Mammal specific protein phosphatase isoform, PPP1CC2, is essential for sperm function and male fertility

A thesis submitted

To Kent State University in partial

fulfillment of the requirements for the

degree of Doctor of Philosophy

By

Nilam Sinha

May, 2012

Dissertation written by

Nilam Sinha

B.S., University Of Calcutta, Kolkata, India, 1999

M.A., University of Calcutta, Kolkata, India, 2002

Ph.D., Kent State University, Kent, Ohio,USA 2012

Approved by

Srinivasan Vijayaraghavan, Ph.D, Chair, Doctoral Dissertation Committee

Douglas W. Kline, Ph.D______, Members, Doctoral Dissertation Committee

Yijing Chen, Ph.D______,

Roger B. Gregory, Ph.D______,

Soumitra Basu, Ph.D______,

Accepted by

James L. Blank, Ph.D______, Chair, Department of Biological Sciences

John R. D. Stalvey, Ph.D______, Dean, College of Arts and Sciences

ii

Table Of Contents

LIST OF FIGURES …………………………………………………………………………………….iv ‐ vii

LIST OF TABLES ………………………………………………………………………………...... viii ‐ ix

ACKNOWLEDGEMENTS ……………………………………………………………………………………………x

INTRODUCTION ………………………………………...………………………………………………….1

MATERIAL AND METHODS ………………………………………………………………………………………….33

RESULTS ………………………...………………………………………………………………..50

DISCUSSIONS …………..……………………………………………………………………………135

CLINICAL SIGNIFICANCE …………………………………………………………………………………………152

BIBLIOGRAPHY ………………………………………………………………………………...... 153

iii

List Of Figures

Fig1 Cycle of rat spermatogenesis………………………………………………………………………………...3

Fig2 Schematic showing the different stages of spermatogenesis in mouse and men………6

Fig3 Mature spermatozoa…………………………………………………………………………………………..15

Fig4 Sperm head………………………………………………………………………………………………………..16

Fig5 Schematics of the different regions of mature spermatozoa…………………………………18

Fig6 Schematic diagram showing the mitochondrial sheath and its underlying ultra‐

structures…………………………………………………………………………………………………………..19

Fig7 Schematic of a cross section of sperm in the principle‐piece region showing the

organization of ODF and microtubule………………………………………………………………….20

Fig8 Schematic of a cross section in the principle piece region showing the fibrous sheath

components and the underlying structures…………………………………………………………21

Fig9 Phosphoprotein phosphatses family…………………………………………………………………...26

Fig10 Mutiple alignment showing conservation among PP1 isoforms……………………………28

Fig11 Conservation of the Ppp1cc2 specific splicing event across all 19 eutherian

mammalian species……………………………………………………………………………………………30

Fig12 Conservation of C’‐terminus tail of PPP1CC2 across all mammalian species………….30

Fig13 Mating scheme for generation of rescue mice………………………………………………………41

Fig14 Mating scheme for generation of conditional deletion of Ppp1cc gene in

Developing germ cells………………………………………………………………………………………...42

Fig15 Schematic of the exon‐intron organization of Ppp1cc genomic locus and

iv

generation of the two isoforms……………………………………………………………………...... 52

Fig16 Tissue northern blots showing differential expression of Ppp1cc isoforms……...... 55

Fig17 Schematic diagram showing the association of different cell types in the

post partum developing testis…………………………………………………………………………….57

Fig18 Northern blot showing the developmental expression of all four PP1 isoforms in

post‐natal developing testis………………………………………………………………………………..58

Fig19 Immunohistochemistry showing the expression and localization of PPP1CC1

isoform in adult testis…………………………………………………………………………………………60

Fig20 Tissue specificity of the 2.6 kb endogenous promoter fragment…………………...... 62

Fig21 Identification of conserved cis‐regulatory elements in the promoter region

of Ppp1cc gene……………………………………………………………………………………………..64‐65

Fig22 EST expression profile for transcription factors Spz1, A‐Myb and Sp1………………….67

Fig23 Promoter deletion fragments used in reporter assay…………………………………………...69

Fig24 Analysis of the Ppp1cc gene promoter using reporter assay………………………………………..69‐70

Fig25 Expression of PP1 regulators closely matches Ppp1cc2 expression in postnatal

developing testis………………………………………………………………………………………………..72

Fig26 The design of the mini gene cassette for Endogenous promoter driven

transgenic expression of PP1CC2………………………………………………………………………..76

Fig27 The design of the mini gene cassette for Pgk2 promoter driven transgenic

expression of PP1CC2…………………………………………………………………………………………78

Fig28 Comparison of levels of transgenically expressed PPP1CC2 in testis across all

transgenic rescue lines……………………………………………………………………………………….80

v

Fig29 Comparison of levels of PPP1CC2 incorporation in spermatozoa across different

transgenic rescue lines……………………………………………………………………………………….82

Fig30 Immunological detection of PPP1CC2 in testis of rescue lines………………………………84

Fig31 Haematoxylin stained testis sections of rescue mice compared to

that of control…………………………………………………………………………………………………….86

Fig32 Morphological abnormalities of spermatozoa as seen in rescue lines

expressing low levels of PP1CC2…………………………………………………………………………88

Fig33 Motility analysis of mature caudal sperm from rescue animals with varying

levels of PPP1CC2……………………………………………………………………………………………...91

Fig34 Over expression of PPP1CC2 in testis of wildtype animal………………………………..94‐95

Fig35 Ppp1r3 expression remains unchanged in wildtype testis overexpressing

Ppp1cc2………………………………………………………………………………………………………………….96

Fig36 Design of the construct for endogenous promoter driven EFP‐PPP1CC2 (C‐tail)

fusion protein…………………………………………………………………………………………………….99

Fig37 The PPP1CC isoforms vary only at their extreme C‐termini………………………………..101

Fig38 The design of the mini gene cassette for endogenous promoter driven transgenic

expression of PPP1CC1……………………………………………………………………………………102

Fig39 Detection of PPP1CC1 across various transgenic rescue lines by

western blot.……………………………………………………………………………………………………104

Fig40 Comparison of PPP1CC1 levels between other transgenic Ppp1cc1

rescue lines.…………………………………………………………………………………………………….105

Fig41 Detection of Ppp1cc1 transgene and its message in rescue lines…………………………107

Fig42 Sperm numbers were significantly lower in rescue line e1Tg‐F2 with low

vi

levels of PPP1CC1…………………………………………………………………………………………….109

Fig43 Gross morphological defects in sperm isolated from e1Tg‐F2 line…………...... 110

Fig44 The plan for generating the Ppp1cc2 isoform specific knock‐out mice…………112‐113

Fig45 Generation of novel Ppp1cc1 transgene construct……………………………………………...114

Fig46 Schematic diagram showing the floxed allele of the Ppp1cc gene………………………..118

Fig47 Confirmation of deletion of Ppp1cc gene in developing germ cells in

testis………………..………………………………………………………………………………………………120

Fig48 PPP1CC2 levels are drastically reduced in testis of GcKO mice…………………….122‐123

Fig49 Immunohistochemistry showing mosaic expression of PPP1CC2 in

the testis of GcKO mice…………………………………………………………………………… ……….125

Fig50 PPP1CC1 level remains unaltered in the testis of GcKO male mice………………………………….127

Fig51 Morphological abnormalities of spermatozoa as seen in GcKO males

with reduced levels of PPP1CC2………………………………………………………………………..131

Fig52 Motility analysis of mature caudal sperm from GcKO mice with drastic reduced

levels of PPP1CC2…………………………………………………………………………………………….132

vii

List Of Tables

Table­1, Testis specific genes for which somatic homologues exist…………………………...8

Table­2, Examples of unique testis specific genes

(do not have somatic homologue)………………………………………………………………9

Table­3, Genes with testis predominant expression………………………………………………….9

Table­4, A list of alternatively expressed gene products in testis…………………………….11

Table­5, Studies on the role of involvement of miRNAs in spermatogenesis…………….13

Table­6, Differences in length of sperm flagellum across mammalian species………….18

Table­7, List of Ca2+ channels present in spermatozoa………………………………………...... 23

Table­8, Primers used for generation of promoter deletion fragments…………………….48

Table­9, List of probes used for Northern blot analysis…………………………………………..49

Table­10, List of transcription factors identified within the evolutionary conserved

region of the Ppp1cc promoter region……………………………………………….66‐67

Table­11, Statistics on founder lines for Endogenous‐Ppp1cc2‐SV40………………………..78

Table­12, Statistics of founder line for Pgk2 –Ppp1cc2 transgenic lines…………………….79

Table­13, Densitometry analysis to determine the PPP1CC2 levels across various

transgenic lines………………………………………………………………………………………82

Table­14, Comparison of PPP1CC2 levels, testis weight, sperm number and morphology

between transgenic rescue lines and control animals………………………………..90

Table­15, Fertility data……………………………………………………………………………………………94

Table­16, Statistics of rescue founder lines for Endogenous promoter driven

Ppp1cc1 lines………………………………………………………………………………………..104

viii

Table­17, Statistics of founder line derived from novel Ppp1cc1 transgene

construct.………………………………………………………………………………………………116

Table­18, Fertility data of GcKO animals………………………………………………………………..129

Table­19, Comparison of testis weight, sperm number and morphology

between GcKO and control animals………………………………………………………...131

ix

This thesis is

Dedicated To “My Parents”

ix

Acknowledgements

I would like to express my sincere gratitude and appreciation for my advisor Dr.

Srinivasan Vijayaraghavan for his support and guidance. Under his training I am able to achieve a firm foothold in field of research. I thank him for allowing me to pursue my interests in molecular genetics works in reproductive biology and for his encouragement.

I would also like to expresses my deepest gratitude to Dr. Douglas Kline and Dr.

Yijing Chen for their encouragement, advice and help whenever I needed the most. I also like to expresses my thankfulness to Dr. Stephen Pilder for his help in writing the manuscript and insightful and critical comment. I would also like to thank my other committee members Dr. Roger Gregory and Dr. Soumitra for their kind help. I am also grateful to Mike Model for his unstinted help with imaging. My gratitude also goes towards my colleagues Rumela Chakrabarty, Pawan Puri, Suraj Kadunganatill, Shandilya Ramdas,

Teja Duduki, Jibiao Lee for their help and support. I also thank my friend Palash Roy for his support and motivation.

Finally, I am deeply indebted and forever grateful to my parents Dipak Kumar Sinha

(father) and Sipra Sinha (mother) for their sacrifice, hardship and determination to provide me with the best education. I am thankful to them for instilling in me moral values and principles. I am also thankful to my wife, Sayantani for being a great friend, help and support. I would also like to thank my brother and sister‐in‐law for their advice and guidance from time to time. Also would like to thank my lovely niece Risha and my little daughter Ritika for their love.

ix

Introduction

1

2

1.1 Spermatogenesis:

Spermatogenesis is a complex testicular process that can be divided into three phases. The first phase in which spermatogonia undergo self renewal, rapid mitotic proliferation, and differentiation leading to the formation of early primary spermatocytes marks the earliest steps of commitment to the spermatogenic pathway. In the second phase mitosis concludes as primary spermatocytes, at the end of a lengthy prophase, undergo meiosis. The final phase, spermiogenesis, is marked by morphological differentiation of the haploid products of meiosis, giving rise to testicular spermatozoa. At the end of spermiogenesis, these structurally mature but functionally immature cells are released from the seminiferous epithelium into the lumen of the seminiferous tubule by a process called spermiation. Complex signaling mechanisms regulating gene expression patterns and post transcriptional/translational processes control each step of spermatogenesis, resulting in the continuous production of testicular spermatozoa from puberty through the reproductive life span of the male (1).

1.1.1 Spermatogenic stages and spermatogenic cycle:

Mammalian spermatogenesis occurs in a synchronized, cyclic pattern where the cellular associations of differentiating germ cells and Sertoli cells are maintained in a progressive and repeated fashion (2, 5). Using this information, the seminiferous epithelium can be categorized into numerous discrete “stages” based upon the cellular complement observed in a given segment of seminiferous tubule. Thorough evaluation of these cellular associations has identified 12 discrete stages of the seminiferous epithelium in mice and 14

3 stages in rats (2), Fig1. Table 1 shows a list of spermatogenic stages, duration of the cycle in various species (3).

Figure1: Diagram of the cycle of rat spermatogenesis. Adapted from Russell et.al.

1.1.2 Sertoli cells are the only somatic component of the seminiferous epithelium. The base of the cell with its nucleus lies on the basement membrane and in direct contact with the extracellular matrix. The inter‐sertoli cell junction at the basal lamina forms the blood­ testis barrier that separates the seminiferous epithelium into distinct physiological compartments. The compartment above it is in direct contact with the extracellular environment and thus modulated by it (1). The below the blood‐testis barrier is cutoff from the external influence and that is how sertoli cells regulate the microenvironment with the adluminal complex. A single sertoli cell can support a finite number of developing germ cells at various stages of development. The various junctions formed by the sertoli cell are

4 tight or occluding junctions, desmosome junctions and gap junctions and each junctional type is associated with germ cell at particular stage of development. The gap junctions allow the sertoli cell to directly communicate with germ cells through physical exchange of signaling and growth factors (2).

1.1.3 Spermatogonial Stem Cell (SSC): SSCs are undifferentiated germ cells that maintain the balance of self‐renewal and differentiation into germ cells so as to ensure continuous production of gametes throughout the reproductive life span of males. Spermatogonia lie at the interface between blood testes barrier and the extracellular milieu. The sertoli cells, myoid cells and Leydig cells creates microenvironment for the proper development of SSCs.

Spermatogonial progenitors undergo a finite number mitosis to give rise to differentiating spermatogonia that is committed to meiosis and germ cell formation. In rodents three different types of SSC’s have been identified based on cell and nuclear morphology. These are Asingle (As); Apaired (Ap); Aaligned (Aal) and A‐intermediate 1‐4 (A1‐A4) and B‐ spermatogonia. The relative levels of As, Ap and Aal (4‐6) remains constant throughout the spermatogenic cycle suggesting they are the true progenitors that serves as the reserve pool. However the abundance of A1‐A4 and B‐spermatogonia varies with cycle and stage of spermatogenesis suggesting their commitment towards germ cell differentiation. It is though that Aal gives rise to A‐intermediate and subsequently to B‐spermatogonia that are committed to germ line differentiation (7, 8). Some well characterized markers of

+ + +/− stem/progenitor spermatogonia (Asingle, Apaired and Aaligned;) are GFRα1 , PLZF , NGN3 and cKIT− and for differentiating spermatogonia (A1–4, Intermediate, B) are GFRα1−, PLZF−,

NGN3+/− and cKIT+)(9‐11). In primates two distinct pools of cells have identified these are

Adark (dark staining nucleus) and Apale (pale staining of nucleus). It is thought that Adark

5

serves as the reserve pool /the true progenitors much alike the As; Apaired in rodents whereas Apale serve as the one which gives rise to the differentiated for B1‐spermatogonia

(12, 13).

1.1.4 Developing germ cells: In rodents and primates the differentiated B‐spermatogonia arising from Aaligned and Apale spermatogonia respectively undergo mitosis to give rise to pre‐leptotene spermatocytes and thus commits itself for the first meiotic division. The prophase of the first meiotic division is the longest step during which the DNA replicates during the pre‐leptotene; leptotene and forms tetrads. It gives rise to early; mid and late pachytene spermatocytes when forming of chiasmata and crossing over takes placing shuffling the paternal genome and then undergo a short diplotene and diakinesis step where reduction division takes place to give rise to haploid secondary spermatocytes. The secondary round spermatids then undergo meiosis‐II (equational division) to give rise to four haploid round spermatids. The round spermatids then embark on a differentiation or morphogenesis step called ‘spermiogenesis’ during which they transform from round cells into elongating spermatids and finally into mature spermatozoa. The landmark changes occurring during speermiogenesis include condensation of chromatin, formation of acrosome; loss of cytoplasm and formation of tail and its apparatus. At each step during spermatogenesis the germ cell at all stages of development remains in intimate contact with the sertoli cells which acts as a nurse cells by providing the mirco‐environment and nourishment to the developing germ cells. Finally the fully differentiated forms are released into lumen of seminiferous tubules by spermiation during which they lose contact with sertoli cell processes (2, 14).

6

Figure 2: Schematic showing the different stages of spermatogenesis in mouse and men

1.2 Regulation of spermatogenesis:

Spermatogenesis occurring in testis is unique in several ways‐ firstly it is part of an adult developmental process that onsets during puberty and continues throughout the reproductive life of an adult male. Secondly it is the only cellular and physiological process that involves meiosis. Meiosis is a unique process of cell division that occurs during gametogenesis that is crucial for maintaining the right chromosomal complement. Thirdly the end product of the process is a cell type (the spermatozoa) that is morphologically and functionally unique from other cell types in the body. Fourthly and finally the process involves cell to cell contact, cross‐talk and elaborate signaling between them. To achieve all of these the process requires tight control and regulation of gene expression. Regulation of spermatogenesis has been suggested to occur at three levels (15);

(i) Intrinsic that governs the expression pattern of genes and protein that are thought to

play important role in the evolutionary conserved process of germ cell development

and differentiation;

(ii) Interactive that is at the level of Sertoli cells, thought to play role in integrating the

external information (hormonal) and modulate it and relaying the information to and

7

fro between the germinal epithelium and between the germ cells at various stages of

development.

(iii) thirdly the extrinsic that involves the hormonal or endocrine regulation of gene

expression and signaling mechanism in somatic cells of seminiferous tubules which in

turn modulates the activity of the germ cells.

Gene expression during spermatogenesis can be regulated at the following levels:

A. Transcription:

B. Post­transcription

C. Translational

1.2.1 Transcriptional control of spermatogenesis:

About one‐third of mouse genome is differentially expressed in testis, some of them are testis specific genes while others are testis predominant. Approximately ~2.3% of rat testicular transcriptome is testis specific while ~4% in mouse (Lee et al). Three major peak of expression in testis have been identified:

(i) Mitotic phase: 0‐8 days post natal; (ii) initiation of meiosis: around 14 days post

partum and (iii) initiation of spermiogenesis (post‐meiotic): around day 20 post

partum.

High levels of transcription in the testis are achieved because of up‐regulation of both general transcription factors and presence of testis specific factors for developmental expression of gene. General factors like TFIID, TFIIB, TBP and RNA Polymerase II are over‐ expressed in testis underscoring the need for transcription. Lot of testis specific factors has been predicted but evidence exists for few. The most significant of them is CREM also called

‘master regulator of transcription in testes. CREM is a c‐AMP responsive element

8 binding protein that binds to CRE (c‐AMP response elements) on promoter of target genes.

Some of the know target of CREM are transition protein‐I (16), ACE (17); CYP51. Various isoform exists for CREM due to alternative splicing however the most prevalent form of

CREM in testis is the activator form. It has been shown that CREM associates with testis specific activators ACT and KIF17b (18) for transcriptional activation of gene expression in testis (19). Other known testis specific factors include Spz1 (20); A‐Myb (21)and TLF that are testis specific. Thus testis specific expression of genes results from recruitment of testis specific elements in the promoter or by utilization of unique transcription factor combinations that acts through pre‐existing promoter elements.

Table­1: Testis specific genes for which somatic homologues exists.

9

Table‐2: Examples of unique testis specific genes (do not have somatic homologue)

Table­3: Genes with testis predominant expression

1.2.2 Post­transcriptional modifications: Two most prevalent methods of post‐

transcriptional modification in testis are:

10

1.2.2.1 Alternative splicing: Alternative splicing is a versatile process of genome

diversification. It vastly enhances the coding capacity of the genome and poses a

new level at which gene expression can be regulated. Testis (along with brain

and thymus) is a hot bed of alternative splicing event. It is estimated that 30% of

human ESTs in testis is thought to undergo alternative splicing. In testis

alternative splicing is achieved by one of the following methods or combinations

of them: i. Use of alternate exons (skipping or inclusion): This can give rise to new

functional domains to serve specific roles in testis. ii. Use of alternate promoters: This results in inclusion (or exclusion) of novel exons

(at 5’ end) or due to alternate transcription start site. This affects the N‐terminal

end of the protein. iii. Use of alternate Poly A adenylation sites (within a intron): That results exon

skipping and effects the 3’ end of the transcript or C‐tail of the protein resulting

from the transcript.

Any combination of i, ii or iii can occur resulting in testis specific variant or protein expression. This is further facilitated by testis specific splicing factors.

11

Table­4: A list of alternatively expressed gene products in testis (not a comprehensive list)

12

1.2.2.2 Micro­RNA mediated gene regulation: miRNA are single stranded non‐coding

RNAs of ~22 nucleotides in length. miRNAs are coded by miRNA genes present in intron and are transcribed as part of the gene within which it resides. miRNA genes are transcribes by RNA PolyII to produce primary‐miRNA that fold in stem loop hairpin structure. Drosha an RNaseIII endonuclease along with DGCR8 trims the ends of both strands to produce a stem‐loop intermediate 5’‐phosphate and ~2nt 3’ overhang called pre‐miRNA. The later is transported to cytoplasm where Dicer (Dcr) another RNaseIII cleaves the hair‐pin structure ~2 helical turns away from the stem‐loop to form a duplex.

The duplex is unwounded by a helicase and the strand with thermodynamically stable

5’end is preferentially loaded onto the RISC comple (RNA induced silencing complex). The

RISC then targets the complementary mRNA and depending on the miRNA‐RISC:mRNA complementarity induces either translation repression or mRNA degradation. Study from

Dicer knock out in germ cells and in sertoli cells have clearly established the role of miRNAs in testis and spermatogenesis Table‐5.

13

Table­5: Studies on the role of involvement of miRNAs in spermatogenesis.

1.2.2.3 Small RNAs (piRNA): Piwi protein has been shown to play role in germline stem cell maintenance and meiosis. A class of small RNAs that are 29‐30 nucleotide long and interact with Piwi proteins have been identified in testis called piwi‐interacting RNAs or piRNAs. Two distinct clusters have been identified that expresses at pre‐pachytene and pachytene stages of spermatogenesis. Some of these piRNAs interact with MILI and MIWI proteins (belong to AGO proteins) (23). Knock out studies involving deletion of Mili ­/­ or

Miwi ­/­ leads to infertility due to blockage at early meiosis‐I or at round spermatid stages respectively, further underscoring the role of piRNAs in germ cell development (24).

14

1.2.3 Uncoupling of transcription from translation: Uncoupling the process of translation from transcription is another mean by which spatial and temporal expression of genes are regulated in testis during spermatogenesis and germ cell development.

Transcription ceases mid‐spermiogenesis in round spermatids. Thus proteins required during late spermatogenesis are expressed during earlier stages and stored as translationally repressed messanger ribonucleoprotein particles (mRNPs). These mRNAs are delayed in their translational and only coded when the specific product are required during late spermiogenesis.

i. Translational delay using variable Poly(A) length: The classic examples are the

expression of transition protein (Tnps) and protamines (Prms). The mRNAs for

these genes are transcribed during round spermatid stage but are tranlationally

delayed till elongating spermatid stage. Translational delay in these messanger

RNA species have been shown to result from alternate use of Poly(A) length

(25). Translationally repressed mRNA (Prm1 and 2; Tnp1 and 2) are associated

with Poly(A) tail of 180 nucleotide long where as the it is shortened to 30

nucleotides to be translationally active (25).

ii. Stablization of mRNA by binding to stabilizers: Another example of uncoupling

occurs for the testis specific protein PGK2. It has been shown that a protein

called polypyrimidine tract binding protein or PTBP2 binds to the 3’UTR of the

Pgk2 mRNA after it is transcribed during meiosis and post meiotic stage and

delays its translation for 2 weeks (26). Another stabilizer named MSY2 that is

abundant in testis and is a DNA/RNA binding protein that marks specific

transcripts for cytoplasmic storage as RNPs. The target genes for MSY2 have

15

been shown to harbor Y‐box element in their promoter. Deletion of Msy2 ‐/‐ in

mice leads to spermatogenic arrest (27).

1.2 Sperm morphology:

The spermatozoa is a highly specialized structure solely aimed for transport outside the

body that is equipped with a well defined motility apparatus for traveling the length of

the female reproductive tract and fertilize the egg. Perhaps the best analogy of a mature

spermatozoon is that with a missile system; the latter is designed to have a payload (the

war head) that is equivalent to the spermatozoon head (containing the genetic material)

and the propulsion system for target delivery of the payload which is analogous to the

motility apparatus in sperm to deliver paternal DNA to the mature ovum. A mammalian

mature spermatozoon is morphologically and structurally subdivided into following

three gross regions: (a) a head; (b) connecting piece; (c) sperm flagellum, Fig 5.

16

Figure 3: Mature spermatozoa. Staining of different regions of spermatozoa using fluorescent dyes. (A)(B) Stained with mitotracker at 10x and 20x. (C) showing staining of mitochondrial sheath with mitotracker and Hoescht (DNA, sperm) that stains sperm head. (D) DIC picture of spermatozoa obtained at 60x.

1.2.1 Spermatozoa head:

The size and shape of the sperm head is species specific but also varies within species. For example rat, mouse and hamster spermatozoa in falciform‐shaped whereas human spermatozoa head is in the shape of spatula, Fig. 6.

Figure 4: Sperm head. Top panel shows differences in head shape in mouse, bull and human. Lower panel is a schematic showing the different segment or regions on sperm head.

The plasma membrane surrounding the head is divided into two regions based on composition and function. These regions or domains are:‐ (a) Acrosomal region, that includes marginal segment; principal segement and equatorial region; (b) Post‐acrosomal region which includes serrated band and posterior ring. The latter separates the head from the connecting piece. The sperm head contains the cytoplasmic organelles nucleus, acrosome and specialized cytoskeletal structures (1).

