UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

Studies on the transcriptional regulation of human sex hormone-binding globulin in testicular germ cells… Meyers, Warren 2016

Your browser doesn't seem to have a PDF viewer, please download the PDF to view this item.

Item Metadata


24-ubc_2016_september_meyers_warren.pdf [ 4.09MB ]
JSON: 24-1.0308668.json
JSON-LD: 24-1.0308668-ld.json
RDF/XML (Pretty): 24-1.0308668-rdf.xml
RDF/JSON: 24-1.0308668-rdf.json
Turtle: 24-1.0308668-turtle.txt
N-Triples: 24-1.0308668-rdf-ntriples.txt
Original Record: 24-1.0308668-source.json
Full Text

Full Text

   STUDIES ON THE TRANSCRIPTIONAL REGULATION OF HUMAN SEX  HORMONE-BINDING GLOBULIN IN TESTICULAR GERM CELLS:  MECHANISTIC AND EVOLUTIONARY INSIGHTS  by  WARREN MEYERS  B.Sc., University of Ottawa, 2011  A THESIS SUBMITTED IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF  MASTER OF SCIENCE in THE FACULTY OF GRADUATE AND POSTDOCTORAL STUDIES (Reproductive and Developmental Sciences)  THE UNIVERSITY OF BRITISH COLUMBIA (Vancouver)  August 2016  © Warren Meyers, 2016 ii  Abstract  In most mammals, the major site of sex hormone-binding globulin (SHBG) synthesis is the liver wherefrom it is secreted into the bloodstream and is the primary determinant of sex steroid access to target tissues. The minor site of SHBG synthesis is the testis and in lower mammals testicular SHBG has long been known to be synthesized and secreted by Sertoli cells. In contrast, human SHBG is expressed in testicular germ cells from an upstream alternative promoter (altP-SHBG).Transcripts arising from this region comprise an alternative first exon (1A) and the resultant protein is confined to the acrosomal compartment of the mature spermatozoa. In Chapters 3 and 4 I dissected the regulatory components of the alternative SHBG promoter and identified motifs that are required for optimal transcriptional activity from this region. Transcriptional activity is driven by two CACCC elements that appear to be functionally redundant. The transcription factor KLF4 interacts with promoter the region spanning these elements in vivo. Knockdown of Klf4 results in decreased altP-SHBG activity, while Klf4 overexpression relieves the effects of knockdown. Examination of KLF4 in testes of transgenic mice harbouring human SHBG transgenes reveals that Klf4 and SHBG have related expression patterns during the seminiferous cycle. The alternative SHBG promoter responds to dbcAMP treatment and this activation is relieved by cotreatment with the protein kinase A inhibitor, H89. Overexpression of CREM isoforms also upregulate altP-SHBG transcriptional activity. Examination testicular SHBG transcripts from members of each major primate family revealed that transcripts containing exon 1A are unique to the catarrhini phylogenetic group. In contrast SHBG transcripts in lemur testes contained exon 1, while no evidence for SHBG expression could be detected in marmoset monkey testes. In general, the exonic identity of primate testicular iii  SHBG transcripts could be predicted based on the structure of their gene’s 5’ regulatory region. This work provides insights into how human SHBG is expressed in a stage-dependent manner throughout the seminiferous cycle and how molecular evolution of higher primate SHBG genes has resulted in distinct changes in how it is expressed in their testes.   iv  Preface  All experiments in Chapters 3-5 were designed and carried out by myself under Dr. Hammond’s supervision. This thesis was also written entirely by myself with editorial input and final approval from Dr. Hammond.   Once completed, finalized versions of Chapters 3 and 5 are intended for publication. Part of the results from Chapter 3 were presented at the 41st Annual meeting of the American Society of Andrology in New Orleans, Louisiana in April 2016.  The secondary use of human serum samples, as described in Chapter 2, was approved by the University of British Columbia Clinical Research Ethics Board under the project title “SHBG: Beyond Plasma Transport” with the certificate number: H12-02373.  The acquisition of primate tissues and testis cDNA samples from collaborating institutions were done so in accordance with the regulations set forth by the Convention on International Trade in Endangered Species of Wild Fauna and Flora (CITES).             v  Table of Contents   Abstract ....................................................................................................................................... ii Preface ........................................................................................................................................ iv Table of Contents ........................................................................................................................ v List of Tables ............................................................................................................................. vii List of Figures .......................................................................................................................... viii List of Abbreviations ................................................................................................................... x Acknowledgements .................................................................................................................. xiii Chapter 1: General Introduction .................................................................................................. 1 1.1 The cellular basis of male fertility in mammals ................................................................ 1 1.2 The spermatogenic wave ................................................................................................... 3 1.3 Gene expression during spermatogenesis .......................................................................... 4 1.4 Hormonal regulation of male fertility ................................................................................ 6 1.5 Sex hormone-binding globulin (SHBG) ............................................................................ 8 1.6 Hypothesis and approach ................................................................................................. 14 Chapter 2: Materials and Methods ............................................................................................ 21 2.1 Animals ............................................................................................................................ 21 2.2 Cell lines and transient transfections ............................................................................... 21 2.3 Plasmids and reporter constructs ..................................................................................... 22 2.4 Transient transfection and luciferase reporter assay........................................................ 23 2.5 Small interfering RNA (siRNA) ...................................................................................... 24 2.6 Whole cell extracts and subcellular fractionation ............................................................ 25 2.7 Antibodies ........................................................................................................................ 27 2.8 Tissue fixation and chromatin immunoprecipitation ....................................................... 27 2.9 Tissue processing and immunohistochemistry/ immunofluorescence ............................ 29 2.10 Polyacrylamide gel electrophoresis (PAGE) and western blotting ............................... 30 2.11 Chemicals and reagents ................................................................................................. 31 2.12 RNA extraction and cDNA synthesis ............................................................................ 31 2.13 Primers and polymerase chain reaction (PCR) .............................................................. 32 2.14 Bioinformatics and phylogenetic analyses .................................................................... 32 vi  2.15 Non-human primate cDNA and tissue biopsies ............................................................ 33 2.16 Purification of antibodies specific to the N-terminal LG4 domain of human SHBG ... 34 Chapter 3: Transcriptional Activation from the Alternative SHBG Promoter by KLF4 .......... 49 3.1 Introduction ..................................................................................................................... 49 3.2 Results ............................................................................................................................. 50 3.3 Discussion and future directions ...................................................................................... 55 Chapter 4: Transcriptional Regulation from the Alternative Sex Hormone-Binding Globulin Promoter by protein kinase A and CREM ................................................................................ 72 4.1 Introduction ..................................................................................................................... 72 4.2 Results ............................................................................................................................. 73 4.3 Discussion and future directions ...................................................................................... 75 Chapter 5: Evolution of Sex Hormone-Binding Globulin gene expression in the primate testis ................................................................................................................................................... 85 5.1 Introduction ..................................................................................................................... 85 5.2 Results ............................................................................................................................. 87 5.3 Discussion and future directions ...................................................................................... 90 Chapter 6: Conclusions, future directions and potential roles for testicular SHBG in reproductive physiology .......................................................................................................... 111 6.1 Top-down and bottom-up approaches reveal multiple modes of transcriptional regulation from the human SHBG alternative promoter ....................................................................... 111 6.2 Distinct patterns of testicular SHBG expression across the major primate groups ....... 114 6.3 Potential roles for testicular SHBG in reproductive physiology ................................... 118 6.4 Final remarks ................................................................................................................. 127 References ............................................................................................................................... 130    vii  List of Tables  Table 2.1 Oligonucleotide primers used in this thesis. ................................................................. 40 Table 5.1 Sources of genomic SHBG/Shbg nucleotide sequences from the public database. ...... 98 Table 5.2 Summary of SHBG genomic and testicular transcript features in relation to seminiferous tubule organization and testicular SHBG location in primates. ............................ 108 Table 6.1 Relative binding affinities (RBA) of SHBG to selected sex steroids across mammals...................................................................................................................................................... 129   viii  List of Figures  Figure 1.1 Structure and principal components of the mammalian seminiferous tubule. ............ 16 Figure 1.2 The arrangement of seminiferous tubular stages in the mammalian testis. ................. 17 Figure 1.3 The mammalian hypothalamic-pituitary-testicular endocrine signalling axis. ........... 19 Figure 1.4 Overview of the human SHBG gene and its products. ................................................ 20 Figure 2.1 Chromatin shearing assay. ........................................................................................... 39 Figure 2.2 Selection of candidate antisera for affinity purification. ............................................. 42 Figure 2.3 Coupling of recombinant LG4 domain of human SHBG to NHS-activated Sepharose column........................................................................................................................................... 43 Figure 2.4 Affinity-purification of total IgG from candidate antisera. ......................................... 44 Figure 2.5 Affinity-purification of anti-LG4 antibodies. .............................................................. 45 Figure 2.6 Affinity-purification efficiency test on mouse serum. ................................................ 46 Figure 2.7 Performance test of affinity-purified antibodies on human serum. ............................. 47 Figure 2.8 Performance test of affinity-purified antibodies on mouse testicular homogenates. .. 48 Figure 3.1 Alternative SHBG transcription units. ......................................................................... 61 Figure 3.2 High transcriptional activity of the human alternative SHBiG promoter in GC-2 cells........................................................................................................................................................ 62 Figure 3.3 Dependence of altP-SHBG transcriptional activity on CACCC elements. ................. 63 Figure 3.4 Expression and localization of Klf4 in GC-2 cells. ..................................................... 64 Figure 3.5 High KLF4 levels are required for optimal altP-SHBG reporter activity in GC-2 cells........................................................................................................................................................ 65 Figure 3.6 KLF4 localizes to the alternative SHBG promoter in vivo. ......................................... 66 Figure 3.7 Co-expression patterns of Klf4 and SHBG in transgenic mice in vivo. ....................... 67 Figure 3.8 High KLF4 protein levels in stage VII/VIII round spermatids. .................................. 68 Figure 3.9 The alternative SHBG promoter responds to all-trans retinoic acid. ........................... 69 Figure 3.10 Retinoid receptors mediate the retinoic acid-induced upregulation of altP-SHBG reporter activity. ............................................................................................................................ 70 Figure 3.11 Expression of Sp1 and Sp3 in GC-2 cells. ................................................................. 71 Figure 4.1 Activated protein kinase A increases altP-SHBG reporter activity. ............................ 81 Figure 4.2 A partial CRE does not mediate altP-SHBG’s response to dbcAMP. ......................... 82 Figure 4.3 Overexpression of CREM isoforms increases altP-SHBG reporter activity. .............. 83 Figure 4.4 Overexpression of CREMτ potentiates altP-SHBG’s response to dbcAMP. .............. 84 Figure 5.1 Distinct testicular cell-type expression profiles of Shbg and SHBG genes across mammals. ...................................................................................................................................... 97 Figure 5.2 Selected species and their positions within the primate order. .................................... 99 Figure 5.3 The FP4/USF element is limited to simiiforme primates. ......................................... 100 Figure 5.4 Comparison of the SHBG exon 1A and flanking promoter sequences amoung primates. ...................................................................................................................................... 103 Figure 5.5 Summary of phylogenetic alignments of SHBG 5’ regulatory regions. .................... 104 Figure 5.6 Alternative SHBG transcripts are present in Old World Monkey testes. .................. 105 ix  Figure 5.7 No evidence for SHBG transcripts in marmoset testes. ............................................. 106 Figure 5.8 Lemur testes express SHBG transcripts from the proximal transcription unit. ......... 107 Figure 5.9 Hypothetical mechanism for FP4 insertion event. .................................................... 109 Figure 5.10 Trends of testicular SHBG gene expression across primates. ................................. 110 Figure 6.1 Locations of testicular SHBG in rodents versus humans. ......................................... 128   x  List of Abbreviations  μL Microlitre μM Micromolar AC Adenylate cyclase ABP Androgen binding protein Alu Abundance transposable element that can be cleaved by Arthrorbacter luteus restriction endonuclease A Amps (amperage) ATF1 Activating transcription factor 1 AR Androgen receptor BLAST Basic local alignment search tool bp Base pair cAMP 3’,5’-cyclic adenosine monophosphate cDNA Complementary deoxyribonucleic acid ChIP  Chromatin immunoprecipitation CRE cAMP response element CREB 3’,5’-cyclic adenosine monophosphate response element binding protein CREM 3’,5’-cyclic adenosine monophosphate response element modulator CBP CREB-binding protein CYP Cytochrome P450 ddH2O Double distilled and deionized water DHT 5α-dihydrotestosterone DMSO Dimethyl sulphoxide DNA Deoxyribonucleic acid DNaseI Deoxyribonuclease I E. coli Escherichia coli EDTA Ethylenediaminetetraacetic acid ER Estrogen receptor FBS Fetal bovine serum FSH Follicle stimulating hormone GAPDH Glyceraldehyde-3-phosphate dehydrogenase GC-2 Immortalized mouse spermatocyte xi  GnRH Gonadotropin releasing hormone GPCR G protein-coupled receptor HNF4α Hepatocyte nuclear factor 4 alpha IgG Immunoglobulin gamma JEG-3 Human placenta choriocarcinoma cell line kb Kilobase (one thousand base pairs of DNA) kDa Kilodalton hr(s) Hour(s) LG Laminin globular-like LH Luteinizing hormone nM Nanomolar Mb Megabase (one million base pairs of DNA) M Molar (molarity) min(s) Minute(s) mL Millilitre mM Millimolar mRNA Messenger ribonucleic acid NaCl Sodium chloride NaHCO3 Sodium bicarbonate NCBI National Center for Biotechnology Information NDS Normal donkey serum no. Number NWM New World Monkey OWM Old World Monkey p300 E1A binding protein PACAP Pituitary adenylate cyclase activating polypeptide PAGE Polyacrylamide gel electrophoresis PAS Polyadenylation sequence PBS  Phosphate buffered saline PCR Polymerase chain reaction PMSF Phenylmethylsulphonyl fluoride PTM Peritubular myoid cell xii  PVDF Polyvinylidene fluoride RAR Retinoic acid receptor RNA Ribonucleic acid SSC Spermatogonial stem cells SD Standard deviation SDS Sodium dodecyl sulphate SHBG Sex hormone-binding globulin siRNA Small-interfering RNA TNP1 Transition protein 1 Tris Tris(hydroxymethyl)aminomethane USF Upstream stimulating factor V Volts (voltage)    xiii  Acknowledgements  Without the support and inspiration from so many people, I would not have been able to complete this thesis.  It was during my last year of undergraduate study where I first developed strong interests in endocrinology and reproductive physiology. Two summer work terms and one year as an honours student in Dr. Barbara Vanderhyden’s laboratory provided me with a rich foundation in laboratory investigation. I thank her and Dr. François Paradis for their guidance and constant encouragement during this early period.  I thank my graduate supervisor, Dr. Geoffrey Hammond for his unrelenting support in the form of time, resources and enthusiastic encouragement throughout this entire project. I am grateful for the academic freedom he provided, which allowed me to come up with my own hypotheses, design my own experiments and construct my own arguments so see this thesis through. I will undoubtedly carry all of these skills forward as I move on to future pursuits.  I also extend my gratitude to all of the other members of the Hammond lab, past and present, for their support, including:  Caroline Underhill, for keeping the whole lab running smoothly and for her outstanding technical assistance with assembling the LG4-NHS affinity column. Drs. Tsung-Sheng (John) Wu and Jason Popesku for training me on many of the essential techniques that I required to complete this project. Dr. Marc Simard, for our many discussions on experimental techniques and results interpretation and for his critical review of an earlier draft of this thesis. Lesley Hill and Phillip Round for their friendship and solidarity during this whole process. I also thank Dr. Wayne Vogl and his team for teaching me everything I have learned about immunohistochemistry/ immuofluorescence and being an invaluble resource for troubleshooting.  Furthermore, I would not have been able to complete this work without the incredible support and encouragement of my parents, Ghiselle & Rick, my sister Emily and aunt Sylvie. And of course I extend thanks to my many close friends in Canada and abroad, including my UBCSK team.     1 Chapter 1: General Introduction  1.1 The cellular basis of male fertility in mammals  The testes are the essential reproductive organs in males. Their two main roles are to produce the main plasma androgen, testosterone, and to continuously produce haploid germ cells, spermatozoa, from puberty until old age [1].  The functional units for germ cell production are the seminiferous tubules which occupy the majority of the space within each testis. Seminiferous tubules are surrounded by a basement membrane and contain two cell types, Sertoli and germ cells, which constitute the seminiferous epithelium. The Sertoli cells are somatic cells that provide essential support for the germ cells throughout their development. Between the seminiferous tubules is an interstitial compartment containing peritubular myoid (PTM) and Leydig cells. Peritubular myoid cells reside outside the seminiferous tubules against their basement membrane, whereas Leydig cells are found in clusters between adjacent tubules. Blood vessels that supply the testis are also found interspersed throughout this space (Figure 1.1).  In sexually mature males, the development of mature germ cells begins at the outer circumference of the seminiferous epithelium with spermatogonial stem cells (SSCs) that divide asymmetrically to self-renew and produce type A spermatogonia. The SSCs divide mitotically to supply a population of cells (type B spermatogonia), which undergo a differentiation program that yields four haploid spermatozoa: spermatogenesis. A type B spermatogonium will divide    2 once by mitosis to produce two type B spermatogonia [2]. Each of these cells will differentiate into a preleptotene primary spermatocyte that moves luminally, crossing the tight junctions formed between adjacent Sertoli cells into the apical compartment of the seminiferous tubule. The tight junctions formed by the Sertoli cells are collectively known as the blood-testis-barrier (BTB) that isolates the apical compartment from the base of the seminiferous tubule. The most basolateral aspect of the tubules is bathed in interstitial fluid which is mainly derived from blood lymph and is conditioned by the surrounding cells [1], [2]. The apical or adluminal compartment of the seminiferous tubules is cut off from this continuous exchange and provides a specialized environment for the rest of spermatogenesis to take place.   It is within the apical compartment where meiosis occurs. As soon as a primary spermatocyte (2N, 2C)1 crosses the blood testis barrier (BTB), its DNA content is doubled and the diploid cell (2N, 4C) divides into two haploid secondary spermatocytes (each 1N, 2C) (meiosis I: reduction division). Within a few hours each secondary spermatocyte divides again into two haploid round spermatids (each 1N, 1C) (meiosis II: mitotic-type division). A round spermatid is not yet a fully matured germ cell. Within this post-meitotic phase the cell undergoes sweeping morphological changes in the final phase of spermatogenesis called spermiogenesis. A few days following meiosis II the spermatid starts to produce the acrosomal granule, a modified part of the Golgi apparatus that progressively lengthens and spreads around the nucleus. Once this is formed the germ cell produces a long flagellar tail at its opposite end. After this occurs most of the histones around which the cell’s DNA is wrapped are exchanged for arginine-rich protamines. Protamine packaging allows for extremely dense compaction of spermatid DNA so that it occupies just 5%                                                            1 N and C refer to ploidy/chromosome number and DNA content, respectively.    3 of the volume that DNA does in somatic cells [3]. As the DNA compacts the cell elongates, shedding most of its cytoplasm, which is phagocytosed by the surrounding Sertoli cell. The fully elongated germ cell is now a spermatozoon and is released into the lumen of the seminiferous tubule through a process called spermiation [4].  Nascent spermatozoa do not yet have the ability to fertilize an oocyte and will travel through the converging efferent ductules to the epididymis. Here, spermatozoa complete their maturation by gaining the ability to fertilize an oocyte, possibly through direct interaction with the principal epithelial cells of the caput epididymis where epididymal-derived proteins are loaded into the sperm plasma membrane [5]. Activation of the flagellar machinery occurs in the epididymal corpus, endowing them the ability to swim forward [6]. Finally, spermatozoa reach the caudal segment of the epididymis where they are stored until ejaculation [7].  1.2 The spermatogenic wave   The time for a spermatozoon to fully develop from a spermatogonial progenitor takes 35 days in mice and 74 days in men [8]–[10]. How then do healthy males remain fertile every day of adulthood with hundreds of millions of sperm produced each day? Microscopic examination of a testicular cross sections reveals that not only are there successive cohorts of differentiating germ cells in one seminiferous tubule (Figure 1.1b), but also that there are germ cells at different morphological stages in adjacent tubules (Figure 1.2a). A single spermatogenic stage refers to the progress of germ cell development and each stage is classified based on germ cell morphology. In mice there are 12 spermatogenic stages [11], whereas in humans there are six    4 [10]. Early investigations involving longitudinal sectioning of single seminiferous tubules revealed that as one moves along a tubule, a segment containing one discrete spermatogenic stage will be followed by another segment exhibiting the next stage of development [12] (Figure 1.2b). For example, in the mouse testis the segment containing early round spermatids that have not yet acquired a proacrosomal granule (stage I) will immediately follow another segment containing round spermatids that display the first evidence of a proacrosomal granule (stage II/III). Similarly, a segment that is undergoing spermiation (stage VIII) will be right behind another segment where this has just taken place (stage IX). Therefore, along the seminiferous tubules there is a progressive wave of continuously differentiating sperm. This spatial and temporal asynchrony of developing germ cells ensures continuous fertility in adult males.  1.3 Gene expression during spermatogenesis  Each phase of spermatogenesis is characterized by a unique pattern of gene expression. The ability of SSCs to self-renew is dependent on the expression of the stem cell marker, OCT4, and expression of this transcription factor falls as their daughter cells differentiate [13]. Primary and secondary spermatocytes express those genes that allow their progression through meiosis. Patterns of gene expression throughout meiosis are conserved throughout phyla. An analysis of 164 genes expressed throughout early, middle and late stages of meiosis in yeast were found to have at least 62% homology in C. elegans, Drosophila or rodents and humans [14]. Round spermatids have a program of post-meiotic gene expression that allows biosynthesis of the acrosome and flagellum, DNA compaction and cell elongation [15].     5 While developing male germ cells express many “housekeeping” genes that are common to somatic cells they also produce many germ cell-specific transcripts [16], [17]. These may be expressed from a gene that is only used by germ cells but is paralogous to another used in somatic cells. A good example of this is expression of the spermatogenic cell-specific Gapds once Gapdh expression falls [18]. While GAPDS performs the same catalytic function as GAPDH does in glycolysis, it also contains a unique proline-rich N-terminal domain that allows it to be anchored to the fibrous sheath of the mature spermatozoon [19]. Other male germ cell-specific genes with no obvious somatic homolog contribute to their unique properties. Examples of these are genes which encode protamines, transition proteins and those which contribute to the spermatozoon’s acrosomal and flagellar structures [20], [21]. Male germ cells are also a rich source of variant transcripts produced from genes that are also used in somatic cells [16], [22], [23]. These vary in type with some using alternative transcriptional start sites, cell type-specific splicing events and/or polyadenylation signals [16], [17], [24]. Alternative transcriptional start sites that are regulated by unique promoters, often allow for stage-dependant expression during spermatogenesis. Male germ cell-specific splicing events allow for the removal or addition of exons resulting in a change in a protein’s functional domains. Spermatogenic cell-specific polyadenylation signals can influence a transcript’s stability and influence translational control. Transcripts for the CREM gene are restricted to pachytene spermatocytes but localization of the CREM protein is only detectable later in round spermatids [25]. The use of an alternative polyadenylation site enhances the CREM transcript’s stability resulting in latent production of CREM protein [26]. Furthermore, since the highly compacted DNA in elongated spermatids is transcriptionally silent many gene products that the cell requires before spermiation are transcribed much earlier [27].    6 1.4 Hormonal regulation of male fertility  While male germ cells follow their own intrinsic program of differentiation, this process is not entirely autonomous. Endocrine signals from other sites of the body influence testicular function. The release of these hormones follows a cascade of chemical messengers known as the hypothalamic-pituitary-gonadal axis (Figure 1.3). Within the brain, pulsatile releases of gonadotropin-releasing hormone (GnRH) from the hypothalamus act on the anterior pituitary gland causing release of the gonadotropins: follicle-stimulating hormone (FSH) and luteinizing hormone (LH). These hormones travel through the bloodstream to the testis where they act on Sertoli and Leydig cells, respectively. Upon stimulation by FSH, Sertoli cells release inhibin into circulation. Once stimulated by LH, Leydig cells synthesize androgens: testosterone and DHT, which diffuse into the bloodstream and surrounding cells. Inhibin and testosterone feed back to the anterior pituitary and hypothalamus to inhibit the release of FSH and LH, respectively, thereby completing the signalling axis [28].  Stimulation of Sertoli cells by FSH also allows them to maintain an optimal environment for spermatogonial proliferation. Male mice lacking FSH or its receptor (FSHR) are fertile but have significantly reduced numbers of germ cells at all phases of spermatogenesis [29]–[31]. Androgens are indirect regulators of spermatogenesis. Classical androgen signalling is mediated through the androgen receptor (AR), a ligand-activated transcription factor that regulates androgen responsive genes through genomic androgen response elements [32]. The roles of AR signalling on spermatogenesis have begun to be revealed with the use of transgenic mice with Ar knocked out of specific cell types [32]–[34]. Males with Ar ablated from their germ cells    7 complete spermatogenesis normally and are fertile [35]. This was not surprising given that AR is not expressed in developing male germ cells in either humans or mice [36]–[39]. The AR gene localizes to the X chromosome that undergoes inactivation during early meiosis [40], [41]. In contrast, AR-signalling in both the Sertoli and PTM cells is required for male fertility. Mice specifically lacking Ar in either of these cell types are infertile [42]–[45].   Androgen receptor signalling in Sertoli cells is necessary for germ cells to complete meiosis. In the testes of the Sertoli cell-specific Ar knockout (SCARKO) mice there are comparable numbers of spermatogonia, a ~50% reduction in spermatocytes and almost complete absence of round spermatids, as compared to wild-type mouse testes [29], [31], [43]–[45]. Dysfunctional formation of the BTB is also observed in SCARKO mice resulting in its increased permeability [46].   Testicular development is normal in mice lacking Ar in their PTM cells (PTM-ARKO) until a reduction in testicular mass is seen from day 15 after birth through adulthood [42]. The reduced number of spermatogonia and all other downstream spermatogenic cells in PTM-ARKO mice suggests that AR-signalling in PTM cells is required for maintaining the normal population of SSCs [42]. Alterations in gene expression and luminal fluid secretion by Sertoli cells as well as abnormal development and function of Leydig cells are also features of PTM-ARKO mice [42]. Recently it was shown that androgen stimulation of cultured PTM cells causes an increase in the secretion of several growth factors that act on SSCs to enhance their survival [47], [48]. Other cells of the testis such as Leydig, vascular smooth muscle and vascular endothelial cells also    8 express Ar. Transgenic mice with Ar ablated from each of these cell types are, however, fertile indicating that AR signalling in these cells is not essential for spermatogenesis [49]–[51].  