (from microsatellite repeats) containing Sry genes have NTDs of various lengths.

Furthermore, the presence of the microsatellite repeats also might hold the key for conserving the Sry gene along with the Y chromosome. Serving as the most representative functional gene on the Y chromosome, Sry proteins with less functional efficiency of Sry-Sox9 regulation might force the organism to find another mechanism for sex determination. This degeneration of SRY process might be due to Y chromosome degeneration associated with its lack of recombination repair. This hypothesis potentially explains those species, including Transcaucasian mole vole

(cricetidae family), Ellobius lutescens (cricetidae family), Zaisan mole vole

(cricetidae family), Ellobius tancrei (cricetidae family), and the Japanese spiny country rats (muridae family), that have lost their entire Y chromosome (Arakawa et al. 2002; Just et al. 2007) and developed an unclear sex determination system. The

233

invasion of microsatellite repeats within Sry compensates and tolerates the degenerated Sry-encoding protein. Thus, these rescued Srys exhibit similar activity in Sry-Sox9 regulation and are associated with the same phenotype (male individuals) although their microsatellite repeats are of various lengths and open reading frames.

A dramatic example of this glutamine tail rescue is in the Okinawa spiny rat

(Tokudaia muenninki), the only member of its genus which has retained its Y chromosome and Sry mediated sex determination system. Remarkably, Tokudaia muenninki’s Sry contains several glutamine-rich tracts in its C-terminal tail which may have allowed it to retain sufficient function to overcome the degeneration of the

Y chromosome. The sufficient function of Sry favors the Y chromosome conservation.

234

Appendix I

Summary of inherited mutations in human SRY and state of characterizations Supplemental Table. Summary of Inherited Mutations in human SRY and State of Characterizationa

Referen Mut. Clinical Assess. Protein Pur. Protein Charact. Protein Partner Cellular Studies Transgenic ces

n

a

o

9

d

r

i

c

t

n

t

r

y

y

y

x

t

s

r

e

r

r

r

e

t

m

n

l

o

e

t

l

y

d

t

o

i

i

a

i

e

i

t

e

o

o

t

o

l

t

n

t

g

r

u

h

S

i

t

n

d

i

x

s

h

p

p

r

e

t

m

a

a

s

i

t

n

e

o

t

a

g

n

b

o

c

t

l

s

s

a

x

t

o

t

1

n

t

a

i

r

a

b

i

a

4

c

o

p

m

u

r

a

s

y

s

b

f

f

l

Y

y

u

P

a

p

c

-

l

o

r

E

t

g

a

a

r

n

e

a

e

f

r

I

n

M

e

i

D

b

c

o

g

e

M

a

s

r

s

r

p

e

b

R

t

r

r

o

a

b

S

s

t

d

- o

a

t

h

r

o

G

-

n

n

l

i

/

o

m

x

d -

a

o

a

G

R

S

n

i

u

n

r

t

c

h

x

i

b

l

e

a

r

C

C

O

i

m

i

e

e

I

/

v

a

r

h

E

A

P

e

d

M

p

e

C

F

h

l

l

e

t

E

g

d

M

e

e

-

t

b

a

s

C

t

n

t

c

c

d

c

i

O

a

s

h

l

n

o

N

p

H

i

t

o

e

a

o

S

u

o

u

h

n

u

d

r

a

D

l

h

S

m

n

W

B

R

o

E

n

l

N

N

I

i

P

F

M

P

e

T

E

G

C

f R30I (1,2)

g V60L (3,4)

h i j I90M (2,5-7)

R76S (8)

235 F109S (9)

Y127F (10)

L163X (11)

a For each inherited clinical mutation, results of prior studies are indicated by or, in the case of possible technical error, by . The results of the present study (and its antecedent in the J. Biol. Chem.) are indicated by . Artifactural conclusions in prior studies corrected in the present study are indicated by . Open spaces signify the absence of published data. Thus, aside from clinical genetics (gray zone at left), scant data are available. bPGD: pure gonadal dysgenesis cModel reporter: luciferase or analogous reporter gene assays using co-transfection of SRY variants and 5’-ATTGTT-dependent reporter plasmid. dModel reporter: luciferase or analogous reporter gene assays using co-transfection of SRY variants and human/mouse TESCO-dependent reporter plasmid. eChimeric SRY/Sry constructions: see Supplemental Figure S6 in companion paper (Chen, Y.-S., Racca, J.D., Phillips, N.B., & Weiss, M.A. Proc. Natl. Acad. Sci. USA (2013)) fStudies employing an artificial TESCO-regulated reporter gene employed described in ref. 2 (below) suggest that inherited mutation R30I exhibits native transcriptional activity, which poses a paradox in light of the proband’s phenotype and data presented in ref. 1 (below). gThe results of ref. 3 indicate the V60L abolishes detectable specific DNA binding as probed by a gel mobility-shift assay (see Appendix), which suggests a transcriptional threshold of >50 in contradiction to the factor of two obtained here in accord with studies of mouse models. The present in vivo and in vitro data demonstrate that the specific DNA-binding affinity of V60L is similar to that of wild-type SRY. h,iThe results of ref. 3 indicate that I90M markedly attenuates specific DNA binding, which is inconsistent with subsequent studies by this and other laboratories (ref 2). Our present in vivo and in vitro data demonstrate that the specific DNA-binding affinity of I90M is similar to that of wild-type SRY. jThe TESCO-dependent reporter gene assay employed in ref. 2 indicated that I90M SRY exhibits enhanced transcriptional activity (see Appendix). The present studies suggest that this result is likely to be an artifact of TF over-expression. Whereas enhanced activity makes sex reversal difficult to understand, the present study resolves this paradox through plasmid dilution and multifactorial characterization of cellular biochemistry, including endogenous Sox9 activation, endogenous TESCO ChIP, analysis of nucleocytoplasmic shuffting, phosphorylation, proteosomal degradation, and and Wnt/b-catenin signaling (multiple green boxes above).

235

References

1. Assumpcao JG, et al. (2002) Novel mutations affecting SRY DNA-binding activity: the

HMG box N65H associated with 46,XY pure gonadal dysgenesis and the familial

non-HMG box R30I associated with variable phenotypes.J Mol Med 80: 782-790.

2. Knower KC, et al. (2011) Failure of SOX9 regulation in 46XY disorders of sex

development with SRY, SOX9 and SF1 mutations. PLoS One 6:e17751.

3. Harley VR, et al.(1992) DNA binding activity of recombinant SRY from normal males

and XY females.Science 255: 453-455.

4. Phillips NB, et al. (2011) Mammalian Testis-determining Factor SRY and the Enigma of

Inherited Human Sex Reversal. J BiolChem 286:36787-36807.

5. Hawkins JR, et al.(1992) Mutational analysis of SRY: nonsense and missense mutations

in XY sex reversal.Human Genetics 88: 471-474.

6. Pontiggia, A. et al. (1994) Sex-reversing mutations affect the architecture of SRY-DNA

complexes. EMBO 13(24):6115-6124.

7. Dörk T. et al (1998) Independent observation of SRY mutation I90M in a patient with

complete gonadal dysgenesis.Human Mutat 11:90-91.

8. Imai A, et al. (1999) A novel sex-determining region on Y (SRY) missense mutation

identified in a 46, XY female and also in the father. Endocrine Jour 46(5):735-739.

9. Jager RJ, et al. (1992) A familial mutation in the testis-determining gene SRY shared by

both sexes.Human genetics 90:250-355.

10. Jordan BK,et al. (2002) Familial mutation in the testis-determining gene SRY shared by

an XY female and her normal father. Jour Clin Endo Metab 87(7):3428-3432.

11. Tajima T, et al.(1994) A novel mutation localized in the 3’ non-HMG box region of the

SRY gene in 46, XY gonadal dysgenesis. Human Mole Gene 3(7) 1187-1189.

236

Appendix II

Model: CH34 cell

Analysis of the altered regulatory properties of inherited SRY variants provides an opportunity to define its threshold biochemical activities; i.e., the extent of perturbation at the edge of function as defined by variable developmental outcomes.

To this end, a functional assay was provided by a ras-immortalized cellular model of the rodent pre-Sertoli cell micro-dissected from the XY gonadal ridge just prior to the onset of Sry expression (Haqq et al. 1994; Haqq and Donahoe 1998; Li et al. 2006).

We are grateful to Prof. Patricia Donahoe and former members of her laboratory (Drs.

Christopher Haqq and Trent Clarke) for kindly providing the CH34 cell line.

Embryonic CH34 cells provide a model of the XY pre-Sertoli cell in which SRY structure-function relationships may be probed in a developmentally appropriate context. Of particular utility, transient transfection of human SRY or chimeric human-mouse constructs activates an endogenous downstream gene-regulatory network (GRN). The resulting pattern of immediate- and delayed gene activation in part recapitulates male-specific developmental regulation of gene expression (Bullejos and Koopman 2005; Park et al. 2011), including robust activation of principal Sry target gene Sox9 (Sekido et al. 2004) and in turn Sox9 target genes Amh/Mis and

Ptgd2 (De Santa Barbara et al. 1998; Wilhelm et al. 2007). GRN regulation requires

(a) specific SRY-DNA recognition as indicated by control studies of non-DNA-binding variants (King and Weiss 1993; Weiss et al. 1997) and (b) nuclear localization as indicated by control studies of NLS variants (Li et al. 2001; Gontan et al. 2009). The extent of activation correlates with SRY occupancy of TESCO sites in the testis-specific enhancer of Sox9 (Sekido and Lovell-Badge 2008) as probed by

ChIP. In these cells SRY functions as a transcriptional activator of Sox9 and as an 237

inhibitor of Wnt/-catenin signaling, presumably via protein-protein interactions

(Sekido and Lovell-Badge 2008; Tamashiro et al. 2008).

238

Appendix III

Floppy Sox

Mutations V60L, V60A, and I90M involve nonpolar substitutions within the hydrophobic cores of the minor (V60) or major (I90) wings. Among globular proteins in general, such substitutions are often observed as neutral species variants

(e.g., among alignments of globin sequences (Kapp et al. 1995)), reflecting the robustness of folding to small changes in side-chain volume (Matthews 1996) and multiple solutions to the problem of core packing (Bowie et al. 1990). Indeed, V60L,

V60A, and I90M each permit native or near-native specific DNA binding and DNA bending (Table 2.1 in the main text). Despite such in vitro findings, V60 and I90 are invariant among mammalian Sry alleles (Gasca et al. 2002; Phillips et al. 2011). Such conservation suggests hidden constraints not apparent in the three-dimensional structure of the SRY-DNA complex (Bewley et al. 1998; Murphy et al. 2001; Phillips et al. 2006). Such constraints are related in part to the flexibility of the sequence-specific HMG box (Love et al. 2004). SRY provides a prototype of “floppy

SOX” (Weiss 2001). The flexibility of the free SRY HMG box affects structure-activity relationships.

Minor wing. In the SRY-DNA complex the side chain of V60 packs within an

“aromatic box” formed by the conserved side chains of H120, Y124, and Y127

(consensus residues 65, 69, and 72 in the HMG box). This cluster of side chains defines the mini-core of the minor wing (Werner et al. 1995; Murphy et al. 2001). In the free domain, however, the minor wing is not well ordered (van Houte et al. 1995;

Weiss 2001), and so substitutions V60L and V60A do not affect its stability (Table 1 in main text; (Phillips et al. 2011)). Such flexibility may facilitate engagement of the 239

N-terminal -strand of the free HMG box (including V60) in protein-protein interactions such as binding to Exportin-4, previously shown to function in nuclear import (Gontan et al. 2009) via a bipartite NLS (VKRPMNAFIVWSRDQRRK; residues 60-77 in human SRY) (Poulat et al. 1995). Although V60L and V60A do not affect the in vitro stability of the SRY domain, these mutations are associated with accelerated proteosome-mediated degradation (Fig. 2.3). The responsible mechanism has not been established. In principle the mutations may either enhance protein-protein interactions leading to ubiquitinylation or interfere with a protein-protein interaction that protects from such degradation. A candidate partner protein that could mask the N-terminal -strand is CaM, previously implicated in the mechanism of SRY nuclear import (Sim et al. 2005). We have found, however, that

V60L and V60A do not affect the binding of SRY to CaM (see Fig. 2.5 and

Supplemental Fig. S2.10). It is possible that altered subcellular localization may in itself lead to differences in rates of proteosome-mediated degradation.

