University of Otago

RESEARCH REPORT

The Role of Lhx9 in the Male Reproductive

System

Lin Song

A thesis submitted in partial fulfilment of the Degree of Bachelor of Biomedical Science with Honours. University of Otago, Dunedin, New Zealand October 2020 Abstract

Leydig cells (LCs) are found within the interstitium of the testes. LCs are crucial to male reproductive function due to their role in testosterone synthesis, a process vital for both organogenesis and secondary sex characteristic development. The generation of LCs differs between foetal and adult testes. During embryonic development, LCs differentiate from progenitors, their numbers peaking at embryonic day 15.5 (E15.5), then degenerating at birth. Postnatally, the stem LC (SLC) population is activated at puberty, providing the precursor pool where the adult LCs (ALCs) will later differentiate from. The exact mechanisms of this process are not well understood. The LIM homeobox 9 gene, Lhx9, is essential for the formation of the gonad, it is also expressed later in development in the testes interstitium in both foetal and adult LCs. The functional role of Lhx9 associated with this expression has been poorly characterised. We hypothesise that the Lhx9 is vital for forming the progenitor population from which LCs later differentiate. The effect of Lhx9 loss on the foetal or adult LC pools has not been studied.

Gene expression was compared between Lhx9+/+ (WT) and Lhx9+/- (HET) mouse testes at two stages: E15.5 (using RT-qPCR), and adult (RNA-Sequencing). In the adult testes, 51 genes were found to be differentially expressed between the WT and HET mice. Additionally, gene ontology analysis revealed enrichment for genes involved in lipid metabolism. This pathway is utilised by LCs for testosterone production. At E15.5, expression of marker genes for progenitor stem cell populations, such as Nestin and Notch, was significantly reduced in the HET testes compared to the WT littermates. Conversely, steroidogenesis genes showed a significant increase in expression; supporting the hypothesis that Lhx9 acts to maintain the LC progenitor population in the testes.

Treatment of testes with ethane dimethanesulfonate (EDS) in vivo causes the death of Leydig cells, allowing analysis of their regenerative capability. Preliminary results of qPCR analysis following EDS treatment in adult HET and WT testes shows altered expression of a few genes, such as a decrease in Nestin in EDS treated HET mice. Together, these results will not only provide further understanding of the role of Lhx9 in foetal and adult testis, but additionally help illustrate delicate balance of progenitor populations and their impact on secondary sex determination and fertility in the adult.

ii Acknowledgements

To begin, I would like to personally thank my supervisor Dr. Megan Wilson. Thank you for not only providing me with this opportunity, but for your constant patience, willingness to teach, and authentic passion and excitement which has motivated me through the year. You work tirelessly and your dedication to the field is inspiring. I am so fortunate to have been under your supervision.

To Stephanie Workman of the Wilson Lab, thank you for being my mentor and immediate support. You have never failed to go above and beyond to help me. The encouragement and friendship that you offer on a daily basis is something I am extremely grateful for and it has not gone unnoticed. It would not have been half the wonderful year I have had without you.

Thank you to the other Wilson Lab members Jeremy, Rebecca, Beri, Ed, Michael, and lab technician Sylvia for welcoming me into the lab and unhesitatingly offering to help whenever I needed. You have made being part of the lab an unforgettable experience. To the office members Rebecca, Rosie, and Jonika, thank you for the constant entertainment and providing comfort through mutual understanding during stressful events. I would also like to thank my partner Ricky, though it is not widely acknowledged, it is unquestionable that I would not have completed this year without your persistent thoughtfulness and care.

Finally, I would like to thank my family for the everlasting support through the duration of my studies, it goes without saying, your unconditional love and concern for me is irreplaceable.

I hope I have made you all proud.

iii Table of Contents

ABSTRACT ...... II

ACKNOWLEDGEMENTS ...... III

TABLE OF CONTENTS ...... IV

LIST OF FIGURES ...... VI

LIST OF TABLES ...... VIII

LIST OF ABBREVIATIONS...... IX

1. INTRODUCTION ...... 1

1.1. UROGENITAL RIDGE DEVELOPMENT ...... 1 1.2. MALE SEX DETERMINATION AND TESTICULAR DEVELOPMENT ...... 2 Anatomical Development ...... 2 Key Genetic Influences ...... 2 1.3. LEYDIG CELLS ...... 4 Development and Function from Embryo to Adult ...... 5 Role in Infertility and Loss of Function Phenotypes ...... 6 Regeneration of Leydig cells ...... 8 1.4. THE LHX9 GENE ...... 9 Proposed Role of Lhx9 in Leydig cells ...... 9 1.5. PROJECT OVERVIEW...... 11

2. METHODS ...... 13

2.1. ANIMAL HUSBANDRY ...... 13 2.2. GENOTYPING ...... 13 DNA Isolation...... 13 Polymerase Chain Reaction (PCR) ...... 14 Gel Electrophoresis ...... 14 2.3. INHIBITOR TREATMENT OF MALE MICE ...... 15 2.4. TISSUE COLLECTION ...... 16 Serum Collection ...... 16 2.5. RNA EXTRACTION ...... 16 RNA Extraction: Smaller Tissues ...... 16 RNA Extraction: Larger Tissues ...... 17 2.6. CDNA SYNTHESIS ...... 17 2.7. REAL-TIME QUANTITATIVE PCR ...... 17 Primer Design & Efficiency Analysis ...... 18 Reaction Set up ...... 18 Data Analysis ...... 19

iv 2.8. HISTOLOGY ...... 19 2.9. IMMUNOHISTOCHEMISTRY...... 19 Antigen Retrieval ...... 19 Primary Antibody Binding ...... 20 Secondary Antibody Binding & Colour Development ...... 20 2.10. RNA SEQUENCING ...... 21 2.11. STATISTICAL ANALYSIS ...... 21

3. RESULTS ...... 22

3.1. LHX9 MRNA EXPRESSION DURING LEYDIG CELL DIFFERENTIATION ...... 22 Selection of Target Genes ...... 22 Primer Design Efficiencies ...... 23 Leydig cell Marker Gene Expression changes in the Perinatal Period ...... 24 3.2. LEYDIG CELL MARKER GENE EXPRESSION IN E15.5 LHX9+/- TESTIS ...... 26 LHX9 Protein Expression in Adult Testis ...... 29 3.3. REGENERATIVE ABILITY OF LHX9+/- MICE ...... 30 Marker Protein Expression in EDS Treated Testis ...... 34 3.4. RNA SEQUENCING DATA ...... 37 Quality Check ...... 39 Analysis ...... 41 Validation ...... 44

4. DISCUSSION ...... 46

4.1. EXPRESSION OF KEY GENES DURING LEYDIG CELL DEVELOPMENT...... 47 4.2. IMPACT OF LHX9+/- HAPLOINSUFFICENCY ...... 48 E15.5 Samples ...... 48 4.3. EDS INJECTION STUDY ...... 50 Cell Characteristics following EDS Treatment...... 51 4.4. RNA SEQUENCING ...... 53 4.5. SIGNIFICANCE ...... 54 4.6. LIMITATIONS ...... 55 4.7. FUTURE DIRECTION ...... 56 4.8. CONCLUSIONS...... 57

REFERENCES ...... 58

APPENDIX 1: BUFFERS AND SOLUTIONS...... 70

APPENDIX 2: RT-QPCR PRIMER SEQUENCES ...... 72

APPENDIX 3: PRIMER EFFICIENCY TESTS ...... 73

APPENDIX 4: RT-QPCR DISSOCIATION CURVES...... 75

APPENDIX 5: DEG LIST FROM RNA-SEQ ANALYSIS...... 77

v List of Figures

Figure 1 – Cellular characteristics and chronological evolution of LPCs to ALCs...... 6

Figure 2 – Schematic of potential pericyte differentiation and differentiation process from LPC to ALC...... 8

Figure 3 – Example Lhx9 genotyping results via gel electrophoresis...... 15

Figure 4 – Representation of RT-qPCR efficiency measures...... 23

Figure 5 – Lhx9 expression in WT testes at varying developmental timepoints...... 24

Figure 6 –Relative expression timeline of key testicular marker genes at varying developmental timepoints in WT testes ...... 25

Figure 7 – Relative gene expression of different LPC marker genes between WT and HET

E15.5 testes...... 27

Figure 8 – Relative gene expression of different steroidogenic enzyme markers in WT vs HET.

...... 27

Figure 9 – Relative expression of LC marker genes in E15.5 WT and HET male testes...... 28

Figure 10 – Schematic of physical arrangement of cells in adult testis...... 29

Figure 11 – LHX9 immunohistochemistry on WT adult (4 months) testis...... 30

Figure 12 – Relative gene expression of marker genes in day 14 preliminary EDS study treatment groups...... 32

Figure 13 – Mean seminal vesicle weight of mice from different treatment groups...... 33

Figure 14 – H&E stain of day 14 testes morphology in different treatment groups from EDS study...... 34

Figure 15 – NESTIN specific IHC on day 14 testes from EDS study...... 35

Figure 16 – NF-H specific IHC on day 14 testes from EDS study...... 36

Figure 17 – Example Bioanalyser results...... 38

Figure 18 – Pipeline figure of steps involved with processing RNA sequencing output...... 39

vi Figure 19 – Example of FastQC quality check results for sample T50...... 40

Figure 20 – RNA-seq analysis for differential gene expression...... 41

Figure 21 – Mapped reads in relation to Rn7sk and Fgf1 genes in UCSC genome browser. . 43

Figure 22 – Results of relative gene expression of Fgf1 to validate RNA sequencing data. ... 44

Figure 23 – Relative gene expression of Fgf1 in preliminary day 14 EDS study samples...... 45

Figure 24 – Basic steroidogenic pathway...... 49

vii List of Tables

Table 1: The function of key genes of early male gonadal pathway development ...... 3

Table 2: Stock and concentration of Lhx9 Genotyping PCR protocol ...... 14

Table 3: Thermal profile for Lhx9 genotyping ...... 14

Table 4: Thermal profile for RT-qPCR reaction...... 19

Table 5: Immunohistochemistry Antibody Dilutions ...... 20

Table 6: Target genes and their function in Leydig cells ...... 22

Table 7: Target gene primer efficiencies ...... 23

Table 8: RNA sample nanodrop results and sequencing RIN results for samples sent for RNA sequencing...... 37

Table 9: Table of mapped reads from sequencing output generated by BGI...... 39

Table 10: Top 4 genes with reduced expression in HET samples compared to WT, sorted by

FC in expression ...... 42

Table 11: Top 4 genes with increased expression in HET samples compared to WT, sorted by

FC in expression ...... 42

Table 12: Results from DAVID overrepresentation test...... 43

Table 13: Forward and reverse RT-qPCR primer sequences...... 72

Table 14: DEGs higher expressed in HET samples (FDR < 0.1) ...... 77

Table 15: DEGs lower expressed in HET samples (FDR < 0.1) ...... 78

viii List of Abbreviations

 micro

Actb actin beta

Amh Anti-Mullerian hormone

Arx aristaless

BLAST Basic Local Alignment Search Tool bp base pairs cDNA complimentary DNA

Ct cycle threshold ddH2O double distilled H2O

DEG differentially expressed gene

Dhh desert hedgehog

DMSO dimethyl sulfoxide

DNA deoxyribonucleic acid dNTP deoxyribonucleotide triphosphate

DSD disorder of sex development

E embryonic day

EDS ethane dimethanesulfonate

EDTA ethylenediaminetetraacetic acid

FDR false discovery rate

Fgf1 fibroblast growth factor-1

FLC foetal Leydig cell g times gravity hr hour

HET heterozygous

ix HSD17 17 - hydroxysteroid dehydrogenase

IHC immunohistochemistry kb kilobase

L litres

LC Leydig cell

Lhx9/2 LIM Homeobox 9/2

Lifr leukaemic inhibitory factor receptor subunit-A

M moles per litre min minutes mRNA micro RNA

NF-H neurofilament heavy peptide ng nanogram

P postnatal day

P450c17 cytochrome P450 17A1

P450scc cholesterol side-chain cleavage enzyme

PBS phosphate buffer saline

PBTx PBS with triton-X

PC pericyte

PCR polymerase chain reaction

Pdgfra platelet derived growth factor receptor subunit-A

PFA paraformaldehyde

Ptch1 Protein patched homolog 1

PTMC peritubular myoid cell

RNA ribonucleic acid

RNase ribonuclease

x Rps29 ribosomal protein S29

RT-qPCR real-time quantitative PCR s seconds

Sf1 steroidogenic factor 1

SLC stem Leydig cell

Sox9 SRY-box transcription factor 9

SRY sex determining region of the Y chromosome

ST seminiferous tubules

SV seminal vesicle

TAE tris-acetate EDTA

UGR urogenital ridge

Vcam vascular cell adhesion molecule 1

WT wild-type

xi 1. Introduction

Reproduction is key to the survival of all species, sexual reproduction in particular allows for genetic variation within a species and subsequently an increased chance of species survival over time. For animals, reproduction is a biologically instinctive process which seldom fails – this misconception commonly transfers across to our current understanding of human reproduction. An estimated 40-50% of all cases of infertility in humans are expected to be a result of “male infertility”, whereby one or more factors, whether lifestyle or genetic, that contribute to the reproductive health of the male result in unsuccessful conception (Durairajanayagam, 2018; Ferlin et al., 2006). Factors can include insufficient sperm, abnormal sperm morphology, or poor sperm motility (Kumar & Singh, 2015). The effects of infertility can have tremendously negative social and mental implications on individuals. Attention towards male infertility can be focused too intently on recently impaired adult function, neglecting potential issues which may arise earlier in embryonic development. The development of the reproductive system is complex both anatomically and genetically, abnormalities which may arise in the embryonic cell populations can have large impacts on adult fertility. This poorly explored area highlights the need for more focused research.

1.1. Urogenital Ridge Development The urogenital ridge (UGR) is the term assigned to the precursor tissue of the kidneys and gonads. It is derived from the intermediate mesoderm, one of the three primary germ layers in the developing embryo that form after gastrulation (Armstrong et al., 1993). The initial stages of genital ridge development are the same for both sexes. The UGR proliferates during early development at approximately embryonic day 9.5 (E9.5) in the mouse to form two bilateral thickenings of epithelial tissue on the posterior wall of the coelomic cavity (Armstrong et al., 1993; Staack et al., 2003). During this period of development, the UGR is being colonised by primordial germ cells (PGCs), the precursors of sperm and oocytes. The PGCs migrate along the hindgut and reach the final destination of the gonad at E10.5 in the mouse (Molyneaux et al., 2001; Rey et al., 2016). The UGR is bipotential at this time, maintaining the ability to differentiate into one of two fates dependant on the genetic influences of the combination of the X and Y sex determining chromosomes.

1 1.2. Male Sex Determination and Testicular Development Anatomical Development Structural changes arise during testicular development with an evolving gene expression profile. Key genes involved are outlined in Table 1. As Sertoli cells develop, anatomically, the seminiferous tubule structure of the testis forms through the process of cellular partitioning. Here, the endothelial cells originating from the surrounding vascular plexus of the mesonephros create a physical barrier between the germ cells and Sertoli cells i.e. defining the interstitium and tubules (Defalco et al., 2014). As the network of tubules begin to expand, an extracellular matrix is deposited by the Sertoli cells into the interstitium, and the physical barrier becomes a layer of peritubular myoid cells (Das et al., 2011; Hadley et al., 1985).

Leydig cells (LCs) begin to form at around E11.5-12.5 in the embryonic testes, secreting testosterone, the hormone critical for the maintenance of the Wolffian duct and mesonephric tubules. (Murashima et al., 2011). This enables structural evolution in the testis, more specifically, the development of the epididymis, vas deferens, and seminal vesicles, eventually contributing to a fully developed testis (Pointis et al., 1980; Rey et al., 2016; Yao et al., 2002).

Key Genetic Influences Gonadal sex determination is triggered by activation of the sex determining region of the Y chromosome (SRY). This occurs in the UGR at E10-10.5 in mice, and at approximately 6 weeks gestation in humans (Kim & Capel, 2006). This activation is localised within the UGR in the Sertoli cell precursors (Bullejos & Koopman, 2001). When SRY is expressed, the bipotential gonad is influenced to follow a testis-specific genetic cascade, becoming morphologically distinguishable from female gonads by E13 (Kim & Capel, 2006; Staack et al., 2003). In mice, SRY must act within a small developmental window, as it is only transiently expressed in the Sertoli cell precursors. This differs from human expression where SRY is constitutively expressed in the Sertoli cells from sex determination through to adulthood (Sinclair et al., 1990). When SRY expression in the male is impaired or insufficient, the developmental pathway of the testes is heavily impacted, as is the downstream development of secondary male sex characteristics and reproductive structures (Goodfellow & Lovell-Badge, 1993). For example, in the presence of SRY, Sertoli cells begin to express anti-Mullerian hormone (AMH), inducing regression of Mullerian ducts which would normally progress into female reproductive organs (Cool et al., 2012).

2 Following the expression of SRY, the target gene SRY-box transcription factor 9 (Sox9) is upregulated in Sertoli cell precursors. Sox9 in turn upregulates other genes such as desert hedgehog (Dhh) which acts to promote further differentiation of the interstitial cells (Bitgood & Mcmahon, 1995; Jeske et al., 1995; Kashimada & Koopman, 2010; Park et al., 2011).

Dhh drives the formation of the peritubular myoid cells (PTMCs), the formation of the physical barrier between interstitium and seminiferous tubule, and LC differentiation (Clark et al., 2000; O’Hara & Smith, 2015; Yao et al., 2002). Dhh knockout mouse models demonstrate the reduction in proliferation and development of both PTMCs and androgen producing Leydig cells (LCs). Similarly, human mutations in DHH result in an XY sex reversal. This illustrates the sensitivity of the complex genetic developmental network and the importance of correctly functioning gonads in reproductive development (O’Hara & Smith, 2015).

SOX9 maintains its own expression by activating fibroblast growth factor 9 (Fgf9) (Lai et al., 2016). This interaction between Fgf9 and Sox9 is therefore essential for ensuring proper function and development of the testis (Arango et al., 1999; Modi et al., 2006).

Table 1: The function of key genes of early male gonadal pathway development

Gene Function in Embryogenesis Function in Testis Development SRY Triggers the testis developmental Presence of SRY enables cascade (Sinclair et al., 1990). differentiation of Sertoli cells from precursors, overriding the ovary granulosa cell fate (Kashimada & Koopman, 2010). Sox9 Sox9 is expressed in sites of Expression in Sertoli cells promotes chondrogenesis, central nervous the male developmental pathway system, and other tissues including the through repression of the ovarian genital ridge. It is required to initiate pathway. (Arango et al., 1999; AMH transcription (Cameron & Chaboissier et al., 2004; Kent et al., Sinclair, 1997; Kent et al., 1996; 1996) Wright et al., 1995).