1.5 Sex hormone-binding globulin (SHBG)  1.5.1 Steroid hormone bioavailability and the free hormone hypothesis  Clusters of Leydig cells are more or less evenly distributed throughout the testis and as result, locally synthesized steroids reach their target cells in the testis by passive diffusion. In contrast, androgen and estrogen-sensitive tissues at other sites of the body receive gonadal steroids via the bloodstream. Steroid hormones do not readily dissolve in this aqueous compartment, however the major plasma protein, albumin, augments their solubility due to its nonspecific but high capacity to weakly bind numerous classes of lipophilic ligands [52]. Another plasma protein, sex hormone-binding globulin (SHBG) selectively binds sex steroids with high affinities that are 3-4 orders of magnitude greater than that of albumin [53], [54]. The highest affinity ligand for SHBG is 5α-dihydrostestosterone (DHT), followed by testosterone and 17β-estradiol; all of which bind to SHBG with nanomolar affinity in contrast to albumin that bind them with micromolar affinity [53]. The classically held view is that steroid hormones that are tightly bound to SHBG cannot diffuse out of the vasculature [54]. Unbound or “free” steroid hormones are biologically active and can passively diffuse into their target cells where they are either converted by steroidogenic enzymes, inactivated by metabolism or effect a cellular response by interacting with their respective nuclear hormone receptors to promote hormone-dependent gene expression [55]. This view forms the basis of the “free hormone hypothesis” [54], [56] and the level of SHBG within the bloodstream will directly influence how sex steroids are distributed between free and protein-   9 bound states. By limiting the entry of sex steroids into their target tissues as well as their rate of metabolic clearance, SHBG is thus a primary regulator of sex steroid hormone action.  1.5.2 The human SHBG gene and its products  The human SHBG gene is located on the short arm of chromosome 17 (p12-p13) [57]. Early sequencing efforts revealed eight exons (numbered 1-8) spanning a 4.3 kb region as well as one alternatively used first exon (1A) positioned ~2 kb upstream of exon 1 [58], [59]. This defined two transcription units that are used separately by the human liver and testis, each under the control of distinct promoter regions (Figure 1.4).   The proximal transcription unit is used by the liver and encodes SHBG that is secreted into the bloodstream. The promoter controlling this region lacks both TATA and CAAT boxes, and instead contains a liver-specific enhancer element and a well characterized binding site for hepatocyte nuclear factor 4 alpha (HNF4A/NR2A1) [58], [60]. Translation initiates from exon 1 which encodes a leucine-rich signal polypeptide that is required for secretion. Exons 2-8 encode two laminin G domains (LG4 and LG5) that are connected by a short linker region [61]. The N-terminal LG4 domain contains the steroid binding and dimerization regions of SHBG [62]–[64], while the function of the LG5 domain is less clear. Secreted SHBG exists as a homodimer in circulation and is glycosylated at three positions along each monomer: one O-linked site at threonine 7 in exon 1 and two N-linked sites at asparagines 351 and 367 in exon 8 (Figure 1.4) [58], [65].  The alternative SHBG transcription unit is used by testicular germ cells [66]. Transcripts originating from exon 1A are spliced directly to exon 2 [58], [59], bypassing exon 1 that encodes    10 the secretion polypeptide, and resulting in a protein that accumulates within the acrosomal compartment [67]. Translation of this sperm SHBG isoform is hypothesized to initiate from methionine 30 along exon 2 (Figure 1.4) [67]. The resultant protein has a truncated LG4 domain that lacks the N-terminal O-glycosylation site present on secreted SHBG, however it possesses a steroid binding ability that is indistinguishable to SHBG in serum [66].  1.5.3 Regulation of hepatic SHBG expression  The major site of SHBG biosynthesis is the liver, wherefrom it is secreted into the bloodstream.  Hepatocytes regulate the abundance of plasma SHBG at the transcriptional level under the influence of numerous hormonal and metabolic signals [68]–[70]. Hepatic expression of SHBG is particularly sensitive to cellular levels of the transcription factor HNF4α [60], which is generally accepted to be the master switch of SHBG expression in this tissue. The levels of HNF4A expression, and in turn SHBG, are tightly linked to the metabolic status of the liver. Clinical studies have consistently found that low SHBG levels within obese individuals are highly predictive of developing the metabolic syndrome, type II diabetes and cardiovascular disease [71]–[73]. Mechanistic studies have shown that consumption of monosaccharides (glucose and fructose) causes hepatic lipogenesis resulting increased cellular levels of palmitate and downregulation of HNF4A resulting in decreased levels of serum SHBG [74].  Elevated SHBG levels are seen in patients with hyperthyroidism [75]. Treatment of hepatocellular carcinoma (HepG2) cells in vitro with thyroid hormone results in an increase in SHBG expression and secretion [76]. Moreover, a two-fold increase in serum SHBG is seen in mice expressing human SHBG transgenes after five day thyroid hormone treatment [76]. While    11 no thyroid hormone response element exists in the proximal SHBG promoter, the same study found that thyroid hormone-mediated increase in SHBG production is actually an indirect mechanism caused by an increase in hepatic HNF4A expression. The increase in HNF4A expression is likely caused by the positive effect that thyroid hormones have on liver metabolism, such as increased β-oxidation of fatty acids resulting in a decrease of palmitate. Given the mounting clinical and mechanistic evidence suggesting that liver fat is a primary determinant of hepatic SHBG expression [69], [70], [77], serum SHBG level is an emerging as a biomarker for hepatic metabolic state.  1.5.4 SHBG expression in other tissues  In an effort to gain insight into how SHBG is expressed in other tissues, two lines of transgenic mice were created harbouring different lengths of the human SHBG locus [78]. The shorter of these transgenes contains the proximal 4.3 kb transcription unit, complete with ~0.9 kb of 5’ regulatory DNA (4K-SHBG). The longer contains an additional 5 kb of 5’ DNA that includes the alternative first exon used in the human testis, as well as another ~1.7 kb of DNA downstream of exon 8 (11K-SHBG). Both lines have high levels of circulating SHBG due to strong hepatic expression of the transgenes and have become established tools for studying the metabolic regulation of the SHBG gene in vivo [68], [70], [74], [76]. Further analyses of these animals also revealed novel sites of SHBG production. Both the 4K and 11K animals produce SHBG in epithelial cells of their kidneys and duodenum [78], [79]. Since this is a feature of both lines we know that SHBG expression in these tissues is a product of the proximal transcription unit and these transcripts contain exon 1. However, a significant fraction of the SHBG synthesized in these epithelial cells remains intracellular which is surprising given    12 that all of it originates from transcripts containing the leader sequence that normally directs its secretion. We now know that intracellular SHBG within the proximal convoluted tubules of the kidney remains intracellular due to incomplete glycosylation [80] and this same phenomenon is expected to cause retention of SHBG within duodenal epithelial cells as well. Recently, SHBG transcripts have also been demonstrated within human renal and duodenal biopsies [79], thus to the best of our knowledge these mice faithfully recapitulate all of the sites where SHBG is expressed in humans.  1.5.5 Testicular SHBG  Early studies of rat testicular and epididymal homogenates revealed the presence of a protein with nanomolar binding affinities for testosterone and DHT [81], [82]. It was noted that this testicular “androgen binding protein” (ABP) bears similar biochemical and steroid-binding properties as SHBG in the bloodstream. Testicular ABP has also been described in the testes of mice, rabbits, sheep, and primates [83]–[85]. In rats, ABP synthesis by Sertoli cells and its secretion into the seminiferous tubules are both regulated by retinoids and FSH [82], [83], [86]. From there it travels to the caput epididymis and is endocytosed by the principal epithelial cells [87]–[92] This transit of ABP through the male reproductive tract has been proposed to be a mode of androgen transport to the epididymis [93], [94], but surprisingly this has never been tested directly.  It has since been shown that in most mammals testicular ABP is produced from the same transcription unit that encodes plasma SHBG [83]. Since a protein with nearly identical properties to plasma SHBG has also been characterized in human testicular homogenates it was    13 widely assumed that expression of SHBG in the human testis occurs in the same manner as in rodents [87], [95]. Examination of SHBG transcripts from human testis cDNA libraries revealed that they contain an alternative first exon followed by exons 2-8 [58], [59]. In testes from 11K-SHBG transgenic mice, SHBG transcripts contain the same alternative first exon [66]. Moreover, northern blots performed on separate Sertoli and germ cell fractions from 11K-SHBG mice revealed that alternative SHBG transcripts are actually a product of their germ cells and not Sertoli cells. Detection of human SHBG within 11K-SHBG mouse testes by immunohistochemistry revealed that SHBG localizes to the acrosome on the heads of mature spermatozoa [66]. These findings were supported by the identification of SHBG within the heads of human spermatozoa [96]. Interestingly the abundance of SHBG in human sperm correlates positively with sperm motility and these levels appear to decline with age [96].  The striking differences of testicular SHBG production between rodents and humans raises the obvious questions of how and why they are different. When a 5.5 kb region of the rat Shbg gene was introduced into mice, transcripts arising from the transgene mirrored the expression endogenous mouse Shbg in Sertoli cells [97]. In contrast, SHBG is not expressed in the testes of transgenic mice carrying the corresponding 4.3 kb region of the human SHBG gene [78]. Phylogenetic comparisons of the human and rodent proximal SHBG promoters revealed the presence of a 38 bp region in the human gene that is absent in rodents (Figure 1.4) [98]. This element had been previously identified as a DNase I foot-printed region (foot-print 4, FP4) of the human SHBG promoter in hepatocytes [60]. When an additional line of SHBG-4K transgenic mice was derived with a modified human transgene that lacked FP4, the resultant males    14 expressed SHBG in their Sertoli cells. Furthermore secretion of SHBG in cultured Sertoli cells from these animals responds to both FSH and retinoid treatments [98].  1.6 Hypothesis and approach  While the biochemical properties of germ cell SHBG have been characterized, little is known about how it is expressed. Within 11K-SHBG males, SHBG transcripts are detectable at multiple spermatogenic stages by in situ hybridization suggesting stage-dependant expression [78]. There is a 10-fold increase in SHBG transcript abundance from the lowest levels during stage IV to peak levels within stage VII seminiferous tubules [78]. This indicated that there is a surge in SHBG expression from stages IV through VII. This observation was later complemented by immunohistochemical evidence demonstrating that the earliest localization of germ cell SHBG is within the proacrosomal granules of stage VII round spermatids [66]. Taken together these findings strongly suggest there exists a developmental trigger that initiates SHBG expression during stages IV-VII of the spermatogenic cycle. I therefore hypothesized that expression of SHBG from an alternative promoter in testicular germ cells is influenced by stage-specific factors and/or events during spermatogenesis.  In order to gain insight into the regulation of the SHBG gene in testicular germ cells I sought to identify what elements and features inherent to its alternative promoter are essential for its transcriptional activity. Moreover, I asked if cellular-physiological mechanisms and signalling pathways that are relevant to post-meiotic sperm development could influence SHBG transcriptional activity. To address these, two approaches were taken:     15 1. Bottom up, where I began with a sequence analysis of the promoter flanking the alternative SHBG transcription unit and screened for known regulatory elements and motifs. The role of these elements was tested by disrupting them in the context of a reporter assay designed to model alternative SHBG promoter activity. I then asked which factors could interact with these elements and if they share related patterns of expression during spermatogenesis in vivo.  2. Top down, where I explored the influence of the transcription factor CREM on alternative SHBG promoter activity. The upstream signalling pathway that activates CREM was manipulated by pharmacological treatment and the roles of two CREM isoforms were tested directly by overexpression.  In addition to mechanistic studies, I present a phylogenetic study of SHBG genes from each major primate group. Using a computer-assisted platform I asked if the genetic elements of the human SHBG gene that contribute to its unique pattern of expression in the testis are shared by other primate species. I complemented these findings by examining SHBG transcripts from whole-testis cDNA samples obtained from several primate species. Inferences on the cell of origin of SHBG in the testis of each species are made. Finally, this thesis concludes with a summary of potential roles for testicular SHBG in male reproductive physiology and offers several routes for how to test them.       16   Figure 1.1 Structure and principal components of the mammalian seminiferous tubule. A. A stylized depiction of the mammalian seminiferous tubule. The interstitial compartment of the testis contains steroidogenic Leydig cells, peritubular myoid cells and is the space where blood vessels which supply the testis are found. Sertoli cells occupy all of the space in between the germ cells and can be thought of as the “glue” that holds them in place. The relative space that Sertoli cells occupy within healthy testes is overly estimated in this panel. In testes where active spermatogenesis is taking place germ cells account for ~90% of the tubular mass [99] as shown in B. B, Histological section of a human seminiferous tubule. Successive cohorts of developing germ cells are arranged in concentric layers. All germ cells within a given cohort are synchronous in their development as they move towards the lumen. A Sertoli cell is partially outlined. SSC & A/B, spermatogonial stem cells and type A/B spermatogonia, PS, pachytene spermatocytes, RS, round spermatids, ES, elongated spermatozoa. Panel A taken from Rato et al, 2012 [100] and is used with permission from the publisher. Panel B is used with permission by Drs. José Luis Calvo, MD and José Enrique Garcia-Mauriño, MD. Department of Cell Biology, Faculty of Medicine, Complutense University of Madrid.     17   Figure 1.2 The arrangement of seminiferous tubular stages in the mammalian testis. A, A micrograph of a human testicular cross section shows that germ cells at different morphological stages exist in adjacent seminiferous tubules. The tubule on the left contains spermatogonial progenitor cells, pachytene spermatocytes, round and elongated spermatids and residual cytoplasmic bodies (stage II), whereas the tubule on the right contains spermatogonia, two layers of pachytene spermatocytes and elongating spermatids (stage V) [101]. B, A stylized depiction of the two modes of seminiferous tubule architecture in mammals is adapted from Luetjens et al 2005 [102]. Seminiferous tubules in rodent and lower primate testes consist of a single spermatogenic stage per cross section. Tubules in higher primate and human testes exhibit a multistaged or “patchy” arrangement exhibiting >1 spermatogenic stage per cross section. Note regarding the upper drawing in B, the actual arrangement of successive spermatogenic stages in    18 single-staged tubules exists as: I, II, III through XII in sequence [12], [103] and not in the order of numerals shown. OWM, Old World Monkey, NWM, New World Monkey. Panel A is used with permission by Drs. José Luis Calvo, MD and José Enrique Garcia-Mauriño, MD. Department of Cell Biology, Faculty of Medicine, Complutense University of Madrid. Panel B is used with permission by the senior author and publisher.       19       Figure 1.3 The mammalian hypothalamic-pituitary-testicular endocrine signalling axis.  Multiple levels of signalling allow for numerous points of control to modulate testicular function. Arrows in the centre and on the left denote the stimulatory (+) signalling cascade, whereas arrows on the right denote the negative (-) feedback responses. All of the hormones indicated here travel from their tissue of origin to their target cells via the bloodstream. An analogous signalling axis that controls reproductive function in females exists as well. GnRH, gonadotropin-releasing hormone, LH, luteinizing hormone, FSH, follicle-stimulating hormone, T, testosterone.        20    Figure 1.4 Overview of the human SHBG gene and its products. The human SHBG gene has two transcription units. The proximal transcription unit encodes SHBG that is synthesized in the liver, kidney and duodenum. The alternative transcription unit encodes SHBG that is produced in testicular germ cells. Each transcription unit is under the control of a distinct promoter region which direct the unique pattern of expression of the SHBG gene. The promoter immediately flanking exon 1 contains the FP4 element which represses the SHBG gene in Sertoli cells. Relative distances between exons are not to scale. Sites of O-linked (filled circle) and N-linked (open circles) glycosylation are indicated on exons 1 and 8, respectively.      21 Chapter 2: Materials and Methods  2.1 Animals  Mice expressing human SHBG transgenes (C57BL/6 X CBA background) [78] and their wild type littermates were maintained under standard conditions at the UBC Centre for Disease Modeling with food and water provided ad libitum. Tissues from adult male mice for immuno-histochemistry, chromatin immunoprecipitation or lysate preparation (all described below) were obtained under a protocol approved by the University of British Columbia Animal Care Committee (A10-0309, SHBG: Beyond Plasma Transport).  2.2 Cell lines and transient transfections  Immortalized mouse spermatocyte cells, GC-2spc(ts) (catalogue number CRL-2196)[104] (GC-2) were obtained from the American Type Culture Collection (ATCC). Cells were maintained in Dulbecco’s Modified Eagle Medium, supplemented with 10% fetal bovine serum (FBS) and penicillin-streptomycin (100 U/mL). Cells were grown in a standard laboratory cell culture incubator at 37oC in 100% humidity, 5% CO2 environment. Cells were routinely passaged 1:5 or 1:6 by detaching with trypsin-EDTA, approximately twice weekly. To minimize the variable effects of cell line drift all experiments performed were with passage numbers below or equal to 15. Human hepatocellular carcinoma cells (HepG2) obtained from ATCC (catalogue number HB-8065) were also maintained as above. Unless otherwise indicated, all cell culture reagents were purchased from Thermo Fisher Scientific.    22 2.3 Plasmids and reporter constructs  The alternative promoter fragment corresponding to -366/+28 nt with respect to the transcriptional start site at exon 1A was PCR amplified from a cosmid containing a fragment of the human chromosome 17 that was used previously for earlier sequencing studies of the SHBG gene [58]. This fragment was cloned into the pGL3 luciferase reporter vector (Promega) by standard cloning methods. The primers used for sticky-end (XhoI/HindIII) cloning are listed in Table 2.1.  The -266/+366 region of the proximal human SHBG promoter incorporated into pGL3 was constructed by Dr. Tsung-Sheng Wu as described [105]. Within the promoter fragment, a stop codon was introduced immediately downstream of the translation initiation codon in exon 1. This was done to disrupt translation from the normal SHBG translation start site in exon 1 to ensure efficient translation initiation of the luciferase gene inserted within exon 2.  The following overexpression constructs for use within mammalian cells were generous gifts from the following laboratories: the mouse Klf4 construct (within pcDNA 3.1) from the laboratory of Dr. J.P. Katz via Dr. Marie-Pier Tetreault at the University of Pennsylvania, Philadelphia, USA; the mouse CREMτ2 construct (within pRC-CMV) from Dr. Joachim Weitzel at the Leibniz Institute for Farm Animal Biology in Dummerstorf, Germany; rat CREMΔCG (within pCMV5) from Dr. William H. Walker at the University of Pittsburgh, Pittsburgh, USA; the mouse Act construct (within pcDNA 3) from the laboratory of Dr. Paolo Sassone-Corsi via Sherry Dilag at the University of California, Irvine, USA.    23 2.4 Transient transfection and luciferase reporter assay  Approximately 16-18 hours (hrs) before transfection, GC-2 cells were trypsinized from maintenance culture into a single cell suspension, counted and 1.4x105 cells/well were seeded into 24 well plates. The next morning, each well was refreshed with 0.5 mL media and transfections were performed as follows.   Transient cotransfection of luciferase and β-galactosidase reporter constructs were performed using Lipofectamine 2000 transfection reagent according to the manufacturer’s protocol. In all experiments a ratio of 2 µL of transfection reagent to 1 µg of nucleic acid was used. After separately diluting the transfection reagent and pool of nucleic acids in Opti-MEM, the two dilutions were combined and incubated for 30 min before pipetting onto cell media. Each well of cells was transfected with 500 ng of luciferase reporter construct plus 45 ng of pRC-CMV-LacZ (β-galactosidase) unless otherwise indicated in figure captions. For experiments involving cotransfection with additional plasmids or siRNA, the transfection complex incubation time was increased to 45-60 min before pipetting them onto cell media. Media was refreshed 6-7 hrs after transfection.  Forty-eight hrs after transfection the cells were washed once with 0.5 mL sterile PBS and 75 uL of 1x Reporter Lysis Buffer (Promega) supplemented with protease inhibitor cocktail lacking EDTA (Thermo Fisher Scientific) was pipetted onto each well. Cells were allowed to swell for 5 min at room temperature before subjecting them to a single freeze/thaw cycle at -80oC. Crude lysates were harvested from their wells by pipetting and transferred to separate microcentrifuge    24 tubes. These were centrifuged at 18,000 g for 10 min at 4oC before transferring supernatants to fresh tubes. Separate 25 µL aliquots of each lysate were plated onto white opaque and clear bottom 96 well plates for measurement of luciferase and β-galactosidase activities, respectively. Using a multichannel pipet, 65 µL of Luciferase Assay Reagent (Promega) was added in unison to the lysates and luminescence activity in counts per second were immediately recorded on a Victor X4 plate reader (Perkin Elmer). Using a multichannel pipet, 35 µL of 1x β-galactosidase assay buffer (Promega) was added in unison to plated lysates and mixtures were incubated at 37oC for 40 min. Reactions were stopped by adding 75 uL of 1 M sodium carbonate and absorbance at 405 nm (A405) was immediately recorded on a Victor X4 plate reader. Each luminescence reading shown in Chapters 3 and 4 was divided by its respective A405 reading to correct for transfection efficiency to produce a relative luminescence value.  2.5 Small interfering RNA (siRNA)  Specific knockdown of mouse Klf4 transcripts was performed with an ON-TARGETplus SMARTpool siRNA against mouse Klf4 (L-040001-01-005, 16600) (Dharmacon, GE Healthcare). This pool of four targeting RNAs targets the following sequences: AGAUUAAGCAAGAGGCGGU, CCAUUAUUGUGUCGGAGGA, CCGAGGAGUUCAACGACCU, CGACUAACCGUUGGCCGUGA. Negative control conditions were performed using an ON-TARGETplus Non-targeting Control siRNA from the same manufacturer.     25 2.6 Whole cell extracts and subcellular fractionation  2.6.1 Extracts from cells in culture  GC-2 cells were lysed by re-suspension in 1x Cell Lysis Buffer (Cell Signalling Technology, Danvers, MA) supplemented with protease inhibitor cocktail (Thermo Fisher Scientific, Waltham, MA) and 1 mM phenylmethane sulphonyl fluoride (PMSF) followed by a single freeze-thaw cycle according the manufacturer’s protocol. Preparation of nuclear and cytoplasmic fractions was performed using the NE-PER Nuclear Protein Extraction Kit (Thermo Fisher Scientific, Waltham, MA) according to the manufacturer’s included protocol. The total protein content of all lysates was quantified in a DC (detergent compatible) Protein Assay (Bio-Rad) against a standard curve of BSA diluted in lysis buffer according to the manufacturer’s instructions.  2.6.2 Preparation of seminiferous tubule and germ cell-enriched lysates from mouse testes  Lysates from whole seminiferous tubules and germ cell-enriched fractions were prepared using a modified protocol from Boucheron and Baxendale [106]. Following isoflurane overdose, male mice were perfused with room temperature PBS for 10-15 min until the perfusate from the animal was no longer visibly red from blood contaminants. The testes were quickly dissected out and placed in ice-cold DMEM for 5-10 min. Each testis was transferred to a dish containing ice cold PBS, decapsulated and tubules were placed into a fresh sterile tube containing 7.5 mL DMEM with 0.8 mg/mL collagenase type 1A, 0.225 mg/mL hyaluronidase, 0.3 mg/mL DNase I (digest media). Tubes were incubated at 37oC shaking gently and were gently flicked every 5    26 min to separate seminiferous tubules from the interstitial cells and material. After 20 min, tubes were removed from the incubator and tubules were allowed to sediment by gravity. Supernatants were removed and tubules were resuspended in another 7.5 mL of fresh digest media before returning to 37oC for an additional 10-15 min. Tubules were allowed to sediment again and supernatants were removed. Tubules were then washed four times by re-suspending in 9-10 mL PBS and then centrifuging at 180 g for 5 min.   Tubules destined for whole tubule lysis were resuspended in 1x lysis buffer (Cell Signalling) supplemented with protease inhibitor cocktail (Roche) and 1mM PMSF and placed on ice for 1-2 hrs to swell with occasional flicking.   To obtain an enrichment of mostly germ cells, washed seminiferous tubules were re-suspended in 3 mL trypsin-EDTA (0.25%) and incubated at 37oC with gentle shaking for 20 min. Tubes were removed and debris was allowed to sediment. Cloudy supernatants containing mostly liberated germ cells were carefully removed and pipetted into 50 mL tubes containing ~30 mL ice cold DMEM containing 10% FBS. A second trypsin digest was performed and that supernatant was pooled with the first. Tubes containing mostly germ cells were centrifuged at 250 g for 10 min. Cells were resuspended with 10 mL PBS containing 0.3 mg/mL DNase I and incubated at 37oC for ~15 min until the viscosity from liberated DNA in solution was reduced. Cells were pelleted and washed four more times with PBS before being resuspended in the same lysis buffer as used for whole seminiferous tubules and allowed to swell as above.     27 Swelled tubule and cell preparations were placed at -80oC overnight. The following day they were thawed, passed ten times through a sterile 30 gauge needle fitted to a sterile 5 mL syringe before centrifuging at 18,000 g for 10 min at 4oC. Supernatants were recovered and stored at -20oC until use. Total protein content in each sample was quantified as described above.  2.7 Antibodies  A rabbit polyclonal antibody to mouse KLF4 (catalogue no. AF3158) used in Chapter 3 for western blotting, chromatin immunoprecipitation and immunohistochemistry was purchased from R&D Systems. Goat anti-GAPDH (V-18, sc-20357), rabbit anti-histone H4 (Lys 8-R, sc-8660-R), rabbit anti-pan-CREM (X-12, sc-440), normal goat IgG (sc-2028), and normal rabbit IgG (sc-2027) antibodies were purchased from Santa Cruz Biotechnology (Santa Cruz, CA).   2.8 Tissue fixation and chromatin immunoprecipitation  Seminiferous tubules from three 11K-SHBG transgenic mouse testes were isolated as described in 2.5.2 and pooled together. Tubules were fixed in freshly prepared 1% paraformaldehyde dissolved in 1x PBS, pH 7.4 for 10 min at room temperature. Tissue lysis and chromatin isolation was performed with a Magna ChIPTM A/G kit (Millipore) according to the manufacturer’s protocol with the exception that following lysis, crude lysates were passed 14 times through a 22 gauge needle fitted to a sterile 5 mL syringe. Chromatin-containing supernatant (600 μL) was divided into 3x 200 μL aliquots and stored at -80oC.     28 All of the following steps for chromatin shearing were performed on ice using ice-cold buffers and tubes. One testis’ worth of chromatin (200 μL) was resuspended in 1500 μL of ice cold Shearing Buffer (Millipore) supplemented with EDTA-free protease inhibitor cocktail (Thermo Fisher Scientific) and 1 mM PMSF and aliquoted into 6x 255 uL in separate 1.5 mL microcentrifuge tubes. Chromatin was sheared using a Microson Ultrasonic Cell Disrupter XL probe sonicator while on ice. The optimal condition to shear all chromatin into 200-300 bp fragments was 14 cycles of: 5 s sonication pulse followed by 25 s rest to prevent over heating. Sample tubes were immersed in an ice bath the entire time. Sheared samples were centrifuged at 18,000 g for 10 min at 4oC and supernatants were aliquoted into 4x 50 μL chilled microcentrifuge tubes and stored at -80oC for subsequent ChIP assays. Remaining volumes (~55 μL) were reverse cross-linked as described in the provided kit protocol (Millipore) by incubating at 62oC overnight in a kit-supplied ChIP Elution buffer supplemented with 0.33 µg/µL proteinase K (Thermo Fisher Scientific). The next morning DNA fragments were cleaned up using a PCR Purification Kit (Qiagen) and eluted in 30 uL. This volume was subjected to 2% agarose gel electrophoresis and visualized using an ImageQuant LAS 4000 system (GE Healthcare Life Sciences, Baie d’Urfe, Quebec) to confirm shearing efficiency. An example of a chromatin shearing optimization assay is shown in Figure 2.1. ChIP analysis was performed using the Magna ChIPTM A/G kit (Millipore) following the included protocol exactly as described. Oligonucleotide primers for PCR analysis are listed in Table 2.1.       29 2.9 Tissue processing and immunohistochemistry/ immunofluorescence  Following isoflurane overdose, male mice were sacrificed by cutting the base of aorta. Testes were dissected out and immediately immersed in excess (10-14 mL) of freshly prepared room temperature 4% paraformaldehyde dissolved in 1x PBS, pH 7.4 (fixative buffer). After 1-2 hrs of gentle rocking at room temperature, fixative buffer was refreshed and testes were further fixed for 20-24 hrs at 4oC with gentle rocking. The next day fixative buffer was replaced with 70% ethanol and tissues were subsequently dehydrated in a series of ethanol solutions and embedded in paraffin by infiltration. Thin sections (5 μm) were mounted into charged Superfrost® Plus slides (VWR, Mississauga ON) and deparaffinized by progressive 3 min immersions in the following: xylene, xylene, (1:1) xylene:100% ethanol, 100% ethanol, 100% ethanol, 95% ethanol, 70% ethanol, 50% ethanol, distilled water, followed by thorough flushing under distilled water for at least 5 min. For heat-induced antigen retrieval, slides were immersed in ~250 mL 10 mM sodium citrate, pH 9.9 (AR buffer) and microwaved for 2x 5 min in a 800 W domestic microwave, with the volume topped up with warmed AR buffer in between heating. While still immersed, slides were allowed to cool for 45-60 min.  For tissue staining, all of the following incubations were performed in individual humid chambers. Slides were washed once with 1x PBS, 0.1% tween 20, 0.5% BSA (wash buffer) for 5 min and blocked with the same buffer containing 5% normal donkey serum (NDS) for 1 hr at room temperature. Slides were incubated with a polyclonal goat anti-mouse KLF4 antibody diluted 1:100 in wash buffer containing 1% NDS overnight at 4oC. Slides were washed 3x 5min in wash buffer before incubation with a donkey anti-goat secondary antibody conjugated to    30 Alexa Fluor® 488 (Thermo Fisher Scientific) diluted 1:200 for 1 hr at 37oC. Slides were washed again for 3x 5 min in wash buffer and incubated with PNA-lectin conjugated to Alexa Fluor® 647 (L32460) for 10 min at room temperature. After 3 final washes slides were mounted with VECTASHIELD® antifade mounting medium containing DAPI (Vector Laboratories) and sealed under glass coverslips with clear nail polish.  Slides were imaged under oil immersion at the UBC Life Sciences Institute Imaging Core with an Olympus FV1000 inverted laser scanning confocal microscope fitted with the following lasers: multi-line argon for 488 nm, 405 nm diode, 633 nm helium-neon.   2.10 Polyacrylamide gel electrophoresis (PAGE) and western blotting  Whole tissue, whole cell lysate or nuclear and cytoplasmic fractions (22 μg) were resolved by 10% or 12% SDS-PAGE and electro-transferred to 0.45 μm PVDF mini membranes using a Trans-Blot Turbo Transfer system (Bio-Rad). Gel to membrane transfers were performed at 25 V, 1.3 A for 10 min. The membranes were blocked in 1x PBS, 0.1% tween 20 (PBS-T) containing 5% skim milk (blocking buffer) for 1 hr at room temperature, followed by overnight incubation at 4oC with primary antibodies diluted in blocking buffer. The next morning membranes were washed once with blocking buffer for 5-10 min and incubated with horseradish peroxidase-conjugated secondary antibodies (Sigma-Aldrich) diluted 1:10 000 in blocking buffer. Immunoreactive proteins were detected in a ImageQuant LAS 4000 system with Amersham ECL Prime chemiluminescent reagent according to the manufacturer’s protocol (GE    31 Healthcare Life Sciences). PAGEs performed under non-denaturing (native) conditions were done so as above with SDS solutions replaced with distilled water in all gels and buffers.  