Major wing. I90 (consensus position 35 in the HMG box) packs in the major-wing core against the side chains of W70, W98, and L101 (consensus positions W15, W43, and L46 in the HMG box). Substitution I90M preserves the nonpolar character of the side chain, but molecular modeling suggests that local reorganization of the core would be required to avoid steric clash. Indeed, this mutation causes a small decrease in the stability of the free domain without change in specific DNA binding or DNA bending (Table 2.1). Despite such preservation of DNA-binding properties, core reorganization is transmitted to the protein-DNA interface as indicated by altered overall pattern of 1H-NMR chemical shifts in the bent DNA site. Despite such structural effects on the free and bound domain, I90M SRY exhibits an unperturbed

240

cellular lifetime (Fig. 2.3).

I90M, despite its internal location, impairs binding of SRY to CRM1, a mediator of nuclear export (see Fig. 2.4) (Fornerod et al. 1997). Although such impairment may reflect an indirect effect of the mutation on the structure of an overlying protein surface, it is also possible that partial transient unfolding of the

HMG box enables I90 to directly contact CRM1. However counter intuitive, the direct model would rationalize the internal positions of each of the four nonpolar residues comprising the putative NES of SRY (consensus positions I35, L39, M45, and L46 of the HMG box) and homologous NES of SOX domains (Gasca et al. 2002). The biophysical plausibility of the direct model is strengthened by the substantial thermal

o unfolding of the SRY HMG box at physiological temperatures (Tmid 40 C; Table 2.1): an equilibrium between folded, partially folded, and unfolded major wings would enable transient exposure of NES residues as a cryptic CRM1-binding surface.

o Because I90M lowers the Tmid of the HMG box to 36 C, it is possible that the defect in CRM1 binding is in part ameliorated by an increase in the fraction unfolded

(relative to wild-type SRY) at body temperature. (We note in passing that box position 45 is divergent in rodent Sry alleles, which by analogy to I90M may account for their more prominent nuclear localization (Sekido 2010).)

Floppy Sox and nucleocytoplasmic trafficking. We envisage that the flexibility of each wing of the HMG box (an apparent “defect” in protein folding) facilitates nucleocytoplasmic trafficking and so represents an evolved property subject to specific selection. Given that non-SOX-related HMG boxes have been characterized with greater structural and thermodynamic stability (Broadhurst et al. 1995), it would be of future interest to identify stabilizing substitutions which may be found in SRY

241

extrinsic to NLS or NES sequences as a means to test whether such an “improvement” might indirectly impair trafficking by delimiting flexibility and hence efficiency of conformational capture by Exportin-4, CRM1, or other components of the nuclear import/export machinery.

242

Appendix IV

Summary of human XY intersex syndromes

The following genetic classification of disorders of sexual development is provided as an aid to biochemists and structural biologists generally interested in transcriptional regulation but without a specific background in this field of human genetics. Because our Discussion of the tenuous transcriptional threshold of SRY seeks to interpret this result in the broader context of the genetic program of gonadogenesis (see main text), the following terms are pertinent.

Swyer's Syndrome. Patients with 46, XY pure gonadal dysgenesis due to deletion or mutation of SRY (located on the short arm of the Y chromosome) exhibit a female somatic phenotype with internal streak gonads. Missense mutations most often arise de novo in paternal spermatogenesis. The present study focuses on a subset of

Swyer’s Syndrome in which the mutations are inherited. In either case the patients exhibit primary amenorrhea but with egg donation, in vitro fertilization, and hormonal support may give birth. Although mutations in additional genes may give rise to similar phenotypes, diverse mutations in SRY (clustering predominantly in the HMG box; see Fig. 2.1A in main text) are a major cause and ordinarily associated with the least extent of testicular differentiation. The majority of mutations raise de novo in spermatogenesis. The association of SRY mutation V60A with ovotestes (rather than pure dysgenesis) is unusual. Translocation of SRY to the X chromosome or to an autosome can give rise to 46, XX males (de la Chapelle Syndrome).

Mutations in WT1. Wilm’s tumor protein 1 (WT1), located on human chromosome

11p13, encodes a zinc finger transcription factor. WT1 expression precedes and is 243

required for the function of SRY in the differentiating gonadal ridge. Homozygous disruption of Wt1 in mice leads to renal agenesis and gonadal failure. Mutations in

WT1 are associated with distinct dominant syndromes depending on the type and location of the genetic alteration. Denys-Drash Syndrome (DDS; due to heterozygous point mutations in the zinc fingers) is characterized by the triad of Wilm’s tumor, pseudohermaphroditism, and mesangial renal sclerosis. Frazier Syndrome (FS; due to altered RNA splicing and in some cases to point mutations in the zinc fingers) is a urogenital anomaly associated with gonadoblastoma. XY patients with FS may also exhibit gonadal dysgenesis and a broad spectrum of intersexual phenotypes. The

WGAR syndrome (due to deletions on chromosome 11p13 spanning WT1, PAX6, and other genes) is characterized by aniridia (A in WGAR), genitourinary malformations

(G), and mental retardation (R) with high risk of Wilm’s tumor (W) and/or gonadoblastoma (G).

Mutations in SF1. Chromosome 9p24.3 contains gene NR5A1, which encodes steroidogenic factor 1 (SF1; also designated nuclear receptor subfamily 5, group A, member 1). SF1, a transcription factor with orphan ligand-binding domain, regulates genes in gonadal development and adrenal steroidogenesis. Missense, nonsense, and micro-deletion mutations in SF1 are associated with XY sex reversal, absence of puberty, and infertility, with or without adrenocortical insufficiency. Patients are usually heterozygous. Gene dosage without biochemical dominance is proposed to cause a pure gonadal phenotype as a syndrome of haploinsufficiency whereas poison polypeptides are proposed to underlie the more severe gonadal-adrenal syndrome.

Rare homozygous patients have been observed with mild missense mutations inherited from unaffected carriers. Chromatin IP assays have provided evidence that

244

Sf1 and Sry in mice regulate Sox9 through joint binding to a testis-specific enhancer element (TES; core sites TESCO). A case report has described a 46, XX female with a heterozygous mutation in SF1 who exhibited normal ovarian development in the setting of primary adrenal failure.

Dosage-Sensitive Sex Reversal. DAX-1 (encoding orphan nuclear receptor NROB1) is a gene on the short (p) arm of the X chromosome between intervals 21.3 and 21.2.

The protein lacks a recognizable DNA-binding domain. In an XX fetus DAX-1 acts as a dominant-negative regulator of SRY and SF1, leading to dosage-sensitive sex reversal. It is unclear whether such duplication is sufficient. Mutations compromising the expression or function of DAX-1 are associated with X-linked syndrome adrenal hypoplasia congenita, which in males leads to cryptorchidism, delayed puberty, and infertility. Females are rarely affected.

Campomelic Dysplasia. A syndrome of congenital abnormalities of cartilage and bone development, sometimes also associated in 46, XY patients with gonadal dysgenesis, due to heterozygous mutations in SOX9 on human chromosome 17.

Patients typically exhibit distinctive faces with large head, small chin, prominent eyes, and flat face; Pierre-Robin dysmorphias include cleft palate, glossoptosis (posterior displacement of the tongue), and micrognathia (small lower jaw). Weakened cartilage in the upper respiratory tract leads to laryngotracheomalacia. About 75% of 46, XY patients exhibit some degree of intersex, ranging from ambiguous genitalia to a

Swyer-like female phenotype; gonads may exhibit pure dysgenesis or exhibit ovotestes.

245

Chromosome 9p Monosomy (Afi Syndrome). A complex set of developmental disorders due to deletions on the short arm of chromosome 9 spanning 9p24, which contains DMRT1, DMRT2, and DMRT3, part of a vertebrate family of genes related by the DM motif to genes encoding Drosophila transcription factor Doublesex and nematode transcription factor Mab3. Depending on the extent of 9p deletion, this syndrome can be associated with delayed psychomotor development, camptodactyly, craniofacial dysmorphism, autism, mental retardation, brain atrophy, cardiovascular and intestinal abnormalities. Approximately 70% of 46, XY patients exhibit partial or complete sex reversal ranging from mild hypospadias to complete gonadal dysgenesis. Rarely 46, XY patients with this syndrome exhibit ovotestes with primitive ovarian histology. XX females with 9p monosomy can exhibit subclinical, mild or marked hypergonadotrophic hypogonadism (i.e., from normally functioning to severely dysgenetic ovaries).

Partial or Complete Androgen Insensitivity Syndrome (PAIS or CAIS). Mutation or deletion of the X-linked gene encoding the androgen receptor (AR; nuclear receptor subfamily 3, group C, member 4 (NR3C4)) can lead to partial or complete loss of biological response to testosterone. Patients ordinarily have 46, XY karyotypes.

Complete loss of AR function leads to the syndrome of testicular feminization with female external genitalia and absence of internal female structures coincide with intra-abdominal testes. Point mutations in the AR with partial retention of biochemical functions are associated with a broad range of phenotypes. Rare XX patients bearing compound heterozygous mutations in AR are minimally affected and fertile, homozygous mutations have not been reported.

246

Appendix V Methods in Detail

Chromatin immunoprecipitation

Following transient transfection by wild-type or variant SRY constructs, CH34 cells were treated with MG132 as described in main text, and subjected to the investigation of chromatin immunoprecipitation (Wittmann et al. 2005). After recovery, the cells were cross-linked by addition of 1% formaldehyde (0.5 ml per well); the plates were then rotated for 15 min at 37 oC. The reactions were quenched by addition of 0.125

M glycine at 37 oC for 5 min. After centrifugation at 2300g for 5 min at 4 oC, pellets were rinsed twice with cold PBS. The cells were then lysed with 1 ml of cell-lysis buffer (5 mM Pipes (pH 8.0), 8.5 mM KCl, and 0.5% NP40; Sigma-Aldrich, Inc., St.

Louis, MO) containing a protease-inhibitor cocktail (Hoffmann LaRoche, Ltd., Nutley,

NJ). The lysates were then centrifuged, and pellets resuspended into nuclei-lysis buffer (50 mM Tris-HCl (pH 8.0) containing 10 mM EDTA and 1% SDS;

Sigma-Aldrich, Inc., St. Louis, MO) supplemented by the protease inhibitor cocktail.

Chromatin lysates were sonicated to generate 300-400-bp fragments using tapered microtips during seven 10-sec bursts separated by 1 min rest periods. Chromatin fragments (50 l) were diluted into IP buffer (16.7 mM Tris-HCl (pH 8.0), 167mM

NaCl, 1.2mM EDTA, 0.01% SDS, and 1.1% Triton X-100; Sigma-Aldrich, Inc., St.

Louis, MO) to a final volume of 1 ml. This step was followed by pre-clearing for 1 247

hour at 4 oC by addition of 80 l of a Protein A/sperm DNA-tRNA mixture solution

(comprised of (a) 1 ml Protein A slurry (Santa Cruz, Inc., Santa Cruz, CA), (b) 10 mg/ml salmon sperm DNA (Clontech Laboratories, Inc., Mountain View, CA), and (c)

24 l of 10 mg/ml tRNA (Hoffmann LaRoche, Ltd., Nutley, NJ)). The pre-cleared supernatants were incubated overnight at 4 oC with 5 g of anti-HA antiserum

(Sigma-Aldrich, Inc., St. Louis, MO); a non-specific antiserum from Santa Cruz

Biotechnologies (Santa Cruz, CA) served as control. Complexes were then incubated with 60 l of Protein A slurry for 1 hour at 4 oC. Slurry beads were centrifuged at 800g and washed at 4 oC with the following buffers in order: (1) IP dilution buffer, (2) TSE wash buffer (20mM Tris-HCl, 150mM NaCl, 2mM EDTA,

0.1% SDS, 1% Triton X-100), (3) LiCl wash buffer (10mM Tris-HCl, 250mM LiCl,

0.5% NP-40, 0.5% deoxycholate, 1mM EDTA) and (4) TE (Tris-HCl, EDTA) wash

buffer. Elution solution (50 mM NaHCO3 and 1% SDS) was used twice to elute complexes from beads for 15 min at room temperature. This was followed by a reversal of cross-linking at 65 oC overnight at a final NaCl concentration of 200 mM.