Sf1 Contributes to formation of Essential for the maturation, steroidogenic gonad and adrenal development of Sertoli cells, and tissues, also expressed in pituitary and spermatogenesis (Kato et al., 2012). hypothalamus (Ikeda et al., 1994)

3 Gene Function in Embryogenesis Function in Testis Development AMH Factor causing the regression of Expression in Sertoli cells required Mullerian ducts in males (XY), for development of both Sertoli cells preventing female (XX) development, and LCs, needed for and enabling growth of functional spermatogenesis (Lee et al., 1996; testicular tissue (Josso et al., 1993; Lee Munsterberg & Lovell-Badge, 1991; et al., 1996). Racine et al., 1998).

Fgf9 Critical for mesenchymal proliferation, Enables the development and regulates the Wnt signaling pathway in proliferation of PTMCs, maintains epithelium, involved in patterning of Sox9 expression thus contributing to embryonic axis (Del Moral et al., 2006; sex determination. Upregulates male Geske et al., 2008). germ cell related markers (Bowles et al., 2010; Park et al., 2011).

Lhx9 Activation of Sf1 and regulation of Involved in gonadogenesis and neural tissue differentiation and limb expressed later in LCs (Mazaud et patterning (Rétaux et al., 1999; al., 2002; Rétaux et al., 1999). Wilhelm & Englert, 2002). KO mouse model shows complete lack of gonadal proliferation (Birk et al., 2000).

Nestin Neural stem cell and proliferating cell Marker of SLCs (Davidoff et al., marker (Jiang et al., 2014). 2004; Jiang et al., 2014). Notch Essential for the morphogenesis of Involved in maintaining a LPC state; vasculature, liver development, and plays a role in spermatogenesis impacts cell ability to react to (Hasegawa et al., 2012; Murta et al., instruction thus cell fate (Krebs et al., 2013) 2000; Weinmaster, 1997; Zong et al., 2009).

Dhh Required for mammalian Drives PTMC and LC development. spermatogenesis, earlies indicator of Contributes to the formation of the male sex determination (Bitgood et al., barrier between the interstitium and 1996). the seminiferous tubules (Clark et al., 2000; O’Hara & Smith, 2015)

1.3. Leydig Cells As the primary androgen source, the correct development of LCs is imperative to the functionality of the male reproductive system. Testosterone is vital to many biological processes. In the reproductive process of spermatogenesis alone, there are a multitude of events which are reliant on testosterone. These include the formation of the seminiferous tubule lumen, maintenance of both spermatogonia population and the blood-testis barrier, spermatocyte meiosis completion, and release of mature sperm (Huhtaniemi & Pelliniemi, 1992).

4 Development and Function from Embryo to Adult Foetal Leydig cells (FLCs) and adult Leydig cells (ALCs) are the two key Leydig populations (Mendis-Handagama & Ariyaratne, 2001). The two populations differ in not only morphology, but also differ functionally with regards to types and volumes of testosterone produced to support spermatogenesis and the developmental needs at that point in time (Ge et al., 2006; Inoue et al., 2016). Both populations arise from a stem LC population which is marked by Nestin (Davidoff et al., 2004; Ge et al., 2006).

FLCs appear flat and oval, and contain many lipid droplets, whereas ALCs appear larger and more dense. FLCs develop prenatally and produce androstenedione, an androgen crucial for the differentiation of male genitalia (Habert et al., 2001). Androstenedione is converted into testosterone within Sertoli cells through 17β- Hydroxysteroid dehydrogenase 1 (HSD171) activity. HSD17 acts through catalysing the conversion between a precursor of androgens to testosterone and estradiol, achieved through oxidation of the carbon 17 position in the pre- converted steroids (He et al., 2016; Huhtaniemi & Pelliniemi, 1992; Kraft et al., 2005).

ALCs are found in the developed testes and produce higher amounts of testosterone which functions to drive spermatogenesis (Benton et al., 1995). Other cellular characteristics noted in Figure 1 illustrate the three stage developmental model which ALCs go through to reach a mature state. The chronological development to reach a mature ALC progresses through Leydig progenitor cell (LPC), immature LC, and finally ALC (Yao et al., 2002). FLCs comparatively do not go through as many developmental stages, as they differentiate directly from the LPC, they have low mitotic activity as a result of directly developing from the stem cell population (Hobert & Westphal, 2000).

There are several points of controversy as to the origin of the foetal and adult Leydig cells. It has been hypothesised by some that they develop from the same progenitor cell, and by others that they differentiate from different progenitor cell populations (Barsoum et al., 2020; DeFalco et al., 2011). A third theory discovered through lineage tracing experiments suggests the dedifferentiation of FLCs to later contribute to the ALC population (Shima & Morohashi, 2017).

5

Cellular Characteristic Progenitor LC Immature LC Adult LC Mitotic Activity High Divides once Likely none Morphology Spindle shape Round containing large Round containing lipid droplets small lipid droplets Testosterone metabolism High Highest Low Testosterone synthesis Low Intermediate High Age where cell is present 21 days 35 days 90 days Androgen receptor High High Low

Figure 1 – Cellular characteristics and chronological evolution of LPCs to ALCs.

The chronological cellular and morphological development between LPC and matured ALC. Differences in testosterone-related cellular characteristics between the developmental stages are listed. Abbreviations: LC = Leydig cell, LPC = Leydig progenitor cell, ALC = adult Leydig cell. Adapted from (Chen et al., 2009). Created with BioRender.com.

Understanding the timeline of reproductive system development is critical to gaining insight into any abnormalities that may arise in the adult. More specifically, understanding mechanisms of the primary steroidogenic cells allows for interpretation of malfunction in the male reproductive system as they drive a majority of the sexual dimorphism.

Role in Infertility and Loss of Function Phenotypes The World Health Organisation defines infertility as the inability of a sexually active couple to conceive a pregnancy in one year, given they are not on contraception (Vander Borght & Wyns, 2018). The impacts of infertility and disorders of sex development (DSDs) can extend beyond physical experiences. Emotionally, issues begin to fester leading to mental stresses. This is further accentuated in the social traditionalism of bearing your own biological children (Greil et al., 2010). Though the reason behind many cases of infertility are undefined, the ones which are understood have a genetic basis which contributes to the complexity of the disease (Ferlin et al., 2007).

6 The effects of atypical development and function of the reproductive system, namely the LCs can seem small physically, but play a large role in development of secondary sex characteristics and consequently, fertility. The dysfunction of LCs can be categorised as a DSD (Rey & Grinspon, 2011).

DSDs are a broad variety of conditions which mainly present at birth/adolescence and affect either process of sex determination or sex differentiation (Ahmed et al., 2011; Biason-Lauber, 2010). DSDs can be further broken down into two general categories, dysgenic and non- dysgenic DSDs.

Dysgenic DSDs are conditions with defects in the process of differentiation of the testis in embryonic development. These can be perpetuated through generations, leading to early onset hypogonadism. In dysgenic DSDs, the function of the whole testicular system is compromised.

Comparatively, non-dysgenic DSDs involve early onset of hypogonadism which can be syndromic, a result of malfunction in the steroidogenic cells such as Leydig or Sertoli cells, or general male hormone organ defects (Ahmed et al., 2011; Warne & Hewitt, 2009; White et al., 2011).

A dysfunction of LCs specifically has the potential to lead to a range of male reproductive illnesses, these include, but are not limited to primary hypogonadism, cryptorchidism (undescended testicle), and hypospadias (misplacement of urethra in male genitalia). Hypogonadism affects approximately 38% of men in the US (Mulligan et al., 2008), being one of the DSDs to affect a large range of individuals.

Hypogonadism is the decrease in functionality of the gonads, thus leading to a decrease in production in hormones. Male hypogonadism is characterised by a reduction in testosterone levels as a result of abnormalities in the testes, hypothalamus, or pituitary gland (Kumar et al., 2010; Mulligan et al., 2008). This deficiency in testosterone results in a failure to produce expected amounts of sperm, absence of secondary sex characteristics, infertility, and can impact cognitive function and development (Kumar et al., 2010). There is little known about the link between the development of LPCs and male infertility.

7 Regeneration of Leydig cells Figure 2 illustrates how LCs regenerate through differentiating from the Nestin-positive LPC/stem cell population to the ALCs (Kumar & DeFalco, 2018). The stem cells/Leydig progenitor cells (LPCs) are capable of differentiating into a number of different cell types, these include adipocytes, peritubular myoid cells, and smooth muscle cells. These cells are all present in the testes (Chen et al., 2019, 2020; Curley et al., 2019). Additional LPC markers include Notch and Neurofilament heavy peptide (NF-H) (Davidoff et al., 2009).

Figure 2 – Schematic of potential pericyte differentiation and differentiation process from LPC to ALC.

Pericytes are believed to be the source of stem Leydig cells/Leydig progenitor cells (LPCs). Pericytes express Nestin, Notch, and NF-H, and are able to differentiate into four main cell types, all of which can be found in the testis. Pericytes differentiate into ALCs through an intermediate immature LC stage, where further expression of steroidogenic enzymes 3HSD and P450scc/c17 enables maturation into ALCs. Abbreviations: LPC = Leydig progenitor cell, ALC = adult Leydig cell, NF-H = Neurofilament heavy peptide. (MJ. Wilson, Original work 2020).

LCs can be directly eliminated through a LC-specific toxicant ethane dimethanesulfonate (EDS). EDS is administered through an intraperitoneal injection, once administered (day 0), the LCs will begin to die and start regeneration at approximately day 4 post injection. It is expected they will be near fully regenerated at day 14 (Davidoff et al., 2004; Teerds & Rijntjes, 2007). Most EDS regeneration studies have been conducted in rats, recent studies have shown that treatment with EDS also causes partial loss of LCs in mice and subsequent regeneration (Jiang et al., 2014; Yang et al., 2017).

8 Based off this knowledge about the differentiative process of LPCs to ALCs, and the LC- specific toxicant which enables specific elimination of LCs, we can create a model for studying different aspects of LC regeneration in vivo.

1.4. The Lhx9 gene Our research focuses on the role of the Lhx9 gene in embryonic and adult reproductive development. Lhx9 belongs to the LIM Homeobox domain transcription factor family. The LHX9 protein is a transcription factor which is associated with many processes in embryogenesis. These include the regulation of cell proliferation in the central nervous system and hindlimb, development of the pineal gland, and retinal growth (Atkinson-Leadbeater et al., 2009; Bertuzzi et al., 1999; Rétaux et al., 1999; Yamazaki et al., 2015). Despite these extensive developmental roles, research conducted by Birk et al (2000) showed Lhx9 to only be essential in the process of gonadogenesis. Through using an Lhx9 knockout (Lhx9-/-) mouse, it was observed that development in the limbs and central nervous system progressed normally, where the only notable defect found was gonadal agenesis. This suggested a functional redundancy in the limb/CNS function, likely accredited to gene Lhx2, the closest genetic relative to Lhx9 (Ottolenghi et al., 2001; Rétaux et al., 1999).

Lhx9 is conserved across a number of species for similar functions such pallial development in tetrapods, and retinal growth in Xenopus laevis and chicken species, there is also a murine orthologue. Reproductively, orthologues can be found in species such as chicken (Gallus gallus domesticus), the American cockerspaniel, and zebrafish (Danio rerio) (Atkinson-Leadbeater et al., 2009; Mazaud et al., 2002; Moreno et al., 2004; Pujar et al., 2005). The research which has contributed to the knowledge of the Lhx9 across several species is indicative of the importance of the role it plays in reproductive development, the high level of conservation highlights incentive to investigate it in humans.

Proposed Role of Lhx9 in Leydig cells The experimental data using in situ hybridisation by Mazaud et al (2002) shows the expression of Lhx9 is present in undifferentiated gonads. The AMH-positive epithelial cells of the undifferentiated gonads progress into Sertoli cells within the seminiferous tubules (Mazaud et al., 2002). Comparatively, Lhx9-positive cells are maintained in the interstitial mesothelial cells of the differentiated testes, these develop into LCs (Schmahl et al., 2000).

9 Postnatally, Lhx9 expression progressively decreases to become near undetectable in rat and chick testis as the majority of the LC differentiation is completed. The expression of Lhx9 in the reproductive system is restricted to the suggested LPCs (DeFalco et al., 2011). This illustrates the inverse relationship between the expression of Lhx9 and the differentiation of cells in the gonads (Hamada et al., 1999). It can therefore be hypothesised that a reduced expression of Lhx9 will result in a reduced pool of LPCs.

As a contributor to a complex network of interactions, it is likely there are a number of factors which may lead to regulation of Lhx9. An important regulatory factor of Lhx9 is the transmembrane receptor Notch (Tang et al., 2008). Notch signalling is involved with the differentiation and maintenance of the LPC pool. The expression of Notch and downstream gene targets, coincides with the development and differentiation of embryonic gonad somatic cells (Tang et al., 2008). Expression of Lhx9 also increases in the presence of Notch signalling inhibitors (Mazaud et al., 2002). This provides supporting evidence to the data reviewed by Martin, L. (2016), exhibiting an increase in LC differentiation in XY gonads fostered from an inhibition of Notch signalling (Martin, 2016).

Correlating with the reviewed data, the constitutive signalling of Notch leads to a decrease in LC population suggesting Notch is responsible for maintaining the LPC population, and prevents differentiation into the mature ALC pathway. Existing LCs and Sertoli cell populations are not impacted by alterations in Notch signalling patterns (Tang et al., 2008). From these observations made, Notch appears to have a “stabilising” effect on the progenitor cell state by demoting differentiation into LCs. When absent, the cells begin to commit down LC line fate. A decrease in Notch therefore results in a decrease in Lhx9 and consequently a reduced LC population.

10 1.5. Project Overview Understanding the developmental processes of LCs, including their regeneration, is important to the study of reproductive health. Aging LCs are likely to be a source of declining fertility because of an age-associated net decline of the LC population, causing an overall decrease in testosterone. Many complex fertility issues can arise as a result of testosterone deficiencies such as hypogonadism and dysfunctional spermatogenesis. With the gonadal primordium as the main model of this project, we are able to study cell fate specification, this could potentially aid in stem cell therapy development and research to contribute to possible fertility treatments.

To understand the full developmental processes which take place in the embryonic testes, the development of steroidogenic LCs must be understood. This project will explore the effects of Lhx9 on the LPC pool, and consequently how this impacts the differentiation into foetal/adult LCs using heterozygous mice (HET = Lhx9+/-). Previously, the Wilson Laboratory had shown that these mice produced reduced levels (50%) of Lhx9 during gonadal development (Workman & Wilson, unpublished).

This project tested this through two hypothesis and a number of different objectives.

Hypotheses: 1. Reduced Lhx9 results in earlier differentiation of progenitor cells to foetal Leydig cells. This also would also decrease the size of the progenitor pool.

2. Reduced expression of Lhx9 will lead to a decrease in progenitor cells and an increase in adult Leydig cells. This will consequently result in a premature loss of Leydig cells in the adult testis.

Objectives: 1. To observe Lhx9 expression over time. - RT-qPCR analysis of a number of marker genes at the following timepoints of development compared to Lhx9 mRNA expression: E15.5, P0, P20, adult (3-5 months old). - Immunohistochemistry to observe the target gene protein levels in adult testis.

11 2. To monitor changes to gene expression upon reduced Lhx9 expression. - RNA sequencing of 4 month old WT and Lhx9+/- testes RNA to observe changes in gene expression. - RT-qPCR of E15.5 WT and Lhx9+/- testes RNA to observe changes in gene expression.

3. To understand the effects of reduced Lhx9 on the regenerative development of Leydig cells. - Intra-peritoneal injection of LC specific toxicant ethane dimethanesulfonate (EDS) administered to mice to eliminate LCs. - RT-qPCR for marker gene expression in testes at different key timepoints in LC regeneration post elimination.

12 2. Methods

2.1. Animal Husbandry C57BL/6 and Lhx9+/- mice were sourced from the Jackson Laboratory (Bar Harbor, Maine, USA) and maintained at the Hercus Taieri Resource Unit (HTRU) and housed in cages fitted with ad libitum water and food access. The cages were stored in the Animal House where the facility operated on 12-hr light/dark cycles.

The Lhx9+/- mouse strain was re-derived from backcrossing frozen sperm and a DBA2J background by the Jackson Laboratories (https://www.jax.org/strain/006812), at Otago it has been backcrossed onto a C57BL/6 background. The line was originally constructed by Birk et al (Birk et al., 2000) deleting exons 2 and 3 of the Lhx9 gene by insertion of a neomycin cassette.

To breed, one male and one female were housed together before the start of the dark cycle. Females were checked the following morning for a copulatory plug indicative of successful copulation (Deb et al., 2006). The time of copulatory plug detection was recorded as 0.5 days post coitum (dpc). The dam was then placed into a separate cage from the sire. When there was an absence of copulatory plug, mating continued until a copulatory plug was observed or until the mating reached four days. If no plug was observed after four days, the mice were separated and the females monitored for signs of pregnancy in case of a missed plug.

Protocols performed all had approval from the University of Otago Animal Ethics Committee.

2.2. Genotyping DNA Isolation DNA was isolated from embryo tail tips or adult ear notch samples. Each tube containing a single sample had a mix of 150 µL lysis buffer, and 1 µL proteinase K. The tubes were incubated in a heat block at 55 °C overnight to degrade the tissue. The tubes were heated to 85 °C for 10 min to deactivate the proteinase K, and genotyping was completed by PCR (see 2.2.2).

13 Polymerase Chain Reaction (PCR) PCR was used for the amplification of a specified region of the DNA extracted from the embryo tail tips and ear notches. To assess the Lhx9 genotype, primers designed to amplify the neomycin cassette inserted to create the knockout line were used. Reagents listed in Table 2 were added and briefly centrifuged to ensure all components were collected at the bottom of the tube. The tubes were run on an genotyping programme in the Applied Biosystems Veriti 96 well thermal cycler as listed in Table 3

Table 2: Stock and concentration of Lhx9 Genotyping PCR protocol

Reagent Concentration µL per Reaction Neomycin Primer - 1 Reaction Buffer 10 x 2 dNTPs 10 mM 0.2 Taq 5 U/µL 0.1 ddH2O - 13.8 DNA - 2

Table 3: Thermal profile for Lhx9 genotyping

Cycle Temperature Time 1 94 °C 3 min 35 94 °C 30 s 60 °C 1 min 72 °C 1 min 1 72 °C 2 min - 4 °C Hold

Gel Electrophoresis Amplified PCR products from the genotyping protocol were visualised through gel electrophoresis. 10 µL of each PCR product and 1 µL of 1 kb DNA Ladder were loaded into wells of a 2% agarose gel and submerged in 1x Tris-acetate-EDTA (TAE) buffer in a gel tank. The gel was left to run for approximately 30 minutes and the DNA visualised in a UVP GelDoc under a UV light. Samples with a band at approximately 492 base pairs indicating the presence of the neomycin cassette were genotyped as Lhx9 heterozygous (Lhx9+/-) (Figure 3).