2.11 Chemicals and reagents  N6,2′-O-Dibutyryladenosine 3′,5′-cyclic monophosphate sodium salt, H89 dihydrochloride and all-trans-retinoic acid were purchased from Sigma Aldrich (St. Louis, MO). The pan-RAR antagonist AGN194310 was a generous gift from Dr. T.M. Underhill at the University of British Columbia.  2.12 RNA extraction and cDNA synthesis  Total RNA was extracted from cells or tissues using the RNeasy Mini Kit (Qiagen, Toronto, Ontario) with a mid-protocol DNase I treatment. Concentrations of eluted RNA were determined using a NanoDrop 2000 spectrophotometer (Thermo Fisher Scientific). Unless used immediately, RNA samples were stored at -80oC. Total RNA (2 500 ng) was reverse transcribed using SuperScript® II Reverse Transcriptase with oligo(dT) primers (Thermo Fisher Scientific) in a 20 uL volume according to the manufacturer’s instructions with the exception that the synthesis incubation period at 42oC was increased from 50 to 60 min. All cDNA samples were stored at -20oC.      32 2.13 Primers and polymerase chain reaction (PCR)  Oligonucleotide/primer sequences for all PCR reactions described in this thesis are listed in Table 2.1. Amplification reactions were performed with either PCR Supermix (Thermo Fisher Scientific, catalogue no. 10572014) containing Taq polymerase, PhusionTM High-Fidelity DNA polymerase for mutagenesis of C-G rich templates (Thermo Fisher Scientific, catalogue no. F541) or AccuPrimeTM Pfx DNA polymerase (Thermo Fisher Scientific, catalogue no. 12344024), each with their included reaction buffers. All PCR reactions were performed in 25 μL reaction volumes with primers diluted according to the instructions included with each polymerase. All reactions were carried out in a iCycler thermocycler (Bio-Rad) with the lid held at 100oC. Whole reaction volumes were analyzed by agarose gel electrophoresis stained with SYBR® (catalogue no. S33102) diluted 1:10 000. Gel images were captured on a ImageQuant LAS 4000 system (GE Healthcare Life Sciences)  2.14 Bioinformatics and phylogenetic analyses  2.14.1 Transcription factor binding site analysis  Two online tools were used to screen for putative transcription factor binding sites within the minimal region of the human alternative SHBG promoter: TRANSFAC® (http://www.gene-regulation.com/pub/databases.html) and ORKAtk (http://cisreg.cmmt.ubc.ca/cgi-bin/ORCAtk/orca). Searches using both programs were performed with a cutoff threshold of 80% match to consensus binding sites. Only hits obtained from both analysis tools were considered as putative binding sites. The partial CRE described in Chapters 3 and 4 matched    33 below 80% consensus CRE, however since this was identified in an earlier analysis [67], it was still considered for downstream investigation.  2.14.2 Primate sequence retrieval and alignments  Primate SHBG sequences were retrieved from the NCBI public database online. For those sequences not annotated on GenBank, Nucleotide BLAST searches were performed using ~300 bp regions of the human gene as query. Search sets were either reference genomic sequences (refseq_genomic) or whole-genome shotgun contigs (wgs) with the project or species/taxa of interest selected. Megablast, discontinuous megablast and blastn stringencies were used to locate sequences from species increasingly phylogenetically distant from human. NCBI accession codes for each species are listed in Table 5.1. Sequence alignments were performed using the Clustal Omega Multiple Sequence Alignment tool (http://www.ebi.ac.uk/Tools/msa/clustalo/). Percent sequence identity calculations were performed using the LALIGN server (http://www.ch.embnet.org/software/LALIGN_form.html) that applies the algorithm developed by Huang and Miller [107].  2.15 Non-human primate cDNA and tissue biopsies  Non-human primate testis cDNA samples were a generous gift from Dr. Rüdiger Behr at the Deutsches Primatenzentrum in Göttingen, Germany. Baboon (Papio hamadryas), Lion-tailed macaque (Macaca silenus), and Marmoset (Callithrix jacchus) cDNA were 100 ng/μL, while that from Rhesus macaque (Macaca mulatta) was 50 ng/μL. RNA samples used as templates for these cDNAs were DNaseI treated prior to being primed by oligo(dT).     34 RNAlater®-treated (Thermo Fisher Scientific) liver and testis biopsies (0.22-0.25 g) from two grey mouse lemurs (Microcebus murinus) were provided by Dr. Erin Ehmke at the Duke Lemur Center in Durham, North Carolina, USA. From each, a 20-30 mg piece was excised and disrupted in 600 μL RLT buffer (Qiagen) supplemented with β-mercaptoethanol using a free-standing rotor-stator tissue homogenizer. Total RNA was extracted according to the RNeasy Mini Kit as above.  2.16 Purification of antibodies specific to the N-terminal LG4 domain of human SHBG  A pool of highly specific antibodies to human SHBG was desired for immunohistochemical detection of SHBG in vivo. First, a total IgG antibody fraction was isolated from crude rabbit antisera raised against human SHBG using a stationary phase of Protein A-linked to Sepharose beads (17-0402-01). Antibodies specific to the N-terminal LG4 domain of SHBG were subsequently obtained by passing the total IgG pool over immobilized LG4 coupled to N-hydroxysuccinimide (NHS)-linked Sepharose beads (17-0906-01). The resultant pool of anti-LG4 antibodies were then subjected to western blot performance tests before use in immunohistochemistry. A description of these methods is provided below.  2.16.1 Selection of antisera to purify antibodies from  Several rabbit antisera raised against human SHBG were available in Dr. Hammond’s laboratory. These were previously produced at different times using different rabbits [108], [109]. In order to select the best antisera to purify antibodies from, serum samples from wild-type and SHBG transgenic mice were diluted 1:1,000 in 1x PBS and 15 µL of each was subjected to western blot    35 analysis under non-denaturing conditions. Following electrophoresis and electrotransfer as described above, samples were probed with three different rabbit anti-human SHBG antisera as primary antibodies. Figure 2.2 shows that that antisera #4 and #5 were best able to identify SHBG in transgenic mouse serum, whereas antiserum #7 exhibited poor identification of SHBG. This suggested that compared to #7, antisera #4 and #5 have a higher titer of specific antibodies to SHBG. Antisera #4 and #5 were selected for downstream antibody purification.   2.16.2 Coupling of LG4 to activated NHS-Sepharose column  Overproduction of N-terminal LG domain of human SHBG was performed in the E. Coli host strain, Rosetta, as described [105]. Bacterial pellets from six litres of culture were lysed and LG4 was isolated using a three-step purification strategy [105]. From this, an aliquot was taken and concentrated to 1.68 mg/mL in 0.9 mL of coupling buffer (0.2 M NaHCO3, 0.5 M NaCl, pH ~8.3). The concentrated LG4 was immobilized to a 1 mL HiTrap NHS-activated HP column according to the manufacturer’s instructions. Figure 2.3 shows that there is a significant decrease in LG4 abundance in the flow through relative to the input. This indicated that the majority of the LG4 was immobilized to the NHS column during coupling. The resultant column was stored in a 20% ethanol solution containing 100 nM DHT at 4oC until use.  2.16.3 Two-step antibody purification protocol  A 50:50 mixture of rabbit anti-human SHBG antisera #4 and #5 was diluted 1:3 in 20 mM sodium phosphate, pH 7.4 to a final volume of 9 mL. Particulates were removed by passing the sample through a 0.45 µm filter fitted to a sterile 30 mL syringe. The total IgG fraction was    36 isolated from the sample using a 1 mL HiTrap Protein A HP column (GE) according to the manufacturer’s instructions (71-7002-00 AP). Seven ~500 µL fractions (eluted in 100 mM citric acid, pH 3.6) were collected into sterilized glass tubes containing a neutralization solution of 100 µL 1 M tris pH 9.0 plus 60 µL 10 mg/mL BSA in ddH2O to stabilize the eluted antibodies. Aliquots from each stage of purification were reserved and subjected to 12% SDS-PAGE followed by protein staining using Bio-Safe Coomassie G-250 Stain (Bio-Rad) according to the manufacturer’s instructions. Figure 2.4 shows that there is significant enrichment of IgG molecules in fractions 2-5 relative to input. To maximize the total yield of IgGs, a second round of IgG purification was performed using the flow-through from the first round and the same column.  Fractions 2-4 from the first IgG purification round and fraction 3 from the second round were pooled and diluted 1:3 in 1x PBS pH 7.4 to a final volume of ~7.5 mL. The sample was then affinity purified against the LG4 column according to the manufacturer’s instructions (71-7006-00 AX) with the one exception that the input was first brought to 37oC and then applied twice to the column before washing. Seven ~500 µL fractions were eluted with 150 mM glycine pH 2.65 into sterilized glass tubes containing a neutralization solution of 20 µL 1 M tris pH 9.0 plus 50 µL 10 mg/mL BSA in ddH2O to stabilize the eluted antibodies. Aliquots from each stage of purification were reserved and subjected to SDS-PAGE followed by protein staining as described above. Figure 2.5 shows enrichment for IgG species in elution fractions 2-7. Fractions 2-4 and 5-7 were pooled separately.      37 2.16.4 Purified antibody performance tests  To confirm the specificity of antibodies eluted off the immobilized LG4 column a series of western blot tests were performed. Diluted serum samples from wild-type and SHBG transgenic mice were subjected to western blotting under non-denaturing conditions using the input IgG, flow-through and pooled elution fractions from Figure 2.5 as the primary antibodies. Figure 2.6 shows that compared to the input IgG pool, the flow-through fraction shows diminished specificity for SHBG in transgenic mouse serum. In contrast, the pool of elution fractions 2-4 shows high specificity for SHBG in transgenic mouse serum and none of the background signal seen in the input and flow-through blots. A replicate experiment using elution fractions 5-7 (from figure 2.5) showed the same high specificity for SHBG as fractions 2-4 but required approximately 10-fold more exposure time to produce a signal of similar intensity (not shown). This indicated there was significantly lower antibody titer in these latter fractions and agrees with their lower staining intensity in Figure 2.5.  Next, to test the specificity of purified anti-LG4 antibodies to SHBG in human serum, serum samples obtained from a woman before and after treatment with a ethinylestradiols (EE2) were subjected to western blotting under denaturing and non-denaturing conditions (Figure 2.7). Ethinylestradiol is known to increase serum SHBG levels and both western blots show specific increases in SHBG abundance after EE2-treatment compared to pretreatment levels with minimal background staining. Replicate blots performed by preincubating the purified anti-LG4 antibodies with human serum albumin did not reduce the minimal background staining that is visible (not shown) suggesting that the non-specific bands are another species.     38 Finally, the ability of the purified anti-LG4 antibodies to recognize the testicular germ cell-specific isoform of human SHBG was tested. Lysates from whole seminiferous tubules and germ cell-enriched cellular fractions from both wild-type and 11K-SHBG transgenic mice were subjected to western blotting under denaturing conditions. Figure 2.8 shows detection of SHBG in only the testicular lysates from the 11K-SHBG mice. Not only does this isoform migrate slightly lower than serum SHBG but also it is more heterogeneous in its glycosylation [66]. Confirmation of germ cell enrichment is indicated by an increase in the ~17 kDa germ cell-specific CREM isoform [110] relative to whole seminiferous tubule lysates. Finally, there is a higher SHBG:GAPDH ratio in the germ cell-enriched fractions, further confirming this SHBG is of testicular germ cell origin since GAPDH production falls in later germ cells [18](Fig 2.8).        39     Figure 2.1 Chromatin shearing assay. Total chromatin from seminiferous tubules obtained from adult 11K-SHBG transgenic mice was subjected to shear cycles by probe sonication as indicated in section 2.7. Sheared chromatin was reverse-crosslinked, purified and subjected to 2% agarose gel electrophoresis and stained with SYBR. Increasing sonication intervals are shown above. 14 shear cycles produced a tight bolus of fragmented chromatin approximately 200-300 bp in size and was considered the optimal size for ChIP analyses. Molecular sizes in bp are indicated. M, molecular size marker.           40 Table 2.1 Oligonucleotide primers used in this thesis.  Chapter Assay / Mutation Primers (5’-3’):  F (sense), R (antisense) Template preparation Amplification Cycles (#) Annealing Temperature (oC) DNA Polymerase Ch. 3 altP-SHBG promoter insert F: ACTGCTCGAGTGAAGAGCCTGAGAGA Template diluted to 100 pg/μL, 2 μL used. 40 50 Taq R: GTCATTCGAAGCACTGCGGGGGAGCC Proximal CACCC mutagenesis F: CCTTATTCCCGACAGCTGGATCTT Same as above. 25 50 Phusion R: CTTGGGGGGGACGCGGCCGT Distal CACCC mutagenesis F: CCTTATTCCCTCGAATTCTGTCGC Same as above. 25 50 Phusion R: CTATGCCCGGACAGAGGATC Full length Klf4 transcript* F: CCTTCTGGGCCCCCACATTA cDNA diluted 1/10 in ddH2O, 2 μL used. 35 60 Taq R: CTGACTTGCTGGGAACTTGAC Gapdh F: GCTGAGTATGTCGTGGAGTC Same as above. 35 58 Taq R: TTGGTGGTGCAGGATGCATT Klf4 PAS I* F: GCAGGCTGTGGCAAAACCTATAC Same as above. 35 60 Taq R: CTGACTTGCTGGGAACTTGAC Klf4 PAS II* F: GCAGGCTGTGGCAAAACCTATAC Same as above. 35 60 Taq R: CACAATTCAAGGGAATCCTGG Klf4 PAS III* F: GCAGGCTGTGGCAAAACCTATAC Same as above. 35 60 Taq R: CTGCTTAAAGGCATACTTGGG Klf4 PAS IV* F: GCAGGCTGTGGCAAAACCTATAC Same as above. 35 60 Taq R: CTCACCTTGAGTATGCAAAATAC Alternative SHBG promoter ChIP F: CCGTACGGGAGGAGAGAGTA 2 μL used from 30 μL elution. 40 63.6 Taq R: CAGGAGCCCAGTCACAAGAT Gapdh promoter ChIP F: GATGATGGAGGACGTGATGG Same as above. 40 60 Taq R: GGCTGCAGGAGAAGAAAATG Sp1 F: CTCTGATCTCCAACCCCAAG cDNA diluted 1/10 in ddH2O, 2 μL used. 35 52 Taq R: GAATGGCCTCTCCCCTGTAT Sp3 F: GACGCCTGTTCAGACACTCA Same as above. 35 54.5 Taq R: GGGTCAGGTTCCTCTTCCTT Ch. 4 CRE mutagenesis F: TGAATTCGAGGGGGTGGG Template diluted to 100 pg/μL, 2 μL used. 25 57 Phusion R: TTCGCAGCAGGGGGCACA Ch. 5    OWM SHBG 1A-8 F: GTGGTGATAACCTGCTTTAGCC cDNA diluted 1/10 in ddH2O, 2 μL used. 40 65 AccuPrime Pfx R: CTTTAATGGGAAGCGTCAGT OWM SHBG 1-8 F: TGGACAGTGGCTGATTATGG Same as above. 40 65 AccuPrime Pfx R: CTTTAATGGGAAGCGTCAGT    41 Chapter Assay / Mutation Primers (5’-3’):  F (sense), R (antisense) Template preparation Amplification Cycles (#) Annealing Temperature (oC) DNA Polymerase Ch. 5   OWM TNP1 F: TGGCAGAACTTACCATGTCG cDNA diluted 1/10 in ddH2O, 2 μL used. 40 60 AccuPrime Pfx R: TAGGCTCCTCTCTGGCTTTG OWM Vimentin F: CCTTCGTGAATACCAAGACCTG Same as above. 40 60 AccuPrime Pfx R: TGATGCTGAGAAGTTTCGTTGA OWM GAPDH F: GCTGAGTATGTCGTGGAGTC Same as above. 35 58 AccuPrime Pfx R: TTGGTGGTGCAGGATGCATT NWM SHBG 1A-8 F: CGCAGTGCTTTTTAAATTGAC Same as above. 40 64 AccuPrime Pfx R: CTTTAATGGGAAGCATCAGTG NWM SHBG 1-8 F: TGGACAGCGGCTGACTATGG Same as above. 40 64 AccuPrime Pfx R: CTTTAATGGGAAGCATCAGTG NWM/human SHBG 3-8 F: TGGTTTATGCTGGGACTTCG Same as above. 40 62 AccuPrime Pfx R: GTCCACATCCAGCCTCTGAC NWM TNP1 F: GGCAGCAAAAGAAAATACCG Same as above. 40 60 AccuPrime Pfx R: GGGGAGAAACAGCCAACATA NWM Vimentin F: CCTTCGTGAATACCAAGACCTG Same as above. 40 60 AccuPrime Pfx R: TGATGCTGAGAAGTTTCGTTGA NWM GAPDH F: GCTGAGTATGTCGTGGAGTC Same as above. 35 58 AccuPrime Pfx R: TTGGTGGTGCAGGATGCATT GML SHBG 1-8 F: TGGACAGCCGCTGATTATGG Same as above. 40 65 AccuPrime Pfx R: CTCCGAGGACAGCTGTGAGT GML TNP1 F: GACCAGCCGCAAATTAAAGA Same as above. 40 60 AccuPrime Pfx R: CTCCTTAGCAGTCCCCCTTC GML Vimentin F: CTACACGACGAGGAAATCCAAG Same as above. 40 60 AccuPrime Pfx R: CTGTCTCCGGTACTCATTGGAC GML GAPDH F: GCTGAGTATGTCGTGGAGTC Same as above. 35 58 AccuPrime Pfx R: TTGGTGGTGCAGGATGCATT Underlined bases indicate recognition sequences for restriction enzymes. All primers used for Phusion mutagenesis were phosphorylated at their 5’ ends. *Primers sequences obtained from Godmann et al 2005 [111] OWM, Old World Monkey, NWM, New World Monkey, GML, grey mouse lemur   42        Figure 2.2 Selection of candidate antisera for affinity purification. Serum samples from wild-type and SHBG-transgenic mice [78] were subjected to non-denaturing PAGE on a 8% gel. The transferred membrane was cut and each piece was probed separately with a different rabbit-anti-human SHBG antisera (#4, #5, #7) diluted 1:6000. Membrane pieces were realigned together for chemiluminescent imaging in parallel. Arrowheads indicate specific staining for human SHBG.                43          Figure 2.3 Coupling of recombinant LG4 domain of human SHBG to NHS-activated Sepharose column. 10 µL each of the input LG4 and flow-through were subjected to 10% PAGE (resolving) with a 5% stacking gel under denaturing conditions. Proteins were stained using Bio-Safe Coomassie G-250 Stain (Bio-Rad) according to the manufacturer’s instructions. Molecular sizes in kDa are indicated. M, molecular weight marker (12 µL).       44         Figure 2.4 Affinity-purification of total IgG from candidate antisera. Total IgG fractions from pooled antisera (#4 and #5 from Figure 2.1) were isolated by protein A affinity-purification. 10 µL aliquots from each stage of purification were subjected to 12% PAGE (resolving) with a 5% stacking gel under denaturing conditions. Proteins were stained using Bio-Safe Coomassie G-250 Stain (Bio-Rad). 1, input, 2, flow-through, 3, wash, 4-10, fractions #1-7, respectively. Input and flow-through samples were diluted 1:12 prior to electrophoresis. Molecular sizes in kDa are indicated. M, molecular weight marker (12 µL).     45         Figure 2.5 Affinity-purification of anti-LG4 antibodies. IgG antibodies specific to the LG4 domain of human SHBG were purified from total serum IgG pools in Figure 2.3. 10 µL aliquots from each stage of purification were subjected to 12% PAGE (resolving) with a 5% stacking gel under denaturing conditions. Proteins were stained using Bio-Safe Coomassie G-250 Stain (Bio-Rad). 1, input, 2, first flow-through, 3, second round flow-through, 4, wash, 5-11, fractions #1-7, respectively. Molecular sizes in kDa are indicated. M, molecular weight marker (12 µL).     46      Figure 2.6 Affinity-purification efficiency test on mouse serum. Serum samples from wild-type and SHBG-transgenic mouse [78] were subjected to non-denaturing PAGE on a 8% gel. The transferred membrane was cut and each piece was probed separately with either input IgG pool, flowthrough IgG off LG4-affinity column or pooled fractions 2-4 as the primary antibody, each diluted 1:6000. Membrane pieces were realigned together for chemiluminescent imaging in parallel. Arrowheads indicate specific staining for human SHBG.       47       Figure 2.7 Performance test of affinity-purified antibodies on human serum. Serum samples obtained from a woman before and after treatment with ethinylestradiol (EE2) were diluted 1:1000 and 15 µL of each was subjected to 10% denaturing and 8% non-denaturing PAGE, followed by western blotting using pooled fractions 2-4 from affinity purification (diluted 1:6 000) as the primary antibody. M and D indicate human SHBG monomers and dimers, respectively. Molecular sizes from the denaturing gel are indicated in kDa.      48     Figure 2.8 Performance test of affinity-purified antibodies on mouse testicular homogenates. 15 µL of transgenic (TG) mouse serum diluted 1:5000 and 22 µg of protein from tissue lysates obtained from whole seminiferous tubules or germ cell-enriched fractions were subjected to SDS-PAGE followed by western blotting. Primary antibodies were diluted as follows: pooled fractions 2-4 from anti-LG4 affinity purification (1:6000), anti-GAPDH (1:400), anti-pan-CREM (1:1000). Open bracket indicates specific detection of human SHBG in either transgenic (TG) mouse serum or in testicular germ cells from 11K-SHBG transgenic mice. SHBG in serum and germ cells each exhibit heterogeneous but distinct glycosylation patterns.          49 Chapter 3: Transcriptional Activation from the Alternative SHBG Promoter by KLF4  3.1 Introduction  The major transcription unit of the human sex hormone-binding globulin (SHBG) gene encodes the SHBG precursor that is processed in the endoplasmic reticulum of hepatocytes and secreted into the bloodstream [65], [68]. It comprises an ~800 bp promoter sequence that flanks exon 1 which contains the AUG translation start site and encodes the secretion signal polypeptide and the two N-terminal residues of the mature SHBG sequence [60], [68], [112]. Analyses of numerous other SHBG transcripts have revealed that many comprise alternative exon 1 sequences located from between 1.9 kb and 17 kb upstream from the major SHBG transcription unit expressed in the liver (Figure 3.1a). At least six different transcriptional start sites have been identified along the SHBG gene that splice directly to exon 2 [58], [59], [113], [114]. Importantly, the only alternative SHBG transcript that results in a truncated SHBG isoform is the one expressed in testicular germ cells containing exon 1A [66], [67], [96]. This feature of the human SHBG gene is recapitulated in the testes of our humanized transgenic mice that carry an 11 kb region of the human SHBG locus (11K-SHBG) [66], [78].  The main SHBG transcript found in testicular germ cells appears to be under the control of the promoter sequence that flanks alternative exon 1A (Figure 3.1b). Compared to the proximal promoter that controls SHBG expression in the liver, this alternative promoter produces very high transcriptional activity in the immortalized mouse spermatocyte cell line (GC-2) within the context of a luciferase reporter assay [67]. The different relative promoter activities in GC-2 cells    50 mimic the situation in vivo, since alternative SHBG transcripts in both human and 11K-SHBG are confined to the testis, are transcribed from the alternative transcription unit containing exon 1A and do not contain the exon 1 sequence that characterizes the SHBG mRNA found in the liver.  I first interrogated the alternative promoter sequence that drives SHBG expression in the testis for known regulatory elements or motifs. Presumed elements of importance were tested by mutagenesis to assess their impact in a luciferase reporter gene assay, as described previously [67]. I then determined whether factors that might interact with these elements share related patterns of expression with SHBG during spermatogenesis in vivo.  3.2 Results  3.2.1 High transcriptional activity from the alternative SHBG promoter in GC-2 cells  First, I sought to confirm data from earlier reporter assay experiments [67]. To do this, I produced a new reporter construct encoding the -336/+28 bp alternative SHBG promoter (altP-SHBG) incorporated into the pGL3 luciferase reporter plasmid. When transfected into mouse GC-2 cell line the reporter activity of altP-SHBG was over 150-fold above background (P<0.0001) while the construct containing the SHBG proximal promoter used in hepatocytes failed to produce significant activity above background (Figure 3.2).       51 3.2.2 Optimal altP-SHBG reporter activity is dependant on two intact CACCC elements  I then asked what features inherent within the alternative SHBG promoter contributed to its basal activity in GC-2 cells. Sequence analysis of this region revealed several motifs and features of interest (Figure 3.3a). The -336/+28 region has 65.5% guanine-cytosine (GC) content and is rich in 5’-cytosine-phosphate-guanine-3’ (CpG) sites with 23 CpG sites on the sense strand and 22 on the antisense. This yielded a CpG frequency of ~14.6% on the sense strand and therefore falls within the accepted definitions of a CpG island [115], [116]. An online CpG island locator tool (http://www.bioinformatics.org/sms2/cpg_islands.html) confirmed this finding by scoring a CpG island from -336 to -55 along this region with an observed/expected ratio of CpG sites at >0.6. Additionally, three candidate regulatory elements were identified. Two identical CACCC elements located at -198/-189 and -111/-102 with 100% match to the consensus CACCC motif and one partial cAMP response element (CRE) located at -180/-173 with a 50% match to consensus CRE [117]. My sequence analyses also identified partial SPZ1 binding sites along this region as has been reported earlier [67]. However, since the SPZ1 motif hits from different analysis programs did not all map to the same location, I did not pursue this further.  To test the importance of the two CACCC elements on the basal transcriptional activity of the altP-SHBG construct, three mutant constructs were generated by site-directed mutagenesis using the wild-type construct as template. The distal, proximal or both of the CACCC elements were altered to be poor in GC content. These were transfected in parallel into GC-2 cells and their reporter activities were compared to the wild-type sequence. Figure 3.3b shows that while no changes in reporter activity are observed when only one CACCC element is mutated, disruption of both elements reduces this activity to ~25% of wild-type (P<0.0001). Since double CACCC    52 element disruption causes a significant reduction in altP-SHBG reporter levels I hypothesized that an endogenous factor within GC-2 cells must bind to either of these elements and drive optimal transactivation of the reporter gene.   3.2.3 Expression and localization of Klf4 in GC-2 cells  The specificity protein (SP) family of transcription factors are well known to interact with CACCC motifs [118], [119]. Within adult mouse testes intense immunolocalization of SP1 and SP3 protein levels are only found in outer spermatogonial cells of the mouse testis where SHBG is not expressed [120]. The Krüppel-like family of transcription factors are also well known to interact directly with CACCC elements [118], [119]. Previous studies have demonstrated that Klf4 is expressed during mouse spermatogenesis in round spermatids [121], [122]. This temporal pattern of expression bears striking similarity to when SHBG transcripts rise during the spermatogenic cycle and when SHBG is first observed in stage VII round spermatids within our 11K-SHBG transgenic mice [66], [78]. To determine whether Klf4 is expressed in GC-2 cells, RT-PCR was performed on GC-2 cell cDNA using primers positioned at the 5’ and 3’ UTRs of the Klf4 gene. Figure 3.4a shows detection of the full length 1613 bp Klf4 open reading frame as previously described in the mouse testis [111], as well as several possible splice variants. Western blots performed on separate nuclear and cytoplasmic fractions from GC-2 cells demonstrated that KLF4 is confined almost entirely within the nucleus (Figure 3.4b). Using a common forward primer positioned at exon 3 and unique reverse primers positioned at the four known polyadenylation signals (PAS) along the Klf4 3’UTR I was able to identify usage of four PAS in GC-2 cells (Figure 3.4c).      53 3.2.4 High KLF4 levels are required for optimal altP-SHBG reporter activity in GC-2 cells  To test the impact of reduced KLF4 protein levels on altP-SHBG transcriptional activity, GC-2 cells were cotransfected with an siRNA targeting Klf4 (si-Klf4) and the altP-SHBG reporter construct. Figure 3.5a clearly shows that a decrease in KLF4 protein levels significantly reduces the activity of altP-SHBG compared to the control siRNA (P<0.001). Moreover, cotransfection of a Klf4 overexpression construct with si-Klf4 rescued the effects of knockdown compared to cotransfection with an empty vector (pcDNA) (Figure 3.5b).  3.2.5 KLF4 localizes to the alternative SHBG promoter in vivo  To test the interaction of KLF4 with the alternative SHBG promoter in vivo, chromatin immunoprecipitation analysis was performed on chromatin preparations from the seminiferous tubules from our 11K-SHBG transgenic mice. Figure 3.6 shows that the -250/-73 region of the alternative SHBG promoter spanning both CACCC elements can be detected following pull down with KLF4-specific antibodies.   3.2.6 Related expression patterns of Klf4 and SHBG in transgenic mice in vivo  Early characterization of 11K-SHBG transgenic mice revealed a stage-dependent expression pattern of SHBG transcripts during the seminiferous cycle [78]. Intriguingly, peak levels of Klf4 transcripts are also found in round spermatids around stage VII [121], which is also when SHBG transcripts reach their maximum. Figure 3.7 shows the overlayed expression profiles of SHBG and Klf4 from these two separate studies indicating that they are coexpressed at stage VII.    54 3.2.7 High KLF4 protein levels in stage VII/VIII round spermatids  To assess the levels of KLF4 protein during spermatogenesis, immunofluorescent staining was performed on testes of our 11K-SHBG transgenic mice. Figure 3.8 shows intense staining for KLF4 in round spermatids during spermatogenic stages VII/VIII, whereas KLF4 was barely detectable in stage IV/V round spermatids.   3.2.8 All-trans retinoic acid induces altP-SHBG reporter activity  A recent report demonstrated that retinoic acid upregulates Klf4 gene expression in vascular smooth muscle cells and is mediated through retinoic acid receptor-alpha (RARα) [123]. To explore if a similar mechanism exists in GC-2 cells, I performed 24 hour doses of increasing all-trans retinoic acid (ATRA) concentrations on transfected GC-2 cells. Figure 3.9a shows that doses of 2-30 uM of ATRA significantly increase altP-SHBG reporter activity in a dose-dependent manner with a maximal response between 5 uM and 10 uM. Western blots indicate that increasing ATRA concentrations result in biphasic decreases in KLF4 protein level between 0.5-5 uM and 5-20 uM doses despite significantly increased reporter activity (Figures 3.9b,c).  3.2.9 Retinoic acid receptor signalling induces altP-SHBG reporter activity  To determine whether ATRA-induced upregulation in Figure 3.9a is mediated through retinoic acid receptors (RARs) transfected GC-2 cells were cotreated with ATRA and the pan-RAR antagonist AGN194310. In the presence of the antagonist, ATRA treatment failed to increase altP-SHBG reporter activity (Figure 3.10).     55 3.3 Discussion and future directions   After the completion of meiosis, haploid testicular germ cells undergo a massive wave of transcription that results in the expression of numerous genes required for spermiogenesis [124], [125]. The burst of SHBG transcription that occurs in stages VI-VII round spermatids [78] are likely part of this wave and their tightly regulated cyclical expression pattern suggests that mechanisms controlling their production are also temporally regulated. Round spermatids in vivo produce a specific environment of factors and mechanisms that permit transcription from exon 1A on the SHBG gene. GC-2 cells recapitulate this feature of round spermatids with high transcriptional activity from the alternative promoter while failing to produce any signal from the proximal SHBG promoter. Basal altP-SHBG reporter activity requires one intact CACCC element. Therefore the presence of two CACCC elements appears to be functionally redundant in the context of my reporter model.  KLF4 plays contrasting roles in different cellular contexts. It is required for the proper formation of the skin barrier as well as goblet cell differentiation in the digestive tract [126], [127], yet is also one of the four Yamanaka factors that can induce pluripotency when introduced with Oct4/Pou5f1, Sox2 and c-Myc [128]. In differentiating sperm, KLF4 is transiently expressed in mid-to-late round spermatids in both the mouse and humans with transcript and protein levels undetectable by the time the cells are elongated [121], [122], [129]. This transient expression pattern suggests that KLF4 plays a role in regulating post-meiotic gene expression. Surprisingly, mice with Klf4 ablated from their germ cells are completely fertile with no aberrations in spermatogenesis, nor differences in litter sizes sired from knockout males [122]. Despite being    56 dispensable for spermatogenesis, significant disturbances in testicular gene expression resulted in the mutant testes. By microarray analysis, the authors found 75 probe sets exhibiting downregulation and 90 probe sets exhibiting upregulation compared to wild-type. This illustrated that KLF4 plays dual roles as both a transcriptional activator and repressor in the testis.  In GC-2 cells KLF4 localizes to the nucleus, consistent with its location in vivo in both mouse and human round spermatids [121], [122], [129]. In the mouse testis, Klf4 transcripts utilize four polyadenylation signals (PAS) [111]. Sertoli cells utilize only the first PAS, while the latter three are considered markers for the presence of post-meiotic germ cells in the mouse testis. The observed pattern of Klf4 expression and localization in GC-2 cells fully resembles its pattern in round spermatids in vivo and therefore supports the use of this cell line as a tool to model testicular germ cell expression of Klf4.   Transcriptional activity from the alternative SHBG promoter is significantly reduced when KLF4 levels are depleted by knockdown. Activity is restored by Klf4 overexpression and therefore strongly suggests that KLF4 is a key transcriptional activator of SHBG in testicular germ cells. In differentiating stem cells KLF4 has been identified as a pioneer transcription factor that can access regions of the genome that are otherwise inaccessible due to ‘closed-state’ higher order chromatin structures (heterochromatin) [130], [131]. Upon engaging exposed motifs on heterochromatin, pioneer factors are thought to further recruit chromatin modifiers, remodelers and other transcriptional coregulators to render these regions permissive for transcription [132], [133]. In support of this model, KLF4 has been shown to recruit the p300 coactivator to the promoters of KLF4-target genes, where it acetylates nearby histone tails to promote chromatin    57 decondensation [134], [135]. The bridging function of KLF4 between promoter elements and transcriptional coregulators implies that it is part of a larger combination of factors that control SHBG expression. Nearly 50% of altP-SHBG reporter activity remains following Klf4 knockdown and approximately 25% remains when both CACCC elements are disrupted suggesting that there may be other factors and elements at work.  