After treatment with proteinase K and RNase (Hoffmann LaRoche, Ltd., Nutley, NJ) for 1 hour at 45 oC and extraction by 1:1 phenol-CIAA solution (the latter contained a

24:1 vol/vol solution of chloroform:isoamyl alcohol), solutions were precipitated by ice-cold ethanol. This was followed by elution in 20 l of deionized distilled water.

248

An expanded high-fidelity PCR protocol was provided by the vender (Hoffmann

LaRoche, Ltd., Nutley, NJ). Primers are listed as follow: set a, (forward) 5’-GGAACTCCAACTACGTAC-3’ and

(reverse) 5’-CCTGTAGTTGGTAGCTGC-3’; set b, (forward) 5’-ATCTCTACAGCTGACTTC-3’ and

(reverse) 5’-TAGCTGGGCTCATATCG-3’; set c (forward) 5’-CTGAGAGCAATCTGAGC-3’ and

(reverse) 5’-CACACCGTGCAAATGTA-3’.

Real-time quantitative reverse-transcriptase PCR protocol

RT-Q-rtPCR was employed to evaluate SRY-mediated transcriptional activation of endogenous male-specific genes in rodent gonadal cell-line CH34. Although Sox9 served as the major probe in this study, three groups (including Sox family, sex-related, and control sets) were also investigated as shown in Supplemental Figures S7 and S8

(see main text for protocol). The following sets of primers were used.

Sox2: 5′-GCCGAGTGGAAACTTTTGTCG and

5′-CGGGAAGCGTGTACTTATCCT

Sox3: 5’-AGCGCCTGGACACGTACAC and

5’-ATGTCGTAGCGGTGCATCT

249

Sox4: 5’- GCAAGAAAAGAAGCCAAGCT and

5’-TGACCAAGAGGCAAAATAAAATCAA

Sox8: 5’-CAGAGCTCAGCAAGACCCT and

5’-GGGTGGTGGCCCAGTT

Sox9: 5’-AGCACTCCGGGCAATCT and

5’-CGGCAGGTATTGGTCAAACT

Sox10: 5’-CAAGGAGGGGCTGCTGCTAT and

5’-ATGGCTCTGGCCTGAGGGGT

Sox17: 5’-CGGTTTCCACGCTCAGCCC and

5’-GTCGGACACCACGGAGGA

HA-tagged SRY: 5’-CAGGATCCTATCCATATGACGTT and

5’-TCCTGGACGTTGCCTTTACT

Amh: 5’-CTATTTGGTGCTGACTGTGCACTT and

5’-AAGGCCTGCAGCTGAGCGAT

Fgf9: 5’-CTATCCAGGGAACCAGGAAAGA and

5’-TCGTTCATGCCGAGGTAGAGT

Ptgds: 5’-GATTTCCACAGACC and

5’-CAGTAGCTCTTTCTTCT

Sry: 5’-TTATGGTGTGGTCCCGTGGA and

250

5’-GGCCTTTTTTCGGCTTCTGT

Wt1: 5’-CGGTCCGACCACCTGAAGAC and

5’-GTTGTGATGGCGGACTAATT

Sf1: 5’-AGCAGAAGAAAGCACAGATTCG and

5’-TAGGGGGTAACATGTAGTC

Wnt5a: 5’-CGCACGAGAAAGGGAACGAATC and

5’-TTACAGGCTACATCTGCCAGGTT

Dmrt1: 5’-GTGCCTGCTCAGACTGGAAACC and

5’-GATCTGGGACATGCTCTGAC

Gata4: 5’-CCTGCGAGACACCCCAATCTC and

5’-AGGTAGTGTCCTGTCCCATCT

GAPDH: 5’-GACATGCCGCCTGGAGAA and

5’- GCCCAGGATGCCCTTTAGT

-actin: 5’- AGCCCAGGATGCCCTTTAGT and

5’- AGCCCAGGATGCCCTTTAGT

TFIID: 5’-CTGAGGGGGCAATGTCTAAC and

5’-GGGCAGCTAGTGAGATGAGC

18S-rRNA: 5’- TTGATTAAGTCCCTGCCCTTT and

5’-CGATCCGAGGGCCTAACT

251

GMSA-based DNA-binding assay

Binding assays were carried out by GMSA using a 36-bp consensus DNA site

(5’-CATACTGCGGGGGTGATTGTTCAGGATCATACTGCG-3’; consensus target site underlined) as described. In brief, the DNA fragments purchased from Oligos,

Etc. (Wilsonville, OR) were labeled with [γ-33P]-ATP in one strand (ICN Biomedicals,

Inc., Irvine, CA). The reaction employed 0-100 nM protein and less than 1 nM

33P-labeled DNA in a buffer containing 10 mM potassium phosphate (pH 7.0), 30

μg/ml bovine serum albumin (BSA), 50 mM KCl, 2.5 mM MgCl2, 5% glycerol, 4 mM dithiothreitol (DTT), 0.7 ng/l poly-dI.dC; the binding reactions were incubated in

12.5 µl reactions for one hour on ice. Only specific binding is observed under these conditions at these protein concentrations tested. Polyacrylamide gels (8%) containing

0.45X Tris-borate-ethylenediamine-tetra-acetic acid buffer (TBE) were pre-run for

45-60 min at 10 V/cm at 4 oC. After loading the samples, gels were run at 15 V/cm for one hour at 4 or 25 oC; the temperature was kept constant by a circulating refrigerated water bath.

SRY-calmodulin binding biophysical assays in vitro

Binding of SRY to calmodulin (CaM; obtained from Sigma-Aldridge, Inc. St. Louis,

252

MO) was monitored by intrinsic Trp fluorescence as described by Harley and colleagues using an ATF 105 fluorimeter (Aviv Biomedical, Lakewood, NJ) and

GMSA as described above. Trp fluorescence spectra were obtained with 2 µM SRY

in 25 mM Tris-HCl (pH 7.0) containing 100 mM KCl and 50 mM CaCl2. For measurements with CaM complexes, 2 µM SRY aliquots were incubated with 2 µM

CaM for 30 min in the same buffer as above at 4 ºC prior to obtaining spectra.

Structures of wild-type and variant SRY/CaM complexes were also probed by CD using an Aviv spectropolarimeter. Free SRY, CaM, and their complexes were made

30 M in 25 mM Tris-HCl (pH 7.0) containing 0.1 M KCl and 50 mM CaCl2; the solutions were incubated at 4 oC for 30 min before spectra were obtained at 20 ºC.

SRY-calmodulin binding GMSA-based assay in vitro

A GMSA-biochemical assay for the binding of the SRY HMG box to CaM was obtained based on competition between such binding and SRY-DNA binding; the protocol was a modification of that described by Harley, V. R., and colleagues.

GMSA employed a 33P-labeled 36-bp DNA site containing a single central consensus site (5’-ATTGTT-3’ and complement) as described below. Incubation buffer

contained 140 mM KCl, 20 mM Tris-HCl (pH 6.8), 1 mM CaCl2, 5 mM MgCl2, 4 mM DTT, 0.25 ng/l poly-dI.dC, and 10% glycerol. Samples were incubated at 4 oC

253

for 1 hour before being loaded onto a 0.5X tris-borate gel; gels were run at 300V for

45-60 min. Bands were visualized by scanning the wet gel with a PhosphoImager

(Molecular Dynamics, Inc.). SRY or variants were either (i) pre-complexed and incubated with DNA prior to the addition of increasing [CaM] or (ii) pre-complexed with the same concentrations of CaM from 0-20 M prior to the addition of DNA.

The concentration of the wild-type or variant SRY domain (V60L and V60A) was 16 nM; the concentration of 33P-labeled DNA was less than 1 nM. Under these conditions the percent shifted protein-DNA complex in the absence of CaM was ca.

80%, allowing visualization of both free and bound bands. Competitor concentrations of CaM tested were 0 nM, 78 nM, 312 nM, 625 nM, 1.25 M, 2.50

M, 5, 10 M, and 20 M.

Circular Dichroism

Far-ultraviolet CD spectra were obtained in a 1-mm path-length quartz cuvette and protein concentration of ca. 25 M. Thermal unfolding curves at 222 nm were obtained from 4-90 oC at 2.5 °C increments as described. Thermal unfolding of the equimolar protein-DNA complexes was likewise monitored using 12-bp DNA duplex

(5’-GTGATTGTTCAG-3’ and complement; target site underlined).

254

Thermodynamic studies of protein stability

Fractional unfolding of SRY domains was investigated by CD (at helix-sensitive wavelength 222 nm) or intrinsic Trp fluorescence as a function of the concentration of guanidine-HCl as described (Phillips et al. 2011). Fluorescence-detected guanidine titrations were monitored at an emission wavelength of 390 nm following excitation at

270 nm at 4 oC. Studies were performed in 140 mM KCl and 10 mM potassium phosphate (pH 7.4) at a protein concentration of 25 M (CD) or 1 M (fluorescence).

Steady-state fluorescence resonance energy transfer

Steady-state FRET was employed to probe protein-directed DNA bending by wild-type or variant SRY HMG boxes; the DNA site consisted of

5’-FAM/TAMRA-labeled 15-bp DNA duplex containing a consensus SRY target site as described (Phillips et al. 2011). The Forster distance of this system lies between the respective donor/acceptor distances in bent and unbent DNA sites, enhancing the sensitivity of the assay to perturbations in the HMG box. Flexible linkage of the dyes to the respective 5’-ends of the upper and lower strands justifies use of the conventional biophysical assumption that 2 equals 0.67. Preparations of three samples of labeled DNA were: (i) containing 5’-fluorescein tethered to the

5’-phosphate of the upper strand by a hexanyl linker (6-FAM); (ii)

255

tetramethylrhodamine tethered to the 5’-phosphate of the lower strand by an analogous linker (TAMRA); and (iii) a double-labeled derivative. HPLC-purified oligonucleotides were from Qiagen Operon, (Alameda, CA) and purified further by

HPLC after annealing using an ion-exchange semipreparative HPLC column (DNA

PAC PA-100, Dionex Corp., Sunnyvale, CA). For FRET study, the duplex DNA probes were made 15 μM (stock solution) in 10 mM potassium phosphate and 10 mM

Tris-HCl (pH 8.4) containing 140 mM KCl, 1 mM EDTA, and 1 mM DTT. The ratio of labeling in the double-labeled sample is close to 1:1.

Stopped-flow kinetic FRET studies

A double-mixing stopped-flow apparatus with a thermo-electric temperature controller was used to measure rates of protein-DNA dissociation by means of competitive stopped-flow FRET. Data were collected using an Aviv ATF 105 spectrofluorometer (Aviv Biomedical, Lakewood, NJ). Kinetic experiments were performed by (a) filling one syringe with a 1:1 complex of 1 M of SRY domain and a 15-bp DNA duplex (in FRET buffer) containing the consensus DNA sequence (see above) labeled at respective 5’ ends with donor and acceptor; and (b) filling the other syringe with a solution containing a 20-fold excess of the free DNA target site (12 bp) lacking fluorescent labels. On mixing, competitive displacement of the bound SRY

256

domain from labeled DNA to unlabeled DNA led to progressive attenuation of the

FRET signal. Fluorescence emission was monitored at 520 nm following excitation at

490 nm. The data were fit to single-exponential equation to obtain observed rate

constants. The reported koff values (see main text and caption to Supplemental Fig.

S5) represent the average of three traces; kon values were calculated from observed koff

and Kd values according to equation: Kd=koff/kon.