17

1.2.2 Sperm nucleus houses the paternal DNA (genetic material), that is highly compacted and condensed and occupies a much lesser volume than average cell. The components of the sperm nucleus include nuclear proteins; nuclear lamina and the nuclear envelope.

Sperm acrosome is a specialized secretory vesicle that caps the anterior part of the nucleus. The components include inner acrosomal membrane that closely overlays the nucleus; outer acrosomal membrane that underlies the plasma membarne. The contents of the acrosomal vescicle include proteases, acid hydrolases and bio‐active peptides. They help in dissolving the zonal pellucid and entry of the sperm through the outer investments of eggs.

1.2.3 The sperm head cytoskeleton play a major role in defining the head shape. It has three distinct regions. They are; (i) Sub‐acrosomal cytoskeleton (between acrosome and nucleus); (ii) Post‐acrosomal cytoskeleton (between nucleus and sperm plasma membrane) posterior to acrosome and (iii) Para‐acrosomal cytoskeleton (between anterior tip of acrosome and plasma membrane (present only in falciform sperm) (1).

1.2.4 Sperm flagellum:

The flagellum of the sperm is chief component that generates the forces responsible for the propulsion of the sperm through the female reproductive tract. The different ultrastructural components of the sperm is responsible for developing various wave pattern and characteristics motility patterns from the tip to end of the tail. Mammalian spermatozoa vary considerably in their length of head and flagellum between various species (see Table‐1).

18

Table­6: Differences in length of sperm flagellum across mammalian species (adapted from Maximilano et. al).

The flagellum of mammalian spermatozoa consists of four well demarcated segments:

The connecting piece: It originates from centrioles and connects the head with the rest of the flagellum. It consists of basal plate, implantation fossa, segmented column and the

capitulum.

i. Mid­piece: The segment of the tail enclosed by

the helically wrapped mitochondria is

19

called as mid‐piece. The length is highly variable between species. The

mitochondria generates ATP solely by glycolysis for sperm motility. Oxidative

phosphorylation does not take place (28).

Figure 5: Schematics of the different regions of mature spermatozoa.

Figure 6: Schematic diagram showing the mitochondrial sheath and its underlying ultra­structures.

ii. Principle­piece: The length of the tail region that is enclosed by the fibrous sheath is known as the principle‐piece. It is the longest region of the flagellum. iii. End­piece: The fibrous sheath ends about ~10 µm from the tip of the tail and the terminal segment of the axoneme thus covered only by the flagellar plasma membrane constitutes the end‐piece, Fig7. iv. Ultra­structural components of the sperm flagellum:

The components are: (a) Axoneme; (b) Outer dense fibers (Odf’s) and the (c) fibrous sheath

(FS). Of these the latter two Odf’s and FS are present exclusively in mammalian spermatozoa.

The axoneme: Runs along the entire length of sperm from the connecting –piece to the tip of the tail (MP, PP and EP). Like all other flagellum constitute 9+2 arrangements of microtubules. Of which the 2 central pair is surrounded by the 9 peripheral doublets. Each doublet is made up of outer ‘A‐tubules’ and inner ‘B‐tubule’. The A tubule bears the dyenin

20 arms ‘inner’ and ‘outer’. The outer arm makes contact with the adjacent B‐tubule of the preceding doublet. The dyenin arms possess ATPase activity necessary for the sliding forces between the tubules. The tubules are made up of α‐tubulin and β‐tubulin. Tetkins forms a major protein of the tubules.

Figure 7: Schematic of a cross section of sperm in the principle­piece region showing the organization of ODF and microtubule.

The outer dense fibers (ODF’s):

Formation of Odf’s starts in spermatids during spermiogenesis. It is a mammal specific feature and though to associate with species with internal fertilization. It confers elastic properties to the sperm. There are 9 Odfs and each corresponds to the nine outer doublets of the axoneme giving a 9+9+2 arrangement, Fig9.

It starts in the mid‐piece and continues almost to the end of the principle‐piece. The proteins components of Odf’s are highly cysteine rich that are cross‐linked by di‐sulphide bonds. The major constituent proteins are ODF1‐4, SPAG4/5. Among other unknown functions it is thought to regulate tail beat by determining its elastic property; provides mechanical protection against shearing forces experienced during epididymal maturation, ejaculation and the highly rough terrain in the female tract. Also known to act as a scaffold for signaling proteins like AKAPs and voltage dependent anion channels.

The fibrous sheath: Is formed throughout spermiogenesis from dital to proximal direction. It is another mammal specific structure of mammalian spermatozoa. It closely underlies the plasma membrane and consists of two longitudinal columns connected by circumferential ribs. The longitudinal ribs overlies peripheral to Odf 3 and 8. The fibrous

21 sheath proteins are crosslinked by di‐sulphide bonds that stabilizes the sheath. Much like

Odf it acts as scaffold for signaling proteins like AKAP4, AKAP3, TAKAP, SP17 and proteins

involved in glycolytic pathway that includes:

HK‐1 (Hexokinase‐1); GAPD

(glyceraldehydes 3‐phosphate

dehydrigenase); PDHB (Pyruvate

dehydrogenase), PGK2 (Phosphoglycerate

kinase). FS controls the flagellar movement by imposing restriction of sliding of axonemal microtubular doublets (1, 28).

Figure 8: Schematic of a cross section in the principle piece region showing the fibrous sheath components and the underlying structures.

1.3 Sperm motility:

Motility initiation and sustenance of motility is a prime requisite for fertilization to occur.

Motility is required for sperm to travel the female reproductive tract and also to penetrate the cumulus oophorous and zona pellucid of the egg. Testiculur spermatozoa released by spermiation into seminiferous tubules are immotile. Motility is absent in caput region of epididymis. However acquisition of motility or the potential for motility develops as sperm moves through the corpus to the caudal region of vasa differentia. Though caudal sperm acquire motility they are held in a quiescent state due to viscosity drag of the caudal fluid.

Ejaculated sperm initiate motility once they are deposited into female tract. Sperm exhibit two types of motility: progressive motility and hyperactivated motility.

Progressive motility is developed during transit through epididymis. Progressive motility is marked by vigorous movement, symmetrical tail beat and rapid forward movement.

22

However motility in immature spermatozoa is markedly different and characterized by erratic or asymmetric beats with static head and devoid of net propulsion.

Hyperactivated motility is marked by high amplitude asymmetrical beats of tail; lateral vigorous movement of tail and circular or erratic trajectory of path. Sperm develops hyperactivated motility during its transit through female reproductive tract in response modulators present in the environment in the female tract (1).

1.4 Biochemical regulators of sperm motility:

1.4.1 Intracellular pH [pH]i: High acidity or low pH inhibits sperm motility. Inhibition is caused by disruption of the ATPase activity of the outer Dyenin arms. Lactic acid present in epididymal fluid enters sperm and lowers pHi. That is how sperm motility is inhibited in bull epididymis. Increase in pHi restores motility. Proposed mechanisms for intracellular alkalinization in sperm include K+‐dependent proton release; Na+‐H+ exchange; Na+‐Cl—

HCO3‐ dependent H+ efflux or some ion dependent mechanism.

1.4.2 Ca2+/ [Ca2+]i: Concentration of intracellular Ca2+ is key factor that regulates various aspects of sperm motility. Various genes that code for Ca2+ ion channel influx and efflux have implicated for sperm function by affecting their motility (29). Some key genes that are known to regulate intracellular Ca2+ levels and role in sperm function are listed in

+ Table‐7. [Ca2 ]i is thought to modulate sperm function by its ability to regulate soluble adenylate cyclase thereby controlling cAMP and PKA activity.

23

Table­7: List of Ca2+ channels present in spermatozoa.

1.4.3 Protein kinase A (PKA):

Cyclic adenosine monophosphate is a key second messenger in the regulation of sperm motility. An increase in cAMP levels occurs when adenylyl cyclase is stimulated to convert

ATP to cAMP. A major result of this in sperm is the activation of cAMP‐dependent kinase A

(PKA), which phosphorylates serine and threonine residues on neighboring proteins to trigger a cascade of protein phosphorylation events. This occurs primarily in the flagellum and results predominantly in phosphorylation of proteins on tyrosine residues. Increase in levels of tyrosine phosphorylation is associated with sperm motility and capacitation. PKA

24 is thought to play role in these by three ways‐ (i) by directly activating tyrosine kinases; (ii) by inhibiting tyrosine phosphatses and (iii) by priming protein for tyrosine phosphorylation by phosphorylating them at Ser/Thr residues. Sperm appear to depend primarily on testis specific soluble adenylyl cyclase (sAC) for its PKA activity (30, 31). The

PKA holoenzyme consists of two catalytic subunit (C) and two regulatory subunits (R). In both human and mouse four genes (RIα, RIβ, RIIα and RIIβ) code for the regulatory region, where as the catalytic subunit is coded by three gene (Cα, Cβ and Cγ) in humans and two subunits (Cα, Cβ) in mouse. In turn the Cα exists as two alternatively spliced variant, somatic form Cα1, is expressed in pre‐meiotic stages of germ cell development and in

Sertoli cells, whereas Cs (Cα2) is expressed in meiotic and post‐meiotic developing germ cells, a situation analogous to the expression of the somatically ubiquitous isoform

(PPP1CC1) and male germ cell restricted isoform (PPP1CC2) of Ppp1cc, respectively (). In addition, utilization of an alternate transcription start site providing testis specific Cs with a unique amino terminus replacing the amino terminus of the somatic form () which is responsible for binding to AKAP’s. The presence of proteins in the fibrous sheath with PKA anchoring sites (AKAP3, AKAP4, and TAKAP‐80) strongly suggests that one of the major roles of this structure is to anchor PKA in the principal piece region of the flagellum. San

Augustin et. al, showed that the average number of Cs subunit/sperm to be in the order of

4 x 105 molecules. The relatively large number of the Cs subunits is consistent to multiple regulatory subunit isoforms (RIα, RIβ, RIIα) located in various sperm structures (ODF, fibrous sheath). The isoform specific knockout of the Cs in mouse lead to male infertility due to failure to hyperactivate (). The high abundance of Cs subunits in sperm is attributed to tethering of the catalytic subunit to sperm tail structures in cAMP independent manner

25

thereby limiting its free diffusion. To overcome this restriction posed by the sperm tail

structure there is need for Cs subunits to be present in relatively high stoichiometry with

that of its substrate molecules. This hypothesis is further supported by the observation that

defective sperm and male sub‐fertility characterize RIα +/‐ haploinsufficient mice, possibly due to

increased unregulated PKA catalytic activity. Significantly, men with mutations in the PRKAR1A

gene have reduced fertility due to defects in sperm morphology and azoo‐ or oligospermia.

­ 1.4.4 Bi­carbonate ions [HCO 3]i : Intracellular bi‐carbonate ions plays crucial role in the

‐ regulation of sperm motility. A requirement of HCO3 for Ca2+ induced elevation of cAMP

have been shown in guinea pig spermatozoa by Garbers et al., (1982). More recently it has

been show using sAC isolated from rat sperm or using recombinant rat and human sAC that

­ this HCO3 acting directly on sAC. The soluble adenylyl cyclase (sAC) of mammalian sperm

− 2+ ­ is activated by HCO3 and this effect is synergized by Ca . HCO 3 increased sAC activity in

vitro in two ways, by increasing enzyme velocity and by relieving substrate inhibition that

− occurs at high ATP‐Mg2+ concentrations (32). The involvement of HCO3 in capacitation

and the acrosome reaction of mammalian spermatozoa have been demonstrated by Lee et

+ − al., (1986). One study showed the existence of an electrogenic Na :HCO3 cotransporters in

mouse sperm plasma membrane involved in membrane hyperpolarization and increase in

intracellular pH (pHi) leading to capacitation (Demarco et al., 2003). Very recently a sperm‐

specific Na+/H+ exchanger (sNHE) has been shown to be critical for expression and in vivo

bicarbonate mediated regulation of the soluble adenylyl cyclase (sAC), Wang et. al., (2007).

1.5 PP1 phosphatases

26

Phosphoprotein phosphatsaes 1 (PP1) are class of ser/thr phosphatses classified based on their specific ability to dephosphorylate the β‐subunit of Phophorylase kinase and also whether they themselves can be inhibited by the heat stable, acid‐stable inhibitory protein I (I‐1) and inhibitor protein‐2 (I‐2). The second major class is PP2 (that comprises

PP2A, PP2B and PP2C), that specifically dephosphorylates α‐subunit of Phophorylase kinase and their insensitive towards I‐1 and I‐2 inhibition.

Figure 9: Phosphoprotein phosphatses family. Note that he entire catalytic core is identical between PP1 and PP2A.

Type 1 serine/threonine protein phosphatases (PP1) belong to the PPP (phospho‐ protein phosphatase) gene family. The PP1 protein phosphatases are present in various eukaryotic organisms ranging from budding yeast to mammals. They play key roles in glycogen metabolism, muscle contraction and cell division (33, 34). Four protein isoforms of PP1 (PPP1CA, PPP1CB, PPP1CC1, and PPP1CC2) are ultimately derived from three genes in mammals (Ppp1ca, Ppp1cb, Ppp1cc). Ppp1cc1 and Ppp1cc2 are differentially spliced products of the Ppp1cc gene (35), an event that does not take place in non‐mammalian

27 species. While three PP1 isoforms, PPP1CA, PPP1CB, and PPP1CC1, are expressed in a wide range of tissues (36), PPP1CC2 is predominantly expressed in testis where its expression is restricted to meiotic and post‐meiotic germ cells (37‐39).

1.5.1 PP1 isoforms are highly conserved: The PP1 isoforms share a high degree of amino acid sequence similarity (~ 90%), and are, thus, among the most evolutionarily conserved proteins known (40, 41). This high level of conservation allows any of the mammalian PP1 isoforms to complement yeast (S. cerevisiae) lacking its native PP1 (Glc7p) (42). The differences between the primary sequences of mammalian PP1 isoforms reside mostly in their extreme C termini, the functions of which are unknown (43).

28

29

Figure 10: Mutiple alignment showing conservation among PP1 isoforms. ClustalW alignment of PPP1CC1, PPP1CC2, PPP1CA and PPP1CB sequences of human (Hu), Mouse (Mo) and Rat. The only difference exists is at their COOH terminal (boxed region).

1.5.2 PPP1CC2 is a mammal specific isoform:

The PPP1CC2 isoform with its unique 22 amino acid COOH‐tail (35, 44) is present only in mammals. Remarkably, this 22 amino acid C‐terminus of PPP1CC2 is virtually unchanged in all mammals for which annotated genomic databases exist (www.ensembl.org), Fig 14. This high degree of conservation suggests a role for the C‐terminus in interactions specific to

PPP1CC2. Not only the aminoacid sequence but the nucleotide sequence flanking the splice donor and acceptor site are as well conserved, suggesting that the splicing event that leads to formation of the Ppp1cc2 are also conserved in the mammals, Fig. 13.

30

Figure 11: Conservation of the Ppp1cc2 specific splicing event across all 19 eutherian mammalian species. Blastn of the Ppp1cc genomic sequence across all eutherian mammals (for which sequence information exists in ENSEMBL) shows that the splice donor (GT, boxed) in exon‐7 and splice acceptor site (AG, boxed) in exon‐8 is 100% conserved at the nucleotide level. Intervening sequences and sequences flanking the donor and acceptor site are as well conserved. The mouse and the rat sequence are shown at the top. ‘//’ denotes the intervening sequence between the two exons that was not shown.

Figure 12: Conservation of C’­terminus tail of PPP1CC2 across all mammalian species. ClustalW alignment was performed for PPP1CC2 sequence for all mammalian species for which information are available in the database. The terminal 22 aminoacids of the –COOH end is virtually conserved. The conserved residues are represented by dots whereas the one letter amino acid code denotes the non‐conserved ones. The mouse sequence is shown in bold atop. The number following the species name is the ENSEMBL ID for the respective protein sequence.

1.6 Role of PPP1CC isoforms in spermatogenesis and sperm motility:

The steady‐state phosphorylation status of a protein is determined by the relative activities of the protein kinases and phosphatases acting on that protein. Our studies have revealed a key role for PPP1CC2 in sperm motility development and possibly in regulating other sperm functions (45). Higher ser/thr protein phosphatase activity correlates with sperm

31 immotility in the caput and corpus epididymis, whereas incremental inhibition of catalytic activity is associated with progressive increases in vigorous motility in bovine and monkey cauda epididymal spermatozoa (46‐48). tion that PPP1CC2 has an isoform‐specific role in the meiotic and/or post‐meiotic phases of spermatogenesis. Targeted disruption of Ppp1cc, eliminating expression of both PPP1CC1 and PPP1CC2, results in aberrant spermiogenesis and male infertility (39). Epididymides of Ppp1cc null mice are virtually devoid of spermatozoa (39). The few spermatozoa found in testes are abnormal, due to a combination of structural defects in developing spermatids and apoptosis (49). While

Ppp1cc +/‐ mice are fertile, levels of PPP1CC2, testis weights, and sperm numbers in the epididymis are reduced by about 10 to 15 % compared to Ppp1cc +/+ mice (unpublished).

Female Ppp1cc ‐/‐ mice are fertile and healthy, suggesting that, PPP1CA and PPP1CB, can substitute for the loss of the PPP1CC isoforms in all cells and tissues, except testis. Only in males is PPP1CC2 or PPP1CC1) indispensable for normal gamete formation and fertility.

32

1.7 The three aims of my dissertation are:

Aim­I: Identification and characterization of the role of Ppp1cc gene promoter for

testis predominant expression of Ppp1cc2 isoform.

Aim­II: Determination of the isoform specific role of PPP1CC2 and PPP1CC1 in

sperm function and male fertility.

Aim­III: Determine the requirement of Ppp1cc gene in pre­meiotic germ cells and

sertoli cells

Methods

RNA isolation, probe generation for northern blot analysis:

Total RNA from testis of adult mice (8 weeks or older) was isolated using TRI reagent

(Sigma Aldrich) following the manufacturer’s protocol. Testes were dissected and pooled from littermate male pups of ages 5, 7, 10, 15, 21 and 30 days for total RNA isolation. A restriction digested and gel purified fragment spanning the full length Ppp1cc2 cDNA

(which should hybridize to messages corresponding to both Ppp1cc isoforms) was used as a probe. Ppp1cc1 mRNA is ~2.3kb whereas PP1cc2 is ~1.4kb. A 327 bp probe from the unique 3’UTR region of the Ppp1ca was PCR amplified from the Ppp1ca cDNA clone (from

ATCC#7490695) using the forward 5’‐CCTCCATGTGCTGCCCTTCTG ‐3’and reverse 5’‐

GAGAATCCAGCTTTGACCTTTATTC‐3’ primer. Similarly a ~410 bp size probe was generated from the unique 3’UTR of the Ppp1cb isoform from the cDNA clone (from ATCC#:

10698998) using the forward 5’‐TAAGGGTTAGCATTAACAAATG‐3’ and the reverse 5’‐

TTACTAAATGAGAAAACTAC‐3’ primers. Ppp1ca and Ppp1cb mRNAs are ~1.3 kb and ~4.3 kb respectively. A probe of length 1091 bp spanning the entire coding sequence Ppp1r7

(Sds22) was amplified from the cDNA clone (ATCC# MGC‐19201) using the primer pair 5’‐

CTCGAGGCCAATATGGCGGCAGAG‐3’ and 5’‐CTCAAGCTTAGGGCTCAGAACCTGACGTA‐3’.

Rsph1 probe was generated from the plasmid pCR4‐TOPO Clone#5 (kind gift from Dr.

Stephen Pilder, Temple University) using the primer pair 5’‐

CTACTCGAGATGTCGGACCTGGGC‐3’ and 5’‐CTCGAATTCTTAATCCTGGAGGTCTG‐3’ as a fragment of length ~900 bp. Probe for Ppp1r11 (I­3) was restriction digested as a ~1.6 kb

33 34

fragment from the pGEX‐4T‐2 vector harboring the full length Ppp1r11 cDNA (50). Ppp1r2

(I­2) probe was generated as a ~750 bp restriction digested (XhoI and BamHI) fragment from the plasmid harboring the entire human I‐2 coding sequence in pBluescript SK II backbone (in house). Northern blot was performed as reported previously (50, 51). The blots were re‐probed with β‐Actin probe to show equal RNA loading in the lanes.

Northern blotting:

A 17 µl mixture containing 2 μl of 10x MOPS (3‐N‐morpholino propanesulfonic acid (MOPS) buffer , 4 µl of HCHO (37% solution; Amresco, Solon, OH), 10 μl of deionized formamide

(Amresco), and 1 μl of ethidium bromide (200 μg/ml) was added to each sample of total

RNA (20 or 25 μg). The mixture was mixed by brief vortexing and then heated to 85˚ C for

10min and then chilled on ice for 10min. 2 µl of 10x gel loading dye (50% glycerol, 10 mM edetic acid [EDTA] [pH 8.0], 0.25% w/v bromophenol blue, and 0.25% w/v xylene cyanol

FF) was added to each sample and mixed before loading. Samples were electrophorectically separated on 1% Agarose/MOPS/HCHO gel at 70 volts for 3.5 to 4 hrs. Following separation, RNA was immobilized on positively charged Hybond‐XL nylon membrane (GE

Healthcare, Piscataway, NJ) by capillary transfer method in 10x SSC transfer buffer.

Transfer was allowed for 16‐18hrs overnight, followed by baking at 85˚C for 1.0 to 1.5 hrs in vaccum oven. Membrane was pre‐hybridized in 8‐10ml of Modified church buffer (1 mM

EDTA [pH 8.0], 0.5 M NaHPO4 [pH 7.2], and 5% SDS) in a sealed nylon bag in water bath at

65˚C. Probe was labeled by incorporating 32P‐dCTP using Rediprime nicktranslation kit from Amersham following manufacturer’s protocol. Probe was purified from unincorporated nucleotides by eluting them through Illustra columns (from GE) in a final

35 eluate of 400 µl of RNAase free TE buffer. The entire volume of 400 µl of the labeled probe was added to the blot along with fresh 8‐10 ml of pre‐hybridization buffer in a sealed bag and allowed to hybridize overnight at 65˚C in rocking waterbath at 50rpm. Following hybridization membrane was washed twice for 5mins in 200 ml of high stringency Wash buffer‐1 (1% 10x SSC + 0.1% SDS) followed by warm low stringency Wash buffer‐2 (0.1%

10x SSC + 0.1% SDS) once. Membrane was wrapped in plastic sheet and exposed to

Phosphor‐Imager screen and finally developed in Typhoon automated film developer (GE

Healthcare). Alternatively membranes were also exposed to X‐ray films and finally developed in Fuji film processor.

Generation of promoter deletion fragments:

All the promoter deletion constructs Del1 to Del6 were generated by PCR from the plasmid clone‐ TOPO‐3.5kb‐Up, mentioned in the preceding section. The detailed information on the sequence of primer pair and restriction site used for subcloning each of these fragments into pGL3‐basic vector is provided in Table‐1. All PCR amplifications were done using proof reading polymerase followed by sequencing of the products.

Cell culture and luciferase reporter assay:

Promoter activities of the deletion fragments (Del1‐Del7) were assayed in C18‐4 spermatogonial stem cell line (kindly provided by Dr. Marie Claude Hoffman, University of

Illinios Urbana‐Champaign) (52, 53). The C18‐4 cells were cultured in Dulbecco’s modified

Eagle’s medium (DMEM from HyClone). The medium was supplemented with 50 U/ml penicillin‐streptomycin and 10% fetal calf serum (FCS). The cells were seeded in 6 well plates at 50‐60% confluence and incubated at 37C in 5% CO2 for 24 hr to reach 80‐85%

36 confluence before transfection. Transient transfection was done using Lipofectamine 1000 reagent (Invitrogen). Renilla luciferase (pRL‐TK from Promega) was co‐transfected for use as an internal control for transfection efficiency. About 500 ul of transfection mixture‐ containing 1.5ug of total plasmid (test plasmid + pRL‐TK) to 2.5 ul of Lipofectamine (in

DMEM) was added to each well of the plate. 48 hrs post transfection the medium was removed and briefly washed with 1xPBS and lysed in 1x cold Passive lysis buffer

(Promega). Following lysis, luciferase activity was measured using Dual Luciferase

Reporter Assay kit (Promega) following the manufactures’ protocol. Transfection and subsequently luciferase activity for all the fragments (Del1 to del7) were performed in one single experiment and overall repeated thrice. Luciferase activity for each deletion construct was normalized against the activity of Renilla luciferase.

In­silico analysis of the Ppp1cc gene promoter region:

Genomic nucleotide sequences 3kb (and 10kb) immediate upstream of the ATG start codon of Ppp1cc gene and all protein sequences corresponding to PPP1CC2 for the species included in the study were obtained from ENSEMBL genome data base

(http://useast.ensembl.org/). The nucleotide sequences were analyzed using the MULAN software suite (http://mulan.dcode.org/). Multiple alignments for amino acid sequence analysis for the carboxy tail of PPP1CC2 was done using ClustalW2 tool from EMBL‐EBI site

(http://www.ebi.ac.uk/Tools/msa/clustalw2/).