Transcriptional activation of SHBG by KLF4 is further supported by their very similar expression patterns during the seminiferous cycle. Separate studies indicate that both SHBG and Klf4 transcripts peak during stage VII seminiferous tubules (Figure 3.7). My immunofluorescence staining shows that KLF4 protein levels in round spermatids rise sharply from stages V-VII, the latter stage being when SHBG protein levels are first detected in round spermatids [66]. Dual immuno-labelling for SHBG and KLF4 at each stage of spermatogenesis on testes from 11K-SHBG mice is underway to confirm their co-expression. The use of an affinity-purified polyclonal antibody to human SHBG (as described in Chapter 2) that does not cross-react with mouse SHBG will allow specific detection of human SHBG during the seminiferous cycle in these animals.  ChIP assays indicate that KLF4 physically interacts with the alternative SHBG promoter in vivo along a region that spans both CACCC elements. Chromatin for these assays was obtained from a mixed population of Sertoli and germ cells. While Klf4 is expressed at low levels in differentiating Sertoli cells of postnatal mouse testes [121], [136], I was only able to detect KLF4 in round spermatids of adult 11K-SHBG mouse testes (Figure 3.8). Therefore immunoprecipitation of KLF4 is expected to be of germ cell origin only. To test the dependence    58 on KLF4 interaction on the CACCC elements, ChIP assays will be performed on the intact and three mutated altP-SHBG reporter constructs (Figure 3.2b) after transfection into GC-2 cells. Following pull-down with KLF4 a decrease in promoter enrichment would be expected from the fully mutated construct compared to the intact. This strategy is an improvement over electrophoretic mobility shift assays since protein-DNA interactions are tested within a context that results in transcriptional activity [137]–[140].  While Klf4 overexpression restored the effects of knockdown it did not consistently augment reporter activities above the control levels. This suggests that the reporter construct is fully saturated with KLF4 or that KLF4 must also be further activated by post-translational mechanisms to elicit an increased response. Interestingly, all-trans retinoic acid (ATRA) has been shown to not only increase Klf4 gene expression in vascular smooth muscle cells (VSMCs) but also increase KLF4 acetylation leading to its increased recruitment to the promoters of downstream target genes [123], [141]. In VSMCs, ATRA-induced upregulation of Klf4 is specifically mediated by RARα. My results indicate that indeed retinoid receptors mediate ATRA-induced upregulation of altP-SHBG transcriptional activity, however this was accompanied by various degrees of decreased KLF4 protein levels. This was surprising given that knockdown of Klf4 consistently resulted in decreased altP-SHBG reporter activity in Figure 3.5a. Klf4 gene expression in GC-2 cells may therefore exhibit complex regulation by retinoid receptors. Potential changes in KLF4 acetylation following ATRA-stimulation are not ruled out, however RAR-dependent increases in altP-SHBG reporter activity may also be mediated by other indirect mechanisms that are entirely independent of KLF4. In support of these data, RARα    59 expression peaks in stages VII-VIII round spermatids of the rat seminiferous cycle, which correspond to stage VII in the mouse [142].   Results in this chapter indicate that transcriptional activity from alternative SHBG promoter is regulated by cis-elements within the -336/+28 region. Another mode of gene regulation implicated in testicular germ cells are changes in the methylation status of 5’ CpG loci within the promoters of temporally expressed genes. The promoter flanking the phosphoglycerate kinase 2 gene is constitutively methylated in somatic tissues but undergoes demethylation in pachytene spermatocytes and round spermatids where it is expressed [143]. Trasler and colleagues [144] found that the promoter flanking the transition protein 1 gene undergoes demethylation in pachytene spermatocytes and this persists throughout meiosis. In both cases, temporally expressed genes are epigenetically repressed by promoter methylation until they are required. Given that demethylation events occur on temporally expressed genes during spermatogenesis, it is possible that this type of regulation also contributes to the temporal expression pattern of germ cell SHBG since the promoter controlling this transcription unit is a CpG island. Since SHBG transcripts peak in stage VII-VIII round spermatids in our 11K-SHBG transgenic mice I hypothesize that the alternative SHBG promoter undergoes prior to its upregulation. During post-natal development, SHBG transcripts in 11K-SHBG mouse testes begin to rise steadily after day 20 which is coincident with the first appearance of post-meiotic germ cells [145], [146] and is also when Klf4 transcripts dramatically increase [121]. Thus the alternative SHBG promoter may be methylated in DNA isolated from <20 days old 11K mouse testes but exhibit decreased methylation in testicular DNA from progressively older mice. The impact of CpG methylation on transcription from the alternative SHBG promoter could also be tested directly. Authors Klug and    60 Rehli have described the utility of a CpG-less luciferase reporter construct (pCpGL) to test the effects of in vitro methylation on promoter activity [147]. My -336/+28 altP-SHBG insert could be incorporated into this vector, subjected to in vitro methylation and assayed for luciferase activity compared to an unmethylated control. The expectation would be that the methylated construct exhibits less transcriptional activity compared to the unmethylated control.       61     Figure 3.1 Alternative SHBG transcription units. A, There are at least six known alternative transcription units on the human SHBG gene used in different tissues. The major transcription unit used by the liver initiates from exon 1 and results in a mature protein secreted into the bloodstream [65], [68]. Transcription from exon 1A occurs in testicular germ cells of humans and humanized 11K-SHBG transgenic mice [58], [59], [66], [67], [78], [96]. The remaining transcriptional start sites are used in either the human prostate or human cancer cell lines [113], [114] where their significance is unknown. B, The proximal and major alternative SHBG transcription units are each controlled by their own distinct promoter regions [68]. Transcripts arising from exon 1A in testicular germ cells splice directly to exon 2, bypassing the secretion polypeptide yet retain the steroid-binding domain and results in a truncated intracellular isoform of SHBG. Relative positions of exons in A and B are not to scale.       62            Figure 3.2 High transcriptional activity of the human alternative SHBiG promoter in GC-2 cells. The basal luciferase reporter activities of the proximal (-266/+366) and alternative             (-336/+28) SHBG promoters were assessed in GC-2 cells. Relative luminescence counts are normalized to a promoterless pGL3 basic reporter vector that was assayed in parallel. Data points are means ± SD of triplicate measurements. P-values and statistical significance were determined by one-way ANOVA.     63   Figure 3.3 Dependence of altP-SHBG transcriptional activity on CACCC elements. A, Sequence analysis revealed two CACCC elements (underlined and bolded) and one partial CRE (bolded) within the minimal region of the alternative SHBG promoter. The transcriptional start site from this region [58] is also shown (underlined and arrow). Bases shown in blue are the first 28 nucleotides of exon 1A. B, The basal luciferase activities of altP-SHBG reporter constructs with single or double mutated CACCC elements were assessed in GC-2 cells. Relative luminescence counts are normalized to a promoterless pGL3 basic reporter vector that was assayed in parallel. Mutagenesis strategy is indicated. Data points are means ± SD of triplicate measurements. P-values and statistical significance were determined by one-way ANOVA.            64      Figure 3.4 Expression and localization of Klf4 in GC-2 cells. A, Full length mouse Klf4 transcripts are detected in GC-2 cells by RT-PCR and electrophoresis in a 1% agarose gel. Molecular size in bp is indicated. Integrity of the cDNA in the sample is assessed by analysis for Gapdh transcripts. B, Western blot for KLF4 are performed on GC-2 cell cytoplasmic and nuclear fractions. Blots for GAPDH and histone H4 were performed on the same membrane to confirm the identity of the cytoplasmic and nuclear fractions, respectively. Molecular mass in kDa is indicated. C, Identification of four Klf4 PAS utilized by GC-2 cells by RT-PCR and electrophoresis in a 1% agarose gel. Molecular size in bp is indicated. RT, reverse transcriptase.             65   Figure 3.5 High KLF4 levels are required for optimal altP-SHBG reporter activity in GC-2 cells. A, The effect of reduced KLF4 levels on altP-SHBG reporter activity was assessed in GC-2 cells by cotransfection of the altP-SHBG reporter construct with either an siRNA targeting Klf4 (si-Klf4) or non-targeting control (si-ctrl). Confirmation of knockdown was assessed by western blots for KLF4 on replicate GC-2 cell lysates from the same experiment. B, The effect of combined Klf4 knockdown and overexpression on altP-SHBG reporter activity was assessed in GC-2 cells. Replicate knockdown experiments in A were cotransfected with either a Klf4 overexpression construct or empty vector (pcDNA). For both A and B data points are means ± SD of triplicate measurements from three independent experiments. Data are normalized to the first mean in each histogram. P-values and statistical significance were determined by Student’s t-test. ns, not significant.        66       Figure 3.6 KLF4 localizes to the alternative SHBG promoter in vivo. Chromatin isolated from seminiferous tubules from 11K-SHBG transgenic mice was fragmented by sonication and subjected to immunoprecipitation (IP) with antibodies against KLF4 followed by PCR amplification of the alternative SHBG promoter (-250/-73). Chromatin integrity was assessed by IP with antibodies against RNAPII followed by PCR amplification of the Gapdh promoter. Input conditions were IPs performed with 1% of the chromatin reserved from IP with specific antibodies. PCR products were analyzed on a 1.5% agarose gel.                          67         Figure 3.7 Co-expression patterns of Klf4 and SHBG in transgenic mice in vivo. The relative levels of SHBG transcripts in 11K-SHBG transgenic mice are plotted ±SD [78]. The shaded bar indicates the stage of the mouse spermatogenic cycle where Klf4 transcripts are at their highest levels (i.e. around spermatogenic stage VII) [121]. These separate studies indicate that SHBG and Klf4 are coexpressed during the mouse seminiferous cycle.            68   Figure 3.8 High KLF4 protein levels in stage VII/VIII round spermatids. Immunofluorescent staining for KLF4 was performed on thin sections from 11K-SHBG transgenic mice. Mouse spermatogenic stages were defined by the progress of acrosome development using a fluorophore-conjugated lectin (PNA-lectin) ([103], [148], [149]. Stage IV/V seminiferous tubules contain earlier round spermatids with proacrosomal granules labelled as single red dots and mature sperm heads with fully formed acrosomes (white arrow heads). Stage VII/VIII tubules contain round spermatids with large acrosomal structures that are spread around nuclei (white arrow heads-right) labelled with KLF4 (merge). No staining for KLF4 in Sertoli cells was observed.    69     Figure 3.9 The alternative SHBG promoter responds to all-trans retinoic acid. A, Effect of 24 hr treatment with six doses of all-trans retinoic acid (ATRA) on altP-SHBG reporter activity in GC-2 cells. Data points are means ± SD of triplicate measurements. P-values and statistical significance were determined by one-way ANOVA. vh indicates DMSO vehicle dosing. B, Whole cell extracts from replicate dosing conditions in A were subjected to western blotting for KLF4. C, The relative protein levels of KLF4 to GAPDH from western blots in B were performed by densitometry using ImageJ. Relative luminescence values are normalized to vehicle controls assayed in parallel. Densitometry analysis is from a single experiment.       70         Figure 3.10 Retinoid receptors mediate the retinoic acid-induced upregulation of altP-SHBG reporter activity. A, The effect of the pan-retinoic acid receptor AGN194310 on ATRA-induced altP-SHBG reporter activity in GC-2 cells. Data points are means ± SD of triplicate measurements. Relative luminescence values are normalized to no treatment controls assayed in parallel. P-values and statistical significance were determined by one-way ANOVA. nt and vh indicate no treatment or DMSO vehicle dosing, respectively.       71         Figure 3.11 Expression of Sp1 and Sp3 in GC-2 cells. RT-PCR analysis for the coding regions of Sp1 and Sp3 transcripts using cDNA derived from GC-2 cells. PCR products were subjected to electrophoresis on a 1.5% agarose gel. RT, reverse transcriptase.      72 Chapter 4: Transcriptional Regulation from the Alternative Sex Hormone-Binding Globulin Promoter by protein kinase A and CREM  4.1 Introduction  In chapter three, sequence analysis of the alternative SHBG promoter alerted me to the presence of partial 3’,5’-cyclic adenosine monophosphate response element (CRE) located 180 bp upstream from the transcription start site. These CREs are known to mediate changes in transcriptional activity of their flanking gene in response to increased cellular levels of 3’,5’-cyclic adenosine monophosphate (cAMP) [117]. Classically, the stimulus begins with the binding of an extracellular ligand to a G-protein coupled receptor (GPCR) on the surface of a cell. GPCRs are large, multi-domained transmembrane proteins that are frequently hormone receptors [150]. Their extracellular domains receive inputs from their environment whereas their cytosolic domains form complexes with the G-proteins (Gα, Gβ, Gγ) and transmit signals within the cell. When a GPCR is stimulated, Gα diffuses away from the receptor’s cytosolic domain and associates with adenylyl cyclase (AC). This interaction permits AC to synthesize cAMP from adenosine triphosphate. Increased levels of cAMP allow activation of protein kinase A (PKA) [151]. Activated PKA can phosphorylate numerous targets within a cell and is thus able to modulate a wide range of cellular activities.  Transcriptional regulation in response to activated PKA is mediated mainly through the transcription factors CREB, CREM and ATF-1. These interact with CREs in the promoter regions of cAMP-responsive genes. In order to influence transcriptional activity they must be    73 phosphorylated by PKA or another kinase to permit their interaction with p300 and CBP co-activators to ultimately activate or repress transcription [151]. The transcription factor CREM is abundantly produced in mammalian round spermatids and male mice lacking the CREM gene are infertile and their germ cells do not progress through spermiogenesis [152]–[156]. Intriguingly, stage VII-VIII round spermatids exhibit intense staining for CREMτ [25] when SHBG transcripts are also highly expressed [78] and first become translated [66]. Using the transient luciferase reporter system described in Chapter three, I tested how cellular PKA activity and CREM overexpression influence transcriptional activity from the alternative SHBG promoter.  4.2 Results  4.2.1 Increased protein kinase activity causes induction of alternative SHBG promoter  To test the cAMP-responsiveness of my alternative SHBG promoter reporter system (altP-SHBG), transfected GC-2 cells were treated with increasing doses of dibutyryl cAMP (dbcAMP) for 24 hours prior to the assay end point. A robust, dose-responsive increase in altP-SHBG activity was observed (Figure 4.1a). Similarly, the AC-activator forskolin also increased reporter activity (not shown). To determine if this effect was mediated through PKA transfected GC-2 cells were co-treated with a fixed dose of dbcAMP in the presence of increasing concentrations of the PKA inhibitor, H89. Increasing doses of H89 overcame the inductive effects of dbcAMP on reporter activity with no changes in cell viability (Figure 4.1b).      74 4.2.2 A partial CRE element in the human alternative Sex hormone-binding globulin promoter does not mediate the activating effects of cAMP  I then asked whether dbcAMP-dependent upregulation of the altP-SHBG is mediated through the partial CRE. A mutant construct was generated by site-directed mutagenesis so that the partial CRE was altered from 50% to 25% of the consensus sequence. Both constructs were transfected into GC-2 cells and their reporter activities were assessed under basal and dbcAMP-stimulated conditions. While the mutated construct exhibited a nearly two-fold higher basal activity compared to control, it displayed the same magnitude of response to treatment with dbcAMP (Figure 4.2).   4.2.3 Overexpression of CREM isoforms induced alternative SHBG promoter activity  To determine whether altP-SHBG reporter activity could be directly influenced by CREM transcription factors, GC-2 cells were co-transfected with the altP-SHBG reporter and an overexpression construct encoding either the tau (τ) or delta-CG (ΔCG) isoforms of CREM. Figure 4.3a shows that overexpression of CREMτ results in an over 1.5-fold increase in altP-SHBG reporter activity (p<0.001) while CREMΔCG overexpression resulted in a nearly three-fold increase (p<0.0001). Increased cellular levels of individual CREM isoforms were demonstrated by western blots in Figure 4.3b. Interestingly CREMτ overexpression also resulted in the production S-CREM, a ~21 kDa repressor isoform of CREM that is translated from an alternative initiation codon within the CREMτ transcript [157].      75 4.2.4 Overexpression of CREMτ potentiates altP-SHBG’s response to dbcAMP  The CREMτ isoform contains a central kinase inducible domain (KID) that can be phosphorylated by PKA as well as other kinases [158], [159]. To test the combined effect of CREMτ overexpression with increased PKA activity, co-transfection of the altP-SHBG reporter construct with CREMτ was repeated under basal and dbcAMP-stimulated conditions. While overexpression of CREMτ resulted in a modest increase in basal reporter activity (Figure 4.4a), combined treatment with dbcAMP and CREMτ resulted in a higher response compared to dbcAMP with no CREMτ. When these responses were re-normalized against their own vehicle controls, dbcAMP treatment in the absence of CREMτ resulted in a 1.5 ± 0.04 fold response, while the same treatment in the presence of CREMτ resulted in a 2.16 ± 0.37 fold response. While this indicated that there is a greater magnitude of response to dbcAMP in the presence of CREMτ, a two-way ANOVA statistical test revealed that the synergistic effect is not significant (P=0.0559).    4.3 Discussion and future directions  When cAMP levels are low, inactive PKA exists as a tetrameric holoenzyme consisting of two regulatory and two catalytic subunits [160]. Upon binding cAMP, the regulatory dimer dissociates from the complex, liberating the catalytic monomers and allowing them to phosphorylate various targets within the cell. With the use of a specific inhibitor to PKA, my results clearly indicate a role for PKA in the dbcAMP-stimulated increase in altP-SHBG reporter activity. In support of this, expression of the four regulatory subunit isoforms of PKA (RIα and β, RIIα and β) have been identified during the rat seminiferous cycle. Transcripts for all isoforms    76 are detectable in stage VII/VIII round spermatids [161]. Furthermore, transcripts for RIIα and RIIβ reach peak levels during these stages. A brief rise in cAMP levels has also been described during late stage VII/VIII tubules [162], [163]. This is important because the stage-dependent expression pattern of SHBG in 11K-SHBG transgenic mice strongly suggests that SHBG is expressed at peak levels in round spermatids during stage VII/VIII of the seminiferous cycle [66], [78]. My findings suggest that increased transcriptional activity from the alternative SHBG promoter results if activation of the cAMP/PKA pathway occurs during this window of spermatogenesis.  Initial characterization of CREM transcripts in the mouse pituitary revealed three splice variants (α, β, γ), all of which bind the consensus CRE and repress PKA-induced transcription from promoters containing a functional CRE [164]. In the testis, CREM transcripts in developing sperm undergo a functional switch in their splicing pattern from repressors in premeiotic cells to an activator isoform (CREMτ) in meioitic and post-meitotic cells [152]. Later reports showed that splicing and processing of testicular CREM transcripts is even more complex as they exhibit the use of alternative polyadenylation and translational start sites [157], [165]–[167]. This also led to the identification of another post-meiotic repressor isoform of CREM that lacks the regulatory kinase-inducible (P-box) and transactivation (Q-rich) domains (CREMΔCG) [110]. Overexpression of CREMτ resulted in significantly increased altP-SHBG reporter activity and, surprisingly, overexpression of CREMΔCG caused an even greater response. This result was unexpected given that functional studies using reporter assays in JEG-3 cells found that CREMΔCG represses PKA-induced transcription on functional CRE elements [110]. The inhibitory potential of CREMΔCG was attributed to the lack of P-box and Q-rich domains in this    77 factor which prevent recruitment of other transcriptional co-regulators [110]. However, one other noteworthy difference between CREMΔCG and τ isoforms is their differential use of basic leucine zipper domains: Ia and Ib, respectively. CREM isoforms containing Ia are better able to heterodimerize with CREB that those containing Ib [168]. Therefore the higher altP-SHBG response to CREMΔCG versus CREMτ overexpression may be due to increased recruitment of CREB by CREMΔCG, resulting in higher levels of transactivation mediated via CREB.   In round spermatids CREMτ has been shown to associate with activator of CREM in the testis (ACT) that acts as a bridging factor to transcriptional machinery [169], [170]. This interaction bypasses the need for upstream signalling kinase activity to phosphorylate CREMτ at serine 117, which was previously considered a prerequisite for transactivation [158], [168]. The literature is not clear on the phosphorylation status of CREM isoforms in round spermatids. With the use of in vitro phosphorylation assays Delmas and colleagues [25] demonstrated that in the presence of cAMP, testicular PKA efficiently phosphorylates recombinant CREMτ but not CREM-Ser117Ala. In contrast, two later reports cite unpublished findings stating that CREM is un-phosphorylated in round spermatids in vivo [169], [170], yet these claims are unsubstantiated and remain unpublished. Clear evidence of CREM’s phosphorylation status in vivo is needed to resolve the molecular mechanisms of CREM signalling in the testis. The combined effects of increased PKA activity and CREMτ overexpression on altP-SHBG reporter activity suggest that these components work together to maximize the transcriptional response, whereas the combined effects of CREMτ and ACT overexpression on my reporter system showed no changes in reporter activity compared to controls (not shown). Others have reported activation of male germ cell-specific promoters by CREMτ overexpression alone implying that ACT is not essential for    78 transcription of these CREM-dependant genes [139], [140], [171]. Consistent with these findings, while CREM-null male mice are infertile, ACT-null males are fertile and exhibit no significant disturbances in testicular gene expression [172], [173].   My results suggest that regulation of the SHBG gene in male germ cells by CREM is highly complex and may involve indirect regulation. Importantly, the effects of CREM isoform overexpression were most pronounced in experiments performed on earlier passage GC-2 cells (Figure 4.3a), while similar experiments performed on later passage cells resulted in only modest increases on reporter activity (4.4a). It is possible that GC-2 cells undergo drifts in their phenotype with subsequent passages and may experience changes in expression levels of co-regulator or intermediate factors that are required for CREM-mediated induction of altP-SHBG. When the GC-2 cell line was derived the authors indicated that these cells express several germ cell-specific genes [104] and this phenotype persisted through 30 passages. However a follow-up report on GC-2 cell phenotype indicated a failure to detect these spermatogenic markers [174]. While not indicated, these further analyses were probably performed on even later-passaged cells suggesting a loss of the germ cell phenotype.  In my earlier experiments, dbcAMP treatment periods of 0.25, 1, 6 and 12 hrs resulted in no or minimal reporter responses (not shown). The optimal dosing period for altP-SHBG activation was for 24 hrs and this time point was incorporated into all subsequent dosing experiments (Figures 4.1, 4.2, 4.4). The immediate transcriptional responses initiated by the cAMP-PKA-CREB/CREM pathway are rapid and occur within minutes of cellular stimuli [175], [176]. Given that my reporter system produced a maximal response within 24 hrs, it is likely that the response    79 observed is an indirect mechanism, possibly requiring the PKA-mediated induction of CREM or one or more other genes.   The consensus CRE is a palindromic octamer consisting to two functional half sites (5’-TGACGTCA-3’) that interact with CREB and CREM. Variant CREs that respond to cAMP and interact with CREB and CREM proteins have at least one intact half site [177]. In light of the fact that the partial CRE in the alternative SHBG promoter lacks a consensus half site and that mutation of this region does not change the altP-SHBG reporter’s response to dbcAMP treatment, my conclusion is that this element is non-functional in this context.    My studies have only begun to look at the effect of CREM on transcriptional activity of the alternative SHBG promoter and these results warrant further investigation. It is known that GC-2 cells are a rich in CREM transcripts [178] and a next step could be to silence all CREM gene products expressed in these cells and then introduce back individual CREM variants or combinations of them by overexpression and to assess their effects using the altP-SHBG reporter system. The requirement for CREMτ phosphorylation on altP-SHBG reporter activity must also be resolved. The phosphorylation status and cellular localization of CREMτ in GC-2 cells should be determined under basal and dbcAMP-stimulated conditions as this will also help clarify the potential link between the upstream and downstream components of this pathway. Furthermore, my experiments in Figure 4.4 that suggest the presence of CREMτ may enhance the response to dbcAMP could be expanded upon on by adding another condition with CREMτ mutated at serine 117 (S117) so that it is phosphorylation deficient. CREMτ that cannot be phosphorylated at S117 may prevent the >2-fold response of altP-SHBG to dbcAMP.    80 Previous analysis of the the alternative SHBG promoter by progressive reporter construct deletions revealed that a minimal region containing the partial CRE yielded high transcriptional activity above background [67]. While this might suggest a role for the CRE in maintaining basal altP-SHBG transcriptional activity, the same minimal region also contains at least one CACCC element that is sufficient to drive optimal reporter activity in this context (Chapter 3, Figure 3).   Since adenylate cyclase (AC)-PKA-cAMP signalling is typically initiated in response to hormone stimulated G-protein coupled receptors (GPCRs), it is interesting to speculate a possible upstream mechanism that initiates this response in testicular germ cells. Round spermatids are insulated from direct endocrine signalling by Sertoli cell tight junctions and therefore the vast majority of endocrine signalling on germ cells is mediated through Sertoli cells. Not surprisingly, GPCRs in round spermatids are poorly characterized. However, an intracellular stimulator of AC, pituitary adenylate cyclase activating polypeptide (PACAP) has been found to be highly expressed in round spermatids [179], [180]. PACAP is abundantly found in spermatid cytosolic fractions and significantly increases AC activity as measured by elevated cAMP production [181]. Induction of AC activity by PACAP therefore represents a possible intracrine mechanism that initiates within spermatids without any stimuli by extracellular ligands.        81      Figure 4.1 Activated protein kinase A increases altP-SHBG reporter activity. A, The effect of increasing 24 hour doses of dbcAMP on altP-SHBG reporter activity in GC-2 cells. B, The effect of a 24 hour fixed dose of 5 mM dbcAMP and increasing concentrations of the PKA inhibitor H89 on altP-SHBG reporter activity in GC-2 cells. Nt and vh indicate no treatment and vehicle dosing, respectively. Data points are means ± SD of triplicate measurements. P-values and statistical significance were determined by one-way ANOVA. *** indicates statistical significance of P<0.0001 compared to vehicle control. § indicates statistical significance of P<0.0001 compared to 5 mM dbcAMP-only in B.      82            Figure 4.2 A partial CRE does not mediate altP-SHBG’s response to dbcAMP. The altP-SHBG reporter construct was used as template to generate a variant construct with the partial CRE mutated to 25% of the consensus CRE. The responses of each to 24 hr, 2 mM dbcAMP treatment was compared in GC-2 cells. Nt and vh indicate no treatment and vehicle dosing, respectively. Data points are means ± SD of triplicate measurements. P-values and statistical significance were determined by one-way ANOVA.      83      Figure 4.3 Overexpression of CREM isoforms increases altP-SHBG reporter activity. A, The altP-SHBG reporter construct was cotransfected with overexpression constructs encoding either CREMτ or CREMΔCG isoforms in GC-2 cells. Cotransfection with an empty pcDNA vector was used as a negative control. B, Cell lysates from experimental replicates in A were subjected to SDS-PAGE and western blotting for CREM and GAPDH. CREMτ was detected under low exposure whereas detection of a band for CREMΔCG was only detectable with a longer blot exposure. CREMτ overexpression also resulted in a second band at ~21 kDa that corresponds to S-CREM [157]. Data points in A are means ± SD of triplicate measurements. P-values and statistical significance were determined by one-way ANOVA. Solid arrowheads in B indicate an increased abundance of CREMτ and S-CREM while an open arrowhead indicates the same for CREMΔCG.   84     Figure 4.4 Overexpression of CREMτ potentiates altP-SHBG’s response to dbcAMP. A, The altP-SHBG reporter construct was cotransfected with a CREMτ overexpression construct and dosed with or without 2 mM dbcAMP for 24 hrs before assay collection. Cotransfection with an empty pcDNA vector with the same dosing strategy was used as a reference. All data points in panel A are normalized to the pcDNA/vehicle mean. B, To show the magnitudes of response to dbcAMP in the presence of either pcDNA or CREMτ, data points from A were normalized against the vehicle control of their own overexpression construct condition. Data presented are means ± SD of three independent experiments, each performed in triplicate. Two-way ANOVA statistical analysis found that there was not a significantly higher response to dbcAMP in the presence of CREMτ.     85 Chapter 5: Evolution of Sex Hormone-Binding Globulin gene expression in the primate testis  5.1 Introduction  It has been long established that sex hormone-binding globulin (also known as testicular androgen binding protein, ABP) is a product of the Sertoli cells in rodents [83]. A protein species with similar physical and steroid-binding properties as SHBG in serum was also demonstrated in human and primate testicular homogenates. However, its site of synthesis and cellular location was never confirmed and was always assumed to be a product of the Sertoli cells [84], [85], [95], [182].   Expression of Shbg in rodent Sertoli cells originates from exon 1 and transcription from this region is under the control of the immediately flanking promoter (Figure 5.1). This was clearly demonstrated when rat Shbg transcripts were abundantly expressed within the Sertoli cells of a transgenic mouse containing a genomic fragment of the rat Shbg gene [97]. In mice harbouring the corresponding region of the human SHBG gene, no testicular SHBG transcripts were detected [78]. Only when an 11 kb fragment of the human SHBG gene was inserted into the mouse genome were SHBG transcripts detected in their testes [78]. Follow up reports indicated that they are a product of the germ cells and comprise an alternative first exon (Figure 5.1) [66], [96]. It was therefore clear that at least two molecular mechanisms were at simultaneously at work: (1) constitutive transcriptional repression of the human SHBG gene in Sertoli cells and (2) temporal activation of transcription of SHBG from an alternative promoter in testicular germ cells.    