FRET-based dissociation constants

Measurements were made at 15 oC. Varying concentrations of wild-type or variant

SRY domains were titrated in FRET buffer (above) at a constant DNA concentration of 25 nM. Fluorescence emission spectra from 500-650 nm were recorded after

excitation at 490 nm. Kd values were determined by plotting changes in fluorescence intensity at 520 nm (ordinate) against total protein concentration (abscissa). Data were fit to a single-site ligand-binding equation as implemented in program Origin 8.0

(OriginLab Corp., Northampton, MA):

where ∆F is the change in donor fluorescence on addition of the SRY domain to DNA

relative to the free DNA fluorescence; ∆F0 is the maximum fluorescence change

obtained when the DNA is fully bound; Kd, (dissociation constant), D0, (DNA 257

concentration), and S (protein concentration).

NMR spectroscopy of SRY-DNA complexes

1H-NMR studies of the I90M SRY domain and its interactions with a consensus DNA site were undertaken at 500 MHz as described (Phillips et al. 2011). The studies employed a 15-bp DNA duplex containing variant core element 5’-TTTGTG-3’ and complement; this core site (which differs from the consensus at two positions and corresponds to a Sox8/9 target site in the promoter of AMH/MIS) was utilized by

Clore and colleagues in multi-dimensional NMR studies of a specific SRY tailed-HMG box/DNA complex (Murphy et al. 2001). The titration employed successive protein:DNA stoichiometries of 2:7, 4:7, 6:7, and 7:7 (1:1). Spectra were obtained in 50 mM KCl and 10 mM potassium phosphate (pH 6.0) and 37 oC.

Resonance assignments were obtained as described by King, C.-Y. and Weiss, M. A.

(King and Weiss 1993).

258

Appendix VI

Screening an appropriate cell models

Today, we have large cell banks for use in cell biology studies. Over the last ten years the topic of how to choose the proper cell model became an important task and necessary to producing reliable data. Available technologies and facilities limited what type of cell models we could maintain and operate. Therefore, we needed to identify the goal and the most important approach in our study and then find appropriate cells.

The core of our study was to observe and characterize a transcription factor

(SRY), and monitor the related downstream gene-regulatory network (initiated by

SRY-SOX9 regulation). Based on this goal, we welcomed any cell type able to express SRY variants and have significant downstream gene regulation responses to exogenous SRY. Although many types of cell lines share a similar basic mechanism, there are particular requirements that are present in only a few lines. Recent cell biology helped us to investigate the cell specification easily by identifying what cell type they were (epithelial, endothelial, fibroblast…..etc), or from which organ they came (bone, colon, kidney, ovary, testis….etc). Some of the most common commercial cell lines have been widely used in many cellular and molecular biology studies. Unfortunately, some important requirements for choosing a cell model,

259

especially in the notoriously difficult field of SRY, are easily overlooked. For studies related to clinical birth defects, animal species and cell stages are very important. In SRY studies, the sex of the cell model (male) will be another critical factor. For example, the human embryonic kidney cell line, HEK 293T, appears to be a very good choice, however, the presence of multiple X chromosomes and the lack of the Y chromosome suggests that the source of the cell was female (Shaw et al.

2002). This cell line also lacks endogenous expression of SF1, which regulates

SRY-triggered SOX9 activation (Sekido and Lovell-Badge 2008). Thus, the significant up-regulation of endogenous SOX9 in response to transfected SRY is not observed, and the luciferase assays using SRY and a 5’ATTGTT-activated reporter gene also need exogenous SF1. For these reasons, HEK 293T would be insufficient for use in our experiments.

Some cell types contain a good genetic environment for SRY-specific study, such as Hs 1.Tes cell line. Its original tissue is normal adult testis. This cell line expresses endogenous SF1 and every necessary male-specific factor. However, its endogenous SOX9 is not able to respond to exogenous SRY. The reason might be that the Hs 1.Tes cell line is not at an appropriate stage and its gene-regulatory network fails to be activated by SRY-mediated stimulation. These observations makes Hs 1.Tes cell a good model for investigating the well-developed male-specific

260

genetic environment, but not for the SRY-initiating gene regulation process.

The reasons that our cell model, CH34, is a good platform for SRY regulatory studies have been discussed in above appendix chapter (Appendix II). The table below organized the validity of SRY-related studies in several popular primate and rodent cell lines. This list provides references for choosing good cell models, and also motivates future studies probing the differences in chromatin organization of

SOX9, or further regulatory genes, that validate a potential cell model.

261

source cell lines description recommended study (gender)

fails to express full-length SRY (able to express 1-151 truncated version); monkey kidney tissue insensitive to SRY structural mutations; insignificant up-regulation of nuclear localization COS-7 (male) endogenous SOX9 in response to SRY; reporter gene assays require co-transfection reporter co-transfection of SF1. human colon cancer able to express full-length SRY; insensitive to SRY structural mutations; nuclear localization DLD-1 (male) insignificant up-regulation of endogenous SOX9 in response to SRY. co-transfection reporter HCT-116 human colon cancer able to express full-length SRY; insensitive to SRY structural mutations; nuclear localization (male) insignificant up-regulation of endogenous SOX9 in response to SRY. co-transfection reporter

262 able to express full-length SRY; structural mutations are degraded rapidly; no nuclear localization human embryonic kidney HEK 293T up-regulation of endogenous SOX9 was observed; co-transfection of SF1 is (female) co-transfection reporter required. a Leydig-like cell; able to express full-length SRY; sensitive to SRY structural human normal testis s 1.Tes mutations; up-regulation of endogenous SOX9 in response to SRY is nuclear localization (male) insignificant; co-transfection of SF1 is not helpful for the reporter gene assays. able to express full-length SRY; sensitive to SRY structural mutations; nuclear localization human testis cancer NT2-D1 endogenous SOX9 is inducible but low fold; endogenous SRY and SF1 are endogenous SOX9 (male) expressed. co-transfection reporter able to express full-length SRY; sensitive to SRY structural mutations; nuclear localization human prostate cancer PC-3 endogenous SOX9 is inducible; no endogenous SRY but SF1 is expressed. endogenous SOX9 (male) co-transfection reporter

262

human cervical cancer cells full-length SRY expression is unstable; no up-regulation of endogenous SOX9 HeLa - (female) was observed; unknown reporter gene assay compability. Chinese hamster ovary able to express full-length SRY; up-regulation of endogenous Sox9 is unknown; CHO co-transfection reporter (female) reporter gene assays require co-transfection of SF1.

263

263

Appendix VII

Improving the transient transfection protocol

Mammalian transcription factors regulating tissue- and stage-specific gene expression are ordinarily present at low abundance (1-100 nM representing 102-104 molecules per nucleus) (Goentoro et al. 2009). Studies in cell culture, however, often employ strong viral promoters (such as that derived from the cytomegalic virus (Qin et al. 2010)), which typically confer high-level transcription factor expression (1-10

M representing 105-106 molecules per nucleus). Such over-expression can alter quantitative patterns of gene expression and qualitative biological responses (Niwa et al. 2000). Our finding that I90M attenuates SRY function at an appropriate level of transcription factor expression (102-104 protein molecules per nucleus; the estimation is shown in Fig. 1 in this Appendix) stands in disagreement with co-transfection studies employing a reporter gene (luciferase) under the control of a human TES element (Knower et al. 2011). The latter studies demonstrated enhanced transcriptional activity of I90M SRY under conditions of its CMV-promoter-driven over-expression in human embryonal carcinoma cell line NT2-D1 (presumably due to enhanced nuclear accumulation) in accordance with the present studies in CH34 and

PC-3 cells the absence of plasmid dilution (Fig. 2.3; see also Fig. 2 in this Appendix).

We have also replicated the findings of Knower et al (Knower et al. 2011) in NT2-D1 cells following co-transfection of luciferase under the control of the minimal mouse

TESCO element (Fig. S2.13). In this cell line the protocol of Knower and colleagues leads in our hands to expression of ca. 1.5 million epitope-tagged SRY molecules per cell (Fig. 3 in this Appendix). Potential artifacts of transcription factor over-expression were mitigated through dilution of the hSRY expression plasmid

(driven by the strong CMV promoter) by the parent empty vector to obtain a level of 264

hSRY expression (102-104 molecules per cell) typical of factors that regulate cell-fate decisions (Goentoro et al. 2009).

265

Figure 1. Effect of Plasmid dilution on HA-tagged SRY expression in CH34 cells.

CH34 cells transfected without dilution (1X). Protein expression is reduced by

320-fold following 50-fold of plasmid dilution (maximum feasible dilution). Top, representative gel images of SRY and -tubulin; the panel was taken from a single gel and blot. Because of the greater abundance of tubulin, the total cellular protein extracts were diluted by 100-fold (100X) to obtain bands of similar intensity. Bottom, summary table providing SRY signal strength relative to -tubulin as a function of plasmid dilution.

Technical note. Plasmid dilution in CH34 cells enables an appropriate intracellular concentration of SRY to be achieved as follows. A typical mammalian cell contains

300 pg of total protein (Lodish, H., et al. (2000) Molecular Cell Biology. 4th Edition.

New York: W. H. Freeman & Co.), and has a tubulin content of 2.5-3.3% of total protein (Hiller and Weber 1978). Relative Western blot intensities enabled estimation of the SRY concentration following transient transfection (either without dilution (1X) or following 50-fold dilution by the empty parent vector (50X; maximum dilution)) by a four-step calculation. This procedure assumes that the affinities of the anti-HA

266

and anti-tubulin antisera are similar; our qualitative conclusions are robust to fivefold errors.

Step 1. We assumed that a typical mammalian cell contains by weight the following amount of tubulin: 300 x 10-12 grams x 3%= 9x10-12 grams;

Step 2. We next estimated the number of tubulin molecules (molecular weight 110 kDa) as [9 x 10-12 grams/110 x 103 grams per mole] x 6.02 x1023 molecules per mole =

5 x 107 molecules;

Step 3. We in turn estimated the number of transfected SRY molecules per cell in the

1X transfection using its Western-blot signal strength relative to -tubulin: 5 x

107/100 (the tubulin-blotting dilution ratio)/3 (relative band intensity)/32.6%

(transfection efficiency), which yields an estimate of 5 x 105 molecules (which are predominantly in the nucleus);

Step 4. Finally, we estimated the number of transfected SRY molecules per cell following 50-fold plasmid dilution using relative signal strength between 1X and 50X

HA-SRY Western blots: 5 x 105 molecules/320, which yields an estimate of between

1,000 and 2,000 molecules per cell. This degree of expression is within the middle of the range of cellular abundances expected of a lineage- and stage-specific transcription factor (102-104 molecules per nucleus).

267

Figure 2. Transient transfection of hSRY in PC-3 cells activates endogenous human

SOX9. (A) Histogram showing results of wild-type and variant rt-Q-PCR assays

(plasmid dose 1 g) with an empty vector and inactive hSRY variant I68A as negative controls. V60L and V60A attenuate SOX9 activation by ca. twofold in accordance with their relative activities in CH34 cells (see Fig. 2.2). At this high plasmid dose

I90M has 30% higher activity than wild-type hSRY; the p-value is 0.035. Statistical comparisons: p-value (*) < 0.05; “ns” indicates p-value > 0.05. (B) Western blots probing extent of total cellular expression of hSRY and variants. Top: similar SRY 268

band intensities are shown in each lane. Bottom: α-tubulin loading controls. We note that PC-3 cells, derived from a human prostate-cancer cell line, has a

Wnt/-catenin-responsive SOX9 gene (Wang et al. 2007) but lacks endogenous SRY

(Dasari et al. 2002). (C) Effect of intermediate plasmid dilution on SRY-dependent transcriptional activation of SOX9: undiluted expression plasmid as in panel A (“1X”; azure) versus 20-fold dilution by empty plasmid (“20X”; sky blue). Whereas such dilution does not affect the ca. twofold attenuation of the activities of V60L and V60A

SRY relative to wild-type SRY, the apparent increase in the activity of I90M SRY relative to wild-type SRY (as observed in panel A on transfection of undiluted plasmids) is mitigated in accordance with intermediate-dilution levels in CH34 cells.