37

Transgene Constructs

Endogenous promoter driven transgenic expression of PPP1CC2

The cDNA (including 5’UTR and 3’UTR) of Ppp1cc2 was PCR amplified after reverse transcription from adult mouse testicular RNA (Omniscript ReverseTranscriptase Kit,

Qiagen) using the forward primer 5’ CTCGAATTCCATCTTGTTCTTCTCGTG‐3’ (introduced

EcoRI site shown in bold letters) and the reverse primer 5’‐

CTCATCGATAGTCTGAAACCATTCTC‐3’ (introduced ClaI site shown in bold letters). The amplified fragment was inserted into a TOPO‐TA cloning vector (Invitrogen) for sequence confirmation. The Ppp1cc2 cDNA fragment released by restriction digestion with EcoRI and

ClaI was subcloned into pBluescript SK II (‐) vector. An ~210 bp SV40 poly A signal was

PCR amplified from pcDNA4.0 plasmid (forward and reverse primers pairs: 5‐

CTCCTCGAGTCTCATGCTGGAGTTCT‐3 and 5‐CTCGGTACCACCATGATTACGCCAAG‐3, respectively), and the resultant fragment was subcloned between the XhoI and KpnI sites downstream of the Ppp1cc2 cDNA. A 3.5 kb genomic region upstream of the Ppp1cc gene transcription start site (TSS) was amplified from a genomic BAC clone (Accession#

AC127266, Clone# RP24‐347B12). The primers used for PCR amplification were forward

5’‐CTAGGTACCCTGGTTGGTTCCTTC‐3’ and reverse 5’‐ CTATGAATTCATGGCCGCCGACTC‐3’

(introduced EcoRI site shown in bold letters). The amplified fragment was subcloned into a

TOPO‐TA vector (TOPO‐3.5kb‐Up) and its sequence was verified. A smaller genomic region spanning 2.6 kb upstream of the transcription start site was amplified from the larger fragment using the forward primer 5’‐GCGGCCGCATTGGATTTCAACATTC‐3’ (introduced

NotI site shown in bold letters) and the same reverse primer. This 2.6kb upstream genomic

38 fragment was subcloned between NotI and EcoRI in the vector containing the Ppp1cc2 cDNA and the SV40‐PolyA tail. The nucleotide sequence of the entire construct was verified before using it for microinjection.

Human Pgk2 promoter driven transgenic expression of PPP1CC2

The entire coding sequence for mouse Ppp1cc2 was amplified by RT‐PCR from mouse testicular RNA using a Pfu proof reading polymerase (Invitrogen), and was subsequently ligated into the pBluescript SK II (‐) vector between the BamHI and XhoI restriction endonuclease sites. Forward and reverse primers were 5’‐GTGGATCCATGGCGGATATCGAC‐

3’ and 5’‐CTCTCGAGTCACTCGTATAGGAC‐3’, respectively. A fragment containing the human Pgk2 promoter (hPgk2) was amplified from the plasmid pCR2.1 (kindly provided by

Dr. John McCarrey, University of Texas, San Antonio, USA) was subcloned into the pBluescript SK II (‐) backbone between the SacI and XbaI restriction endonuclease sites.

Forward and reverse primers used were 5‐CTCGAGCTCGAGGTTTTTACATATCA‐3 and 5‐

CTCTCTAGAGACAATATAAAGACATA‐3’, respectively. Finally the SV40 PolyA signal sequence was isolated from the endogenous promoter plasmid (described above) and introduced into the backbone between the XhoI and KpnI sites. All PCR amplifications were carried out using the proofreading Pfu polymerase (Invitrogen) and the resulting sequences were verified.

Endogenous promoter driven transgenic expression of PPP1CC1

The full length Ppp1cc1 c‐DNA (including the 5’UTR and the entire 3’UTR) was PCR amplified from the ATCC clone using the forward 5’‐CTCGAATTCCATCTTGTTCTTCTCGTG ‐

3’ (with EcoRI site introduced) and reverse primer 5’‐CTCATCGATAGTCTGAAACCATTCTC‐

39

3’ (ClaI site introduced). The PCR fragment was subcloned in pBluescript SK II (‐) vector backbone between EcoRI and ClaI site. Site directed mutagenesis was used to mutate the

‘splice acceptor site’ and ‘spice donor’ site in two steps using the QuikChange II Site‐

Directed Mutagenesis Kit (from Stratagene) following manufacturers’ protocol. In the first step mutagenesis was introduced in the acceptor site by replacing the original nucleotides

‘CAG’ with ‘ATT’ by the mutagenic primer pair: forward 5’‐

GAGTGATCTTTTTAATTTTGGATCAG‐3’ and reverse 5’‐

GAGGCCTGATCCAAAATTAAAAAGAT‐3’ to create the mutant plasmid PP1mut. In the second step mutation was introduced in the splice donor site by replacing the donor site

‘GGT’ with ‘GGC’ by the second mutagenic primer pair 5’‐

CACCACCACGGGGCATGATCACAAAGC‐3’ and 5’‐GCTTTGTGATCATGCCCCGTGGTGGTG‐3’ to give rise to Ppp1cc1‐dblMut plasmid. At every step after introduction of mutation the mutated fragment was restriction digested and subcloned into a new pBluescript SK II (‐) vector backbone between EcoRI and ClaI site and sequence was verified to confirm the mutation. The 2.6 kb upstream genomic fragment for use as putative endogenous promoter of Ppp1cc gene was excise from the Ppp1cc2‐endogenous promoter plasmid (above) and ligated between NotI and EcoRI site of the Ppp1cc1‐dblMut plasmid. Finally the SV40 PolyA signal sequence was isolated from the endogenous promoter plasmid (described above) and introduced into the backbone between the XhoI and KpnI sites. The full length of the endogenous promoter driven Ppp1cc1 mini‐gene cassette is ~ 5.1 kilobases.

40

Generation of transgenic mice

The endogenous promoter‐Ppp1cc2 cDNA ‐SV40 PolyA construct was digested with NotI and KpnI to release a ~ 4.2 kb fragment. The hPgk2‐ Ppp1cc2 ‐SV40 PolyA construct was digested with KpnI and SacI to release the ~1.7kb transgene. Likewise endogenous promoter‐Ppp1cc1‐dblMut‐cDNA‐SV40 construct was restriction digested between NotI and KpnI site to relive the mini gene cassette. The excised fragments were gel purified and micro‐injected into the pro‐nuclei of fertilized B6SJL eggs, and the injected eggs were implanted into the uteri of pseudopregnant mothers. Both microinjection and embryo implantation were carried out at the Transgenic Facility of Case Western Reserve

University (Cleveland, Ohio). Transgenic mouse production and use at Kent State

University follows approved institutional animal care and use committee protocols adapted from the National Research Council publication Guide for the Care and Use of Laboratory

Animals.

Genotyping and developing rescue lines:

Genotyping was performed by PCR with genomic DNA isolated by alkaline lysis of ear punches. The transgene in each line was detected using the 5’‐ GTGGTTGAAGATGGCTATGA

‐‘3 (exon 6) forward and 5’‐ AAGCTGCAATAAACAAGTTGG‐3’ (SV40 internal) reverse primer pair. The transgene positive B6SJL founder mice were mated with either, Ppp1cc­/­

CD1 females or Ppp1cc+/− males. Males with genotype Ppp1cc+/−; Tg+ resulting from the above cross were subsequently mated with Ppp1cc­/­ null females to derive rescue males

(Ppp1cc ‐/‐; Tg+), [see mating scheme in Fig ]. Ppp1cc­null founder mice were originally the gift of Dr. Susan Varmuza, University of Toronto, Toronto, ON).

41

Figure 13: Mating scheme employed for generation of rescue mice.

42

Figure 14: Mating scheme employed for generation of conditional deletion of Ppp1cc gene in developing germ cells.

Genotyping for detection of floxed allele and Stra­8 Cre transgene:

A transgenic mice harboring a mini‐casette expressing the CRE driven by the 1.4kb mouse

Stra‐8 promoter was purchased from the Jackson laboratories [008208 STOCK Tg(Stra8‐ cre)1Reb/J]. Forward primer 5’‐ GTG CAA GCT GAA CAA CAG GA‐3’ and reverse 5’‐ AGG

GAC ACA GCA TTG GAG TC‐3’ primer were used for detecting the trangene as ~179 bp band on 1.5% agarose gel as suggested. The Ppp1cc floxed mice were a kind gift from Dr. Angus

Nairn from Rockefeller University. The primers were RFK‐For: 5’­CTAACATAGCTTGAA

43

GTATACAACTG ­3’; RFK­87rev: 5’­CATATCTTGAGTGGTGCTTC­3’ and RFK­88rev 5’­

GAGACAGCTGACTCTACAC­3’ to distinguish between floxed allele and wildtype allele following the PCR condition suggested by the Dr. Nairns lab. The presence of a single copy of the floxed allele (+/flox) was identified as doublet band at positions 750bp and 600bp corresponding to the floxed allele and the wildtype allele respectively, where as absence of floxed allele in widtype mice (+/+) is denoted by a single band of 600bp and finally animals homozygous for the floxed allele (floxed/floxed) show a single band of size 750 bp.

Preparation of mouse testis and sperm protein extracts:

Tissues (testis, brain, spleen, liver, heart, stomach, tongue and lungs) were isolated from 8 week or older rescue mice (Ppp1cc ‐/‐, Tg+) and Ppp1cc +/‐ littermates. Tissues were homogenized in HB buffer (10 mM Tris‐HCl, pH 7.0, 1 mM EDTA, 1 mM EGTA, 10 mM benzamidine‐HCl, 1 mM PMSF, 0.1 mM N‐tosyl‐L‐phenylalanine chloromethyl ketone

[TPCK], 0.1% [V/V] β‐mecaptoethanol). The homogenates were centrifuged at 16,000×g for

20 min to remove insoluble material. Following protein estimation (as mentioned below), supernatants were mixed with 6x Laemmli sample buffer to a final concentration of 1X, and boiled for 5 min, then stored at ‐20ºC until use. Cauda epididymal sperm were collected in phosphate buffered saline as previously described (39, 51). Twice washed sperm pellets were boiled in 1% SDS. The dispersed pellets were sonicated on ice with three 10 sec bursts at 50% duty. The resulting sperm sonicates were boiled again for 5 min followed by brief centrifugation at 12,000xg for 1 min. The resulting supernatants (whole sperm extracts) were mixed with 6x Laemmli sample buffer to a final concentration of 1X, and stored at ‐20ºC until use.

44

Western Blot Analysis

Protein extracts boiled in Laemmli sample buffer and separated by 12% SDS‐PAGE were electrophoretically transferred to Immobilon‐P PVDF membranes (Millipore Corp.,

Billerica, MA, USA). After blocking non‐specific binding sites with 5% nonfat dry milk in

Tris‐buffered saline (TBS: 25 mM Tris‐HCl, pH 7.4, 150 mM NaCl), blots were incubated with primary antibodies (1:5000 dilution) overnight at 4°C. The primary antibodies anti‐

PPP1CC1 and anti‐PPP1CC2 were commercially prepared and their ability to recognize respective isoforms are well documented (39, 50, 51). After washing, blots were incubated with anti‐rabbit/mouse secondary antibody (1:5000 dilution) conjugated to horseradish peroxidase (GE Healthcare, Piscataway, NJ, USA) for 1 hr at room temperature. Blots were washed with TTBS twice for 15 min each and four times for 5 min each, and then developed with homemade ECL chemiluminescence preparation.

Quantitation of Morphology of Cauda Epididymal Sperm:

The caudae epididymides and the ductus deferens were carefully removed and placed in a petri dish containing 1x PBS. After adipose tissue and blood vessels were removed, tissues were transferred to a new petri dish containing 1ml of 1x PBS. The ductus deferens was squeezed gently with forceps to extrude sperm. The cauda was lightly minced and squeezed gently, and then incubated at 37ºC to allow remaining sperm to swim out into the medium. Extruded sperm were collected in a micro centrifuge tube and kept on ice for 30 min. The tubes were centrifuged at 600xg for 5min to pellet sperm, and the pellet was re‐

45

suspended in double distilled H20 (ddH20) and further incubated on ice to inhibit motility.

All steps involving pipetting of sperm were performed using large bore pipet tips.

For counting, sperm were diluted (1:10) in ddH20 and 10 µl of the diluted sperm were loaded onto a Neubaumer Haematocytometer. For fixation, the sperm pellet was resuspended in freshly prepared 3.75% paraformaldehyde in 1x PBS and incubated on ice for 1 hr. Fixed spermatozoa were mounted on clean slides and sealed under coverslips.

Sperm morphology was analyzed under 20x and 60x objectives with a 1x70 Olympus microscope (Melville, NY, USA) using differential interference contrast (DIC) optics.

From each slide, randomly selected fields were observed and the proportion of defective sperm was counted using a cell counter. Spermatozoa with the following morphological characteristics were counted as defective: head jack‐knifed at the connecting piece or tail with 180° hairpin bend at the midpiece‐principal piece junction, malformed heads, and shortened mitochondrial sheath (Fig 4).

Histology and immunohistochemistry

After removal, testes were immediately fixed in freshly 4% paraformaldehyde in 1x PBS for

12‐16 hrs. Fixed tissues were dehydrated by washing in a graded series of ethanols (70%,

95%, and 100%) for 45mins each, and then permeabilized in CitriSolv (Fischer Scientific) for 30 min. Tissues were then embedded in paraffin and 8 µm thick sections were transferred to poly L‐lysine coated slides. For haematoxylin staining and IHC, testis sections were de‐paraffinized in Citrisolv and rehydrated by washing in a graded series of decreasing ethanol concentration (100%, 95%, 80% and 70%) as previously reported (39,

50). Deparaffinized, rehydrated sections were stained for 2 mins with haematoxylin stain

46

(Sigma‐Aldrich, St. Louis, MO, USA). Slides were washed with distilled water twice for 2 min followed by brief immersion in freshly prepared acid rinse solution (1 ml glacial acetic acid in 49 ml distilled water) 10 times, then transferred to bluing solution (0.1% NaHCO3) for one minute, brief immersion in fresh distilled water 10 times. Sections were dehydrated through a single change of 70%, 90% and 100% ethanol each and finally immersed in

Citrisolv for one minute before mounting. Citrate Buffer (10mM Citric acid, 0.05%

Tween20, pH 6.0) was used for antigen retrieval of de‐paraffinized and rehydrated testis sections, followed by incubation in blocking media (10% goat serum, 5% BSA in 1x PBS) for

1hr at room temperature. Sections were incubated with aforementioned anti‐PPP1CC1 and anti‐PPP1CC2 primary antibodies at 1:250 dilutions at 4OC overnight followed by three brief washes in 1x PBS. Negative control sections were incubated in blocking solution instead of primary antibody. Sections were subsequently incubated with corresponding

Cy3‐conjugated secondary antibodies for 1hr at room temperature. Finally sections were washed several times for 10 min each time in 1x PBS and mounted with Vectashield

(Vector Laboratories, Burlingame, CA, USA) mounting media, then examined using a Fluo

View 500 Confocal Fluorescence Microscope (Olympus, Melville, NY, USA).

Sperm motility analysis

Cauda epididymal sperm were collected as described above into pre‐warmed modified human tubal fluid medium, m‐HTF (Irwine Scientific) without HEPES, supplemented with

BSA (5mg/ml). Sperm were incubated for 10min at 37º C with 5% CO2 to allow swim out and dispersion into the media. At the end of 10min, the petri dish containing sperm were swirled gently to allow further dispersion before measurement of motility parameters.

47

Sperm were diluted 1:10 times in m‐HTF medium (representing ~ 2 x 106 sperm/ml) and from the diluted sperm suspension 25 µl was loaded onto a chamber of 100µm Leja

Chamber 2 slide, previouslywarmed to 37ºC using a MiniTherm stage warmer. Large bore pipet tips were used for pipetting sperm suspensions. Sperm motility analysis was performed using CASA (Computer Assisted Sperm Analyzer) equipped with the CEROS sperm analysis system (software version 12.3, Hamilton Thorne Biosciences, Beverly, MA).

Motility was analyzed using the default Mouse‐2 settings from Hamilton Thorne except for minor adjustments (54). The settings for analysis were: 60frames/sec, 90 frames acquired, minimum contrast: 30, default cell size: 13 pixels, minimum cell size: 4 pixels, default cell intensity: 75, cells progressive if VAP >50 μm/sec and STR > 50%, slow cells were counted as motile, low VAP cut off: 10 μm/sec, low VSL cutoff = 0 μm/sec. For each chamber 10‐12 random non‐overlapping fields were recorded for analysis purposes. For each transgenic rescue line and control animal, motility was recorded independently from two animals (8 weeks or older) and the values for each motility parameter were expressed as mean of the two.

Protein concentration estimation and densitometry analysis:

Protein concentration of samples was estimated by the Bradford method using a DC

Protein Assay Kit (Bio‐Rad, USA). Each sample (including standards) was measured in triplicate and the resulting mean absorbance was used for developing a standard curve and subsequent protein concentration determination. To quantify and compare intensities of protein bands between rescue and control animals, Multi Gauge‐ver3.X (from Fujifilm Inc)

48 image analysis software was used for densitometry. Bands that appeared to be of equal intensities by visual inspection were selected for densitometry analysis.

Statistical Analysis:

All statistical analysis involving Student t‐test (unpaired), Mann‐Whitney Rank Sum test and one‐way ANOVA was performed using Sigma Plot 11 software suite. In all cases significance was considered if p <0.05.

Table­8: Primers used for generation of promoter deletion fragments.

49

Table­9: List of probes used for Northern blot analysis.

Results

Aim‐I

Identification and characterization of the role of Ppp1cc gene promoter for testis predominant expression of Ppp1cc2 isoform.

50 51

Background and rationale:

Previous studies have documented the tissue distribution of the PP1 isoforms and expression of Ppp1cc2 in rat testis (37). In mouse the Ppp1cc gene is expressed as a pre‐ mRNA transcript of length ~16 kb in size, Fig. 16. The pre‐mRNA message comprises of

5’UTR, exon 1 to exon8 with the intron 1 to intron7 incorporated and the 3’UTR. The pre‐ mRNA undergoes processing that gives rise to ~2.3 kb message corresponding to Ppp1cc1 that retains the 5’UTR, exon1 to exon7. However the entire inton7 along with exon8 and its downstream 3’ end becomes the part of the 3’UTR for Ppp1cc1 message(41). In contrast the Ppp1cc2 message originates from an alternative splicing event that involves removal of an intervening sequence of length 1.1 kb between a internal splice donor site (GT) within exon7 and acceptor site ‘CAG’ present at the 5’ end of exon8. This results in a shorter variant of mRNA species of length ~1.7 kb that corresponding to Ppp1cc2 and is highly expressed in testis, Fig. 16 (35, 37). Protein expression pattern of the PPP1CC isoforms across tissues also parallels their mRNA pattern in testis and other tissues in rat (55). It is known that in mouse testis PPP1CC2 protein levels are predominant in developing germ cells (39). However these previous studies do not shed light on whether high PPP1CC2 levels in testis is the result of tissue specific expression and/or the presence of splicing mechanisms to generate Ppp1cc2 mRNA in testis. To further define the role of the Ppp1cc2 splice variant in spermatogenesis and sperm development we examined the requirement and basis for the testis specific expression of PPP1CC2 in mouse testis. The goal of this aim was to determine expression patterns of the two Ppp1cc isoforms. Next we also looked to

52 characterize in vitro and in vivo regulatory elements in the promoter of Ppp1cc gene responsible for tissue specific expression.

Figure 15: Schemtic of exon­intron organization of Ppp1cc isoforms. (A) PPP1CC2 is derived as result of alternative splicing that occurs only in mammal specific event. The alternate splicing involves utilization of exon7. Ppp1cc1 retains compete exon7 whereas Ppp1cc2 is formed by joining of a portion of exon7 (internal splice donor site, GT) with that of exon8 resulting from the splicing event. (B) The alternate splicing event generates a unique 21 aminoacid C‐tail for Ppp1cc2 isoform coded by the exon8.

53

Sub­aim 1:

1.1 Characterization of the differential expression and localization of PPP1CC isoforms in testis and other tissues.

1.1.1 Ppp1cc2 message is highly abundant in testis compared to other somatic

tissues:

Northern blot analysis was performed to determine the expression of PP1 isoforms in adult and developing mouse testis compared to other tissues. A probe corresponding to the full length of Ppp1cc2 cDNA, Table‐9, capable of detecting both Ppp1cc isoforms was used. Data from Fig. 17(A), show high level of Ppp1cc2 message in testis compared to other tissues included in the study (brain, spleen, kidney and liver). A relatively low level of

Ppp1cc2 mRNA, compared to testis is detected in all other tissues. These findings are consistent with similar observations made with rat tissue blot where Ppp1cc2 mRNA being the most predominant form in testis (37, 41). Since the same probe (full length) was used to detect both isoforms the relative levels of intensities of the message corresponding to the isoforms is also a measure of their relative abundance. Comparison of the intensities of the two messages reveals that the relative abundance of the transcripts for the two isoforms is more or less equal in all tissues except testis, Fig. 17B. In testis the expression of

Ppp1cc2 message is almost ~6 times compared to spleen and kidney, Fig 17B. This suggests that the splicing mechanism is functional at a basal rate in all tissue except testis.

Interestingly in testis both high message levels resulting from high promoter activity of the

54

Ppp1cc gene and subsequent splicing of the message are the two most important mechanisms that control higher abundance of Ppp1cc2 message levels in the testis. Next we determined the relative abundance of the two isoform specific transcripts in reproductive tissues other than testis by examine the epididymis. For this purpose we collected caput, corpus and caudal region of the epididymis and isolated total RNA for northern analysis. As shown in figure, Fig. 17(C), message levels in different regions of vasa differentia is comparable to that of other somatic tissues, further suggesting the requirement of Ppp1cc2 isoform in the testis. As evident from Fig. 17(D) , the message levels of Ppp1cc2 isoform is reduced to almost half in heterozygous (+/‐) because of the loss of one copy of the Ppp1cc gene in the +/‐ animal providing the correspondence between gene copy number and transcript levels. However in the Ppp1cc null animals (and also in +/‐ animal) a shorter truncated message corresponding to position 0.9 kb can be detected resulting from splicing between exon4 and exon7 in the mutant pre‐mRNA arising from transcription from the mutant Ppp1cc genomic region harboring the Neomycin cassette, Fig. 17(C), as reported by other workers (49).

,

55

Figure 16: Tissue northern blot showing differential expression of Ppp1cc isoforms: (A) 20ug of total RNA extracted from brain, spleen kidney, liver and testis were loaded on to each lane. Ppp1cc1 isoform is expressed at basal level in all tissues whereas Ppp1cc2 is expressed only in testis in higher amounts in testis. (B) Density values of the bands were analyzed using densitometry software and expressed asrelative expression levels. Values were normalized to spleen Ppp1cc1. (C) RNA isolated from different regions of the epididymis (caput, corpus, cauda) was also probed to show that the characteristic pattern of expression of PPP1C isoforms is only limited to testis and not any other parts of the reproductive tract. (D) Northern blot showing that mRNA levels correlates with that of PPP1CC gene copy number. Message levels Ppp1cc1 and Ppp1cc2 in heterozygote are significantly lower compared to wildtype, whereas these mRNAs are absent in testis of knockout mice. Message levels are almost restored to wildtype levels in +/‐(t) mice.

56

1.1.2 Spatio­temporal expression of the PP1 isoforms in postnatal developing testis:

To determine the exact time, place and relative abundance of the Ppp1cc transcripts and other PP1 isoforms in testis we studied the developmental expression of PP1 isoforms in postnatal developing testis (Fig. 18). For this purpose total RNA was isolated from testis at different days postpartum (dpp) and northen blot analysis was performed using isoform specific probe for determining spatio‐temporal expression pattern, (see Table‐9). As evident from Fig. 19, expression of the Ppp1ca and Ppp1cc1 isoforms are similar ‐ relatively high levels were observed in testes between day 5 to 10 and the levels slightly tapers off with further maturation of testis. At day 5 and day 10, expression of Ppp1cc1 and Ppp1ca are relatively high when the Sertoli cells and differentiating germ cells mostly occupy the testis respectively. However following day 15 with the emergence of secondary spermatocytes (early, mid and late pachytene spermatocytes) their level drops and stay low for the remaining stages, suggesting lack of new synthesis of these messages or very low basal level of expression in the developing germ cells. The message for Ppp1b is detectable at low levels throughout the developmental stages suggesting that the message is produced at low levels at all stages of germ cell development including Sertoli cells, spermatogonia and Leydig cells. Interestingly the expression pattern for Ppp1cc2 is distinct and non‐overlapping with Ppp1cc1 and other PP1 isoforms in testis, Fig. 19. A low level of expression was observed between 5 to 10 dpp that is comparable to levels in other tissues (spleen, kidney, liver). At 15 dpp, which coincides with the emergence of pachytene secondary spermatocytes, there is a dramatic increase in expression of the message for

Ppp1cc2. Thereafter its level increases with emergence of late pachytene spermatocytes,

57 round and elongating spermatids (day 20) and remains elevated in testis of adult mice (day

30).

Figure 17: Schematic diagram showing the association of different cell types in the post partum developing testis. The development of testis is marked by the emergence of specific cell types at distinct time points during maturing of testis. Thus the cellular association of testis at specified time points can be used to identify the temporal expression pattern of protein or RNA species within testis.

58

Figure 18: Northern blot showing the developmental expression of all four PP1 isoforms in post­natal developing testis. . 25ug of total RNA was loaded in each lane. High levels of Ppp1cc2 expression coincide with onset of meiosis in developing germ cells and expression remains elevated thereafter. Ppp1cc1 and Ppp1ca expression pattern appears restricted to Sertoli cells, spermatogonia and preleptotene spermatocytes (day 5‐ 10). Ppp1cb was detected in low amounts in all the stages. β‐actin was used as a loading control.