14

Figure 3 – Example Lhx9 genotyping results via gel electrophoresis.

The presence of bands in lanes 2, 3, 4, 7, and 8 at the 492 bp mark indicate the presence of the neomycin cassette and therefore heterozygosity in Lhx9. The lack of band in the negative control (-ve) lane ensures confidence the results are not due to experimental error. ddH2O used as negative control.

2.3. Inhibitor Treatment of Male Mice Ethane dimethanesulfonate (EDS) is a Leydig cell specific toxicant. 250 mg stock EDS was dissolved in 1.7mL of vehicle solution (30% Dimethyl sulfoxide (DMSO) and 70% H2O) to give a 30 mg/mL working solution. A final dosage of 160mg/kg was injected.

Mice were weighed to calculate the injection volume of either EDS/vehicle through the following equation:

(푊푒𝑖𝑔ℎ푡 𝑖푛 푘𝑔 × 160) = mL Injected 30

Room temperature EDS/vehicle was drawn into 1mL syringes with needles gauge size 25-27. Mice were restrained via the scruff technique and tilted at an angle where the head was lower than the body allowing organs to fall away from the injection site. The needle was inserted bevel up into the lower right quadrant of the mouse and an aspiration was performed before the solution was injected intraperitoneally. Each mouse was monitored daily for any negative body weight, behavioural, or overall health changes.

15 2.4. Tissue Collection Adult mice were euthanised at days zero, four, or fourteen post EDS injection through cervical dislocation. The abdomen of the mice were opened after being sprayed with 70% ethanol. Blood samples were drawn via cardiac puncture and left to rest for serum collection for a minimum of 30 minutes. The seminal vesicles were also dissected out for weight measurements. The adult testes were removed and immediately placed into 4% PFA for fixation, or in RNALater (Sigma Aldrich, USA) for RNA extraction.

Adult mice were euthanised for embryo collection eighteen days after plug observation to collect E18.5 embryos. Embryos were immediately placed in a sterile petri dish with PBS after being removed from the mother followed by decapitation. The embryos were dissected under the dissecting microscope after being removed from the extra-embryonic membranes. The embryo tail tips were removed for necessary genotyping to determine either the sex or the Lhx9 genotype of the embryo (see 2.2.1). Gonads were removed and placed immediately into RNALater. WT embryos and heterozygous embryos were collected from the same litter.

Serum Collection Blood drawn via cardiac puncture was left to rest at room temperature to allow for blood to coagulate for 30 mins to 1 hr. Blood was centrifuged at 3500 g for 10 mins at 4 °C to allow the separation of the serum (top layer) from the red blood cells (bottom layer). The serum was carefully collected without disturbance of the red blood cells into a new tube before storing at -20 °C.

2.5. RNA Extraction Two methods of RNA extraction were used depending on the size of the tissue sample. Smaller tissue such as embryonic testes were extracted through the Bioline Isolate II RNA Mini Kit, where a Trizol-based protocol was used for larger adult testis samples.

RNA Extraction: Smaller Tissues The Bioline Isolate II RNA Mini Kit was used for RNA extraction from embryonic testes. Tissue was homogenised using a pestle in 350 µL of the supplied Lysis Buffer RLY. RNA extraction and purification was carried out according to the protocol provided by the manufacturer. The RNA was eluted into 60 µL of RNase-free water.

16 RNA Extraction: Larger Tissues The RNA extraction from the adult testes was completed using a Trizol-based method. 500 µL of Trizol was added into a glass homogeniser containing the tissue. The tissue was ground until homogenous, followed by the addition of another 500 µL of Trizol which was incubated at room temperature for 5 minutes to allow for the dissociation of the cells. 200 µL of chloroform was then added and the sample was shaken vigorously for 15 s followed by a 3 min rest period at room temperature. Following the rest period, the samples were then centrifuged at 15,000 g for 15 min at 4 °C to enable the separation of the sample into three layers; a top clear aqueous layer of RNA, a middle white layer of DNA, and a bottom pink layer of organic proteins. The top aqueous layer was removed carefully without disturbing the other layers and pipetted into a new sample tube containing 500 µL of isopropanol. The sample was then vortexed and rested for 10 min at room temperature before being centrifuged at 12,000 g for 10 min at 4 °C. Following centrifugation, the supernatant was removed and the remaining pellet resuspended in 1 mL of 100% ethanol by brief vortexing. The sample was centrifuged at 7,500 g for 5 min at 4 °C. The supernatant was removed following the centrifuge and the pellet left to dry for 30 min to 1 hr.

The pellet was then resuspended in 50 µL of RNase-free water and incubated on a heat block for 10 min at 57.5 °C before storage at -20 °C.

2.6. cDNA Synthesis Complimentary DNA (cDNA) was synthesised using qScript cDNA SuperMix (5x reaction buffer) from QuantaBio. The reaction set up was as follows: qScript (4 µL), RNA template

(calculated by 200/nanodrop concentration), and RNase-free H2O (variable amount to make reaction volume to 20 µL) was added. cDNA synthesis was completed using the Applied Biosystems Veriti 96 well thermal cycler with the following programme: 25 °C (5 min), 42 °C (60 min), 85 °C (5 min), and 4°C (hold).

2.7. Real-Time Quantitative PCR The relative expression of genes in the testes samples were quantified through real-time quantitative PCR (RT-qPCR). The cycles were normalised against two reference genes, ribosomal protein S29 (Rps29) and actin beta (Actb).

17 Primer Design & Efficiency Analysis Primers were designed through Primer BLAST after sourcing the gene mRNA transcripts from NCBI. Primer BLAST parameters altered from the default setting were set to the following: 100-120 bp PCR Product size, exon junction span to span an exon-exon junction, and organism set to Homo sapiens (taxid: 9606).

Primer sequences can be found in Appendix 2. The efficiencies of the primers were tested in a 5-fold serial dilution of cDNA in RT-qPCR master mix The plate was sealed off with a plastic sheet and contents were centrifuged down at 2000 g for 15 s.

The Applied Biosystems QuantStudio3 Real-Time PCR Systems machine was used to run the thermal profile outlined in Table 4. A non-linear regression plot was generated in PRISM (Appendix 3) based on the Ct values calculated for each replicate, determining the slope which was used to calculate the primer efficiency using the equation as follows:

(− 1 )−1 (10 푠푙표푝푒 ) × 100

Primers with efficiencies between 85-120% were of deemed suitable for use in RT-qPCR experiments.

Reaction Set up The reactions for RT-qPCR were set up in a 96-well plate. Each well contained 1 µL of sample cDNA, and 9 µL of RT-qPCR reaction master mix. The master mix was made up of 5 µL of

SYBR Green Fast Mix, 3 µL of RNase-free H2O, and 1 µL of primer (forward and reverse). Each test sample was carried out in triplicate to act as technical replicates. The plate was sealed off with a plastic sheet and contents were centrifuged down at 2000 g for 15 s.

The Applied Biosystems QuantStudio3 Real-Time PCR Systems machine was used to run the thermal profile outlined in Table 4

18 Table 4: Thermal profile for RT-qPCR reaction.

Temperature Time Cycles 55 °C 2 min 1 95 °C 2 min 40 95 °C 5 sec 60 °C 19 sec 1 Machine specific melt curve programme

Data Analysis The melt curve was analysed for a single peak in each sample to ensure specificity of amplification of a single product (Appendix 4), deeming the data suitable for relative gene expression analysis. The mean Ct value of three technical replicates were compared to the mean Ct value of two reference genes Rps29 and Actb, with the relative expression being calculated as shown.

푅푒푙푎푡𝑖푣푒 퐸푥푝푟푒푠푠𝑖표푛 = 2− (푚푒푎푛 퐶푡 푔푒푛푒 − 푚푒푎푛 퐶푡 푟푒푓 푔푒푛푒)

2.8. Histology Tissue was fixed in 4% PFA overnight and washed in PBS to be embedded by the Otago Histology Services Unit in paraffin wax. Tissues were sectioned at 5 µm on a microtome and placed onto slides to dry and set at 60 °C for 60 min. Sections to be used for histological analysis were stained in the Otago Histology Services Unit in haematoxylin and eosin (H&E) using a standard protocol (Fischer et al., 2008).

2.9. Immunohistochemistry Sectioned 5 µm tissue samples prepared using a microtome were used for immunohistochemistry.

Antigen Retrieval Sections were deparaffinised and rehydrated with a series of washes as follows: 2x 100% xylene for 3 mins, then ethanol/H2O dilutions (100% x2, 90% x1, and 50% x1) for 3 min each.

19 Slides were then placed into a plastic slide holder and covered in sodium citrate buffer in a vented microwaveable container. The container was microwaved at medium-power for 30 min, adding buffer when necessary to ensure there was enough to prevent the slides from drying out. Container was cooled at room temperature for 15 min before rinsing the slides in RNase- free water.

Primary Antibody Binding Slides were washed in PBS with 0.025% Triton-X (PBTx) three times for 10 min each. They were then blocked in PBTx + 10% heat inactivated sheep serum + 1% bovine serum albumin solution for 2 hr at room temperature. The slides were then incubated overnight at 4 °C in a wash with primary antibodies diluted in the blocking solution. Slides were also incubated with a solution containing no antibody as a control.

Table 5: Immunohistochemistry Antibody Dilutions

Target Primary antibody Dilution Secondary antibody Dilution Generated by the Goat Anti-Rabbit LHX9 Wilson lab (Y Yang, 1 in 2000 IgG H&L (HRP) 1 in 2000 2018). ab205718 RC2-s (Developmental Goat anti-mouse Nestin studies hybridoma 1 in 25 1 in 600 (HRP) bank)

Secondary Antibody Binding & Colour Development To remove any unbound antibodies, slides were washed in PBTx three times (5 min per wash) at room temperature, then incubated with 3% H2O2 for 5 min to inhibit any endogenous horseradish peroxidase (HRP) activity. Following the incubation, blocking solution was added and slides were blocked again for 20 min. The secondary antibody/blocking solution dilution was then added to the slides for 20 min at room temperature. Unbound secondary antibodies were washed off with three 5 min washes in PBTx.

DAB substrate kit (Abcam - ab64238) was used to develop the colour according to manufacturer instructions. Sections were then counterstained in 50% Haematoxylin for 10 s.

20 2.10. RNA Sequencing The RNA was isolated (see 2.5) and quality check was performed via nanodrop. The RNA was dried down and treated with RNA Stable before being sent to Beijing Genomics Institute (BGI), Hong Kong, for sequencing and RNA integrity number (RIN). The genomic library was prepared by BGI using their own sequencing platform – DNBseq. 100bp paired end reads were sequenced and returned. The clean paired end reads were mapped and analysed via the following Galaxy software tools:

- Feature counts: galaxy version 1.6.4+galaxy1 – tool version v1.6.4 - MultiQC: galaxy version 1.8+galaxy0 - HISAT2 Software: 2.1.0+galaxy5 - FastQC: 0.72+galaxy1 - BAM to BigWig conversion: 1.0.2

Differential gene expression analysis was done by Dr Megan Wilson using RUVSeq/DESeq2 R packages, genes differentially expressed with a FDR of 0.1 were considered further (due to the low sample number).

Gene ontology analysis was done using DAVID (Version 6.8), and Panther (Version 15.0) using default settings.

2.11. Statistical Analysis Statistical analyses was carried out using either an unpaired t-test or a one-way ANOVA with Tukey’s multiple comparison test where appropriate.

21 3. Results

3.1. Lhx9 mRNA Expression during Leydig cell Differentiation Selection of Target Genes A series of target genes to be used for expression analysis were chosen based on their role in the development and function of LCs. Markers of Leydig progenitor cells (LPCs) used were Nestin, Notch, Patched-1(Ptc1), and Aristaless (Arx). Mature LC markers used include cholesterol side-chain cleavage enzyme (P450scc), cytochrome P450 17A1 (P450c17), 3- hydroxysteroid dehydrogenase (3HSD), platelet derived growth factor receptor subunit-A (Pdgfra), leukemia inhibitory factor receptor (Lifr), and vascular cell adhesion molecule (Vcam). RT-qPCR assays are shown relative to the expression of reference genes ribosomal protein S29 (Rps29), and actin beta (Actb). Together, these sets of markers can provide insight into the proportion of LPCs and ALCs in the Lhx9+/- (HET) testes and whether that proportion is altered in comparison to Lhx9+/+ (WT) littermates. This can begin to build evidence as to what the role of Lhx9 may be during embryogenesis and LC function. A selection of these genes were also used to generate a timeline of their expression over the perinatal (E15.5 – P4) and the adult period, giving insight into how the markers are expected to change during this time and what the implications of change in a Lhx9+/- may be beyond E15.5. The specific function of the genes used in RT-qPCR is listed in Table 6 below.

Table 6: Target genes and their function in Leydig cells

Target gene Function in LCs Nestin Marker of LPC (Davidoff et al., 2004). Notch Maintains the progenitor state of the LPCs, expression prevents differentiation into mature LCs (Tang et al., 2008). Arx Functions to promote the differentiation of FLCs through paracrine signaling (Yu et al., 2014). 3HSD Marker of differentiated progenitor cells, involved with steroidogenesis (Rasmussen et al., 2013; Teerds et al., 2007). Patched-1 Triggers LC differentiation through upregulation of P450scc (Yao et al., 2002). P450scc Catalyses cholesterol to steroid hormone synthesis (Wang et al., 2017). P450c17 Contributes to steroid synthesis pathway (Chung et al., 1987). Pdgfra Receptor found on ALC surface for PDGF, responsible for LC differentiation (Basciani et al., 2010). Lifr Marker of stem cell proliferation and differentiation (Curley et al., 2018). Vcam Adhesion promoting molecule expressed in ALC (O’Shaughnessy et al., 2002).

22 Primer Design Efficiencies Forward and reverse primers were either sourced or designed for each marker gene using NCBI Primer BLAST and tested for efficiency as outlined in 2.7.1. Each reaction produced a melt curve which was analysed to ensure the amplification of a single product (see example in Figure 4, complete list of figures in Appendix 4). A table of primer sequences can be found in Appendix 2.

A. B.

Figure 4 – Representation of RT-qPCR efficiency measures. A. Efficiency curve for gene Aristaless, efficiency of 96.99% is suitable for RT-qPCR. B. Melt curve displaying the changes of fluorescence relative to temperature, where a single peak indicates amplification of a single product and suitability for downstream analysis. Melt curves for full list of genes can be found in Appendix 4. RT-qPCR primer efficiencies were run in serial dilutions (2.7.2). Primer efficiencies are listed below (Table 7).

Table 7: Target gene primer efficiencies

Gene target Efficiency Gene Efficiency

3HSD 106.76% Pdgfra 87.63%

Arx 96.99% Vcam 93.32 Lifr 118.5% Fgf1 120% Nestin 114.46% Rps29 114.1% Notch 112.23% Actb 99.5% P450c17 107.28% Patched-1 108.68%

P450scc 112.91% Lhx9 109.13%

23 Leydig cell Marker Gene Expression changes in the Perinatal Period The relative expression of the LC marker genes of interest were calculated as outlined in 2.7.3, using the RNA extracted from the testes. The expression of the genes was normalized against two reference genes, Rps29 and Actb. Timelines were created for a total of 6 marker genes selected to represent function and various cell types in the testes.

Figure 5 shows the changes of Lhx9 in WT mice across different points of development. The expression of Lhx9 quantified for RT-qPCR appears to peak at birth/postnatal day 0 (P0), before declining at P20 and even further so at adult age (3-5 months of age). Expression was significantly different between P0 and adult stages (p = 0.02, one-way ANOVA with Tukey’s multiple comparison test). This may represent the developmental requirements of LPCs at each timepoint.

Lhx9

0.15

n B-D

o

i

s

s

e r

p 0.10

x

E

e

n

e G

0.05

e

v

i

t

a

l

e R 0.00 E15.5 P0 P20 adult Age

Figure 5 – Lhx9 expression in WT testes at varying developmental timepoints.

For each timepoint, Lhx9 expression is shown relative to the reference genes (Actb and Rsp29). RT- qPCR assays were carried out on biological replicates: E15.5 (n=3), P0 (n=3), P20 (n=3) and adult testis (n=4). Data shown is mean +/- SEM. Abbreviations: E15.5 = Embryonic day 15.5, P0 = Postnatal day 0, P20 = Postnatal day 20, and adult = 3-5 months of age. Statistical analysis used was a one-way ANOVA with Tukey’s multiple comparison test. Within each graph, labels above bars indicate statistical significance between the lettered data groups, data is represented as follows: A = E15.5, B = P0, C = P20, and D = Adult.

Additionally, Figure 6 illustrates the trends of key marker genes investigated in this project alongside Lhx9. LPC marker Nestin (Figure 6A) appears to peak at E15.5 and declines to minimal relative expression in the post-natal P20 and adult testes. Expression was significantly different between E15.5 and P20 stages (p = 0.04, one-way ANOVA with Tukey’s multiple comparison test). Similarly, Notch follows a parallel expression pattern where peak levels are seen at E15.5 a period of peak FLCs differentiation (Chen et al., 2010) (Figure 6B).

24 The expression was significantly different between E15.5 and each of the following timepoints, P0 (p = 0.009), P20 (p = 0.004), and adult (p = 0.007), all of which were calculated through one-way ANOVA with Tukey’s multiple comparison test.

Steroidogenic enzyme markers were examined by RT-qPCR were 3HSD, P450c17, and P450scc (Figure 6C, D, & E).

P0

Figure 6 –Relative expression timeline of key testicular marker genes at varying developmental timepoints in WT testes .

For each timepoint, expression of genes of interest is shown relative to the reference genes (Actin and Rps29). RT-qPCR assays were carried out on the following biological replicates: A. Nestin: E.15.5 (n=2), P0 (n=2), P20 (n=3), 4 month adult testis (n=1), B. Notch: E.15.5 (n=2), P0 (n=3), P20 (n=3), adult testis (n=2), C. 3HSD: E.15.5 (n=3), P0 (n=3), P20 (n=3), adult testis (n=2), D. P450c17: E.15.5 (n=3), P0 (n=0), P20 (n=2), adult testis (n=4), E. P450scc: E.15.5 (n=7), P0 (n=1), P20 (n=3), adult testis (n=3), F. Patched1: E.15.5 (n=3), P0 (n=3), P20 (n=2), adult testis (n=3).