This shared trend suggests that the apparent increase in the activity of I90M SRY (in undiluted transfections) represents in both cell lines an artifact of TF over-expression as indicated in the main text.

Technical note: Because the magnitude of wild-type SRY-dependent activation of the endogenous human SOX9 in PC-3 cells is smaller than the magnitude of corresponding SRY-dependent activation of the endogenous rat Sox9 gene in CH34 cells (presumably due to baseline differences between the two cell lines in stage- and tissue-specific patterns of gene expression and SOX9/Sox9 chromatin structure), it was not technically feasible to test the effects of further dilutions in PC-3 cells (i.e., >

20-fold). The increased efficiency of SRY-dependent Sox9 activation in CH34 cells by contrast enabled studies of 50-fold dilution (see main text). This technical difference highlights the utility of using a mammalian cell line that derives from the appropriate embryonic stage and site of hSRY/mSry expression as emphasized by Haqq, C.M. and

Donahoe, P.K..

269

Figure 3. Expression of HA-tagged SRY following transient transfection in cell line

NT2-D1. Cells were transfected without dilution (100 ng; “1X”) or following dilution with parent empty plasmid as indicated. Protein expression was reduced by 20-fold following 50-fold plasmid dilution (maximum feasible dilution). Top, representative gel images of SRY and -tubulin (housekeeping loading control); the panel was taken from a single gel and blot. Because of the greater abundance of tubulin, total cellular protein extracts were diluted by 50-fold (“50X”) to obtain bands intensities differing by ca. twofold. Bottom, summary table providing SRY signal strength relative to tubulin as a function of plasmid dilution.

Technical note. A calculation formally analogous to that outline in the caption to

Figure 1 indicates that under 1X transfection conditions in NT2-D1 cells (as employed in (Knower et al. 2011)) the abundance of HA-tagged SRY is ca. 1.5 million molecules per cell.

270

Bibliography

Albrecht KH, Young M, Washburn LL, Eicher EM. 2003. Sry expression level and

protein isoform differences play a role in abnormal testis development in

C57BL/6J mice carrying certain Sry alleles. Genetics 164: 277-288.

Arakawa Y, Nishida-Umehara C, Matsuda Y, Sutou S, Suzuki H. 2002.

X-chromosomal localization of mammalian Y-linked genes in two XO species

of the Ryukyu spiny rat. Cytogenet Genome Res 99: 303-309.

Atanesyan L, Gunther V, Dichtl B, Georgiev O, Schaffner W. 2012. Polyglutamine

tracts as modulators of transcriptional activation from yeast to mammals. Biol

Chem 393: 63-70.

Badenhorst D, Dobigny G, Adega F, Chaves R, O'Brien PC, Ferguson-Smith MA,

Waters PD, Robinson TJ. 2011. Chromosomal evolution in Rattini (Muridae,

Rodentia). Chromosome Res 19: 709-727.

Baek WY, Kim JE. 2011. Transcriptional regulation of bone formation. Front Biosci

(Schol Ed) 3: 126-135.

Baker ME. 2011. Insights from the structure of estrogen receptor into the evolution of

estrogens: implications for endocrine disruption. Biochem Pharmacol 82: 1-8.

Barabasi AL, Oltvai ZN. 2004. Network biology: understanding the cell's functional

organization. Nat Rev Genet 5: 101-113.

Barbaro M, Balsamo A, Anderlid BM, Myhre AG, Gennari M, Nicoletti A, Pittalis

MC, Oscarson M, Wedell A. 2009. Characterization of deletions at 9p

affecting the candidate regions for sex reversal and deletion 9p syndrome by

MLPA. Eur J Hum Genet 17: 1439-1447.

Barmina O, Kopp A. 2007. Sex-specific expression of a HOX gene associated with

rapid morphological evolution. Dev Biol 311: 277-286.

271

Baron-Cohen S, Lutchmaya S, Knickmeyer R. 2004. Prenatal testosterone in mind:

amniotic fluid studies. The MIT Press, Cambridge, MA.

Barrionuevo F, Scherer G. 2010. SOX E genes: SOX9 and SOX8 in mammalian testis

development. Int J Biochem Cell Biol 42: 433-436.

Beitel LK, Kazemi-Esfarjani P, Kaufman M, Lumbroso R, DiGeorge AM, Killinger

DW, Trifiro MA, Pinsky L. 1994. Substitution of arginine-839 by cysteine or

histidine in the androgen receptor causes different receptor phenotypes in

cultured cells and coordinate degrees of clinical androgen resistance. J Clin

Invest 94: 546-554.

Bennett NC, Gardiner RA, Hooper JD, Johnson DW, Gobe GC. 2010. Molecular cell

biology of androgen receptor signalling. Int J Biochem Cell Biol 42: 813-827.

Bergstrom DE, Young M, Albrecht KH, Eicher EM. 2000. Related function of mouse

SOX3, SOX9, and SRY HMG domains assayed by male sex determination.

Genesis 28: 111-124.

Bernard P, Sim H, Knower K, Vilain E, Harley V. 2008. Human SRY inhibits

-catenin-mediated transcription. Int J Biochem Cell Biol 40: 2889-2900.

Berta P, Hawkins JR, Sinclair AH, Taylor A, Griffiths BL, Goodfellow PN, Fellous M.

1990. Genetic evidence equating SRY and the testis-determining factor.

Nature 348: 448-450.

Beverdam A, Koopman P. 2006. Expression profiling of purified mouse gonadal

somatic cells during the critical time window of sex determination reveals

novel candidate genes for human sexual dysgenesis syndromes. Hum Mol

Genet 15: 417-431.

Bewley CA, Gronenborn AM, Clore GM. 1998. Minor groove-binding architectural

proteins: structure, function, and DNA recognition. Annu Rev Biophys Biomol

272

Struct 27: 105-131.

Bianchi NO. 2002. Akodon sex reversed females: the never ending story. Cytogenet

Genome Res 96: 60-65.

Bogarad LD, Deem MW. 1999. A hierarchical approach to protein molecular

evolution. Proc Natl Acad Sci U S A 96: 2591-2595.

Booth LN, Tuch BB, Johnson AD. 2010. Intercalation of a new tier of transcription

regulation into an ancient circuit. Nature 468: 959-963.

Bornberg-Bauer E, Beaussart F, Kummerfeld SK, Teichmann SA, Weiner J, 3rd. 2005.

The evolution of domain arrangements in proteins and interaction networks.

Cell Mol Life Sci 62: 435-445.

Bowie JU, Reidhaar-Olson JF, Lim WA, Sauer RT. 1990. Deciphering the message in

protein sequences: tolerance to amino acid substitutions. Science 247:

1306-1310.

Bowles J, Cooper L, Berkman J, Koopman P. 1999. SRY requires a CAG repeat

domain for male determination in Mus musculus. Nat Genet 22: 405-408.

Broadhurst RW, Hardman CH, Thomas JO, Laue ED. 1995. Backbone dynamics of

the A-domain of HMG1 as studied by 15N NMR spectroscopy. Biochemistry

34: 16608-16617.

Bullejos M, Koopman P. 2005. Delayed Sry and Sox9 expression in developing mouse

gonads underlies B6-YDOM sex reversal. Dev Biol 278: 473-481.

Bylund M, Andersson E, Novitch BG, Muhr J. 2003. Vertebrate neurogenesis is

counteracted by Sox1-3 activity. Nat Neurosci 6: 1162-1168.

Camargo S, Viana RL, Anteneodo C. 2012. Intermingled basins in coupled Lorenz

systems. Phys Rev E Stat Nonlin Soft Matter Phys 85: 036207.

Chang YF, Imam JS, Wilkinson MF. 2007. The nonsense-mediated decay RNA

273

surveillance pathway. Annu Rev Biochem 76: 51-74.

Charlesworth B. 2003. The organization and evolution of the human Y chromosome.

Genome Biol 4: 226.

Chen YS, Racca JD, Phillips NB, Weiss MA. 2013a. Inherited human sex reversal due

to impaired nucleocytoplasmic trafficking of SRY defines a male

transcriptional threshold. Proc Natl Acad Sci U S A in second review.

Chen YS, Racca JD, Phillips NB, Weiss MA. 2013b. A microsatellite-encoded domain

in rodent Sry functions as a genetic capacitor to enable the rapid evolution of

biological novelty Proc Natl Acad Sci U S A in second review.

Chu J, Zhang r, Zhao Z, Zou W, Yefei H, Qi Q, Zhang H, Wang J, Tao S, Liu X et al.

2002. Male fertility is compatible with an Arg840Cys substitution in the AR in

a large Chinese family affected with divergent phenotypes of AR insensitivity

syndrome. J Clin Endocrinol Metab 87: 347-351.

Civetta A, Singh RS. 1998. Sex-related genes, directional sexual selection, and

speciation. Mol Biol Evol 15: 901-909.

Clarkson MJ, Harley VR. 2002. Sex with two SOX on: SRY and SOX9 in testis

development. Trends Endocrinol Metab 13: 106-111.

Cooper MD, Alder MN. 2006. The evolution of adaptive immune systems. Cell 124:

815-822.

Coward P, Nagai K, Chen D, Thomas HD, Nagamine CM, Lau YF. 1994.

Polymorphism of a CAG trinucleotide repeat within Sry correlates with

B6.YDom sex reversal. Nat Genet 6: 245-250.

Darwin C, ed. 1871. The Descent of Man, and Selection in Relation to Sex. Appleton,

New York.

Dasari VK, Deng D, Perinchery G, Yeh CC, Dahiya R. 2002. DNA methylation

274

regulates the expression of Y chromosome specific genes in prostate cancer. J

Urol 167: 335-338.

De Santa Barbara P, Bonneaud N, Boizet B, Desclozeaux M, Moniot B, Sudbeck P,

Scherer G, Poulat F, Berta P. 1998. Direct interaction of SRY-related protein

SOX9 and steroidogenic factor 1 regulates transcription of the human

anti-Mullerian hormone gene. Mol Cell Biol 18: 6653-6665.

De Vries GJ, Rissman EF, Simerly RB, Yang LY, Scordalakes EM, Auger CJ, Swain A,

Lovell-Badge R, Burgoyne PS, Arnold AP. 2002. A model system for study of

sex chromosome effects on sexually dimorphic neural and behavioral traits. J

Neurosci 22: 9005-9014.

Denny P, Swift S, Connor F, Ashworth A. 1992. An SRY-related gene expressed

during spermatogenesis in the mouse encodes a sequence-specific

DNA-binding protein. EMBO J 11: 3705-3712.

Desclozeaux M, Poulat F, de Santa Barbara P, Capony JP, Turowski P, Jay P, Mejean

C, Moniot B, Boizet B, Berta P. 1998. Phosphorylation of an N-terminal motif

enhances DNA-binding activity of the human SRY protein. J Biol Chem 273:

7988-7995.

Dork T, Stuhrmann M, Miller K, Schmidtke J. 1998. Independent observation of SRY

mutation I90M in a patient with complete gonadal dysgenesis. Hum Mutat 11:

90-91.

Dubin RA, Ostrer H. 1994. Sry is a transcriptional activator. Mol Endocrinol 8:

1182-1192.

Dubois CL, Shih HP, Seymour PA, Patel NA, Behrmann JM, Ngo V, Sander M. 2011.

Sox9-haploinsufficiency causes glucose intolerance in mice. PLoS One 6:

e23131.

275

Dunah AW, Jeong H, Griffin A, Kim YM, Standaert DG, Hersch SM, Mouradian MM,

Young AB, Tanese N, Krainc D. 2002. Sp1 and TAFII130 transcriptional

activity disrupted in early Huntington's disease. Science 296: 2238-2243.

Eicher EM, Shown EP, Washburn LL. 1995. Sex reversal in C57BL/6J-YPOS mice

corrected by a Sry transgene. Philos Trans R Soc Lond B Biol Sci 350:

263-268; discussion 268-269.