59

1.1.3 Immunological detection of PPP1CC1 isoform localization in testis cross­

section of +/+ mice:

The observation made from the developmental expression in maturing testis that

Ppp1cc1 mRNA is expressed in Sertoli cells, spermatogonia and pre‐leptotene spermatocytes were further confirmed by immunohistochemical studies on 10 µm thick paraffin embedded testis section of adult mouse testis. PPP1CC1 ant‐rabbit primary antibody was used to incubate the sections followed by Cy3 conjugated rabbit secondary antibody was used for fluorescent detection using confocal microscopy. Fig. 20, IHC clearly demonstrated PPP1CC1 expression in Sertoli cells, spermatogonia along the periphery of the seminiferous tubules. Confirmation of staining of Sertoli cell is made by the presence of distinct cellular process specific to the morphology of Sertoli cell type. Pre‐leptotene and leptotene spermatocytes are marked by round large cells among the developing germ cells that are closest to the periphery of the tubules but not in direct contact with the basement membrane also were positive for PPP1CC1 staining, Fig. 20(a’). However no

PPP1CC1 signal could be detected in the secondary spermatocyte stage onward towards the lumen of the tubules suggesting its exclusion from these cell types, Fig. 20(a,c and e).

The immunofluorescent signaling pattern for PPP1CC1 is in complete contrast to that of

PPP1CC2 pattern in adult testis crossection. From our earlier analysis we know that

PPP1CC2 is absent in sertoli cells, spermatogonia along the periphery and also in primary spermatocytes and is detected only from secondary spermatocyte onwards and signal remains elevated thereafter in complete agreement to the Ppp1cc2 expression pattern in postpartum developing testis, Fig. 21 (39).

60

Figure 19: Immunohistochemistry showing the expression and localization of PPP1CC1 isoform in adult testis. (a and b) At 20x magnification the PPP1CC1 staining of the sertoli cell appear like radial spikes within the seminiferous tubules. (a’) The magnified region in white box clearly shows the staining of the body of the Sertoli cells and its processes. It is to be noted that the bracketed region (white brackets) along the periphery shows PPP1CC1 staining for spermatogona and pre‐leptotene primary spermatocytes. (c) PPP1CC1 staining pattern at 60x and it corresponding merged image with DAPI staining (d). (d’) The magnified region in the boxed area show the sertoli staining and its nucleus stained with DAPI. (e) Additional images of PPP1CC1 staining and its corresponding DIC image superimposed (f).

61

Sub­aim 2:

1.2.1 In­vivo confirmation of the test specific promoter activity of the Ppp1cc gene.

In order to examine the expression of Ppp1cc2 and its role in spermatogenesis we transgenically expressed PPP1CC2 in Ppp1cc null mice driven by its endogenous promoter.

A genomic fragment of length 2.6 kb that is upstream of the transcription start site of

Ppp1cc gene was used as a putative endogenous promoter to drive the expression of full length Ppp1cc2‐cDNA in transgenic mice. Another purpose of this study was to test whether the role of the 2.6 kb fragment used as endogenous promoter in its ability to direct testis predominant mode of expression of PPP1CC2 and restore its correct spatio‐temporal expression within testis. Results from all the transgenic founder lines that were opened showed that the 2.6 fragment is enough to direct testis specific expression of PPP1CC2, as shown in Fig. 21. Absence of immunoreactivity for PPP1CC1 in testis and brain (Fig. 21c,d) confirms the Ppp1cc null background of the transgenic mice compared to control heterozygous (+/‐) littermate. As seen in Fig. 21(b), PPP1CC2 expression was high in testis and with low but detectable levels in brain. This expression pattern is similar to that of the endogenous gene in control (+/‐), Fig. 21(a) and wildtype mice [7]. PPP1CC2 immunoreactivity was not detected in any other tissue studied suggesting that cis‐ regulatory elements controlling testis specific expression pattern for Ppp1cc2 are harbored within the 2.6kb fragment. Immunohistochemistry data further confirmed that the 2.6 kb fragment is efficient in restoring the correct spatio‐temporal expression pattern of

PPP1CC2 in the developing germ cells in early pachytene spermatocyte and continued till

62 the elongating spermatids (Fig. 21 e, f, g). Staining for PPP1CC2 was absent in Sertoli cells,

Leydig cells, spermatogonia and pre‐leptotene spermatocytes as seen in wildtype (39).

Figure 20: Testis specificity of the 2.6 kb endogenous promoter fragment. (a, b) Tissue western blot showing the testis specificity of the Ppp1cc gene promoter. PPP1CC2 expression is detected in high amounts in testis and relatively low amounts in brain (Br) and spleen (Spl) in rescue males and is comparable to control pattern of expression. No PPP1CC2 reactivity was observed in other somatic tissues studied, including tongue (Tn), stomach (Sto), lungs (Lng), heart (Hrt) and liver (Liv). (c, d) Multi‐tissue western blot analysis of PPP1CC expression in rescue lines vs. control animals. The absence of PPP1CC1 expression in the testis and brain confirms the Ppp1cc null background of the rescue mice in comparison to control mice where the isoform is expressed. (e, f, g) Immunohistochemistry of testis sections from endogenous promoter‐ transgenic rescue lines (eTg), showing a wild‐type pattern of PPP1CC2 localization in the testis. Images were obtained at 20x (i and ii) and 60x (iii) using confocal fluorescent microscopy. These images are representative observations taken from multiple sections of all three eTg lines (M7, F1 and F10) and several different testis preparations.

63

Sub­aim 3:

1.3 In­silico analysis of the Ppp1cc gene promoter.

1.3.1 Identification of putative cis­regulatory element in the promoter region using

promoter foot printing analysis:

To identify evolutionary conserved regions (ECR), we used the Mulan software suite

(56‐58) to perform local sequence alignment for in­silico analysis. We analyzed a 3.0 kb upstream genomic region from the ATG start codon for the mammalian species (mouse, rat, human, chimp, cat, monkey and gorilla) for which the genomic sequence has been annotated. The output of the analysis is shown in a graphical format (smoothed graph plot) in Fig. 22. Three conserved regions (ECR) with more than 70% homology were identified

(colored boxed area in the upper panel) labeled as ECR1 (blue), ECR2 (yellow) and ECR3

(green). Of these, ECR1 and ECR2 span the genomic region between +283 to ‐280 bp [Fig.

22(A), upper panel]. ECR3 mapped the farthest from the transcription start site, between ‐

1967 to ‐2267 bp. Potential binding sites for a set of 28 unique transcription factors were identified that mapped within the three ECR regions across the species (see Table‐10 for complete list of factors). Among these 28 factors three factors were selected based on EST expression profiles (Fig. 23) and previously published information regarding their importance in regulating testis‐specific gene expression (59‐62). These include a binding site for the transcription factor Spz1 (+272/+257), two non‐overlapping sites for Sp1 (‐

53/‐64 and ‐76/‐84) and a Myb (A‐Myb/ B‐Myb) binding site at ‐169/‐178 [Fig 22(B),

64 lower panel]. Of these Spz1 and A‐Myb are predominantly expressed in developing germ cells in testis (21, 63, 64).

65

Figure 21: Identification of conserved cis­regulatory elements in the promoter region of Ppp1cc gene: The region ~3.0kb upstream of the translation start site for Ppp1cc from various species ‐ mouse, human, rat, chimapanzee, gorilla, macaque and cat ‐were analyzed. Upper panel shows the homology of the sequence in pairwise alignment for each species with that of the query sequence (mouse). Sequences with homology 50% (lower cutoff) and above are only depicted in the smoothed plot (in red). Three regions sharing high homology (~70% and above) and conserved across all the species were detected. The evolutionary conserved regions, ECR’s are highligted with colored boxes (ECR1‐blue; ECR‐2 in yellow and ECR‐3 in green). (B) The lower panel shows the conserved binding site for the transcription factors Spz1, A‐Myb and Sp1 in the promoter region of Ppp1cc gene for all the speccies. The numbers at the lower bottom corner of the box denotes the genomic position with respect to the transcription start site of the mouse Ppp1cc gene.

Table­10: Complete list of conserved cis­binding factors identified by MULAN tool.

Plus' or Unique Total Binding motif within the 'minus' Postion factors Factors sequence strand

1 AP2 AP2 ­ 2970­2981 ctgCTGCGGGag 2 AP4 AP4 + 2980­2997 agggtCGGCGGCGGGAcg 3 CACD CACD + 2634­2641 CCCCGCCC 4 CETS168 CETS168 + 2807­2814 GAGGAAGc 5 DEAF1 DEAF1 ­ 623­647 accttCAGCAAACAACCCAGccttt E2F ­ 2544­2554 gTTTGGCGgaa E2F ­ 2635­2640 ccCGCC E2F ­ 2657­2662 ccCGCC E2F ­ 2670­2675 cgCGCG 6 E2F E2F + 2722­2727 CGCGcg E2F ­ 2722­2727 cgCGCG E2F ­ 2848­2853 ccCGCC E2F1 + 2544­2553 gTTTGGCGga E2F1 ­ 2662­2677 ccCCTTCCCGCGCgct 7 ELF ELF1 + 2803­2814 ggaggAGGAAgc ETF ­ 2636­2642 CCGCCCC ETF ­ 2658­2664 CCGCCCC 8 ETF ETF ­ 2849­2855 CCGCCGC ETF + 2946­2952 GCGGCGG 9 ETS ETS ­ 2807­2814 gaGGAAgc 10 GC GC ­ 2653­2666 cggCCCCGCCCcct HIC1 ­ 2704­2721 cgggccgctGCActcccg 11 H1C1 HIC1 ­ 2718­2735 cccgcgcgcGCAtccgtg 12 HIF1 HIF1 + 2727­2740 gcatCCGTGCgccg

66

13 HNF4 HNF4 + 1959­1964 AGTTCA HSF 14 HSF + 2779­2791 TTCTCGTGGTTCC

KROX + 2657­2670 ccCGCCCCCttccc

KROX KROX + 2872­2885 gcCGCCGCCgccac 15 KROX ­ 2981­2994 gggtcGGCGGCGgg MAZ ­ 2657­2664 CCCGCCCc 16 MAZR MAZR + 2837­2849 gggGGGGTGGacc 17 MYB MYB ­ 2540­2549 gcccGTTtgg 18 NFMUE1 NFMUE1 + 2765­2773 CGGCCATCT 19 NRF1 NRF1 ­ 2724­2733 cGCGCATCCG 20 PAX4 PAX4 + 2335­2346 caacccCACCCc SP1 ­ 2632­2644 aaccCCGCCCctt SP1 + 2634­2643 cCCCGCCCct 21 SP1 SP1 ­ 2654­2666 ggccCCGCCCcct SP1 + 2656­2665 cCCCGCCCcc 22 SPZ1 SPZ1 + 2975­2989 gcgGGAGGGTcggcg SREBP1 + 2341­2347 CACCCCA 23 SREBP1 SREBP1 ­ 2904­2910 TGGCGTG 24 VBP VBP ­ 1943­1952 attatATAAc 25 WT1 WT1 ­ 2975­2983 GCGGGAGgg

26 XVENT1 XVENT1 + 2526­2538 ggtctATTTCccc 27 YY1 YY1 + 2767­2775 GCCATCTtg ZF5 ­ 2634­2641 ccCCGCCC ZF5 ­ 2656­2663 ccCCGCCC 28 ZF5 ZF5 ­ 2658­2665 ccGCCCCC ZF5 ­ 2667­2674 tcCCGCGC ZF5 ­ 2731­2738 ccGTGCGC

* 28 unique factors with conserved binding sites have been identified.

* Of these 28 factors, for some multiple binding sites are present within the 3.0 kb upstream genomic region being studied. For eg, SP1 there are two unique binding sites at location 2632‐2644 and 2654‐2666 (excluding the binding site on – strands).

‘+’ denotes plus strand whereas (‐) denotes minus strand of DNA.

67

EST expression of transcription factors Spz1, A­Myb and Sp1 from UNIGENE database:

Figure 22: EST expression profile for transcription factors Spz1, A­Myb and Sp1. Note all the three factors are expressed in testis (red boxed). Of the three Spz1 and A‐Myb are associated with phenotype in testis due to overexpression or deletion of the respective gene.

68

1.3.2 In­vitro identification of the core promoter of the Ppp1cc gene:

Following the in‐vivo confirmation of the testis specific activity of the Ppp1cc gene promoter within the 2.6 kb genomic and identification of putative cis‐elements, we next attempted to characterize the core promoter in‐vitro using a luciferase reporter system. A series of deletion constructs (Fig. 24) were developed and tested for promoter activity. We used spermatogonial stem cell line C18‐4 (52, 53) for these experiments along with somatic

HEK‐293 cells. Results from these reporter assays are summarized in Fig. 25(A,B). As evident from Fig. 25A, progressive deletion of the distal upstream region of the promoter from ‐2.8kb (Del1) to ‐467 bp (Del3) led to a gradual increase in expression of the luciferase activity. Maximum luciferase activity was seen with the deletion construct, Del4

(+69bp to ‐173bp). Further deletion in Del5 that removes binding sites for Myb and Sp1 factors led to lowering of luciferase activity by 75% compared to Del4. Reporter activity was almost completely abolished in the fragment Del6 (+73 to +306 bp) that spans most part of the 5’UTR region of the Ppp1cc gene. Similar experiments with the same set of deletion fragments (Del1 to Del6) were used to study the promoter activity in HEK‐293 cells representing the somatic cells, Fig. 25(B). The results from the study in Hek‐293 cells show similar trend in activity of the promoter fragments with Del 4 showing the highest activity for luciferase activity thus further confirming the localization of the core promoter in this region.

69

Figure 23: Promoter deletion fragments used in reporter assay. (A) Schematic diagram showing the DNA fragments used for promoter analysis. Serial deletion of promoter fragments was done by PCR amplification (Del1 to Del6). Following amplification the fragments were individually subcloned into pGL3‐basic vector for luciferase reporter assay. The number within the parenthesis on the right denotes the length of the amplicon in base pairs.

70

Figure 24: Analysis of the Ppp1cc gene promoter using reporter assay: (A) Reporter activity of the promoter deletion constructs. C18‐4 cells were transfected with the deletion constructs. 48 hrs post‐transfection cells were harvested and assays were performed for luciferase activity. Renilla luciferase was used as internal control. (B) Promoter activity of the fragments were also carried out in HEK‐293 cells. The activity of the fragments show similar trend as seen in C18‐4 cells with Del4 showing the highest activityThe activity for each construct was normalized to that of empty pGL3‐ Basic vector. The data shown are from three independent experiments (n=3), the mean value were derived and expressed relative to the mean luciferase activity of the full length Del1 construct (+69 to ‐2759 bp), which was considered as 100%. The error bars represent standard error of mean (± SD). The location of the transcription factors (Spz1, Sp1 and Myb) relative to the start site is represented as colored bars.

1.3.3 Expression of sperm PPP1CC2 regulators closely follow testis expression

pattern of Ppp1cc2:

Strikingly high levels of Ppp1cc2 are present in testis and spermatozoa. It was of interest to determine whether postnatal testicular expression pattern of the PP1 regulatory proteins known to be present in sperm match with that of Ppp1cc2. The three known regulators of

PP1 present in sperm are Ppp1r7 (sds22), Ppp1r11 (I‐3) and Ppp1r2 (I‐2) (50, 65). In order

71 to determine the spatio‐temporal expression of the PP1 regulators in testis we performed northern blot analysis for each of the regulators in post natal developing testis, Fig. 26.

Notably expression of I‐3 was not detected between 5‐10 dpp, suggesting its complete absence in sertoli cells, spermatogonia, Leydig cells and pre‐leptotene spermatocytes.

Message for I‐3 was first detected in 15 dpp testis that corresponds with pachytene phase of meiosis in spermatogenic cells, Fig. 26A. Its expression gradually increases from day 15 onwards, reaching a maximum on day 30 and in adult testis (105 days) that coincides with the emergence of elongating spermatids. A similar pattern of expression was also observed for the PP1 regulator, Ppp1r7 (Sds22), Fig. 26C. Expression of sds22 is low at 10 dpp and its levels gradually increases and reaching a maximum in day 30 testis. The message for sds22 is present as a single band slightly above 1.5 kb, which is drastically different compared to the somatic form (66). However we were not able to detect the larger somatic and ubiquitous form in the testis whose reported size is ~3.6 kb (from NCBI). The most likely reason for not able to detect the higher form is probably due to low abundance of the message which is beyond the detection by northern blot analysis. Previous reports have shown the expression of inhibitor, I­2 (Ppp1r2) in maturing rat and adult rabbit testis (67,

68). We wanted to determine the age dependent expression of I­2 in mouse testis during post natal testicular development. We were able to detect messages corresponding to positions ~2.4 kb, ~1.7 kb and 0.9 – 1.0 kb using the entire human I­2 coding region as a probe, Fig. 26B. The higher molecular size messages are thought to be due to alternate polyadenylation and were found to be present at all stages of testicular development albeit at low levels (68). Interestingly, the shorter messenger RNA species (1.1 ‐1.2 kb) was the most abundant and expressed only at day 30 and adult testis (day105). Expression pattern

72 of a testis expressed protein RSPH1, is known to be expressed during meiosis and post meiosis, is shown for comparison.

Figure 25: Expression of PP1 regulators closely matches Ppp1cc2 expression in postnatal developing testis: (A) Northern blot analysis showing an age dependent expression pattern for Ppp1r11 in the testis with maximum levels at 30 dpp and in adult testis. No expression for I­3 was observed between day 5 to day 10, suggesting its absence in Sertoli cells, Leydig cells, spermatogonia and pre‐leptotene spermatocytes. (B) Expression of PP1 inhibitor, Ppp1r2/I­2 in post partum developing testis. Both the splice variants of I­2 namely I­2α1 and I­2α2 were detected at positions corresponding ~2.4 kb and ~1.4 kb respectively, uniformly but at low levels during all stages of post natal development. A third isotype, I­2β at position between 0.9 ‐1.0 kb is highly expressed in stage specific manner only at day 30 and in adult testis. (C) Postnatal expression of Ppp1r7/Sds22, is detected at day10 but its levels increases post‐meiotically reaching a maximum at 30 dpp. The band corresponding to Sds22 is localized between 0.95 ‐ 1.1 kb. The testis specific protein RSPH1, known to be expressed during meiotic and post meiotic germ cells, is shown for comparison

Aim‐II

Determination of the isoform specific role of PPP1CC2 and PPP1CC1 in sperm function and male fertility.

73 74

Sub­aim I:

2.1 Isoform specific role of PPP1CC2 in sperm development, spermatogenesis and male fertility.

2.1.1 Background and rationale:

Targeted disruption of the Ppp1cc gene, that results in the elimination of both PP1 isoforms, Ppp1cc1 and Ppp1cc2, leads to male infertility due to impaired spermatogenesis, sperm morphogenesis, spermiation, and increased testicular apoptosis (39, 49). This suggests an indispensible role for either PPP1CC isoforms or both in testis and male reproduction. However female mice were normal and fertile suggesting that PPP1CA and

PPP1CB can substituefor the loss of both PPP1CC isoforms in all tissues except testis. The goal of this aim was to determine whether the Ppp1cc2 isoform is capable by itself of correcting these testicular defects as well as rescuing male fertility, or whether the

Ppp1cc1 isoform was also essential in the testes to restore fertility, we transgenically expressed the Ppp1cc2 isoform driven by the well‐characterized testis specific Pgk2 promoter in the Ppp1cc null background (69, 70). Results of our early transgenic rescue experiments showed that PPP1CC2 was able to rescue apoptosis and spermiation, but not fertility (51). Mature spermatozoa were isolated from the caudae epididymides of rescue males, but demonstrated poor forward motility at best due to a wide range of morphological abnormalities. We speculated that there were three possible reasons for the lack of complete rescue: (1) the relatively low levels of transgenic expression of PPP1CC2;

75

(2) the fact that the Pgk2 promoter could not recapitulate precisely the endogenous spatio‐ temporal expression pattern of PPP1CC2 in testis; and (3) PPP1CC1 expression was also required, and (4) any combination of 1, 2, and/or 3. In this aim we aim we attempted to distinguish between these possibilities.

2.1.2 Generation of Endogenous promoter driven PPP1CC2 rescue lines:

In the first transgenic line that we created we used a 2.6 kb upstream (from the transcription start site) genomic fragment to drive the transgenic expression of PPP1CC2 in testis. The design for the mini‐transgene cassette used for pro‐nuclear injection is shown in, Fig. 27(a,b). , and the steps outlining its construction is detailed in materials and methods section. The injection of the linearized contruct (Fig. 27c), resulted in 33 pups, of which four were transgene positive founders. The genotyping PCR to identify the founder animals is shown in Fig???. These transgene positive animals were crossed with CD1‐

Ppp1cc ‐/‐ animals to derive founder lines eTg‐M7; eTg‐F1; eTg‐F3; eTg‐F10 (where ‘eTg’ denotes the endogenous promoter driven transgene, and the alphanumeric number following it denotes the sex and serial number of the founder animal, respectively). The F1 males (Ppp1cc +/­; Tg+) from the founder lines were further crossed with CD1‐Ppp1cc ­/­ females to derive Ppp1cc transgenic rescue males (Ppp1cc­/­; Tg+) [see Fig. 15]. Of the four positive animals three lines were established (eTg‐M7; eTg‐F1; eTg‐F10) with confirmed transgenic protein expression.

76

Figure 26: The design of the mini gene cassette for Endogenous promoter driven transgenic expression of PPP1CC2. (a) Linear restriction map of the plasmid construct. (b) The circular map of the mini gene. (c) The gel picture showing the position and size of the relieved mini gene on double digestion with NotI and KpnI in contrast to the single or uncut plasmid.

Table­11: Statistics of founder lines for Endogenous­Ppp1cc2­SV40 PolyA

77

2.1.3 Generation of Pgk2 promoter driven PPP1CC2 rescue lines:

A second transgenic line was developed in parallel to the above where the transgenic expression of PPP1CC2 was driven by the testis specific human Pgk2 promoter

(hPgk2). The design for the mini‐transgene cassette used for pro‐nuclear injection is shown in, Fig. 28, and the steps outlining its construction is detailed in materials and methods section. Briefly, transgenic expression of Ppp1cc2 was placed under the control of the hPgk2 promoter and that for the construct we used coding sequence (Cds) of Ppp1cc2‐ cDNA corresponding to mouse Ppp1cc2 mRNA, Fig. 28. Microinjection of the construct yielded seven transgene positive founders (see Table‐13) of which three were used to establish transgenic rescue lines: pTg‐M1, pTg‐M26 and pTg‐M30 (where ‘pTg’ denotes the

Pgk2 promoter driven transgene, and the alphanumeric number following it denotes the sex and serial number of the founder animal, respectively). The same mating scheme as outlined in Fig. 15, was used to generate the rescue animals. Transgene transmission for both the endogenous and Pgk2 promoter driven Ppp1cc2 transgenic lines followed expected Mendelian ratios.

78

Figure 27: The design of the mini gene cassette for Pgk2 promoter driven transgenic expression of PPP1CC2. (a) Linear restriction map of the plasmid construct. (b) The circular map of the mini gene. (c) The gel picture showing the position and size of the relieved mini gene on double digestion with NotI and SacI in contrast to the single or uncut plasmid.

Table­12: Statistics of founder line for Pgk2 –Ppp1cc2 transgenic lines.

79

2.1.4 Comparison of PPP1CC2 expression levels in testis across different transgenic

rescue lines:

One of the chief objectives of this aim was to attempt to improve the expression of

PPP1CC2 in rescue animals and to test whether increased levels of PPP1CC2 is capable of completely rescuing the infertility phenotype of Ppp1cc null (‐/‐) male mice. Thus the task of determining the PPP1CC2 levels accurately in testis and compare it with other transgenic rescue line was undertaken. In order to do so I made testis extracts of rescue animal and that of its heterozygous litter mate control (+/‐) and the extracts were subjected to protein estimation by Bradford assay. Following estimation, protein adjusted amounts of testis extracts from rescue and control animals were serial diluted and subjected to western blot analysis. For each rescue animals the following dilutions were made: 20 µg, 10 µg, 5 µg and 2.5 µg and in the same gel for the control the dilutions that were made include: 20 µg, 10 µg, 7.5 µg, 5 µg and 2.5 µg. Intensities of band from rescue lane that matched closely with that from control lane by visual inspection were further subjected to densitometry analysis for quantification. To confirm the levels within a given rescue line, western blot was replicated with extracts from at least two (or three) different animals from the line, Table‐14. Figure. 29, shows representative western blot for each rescue line showing the PPP1CC2 levels compared to control. Using western blot followed by the densitometry analysis we determined the relative levels of PPP1CC2 for each rescue line as given in the Table‐14. In the endogenous promoter driven lines (eTg‐M7, eTg‐F10 and eTg‐F1) the levels of expressed PPP1CC2 were approximately 25%, 25%, and 50%, respectively of levels expressed in the testes of Ppp1cc+/­ controls. The hPgk2 promoter lines, pTg‐M3, pTg‐M26

80 and pTg‐M30, also showed varying levels of PPP1CC2 expression (approximately 12.5%, 75%, and

100%, respectively) compared to positive control lines

Figure 28: Comparison of levels of transgenically expressed PPP1CC2 in testis across all transgenic rescue lines. Western blot showing the levels of PPP1CC2 expressed in the testis of rescue lines compared with littermate controls. In the left panel protein estimated testis extracts from rescue animals were serially diluted to 20 µg, 10 µg, 5.0g, 2.5 µg and for heterozygous control animal 20 μg, 10 μg, 7.5 μg, 5.0 μg and 2.5 μg respectively. (*) denotes the lane in which the band corresponding to rescue mouse PPP1CC2 expression is comparable in intensity to a band in a control mouse lane. In the right panels, (*) lanes are shown adjacent to each other for comparison. b‐Actin was used as loading control. Each

81 blot was repeated with preparations from different animals of the same line to confirm the original data.