86 The genetic determinants that underlie the unique expression pattern of human SHBG in the testis have begun to be explored. Constitutive repression of SHBG in Sertoli cells is entirely dependent on the presence of a 38 bp element that was previously identified to be a DNase foot-printed region in hepatocytes (foot-print 4, FP4) [60]. The centre of FP4 contains a binding site for upstream stimulatory factors (USF), and intriguingly, this element is the only major structural difference between the proximal promoters of the human and rodent SHBG/Shbg genes [98]. When a line of transgenic mice carrying a modified human proximal SHBG transcription unit that lacked FP4 was derived, repression of SHBG was reversed and high levels of SHBG production resulted in their Sertoli cells [98].  The mechanisms underlying SHBG expression in testicular germ cells have received less attention. It is clear that the alternative transcription unit is regulated in a stage-dependent manner under the control of an upstream alternative promoter [66], [78]. My studies in Chapter 3 strongly suggest that KLF4 is a novel transcriptional regulator at this region. In their phylogenetic analyses, Pinos and colleagues identified regions of near-perfect sequence identity to the human exon 1A within the SHBG genes of several higher primates [113]. Their findings implied that these could be functional alternative first exons but analysis of testicular SHBG transcripts in these species is lacking. In addition to a putative exon 1A the chimpanzee SHBG gene also contains the FP4 element [98]. Taken together this strongly suggests that SHBG expression in the chimpanzee testis is repressed in their Sertoli cells and is instead a product of the germ cells as in humans.     87 Considering the importance of these two genomic regions in directing the unique expression pattern of the human SHBG gene in the testis, I asked if they were common to all primates. With the aid of public nucleotide sequence databases and computer-based alignment tools I constructed a phylogenetic comparison of SHBG 5’ regulatory regions using sequences from all major primate groups with the mouse Shbg gene as an outgroup. In particular, the regions responsible for repression of human SHBG in Sertoli cells and expression in germ cells were examined. This analysis is strengthened with RT-PCR studies for proximal and alternative SHBG transcripts using biopsies and cDNA obtained from several primate testes. I present novel data on the nature of testicular SHBG gene expression throughout the major primate groups and show that macromutation events in their SHBG genes are associated with changes in how it is expressed.  5.2 Results  5.2.1 Alignment of SHBG 5’ regulatory regions across primates  Building on earlier species comparisons [98] the region of the proximal human SHBG promoter that contains the FP4/USF element was aligned against the corresponding regions of ten other primate SHBG genes. Sources of nucleotide sequences used in alignments are listed in Table 5.1. The respective positions of these species in the primate order are indicated in Figure 5.2. Figure 5.3 clearly shows that while the FP4/USF element is present in all higher primate SHBG genes, it is absent in the tarsier and lemur sequences like the mouse. I then asked which primates had regions of their SHBG genes that resembled the human exon 1A and its flanking promoter. Alignment of the human alternative SHBG promoter and exon 1A with upstream regions of ten    88 primate SHBG genes revealed that all simians contain regions with ≥90% sequence identity to the human exon 1A (Figure 5.4, 5.5). The calculated sequence identities for exon 1A differ slightly from Pinos et al 2009 [113]. This is probably because the bioinformatics server used for this study uses a different algorithm to determine sequence identity. Since the CACCC elements characterized in Chapter 3 are required for alternative SHBG promoter activity in testicular germ cells and are most likely binding elements for KLF4, their presence was also examined. While all species contained the proximal CACCC or functionally similar CGCCC element [183], [184], the distal element is restricted to higher primates. The most striking finding of this alignment was that the exon 1A regions in both New World Monkey SHBG genes are interrupted by a 312 or 314 bp Alu element.   The sequence alignments in Figures 5.3 and 5.4 are summarized in Figure 5.5. All simiiforme SHBG genes contain the FP4 element which is sufficient criteria to suspect that SHBG is repressed in their Sertoli cells. Since exon 1A and its associated promoter in all catarrhini SHBG loci closely resemble those in the human, I hypothesized that these species express alternative SHBG transcripts in their testes and contain exon 1A. In the New World Monkey (NWM) the situation is less obvious. It is not clear if the presence of the Alu element interferes with transcription from this region since the exon 1A transcriptional start site and splice donor GT remain intact. Like the mouse Shbg gene, the lemur and tarsier lack the FP4 element. Their upstream regions corresponding to the human exon 1A have only 64-68% sequence identity. Based on their similarities to the mouse Shbg gene, I predict that lemur and tarsier testicular SHBG transcripts only contain exon 1.    89 To validate my conclusions from these sequence comparisons, I obtained whole-testis cDNA from several higher primates: cDNA from hamadryas baboon (Papio hamadryas), rhesus (Macaca mulatta) and lion-tailed (Macaca silenus) macaques represented Old World Monkeys while cDNA from three common marmosets (Callithrix jacchus) represented New World Monkeys. I also obtained liver and testicular biopsies from two grey mouse lemurs (Microcebus murinus) to represent the lower primates. With these samples I completed RT-PCR analyses for their SHBG transcripts.   5.2.2 Old World Monkeys express alternative SHBG transcripts in their testes  Using primers with perfect complementarity to exons 1A, 1 and 8 on the baboon and macaque SHBG genes, I tested if OWM testes contain proximal or alternative SHBG transcripts. Figure 5.6 shows that both macaque and baboon testes are positive for SHBG transcripts containing exon 1A. The double banding pattern in the 1A-8 PCR assay resembles the doublet seen for SHBG transcripts in both human and 11K transgenic mice testes [66], [96] where the upper band is the full length transcript and the lower lacks exon 7 [58], [66], [96].  After 40 cycles of PCR both macaque testes were negative for proximal SHBG transcripts while the baboon testis produced a very low abundance of SHBG transcripts containing exon 1. Human hepatocellular carcinoma cell (HepG2) cDNA was used as a positive control for proximal (exons 1-8) SHBG transcripts. However, HepG2 cells have previously been shown to produce low levels of exon 1A-8 transcripts [114]. Isolation and sequencing of the upper band from the rhesus 1A-8 assay revealed that it contains exon 1A followed by exons 2-8.     90 5.2.3 No evidence for SHBG transcripts in the New World Monkey testis  Primers designed to marmoset SHBG exons 1A, 1 and 8 failed to detect alternative or proximal SHBG transcripts in the their testes. Since these PCR assays lacked an appropriate positive controls and do not rule out the possibility of a novel SHBG transcriptional start site used in their testis, I designed primers spanning the SHBG coding region (exons 3 and 8) with perfect complementarity to both the human and marmoset. Compared to the strong signal by HepG2 cDNA, none could be detected in any of the marmoset testes despite their abundant signals for both transition protein 1 and vimentin transcripts indicating that these cDNA were derived from testes containing germ and Sertoli cells, respectively (Figure 5.7).  5.2.4 High abundance of proximal SHBG transcripts in the lemur testis  Since the 5’ regulatory region of the lemur SHBG gene does not contain the FP4/USF repressor element, I expected that their testicular SHBG transcripts are produced from the same transcription unit used by the liver. Figure 5.8 shows that both lemur testes produce transcripts containing exon 1 as does the liver. Testing for alternative SHBG transcripts in the lemur testes was not performed because an appropriate positive control was lacking. A primer could not be designed with shared complementarity to the human exon 1A and the corresponding region of the lemur gene in order to use 11K-SHBG mouse testis cDNA as a control.  5.3 Discussion and future directions  Results from this study are summarized in Table 5.2.    91 Primate SHBG genes containing the FP4/USF element are limited to all simians within the haplorhini (Figure 5.2). The simiiforme infraorder is composed of the catarrhini (Hominoids and Old World Monkeys) and platyrrhini (New World Monkeys) parvorders but does not include tarsiiformes [185]. On this criterion alone I hypothesized that these species do not contain SHBG transcripts containing exon 1 in their testes. This hypothesis was confirmed in macaques and marmosets. The baboon testis exhibited a very low abundance of exon 1-containing SHBG transcripts. However, given the PCR product’s faint intensity compared to HepG2 cells after 40 amplification cycles, the presence of this transcript could represent transcriptional noise. Follow up analyses for exon 1-containing SHBG transcripts are therefore warranted in other baboon testes. The lemur SHBG locus does not contain the FP4/USF repressor element. Consistent with this, their testes are positive for transcripts arising from exon 1 indicating that the proximal transcription unit is a major source of SHBG in their testes. To confirm that this is the only SHBG transcription unit used in their testes, real-time PCR should be performed on these cDNA to quantify the levels of SHBG transcripts containing exons 1-2 versus another part of coding region (such as exons 2-3) with the expectation that they would be the same. Since the 5’ regulatory region of the lemur SHBG locus closely resembles that on the mouse gene and that their testes utilize the proximal transcription unit it is very likely these transcripts are a product of their Sertoli cells just as in rodents.  All catarrhine SHBG genes contain regions with near-perfect sequence similarity to the human exon 1A. Since these species also have two intact CACCC activator elements it is likely that they all produce alternative SHBG transcripts in their testes. All OWM samples tested contain an abundance of 1A-8 SHBG transcripts that are expected to be a product of their germ cells.    92 Isolation and sequencing of the lower transcript is underway and is expected to be a splice variant of the 1A-8 transcript that is also present in human and 11K-SHBG mouse testes [66], [96].  Failure to detect any SHBG transcripts in marmoset testes confirms that this gene is not expressed in this tissue from any transcription unit (Figure 5.7). Strikingly, this suggests that testicular SHBG is not required for marmoset fertility. Spermatogenesis in marmosets has many shared features with other higher primates, such as a multistaged seminiferous epithelium (Figure 1.2b from Chapter 1)[186]–[188]. In contrast, endocrine control of spermatogenesis as well as steroid hormone profiles in marmosets are much more distinct [188].  Moreover, they also differ in other aspects of reproductive endocrine control. For instance, gonadotropin regulation of marmoset testicular steroidogenesis is mediated through chorionic gonadotropin (CG) instead of luteinizing hormone (LH) as in most other mammals [189], [190]. CGβ but not LHβ is expressed in the marmoset anterior pituitary [191]. Exon 10 is spliced out of the marmoset luteinizing hormone receptor (LHR) transcripts [192], resulting in a lost capacity to be stimulated by LH yet preserving stimulation capacity by CG [193].   All primate species studied to date, including marmosets, contain SHBG in their bloodstream [194]. Marmoset SHBG is similar to human SHBG based on electrophoretic and relative steroid-binding properties [195]. Pugeat et al [196] determined that plasma SHBG in NWM has lower affinity for testosterone (T) by about one order of magnitude as well as 3-6 fold higher levels compared to catarrhines. This may, in part, account for their 2-4 fold higher concentrations of free testosterone compared to other simians [196]. The activity of 5α-reductase is drastically    93 reduced in squirrel monkey androgen-sensitive tissues [197]. This enzyme is required for the conversion of testosterone to 5α-dihydrotestosterone and reduced activity may help protect target-organs from excessive androgenization. NWM also exhibit an excess of total and free plasma glucocorticoid levels compared to catarrhines despite no physiological signs of glucocorticoid hormone excess [198]. This is most likely due to a decreased affinity of the glucocorticoid receptor for its ligands [198] causing decreased negative feedback and increasing outflow from the hypothalamic-pituitary-adrenal axis. Interestingly, cloning and characterization of squirrel monkey corticosteroid binding globulin (CBG) revealed that it has a much lower binding affinity for glucocorticoids compared to human CBG [199]. Increased circulating steroid hormone levels, decreased carrier protein affinity and either decreased nuclear receptor abundance or affinity to ligands are all features of the glucocorticoid, mineralocorticoid, sex steroid and vitamin D systems in NWM [200]. It is not known what evolutionary pressures or advantages favoured their “generalized steroid hormone resistance” but is an important consideration when selecting NWM for studies relating to endocrine physiology.  Estimates of the diversification times of the major primate groups allows for approximation of when macromutation events occurred in SHBG loci during primate evolution. Since the FP4/USF element is common to all simiiformes it must have arisen in their common ancestor. Studies estimate that the simian-tarsiiforme divergence occurred either before 61-70 million years ago (Ma) [185], [201], [202] or before 81-82 Ma [185], [203], [204]. Based on our knowledge that this element is sufficient to repress SHBG expression in Sertoli cells [98], its insertion into the genome of an early simian would have soon resulted in a phenotype of SHBG repression in the Sertoli cells of subsequent male offspring. The FP4 region is flanked by quadruplicate guanines    94 (G4), and interestingly, there is a single G4 in the corresponding regions of the lower primates and mouse (Figure 5.3). These repeat sequences allow me to hypothesize that the single G4 was an ancestral insertion site for the FP4/USF element within an early simian. A depiction of this event is suggested in Figure 5.9. This element is also remarkably small. Short interspersed nuclear elements (SINEs) are a type of small transposable elements on the order of 80-500 bp, whereas other mobile elements range of one to many kb in length [205], [206]. Furthermore the origin of this element is unclear. A BLAST alignment of the FP4/USF element against the human genome returned no other matches other than itself in the SHBG gene, suggesting that it had not been copied from another part of the genome. No hits were also returned when it was aligned against the Philippine tarsier genome, suggesting that it is not a “jumping” mobile element that ancestrally existed at another site. Insertion of the FP4/USF element into the SHBG locus may therefore have arisen from some type of de novo virus-based or other unknown mechanism.  No evidence for SHBG expression in marmoset testes is attributed to exon 1A disruption by an Alu. This element is a feature in both squirrel monkeys and marmosets, two species of the Cebidae clade, one of the three major branches within NWM. Most analyses relying on fossil and molecular data suggest cebidae divergence from the rest of the NWM occurred before 20 Ma indicating that this insertion is at least that old [185]. Loss of testicular SHBG expression can therefore only be concluded for squirrel monkeys and marmosets. Examination of SHBG loci for Alu insertion in exon 1A in the remaining NWM families (Atelidae and Pitheciidae) will allow me to determine if this is a feature of all platyrrhines. If they contain this Alu as well then loss of testicular SHBG expression can be estimated to be much older; before 35-45 Ma when    95 platyrrhines diverged from catarrhines [185]. Compared to other primates, marmosets have the highest incidence of Alu elements per megabase in their genomes (188 Alu counts/Mb), whereas humans and lemurs have 104 and 55 Alu counts/Mb respectively [207]. The widely accepted mechanism for Alu insertion first involves transcription from another genomic Alu; the resulting transcript seeks out a nicked poly-thymine which is used to prime genomic extension along the Alu transcript [208], [209]. The insertion site utilized within NWM exon 1A is the 5’-TTTTTAAA-3’ that immediately precedes the Alu (Fig 5.4, page 2 of 3). Figure 5.5 indicates the high (~90%) sequence identity of NWM exon 1A regions on either side of the Alu, therefore it is likely that before disruption this was a functional alternative first exon.  Another interesting observation from these data is that the presence of the FP4/USF element (and hence repression of Sertoli cell SHBG) is associated with primate species that exhibit a multi-staged arrangement of their seminiferous tubules [187]. A given cross-section through a simiiforme testis will reveal >1 spermatogenic stage per seminiferous tubule, whereas lemur or rodent testes will show only a single-staged arrangement (Fig 1.2b from Chapter 1) [102]. Tarsiiformes are the only infraorder in the haplorhines whose seminiferous tubule arrangement has not been characterized [102], [187] so it will be interesting to learn whether their tubules resemble those in lemurs or the rest of the haplorhines.  In this study, testicular transcript analyses were performed on heterogeneous cell populations containing both Sertoli and germ cells. In general, testis expression profiles of SHBG could be predicted based on the structure of that specie’s SHBG 5’ regulatory region. Based on our pre-existing knowledge of where SHBG and Shbg are produced in mammalian testes and the    96 genomic and transcriptional trends from this study, I therefore infer that (1) expression of SHBG in testicular germ cells from an alternative first exon is a feature of all catarrhini and (2) both lemurs and tarsiers express SHBG in their Sertoli cells from exon 1. This model is stylized in Figure 5.10.  This study does not inform the origin of exon 1A. This transcription unit probably arose from molecular evolution of exon 1A so that it gained a splice donor site and its flanking promoter. Selection pressures in favour of germ cell SHBG are difficult to hypothesize, however its correlation with multistaged seminiferous tubules in simiiformes indicates that there were significant genetic and morphological changes in testis biology occurring around the same time during primate evolution. Knowledge on the arrangement of tarsier seminiferous tubules is absent in the literature and will be needed to test the strength of this association.      97      Figure 5.1 Distinct testicular cell-type expression profiles of Shbg and SHBG genes across mammals. In rodents and all lower eutherians known Shbg is expressed in their Sertoli cells from exon 1 [83]. In contrast, in the human and 11K-SHBG transgenic mouse testis expression of SHBG occurs from an upstream alternative first exon in testicular germ cells [58], [59], [66], [98]. Expression of SHBG in Sertoli cells is constitutively repressed by USF1/2 from the FP4 element that is unique to humans and chimpanzees [98].      98  Table 5.1 Sources of genomic SHBG/Shbg nucleotide sequences from the public database.   Species  (Common name) NCBI Accession Code Homo sapiens  (Human) NG_011981.1 Pan troglodytes  (Chimpanzee) NW_003458253.1 Gorilla gorilla  (Gorilla) NW_004006353.1 Pongo abelii  (Sumatran Orangutan) NW_002888527.1 Nomascus leucogenys  (Gibbons) NC_019834.1,  NW_003501397.2 Papio anubis  (Olive Baboon) NC_018167.1, NW_003878344.1 Macaca mulatta  (Rhesus Macaque) NW_001102932.1 Callithrix jacchus  (Marmoset) ACFV01148737.1 Saimiri boliviensis  (Bolivian Squirrel Monkey) AGCE01058238 Tarsius syrichta (Philippine Tarsier) ABRT02392649.1 Microcebus murinus (Grey Mouse Lemur) ABDC01224946.1 Mus musculus  (Mouse) AL731687.13     99     Figure 5.2 Selected species and their positions within the primate order. Species whose SHBG loci are included in this study are shown within their respective taxononic groups. This cladogram is not an exhaustive map of the primate order, for example: (1) bonobos were not included in the hominid group and (2) the two NWM species listed represent only two of the five platyrrhini families.         100   human     -130 ATCCCCAGAGGGGTGATAGCTGAGTCTTGTGACTGGGCCCCTGGGCAGGGGTCAAGGGTC chimp          ATCCCCAGAGGGGTGATAGCTGAGTCTTGTGACTGGGCCCCTGGGCAGGGGTCAAGGGTC gorilla        ATCCCCAGAGGGGTGATAGCTGAGTCTTGTGACTGGGCCCCTGGGCAGGGGTCAAGGGTC orangutan      ATCCCCAGAGGGGTGGTAGCTGGGTCTTGTGACTGGGCCCCTGGGCAGGGGTCAAGGGTC gibbons        ATCCCCAGAGGGGTTATAGCTGAGTCTTGTGACTGGGCCCCTGGGCAGGGGTCAAGGGTC macaque        ATCCCCAGAGGAGTGGTAGCTGGGTCTTGTGACTGGGCCCCTGGGCAGGGGTCAAGGGTC baboon         ATCCCCAGAGGGGTGGTAGCTGGGTCTTGTGACTGGGCCCCTGGGCAGGGGTCAAGGGTC marmoset       ATCCCCAGAGTGTTGGCAGCTGTGTCTTGTGACTGGGCCCCCGGGCAGGGGTCAAGGGTC sq.monkey      ATCCCCAGAGGGTTGGCAGCTGTGTCTTGTGACTTGGTCCCTGGGCAGGGGTCAAGGGTC tarsier        GTCCACAGAGG-------------------------------------GGACACAGGGTC lemur          GTCCTCATTGG-------------------------------------GGCCCGCAGGGT mouse          GCCTTCAGAGG-------------------------------GGCCGCACGGTCAGGGTC                                                                             human          AGTGCCCCTGTTTCCTTTACCCCCTCCTCCCC-------GGGCAACCTTTAACCCTCCAC chimp          AGTGCCCCTGTTTCCTTTACCCCCTCCTCCCC-------GGGCAACCTTTAACCCTCCAC gorilla        AGTGCCCCTGTTTCCTTTACCCCCTCCTCCCT-------GGGCAACCTTTAACCCTCCAC orangutan      AGTGCCCCTGTTTCCTTTACCCCCTCCTCCCC-------GGGCAACCTTTAACCCTCCAC gibbons        AGTGCCCCG-TTTCCTTTACCCCCTCCTCCCC-------GGGCAACCTTTAACCCTCCAC macaque        AGTGCCCCTGTTTCCTTTACCCCCTCCCTCC--------GGGCAACCTTTAACCCTCCAC baboon         AGTGCCCCTGTTTCCTTTACCCCCTCCCCCC--------GGGCAACCTTTAACCCTCCAC marmoset       AGTGCCCCTGTTTCCTTTACCCTGCTCCCCCCTCCCCGGGGGCAACCTTTAACCCTCCAC sq.monkey      AGTGCCCCTGTTTCCTTTACCCTGCCCCCCC-TCCCCGGGGGCAACCTTTAACCCTCCGC tarsier        AGTGCCCT-----CATCTCCCGCCCCCCTTCTTCCCCCAGGGCAACCTTTAACCCTCCAC lemur          CAGTGCCC-----CATCTCCTGCCCCCCTTTTTCCCCCGGTGCAACCTTTAACCCTCCAC mouse          AGTGTCCC-----TATCTCCTGCCCCCCTTCTTCCCCCAGAGCAACCTTTAACCCTCCAC  human          CGCCCACACGCAAGGCTG------ +1 chimp          CGCCCACACGCAAGGCTG------ gorilla        CGCCCACATGCAAGGCTG------ orangutan      CGCCCACATGCAAGGCTG------ gibbons        CGCCCACACGCAAGGCTG------ macaque        CGCCTACACACAAGGCTG------ baboon         CGCCTACACACAAGGCTG------ marmoset       CGCCCACATGCAAGTCTG------ sq.monkey      CGCCCACATGCAAGTCTG------ tarsier        CACTTATGCGCAAGGCCACCTGCC lemur          CGCCATGCACAAGGCTGC------ mouse          CACCCATGTGCGCCAGGT------  Figure 5.3 The FP4/USF element is limited to simiiforme primates. A minimal region of the human SHBG proximal promoter, from -130/+1 with respect to the transcriptional start site used by the liver [60] is aligned with the corresponding regions from the chimpanzee, gorilla, orangutan, gibbons, macaque, baboon, marmoset, squirrel monkey, tarsier, lemur and mouse SHBG/Shbg genes (Table 5.1). The region contaning the entire FP4 element in the human sequence is shaded in grey. The central binding site for USF1/2 is bolded for emphasis [98] as are the quadruplicate guanines that form element’s boundaries.       101 human     -336 TGAAGAGCCTGAGAGA------------------GCGGGGTGGCGGGAGTCGGGGGGGAC chimp          TGAAGAGCCTGGGAGA------------------GCGGGGTGGCGGGAGTCGGGGGGGAC gorilla        TGAAGAGCCTGGGAGA------------------GCGGGGTGGCGGGAGTCG-GGGGGGC orangutan      TGAAGAGCCTGGGAGA------------------GCGGGGTGGCGGGAGTGG-CGGGGGC gibbons        TGAAGAGCCTGGGAGA------------------GCGGGGTGGCGGGAGTGG-CGGGGGC macaque        TGAAGAGCCTGGGAGA------------------GCGGGGTGGCGGGAGTGG-CCGGGAC baboon         TGAAGAGCCTGGGAGA------------------GCGGGGTGGCGGGAGTGG-CCGGGAC marmoset       CAAAGA----------------------------GCCTGAGAGAGCGGGGTGGCGAGGGC sq-monkey      CGAAGAGCCT------------------------GGGAAAGTGGGGTGGGTGGCGAGGGC tarsier        CGAAAAGCCCGGGA-------------------------------------AAGCGGGAA lemur          CGAAAAGCCTGGGA-------------------------------------GAGCGGGGC mouse          TGCAGAACCGCGGAGGTCTGGTCTCAAAGCTCAGGTCTCAAAGTGCTGGGAGCGAGGGGC  human          GGCGGGGTAGCCGCGGCCTGGTAAGTGGAGCTGGGATTCCGGCGCCGTACG-----GGAG chimp          GGCAGGGTAGCCGCGGCCTGGTAAGTGGAGCTGGGATTCCGGCGCCGTACG-----GGAG gorilla        GGCGGGGTAGCCGCGGCCTGGTAAGTGGAGCTGGGATTCCGGCGCCGTACG-----GGAG orangutan      GACGGGGTAGCCGCGGCCTGGTAAGTGGAGCTGGGATTCCGGCGCCGTGGG-----GGCG gibbons        GGCGGGGTAGCCGCGGCCTGGTAAGTGGAGCTGGGATTCCGGCGCCGTGAG-----GGCG macaque        GGCGGGGTAGCAGCAGCCTGCGAAGTGGGGCTGGGATTCCGGTGCCGTCAG-----GACG baboon         GGCGGGGTAGCAGCGGCCTGCGAAGTGGGGCTGGGATTCCGGTGCAGTCAG-----GGCG marmoset       TGCAGGGGCGCAGCGGCCTGGGAAATGGAGCTGGGATTCACGCGCGGTCGG---GGGGCG sq-monkey      GGCGGGGCCGCAGCGGTCTAGGAAATGGAGCTGTGATTCACTCGCGGTCGG-----GGCG tarsier        TGCCGGCGTGCCGCAGGCTGGGAAGCGGGACCCAGATCCCGGTACCCTCCA-----AGCA lemur          TGCGGGGGCG-----------------CCCAGGCTGGGCAGCGGCGCTGGC---TGGGCG mouse          AGAAGAAATGCTAGAAGCT------TCTAGACAAGATTCTGGCACCAACCGGTCTCAGCT  human          GAGAGAGTAGGCCAGCGAGGCGATCCTCTGTCCG-GGCATAGCCCCACCCCCTCGAATTC chimp          GAGAGAGTAGGCCAGCGAGGAGATCCTCTGTCCG-GGCAAAGCCCCACCCCCTCGAATTC gorilla        GAGAGAGTAGGCCAGCGAGGCGATCCTCTGTCCG-GGCAAAGCCCCACCCCCTCGAATTC orangutan      GAGACAGTAGGCCAGCGAGGCGGTCCTCTGTCCT-GGCAAAGCCCCACCCCCTCGAATTC gibbons        GAGAGAGTAGGCCAGCGAGGCGGTCCTCTGTCCG-GGCAAAGCCCCACCCTCTCGAATTC macaque        GAGAGAGTGGGCCAGCGAGGAGGTCCTCACTGCGGGGCAAAGCCCCACCCCCTCGAATTC baboon         GAGAGAGTGGGCCAGCGAGGAGGTCCTCACTGCGGGGCAAAGCCCCACCCCCTCGAATTC marmoset       GAGAGAGTGGGCCTGCGAGGCTGTCCTCACTGCCGGGCAAAGCCCCGCCCCCTCGAATTC sq-monkey      GAGAGAGTGGGCGGGCGAGGCTGTCCCCACTGCCGGGCAAAGCCCCGCCCCCTCGAAGTC tarsier        AAGAGGGCGGGTGGCTGAGGTGGTCCTCACTGCGGGAGTAAGCCCC-------TGAATTC lemur          GAGAAAACGGGCCGGCGAGGCGGTCCTCACCGCCGGGGTAAGCCCC--------TAAATT mouse          GGGAGGATTAGCCGGCCTGGTGGTCTCTTCTGCCAGAAAAGGCCT--------TAAATTC  human          TGTCGCAGCAGGGGGCACAACTGTCAGCCAATCAGCTTGGAGAACAGGCACGGCCGCGTC chimp          TGACGCAGCAGGGGGCACAACCGTCAGCCAATCAGCTTGGAGAACAGGCACGGCCGCGTC gorilla        TGTCGCAGCAGGGGGCACAACCGTCAGCCAATCAGCTTGGAGAACAGGCACGGCCGCGTC orangutan      TGTCGCAGCAGGGGGCACAACCGTCAGCCAATCAGCTTGGAGAACAGGCGCAGCCGCGTT gibbons        TGTCGCAGCAGGGGGCACAACCGTC-GCCAATCAGCTTGGAGAACAGGCACAGCCGCGTC macaque        TGTCGCAGCAGGGGGCACAACCGTCAGCCAATCAGCTTGGAGAACAGGCGCAGCCCCGCC baboon         TGTCGCAGCAGGGGGCACAACCGTCAGCCAATCAGCTTGGAGAACAGGCGCAGCCCCGCC marmoset       TGCCGCAGCAGGGGGCACACCCATCAGCCAATAAGCTCGGTTAACAGGCGCAGCCCCGCC sq-monkey      CGCCGCCGCAGGGCGCACACCCATCAGCCAATAAGCTTGGAGAACAGGCGCAGCCCCGCC tarsier        TGCCGC-TGCAGGAGCATACCCGCCAGCCAATCAGCTTGCAGACTAGGTGAATTTCTC-C lemur          CCGCCGCTGCGGGCGCACACCCGCTAGCCAATCAGCTTGCAGAACAGGCGCATCCGCGCC mouse          -TGAGCCCCCTCCGGGCCTCCCTGCAGCCAATCAGCTTGCTGCTTACGACCATTTCCCCA  human          CCCCCC---AAGCCCCACCCCCGACAGCTGGATCTTGTGACTGGGCTCCTGGG--TAGAG chimp          CCCCCCC--AAGCCCCACCCCCGACAGCTGGATCTTGTGACTGGGCTCCTGGG--TAGAG gorilla        CCCCC----AAGCCCCACCCCCGACAGCT-GATCTTGTGACTGGGCTCCTGGG--TAGAG orangutan      CCCC-----AAGCCCCACCCCCGACAGCTGGATCTTGTGACTGGGCCCCTGGG--TAGAG gibbons        CCCC-----AAGCCCCACCCCCGACAGCTGGATCTTGTGACTGGGCCCCTGGG--TAGAG macaque        CCCCCA----AGTCCCACCCCCGACAGCTGGATCTTGTGACTGGGCCCCTGGG--TAGAG baboon         CCCCCCA---AGTCCCACCCCCGACAGCTGGATCTTGTGACTGGGCCCCTGGG--TAGAG marmoset       TCCCG-------TCCCACCCCCGACAGCTGGGTCTTGTGACTG-GGCCCTGGG--TGGGG sq-monkey      TCCCC-----GTCCCACCCCCCGGCCGCTGGGTCTTGTGACTG-GGCCCTGGG--TAGGG tarsier        GCCTCCCTAG--CCCCGCCCCCGACAGCCCGGTCTTGTGACTG-GGCCCCGAG--CAGGG lemur          CC--------AGCCCCGCCCACAACGCCCCAGTCCTGTGACCAGGCCCCTGGG--CAGGG mouse          CCCCCACCCCCTGGTCCCGCCTCGCCTCGCTGTCCAGTGACTGGACTCTGCGCCGGACTA Figure 5.4 (page 1 of 3)    102 human          TTCAAGGTTGGAGTGAAGCGGCTTCCTTGCGGTTGTGTGGGTGTCCCAACCTGGGTCGAG chimp          TTCAAGGTTGGAGTGAAGCGGCTTCCTTGCGGTTGTGTGGGTGTCCCAACCTGGGTCGAG gorilla        TTCAAGGTTGGAGTGAAGCGGCTTCCTTGCGGTTGTGTGGGTGTCCCAACCTGGGTCGAG orangutan      TTCAAGGTTGGAGTGCGGCGGCTTCCTTGCGGTTGTGTGGGTGTCCCAACCTGGGTCGAG gibbons        TTCAAGGTTGGAGTGCGGCGGCTTCCTTGCGGTTGTGTGAGTGTCCCAACCTGGGTCGAG macaque        GTCAAGGTTGGAGTGTGGCGGCTTCCTTGAGGTTGTGTGGGTGTCCCAACCTCGATCGAG baboon         GTCAAGGTTGGAGTGCGGCGGCTTCCTTGCGGTTGTTTGGGTGTCCCAACCTCGATCGAG marmoset       GTCAAGGTTGGAGTGCGGCGGCTTCCTAGCGGTTGTATGGGTGTCCCAACATTAATCGAG sq-monkey      GTCAAGGTTGGAGTGCGGCGGCTTTCTAGCGGTTGTATGGGTGTCCCAGCCTTCATCGAG tarsier        TTCAAGGTTGGAGGGTGGAGTCTTCCTTGCAGTTGTTGGGGTGTCCCTGCCTCCAGCGGA lemur          GTCAAAGTTGGAGTGCGGAGGCTTCTTTGCTGTTGTTTGGTCGTCCCAGCCTCCATCGGC mouse          CTTAAGCTTCCTGTTTGTTTGTTTATTTTTTGTATTTTGTATTTTATTTATTTATTTTGG  human          ATACCCCGCGGTTCAAAGGCTCCCCCGCAGTGCTTTTTAAATTGACA------------- chimp          ATACCCCGCGGTTCAAAGGCTCCCCCGCAGTGCTTTTTAAATTGACA------------- gorilla        ATACCCCGCGGTTCAAAGGCTCCCCCGCAGTGCTTTTTAAATTGACA------------- orangutan      ATACCCCGCGGTTCAAAGGCTCCCCCGCAGTGCTTTTTAAATTGACA------------- gibbons        TTACCCCGCCGTTCAAAGGCTCCCCCGCAGTGCTTTTTAAATTGACA------------- macaque        TTACCCTGCGGTTCAAAGGCTCCCCGGCAGTGCTTTTTAAATTGACG------------- baboon         TTACCCCGTGGTTCAAAGGCTCCCCGGCAGTGCTTTTTAAATTGACG------------- marmoset       ACACCCCACAGTTCAAAGGTTCCCTCGCAGTGCTTTTTAAATTGACGAATGGAGGCCGGG sq-monkey      ACATCCCACAGTTCAAAGGTTCCCCCGCAGTGCTTTTTAAATTGACGAGCGGAGGCCGGG tarsier        AATACCCTACGGCCCAGAGACTGACCTGGCC-CTTCTTTAATTGAAG------------- lemur          TGGATAGCTTCCCC---------CACCTCTTGCTTCTTGAATTGAAG------------- mouse          TTTTT----CGAGACAGGGTTTCTCTGTGTAGCCTTGGCTATCCAGTCTGTGTCGTACAT  human          ------------------------------------------------------------ chimp          ------------------------------------------------------------ gorilla        ------------------------------------------------------------ orangutan      ------------------------------------------------------------ gibbons        ------------------------------------------------------------ macaque        ------------------------------------------------------------ baboon         ------------------------------------------------------------ marmoset       CGTGGTGGCTCACGCTTGTAATCCTAGCACTTTGGGAGGCCGAGGCGGGTGGATTACCTG sq-monkey      CGTGGTGGCTCACGCTTGTAATCCTAGCAATTTGGGAGGCCGGGACGGGTGGATTACTTG tarsier        ------------------------------------------------------------ lemur          ------------------------------------------------------------ mouse          ACCTCAAGCTTAAAAGTCCTCCCCTACCCTGCTTTAGTTAGAG-----------------                                                                                 human          ------------------------------------------------------------ chimp          ------------------------------------------------------------ gorilla        ------------------------------------------------------------ orangutan      ------------------------------------------------------------ gibbons        ------------------------------------------------------------ macaque        ------------------------------------------------------------ baboon         ------------------------------------------------------------ marmoset       AGCTCAGGAATTCGAGACCAGCCTGAGAACACGGTGAAACCCCGTCTCTACTAAAAAATA sq-monkey      AGCTCAGGAGTTTGAGACCAGCCTGAGAACACGGTGAAACCCCGTCTCTACTAAAAAATA tarsier        ------------------------------------------------------------ lemur          ------------------------------------------------------------ mouse          ------------------------------------------------------------                                                                                  human          ------------------------------------------------------------ chimp          ------------------------------------------------------------ gorilla        ------------------------------------------------------------ orangutan      ------------------------------------------------------------ gibbons        ------------------------------------------------------------ macaque        ------------------------------------------------------------ baboon         ------------------------------------------------------------ marmoset       CAAAAAATTAGCCGGGCATGGCGGCGTGTGCCTATAATCCGAGCTACTCGGGAGGCTGAG sq-monkey      CAAAAAATTAGCCGGGCATGGCGGCGTGTGCTTGTAATCCCAGCTACTCGAGAGGTTGAG tarsier        ------------------------------------------------------------ lemur          ------------------------------------------------------------ mouse          ------------------------------------------------------------ Figure 5.4 (page 2 of 3)    103 human          ------------------------------------------------------------ chimp          ------------------------------------------------------------ gorilla        ------------------------------------------------------------ orangutan      ------------------------------------------------------------ gibbons        ------------------------------------------------------------ macaque        ------------------------------------------------------------ baboon         ------------------------------------------------------------ marmoset       ACAGGAGAATCGCTTGAAACTGGGAGGTGGAGGTTGCAGCGATCCGAGATCGGGCCACTG sq-monkey      GCAGGAGAATCGCTTGAAACTGGGAGGCGGAGATTGCAGCGAGCCAGGATCGGGCCACTG tarsier        ------------------------------------------------------------ lemur          ------------------------------------------------------------ mouse          ------------------------------------------------------------                                                                                  human          ------------------------------------------------------------ chimp          ------------------------------------------------------------ gorilla        ------------------------------------------------------------ orangutan      ------------------------------------------------------------ gibbons        ------------------------------------------------------------ macaque        ------------------------------------------------------------ baboon         ------------------------------------------------------------ marmoset       CACTCCAGCCTGGGTGAC-AGAGCGAGACTCCATCTCAAAAAAATAAAAAATACA-TTGG sq-monkey      CACTCCAGCCTGGGTGGACAGAGCGAGACTCCATCCCAAAAAAATAAAAAATACATTGAC tarsier        ------------------------------------------------------------ lemur          ------------------------------------------------------------ mouse          ------------------------------------------------------------                                                                     +78         human          TATGCAGTGATAACCTGCTTTAGCCTCAGGCTCACTCACCCGCCCAGACCCTGGGTAAGC chimp          TATGCAGCGATAACCTGCTTTAGCCTCAGGCTCACTCACCCGCCCAGACCCTGGGTAAGC gorilla        TATGCAGTGATAACCTGCTTTAGCCTCAGGCTCACTCACCCGCCCAGACCCTGGGTAAGC orangutan      TATGCAGTGATAACCTGCTTTAGCCTCAGGCTCACTCACCCGCCCCGACCCTGGGTAAGC gibbons        TATGCAGTGATAACCTGCTTTAGCCTCAGGCTCACTCACCGGTCCAGACCCTGGGTAAGC macaque        TATGTGGTGATAACCTGCTTTAGCCTCAGGCTCACTCACCCCCCCAGACCCTGGGTAAGC baboon         TATGCGGTGATAACCTGCTTTAGCCTCAGGCTCACTCAGCCCCCCAGACCCTGGGTAAGC marmoset       CTATACAGTGATTATCTGCTTTAGCCTCAGGCCCCCTCACCTGCCCAGACCCTGGGTAAG sq-monkey      GTATACAGTGATTATCTGCTTTAGCCTCGGGCCCACTCACCCGCCCAGATCCTGGGTAAG tarsier        TATGCAGAGATTATCTCCTTTACCCTCGCTCTCGCTCGTCCACCCAGGCACTGGGTGAGC lemur          ------------TGTGCACTTAGCCTCACTCTCACTCA----CCCAGACCCAGGGGCCCC mouse          ----------TACACTCCAGTTAGCTGCTTTAGTTTTCA---CTCTTAGCCGTCTAATCA  Figure 5.4 Comparison of the SHBG exon 1A and flanking promoter sequences amoung primates. The human alternative first exon and 336 bp of its 5’ flanking promoter used in testicular germ cells is aligned with the corresponding regions of the chimpanzee, gorilla, orangutan, gibbons, macaque, baboon, marmoset, squirrel monkey, tarsier, lemur and mouse SHBG/Shbg genes (Table 5.1). The marmoset and squirrel monkey exon 1A regions are each interrupted by an Alu element (312 and 314 bp, respectively). Lightly shaded bases indicate the human exon 1A and possible alternative first exons in other species. These regions were used to calculate the percent sequence identities listed in Figure 5.3 and Table 5.2. Intensely shaded bases indicate the CACCC regulatory elements described in Chapter 3. Bolded bases indicate the partial CRE described in Chapter 4. The underlined guanine indicates the +78 position and 3’ boundary of the human alternative exon 1A.    