Eicher EM, Washburn LL. 2001. Does one gene determine whether a C57BL/6J-YPOS

mouse will develop as a female or as an hermaphrodite? J Exp Zool 290:

322-326.

Eicher EM, Washburn LL, Schork NJ, Lee BK, Shown EP, Xu X, Dredge RD, Pringle

MJ, Page DC. 1996. Sex-determining genes on mouse autosomes identified by

linkage analysis of C57BL/6J-YPOS sex reversal. Nat Genet 14: 206-209.

Eldredge N, Gould SJ. 1972. Punctuated equilibria: an alternative to phyletic

gradualism. in Models in Paleobiology (ed. TMJ Schopf), pp. 82-115.

Freeman Cooper, San Francisco.

Evans BAJ, Hughes IA, Bevan CL, Patterson MN, Gregory JW. 1997. Phenotypic

diversity in siblings with partial androgen insensitivity syndrome. Arch Dis

Child 76: 529-531.

Ezaz T, Stiglec R, Veyrunes F, Marshall Graves JA. 2006. Relationships between

vertebrate ZW and XY sex chromosome systems. Curr Biol 16: R736-743.

Ferguson-Smith MA, Trifonov V. 2007. Mammalian karyotype evolution. Nat Rev

Genet 8: 950-962.

Fornerod M, Ohno M, Yoshida M, Mattaj IW. 1997. CRM1 is an export receptor for

lecucine-rich nuclear export signals. Cell 90: 1051-1060.

Gasca S, Canizares J, De Santa Barbara P, Mejean C, Poulat F, Berta P,

276

Boizet-Bonhoure B. 2002. A nuclear export signal within the high mobility

group domain regulates the nucleocytoplasmic translocation of SOX9 during

sexual determination. Proc Natl Acad Sci U S A 99: 11199-111204.

Gatewood JD, Wills A, Shetty S, Xu J, Arnold AP, Burgoyne PS, Rissman EF. 2006.

Sex chromosome complement and gonadal sex influence aggressive and

parental behaviors in mice. J Neurosci 26: 2335-2342.

Gibson JR, Chippindale AK, Rice WR. 2002. The X chromosome is a hot spot for

sexually antagonistic fitness variation. Proc Biol Sci 269: 499-505.

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: 3368-3372.

Goentoro L, Shoval O, Kirschner MW, Alon U. 2009. The incoherent feedforward

loop can provide fold-change detection in gene regulation. Mol Cell 36:

894-899.

Gontan C, Guttler T, Engelen E, Demmers J, Fornerod M, Grosveld FG, Tibboel D,

Gorlich D, Poot RA, Rottier RJ. 2009. Exportin 4 mediates a novel nuclear

import pathway for Sox family transcription factors. J Cell Biol 185: 27-34.

Graves JA. 2002. Evolution of the testis-determining gene--the rise and fall of SRY.

Novartis Found SympS 244: 86-97; discussion 97-101, 203-106, 253-107.

Graves JA. 2004. The degenerate Y chromosome--can conversion save it? Reprod

Fertil Dev 16: 527-534.

Graves JAM, Shetty S. 2000. Comparative genomics of vertebrates and the evolution

of sex chromosomes. in Comparative Genomics (ed. M Clark), pp. 153-207.

Kluwer Academic Publishers, Norwell, MA.

Griffiths R, Tiwari B. 1993. Primers for the differential amplification of the

277

sex-determining region Y gene in a range of mammal species. Mol Ecol 2:

405-406.

Gubbay J, Vivian N, Economou A, Jackson D, Goodfellow P, Lovell-Badge R. 1992.

Inverted repeat structure of the Sry locus in mice. Proc Natl Acad Sci U S A 89:

7953-7957.

Guth SI, Wegner M. 2008. Having it both ways: Sox protein function between

conservation and innovation. Cell Mol Life Sci 65: 3000-3018.

Haqq CM, Donahoe PK. 1998. Regulation of sexual dimorphism in mammals.

Physiol Rev 78: 1-33.

Haqq CM, King CY, Ukiyama E, Falsafi S, Haqq TN, Donahoe PK, Weiss MA. 1994.

Molecular basis of mammalian sexual determination: activation of Müllerian

inhibiting substance gene expression by SRY. Science 266: 1494-1500.

Harley VR, Jackson DI, Hextall PJ, Hawkins JR, Berkovitz GD, Sockanathan S,

Lovell-Badge R, Goodfellow PN. 1992. DNA binding activity of recombinant

SRY from normal males and XY females. Science 255: 453-456.

Hawkins JR, Taylor A, Berta P, Levilliers J, Van der Auwera B, Goodfellow PN. 1992.

Mutational analysis of SRY: nonsense and missense mutations in XY sex

reversal. Hum Genet 88: 471-474.

Hersmus R, de Leeuw BH, Stoop H, Bernard P, van Doorn HC, Bruggenwirth HT,

Drop SL, Oosterhuis JW, Harley VR, Looijenga LH. 2009. A novel SRY

missense mutation affecting nuclear import in a 46,XY female patient with

bilateral gonadoblastoma. Eur J Hum Genet 17: 1642-1649.

Hiller G, Weber K. 1978. Radioimmunoassay for tubulin: a quantitative comparison of

the tubulin content of different established tissue culture cells and tissues. Cell

14: 795-804.

278

Hiort O, Gramss B, Klauber GT. 1995. True hermaphroditism with 46, XY karyotype

and a point mutation in the SRY gene. J Pediatr 126: 1022.

Hodgkin J. 2005. Karyotype, ploidy, and gene dosage. WormBook: 1-9.

Huang B, Wang S, Ning Y, Lamb AN, Bartley J. 1999. Autosomal XX sex reversal

caused by duplication of SOX9. Am J Med Genet 87: 349-353.

Huchon D, Madsen O, Sibbald MJ, Ament K, Stanhope MJ, Catzeflis F, de Jong WW,

Douzery EJ. 2002. Rodent phylogeny and a timescale for the evolution of

Glires: evidence from an extensive taxon sampling using three nuclear genes.

Mol Biol Evol 19: 1053-1065.

Hughes JF, Skaletsky H, Brown LG, Pyntikova T, Graves T, Fulton RS, Dugan S,

Ding Y, Buhay CJ, Kremitzki C et al. 2012. Strict evolutionary conservation

followed rapid gene loss on human and rhesus Y chromosomes. Nature 483:

82-86.

Imasaki K, Hasegawa T, Okabe T, Sakai Y, Haji M, Takayanagi R, Nawata H. 1994.

Single amino acid substitution (840ArgHis) in the hormone-binding domain

of the androgen receptor leads to incomplete androgen insensitivity syndrome

associated with a thermolabile androgen receptor. Eur J Endocrinol 130:

569-574.

Jäger RJ, Harley VR, Pfeiffer RA, Goodfellow PN, Scherer G. 1992. A familial

mutation in the testis-determining gene SRY shared by both sexes. Hum Genet

90: 350-355.

Jones DN, Searles MA, Shaw GL, Churchill ME, Ner SS, Keeler J, Travers AA,

Neuhaus D. 1994. The solution structure and dynamics of the DNA-binding

domain of HMG-D from Drosophila melanogaster. Structure 2: 609-627.

Jordan BK, Jain M, Natarajan S, Frasier SD, Vilain E. 2002. Familial mutation in the

279

testis-determining gene SRY shared by an XY female and her normal father. J

Clin Endocrinol Metab 87: 3428-3432.

Josso N, Picard JY, Rey R, di Clemente N. 2006. Testicular anti-Mullerian hormone:

history, genetics, regulation and clinical applications. Pediatr Endocrinol Rev

3: 347-358.

Just W, Baumstark A, Suss A, Graphodatsky A, Rens W, Schӓfer N, Bakloushinskaya

I, Hameister H, Vogel W. 2007. Ellobius lutescens: sex determination and sex

chromosome. Sex Dev 1: 211-221.

Kaighn ME, Narayan KS, Ohnuki Y, Lechner JF, Jones LW. 1979. Establishment and

characterization of a human prostatic carcinoma cell line (PC-3). Invest Urol

17: 16-23.

Kalderon D, Roberts BL, Richardson WD, Smith AE. 1984. A short amino acid

sequence able to specify nuclear location. Cell 39: 499-509.

Kamachi Y, Uchikawa M, Kondoh H. 2000. Pairing SOX off: With partners in the

regulation of embryonic development. Trends Genet 16: 182-187.

Kao HY, Verdel A, Tsai CC, Simon C, Juguilon H, Khochbin S. 2001. Mechanism for

nucleocytoplasmic shuttling of histone deacetylase. J Biol Chem 276:

47496-47507.

Kapp OH, Moens L, Vanfleteren J, Trotman CN, Suzuki T, Vinogradov SN. 1995.

Alignment of 700 globin sequences: extent of amino acid substitution and its

correlation with variation in volume. Protein Sci 4: 2179-2190.

Kashimada K, Koopman P. 2010. SRY: the master switch in mammalian sex

determination. Development 137: 3921-3930.

Kasimiotis H, Myers MA, Argentaro A, Mertin S, Fida S, Ferraro T, Olsson J, Rowley

MJ, Harley VR. 2000. Sex-determining region Y-related protein SOX13 is a

280

diabetes autoantigen expressed in pancreatic islets. Diabetes 49: 555-561.

Katoh K, Miyata T. 1999. A heuristic approach of maximum likelihood method for

inferring phylogenetic tree and an application to the mammalian SOX-3 origin

of the testis-determining gene SRY. FEBS Lett 463: 129-132.

Keren H, Lev-Maor G, Ast G. 2010. Alternative splicing and evolution: diversification,

exon definition and function. Nat Rev Genet 11: 345-355.

King CY, Weiss MA. 1993. The SRY high-mobility-group box recognizes DNA by

partial intercalation in the minor groove: a topological mechanism of sequence

specificity. Proc Natl Acad Sci U S A 90: 11990-11994.

Knower KC, Kelly S, Harley VR. 2003. Turning on the male--SRY, SOX9 and sex

determination in mammals. Cytogenet Genome Res 101: 185-198.

Knower KC, Kelly S, Ludbrook LM, Bagheri-Fam S, Sim H, Bernard P, Sekido R,

Lovell-Badge R, Harley VR. 2011. Failure of SOX9 regulation in 46XY

disorders of sex development with SRY, SOX9 and SF1 mutations. PLoS One

6: e17751.

Knower KC, Sim H, McClive PJ, Bowles J, Koopman P, Sinclair AH, Harley VR.

2007. Characterisation of urogenital ridge gene expression in the human

embryonal carcinoma cell line NT2/D1. Sex Dev 1: 114-126.

Kobayashi Y, Nakamura M, Sunobe T, Usami T, Kobayashi T, Manabe H,

Paul-Prasanth B, Suzuki N, Nagahama Y. 2009. Sex change in the Gobiid fish

is mediated through rapid switching of gonadotropin receptors from ovarian to

testicular portion or vice versa. Endocrinology 150: 1503-1511.

Koopman P. 1995. The molecular biology of SRY and its role in sex determination in

mammals. Reprod Fertil Dev 7: 713-722.

Koopman P. 1999. Sry and Sox9: mammalian testis-determining genes. Cell Mol Life

281

Sci 55: 839-856.

Koopman P. 2001. Sry, Sox9 and mammalian sex determination. EXS 91: 25-56.

Koopman P, Bullejos M, Bowles J. 2001. Regulation of male sexual development by

Sry and Sox9. J Exp Zool 290: 463-474.

Koopman P, Gubbay J, Vivian N, Goodfellow P, Lovell-Badge R. 1991. Male

development of chromosomally female mice transgenic for Sry. Nature 351:

117-121.

Kuo PD, Eliasmith C. 2005. Integrating behavioral and neural data in a model of

zebrafish network interaction. Biol Cybern 93: 178-187. la Cour T, Kiemer L, Molgaard A, Gupta R, Skriver K, Brunak S. 2004. Analysis and

prediction of leucine-rich nuclear export signals. Protein Eng Des Sel 17:

527-536.