Table­13: Densitometry analysis to determine the PPP1CC2 levels across various transgenic lines.

Student t‐test (unpaired) was performed to compare the mean band intensities for a transgenic rescue line with that of control (+/‐). Differences were considered significant if P < 0.05 at a confidence interval of 95%.

‘a’ denotes that no statistical significant difference was observed for mean intensity value of bands between test group and control group.

2.1.4 Determination of PPP1CC2 expression levels in spermatozoa across different

transgenic rescue lines.

To assess the levels of PPP1CC2 incorporated into mature spermatozoa, whole sperm soluble protein extracts from 2 x 106 caudal spermatozoa for each transgenic line

82

and Ppp1cc +/­ controls were compared by western blot analysis. The levels of PPP1CC2

incorporated into mature spermatozoa roughly paralleled its testis levels of expression, Fig.

30.

Figure 29: Comparison of levels of PPP1CC2 incorporation in spermatozoa across different transgenic rescue lines. Western blot of whole sperm protein extract demonstrate PPP1CC2 levels in spermatozoa. Protein extract from 2 x 106 sperm was loaded in each lane for each rescue line and its littermate control. Upper panel represents PPP1CC2 levels in endogenous promoter driven rescue lines (eTg) while the lower panel represents PPP1CC2 levels in the Pgk2 promoter driven rescue lines (pTg). Note that the PPP1CC2 levels in spermatozoa conform to their levels in the testis.

83

2.1.5 Detection of PPP1CC2 localization within the cross section of seminiferous

tubules:

To detect the spatial expression pattern of transgenically expressed PPP1CC2 within testis of endogenous and Pgk2 promoter driven transgenic rescue lines we performed immunohistochemistry on paraffin embedded testis section. As SEEN IN Fig. 31 (a, b, c, d) and also reported in previous section endogenous promoter was adequate in driving the expression of PPP1CC2 from early pachytene spermatocytes as seen in wildtype and heterozygous littermate control. The same pattern was observed for all the three endogenous transgenic lines (eTg‐M7, eTg‐F1 and eTg‐F10), suggesting once again that the cis‐elements responsible for testis pattern of expression are harbored within the 2.6 kb genomic fragment. As expected, the Pgk2 was successful in driving testis specific expression of PPP1CC2. Within testis the expression of PPP1CC2 onset from secondary round spermatocytes onwards and continues to the elongating spermatids and mature spermatozoa, Fig. 31 (e, f). The same pattern of expression was observed for the rest of the

Pgk2 founder lines pTg‐M26 and pTg‐M30.

84

Figure 30: Immunological detection of PPP1CC2 in testis of rescue lines. Paraffin embedded testis section was incubated with PPP1CC2 primary antibody to detect the localization of the signal within testis. (a) IHC of testis section from rescue line eTg‐F10 showing PPP1CC2 is expression from early pacytene spermatocyte onwards driven by the endogenous promoter. (b) The control without primary. (c) PPP1CC2 detection in rescue line eTg‐F1 is similar for all the endogenous line with detection from primary spermatocytes, and its corresponding control (d). (e, f) IHC of testis section of Pgk2 driven transgenic line pTg‐M3 and its control respectively.

85

2.1.6 Testis architecture, spermatogenesis and spermiation is restored in rescue

lines irrespective of the promoter and PPP1CC2 levels:

Spermatogenesis, as determined by the number of mature spermatozoa, appeared largely restored in rescue mice compared to controls. Significant numbers of spermatozoa were recovered from the caudae epididymides and ductus deferens of rescue mice (eTg‐

M7: 0.4‐5.0 x 107; Tg‐F1: 1.4‐3.7 x 107; eTg‐F10: 1.3‐3.6 x 107) compared to (2.8 – 4.5 x

107) in Ppp1cc (+/‐) control mice, Table‐15. Paraffin embedded sections of testis from adult rescue mice (≥ 8 weeks) were haematoxylin stained for comparison of morphology of testis sections, seminiferous tubules, and presence of adluminal testicular sperm with that of control mice. As seen in Fig. 32, morphology of the testis sections appeared normal and comparable to testis from wild‐type mice. Unlike those of Ppp1cc ‐/‐ mice [8, 21] the lumen of the seminiferous tubules of rescue mice were replete with spermatozoa, Fig. 32. A full complement of germ cells at various stages of development including round and early and late elongating spermatids were evident in testes of rescue mice. In addition, mean testis weights of all rescue mice from endogenous lines, except for two lines (eTg‐F1 and pTg‐

M3) were significantly higher than the mean testis weight of Ppp1cc null mice (68.0 mg) and comparable to testis weights from Ppp1cc +/‐ controls (115.0 mg), Table‐15. Only the pTg‐M3 (???) and the eTg‐F1 (101.0 mg) lines had mean testis weights that were significantly different from that of positive controls.

As expected and unlike endogenous promoter driven expression of Ppp1cc2, hPgk2 driven expression of mouse Ppp1cc2 was testis specific. Thus, PPP1CC2 immuno‐reactivity was

86 not detected in tissues other than testis (data not shown). As seen with the endogenous promoter lines, spermatogenesis was restored, and higher epididymal sperm counts corresponded to higher PPP1CC2 expression in the testis of hPgk2 driven lines, Table‐15.

Figure 31: Haematoxylin stained testis sections of rescue mice compared to that of control.

87

2.1.7 Rescue animals with lower levels of transgenic expression of PPP1CC2 showed wide range of morphological anomalies:

We determined the morphology of mature spermatozoa from various rescue lines with varying levels of PPP1CC2 expression. For this purpose mature spermatozoa from the epididymis were collected and fixed in 4% paraformaldehyde. Fixed sperm were mounted on slide and viewed at 20x magnification using the Olympus 70x microscope. For each rescue animal that was opened approximately 20‐40 (n) non‐overlapping fields were observed and the number of sperm with abnormal versus normal morphology were counted. Table‐15, summarizes the statistics from the results of these morphological analysis. Three broad categories of morphological abnormalities were seen in rescue lines with reduced levels of PPP1CC2 expression. These are sperm with ‘kink’ (bend) between mid‐piece and principle piece. As seen in Fig. 33, this kink varied in angle from 45° to 180°, in the later the sperm is completely folded back upon the tail. The second major deformity that was noticed was the uncharacteristic bending of head at the neck region between the capitulum and the mid‐piece. The third category of defects include deformity of head resulting from malformation [Fig. 33(k,m)] and disorganized mitochondrial sheath. The mitochondrial sheaths in some cases showed signs of thinning (Fig. 33m), in some it is stunted and in some cases it is completely absent (Fig. 33j). Results from Table‐15, suggests that, sperm morphology and thus morphogenesis is clearly dependent on the levels of

PPP1CC2 expressed in testis and spermatozoa. Rescue lines with low levels of PPP1CC2 show a high proportion of abnormal sperm (eTg‐F10 about 68% and for pTg‐M3 about

54%), consequently lower proportion of normal spermatozoa (32% and 46% in eTg‐F10

88 and pTg‐M3 respectively). Sperm morphology showed marked improvement in lines with

PPP1CC2 expression equal to or above 50% of that of control (eTg‐F1 about 75% and pTg‐

M26 about 88%). In line pTg‐M30 sperm morphology is completely restored and comparable to that of control (93% compared to 95% in heterozygous littermate).

Figure 32: Morphological abnormalities of spermatozoa as seen in rescue lines expressing low levels of PPP1CC2. Figs (a, b, c, d, e) show mature epididymal spermatozoa with aberrant bending between mid‐piece and principle piece. The angular variation of the bend ranges from 45° (b), 90° (d, e) to 180 ° (a, c). Figs (f­j), show another frequently observed morphological defect involving jack‐knifing of the head between the capitulum and the proximal mid‐piece. Figs (k­n), demonstrate other less common deformities, including shortened mitochondrial sheaths (k, n), and malformed heads (l, n, m). All micrographs were photographed under DIC optics with an Olympus 70 light microscope. Abnormal regions of the spermatozoa are denoted by white arrows and its

89 corresponding normal regions are denoted by black arrows. These observations are representative of multiple samples from caudal preparations of different rescue animals.

Table­14: Comparison of PPP1CC2 levels, testis weight, sperm number and morphology between transgenic rescue lines and control animals

Transgenic lines Normalized Mean testis weight Mean Sperm number Sperm Morphology PPP1CC2 levels ±SEM (in mg) ±SEM x 107/ml compared to Total number of Normal Defective Ppp1cc +/­ level sperm counted (%)±SEM (%)±SEM

Ppp1cc +/­ n=2, 1c n=10; 114.4 ± 1.9 c n=6; 3.9 ± 0.3 b n=1; 368 95 5 c

Ppp1cc ­/­ N/A n=6; 68.4 ± 5.4 a N/A N/A N/A N/A

eTg­F10 n=2; 0.125a n=3; 110.5 ± 1.9 b,c n=3; 2.1 ± 0.8 a n=2; 323 32+2.8 68+2.8 a,b

pTg­M3 n=2; 0.125a n=2; 79.9 ± 2.9 a,b n=2; 1.4 ± 0.05 a n=2; 981 47+1.2 53+1.2 a,b

eTg­M7 n=2; 0.25a,b n=8; 107.4 ± 4.6 b,c n=5; 1.7 ±0. 9 a n=2; 566 24+9.9 76+9.9 a

eTg­F1 n=5; 0.5a,b,c n=9; 102.7 ± 2.9 a,b n=5; 2.6 ± 0.5 a,b n=3; 1194 75+0.2 25+0.2 a,b,c

pTg­M26 n=3; 0.75b,c n=5; 108.2 ± 4.2 b,c n=4; 3.7 ± 0.3 a,b n=2; 986 88+2.3 12+2.3 b,c

pTg­M30 n=3; 1c n=4; 108.2 ± 0.5 b,c n=3; 3.7 ± 0.5 a,b n=2; 431 94+1.1 6+1.1 c

The Kruskal‐Wallis non‐parametric one‐way ANOVA by rank was performed for testing whether samples originated from the same distribution. Results of Kruskal‐Wallis were analyzed post­hoc by Dunn’s procedure for performing two‐tailed multiple pair wise comparisons. Differences were considered significant if p < 0.05 at a confidence interval of 95%. a, b, c… denote significantly different groups.

SEM denotes Standard Error of the Mean. n denotes number of samples/group

90

2.1.8 Motility of mature spermatozoa was restored to normal only in rescue lines

with PPP1CC2 expression 50% or above to that of control (+/­) animal:

In order to find how PPP1CC2 levels effect sperm motility and other sperm velocity parameters we next undertook computer assisted sperm motility analysis or CASA. Mature sperm was extruded into modified HTF media (human tubular fluid) and incubated in 5%

CO2 for 10min before analysis using CASA on 100μm Leja slides. Both percent of motile sperm and progressive motility increased in direct proportion to increasing levels of

PPP1CC2 in both testis and sperm. As seen in Fig. 34A, sperm isolated from transgenic lines with PPP1CC2 levels equal to 50% or less than control levels (eTg‐M7 and eTg‐F1) showed no more than~57.5 ± 7.5% motile sperm compared to ~84.7 ± 1.7% for controls.

Additionally, progressive motility recorded in these lines were significantly less (eTg‐M7:

9.0 ± 2.0 % and eTg‐F1: 13 ± 4.0 %) compared to ~32.0 ± 2.1 % for control sperm. With increasing levels of PPP1CC2 (lines pTg‐M26, pTg‐M30) percent of motile sperm increased to 78.3 ± 2.6 % and 76.3 ± 7.9 %, and progressive motility to 23.3 ± 4.9 % and 27 ± 6.9 %, respectively. Other sperm velocity parameter including average path velocity (VAP), straight line velocity (VSL) and the curvilinear velocity (VCL), also increased with increasing testis levels of PPP1CC2 (Fig. 34B). VSL recorded for rescue lines eTg‐M7 (23.7 ±

0.9 µm/sec) and eTg‐F1 (25.5 ± 3.8 µm/sec) were significantly lower compared to control

(52.7 ± 2.8) but improved with increasing levels of PPP1CC2 in transgenic lines pTg‐M26

(43.4 ± 5.3 µm/sec) and pTg‐M30 (49.1± 8.5 µm/sec) with values comparable with that of control. A measure of VCL also showed similar trends with significant deviation between mean values for eTg‐M7 (74.2 ± 8.5 µm/sec) and eTg‐F1 (89.0 ± 7.0 µm/sec) with that of control mice.

91

Figure 33: Motility analysis of mature caudal sperm from rescue animals with varying levels of PPP1CC2. Computer assisted sperm analysis of freshly prepared mature caudal epididymal spermatozoa was performed on sperm from adult (8‐12 weeks old) mice from different transgenic rescue lines. (A) Both total percent motile (black bars) and progressive motility (grey bars) increased as levels of PPP1CC2 increased. Significant difference in mean values of progressive motility was observed in lines eTg‐M7 and eTg‐F1 when compared to control group (+/‐). (B) The velocity parameters VAP (average path velocity; grey bars), VSL (straight line velocity; white bars) and VCL (curvilinear velocity; black bars) showed steady increase with increasing expression of PPP1CC2. Mean values for VAP, VSL and VCL showed significant difference when compared to control (+/‐) in animals with low testis levels of PPP1CC2 (eTg‐M7 and eTg‐F1).

92

2.1.9 Fertility is restored only in transgenic rescue lines with PPP1CC2 levels equal

to or above 50% of that of control (+/­) animals:

To confirm the relevance of these data, we tested the fertility of adult male rescue mice for their ability to sire pups. The results are summarized in Table‐16. Males from transgenic line eTg‐M7, with testis levels of PPP1CC2 less than 50% of control were all infertile, whereas males from transgenic rescue line eTg‐F1 were subfertile. As seen in

Table‐16, out of thirteen rescue males from this line that were tested only four (~31 %) were fertile and sired pups. In addition, for the fertile members of this line, the average litter size of 7.2 ± 1.4 pups, is significantly lower than control groups. Besides the average elapsed time between placing a male with a female and birth of pups and also the interval between two successive litters was longer (23.1 ± 0.7 days) compared to higher expressing lines (pTg‐M26 and pTg‐M30) and controls. However males from the eTg‐F1 line

(subfertile) showed normal mating behavior as evidenced by the presence of vaginal plugs in females. Thus, as the results demonstrate, many of these copulatory events were unsuccessful in fertilizing females. Transgenic lines with higher PPP1CC2 levels (pTg‐M26 and pTg‐M30) were fertile with mean litter sizes of 9.6 ± 1.0 and 8.7 ± 0.7 respectively that are comparable to control (9.0 ± 0.71pups). The average number of litter’s sired (data not shown) and the average interval between deliveries of litters (22.3 ± 0.5 days for pTg‐M26 and 20.4 ± 0.7 days for pTg‐M30) were comparable to Ppp1cc +/‐ controls (20.9 ± 0.3 days).

93

Table­15: Fertility data.

Transgenic Number of males No of males Mean time (in days) elapsed Average litter Fertility lines tested for actually between two consecutive size status fertility (n) fertile litters ± SEM (min – max)

eTg-M7 10 0 0 0 Infertile

Sub- eTg-F1 13 4 23.1 ± 0.7 (21-27) 7.2 ± 1.4 fertile

pTg-M26 5 5 22.3 ± 0.5 (19-25) 9.6 ± 1.0 Fertile

pTg-M30 3 3 20.4 ± 0.7 (18-23) 8.7 ± 0.7 Fertile

Control (+/-) 3 3 20.9 ± 0.3 (20-21) 9.0 ± 0.71 Fertile

The Kruskal‐Wallis non‐parametric one‐way ANOVA by rank was employed to test whether samples originated from the same distribution. Results of Kruskal‐Wallis were analyzed post­hoc by Dunn’s procedure for performing two‐tailed multiple pairwise comparisons. Differences were considered significant if p<0.05. a, b…denote significantly different groups.

SEM denotes Standard Error of the Mean.

2.1.10 Ppp1cc2 message levels and protein levels increases with overexpression of PPP1CC2 in testis:

The analysis of the rescue lines clearly showed PPP1CC2 above a high threshold level of 50% and above of that of heterozygous control is critical for sperm morphogenesis, initiation of motility and fertility in male. PPP1CC2 levels below the threshold adversely affect sperm development and its function. Here we set to determine whether overexpressing PPP1CC2 in wildtype (+/+) testis affected spermatogenesis and sperm function. Wild type mice were crossed with Ppp1cc2 transgenic positive mice (of pTg‐M30

94 line, with expression equal to control; +/‐; Tg+) to derive lines with three copies of Ppp1cc2 gene (two endogenous and one transgenic copy; +/+, Tg+). Analysis of +/+;Tg+ adult mice show that they are phenotypically indistinguishable from +/+ and +/‐ control littermates in terms of sperm morphology, motility and number. Quantitative western blot analysis was perfprmed to detect any over expression as compared to controls. As evident from Fig. 35

(A, B), introduction of an additional copy of the Ppp1cc2 gene in form mini‐transgene resulted in 20% and 35% increase in expression of PPP1CC2 over wildtype and heterozygous control respectively. Northern blot analysis was performed in parallel to detect over expression of Ppp1cc2 message levels and how its level compares to that of the protein levels. Fig. 35 (C, D) show that there is dramatic increase in Ppp1cc2 mRNA levels in testis of +/+; Tg+ compared to +/+ mice and +/‐ mice. There is about ~35% increase in message levels that translate into 20% over expression of PPP1CC2 levels compared to wildtype, similarly ~70% increase in mRNA levels that that causes 35% increase of the protein over +/‐ control littermate. Thus the result suggests that there is linear correlation between gene copy number and message levels to protein levels.

95

Figure 34: Linear correlation between Ppp1cc2 gene copy number and message and also protein levels. (A) Western blot showing more or less equivalent levels of PPP1CC2 expression in wild type (+/+), Tg positive wild type (+/+; Tg+) and heterozygous (+/‐) animal having two copies, three copies and one copy of gene coding PPP1CC2 respectively. (B) The corresponding densitometry bar graph, whicg shows 20% and 35% increase in PPP1CC2 levels above +/+ and +/‐ for +/+;Tg+. (C) Northern blot showing the message levels corresponding to Ppp1cc2 isoform in the testis of +/+; +/+;Tg+ and +/‐ animal. (D) 35% and 70% increase in message levels in +/+;Tg+ animal with 3 copies of Ppp1cc2 gene is noted compared to +/+ and +/‐ animal. Increase in Ppp1cc2 m‐RNA levels does not accompany corresponding increase in levels of PPP1CC2 protein in the testis.

2.1.11 Effect of over expression of Ppp1cc2 on PP1 regulator, Ppp1r3 (I­3) in testis:

Next I tested for the expression level of PP1 regulator, Ppp1r3 (I‐3) in mice overexpressing Ppp1cc2 in testis. The purpose is to detect whether increased levels of the

Ppp1cc2 message in testis in some way regulate the expression of the I‐3, since we have shown previously that I­3 levels decrease in Ppp1cc null mice compared to control (+/+). As shown in the Fig. 36, we could not detect any difference in expression levels of I‐3 message netween +/+; +/+, Tg+ and +/‐ mice.

96

Figure 35: Ppp1r3 expression does not changes with Ppp1cc2 overexpression in testis. We used the probe against the full length I‐3 probe that detects the testicular isoform at position ~0.7 kb.

2.1.12 Designing and developing mini EGFP fusion construct to test the minimal promoter region of Ppp1cc gene promoter to drive testis predominant expression of

PPP1CC2 and the function of its C’­tail.

The essential requirement for PPP1CC2 among all the PP1 isoforms is surprising given the high degree of homology in the amino acid sequences of the four isoforms

(~90%). The catalytic core of the enzyme is conserved, while the major differences between the isoforms reside in their C‐terminal regions. Indeed, PPP1CC2 has a unique

~22 amino acid carboxy terminal tail, absent from other PP1 isoforms, and it is significant that the carboxy terminal tail is nearly completely conserved in its entirety among all the mammalian species for which annotated genomic databases exists. This indicates a

97 function specific to PPP1CC2, but to date the role of this carboxy tail has not been elucidated. To answer this longstanding question I attempted to force express PPP1CC1 in developing germ cells to see whether PPP1CC1 can substitute for PPP1CC2. Two possible outcomes are expected from the study:

(a) PPP1CC1 completely restores spermatogenesis and fertility. This will suggest that the

C‐tail of PPP1CC2 is redundant it terms of unique function, since PPP1CC1 can do the

same role without it.

(b) PPP1CC1 cannot restore fertility but partially restores spermatogenesis. This finding

will suggest an isoform specific role for PPP1CC2 that can be attributed to its carboxy

C‐tail.

Currently studies are on going to answer this questions (outline in Aim‐II) novel construct expressing PPP1CC1. However these approaches are indirect way of finding the role for

PPP1CC2 C‐tail. Thus in parallel to this we attempted to design an experiment where we can answer the role of the C‐tail in a more direct fashion. For this purpose I have designed a unique transgene construct, Fig 53. In this construct we have fused the 22 amino acid C‐tail of PPP1CC2 to the 3’‐end of the coding sequence of enhanced green fluorescent protein,

EGFP in one reading frame. The expression of the transgene cassette was directed under the endogenous promoter of the Ppp1cc gene. However we have used a much shorter fragment for use as a promoter region, 0.725 kb instead of 2.6 kb (used earlier). The transgene is already made and ready to be injected to derive founder lines. From analysis of the founder lines derived from this transgenic experiment we attempt to answer the following questions:

98

1. Testis specific region of the Ppp1cc promoter: Whether the regulatory elements

predicted from in‐silico analysis (A‐Myb, Sp1 and Spz1) that is present between +272 to

– 180 bp relative to the transcription start site is sufficient enough to drive testis

specific expression of the transgene.

2. Structural role of C’tail: Secondly we also want to find whether any structural role

played by the C‐tail in anchoring PPP1CC2 to specific structures within the sperm, for

eg: fibrous sheath, ODF, mitochondrial sheath etc. By tracking the EGFP fluorescence in

mature spermatozoa we will be able locate the scaffolding role of C‐tail if any.

3. Role of C’­tail in binding to PPP1CC2 specific regulators and/or substrates: Thirdly

if proper spatio‐temporal expression of EGFP is restored by the promoter fragment will

result in over expression of the C‐tail in testis. We speculate that if C‐tail play role in

binding to specific partners, then transgenic expression of the EGFP‐C‐tail will compete

with wildtype PPP1CC2 in binding with these interactors and disrupt the binding. If the

later is true we might see a graded series of phenotype like the ones in rescue studies

involving low levels of PPP1CC2 depending upon the extent of disruption of PPP1CC2

interactions with its interactors.

99

Figure 36: Design of the construct for endogenous promoter driven EFP­PPP1CC2 (C­ tail) fusion protein. (A) Shows the linearized plasmid map with the all the restriction sites. The promoter region of the Ppp1cc gene spans from +235bp to ‐517 bp relative to transcription start site. The EGFP coding sequence is in frame with the 81 nucleotide that corresponds to the 22 amino acid C‐tail. (B) Circular map of the plasmid. (C) Restriction digestion of the final plasmid. The transgene of length ~1.79kb is released by digestion with KpnI and SacI.

100

Sub­aim II:

2.2 Isoform specific role of PPP1CC 1 is sperm development, spermatogenesis and male fertility.

Background and rationale:

In the preceding sections under Aim‐I we have clearly demonstrated the spatio‐ temporal expression pattern for the two PPP1CC isoforms in testis. Using northern analysis in developing testis and IHC we have shown that Ppp1cc1 message is present during days 5 to day 10 when sertoli cells, spermatogonia and pre‐leptotene spermatocytes are the most prevalent cell types, where as PPP1CC2 is expressed almost exclusively in developing germ cells and absent in sertoli cells. This expression pattern suggests a crucial role for the

PPP1CC2 in germ cell development. To test this we transgenically expressed PPP1CC2 in the developing germ cells driven by the endogenous promoter and the Pgk2 promoter respectively. Results from these experiments showed that PPP1CC2 when expressed only in the developing germ cells in testis of Ppp1cc ‐/‐ mice at high threshold levels is sufficient enough to restore spermatogenesis and fertility in otherwise infertile male mice. These studies also demonstrate for the first time that PPP1CC2 in absence of PPP1CC1 is adequate for reversing male infertility phenotype in Ppp1cc ‐/‐ mice. This posed an interesting question as to whether PPP1CC1 itself can restore spermatogenesis and male fertility when forced to express at high levels in the developing germ cells in place of

101

PPP1CC2. The rationale for the hypothesis is that PPP1CC1 and PPP1CC2 isoforms are virtually identical except at their extreme C’‐termini. The entire catalytic core that determines its substrate specificity and binding of regulators are identical to both isoforms except that PPP1CC2 has an additional 22 amino‐acid COOH‐tail absent in PPP1CC1, Fig. 37.

Thus we speculated that PPP1CC1 should be able to restore spermatogenesis and fertility in Ppp1cc ‐/‐ mice when expressed at sufficient levels in the developing germ cells.

Figure 37: The PPP1CC isoforms vary only at their extreme C­termini.

