1 THE ORIGIN AND ROLE OF MEIOSIS SPECIFIC ADENINE NUCLEOTIDE TRANSLOCASE 4 By CHAE HO LIM A DISSERTATION PRESENTED TO THE GRADUATE SCHOOL OF THE UNIVERSITY OF FLORIDA IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEG REE OF DOCTOR OF PHILOSOPHY UNIVERSITY OF FLORIDA 2013
2 2013 Chae Ho Lim
3 To my family for their love and support
4 ACKNOWLEDGMENTS I would like to first thank Dr. Terada and our lab members. Dr. Terada allowed me to join the lab and always encourages me to think broadly and deeply and create new ideas. He is always open to discuss my projects as well as any scientific subjects. Aside from laboratory life, he and his wife, Yoshie Takamatsu Terada, kindly invite me to join them for holidays and sometimes even for regular dinners like family. For this, I sincerely appreciate Yoshie too. Charlie, our previous lab member, always made time to discuss my projects with me and helped me with experimental design a nd problems even though he was busy with his own experiments. In addition, he taught me molecular cloning techniques needed for the generation of transgenic mice. Katherine, a research coordinator, always supported me in ordering materials and technical ad vice for my experiments. In addition, she regularly edits and revises my English writing without any complaint. Milena, a graduate student, has been a good conversation partner with me. She is erudite on topics of history and talking with her is both infor mative and entertaining. Wai Yee, a former undergraduate student, made time to read a book and talk with me to improve my English speaking skills. Lastly, I would like to thank other lab members including Joon and undergraduate volunteers, Michala, Crysta, and Anastasiya for their contribution to the lab. I would like to thank my committee members. Dr. Sugrue encouraged me to think from different points of view. Dr. Resnick taught me BAC DNA techniques and allows me to use his materials and equipments free ly. Dr. Oh discussed my projects and suggested solutions all the time. Dr. Braun supports me with genomic analysis and helped to make our PLoS One have finished these projects.
5 I would like to thank Rya n for pronuclear injection of transgene constructs and Marda for assisting me with histology and immunostaining. I would like to thank Korean and discussions with my e xperiments. I sincerely appreciate Drs. Son and Park and Dong Sun. Dr. Son allowed me to join her lab to learn biological experiments although I had no prior knowledge and experience. She also allowed me to work with Dr. Park. Dr. Park is an incredible re searcher and technician and taught me everything about cell biology from the bottom up. She is so strict but accurate so I was able to learn how to do experiments properly. Dong Sun was my savior whenever I had problems with my experiments. He never told m e what I should do to solve the problems but encourages me to think of a solution. Most of all, Dr. Park and Dong sun made me realize that biological research is so Lastly, I would like to thank my parents and two sisters, Chae Hee and Young A. They always love, support, and believe me without condition. Also, I thank their husbands, Sang Joon and Joon Young. They are always good to my parents like sons in my absence. I thank my niece and two neph ew, Se Bin, Jong Hyun, and Se Hyun for making us all happy. In addition, I would like to thank my best friends, Sung Hee, Hyun Sook, Chul Woo, Ki Jung, and Jong Hwa for their long and true friendships with me.
6 TABLE OF CONTENTS page ACKNOWLEDGMENTS ................................ ................................ ................................ .. 4 LIST OF TABLES ................................ ................................ ................................ ............ 8 LIST OF FIGURES ................................ ................................ ................................ .......... 9 LIST OF ABBREVIATIONS ................................ ................................ ........................... 10 ABSTRACT ................................ ................................ ................................ ................... 12 CHAPTER 1 INTRODUCTION ................................ ................................ ................................ .... 14 Adenine Nucleotide Translocase ................................ ................................ ............ 14 Functions of Ant ................................ ................................ ................................ ...... 14 ATP/ADP Antiporter ................................ ................................ ......................... 14 Mitochondrial Membrane Transition Pore Complex ................................ .......... 16 Uncoupler ................................ ................................ ................................ ......... 17 Adenine Nucleotide Translocase 4 ................................ ................................ ......... 18 2 E VOLUTIONARY GENOMICS IMPLIES A SP ECIFIC FUNCTION OF ANT4 IN MAMMALIAN AND ANOLE LIZARD MALE GAMETES ................................ .......... 21 Background ................................ ................................ ................................ ............. 21 Materials and Methods ................................ ................................ ............................ 24 Database Mining, Sequence Alignment and Phylogenetic Analysis ................. 24 Chromosome Synteny ................................ ................................ ...................... 25 Animals and Sample Preparation ................................ ................................ ..... 26 RNA Expression (qRT PCR) ................................ ................................ ............ 26 Gene Dosage Analysis (qPCR of G enomic DNA) ................................ ............ 27 CpG Methylation Analysis (Bisulfite Sequencing) ................................ ............. 28 Results ................................ ................................ ................................ .................... 28 Anole Lizard H as t he Ant4 Ortholog ................................ ................................ 28 Anole Ant4 I s E xpressed in Testis ................................ ................................ .... 30 Chicken H as a D egenerate DNA Fragment t hat C orresponds t o a L ikely Ant4 Pseudogene ................................ ................................ .......................... 30 Anole Ant2 I s N ot on a Heterogametic Sex Chromosome ................................ 33 D iscussion ................................ ................................ ................................ .............. 34 3 TRANSGENE ANT4 EXPRESSION DOES NOT RESTORE THE IMPAIRED SPERMATOGENESIS IN ANT4 NULL TESTIS ................................ ..................... 48 Background ................................ ................................ ................................ ............. 48
7 Materials and Methods ................................ ................................ ............................ 49 Mouse H usbandry ................................ ................................ ............................ 49 Generation of BAC Transgenic M ice ................................ ................................ 49 Transgenic c onstructs and t argeting of BAC DNA ................................ ..... 49 Purification of the targeted BACs ................................ ............................... 51 Pronuclear i njection ................................ ................................ ................... 53 Histology and Immunohistochemistry ................................ ............................... 53 GFP Expression ................................ ................................ ............................... 54 Galactosidase Staining ................................ ................................ ................. 55 RNA Isolation ................................ ................................ ................................ ... 55 qRT PCR ................................ ................................ ................................ .......... 55 RT PCR ................................ ................................ ................................ ............ 56 Western Blotting ................................ ................................ ............................... 56 Results ................................ ................................ ................................ .................... 56 BAC Transgenic Ant4 Promoter Ac hieves Comparable Gene Expression Specificity t o Endogenous Ant4 i n Mice. ................................ ....................... 56 BAC Transgenic Ant4 Promoter Achieves Comparable Gene Expression Levels to Endogenous Ant4 in Mice ................................ .............................. 59 HA t agged Transgene Ant4 Protein Shows More Precise Expression Pattern Of BAC Transgenic Ant4 Promoter ................................ ................... 60 Transgenic Ant4 Localizes in Mitoch ondria of Male Germ Cells. ...................... 60 BAC Ant4 Tg Mice Do Not Restore t he Impaired Spermatogenesis When Crossed w ith Ant4 / Mice ................................ ................................ ............. 61 Dis cussion ................................ ................................ ................................ .............. 62 4 ANT4 IS EXPRESSED IN MEIOTIC FETAL OVARY ................................ ............. 77 Background ................................ ................................ ................................ ............. 77 Materials a nd Methods ................................ ................................ ............................ 78 Results ................................ ................................ ................................ .................... 79 Ant4 and Ant2 are Expressed in Fetal Ovary. ................................ .................. 79 Female Ant4 / Mice are Fertile with Normal Ovary Morphology. ..................... 79 Discussion ................................ ................................ ................................ .............. 80 5 CONCLUSIONS AND DISCU SSION ................................ ................................ ...... 87 LIST OF REFERENCES ................................ ................................ ............................... 93 BIOGRAPHICAL SKETCH ................................ ................................ .......................... 103
8 LIST OF TABLES Table page 5 1 Ant4 and CatSper genes share an identical ortholog conservation pattern in vertebrates ................................ ................................ ................................ .......... 92
9 LIST OF FIGURES Figure page 2 1 Anole lizard has orthologs of the Ant1, 2, 3 and 4 genes. ................................ 42 2 2 The maximum likelihood (ML) estimate of Ant phylogeny supports an ancient origin of Ant4 ................................ ................................ ................................ ..... 43 2 3 Anole Ant4 is specifically expressed in testis. ................................ ..................... 44 2 4 Degenerate DNA fragment of putative Ant4 gene in the synteni c region of chicken genome. ................................ ................................ ................................ 45 2 5 Gene dosage analysis in female and male. ................................ ........................ 46 2 6 CpG methylation of the anole and mouse Ant2 g enes.. ................................ ..... 47 3 1 Generation of BAC Ant4 IRES GFP transgenic mice. ................................ ........ 69 3 2 Testis specific GFP expression regulated by the Ant4 promo ter. ....................... 70 3 3 Meiosis specific GFP expression in spermatogenic cells ................................ .. 71 3 4 Timely expression of a transgene in postnatal testes ................................ ........ 72 3 5 Transgene expression level in testes. ................................ ................................ 73 3 6 HA tagged transgene expression by the Ant4 promoter ................................ .... 74 3 7 Transgenic Ant4 localizes in mitochondria during spermatogenesis. ................. 75 3 8 Tg Ant4 expression does not restore impaired spermatogenesis in Ant4 n ull testis. ................................ ................................ ................................ .................. 76 4 1 Ant4 is expressed in fetal ovaries. ................................ ................................ ...... 83 4 2 Ant4 / ovary is morphologically normal. ................................ ............................. 84 4 3 Ant4 / ovary is histologically normal. ................................ ................................ 85 4 4 The number of offsprings from Ant4+/ and Ant4 / females. .............................. 86
10 LIST OF ABBREVIATIONS A AC ATP/AD P carrier A DP Adenosine diphosphate A NT Adenine nucleotide translocase A TP Adenosine triphosphate B AC Bacterial artificial chromosome C ATR C arboxyatractyloside C Y D C yclophilin D D API 4',6 diamidino 2 phenylindole E MS E thyl metha nesulfonate F CCP C arbonyl cyanide p (trifluoromethoxy)phenylhydrazone G FP G reen fluorescent protein G LUT Glucose transporter I M I nner mitochondrial membrane I RES Inter nal ribosome entry site K o Knockout L DH Lactate dehydrogenase M CT M onocarboxylate transpo rter M PTP Mitochondrial permeability transition pore M SCI Meiotic sex chromosome inactivation M YA M illion years ago O M Outer mitochondrial membrane P A Polyadenylation P IWI P element induced wimpy testis P T Permeability transition
11 P TPC P ermeability transiti on pore complex R OS R eactive oxygen species T CA T ricarboxylic acid T G T ransgeni c 5 U TR 5 Untranslated region V DAC V oltage dependent anion channel
12 Abstract of Dissertation Presented to the Graduate School of the University of Florida in P artial Fulfillment of the Requirements for the Degree of Doctor of Philosophy THE ORIGIN AND ROLE OF MEIOSIS SPECIFIC ADENINE NUCLEOTIDE TRANSLOCASE 4 By Chae Ho Lim August 2013 Chair: Naohiro Terada Major: Medical Sciences Molecular Cell Biology Ad enine nucleotide translocase (Ant) is an inner mitochondrial membrane protein that catalyzes the exchange between ATP and ADP. Among four isoforms in humans, which are expressed in a tissue dependent manner, Ant4 is conserved in mammals and exclusively exp ressed in testicular germ cells. The testicular specific expression of Ant4 has been suggested to have a specialized function in mammalian male germ cells. On the other hand, localization of the Ant2 gene on the X chromosome in mammals implies that Ant4 ma y act as a compensat ory Ant for the loss of Ant2 gene ex pression during male meiosis due to meiotic sex chromosome inactivation (MSCI). In the present study, I demonstrate that the orthologous Ant4 gene was found in the reptilian anole lizard, which has a heterogametic sex determination system like mammals. Interestingly, gene dosage analysis with genomic DNA and bisulfite sequencing patterns indicated that the anole Ant2 was not on the X chromosome suggesting that the gene is free from MSCI Nonetheless, A nt4 was still selectively expressed in anole testis. These data impl y that Ant4 may not be conserved simply for compensating for the loss of Ant2 gene expression during male meiosis. Ant4 expression may be advantageous in spermatogenesis and/or for sperm f unction in certain species To further investigate the
13 hypothesis that Ant4 has a specializ ed function for male germ cells, I utilized Ant4 null mice that have defects in spermatogenesis with increased apoptosis and infertil ity Using a genetic approach wi th BAC trangenes Ant2 or Ant4 wa s forcefully expressed in Ant4 null testes that were then examine d to see if Ant2 was able to restore the impaired spermatogenesis. Unexpectedly, the control Ant4 transgene expression did not rescue the Ant4 null testis phe notype even though the transgene was expressed comparably to the endogenous Ant4 gene. Here I fully discuss possibilities why the designed experiments did not work. In order to obtain further insight regarding the function of Ant4 in gametogenesis, I also closely examined the phenotype of Ant4 null female mice. Ant4 null female mice were fertile with normal morphology, size, and weight of ovaries comparable to those of wild type although Ant4 was expressed during female meiosis in the embryonic ovary. This suggest ed that Ant 4 is dispensable during female meiosis when Ant2 is expressed. Lastly, I will demonstrate a gene conservation profile of Ant4 orthologs in vertebrates through updated gene databases. There I highlight an interesting similarity between or tholog conservation profiles of Ant4 and CatSper the latter of which is known to be essential for sperm hyperactivation. These data provide us with new insights regarding specific and redundant function of Ant4 during mammalian gametogenesis
14 CHAPTE R 1 INTRODUCTION Adenine Nucleotide Translocase Adenine nucleotide translocase (Ant), also called ADP/ATP carrier (Aac) is a protein embedded in the inner mitochondrial membrane that exchanges cytoplasmic ADP for mitochondrial ATP in eukaryotes. Thus, Ant is essential for energy metabolism because most of the energy utilized by eukaryotic cells is produced by hydrolysis of A T P into A D P. This protein functions as an antiporter to allow one molecule of ADP to enter only if one molecule of ATP exits. Ant membe rs are nuclear encoded proteins of 30 35 kDa and are the most abundant proteins in the inner mitochondrial membrane  Ant consists of six transmembrane domains and has a conserved sequence motif, RRRMMM that is essential for ADP/ATP exchange activity  To date, four isoforms have been identified in mammals and are expressed in a tissue dependent manner. Ant1 is specifically expressed in heart and skeletal muscle and Ant2 is ubiquitously expressed at a low level, but is induced in proliferating cells such as lymphocytes and cancer cells Ant3 is ubiquitously and constitutively expressed whereas Ant4 is specifically expressed in testis [3 5] Rodent s including mouse ha ve Ant1, 2 and 4 but not Ant3 Functions o f Ant ATP / ADP Antiporter Collectively Ant isoforms exclusively exchange cytoplasmic ADP for mitochondrial ATP. However, it is not cl e ar whether individual Ant s ha ve similar biochemical properties. De Marcos Lousa et al examined this question by expressing thre e human ANT isoforms, ANT1, ANT2 and ANT3 in a yeast strain whose Aac
15 isoforms were disrupted  In the ir system, human ANT1, 2 or 3 were inserted between the yeast Aac2 promoter and terminator. Human ANT3 most efficiently restored the growth of Aac tr iple mutant yeast on medium including nonfermentable carbon sources such as glycerol or lactate while human ANT1 or 2 expressing yeast were less efficient. In addition, they measured the kinetics of ATP/ADP exchange (K m and V max ) in isolated mitochondria f rom human ANT expressing yeast. C onsistent with the observation for growth, K m (8.4uM) and V max (80.5nmol ADP/min/mg protein) were highest in human ANT3 expressing yeast mitochondria whereas K m (2.5 3.7uM) and V max (30 40nmol ADP/min/mg protein) were lower in human ANT1 or ANT2 expressing mitochondria. The kinetics of Ant4 the most recently identified Ant, was examined by Dolce et al  They measured K m and V max in liposomes reconstituted with human ANT4. K m and V max were 72 8uM and 1300 200nmol ADP/min/mg protein, respectively. Howe ver, since the two groups used two different systems, yeast mitochondria and reconstituted liposomes, to measure the kinetics of ATP/ADP exchange, it is difficult to directly compare the ir results in somatic ANTs and germline ANT4. To solve the issue, Hama zaki et al attempted to generate individual human ANT expressing yeast lacking three yeast AAC isoforms  Unexpectedly, human ANT4 expressing yeast did n o t grow in medium includin g nonfermentable carbon source regardless of the copy number of ANT4 Only after mutagenesis with mutagen e thyl methanesulfonate (EMS), they could obtain human ANT4 mutant yeast strains bearing mutations (A30V, P95S and S202L) in the Ant4 coding sequence, which we able to grow on nonfermentable carbon source s In isolated mitochondria from those mutants, K m and V max were measured but displayed a wide variation in the kinetic values depending on the specific mutation Thus,
16 a proper system should be develope d to analyze individual biochemical properties of Ant. Mitochondrial Membrane Transition Pore Complex Apart from the function of ATP/ADP exchange, Ant has been implicated in the mitochondrial permeability transition pore (MPTP). Classically Ant was consid ered to be an inner mitochondrial membrane (IM) component of the MPTP complex which also contains voltage dependent anion channel (VDAC) in the outer mitochondrial membrane (OM) and cyclophilin D (CyD) in the matrix  Permeability transition (PT) is a sudden increase of IM permeability for solute s with molecular mass up to 1.5kDa and is caused by MPTP opening. In physiological condition s the permeability transition pore complex (PTPC) involves an exchange of small molecules such as ADP and ATP between cytoplasm and mitochondrial matrix. When cells encounter lethal signals such a s increase of reactive oxygen species (ROS) and Ca 2+ overload, MPTP open s to allow small molecules to enter freely into the mitochondrial matrix resulting in an influx of water into the mitochondria matrix by osmotic pressure. This increased water content induces swelling of mitochondria and eventually rupture of the OM leading to the release of apoptosis inducing molecules including cytochrome c. A ctivation of the caspase cascade by the released molecules is followed by cell death  Although studies with b ongkrek ic acid an inhibitor of Ant  and sanglifehrin A and cyclosporin A, inhibitors of CyD  suggests that Ant and CyD are components of PTPC, genetic studies has since raise d questions. Kokoszka et al generated conditional Ant2 ko mice and mated the mice with Ant1 / mice  In the double mutant mice, the Ant2 gene was specifically deleted in a liver by albumin promoter induc ing cre recombinase and MPTP opening was analyzed in the isolated mitochondria from the Ant1/2 deficient liver.