Abbreviations: WT = Lhx9+/+, E15.5 = Embryonic day 15.5, P0 = Postnatal day 0, P20 = Postnatal day 20, and adult = 3-5 months of age. Statistical analysis used was a one-way ANOVA with Tukey’s multiple comparison test. Within each graph, labels above bars indicate statistical significance between the lettered data groups, data is represented as follows: A = E15.5, B = P0, C = P20, and D = Adult.

25 3HSD (Figure 6C) and P450c17 (Figure 6D) appear to increase in expression with age, likely indicative of the increased steroidogenesis occurring in the adult testes. Relative expression is lowest in the younger tissues such as in E15.5, and noticeably increases by the time the adult tissues are developed. Due to time restraints, there is no data for the P0 P450c17 timepoint therefore the exact pattern of expression cannot yet be determined.

The expression calculated for steroidogenic enzyme markers displayed statistical significance, where 3HSD E15.5 samples has significantly lower expression than both P20 (p = 0.0053) and adult samples (p = 0.0044). Similarly, the 3HSD expression in P0 samples were also significantly lower than in P20 (p = 0.027) and adult samples (p = 0.018).

For P450c17, a significant increase in expression is shown in adult samples from E15.5 (p = 0.02). Expression of P450scc (Figure 6E) was found to be significantly increased in the adults compared to all other samples of E15.5 (p = 0.0009), P0 (p = 0.0053), and P20 (p = 0.0008), the pattern of expression declines at birth before increasing from P0 onwards.

3.2. Leydig cell Marker Gene Expression in E15.5 Lhx9+/- Testis As Lhx9 is likely to have an important role in the development of LCs, the LC differentiation between WT and HET mice was examined between the adult and embryonic stage tissues through closer observation of LC marker expression. The expression levels of 11 selected marker genes were quantified and calculated through RT-qPCR as outlined in 2.7.3. These genes represented a variation of cell types and process which may be expected to occur differently between WT and HET E15.5 samples. Gene expression was normalized against two reference genes Rps29 and Actb. All statistics were calculated through an unpaired t-test.

Usually acting to keep LPCs stable in the progenitor state (Barsoum & Yao, 2010), Notch expression was significantly reduced in the HET testis compared to WT littermates by a -4.3 fold change (p = 0.0189) (Figure 7A). Similarly, LPC marker Nestin, displayed a reduced expression in the HET compared to the WT (-1.8 fold-change) although this was not significant, (p = 0.08) (Figure 7B). This may be representative of a small sample size. The reduced expression in the progenitor marker Notch in HET samples (Figure 7) imply a decrease in LPCs, specifically in foetal testis haploinsufficient for Lhx9.

26

Figure 7 – Relative gene expression of different LPC marker genes between WT and HET E15.5 testes.

All gene expression was relative to reference genes Rps29 and Actb. RT-qPCR assays were carried out with the following biological replicates A. Notch: WT (n=2), HET (n=3), B. Nestin: WT (n=2), HET (n=3). Statistical significance determined by unpaired t-test * p <0.05. Abbreviations: WT = Lhx9+/+, HET = Lhx9+/-.

Figure 8 – Relative gene expression of different steroidogenic enzyme markers in WT vs HET.

RT-qPCR results for HET and WT E15.5 testis gene expression relative to reference genes (Rps29 and Act). Data shown is mean of the relative expression, error bars represent standard error of the mean. A. P450c17: WT (n=3), HET (n=2), B. 3bHSD: WT (n=2), HET (n=3), C.P450scc: WT (n=6), HET (n=2). Statistical significance determined by unpaired t-test, * p <0.05, *** p<0.001. HET = Lhx9+/-.

27 Steroidogenic enzyme marker P450c17 (Figure 8A) shows a statistically significant 8.3 fold increase in the E15.5 HET testis compared to WT littermates (p = 0.0001). Marker of the same steroidogenic process 3HSD (Figure 8B) also displays a significant increase in the HET versus the WT samples (2.9 fold-change, p = 0.041). P450scc, a gene upstream of 3HSD and P450c17 in the same steroidogenic pathway, showed an unexpected significant decrease in the HET compared to the WT (-2.3 fold-change, p = 0.024) (Figure 8C). This differs from the significant increase in the HET in the other steroidogenic enzyme markers P450c17 and 3HSD.

Figure 9 – Relative expression of LC marker genes in E15.5 WT and HET male testes.

Mean of the relative expression compared against reference genes (Rps29 and Actb), error bars represent standard error of the mean. RT-qPCR assays were carried out with the following biological replicates: A. Vcam: WT (n=3), HET (n=3), B. Pdgfra: WT (n= 3), HET (n=3), C. Patched-1: WT (n=3), HET (n=2), D. Lhx9: WT (n=4), HET (n=4), E. Lifr: WT (n=2), HET (n=2), F. Aristaless: (n=2), HET (n=2). Statistical significance determined by unpaired t-test. Where * p <0.05. HET = Lhx9+/-. Vcam, a marker for vascular endothelium (Kong et al., 2018) showed significantly increased expression in the HET testes compared to WT littermates by 1-fold, (p = 0.026) (Figure 9A).

28 Cell proliferation marker Pdgfra (Odeh et al., 2014) (Figure 9B) was also significantly upregulated in the HET testes by 2-fold (p = 0.012) compared to the WT samples, this marker is essential for LC differentiation. Expression of Lifr in the HET testis increased in some HET samples, however this was not statistically significant (p = 0.34) or reproducible across biological replications (as indicated by the large error bars).

Patched-1 is required for LC differentiation (Mendis-Handagama & Ariyaratne, 2001). Two genes Patched-1 (Figure 9C) and Lhx9 (Figure 9D) displayed no significant changes in expression in HET samples. An additional gene, Arx also functions to promote LC differentiation (Miyabayashi et al., 2013). Arx (Figure 9F) has a slightly more prominent decrease in HET with a 3-fold change, however, this was also insignificant. Smaller changes to gene expression would be difficult to detect with smaller sample sizes.

LHX9 Protein Expression in Adult Testis The physical arrangement of cells can be visualized in the below schematic (Figure 10), where the cluster of ALCs have blood vessels through it, these are important as they are thought to contribute to the LPC population (Davidoff et al., 2004). Pericytes (PCs) are another suggested stem LC alongside LPCs to the ALCs (Eliveld et al., 2020), these can be found embedded within the peritubular myoid cells (PTMCs) or surrounding blood vessels (Curley et al., 2019; Kumar & DeFalco, 2018).

Figure 10 – Schematic of physical arrangement of cells in adult testis.

A schematic illustrating the physical arrangement of testicular cells with relation to the ALCs. Suspected progenitor cells PCs are seen embedded within the PTMC and wrapped around the vasculature. Abbreviations: PTMC = Peritubular myoid cell. PC = pericyte, ALC = adult Leydig cell. (MJ. Wilson., Original work, 2020).

29 Figure 11 visualizes the expression of LHX9 through an immunohistochemistry (IHC). The staining (brown precipitate) is localized in the LCs, showing the expression of LHX9 in WT adult testes is primarily expressed in the interstitial cells. Some fainter staining was also observed in the spermatogonia (SP) within the seminiferous tubules (Figure 11B), suggesting additional expression of LHX9 in these cells. No background staining was observed on the secondary (anti-rabbit-HRP) only control (data not shown), suggesting that staining observed is due to binding of the primary to LHX9.

A. B. LC LC

SP ST

ST

Figure 11 – LHX9 immunohistochemistry on WT adult (4 months) testis.

Expression of LHX9 localised in the interstitial Leydig cells (LC), outside of the seminiferous tubules (ST) can be visualised through brown staining. Faint spermatogonia (SP) staining can be seen. A. 10x magnification, scale bar = 100 m. B. Magnification 40x, scale bar = 50 m.

Adult testes from HET mice were unable to be stained through IHC due to time limitations. This would be done in future to create a comparison of LHX9 protein staining to WT. A western blot could additionally be performed to confirm whether the quantified amount of LHX9 protein in an adult HET is half of an adult WT. This has been proven previously in the embryo (Workman, unpublished).

3.3. Regenerative Ability of Lhx9+/- Mice Due to the previously mentioned link between Lhx9, LPCs, Notch signaling and adult Leydig cell differentiation (Table 6), an RT-qPCR assay was done to quantify the effects of EDS treatments on LC regeneration in HET mice.

30 The original proposal for this experiment was to collect EDS treated samples at all three timepoints of LC regeneration, day of injection (day 0), proliferation of LCs (day 4), and the near complete restoration of LCs (day 14) based on previous literature (Jiang et al., 2014; Yang et al., 2017). Unfortunately, as a result of time restraints, only one litter of male mice of the correct age was obtained. This litter was only sufficient for a pilot study. EDS (final concentration of 160 mg/body weight in kg) was administered via intra-peritoneal injection to the mice n=1 (WT), n=1 (HET) to begin the partial elimination of LCs. Following this, the testes, blood, seminal vesicle weight, and body weight data was collected at day 14. As controls, samples collected were a day 0 (no injection) and a day 14 (injected with the vehicle/buffer only).

Using these tissues, 5 genes of interest picked based on their important roles in the testis and the cell types which they marked were quantified (Figure 12) relative to two reference genes, ribosomal protein S29 (Rps29) and actin beta (Actb). Nestin, Notch, and Lhx9 were picked as they are markers of LPCs, the main cell population of interest. Where P450scc and 3HSD were picked based on their role in steroidogenesis, the function of the cells of interest.

There were differences in expression levels after the 14 days post-EDS injection. Expression levels for Nestin, Lhx9, and 3HSD were all highest in the EDS treated WT mice. This is to be expected when in contrast with HET treatment groups as the WT are expected to have a higher regenerative ability. Nestin and Notch are expected to be increased due to their role in localised co-expression in LPCs, an increase in these two markers are likely indicative of the proportionate increase in proliferative activity occurring in the LPCs (Chen et al., 2017; Murta et al., 2013). Downstream of an increase in LPCs, we can expect to see an increase in the number of LCs differentiated and a subsequent increase in the amount of active steroidogenesis occurring (Chen et al., 2019; Hu et al., 2007). This justifies the rise in P450scc and 3HSD expression.

The expression levels for Notch, and P450scc are all similar across all three treatment groups. However, with the low sample size any subtle differences in gene expression between the groups would not be obvious.

31

Figure 12 – Relative gene expression of marker genes in day 14 preliminary EDS study treatment groups.

RT-qPCR assay on mice testes is relative to reference genes (Rps29 and Actb) A. Nestin, B. Lhx9, C. Notch, D. 3bHSD, E. P450scc. Where WT control = Day 0, and other treatment groups (WT no EDS, WT + EDS, HET + EDS) are day 14. Preliminary study therefore n=1 for all treatment groups. EDS = ethane dimethanesulfonate, HET = Lhx9+/-.

Seminal vesicle (SV) weight is an effective indicator of the amount of testosterone which is being produced in a mouse (Eleftheriou & Lucas, 1974; Kerr et al., 1987). The SVs were dissected out and weighed from each mouse of the EDS (Figure 13). The SV weight was divided by the overall body weight of the mouse to normalize for differences in body weight. The graphed results show no significant difference in SV between samples but again, it must be acknowledged that n=1 and therefore, no significant conclusions can be drawn. Similarly, with results from Figure 12, subtle differences between the treatment groups can be acknowledged.

32

Figure 13 – Mean seminal vesicle weight of mice from different treatment groups.

Error bars represent standard error of the mean. SV weight calculated by SV/Body weight to normalise for body weight differences between mice (n=1). Abbreviations: SV = Seminal vesicle, EDS = Ethane dimethanesulfonate, HET = Lhx9+/-.

This brings light to limitations of this study given the restrictive time allowance for my overall project. To acquire a full set of data which enables insight across all three timepoints of LC regeneration, it would require much more time given that 2 weeks are needed for the LCs to regenerate, following injection of EDS, and required the same age of mice (4-5 month old mice), unfortunately no mice of the correct age were ready in time. This study has potential based on the presented preliminary results, indicating a different response in HET mice to EDS- induced LC regeneration.

33 Marker Protein Expression in EDS Treated Testis Testes were dissected and sectioned for IHC and H&E staining (Figure 14). The testis in the WT control (vehicle only) appear as normal, where the seminiferous tubules are round and abundant of spermatozoa (Figure 14). The interstitium contains LCs which appear as regular with vasculature running between them (orange arrowhead).

The panel in the WT + EDS treatment image displays distorted looking seminiferous tubules; the spermatozoa appear detached from each other where there are many crack-like separations running through and do not appear as cohesive as the WT control image (Figure 14). Similarly, the LCs/interstitial cells are not as aggregated as the WT control, there is evidence of LCs detaching from the interstitial mass (red arrowhead).

1 2

Figure 14 – H&E stain of day 14 testes morphology in different treatment groups from EDS study.

LC population and seminiferous tubule structure appear regular in WT control. Structure of LCs appear to degrade with EDS treatment in WT sample, and worsen with EDS treatment in HET. Abbreviations: WT Control = Vehicle treatment, EDS = ethane dimethanesulfonate, HET = Lhx9+/-.

The HET + EDS treated testis images in Figure 14 display interstitial cell masses which are uncharacteristic of WT testis interstitial cell mass features. Normal structure and morphology can be seen in the WT control image (Figure 14). In the HET + EDS1 image, vasculature can be seen running through the center of the mass (green arrowhead).

The LCs in HET + EDS2 image appear poorly organized surrounding the central vasculature (blue arrowhead) and morphologically less round and more elongated than that of the WT images. This is of close resemblance to the morphology of LPCs or pericytes (Figure 10), potentially suggesting an incomplete maturation through to ALC stage. This may be due to failed proliferation as there are also fewer LPCs observed, this would be in support of the results observed in the RT-qPCR where no increased expression of Nestin was seen, indicative

34 of no proliferation of LPCs. These results together can contribute to the explanation suggesting a reduced expression of Lhx9 may contribute to the lessened regenerative capacity of LCs.

The testis dissected from the various treatment groups were stained using a NESTIN antibody. The results imaged in Figure 15 display the subtle changes which can be seen in NESTIN expression. Though all expression is localized to the interstitial cell masses, it is particularly noticeable how dark the staining in the HET + EDS treatment is (Figure 15C & F).

The staining in the WT panels A, B, D, and E are all stained evenly, in comparison the HET + EDS treated stains. The HET + EDS treated testis stain was uneven and more evident in certain cells which appear to have an elongated morphology to them.

WT CONTROL WT + EDS HET + EDS A. B. C.

D. E. F.

Figure 15 – NESTIN specific IHC on day 14 testes from EDS study.

The WT control images (A&D) show staining in the interstitial cell masses where LPCs would be found, the staining is also present in WT + EDS. Staining appears darker and more concentrated in HET + EDS slide.

Fig A-C images at 20x magnification, scale bars 100 m. Fig D&E imaged at 40x magnification, scale bars = 100 m. Fig E imaged at 60x magnification, scale bar = 100 m. Abbreviations: EDS = ethane dimethanesulfonate, HET = Lhx9+/-/

35 Neurofilament heavy peptide (NF-H) is a co-expressed marker of LPCs with Nestin, therefore, an IHC was performed to visualize the staining of an additional marker (Figure 16). The staining can be seen as even and localized in the interstitial cells where LPCs are located. The staining in the EDS treated WT shows an equally even staining, however also shows darker staining of an elongated cell (red arrowhead, Figure 16), this is likely a LPC based off the morphology (Figure 1).

WT CONTROL WT + EDS HET + EDS 2 Ab CONTROL

DAY 14 DAY

Figure 16 – NF-H specific IHC on day 14 testes from EDS study.

Neurofilament heavy peptide (NF-H) is a LPC marker co-expressed with Nestin. The WT control image displayed localised staining in the interstitial cells where LPCs are located. WT + EDS image shows darker staining on elongated cell, a potential LPC. Secondary Antibody (2 Ab) control image demonstrates background staining trapped under cells which are damaged in absence of primary antibody. The HET + EDS staining is therefore likely to be a combination of normal expression staining in the LPCs and darker staining from damaged cells.

The secondary antibody control image shows there is non-specific staining present in the absence of a primary antibody, this is likely due to damaged tissues which trap the secondary antibody resulting in precipitation forming, alternatively, blood on sections can result in the release of endogenous peroxidases. The endogenous peroxidases may not be entirely suppressed by the concentration of hydrogen peroxidase used in the protocol, and therefore will not be properly blocked and cause non-specific staining (Kim et al., 2016). Additional time would enable the testing of different blockers for optimal staining.

36 3.4. RNA Sequencing Data Six samples of RNA were extracted (see 2.5) from HET and WT 3-4-month-old adult testes. The quality of the RNA was checked using a nanodrop blanked against 1L of RNase-free water, results can be found in Table 8. RNA of good quality can be indicated through the measurements from the nanodrop. For good quality RNA, we expect a 260/280 figure of approximately 1.8-2, where the closer it is to 2, the “purer” the RNA is. A 260/230 number of 2-2.2 can be expected for a high level of RNA purity.

RNA samples were then sent to BGI sequencing Hong Kong (see 2.10) for further quality analysis before library construction. The data returned from BGI contained RNA Integrity Numbers (RIN) shown in Table 8. The RIN number acts as a measure of quality, the scale ranges between 1 – 10, 10 being the least degraded and therefore the highest quality (Schroeder et al., 2006).

Table 8: RNA sample nanodrop results and sequencing RIN results for samples sent for RNA sequencing.

Sample Nanodrop Concentration 260/280 260/230 RIN Number (ng/L) T4m 12390.2 2.02 1.74 8.4 T601 17580.6 1.85 1.94 8.5 T602 14884.8 1.98 1.98 8.8 T50 2428.9 2.03 2.17 9.1 T66 1568.4 1.97 1.05 3.1 T67 1267.5 1.97 0.94 2.4

Of the six samples sent for sequencing, only four were of high enough quality to continue sequencing with (Table 8). Two samples T66 (RIN 3.1) and T67 (RIN 2.4) were not of high enough quality as they had particularly low RIN numbers compared to the other samples such as T50 (RIN 9.1).

Bioanalyser results produced from BGI sequencing additionally provides information on the quality of the RNA. An example set of results can be found in Figure 17. The graphs can be interpreted through two key regions on the x-axis, the baseline signal (25 on x-axis), and the 18S to 28S ribosomal band region (approximately 2000 on x-axis) (Mueller et al., 2004). The peaks expected to be found in high quality RNA are seen in sample T50 of Figure 17.