2.2.1 Generation of PPP1CC1 transgenic rescue lines:

To test whether PPP1CC1 can substitute PPP1CC2 in developing germ cells we designed a construct whereby the expression of Ppp1cc1 was driven by the 2.6 kb endogenous promoter of the Ppp1cc gene. The outline of the construct is shown in Fig. 38, in which two mutation were introduced in the transgene to disrupt the splice ‘donor site’

(GT to GC) and the splice ‘acceptor site’ (CAG to ATT). The site directed mutagenesis thus introduced was aimed to inhibit the splicing event that gives rise to Ppp1cc2 message from the Ppp1cc1 pre‐mRNA, thereby forcing the expression of Ppp1cc1 in developing germ cells.

102

A genomic fragment +1 to ‐2.6kb upstream was used as endogenous promoter (the same one used for driving transgenic PPP1CC2 expression) to drive the expression of the transgene. Injection of the transgene in one‐celled embryo and its implantation gave rise to

83 animals. These animals were subsequently genotyped to identify eleven transgene positive founder animals. Each of these founder animals were crossed in a scheme identical to the one mentioned above to establish Ppp1cc1 transgenic rescue animals (Ppp1cc ‐/‐;

Tg+), Table‐17.

Figure 38: The design of the mini gene cassette for endogenous promoter driven transgenic expression of PPP1CC1. (a) Linear restriction map of the plasmid construct. (b) The circular map of the mini gene. (c) The gel picture showing the position and size of the relieved mini gene on double digestion with NotI and SacI in contrast to the single or uncut plasmid.

103

Table­16: Statistics of rescue founder lines for Endogenous promoter driven Ppp1cc1 lines.

2.2.2 Expression and detection of PPP1CC1 in transgenic rescue line:

To detect the levels of the transgenically expressed PPP1CC1 levels in the different rescue founder lines testis extract were made and westernblot was performed. The first among the rescue line that was opened was e1Tg‐M20. We could not detect any PPP1CC1 expression in these animals in testis or any other somatic tissues included in the study

(brain, spleen, kidney, liver, heart) compared to control. Similar outcome with almost no detectable levels of PPP1CC1 was observed for other lines as well, Fig. 39(B, C). The only lines for which PPP1CC1 expression could be detected by western blot were for lines e1Tg‐

F2 and e1Tg‐M37 (Fig. 39D). Of these in the line e1Tg‐F2, PPP1CC1 signal were detectable both in brain and testis but at levels which is clearly lower compared to corresponding tissues in control (+/‐) animal. In line e1Tg‐M37, though levels were somewhat detectable in brain but the PPP1CC1 signal was weak in testis. Thus we selected the line e1Tg‐F2 which is the highest expresser of PPP1CC1 among all the other founder lines for our subsequent analysis of phenotype. In order to determine the levels of PPP1CC1 expressed in this line we compared it with that of the other PPP1CC2 transgenic rescue line (N‐line and AK‐line). In Fig. 40, the comparison of the levels where protein estimated quantities of testis extract were loaded onto gel for western blot analysis. The intensities of 100ug band

104 in lane1 (of e1Tg‐F2) corresponds to 50 ug band in lane5 (corresponding to AK‐line) and

25 ug band in lane9 that corresponds to N‐line by visual inspection. This suggests that

PPP1CC1 levels in e1Tg‐F2 rescue line one‐half of that of AK‐line (50%) and approximately about one‐eigth (~12.5%) of that of N‐line, Fig. 40.

105

Figure 39: Detection of PPP1CC1 across various transgenic rescue lines by western blot. (A) No expression of the transgene was noted in rescue line e1Tg‐M20 in testis or in any other tissue studied. 30 ug of total protein was loaded in each lane. (B, C) Similarly no expression was detected in testis or in brain of lines e1Tg‐M24 and e1Tg‐M27. (D) In lines e1Tg‐M15, e1Tg‐F2 and e1Tg‐M37 expression was detected for the first time in brain albeit in much lower amounts compared to control. However only in e1Tg‐F2, testis expression was observed at detectable levels though quite low compared to testis. (Br‐brain; Ts‐testis; Spl‐spleen, Kid‐kidney; Liv‐liver; Hrt‐heart).

Figure 40: Comparison of PPP1CC1 levels between other transgenic Ppp1cc1 rescue lines. AK­line, where Ppp1cc trangene expresses both PPP1CC1 and PPP1CC2 driven by Pgk2 promoter. N­line, different transgenic line where PPP1CC1 expression alone was driven by the same Pgk2 promoter. Protein estimated extracts were loaded for each samples. The band intensities were matched by visual inspection where 100 µg in lane1 is comparable to 50 µg in lane5 and 25 µg in lane9.

The low levels of the transgene expression raised the speculation as to whether the complete transgene was excised from the vector backbone by restriction digestion and if the full length mini‐gene cassette was integrated in the genome during transgenic mouse production. Though all the rescue animals were checked by genomic PCR for the presence of the transgene, however we used a primer pair (see materials and method) that cover

106 only a portion of the transgene cassette (Exon6 forward and SV40 Rev). To test these possibilities we digested tail biopsies and did genomic PCR using primer pairs‐ ‘ATG forward’ (that marks the beginig of the ATG initiation codon) and ‘SV40 reverse’. The primer pair is supposed to detect the full length of the coding region along with 3’UTR and the SV40 Poly A tail. Figure. , shows that PCR was successful in detecting the targeted region by the presence of the amplicon of size that matches with the expected position.

This suggests the presence of the complete coding region of the transgene cassette. Next we performed RT‐PCR on testicular cDNA to detect message levels if any for the line e1Tg‐M15 for which no PPP1CC1 was detected and also for lines e1Tg‐F2 and e1Tg‐M37 for which transgene expression was confirmed by western blot. We used the primer pairs ‘Intron7‐ forward’ (internal) and ‘3’UTR‐reverse’. RT‐PCR showed bands for tall the three lines suggesting the presence of the message for the transgene in these lines. We also performed

Northern analysis for detection of the Ppp1cc1 message in lines e1Tg‐M20 and e1Tg‐M27, but we were not able to detect the message probably due to low sensitivity of the northern technique in detecting low abundant transcripts.

107

Figure 41: Detection of Ppp1cc1 transgene and its message in rescue lines. (A) Genomic PCR was carried out on tail DNA using primer pair ‘ATG‐forward’ and ‘SV40‐ reverse’. PCR showed an expected band size of ~2.1 kb corresponding to the full coding region + 3’UTR + SV40PolyA tail. Plasmid DNA construct for the transgene was used as positive control. Whereas germ cell knock out tail DNA was used as –ive control in the PCR. (B) RT‐PCR was performed on testicular (and brain) c‐DNA derived from total RNA. Primer pair ‘Intron‐7 forward (internal)’ and ‘3’UTR‐reverse’ was used for amplification. –RT denotes negative RT control to show absence of genomic DNA contamination.

108

2.2.3 Rescue line e1­Tg­F2, with low levels of transgenic PPP1CC1 expression has very low sperm number with gross deformity in morphology:

Rescue lines with any detectable to barely detectable levels of PPP1CC1 were opened to see for signs of restoration of spermatogenesis. For these adult mice of 8 weeks or older were opened and epididymis was dissected and squeezed to release mature epididymal spermatozoa. However in these lines no spermatozoa could be isolated from the epididymis and the phenotype of these mice are reminiscent of Ppp1cc knock out (‐/‐) mice with no expression of either PPP1 isoforms. Thus for all subsequent studies we selected rescue line e1Tg‐F2 with low but detectable levels of PPP1CC1. In these animals though we could isolate spermatozoa from the epididymis but the sperm numbers were significantly lower (~ 1 x 105 sperm/ml), about 99% lower than that of control +/+ (~ 3.9 x

107 sperm/ml, Table‐15). As shown in the Fig, 42, DIC picture of fixed epididymal content at 20x one can barely detect any sperm due to its very low numbers. Mostly round and elongating spermatids are visible much like the Ppp1cc null mice. A few sperm that could be isolated were analyzed for morphology at 60x magnification. As it is evident from Fig.

43, most of sperm bore gross morphological defects. These defects include complete absence or highly redundant mitochondrial sheath, or in some cases it is the mitochondrial sheath is frizzled in appearance. The head in most of them are either malformed or incomplete or with gross abnormality. This abnormalities of the sperm thus seen is similar to other transgenic PPP1CC1 line (AK‐line and N‐line) but different from those of rescue line with low levels of PPP1CC2 (se Fig ). As expected for these lines males were infertile.

109

Figure 42: Sperm numbers were significantly lower in rescue line e1Tg­F2 with low levels of PPP1CC1. DIC pictures of fields at 20x magnification show negligible sperms (white arrow) in these animals compared to control (+/‐). It is to be noted that round cells (representing round spermatids (red arrow) and elongating cells (elongating spermatids, red arrow) mostly populate the epididymis. The average sperm count is ~ 1 x 105 sperm/ml in these lines which is significantly lower than that of control.

110

Figure 43: Gross morphological defects in sperm isolated from e1Tg­F2 line: Mitocondrial sheath show wide range of abnormalities. It is either absent (a, f), or degenerated (b, d) with fuzzy or loosely bound appearance or stunted (c, e). The head also varies considerably in abnormalities, ranging from malformation (b, e, f) or deformed (c, d).

2.2.4 Alternate approaches to increase expression of PPP1CC1 in developing germ cells:

The generation of the transgenic rescue lines with endogenous promoter driven expression of PPP1CC1 in the developing germ cells was the third in our attempt to overexpress PPP1CC1 in testis. In all our previous effort including the present one we were not successful in driving adequate expression of PPP1CC1 in the germ cells. Thus to answer the question whether PPP1CC1 can substitute PPP1CC2 when expressed at sufficient levels in spermatogenic cells remained largely unanswered. Thus to better the expression we have designed alternative approaches.

111

2.2.4.1 Designing and developing replacement construct to create isoform specific

knock out of Ppp1cc gene:

We planned to create PPP1CC2 isoform specific knock out of the Ppp1cc gene. For this I designed a replacement construct, Fig. 44. The plan for generation of knock‐in mouse that specifically expresses Ppp1cc1 and not Ppp1cc2 in developing germ cells is outlined in the top panel of Fig. 44. The design of the knock‐in replacement construct is shown in the bottom panel of Fig. 44. It was designed to have a long and a short arm of homology and also carrying a mutation in the splice acceptor site before exon8 (CAG to ATT) of Ppp1cc1 cDNA, besides also carrying the selectable marker Neomycin (driven by Pgk promoter) and

Thymidine kinase (driven by CMV promoter). The objective is to perturb the splice acceptor site for exon 8 in the Ppp1cc genomic locus, by incorporating a fragment bearing mutation for the acceptor site introduced by homologous recombination in the ES cells. The

ES cells that are positive for desired recombination event will be injected into inner cell mass of blastocysts, and later implanted in to surrogate females. Progeny derived from the female will be screened for chimeras that are positive for germ line transmission. Thus mice homozygous for the altered Ppp1cc genomic locus will express PPP1CC1 as in wildtype at the same time will express the protein in tissues and cells supposed to express

PPP1CC2. The later is because of the inability of the pre‐mRNA to undergo splicing because of the mutation in the splice donor site ‘CAG to ATT’. The construct has been made and ready to be injected to produce genetically altered mice.

112

The Final construct:

113

Figure 44: The plan for generating the Ppp1cc2 isoform specific knock out mice. (A) The long homology arm of size ~3.5 kb spans from intron3 to middle of inron7. This is followed by the fragment carrying the mutation for the splice acceptor site (CAG to ATT) and finally the short homology arm that is 2.5 kb long. (B) The final construct that was developed using the pEasy‐Flox vector backbone that is ready to be injected. (C) Shows the restriction map of the final plasmid. In C1, the 1st lane shows the plasmid of length~ 13.5 kb linearized with NotI. The 2nd lane shows the restriction fragments generated by digestion with NotI and SalI. The 5.2 kb fragment (lower blue arrow) encompasses the full length of short arm + the Neo cassette and the whole of mutant fragment containing exon 8. The 8.2 kb (top blue arrow) fragment is rest of the backbone that also includes the long arm. In C2, the entire short arm (green arrow) is released on digestion with NotI and BamHI and the rest of the backbone (red arrow). C3 represents the restriction fragments produced by digestion with HindIII and SalI. It releases the entire long arm (excluding the mutant fragment) of size 3.5 kb (orange arrow) and other short fragments are also produced of expected size due to presence of multiple HindIII internal sites (white arrow) within the backbone.

114

2.2.4.2 Generation of a novel PPP1CC1 expression construct that mimics the Ppp1cc2 cDNA.

We took a second approach whereby a novel transgene mini‐cassette (fourth in the

Ppp1cc1 series) was designed so that it codes for PPP1CC1 by mimicking to a large extent

Ppp1cc2 cDNA in its coding region. The purpose of this novel transgene is twofold; (1) firstly to get rid of intron7 which we predict might be responsible for contributing towards instability of the Ppp1cc1 mRNA. (2) Secondly to fool the system in recognizing the message coded by the transgene as Ppp1cc2 though it expresses Ppp1cc1. Thus I created the fourth transgene construct as shown in the Fig. 45.

Figure 45: Generation of novel Ppp1cc1 transgene construct. The top panel shows the intron/exon organization of the Ppp1cc gene. Exon 7 (black/grey) and 8 (green) are color‐

115 coded. It also shows the cDNA of the Ppp1cc2 isoform arising due to the splicing event between internal spice donor site in exon7 and the splice acceptor site before exon8. The bottom panel shows the map of the new transgene mini cassette. Here the exon8 and the entire 3’UTR were flushed with exon7. However the message coded by the transgene cannot translate Ppp1cc2 due to presence of Stop codon (TAG) at the end of exon7 and thus will only express Ppp1cc1. Pgk2 promoter was used to drive the expression of the transgene.

The above transgene construct was purified and injected into one‐celled embryo (at CASE

Western Transgenic Core) and pups have been derived from them. Genotyping PCR have identified 19 positive founders, Table‐18. Some of these positive founders are already set for breeding to transmit the transgene and efforts are underway to derive rescue animals for each of these founder lines.

Table­17: Statistics of founder line derived from novel Ppp1cc1 transgene.

Aim‐III

Determine the requirement of Ppp1cc gene in pre­meiotic germ cells and Sertoli cells

116 117

3.1 Background and rationale:

Thus far we have shown that PPP1CC2 is the sole PPP1CC isoform restricted to developing germ cells whereas PPP1CC1 is localized mainly to Sertoli cells, spermatogonia, pre‐leptotene spermatocytes but absent in developing germ cells. The goal of this study was to determine whether presence of Ppp1cc gene and its product PPP1CC1 in the somatic

Sertoli cells, steroidogenic Leydig cells and in undifferentiated spermatogonia play any role in germ cell development by providing signals necessary for cross‐talk between these cell types and for sustenance and development of germ cells. To test this we attempted to conditionally knock down the Ppp1cc gene in the developing germ cells from primary spermatocytes onwards however keeping the gene intact in all other tissue in testis and in the whole body.

3.1.1 Selective deletion of Ppp1cc gene in developing germ cells to generate germ

cell knock out mice (Ppp1cc∆flox/∆flox; TgStra8­Cre+ ):

In‐order to generate germ cell knock out mice we obtained the Ppp1cc+/flox mouse from Dr. Angus Nairn from Rockefeller University, New York. These mice harbors a pair of

34 bp LoxP sites flanking exon2 and exon3 (floxed) introduced by homologous recombination, Fig. 46. The Ppp1cc+/flox mice were bred to produce mice homozygous for the floxed allele, Ppp1ccflox/flox. Transgenic mice expressing CRE recombinase driven by the

1.4kb Stra8 promoter was purchased from Jackson lab (deposited by Dr. Robert Braun).

Female Stra8‐Cre mice were crossed with Ppp1ccflox/flox male mice to obtain Ppp1cc+/Δflox;

118

TgStra8­Cre+ mice. In the next generation Ppp1cc+/Δflox; TgStra8­Cre+ mice was crossed with

Ppp1ccflox/flox mice to obtain germ cell knockout mice (GcKO) with the genotype

Ppp1ccΔflox/Δflox; TgStra8­Cre+ and Ppp1ccflox/flox mice for use as littermate control (see the mating scheme in materials and methods).

Figure 46: Schematics diagram showing the floxed allele of the Ppp1cc gene. (A) Shows the restriction map and the strategy used to generate the floxed allele using

119 homologous recombination (from Angus Nairn’s Lab). (B) A simple schematic showing the exon/intron (solid bar/open box) diagram of the floxed allele generated as a result of genetic manipulation. The loxP sites are depicted as solid green arrows.

Genotyping for the floxed allele was done using the primers mentioned in the materials and method section. As shown in the Fig. 47B, presence of a single copy of the floxed allele

(+/flox) was identified as doublet band at positions 750bp and 600bp corresponding to the floxed allele and the wildtype allele respectively, where as absence of floxed allele in widtype mice (+/+) is denoted by a single band of 600bp and finally animals homozygous for the floxed allele (floxed/floxed) show a single band of size 750 bp (see Fig. 47B). After confirming the genotype of the floxed allele the animals were subsequently genotyped for the iCre transgene using the primers mentioned in the materials and method section. The presence of the Stra8‐Cre transgene was detected by a single 179 bp band, Fig. 47C, confirms the specificity of the expression of iCre message driven by the Stra8 promoter.

Reverse transcriptase PCR detects iCre message only in testis and was absent from all other tissues examined (brain, spleen, kidney, liver), Fig. 47C. In the above genetic crosses trangene transmission occurred at expected Mendelian ratio. For all analysis adult male

GcKO mice of age 8‐12 weeks were used along with age matched littermate as control

(Ppp1ccflox/flox). We next performed northern blot to detect relative message levels of

Ppp1cc2 isoform in the GcKO animal compared to that of control to confirm the deletion of the Ppp1cc locus. As shown in Fig. 47D, CRE mediated recombination event resulted in marked reduction of the Ppp1cc2 message compared to control (single headed arrow in Fig.

47D). However Ppp1cc1 message remained unaltered in control compared to GcKO animal,

120

Fig. 47D. We could also detect a lower message of size ~1.0 kb corresponding to the truncated message generated as result of splicing occurring between exon1 and exon4

(double headed arrow, Fig. 47D).

Figure 47: Confirmation of deletion of Ppp1cc gene in developing germ cells in testis. (A) Schematic representation of the wildtype allele, floxed allele and the deleted allele of Ppp1cc gene. (B) Genomic PCR of the floxed allele. The wildtype +/+ allele gives a band at 600 bp, where as +/flox allele gives a band at 600 bp and 750 bp respectively (the middle band around 675bp is a non‐specific band; the homozygoys floxed allele give a single band at 750 bp. (C) Reverse transcriptase‐PCR, showing the specificity of Cre expression only in testis compared to the somatic tissues where it not detected. (D) Northern blot showing the effective deletion of Ppp1cc allele result in drastic lowering of Ppp1cc2 message (single headed arrow) compared to control (flx/flx). The truncated message resulting from

121 splicing between exon1 and exon4 is present as lower band at around ~1.0 kb region (double headed arrow). However the Ppp1cc1 message levels are indentical in GcKO and in control.

3.1.2 Conditional deletion of Ppp1cc gene drastically reduces PPP1CC2 levels in

testis:

Data from northern blot demonstrated that the message levels of Ppp1cc2 were significantly lowered in GcKO male testis but not completely absent. In order to determine whether PPP1CC2 protein levels in these germ cell knock out animals were effectively ablated we ran western blot on protein estimated testis extracts for GcKO and control animals. As evident from the Fig. 2BA, CRE mediated knockdown of Ppp1cc gene resulted in drastic reduction but not complete loss of PPP1CC2. In all three GcKO that were analyzed the band intensities from lane corresponding to 40ug and 20ug of that of GcKO animals#1 and 2, testis were lower compared to control (flx/flx) lane with 2.5 ug of total protein. A

GcKO #3 testis level of PPP1CC2 was barely detectable. This low residual level of PPP1CC2 in the testis is due to incomplete deletion of the Ppp1cc gene resulting from lack of complete penetrance of the CRE expression in the pre‐meiotic germ cells. Fig. 28B show that PPP1CC2 levels in testis of new GcKO animal is about 12.5 % (1/8th) of that of its littermate control. PPP1CC2 levels form the same GcKO animal was also compared with that of other rescue lines (eTg and pTg). For comparison by western blot, 10 µg of total protein form testis extract from each animal were loaded to corresponding lane. In Fig.

28C, analysis of the band intensities by visual inspection (also by densitometry) confirms

122 that PPP1CC2 levels in the GcKO animal in comparable to rescue line pTg‐M3(12.5% of that of control heterozygous animal). These results suggests that due to incomplete deletion of

Ppp1cc gene there is not complete loss of PPP1CC2 expression, however its level is markedly reduced to about 12.5% or less of that of control Ppp1cc (flx/flx) mice. However it also to be noted that GcKO animals varied among themselves in their level for PPP1CC2 expression within this range for all the animals studied so far. As expected PPP1CC2 levels remains unchanged in brain between GcKO and control animals further confirming the specificity of CRE mediated recombination event in testis, Fig. 28D.

123

Figure 48: PPP1CC2 levels are drastically reduced in testis of GcKO mice. (A) Western blot analysis showing that PPP1CC2 levels in GcKO is significantly reduced compared to control. (B) PPP1CC2 levels in GcKO testis is reduced to about less than 12.5 % (1/8th) that of control mice. (C) Western blot data comparing the PPP1CC2 levels in testis of GcKO mice to that of different rescue lines (eTg and pTg). (D) Another western blot data showing that PPP1CC2 levels in brain. It is to be noted that in brain its levels remains unchanged compared to control, thereby confirming the specificity of CRE mediated deletion of Ppp1cc gene.

3.1.3 PPP1CC2 exhibit a mosaic expression pattern within the cross section of testis

and seminiferous tubules of germ cell knockout animals.

To confirm the partial deletion of the Ppp1cc gene in developing germ cells we did immunohistochemistry to see if PPP1CC2 signal within the seminiferous tubules of GcKO mice can be detected. The IHC result showed that a vast majority of the tubules were negative for PPP1CC2 signal thus confirming its absence in these tubules and hence explains the drastic reduction of PPP1CC2 levels. Interestingly however a very small number of tubules were positive for PPP1CC2 signal, thus explaining the persistently low

(≤ 12.5%) but detectable levels that can be seen in western blot. A closer examination of these tubules at higher magnification revealed that in these tubules expression pattern of

PPP1CC2 is very different from that of wildtype and exhibit a mosaic pattern of expression.

In the control PPP1CC2 is present in all the developing germ cells uniformly starting from

124 early pachytene spermatocytes to elongating spermatids from the periphery to the lumen of the tubule, Fig. 49F. In GcKO animals, the tubules expressing PPP1CC2 vary among themselves in terms of the size of the area positive for PPP1CC2 signal. As shown in Fig. 49

(A‐F), the tubules vary from 1/4th (double head arrow, B), ½ (blue arrow, B,E) to about complete expression (white arrow, A) of PPP1CC2 in the germinal epithelium as compared to control. Within a given GcKO animal the proportion of PPP1CC2 positive tubules is stochastically determined by the Ppp1cc gene deletion event that occurs randomly, ranging from 100% in efficiency to nearly zero in certain tubules. This explains the variable expression of the PPP1CC2 levels between the testes of GcKO animals.

125

Figure 49: Immunohistochemistry showing mosaic expression of PPP1CC2 in the testis of GcKO mice. (A, A’), (B, B’) are sections of testis at 10x magnification. (D, D’) and (E, E’) are sections at 20x. (C, C’) and (F, F’) are negative control and positive control respectively. It is to be noted that some tubules are positive for PPP1CC2 signal and the expression is mosaic in these tubules.

126

3.1.4 PPP1CC1 levels remains unchanged in testis of GcKO mice:

Next we looked into the expression of the PPP1CC1 in testis. Given the predominant localization of PPP1CC1 expression in Sertoli cells, spermatogonia and pre‐leptotene spermatocytes we reasoned that targeted deletion of Ppp1cc gene in developing germ cells would not affect the levels or distribution of PPP1CC1 isoform within the testis of GcKO animals. Fig. 50(top panel), PPP1CC1 levels were found to be more or less constant in GcKO testis when compared to control suggesting that PPP1CC1 expression is not affected by

Ppp1cc gene deletion in developing germ cells, Fig. . To confirm these findings we performed immunohistochemistry of testis cross‐section of GcKO mice to detect PPP1CC1 localization within the seminiferous cross section. Fig, 50(lower panel), shows that in GcKO male testis PPP1CC1 signal could be detected within the Sertoli cells, spermatogonia and pre‐leptotene spermatocytes as in control (in Fig. 20).

127

Figure 50: PPP1CC1 levels remain unaltered in the testis of GcKO male mice. (A) Western blot on protein estimated testis extracts showing more or less equal PPP1CC1 levels in these animals compared to control. 20 µg of total protein from each of GcKO animal and control were loaded on to each lane. β‐Actin show equal loading for each lane. (B)Immuno‐histochemistry showing detection of PPP1CC1 signal in sertoli cells and in spermatoginia as in control (Fig. 20).

3.1.5 Reduced levels of PPP1CC2 in germ cell knockout animal results in infertility.

In order to test the effect of reduced levels of PPP1CC2 in the testis resulting from selective ablation of Ppp1cc gene in spermatogenetic cell, male mice of age 8 weeks or older were subsequently tested for their ability to sire pups. Adult male mice were housed with adult

CD1 females (one or two/cage) for an average of three months. During this time females were checked for copulatory plug and signs of pregnancy and the number of litters born

128 during the period were recorded. Ppp1ccflx/flx males of similar age were used as control. As evident from Table‐19, all GcKO males (n=18) tested so far were infertile for their inability to sire pups. Whereas all control male mice (n=7) that were included in the study were fertile and the average number of pups born per litter were 8.0 ± 0.7(SEM). Together these results suggests for the first time that presence of the intact Ppp1cc gene and its product

(mostly PPP1CC1) has no discernible role in male fertility in absence of PPP1CC2 in developing germ cells.