17 Surprisingly, uncoupler carbonyl cyanide p (trifluoromethoxy)phenylhydrazone (FCCP) induced MPTP opening occurred in Ant1/2 deficient liver mit ochondria like wild type In addition, increased Ca 2+ activated MPTP in the Ant1/2 deficient liver mitochondria although three fold more Ca 2+ was required to induce the M PTP than wild type mitochondria. In both conditions, c yclosporin A could inhibit the F CCP or Ca 2+ induced MPTP opening, suggesting that Ant is not essential for MPTP opening but may act as a regulator which is sensitive to Ca 2+ induced MPTP opening  In addition, VDAC has been reported to be dispensable for the MPTP opening  Mitochondria from Vdac1 / Vdac3 / and Vdac1/3 null mouse hearts showed similar level s of Ca 2+ induced MPTP opening compared to wild type mitochondria while cyclosporine A inhibited Ca 2+ or oxidative stress (H 2 O 2 ) induced MPTP opening. In contrast, Ppif the gene en coding CyD, null mitochondria were resistant to Ca 2+ or H 2 O 2 induced MPTP opening, suggesting that CyD is an essential component of MPTP  Thus, it remains to be elucidated what molecules are true components of PTPC Uncoupler In addition to its antiporter and potential MPTP functions Ant is involved in uncoupling. Uncoupling is respiration that is not coupled with ATP synth esis by oxidative phosphorylation in inner mitochondrial membrane. I t utilizes the electrochemical proton gradient established across the inner mitochondrial membrane for thermogenesis instead of ATP generation by ATP synthase and it also prevents an exces sive increase of ATP by hyperpolarization that would inhibit respiration  A number of mitochondrial uncoupling protein s are invo lved in uncoupling. A well known example is uncoupling protein 1 (Ucp1) which functions in brown adipose tissue. However, in addition to brown adipose tissue, various other tissue s also become
18 uncoupled under the presence of fatty acids  Thus, it has been suggested that other inner mitochondrial membrane transporter s including Ant function as an uncoupler. Surprisingly, basal uncoupling and fatty acid induced uncoupling was found in Ucp1 / brown fat cells  O leate induced uncoupling was inhibited by carboxyatractyloside (CATR), an Ant inhibitor in Ucp1 / fat and liver mitochondria wherea s basal uncoupling was inhibited by CATR in Ucp1 / fat mitochondria but not in Ucp1 / liver mitochondria. While Ant2 is predominantly expressed in liver, both Ant1 and Ant2 are expressed in brown fat tissue. Thus, it is thought that Ant1 and Ant2 may be responsible for basal uncoupling and fatty acid induced uncoupling, respectively. However, it is still questionable whether Ant1 and Ant2 function as uncoupler s in other tissues and uncertain whether Ant3 and Ant4 also mediate uncoupling. Adenine Nucleoti de Translocase 4 Unlike other somatic Ants Ant4 is a germ line specific Ant which is exclusively expressed in male germ cells and loca lizes in the inner membrane of mitochondria  Mitochondrial morphology is known to be relevant t o the metabolic status of cells. In this regard, t here are two major morphological conformations: orthodox and condensed. The orthodox conformation is characterized by linear tube like cristae with one or at most two connections to the periphery wher e as t he condensed conformation have larger cristae with multiple connections to the periphery and an irregular pattern of inner mitochondrial membrane  The condensed conformation represent higher mitochondrial respiration compared to the orthodox conformation  During spermatogenesis, mitochondrial morphology is dramatically changed. O rthodox type mitochondria ar e observed in spermatogonia and preleptotene and leptotene spermatocytes and intermediate mitochondria form in zygotene speramtocytes.
19 Pachytene and secondary spermatocytes and early spermatids have condensed mitochondria. Intermediate mitochondria appear again in late spermatids and spermatozoa  Thus, mitochondria are highly activated in mainly meiotic male germ cells to provide energy This mitochondrial respiration activity change is relevant to lactate metabolism in male germ cells [23,24] In Sertoli cells, glucose is taken up via specific glucose transporters (Gluts) and is converted to pyruvate through glycolysis. Pyruvate is converted to lactate by lactate dehydrogenase s (Ldhs). Lactate is transported out of Sertoli cells and in meiotic and postmeiotic germ cells via monocarboxylate transporters (MCTs). Lactate is then oxidized to pyruvate by LdhC and enters the tricarboxylic acid (TCA) cycle in male germ cells. Thus, high mitochondrial respiration activity is required for male germ cells to produce ATP using pyruvate during meiotic and early postmeiotic periods. Interestingly, this observation is also relevant to the Ant4 expression pattern  Ant4 transcript level s increase through leptotene and zygotene stages of spermatocytes peak in early pachytene spermatocytes and then decrease in late pachytene spermatocytes and round spermatids. This suggests that Ant4 likely play s an i mportant role in male germ cell metabolism during meiosis. As speculated, genetic deletion of Ant4 causes early meiotic arrest in male germ cells, leading to male infertility [25,26] However, it should be noted tha t there is no direct evidence to demonstrate that a defect in the energy supply in Ant4 null testes results in impaired spermatogenesis. The present study focused on investigating the hypothesis that Ant4 has a specialized function in spermatogenesis and/ or sperm functions. Based on the tissue specific gene expression of Ant isoforms, it has been suggested that an individual Ant
20 plays a specialized role in the tissue where it is expressed. Thus, Ant4 may be specialized and optimized to function in male ger m cells On the other hand, localization of the Ant2 gene on the X chromosome in mammals implies that Ant4 might only compensate for the loss of Ant2 gene expression during male meiosis due to meiotic sex chromosome inactivation (MSCI). In the present stud y, I will initially demonstrate that Ant4 is also conserved beyond mammalian species, in anole lizard, which gave us a further insight regarding the origin and specific role of Ant4 in male germ cells (Chapter 2). In C hapter 3 I will attempt to elucidate t he hypothesis by adding back orthologous Ant4 or paralogous Ant2 genes in Ant4 null mice Further I will scrutinize the role of Ant4 during female germ cell meiosis in order to obtain additional insight regarding the hypothesis (Chapter 4). In C hapter 5 I will discuss overall conclusions based on the data obtained from the present study. In addition, I will introduce newly obtained gene conservation data in vertebrates and discuss the potential roles of Ant4 in sperm hyperactivat ed motility
21 CHAPTE R 2 E VOLUTIONARY GENOMICS IMPLIES A SPECIFIC FUNCTION OF ANT4 IN MAMMALIAN AND ANOLE LIZARD MALE GAMETES Background The adenine nucleotide translocase (Ant), also called ADP/ATP carrier (Aac), mediates the exchange of ADP and ATP across the inner mitochon drial membrane, thus playing an essential role in energy metabolism in eukaryotic cells [28 30] Under respiring conditions, ATP produced within the mitochondria is exported to the cytosol through Ant to support cel lular activities. In exchange, ADP is imported to provide a substrate for the conversion of ADP to ATP by ATP synthase. Ant belongs to the mitochondrial carrier family that supports a variety of transport activities across the mitochondrial inner membrane [30,31] A typical Ant molecule comprises 300 320 amino acid residues that form six transmembrane helices. The Ant family proteins are encoded by the nuclear genome. Since free living bacteria do not have Ant like m olecules, Ant proteins are thought to have been derived from a broad specificity transport family of eukaryotic origin  Most eukaryotes, including unicellular eukaryotes, have multiple Ants. For example, the bu dding yeast Saccharomyces cerevisiae has three genes ( AAC1, AAC2 and AAC3 ) that encode Ant proteins  Among them, Aac2p is a major isoform that is abundantly expressed during respiration and repressed du ring fermentation  Aac1p and Aac3p are expressed almost exclusively in aerobic and anaerobic conditions, respectively [33, 35] Moreover, Aac2p and Aac3p have an ability to import ATP that is sufficient for survival of the yeast after loss of mitochondrial genome whereas Aac1p does not  Thus, Reprinted with permission from 27. Lim CH, Hamazaki T, Braun EL Wade J, Terada N (2011) Evolutionary genomics implies a specific function of Ant4 in mammalian and anole lizard male germ cells. PLoS One 6: e23122.
22 different Ants are likely utilized for ATP export and import, in order to efficiently cope with varying external nutrient and oxygen conditions. Provocatively, the S. cerevisiae AAC1 gene appears to have accumulated substitutions at a higher rate that AAC2/3  consistent with the idea that the AAC1 gene underwent a functional change after duplication. It is interesting to establish whether a phylogenetic approach can also provide insights into the functions of Ant genes in vertebrates. In multicellul ar organisms, differential expression of Ant genes depends on a variety of factors, including tissue type, developmental stage, and cellular proliferation state. Most vertebrates possess three distinct Ant paralogs that exhibit a relatively high degree of sequence identity. Of these, Ant1 ( Slc25a4 ) is expressed primarily in the heart and skeletal muscle, and is presumed to be suitable for rapid ATP metabolism in heart and skeletal muscles  Genetic inactivation of Ant1 ( Ant1 / ) resulted in viable mice  However, these animals developed mitochondrial myopathy and severe exercise intolerance along with a hypertrophic cardiomyopathy as young adults  Ant2 ( Slc25a5 ) and Ant3 ( Slc25a6 ) are expressed ubiquitously in somatic tissues; however, Ant2 expression is higher in rapidly growing cells and is inducible in mammals whereas Ant3 appears to be constitutively expressed in all tissues [3,40] It shoul d be noted that rodents do not have the Ant3 ortholog; instead, the mouse Ant2 ortholog seems to have the combined functions of human Ant2 and Ant3 [41,42] Utilizing various approaches, we and others identified a f ourth member of the Ant gene family, Ant4 (also called Slc25a31, Aac4 and SFEC ) in both humans and mice [5,7,19] The human ANT4 gene was predicted to encode a 315 amino acid protein with a relatively high degree o f amino acid sequence identity to the previously identified ANT
23 proteins (73%, 71% and 72% overall amino acid identity to ANT1, ANT2 and ANT3, respectively). Like all other ANT genes, ANT4 encodes a protein that c ontains three tandem repeats of an approxim ately 100 residue domain, where each domain includes two transmembrane regions, This is a characteristic shared by all members of the solute carrier family  The amino acid conservation predicted that Ant4 was a specific carrier of ADP/ATP. The Ant4 protein contains a RRRMMM sequence at the end of the fi fth helix transmembrane domain, a sequences that is conserved in all of the Ant proteins (but not in other Slc25s) that is essential for ADP/ATP transport activity  Indeed, Dolce et al  reported that Ant4 specifically exchanges ADP and ATP, but not other solutes, by an electrogenic antiport mechanism. We have determined that Ant4 is expressed exclusively in testicular germ cells in adult mice, with this expression being particularly high during meiosis  It should be noted here that Ant4 appears to be expressed in e mbryonic ovaries as well  (Lim et al ., unpublished observation). Further, Ant4 is essential f or male germ cell development in mice  Ant4 null male mice are infertile and they exhibit a meiotic arrest at or after the leptotene stage [25,26] The Ant2 gene is alwa ys located on the X chromosome of mammals; thus, Ant2 gene expression is repressed during male germ cell meiosis due to meiotic sex chromosome inactivation (MSCI)  On the other hand, the Ant4 gene is always loc ated an autosome. A plausible hypothesis regarding primary function of Ant4 is that it acts to compensate for the loss of Ant2 expression during male meiosis in mammals  The meiotic arrest phenotype in Ant4 nul l male mice is indeed consistent with this compensation theory  and the compensation theory would be consistent with presence of Ant4 only in mammals. Since Ant4 is present in both eutherian and
24 metatherian mammals this hypothesis would place the duplication sometime between 150 million years ago (MYA), when the eutherian metatherian divergence o ccurred [45,46] and 300 MYA, when synapsids (Synapsida, the clade that includes mammals) diverged from the birds and true reptiles (clade Reptilia)  The details of thi s hypothesis were, however, indirectly challenged by phylogenetic analyses that suggested a more ancient origin of Ant4  However, it is unclear whether the results of those phylogenetic analyses reflect any of the difficulties associated with examining ancient evolutionary events [47,48] Specifically, branch lengths for Ant4 homologs were much longer than those for Ant1, 2, and 3 [ 25] consistent with the hypothesis that Ant4 has a function distinct from the other vertebrate Ant genes but also consistent with the idea that Ant4 may be misplaced in estimates of phylogeny. Here we present direct evidence for a more ancient origin of Ant4 Briefly, the anole lizard has an ortholog of Ant4 that exhibits with a testis specific expression similar to that seen in mammals; however the anole Ant2 gene was unlikely to be located on a sex chromosome. Moreover, additional phylogenetic analyses confirm that the Ant gene family in animals has a complex history of gene duplications and losses and further indicate that the Ant4 subgroup is likely to have had an ancient origin. Here we will introduce our new hypothesis, which is not necessarily mutua lly exclusive with the compensation theory, that there is a specific advantage for mammals and lizards to express Ant4 during spermatogenesis and/or in their sperm. Materials and Methods Database M ining, S equence A lignment and P hylogenetic A nalysis Genomic sequences encoding Ant proteins from various species and their inferred amino acid sequences, including the anole lizard, were obtained from the
25 Ensembl ( http://useast.ensembl.org/index.html ) NCBI ( http://www.ncbi.nlm.nih.gov ), and JGI ( http://genome.jgi psf.org ) databases Sequences were aligned using Mafft [49,50] and imported into MacClad e 4.08 ( http://macclade.org ) for manual adjustment. The maximum likelihood (ML) estimate of phylogeny was obtained using RAxML  version 7.2.8, using the LG model  with distributed rates across sites and empirical amino acid frequencies. We chose the LG+ +F model because was the best fitting model from the candidate set of empirical models. Model fit was assessed using the Akaike information criterion  using a neighbor joining (NJ) tree estimated using uncorrected distances ( p distances). To assess support for specific groups in these we conduc ted 500 bootstrap replicates, using a full ML search for each replicate. Chromosome S ynteny Syntenic chromosomal regions including Ant4 gene among ( Homo sapiens ), anole lizard ( Anolis carolinensis ) chicken ( Gallus gallus ) turkey ( Meleagris gallopavo ), an d zebra finch ( Taenopygia guttata ) were examined co mparative genomics information Ensembl and UCSC genome browser  For the identification of degenerate DNA fragment of putative Ant4 gene in chicken, the putative degenerated Ant4 region o f chicken was translated into an amino acid sequence and the amino acid sequence was compared to the protein encoded by the human Ant4 gene. To determine when the shifts in the constrains on the avian Ant4 orthologs occurred, we used PAML 4.2  to estimate the rat io of nonsynonymous substitutions per nonsynonymous site ( K A ) to synonymous substitutions per synonymous site ( K S ). The K A / K S ratio mouse ( Mus musculus ), and dog ( Canis familiaris ) sequences in addition to the avian pseudogene sequences. The set of codon models that was examined for this question