Levine M, Davidson EH. 2005. Gene regulatory networks for development. Proc Natl

Acad Sci U S A 102: 4936-4942.

Li B, Phillips NB, Jancso-Radek A, Ittah V, Singh R, Jones DN, Haas E, Weiss MA.

2006. SRY-directed DNA bending and human sex reversal: reassessment of a

clinical mutation uncovers a global coupling between the HMG box and its tail.

J Mol Biol 360: 310-328.

Li B, Zhang W, Chan G, Jancso-Radek A, Liu S, Weiss MA. 2001. Human sex

reversal due to impaired nuclear localization of SRY. A clinical correlation. J

Biol Chem 276: 46480-46484.

Li WH, Ellsworth DL, Krushkal J, Chang BH, Hewett-Emmett D. 1996. Rates of

nucleotide substitution in primates and rodents and the generation-time effect

hypothesis. Mol Phylogenet Evol 5: 182-187.

Lin L, Philibert P, Ferraz-de-Souza B, Kelberman D, Homfray T, Albanese A, Molini

282

V, Sebire NJ, Einaudi S, Conway GS et al. 2007. Heterozygous missense

mutations in steroidogenic factor 1 (SF1/Ad4BP, NR5A1) are associated with

46,XY disorders of sex development with normal adrenal function. J Clin

Endocrinol Metab 92: 991-999.

Lindquist S. 2009. Protein folding sculpting evolutionary change. Cold Spring Harb

Symp Quant Biol 74: 103-108.

Liu Y, Wilson SH. 2012. DNA base excision repair: a mechanism of trinucleotide

repeat expansion. Trends Biochem Sci 37: 162-172.

Love JJ, Li X, Chung J, Dyson HJ, Wright PE. 2004. The LEF-1 high-mobility group

domain undergoes a disorder-to-order transition upon formation of a complex

with cognate DNA. Biochemistry 43: 8725-8734.

Lovell-Badge R, Canning C, Sekido R. 2002. Sex-determining genes in mice: building

pathways. John Wiley & Sons Ltd., West Sussex, UK.

Lundrigan BL, Tucker PK. 1994. Tracing paternal ancestry in mice, using the

Y-linked, sex-determining locus, Sry. Mol Biol Evol 11: 483-492.

Maatouk DM, DiNapoli L, Alvers A, Parker KL, Taketo MM, Capel B. 2008.

Stabilization of -catenin in XY gonads causes male-to-female sex-reversal.

Hum Mol Genet 17: 2949-2955.

Malki S, Boizet-Bonhoure B, Poulat F. 2010. Shuttling of SOX proteins. Int J

Biochem Cell Biol 42: 411-416.

Malki S, Nef S, Notarnicola C, Thevenet L, Gasca S, Mejean C, Berta P, Poulat F,

Boizet-Bonhoure B. 2005. Prostaglandin D2 induces nuclear import of the

sex-determining factor SOX9 via its cAMP-PKA phosphorylation. EMBO J 24:

1798-1809.

Mallet D, Bretones P, Michel-Calemard L, Dijoud F, David M, Morel Y. 2004.

283

Gonadal dysgenesis without adrenal insufficiency in a 46, XY patient

heterozygous for the nonsense C16X mutation: a case of SF1

haploinsufficiency. J Clin Endocrinol Metab 89: 4829-4832.

Marcelli M, Zoppi S, Wilson CM, Griffin JE, McPhaul MJ. 1994. Amino acid

substitutions in the hormone-binding domain of the human androgen receptor

alter the stability of the hormone receptor complex. J Clin Invest 94:

1642-1650.

Marchal JA, Acosta MJ, Bullejos M, Díaz de la Guardia R, Sánchez A. 2003. Sex

chromosomes, sex determination, and sex-linked sequences in Microtidae.

Cytogenet Genome Res 101: 266-273.

Marin I, Baker BS. 1998. The evolutionary dynamics of sex determination. Science

281: 1990-1994.

Masel J, Siegal ML. 2009. Robustness: mechanisms and consequences. Trends Genet

25: 395-403.

Matthews BW. 1996. Structural and genetic analysis of the folding and function of T4

lysozyme. FASEB J 10: 35-41.

McClelland K, Bowles J, Koopman P. 2012. Male sex determination: insights into

molecular mechanisms. Asian J Androl 14: 164-171.

McPhie-Lalmansingh AA, Tejada LD, Weaver JL, Rissman EF. 2008. Sex

chromosome complement affects social interactions in mice. Horm Behav 54:

565-570.

Megias-Nogales B, Marchal JA, Acosta MJ, Bullejos M, Diaz de la Guardia R,

Sanchez A. 2003. Sex chromosomes pairing in two Arvicolidae species:

Microtus nivalis and Arvicola sapidus. Hereditas 138: 114-121.

Michala L, Goswami D, Creighton SM, Conway GS. 2008. Swyer syndrome:

284

presentation and outcomes. BJOG 115: 737-741.

Miller KE, Lundrigan BL, Tucker PK. 1995. Length variation of CAG repeats in Sry

across populations of Mus domesticus. Mamm Genome 6: 206-208.

Miyamoto Y, Taniguchi H, Hamel F, Silversides DW, Viger RS. 2008. A GATA4/WT1

cooperation regulates transcription of genes required for mammalian sex

determination and differentiation. BMC Mol Biol 9: 44.

Moniot B, Declosmenil F, Barrionuevo F, Scherer G, Aritake K, Malki S, Marzi L,

Cohen-Solal A, Georg I, Klattig J et al. 2009. The PGD2 pathway,

independently of FGF9, amplifies SOX9 activity in Sertoli cells during male

sexual differentiation. Development 136: 1813-1821.

Monod J. 1972. Chance and Necessity: An Essay on the Natural Philosophy of

Modern Biology. Vintage Books United Kingdom.

Murata C, Yamada F, Kawauchi N, Matsuda Y, Kuroiwa A. 2010. Multiple copies of

SRY on the large Y chromosome of the Okinawa spiny rat, Tokudaia

muenninki. Chromosome Res 18: 623-634.

Murphy EC, Zhurkin VB, Louis JM, Cornilescu G, Clore GM. 2001. Structural basis

for SRY-dependent 46-X,Y sex reversal: modulation of DNA bending by a

naturally occuring point mutation. J Mol Biol 312: 481-499.

Nagamine CM, Taketo T, Koo GC. 1987. Morphological development of the mouse

gonad in tda-1 XY sex reversal. Differentiation 33: 214-222.

Ner SS. 1992. HMGs everywhere. Curr Biol 2: 208-210.

Niwa H, Miyazaki J, Smith AG. 2000. Quantitative expression of Oct-3/4 defines

differentiation, dedifferentiation or self-renewal of ES cells. Nat Genet 24:

372-376.

Nott A, Le Hir H, Moore MJ. 2004. Splicing enhances translation in mammalian cells:

285

an additional function of the exon junction complex. Genes Dev 18: 210-222.

Nykter M, Price ND, Aldana M, Ramsey SA, Kauffman SA, Hood LE, Yli-Harja O,

Shmulevich I. 2008. Gene expression dynamics in the macrophage exhibit

criticality. Proc Natl Acad Sci U S A 105: 1897-1900.

Page DC, Bieker K, Brown LG, Hinton S, Lathrop M, Nystrom-Lahti M, de la

Chapelle A, White R. 1987. Linkage, physical mapping, and DNA sequence

analysis of pseudoautosomal loci on the human X and Y chromosomes.

Genomics 1: 243-256.

Pamilo P, O'Neill RJ. 1997. Evolution of the Sry genes. Mol Biol Evol 14: 49-55.

Pannetier M, Tilly G, Kocer A, Hudrisier M, Renault L, Chesnais N, Costa J, Le

Provost F, Vaiman D, Vilotte JL et al. 2006. Goat SRY induces testis

development in XX transgenic mice. FEBS Lett 580: 3715-3720.

Park S, Zeidan K, Shin JS, Taketo T. 2011. SRY upregulation of SOX9 is inefficient

and delayed, allowing ovarian differentiation, in the B6.Y(TIR) gonad.

Differentiation 82: 18-27.

Pearson CE, Nichol Edamura K, Cleary JD. 2005. Repeat instability: mechanisms of

dynamic mutations. Nat Rev Genet 6: 729-742.

Persengiev SP, Saffer JD, Kilpatrick DL. 1995. An alternatively spliced form of the

transcription factor Sp1 containing only a single glutamine-rich transactivation

domain. Proc Natl Acad Sci U S A 92: 9107-9111.

Peters R, King CY, Ukiyama E, Falsafi S, Donahoe PK, Weiss MA. 1995. An SRY

mutation causing human sex reversal resolves a general mechanism of

structure-specific DNA recognition: application to the four-way DNA junction.

Biochemistry 14: 4569-4576.

Phillips NB, Jancso-Radek A, Ittah V, Singh R, Chan G, Haas E, Weiss MA. 2006.

286

SRY and human sex determination: The basic tail of the HMG box functions

as a kinetic clamp to augment DNA bending. J Mol Biol 358: 172-192.

Phillips NB, Nikolskaya T, Jancso-Radek A, Ittah V, Jiang F, Singh R, Haas E, Weiss

MA. 2004. Sry-directed sex reversal in transgenic mice is robust to enhanced

DNA bending: comparison of human and murine in HMG boxes. Biochemistry

43: 7066-7081.

Phillips NB, Racca J, Chen YS, Singh R, Jancso-Radek A, Radek JT, Wickramasinghe

NP, Haas E, Weiss MA. 2011. Mammilian testis-determining factor SRY and

the enigma of inherited human sex reversal. J Biol Chem 286: 36787-36807.

Piprek RP. 2009. Genetic mechanisms underlying male sex determination in mammals.

J Appl Genet 50: 347-360.

Polanco JC, Koopman P. 2007. Sry and the hesitant beginnings of male development.

Dev Biol 302: 13-24.

Pontiggia A, Rimini R, Harley VR, Goodfellow PN, Lovell-Badge R, Bianchi ME.

1994. Sex-reversing mutations affect the architecture of SRY-DNA complexes.

EMBO J 13: 6115-6124.

Pontiggia A, Whitfield S, Goodfellow PN, Lovell-Badge R, Bianchi ME. 1995.

Evolutionary conservation in the DNA-binding and -bending properties of

HMG-boxes from SRY proteins of primates. Gene 154: 277-280.

Poulat F, Girard F, Chevron MP, Goze C, Rebillard X, Calas B, Lamb N, Berta P.

1995. Nuclear localization of the testis determining gene product SRY. J Cell

Biol 128: 737-748.

Poux C, van Rheede T, Madsen O, de Jong WW. 2002. Sequence gaps join mice and

men: phylogenetic evidence from deletions in two proteins. Mol Biol Evol 19:

2035-2037.

287

Pritchett J, Athwal V, Roberts N, Hanley NA, Hanley KP. 2011. Understanding the

role of SOX9 in acquired diseases: lessons from development. Trends Mol

Med 17: 166-174.

Qin JY, Zhang L, Clift KL, Hulur I, Xiang AP, Ren B-Z, Lahn BT. 2010. Systematic

Comparison of Constitutive Promoters and the Doxycycline-Inducible

Promoter. PLoS One 5: e10611.

Raymond CS, Shamu CE, Shen MM, Seifert KJ, Hirsch B, Hodgkin J, Zarkower D.

1998. Evidence for evolutionary conservation of sex-determining genes.

Nature 391: 691-695.

Read CM, Cary PD, Crane-Robinson C, Driscoll PC, Norman DG. 1993. Solution

structure of a DNA-binding domain from HMG1. Nucleic Acids Res 21:

3427-3436.

Reenan RA. 2005. Molecular determinants and guided evolution of species-specific

RNA editing. Nature 434: 409-413.

Rhen T, Crews D. 1999. Embryonic temperature and gonadal sex organize

male-typical sexual and aggressive behavior in a lizard with

temperature-dependent sex determination. Endocrinology 140: 4501-4508.

Rossignol T, Dulau L, Julien A, Blondin B. 2003. Genome-wide monitoring of wine

yeast gene expression during alcoholic fermentation. Yeast 20: 1369-1385.