Table­18: Fertility data for GcKO animals.

Transgenic lines Number of Number of males Average litter size Fertility status males tested fertile + for fertility SEM (n)

Control 7 7 8 ± 0.7 Fertile Ppp1cc flx/flx

GcKO 18 0 0 Infertile Ppp1cc flx/flx ; Stra8­Cre+

3.1.6 Infertility in GcKO male mice results due to of low sperm number and high

proportion of sperm morphological abnormality.

In order to examine potential causes of infertility in male Ppp1cc germ cell knock‐out mice we attempted to recover mature spermatozoa from the caudae epididymides and ductus deferens of

129 these mice and that of control. As evident from Table‐20, statistical analysis of sperm numbers, recovered from the caudae epididymides and vas deferens of GcKO sterile male mice showed significantly lower sperm counts than those of Ppp1ccflx/flx positive control mice (p<0.0004). In the sterile males the sperm count was on an average ~90% lower than that of control signifying an oligospermic condition. We next looked for sperm morphology in the GcKO sterile males using the method used previously for PPP1CC2 transgenic rescue lines. Sperm morphology data are demonstrated in Fig. 51, and summarized in Table‐20. Sperm showed a great degree of different morphological abnormalities (66%) that is significantly different from control (7%) [see Table‐20] and comparable to rescue lines eTg‐F10 and pTg‐M3 (68% and 76% of defective sperm, Table‐10) with low levels (12.5%) of PPP1CC2. However these males though bear overall similarity in morphological defects also differ in terms of severity and range of the abnormalities. A significant proportion of these spermatozoa bear defective mitochondrial sheath (stunted or thinning), Fig.

51(g, h, i) and among the new ones include pin head like sperm (Fig. ), sperm with multiple tail and prematurely released sperm without fully formed head (Fig, 51d,e). Testis weight though reduced in GcKO male mice (104.2 ± 3.7 mg) were not significantly different (p = 0.082) from control ones

(116.1 ± 3.3 mg) suggesting once again that PPP1CC2 apparently have no significant correlation with that of testis weight. The results from study of sperm count and sperm morphology indicates that spermatozoa isolated from GcKO males exhibit both oligo‐ and teratozoospermic condition.

130

Table­19: Comparison of testis weight, sperm number and sperm morphology between GcKO and control animals.

Transgenic Mean testis Mean Sperm number Sperm Morphology lines weight ±SEM (in ±SEM x 107/ml mg) Total number of Normal Defective sperm counted (%)±SEM (%)±SEM (range) (range)

93 + 1.2 7 + 1.2 n=4; 116.1 ± 3.3 n=4; 4.1 ± 0.5 n=4; 1272 Control (90 – 96) (4 – 10) Ppp1cc flx/flx

GcKO 34 + 6.5b 66 + 6.5 b Ppp1cc flx/flx; n=9; 104.2 ± 3.7a n=8; 0.4 ±0.08b n=4;1651 (24 – 53) (47 – 76) Stra8­Cre+

Student t‐test (unpaired) was performed to compare the mean testis weight and sperm numbers between different transgenic lines and control (+/‐) or (‐/‐) groups. Differences were considered significant if P < 0.05 at a confidence interval of 95%.

‘a’ denotes that no statistical significant difference was observed for mean value of between test group and control group.

‘b’ denotes statistical significant difference between test group and control group.

131

Figure 51: Morphological abnormalities of spermatozoa as seen in GcKO males with reduced levels of PPP1CC2. (a, b, c, h) Shows a high proportion of sperm with head bent between capitulum and mid‐piece. White arrows indicate the morphologically abnormal regions and black arrows the corresponding normal region in. (d, e, f, g) Deformed head morphology as seen in these spermatozoa. (g, i) A vast majority of sperm are defective in their organization of the mitochondrial sheath. Stunted mitochondria with regions of thinning are frequently observed in these sperms.

3.1.7 Spermatozoa from GcKO mice are asthenospermic with poor motility

characteristics.

Next we measured the sperm motility parameters in GcKo male mice and that of control as described in Materials and Methods. Sperm isolated from sterile germ cell knockout mice showed poor motility characteristics, Fig. 52. Both percent of motile sperm (40%) and of progressively motile sperm (10.5 %) were significantly less compared to positive controls (83 % and 31%

132 respectively). Sperm velocity parameters including average path velocity (VAP), straight line velocity (VSL) and curvilinear velocity (VCL), were also significantly decreased compared to control, Fig. 52B. VSL recorded for GcKO line (38.4 µm/sec) along with VAP (60.5 µm/sec) and VCL

(113.7 µm/sec) were significantly lower compared to positive control (VSL: 54.5 µm/sec; VAP:

88.3 µm/sec; VCL: 173.8 µm/sec).

Figure 52: Motility analysis of mature caudal sperm from GcKO mice with drastic reduced levels of PPP1CC2. (A) Both total percent motile (black bars) and progressively motile (grey bars) show marked drop in motility a condition called asthenozoospermia. Significant differences from positive control mean values of percent motile sperm were

133 observed for males of GcKO lines, and for percent progressively motile sperm. (B) The velocity parameters VAP (average path velocity; grey bars), VSL (straight line velocity; white bars) and VCL (curvilinear velocity; black bars) also showed concomitant decrease in values. Mean values for VSL and VCL were significantly lower than positive control values in animals.

134

Discussions

A. Promoter activity and splicing maintains differential distribution of PPP1CC1 and

PPP1CC2 within the testis.

We document for the first time in this study, a comparative abundance of messages for the two alternatively spliced Ppp1cc isoforms in various mouse tissues. Messenger RNA for both Ppp1cc isoforms are detected in all tissues whereas Ppp1cc2 is most abundant compared to the Ppp1cc1 in adult testis. It may be noted that signal levels for both isoforms are similar in tissues other than testis suggesting that the high level of Ppp1cc2 in testis could be due to the combined result of testis specific promoter activity and efficient splicing of the pre‐Ppp1cc mRNA (Fig.17A). The fact that Ppp1cc2 message is detected in all tissues suggests basal levels of splicing in these tissues as well. Since the same hybridization probe was used to detect both Ppp1cc1 and Ppp1cc2 (see Table‐1) the relative intensities of the signals in the northern blot are a measure of the levels of the messages. Post natal testicular development is associated with emergence of various different cell types at specific time point during maturation starting from day 0 to 30. This developmental feature of testis serves as a useful means for detecting the spatio‐temporal expression of gene or its product. Result from Figure 19, show the distribution and abundance of all four Ppp1 isoforms in post partum maturing testis. The levels of the PP1 isoforms, Ppp1ca and Ppp1cc1 are relatively high at days 5‐10 post partum when mostly sertoli cells, spermatogonia and Leydig cell populate the testis. This suggests expression of these isoforms in these testicular cell types. With maturation of testis the expression levels

135 of both Ppp1cc1 and Ppp1ca tend to gradually drop with emergence of developing germ cells in the testis from day10 onwards further suggesting their absence in the spermatogenic cells. In contrast the message for Ppp1cc2 rises dramatically at around 15 dpp reaching a maximum in adult testis. This pattern of Ppp1cc2 mRNA expression parallel levels for PPP1CC2 protein found in postnatal developing testis (39). It is notable that basal levels of Ppp1cc2 message, similar to those seen in other tissues, are found in day 5 to day 10 testis, suggesting that the splicing also increases in parallel with increased transcription in developing germ cells. Low levels of postnatal expression of Ppp1cb at all stages suggests that it may be may be expressed in Sertoli cells, spermatogonia and pre‐ leptotene spermatocytes and developing germ cells at basal levels. However, it should be noted that due to difference in probe length that was used for detection of Ppp1ca and

Ppp1cb, relative levels of the messages cannot be compared between one another or to the

Ppp1cc isoforms (see Table‐1). The unique and non‐overlapping expression of the Ppp1cc2 compared to other isoforms in testis suggests a role for this isoform in developing germ cells. Immunohistochemical staining and the western blot (data not shown) analysis confirmed the localization of PPP1CC1 in Sertoli cells, in spermatogonia and pre‐leptotene spermatocytes (Fig. 20). A notable observation made from the IHC is that PPP1CC1 is excluded from early and mid pachytene stage with the onset of meiosis at around 15 dpp and remain absent at all stages of spermatogensis. However the possibility of PPP1CC1 expression in Leydig cells cannot be ruled out though we did not detect any consistent signal in our IHC. In contrast PPP1CC2 expression is detected from early pachytene stage onwards and its expression stays elevated during all the later stages of spermatogenesis,

(Fig 21). The data further demonstrates the Ppp1cc gene promoter and splicing mechanism

136 in controlling the testis specific expression and distribution of Ppp1cc isoforms within the seminiferous tubules. Onset of meiosis following pre‐leptotene stage is marked by increased promoter activity of the Ppp1cc gene resulting in increased Ppp1cc pre‐mRNA and concomitant increase in the splicing mechanism to convert the entire Ppp1cc pre‐ mRNA into Ppp1cc2 message.

B. The Ppp1cc gene promoter is testis specific and functions to maintain testis

predominant expression of Ppp1cc2:

In order to test the efficacy of the promoter to direct testis predominant expression of Ppp1cc2 we developed transgenic mice, in which PPP1CC2 expression was driven by the

2.6 kilobases of genomic region upstream of transcription start site. This study was part of the goal in Aim‐II to determine if PPP1CC2 expressed, as a transgene will restore spermatogenesis and fertility in Ppp1cc null mice. Our results demonstrate that the 2.6 kb upstream fragment is able to drive testis specific expression of Ppp1cc2 transgene for all the rescue line driven by the promoter (Fig 21). Western blot analysis of different tissue extracts and immuno‐histochemistry of testis sections confirmed that indeed the Ppp1cc promoter is specific in maintaining abundant expression of PPP1CC2 in testis. Besides the in‐vivo results also confirmed that the cis‐regulatory elements controlling the promoter activity in a testis specific‐manner are harbored within the 2.6 kb genomic fragment.

Analysis of the genomic region surrounding the transcription start site of the mouse Ppp1cc gene shows that the Ppp1cc promoter lacks a TATA box and is GC rich (data not shown).

137

Absence of a canonical TATA box and high GC content is a feature common to genes that are either testis specific or regulated in a testis specific manner (71, 72).

C. Identification of putative conserved factors that bind to the core promoter of the

Ppp1cc gene (+69 to ­173 bp):

To further identify regulatory elements in the promoter, we undertook phylogenetic shadowing analysis using Mulan software tool. This analysis indicated three evolutionary conserved regions across all the species, ECR‐1, 2, and 3 which had an overall homology of

70% or above (Fig. 22). Binding sites for 28 unique transcription factors were identified within these regions (Table‐11). Based on documented expression in testis and from EST database [Fig. 23 (63, 64)], their roles in driving testis specific expression of other genes and male fertility defects resulting from the targeted deletion of these products (21, 73), we suggest three of these factors may be involved in Ppp1cc regulation in testis. A single binding site for the transcription factor Myb (‐169 to ‐178 bp) was identified within the

ECR‐2 region. This site could potentially bind both Myb isoforms A (Mybl1) and B (Mybl2).

Both Myb isoforms are expressed in testis, however spatial and temporal expression of the two isoforms in testis are distinct. B‐myb largely restricted to gonocytes and spermatogonia where as A‐myb expression starts from day 10 post partum testis and progress during the first meiotic division in differentiating spermatocytes (74). Targeted deletion of A­myb gene causes male infertility due to meiotic arrest (21). The similarities in developmental expression pattern of Ppp1cc2 and A­myb, and the role of the later in meiotic progression suggest that Ppp1cc promoter might be a target for A‐Myb. The second

138 conserved binding site identified was that for the universal factor, specificity factor 1(Sp1) at positions ‐53/‐64 and ‐76/‐84 within ECR‐1. Presence of multiple Sp1 sites and their role in binding Sp1 factors and directing tissue‐specific gene expression including testis are well documented (62, 75). Our attempt to further delineate the properties of the promoter using reporter analysis showed that core promoter activity resides within +69 to ‐173 bp

(Del 4) relative to the transcription start site (Figure 5). Further deletion of this region that causes removal of the putative A‐Myb (‐169 to ‐178 bp) and the Sp1 binding sites (‐53/‐64 and ‐76/‐84) leads to a 75% decrease in transcription activity, suggesting the possible involvement of these factors in core promoter regulation. The third factor that was conserved across species identified in the ECR‐1 is the basic helix loop helix factor, Spz1

(+272/+257), located immediately upstream of the ATG initiation codon within the 5’UTR.

Expression of Spz1 is predominant in testis and also known to interact in an isoform specific manner with PPP1CC2 in a yeast‐two hybrid system (20). Taken together our observations suggests a transcriptional module in which testis specific factor A‐Myb in association with the universal factor Sp1 initiates high rates of transcription of the Ppp1cc during the onset of meiotic division which occurs around day 15 in post partum testis. It is possible that high levels of expression at late meiotic and post meiotic stages may involve the participation of Spz1 which is expressed in post meiotic developing germ cells. Further studies are required to understand the exact nature of regulation by these factors. Our attempt to further characterize the binding of these factors to the promoter region using chromatin immunoprecipitation (ChIP) is limited due to unavailability of suitable cell culture system for developing germ cells and specific antibody against Spz1. It may also be possible that the splicing mechanism that generates Ppp1cc2 mRNA could be involved in

139 increased transcription of Ppp1cc in testis. It may be noted that an attempt to identify homology in the upstream genomic region (3.0 kb upstream from ATG start site) for

Ppp1ca was not successful due to the lack of significant sequence similarities across the species. The presence of conserved putative transcription regulatory sites in the Ppp1cc but not in Ppp1ca suggests a conserved mechanism. Most likely the function of these mechanism would be to ensure utilization of unique testis specific transcription factor to act on pre‐existing cis‐elements in the Ppp1cc promoter so that it can be regulated to express both ubiquitously and also in a testis specific manner during the onset of spermatogenesis.

D. Significant expression levels of transgenic PPP1CC2 in testis and sperm are

required to overcome the male infertility phenotype of Ppp1cc null mice:

Results from Aim‐I clearly demonstrate that testis expression of PPP1CC1, PPP1CA, and PPP1CB is confined primarily to Sertoli cells, spermatogonia and pre‐leptotene spermatocytes. In contrast, PPP1CC2 expression is restricted principally to meiotic and post‐meiotic stages of developing male germ cells. Thus, infertility of Ppp1cc null males resulting in the disruption of the spermatogenic process, particularly during the spermiogenic phase, has suggested critical roles for PPP1CC2 in the morphological differentiation and spermiation of male germ cells (39, 49). To determine these roles (and a possible function for PPP1CC1 in spermiation) we employed expression of transgenic mouse Ppp1cc2 cDNA, driven either by its endogenous promoter or by the testis specific hPgk2 promoter, in the Ppp1cc null background. Indeed, our results demonstrate spermiogenic roles for PPP1CC2 that are not only isoform specific, but critical for male

140 fertility, while the PPP1CC1 isoform appears to be non‐essential. Furthermore our studies show that, irrespective of the promoter used for transgenic expression, complete restoration of spermatogenesis and fertility occurs provided that the PPP1CC2 expression level is greater than 50% of the level of PPP1CC2 expression in the testis of a Ppp1cc +/­ control male.

Data from sub‐aim‐I of Aim‐II demonstrate, for the first time, a relationship between testis levels of PPP1CC2 expression and the morphological differentiation of spermatids, spermiation, and sperm motility. Release of sperm into the lumen of seminiferous tubules, as evidenced by the appearance of mature spermatozoa in the cauda epididymis occurred in all lines regardless of PPP1CC2 expression levels (Fig. 32), while high numbers of sperm released, normal sperm morphology, progressive sperm motility, and fertility were dependent on relatively high PPP1CC2 expression in the testis (see Table‐15). At lower levels of PPP1CC2 expression (below 50% of positive control levels), males were mildly oligozoospermic, and a significant proportion of sperm released (~53‐77%) were teratozoospermic. Spermatozoal populations from these same males also exhibited asthenozoospermia, as their progressive motility was extremely poor. However, in general, the percentages of motile sperm and progressively motile sperm both increased with increasing levels of PPP1CC2 expression in the testis. Thus, it appears that a threshold level of PPP1CC2 in the testis is required for normal sperm morphogenesis and spermiation, and that the compendium of morphological aberrations (Fig. 33) in low PPP1CC2 expressers has a direct bearing on both the number of motile spermatozoa and the quality of their motility.

141

Complete fertility was restored in transgenic lines where PPP1CC2 expression levels were equal to or greater than ~75% of levels in the testis of positive control mice. It is significant that the transgenic line eTg‐F1 was sub‐fertile where testis expression of

PPP1CC2 is about 50% of positive control expression (see Table‐16). We predict that rendering the transgene homozygous to increase PPP1CC2 expression in this line will restore full fertility to the line. These studies are currently in progress.

The requirement for an adequate level of PPP1CC2 may reflect stoichiometric interactions of the enzyme with its regulators and/or its protein substrates in testis and spermatozoa. It is possible that these interactions target PPP1CC2 to specific structures in developing and mature spermatozoa. A requirement for stoichiometric amounts of the testis specific isoform of the PKA catalytic subunit Cs (Cα2) has also been suggested (76).

The somatic form of the PKA catalytic subunit, Cα1, is expressed in pre‐meiotic stages of germ cell development and in Sertoli cells, whereas Cs is expressed in meiotic and post meiotic developing germ cells, a situation reminiscent of the expression of the somatically ubiquitous isoform (PPP1CC1) and male germ cell restricted isoform (PPP1CC2) of Ppp1cc

(76). In addition, utilization of an alternate transcription start site providing testis specific

Cs with a unique amino terminus replacing the amino terminus of the somatic form (77) is akin to the alternative splicing event that provides PPP1CC2 with a unique, mammal specific carboxy terminus.

Morphologically defective sperm and male subfertility characterize RIα (a PKA regulatory subunit also known as PRKAR1A) +/‐ mice, possibly due to increased unregulated PKA catalytic activity. Significantly men with mutations in the PRKAR1A gene

142 have reduced fertility due to defects in sperm morphology and azoo‐ or oligospermia (78).

While Ppp1cc +/­ mice are not similarly affected, a further reduction in PPP1CC2 activity in the testis probably leads to an increased steady state level of protein phosphorylation, a situation analogous to that which occurs in the testis of haploinsufficient RIα +/‐ mice.

Significantly, sperm structural abnormalities seen in RIα +/‐ mice, bear striking similarities to those of defective spermatozoa from rescue mice expressing reduced levels of PPP1CC2.

Requirement for optimal levels of protein for stoichiometric interactions (binding to DNA) and proper chromatin condensation has been clearly and unequivocally demonstrated for the protamine proteins PRM1 and PRM2 (79). Haploinsufficiency of PRM1 or PRM2 results in male infertility due to disruption of proper chromatin condensation, DNA damage and nuclear formation (80).

Spermatozoa have little cytoplasm with limited ability of proteins to diffuse within the cell. Consequently each PPP1CC2 catalytic subunit has access to a limited number of substrate molecules and therefore needs to be present at a relatively high stoichiometry of enzyme to substrate ratio. It may be noted that lack of PPP1CC isoforms has been associated with gross morphological and ultrastructural abnormalities of spermatozoa, including disorganized and supernumerary of outer dense fibers, malformed mitochondrial and fibrous sheaths, and defective sperm head morphology (39). These morphological and ultra‐structural deformities were also observed in sperm from the initial Pgk2 promoter driven transgenic Ppp1cc2 rescue experiments where there was limited expression of

PPP1CC2 (51). Our data now show that these sperm morphological defects are caused by either the complete absence or presence of sub‐optimal levels of PPP1CC2 in developing germ cells and spermatozoa.

143

E. Spatio­temporal expression of PP1 regulators, I­2, I­3 and sds­22 suggests tight

regulation and stoichiometric interactions with PPP1CC2:

A large number of PP1 interacting proteins have been identified in a wide range of somatic tissues and organisms (33), of which Ppp1r7 (Sds22), Ppp1r11 (I‐3) and Ppp1r2 (I‐

2) are the only PP1 regulators identified in spermatozoa (50). It is intriguing that the regulators of the mammal and sperm‐specific isoform PPP1CC2 are ubiquitously expressed and are among the evolutionarily ancient of the PP1 regulators (33). Since PPP1CC2 has been shown to form large multimeric complexes with Sds22 and I‐3, known inhibitory regulators of PP1, in both testis and sperm (50) we wanted to determine if the spatio‐ temporal expression patterns of these regulatory proteins are similar to that of Ppp1cc2.

Northern blot data in Figure. 26, show that mRNA levels for the three regulators increases with the onset of spermatogenesis. Similar to Ppp1cc2, mRNA levels for its regulators I‐3,

Sds22 and I‐2 are predominant in germ cells with their highest levels coinciding with the appearance of post‐meiotic round and elongating spermatids. It is remarkable that testis specific isoforms for both I­2 and Sds22 are modified at their extreme C‐termini (due to intron retention or alternate splicing) and both are highly expressed at late stages of spermatogenesis (66, 68). In did we discovered that a testis specific isoform of inhibitor I­3 may exist in testis as well (data not shown). How testis specific expression regulatory mechanisms for these inhibitors, found in a wide range of tissues and species, had evolved to play a role in mammalian spermatogenesis is an intriguing question. Are testis specific regulatory mechanisms such as cis‐elements similar to those in Ppp1cc also present in the

144 promoters of the genes for these regulators of PP1? Thus it would be important to determine if there exist conserved transcriptomes that are responsible for coordinated expression of PPP1CC2 and its regulators and also shared by other meiotically and post meiotically expressed genes. Alternatively, does Ppp1cc2 participate in feedback mechanisms leading to regulation of expression of its own regulators? Further work is required to answer these questions and to determine whether the functions of these regulators during spermiogenesis and in mature sperm are distinct from their roles in somatic cells. As noted earlier Spz1 seems a likely member of such conserved module, which besides controlling Ppp1cc2 expression also likely to control expression of its regulators. As an added level of control it is possible that PPP1CC2 regulates expression of its own regulators by virtue of its interaction with Spz1.

F. Copy number of Ppp1cc2 gene in testis is linearly correlated to both its message

levels as well as protein:

We attempted to over express PPP1CC2, by expressing the pTg‐M30 transgene

(with PPP1CC2 levels almost equal to control +/‐) on the wild‐type background (Ppp1cc

+/+; Tg+). Results from quantitative western blot followed by densitometry analysis suggests that introduction of an third copy of the Ppp1cc2 gene in form mini‐transgene resulted in 20% over‐expression of the protein with concomitant 35% increase in message levels compared to wildtype (+/+), whereas ~35% increase in protein and ~70% in mRNA levels of that of heterozygous control (Fig. 35). There was no discernible phenotype observed for the +/+;Tg+ mice in terms of sperm morphology, number and motility. This suggests that PPP1CC2 levels above the wild‐type are well tolerated in the testis. It will be

145 interesting to see whether PPP1CC2 levels can be further enhanced, and if so to what extent by rendering the wildtype mice homozygous for the transgene (+/+; Tg+/Tg+). This preliminary study indicates that though there is strict lower threshold requirement for

PPP1CC2 in testis, however no such threshold exists for the upper limit. The mechanism by which this regulation is mediated is unknown, but we speculate that Spz1, might be a crucial component of the control pathway. It has been shown that Spz1 binds to PPP1CC2 in an isoform specific manner by its unique carboxy terminal tail (20). We speculate that

PPP1CC2 auto‐regulates its own expression by binding to Spz1, thereby modulating its own expression at the level of translational.

G. Ppp1cc gene in sertoli cell and pre­meiotic cell is dispensible for spermatogenesis

and sperm function provided adequate level of PPP1CC2 is maintained in germ

cells:

Previous attempts to globally delete Ppp1cc gene implicated a role for both PP1 isoforms, PPP1CC1 and PPP1CC2 in spermatogenesis and male fertility. Data from transgenic rescue animals (in Aim‐II) where PPP1CC2 expression was transgenically expressed in developing germ cells confirmed that PPP1CC2 alone can restore spermatogenesis, sperm morphology, motility and fertility. Thus far, the potential role of the Ppp1cc gene in pre‐meiotic germ cells, sertoli cells in supporting spermatogenesis and sperm function in its absence in developing germ cells in testis remains largely unexplored.

In Aim‐III of my dissertation work I performed targeted knockdown of the Ppp1cc gene in developing germ cells from differentiated spermatogonia and pre‐leptotene spermatocyte onwards. For this purpose we crossed mice homozygous for floxed allele (Ppp1ccflx/flx) with

146 transgenic mice expressing CRE driven by the Stra8­Cre promoter (TgStr8­Cre+) to develop germ cell knock out mice Ppp1ccflx/flx; TgStr8­Cre+ (GcKO). The GcKO mice showed drastic reduction for Ppp1cc2 message levels and concomitant decrease in PPP1CC2 levels. Though

PPP1CC2 levels varied slightly between different GcKO animals however their levels were consistently lower (≤12.5%) of control (Ppp1ccflx/flx). These residual but variable levels of

PPP1CC2 resulted from incomplete deletion of Ppp1cc gene due to reduced penetrance of

CRE trangene expression. This was further supported by the observation from The data from the study of the GcKO, showed for the very first time the role of PPP1CC1 the main

Ppp1cc gene product in pre‐meiotic germ cells and in sertoli cells is dispensable and alone cannot sustain spermatogenesis and sperm function. The results also confirmed once again our previous finding on the crucial role PPP1CC2 in spermatogenesis and sperm function.