26  and branch models  branches). Akaike weights  which can be interpreted as the probability that a specific model is the member of the candidate set with the smallest distance to the (unknown) true model, were calculated to compare the fit of the candidate models. Animals and Sample P reparation Anole lizards were rapidly decapitated, and tissue samples were immediately frozen on dry ice. They were stored at 80 o C until use. For the extraction of total RNA and genomic DNA, the tissues were ground in the liquid nitro gen using a mortar and pestle prechilled in liquid nitrogen. Murine tails were cut from 3 week old C57BL/6J when they were weaned and the tails were kept at 20 o C. RNA E xpression (qRT PCR) Total RNA was isolated from heart, liver and testis of anole lizards by DNA contaminated during the RNA isolation was removed by TURBO DNA free TM kit (Abion). Complementary cDNA was synthesized from the total mRNA by reverse transcriptase and random primers (Applied Biosystems) under the following synthetic conditions; 25 o C for 10min, 37 o C for 120min, 85 o C for 5sec. The qRT PCR was performed with the SYBR Green assay (Applied Biosystems) under the fol lowing thermal cycling conditions; 95 o C for 10min and 40 cycles at 95 o C for 15sec and 60 o C for 1 min. The levels of Ant mRNAs in each tissue were normalized to the housekeeping gene, actin and relatively quantified by the 2 Ct method. T emplate equivalent to 2.5ng of total RNA was used for the amplification and each reaction was performed in triplicate. The primer sets used were: Ant1
27 AAGAAAGCCTTGCCTCCTTC, Ant2 TCCAGCTGATCAAAATGTGG, Ant3: TTGC TCTGGGAGCATACCTTTTGC, Ant4 AAGGCTCCTCGAAAGAAAGC, and actin GCAGGACTCCATACCCAAAA. Gene Dosage A nalysis (qPCR of G enomic DNA) The relative gene dosages of Ant1 2 3 and 4 as well as Slc25A43 which is a neighboring gene of Ant2, between female and male genome, was determined by qPCR of genomic DNAs. Wizard Genomic DNA Purification Kit (Promega) according to the e lizards and tails of male and female mouse as controls. A 2.5ng sample of gDNA from each tissue was used as the template and qPCR was performed under the same thermal cycling conditions as described above. Each amplification was normalized to the actin of anole lizard and mouse, respectively and relative gene copy number was calculated by the 2 Ct method. Finally, the r elative gene dosages of female obtained by dividing the relative gene copy numbers in female by in male. The primer sets are listed be low. Primer set I for anole lizard was Ant1 TGCCACTTCCGAGACCTCTA, Ant2 AAACGGAAACACGGATACGA, Ant3 GCCATACAGCCCAGAAAACA, Ant4 GATTTG GCAGCATTCCCTAA and actin GGGGTGTTGAAGGTCTCAAA. Primer set II for anole lizard was Ant2 ATGGTGCCCGAGTACATGAT, Ant4
28 CCGAAAGAGATGGGATCAAA, and actin CATTAGCCCT GGATACCGCAGGACTCCATA. Primer sets for mouse was Ant2 TATCTGCCGTGATTTGCTTG, Ant4 TATGGGATGCTAAGGCCAAG and actin GGGGTGTTGAAGGTCTCAAA. CpG Methyl ation Analysis (Bisulfite S equencing) The bisulfite conversion of genomic DNA extracted from the tissues as described above was carried out using EZ DNA Methylation Gold TM Kit (Zymo Research). Putative Ant2 promoter regions in the bisulfite converted gDNAs of anole lizard and mouse were amplified by PCR with the following primer sets: Ant2 CCTAAAACAAAAACTTAATCCTCTC and Ant2 AAAATACCCCCTTTCTATACAAATC. The ampli cons were gel purified and cloned into bacterial cells using TOPO TA cloning (Invitrogen). C o lonies from each sample were chosen and sequenced. R esults Anole Lizard H as t he Ant4 O rtholog The green anole lizard, Anolis carolinensis is the first non avian r eptile genome to be sequenced. When we examined the conservation of Ant gene orthologs in the anole we found likely orthologs of all four Ant1 2 3 and 4 genes present in humans. The amino acid identity of the proteins encoded by the anole Ant1, 2, 3 and 4 genes to the proteins that are encoded by their human orthologs was 90%, 93%, 90% and 79% respectively. Moreover, their intron exon structure was conserved between the anole
29 and mammals, with four exons observed in Ant1, 2 and 3 and exons in Ant4 just a s seen in mammals ( Figure 2 1A ). Indeed, the deduced amino acid sequence of anole Ant4 had unique features only seen in mammalian Ant4, including N terminal and C terminal extensions and RRRMMMQSGE at the end of the helix H5 region ( Figure 2 1B ). To the be st of our knowledge, this was the first time a clear ortholog of Ant4 has been identified outside of the mammals. Consistent with the intron exon structure and the signature sequences, phylogenetic analyses ( Figure 2 2A ) verified the relationship between anole and mammalian Ant4 and further suggested that the Ant4 lineage is more closely related to specific invertebrate Ant genes than it is to Ant1 2 or 3 These analyses suggested that there have been a large number of gene duplications (and gene losses, given the inclusion of many organisms with relatively complete genome sequences that were included in our analysis) and further indicated that vertebrate Ant genes are found in a was united by a long branch with a high degree of bootstrap support. Within this group, however, bootstrap support was relatively limited. Since the inclusion of very divergent outgroup sequences has the potential to distort the ingroup topology [59,60] W e excluded the outgroup sequences (along with two very divergent ingroup sequences) and conduct ed another phylogenetic analysis ( Figure 2 2B ) These analyses confirmed the relationship between a set of tunicate Ant 4 like genes and vertebrate Ant4 orthologs. Although the relevant tunicate genes are intronless and lack the signature sequences of vertebrate Ant4 orthologs, the phylogenetic analyses support the
30 hypothesis that that the divergence between the Ant4 group and the Ant1 3 group reflects an ancient duplication, substantially predating the origin of vertebrates. Anole Ant4 I s E xpressed in T estis We next examined the gene expression pattern of Ant1 2 3 and 4 in anole lizard. Heart, liver and testis were harves ted from male anole lizard, and total RNA was extracted. The amount of Ant mRNA in these organs was then examined using quantitative RT PCR ( Figure 2 3 ). Ant1 which is selectively expressed in heart and skeletal muscle in mammals, was also expressed in an ole heart but not detectable in anole liver and testis. Ant2 and Ant3 which are expressed ubiquitously in mammals and they were also expressed in all of the anole tissues examined. Ant4 which is expressed exclusively in testis in adult mammals, was also expressed in anole testis but not in any somatic organs examined. These data indicate that all Ant orthologs in anole and mammals. In particular, testis specific expression of Ant4 suggests that the gene is likely to play a role in anole male germ cells s imilar to the role that Ant4 plays in mice. Chicken H as a D egenerate DNA F ragment t hat C orresponds t o a L ikely Ant4 P seudogene Based on phylogeny and intron exon structure, the anole Ant4 ( Slc25a31 ) gene appears to be an ortholog of mammalian Ant4 genes. Moreover, the gene expression profile suggests that its function in the anole is similar to its function in mammals. This implies that the common ancestor of mammalian and reptilian species had the Ant4 gene. Then, what happened to the Ant4 gene in birds? To answer this question, we examined chromosomal regions harboring Ant4 in other vertebrates As shown in Figure 2 4A human ANT4 gene is located on chromosome 4q28.1, between the INTU gene and the HSPA4L PLK4 and MFSD8 genes. Chromosomal synten y is con served
31 between human and anole genomes and t end of the Ant4 gene in the anole as well These data further confirm that the anole Ant4 gene is the authentic ortholog of the mammalian Ant4 genes. When the chicken genome was e xamined, the syntenic region including Intu Hspa4l Plk4 Mfsd8 genes were present in this order was found in chicken chromosome 4, although an Ant4 ortholog was not present in this region ( Figure 2 4A ). However, it was possible to identify DNA fragments of similar to the Ant4 gene between Intu and Hspa4l genes in the chicken genome and other avian genomes For example, the sequence from Chromosome 4:35430626 to 35431987 in the chicken genome includes two genomic regions which could be translated to 23 and 40 amino acid fragments with a high degree of similarity to human and anole Ant4 proteins. However, there is one nonsense codon and a two bp insertion that creates a frameshift in these regions. Similar degenerate Ant4 like DNA sequences were also evident in the turkey and zebra finch genomes, although the zebra finch sequence only includes a region similar to the second of these exon s An intron like sequence was present at a site identical to the first intron of the human and anole Ant4 genes in the avia n pseudo coding regions, and these regions retain the GT sequence after the splice donor site and the AG before the splice acceptor (only the AG is retained in the zebra finch since the first exon like region appears to have been deleted in this taxon). Mo reover, the second exon is followed by a GT sequence in the zebra finch (this sequence is AT in the chicken and turkey, suggesting a mutation occurred in these lineages after their divergence from the finch lineage). These data confirm that the common ance stor of reptilian and mammalian species had the Ant4 gene, as suggested by the phylogenetic
32 analyses ( Figure 2 2 ), and indicate that the Ant4 gene has been degraded during avian evolution. Since we were not able to identify obvious synteny of the region in the zebrafish ( Danio rerio ) and frog ( Xenopus tropicalis ) genome, we were not able to elucidate if degenerate Ant4 pseudogenes are present outside the amniotes. It seems clear that the Ant4 gene was inactivated in the common ancestor of the zebra finch (a member of the Passeriformes) and the chicken and turkey (both members of Galliformes). The avian supergroup (Neognathae) that includes both of these orders is very diverse, including virtually all orders of birds [6 1] Given this constraint, it remains possible that the Ant4 was inactivated at any point before the divergence of these lineages ( Figure 2 4B ). To place the timing of this inactivation in a framework that can be tested using phylogenetic methods we simpl ified this to two models, an early inactivation model and a late inactivation model (indicated by the arrows in Figure 2 4B ). Fitting a set of models of codon evolution to the nucleotide sequence data for Ant4 including the two avian Ant4 pseudogenes, all ows us to test these si describes the relative rate s of nonsynonymous and synonymous substitution s will increase to a value of one upon gene inactivation. As a heuristic, we present a comparison of trees with branch lengths proportional to the amounts of nonsynonymous and synonymous change ( Figure 2 4C ). As expected for pseudogenes, the synonymous branch lengths were similar for birds and other taxa whereas the nonsynonymous branch lengths were much longer in birds ( Figure 2 4C ). The length of the branch leading to the birds in the nonsynonymous tree was also long, a finding more consistent with the early inactivation model.
33 W e tested five different models of codon evolution using th is Ant4 data All of th ese models involved shifts in the and the model set included one model with two different s that was more consistent with the early inactivation model and one that was more consistent with late inactivation This allowed us to put t he branch length heuristic in more rigorous framework. The best fitting model based upon the AIC was the two was not excluded from the 95% credible set of models Nonetheless, the early inactivatio n model is better supported by th from other alternative models Taken as a whole, the majority of the data suggest an early inactivation. Anole Ant2 I s N ot o n a Heterogametic Sex Chromosome We have previously hypothesized tha t mammalian Ant4 genes may have been conserved to compensate the loss of X linked Ant2 ( Slc25a5 ) gene expression during male meiosis. Of interest, anole lizards are considered male heterogametic reptiles, having XY chromosome differentiation [62,63] Thus, we thought that the anole Ant2 gene might also be localized on the X chromosome. Since chromosomal localization data are not available for a number of green anole genes, here we determined relative copy numbers of genes between male and female animals by quantitative genomic DNA PCR ( Figure 2 5 ). First, in order to verify the methods, we compared dosages of the Ant2 genes in male and female mice. When compared to the control actin gene, relative copy number of the X linked Ant2 gene was approximately double in female mice when compared to male mice. In contrast, dosage of the autosomal Ant4 gene was similar between male and female mice. These indicate that the method is valid. When we applied the method to anole li zard, relative copy number ratios were
34 approximately one for all Ant genes including Ant2 Indeed, multiple primer sets were used to confirm the results for the Ant2 and Ant4 genes. Similarly, relative copy number ratio between male and female were examin ed for Slc25a43 another gene encoding a solute carrier protein that is located right next to Ant2 gene in syntenic region of both the mouse and the anole genome. As seen with the Ant2 gene, relative copy number ratio of the Slc25a43 gene in female/male we re approximately two in mice but approximately one in anole ( Figure 2 5B ). Taken together, these data indicate that anole Ant2 gene is located not at a heterologous region of X chromosome, but either at autosomes or the pseudoautosomal regions of the anole X and Y chromosomes. We previously demonstrated that the promoter region of murine Ant2 gene is partially methylated at CpGs in female somatic tissues. This is due to somatic X chromosome inactivation of one allele in female. Here we examined CpG methylat ion of the promoter region of Ant2 gene in male and female tissue samples of anole lizards and mice. As we previously described, the murine Ant2 pr omoter was unmethylated in male whereas female showed hypermethylation of the Ant2 promoter among approxima tely half samples examined ( Figure 2 6 ). In contrast, the anole Ant2 promoter was largely unmethylated in both male and female livers. These data are consistent with the idea that Ant2 is localized at autosomes or undifferentiated regions of sex chromosome s in anole lizard. D iscussion Previously we described that the emergence of Ant4 occurred at least 150 million years ago  based on the conservation of the gene in both eutherian and metatherian mammals. Here we demonstrated that anole lizard has an authentic Ant4 ortholog based upon a number of criteria, including phylogeny, intron exon structure,
35 synteny, and patterns of gene expression. Further, degenerate DNA fragments of a putative Ant4 pseudogene were ident ified in the syntenic region of avian genomes. These indicate the presence of the Ant4 gene in the common amniote ancestor. This indicates that origin of the Ant4 gene clearly pr edates the divergence of mammals and reptiles, more than 300 MYA. Phylogenetic analyses suggest that the gene duplication that led to the Ant4 lineage and the Ant1 2 and 3 lineage may actually be much more ancient than the origin of amniotes. A number of invertebrate lineages were nested within the diversity of vertebrate Ant gene s ( Figure 2 2A ), a topology similar to the estimate of phylogeny obtained by Brower  However, there are a number of reasons that one can obtain an incorrect estimate of phylogeny [47,48] and the branch length heterogeneity evident in the large scale animal Ant tree is a major cause for concern (especially for the Brower  tree that was based upon NJ of protein distances that we re corrected by the Kimura  formula). To address concerns regarding the phylogeny we used ML with the best fitting model ( model adequacy. We also examined two sets of sequences, one of which excluded the divergent outgroups and other long branches ( Figure 2 2B ) because divergent can distort the topology [59,60,65] Hi s estimate of phylogeny appeared quite robust to the inclusion of different sets of taxa, although the exclusion of divergent sequences did appear to improve the estimate of p hylogeny. Specifically, the topology for both the Ant1 and Ant3 subgroups were more congruent with the species tree (e.g., analyses of the large taxon sample placed the root within tetrapods whereas analyses of the smaller taxon sample placed it between fi sh and