Rutherford SL, Lindquist S. 1998. Hsp90 as a capacitor for morphological evolution.

Nature 396: 336-342.

Sachidanandam R, Weissman D, Schmidt SC, Kakol JM, Stein LD, Marth G, Sherry S,

Mullikin JC, Mortimore BJ, Willey DL et al. 2001. A map of human genome

sequence variation containing 1.42 million single nucleotide polymorphisms.

Nature 409: 928-933.

288

Schafer AJ, Goodfellow PN. 1996. Sex determination in humans. Bioessays 18:

955-963.

Schlötterer C. 2000. Evolutionary dynamics of microsatellite DNA. Chromosoma 109:

365-371.

Sears KE, Goswami A, Flynn JJ, Niswander LA. 2007. The correlated evolution of

Runx2 tandem repeats, transcriptional activity, and facial length in carnivora.

Evol Dev 9: 555-565.

Seidman JG, Seidman C. 2002. Transcription factor haploinsufficiency: when half a

loaf is not enough. J Clin Invest 109: 451-455.

Sekido R. 2010. SRY: a transcriptional activator of mammalian testis determination.

Int J Biochem Cell Biol 42: 417-420.

Sekido R, Bar I, Narvaez V, Penny G, Lovell-Badge R. 2004. SOX9 is up-regulated

by the transient expression of SRY specifically in Sertoli cell precursors. Dev

Biol 274: 271-279.

Sekido R, Lovell-Badge R. 2008. Sex determination involves synergistic action of

SRY and SF1 on a specific Sox9 enhancer. Nature 453: 930-934.

Sekido R, Lovell-Badge R. 2009. Sex determination and SRY: down to a wink and a

nudge? Trends Genet 25: 19-29.

Seymour PA, Freude KK, Dubois CL, Shih HP, Patel NA, Sander M. 2008. A

dosage-dependent requirement for Sox9 in pancreatic endocrine cell formation.

Dev Biol 323: 19-30.

Shaw G, Morse S, Ararat M, Graham FL. 2002. Preferential transformation of human

neuronal cells by human adenoviruses and the origin of HEK 293 cells.

FASEB J 16: 869-871.

Siegal ML, Bergman A. 2002. Waddington's canalization revisited: developmental

289

stability and evolution. Proc Natl Acad Sci U S A 99: 10528-10532.

Siegal ML, Promislow DE, Bergman A. 2007. Functional and evolutionary inference

in gene networks: does topology matter? Genetica 129: 83-103.

Sim H, Argentaro A, Harley VR. 2008. Boys, girls and shuttling of SRY and SOX9.

Trends Endocrinol Metab 19: 213-222.

Sim H, Rimmer K, Kelly S, Ludbrook LM, Clayton AH, Harley VR. 2005. Defective

calmodulin-mediated nuclear transport of the sex-determining region of the Y

chromosome (SRY) in XY sex reversal. Mol Endocrinol 19: 1884-1892.

Sinclair AH, Berta P, Palmer MS, Hawkins JR, Griffiths BL, Smith MJ, Foster JW,

Frischauf AM, Lovell-Badge R, Goodfellow PN. 1990. A gene from the

human sex-determining region encodes a protein with homology to a

conserved DNA-binding motif. Nature 346: 240-244.

Sitnikova NA, Romanenko SA, O'Brien PC, Perelman PL, Fu B, Rubtsova NV,

Serdukova NA, Golenishchev FN, Trifonov VA, Ferguson-Smith MA et al.

2007. Chromosomal evolution of Arvicolinae (Cricetidae, Rodentia). I. The

genome homology of tundra vole, field vole, mouse and golden hamster

revealed by comparative chromosome painting. Chromosome Res 15:

447-456.

Smith JM, Koopman PA. 2004. The ins and outs of transcriptional control:

nucleocytoplasmic shuttling in development and disease. Trends Genet 20:

4-8.

Soskine M, Tawfik DS. 2010. Mutational effects and the evolution of new protein

functions. Nat Rev Genet 11: 572-582.

Soullier S, Hanni C, Catzeflis F, Berta P, Laudet V. 1998. Male sex determination in

the spiny rat Tokudaia osimensis (Rodentia: Muridae) is not Sry dependent.

290

Mamm Genome 9: 590-592.

Storrs-Mendez SI, Semlitsch RD. 2010. Intersex gonads in frogs: understanding the

time course of natural development and role of endocrine disruptors. J Exp

Zool B Mol Dev Evol 314: 57-66.

Su H, Lau YF. 1993. Identification of the transcriptional unit, structural organization,

and promoter sequence of the human sex-determining region Y (SRY) gene,

using a reverse genetic approach. Am J Hum Genet 52: 24-38.

Tamashiro DAA, Alarcon VB, Marikawa Y. 2008. Ectopic expression of mouse SRY

interferes with WNT/βcatenin signaling in mouse embryonal carcinoma cell

lines. Biochim Biophys Acta 1780: 1395-1402.

Taylor JP, Hardy J, Fischbeck KH. 2002. Toxic proteins in neurodegenerative disease.

Science 296: 1991-1995.

Thevenet L, Mejean C, Moniot B, Bonneaud N, Galeotti N, Aldrian-Herrada G, Poulat

F, Berta P, Benkirane M, Boizet-Bonhoure B. 2004. Regulation of human SRY

subcellular distribution by its acetylation/deacetylation. EMBO J 23:

3336-3345.

Tobaben S, Varoqueaux F, Brose N, Stahl B, Meyer G. 2003. A brain-specific isoform

of small glutamine-rich tetratricopeptide repeat-containing protein binds to

Hsc70 and the cysteine string protein. J Biol Chem 278: 38376-38383.

Tomiuk J, Bachmann L, Bauer C, Rolfs A, Schols L, Roos C, Zischler H, Schuler MM,

Bruntner S, Reiss O et al. 2007. Repeat expansion in spinocerebellar ataxia

type 17 alleles of the TATA-box binding protein gene: an evolutionary

approach. Eur J Hum Genet 15: 81-87.

Tucker PK, Lundrigan BL. 1993. Rapid evolution of the sex determining locus in Old

World mice and rats. Nature 364: 715-717.

291

Ukiyama E, Jancso-Radek A, Li B, Milos L, Zhang W, Phillips NB, Morikawa N,

King CY, Chan G, Haqq CM et al. 2001. SRY and architectural gene

regulation: the kinetic stability of a bent protein-DNA complex can regulate its

transcriptional potency. Mol Endocrinol 15: 363-377.

Uller T, Helantera H. 2011. From the origin of sex-determining factors to the

evolution of sex-determining systems. Q Rev Biol 86: 163-180. van Eyk CL, O'Keefe LV, Lawlor KT, Samaraweera SE, McLeod CJ, Price GR,

Venter DJ, Richards RI. 2011. Perturbation of the Akt/Gsk3-beta signalling

pathway is common to Drosophila expressing expanded untranslated CAG,

CUG and AUUCU repeat RNAs. Hum Mol Genet 20: 2783-2794. van Houte LP, Chuprina VP, van der Wetering M, Boelens R, Kaptein R, Clevers H.

1995. Solution structure of the sequence-specific HMG box of the lymphocyte

transcriptional activator Sox-4. J Biol Chem 270: 30516-30524.

Veyrunes F, Waters PD, Miethke P, Rens W, McMillan D, Alsop AE, Grutzner F,

Deakin JE, Whittington CM, Schatzkamer K et al. 2008. Bird-like sex

chromosomes of platypus imply recent origin of mammal sex chromosomes.

Genome Res 18: 965-973.

Vilain E, McCabe ER. 1998. Mammalian sex determination: from gonads to brain.

Mol Genet Metab 65: 74-84.

Vom Saal FS. 1981. Variation in phenotype due to random intrauterine psotioning of

male and female fetuses in rodents. J Reprod Fertil 62: 633-650.

Vom Saal FS. 1989. Sexual differentiation in litter-bearing mammals: influence of sex

of adjacent fetuses in utero. J Anim Sci 67: 1824-1840.

Vom Saal FS, Clark MM, Galef BG, Drickamer LC, Vandenbergh JG, ed. 1999.

Intrauterine position phenomenon. Academic Press, San Diego, Calif.,.

292

Waddington CH. 1959. Canalization of development and genetic assimilation of

acquired characters. Nature 183: 1654-1655.

Walker FO. 2007. Huntington's disease. Lancet 369: 218-228.

Wang H, McKnight NC, Zhang T, Lu ML, Balk SP, Yuan X. 2007. SOX9 is expressed

in normal prostate basal cells and regulates androgen receptor expression in

prostate cancer cells. Cancer Res 67: 528-536.

Washburn LL, Albrecht KH, Eicher EM. 2001. C57BL/6J-T-associated sex reversal in

mice is caused by reduced expression of a Mus domesticus Sry allele.

Genetics 158: 1675-1681.

Weir HM, Kraulis PJ, Hill CS, Raine AR, Laue ED, Thomas JO. 1993. Structure of

the HMG box motif in the B-domain of HMG1. EMBO J 12: 311-319.

Weiss MA. 2001. Floppy SOX: mutual induced fit in hmg (high-mobility group)

box-DNA recognition. Mol Endocrinol 15: 353-362.

Weiss MA. 2005. Molecular mechanisms of male sex determination: the enigma of

SRY. in DNA Conformation and Transcription (ed. T Ohyama), pp. 1-15.

Landes Bioscience, Georgetown, TX.

Weiss MA, Ukiyama E, King CY. 1997. The SRY cantilever motif discriminates

between sequence- and structure- specific DNA recognition: alanine

mutagenesis of an HMG box. J Biomol Struct Dyn 15: 177-184.

Werner MH, Bianchi ME, Gronenborn AM, Clore GM. 1995. NMR spectroscopic

analysis of the DNA conformation induced by the human testis determining

factor SRY. Biochemistry 34: 11998-12004.

Whitfield LS, Lovell-Badge R, Goodfellow PN. 1993. Rapid sequence evolution of

the mammalian sex-determining gene SRY. Nature 364: 713-715.

Wilhelm D, Hiramatsu R, Mizusaki H, Widjaja L, Combes AN, Kanai Y, Koopman P.

293

2007. SOX9 regulates prostaglandin D synthase gene transcription in vivo to

ensure testis development. J Biol Chem 282: 10553-10560.

Wilhelm D, Koopman P. 2006. The makings of maleness: towards an integrated view

of male sexual development. Nat Rev Genet 7: 620-631.

Wilhelm D, Martinson F, Bradford S, Wilson MJ, Combes AN, Beverdam A, Bowles

J, Mizusaki H, Koopman P. 2005. Sertoli cell differentiation is induced both

cell-autonomously and through prostaglandin signaling during mammalian sex

determination. Dev Biol 287: 111-124.

Wilhelm D, Washburn LL, Truong V, Fellous M, Eicher EM, Koopman P. 2009.

Antagonism of the testis- and ovary-determining pathways during ovotestis

development in mice. Mech Dev 126: 324-336.

Wilson DS, Wilson EO. 2007. Rethinking the theoretical foundation of sociobiology.

Q Rev Biol 82: 22.

Wilson MA, Makova KD. 2009. Genomic analyses of sex chromosome evolution.

Annu Rev Genomics Hum Genet 10: 333-354.

Wittmann BM, Fujinaga K, Deng H, Ogba N, Montano MM. 2005. The breast cell

growth inhibitor, estrogen down regulated gene 1, modulates a novel

functional interaction between estrogen receptor α and transcriptional

elongation factor cyclin T1. Oncogene 24: 5576-5588.

Wong E, Cuervo AM. 2010. Integration of clearance mechanisms: the proteasome and

autophagy. Cold Spring Harb Perspect Biol 2: a006734.

Wright E, Hargrave MR, Christiansen J, Cooper L, Kun J, Evans T, Gangadharan U,

Greenfield A, Koopman P. 1995. The Sry-related gene Sox9 is expressed

during chondrogenesis in mouse embryos. Nat Genet 9: 15-20.

294