Figure 5.4 (page 3 of 3)    104   Figure 5.5 Summary of phylogenetic alignments of SHBG 5’ regulatory regions. All simiiformes (catarrhini and platyrrhini) contain the FP4/USF element which represses SHBG expression in Sertoli cells [98]. All catarrhini contain a region with very high (>90%) sequence identity to the human exon 1A. Platyrrhini (NWM) also have a region that resembles exon 1A but it is interrupted by a 312/314 bp Alu element. The two C(A/G)CCC elements required for basal transcriptional activity from the alternative SHBG promoter (identified in Chapter 3) are conserved in all simiiformes.     105       Figure 5.6 Alternative SHBG transcripts are present in Old World Monkey testes. RT-PCR assays for SHBG transcripts (exons 1A-8 and 1-8) were performed on cDNA derived from HepG2 cells, rhesus macaque, lion-tailed macaque and baboon testes. Analyses for TNP1 and Vimentin transcripts indicate the presence of germ and Sertoli cells, respectively, in testis-derived cDNA. The integrity of all cDNA samples is assessed by analysis for GAPDH transcripts. All PCR products were subjected to electrophoresis on either 1% (for SHBG) or 1.5% (for TNP1, Vimentin and GAPDH) agarose gels. Molecular sizes in bp are indicated for the SHBG assays.      106         Figure 5.7 No evidence for SHBG transcripts in marmoset testes. RT-PCR assays for SHBG transcripts (exons 3-8) were performed on cDNA derived from HepG2 cells, and three marmoset testes. Analyses for TNP1 and Vimentin transcripts indicate the presence of germ and Sertoli cells, respectively, in testis-derived cDNA. The integrity of all cDNA samples is assessed by analysis for GAPDH transcripts. All PCR products were subjected to electrophoresis on either 1% (for SHBG) or 1.5% (for TNP1, Vimentin and GAPDH) agarose gels. Molecular sizes in bp are indicated for the SHBG assay.      107        Figure 5.8 Lemur testes express SHBG transcripts from the proximal transcription unit. RT-PCR assays for SHBG transcripts (exons 3-8) were performed on cDNA derived from liver and testes from two grey mouse lemurs (Microcebus murinus). Analyses for TNP1 and Vimentin transcripts indicate the presence of germ and Sertoli cells, respectively, in testis-derived cDNA. The integrity of all cDNA samples is assessed by analysis for GAPDH transcripts. All PCR products were subjected to electrophoresis on either 1% (for SHBG) or 1.5% (for TNP1, Vimentin and GAPDH) agarose gels. Molecular sizes in bp are indicated for the SHBG assay. RT, reverse transcriptase.                 108 Table 5.2 Summary of SHBG genomic and testicular transcript features in relation to seminiferous tubule organization and testicular SHBG location in primates.  Taxonomic group  SHBG gene features     Testicular SHBG transcripts  Known or inferred location of testicular SHBG  Seminiferous tubule organization [i]  FP4/USF element Sequence identity to human exon 1A (%) Two intact C(A/G)CCC elements in alternative promoter Other  Contains exon 1A Contains exon 1   Human  Yes  Yes   Yes [ii] No [ii]  Germ cell [ii]  MS Other Hominoids  Yes 97 to 99 Yes   nt nt  Germ cell  MS Old World Monkey  Yes ~94 Yes   Yes No**  Germ cell  IM/MS New World Monkey  Yes ~90* Yes Exon 1A interrupted by Alu  No No  None?  MS Tarsier  No 68.8 No   nt nt  Sertoli cell  Unknown Lemur  No 63.5 No   No Yes  Sertoli cell  SS Mouse  No 63.1 No   No Yes  Sertoli cell [iii]  SS   [i] Wistuba et al [187], Luetjens et al [102]. [ii] Selva et al [66], [96]. [iii] French and Ritzen [81], Munnel et al [210]. * Sequence identity not including Alu element. ** Baboon testis contains SHBG transcripts with exon 1 at very low levels. nt: not tested, MS: multistaged, IM: intermediate staged, SS: single staged. Italicised entries indicate inferred or hypothesized location of testicular SHBG based on RT-PCR analyses and genomic trends described in this chapter.    109      Figure 5.9 Hypothetical mechanism for the FP4 insertion event. The ancestral SHBG promoter did not contain the FP4/USF element (pre-insertion). The quadruplicate guanines were probably the site of a recombination-mediated insertion event. The resultant region contains the FP4/USF element flanked by quadruplicate guanines at either end (post-insertion). Since the presence of this element is common to all simiiformes, FP4/USF insertion may have occurred soon after simians diverged from the rest of the haplorhines (See Figures 5.2, 5.3). Most estimates suggest this occurred either 61-70 Ma or 81-82 Ma [185]. The source and origin of this 38 element is unknown.                110         Figure 5.10 Trends of testicular SHBG gene expression across primates. Sertoli cells are the ancestral location of testicular SHBG. Alternative SHBG transcripts containing exon 1A have only been detected in hominoids and OWM (catarrhini) and are therefore the only groups where germ cell SHBG is expected to be synthesized. NWM do not express SHBG in their testes but probably did before Alu disruption on exon 1A. Each change in how testicular SHBG is expressed is always preceded by a structural change in the SHBG 5’ regulatory region of that group. Numbers indicate approximate period of a macromutation event in the SHBG gene: 1, insertion of the FP4 element. 2, insertion of an Alu element in the NWM alternative first exon. OWM, Old World Monkey, NWM, New World Monkey.       111 Chapter 6: Conclusions, future directions and potential roles for testicular SHBG in reproductive physiology  6.1 Top-down and bottom-up approaches reveal multiple modes of transcriptional regulation from the human SHBG alternative promoter  The use of a transient reporter construct allowed me to take two different approaches to investigate transcriptional regulation. In Chapter 3, sequence analysis and dissection of the human alternative SHBG promoter sequence led to the identification of two CACCC elements that are required for its basal transcriptional activity. Transactivation from either or both of these motifs is likely mediated through KLF4. These data agree well with observations in vivo that indicate that Klf4 and SHBG are co-expressed in round spermatids during the spermatogenic cycle.  In Chapter 4 I found that increased levels of cAMP and PKA activity cause induction of altP-SHBG reporter activity, yet these effects are not mediated through a partial CRE. CREM isoforms that are co-expressed with SHBG in vivo also upregulate altP-SHBG reporter activity and CREMτ may synergize with cAMP to maximize its response. These data reinforce the importance of performing functional studies, such as reporter assays, to confirm promoter sequence analyses. In their recent paper, Ma et al [211] stated that CREB/CREM control SHBG expression in testicular germ cells after citing an earlier report that only implicated these factors due to the identification of a partial CRE in the alternative SHBG promoter [67].    112 Taking the results of Chapters 3 and 4 together, my findings indeed support the hypothesis that expression of SHBG from an alternative promoter in testicular germ cells is influenced by stage-specific factors and/or events during spermatogenesis.    When designing a series of in vitro regulation studies as I have for Chapters 3 and 4, it is crucial to use a cell line that recapitulates the environment and physiology of the endogenous tissue in order to obtain relevant results. Even though they were first characterized as immortalized premeiotic spermatocytes [104], GC-2 cells clearly have characteristics of round spermatids based on their profile of Klf4 expression (Chapter 3). On the other hand, round spermatids at similar stages also produce high levels of CREM. While GC-2 cells express a wide range of Crem isoform transcripts [178] basal levels of CREMτ protein were detected at very low abundance compared to that produced by overexpression (Chapter 4).   Luciferase reporter constructs are a convenient and versatile tool for modelling transcriptional activity in vitro. A variety of experimental manipulations can be performed on a relatively short timescale since co-treatments are easily incorporated into transient reporter experiments. Small-interfering RNAs and various pharmacological agents used in Chapters 3 and 4 are examples of these that all fit into the same three-day seed-transfect-assay protocol described in Chapter 2. Reporter constructs are also amenable to site-directed mutagenesis to test the impact of candidate regulatory elements and polymorphic regions [105], [112] on transcriptional activity. On the other hand, there are limitations to these types of experiments. Reporter constructs containing promoter regions are limited by their size and subsequent regulation studies may exclude the effects of long-range cis-regulatory elements such as enhancers or insulators. When selecting a    113 regulatory element to insert into a reporter construct it is common practice to clone various lengths of that region and then assay them in parallel to pinpoint key promoter regions that confer activation or repression. This methodology has been widely used to uncover regulatory motifs in the proximal SHBG promoter [60], [112] and to a lesser degree in the alternative promoter [67].   Furthermore, the use of transient reporter constructs does not take into account the effects that higher order chromatin structures have on endogenous gene regulation [212].  Since reporter constructs are naked DNA that do not adopt chromatin structure, they are constitutively derepressed and may not be the best model for physiologically relevant gene regulation. With these considerations in mind, my initial strategy for modelling the expression of alternative SHBG transcripts in vitro involved incorporating the full 11 kb SHBG genomic locus into an episomal vector (pEBNA-DEST). Once incorporated into host mammalian cells, this vector adopts a chromatin structure and expresses the Epstein-Barr nuclear antigen-1 (EBNA-1) viral protein which facilitates its extrachromosomal persistence as well as replication from its Epstein-Barr virus origin of replication (oriP) [213]–[217]. The goal of this system was to establish a line of GC-2 cell clones containing episomes that stably express alternative SHBG transcripts and to use this platform to study transcriptional regulation using all of the manipulations described in Chapters 3 and 4. I experienced significant challenges identifying cell clones that positively expressed SHBG and in designing a convenient way to screen large numbers of clones in parallel. I was unable to detect SHBG transcripts by RT-PCR in populations of uncloned antibiotic-selected cells. These roadblocks led to the subsequent design and construction of the altP-SHBG-pGL3 reporter construct used in Chapters 3 and 4.     114 Future attempts to engineer an episomal system to model the alternative SHBG transcription unit might benefit from modifications to my initial strategy. Detection of SHBG-positive GC-2 clones following resistance selection was a major challenge given that my desired cell clones were expected to produce an intracellular isoform of SHBG. Other cell lines engineered to stably express the proximal SHBG transcription unit benefitted from the fact that positive cell clones secreted SHBG [80] are easily detected by an ultrasensitive immunofluorometric assay to human SHBG [218]. One improvement to my pEBNA-DEST-SHBG could be to incorporate an internal ribosomal entry site coupled to a green fluorescent protein cDNA (IRES-GFP) into the same reading frame as the SHBG transgene. Following transfection and selection for antibiotic-resistant cells, the entire pool could be subjected to fluorescence activated cell sorting (FACS) to isolate cells expressing GFP. GFP positive cells would be expected to also express the SHBG transgene. Conveniently, expression of GFP and the construct’s selection marker would be useful factors to normalize SHBG expression against, thereby controlling for episome copy number across separate clones.  6.2 Distinct patterns of testicular SHBG expression across the major primate groups  In Chapter 5, I provided evidence that SHBG transcripts in selected Old World Monkey (OWM) testes comprise the same alternative first exon used in human testes [58], [59]. This was anticipated since the structure of their SHBG 5’ regulatory regions resemble that of the human gene. These transcripts encode a truncated form of SHBG that is produced in testicular germ cells of humans and 11K-SHBG transgenic mice [66], [96]. Therefore, transcripts arising from exon 1A in OWM testes are expected to be of germ cell origin.     115 The 5’ regulatory region of the lemur SHBG gene bears much closer resemblance to the rodent gene as opposed to the human. Therefore not surprisingly, an abundance of SHBG transcripts originating from exon 1 are present in their testes and resemble those found in the liver. Sertoli cells are the location of testicular SHBG in most mammals [83] and are also the expected source in lemurs.  Unexpectedly, I was unable to demonstrate any evidence for SHBG expression in marmoset testes. The exon 1A region of the marmoset SHBG locus is disrupted by an Alu element. Insertion of this element was probably the driving factor that abolished transcription from exon 1A. The consequences of lost SHBG expression in their testes are not clear but suggest that there are fundamental differences in reproductive physiology between New World Monkeys (NWM) and other primates. Given that the distribution and bioavailability of steroid hormones in NWM is quite distinct to other primates (discussed in section 5.3), they may have no need for SHBG in their testes.  The intrinsic regulatory mechanisms within developing sperm are generally conserved throughout eutheria. This is quite clear given our ability to model the testicular expression pattern of human SHBG in mice by transgenesis in a way that reflects the endogenous expression pattern in humans [66], [78], [96]. Therefore it is not surprising that the expression pattern of numerous testis genes are mirrored in both mice and humans. For example, KLF4 and CREM are both found in round spermatids of both species [121], [122], [129], [219] even though there is 65-110 Ma of evolutionary divergence between them [220]. The discovery that testicular expression of human SHBG differs from rodent Shbg [66], [98], and my follow up findings that    116 suggest germ cell SHBG expression is common to all catarrhini informs us that testis gene expression patterns have not remained static throughout evolution. This should inspire us to ask if any other testis genes have undergone abrupt changes in expression patterns over a similar timeframe. Moreover, are there any genes that have undergone the same switch from Sertoli to germ cell or lost expression entirely like SHBG has?  Regulation of spermatogenesis by gonadotropins is not uniform across eutheria. It is generally accepted that spermatogenesis in rodents is always operating at full capacity since FSH-stimulation of germ cell numbers is not modulated by inhibin [99]. Sexual development is normal and fertility is preserved in male mice lacking Fshb indicating that FSH is not a prime modulator of spermatogenic efficiency [221]–[224]. In contrast, studies using macaques indicate that spermatogenesis is not normally maintained at maximum. In these animals FSH levels are tightly controlled by inhibin in response to the number of testicular stem cells that are supported by Sertoli cells [225]–[227]. FSH stimulation augments spermatogonial proliferation and spermatogenic output in intact and FSH-depleted monkeys [228], [229]. The absolute dependence of human spermatogenesis on FSH became obvious as men with inactivating mutations in their FSHB gene presented with azoospermia [230]–[232]. These studies inform us that among OWM and humans, similarities in endocrine regulation of spermatogenesis correlate with shared patterns of testicular SHBG expression. With these in mind, macaques are probably good models for hormonal regulation of male fertility that are applicable to humans.   Despite minimal data on endocrine regulation of spermatogenesis in marmosets they have become widely used as a preclinical model for reproductive physiology and toxicology [233].    117 Failure to detect SHBG transcripts in marmoset testes informs us that they may not recapitulate all of the features of human reproductive physiology, particularly in the context of steroid hormone bioavailability in the reproductive tract since this is the suspected role of testicular SHBG (discussed in section 6.3).   My results from Chapter 5 therefore reinforce two key research agendas.  The first is to further investigations on primate reproductive physiology. Careful attention must be paid to how endocrine regulation of fertility varies across all major primate groups, especially those primates that are more phylogenetically distant to humans. The development and testing of novel strategies for fertility treatments and contraceptives rely on using biological models that provide relevant results. Our understanding of a “relevant” model comes only after its thorough characterization.   The second is to identify the role(s) of testicular SHBG in reproductive physiology. Since the testis is only the minor source of SHBG production in most mammals, functional studies have been limited to plasma SHBG. SHBG is produced in different testicular cell types across mammals. While this may indicate diverse functions, these differences may instead reflect differences in their endocrine status and regulation described above.        118 6.3 Potential roles for testicular SHBG in reproductive physiology  Under the free hormone hypothesis, high affinity steroid-binding proteins in the blood plasma are the primary limiting factors of steroid hormone movement into target tissues [54]–[56]. While this is the prevailing model for SHBG in the serum transport of sex steroids, emerging roles indicate that in other contexts SHBG can 1) initiate second messenger signalling within target cells via interaction with membrane receptors [234]–[236], 2) bind to and be taken up into cells in culture by receptor-mediated endocytosis [237]–[239], 3) be sequestered from the vasculature into the stroma of steroid-sensitive tissues [240] and 4) maintain a local reservoir of androgens or estrogens in cells where it is synthesized and/or retained [80].   Despite an extensive body of literature characterizing SHBG (also known as ABP) in the male rodent reproductive tract [83], [210] studies directly interrogating the physiological role of testicular SHBG in men are absent in the literature and therefore its role in human reproduction remains unknown. Men with homozygous SHBG mutations that lack steroid binding have never been reported and any potential phenotype on reproductive physiology would likely be complicated by dysregulated hormone transport by SHBG in the bloodstream.   6.3.1 Sertoli cell SHBG  In rodents and all sub-primate eutheria known, SHBG is produced and secreted by Sertoli cells into the seminiferous tubular lumen (Figure 6.1). SHBG travels with spermatozoa through the converging tubules and efferent ducts to the epididymis. Virtually all of the luminal SHBG is taken up into the epithelium of the proximal caput epididymis [83]. Given the very high    119 concentrations of testosterone in the luminal space [94], SHBG is almost certainly bound to testosterone and its movement into the epididymal epithelium is generally accepted to be a mode of androgen delivery to these cells [210]. In support of this model incubation of segments from the rat caput epididymis with increasing concentrations of SHBG enhances their uptake of testosterone [241].   Sperm mature as they move through the epididymis. Direct contact with the caput epithelium is required for remodelling of the sperm head plasma membrane which is needed for downstream sperm-oocyte interactions [5]. The development and function of the caput epididymis is exquisitely sensitive to androgen signalling. The initial segment of the epididymis fails to develop in transgenic mice where the androgen receptor gene (Ar) is specifically ablated from their caput epithelium (CEARKO) [242], [243]. Sperm transport into the epididymides of CEARKO mice is obstructed, resulting in azoospermic infertility. Luminal fluid from the testes is also obstructed in these animals resulting in significant back-pressure causing testicular hypertrophy, seminiferous tubule distension and sloughing of the seminiferous epithelium [242], [243]. Given the absolute dependence of androgen signalling in the formation and function of the caput epididymis, it is tempting to speculate that in rodents, SHBG-mediated androgen transport into the caput epithelium is required to maintain sufficient local androgen levels for optimal AR-dependent signalling.   A similar phenotype is observed in adult rats that have undergone ligation of the efferent ductules, which cuts off the flow of sperm and testicular fluid into the epididymis [244]. Within two weeks following ligation there is regression of the epididymal initial segment and a    120 decreased height of its epithelium, consistent with the CEARKO phenotype [242], [243]. Surprisingly, high systemic doses of exogenous testosterone fail to restore normal epithelial morphology following ligation suggesting that luminal androgen uptake via SHBG is required to maintain high levels of AR-mediated signalling within the caput epithelium. In support of this model, stably transfected renal epithelial cells that retain SHBG in their cytoplasm dramatically enhance AR-dependent gene expression by maintaining a local reservoir of steroid hormone due to the intracellular accumulation of SHBG [80]. Secreted growth factors from the testis that travel through the lumen to stimulate and maintain the caput epithelium integrity have also been suggested [245]. Efferent ductule ligation would also block their transit so they are not ruled out.  6.3.2 Germ cell SHBG  In humans, a SHBG isoform is found in testicular germ cells (Figure 6.1) [66], [96]. Intracellular retention of germ cell SHBG suggests that it may mediate local steroid hormone actions at the level of individual germ cells [66], [96]. This isoform is produced in round spermatids where the androgen receptor is not expressed [36]–[39] and therefore rules out a role for SHBG in enhancing androgenic activity in these cells. There are numerous reports of estrogen receptor and CYP19 (aromatase) expression in most testicular cell types, including germ cells [39], [246]. Global knockout of ESR1 (ERα) results in infertility in mice [247] however germ cell transplantation studies indicate that ERα in only somatic cells is required for mouse spermatogenesis [248], [249]. It is not known if humans share the same cell-type dependence of ER-signalling for spermatogenesis as mice or if ERα in human testicular germ cells is required for their development. Interestingly, SHBGs in humans and other primates, in which SHBG expression may occur in germ cells rather than Sertoli cells, have a much higher relative binding    121 affinity for estrogens than SHBG in sub-primate species (Table 6.1). Despite the very high concentrations of intratesticular testosterone vs estradiol (>40-fold greater) [250]–[252] the presence of SHBG in human testicular germ cells may therefore serve to influence their local estrogen levels.   After they are deposited into the female reproductive tract, spermatozoa undergo a series of physiological changes that result in hyperactive motility and render the sperm head permissible to acrosome reactions [253], [254]. These changes are called “capacitation” and are associated with global increases in tyrosine phosphorylation within spermatozoa [255]. It has recently been shown that estrogens stimulate capacitation and tyrosine phosphorylation but inhibit the incidence of acrosome reactions in mammalian spermatozoa in vitro [256], [257]. It was therefore suggested that the high estrogen levels in the female reproductive tract stimulate sperm capacitation through non-genomic mechanisms and prevent premature acrosome reactions until they reach the oocyte [256], [257]. Estrogen receptors alpha and beta have been shown to localize to human and mouse sperm heads, respectively [256], [258]. This is very close to where SHBG is detected in the sperm heads of humans and 11K-SHBG transgenic mice [66], [96]. Intracellular SHBG enhances steroid hormone actions by sequestering extracellular androgens and estrogens, providing a local reservoir of steroids that can constitutively stimulate steroid receptors [80]. Since SHBG is retained in the sperm head of ejaculated sperm, upon entry into the female reproductive tract it would most likely be occupied by estrogens and possibly serve to maintain a local pool of estrogen within the sperm head resulting in sustained local non-genomic ER-signalling. Under this hypothesis, those sperm with higher SHBG levels in their heads would more effective at maintaining ER-signalling resulting in prolonged capacitation and protection    122 from premature acrosome reaction. In support of this hypothesis, the abundance of intracellular SHBG in human sperm samples correlates positively with their proportion of grade A motile sperm. Furthermore, these levels show progressive decline with advancing age [96].  It is very possible that the steroid-occupancy of germ cell SHBG changes during its journey from initial synthesis in the seminiferous epithelium, transit through the male reproductive tract and finally passage through the female reproductive tract. The major factors determining its occupancy are the relative abundances of 1) sex steroids and 2) free zinc ions (Zn2+) within the sperm’s local environment. While androgens are the preferred ligands for human SHBG, it’s relative binding affinity for estrogen is markedly higher compared to sub-primate eutheria (Table 6.1). Intriguingly, when human SHBG is saturated by zinc ions, its relative binding affinity for estrogen decreases by about five to six-fold yet its affinity for androgens is preserved [259]. All mammals tested to date, including humans, have detectable amounts of aromatase in their germ cells [246]. Locally synthesized estrogens could therefore become bound to germ cell SHBG and remain occupied as the cell transits into and through the epididymis. Throughout this period acrosomal SHBG may even sequester estrogen away from ERs in the sperm head to prevent premature capacitation reactions. Total zinc levels in human testicular and epidymal homogenates are quite similar [260], however the abundance of free zinc can be modulated by relative distribution of metallothionein in the testis and reproductive tract [261], [262] which can sequester free zinc ions. During the pre-ejaculatory emission phase spermatozoa become surrounded by seminal fluids derived from the prostate and seminal vesicles. Seminal fluid is rich in zinc levels [263] which may bind to germ cell SHBG and result in loss of estrogen occupancy. In the female reproductive tract the occupancy of germ cell SHBG may change again    123 depending on where in the woman is in her ovulatory cycle. Gas chromatography-mass spectrometry measurements of sex steroids in human cervical mucus have shown that estrogen levels in the reproductive tract vary greatly depending on where she is in her ovulatory cycle [264]. Estradiol levels in cervical mucus peak (399 pg/mL) one day after ovulation before falling to preovulatory levels (160-170 pg/mL) within three days, despite testosterone levels remaining constant (30-40 pg/mL) over the same period [264]. Therefore the steroid-occupancy of germ cell SHBG and the extent that it may influence ER-signalling within the sperm head is probably a direct result of the steroids and ions in its environment throughout the male and female reproductive tracts.  6.3.3 Species-specific locations of SHBG in the male reproductive tract. Possible conservation of roles across mammals?  While Shbg and SHBG are expressed in distinct cell types in the rodent and human testes (Figure 6.1), several studies have indicated other locations of SHBG in the mouse and human male reproductive tracts that may make up for the differences in species-specific expression patterns, suggesting conserved roles across mammals. Rodent SHBG in germ cells  Using immuno-electron microscopy and co-culture experiments Gerard and colleagues have demonstrated that in rats SHBG is transferred from Sertoli cells to germ cells both in vitro and in vivo [265], [266]. In their latter studies SHBG was found to associate with nuclei in spermatocytes and round spermatids. In elongated spermatids there was intense cytoplasmic labelling in the sperm head that was associated with vesicular bodies. The apparent loading of    124 Sertoli-derived SHBG into the sperm head suggests that it may perform a related function to the acrosomal human SHBG isoform. Human SHBG in the epididymal stroma  Since human testicular SHBG is retained within the acrosome of mature sperm heads, the conclusions from rodent studies that SHBG secreted from Sertoli cells is required to maintain the androgen microenvironment in the caput epididymis do not apply to humans. However, studies using mice that express human SHBG transgenes have shown that SHBG in the vasculature can be sequestered into the uterine stroma by the fibulin family of extracellular matrix proteins, and SHBG is also present in epididymal stroma [240]. Uptake of SHBG into these tissues suggested that this could be a mode of steroid hormone delivery from the vasculature. Androgen and estrogen receptor signalling within the epididymis are essential for mouse fertility [242], [243], [267] and AR and ER both localize to the epididymal principal cells and stroma in rodents [39], [268]. High levels of zinc [260] as well as localized regions of metallothionein in the basal epithelial cells [261] are found throughout the epididymis. Human SHBG in this tissue may therefore serve dual roles influencing local androgens and estrogens. While human SHBG is found within the epididymal stroma of transgenic mice, localization of SHBG in human epididymides is lacking. Fibulin-2 has been found to be expressed in the human cauda epididymis [269].         125 6.3.4 Research strategies: SHBG in the reproductive tract Interrogating Sertoli cell SHBG  Since adult mice and rats do not have circulating SHBG and its production is confined to their Sertoli cells, a drug that specifically blocks or displaces steroids from SHBG could be conveniently tested in these species to evaluate the specific effects of SHBG deficiency in the testis and reproductive tract. Novel ligands that displace steroids from SHBG are an area of active investigation [270]. Chronic administration of such a drug in male rodents for at least 35 days (one full round of spermatogenesis) would allow evaluation of SHBG function on both spermatogenesis and in the downstream reproductive tract.   A Shbg knockout rodent has never been described in the literature. Global gene knockouts often result in phenotypes in all tissues where the ablated gene is normally expressed [122], [126], [127]. With this in mind a global Shbg knockout mouse may result in phenotypes beyond testicular function. Fetal and neonatal mice express hepatic Shbg during a window of development where gonadal differentiation occurs [78], [271], [272] and the absence of SHBG during this period may defeat later examination of adult gonadal structure and function. To isolate the effects of Sertoli cell SHBG by gene ablation, a Sertoli cell-specific knockout is preferable.   