Deletion of PPP1CC2 in developing germ cells resulted in oligospermia due to low sperm count (0.4 x 106 ± 0.8), and was teratozoospermic because of high incidence of sperm morphological defects (66% of deformity) and finally asthenospermic due to poor motility

(40% total motility and 10.5% of progressive motility). Consequently all males tested were infertile due to the combined effect of oligo‐terato‐asthenospermic condition as seen previously with rescue animals with low levels of PPP1CC2.

Though sperm recovered from vasa differentia of GcKO males were on an average 10% of that of control, even then a sperm count of 0.4 x 106 ± 0.8, seemed significantly higher compared to Ppp1cc null (‐/‐) mice, inspite of the fact that PPP1CC2 expression is lost from vast majority of the tubules and levels significantly reduced. In order to find answers to this apparent mismatch we went back on our analysis of the IHC data. Immnunohistochemistry showed mosaic expression of PPP1CC2 in the testis and seminiferous tubules. These

147

PPP1CC2 positive tubules though were far less in number were however variable in terms of PPP1CC2 expression. Some tubules showed almost full expression within the germinal epithelium where in others the expression was restricted to part of the tubules from periphery to the lumen. From analysis of the IHC data we speculate three possible reasons for the seemingly higher sperm count in GcKO males. (1) Firstly tubules that are almost intact for PPP1CC2 expression (Fig 49A) underwent normal spermatogenesis and sperm production as in wildtype without disruption thereby adding to the pool of normal spermatozoa and towards the total pool of sperm. (2) Both spermatogonia and spermatocytes (pre‐leptotene to spermatids) at all stages of development are connected by intercellular bridges (cytoplasmic bridges) forming syncytium due to incomplete cytokinesis (1). The later is important for synchronous growth and differentiation during spermatogenesis. The clonal siblings in the syncytium share their cytoplasm, exchange mRNA and signaling molecules through the cytoplasmic bridges (81). The CRE mediated knockdown in our system is a stochastic event rendered due to reduced penetrance of Cre expression in spermatogonia or pre‐leptotene spermatocytes. Stochasticity resulted in deletion of Ppp1cc gene in vast majority of the tubules, however randomly leaving Ppp1cc gene intact in a very small population of spermatogonia or pre‐leptotene spermatocytes.

This small population of Ppp1cc gene intact germ cell during early pachytene stage starts expressing PPP1CC2 normally (Fig. 49B,C,D). However due to presence of intercellular bridges, PPP1CC2 diffuses between the clonal subpopulation and in turn effectively dilutes and lower the PPP1CC2 levels on per cell basis within the syncytium. This in turn creates a situation analogous to rescue mice with low levels of PPP1CC2 resulting from low expression within a syncytium. This results in suboptimal levels of the PPP1CC2 within a

148 cell in the syncytium, that is enough to support spermatogenesis but inadequate for proper sperm morphogenesis, spermiation and motility development. (3) Alternatively, suboptimal levels of PPP1CC2 in Ppp1cc gene intact germ cell can also result from diffusion of PPP1CC2 via sertoli cell‐germ cell junctions to adjacent non‐germ cells (sertoli cells) and germ cells. We speculate two combinations of the above mentioned possibilities, either 1,2 or 1,2,3 that explains the higher sperm number in GcKO compared to Ppp1cc null mice and presence of significant population normal spermatozoa (34 ± 6.5%) in these animals.

Nonetheless the conditional deletion of the Ppp1cc gene in developing germ cell further reinforces our earlier observation on the correlation between PPP1CC2 levels and sperm morphology and motility development.

H. Transgenic rescue study suggests that PPP1CC1 can partially restore

spermatogenesis but not sperm morphology, motility and fertility.

In sub‐aim‐II of Aim‐I, we attempted to force express PPP1CC1 at high levels using the endogenous promoter of Ppp1cc gene in the developing germ cells. The hypothesis behind this experiment was to show whether PPP1CC1 can substitute for PPP1CC2 when expressed at sufficiently highly levels in the developing germ cells. Unfortunately we could not achieve high expression in these lines. The only rescue line for which PPP1CC1 was detectable was much lower. Consequently the sperm produced were 99% lower compared to control (+/‐) animals and bore wide range of deformities (Fig. 43) and were immotile.

The endogenous promoter driven transgenic line was third transgenic experiment attempted by our lab to over express PPP1CC1 in the developing germ cells. In the previous two transgenic experiments the hPgk2 promoter was used to drive either the coding region

149

(with a part of the intron7 at the 3’‐end) or the complete cDNA respectively. In both instances PPP1CC1 expression was low but higher than the endogenous driven line (Fig.

40). The result from these studies so far suggests that PPP1CC1 can partially restore spermatogenesis but not sperm morphogenesis, motility and male fertility. The consistent low expression of Ppp1cc1 in all these different transgenic line raised the speculation of possible instability of mRNA rendered due to the design of the transgene cassette. Closer examination revealed that the only common non‐coding region in all the three transgenic constructs is the partial or complete presence of intron‐7. Analysis of the nucleotide sequence of the intron‐7 showed the presence of multiple copies of the mRNA instability signal “AUUUA” in the region (82) (data not shown). To test the instability created by intron‐7 and to better the expression of PPP1CC1 we thus created a novel transgene construct that mimics in its design the Ppp1cc2 cDNA though it expresses Ppp1cc1 (Fig. 45).

The study for this new transgenic line has just begun and evidence for over expression of

PPP1CC1 is still awaiting. In the eventuality this strategy to force express PPP1CC1 do not work we plan to develop isoform specific knock‐out of PPP1CC1 as outlined in section 2.4.1

(Fig. 44).

I. Why only PPP1CC2 and not any other PP1 isoform?

The essential requirement for PPP1CC2 among all the PP1 isoforms is surprising given the high degree of homology in the amino acid sequences of the four isoforms

(~90%). The catalytic core of the enzyme is conserved, while the major differences between the isoforms reside in their C‐terminal regions. Indeed, PPP1CC2 has a unique

~22 amino acid carboxy terminal tail, absent from other PP1 isoforms, and it is significant

150 that the carboxy terminal tail is nearly completely conserved in its entirety among all the mammalian species for which annotated genomic databases exists. This indicates a function specific to PPP1CC2, but to date the role of this carboxy tail has not been elucidated. Efforts are currently underway in our lab to determine the function(s) of this C‐ terminus using several transgenic approaches, including transgenic expression of the C‐ terminus of PPP1CC2 fused to Enhanced GFP (see section 2.1.12).

151

Major findings and conclusion of this dissertation work are:

1. Ppp1cc gene promoter is regulated in a testis specific manner to maintain high levels

of PPP1CC2 in developing germ cells in testis.

2. In‐vivo results confirm for the first time that Ppp1cc promoter is testis specific and

the specific property lies within the 2.6 kb upstream genomic region.

3. In‐silico analysis identified putative transcription factors Spz1, A‐Myb and Sp1 with

conserved binding site in the promoter region.

4. Testis specific factors like Spz1 and A‐Myb in coordination with ubiquitous factor

Sp1 might regulate the core promoter in testis specific manner.

5. PPP1CC2 is the sole PP1 isoform capable of restoring male fertility.

6. High threshold levels of PPP1CC2 (≥50% of that of control+/‐) in testis is an

essential requisite to overcome male infertility phenotype of Ppp1cc null mice.

7. Sperm morphogenesis and motility is strongly correlated with PPP1CC2 levels.

8. Expression of sperm PPP1CC2 regulators, I‐2, I‐3 and sds‐22 closely follow testis

expression pattern of Ppp1cc2 in late meiotic and post‐meiotic germ cells.

9. We speculate that high levels of PPP1CC2 are required for maintaining

stoichiometric interactions with its regulators and substrates in testis and

spermatozoa.

10. PPP1CC1 and other PP1 isoforms in sertoli cells, pre‐meiotic germ cells and Leydig

cells are not essential for spermatogenesis and sperm function.

Clinical Significance

. A novel finding of this work is that high threshold requirement of PPP1CC2 levels or

its activity is strongly correlated to sperm morphogenesis, motility and fertility. A

low level of PPP1CC2 is associated with pathological condition that represents oligo‐

terato‐asthenozoospermia. Thus a potential clinical implication of this study is that

both protein and catalytic activity levels of PPP1CC2 could be used as a biomarker

for assessment of male fertility in a subset of male population.

. Designing of male contraceptive targeting PPP1CC2 activity. This can be in form of a

gel for use in female tract that targets and inhibits PPP1CC2 activity thereby

inhibiting sperm motility and function.

152

Bibilography

1. Knobil E & Neill JD (1994) The Physiology of reproduction (Raven Press, New York)

2nd Ed.

2. Russell LD (1990) Histological and histopathological evaluation of the testis (Cache

River Press, Clearwater, Fl) 1st Ed pp xiv, 286 p.

3. Leblond CP & Clermont Y (1952) Definition of the stages of the cycle of the

seminiferous epithelium in the rat. Ann N Y Acad Sci 55(4):548‐573.

4. Tegelenbosch RA & de Rooij DG (1993) A quantitative study of spermatogonial

multiplication and stem cell renewal in the C3H/101 F1 hybrid mouse. Mutat Res

290(2):193‐200.

5. de Rooij DG (1973) Spermatogonial stem cell renewal in the mouse. I. Normal

situation. Cell Tissue Kinet 6(3):281‐287.

6. Oakberg EF (1971) A new concept of spermatogonial stem‐cell renewal in the

mouse and its relationship to genetic effects. Mutat Res 11(1):1‐7.

7. Huckins C (1971) The spermatogonial stem cell population in adult rats. I. Their

morphology, proliferation and maturation. Anat Rec 169(3):533‐557.

8. Lok D, Weenk D, & De Rooij DG (1982) Morphology, proliferation, and

differentiation of undifferentiated spermatogonia in the Chinese hamster and the

ram. Anat Rec 203(1):83‐99.

9. Buaas FW, et al. (2004) Plzf is required in adult male germ cells for stem cell self‐

renewal. Nat Genet 36(6):647‐652.

153 154

10. Schrans‐Stassen BH, van de Kant HJ, de Rooij DG, & van Pelt AM (1999) Differential

expression of c‐kit in mouse undifferentiated and differentiating type A

spermatogonia. Endocrinology 140(12):5894‐5900.

11. Costoya JA, et al. (2004) Essential role of Plzf in maintenance of spermatogonial

stem cells. Nat Genet 36(6):653‐659.

12. Grisanti L, et al. (2009) Identification of spermatogonial stem cell subsets by

morphological analysis and prospective isolation. Stem Cells 27(12):3043‐3052.

13. Waheeb R & Hofmann MC (2011) Human spermatogonial stem cells: a possible

origin for spermatocytic seminoma. Int J Androl 34(4 Pt 2):e296‐305; discussion

e305.

14. Grootegoed JA, Siep M, & Baarends WM (2000) Molecular and cellular mechanisms

in spermatogenesis. Baillieres Best Pract Res Clin Endocrinol Metab 14(3):331‐343.

15. Eddy EM (2002) Male germ cell gene expression. Recent Prog Horm Res 57:103‐128.

16. Kistler MK, Sassone‐Corsi P, & Kistler WS (1994) Identification of a functional cyclic

adenosine 3',5'‐monophosphate response element in the 5'‐flanking region of the

gene for transition protein 1 (TP1), a basic chromosomal protein of mammalian

spermatids. Biol Reprod 51(6):1322‐1329.

17. Zhou Y, Overbeek PA, & Bernstein KE (1996) Tissue specific expression of testis

angiotensin converting enzyme is not determined by the ‐32 nonconsensus TATA

motif. Biochem Biophys Res Commun 223(1):48‐53.

18. Macho B, et al. (2002) CREM‐dependent transcription in male germ cells controlled

by a kinesin. Science 298(5602):2388‐2390.

155

19. Monaco L, et al. (2004) Specialized rules of gene transcription in male germ cells:

the CREM paradigm. Int J Androl 27(6):322‐327.

20. Hrabchak C & Varmuza S (2004) Identification of the spermatogenic zip protein

Spz1 as a putative protein phosphatase‐1 (PP1) regulatory protein that specifically

binds the PP1cgamma2 splice variant in mouse testis. J Biol Chem 279(35):37079‐

37086.

21. Toscani A, et al. (1997) Arrest of spermatogenesis and defective breast development

in mice lacking A‐myb. Nature 386(6626):713‐717.

22. Ro S, Park C, Sanders KM, McCarrey JR, & Yan W (2007) Cloning and expression

profiling of testis‐expressed microRNAs. Dev Biol 311(2):592‐602.

23. Girard A, Sachidanandam R, Hannon GJ, & Carmell MA (2006) A germline‐specific

class of small RNAs binds mammalian Piwi proteins. Nature 442(7099):199‐202.

24. Kuramochi‐Miyagawa S, et al. (2004) Mili, a mammalian member of piwi family

gene, is essential for spermatogenesis. Development 131(4):839‐849.

25. Kleene KC, Distel RJ, & Hecht NB (1984) Translational regulation and deadenylation

of a protamine mRNA during spermiogenesis in the mouse. Dev Biol 105(1):71‐79.

26. Xu M & Hecht NB (2007) Polypyrimidine tract binding protein 2 stabilizes

phosphoglycerate kinase 2 mRNA in murine male germ cells by binding to its 3'UTR.

Biol Reprod 76(6):1025‐1033.

27. Yang J, Medvedev S, Reddi PP, Schultz RM, & Hecht NB (2005) The DNA/RNA‐

binding protein MSY2 marks specific transcripts for cytoplasmic storage in mouse

male germ cells. Proc Natl Acad Sci U S A 102(5):1513‐1518.

156

28. Fawcett DW (1981) The cell (W. B. Saunders Co., Philadelphia) 2d ed Ed pp viii, 862

p.

29. Darszon A, Nishigaki T, Beltran C, & Trevino CL (2011) Calcium channels in the

development, maturation, and function of spermatozoa. Physiol Rev 91(4):1305‐

1355.

30. Vijayaraghavan S, Critchlow LM, & Hoskins DD (1985) Evidence for a role for

cellular alkalinization in the cyclic adenosine 3',5'‐monophosphate‐mediated

initiation of motility in bovine caput spermatozoa. Biol Reprod 32(3):489‐500.

31. Vijayaraghavan S & Hoskins DD (1986) Regulation of bovine sperm motility and

cyclic adenosine 3',5'‐monophosphate by adenosine and its analogues. Biol Reprod

34(3):468‐477.

32. Hoskins DD, Acott TS, Critchlow L, & Vijayaraghavan S (1983) Studies on the roles of

cyclic AMP and calcium in the development of bovine sperm motility. J Submicrosc

Cytol 15(1):21‐27.

33. Cohen PT (2002) Protein phosphatase 1‐‐targeted in many directions. J Cell Sci

115(Pt 2):241‐256.

34. Garcia A, et al. (2003) Serine/threonine protein phosphatases PP1 and PP2A are key

players in apoptosis. Biochimie 85(8):721‐726.

35. Okano K, Heng H, Trevisanato S, Tyers M, & Varmuza S (1997) Genomic organization

and functional analysis of the murine protein phosphatase 1c gamma (Ppp1cc) gene.

Genomics 45(1):211‐215.

36. da Cruz e Silva EF, et al. (1995) Differential expression of protein phosphatase 1

isoforms in mammalian brain. J Neurosci 15(5 Pt 1):3375‐3389.

157

37. Kitagawa Y, et al. (1990) Protein phosphatases possibly involved in rat

spermatogenesis. Biochem Biophys Res Commun 171(1):230‐235.

38. Shima H, et al. (1993) Protein phosphatase 1 gamma 2 is associated with nuclei of

meiotic cells in rat testis. Biochem Biophys Res Commun 194(2):930‐937.

39. Chakrabarti R, et al. (2007) Analysis of Ppp1cc‐null mice suggests a role for

PP1gamma2 in sperm morphogenesis. Biol Reprod 76(6):992‐1001.

40. Lin Q, Buckler ESt, Muse SV, & Walker JC (1999) Molecular evolution of type 1

serine/threonine protein phosphatases. Mol Phylogenet Evol 12(1):57‐66.

41. Sasaki K, et al. (1990) Identification of members of the protein phosphatase 1 gene

family in the rat and enhanced expression of protein phosphatase 1 alpha gene in

rat hepatocellular carcinomas. Jpn J Cancer Res 81(12):1272‐1280.

42. Gibbons JA, Kozubowski L, Tatchell K, & Shenolikar S (2007) Expression of human

protein phosphatase‐1 in Saccharomyces cerevisiae highlights the role of

phosphatase isoforms in regulating eukaryotic functions. J Biol Chem

282(30):21838‐21847.

43. Bollen M, Peti W, Ragusa MJ, & Beullens M (2010) The extended PP1 toolkit:

designed to create specificity. Trends Biochem Sci 35(8):450‐458.

44. Barker HM, Craig SP, Spurr NK, & Cohen PT (1993) Sequence of human protein

serine/threonine phosphatase 1 gamma and localization of the gene (PPP1CC)

encoding it to chromosome bands 12q24.1‐q24.2. Biochim Biophys Acta

1178(2):228‐233.

158

45. Chakrabarti R, Cheng L, Puri P, Soler D, & Vijayaraghavan S (2007) Protein

phosphatase PP1 gamma 2 in sperm morphogenesis and epididymal initiation of

sperm motility. Asian J Androl 9(4):445‐452.

46. Vijayaraghavan S, et al. (1996) Sperm motility development in the epididymis is

associated with decreased glycogen synthase kinase‐3 and protein phosphatase 1

activity. Biol Reprod 54(3):709‐718.

47. Smith GD, et al. (1996) Primate sperm contain protein phosphatase 1, a biochemical

mediator of motility. Biol Reprod 54(3):719‐727.

48. Smith GD, Wolf DP, Trautman KC, & Vijayaraghavan S (1999) Motility potential of

macaque epididymal sperm: the role of protein phosphatase and glycogen synthase

kinase‐3 activities. J Androl 20(1):47‐53.

49. Varmuza S, et al. (1999) Spermiogenesis is impaired in mice bearing a targeted

mutation in the protein phosphatase 1cgamma gene. Dev Biol 205(1):98‐110.

50. Cheng L, Pilder S, Nairn AC, Ramdas S, & Vijayaraghavan S (2009) PP1gamma2 and

PPP1R11 are parts of a multimeric complex in developing testicular germ cells in

which their steady state levels are reciprocally related. PLoS One 4(3):e4861.

51. Soler DC, et al. (2009) Expression of transgenic PPP1CC2 in the testis of Ppp1cc‐null

mice rescues spermatid viability and spermiation but does not restore normal

sperm tail ultrastructure, sperm motility, or fertility. Biol Reprod 81(2):343‐352.

52. Hofmann MC, Braydich‐Stolle L, Dettin L, Johnson E, & Dym M (2005)

Immortalization of mouse germ line stem cells. Stem Cells 23(2):200‐210.

159

53. He Z, et al. (2008) Gdnf upregulates c‐Fos transcription via the Ras/Erk1/2 pathway

to promote mouse spermatogonial stem cell proliferation. Stem Cells 26(1):266‐

278.

54. Goodson SG, Zhang Z, Tsuruta JK, Wang W, & O'Brien DA (2011) Classification of

mouse sperm motility patterns using an automated multiclass support vector

machines model. Biol Reprod 84(6):1207‐1215.

55. Strack S, Kini S, Ebner FF, Wadzinski BE, & Colbran RJ (1999) Differential cellular

and subcellular localization of protein phosphatase 1 isoforms in brain. J Comp

Neurol 413(3):373‐384.

56. Ovcharenko I, et al. (2005) Mulan: multiple‐sequence local alignment and

visualization for studying function and evolution. Genome Res 15(1):184‐194.

57. Loots GG & Ovcharenko I (2005) Dcode.org anthology of comparative genomic tools.

Nucleic Acids Res 33(Web Server issue):W56‐64.

58. Loots GG & Ovcharenko I (2007) Mulan: multiple‐sequence alignment to predict

functional elements in genomic sequences. Methods Mol Biol 395:237‐254.

59. Horvath GC, Kistler MK, & Kistler WS (2009) RFX2 is a candidate downstream

amplifier of A‐MYB regulation in mouse spermatogenesis. BMC Dev Biol 9:63.

60. Widlak W, et al. (2007) The GC‐box is critical for high level expression of the testis‐

specific Hsp70.2/Hst70 gene. Acta Biochim Pol 54(1):107‐112.

61. Tang H & Goldberg E (2009) Homo sapiens lactate dehydrogenase c (Ldhc) gene

expression in cancer cells is regulated by transcription factor Sp1, CREB, and CpG

island methylation. J Androl 30(2):157‐167.

160

62. Maeda T, Hirayama M, Kobayashi D, & Tamai I (2005) Regulation of testis‐specific

carnitine transporter (octn3) gene by proximal cis‐acting elements Sp1 in mice.

Biochem Pharmacol 70(6):858‐868.

63. Hsu SH, Shyu HW, Hsieh‐Li HM, & Li H (2001) Spz1, a novel bHLH‐Zip protein, is

specifically expressed in testis. Mech Dev 100(2):177‐187.

64. Mettus RV, et al. (1994) Murine A‐myb: evidence for differential splicing and tissue‐

specific expression. Oncogene 9(10):3077‐3086.

65. Huang Z, Khatra B, Bollen M, Carr DW, & Vijayaraghavan S (2002) Sperm

PP1gamma2 is regulated by a homologue of the yeast protein phosphatase binding

protein sds22. Biol Reprod 67(6):1936‐1942.

66. Chun YS, et al. (2000) A sds22 homolog that is associated with the testis‐specific

serine/threonine protein phosphatase 1gamma2 in rat testis. Biochem Biophys Res

Commun 273(3):972‐976.

67. Park IK, Roach P, Bondor J, Fox SP, & DePaoli‐Roach AA (1994) Molecular

mechanism of the synergistic phosphorylation of phosphatase inhibitor‐2. Cloning,

expression, and site‐directed mutagenesis of inhibitor‐2. J Biol Chem 269(2):944‐

954.

68. Osawa Y, Nakagama H, Shima H, Sugimura T, & Nagao M (1996) Identification and

characterization of three isotypes of protein phosphatase inhibitor‐2 and their

expression profiles during testis maturation in rats. Eur J Biochem 242(3):793‐798.

69. Robinson MO, McCarrey JR, & Simon MI (1989) Transcriptional regulatory regions

of testis‐specific PGK2 defined in transgenic mice. Proc Natl Acad Sci U S A

86(21):8437‐8441.

161

70. Zhang LP, Stroud J, Eddy CA, Walter CA, & McCarrey JR (1999) Multiple elements

influence transcriptional regulation from the human testis‐specific PGK2 promoter

in transgenic mice. Biol Reprod 60(6):1329‐1337.

71. Somboonthum P, et al. (2005) cAMP‐responsive element in TATA‐less core

promoter is essential for haploid‐specific gene expression in mouse testis. Nucleic

Acids Res 33(10):3401‐3411.

72. Horowitz E, et al. (2005) Patterns of expression of sperm flagellar genes: early

expression of genes encoding axonemal proteins during the spermatogenic cycle

and shared features of promoters of genes encoding central apparatus proteins. Mol

Hum Reprod 11(4):307‐317.

73. Hsu SH, Hsieh‐Li HM, & Li H (2004) Dysfunctional spermatogenesis in transgenic

mice overexpressing bHLH‐Zip transcription factor, Spz1. Exp Cell Res 294(1):185‐

198.

74. Latham KE, et al. (1996) Temporal patterns of A‐myb and B‐myb gene expression

during testis development. Oncogene 13(6):1161‐1168.

75. Shell SA, Fix C, Olejniczak D, Gram‐Humphrey N, & Walker WH (2002) Regulation of

cyclic adenosine 3',5'‐monophosphate response element binding protein (CREB)

expression by Sp1 in the mammalian testis. Biol Reprod 66(3):659‐666.

76. San Agustin JT & Witman GB (2001) Differential expression of the C(s) and Calpha1

isoforms of the catalytic subunit of cyclic 3',5'‐adenosine monophosphate‐

dependent protein kinase testicular cells. Biol Reprod 65(1):151‐164.

162

77. Agustin JT, Wilkerson CG, & Witman GB (2000) The unique catalytic subunit of

sperm cAMP‐dependent protein kinase is the product of an alternative Calpha

mRNA expressed specifically in spermatogenic cells. Mol Biol Cell 11(9):3031‐3044.

78. Burton KA, et al. (2006) Haploinsufficiency at the protein kinase A RI alpha gene

locus leads to fertility defects in male mice and men. Mol Endocrinol 20(10):2504‐

2513.

79. Cho C, et al. (2001) Haploinsufficiency of protamine‐1 or ‐2 causes infertility in mice.

Nat Genet 28(1):82‐86.

80. Cho C, et al. (2003) Protamine 2 deficiency leads to sperm DNA damage and embryo

death in mice. Biol Reprod 69(1):211‐217.

81. Braun RE, Behringer RR, Peschon JJ, Brinster RL, & Palmiter RD (1989) Genetically

haploid spermatids are phenotypically diploid. Nature 337(6205):373‐376.

82. Akashi M, et al. (1994) Number and location of AUUUA motifs: role in regulating

transiently expressed RNAs. Blood 83(11):3182‐3187.