36 tetrapods, with the latter being more likely to be correct). Significantly, we found a higher degree of bootstrap support for the clade including vertebrate Ant4 and tunicate Ant4 like sequences when the more limited taxon sample was analyzed, consi stent with our hypothesis that Ant4 group is the result of an ancient duplication. It is clear that all estimates of Ant phylogeny demand a number of duplications and losses to reconcile them with the animal tree of life. Some of the observed incongruence between expectation based upon the species tree and the estimates of Ant phylogeny may reflect the lack of power associated with relatively short protein sequences  Indeed, fairly long sequences can be necessary to have a high probability of obtaining an accurate estimate of phylogeny [67,68] In addition to issues associated with the power of phylogenetic analyses to distinguish among alternative trees, the observed branch lengths heterogeneity ( Figure 2 2 ) suggests that the patter ns of evolution for Ant genes have undergone substantial changes during animal evolution. More specifically, the branch length for the Ant4 group are somewhat longer than those for the Ant1 2 3 group, suggesting that the latter group has retained more of the ancestral functions of animal Ant proteins. Indeed the distance between the anole and human Ant4 (0.3608 amino acid substitutions per site, based upon patristic distances calculated from the tree in Figure 2 2B ) indicates that the rate of amino acid e volution was approximately 2.4 fold faster for the product of the Ant4 than it was for the Ant1 2 and 3 gene products (the mean anole human distance for th e Ant1 3 group was 0.1484 amino acid substitutions per site). It remains possible that these change s in the patterns of evolution for Ant gene may also have had an impact upon the estimate of Ant phylogeny, although the heterogeneity does not appear so extreme as to radically
37 alter our conclusions regarding the evolution of the vertebrate Ant gene famil y. The basis for the higher rate of nonsynonymous sequence evolution evident in the Ant4 group is unclear, although it is known that genes with sex biased expression often exhibit higher rates of evolution  The basis for the observed rate differences for sex biased genes are complex  but it is important to note that our observation that Ant4 exhibits a higher rate of evolution is consistent with a role for this gene in male repro duction in various amniotes. Alternatively, the higher rate of Ant4 evolution may reflect changes in selection unrelated to sex biased expression either the relaxation of purifying selection the action of positive section having acted upon Ant4 or both phenomena Indeed, when the branch length heterogeneity observed in our Ant trees ( Figure 2 2 ) is combined with the evidence for the action of natural selection upon other mitochondrial proteins [71,72] that various Ant proteins have been subject to episodes of positive selection during animal evolution. Regardless of the specifics, both hypotheses regarding Ant4 (relaxed purifying selection or positive selection) are consistent with the idea that Ant4 has undergone a shift in function. Pseudogenes exhibit relaxed selection on nonsynoymous sites after the relaxation of purifying  Although the fact that the ML estimates 3921 for exon 2 alone ) might seem more consistent a relatively recent the relaxation of purifying selection tends to be quite wide  Instead, the best fitting models suggest that the relaxation of purifying selection upon Ant4 in birds occurred substantially earlier than the Passeriformes Galliformes divergence at least 87 MYA
38  It will be of interest to determine whether a functional Ant4 gene is present in any other extant archosaurs (birds and crocodilians) and whether there are any phenotypic correlates of the Ant4 gene loss The pre sent study also urged us to reconsider our original hypothesis that was based upon the conservation of Ant4 only in mammals, since it is clearly present both in mammals and anole lizards. Based on the unique chromosomal localization pattern of the Ant gene s as described in Introduction we originally hypothesized that the conservation of Ant4 is driven by the sex chromosomal localization of the Ant2 gene and the subsequent inactivation of the Ant2 gene during male meiosis in mammals. When we identified the authentic Ant4 ortholog in anole lizard, we had first expected that the anole Ant2 gene might be localized to its sex Among reptiles that have various sex determination systems, anol es are considered to have male heterogametic sex determination as mammals [62,63] Although the scenario was rather unlikely since sex chromosomal differentiation has been presumably developed independently in mamma ls and reptiles after the bifurcation of the two species, it is possible that the Ant2 gene can be localized to sex chromosomes in both mammals and anoles by chance. Of interest, an Ant gene homolog was isolated from heteromorphic sex chromosomes in Japane se frog Rana rugosa  an unusual species where different populations have three distinct types of sex chromosomes (homomorphic XY chromosomes, heteromorphic XY chromosomes, and heteromorphic ZW chromosomes). Alt hough the gene was originally annotated as Ant3 in the report  it actually shows greater sequence identity to Ant2 than Ant3 (91% and 88% to human Ant2 and Ant3 respectively) Placement of the Rana rugosa sequ ences using
39 phylogenetic criteria was more equivocal, since they formed a clade with putative Xenopus Ant3 ortholog (data not shown). However, the putative Xenopus Ant3 sequence was placed either within fish ( Figure 2 2 ) or sister to fish (when the Rana se quence was added); this topology suggests either a local rearrangement in the tree (probably due to the relatively short length of the Ant sequences) or additional gene duplications and losses. The absence of bootstrap support for separating Ant2 and Ant3 ( Figure 2 2 ) is consistent with the first hypothesis, which postulates that the relevant branches in our estimate of phylogeny are rearranged relative to the actual evolutionary history Regardless, it is clear that the sex chromosomal localization of a me mber of the Ant2/3 group is not unique to mammals. There could be a yet unidentified common mechanism or reason why members of this subgroup of Ant genes have a higher probability of moving to a sex chromosome after their origin. However, it should be note d that Ant2 gene localizes at the autosomes in birds (chicken and zebra finch); thus the sex linkage of Ant2 is not a universal phenomenon in vertebrates. Chromosome maps are not completed to date for Anolis carolinensis making it difficult to determine whether or not the anole Ant2 gene is localized at sex chromosome. Instead here we investigated relative gene dosages of the gene between females and males. In mice, dosages of Ant2 gene as well as its neighbor Slc25a43 gene were approximately 2:1 between females and males, consistent to the fact that those genes are localized at heterologous regions of X chromosomes. In contrast, in anoles, both Ant2 and Slc25a43 genes dosages were 1:1 between females and males. These data indicate that the Ant2 Slc25a43 r egion is not likely on a heterologous region of the sex chromosomes. One of two X chromosomes in the somatic cells of female
40 mammals is known to be inactivated accompanying CpG methylation. Indeed, we previously showed that the mouse Ant2 gene was partiall y methylated in females but not in males  Here, we demonstrated that anole Ant2 gene was unmethylated equally both in females and males, which is consistent with the idea that Ant2 gene is not on the anole X ch romosome. It should be noted, however, that there is no information available regarding the mechanism (or even the existence) of dosage compensation for sex linked genes in anoles. Thus, it is unclear whether X chromosomes undergo methylation and/or inacti vation in anoles that is similar to the dosage compensation mechanism in mammals, although it is known that reptilian genomes typically have comparable a CpG methylation ratio comparable to that of other amniotes  Alth ough a definitive conclusion regarding the chromosomal location of Ant2 in Anolis carolinensis will require an improved anole genome assembly, our present data suggest that the Ant2 gene is not likely localized to a heterogametic sex chromosome in anoles. Our results imply the conservation of Ant4 may not be simply driven by the sex chromosomal localization of Ant2 gene and its subsequent inactivation during male meiosis. Importantly, our data suggest that testis specific expression of Ant4 existed prior to and is independent of the sex chromosome linkage of the Ant2 gene in mammals. This prompted us to speculate that Ant4 has been functioning originally in testicular germ cells and sperm, in the amniote ancestor and potentially even earlier in evolutionary phylogeny. The degeneration of Ant4 in birds indicates that this Ant paralog is unnecessary for the development of testicular germ cells in this lineage. To this end, one alternative hypothesis could be that Ant4 plays a critical additional role during
41 spe rmatogenesis and/or in sperm function This role would be shared by specific groups of organisms such as mammals and anoles (and probably by at least some other squamates) and absent in other groups of organisms, such as birds. Likewise, the absence of Ant 4 orthologs in the frog genome and a ll teleost fish genomes suggest it is absent from th e s e lineage s as well and the existence of multiple fish genome sequences strongly suggests this absence is real rather than a problem with the completeness (or assembl y and annotation) of the available draft genome sequences Our phylogeny implies that the duplication leading to Ant4 lineage predates the origin of vertebrates, indicating that the absence of Ant4 in this lineage is also likely to reflect gene loss. The l oss of Ant4 in birds and fish could reflect the relaxation of selection on this paralog or the existence of a specific advantage linked to the loss of this gene, since some evolutionary innovations can be linked to gene loss  Regardless, it seems clear that conservation of Ant4 in anoles and mammals cannot be explained by the simplest version of the compensation theory, which postulates selection to retain the Ant4 gene after duplication reflects a need to compensate for the loss of Ant2 expression during male meiosis in mammals. Instead it suggests that vertebrate Ant4 orthologs may have additional specific func tions associated with male gametes prior to the need for compensation.
42 Figure 2 1. Anole lizard has orthologs of the Ant1, 2, 3 and 4 genes. (A) The schematic diagram from the Ensemble database shows the configuration of the genes encoding Ant protein s in human and anole lizard. Exons and introns are shown as black boxes and lines, respectively. Numbers represent the amount of amino acid sequence identity that the products of each Ant gene in the human and anole lizard exhibit. (B) Alignment of the inf erred amino acid sequences of the products of the Ant1, 2, 3 and 4 genes in mammals (human, cow, mouse) and reptile (anole lizard). Ant amino acid sequences of the selected species were aligned using ClustalW2. Amino acids sequence surrounding a signature motif of in the Ant4 protein (RRRMMM) and the N and C terminal parts of the proteins are shown. Amino acid residues are numbered from the initiation codon (M) to the termination codon in each Ant protein.
43 Figure 2 2. The maximum likelihood (ML) estima te of Ant phylogeny supports an ancient origin of Ant4 (A) Large scale phylogeny of animal Ant homologs evolution. Support based upon 500 bootstrap replicates is shown as a percenta ge adjacent to the relevant branches when it exceeds 50%. This phylogeny included a fungal (Saccharomyces cerevisiae) outgroup, and the root of the tree was assumed to lie between fungi and the animal choanoflagellate clade (the latter group is represented by Monosiga brevicollis). This phylogeny suggests a number of ancient gene duplications, Ant homologs. based upon 500 bootstrap replicates is shown as a percentage adjacent to the relevant branches when it exceeds 50%; support for some clades within well supported groups (e.g., branches within teleost fish) was omitted in the interest of simplicity. Analyses using this m ore limited taxon sample were conducted to test the possibility that the position of Ant4 was influenced by the inclusion of divergent sequences. The position of Ant4 was robust to the taxon set analyzed; in fact, the bootstrap support for an ancient origi n (the Ant4 tunicate clade) increased for the smaller taxon sample.
44 Figure 2 3 Anole Ant4 is specifically expressed in testis. Ant1, 2, 3 and 4 gene expression in heart, liver, and testis of anole lizard was examined by qRT PCR. Ant mRNA amounts were n ormalized based upon the expression of the actin mRNA. Error bars indicate standard deviations of triplicate samples.
45 Figure 2 4. Degenerate DNA fragment of putative Ant4 gene in the syntenic region of chicken genome. (A) The syntenic region that inc ludes Ant4 in the human, anole lizard and chicken genomes. The schematic diagram shows Ant4 (red boxes) with chromosomal location flanked by neighboring genes (blue boxes). Translated amino acid sequence of degenerate DNA region of the putative Ant4 gene l oci in chicken genome was aligned with amino acid sequence of Ant4 of human and anole lizard. The asterisk in the chicken sequence is a stop codon; additional inactivating mutations include a 2 bp frameshift between nucleotide that would encode the D and K at the end of the chicken sequence and a stop codon immediately after the K. (B) Vertebrate phylogeny showing approximate divergence times (in millions of years before present) inactivation models. (C) Phylogeny with branch lengths reflecting numbers of synonymous and non synonymous substitutions. The light gray lines are included to make it easier to identify the taxon associated with each terminal, they have no biological sign ificance.
46 Figure 2 5 Gene dosage analysis in female and male. Relative copy number ratio of Ant1, 2, 3 and 4 genes and Slc25a43 gene in male and female animals were examined by quantitative PCR analysis of genomic DNA of anole lizard and mouse. PCR amp lification was normalized to the control actin gene. Error bars indicate standard deviations of three independent experiments. (A) Relative gene dosages of the Ant 1, 2, 3 and 4 between female and male in anole lizard and mouse. The result was confirmed with two different sets of primers for anole Ant2 and Ant4 genes. (B) Relative gene dosage of the Slc25a43 between female and male in anole lizard and mouse. The Slc25a43 gene localizes adjacent to the Ant2 gene in both anole and mouse genome (left panel).
47 Figure 2 6 CpG methylation of the anole and mouse Ant2 genes. The CpG methylation of the Ant2 gene promoter regions in female and male of anole lizard and mouse was examined by a bisulfite sequencing analysis. The schematic diagrams show CpG islands in the Ant2 gene promoter regions examined in anole lizard and mouse. Each row of the circles represents an individual clone of a PCR amplicon in bisulfate sequencing analysis.