Cell type-specific knockout animals have been used extensively to dissect the roles of AR- signalling in the testes [32]–[34]. To date, the murine Ar has been separately ablated from all of the major cells of the testis. The dependence of AR-signalling in androgen-sensitive cells in the    126 downstream male reproductive tract have also been examined [242], [243]. A natural continuation of this theme is to use this approach to test potential regulators of the sex steroid microenvironment in these tissues (such as SHBG). Cell-specific knockouts made using Cre-lox technology involve crossing two differently engineered mouse strains. One strain must have some part of the gene of interest (usually a key exon) flanked by loxP recombination sites. The other mouse must harbour the Cre-recombinase enzyme under the control of a tissue specific promoter. Sertoli cell specific knockouts have been generated with the use of Cre under the control of the promoter flanking the anti-müllerian hormone gene [43].  Either method of interrogating Sertoli cell SHBG would benefit from incorporating the ARE-luc reporter mouse into their experimental approach [273], [274]. These mice carry a firefly luciferase transgene under the control of an androgen response element (ARE). The reporter gene is expressed in proportion to the level of AR-mediated activity in that tissue and can be measured in a typical luciferase assay using tissue homogenates or measured directly using a camera that detects bioluminescence in vivo. The impact of androgen-dependent signalling could therefore be rapidly assessed in those tissues where SHBG is synthesized or sequestered. This method would be particularly useful in assessing AR-signalling in those tissues where androgen-responsive genes have not yet been identified. Interrogating germ cell SHBG  It is expected that SHBG localized to the sperm head would sequester estrogens from the environment. Incubation of spermatozoa from wild-type and 11K-SHBG transgenic with labelled [3H]estrogen would be predicted to result in higher radioactive counts within the transgenic    127 sperm, and this could be easily tested. Moreover to test the influence of sperm SHBG on non-genomic estrogen receptor signalling, sperm from wild-type and 11K-SHBG transgenic mice could be incubated with estrogens and the incidence of tyrosine phosphorylation could be assessed [256]. Assays for spontaneous acrosome reaction and forward motility following estrogen incubation are also warranted [256], [275]. Co-administration with a specific SHBG antagonist may counter any effects attributed to sperm SHBG.  6.4 Final remarks  This thesis expands our understanding of how sperm SHBG is expressed during the seminiferous cycle and how this fits within the unique paradigm of gene expression in testicular germ cells. It also sheds light into how testicular SHBG expression differs across primates and allows us to identify those species that would be appropriate animals models to study germ cell SHBG in vivo. It is possible that there are related roles of testicular SHBG across mammals despite being synthesized in different locations. Direct interrogation of testicular SHBG in vivo by pharmacological means and/or gene manipulation is clearly warranted and still remains a gaping hole in the literature after over 40 years of study and speculation. Whether or not testicular SHBG will emerge as a viable contraceptive target remains to be seen.        128      Figure 6.1 Locations of testicular SHBG in rodents versus humans. In rodents and known sub-primate species, testicular SHBG is produced in the Sertoli cells. The majority of this SHBG is secreted into the seminiferous tubule lumen where it travels with released spermatozoa to the caput epididymis [83]. Gerard and colleagues [265], [266] have shown that SHBG is transferred from Sertoli cells to germ cells and represents a minor destination of Sertoli SHBG. In humans and 11K-SHBG transgenic mice, human SHBG is produced in and confined to the germ cells where it accumulates in the acrosomal compartment within the sperm head [66], [96]. This isoform of human SHBG remains in the germ cell following sperm release. PTM, peritubular myoid cell. Figure is adapted from Calvel et al [276].      129       Table 6.1 Relative binding affinities (RBA) of SHBG to selected sex steroids across mammals.  Steroid Rat [277], [278] Rabbit [279] Dog [280] Ram [281] Human [282] Rhesus Macaque [282] Baboon [282], [283] Marmoset [282] Squirrel Monkey [282] DHT 100 100 100 100 100 100 100 100 100 T 50 33 42 32-35 79 87 83-89 77 81 E2 6 4.6 2.5 4-6 33 55 38-69 79 60  Data taken from Westphal 1986 [53]. Original citations indicated. Note that RBA of estrogen is markedly higher in human and non-human primates shown (shaded values). DHT, 5α-dihydrotestoserone, T, testosterone, E2, 17β-estradiol.      130 References  [1] P. Y. Liu and J. D. Veldhuis, “Chapter 13 - The Hypothalamo-Pituitary Unit, Testis, and Male Accessory Organs A2  - Strauss, Jerome F.,” R. L. B. T.-Y. & J. R. E. (Seventh E. Barbieri, Ed. Philadelphia: W.B. Saunders, 2014, pp. 272–286.e8. [2] C. Y. Cheng and D. D. Mruk, “The Blood-Testis Barrier and Its Implications for Male Contraception,” Pharmacol. Rev. , vol. 64 , no. 1 , pp. 16–64, Jan. 2012. [3] H. White-Cooper and N. Bausek, “Evolution and spermatogenesis,” Philos. Trans. R. Soc. London B Biol. Sci., vol. 365, no. 1546, pp. 1465–1480, Apr. 2010. [4] L. O’Donnell, P. K. Nicholls, M. K. O’Bryan, R. I. McLachlan, and P. G. Stanton, “Spermiation,” Spermatogenesis, vol. 1, no. 1, pp. 14–35, Jan. 2011. [5] B. T. B. T. Hinton, “Epididymal epithelium: its contribution to the formation of a luminal fluid microenvironment.,” Microsc. Res. Tech., vol. 30, no. 1, pp. 67–81, Jan. 1995. [6] J.-L. Dacheux, C. Belleannée, R. Jones, V. Labas, M. Belghazi, B. Guyonnet, X. Druart, J. L. Gatti, and F. Dacheux, “Mammalian epididymal proteome.,” Mol. Cell. Endocrinol., vol. 306, no. 1–2, pp. 45–50, Jul. 2009. [7] B. Robaire, B. Hinton, and M. Ogrebin-Crist, “The epididymis,” in Knobil and Neill’s Physiology of Reproduction, J. Neill, Ed. St Louis, MO: Elsevier Academic Press, 2006, pp. 1071–1148. [8] E. F. Oakberg, “Duration of spermatogenesis in the mouse and timing of stages of the cycle of the seminiferous epithelium,” Am. J. Anat., vol. 99, no. 3, pp. 507–516, Nov. 1956. [9] C. G. Heller and Y. Clermont, “Spermatogenesis in Man: An Estimate of Its Duration,” Science (80-. )., vol. 140, no. 3563, pp. 184–186, Apr. 1963. [10] C. Heller and Y. Clermont, “Kinetics of the Germinal Epithelium in Man,” Recent Prog. Horm. Res., vol. 20, pp. 545–575, 1964. [11] E. F. Oakberg, “A description of spermiogenesis in the mouse and its use in analysis of the cycle of the seminiferous epithelium and germ cell renewal,” Am. J. Anat., vol. 99, no. 3, pp. 391–413, Nov. 1956. [12] B. Perey, Y. Clermont, and C. P. Leblond, “The wave of the seminiferous epithelium in the rat,” Am. J. Anat., vol. 108, no. 1, pp. 47–77, Jan. 1961. [13] C. T. Dann, A. L. Alvarado, L. A. Molyneux, B. S. Denard, D. L. Garbers, and M. H. Porteus, “Spermatogonial Stem Cell Self-Renewal Requires OCT4, a Factor Downregulated During Retinoic Acid-Induced Differentiation,” Stem Cells, vol. 26, no. 11, pp. 2928–2937, Nov. 2008. [14] S.-Y. Hwang, B. Oh, B. B. Knowles, D. Solter, and J.-S. Lee, “Expression of genes involved in mammalian meiosis during the transition from egg to embryo,” Mol. Reprod. Dev., vol. 59, no. 2, pp. 144–158, Jun. 2001. [15] J.-P. Dadoune, J.-P. Siffroi, and M.-F. Alfonsi, “Transcription in Haploid Male Germ Cells,” vol. Volume 237, B. T.-I. R. of Cytology, Ed. Academic Press, 2004, pp. 1–56.     131 [16] E. M. Eddy, “Regulation of gene expression during spermatogenesis,” Semin. Cell Dev. Biol., vol. 9, no. 4, pp. 451–457, Aug. 1998. [17] E. M. E. M. Eddy, “Male germ cell gene expression.,” Recent Prog. Horm. Res., vol. 57, pp. 103–128, 2002. [18] C. Mori, J. E. Welch, Y. Sakai, and E. M. Eddy, “In situ localization of spermatogenic cell-specific glyceraldehyde 3-phosphate dehydrogenase (Gapd-s) messenger ribonucleic acid in mice.,” Biol. Reprod. , vol. 46 , no. 5 , pp. 859–868, May 1992. [19] D. O. Bunch, J. E. Welch, P. L. Magyar, E. M. Eddy, and D. A. O’Brien, “Glyceraldehyde 3-phosphate dehydrogenase-S protein distribution during mouse spermatogenesis.,” Biol. Reprod. , vol. 58 , no. 3 , pp. 834–841, Mar. 1998. [20] K. Kitamura, Y. Miyagawa, N. Iguchi, H. Nishimura, H. Tanaka, and Y. Nishimune, “Molecular cloning and characterization of the human orthologue of the oppo 1 gene encoding a sperm tail protein,” Mol. Hum. Reprod. , vol. 9 , no. 5 , pp. 237–243, May 2003. [21] N. Baek, J.-M. Woo, C. Han, E. Choi, I. Park, D. H. Kim, E. M. Eddy, and C. Cho, “Characterization of eight novel proteins with male germ cell-specific expression in mouse,” Reprod. Biol. Endocrinol., vol. 6, no. 1, pp. 1–12, 2008. [22] D. J. D. J. Wolgemuth, “List of cloned mouse genes with unique expression patterns during spermatogenesis.,” Mamm. genome, vol. 1, no. 4, pp. 283–288, 1991. [23] K. Willison and A. Ashworth, “Mammalian spermatogenic gene expression,” Trends Genet., vol. 3, pp. 351–355, May 1987. [24] J. A. MacLean II and M. F. Wilkinson, “Gene Regulation in Spermatogenesis,” vol. Volume 71, B. T.-C. T. in D. Biology, Ed. Academic Press, 2005, pp. 131–197. [25] V. Delmas, F. van der Hoorn, B. Mellström, B. Jégou, and P. Sassone-Corsi, “Induction of CREM activator proteins in spermatids: down-stream targets and implications for haploid germ cell differentiation.,” Mol. Endocrinol., vol. 7, no. 11, pp. 1502–1514, Nov. 1993. [26] N. S. N. S. Foulkes, “Pituitary hormone FSH directs the CREM functional switch during spermatogenesis.,” Nat., vol. 362, no. 6417, pp. 264–267, Jan. 1993. [27] K. Kleene, “Patterns, mechanisms, and functions of translation regulation in mammalian spermatogenic cells,” Cytogenet. Genome Res., vol. 103, no. 3–4, pp. 217–224, 2003. [28] D. O. Norris, Vertebrate Endocrinology, 4th ed. Burlington, MA: Elsevier Academic Press, 2010. [29] M. H. Abel, P. J. Baker, H. M. Charlton, A. Monteiro, G. Verhoeven, K. De Gendt, F. Guillou, and P. J. O’Shaughnessy, “Spermatogenesis and Sertoli Cell Activity in Mice Lacking Sertoli Cell Receptors for Follicle-Stimulating Hormone and Androgen,” Endocrinology, vol. 149, no. 7, pp. 3279–3285, Jul. 2008. [30] P. J. O’Shaughnessy, A. Monteiro, and M. Abel, “Testicular Development in Mice Lacking Receptors for Follicle Stimulating Hormone and Androgen,” PLoS One, vol. 7, no. 4, p. e35136, Apr. 2012. [31] P. J. O’Shaughnessy, “Hormonal control of germ cell development and spermatogenesis,” Semin. Cell Dev. Biol., vol. 29, pp. 55–65, May 2014.    132 [32] L. O’Hara and L. B. Smith, “Androgen receptor roles in spermatogenesis and infertility,” Best Pract. Res. Clin. Endocrinol. Metab., vol. 29, no. 4, pp. 595–605, Aug. 2015. [33] R.-S. Wang, S. Yeh, C.-R. Tzeng, and C. Chang, “Androgen Receptor Roles in Spermatogenesis and Fertility: Lessons from Testicular Cell-Specific Androgen Receptor Knockout Mice,” Endocr. Rev., vol. 30, no. 2, pp. 119–132, Dec. 2008. [34] G. Verhoeven, A. Willems, E. Denolet, J. V Swinnen, and K. De Gendt, “Androgens and spermatogenesis: lessons from transgenic mouse models,” Philos. Trans. R. Soc. London B Biol. Sci., vol. 365, no. 1546, pp. 1537–1556, Apr. 2010. [35] M.-Y. Tsai, S.-D. Yeh, R.-S. Wang, S. Yeh, C. Zhang, H.-Y. Lin, C.-R. Tzeng, and C. Chang, “Differential effects of spermatogenesis and fertility in mice lacking androgen receptor in individual testis cells,” Proc. Natl. Acad. Sci. , vol. 103 , no. 50 , pp. 18975–18980, Dec. 2006. [36] H. Takeda, G. Chodak, S. Mutchnik, T. Nakamoto, and C. Chang, “Immunohistochemical localization of androgen receptors with mono- and polyclonal antibodies to androgen receptor,” J. Endocrinol. , vol. 126 , no. 1 , p. 17–NP, Jul. 1990. [37] M. SAR, D. B. LUBAHN, F. S. FRENCH, and E. M. WILSON, “Immunohistochemical Localization of the Androgen Receptor in Rat and Human Tissues,” Endocrinology, vol. 127, no. 6, pp. 3180–3186, Dec. 1990. [38] C. A. Suárez-Quian, F. Martínez-García, M. Nistal, and J. Regadera, “Androgen Receptor Distribution in Adult Human Testis,” J. Clin. Endocrinol. Metab., vol. 84, no. 1, pp. 350–358, Jan. 1999. [39] Q. ZHOU, R. NIE, G. S. PRINS, P. T. K. SAUNDERS, B. S. KATZENELLENBOGEN, and R. E. X. A. HESS, “Localization of Androgen and Estrogen Receptors in Adult Male Mouse Reproductive Tract,” J. Androl., vol. 23, no. 6, pp. 870–881, Nov. 2002. [40] E. Lifschytz and D. L. Lindsley, “The Role of X-Chromosome Inactivation during Spermatogenesis,” Proc. Natl. Acad. Sci. U. S. A., vol. 69, no. 1, pp. 182–186, Jan. 1972. [41] J. R. McCarrey, W. M. Berg, S. J. Paragioudakis, P. L. Zhang, D. D. Dilworth, B. L. Arnold, and J. J. Rossi, “Differential transcription of Pgk genes during spermatogenesis in the mouse,” Dev. Biol., vol. 154, no. 1, pp. 160–168, 1992. [42] M. Welsh, P. T. K. Saunders, N. Atanassova, R. M. Sharpe, and L. B. Smith, “Androgen action via testicular peritubular myoid cells is essential for male fertility,” FASEB J. , vol. 23 , no. 12 , pp. 4218–4230, Dec. 2009. [43] K. De Gendt, J. V Swinnen, P. T. K. Saunders, L. Schoonjans, M. Dewerchin, A. Devos, K. Tan, N. Atanassova, F. Claessens, C. Lécureuil, W. Heyns, P. Carmeliet, F. Guillou, R. M. Sharpe, and G. Verhoeven, “A Sertoli cell-selective knockout of the androgen receptor causes spermatogenic arrest in meiosis,” Proc. Natl. Acad. Sci. United States Am. , vol. 101 , no. 5 , pp. 1327–1332, Feb. 2004. [44] C. Chang, Y.-T. Chen, S.-D. Yeh, Q. Xu, R.-S. Wang, F. Guillou, H. Lardy, and S. Yeh, “Infertility with defective spermatogenesis and hypotestosteronemia in male mice lacking the androgen receptor in Sertoli cells,” Proc. Natl. Acad. Sci. United States Am. , vol. 101 , no. 18 , pp. 6876–6881, May 2004.    133 [45] R. W. Holdcraft and R. E. Braun, “Androgen receptor function is required in Sertoli cells for the terminal differentiation of haploid spermatids,” Development, vol. 131, no. 2, pp. 459–467, Dec. 2003. [46] A. Willems, S. R. Batlouni, A. Esnal, J. V Swinnen, P. T. K. Saunders, R. M. Sharpe, L. R. França, K. De Gendt, and G. Verhoeven, “Selective Ablation of the Androgen Receptor in Mouse Sertoli Cells Affects Sertoli Cell Maturation, Barrier Formation and Cytoskeletal Development,” PLoS One, vol. 5, no. 11, p. e14168, Nov. 2010. [47] L.-Y. Chen, P. R. Brown, W. B. Willis, and E. M. Eddy, “Peritubular Myoid Cells Participate in Male Mouse Spermatogonial Stem Cell Maintenance,” Endocrinology, vol. 155, no. 12, pp. 4964–4974, Sep. 2014. [48] S. Windschüttl, C. Kampfer, C. Mayer, F. Flenkenthaler, T. Fröhlich, J. U. Schwarzer, F. M. Köhn, H. Urbanski, G. J. Arnold, and A. Mayerhofer, “Human testicular peritubular cells secrete pigment epithelium-derived factor (PEDF), which may be responsible for the avascularity of the seminiferous tubules,” Sci. Rep., vol. 5, p. 12820, Sep. 2015. [49] L. O’Hara, K. McInnes, I. Simitsidellis, S. Morgan, N. Atanassova, J. Slowikowska-Hilczer, K. Kula, M. Szarras-Czapnik, L. Milne, R. T. Mitchell, and L. B. Smith, “Autocrine androgen action is essential for Leydig cell maturation and function, and protects against late-onset Leydig cell apoptosis in both mice and men,” FASEB J. , vol. 29 , no. 3 , pp. 894–910, Mar. 2015. [50] M. Welsh, R. M. Sharpe, L. Moffat, N. Atanassova, P. T. K. Saunders, S. Kilter, A. Bergh, and L. B. Smith, “Androgen Action via Testicular Arteriole Smooth Muscle Cells Is Important for Leydig Cell Function, Vasomotion and Testicular Fluid Dynamics,” PLoS One, vol. 5, no. 10, p. e13632, Oct. 2010. [51] L. O’Hara and L. B. Smith, “Androgen receptor signalling in Vascular Endothelial cells is dispensable for spermatogenesis and male fertility,” BMC Res. Notes, vol. 5, no. 1, pp. 1–7, 2012. [52] M. E. Baker, “Albumin, steroid hormones and the origin of vertebrates,” J. Endocrinol. , vol. 175 , no. 1 , pp. 121–127, Oct. 2002. [53] U. Westphal, “Steroid-Protein Interactions II,” Monogr. Endocrinol., vol. 27, pp. 1–603, 1986. [54] P. K. P. K. Siiteri, “The serum transport of steroid hormones.,” Recent Prog. Horm. Res., vol. 38, pp. 457–510, 1982. [55] G. L. Hammond, “Access of reproductive steroids to target tissues,” Obstet. Gynecol. Clin. North Am., vol. 29, pp. 411–324, 2002. [56] C. M. Mendel, “The Free Hormone Hypothesis : A Physiologically Based Mathematical Model,” Endocr. Rev., vol. 10, no. 3, pp. 232–274, 1989. [57] D. Berube, G. E. Seralini, R. Gagne, and G. L. Hammond, “Localization of the human sex hormone-binding globulin gene (SHBG) to the short arm of chromosome 17 (17p12----p13),” Cytogenet Cell Genet, vol. 54, no. 1–2. pp. 65–67, 1990. [58] G. L. Hammond, D. A. Underhill, H. M. Rykse, and C. L. Smith, “The human sex hormone-binding globulin gene contains exons for androgen-binding protein and two other testicular messenger RNAs.,” Molecular endocrinology (Baltimore, Md.), vol. 3, no. 11. pp. 1869–76, 1989.     134 [59] S. Gershagen, A. Lundwall, and P. Fernlund, “Characterization of the human sex hormone binding globulin (SHBG) gene and demonstration of two transcripts in both liver and testis.,” Nucleic Acids Res., vol. 17, no. 22, pp. 9245–9258, Nov. 1989. [60] M. Jänne, “Hepatocyte Nuclear Factor-4 Controls Transcription from a TATA-less Human Sex Hormone-binding Globulin Gene Promoter,” J. Biol. Chem., vol. 273, no. 51, pp. 34105–34114, 1998. [61] D. R. Joseph and M. E. Baker, “Sex hormone-binding globulin, androgen-binding protein, and vitamin K-dependent protein S are homologous to laminin A, merosin, and Drosophila crumbs protein.,” FASEB J. , vol. 6 , no. 7 , pp. 2477–2481, Apr. 1992. [62] G. L. Hammond and W. P. Bocchinfuso, “Sex hormone-binding globulin/androgen-binding protein: Steroid-binding and dimerization domains,” J. Steroid Biochem. Mol. Biol., vol. 53, no. 1–6, pp. 543–552, 1995. [63] C. Hildebrand, W. P. Bocchinfuso, D. Dales, and G. L. Hammond, “Resolution of the steroid-binding and dimerization domains of human sex hormone-binding globulin by expression in Escherichia coli,” Biochemistry, vol. 34, no. 10, pp. 3231–3238, Mar. 1995. [64] G. L. Hammond and W. P. Bocchinfuso, “Sex hormone-binding globulin: gene organization and structure/function analyses.,” Horm. Res., vol. 45, no. 3–5, pp. 197–201, 1996. [65] G. L. Hammond, D. A. Underhill, C. L. Smith, I. S. Goping, M. J. Harley, N. A. Musto, C. Y. Cheng, and C. W. Bardin, “The cDNA-deduced primary structure of human sex hormone-binding globulin and location of its steroid-binding domain,” FEBS Lett., vol. 215, no. 1, pp. 100–104, 1987. [66] D. M. Selva, K. N. Hogeveen, K. Seguchi, F. Tekpetey, and G. L. Hammond, “A human sex hormone-binding globulin isoform accumulates in the acrosome during spermatogenesis,” J. Biol. Chem., vol. 277, no. 47, pp. 45291–45298, 2002. [67] D. M. Selva and G. L. Hammond, “Human sex hormone-binding globulin is expressed in testicular germ cells and not in sertoli cells,” Horm Metab Res, vol. 38, no. 4, pp. 230–235, 2006. [68] G. L. Hammond, “Diverse roles for sex hormone-binding globulin in reproduction.,” Biol. Reprod., vol. 85, no. 3, pp. 431–441, 2011. [69] M. Pugeat, N. Nader, K. Hogeveen, G. Raverot, H. Déchaud, and C. Grenot, “Sex hormone-binding globulin gene expression in the liver: Drugs and the metabolic syndrome,” Mol. Cell. Endocrinol., vol. 316, no. 1, pp. 53–59, Mar. 2010. [70] R. Simó, C. Sáez-López, A. Barbosa-Desongles, C. Hernández, and D. M. Selva, “Novel insights in SHBG regulation and clinical implications,” Trends Endocrinol. Metab., vol. 26, no. 7, pp. 376–383, May 2016. [71] N. Stefan, F. Schick, and H.-U. Häring, “Sex Hormone–Binding Globulin and Risk of Type 2 Diabetes,” N. Engl. J. Med., vol. 361, no. 27, pp. 2675–2678, Dec. 2009. [72] A. Peter, K. Kantartzis, J. Machann, F. Schick, H. Staiger, F. Machicao, E. Schleicher, A. Fritsche, H.-U. Häring, and N. Stefan, “Relationships of Circulating Sex Hormone–Binding Globulin With Metabolic Traits in Humans,” Diabetes, vol. 59, no. 12, pp. 3167–3173, Nov. 2010.     135 [73] A. R. GLASS, R. S. SWERDLOFF, G. A. BRAY, W. T. DAHMS, and R. L. ATKINSON, “Low Serum Testosterone and Sex-Hormone-Binding-Globulin in Massively Obese Men,” J. Clin. Endocrinol. Metab., vol. 45, no. 6, pp. 1211–1219, Dec. 1977. [74] D. M. Selva, K. N. Hogeveen, S. M. Innis, and G. L. Hammond, “Monosaccharide-induced lipogenesis regulates the human hepatic sex hormone–binding globulin gene,” J. Clin. Invest., vol. 117, no. 12, pp. 3979–3987, Dec. 2007. [75] D. C. ANDERSON, “SEX-HORMONE-BINDING GLOBULIN,” Clin. Endocrinol. (Oxf)., vol. 3, no. 1, pp. 69–96, Jan. 1974. [76] D. M. Selva and G. L. Hammond, “Thyroid hormones act indirectly to increase sex hormone-binding globulin production by liver via hepatocyte nuclear factor-4??,” J. Mol. Endocrinol., vol. 43, no. 1, pp. 19–27, 2009. [77] M. A. Thaler, V. Seifert-Klauss, and P. B. Luppa, “The biomarker sex hormone-binding globulin &#x2013; From established applications to emerging trends in clinical medicine,” Best Pract. Res. Clin. Endocrinol. Metab., vol. 29, no. 5, pp. 749–760, May 2016. [78] M. Jänne, H. K. Deol, S. G. A. Power, S. Yee, and G. L. Hammond, “Human Sex Hormone-Binding Globulin Gene Expression in Transgenic Mice,” Mol. Endocrinol., vol. 12, no. 10, pp. 123–136, 1998. [79] E.-J. Hong, “REGULATION OF ANDROGEN ACTION BY SEX HORMONE-BINDING GLOBULIN,” University of British Columbia, 2011. [80] E.-J. Hong, B. Sahu, O. A. Jänne, and G. L. Hammond, “Cytoplasmic Accumulation of Incompletely Glycosylated SHBG Enhances Androgen Action in Proximal Tubule Epithelial Cells.,” Mol Endocrinol, vol. 25, no. 2, pp. 269–281, 2011. [81] E. M. Ritzén, M. C. Dobbins, D. J. Tindall, F. S. French, and S. N. Nayfeh, “Characterization of an androgen binding protein (ABP) in rat testis and epididymis,” Steroids, vol. 21, no. 4, pp. 593–607, Apr. 1973. [82] F. S. FRENCH and E. M. RITZÉN, “ANDROGEN-BINDING PROTEIN IN EFFERENT DUCT FLUID OF RAT TESTIS,” J. Reprod. Fertil. , vol. 32 , no. 3 , pp. 479–483, Mar. 1973. [83] D. R. D. R. Joseph, “Structure, function, and regulation of androgen-binding protein/sex hormone-binding globulin.,” Vitam. Horm., vol. 49, pp. 197–280, 1994. [84] H. S. KEEPING, S. J. WINTERS, B. ATTARDI, and P. TROEN, “Developmental Changes in Testicular Inhibin and Androgen-Binding Protein during Sexual Maturation in the Cynomolgus Monkey, Macaca fascicularis,” Endocrinology, vol. 126, no. 6, pp. 2858–2867, Jun. 1990. [85] H. S. KEEPING, S. J. WINTERS, and P. TROEN, “Identification of Androgen-Binding Protein from Testis Cytosol and Sertoli Cell Culture Medium of the Cynomolgus Monkey, Macaca fascicularis,” Endocrinology, vol. 117, no. 4, pp. 1521–1529, Oct. 1985. [86] V. V Hansson, “FSH stimulation of testicular androgen binding protein.,” Nature. New Biol., vol. 246, no. 150, pp. 56–58, Jan. 1973. [87] L. J. PELLINIEMI, M. DYM, G. L. GUNSALUS, N. A. MUSTO, C. W. BARDIN, and D. O. N. W. FAWCETT, “Immunocytochemical Localization of Androgen-Binding Protein in the Male Rat Reproductive Tract,” Endocrinology, vol. 108, no. 3, pp. 925–931, Mar. 1981.    136 [88] A. ATTRAMADAL, C. W. BARDIN, G. L. GUNSALUS, N. A. MUSTO, and V. HANSSON, “Immunocytochemical Localization of Androgen Binding Protein in Rat Sertoli and Epididymal Cells,” Biol. Reprod. , vol. 25 , no. 5 , pp. 983–988, Dec. 1981. [89] A. GERARD, J. KHANFRI, J. L. GUEANT, S. FREMONT, J. P. NICOLAS, G. GRIGNON, and H. GERARD, “Electron Microscope Radioautographic Evidence of in Vivo Androgen-Binding Protein Internalization in the Rat Epididymis Principal Cells,” Endocrinology, vol. 122, no. 4, pp. 1297–1307, Apr. 1988. [90] J. L. Guéant, S. Fremont, J. Khanfri, A. Gérard, G. Grignon, J. P. Nicolas, and H. Gérard, “Biochemical evidences for a receptor-mediated uptake of rat androgen binding protein by epididymis,” Steroids, vol. 52, no. 4, pp. 347–348, Oct. 1988. [91] F. Felden, B. Leheup, S. Fremont, R. Bouguerne, M. Egloff, J. P. Nicolas, G. Grignon, and J. L. Gueant, “The plasma membrane of epididymal epithelial cells has a specific receptor which binds to androgen-binding protein and sex steroid-binding protein,” J. Steroid Biochem. Mol. Biol., vol. 42, no. 3–4, pp. 279–285, May 1992. [92] J.-L. Guéant, S. Fremont, F. Felden, J.-P. Nicolas, A. Gerard, B. Leheup, H. Gerard, and G. Grignon, “Evidence that androgen-binding protein endocytosis in vitro is receptor mediated in principal cells of the rat epididymis,” J. Mol. Endocrinol. , vol. 7 , no. 2 , pp. 113–122, Oct. 1991. [93] J. Larriva-Sahd, H. Orozco, R. Hernandez-Pando, R. M. Oliart, N. A. Musto, and F. Larrea, “Immunohistochemical demonstration of androgen-binding protein in the rat prostatic gland,” Biol Reprod, vol. 45, no. 3, pp. 417–423, 1991. [94] T. T. TURNER, C. E. JONES, S. S. HOWARDS, L. L. EWING, B. ZEGEYE, and G. L. GUNSALUS, “On the Androgen Microenvironment of Maturing Spermatozoa,” Endocrinology, vol. 115, no. 5, pp. 1925–1932, Nov. 1984. [95] A.-F. Hsu and P. Troen, “An Androgen Binding Protein in the Testicular Cytosol of Human Testis: COMPARISON WITH HUMAN PLASMA TESTOSTERONE-ESTROGEN BINDING GLOBULIN,” J. Clin. Invest., vol. 61, no. 6, pp. 1611–1619, 1978. [96] D. M. Selva, L. Bassas, F. Munell, A. Mata, F. Tekpetey, J. G. Lewis, and G. L. Hammond, “Human sperm sex hormone-binding globulin isoform: Characterization and measurement by time-resolved fluorescence immunoassay,” J. Clin. Endocrinol. Metab., vol. 90, no. 11, pp. 6275–6282, 2005. [97] J. Reventos, P. M. Sullivan, D. R. Joseph, and J. W. Gordon, “Tissue-specific expression of the rat androgen-binding protein/sex hormone-binding globulin gene in transgenic mice,” Mol. Cell. Endocrinol., vol. 96, no. 1–2, pp. 69–73, 1993. [98] D. M. Selva, K. N. Hogeveen, and G. L. Hammond, “Repression of the human sex hormone-binding globulin gene in sertoli cells by upstream stimulatory transcription factors,” J. Biol. Chem., vol. 280, no. 6, pp. 4462–4468, 2005. [99] S. Schlatt and J. Ehmcke, “Regulation of spermatogenesis: An evolutionary biologist’s perspective,” Semin. Cell Dev. Biol., vol. 29, pp. 2–16, May 2014. [100] L. Rato, M. G. Alves, S. Socorro, A. I. Duarte, J. E. Cavaco, and P. F. Oliveira, “Metabolic regulation is important for spermatogenesis,” Nat Rev Urol, vol. 9, no. 6, pp. 330–338, Jun. 2012.    137 [101] Y. Clermont, “The cycle of the seminiferous epithelium in man,” Am. J. Anat., vol. 112, no. 1, pp. 35–51, Jan. 1963. [102] C. Marc Luetjens, G. F. Weinbauer, and J. Wistuba, “Primate spermatogenesis: new insights into comparative testicular organisation, spermatogenic efficiency and endocrine control,” Biol. Rev., vol. 80, no. 03, pp. 475–488, 2005. [103] L. Russell, R. . Ettlin, A. P. Sinha Hikim, and E. D. Clegg, Histological and Histopathological Evaluation of the Testis. Clearwater, Florida: Cache River Press, 1990. [104] M. C. Hofmann, R. A. Hess, E. Goldberg, and J. L. Millán, “Immortalized germ cells undergo meiosis in vitro.,” Proc. Natl. Acad. Sci. U. S. A., vol. 91, no. 12, pp. 5533–5537, Jun. 1994. [105] T.-S. Wu, “Functional characterization of sex hormone-binding globulin genetic polymorphism,” University of British Columbia, 2015. [106] C. Boucheron and V. Baxendale, “Isolation and Purification of Murine Male Germ Cells BT  - Germline Development: Methods and Protocols,” W.-Y. Chan and A. Le Blomberg, Eds. New York, NY: Springer New York, 2012, pp. 59–66. [107] X. Huang and W. Miller, “A time-efficient, linear-space local similarity algorithm,” Adv. Appl. Math., vol. 12, no. 3, pp. 337–357, 1991. [108] G. L. Hammond and P. A. Robinson, “Characterization of a monoclonal antibody to human sex hormone binding globulin,” FEBS Lett., vol. 168, no. 2, pp. 307–312, 1984. [109] G. L. Hammond, M. S. Langley, and P. A. Robinson, “A liquid-phase immunoradiometric assay (IRMA) for human sex hormone binding globulin (SHBG),” J Steroid Biochem, vol. 23, no. 4. pp. 451–460, 1985. [110] W. H. Walker, B. M. Sanborn, and J. F. Habener, “An isoform of transcription factor CREM expressed during spermatogenesis lacks the phosphorylation domain and represses cAMP-induced transcription.,” Proc. Natl. Acad. Sci. U. S. A., vol. 91, no. 26, pp. 12423–12427, Dec. 1994. [111] M. Godmann, I. Kromberg, J. Mayer, and R. Behr, “The mouse Krüppel-like Factor 4 (Klf4) gene: Four functional polyadenylation sites which are used in a cell-specific manner as revealed by testicular transcript analysis and multiple processed pseudogenes,” Gene, vol. 361, pp. 149–156, Nov. 2005. [112] K. N. Hogeveen, M. Talikka, and G. L. Hammond, “Human Sex Hormone-binding Globulin Promoter Activity Is Influenced by a (TAAAA)n Repeat Element within an Alu Sequence,” J. Biol. Chem., vol. 276, no. 39, pp. 36383–36390, 2001. [113] T. Pinós, A. Barbosa-Desongles, A. Hurtado, A. Santamaria-Martínez, I. de Torres, J. Morote, J. Reventós, and F. Munell, “Identification, characterization and expression of novel Sex Hormone Binding Globulin alternative first exons in the human prostate.,” BMC Mol. Biol., vol. 10, p. 59, 2009. [114] A. M. Nakhla, D. J. Hryb, W. Rosner, N. A. Romas, Z. Xiang, and S. M. Kahn, “Human sex hormone-binding globulin gene expression- multiple promoters and complex alternative splicing,” BMC Mol. Biol., vol. 10, no. 1, pp. 1–18, 2009. [115] M. Gardiner-Garden and M. Frommer, “CpG Islands in vertebrate genomes,” J. Mol. Biol., vol. 196, no. 2, pp. 261–282, 1987.    138 [116] F. Larsen, G. Gundersen, R. Lopez, and H. Prydz, “CpG islands as gene markers in the human genome,” Genomics, vol. 13, no. 4, pp. 1095–1107, Aug. 1992. [117] M. R. Montminy, K. A. Sevarino, J. A. Wagner, G. Mandel, and R. H. Goodman, “Identification of a cyclic-AMP-responsive element within the rat somatostatin gene,” Proc. Natl. Acad. Sci. , vol. 83 , no. 18 , pp. 6682–6686, Sep. 1986. [118] A. R. Black, J. D. Black, and J. Azizkhan-Clifford, “Sp1 and krüppel-like factor family of transcription factors in cell growth regulation and cancer,” J. Cell. Physiol., vol. 188, no. 2, pp. 143–160, Aug. 2001. [119] J. Kaczynski, T. Cook, and R. Urrutia, “Sp1- and Krüppel-like transcription factors,” Genome Biol., vol. 4, no. 2, p. 206, Feb. 2003. [120] W. Ma, G. C. Horvath, M. K. Kistler, and W. S. Kistler, “Expression Patterns of SP1 and SP3 During Mouse Spermatogenesis: SP1 Down-Regulation Correlates with Two Successive Promoter Changes and Translationally Compromised Transcripts,” Biol. Reprod. , vol. 79 , no. 2 , pp. 289–300, Aug. 2008. [121] R. Behr and K. H. Kaestner, “Developmental and cell type-specific expression of the zinc finger transcription factor Krüppel-like factor 4 (Klf4) in postnatal mouse testis,” Mech. Dev., vol. 115, no. 1–2, pp. 167–169, Jul. 2002. [122] M. Godmann, I. Gashaw, J. P. Katz, A. Nagy, K. H. Kaestner, and R. Behr, “Krüppel-like Factor 4, a ‘Pluripotency Transcription Factor’ Highly Expressed in Male Postmeiotic Germ Cells, Is Dispensable for Spermatogenesis in the Mouse,” Mech. Dev., vol. 126, no. 8–9, pp. 650–664, Jun. 2009. [123] J. Shi, B. Zheng, S. Chen, G. Ma, and J. Wen, “Retinoic Acid Receptor α Mediates All-trans-retinoic Acid-induced Klf4 Gene Expression by Regulating Klf4 Promoter Activity in Vascular Smooth Muscle Cells,” J. Biol. Chem., vol. 287, no. 14, pp. 10799–10811, Mar. 2012. [124] E. E. Schmidt and U. Schibler, “High accumulation of components of the RNA polymerase II transcription machinery in rodent spermatids,” Development, vol. 121, no. 8, pp. 2373–2383, Aug. 1995. [125] P. Sassone-Corsi, “Unique Chromatin Remodeling and Transcriptional Regulation in Spermatogenesis,” Science (80-. )., vol. 296, no. 5576, pp. 2176–2178, Jun. 2002. [126] J. A. Segre, C. Bauer, and E. Fuchs, “Klf4 is a transcription factor required for establishing the barrier function of the skin,” Nat Genet, vol. 22, no. 4, pp. 356–360, Aug. 1999. [127] J. P. Katz, N. Perreault, B. G. Goldstein, C. S. Lee, P. A. Labosky, V. W. Yang, and K. H. Kaestner, “The zinc-finger transcription factor Klf4 is required for terminal differentiation of goblet cells in the colon,” Development, vol. 129, no. 11, pp. 2619–2628, Jun. 2002. [128] K. Takahashi and S. Yamanaka, “Induction of Pluripotent Stem Cells from Mouse Embryonic and Adult Fibroblast Cultures by Defined Factors,” Cell, vol. 126, no. 4, pp. 663–676, Aug. 2006. [129] R. Behr, C. Deller, M. Godmann, T. Müller, M. Bergmann, R. Ivell, and K. Steger, “Krüppel-like factor 4 expression in normal and pathological human testes,” Mol. Hum. Reprod. , vol. 13 , no. 11 , pp. 815–820, Nov. 2007.     139 [130] A. Soufi, G. Donahue, and K. S. Zaret, “Facilitators and Impediments of the Pluripotency Reprogramming Factors’ Initial Engagement with the Genome,” Cell, vol. 151, no. 5, pp. 994–1004, Nov. 2012. [131] T. Chen and S. Y. R. Dent, “Chromatin modifiers and remodellers: regulators of cellular differentiation,” Nat Rev Genet, vol. 15, no. 2, pp. 93–106, Feb. 2014. [132] K. S. Zaret and J. S. Carroll, “Pioneer transcription factors: establishing competence for gene expression,” Genes Dev., vol. 25, no. 21, pp. 2227–2241, Nov. 2011. [133] M. Iwafuchi-Doi and K. S. Zaret, “Pioneer transcription factors in cell reprogramming,” Genes Dev., vol. 28, no. 24, pp. 2679–2692, Dec. 2014. [134] P. M. Evans, W. Zhang, X. Chen, J. Yang, K. K. Bhakat, and C. Liu, “Krüppel-like Factor 4 Is Acetylated by p300 and Regulates Gene Transcription via Modulation of Histone Acetylation,” J. Biol. Chem. , vol. 282 , no. 47 , pp. 33994–34002, Nov. 2007. [135] B. Zheng, M. Han, and J.