37 There are two peaks at the 18S – 28S region, and one small peak at the baseline. The smaller the baseline signal peak is, the higher quality the RNA is, this is the reason sample T50 only has a RIN of 9.1, not 10, as there is a peak present at the baseline. The more degraded the RNA is, the closer to the baseline signal the peaks get and the larger the baseline signal peak becomes, this can be seen in sample T67. The gel electrophoresis image column on the right of the graph is another indication of degradation, the more degraded RNA becomes, the smaller the bands fragment into and thus the further down the column they will travel. Sample T67 displays a smeared column with a single band resting close to the baseline signal (green band) indicating the degradation of RNA. Sample T50 however shows a definitive band halfway down the column, exhibiting non-degraded RNA (Mueller et al., 2004).

Figure 17 – Example Bioanalyser results.

With the degradation of RNA, the 18S and 28S ratio (region around 2000 on x-axis) decreases, and signal at baseline marker RNA increases (25 on x-axis). Sample T67 (Top) displays low quality RNA (RIN = 2.4). Band on gel electrophoresis column on right can be seen as degraded. Graph shows lack of peaks at the 18S and 28S points. Sample T50 (Bottom) displays high quality RNA (RIN = 9.1), band can be seen midway down gel electrophoresis column indicating RNA has not degraded. Graph shows two peaks at 18S and 28S indicting high-quality non-degraded RNA.

38 Quality Check The four samples of RNA quality were used to prepare a library and sequenced using a pipeline of online tools (Figure 18) This involved using a series of online tools from Galaxy.

Transfer files to galaxy •Filezilla •Transfer files to galaxy

Mapping to mouse genome •Confirm quality •HISAT2 Mapping

Counts •Feature count to count reads mapped to genes •Generation of a count table for each sample

Combine counts •Combine count columns for each sample to one column •Downloaded and ready for differential gene expression calling

Figure 18 – Pipeline figure of steps involved with processing RNA sequencing output.

The reads produced were mapped against the reference mouse genome (mm10 assembly) using HISAT tool. The sequencing returned approximately 23 million reads, 93% of which were uniquely mapped to the reference genome. The final table of aligned reads can be found below in Table 9.

Table 9: Table of mapped reads from sequencing output generated by BGI.

RNA sample FastQC file ID Clean Raw Reads Mapped Reads % Uniquely Mapped WT_2 T4m_2 24489305 22 673 252 92.6 T4m_1 WT_1 T601_1 25653754 23 965 844 93.2 T601_2 HET_2 T602_1 25727527 23 924 079 93.3 T602_2 HET_1 T50_1 25604916 23 542966 91.9 T50_2

39 Following on, the quality of the sequences was then checked against the FastQC tool on Galaxy and shown to be of high quality (Figure 19). Figure 19A is a visual representation of quality across all bases. The bases were all within the green zone of the quality check graph, falling no lower than the quality score of approximately 30 for each position in the read, an indication of high quality. Figure 19B graphs the quality score distribution across all sequences, the average quality scores per sequence is seen to peak at an average within the range of 35-38 on the Phred scale.

Figure 19 – Example of FastQC quality check results for sample T50.

A. Checking sequence quality per base, all bases are in the green zone above quality score 30 therefore are high quality zones for each position in the read. B. Quality score distribution across all sequences graphed, where average quality per read peaks around 35-38 on the Phred score, indicating high quality.

40 Analysis A PCA plot was generated to visualise the amount of variation between the samples and within each sample group (Figure 20A) using RUVSeq R package. The sample groups (HET and WT) are displaying variation between them which is expected as they are samples of differing genotypes. There is clustering appearing within the replicates of each sample group, this is indicative that there is consistency within each of the genotypes but variation between them.

A. B.

Figure 20 – RNA-seq analysis for differential gene expression.

A. PCA plot displaying variation between the sample groups and between replicates, HET and WT sample groups have a large variation (99.85%), where there is clustering with minimal variation within the replicates in the sample groups, HETs cluster together with each other and WT cluster together, maximum variation of 0.08%. B. Volcano plot illustrating the differentially expressed genes in HET samples relative to WT by fold change. Red= significantly DEG (FDR<0.1). Data provided by Dr Megan Wilson.

The differentially expressed genes were called using the edgeR package with a false discovery rate (FDR) of 0.1 due to the small sample size. The volcano plot generated was a visual representation of DEGs which are differentially expressed between HET and WT samples, this was calculated by fold change (FC) compared to expression value (counts per million, CPM). The DEGs which were significantly differentially expressed (FDR < 0.1) are shown as a red dot on the volcano plot (Figure 20B).

41 Of the significant DEGs, 30 genes were expressed higher in the HET (positive FC), and 21 were expressed lower in the HET (negative FC). A table of differentially expressed genes was generated with their assigned ENTREZ ID numbers (Table 14 & Table 15; Appendix 5). The ENTREZ ID numbers were translated into a list of official gene symbols and names, this list was then used for gene ontology (GO) analysis. Prior to the GO analysis, the gene list was split into two lists, one for each of the genes with higher expression in HET or higher expression in WT. The top genes with reduced expression in HET samples can be found below in Table 10, and top genes with increased expression in HET samples in Table 11.

Table 10: Top 4 genes with reduced expression in HET samples compared to WT, sorted by FC in expression

Gene ID log2(fold adjusted Gene symbol/name Known role in the testis change) P-value edgeR 14164 2.1994359 0.0073572 Fibroblast growth Key regulator of lipid metabolism (Jiang 9 7 factor 1 (Fgf1) et al., 2013). 12350 2.4220689 1.23E-11 Carbonic Catalyses production of bicarbonate 3 anhydrase 3 necessary for optimal sperm motility and (Car3) fertilisation ability (Wandernoth et al., 2015). 19817 2.1589072 7.46E-11 7SK small nuclear PGC growth and proliferation in testis 5 ribonucleoprotein (Okamura et al., 2012). (Rn7sk) 71052 2.1728296 0.0100138 RIKEN cDNA Prominent foetal testis marker 4318753 9 (O’Shaughnessy et al., 2003).

Table 11: Top 4 genes with increased expression in HET samples compared to WT, sorted by FC in expression

Gene ID log2(fold adjusted Gene symbol/name Known role in the testis change) P-value edgeR (FDR) 232413 -2.0734481 0.0038083 C-type lectin Low levels of expression in the testes, domain family 12, expressed in macrophages/immune cells member a (Clec12a) (Marshall et al., 2004). 100042355 -2.0681159 1.23E-11 predicted gene Not been studied in any detail. Low levels 10705 (Gm10705) of expression previously detected in testis (Papatheodorou et al., 2020). 100736249 -2.0288927 0.07342035 mistral long non- Found within HOX complex (Clark et al., coding RNA (Mira) 2016). It has not been studied with respect to adult testis. 22526 -1.981199 0.00118482 predicted gene 4836 Not been studied in the testis. Located on X (Gm4836) chromosome (Clark et al., 2016).

42 BigWig data tracks were produced using the DEG list and visualized in the UCSC genome browser, example tracks in Figure 21 display the mapped reads of the DEGs specifically for two genes, Rn7sk (top), and Fgf1 (bottom). There is a clear difference in the number of reads mapped against the WT samples than the HET.

Figure 21 – Mapped reads in relation to Rn7sk and Fgf1 genes in UCSC genome browser.

Reads mapped against Rn7sk (top) and Fgf1 (bottom) both more prominent in WT samples than HET samples. HET = Lhx9+/-. Data is shown as bigwig data format.

For each of the DEG lists, two overrepresentation tests were performed using two different online GO tools, DAVID (Huang et al., 2009) and Panther (Thomas et al., 2006). For the list of genes with lower expression in the HET samples relative to the WT, both overrepresentation tests from the two GO tools provided the same results. The results discovered that from the genes identified from the ENTREZ ID numbers, the biological pathways significantly overrepresented appear to be involved with the lipid metabolism pathway. The table generated from DAVID (Table 12) is inserted as an example.

Table 12: Results from DAVID overrepresentation test.

Category Term Count % PValue Genes Fold Enrichment GOTERM_BP_ GO:0006631~fatty acid 3 10.34482759 0.01148599 14600, 17.3865385 DIRECT metabolic process 11770, 20249 GOTERM_BP_ GO:0050872~white fat 2 6.896551724 0.01668732 11770, 1.13E+02 DIRECT cell differentiation 20249 GOTERM_BP_ GO:0050873~brown fat 2 6.896551724 0.03514514 11770, 53.1823529 DIRECT cell differentiation 20249 GOTERM_BP_ GO:0006641~triglyceri 2 6.896551724 0.04021377 20249, 4.64E+01 DIRECT de metabolic process 13106

43 There were no significantly overrepresented biological pathways found in the genes higher expressed in HET testis than WT. This is likely due to the small number of genes differentially expressed higher in HET.

Validation To validate the levels of expression from the results in section 3.4.2, primers were designed for two of the genes significantly differentially expressed between the HET/WT samples. these primers were then used to validate the results through RT-qPCR using different biological replicates to the RNA sequencing samples.

Of the primers ordered, only one set was efficient at 120%, this was for fibroblast growth factor 1 (Fgf1). The Fgf1 mRNA expression levels were then quantified in 4-month-old HET/WT samples different to the samples sent for RNA sequencing.

Figure 22 – Results of relative gene expression of Fgf1 to validate RNA sequencing data.

Validation completed using different biological replicates to RNA sequencing samples. Gene expression results from RT-qPCR assay measured against reference genes (Rps29 and Actb). Mean of the relative expression, error bars represent standard error of the mean. WT (n = 4), HET (n = 3). Statistical significance determined by unpaired t-test. *p<0.05. Fgf1 = Fibroblast growth factor-1, HET = Lhx9+/-.

Figure 22 displays the results from the RT-qPCR quantification, the results indicate a significant -1.3-fold difference (p = 0.025) in expression of Fgf1 between HET and WT samples. The RNA sequencing data also displayed a significant 2-fold difference (Table 10).

44 Additionally, expression of Fgf1 was quantified in the day 14 EDS treated mice (Figure 23) as it is a gene which is differentially expressed in Lhx9+/- RNA-seq data (Table 10). The expression of Fgf1 is highest in the WT Control (no injection, day 0 tissue collection) and WT no EDS samples (vehicle injection; Figure 23). In comparison, Fgf1 expression is lower in both the WT EDS day 14 treatment and in the HET EDS sample.

FGF1 is a promoter of LPC proliferation but not differentiation, the reduction of Fgf1 expression is likely to be required to enable differentiation of LCs (Chen et al., 2019). Based on this knowledge we can expect the overall expression of Fgf1 transcript to reduce at day 14 given the LCs are forming from the LPCs. The EDS treated WT and HET mice both have lower expression of Fgf1 than both WT control (day 0) and WT with no EDS (vehicle) treatment. This is expected as EDS reduces the LC populations. The lower expression in the HET EDS treated sample was also expected as it begins with a lower baseline expression of Fgf1 than WT samples as discovered in the RNA sequencing data (Table 10). The results from this quantification are interesting and warrant further experiments to increase the sample size and ensure it is reproducible.

Figure 23 – Relative gene expression of Fgf1 in preliminary day 14 EDS study samples.

RT-qPCR assay was normalised against reference genes (Rps29 and Actb). Where WT control = Day 0, and other treatment groups (WT no EDS, WT + EDS, HET + EDS) are day 14. Preliminary study therefore n=1 for all treatment groups. Abbreviations: Fgf1 = Fibroblast growth factor-1, EDS = ethane dimethanesulfonate, HET = Lhx9+/-.

45 4. Discussion

With the ongoing rise in fertility difficulties in men, the issues uncovered with earlier developmental stages of the male reproductive system become more prominent. There is already ample understanding about the testes and the associated genetic components, however there is much more to be explored, for example the impact of Lhx9 on the development of the testosterone producing Leydig cells. This project aimed to investigate the broad hypothesis that reduced expression of Lhx9 would lead to a smaller Leydig progenitor cell pool, potentially being the cause of reduced fertility observed previously (Wilson, unpublished data). The research conducted in this project has begun to prove the Lhx9+/- mice trend towards smaller progenitor cell pools, and therefore we can assume that correlates with less adult LCs maturing, and predict reduced testosterone causing subfertility. Blood was collected to test testosterone levels but due to the shortened project time frame, this had to be left out.

RT-qPCR results revealed changes in overall patterns of Leydig cell marker expression. There was a general trend of decreased LPC marker genes mRNA such as Nestin and Notch, alongside a general significant difference between the expression of steroidogenic enzyme markers in WT and HET samples.

Preliminary EDS injection experiments similarly show changes in the expression of progenitor cell pool markers and steroidogenic markers between experimental WT and HET mice. The conclusions which can be drawn are however limited due to n=1 and time restraints, this study needs to be replicated with a larger sample size.

RNA sequencing results identified 51 genes which were differentially expressed between the adult WT/HET samples. An overrepresentation test performed during the gene ontology analysis uncovered an overrepresentation of the differentially expressed genes in pathways associated with lipid metabolism. Fgf1 was of particular interest as it is a known promoter of LPC proliferation (Chen et al., 2019).Validation of this analysis for Fgf1 confirmed a significant decrease in the HET expression (Figure 22).

46 4.1. Expression of Key Genes during Leydig cell Development RT-qPCR was carried out on the Lhx9 WT RNA extracted from mouse testes of different developmental stages ranging from embryonic day 15.5 to adult aged. The aim of this was to observe the global gene expression of a number of key marker genes of Leydig cells.

Progenitor cell pool markers such as Nestin and Notch play a role in the maintenance of the LPC pool. With this knowledge, the decline in expression of such markers between the post- natal/pre-pubertal period is unsurprising. One of the reasons being the increasing number of ALCs, which the LPCs differentiate into around that developmental period (Zirkin et al., 2010).

With the decrease in LPCs around the pubertal period we can expect this naturally implies there is an increase in ALC population. The steroidogenic enzyme markers increase over the developmental period to peak at the adult stage where the LCs are most active. This occurs as the reproductive peak where functions such as spermatogenesis are most crucial. This aligns with previous research indicating that there is an increase in steroidogenic function and a decrease in proliferative function along the developmental timeline (Ye et al., 2017) justifying the decrease in Lhx9, Nestin, and Notch, and the increase in 3HSD, and P450c17.

Interestingly, the expression pattern of 3HSD differs from that of P450scc though they are in the same pathway (Figure 24). 3HSD begins with lowest expression in the E15.5 samples, progressively increasing with development, but still having peak expression occur in the adult tissues. Though there is no data for one timepoint in P450c17, it can still be observed that the peak expression occurs in adult tissues (Figure 6).

The expression of P450scc interestingly does not display a dosage dependent-like pattern across the developmental timeline, the literature however suggests there is a decrease in expression of P450scc before returning to high expression levels as an adult post-puberty (Hu et al., 2007). The results of the P450scc expression pattern can therefore be expected. Based on the pre-existing knowledge regarding the developmental timeline of LCs, it is known that LC populations decline at birth, and increase again at puberty for the production of testosterone necessary for reproductive function (Zirkin et al., 2010). With the known steroidogenic function of P450scc, the depleted expression at P0 from E15.5 can be linked to the degenerating FLC population, while expression reaches a high in the adult testis (Figure 6). This could

47 possibly be a correlation with the increase in demand for steroidogenesis as the testosterone requiring biological processes become active. The proportional increase in testosterone synthesis with LC differentiation has been previously explored in the literature to be a result of enzymatic activity including P450scc and P450c17 (Shan, 1993). This is indicative that the enzymes may play different roles through the developmental period, however are all expressed highest in adulthood when reproduction is expected.

The expression of Lhx9 compared to the steroidogenic enzymes do not resemble each other’s patterns, but appear inverse. The peak expression of Lhx9 occurs at P0, this timepoint coincides with the peak of LCs needed for the activation of androgen responsiveness in the hypothalamus. The responsiveness is crucial at that point in development as it an important period for the development of secondary sex characteristics highly driven by testosterone (Nef & Parada, 2000). This decline of Lhx9 is likely a result of the decrease in the Lhx9 expressing progenitor population guiding the necessary differentiation into mature LCs occurring (Clarkson & Herbison, 2016; Zirkin et al., 2010).

4.2. Impact of Lhx9+/- Haploinsufficency RT-qPCR and RNA sequencing were carried out on WT and HET samples at two timepoints, E15.5 and adult stage tissues. This was to quantify marker genes which would indicate whether the LPC pool size changed based on genotype as hypothesized.

E15.5 Samples The results of the differentially expressed genes illustrated the changes which occurred between the E15.5 WT and HET mice.

Firstly, the significant decrease in the LPC marker Notch and trending-towards significant decrease of Nestin in HET mouse was interesting. This observation supports the hypothesis suggesting reduced expression in Lhx9 reduces the LPC pool (Figure 7). Nestin differential expression was not significant, however could be due to sample size (Figure 7). Given the established roles of markers Nestin and Notch in the progenitor cell pool, the prominent decrease provides evidence contributing to the hypothesis whereby reduced Lhx9 results in a smaller LPC pool.

48 Similarly, with the established roles of the steroidogenic enzymes, the decrease in HET P450scc expression was not surprising as we can understand how less LCs directly lead to less production of testosterone.

The opposite significant increase in 3HSD and P450c17 HET samples was however interesting as it was expected to follow the same trend of P450scc as they are part of the same pathway as seen in Figure 24. The action of the enzymes is sequential (Martin & Touaibia, 2020). This raises the question as to whether Lhx9 has a specific region of the steroidogenic pathway in which it targets, potentially downstream of the P450scc action which may contribute to a reduction in fertility in the affected mice.

Figure 24 – Basic steroidogenic pathway.

The simplified steroidogenic pathway for testosterone production, primarily occurring in the LCs. Orange arrows indicate direction of change of gene expression in HET compared to WT; decreased for P450scc and increased for P450c17 and 3HSD.

Enzymes involved in process are in orange text, by-products are in black text. Where DHEA = dehydroepiandrosterone. 17HSD was not investigated in this study however is integral to the process.

Alternatively, the increase in 3HSD and P450c17 which opposes the decrease in P450scc could be considered a physiological attempt at compensating for the decrease in production of pregnenolone to maintain a normal production of testosterone. This pathway can have a large impact on the development of individuals. It has been discovered previously in literature that a deficiency in pregnenolone due to a heterozygous mutation in P450scc was found in genetically 46 XY female. It was discussed that this mutation in P450scc resulted in phenotypical sex reversal (Tajima et al., 2001).

Understanding the potential downstream effects which reduced Lhx9 enables allows us to consider biomedical applications in which the information can be proactively applied. As for the example mentioned, the expression of Lhx9 could be investigated for any heterozygous mutations which may develop a link to upstream causes of mutations in P450scc.