48 CHAP T ER 3 TRANSGENE ANT4 EXPRESSION DOES NOT RESTORE THE IMPAIRED SPERMATOGENES IS IN ANT4 NULL TESTIS Background In the mouse genome Ant4 localizes on chromosome 3 and Ant2 is linked to the X chromosome. The autosomal localization of Ant4 implies it may play a possible compensatory role for Ant2 repression in germ cells during male meiosis when unsynapsed XY sex chromosome s are silenced by meiotic sex chromosome inactivation (MSCI) [ 1 2] MSCI is the process of transcriptional silencing of the sex chromosomes that occurs during the meiosis in species which have a heterogametic sex d etermination mechanism. During meiosis, homologous chromosomes (autosomes) pair, synapse and exchange genetic material by homologous recombination. However, for sex chromosomes, for example, X and Y chromosomes do not pair except for their pseudoautosomal region. This lack of pairing induces meiotic silencing in unsynapsed chromatin. Indeed, Ant4 and Ant2 transcript levels measured by real time PCR show ed that Ant4 is actively expressed while Ant2 is repressed in germ cells during male meiosis  Interestingly, our anole Ant4 study ( C hapter 2) did not support the compensatory role of Ant4 as we showed Ant4 is exclusively expressed in testis even though Ant2 is not localized on the X chromosome Rather, this finding impl ies that Ant4 p erforms a specialized function in spermatogenesis and/ or sperm. To prove this, we forcefully expressed Ant2 in Ant4 deficient testis using a bacterial artificial chromosome (BAC) transgenic approach. BAC is an engineered DNA fragment with a length of sever al hundred kilo base pairs including long range cis regulatory elements required for correctly regulated gene expression  BACs have been increasingly used for
49 transgenesis because it allows for comparable transgene expression to its endogenous counterpart and displays resistance to positional effects compared to shorter transgene construct s  Moreover, BACs can be easily and genetically manipulated by recombination mediated genetic engineering (recombineering)  During recombineering, a d efective prophage is integrated into the bacterial genome. The expression of recombination proteins exo, bet, and gam is regulated by the P L promoter and temperature sensitive repressor Cl857. The P L promoter is repressed by the Cl857 repressor at 32 o C and activated at 42 o C when the repressor is deactivated. Thus, unwanted recombination can be prevented by the co ntrol of temperature. Moreover, the recombination is highly efficiently achieved with short homology arms (50bp arm for a 1kb fragment). In the current study, transgene Ant2 and Ant4 expressing mice were generated by genetically manipulated BACs of two dif ferent lengths, ~168 and ~139kbps. The transgenic mice were mated with Ant4 ko mice to generate transgene Ant2 and Ant4 expressing mice on an Ant4 null background. In the Ant4 tg mice used as a control, transgene Ant4 expression and the phenotype of tg Ant4 +; Ant4 / testis were examined. Materials and Methods Mouse H usbandry Mice were bred and maintained on a 12hr light/dark cycle and given ad libitum access to food and water. All in vivo experimental procedure carried out in the study were approved by the Institutional Animal Care and Use Committee at the University of Florida. Generation of BAC Transgenic M ice Transgenic c onstructs and t argeting of BAC DNA
50 Two BAC clones including the Ant4 locus, RP24 324D19 (~168kbps) (CHORI USA) and bMQ363n10 (~139kbp s) (Source Bioscience) were genetically manipulated for the generation of BAC Ant4 IRES GFP and BAC HA Ant4 transgenic mice, respectively. RP24 324D19 (~168kbps) has 50kbps upstream and 101kbps downstream of Ant4 and bMQ363n10 (~139kbps) has 58kbps upstre am and 64kbps downstream of Ant4 First, each transgenic construct w as generated in pCR2.1 TOPO (Invitrogen) vector for the genetic modification. For BAC Ant4 IRES GFP tg mice, the 5 homologous arm which is 1kbp upstream from the ATG initiation site of th e Ant4 gene and the Ant4 cDNA sequence flanked by loxP sites were fused by PCR and introduced in to the pCR2.1 TOPO vector. IRES GFP cassette, SV40 polyadenylation (pA) signal and kanamycin resistance gene cassette flanked by FRT sequences, were then insert ed downstream of the Ant4 cDNA sequence. Lastly, the 3 homologous arm which is 1kbps downstream of exon 1 of the Ant4 gene was introduced downstream of the kanamycin resistance gene cassette ( Figure 3 1A). For BAC HA Ant4 tg mice, the same construction steps were performed except that Ant4 cDNA with an HA tag was fused with the 5 homologous arm (1kbp) by PCR and the fused product was introduced into the TOPO vector without an IRES GFP cassette ( Figure 3 6A). To recombine the transgenic construct s with B AC DNAs, E. coli strain EL250 which can produce r ecombination proteins by heat shock at 42 o C and flippase under 0.1% L arabinose was used for recombineering The recombineering was performed as described by the Frederick National Laboratory for cancer rese arch ( http://ncifrederick.cancer.gov/research/brb/recombineeringInformation.aspx ). Briefly, each BAC DNA was transformed into EL250 by electroporation and selected o n
51 c hloramphenicol (12.5ug/ml) containing plates. After overnight culture at 32 o C in 5ml of low salt containing Lysogeny broth (LB) medium, the selected BAC DNA containing EL250 was incubated for 15min at 42 o C for activation of homologous recombination prot eins. After immediate cooling on a wet ice, the EL250 strains were washed with ice cold water then each transgenic construct was transformed into the corresponding BAC DNA containing EL250 strains by electroporation and screened on kanamycin (50ug/ml) cont aining plates. After 2 3day culture, several clones were picked up to screen transformants for recombineered BAC DNAs. To screen for recombineered BACs, genotype was checked by PCR (WT Ant4 : 5 GCTGTGCACTGATTGAGCAT 5 TGTCAACGTCACCTCCTCTG and Tg Ant4 :5 G CTGTGCACTGATTGAGCAT 5 TGCCACGCCAATAACTTAAA and thermocycling condition: 95 o C for 5 min and 35 cycles at 95 o C for 30 sec and 55 o C for 30sec and 72 o C for 45sec and 72 o C for 10min) and the positive clones by PCR were confirmed by Southern blot analysis with a DIG labeled DNA probe. To confirm the integrity of the recombineered BAC DNAs, restriction patterns of the BACs were compared to that of the corresponding original BACs using EcoRI. T o remove the kanamycin resistance gene cassette, clones were first cultur ed overnight at 32 o C in 5ml of medium Next, they were incubated for 1hr in 0.1% L arabinose containing medium to induce flippase expression Afterwards, they were spread on c hloramphenicol containing plates After 2 3day s several clones were isolated and checked by PCR and Southern blot analysis to confirm cassette deletion The selected clones were cryopreserved at 80 o C until further processes. Purification of the targeted BACs
52 The targeted BACs were purified by Cesium Chloride (CsCl) d ensity g radient c entrifugation for pronuclear injection. To amplify BAC DNAs, BACs were cultured overnight at 32 o C in 3ml of low salt containing LB medium T he next day, 1ml of the starter volume was inoculated into 1.5L of low salt LB followed by overnight culture at 32 o C Next the cultures were collected and lysed to extract the amplified BAC DNAs. The extracted BAC DNAs were precipitate d by adding 100% ethanol and spun down at 10,000rpm for 10min. T he resulting pellets were washed with 70% ethanol then air dried at 37 o C For rehydration, 8ml of Tris EDTA (TE) buffer with RNase was added. To carry out CsCl density gradient centrifugation, CsCl was dissolved in the BAC DNAs/TE mixture in order to make a final concentration of 1.55g/ml Next, the BACs/TE/CsCl mixture w as di vided into 2 ultracentrifuge tubes (Beckman) and 100ul of ethidium bromide (10mg/ml) was added to each tube to visualize the BAC DNAs. Three consecutive ultracentrifugations were performed with an NVT 90 fixed angle rotor at 25 o C: 1) 55,000rpm overnight 2 ) 78,000rpm for 4hrs, and 3) 55,000rpm overnight. After each ultracentrifugation, red bands (DNA) were extracted using an 18gauge needle and were transferred to a new ultracentrifuge tube. T he tube was filled with pre made 1.55g/ml TE RNAse solution. After the third ultracentrifugation, the red bands (~1.5ml) were collected into a 15ml tube and combined with salt water saturated butanol to remove the ethidium bromide from the purified BAC DNA solution. 2ml of butanol was added to the BAC DNA solution and mi xed by inverting. The tube was placed upright for a few seconds until the aqueous layer (bottom) separated from the butanol layer (upper) Next, the
53 upper butanol layer was removed and 2 ml of butanol was added to the tube again. These steps were repeated u ntil the lower BAC DNA solution was completely clear. To extract CsCl from the purified BAC DNA solution, dialysis was performed. The BAC DNA solution was transferred into a dialysis tube (MWCO 12,000 14,000, Spectra/Por) and two consecutive dialyses were performed. For the first, the dialysis tube was immersed with gentle stirring in 1L of injection buffer (10mM tris pH 7.5, 0.1mM EDTA, 100mM NaCl) for 7hrs. The tube was then transferred into 3L of fresh injection buffer and stirred overnight at 4 o C The n ext day, the CsCl free BAC DNA solution was transferred into a microtube and the concentration was measured by a spectrophotometer ( Synergy HT Multi Mode Microplate Reader BioTek). Finally, the purified BAC DNAs was stored at 80 o C until pronuclear inject ion. Pronuclear i njection The BAC DNAs were injected into 200 300 embryos obtained from B6D2F1/J (C57BL/6JXDBA2J) at the Mouse Model Core in the Department of Animal Care Services at the University of Florida. The BAC injected embryos were implanted to rec ipient mice. After 6 weeks, tail snips from new born mice were collected and genotyped by PCR as described in the section of Transgenic constructs and targeting of BAC DNA to select transgenic founders. Histology and I mmunohistochemistry Testes were coll ected from 6 week or 14 week old mice, washed in PBS and fixed overnight in 4% paraformaldehyde in PBS. The next day, the testes were washed with PBS, dehydrated in 70% eth a nol infiltrated and embedded in paraffin. The paraffin embedded testes were cut in to 5um slice s and the sections were placed on glass slides and dried until further staining. For immunohistochemistry, HA tagged Ant4 was
54 visualized using an anti HA tag antibody (1:400, rabbit, C ell S ignaling Technology ) and VECTASTAIN Elite ABC kit (Vect or Lab) according to Briefly, the paraffin sections were deparaffinized in xylene and hydrated through an ethanol series and water. After heat induced epitope retrieval (DAKO), pre blocking w as carried out by treatment with Per oxo Block (Invitrogen) for 45sec onds, goat serum for 20min and with the A v idin/ B iotin B locking K it (Vector Lab) sequentially, according to After overnight incubation with the anti HA tag antibody, the sections were incubated f or 30min with a biotinylated secondary antibody then for 30min with ABC reagent. Finally HA tagged Ant4 w as visualiz e d by the perxosidase d iaminobenzidine (DAB, Vector Lab) reaction. A fter rins ing in water, the sections were mounted and observed under the microscope (Olympus IX70). All staining steps were performed at room temperature (RT) except for the incubation with HA tag antibody at 4 o C GFP Expression The extracted testes were fixed overnight in 4% paraformaldehyde in PBS. T he next day, the tissues were washed three times with PBS and incubated at 4 o C in 30% sucrose in PBS. After overnight incubation, the tissues were embedded in Optimal Cutting Temperature (OCT, Tissue Tek) compound on dry ice cut into 10um slice s, and placed on glass slide s The slides were rinsed with PBS and stained with DAPI ( 4',6 diamidino 2 phenylindole ) for 10 min. After rinsing with PBS, the slides were covered with cover glass and anti fading mounting medium (Vector labs) and observed under the microscope (Olympus IX70).
55 Galactosidase Staining Postnatal testes were collected at P7, P15, P21 and P68 and fixed overnight in 0.2% paraformaldehyde in PBS at 4 o C The next day, the fixed testes were washed 3times for 10min with concentrated rinse buffer (0.1M sodium phosphate [ pH 7.4], 0.1% sodium deoxycholate, 2mM MgCl 2 0.2% NP 40) Samples were develop ed overnight without light in staining solution (1mg/ml X Gal in DMF, 5mM K 3 Fe(CN) 6 5mM K 4 Fe(CN) 6 in concentrated rinse buffer). Samples were then washed 3 times in PBS for 10 min and placed in 4% paraformaldehyde in PBS for 1hr to stop the staining reaction. After the fixation, the testes were stored in PBS at RT until further process. RNA Isolation Total RNA was isolated from a variety of murine tissues as shown in Figure 3 1 B Genomic DNA contamination during the RNA isolation was removed using the TURBO DNA free TM Kit (Ambion). Complementary cDNA was synthesized from the total mRNA by reverse trans criptase and random primers using the High Capacity cDNA Reverse Transcription Kit (Applied Biosystems) under the following thermocycler conditions: 25 o C for 10min, 37 o C for 120min, and 85 o C for 5sec. qRT PCR The qRT PCR was performed using Taqman Gene E xpression Master Mix for the Ant4 and actin expressions and Power SYBR Green PCR Master Mix for the Sycp3 expression (Applied Biosystems) under the following conditions: 95 o C for 10min and 40cycles at 95 o C for 15sec and 60 o C for 1 min. The levels of trans cript s in each tissue were normalized to the housekeeping gene, actin and relatively quantified by
56 the 2 Ct method. Template equivalent to 2.5ng of total RNA was used for the amplification and each reaction was performed in triplicate. The primer sets used were: An 4 :Taqman probe (Mm00617754_m1, Applied Biosyste ms), Sycp3 : GCAGCAGTGGGAACTGGATA and TTCATTCTCTGGCTCTGAACAA actin : Taqman probe (4352341E, Applied Biosystems ) RT PCR cDNA equivalent to 2.5ng of total RNA was used for the PCR (5 Prime Taq DNA polymerase, 5 Prime) under the following conditions: 95 o C for 2 min and 30 35 cycles at 95 o C for 30 sec 55 o C for 30sec and 72 o C for 45sec. For electrophoresis, 2% agarose gel was used. The primer sets used were: GFP : 5 TGGCTCTCCTCAAGCGTATT and 5 G AACTTCAGGGTCAGCTTGC Actin : ATGGATGACGATATCGCTG and ATGAGGTAGTCTGTCAGGT Western Blotting Protein was extracted using standard RIPA buffer (50mM Tris Cl [pH 8], 150mM NaCl, 1% NP 40, 0.5% Na deoxycholate, 0. 1% SDS)with proteinase inhibitor cocktail ( S igma ) and denatured at 95 o C for 10min. After electrophoresis immunoblotting was performed with anti Ant4 (1:5000, self generated ) and anti Actb (1:5000, S anta C ruz Biotechnology Inc. ). Then, the Ant4 band was normalized to the Actin band in an exposed fil m using GeneTool software (SynGene). Results BAC Transgenic Ant4 Promoter Achieves Comparable Gene Expression Specificity t o Endogenous Ant4 i n Mice In order to test if the Ant4 promoter can provide meiosis specific expression of transgenes, we initially introduced a GFP reporter gene under the promoter of Ant4 in a
57 BAC construct. The structure of a long BAC transgenic (tg) construct(~168kbps) including the upstream promoter region (~ 50 kbps)of the Ant4 gene is illustrated in Figure 3 1A. An IRES GFP casset te was inserted between Ant4 cDNA and SV40 polyadenylation signal sequences to trace the transgene expression. After pronuclear injection of the BAC Ant4 IRES GFP tg construct, one male and one female founder were obtained with an overall success rate of 1 8% (2 out of 11) in generating transgenic mice but the BAC transgene was not germ line transmissible in the female founder. Thus, the male was used to expand the BAC Ant4 IRES GFP transgenic line. We and others previously demonstrated that Ant4 is exclusi vely expressed in testis [5,19] W ithin the testis, Ant4 transcript level s increase through leptotene and zygotene meiosis stages of spermatocytes reach a peak in early pachytene spermatocytes and then decrease in late pachytene spermatocytes and round spermatids Expression is very low in Sertoli cells  In Figure 3 1b, we confirmed that expression of Ant4 transcript was only detected in testis but not in other somatic tissues as seen with a nother meiosis specific gene Sycp3 a structural component of synaptonemal complex. Consistently, the BAC Ant4 promoter driven GFP transcript was found by RT PCR only in testis of BAC Ant4 IRES GFP tg mice ( Figure 3 1C). Further, GFP expression was examine d in cryosectioned samples of eight different tissues As shown in Figure 3 2, only testis of the transgenic mice presented a significant increase in GFP expression beyond wild type controls. It should be noted that interstitial tissues between seminiferou s tubules showed high autofluorescence in both wild type and transgenic testes As shown in Figure 3 3, the increase in GFP expression in testis of the transgenic animal was apparent at lowest magnification (x4), where it was
58 expressed in seminiferous tubu les but not in the vessels of transgenic testes. Higher magnification images revealed the GFP was expressed highly in spermatocytes and spermatids whereas spermatogonia expressed GFP at a low level, indicating that the GFP expression pattern shown in BAC Ant4 tg testes is consistent with endogenous Ant4 expression we described previous ly  ( Figure 3 3). In mice male meiosis is initiated around P10 and preleptotene primary spermatocytes are differentiated from the type B spermatogoni a  As expected, GFP was not detected in type A intermediate and type B spermatogonia at P7 during mitotic proliferation Expression was detected in primary spermatocytes located in the center of seminiferous tubules at P15 after t he onset of meiosis in testes. By P21, GFP expression extended to newly formed spermatocytes along with developing testes. A fter sexual maturity, fully developed testes (P63) maintained high GFP expression in spermatocytes and spermatids and low expression in spermatogonia. To examine the similar ity of GFP reporter expression to endogenous Ant4 protein expression during development of postnatal testes, galactosidase staining was performed in age matched postnatal testes of heterozygous Ant4 knockout male s carrying one copy of the galactosidase expression cassette  We found BAC Ant4 promoter driven GFP expression was expressed in a similar manner to endogenous Ant4 promoter driven galactosidase expression ( Figure 3 4). Thus, we conclude the pattern and timing of gene expression driven by BAC Ant4 and endogenous Ant4 promoters is consistent during postnatal testis maturation.
59 BAC Transgenic Ant4 Promoter Achieves Comparable Gene Expression Levels t o Endogenous Ant4 i n Mice In the BAC DNA construct described above the Ant4 coding sequence (cDNA) was also inserted under the translation initiation site (ATG) of the BAC Ant4 gene to assess the level of transgene transcript and protein expression produced from the BAC Ant4 promoter co mpared to the endogenous Ant4 promoter ( Figure 3 1A). T ranscript and protein levels were measured by quantitative RT PCR (qRT PCR) and western blotting, respectively. Since primers for qRT PCR and the antibody for western blotting detect both tg and endoge nous Ant4 gene expression level s regulated by BAC Ant4 and endogenous Ant4 promoters can be directly compared. At the transcript level, total Ant4 levels were a pproximately three fold higher in tg testis when compared to the level of Ant4 in a non transgen ic control which is heterozygous for the Ant4 gene ( Figure 3 5A). We a ls o estimated using q uantitative PCR of genomic DNA that one copy of the BAC Ant4 tg construct was likely integrated in the host genome (data not shown). Th i s implies that in the transge nic line, transcription efficiency is approximately 2 fold higher for tg Ant4 compared to endogenous Ant4 expressed by one allele Ant4 protein level s were quantitated by immunoblot analysis using serially diluted testicular lysates ( Figure 3 5B). T he dens ity of protein bands were also measured using GeneTool software and Ant4 protein was normalized to actin. When protein lysates were diluted sufficiently, an approximately t wo fold increase in protein level was seen in tg testes compared to Ant4 heterozyg ous testes although no difference was detected when higher amounts of total protein were loaded Thus, in th is particular transgenic line, tg Ant4 protein was expressed at approximately the same level as that expressed by one allele of endogenous Ant4 gen e.