-K. Wen, “Role of Krüppel-like factor 4 in phenotypic switching and proliferation of vascular smooth muscle cells,” IUBMB Life, vol. 62, no. 2, pp. 132–139, Feb. 2010. [136] M. Godmann, J. P. Katz, F. Guillou, M. Simoni, K. H. Kaestner, and R. Behr, “Krüppel-like factor 4 is involved in functional differentiation of testicular Sertoli cells,” Dev. Biol., vol. 315, no. 2, pp. 552–566, Mar. 2008. [137] J. L. Lavrrar and P. J. Farnham, “The Use of Transient Chromatin Immunoprecipitation Assays to Test Models for E2F1-specific Transcriptional Activation,” J. Biol. Chem. , vol. 279 , no. 44 , pp. 46343–46349, Oct. 2004. [138] A. Wulf, A. Harneit, M. Kröger, M. Kebenko, M. G. Wetzel, and J. M. Weitzel, “T3-mediated expression of PGC-1α via a far upstream located thyroid hormone response element,” Mol. Cell. Endocrinol., vol. 287, no. 1–2, pp. 90–95, Jun. 2008. [139] M. Rajković, R. Middendorff, M. G. Wetzel, D. Frković, S. Damerow, H. J. Seitz, and J. M. Weitzel, “Germ Cell Nuclear Factor Relieves cAMP-response Element Modulator τ-mediated Activation of the Testis-specific Promoter of Human Mitochondrial Glycerol-3-phosphate Dehydrogenase,” J. Biol. Chem. , vol. 279 , no. 50 , pp. 52493–52499, Dec. 2004. [140] M. Rajković, K. A. H. Iwen, P. J. Hofmann, A. Harneit, and J. M. Weitzel, “Functional cooperation between CREM and GCNF directs gene expression in haploid male germ cells,” Nucleic Acids Res. , vol. 38 , no. 7 , pp. 2268–2278, Apr. 2010. [141] F. Meng, M. Han, B. Zheng, C. Wang, R. Zhang, X. Zhang, and J. Wen, “All-trans retinoic acid increases KLF4 acetylation by inducing HDAC2 phosphorylation and its dissociation from KLF4 in vascular smooth muscle cells,” Biochem. Biophys. Res. Commun., vol. 387, no. 1, pp. 13–18, Sep. 2009. [142] K. M. Akmal, J. M. Dufour, and K. H. Kim, “Retinoic acid receptor alpha gene expression in the rat testis: potential role during the prophase of meiosis and in the transition from round to elongating spermatids.,” Biol. Reprod. , vol. 56 , no. 2 , pp. 549–556, Feb. 1997. [143] M. Ariel, J. McCarrey, and H. Cedar, “Methylation patterns of testis-specific genes.,” Proc. Natl. Acad. Sci. U. S. A., vol. 88, no. 6, pp. 2317–2321, Mar. 1991.    140 [144] J. M. Trasler, L. E. Hake, P. A. Johnson, A. A. Alcivar, C. F. Millette, and N. B. Hecht, “DNA methylation and demethylation events during meiotic prophase in the mouse testis.,” Mol. Cell. Biol., vol. 10, no. 4, pp. 1828–1834, Apr. 1990. [145] M. Jänne, K. N. Hogeveen, H. K. Deol, and G. L. Hammond, “Expression and Regulation of Human Sex Hormone-Binding Globulin Transgenes in Mice during Development,” Endocrinology, vol. 140, no. 9, pp. 4166–4174, 1999. [146] A. R. Bellvé, C. F. Millette, Y. M. Bhatnagar, and D. A. O’Brien, “Dissociation of the mouse testis and characterization of isolated spermatogenic cells.,” J. Histochem. Cytochem. , vol. 25 , no. 7 , pp. 480–494, Jul. 1977. [147] M. Klug and M. Rehli, “Functional Analysis of Promoter CPG-Methylation using a CpG-Free Luciferase Reporter Vector,” Epigenetics, vol. 1, no. 3, pp. 127–130, Jul. 2006. [148] P.-A. Koenig, P. K. Nicholls, F. I. Schmidt, M. Hagiwara, T. Maruyama, G. H. Frydman, N. Watson, D. C. Page, and H. L. Ploegh, “The E2 Ubiquitin-conjugating Enzyme UBE2J1 Is Required for Spermiogenesis in Mice,” J. Biol. Chem. , vol. 289 , no. 50 , pp. 34490–34502, Dec. 2014. [149] H. Nakata, T. Wakayama, Y. Takai, and S. Iseki, “Quantitative Analysis of the Cellular Composition in Seminiferous Tubules in Normal and Genetically Modified Infertile Mice,” J. Histochem. Cytochem. , vol. 63 , no. 2 , pp. 99–113, Feb. 2015. [150] A. P. N. Themmen and I. T. Huhtaniemi, “Mutations of Gonadotropins and Gonadotropin Receptors: Elucidating the Physiology and Pathophysiology of Pituitary-Gonadal Function,” Endocr. Rev., vol. 21, no. 5, pp. 551–583, Oct. 2000. [151] P. Sassone-Corsi, “The Cyclic AMP Pathway,” Cold Spring Harb. Perspect. Biol. , vol. 4 , no. 12 , Dec. 2012. [152] N. S. Foulkes, B. Mellstrom, E. Benusiglio, and P. Sassone-Corsi, “Developmental switch of CREM function during spermatogenesis: from   antagonist to activator,” Nature, vol. 355, no. 6355, pp. 80–84, Jan. 1992. [153] R. Behr and G. F. Weinbauer, “CREM activator and repressor isoforms in human testis: sequence variations and inaccurate splicing during impaired spermatogenesis,” Mol. Hum. Reprod. , vol. 6 , no. 11 , pp. 967–972, Nov. 2000. [154] R. Behr and G. F. Weinbauer, “Germ Cell-Specific Cyclic Adenosine 3′,5′-Monophosphate Response Element Modulator Expression in Rodent and Primate Testis Is Maintained Despite Gonadotropin Deficiency,” Endocrinology, vol. 140, no. 6, pp. 2746–2754, Jun. 1999. [155] J. A. Blendy, K. H. Kaestner, G. F. Weinbauer, E. Nieschlag, and G. Schutz, “Severe impairment of permatogenesis in mice lacking the CREM gene,” Nature, vol. 380, no. 6570, pp. 162–165, Mar. 1996. [156] F. Nantel, L. Monaco, N. S. Foulkes, D. Masquilier, M. LeMeur, K. Henriksen, A. Dierich, M. Parvinen, and P. Sassone-Corsi, “Spermiogenesis deficiency and germ-cell apoptosis in CREM-mutant mice,” Nature, vol. 380, no. 6570, pp. 159–162, Mar. 1996.      141 [157] V. Delmas, B. M. Laoide, D. Masquilier, R. P. de Groot, N. S. Foulkes, and P. Sassone-Corsi, “Alternative usage of initiation codons in mRNA encoding the cAMP-responsive-element modulator generates regulators with opposite functions.,” Proc. Natl. Acad. Sci. , vol. 89 , no. 10 , pp. 4226–4230, May 1992. [158] R. P. de Groot, J. den Hertog, J. R. Vandenheede, J. Goris, and P. Sassone-Corsi, “Multiple and cooperative phosphorylation events regulate the CREM activator function.,” EMBO J., vol. 12, no. 10, pp. 3903–3911, Oct. 1993. [159] P. Brindle, S. Linke, and M. Montminy, “Protein-kinase-A-dependent activator in transcription factor CREB reveals new role for CREM repressers,” Nature, vol. 364, no. 6440, pp. 821–824, Aug. 1993. [160] V. Hansson, B. S. Skålhegg, and K. Taskén, “Cyclic-AMP-dependent protein kinase (PKA) in testicular cells. Cell specific expression, differential regulation and targeting of subunits of PKA,” J. Steroid Biochem. Mol. Biol., vol. 73, no. 1–2, pp. 81–92, May 2000. [161] P. Lönnerberg, M. Parvinen, T. Jahnsen, V. Hansson, and H. Persson, “Stage- and cell-specific expression of cyclic adenosine 3’,5'-monophosphate-dependent protein kinases in rat seminiferous epithelium.,” Biol. Reprod. , vol. 46 , no. 6 , pp. 1057–1068, Jun. 1992. [162] M. Kangasniemi, A. Kaipia, J. Toppari, P. Mali, I. Huhtaniemi, and M. Parvinen, “Cellular regulation of basal and FSH-stimulated cyclic AMP production in irradiated rat testes,” Anat. Rec., vol. 227, no. 1, pp. 32–36, May 1990. [163] M. Kangasniemi, A. Kaipia, P. Mali, J. Toppari, I. Huhtaniemi, and M. Parvinen, “Modulation of basal and FSH-dependent cyclic AMP production in rat seminiferous tubules staged by an improved transillumination technique,” Anat. Rec., vol. 227, no. 1, pp. 62–76, May 1990. [164] N. S. Foulkes, E. Borrelli, and P. Sassone-Corsi, “CREM gene: Use of alternative DNA-binding domains generates multiple antagonists of cAMP-induced transcription,” Cell, vol. 64, no. 4, pp. 739–749, Feb. 1991. [165] W. H. Walker and J. F. Habener, “Role of transcription factors CREB and CREM in cAMP-regulated transcription during spermatogenesis,” Trends Endocrinol. Metab., vol. 7, no. 4, pp. 133–138, May 1996. [166] J. Don and G. Stelzer, “The expanding family of CREB/CREM transcription factors that are involved with spermatogenesis,” Mol. Cell. Endocrinol., vol. 187, no. 1–2, pp. 115–124, Feb. 2002. [167] V. Delmas and P. Sassone-Corsi, “The key role of CREM in the cAMP signaling pathway in the testis,” Mol. Cell. Endocrinol., vol. 100, no. 1–2, pp. 121–124, Apr. 1994. [168] B. M. Laoide, N. S. Foulkes, F. Schlotter, and P. Sassone-Corsi, “The functional versatility of CREM is determined by its modular structure.,” EMBO J., vol. 12, no. 3, pp. 1179–1191, Mar. 1993. [169] G. M. Fimia, D. De Cesare, and P. Sassone-Corsi, “CBP-independent activation of CREM and CREB by the LIM-only protein ACT,” Nature, vol. 398, no. 6723, pp. 165–169, Mar. 1999.      142 [170] D. De Cesare, G. M. Fimia, and P. Sassone-Corsi, “Signaling routes to CREM and CREB: plasticity in transcriptional activation,” Trends Biochem. Sci., vol. 24, no. 7, pp. 281–285, Jul. 1999. [171] M. Rodova, A.-N. Nguyen, and G. Blanco, “The transcription factor CREMτ and cAMP regulate promoter activity of the Na,K-ATPase α4 isoform,” Mol. Reprod. Dev., vol. 73, no. 11, pp. 1435–1447, Nov. 2006. [172] N. Kotaja, D. De Cesare, B. Macho, L. Monaco, S. Brancorsini, E. Goossens, H. Tournaye, A. Gansmuller, and P. Sassone-Corsi, “Abnormal sperm in mice with targeted deletion of the act (activator of cAMP-responsive element modulator in testis) gene,” Proc. Natl. Acad. Sci. United States Am. , vol. 101 , no. 29 , pp. 10620–10625, Jul. 2004. [173] A. Lardenois, F. Chalmel, P. Demougin, N. Kotaja, P. Sassone-Corsi, and M. Primig, “Fhl5/Act, a CREM-binding transcriptional activator required for normal sperm maturation and morphology, is not essential for testicular gene expression,” Reprod. Biol. Endocrinol., vol. 7, no. 1, pp. 1–9, 2009. [174] M. J. Wolkowicz, S. A. Coonrod, P. P. Reddi, J. L. Millan, M. C. Hofmann, J. C. Herr, and S. M. Coonrod, “Refinement of the differentiated phenotype of the spermatogenic cell line GC-2spd(ts),” Biol. Reprod. , vol. 55 , no. 4 , pp. 923–932, Oct. 1996. [175] E. R. Kandel, “The molecular biology of memory: cAMP, PKA, CRE, CREB-1, CREB-2, and CPEB,” Mol. Brain, vol. 5, p. 14, May 2012. [176] G.-Y. Wu, K. Deisseroth, and R. W. Tsien, “Activity-dependent CREB phosphorylation: Convergence of a fast, sensitive calmodulin kinase pathway and a slow, less sensitive mitogen-activated protein kinase pathway,” Proc. Natl. Acad. Sci. U. S. A., vol. 98, no. 5, pp. 2808–2813, Feb. 2001. [177] N. S. Foulkes, B. M. Laoide, F. Schlotter, and P. Sassone-Corsi, “Transcriptional antagonist cAMP-responsive element modulator (CREM) down-regulates c-fos cAMP-induced expression.,” Proc. Natl. Acad. Sci. U. S. A., vol. 88, no. 12, pp. 5448–5452, Jun. 1991. [178] B. M. SANBORN, J. L. MILLAN, M. L. MEISTRICH, and L. C. MOORE, “Alternative Splicing of CREB and CREM mRNAs in an Immortalized Germ Cell Line,” J. Androl., vol. 18, no. 1, pp. 62–70, Jan. 1997. [179] S. Shioda, G. Legradi, W. C. Leung, S. Nakajo, K. Nakaya, and A. Arimura, “Localization of pituitary adenylate cyclase-activating polypeptide and its messenger ribonucleic acid in the rat testis by light and electron microscopic immunocytochemistry and in situ hybridization.,” Endocrinology, vol. 135, no. 3, pp. 818–825, Sep. 1994. [180] H. Yanaihara, S. Vigh, T. Kozicz, A. Somogyvári-Vigh, and A. Arimura, “Immunohistochemical demonstration of the intracellular localization of pituitary adenylate cyclase activating polypeptide-like immunoreactivity in the rat testis using the stamp preparation,” Regul. Pept., vol. 78, no. 1–3, pp. 83–88, Nov. 1998. [181] M. Li, H. Funahashi, M. Mbikay, S. Shioda, and A. Arimura, “Pituitary adenylate cyclase activating polypeptide-mediated intracrine signaling in the testicular germ cells,” Endocrine, vol. 23, no. 1, pp. 59–75, 2004.     143 [182] R. A. Vigersky, D. L. Loriaux, S. S. Howards, G. B. Hodgen, M. B. Lipsett, and A. Chrambach, “Androgen binding proteins of testis, epididymis, and plasma in man and monkey.,” J. Clin. Invest., vol. 58, no. 5, pp. 1061–1068, Nov. 1976. [183] D. T. Dang, J. Pevsner, and V. W. Yang, “The biology of the mammalian Krüppel-like family of transcription factors,” Int. J. Biochem. Cell Biol., vol. 32, no. 11–12, pp. 1103–1121, Dec. 2000. [184] R. Pearson, J. Fleetwood, S. Eaton, M. Crossley, and S. Bao, “Krüppel-like transcription factors: A functional family,” Int. J. Biochem. Cell Biol., vol. 40, no. 10, pp. 1996–2001, 2008. [185] R. L. Raaum, “Molecular Evidence on Primate Origins and Evolution,” in Handbook of Paleoanthropology, Second., W. Henke and I. Tattersall, Eds. Berlin, DE: Springer-Verlag., 2014, pp. 1083–1133. [186] M. R. Millar, R. M. Sharpe, G. F. Weinbauer, H. M. Fraser, and P. T. K. Saunders, “Marmoset spermatogenesis: organizational similarities to the human,” Int. J. Androl., vol. 23, no. 5, pp. 266–277, Oct. 2000. [187] J. Wistuba, A. Schrod, B. Greve, J. K. Hodges, H. Aslam, G. F. Weinbauer, and C. M. Luetjens, “Organization of Seminiferous Epithelium in Primates: Relationship to Spermatogenic Efficiency, Phylogeny, and Mating System,” Biol. Reprod. , vol. 69 , no. 2 , pp. 582–591, Aug. 2003. [188] L.-H. Li, J. M. Donald, and M. S. Golub, “Review on testicular development, structure, function, and regulation in common marmoset,” Birth Defects Res. Part B Dev. Reprod. Toxicol., vol. 74, no. 5, pp. 450–469, Oct. 2005. [189] A. H. Payne and M. P. Hardy, The Leydig Cell in Health and Disease. Totowa, NJ: Humana Press Inc., 2007. [190] Bluemel, Korte, Schenk, and G. F. Weinbauer, The Nonhuman Primate in Nonclinical Drug Development and Safety Assessment, 1st ed. London, UK: Elsevier Academic Press, 2015. [191] T. Muller, M. Simoni, E. Pekel, C. M. Luetjens, R. Chandolia, F. Amato, R. J. Norman, and J. Gromoll, “Chorionic gonadotrophin beta subunit mRNA but not luteinising hormone beta subunit mRNA is expressed in the pituitary of the common marmoset (Callithrix jacchus),” J. Mol. Endocrinol. , vol. 32 , no. 1 , pp. 115–128, Feb. 2004. [192] F.-P. Zhang, A. S. Rannikko, P. R. Manna, H. M. Fraser, and I. T. Huhtaniemi, “Cloning and Functional Expression of the Luteinizing Hormone Receptor Complementary Deoxyribonucleic Acid from the Marmoset Monkey Testis: Absence of Sequences Encoding Exon 10 in Other Species,” Endocrinology, vol. 138, no. 6, pp. 2481–2490, Jun. 1997. [193] T. Müller, J. Gromoll, A. P. Simula, R. Norman, R. Sandhowe-Klaverkamp, and M. Simoni, “The Carboxyterminal Peptide of Chorionic Gonadotropin Facilitates Activation of the Marmoset LH Receptor,” Exp Clin Endocrinol Diabetes, vol. 112, no. 10, pp. 574–579, 2004. [194] W. ROSNER, M. M. PUGEAT, G. P. CHROUSOS, and M. S. KHAN, “Steroid-Binding Proteins in Primate Plasma,” Endocrinology, vol. 118, no. 2, pp. 513–517, Feb. 1986. [195] J. K. Hodges, S. A. K. Eastman, and N. Jenkins, “Sex steroids and their relationship to binding proteins in the serum of the marmoset monkey (Callithrix jacchus),” J. Endocrinol. , vol. 96 , no. 3 , pp. 443–450, Mar. 1983.     144 [196] M. Pugeat, B. Rocle, G. P. Chrousos, J. F. Dunn, M. B. Lipsett, and B. C. Nisula, “Plasma testosterone transport in primates,” J. Steroid Biochem., vol. 20, no. 1, pp. 473–478, 1984. [197] P. K. Siiteri, “High Plasma Steroid Levels in the Squirrel Monkey: Deficient Receptors or Metabolism?,” in Steroid Hormone Resistance: Mechanisms and Clinical Aspects, G. P. Chrousos, D. L. Loriaux, and M. B. Lipsett, Eds. Plenum, NY: Springer, 1986, pp. 279–289. [198] G. P. Chrousos, D. Renquist, D. Brandon, C. Eil, M. Pugeat, R. Vigersky, G. B. Cutler, D. L. Loriaux, and M. B. Lipsett, “Glucocorticoid hormone resistance during primate evolution: receptor-mediated mechanisms,” Proc. Natl. Acad. Sci. , vol. 79 , no. 6 , pp. 2036–2040, Mar. 1982. [199] G. L. Hammond, C. L. Smith, P. Lähteenmäki, A. Grolla, S. Warmels-Rodenhiser, H. Hodgert, J. T. Murai, and P. K. Siiteri, “Squirrel monkey corticosteroid-binding globulin: primary structure and comparison with the human protein.,” Endocrinology, vol. 134, no. 2, pp. 891–8, 1994. [200] M. B. Lipsett, “The defective glucocorticoid receptor in man and nonhuman primates.,” Recent Prog. Horm. Res., vol. 41, pp. 199–247, 1985. [201] M. S. Springer, R. W. Meredith, J. Gatesy, C. A. Emerling, J. Park, D. L. Rabosky, T. Stadler, C. Steiner, O. A. Ryder, J. E. Jane?ka, C. A. Fisher, and W. J. Murphy, “Macroevolutionary Dynamics and Historical Biogeography of Primate Diversification Inferred from a Species Supermatrix,” PLoS One, vol. 7, no. 11, p. e49521, Nov. 2012. [202] A. Matsui, F. Rakotondraparany, I. Munechika, M. Hasegawa, and S. Horai, “Molecular phylogeny and evolution of prosimians based on complete sequences of mitochondrial DNAs,” Gene, vol. 441, no. 1–2, pp. 53–66, Jul. 2009. [203] O. R. P. Bininda-Emonds, M. Cardillo, K. E. Jones, R. D. E. MacPhee, R. M. D. Beck, R. Grenyer, S. A. Price, R. A. Vos, J. L. Gittleman, and A. Purvis, “The delayed rise of present-day mammals,” Nature, vol. 446, no. 7135, pp. 507–512, Mar. 2007. [204] P. Perelman, W. E. Johnson, C. Roos, H. N. Seuánez, J. E. Horvath, M. A. M. Moreira, B. Kessing, J. Pontius, M. Roelke, Y. Rumpler, M. P. C. Schneider, A. Silva, S. J. O’Brien, and J. Pecon-Slattery, “A Molecular Phylogeny of Living Primates,” PLoS Genet, vol. 7, no. 3, p. e1001342, Mar. 2011. [205] T. Wicker, F. Sabot, A. Hua-Van, J. L. Bennetzen, P. Capy, B. Chalhoub, A. Flavell, P. Leroy, M. Morgante, O. Panaud, E. Paux, P. SanMiguel, and A. H. Schulman, “A unified classification system for eukaryotic transposable elements,” Nat Rev Genet, vol. 8, no. 12, pp. 973–982, Dec. 2007. [206] M. Muñoz-López and J. L. García-Pérez, “DNA Transposons: Nature and Applications in Genomics,” Curr. Genomics, vol. 11, no. 2, pp. 115–128, Apr. 2010. [207] G. E. Liu, C. Alkan, L. Jiang, S. Zhao, and E. E. Eichler, “Comparative analysis of Alu repeats in primate genomes,” Genome Res. , vol. 19 , no. 5 , pp. 876–885, May 2009. [208] M. A. Batzer and P. L. Deininger, “Alu repeats and human genomic diversity,” Nat Rev Genet, vol. 3, no. 5, pp. 370–379, May 2002. [209] P. Deininger, “Alu elements: know the SINEs,” Genome Biol., vol. 12, no. 12, pp. 1–12, 2011.     145 [210] F. MUNELL, C. A. SUÁREZ-QUIAN, D. M. SELVA, O. M. TIRADO, and J. REVENTÓS, “Androgen-Binding Protein and Reproduction: Where Do We Stand?,” J. Androl., vol. 23, no. 5, pp. 598–609, Sep. 2002. [211] Y. Ma, H.-Z. Yang, L.-M. Xu, Y.-R. Huang, H.-L. Dai, and X.-N. Kang, “Testosterone regulates the autophagic clearance of androgen binding protein in rat Sertoli cells,” Sci. Rep., vol. 5, p. 8894, Mar. 2015. [212] C. L. Smith and G. L. Hager, “Transcriptional Regulation of Mammalian Genes in Vivo : A TALE OF TWO TEMPLATES,” J. Biol. Chem. , vol. 272 , no. 44 , pp. 27493–27496, Oct. 1997. [213] G. Kennedy and B. Sugden, “EBNA-1, a Bifunctional Transcriptional Activator,” Mol. Cell. Biol. , vol. 23 , no. 19 , pp. 6901–6908, Oct. 2003. [214] J. Black and J.-M. Vos, “Establishment of an oriP/EBNA1-based episomal vector transcribing human genomic beta-globin in cultured murine fibroblasts.,” Gene Ther., vol. 9, no. 21, pp. 1447–54, 2002. [215] T. A. Gahn and C. L. Schildkraut, “The Epstein-Barr virus origin of plasmid replication, <em>oriP</em>, contains both the initiation and termination sites of DNA replication,” Cell, vol. 58, no. 3, pp. 527–535, May 2016. [216] S. C. Hung, M.-S. Kang, and E. Kieff, “Maintenance of Epstein–Barr virus (EBV) oriP-based episomes requires EBV-encoded nuclear antigen-1 chromosome-binding domains, which can be replaced by high-mobility group-I or histone H1,” Proc. Natl. Acad. Sci. , vol. 98 , no. 4 , pp. 1865–1870, Feb. 2001. [217] M.-S. Kang, S. C. Hung, and E. Kieff, “Epstein–Barr virus nuclear antigen 1 activates transcription from episomal but not integrated DNA and does not alter lymphocyte growth,” Proc. Natl. Acad. Sci. , vol. 98 , no. 26 , pp. 15233–15238, Dec. 2001. [218] S. Niemi, O. Maentausta, N. J. Bolton, and G. L. Hammond, “Time-resolved immunofluorometric assay of sex-hormone binding globulin,” Clin. Chem., vol. 34, no. 1, pp. 63–66, 1988. [219] G. F. Weinbauer, R. Behr, M. Bergmann, and E. Nieschlag, “Testicular cAMP responsive element modulator (CREM) protein is expressed in round spermatids but is absent or reduced in men with round spermatid maturation arrest,” Mol. Hum. Reprod., vol. 4, no. 1, pp. 9–15, 1998. [220] R. D. Emes, L. Goodstadt, E. E. Winter, and C. P. Ponting, “Comparison of the genomes of human and mouse lays the foundation of genome zoology,” Hum. Mol. Genet. , vol. 12 , no. 7 , pp. 701–709, Apr. 2003. [221] N. G. Wreford, T. Rajendra Kumar, M. M. Matzuk, and D. M. de Kretser, “Analysis of the Testicular Phenotype of the Follicle-Stimulating Hormone β-Subunit Knockout and the Activin Type II Receptor Knockout Mice by Stereological Analysis,” Endocrinology, vol. 142, no. 7, pp. 2916–2920, Jul. 2001. [222] T. R. Kumar, Y. Wang, N. Lu, and M. M. Matzuk, “Follicle stimulating hormone is required for ovarian follicle maturation but not male fertility,” Nat Genet, vol. 15, no. 2, pp. 201–204, Feb. 1997.      146 [223] H. Johnston, P. J. Baker, M. Abel, H. M. Charlton, G. Jackson, L. Fleming, T. R. Kumar, and P. J. O’Shaughnessy, “Regulation of Sertoli Cell Number and Activity by Follicle-Stimulating Hormone and Androgen during Postnatal Development in the Mouse,” Endocrinology, vol. 145, no. 1, pp. 318–329, Jan. 2004. [224] E. T. Siegel, H.-G. Kim, H. K. Nishimoto, and L. C. Layman, “The Molecular Basis of Impaired Follicle-Stimulating Hormone Action: Evidence From Human Mutations and Mouse Models ,” Reprod. Sci. , vol. 20 , no. 3 , pp. 211–233, Mar. 2013. [225] L. Foppiani, S. Schlatt, M. Simoni, G. F. Weinbauer, U. Hacker-Klom, and E. Nieschlag, “Inhibin B is a more sensitive marker of spermatogenetic damage than FSH in the irradiated non-human primate model,” J. Endocrinol. , vol. 162 , no. 3 , pp. 393–400, Sep. 1999. [226] S. Schlatt, L. Foppiani, C. Rolf, G. F. Weinbauer, and E. Nieschlag, “Germ cell transplantation into X-irradiated monkey testes,” Hum. Reprod. , vol. 17 , no. 1 , pp. 55–62, Jan. 2002. [227] S. Ramaswamy and T. M. Plant, “Operation of the follicle-stimulating hormone (FSH)–inhibin B feedback loop in the control of primate spermatogenesis,” Mol. Cell. Endocrinol., vol. 180, no. 1–2, pp. 93–101, Jul. 2001. [228] G. R. Marshall, D. S. Zorub, and T. M. Plant, “Follicle-stimulating hormone amplifies the population of differentiated spermatogonia in the hypophysectomized testosterone-replaced adult rhesus monkey (Macaca mulatta).,” Endocrinology, vol. 136, no. 8, pp. 3504–3511, Aug. 1995. [229] M. M. A. VAN ALPHEN, H. J. G. VAN DE KANT, and D. G. DE ROOIJ, “Follicle-Stimulating Hormone Stimulates Spermatogenesis in the Adult Monkey,” Endocrinology, vol. 123, no. 3, pp. 1449–1455, Sep. 1988. [230] M. Phillip, J. E. Arbelle, Y. Segev, and R. Parvari, “Male Hypogonadism Due to a Mutation in the Gene for the β-Subunit of Follicle-Stimulating Hormone,” N. Engl. J. Med., vol. 338, no. 24, pp. 1729–1732, Jun. 1998. [231] G. Lindstedt, E. Nyström, and C. Matthews, “Deficiency in an Infertile Male due to FSHβ Gene Mutation. A Syndrome of Normal Puberty and Virilization but Under-developed Testicles with Azoospermia, Low FSH but High Lutropin and Normal Serum Testosterone Concentrations.,” Clin. Chem. Lab. Med., vol. 36, no. 8, pp. 663–665, 1998. [232] L. C. Layman, A. L. A. Porto, J. Xie, L. A. C. R. da Motta, L. D. C. da Motta, W. Weiser, and P. M. Sluss, “FSHβ Gene Mutations in a Female with Partial Breast Development and a Male Sibling with Normal Puberty and Azoospermia,” J. Clin. Endocrinol. Metab., vol. 87, no. 8, pp. 3702–3707, Aug. 2002. [233] K. Mansfield, “Marmoset Models Commonly Used in Biomedical Research,” Comp. Med., vol. 53, no. 4, pp. 383–392, 2003. [234] W. Rosner, D. J. Hryb, M. S. Khan, A. M. Nakhla, and N. A. Romas, “Sex hormone-binding globulin mediates steroid hormone signal transduction at the plasma membrane,” J. Steroid Biochem. Mol. Biol., vol. 69, no. 1–6, pp. 481–485, Apr. 1999. [235] W. Rosner, D. J. Hryb, S. M. Kahn, A. M. Nakhla, and N. A. Romas, “Interactions of sex hormone-binding globulin with target cells,” Mol. Cell. Endocrinol., vol. 316, no. 1, pp. 79–85, Mar. 2010.    147 [236] W. Rosner, D. J. Hryb, M. S. Khan, A. M. Nakhla, and N. A. Romas, “Androgen and estrogen signaling at the cell membrane via G-proteins and cyclic adenosine monophosphate,” Steroids, vol. 64, no. 1–2, pp. 100–106, Jan. 1999. [237] C. S. Porto, N. A. Musto, C. W. Bardin, and G. L. Gunsalus, “Binding of an extracellular steroid-binding globulin to membranes and soluble receptors from human breast cancer cells (MCF-7 cells).,” Endocrinology, vol. 130, no. 5, pp. 2931–2936, May 1992. [238] C. S. PORTO, G. L. GUNSALUS, C. W. BARDIN, D. M. PHILLIPS, and N. A. MUSTO, “Receptor-Mediated Endocytosis of an Extracellular Steroid-Binding Protein (TeBG) in MCF-7 Human Breast Cancer Cells,” Endocrinology, vol. 129, no. 1, pp. 436–445, Jul. 1991. [239] A. Hammes, T. K. Andreassen, R. Spoelgen, J. Raila, N. Hubner, H. Schulz, J. Metzger, F. J. Schweigert, P. B. Luppa, A. Nykjaer, and T. E. Willnow, “Role of Endocytosis in Cellular Uptake of Sex Steroids,” Cell, vol. 122, no. 5, pp. 751–762, May 2005. [240] K. M. Ng, M. G. Catalano, T. Pin??s, D. M. Selva, G. V. Avvakumov, F. Munell, and G. L. Hammond, “Evidence that fibulin family members contribute to the steroid-dependent extravascular sequestration of sex hormone-binding globulin,” J. Biol. Chem., vol. 281, no. 23, pp. 15853–15861, 2006. [241] T. T. Turner and M. S. Roddy, “Intraluminal androgen binding protein alters 3H-androgen uptake by rat epididymal tubules in vitro.,” Biol. Reprod. , vol. 43 , no. 3 , pp. 414–419, Sep. 1990. [242] L. O’Hara, M. Welsh, P. T. K. Saunders, and L. B. Smith, “Androgen Receptor Expression in the Caput Epididymal Epithelium Is Essential for Development of the Initial Segment and Epididymal Spermatozoa Transit,” Endocrinology, vol. 152, no. 2, pp. 718–729, Dec. 2011. [243] A. Krutskikh, K. De Gendt, V. Sharp, G. Verhoeven, M. Poutanen, I. Huhtaniemi, K. De Gendt, and V. Sharp, “Targeted inactivation of the androgen receptor gene in murine proximal epididymis causes epithelial hypotrophy and obstructive azoospermia.,” Endocrinology, vol. 152, no. 2, pp. 689–96, Feb. 2011. [244] D. O. N. W. FAWCETT and A. P. HOFFER, “Failure of Exogenous Androgen to Prevent Regression of the Initial Segments of the Rat Epididymis after Efferent Duct Ligation or Orchidectomy,” Biol. Reprod. , vol. 20 , no. 2 , pp. 162–181, Mar. 1979. [245] B. T. Hinton, Z. J. Lan, R. J. Lye, and J. C. Labus, “The Testis: From Stem Cell to Sperm Function,” E. Goldberg, Ed. New York, NY: Springer New York, 2000, pp. 163–173. [246] S. Carreau and R. a Hess, “Oestrogens and spermatogenesis.,” Philos. Trans. R. Soc. Lond. B. Biol. Sci., vol. 365, no. 1546, pp. 1517–1535, 2010. [247] E. M. Eddy, T. F. Washburn, D. O. Bunch, E. H. Goulding, B. C. Gladen, D. B. Lubahn, and K. S. Korach, “Targeted disruption of the estrogen receptor gene in male mice causes alteration of spermatogenesis and infertility.,” Endocrinology, vol. 137, no. 11, pp. 4796–4805, Nov. 1996. [248] D. Mahato, E. H. Goulding, K. S. Korach, and E. M. Eddy, “Spermatogenic Cells Do Not Require Estrogen Receptor-α for Development or Function,” Endocrinology, vol. 141, no. 3, p. 1273, Mar. 2000.      148 [249] D. Mahato, E. H. Goulding, K. S. Korach, and E. M. Eddy, “Estrogen receptor-α is required by the supporting somatic cells for spermatogenesis,” Mol. Cell. Endocrinol., vol. 178, no. 1–2, pp. 57–63, Jun. 2001. [250] A. D. Coviello, W. J. Bremner, A. M. Matsumoto, K. L. Herbst, J. K. Amory, B. D. Anawalt, X. Yan, T. R. Brown, W. W. Wright, B. R. Zirkin, and J. P. Jarow, “Intratesticular Testosterone Concentrations Comparable With Serum Levels Are Not Sufficient to Maintain Normal Sperm Production in Men Receiving a Hormonal Contraceptive Regimen,” J. Androl., vol. 25, no. 6, pp. 931–938, Nov. 2004. [251] J. P. JAROW and B. R. ZIRKIN, “The Androgen Microenvironment of the Human Testis and Hormonal Control of Spermatogenesis,” Ann. N. Y. Acad. Sci., vol. 1061, no. 1, pp. 208–220, Dec. 2005. [252] S. S. Carreau, “Testicular and blood steroid levels in aged men.,” Reprod. Biol., vol. 4, no. 3, pp. 299–304, Jan. 2004. [253] L. J. D. Zaneveld, C. J. De Jonge, R. A. Anderson, and S. R. Mack, “OPINION: Human sperm capacitation and the acrosome reaction ,” Hum. Reprod. , vol. 6 , no. 9 , pp. 1265–1274, Oct. 1991. [254] E. E. Baldi, “Human sperm activation during capacitation and acrosome reaction: role of calcium, protein phosphorylation and lipid remodelling pathways.,” Front. Biosci., vol. 1, pp. d189–205, Jan. 1996. [255] F. Urner and D. Sakkas, “Protein phosphorylation in mammalian spermatozoa,” Reprod. , vol. 125 , no. 1 , pp. 17–26, Jan. 2003. [256] N. Sebkova, M. Cerna, L. Ded, J. Peknicova, and K. Dvorakova-Hortova, “The slower the better: how sperm capacitation and acrosome reaction is modified in the presence of estrogens,” Reprod. , vol. 143 , no. 3 , pp. 297–307, Mar. 2012. [257] P. Vigil, A. Toro, and A. Godoy, “Physiological action of oestradiol on the acrosome reaction in human spermatozoa,” Andrologia, vol. 40, no. 3, pp. 146–151, Jun. 2008. [258] S. Solakidi, A.-M. G. Psarra, S. Nikolaropoulos, and C. E. Sekeris, “Estrogen receptors α and β (ERα and ERβ) and androgen receptor (AR) in human sperm: localization of ERβ and AR in mitochondria of the midpiece,” Hum. Reprod. , vol. 20 , no. 12 , pp. 3481–3487, Dec. 2005. [259] G. V. Avvakumov, Y. A. Muller, and G. L. Hammond, “Steroid-binding specificity of human sex hormone-binding globulin is influenced by occupancy of a zinc-binding site,” J. Biol. Chem., vol. 275, no. 34, pp. 25920–25925, 2000. [260] M. O. Suescun, S. Campo, M. A. Rivarola, F. González-echeverría, C. Scorticati, J. Ghirlanda, J. Tezón, J. A. Blaquier, and R. S. Calandra, “Testosterone, Dihydrotestosterone, and Zinc Concentrations in Human Testis and Epididymis,” Arch. Androl., vol. 7, no. 4, pp. 297–303, Jan. 1981. [261] T. SUZUKI, H. YAMANAKA, K. NAKAJIMA, K. KANATANI, K. SUZUKI, M. KIMURA, and N. OTAKI, “IMMUNOHISTOCHEMICAL DEMONSTRATION OF METALLOTHIONEIN IN HUMAN MALE EXCRETORY DUCTS OF THE TESTIS,” Acta Histochem. Cytochem., vol. 26, no. 2, pp. 85–92, 1993.     149 [262] V. Elgazar, V. Razanov, M. Stoltenberg, M. Hershfinkel, M. Huleihel, Y. B. Nitzan, E. Lunenfeld, I. Sekler, and W. F. Silverman, “Zinc-regulating Proteins, ZnT-1, and Metallothionein I/II Are Present in Different Cell Populations in the Mouse Testis,” J. Histochem. Cytochem. , vol. 53 , no. 7 , pp. 905–912, Jul. 2005. [263] S.-E. CHIA, C.-N. ONG, L.-H. CHUA, L.-M. HO, and S.-K. TAY, “Comparison of Zinc Concentrations in Blood and Seminal Plasma and the Various Sperm Parameters Between Fertile and Infertile Men,” J. Androl., vol. 21, no. 1, pp. 53–57, Jan. 2000. [264] D. A. Adamopoulos, N. Kapolla, A. Abrahamian, A. Dessypris, S. Nicopoulou, and G. Giannacodemos, “Sex steroids in cervical mucus of spontaneous or induced ovulatory cycles,” Steroids, vol. 65, no. 1, pp. 1–7, Jan. 2000. [265] A. Gerard, “Endocytosis of androgen-binding protein (ABP) by spermatogenic cells,” J. Steroid Biochem. Mol. Biol., vol. 53, no. 1–6, pp. 533–542, Jun. 1995. [266] H. Gerard, A. Gerard, A. En Nya, F. Felden, and J. L. Gueant, “Spermatogenic cells do internalize Sertoli androgen-binding protein: a transmission electron microscopy autoradiographic study in the rat.,” Endocrinology, vol. 134, no. 3, pp. 1515–1527, Mar. 1994. [267] R. A. Hess, D. Bunick, K.-H. Lee, J. Bahr, J. A. Taylor, K. S. Korach, and D. B. Lubahn, “A role for oestrogens in the male reproductive system,” Nature, vol. 390, no. 6659, pp. 509–512, Dec. 1997. [268] L.-J. Zhu, M. P. Hardy, I. V Inigo, I. Huhtaniemi, C. W. Bardin, and A. J. Moo-Young, “Effects of Androgen on Androgen Receptor Expression in Rat Testicular and Epididymal Cells: A Quantitative Immunohistochemical Study,” Biol. Reprod. , vol. 63 , no. 2 , pp. 368–376, Aug. 2000. [269] J.-S. Zhang, Q. Liu, Y.-M. Li, S. H. Hall, F. S. French, and Y.-L. Zhang, “Genome-wide profiling of segmental-regulated transcriptomes in human epididymis using oligo microarray,” Mol. Cell. Endocrinol., vol. 250, no. 1–2, pp. 169–177, May 2006. [270] G. V. Avvakumov, A. Cherkasov, Y. A. Muller, and G. L. Hammond, “Structural analyses of sex hormone-binding globulin reveal novel ligands and function,” Mol. Cell. Endocrinol., vol. 316, no. 1, pp. 13–23, 2010. [271] N. Arango and P. Donahoe, “Sex differentiation in mouse and man and subsequent development of the female reproductive organs.,” Harvard Stem Cell Institute, Cambridge (MA, 2008. [272] B. J. Schlomer, M. Feretti, E. Rodriguez Jr., S. Blaschko, G. Cunha, and L. Baskin, “Sexual Differentiation in the Male and Female Mouse from Days 0 to 21: A Detailed and Novel Morphometric Description,” J. Urol., vol. 190, no. 4, Supplement, pp. 1610–1617, Oct. 2013. [273] P. Pihlajamaa, F.-P. Zhang, L. Saarinen, L. Mikkonen, S. Hautaniemi, and O. A. Jänne, “The Phytoestrogen Genistein Is a Tissue-Specific Androgen Receptor Modulator,” Endocrinology, vol. 152, no. 11, pp. 4395–4405, Aug. 2011. [274] D. A. Dart, J. Waxman, E. O. Aboagye, and C. L. Bevan, “Visualising Androgen Receptor Activity in Male and Female Mice,” PLoS One, vol. 8, no. 8, p. e71694, Aug. 2013.      150 [275] S. G. Goodson, Z. Zhang, J. K. Tsuruta, W. Wang, and D. A. O’Brien, “Classification of Mouse Sperm Motility Patterns Using an Automated Multiclass Support Vector Machines Model,” Biol. Reprod., vol. 84, no. 6, pp. 1207–1215, Jun. 2011. [276] P. Calvel, A. D. Rolland, B. Jégou, and C. Pineau, “Testicular postgenomics: targeting the regulation of spermatogenesis,” Philos. Trans. R. Soc. London B Biol. Sci., vol. 365, no. 1546, pp. 1481–1500, Apr. 2010. [277] B. J. DANZO, C. A. TAYLOR, and W. N. SCHMIDT, “Binding of the Photoaffinity Ligand 17β-Hydroxy-4,6- Androstadien-3-One to Rat Androgen-Binding Protein: Comparison with the Binding of 17β-Hydroxy-5α- Androstan-3-One,” Endocrinology, vol. 107, no. 4, pp. 1169–1175, Oct. 1980. [278] W. N. SCHMIDT, C. A. TAYLOR, and B. J. DANZO, “The Use of a Photoaffinity Ligand to Compare Androgen-Binding Protein (ABP) Present in Rat Sertoli Cell Culture Media with ABP Present in Epididymal Cytosol,” Endocrinology, vol. 108, no. 3, pp. 786–794, Mar. 1981. [279] J. A. MAHOUDEAU and P. CORVOL, “Rabbit Testosterone-Binding Globulin. I. Physico-Chemical Properties,” Endocrinology, vol. 92, no. 4, pp. 1113–1119, Apr. 1973. [280] J. Y. Dubé, R. R. Tremblay, F. T. Dionne, and P. Chapdelaine, “Binding of androgens in dog prostate cytosol and in plasma,” J. Steroid Biochem., vol. 10, no. 4, pp. 449–458, 1979. [281] B. Jégou, J. L. Dacheux, D. H. Garnier, M. Terqui, G. Colas, and M. Courot, “Biochemical and physiological studies of androgen-binding protein in the reproductive tract of the ram,” J. Reprod. Fertil. , vol. 57 , no. 2 , pp. 311–318, Nov. 1979. [282] J. M. Renoir, “Hormonal and immunological aspects of sex steroid-binding plasma protein of primates.,” J. Reprod. Fertil. Suppl., vol. Suppl 28, pp. 113–119, 1980. [283] J. P. Karr, R. Y. Kirdani, G. P. Murphy, and A. A. Sandberg, “Sex Hormone Binding Globulin and Transcortin in Human and Baboon Males,” Arch. Androl., vol. 1, no. 2, pp. 123–129, Jan. 1978.  


Citation Scheme:


Citations by CSL (citeproc-js)

Usage Statistics



Customize your widget with the following options, then copy and paste the code below into the HTML of your page to embed this item in your website.
                            <div id="ubcOpenCollectionsWidgetDisplay">
                            <script id="ubcOpenCollectionsWidget"
                            async >
IIIF logo Our image viewer uses the IIIF 2.0 standard. To load this item in other compatible viewers, use this url:


Related Items