49 The increase in expression is interesting as it would be expected that the downstream enzymes, 3HSD and P450c17, would have lower expression levels than the upstream P450scc if the target of Lhx9 was in the pathway between P450scc and P450c17 (Figure 24).

Genetic markers Pdgfra and Vcam were significantly increased in HET samples. Pdgfra plays a role in the differentiation of LPCs (Odeh et al., 2014). It has been previously established in literature that the absence of Pdgfra eliminates the ability to form ALCs (Gnessi et al., 2000), with this knowledge, there could be many reasons as to why there is such a significant difference in expression between the two genotypes. One possibility includes the sudden overexpression in the HET samples in attempt to match the expected levels of differentiated LCs needed, with attempts to match the essential level of testosterone production for optimal reproductive function.

With Arx being another positive factor for LPC differentiation (Miyabayashi et al., 2013), the expression was trending towards significantly decreased, this was surprising given the two markers are both markers of proliferation. Comparing these two results, we can consider whether it is a possibility that they are acting in the same pathway as the target of Lhx9 leading to the differing changes in expression.

Lifr expression is increased in the HET mice (Figure 9). Lifr signaling has been characterized as a regulator of testicular function, with the primary source being from peritubular myoid cells (PTMCs) (Curley et al., 2018). PTMCs are Nestin-positive cells found in the testes which give rise to LCs (Kumar & DeFalco, 2018). The decrease in Nestin indicates what is an assumed decline in LPCs, this would be expected to lead to a decrease in the expression of Lifr being expressed. The expression differences of Lifr are counterintuitive and would be interesting to further research.

4.3. EDS Injection Study The process of LC regeneration was studied through the elimination of the population by injection of EDS. The results (Figure 12) were indicative that there was high expression of Nestin, Lhx9, and 3HSD in the day 14 WT EDS treated mouse group. The strong increase in these factors in the WT EDS treatment group is likely an indication of the superior regenerative ability in the WT mouse versus the HET.

50 The regeneration of LCs begins with regeneration and proliferation of the LPCs to renew the progenitor pool from which the LCs will differentiate (Chen et al., 2017). The increase in Nestin and Lhx9 is likely the quantifiable proof which illustrates the increase in LPCs and the beginning of differentiation into ALCs which can be identified by the increase in steroidogenic enzyme 3HSD. Additionally, the increase in 3HSD can also account for the functional differences between the EDS treated WT and HET. There is less expression of these three markers in the EDS treated HET samples, the decrease in expression can be linked back to the hypothesis suggesting the reduced starting LPC pool size which comes with less Lhx9. With less LPCs to begin with, it is likely there is less potential proliferation and differentiation for the same developmental window than the WT samples with an assumed full LPC pool. The expression of Nestin, Lhx9, and 3HSD in the WT control is low, but present. This could be as they were not treated with EDS, thus there was no elimination of LCs. With this, we can consider there was therefore no regeneration required and so the expression levels were neither elevated as the WT EDS treated group, or as reduced as the WT control group because the necessary ALCs were already differentiated and stably functioning.

Expression between the three EDS treatment groups for Notch and P450scc are all relatively similar to each other. Based on the results mentioned above, we could expect that the expression for P450scc to be higher in the WT EDS treatment group. Though it is not, the low expression of Notch can be justified. Notch expression has been established as responsible for the maintenance of the LPC state (Tang et al., 2008), the decrease in expression observed could therefore be correlated to the increase in differentiation in the WT EDS treated mice, the more differentiation into ALCs which occurs, the less Notch which will be required to hold the LPC state.

The SV weights were also measured to get an interpretation of how the function of the LCs are working, the results did not show any notable differences between the treatment groups, this could very likely be a result of a lack of replicates.

Cell Characteristics following EDS Treatment The H&E staining of the EDS samples revealed interesting characteristics (Figure 14) between the WT and HET EDS treated testes. LCs surround a central blood vessel due to their ability to stimulate angiogenesis and the vasculature as a source of their LPCs (Collin & Bergh, 1996).

51 From this, we can expect healthy LC masses to appear as elongated LPCs surrounding vasculature with mature LCs also in the population. This is as they appear in the WT control image of Figure 14. The WT EDS image displays cells appearing to lose their adhesive properties (Lin et al., 2008), the red arrowhead draws attention to the LC which is dissociating from the mass. The seminiferous tubules also appear to be affected, displaying less abundance of cells within the lumen. This however still appears healthier than the HET treated samples. The HET EDS treated testis display a lack of organization surrounding the central vasculature, the cells are displaced in the interstitium and appear poorly with more elongated cells than the WT, and the seminiferous tubule appear to lack a distinctive lumen.

The elongated cells resembling LPCs (Figure 1) potentially suggest successful proliferation of LPCs, but failure to continue development through maturation. With the understanding of the presence of Lhx9 in the LPCs (Birk et al., 2000), the reduced expression is more likely than not to be involved in the poor physical health of the testicular cells, potentially impacting the maturation process more than the proliferation and differentiation.

A NESTIN IHC was performed to visualize the location of expression in the testis. There was an abundance of background staining and so was difficult to conclude, however it can be acknowledged that there was particularly concentrated staining in the HET EDS treated sections. It is known that IHC staining can become particularly dark in dying tissue (Kim et al., 2016). IHC staining with NF-H was more successful, with some staining of elongated cells (same morphology of LPCs) near the regenerating LC populations were observed in the WT+EDS samples. This staining was not detected in the HET+EDS or WT control samples, possibly eluding that the HET struggles to proliferate and mature new cells from LPC to ALC. This aligns with the observations made in the H&E staining (Figure 14).

This portion of the study needs to be repeated as it can be difficult to distinguish between pericyte cells and LPCs through H&E staining. This is due to morphological similarities between pericytes associated with the blood vessels, and LPCs which similarly are detached from the blood vessels. A stain specific for LPC markers such as NESTIN or NF-H would be the stain of preference for differentiating between the two.

52 4.4. RNA Sequencing Six samples were sent to BGI for sequencing, two of which were lacking in quality and therefore could not be sequenced (Table 8). The RIN numbers were not high enough and showed much of the RNA was degraded. The nanodrop values (260/230) attained prior to sequencing were poor, this could have been a result of poor RNA extraction (Table 8).

The results from the sequencing output from BGI sequencing were aligned to the reference mouse genome, there were high levels of mapping, all of which were >90% uniquely mapped.

The PCA plot generated to check the variability between the four sequenced samples was expected. The clustering of the HET and WT biological replicates within each sample groups indicated the replicates were similar enough (Figure 20). Most of the sample variation (98.8%, Figure 20A, x-axis) was between the HET and WT sample groups as anticipated.

The decreased expression of Fgf1 in HET samples was validated by RT-qPCR on separate biological replicates (Figure 22). This was fascinating to discover given the already established roles of Fgf1 in lipid metabolism and its localized expression in the LCs (Jiang et al., 2013). The role of Fgf1 is to provide the LCs with the synthesis of testosterone precursor, cholesterol (Figure 24), the cholesterol can then be converted into testosterone and used for other biological processes such as spermatogenesis and bone maintenance (Ferlin et al., 2013; Jiang et al., 2013; Struik et al., 2019; Wang et al., 2015). The decrease in Fgf1 expression in the HET samples as a result of a decreased Lhx9 expression in the mice is a viable explanation. Based on the hypothesis formed, less Lhx9 could result in a reduced expression of Fgf1 further downstream, leading to less cholesterol synthesis and therefore less testosterone synthesis contributing to subfertility. The relationship between Lhx9 and Fgf1 could be further investigated, specifically whether Fgf1 is directly downstream of Lhx9.

The results from the preliminary EDS study displayed similar trends of expression from the Fgf1 RT-qPCR assay (Figure 23). The EDS treated HET sample appears to have a reduced expression of Fgf1 compared to the WT control and vehicle. The levels of expression are very similar to the WT EDS treated levels. It would be expected that the HET and WT EDS treated expression would differ due to the difference in Lhx9, however n=1, thus the sample size is very restricting for forming conclusions. Based off the results, it would be interesting

53 to prolong the time allowed for regeneration to exceed 14 days as this may have an impact on the results between the WT EDS and HET EDS expression levels. Regardless, the overall trend of Figure 23 support the findings that reduced Lhx9 expression is potentially linked to a decline in Fgf1 expression, and potentially acting in the same pathway.

It is important to recognize that LCs do degenerate both overtime and with age (Beattie et al., 2015), to manage this, the day 0 WT vehicle administered mice can act as a control, in comparison to the day 0 WT control, there is little difference in the amount of Fgf1 expression, and thus implies the decrease in expression is likely a result of the Lhx9 reduced.

4.5. Significance With the increasing infertility rates and approximately 40-50% of global infertility rates a result of male factors (N. Kumar & Singh, 2015), it is an area of biological research which demands more attention. Clinical uses for LC regeneration have already begun to be investigated for issues related to adult-onset hypogonadism with hopes to increase testosterone production (Lamb, 2019).

Additionally, developmental abnormalities related to mutations in LHX9 have been recently discovered, with links to disorders of sex development (DSDs) (Kunitomo et al., 2020). LHX9 has also been linked to cancer in a number of cases. For example, FGF has been discovered to induce LHX9 and further regulate the progression of osteosarcoma (Li et al., 2019). Though it is not yet directly linked to testicular cancer, it is interesting that the gene regulating LHX9 in osteosarcoma progression (FGF), is from the same family as Fgf1, the gene found most differentially expressed in a reduced Lhx9 testis (Table 10). The LHX9 promoter has also been found to have increased methylation in cervical cancer (Bhat et al., 2017). With a more comprehensive insight into the mechanisms of LC development and regeneration, it is unquestionable that research regarding FGF1 and LHX9 is likely to have great biomedical relevance, and a myriad of not only DSDs, but also testosterone deficiency related issues could be relieved (Kunitomo et al., 2020). With enough development, the mechanisms involving specific genes such as LHX9 and FGF1 may even be transferrable across human diseases inclusive of cancers as previously mentioned.

54 4.6. Limitations Whilst a fulfilling year was carried out, it must be acknowledged that there are several limitations which have hindered carrying out this research to its full potential. Firstly, the obvious time restraints caused by the global pandemic this year very suddenly stripped crucial time from this tightly planned project. The loss of time was not only inclusive of the 4-week lockdown, but also the additional restrictive access to the laboratory during level 3.

With regards to the project itself, there are few flaws to be mentioned. Regarding the Leydig cell elimination study, there is only partial elimination in mice, where most studies induce near complete elimination in rats, this however could not be avoided as a mouse line was the model organism to be used. Additionally, the mouse line used does not generate a Lhx9-/- knockout (KO) line as it is embryonic lethal (Workman, Wilson, pers. comm.). This is unfortunate as it would be beneficial to understand the expression profile across all three genotypes. One possible solution is to generate a conditional-KO line where Lhx9 is knocked out in the testis, post-gonadal formation. This however was not possible based on the short timeframe given to complete this project.

The staining in the NESTIN IHC revealed issues regarding background staining. To obtain more precise knowledge on the location of expression of NESTIN, ranging conditions of staining such as differing buffers and blocking times could contribute to more solid results for analysis.

Finally, the RNA sequencing data only enabled two replicates rather than the standard minimum of three. This was likely due to poor RNA extraction technique of the poor-quality samples. The RNA sequencing validation required more RT-qPCR validation than the single Fgf1 gene. This was however unavoidable as the primers designed for the other DEGs were inefficient and there was insufficient time to design and order extra sets to test.

55 4.7. Future Direction There is great potential in this project based on the results obtained in this short period. There is much more to potentially invest time into exploring. To begin, the investigation of LPC marker gene expression in Lhx9+/- testes could be more comprehensively mapped if expanded into additional timepoints, specifically prenatally and beyond 3-4 months of age in adulthood. Similarly, for the EDS LC elimination study, it would be intriguing to further observe effects of changes in gene expression for the regeneration of LCs beyond 14 days, and with more replicates.

To further develop the RNA sequencing results, the investigation of the relationship between Lhx9 and lipid synthesis in the LC steroidogenic pathway would provide great insight into the underlying mechanisms of testosterone synthesis and LC regeneration. It would be interesting to complete more replicates to potentially observe differences in lowly expressed genes. This would enable validation of more genes that can be classed as differentially expressed in the bioinformatics. From this, direct targets of LHX9 could then be determined through chromatin immunoprecipitation, where LHX9 protein binding sites would be identified.

Samples for the measurement of serum testosterone levels between Lhx9+/- and WT mice were collected through the duration of this project as it was initially to be done by an ELISA assay. This is important as testosterone is the direct quantifiable product of the LCs and would be indicative of differences in functionality between the genotypes. Unfortunately, due to the time restraints placed on this project, this experiment was not able to be carried out but must be considered for future direction.

Finally, though not the focus of this study, the compelling observation of faint LHX9 expression in spermatogonia (Figure 5) could be further investigated. Given that LHX9 is also expressed in primary oocytes (Workman, unpublished), it would be fascinating as to uncover the function of Lhx9 in early germ cells.

56 4.8. Conclusions To conclude, this research into the role of Lhx9 in the testis has contributed to a greater understanding of the underpinning mechanisms whereby reduced Lhx9 negatively impacts fertility.

The results from this research have demonstrated factors which may contribute to subfertility in HET mice, more specifically, downregulation of Leydig progenitor cell markers (Nestin and Notch), and changes in steroidogenic enzymes, both of which indicate a decrease in progenitor cell population as hypothesised. From the RNA sequencing data, we observe 51 differentially expressed genes such as Fgf1 between the WT and HET samples. The biological processes overrepresented in these genes were involved in lipid metabolism pathways. Finally, preliminary LC elimination studies using EDS additionally show interesting changes in the LC marker genes, these are promising results for further pursuit.

These results have contributed to defining the genetic network which underlies the development and function of testes. This is vital, as a comprehensive understanding of male fertility enables the possibility for contribution towards potential research which may progress clinically.

57 References

Ahmed, S. F., Achermann, J. C., Arlt, W., Balen, A. H., Conway, G., Edwards, Z. L., ... & Miles, H. L. (2011). UK guidance on the initial evaluation of an infant or an adolescent with a suspected disorder of sex development. Clinical Endocrinology, 75(1), 12-26.

Arango, N. A., Lovell-Badge, R., & Behringer, R. R. (1999). Targeted mutagenesis of the endogenous mouse Mis gene promoter: in vivo definition of genetic pathways of vertebrate sexual development. Cell, 99(4), 409-419.

Atkinson‐Leadbeater, K., Bertolesi, G. E., Johnston, J. A., Hehr, C. L., & McFarlane, S. (2009). FGF receptor dependent regulation of Lhx9 expression in the developing nervous system. Developmental Dynamics, 238(2), 367-375.

Barrionuevo, F., Georg, I., Scherthan, H., Lécureuil, C., Guillou, F., Wegner, M., & Scherer, G. (2009). Testis cord differentiation after the sex determination stage is independent of Sox9 but fails in the combined absence of Sox9 and Sox8. Developmental biology, 327(2), 301-312.

Barsoum, I. B., & Yao, H. H. C. (2010). Fetal Leydig cells: progenitor cell maintenance and differentiation. Journal of andrology, 31(1), 11-15.

Barsoum, I. B., Kaur, J., Ge, R. S., Cooke, P. S., & Yao, H. H. C. (2013). Dynamic changes in fetal Leydig cell populations influence adult Leydig cell populations in mice. The FASEB Journal, 27(7), 2657-2666.

Basciani, S., Mariani, S., Spera, G., & Gnessi, L. (2010). Role of platelet-derived growth factors in the testis. Endocrine reviews, 31(6), 916-939.

Beattie, M. C., Adekola, L., Papadopoulos, V., Chen, H., & Zirkin, B. R. (2015). Leydig cell aging and hypogonadism. Experimental gerontology, 68, 87-91.

Benson, D. A., Cavanaugh, M., Clark, K., Karsch-Mizrachi, I., Lipman, D. J., Ostell, J., & Sayers, E. W. (2012). GenBank. Nucleic acids research, 41(D1), D36-D42.

Bertuzzi, S., Porter, F. D., Pitts, A., Kumar, M., Agulnick, A., Wassif, C., & Westphal, H. (1999). Characterization of Lhx9, a novel LIM/homeobox gene expressed by the pioneer neurons in the mouse cerebral cortex. Mechanisms of development, 81(1-2), 193-198.

Bhat, S., Kabekkodu, S. P., Varghese, V. K., Chakrabarty, S., Mallya, S. P., Rotti, H., ... & Satyamoorthy, K. (2017). Aberrant gene-specific DNA methylation signature analysis in cervical cancer. Tumor Biology, 39(3), 1010428317694573.

Biason-Lauber, A. (2010). Control of sex development. Best practice & research Clinical endocrinology & metabolism, 24(2), 163-186.

Bitgood, M. J., & McMahon, A. P. (1995). Hedgehog and Bmp genes are coexpressed at many diverse sites of cell-cell interaction in the mouse embryo. DEVELOPMENTAL BIOLOGY-ACADEMIC PRESS-, 172, 126-126.

58 Bitgood, M. J., Shen, L., & McMahon, A. P. (1996). Sertoli cell signaling by Desert hedgehog regulates the male germline. Current biology, 6(3), 298-304.

Bowles, J., Feng, C. W., Spiller, C., Davidson, T. L., Jackson, A., & Koopman, P. (2010). FGF9 suppresses meiosis and promotes male germ cell fate in mice. Developmental cell, 19(3), 440-449.

Brennan, J., & Capel, B. (2004). One tissue, two fates: molecular genetic events that underlie testis versus ovary development. Nature Reviews Genetics, 5(7), 509-521.

Bullejos, M., & Koopman, P. (2001). Spatially dynamic expression of Sry in mouse genital ridges. Developmental dynamics: an official publication of the American Association of Anatomists, 221(2), 201-205.

Cameron, F. J., & Sinclair, A. H. (1997). Mutations in SRY and SOX9: testis-determining genes. Human mutation, 9(5), 388.

Carson, C. C. (2002). Prevalence, diagnosis and treatment of hypogonadism in primary care practice. Focus on Sexual Health medicine, 3, 1-3.

Chamindrani Mendis-Handagama, S. M. L., & Siril Ariyaratne, H. B. (2001). Differentiation of the adult Leydig cell population in the postnatal testis. Biology of reproduction, 65(3), 660-671.

Chen, H., Ge, R. S., & Zirkin, B. R. (2009). Leydig cells: from stem cells to aging. Molecular and cellular endocrinology, 306(1-2), 9-16.