60 HA t agged Transgene Ant4 Protein Shows More Precise Expression Pattern Of BAC Transgenic Ant4 Promoter In order to further examine the tg Ant4 expression pattern, we generated an additional BAC tg construct using ~138kbp BAC DNA. In th is construct, the h uman influenza hemagglutinin (HA) sequence was inserted between the 5 UTR and the Ant4 coding sequence which allowed us to trace Ant4 expression by immunohistochemistry using an anti HA antibody ( Figure 3 6A). After pronuclear injection with the BAC HA A nt4 tg construct, one male and two females were identified as transgenic founders with a 20% success rate (3 out of 15). T he male descendant s from two female founders were used for immunohistochemical analysis because heterogeneous transgene expression was observed in the male founder line. During spermatogenesis, HA tagged Ant4 expression was not detected or detected at a very low level in spermatogonia. The expression level was the highest in spermatocytes and similar or slightly less in round spermatids. Further, comparable to wt Ant4 protein, we observed that HA tagged Ant4 protein localized in the spermatozoon tail midpiece where mitochondria reside (arrow in the middle of Figure 3 6B). In addition, tg Ant4 expression was absent in Sertoli cells (inset in the bottom of Figure 3 6B) and interstitial tissues including Leydig cells although nonspecific binding of the antibody was detected in the interstitial tissues ( Figure 3 6B). Transgenic Ant4 Localizes i n Mitochondria o f Male Germ Cells. Since Ant4 is a mitochondrial inner membrane protein, we utilized immun ostaining to examine whether tg Ant4 localizes i n mitochondria of male germ cells from tg Ant4 +; Ant4 +/ mice generated by mating of BAC HA Ant4 tg mice with Ant4 ko mice. Immunohistochemistry with an anti HA tag antibody showed punctate
61 staining patterns similar to other mitochondrial proteins in spermatogenic cells of the seminiferous tubules (arrows in right panel of Figure 3 7A). For further confirmation, sperm were collected from cauda epidid y mis a nd a part of the ductus deferen s and were stained with anti Ant4 and anti HA tag antibodies. As expected, Ant4 was observed in the midpiece of sperm wh ere mitochondria reside (top in Figure 3 7B) and tg Ant4 was similarly localized These data suggests tha t tg Ant4 localizes in mitochondria of male germ cells comparable to engodenous Ant4 BAC Ant4 Tg Mice Do Not Restore t he Impaire d Spermatogenesis When Crossed w ith Ant4 / Mice In order to examine if tg Ant4 expression can recover impaired spermatogenesis in Ant4 null testis, tg Ant4 +; Ant 4 / mice were obtained after mating of BAC Ant4 IRES GFP tg mice with Ant4 ko mice. Spermatogenesis was assessed by hematoxylin and eosin staining in testes from mice with four different genotypes : tg Ant4 ; Ant4 +/ tg Ant 4 ; Ant4 / tg Ant4 +; Ant4 +/ and tg Ant4 +; Ant4 / Tg Ant4 expression did not disturb spermatogenesis in Ant4 +/ testis (left panel in Figure 3 8). However, tg Ant4 +; Ant4 / testis still displayed abnormal spermatogenesis (right panel in Figure 3 8) altho ugh tg Ant4 expression level was sufficient to restore the defect in spermatogenesis ( Figure 3 5). As seen in tg Ant4 ; Ant4 / testis, spermatogonial stem cells which reside in the outermost layers of seminiferous tubules and some spermatocytes were found but round and elongated spermatids and spermatozoa were not observed in tg Ant4 +; Ant4 / testis, indicating that tg Ant4 expression cannot rescue the impaired spermatogenesis of Ant4 / testis. These data were confirmed in tg Ant4 +; Ant4 / testis from BAC HA Ant4 tg mice (data not shown).
62 Discussion In the present study, BAC Ant transgenic mice were generated to examine whether Ant2 can compensate for the loss of Ant4 in testis. BAC Ant4 tg mice were used as a control. Transgene Ant4 expression was analogo us to endogenous Ant4 expression in pattern, level, and mitochondrial localization. However, unexpectedly, tg BAC Ant4+; Ant4 / testes generated from mating BAC Ant4 tg mice with Ant4 ko mice still showed impaired spermatogenesis. This result was addition ally confirmed in homozygote tg BAC Ant4+; Ant4 / testes where tg Ant4 expression level is expected two fold higher than hemizygote tg BAC Ant4+; Ant4 / testes (data not shown). Thus, transgene expression may not be the cause of this issue. This suggest s that impaired spermatogenesis in Ant4 / testes may be affected by factors other than Ant4 deletion alone One possible factor is downregulation of downstream genes in targeted Ant4 locus. Previously, unexpected downregulation of downstream gene s in tar geted ko studies have been described. In one such case, Braun and Arnold showed that expression of Myf 5 a downstream gene of Myf 6 decreased in homozygous Myf 6 mutant mice  In the study, the nucleotides 5 to +207 relative to the tra nscriptional initiation site of Myf 6 were replaced by the pgk neo cassette in the sense orientation, resulting in the disruption of the Myf 6 gene expression. Unexpectedly, the Myf 5 gene was not detected by Northern blot analysis in Myf 6 null skeletal m uscle. Therefore, a severe rib defect in Myf 6 null mice was most likely due to the drastically reduced expression of Myf 5 which has been previously reported  The authors suggested two possib le explanations for the issue. First, a cis regulatory element for Myf 5 expression is affected by the insertion of the pgk neo cassette into the Myf 6 locus.
63 Second, Myf 5 expression may be regulated by Myf 6 directly, which acts as a transcription factor In a followup study, it was proved using Myf 5+/ ; Myf 6+/ mutant mice that the former was to the cause of this issue  Similar to the Myf 6 case, we found Hspa4l ( Heat shock 7 0 kDa protein 4L) a downstream gene of Ant4 expressi on was decreased 50% in the kidney of homozygous Ant4 null mi ce compared to wild type mice. Expression could not be examined in Ant4 / testes due to the lack of spermatogenic cells except for spermatogonia and some primary spermatocytes. Hspa4l which belo ngs to the HSP110 heat shock gene family is expressed ubiquitously but predominantly in testis. Interestingly, the protein is highly expressed from late pachytene spermatocytes to spermatids like Ant4 In addition to the expression pattern, the deletion of the gene caused male infertility in approximately 42% of homozygous male mutants  This phenotype is similar our Ant4 ko mouse phenotype although Hspa4l ko mice ha ve a m ilder phenotype (100% infertility in Ant4 / male mice). This suggest s impaired spermatogenesis in Ant4 null testes may be affected by the reduced expression of Hspa4l In our current study, the long BAC DNA included two downstream genes, Hspa4l and plk4 However, tg Ant4 expression with the expression of the two downstream genes did not restore impaired spermatogenesis of Ant4 ko testes even though transgene expression did not disrupt normal spermatogenesis ( Figure 3 3). Thus, it remains to be clarified if further downstream gene expression in Ant4 locus beyond Hspa4l and plk4 contributes to the Ant4 ko phenotype. The other possibility is the effect of non coding RNAs including long non coding RNAs and small RNAs such as microRNAs and piRNAs during spermat ogenesis 
64 Testis speci fic miRNAs have been identified in different developmental stages of germ cells using RT PCR, microarray or small RNA sequencing. During spermatogenesis, miR 18 which belongs to the miR 17~92 cluster is highly expressed in spermatocytes and represses heat shock factor 2 ( HSF2 ), a transcription factor, involved in embryogenesis and gametogenesis  miR 425 and miR 1 91 are highly expressed in testis and downregulated in severe teratozoospermia  Not surprisingly, the knockout of Dicer 1 an RNase III endonuclease required for the cleavage of pre miRNA to generate miRNA, by Stra8 Cre in postnatal testes showed increased leptotene and zygotene spermatocytes and decreased spermatocytes in pachytene, diplotene and metaphase I stages, indicating early meiotic arrest during spermatogenesis. As a result, sperm number was dramatically reduced by 94%, resulting in male infertility  To generate Ant4 ko mice, exons 2, 3, and 4 of Ant4 including surrounding introns were replaced by IRES LacZ and pgk neo cassettes. Although the removed regions may include potential miRNAs that are essential for spermatogenesis, BAC DNAs used in this study entirely c over the deleted areas, suggesting that miRNA expression is not likely altered except for mirtron. M irtron is located in introns of mRNA and generated by splicing of the introns  Ant4 cDNA with SV40 p A signal sequences was inserted Ant4 in BAC DNA. In addition, splicing donor sequences in order to prevent the generation of a potential fusi on protein. Thus, transcription does not occur beyond the pA sequences. Moreover, even though transcription occurs at a low level beyond the pA signal, splicing does not occur properly due to the removal of the splicing donor sequence. In sum, if there are mirtrons in intronic regions of Ant4 that
65 are essential for spermatogenesis, BAC DNA cannot recover expression of the mirtron s This should be further elucidated in the future. Piwi interacting RNAs (piRNAs) are single stranded RNAs of 24 31 nucleotides which are predominantly expressed in testis. Piwi (P element induced wimpy testis) which belongs to the argonaute family is involved in gene silencing of retrotransposons triggered by the association with piRNAs  In the germ line of Drosophila the complex of piRNAs and Piwi re press target RNAs by cleavage activity  Miwi, a murine homologue of P iwi, has been previously demonstrated to play an essential role in spermatogenesis  Homozygous M iwi mut ants show early spermatogenic arrest with increased apoptotic cells, leading to a complete lack of elongating spermatids and resulting male infertility. piRNA expression is significantly reduced in murine M iwi / testes at 24 d ay postpartum, implying an es sential role of piRNAs association with M iwi for spermatogenesis  Thus, piRNA may be disrupted in the Ant4 null allele and could affect the phenotype of Ant4 ko mice although there is no known piRNA in Ant4 locus (http://pirnabank.ibab.ac.in/). To clarify the issues mentioned above, a knockout mouse can be generated with minimal point mutations in Ant4 exons to circumvent the effect of potential factors which may influ ence Ant4 knockout phenotype. Regardless of the unexpected failure to rescue the Ant4 ko phenotype by BAC Ant4 tg mice, transgene Ant4 expression using the BAC DNA approach recapitulated the temporal and spatial expression of endogenous Ant4 in spermatogen ic cells, indicating the Ant4 promoter is useful to achieve the expression of recombinases such as Cre recombinase or reporter genes such as GFP during male meiosis.
66 For successful recombination in vivo Cre recombinase must be expressed specifically in a cell type of interest at a sufficient level to be able to induce recombination. Cre expressing transgenic mice have been generated in two ways. The first is to use a conventional transgenic vector with pronuclear injection  The transgenic vector is generated by placing c re recombinase sequence downstream of a c ell type specific promoter prior to injection into the pronucleus of embryos. This approach is advantageous because the transge nic vector is relatively simple and easily constructed and transgenic mice are generated faster. However, the transgene may not b e properly expressed due to the use of a short promoter and the random insertion into a potentially heterochr oma tic region of the mouse genome. Thus, this approach may not guarantee proper regulation of transgene expression and as a result requires laborio us work to screen for appropriate transgenic founders. Second, transgenic mice can be generated by knockin approach using targeted ES cells. In this case, a transgene is inserted downstream of a promoter of interest in ES cells using homologous recombinati on. This method promises transgene expression will be comparable to the expression of the endogenous gene due to regulation by the promoter. However, the loss of one allele of the endogenous gene may result in an unexpected phenotype due to haploinsufficie ncy. In addition, it is tedious and laborious to obtain correctly targeted ES cells because of low targeting efficiency. An alternative approach is to use BAC DNA, which can be easily manipulated i n engineered E. coli by the recombineering (recombination m ediated genetic engineering) technique with high targeting efficiency. BAC transgenesis promises an endogenous gene expression of a transgene while overcoming potential transgene repression due to random insertion. In addition, levels
67 of transgene expressi on can be controlled by adjusting BAC DNA copy number because BAC transgene expression is proportional to BAC transgene copy number  Therefore, the BAC approach for transgenesis is a powerful alternative. Several cre mice have been generated to target spermatocytes. Synaptonemal Complex Protein 1 (Sycp1) is expressed from leptotene to early pachytene s tages of male meiosis. Previou sl y, a Sycp1 cr e of a promoter enhancer complex  Unexpectedly, Cre loxP mediated recombination was inhibited by the methylation of the cytosine residues of the loxP sites when the mice carrying Sycp1 cre rec ombinase we re mated with Rosa26 flox/flox mice in the second generation  The cause of this phenomenon remains to be answered although it might be due to cre recombinase expression in meiotic cells as the authors mentioned. To circumvent this issue, the authors suggested the Sycp1 cre transgene can be carried in female double transgenic mice ( Sycp1 cre ; Rosa26 flox/folx ) throughout generation because the partial region of promoter enhancer complex of Sycp1 used in t he transgenic mice cannot induce cre expression during meiosis in females. However, it is questionable whether the Sycp1 cre mice are a good candidate for meiosis specific deletion of a gene of interest Phosphoglycerate kinase 2 (pgk2) cre mice has been r eported and are a potential alternative to Sycp1 cre Pgk2 catalyzes the conversion of 1, 3 bisphosphoglycerate to 3 phosphoglycerate in the glycolytic pathway and is specifically expressed in testis. Pgk2 transcripts are first detected in preleptotene spe rmatocytes and expression continues to increase in spermatids  To generate a pgk2 cre transgenic line, a 1.4 kbps DNA fragment including the murine pgk2 promo ter was used  Unpredictably, galactosidase expression was very heterogeneous
68 among seminiferous tubules in testes carrying pgk2 cre and reporter (CAG CAT LacZ ) alleles, resulting in galactosidase expression only in 70% of spermatogenic cells. Although the authors argued Cre/loxP recombination occurs only in a part of spermatogenic cells for an unknown reason, c re recombinase expression should be thoroughly ex amined in male germ cells at each developmental stage during spermatogenesis because the 1.4kbps pgk2 promoter fragment may not be sufficient to produce cre expression comparable to endogenous pgk2 expression. Further, another pgk2 cre transgenic mouse gen erated with a 450bp pgk2 promoter fragment showed ectopic cre expression in embryonic tissues as well as adult non meiotic tissues including skeletal muscle, brain and heart  implying that the short pgk2 promoter fragment could not correctly regulate the expression of transgenic cre recombinase like endogenous pgk2 Testicular heat shock protein 2 ( Hspa2 ) was also used to develop a cre line to target spermatocytes  However, cre expression by a 907 bp Hspa 2 promoter fragment was observed in brain while the cre recombinase was predominantly detected in spermatocytes and spermatids. Therefore, the use of Hspa2 cre mice is limited when a targeted gene is expressed in testes as well as brain. The other t wo cre lines to target spermatocytes c kit cre  and Synapsin1 cre  have limitations in their utility because c kit and Synapsin 1 also are expressed in embryonic tissues and neurons, respectively. Therefore, Ant4 with BAC transgenesis approach is a promising strategy to target s permatogenic cells. Ant4 expression is limited to testicular germ cells and BAC DNA circumvents the problems of the previously reported cre lines for spermatogenic cells fa c e d as described above.
69 Figure 3 1. Generation of BAC Ant4 IRES GFP transgenic mi ce (A) Strategy to target BAC DNA (B) Ant4 transcript expression by qRT PCR (C) GFP expression by RT PCR. 3T3 with GFP and without GFP were used as positive and negative controls, respectively. [Photos courtesy of Chae Ho Lim]
70 Figure 3 2. Testis spec ific GFP expression regula ted by the Ant4 promoter H/E: Hem a toxylin and eosin staining. [Photos courtesy of Chae Ho Lim]
71 Figure 3 3. Meiosis specific GFP expression in spermatogenic cells [Photos courtesy of Chae Ho Lim]
72 Figure 3 4. Timely expres sion of a transgene in postnatal testes [Photos courtesy of Chae Ho Lim]
73 Figure 3 5. Transgene expression level in testes (A) Tg Ant4 transcript expression by qRT PCR (B) Tg Ant4 protein expression by Western blot analysis
74 Figure 3 6. HA tagged transgene expression by the Ant4 promoter (A) Strategy to target BAC DNA (B) Tg Ant4 expression detected by anti HA antibody in testis [Photos courtesy of Chae Ho Lim]
75 Figure 3 7. Transgenic Ant4 localizes in mitochondria during spermatogenesis. (A) Immunohistochemistry with an anti HA antibody in testis (B) Immunocytochemistry with anti Ant4 and anti HA antibodies in sperm. [Photos courtesy of Chae Ho Lim]
76 Figure 3 8. Tg Ant4 expression does not restore impaired spermatogenesis in Ant4 null testi s H/E staining was performed. [Photos courtesy of Chae Ho Lim]
77 CHAPTER 4 ANT4 IS EXPRESSED IN MEIOTIC FETAL OVARY Background In mammals, meiosis for oogenesis and spermatogenesis begin at different life stages. In mice, primordial germ cells reach the go nards at ~11dpc  In male mice, the germ cells arrest in mito sis at 14dpc. After birth, mitosis resumes and t ype A spermatogonia begin to appear at P3 7 and differentiate into intermediate spermatogonia. T he intermediate sperm atogonia develop into type B spermatogonia which differentiate into primary spermatocytes. The primary spermatocytes enter meiosis I at P10  and continue to generate further differentiated lineages such as spermatids and spermatozoa Once spermatogenesis initiate s it continues to generate male germ cells throughout life. In female mice, oogonia differentiate into primary oocytes by mitosis and the oocytes enter meiotic prophase I at approximately 13.5dpc  The primary oocytes undergo leptotene, zygotene and pachytene stages with extensive apoptosis then arrest at the diplotene stage  The primary oocytes remain in a dormant state until they reach puberty. After birth, each oocyte is enclosed by somatic granulosa cells, which provide metabolic and signaling molecules to the oocytes to form primordial follicles T he primordial follicles begin to grow and matur e with increased size until puberty, resulting in antral follicles. During any one menstrual cycle, selected oocytes reenter the final stage of meiosis I and complete the first meiotic division with formation of the first polar body prior to ovulation. After ovulat ion the oocytes arrest in metaphase II of meio sis II. Once fertiliz ed the second meiotic division is completed with formation of the second polar body.