Chen, H., Stanley, E., Jin, S., & Zirkin, B. R. (2010). Stem Leydig cells: from fetal to aged animals. Birth Defects Research Part C: Embryo Today: Reviews, 90(4), 272-283.

Chen, H., Wang, Y., Ge, R., & Zirkin, B. R. (2017). Leydig cell stem cells: Identification, proliferation and differentiation. Molecular and cellular endocrinology, 445, 65-73.

Chen, L., Li, X., Wang, Y., Song, T., Li, H., Xie, L., ... & Lv, Y. (2019). Fibroblast growth factor 1 promotes rat stem Leydig cell development. Frontiers in endocrinology, 10, 118.

Chen, P., Guan, X., Zhao, X., Chen, F., Yang, J., Wang, Y., ... & Chen, H. (2019). Characterization and differentiation of CD51+ Stem Leydig cells in adult mouse testes. Molecular and cellular endocrinology, 493, 110449.

Chen, P., Zirkin, B. R., & Chen, H. (2020). Stem Leydig cells in the adult testis: characterization, regulation and potential applications. Endocrine Reviews, 41(1), 22- 32.

Chung, B. C., Picado-Leonard, J., Haniu, M., Bienkowski, M. H. P. F., Hall, P. F., Shively, J. E., & Miller, W. L. (1987). Cytochrome P450c17 (steroid 17 alpha-hydroxylase/17, 20 lyase): cloning of human adrenal and testis cDNAs indicates the same gene is expressed in both tissues. Proceedings of the National Academy of Sciences, 84(2), 407-411.

59

Clark, A. M., Garland, K. K., & Russell, L. D. (2000). Desert hedgehog (Dhh) gene is required in the mouse testis for formation of adult-type Leydig cells and normal development of peritubular cells and seminiferous tubules. Biology of reproduction, 63(6), 1825-1838.

Clarkson, J., & Herbison, A. E. (2016). Hypothalamic control of the male neonatal testosterone surge. Philosophical Transactions of the Royal Society B: Biological Sciences, 371(1688), 20150115.

Collin, O., & Bergh, A. (1996). Leydig cells secrete factors which increase vascular permeability and endothelial cell proliferation. International journal of andrology, 19(4), 221-228.

Curley, M., Gonzalez, Z. N., Milne, L., Hadoke, P., Handel, I., Péault, B., & Smith, L. B. (2019). Human Adipose-derived Pericytes Display Steroidogenic Lineage Potential in Vitro and Influence Leydig Cell Regeneration in Vivo in Rats. Scientific reports, 9(1), 1-13.

Curley, M., Milne, L., Smith, S., Atanassova, N., Rebourcet, D., Darbey, A., ... & Smith, L. B. (2018). Leukemia inhibitory factor-receptor is dispensable for prenatal testis development but is required in sertoli cells for normal spermatogenesis in mice. Scientific reports, 8(1), 1-13.

Curley, M., Milne, L., Smith, S., Atanassova, N., Rebourcet, D., Darbey, A., ... & Smith, L. B. (2018). Leukemia inhibitory factor-receptor is dispensable for prenatal testis development but is required in sertoli cells for normal spermatogenesis in mice. Scientific reports, 8(1), 1-13.

Davidoff, M. S., Middendorff, R., Enikolopov, G., Riethmacher, D., Holstein, A. F., & Müller, D. (2004). Progenitor cells of the testosterone-producing Leydig cells revealed. The Journal of cell biology, 167(5), 935-944.

Davidoff, M. S., Middendorff, R., Mueller, D., & Holstein, A. F. (2009). The neuroendocrine Leydig cells and their stem cell progenitors, the pericytes (Vol. 205). Springer Science & Business Media.

De Santa Barbara, P., Moniot, B., Poulat, F., & Berta, P. (2000). Expression and subcellular localization of SF‐1, SOX9, WT1, and AMH proteins during early human testicular development. Developmental dynamics: an official publication of the American Association of Anatomists, 217(3), 293-298.

Deb, K., Reese, J., & Paria, B. C. (2006). Methodologies to study implantation in mice. In Placenta and Trophoblast (pp. 9-34). Humana Press.

Del Moral, P. M., De Langhe, S. P., Sala, F. G., Veltmaat, J. M., Tefft, D., Wang, K., ... & Bellusci, S. (2006). Differential role of FGF9 on epithelium and mesenchyme in mouse embryonic lung. Developmental biology, 293(1), 77-89.

Dressler, G. R. (2002). Development of the excretory system. In Mouse development (pp. 395- 420). Academic Press.

60

Durairajanayagam, D. (2018). Lifestyle causes of male infertility. Arab Journal of Urology, 16(1), 10-20.

Eggers, S., Ohnesorg, T., & Sinclair, A. (2014). Genetic regulation of mammalian gonad development. Nature Reviews Endocrinology, 10(11), 673.

Eleftheriou, B. E., & Lucas, L. A. (1974). Age-related changes in testes, seminal vesicles and plasma testosterone levels in male mice. Gerontology, 20(4), 231-238.

Eliveld, J., van Daalen, S. K., de Winter-Korver, C. M., van der Veen, F., Repping, S., Teerds, K., & van Pelt, A. M. (2020). A comparative analysis of human adult testicular cells expressing stem Leydig cell markers in the interstitium, vasculature and peritubular layer. Andrology.

Ferlin, A., Arredi, B., & Foresta, C. (2006). Genetic causes of male infertility. Reproductive toxicology, 22(2), 133-141.

Ferlin, A., Raicu, F., Gatta, V., Zuccarello, D., Palka, G., & Foresta, C. (2007). Male infertility: role of genetic background. Reproductive biomedicine online, 14(6), 734-745.

Ferlin, A., Selice, R., Carraro, U., & Foresta, C. (2013). Testicular function and bone metabolism—beyond testosterone. Nature Reviews Endocrinology, 9(9), 548-554.

Fischer, A. H., Jacobson, K. A., Rose, J., & Zeller, R. (2008). Hematoxylin and eosin staining of tissue and cell sections. Cold spring harbor protocols, 2008(5), pdb-prot4986.

Gao, F., Maiti, S., Alam, N., Zhang, Z., Deng, J. M., Behringer, R. R., ... & Huff, V. (2006). The Wilms tumor gene, Wt1, is required for Sox9 expression and maintenance of tubular architecture in the developing testis. Proceedings of the National Academy of Sciences, 103(32), 11987-11992.

Geske, M. J., Zhang, X., Patel, K. K., Ornitz, D. M., & Stappenbeck, T. S. (2008). Fgf9 signaling regulates small intestinal elongation and mesenchymal development. Development, 135(17), 2959-2968.

Gnessi, L., Basciani, S., Mariani, S., Arizzi, M., Spera, G., Wang, C., ... & Betsholtz, C. (2000). Leydig cell loss and spermatogenic arrest in platelet-derived growth factor (PDGF)-A– deficient Mice. The Journal of cell biology, 149(5), 1019-1026.

Goluža, T., Boscanin, A., Cvetko, J., Kozina, V., Kosović, M., Bernat, M. M., ... & Ježek, D. (2014). Macrophages and Leydig cells in testicular biopsies of azoospermic men. BioMed Research International, 2014.

Goodfellow, P. N., & Lovell-Badge, R. (1993). SRY and sex determination in mammals. Annual review of genetics, 27(1), 71-92.

Greil, A. L., Slauson‐Blevins, K., & McQuillan, J. (2010). The experience of infertility: a review of recent literature. Sociology of health & illness, 32(1), 140-162.

61 Hadley, M. A., Byers, S. W., Suárez-Quian, C. A., Kleinman, H. K., & Dym, M. (1985). Extracellular matrix regulates Sertoli cell differentiation, testicular cord formation, and germ cell development in vitro. The Journal of cell biology, 101(4), 1511-1522.

Haines, D. M., & Chelack, B. J. (1991). Technical considerations for developing enzyme immunohistochemical staining procedures on formalin-fixed paraffin-embedded tissues for diagnostic pathology. Journal of Veterinary Diagnostic Investigation, 3(1), 101-112.

Hamada, Y., Kadokawa, Y., Okabe, M., Ikawa, M., Coleman, J. R., & Tsujimoto, Y. (1999). Mutation in ankyrin repeats of the mouse Notch2 gene induces early embryonic lethality. Development, 126(15), 3415-3424.

Hasegawa, K., Okamura, Y., & Saga, Y. (2012). Notch signaling in Sertoli cells regulates cyclical gene expression of Hes1 but is dispensable for mouse spermatogenesis. Molecular and cellular biology, 32(1), 206-215.

He, W., Gauri, M., Li, T., Wang, R., & Lin, S. X. (2016). Current knowledge of the multifunctional 17β-hydroxysteroid dehydrogenase type 1 (HSD17B1). Gene, 588(1), 54-61.

Hobert, O., & Westphal, H. (2000). Functions of LIM-homeobox genes. Trends in genetics, 16(2), 75-83.

Hu, L., Monteiro, A., Johnston, H., King, P., & O’Shaughnessy, P. J. (2007). Expression of Cyp21a1 and Cyp11b1 in the fetal mouse testis. Reproduction, 134(4), 585-591.

Hutchison, G. R., Scott, H. M., Walker, M., McKinnell, C., Ferrara, D., Mahood, I. K., & Sharpe, R. M. (2008). Sertoli cell development and function in an animal model of testicular dysgenesis syndrome. Biology of Reproduction, 78(2), 352-360.

Ikeda, Y., Shen, W. H., Ingraham, H. A., & Parker, K. L. (1994). Developmental expression of mouse steroidogenic factor-1, an essential regulator of the steroid hydroxylases. Molecular endocrinology, 8(5), 654-662.

Inoue, M., Shima, Y., Miyabayashi, K., Tokunaga, K., Sato, T., Baba, T., ... & Morohashi, K. I. (2016). Isolation and characterization of fetal Leydig progenitor cells of male mice. Endocrinology, 157(3), 1222-1233.

Jameson, S. A., Natarajan, A., Cool, J., DeFalco, T., Maatouk, D. M., Mork, L., ... & Capel, B. (2012). Temporal transcriptional profiling of somatic and germ cells reveals biased lineage priming of sexual fate in the fetal mouse gonad. PLoS Genet, 8(3), e1002575.

Jeske, Y. W., Bowles, J., Greenfield, A., & Koopman, P. (1995). Expression of a linear Sry transcript in the mouse genital ridge. Nature genetics, 10(4), 480-482.

Jiang, M. H., Cai, B., Tuo, Y., Wang, J., Zang, Z. J., Gao, Y., ... & Jiao, J. (2014). Characterization of Nestin-positive stem Leydig cells as a potential source for the treatment of testicular Leydig cell dysfunction. Cell research, 24(12), 1466-1485.

62 Jiang, X., Skibba, M., Zhang, C., Tan, Y., Xin, Y., & Qu, Y. (2013). The roles of fibroblast growth factors in the testicular development and tumor. Journal of diabetes research, 2013.

Joseph, A., Yao, H., & Hinton, B. T. (2009). Development and morphogenesis of the Wolffian/epididymal duct, more twists and turns. Developmental biology, 325(1), 6- 14.

Josso, N., Lamarre, I., Picard, J. Y., Berta, P., Davies, N., Morichon, N., ... & Jeny, R. (1993). Anti-Müllerian hormone in early human development. Early human development, 33(2), 91-99.

Kanai, Y., Hiramatsu, R., Matoba, S., & Kidokoro, T. (2005). From SRY to SOX9: mammalian testis differentiation. Journal of biochemistry, 138(1), 13-19.

Kashimada, K., & Koopman, P. (2010). Sry: the master switch in mammalian sex determination. Development, 137(23), 3921-3930.

Kato, T., Esaki, M., Matsuzawa, A., & Ikeda, Y. (2012). NR5A1 is required for functional maturation of Sertoli cells during postnatal development. Reproduction, 143(5), 663- 672.

Kent, J., Wheatley, S. C., Andrews, J. E., Sinclair, A. H., & Koopman, P. (1996). A male- specific role for SOX9 in vertebrate sex determination. Development, 122(9), 2813- 2822.

Kerr, J. B., Bartlett, J. M. S., Donachie, K., & Sharpe, R. M. (1987). Origin of regenerating Leydig cells in the testis of the adult rat. Cell and tissue research, 249(2), 367-377.

Kim, S. W., Roh, J., & Park, C. S. (2016). Immunohistochemistry for pathologists: protocols, pitfalls, and tips. Journal of pathology and translational medicine, 50(6), 411.

Kitamura, K., Yanazawa, M., Sugiyama, N., Miura, H., Iizuka-Kogo, A., Kusaka, M., ... & Matsuo, M. (2002). Mutation of ARX causes abnormal development of forebrain and testes in mice and X-linked lissencephaly with abnormal genitalia in humans. Nature genetics, 32(3), 359-369.

Kong, D. H., Kim, Y. K., Kim, M. R., Jang, J. H., & Lee, S. (2018). Emerging roles of vascular cell adhesion molecule-1 (VCAM-1) in immunological disorders and cancer. International journal of molecular sciences, 19(4), 1057.

Kraft, P., Pharoah, P., Chanock, S. J., Albanes, D., Kolonel, L. N., Hayes, R. B., ... & Burtt, N. P. (2005). Genetic variation in the HSD17B1 gene and risk of prostate cancer. PLoS Genet, 1(5), e68.

Krebs, L. T., Xue, Y., Norton, C. R., Shutter, J. R., Maguire, M., Sundberg, J. P., ... & Smith, G. H. (2000). Notch signaling is essential for vascular morphogenesis in mice. Genes & development, 14(11), 1343-1352.

63 Kumar, D. L., & DeFalco, T. (2018). A perivascular niche for multipotent progenitors in the fetal testis. Nature communications, 9(1), 1-13.

Kumar, N., & Singh, A. K. (2015). Trends of male factor infertility, an important cause of infertility: A review of literature. Journal of human reproductive sciences, 8(4), 191.

Kumar, P., Kumar, N., Thakur, D. S., & Patidar, A. (2010). Male hypogonadism: Symptoms and treatment. Journal of advanced pharmaceutical technology & research, 1(3), 297.

Kunitomo, M., Khokhar, A., Kresge, C., Edobor‐Osula, F., & Pletcher, B. A. (2020). 46, XY DSD and limb abnormalities in a female with a de novo LHX9 missense mutation. American Journal of Medical Genetics Part A.

Lai, M. S., Wang, C. Y., Yang, S. H., Wu, C. C., Sun, H. S., Tsai, S. J., ... & Huang, B. M. (2016). The expression profiles of fibroblast growth factor 9 and its receptors in developing mice testes. Organogenesis, 12(2), 61-77.

Lamb, D. J. (2019). An approach that someday may boost testosterone biosynthesis in males with late-onset hypogonadism (low testosterone). Proceedings of the National Academy of Sciences, 116(46), 22904-22906.

Lee, M. M., Donahoe, P. K., Hasegawa, T., Silverman, B., Crist, G. B., Best, S., ... & MacLaughlin, D. T. (1996). Mullerian inhibiting substance in humans: normal levels from infancy to adulthood. The Journal of Clinical Endocrinology & Metabolism, 81(2), 571-576.

Li, S. Q., Tu, C., Wan, L., Chen, R. Q., Duan, Z. X., Ren, X. L., & Li, Z. H. (2019). FGF- induced LHX9 regulates the progression and metastasis of osteosarcoma via FRS2/TGF-β/β-catenin pathway. Cell Division, 14(1), 1-17.

Li, Y., Li, N., Yu, X., Huang, K., Zheng, T., Cheng, X., ... & Liu, X. (2018). Hematoxylin and eosin staining of intact tissues via delipidation and ultrasound. Scientific reports, 8(1), 1-8.

Lin, H., Ge, R. S., Chen, G. R., Hu, G. X., Dong, L., Lian, Q. Q., ... & Hardy, M. P. (2008). Involvement of testicular growth factors in fetal Leydig cell aggregation after exposure to phthalate in utero. Proceedings of the National Academy of Sciences, 105(20), 7218- 7222.

Marshall, A. S., Willment, J. A., Lin, H. H., Williams, D. L., Gordon, S., & Brown, G. D. (2004). Identification and characterization of a novel human myeloid inhibitory C-type lectin-like receptor (MICL) that is predominantly expressed on granulocytes and monocytes. Journal of Biological Chemistry, 279(15), 14792-14802.

Martin, L. J. (2016). Cell interactions and genetic regulation that contribute to testicular Leydig cell development and differentiation. Molecular reproduction and development, 83(6), 470-487.

64 Martin, L. J., & Touaibia, M. (2020). Improvement of Testicular Steroidogenesis Using Flavonoids and Isoflavonoids for Prevention of Late-Onset Male Hypogonadism. Antioxidants, 9(3), 237.

Mazaud, S., Oreal, E., Guigon, C. J., Carre-Eusebe, D., & Magre, S. (2002). Lhx9 expression during gonadal morphogenesis as related to the state of cell differentiation. Gene expression patterns, 2(3-4), 373-377.

Miura, H., Yanazawa, M., Kato, K., & Kitamura, K. (1997). Expression of a novel aristaless related homeobox gene ‘Arx’in the vertebrate telencephalon, diencephalon and floor plate. Mechanisms of development, 65(1-2), 99-109.

Miyabayashi, K., Katoh-Fukui, Y., Ogawa, H., Baba, T., Shima, Y., Sugiyama, N., ... & Morohashi, K. I. (2013). Aristaless related homeobox gene, Arx, is implicated in mouse fetal Leydig cell differentiation possibly through expressing in the progenitor cells. PLoS One, 8(6), e68050.

Molyneaux, K. A., Stallock, J., Schaible, K., & Wylie, C. (2001). Time-lapse analysis of living mouse germ cell migration. Developmental biology, 240(2), 488-498.

Moreno, N., Bachy, I., Rétaux, S., & González, A. (2004). LIM‐homeodomain genes as developmental and adult genetic markers of Xenopus forebrain functional subdivisions. Journal of Comparative Neurology, 472(1), 52-72.

Mueller, O., Lightfoot, S., & Schroeder, A. (2004). RNA integrity number (RIN)– standardization of RNA quality control. Agilent application note, publication, 1, 1-8.

Mulligan, T., Frick, M. F., Zuraw, Q. C., Stemhagen, A., & McWhirter, C. (2006). Prevalence of hypogonadism in males aged at least 45 years: the HIM study. International journal of clinical practice, 60(7), 762-769.

Munsterberg, A., & Lovell-Badge, R. (1991). Expression of the mouse anti-mullerian hormone gene suggests a role in both male and female sexual differentiation. Development, 113(2), 613-624

Murashima, A., Miyagawa, S., Ogino, Y., Nishida-Fukuda, H., Araki, K., Matsumoto, T., ... & Kato, S. (2011). Essential roles of androgen signaling in Wolffian duct stabilization and epididymal cell differentiation. Endocrinology, 152(4), 1640-1651.

Murta, D., Batista, M., Silva, E., Trindade, A., Henrique, D., Duarte, A., & Lopes-da-Costa, L. (2013). Dynamics of Notch pathway expression during mouse testis post-natal development and along the spermatogenic cycle. PLoS One, 8(8), e72767.

Nef, S., & Parada, L. F. (2000). Hormones in male sexual development. Genes & Development, 14(24), 3075-3086.

Nishimori, K., & Matzuk, M. M. (1996). Transgenic mice in the analysis of reproductive development and function. Reviews of Reproduction, 1(3), 203-212.

65 O’Neill, M., Zhelyazkova, B., White, J. T., Thirumavalavan, N., & Lamb, D. J. (2019). Developmental Genetics of the Male Reproductive System. In Human Reproductive and Prenatal Genetics (pp. 3-25). Academic Press.

O’Shaughnessy, P. J., Morris, I. D., & Baker, P. J. (2008). Leydig cell re-generation and expression of cell signaling molecules in the germ cell-free testis. Reproduction, 135(6), 851-858.

O’Shaughnessy, Peter. (2017). The Human Leydig Cell. 10.1007/978-3-319-53298-1_2.

O’Shaughnessy, P. J., Fleming, L., Baker, P. J., Jackson, G., & Johnston, H. (2003). Identification of developmentally regulated genes in the somatic cells of the mouse testis using serial analysis of gene expression. Biology of reproduction, 69(3), 797-808.

O’Shaughnessy, P. J., Willerton, L., & Baker, P. J. (2002). Changes in Leydig cell gene expression during development in the mouse. Biology of Reproduction, 66(4), 966-975.

Odeh, H. M., Kleinguetl, C., Ge, R., Zirkin, B. R., & Chen, H. (2014). Regulation of the proliferation and differentiation of Leydig stem cells in the adult testis. Biology of reproduction, 90(6), 123-1.

Okamura, D., Maeda, I., Taniguchi, H., Tokitake, Y., Ikeda, M., Ozato, K., ... & Matsui, Y. (2012). Cell cycle gene-specific control of transcription has a critical role in proliferation of primordial germ cells. Genes & development, 26(22), 2477-2482.

Ottolenghi, C., Moreira-Filho, C., Mendonça, B. B., Barbieri, M., Fellous, M., Berkovitz, G. D., & McElreavey, K. (2001). Absence of mutations involving the LIM homeobox domain gene LHX9 in 46, XY gonadal agenesis and dysgenesis. The Journal of Clinical Endocrinology & Metabolism, 86(6), 2465-2469.

Papatheodorou, I., Moreno, P., Manning, J., Fuentes, A. M. P., George, N., Fexova, S., ... & Huerta, L. (2020). Expression Atlas update: from tissues to single cells. Nucleic Acids Research, 48(D1), D77-D83.

Park, S. Y., Tong, M., & Jameson, J. L. (2007). Distinct roles for steroidogenic factor 1 and desert hedgehog pathways in fetal and adult Leydig cell development. Endocrinology, 148(8), 3704-3710.

Park, S., Zeidan, K., Shin, J. S., & Taketo, T. (2011). SRY upregulation of SOX9 is inefficient and delayed, allowing ovarian differentiation, in the B6. YTIR gonad. Differentiation, 82(1), 18-27.

Payne, A. H., & Hales, D. B. (2004). Overview of steroidogenic enzymes in the pathway from cholesterol to active steroid hormones. Endocrine reviews, 25(6), 947-970.

Pointis, G., Latreille, M. T., & Cedard, L. (1980). Gonado-pituitary relationships in the fetal mouse at various times during sexual differentiation. Journal of Endocrinology, 86(3), 483-488.

66 Pujar, S., Kothapalli, K. S., Kirkness, E., Van Wormer, R. H., & Meyers-Wallen, V. N. (2005). Exclusion of Lhx9 as a candidate gene for Sry-negative XX sex reversal in the American cocker spaniel model. Journal of Heredity, 96(4), 452-454.

Racine, C., Rey, R., Forest, M. G., Louis, F., Ferré, A., Huhtaniemi, I., ... & Di Clemente, N. (1998). Receptors for anti-Müllerian hormone on Leydig cells are responsible for its effects on steroidogenesis and cell differentiation. Proceedings of the National Academy of Sciences, 95(2), 594-599.

Rasmussen, M. K., Ekstrand, B., & Zamaratskaia, G. (2013). Regulation of 3β-hydroxysteroid dehydrogenase/Δ5-Δ4 isomerase: a review. International journal of molecular sciences, 14(9), 17926-17942.

Rétaux, S., Rogard, M., Bach, I., Failli, V., & Besson, M. J. (1999). Lhx9: a novel LIM- homeodomain gene expressed in the developing forebrain. Journal of Neuroscience, 19(2), 783-793.

Rey, R. A., & Grinspon, R. P. (2011). Normal male sexual differentiation and aetiology of disorders of sex development. Best Practice & Research Clinical Endocrinology & Metabolism, 25(2), 221-238.

Roy, A., & Matzuk, M. M. (2006). Deconstructing mammalian reproduction: using knockouts to define fertility pathways. Reproduction, 131(2), 207-219.

Schmahl, J., Eicher, E. M., Washburn, L. L., & Capel, B. (2000). Sry induces cell proliferation in the mouse gonad. Development, 127(1), 65-73.

Schroeder, A., Mueller, O., Stocker, S., Salowsky, R., Leiber, M., Gassmann, M., ... & Ragg, T. (2006). The RIN: an RNA integrity number for assigning integrity values to RNA measurements. BMC molecular biology, 7(1), 1-14.

Scientific, T. (2012). Interpretation of nucleic acid 260/280 ratios. T123 Technical Bulletin. Wilmington, DE, USA: Thermo Scientific.

Shan, L. X., Phillips, D. M., Bardin, C. W., & Hardy, M. P. (1993). Differential regulation of steroidogenic enzymes during differentiation optimizes testosterone production by adult rat Leydig cells. Endocrinology, 133(5), 2277-2283.

Sharpe, P. T. (1999). The mouse as a developmental model. In Molecular Embryology (pp. 3- 5). Humana Press, Totowa, NJ.

Sherman, B. T., & Lempicki, R. A. (2009). Systematic and integrative analysis of large gene lists using DAVID bioinformatics resources. Nature protocols, 4(1), 44.

Sinclair, A. H., Berta, P., Palmer, M. S., Hawkins, J. R., Griffiths, B. L., Smith, M. J., ... & Goodfellow, P. N. (1990). A gene from the human sex-determining region encodes a protein with homology to a conserved DNA-binding motif. Nature, 346(6281), 240- 244.

67 Sohni, A., Tan, K., Song, H. W., Burow, D., de Rooij, D. G., Laurent, L., ... & Wilkinson, M. F. (2019). The neonatal and adult human testis defined at the single-cell level. Cell reports, 26(6), 1501-1517.

Staack, A., Donjacour, A. A., Brody, J., Cunha, G. R., & Carroll, P. (2003). Mouse urogenital development: a practical approach. Differentiation, 71(7), 402-413.

Struik, D., Dommerholt, M. B., & Jonker, J. W. (2019). Fibroblast growth factors in control of lipid metabolism: from biological function to clinical application. Current opinion in lipidology, 30(3), 235.

Szczepny, A., Hime, G. R., & Loveland, K. L. (2006). Expression of hedgehog signalling components in adult mouse testis. Developmental dynamics, 235(11), 3063-3070.

Tajima, T., Fujieda, K., Kouda, N., Nakae, J., & Miller, W. L. (2001). Heterozygous mutation in the cholesterol side chain cleavage enzyme (p450scc) gene in a patient with 46, XY sex reversal and adrenal insufficiency. The Journal of Clinical Endocrinology & Metabolism, 86(8), 3820-3825.

Teerds, K. J., Rijntjes, E., Veldhuizen-Tsoerkan, M. B., Rommerts, F. F., & de Boer-Brouwer, M. (2007). The development of rat Leydig cell progenitors in vitro: how essential is luteinising hormone?. Journal of Endocrinology, 194(3), 579-593.

Teerds, K., & Rijntjes, E. (2007). Dynamics of Leydig cell regeneration after EDS. In The Leydig cell in health and disease (pp. 91-116). Humana Press.

Thomas, P. D., Kejariwal, A., Guo, N., Mi, H., Campbell, M. J., Muruganujan, A., & Lazareva- Ulitsky, B. (2006). Applications for protein sequence–function evolution data: mRNA/protein expression analysis and coding SNP scoring tools. Nucleic acids research, 34(suppl_2), W645-W650.

Vander Borght, M., & Wyns, C. (2018). Fertility and infertility: Definition and epidemiology. Clinical biochemistry, 62, 2-10.

Wandernoth, P. M., Mannowetz, N., Szczyrba, J., Grannemann, L., Wolf, A., Becker, H. M., ... & Wennemuth, G. (2015). Normal fertility requires the expression of carbonic anhydrases II and IV in sperm. Journal of Biological Chemistry, 290(49), 29202- 29216.

Wang, W., Wei, S., Li, L., Su, X., Du, C., Li, F., ... & Xu, G. (2015). Proteomic analysis of murine testes lipid droplets. Scientific reports, 5, 12070.

Wang, Y., Chen, F., Ye, L., Zirkin, B., & Chen, H. (2017). Steroidogenesis in Leydig cells: effects of aging and environmental factors. Reproduction, 154(4), R111-R122.

Warne, G. L., & Hewitt, J. K. (2009). Disorders of sex development: current understanding and continuing controversy. Med J Aust, 190(11), 612-613.

68 Waterston, R. H., Lindblad-Toh, K., Birney, E., Rogers, J., Abril, J. F., Agarwal, P., ... & Antonarakis, S. E. (2002). Initial sequencing and comparative analysis of the mouse genome. Nature, 420(6915), 520-562.

Weinmaster, G. (1997). The ins and outs of notch signaling. Molecular and Cellular Neuroscience, 9(2), 91-102.

White, S., Ohnesorg, T., Notini, A., Roeszler, K., Hewitt, J., Daggag, H., ... & Miles, D. (2011). Copy number variation in patients with disorders of sex development due to 46, XY gonadal dysgenesis. PloS one, 6(3), e17793.

Wilhelm, D., & Englert, C. (2002). The Wilms tumor suppressor WT1 regulates early gonad development by activation of Sf1. Genes & development, 16(14), 1839-1851.

Wright, E., Hargrave, M. R., Christiansen, J., Cooper, L., Kun, J., Evans, T., ... & Koopman, P. (1995). The Sry-related gene Sox9 is expressed during chondrogenesis in mouse embryos. Nature genetics, 9(1), 15-20.

Yamazaki, F., Møller, M., Fu, C., Clokie, S. J., Zykovich, A., Coon, S. L., ... & Rath, M. F. (2015). The Lhx9 homeobox gene controls pineal gland development and prevents postnatal hydrocephalus. Brain Structure and Function, 220(3), 1497-1509.

Yang, Y. (2018). The regulation and function of the transcription factor Lhx9 during mouse limb and urogenital development. (Thesis, Doctor of Philosophy). University of Otago. Retrieved from http://hdl.handle.net/10523/7841

Yang, Y., Li, Z., Wu, X., Chen, H., Xu, W., Xiang, Q., ... & Huang, Y. (2017). Direct reprogramming of mouse fibroblasts toward Leydig-like cells by defined factors. Stem cell reports, 8(1), 39-53.

Ye, L., Li, X., Li, L., Chen, H., & Ge, R. S. (2017). Insights into the development of the adult Leydig cell lineage from stem Leydig cells. Frontiers in physiology, 8, 430.

Yu, H., Pask, A. J., Hu, Y., Shaw, G., & Renfree, M. B. (2014). ARX/Arx is expressed in germ cells during spermatogenesis in both marsupial and mouse. Reproduction, 147(3), 279- 289.

Zarkower, D. (2001). Establishing sexual dimorphism: conservation amidst diversity?. Nature Reviews Genetics, 2(3), 175-185.

Zirkin, B., Papadopoulos, V., & Hardy (deceased), M. (2009). Leydig cell development and function. In L. Lipshultz, S. Howards, & C. Niederberger (Eds.), Infertility in the male (pp. 29-47). Cambridge: Cambridge University Press. doi: 10.1017/CBO9780511635656.004

Zong, Y., Panikkar, A., Xu, J., Antoniou, A., Raynaud, P., Lemaigre, F., & Stanger, B. Z. (2009). Notch signaling controls liver development by regulating biliary differentiation. Development, 136(10), 1727-1739.

69 Appendix 1: Buffers and Solutions

10x PBS

18.6 mM Na2H2PO4

84.1 mM Na2HPO4 1.75 M NaCl

Diluted to 1x with dH2O (1:10)

For PBT add 0.025% Tween-20

2% Agarose 2 g molecular grade agarose 100 mL 1x TAE buffer Dissolve in microwave Cool, add 4 L RedSafe

50x Tris-acetate-EDTA (TAE) buffer 49 mM Tris base 20 mM acetic acid 1 mM EDTA in H2O

Diluted to 1x with ddH2O (1:50)

Sodium Citrate buffer 2.94 g Trisodium citrate

Make up to 1 L with H2O Adjust pH to 6.0 with 1M HCl

Proteinase K Buffer 2.5 mL Tris-Cl pH 7.5 0.5 mL EDTA

Make up to 50 mL with H2O

70

Lysis Buffer 2.5mL Tris pH 8 0.093g EDTA 0.585g NaCl 100 L SDS Make up to 50mL with water 200 L with 1.4 L proteinase K.

71 Appendix 2: RT-qPCR Primer Sequences

Table 13: Forward and reverse RT-qPCR primer sequences Gene Target Forward Primer (5’ – 3’) Reverse Primer

3-HSD CCAGGGAGCAATTCTTCAACCT GTGGATAACAACAGAGATGCCC

Actb GGCTGTATTCCCCTCCATCG CCAGTTGGTAACAATGCCATGT

Arx GAGCTCCGGCTGAGGAGA TCTCAGAGCAGCCCTCTTCC

Fgf1 AGGATCCTTCCTGATGGCAC GGTGTCTGCGAGCCGTATAA

Lhx9 GCCAAGGACGGTAGCATTTA AGCTCAGATGGTAGACAGAGT

Lifr GACTGCTCCTTCAGGACTGC GAACACAGTTTCCACCAGCC

Nestin TTTCCTGACCCCAAGCTGAAG GGGTATTAGGCAAGGGGGAAG

Notch ACAGTGCAACCCCCTGTATG TCTAGGCCATCCCACTCACA

P450c17 GAGTTTGCCATCCCGAAGGA GAAGCGCTCAGGCATAAACC

P450scc GGCACTTTGGAGTCAGTTTA CACCTCTTGGTTTAGGACGA

Patched1 GGCCCCGGGAAATTAATAAAAGG CCAGTAGCCTTCCCCATAGC

Pdgfra GGAACCTCAGAGAGAATCGGC CATAGCTCCTGAGACCCGCT

Rps29 TGAAGGCAAGATGGGTCAC GCACATGTTCAGCCCGTATT

Vcam TACTGTTTGCAGTCTCTCAAGC CGTAGTGCTGCAAGTGAGGG

72 Appendix 3: Primer Efficiency Tests

73

74 Appendix 4: RT-qPCR Dissociation Curves

Actb Lhx9

Fgf1 Rps29

P450scc Pdgfra

Vcam P450c17

75 Notch Nestin

76 Appendix 5: DEG List from RNA-Seq Analysis

Table 14: DEGs higher expressed in HET samples (FDR < 0.1)

ENTREZ_GENE_ID logFC PValue FDR 232413 -2.0734481 2.65E-06 0.0038083

100042355 -2.0681159 8.59E-16 1.23E-11

100736249 -2.0288927 0.00017559 0.07342035

22526 -1.981199 7.25E-07 0.00118482

20359 -1.9495702 8.83E-05 0.05145929 17068 -1.7726812 0.0001074 0.05679469

14129 -1.7470464 0.00014873 0.07037115 100504361 -1.6050829 0.00024542 0.09193148

12608 -1.5640013 7.06E-06 0.00705331

12310 -1.5540218 0.00021638 0.08645482 68195 -1.324456 4.88E-05 0.03376457

12259 -1.2446698 3.62E-05 0.02710474 11522 -1.2440153 6.75E-05 0.04335703

80891 -1.221175 1.02E-05 0.00912645

100271704 -1.2164418 0.00022788 0.087178

30794 -1.1973013 0.00025125 0.09219362

75140 -1.1102941 9.30E-05 0.05145929 665155 -1.1052984 6.09E-05 0.0405286

66995 -0.9990849 0.00017102 0.07342035

68267 -0.9940838 0.00016913 0.07342035 18826 -0.9568896 7.79E-05 0.0482836

77 Table 15: DEGs lower expressed in HET samples (FDR < 0.1)

ENTREZ_GENE_ID logFC PValue FDR

13984 0.76937415 9.44E-05 0.05145929

20249 0.82276239 2.69E-05 0.02106363 14600 0.90228994 0.00018334 0.07492095 277468 0.91229317 0.00022697 0.087178 100862375 1.0576976 0.00013148 0.06566652

57390 1.09129466 0.00025762 0.09264095

16011 1.15840785 3.68E-06 0.00440579 67522 1.17727851 3.66E-06 0.00440579 258438 1.22328913 0.00014815 0.07037115

14311 1.25714774 0.00016961 0.07342035 73889 1.26070128 2.41E-05 0.01966604

11770 1.43711508 3.49E-10 1.26E-06 619991 1.44524338 2.75E-06 0.0038083

15439 1.52733748 6.90E-07 0.00118482 16956 1.53285835 1.02E-07 0.0002045 69629 1.58412382 0.00026846 0.09464673

231004 1.67026852 5.78E-06 0.0061164

110257 1.69838116 5.06E-06 0.00568701 13106 1.77812954 1.88E-09 4.83E-06 75342 1.84156136 0.00012634 0.0649028 11537 1.92374945 4.30E-11 1.93E-07

100502669 2.03654264 9.34E-05 0.05145929

100038453 2.15098014 0.00017468 0.07342035 19817 2.15890725 1.24E-14 7.46E-11 71052 2.17282964 1.17E-05 0.01001389 14164 2.19943599 7.77E-06 0.00735727

100038508 2.36930831 4.51E-05 0.03242912 170942 2.40110091 3.59E-08 8.07E-05 12350 2.42206893 1.37E-15 1.23E-11 380994 2.80536032 5.43E-10 1.63E-06

78