78 M ammalian female primordial germ cells have two X chromosomes whereas male primordial germ cells have heterogametous X and Y chromosomes. During male meiosis, sex chromosomes (XY) are subject to meiotic sex chromosome inactivation (MCSI), which may originate from an ancient defensive mechanism o f the genome against invading viruses or transposons  Unpaired XY chromosomes are repressed because they lack a pa ired counterpart during meiosis but paired XX chromosomes can escape from the MCSI  Thus, X linked genes including Ant2 can be repressed during male meiosis but are expressed during female meiosis. Ant4 is exclusively expressed in te stis with high est expression in primary spermatocytes that undergo prophase I of meiosis I  However, we were interested to examine whether Ant4 is also expressed during female meiosis. In the ovary, prophase I of meiosis I occurs du ring fetal development as described above. Thus, we examined Ant4 expression in fetal ovary Our previous study demonst r ated that there is no Ant4 expression in adult ovary  In addition when Ant4 is expressed during female meiosis Ant4 / female mice may have compromised fertility In the present study, we examined Ant4 and Ant2 expression in fetal ovaries at various time points. In addition, the fertility and ovary histology of female Ant4 / mice were closely examined Materials a nd Methods qRT PCR and Hematoxylin and Eosin staining were performed as described in C hapter 3. To examine the morphology of adult ovaries, they were extracted from various ranges ages of Ant4 +/ and Ant4 / females (7 13 months old). After fix ation in 4% paraformaldehyde in PBS, the ovaries were weighed and t aken images with a ruler for scale. To analyze offspring number Ant4 +/ and Ant4 / females were mated with
79 Ant4 +/ males, respectively. The number of offspring from 22 Ant4 +/ and 17 Ant4 / females was counted and recorded. Results Ant4 a nd Ant2 a re Expressed i n Fetal Ovary. To investigate if Ant4 and Ant2 are expressed in fetal ovary, murine ovaries were collected at different embryonic days (E13.5d, E14.5d, E15.5d, E16.5d, E17.5d and E18.5d) Tran script levels of each gene were detected by qRT PCR ( Figure 4 1). Ant4 transcripts were found in the fetal ovaries over the selected time points except for 14 w ee k old adult ovary. 14 w ee k old testis was used as a positive control for Ant4 expression. In t estis, Ant4 expression is high in spermatocytes undergoing meiosis ( Figure 3 6B). Like the expression pattern in testis, in ovary, Ant4 was detected during embryonic development when meiotic prophase I occurs as seen with Sycp3 expression whereas adult ov ary did not express Ant4 This indicat es Ant4 is specifically expressed during meiosis in both female and male reproductive organs. Since the h omogametic sex chromosome (XX) in female m ice can escape meiotic sex chromosome inactivation  we expected X linked Ant2 gene express ion in fetal ovary. Indeed, Ant2 expression was found in the fetal and adult ovary (14 wk old) regardless of meiotic stages ( Figure 4 1). 14 wk old liver was used as a positive control for the Ant2 expression. Female Ant4 / Mice a re Fertile w ith Normal Ovary Mor phology In order to examine if Ant4 / ovary is morphologically abnormal adult ovaries (7 13 months old) were extracted from Ant4 +/ and Ant4 / females. We found the ovaries examined were not different in morphology or size between Ant 4+/ and Ant4 / mice regardless of age even though there was a difference in size among individuals ( Figure
80 4 2A). Weight of ovaries also was similar between two groups (Mean value: 4.21mg for Ant4 +/ and 4.31mg for Ant4 / p=0.84) ( Figure 4 2B).For further examination of ovary structure, H&E staining was carried out with extracted ovaries (12 months old). Developing follicles with normal morphology were found in both genotypes (arrows) and no clear abnormalities were observed in the follicles or surrounding tissues ( Figure 4 3). In addition, t he number of offspring from 22 Ant4 +/ and 17 Ant4 / females was counted. Mean value was 5.7 for Ant4 +/ and 4.5 for Ant4 / (p=0.01) ( Figure 4 4 ). On average, Ant4 +/ female s ha d one additional offspring compared to Ant4 / female s Discussion This stud y demonstrated that Ant4 / ovary is normal in morphology, size, and weight. However, it should be noted that average litter size was slightly smaller in Ant4 / female s compared to Ant4 +/ female s implying that Ant4 may play a non essential but somewhat beneficial role in female fertility as well Nonetheless, Ant4 / female s are fertile unlike Ant4 / male s One possible explanation is that Ant2 compensate s for the loss of Ant4 during female meiosis Gene expression profile s showed Ant4 expression is lim ited to the meiotic fetal ovaries while Ant2 is expressed in the fetal and adult ovaries ( Figure 4 1). Thus, unlike testis where Ant2 is repressed by MSCI, Ant2 expression may be sufficient to compensate for the absence of Ant4 in fetal ovary. The other po ssibility is that Ant activity may not be required for embryonic ovary development. To clarify these issues, Ant2 conditional knockout mice were generated by Dr. S. Paul Oh at UF by targeting exon 2 and 3 of Ant2. T he se mice are currently being mat ed with Ant4 ko mice to generate Ant2/4 double ko mice for future study We will utilize a Stra8 cre mouse for germ cell specific deletion of Ant2 during and after meiosis The fetal Ant2/4 double null mutant ovaries will be examined for morphology
81 size and cell death to determine if the developmental process is normal in fetal ovaries lacking both Ant2 and Ant4 expression. Further, it will be interesting to investigate if Ant2 is required for oogenesis both in embryonic and adult stage s The X chromosome contai ns multiple genes encoding metabolic enzyme s that are essential for spermatogenesis. Thus, male germ cells need compensatory system s to compensate for X linked gene silencing by MSCI. One of such compensatory mechanisms is through retrogene s, which are for med by a retrotransposition process  After the progenitor gene of a retrogene is transcribed and spliced to form mature RNA, the mature RNA is reverse transcribed by reverse transcriptase. T he product of such revers e transcription is integrated into the genome, specifically into autosome s in the case of X linked gene s Thus, in general, retrogene s lack introns unlike the ir progenitor gene s although some retrogenes can retain an intron from partially processed mRNA or aquire an intron after retrotransposition. The phosphoglycerate kinase family ( Pgk1 and Pgk2 ) is a typical example of a retrogene  Pgk is a glycolytic enzyme that convert s 1, 3 diphosphoglycerate to 3 phosphoglycerat e. X linked Pgk1 is constitutively expressed in somatic cells and premeiotic germ cells while Pgk2 is an autosomal linked gene which is exclusively expressed in meiotic germ cells during spermatogenesis. The Pgk2 gene lacks introns unlike Pgk1 contains 10 introns. Thus, Pgk2 is thought to be a retrogene of Pgk1 So far, 20 retrogen es derived from 16 X linked progenitor genes have been identified in human s and m ice Interestingly, 14 of these are specifically expressed in testis  However, it is unlikely that autosomal Ant4 is a retrogene of X linked Ant 2 because Ant4 has 6exons with su rrounding introns whereas Ant2 has 4exons. Nonetheless, if exchanging cytoplasmic ADP for
82 mitochondrial ATP is essential for meiosis, Ant2 or Ant4 alone may be sufficient to accomplish the meiosis although Ant2 ma y not be optimal. We still question why Ant4 is conserved in mammals and anole lizard which is a re p tile What function of Ant4 is required for germ cells? The one clue com es from the differences between male and female germ cells. Male germ cells, specif ically spermatozoa are motile unlike female germ cells suggesting Ant4 may be specialized to provide energy for motility in spermato z oa that contain densely packed mitochondria. Aside from general sperm motility, Ant4 may be specialized for hyperactivated sperm motility. Hyperactivated sperm motility is characterized by the increased amplitude of the flagellar bend of sperm with highly asymmetrical beating pattern which is required to penetrate the zona pellucida of an ovum. This will be discussed in furthe r detail in C hapter 5.
83 Figure 4 1. Ant4 is expressed in fetal ovaries Ant 4, 2 and Sycp3 gene expression was examined in fetal and adult ovaries by qRT PCR. Each mRNA amounts were normalized based upon the expression of the actin mRNA. Error b ars indicate standard deviations of triplicate samples
84 Figure 4 2. Ant4 / ovary is morphologically norma l. (A) Gross morphology of fetal ovaries from Ant4 +/ and Ant 4 / females. (B) Weight of adult ovaries shown in Figure 4 2A (p=0.84). [Photos cour tesy of Chae Ho Lim]
85 Figure 4 3 Ant4 / ovary is histolo gically norma l. Hematoxylin and eosin staining (X4). Arrows indicates developing follicles. [Photos courtesy of Chae Ho Lim]
86 Figure 4 4 The number of offsprings from Ant4+/ and Ant4 / fema les. Asterisk indicate s p<0.05 in t test (p=0.01).
87 CHAPTER 5 CONCLUSIONS AND DISCUSSION Prior to this work, Ant4 was thought to be conserved only in mammals. Through close examination of an updated genomic sequencing database, we found that the reptilian anole lizard, which utilizes the heterogametic sex (XY) determination system has the Ant4 ortholog Exon intron structures and testis specific expression of anole Ant4 were analogous to those of human and murine Ant4 Previously, we thought that autosomal Ant4 acts as a compensatory Ant for the loss of X linked Ant2 gene ex pression during male meiosis due to meiotic sex chromosome inactivation (MSCI). However, g ene dosage analysis with genomic DNA and bisulfite sequencing patterns indicated that anole Ant2 is not locate d o n the X chromosome. Interestingly, even though autosomal Ant2 is free from MSCI, Ant4 was still highly expressed in anole testis. This implies Ant4 may not simply be conserved as a compensatory gene for the Ant2 silencing during male meios is. Instead, Ant4 may have a specialized function to facilitate spermatogenesis and/or sperm functions in these species To examine the hypothesis that Ant4 is a specialized Ant for male germ cells, Ant2 or Ant4 was expressed in Ant4 null testes using a BA C transgenic approach to see if Ant2 could restore the Ant4 null phenotype. Unexpectedly, Ant4 expression in the Ant 4 / testes did not recover impaired spermatogenesis even though transgene Ant4 expression pattern and level seemed comparable to the endoge nous Ant4 gene Thus, it is likely additional unknown factors such as effects on downstream gene s or small non coding RNA depletion may be involved in the Ant4 ko mouse phenotype beyond simple deletion of exons and intron of the Ant4 gene, may contribute t o the Ant4 ko mouse phenotype. Finally, we closely examined the phenotype of homozygous Ant4 ko ovar ies Ant4 was
88 detected in fetal ovaries but not in adult ovary while Ant2 was expressed in both fetal and adult ovaries. Unlike male Ant4 null testis, there were no differences in gross morphology, size, and weight of adult ovaries from Ant4 +/ and Ant4 / females although Ant4 / female s has a slightly smaller litter size than Ant4 + / females This suggests that Ant2 is sufficient for embryonic ovary develop ment although Ant2 may not be optimal. Base d on the apparent ly normal phenotype of Ant4 null ovary, meiosis in general may not require a specific Ant paralog I n this case what can be a specialized function of Ant4 ? Unlike oocytes and other somatic cell t ypes, s permatozoa are highly motile. Thus, i ntuitively Ant4 may play a role in sperm motility. H yperactivated sperm motility a specialized type of motility, is acquired in the oviduct of female reproductive organs and is characterized by highly asymmetri cal and increased amplitude of the flagellar bend of sperm in mammals  The asymmetrical beating pattern makes sperm swim in circles or a figure eight pattern with extreme beating whereas normal sperm swim nearly straight on glass slides. The extraordinarily active movement of sperm is proposed to be required for penetrating through thick mucus secretions in oviductal lumen and the cumulus oophorus and zona pellucida of oocytes  Without the gain of hyperactivated motility, sperm cannot pass through the highly viscoelastic environment to reach oocytes for fertilization resulting in infertili ty even though sperm are still able to move progressively Ca 2+ is a well studied inducer to trigger hyperactivation. The increase of intracellular Ca 2+ level by a calcium ionophore A23197 in boar sperm was previously use d to trigger hyperactivated motilit y in vitro  CatSper is a flagellum specific Ca 2+ channel protein to mediate Ca 2+ influx through the plasma membrane of sperm tails. Four CatSp er genes have been identified in the
89 principal piece of mammalian sperm and all of them have been reported to be required for normal and hyperactivated sperm motility [112 115] Very intriguingly, the gene conservat ion patterns of CatSper and Ant4 in a variety of vertebrate species are identical (Table 5 1). On the other hand, other sperm specific channels such as Kspe r P2RX2 and Hv1 which are not related to sperm hyperactivation are not conserved with CatSper and Ant4 This identical gene conservation pattern between CatSper and Ant4 implies Ant4 is involved in sperm hyperactivation. Indeed, hyperactivated sperm motility triggered by increased Ca 2+ level s is depend on ATP availability  and mitochondrial respiratory activity determined by oxygen consumption is remarkably high in sperm selected by the swim up method and the highly upregulated mitochondrial respiratory activity is maintained under in vitro condition mimic to i n vivo capacitation that is coincided with hyperactivation  However, it has been reported that mitochondrial oxidative phosphorylation is not required for hyperactivat ed sperm motility in the r hesus m acaque  Thus, it will be interesting to examine sperm hyperactivation with Ant inhibitors such as CATR or bon g krekic acid under a hyperact i vation inducing condition. How does Ant4 regulate sperm motility including hyperactivated spe rm motility? During the last three decades, the metabolic compartmentation theory has been intensively studied with experimental and theoretical considerations  In the classical view, metabolism occurs through a series of enzymes d ispersed freely in the intracellular aqueous space. Metabolic intermediates catalyzed by one enzyme are released in the aqueous phase and move to the next enzyme in the metabolic sequence by passive diffusion. However, living cells are filled with solutes, macromolecules such as nucleic acids and proteins including metabolic enzymes and organelles. The
90 extreme complexity of cells hinders the mobility of metabolites by diffusion. To successfully achieve metabolism, sequential metabolic enzymes interact each other or with structural proteins and membranes. The interactions of sequential metabolic enzymes facilitate metabolic channeling which directly transfer intermediate metabolites from one enzyme to an adjacent enzyme without diffusion. This metabolic comp artmentation phenomenon has been found in plant as well as animals. Particularly, glycolytic enzymes in Arabidopsis thaliana are associated dynamically with the mitochondrial outer membrane protein, VDAC to provide pyruvate for mitochondrial respiration  Since metabolic intermediates catalyzed in glycolysis are in high demand for a number of metabolic pathways including oxidative phosphorylation, the interaction between glycolytic enzymes and VDAC ensure sufficient pyruvate supply for mitochondrial oxidative phosphorylation without competing for consumption of upstream metabolites by other pathways. In addition, tricarboxylic acid cycle enzymes ha ve been reported to form a diffusible multienzyme complex i n the mitochondria of mammalian cells such as CHO and COS7 cells  Interestingly, Odet et al reported LdhC associated with Ant4 plays an essential role in the regulation of glycolys is in sperm flagellum, which is critical for providing energy for sperm motility  They proposed the glycosome model as a potential mechanism where glycosomes including LdhC and Ant4 in sperm flagellum provide ATP for sperm motilit y. However, immunostaining with anti Ant4 and anti HA tag antibodies showed Ant4 is localized to the murine sperm midpiece where mitochondria reside ( Figure 3 7). In addition, although LdhC is highly expressed in sperm flagellum it is also expressed in the sperm midpiece  Thus, I suggest that Ant4 with VDAC forms a multienzyme complex including glycolytic and
91 tricarboxylic acid cycle enzymes in sperm mitochondria to facilitate highly demanded ATP production for progressive and hype ractivated sperm motility. It will be very interesting to examine what proteins are associated with Ant4 using a pull down assay with an anti HA tag antibody in HA tagged Ant4 expressing testis.
92 Table 5 1. Ant4 and CatSper genes share an identical ortho log conservation pattern in vertebrates ( 1 to 1: ortholog identified to human gene ND: not detected PO: possible orthologue )
93 LIST OF REFERENCES 1. Klingenberg M (2008) The ADP and ATP transport in mitochondria and its carrier. Bioc him Biophys Acta 1778: 1978 2021. 2. Pebay Peyroula E, Dahout Gonzalez C, Kahn R, Trezeguet V, Lauquin GJ, et al. (2003) Structure of mitochondrial ADP/ATP carrier in complex with carboxyatractyloside. Nature 426: 39 44. 3. Stepien G, Torroni A, Chung AB, Hodge JA, Wallace DC (1992) Differential expression of adenine nucleotide translocator isoforms in mammalian tissues and during muscle cell differentiation. J Biol Chem 267: 14592 14597. 4. Levy SE, Chen YS, Graham BH, Wallace DC (2000) Expression and sequ ence analysis of the mouse adenine nucleotide translocase 1 and 2 genes. Gene 254: 57 66. 5. Rodic N, Oka M, Hamazaki T, Murawski MR, Jorgensen M, et al. (2005) DNA methylation is required for silencing of ant4, an adenine nucleotide translocase selectivel y expressed in mouse embryonic stem cells and germ cells. Stem Cells 23: 1314 1323. 6. De Marcos Lousa C, Trezeguet V, Dianoux AC, Brandolin G, Lauquin GJ (2002) The human mitochondrial ADP/ATP carriers: kinetic properties and biogenesis of wild type and m utant proteins in the yeast S. cerevisiae. Biochemistry 41: 14412 14420. 7. Dolce V, Scarcia P, Iacopetta D, Palmieri F (2005) A fourth ADP/ATP carrier isoform in man: identification, bacterial expression, functional characterization and tissue distributio n. FEBS Lett 579: 633 637. 8. Hamazaki T, Kehoe SM, Nakano T, Terada N (2006) The Grb2/Mek pathway represses Nanog in murine embryonic stem cells. Mol Cell Biol. United States. pp. 7539 7549. 9. Baines CP (2009) The molecular composition of the mitochondri al permeability transition pore. J Mol Cell Cardiol 46: 850 857. 10. Zhivotovsky B, Galluzzi L, Kepp O, Kroemer G (2009) Adenine nucleotide translocase: a component of the phylogenetically conserved cell death machinery. Cell Death Differ 16: 1419 1425. 11 Haworth RA, Hunter DR (2000) Control of the mitochondrial permeability transition pore by high affinity ADP binding at the ADP/ATP translocase in permeabilized mitochondria. J Bioenerg Biomembr 32: 91 96. 12. Clarke SJ, McStay GP, Halestrap AP (2002) San glifehrin A acts as a potent inhibitor of the mitochondrial permeability transition and reperfusion injury of the heart by binding to cyclophilin D at a different site from cyclosporin A. J Biol Chem 277: 34793 34799.
94 13. Kokoszka JE, Waymire KG, Levy SE, Sligh JE, Cai J, et al. (2004) The ADP/ATP translocator is not essential for the mitochondrial permeability transition pore. Nature. England. pp. 461 465. 14. Baines CP, Kaiser RA, Sheiko T, Craigen WJ, Molkentin JD (2007) Voltage dependent anion channels are dispensable for mitochondrial dependent cell death. Nat Cell Biol 9: 550 555. 15. Baines CP, Kaiser RA, Purcell NH, Blair NS, Osinska H, et al. (2005) Loss of cyclophilin D reveals a critical role for mitochondrial permeability transition in cell death Nature 434: 658 662. 16. Rousset S, Alves Guerra MC, Mozo J, Miroux B, Cassard Doulcier AM, et al. (2004) The biology of mitochondrial uncoupling proteins. Diabetes 53 Suppl 1: S130 135. 17. Skulachev VP (1998) Uncoupling: new approaches to an old proble m of bioenergetics. Biochim Biophys Acta 1363: 100 124. 18. Shabalina IG, Kramarova TV, Nedergaard J, Cannon B (2006) Carboxyatractyloside effects on brown fat mitochondria imply that the adenine nucleotide translocator isoforms ANT1 and ANT2 may be respon sible for basal and fatty acid induced uncoupling respectively. Biochem J 399: 405 414. 19. Kim YH, Haidl G, Schaefer M, Egner U, Mandal A, et al. (2007) Compartmentalization of a unique ADP/ATP carrier protein SFEC (Sperm Flagellar Energy Carrier, AAC4) w ith glycolytic enzymes in the fibrous sheath of the human sperm flagellar principal piece. Dev Biol 302: 463 476. 20. Hackenbrock CR (1966) Ultrastructural bases for metabolically linked mechanical activity in mitochondria. I. Reversible ultrastructural ch anges with change in metabolic steady state in isolated liver mitochondria. J Cell Biol 30: 269 297. 21. De Martino C, Floridi A, Marcante ML, Malorni W, Scorza Barcellona P, et al. (1979) Morphological, histochemical and biochemical studies on germ cell m itochondria of normal rats. Cell Tissue Res 196: 1 22. 22. Ramalho Santos J, Varum S, Amaral S, Mota PC, Sousa AP, et al. (2009) Mitochondrial functionality in reproduction: from gonads and gametes to embryos and embryonic stem cells. Hum Reprod Update 15: 553 572. 23. Boussouar F, Benahmed M (2004) Lactate and energy metabolism in male germ cells. Trends Endocrinol Metab 15: 345 350. 24. Rato L, Alves MG, Socorro S, Duarte AI, Cavaco JE, et al. (2012) Metabolic regulation is important for spermatogenesis. Nat Rev Urol 9: 330 338.
95 25. Brower JV, Rodic N, Seki T, Jorgensen M, Fliess N, et al. (2007) Evolutionarily conserved mammalian adenine nucleotide translocase 4 is essential for spermatogenesis. J Biol Chem 282: 29658 29666. 26. Brower JV, Lim CH, Jorgens en M, Oh SP, Terada N (2009) Adenine nucleotide translocase 4 deficiency leads to early meiotic arrest of murine male germ cells. Reproduction 138: 463 470. 27. Lim CH, Hamazaki T, Braun EL, Wade J, Terada N (2011) Evolutionary genomics implies a specific function of Ant4 in mammalian and anole lizard male germ cells. PLoS One 6: e23122. 28. Klingenberg M (1989) Molecular aspects of the adenine nucleotide carrier from mitochondria. Arch Biochem Biophys 270: 1 14. 29. Nelson DR, Felix CM, Swanson JM (1998) H ighly conserved charge pair networks in the mitochondrial carrier family. J Mol Biol 277: 285 308. 30. Fiore C, Trezeguet V, Le Saux A, Roux P, Schwimmer C, et al. (1998) The mitochondrial ADP/ATP carrier: structural, physiological and pathological aspects Biochimie 80: 137 150. 31. Palmieri L, Lasorsa FM, Vozza A, Agrimi G, Fiermonte G, et al. (2000) Identification and functions of new transporters in yeast mitochondria. Biochim Biophys Acta 1459: 363 369. 32. Amiri H, Karlberg O, Andersson SG (2003) Deep origin of plastid/parasite ATP/ADP translocases. J Mol Evol 56: 137 150. 33. Kolarov J KN, Nelson N. (1990) A third ADP/ATP translocator gene in yeast. J Biol Chem 265: 12711 12716. 34. Betina S GG, Haviernik P, Sabov L, Kolarov J. (1995) Expression of t he AAC2 gene encoding the major mitochondrial ADP/ATP carrier in Saccharomyces cerevisiae is controlled at the transcriptional level by oxygen, heme and HAP2 factor. Eur J Biochem 229: 651 657. 35. Sabova L, Zeman I, Supek F, Kolarov J (1993) Transcription al control of AAC3 gene encoding mitochondrial ADP/ATP translocator in Saccharomyces cerevisiae by oxygen, heme and ROX1 factor. Eur J Biochem 213: 547 553. 36. Smith CP, Thorsness PE (2008) The molecular basis for relative physiological functionality of t he ADP/ATP carrier isoforms in Saccharomyces cerevisiae. Genetics 179: 1285 1299. 37. Mentel M, Piskur J, Neuveglise C, Rycovska A, Cellengova G, et al. (2005) Triplicate genes for mitochondrial ADP/ATP carriers in the aerobic yeast Yarrowia lipolytica are regulated differentially in the absence of oxygen. Mol Genet Genomics 273: 84 91.
96 38. Huizing M, Ruitenbeek W, van den Heuvel LP, Dolce V, Iacobazzi V, et al. (1998) Human mitochondrial transmembrane metabolite carriers: tissue distribution and its implic ation for mitochondrial disorders. J Bioenerg Biomembr 30: 277 284. 39. Graham BH WK, Cottrell B, Trounce IA, MacGregor GR, Wallace DC. (1997) A mouse model for mitochondrial myopathy and cardiomyopathy resulting from a deficiency in the heart/muscle isofo rm of the adenine nucleotide translocator. Nat Genet 16: 226 234. 40. Lunardi J, Hurko O, Engel WK, Attardi G (1992) The multiple ADP/ATP translocase genes are differentially expressed during human muscle development. J Biol Chem 267: 15267 15270. 41. Elli son JW, Salido EC, Shapiro LJ (1996) Genetic mapping of the adenine nucleotide translocase 2 gene (Ant2) to the mouse proximal X chromosome. Genomics 36: 369 371. 42. Ceci JD (1994) Mouse chromosome 8. Mamm Genome 5 Spec No: S124 138. 43. Belzacq AS, Vieir a HL, Kroemer G, Brenner C (2002) The adenine nucleotide translocator in apoptosis. Biochimie 84: 167 176. 44. Rolland AD, Lehmann KP, Johnson KJ, Gaido KW, Koopman P (2011) Uncovering gene regulatory networks during mouse fetal germ cell development. Biol Reprod 84: 790 800. 45. McCarrey JR TK (1987) Human testis specific PGK gene lacks introns and possesses characteristics of a processed gene. Nature 326: 501 505. 46. Wang PJ (2004) X chromosomes, retrogenes and their role in male reproduction. Trends End ocrinol Metab 15: 79 83. 47. Katsu Y, Braun EL, Guillette LJ, Jr., Iguchi T (2009) From reptilian phylogenomics to reptilian genomes: analyses of c Jun and DJ 1 proto oncogenes. Cytogenet Genome Res 127: 79 93. 48. Philippe H, Laurent J (1998) How good are deep phylogenetic trees? Curr Opin Genet Dev 8: 616 623. 49. Katoh K, Misawa K, Kuma K, Miyata T (2002) MAFFT: a novel method for rapid multiple sequence alignment based on fast Fourier transform. Nucleic Acids Res 30: 3059 3066. 50. Katoh K, Toh H (2008) Recent developments in the MAFFT multiple sequence alignment program. Brief Bioinform 9: 286 298. 51. Stamatakis A (2006) RAxML VI HPC: maximum likelihood based phylogenetic analyses with thousands of taxa and mixed models. Bioinformatics 22: 2688 2690.
97 5 2. Le SQ, Gascuel O (2008) An improved general amino acid replacement matrix. Mol Biol Evol 25: 1307 1320. 53. Akaike H. Information theory and an extension of the maximum likelihood principle.; 1974; Budapest. pp. 267 281. 54. Fujita PA, Rhead B, Zweig AS Hinrichs AS, Karolchik D, et al. (2011) The UCSC Genome Browser database: update 2011. Nucleic Acids Research 39: D876 D882. 55. Yang Z (2007) PAML 4: phylogenetic analysis by maximum likelihood. Mol Biol Evol 24: 1586 1591. 56. Yang Z, Nielsen R, Hasega wa M (1998) Models of amino acid substitution and applications to mitochondrial protein evolution. Mol Biol Evol 15: 1600 1611. 57. Yang Z (1998) Likelihood ratio tests for detecting positive selection and application to primate lysozyme evolution. Mol Bio l Evol 15: 568 573. 58. Burnham KP, Anderson DR (2002). Model selection and multimodel inference: a practical information theoretic approach. New York: Springer Verlag. pp. 488. 59. Lin YH, McLenachan PA, Gore AR, Phillips MJ, Ota R, et al. (2002) Four new mitochondrial genomes and the increased stability of evolutionary trees of mammals from improved taxon sampling. Mol Biol Evol 19: 2060 2070. 60. Holland BR, Penny D, Hendy MD (2003) Outgroup misplacement and phylogenetic inaccuracy under a molecular cloc k -a simulation study. Syst Biol 52: 229 238. 61. Hackett SJ, Kimball RT, Reddy S, Bowie RC, Braun EL, et al. (2008) A phylogenomic study of birds reveals their evolutionary history. Science 320: 1763 1768. 62. GORMAN GC (1973) The chromosomes of the Repti lia,a cytotaxonomic interpretation. A B Chiarelli and E Capanna (eds), Cytotaxonomy and Vertebrate Evolution. New York: Academic Press. pp. 349 424. 63. Ezaz T, Sarre SD, O'Meally D, Graves JA, Georges A (2009) Sex chromosome evolution in lizards: independ ent origins and rapid transitions. Cytogenet Genome Res 127: 249 260. 64. Kimura M (1983). The Neutral Theory of Molecular Evolution. Cambridge: Cambridge University Press. pp. 75. 65. MA C, FA M Polyhedral Geometry of Phylogenetic Rogue Taxa. Bull Math Bi ol published 17 July 2010; PMID: 20640527. 66. Braun EL, Grotewold E (2001) Fungal Zuotin proteins evolved from MIDA1 like factors by lineage specific loss of MYB domains. Mol Biol Evol 18: 1401 1412.
98 67. Chojnowski JL, Kimball RT, Braun EL (2008) Introns outperform exons in analyses of basal avian phylogeny using clathrin heavy chain genes. Gene 410: 89 96. 68. Spinks PQ, Thomson RC, Lovely GA, Shaffer HB (2009) Assessing what is needed to resolve a molecular phylogeny: simulations and empirical data from emydid turtles. BMC Evol Biol 9: 56. 69. Ellegren H, Parsch J (2007) The evolution of sex biased genes and sex biased gene expression. Nat Rev Genet 8: 689 698. 70. Meisel RP (2011) Towards a more nuanced understanding of the relationship between sex biase d gene expression and rates of protein coding sequence evolution. Mol Biol Evol: In press. 71. Grossman LI, Wildman DE, Schmidt TR, Goodman M (2004) Accelerated evolution of the electron transport chain in anthropoid primates. Trends Genet 20: 578 585. 72. Castoe TA, de Koning AP, Kim HM, Gu W, Noonan BP, et al. (2009) Evidence for an ancient adaptive episode of convergent molecular evolution. Proc Natl Acad Sci U S A 106: 8986 8991. 73. Sassi SO, Braun EL, Benner SA (2007) The evolution of seminal ribonucl ease: pseudogene reactivation or multiple gene inactivation events? Mol Biol Evol 24: 1012 1024. 74. Miura I, Ohtani H, Nakamura M, Ichikawa Y, Saitoh K (1998) The origin and differentiation of the heteromorphic sex chromosomes Z, W, X, and Y in the frog R ana rugosa, inferred from the sequences of a sex linked gene, ADP/ATP translocase. Mol Biol Evol 15: 1612 1619. 75. Brower JV, Lim CH, Han C, Hankowski KE, Hamazaki T, et al. (2009) Differential CpG island methylation of murine adenine nucleotide transloca se genes. Biochim Biophys Acta 1789: 198 203. 76. Varriale A, Bernardi G (2006) DNA methylation in reptiles. Gene 385: 122 127. 77. Braun EL (2003) Innovation from reduction: gene loss, domain loss and sequence divergence in genome evolution. Appl Bioinfor matics 2: 13 34. 78. Heintz N (2001) BAC to the future: the use of bac transgenic mice for neuroscience research. Nat Rev Neurosci 2: 861 870. 79. Giraldo P, Montoliu L (2001) Size matters: use of YACs, BACs and PACs in transgenic animals. Transgenic Res 1 0: 83 103. 80. Copeland NG, Jenkins NA, Court DL (2001) Recombineering: a powerful new tool for mouse functional genomics. Nat Rev Genet 2: 769 779.
99 81. Bellve AR, Millette CF, Bhatnagar YM, O'Brien DA (1977) Dissociation of the mouse testis and characteri zation of isolated spermatogenic cells. J Histochem Cytochem 25: 480 494. 82. Braun T, Arnold HH (1995) Inactivation of Myf 6 and Myf 5 genes in mice leads to alterations in skeletal muscle development. Embo j 14: 1176 1186. 83. Braun T, Rudnicki MA, Arnol d HH, Jaenisch R (1992) Targeted inactivation of the muscle regulatory gene Myf 5 results in abnormal rib development and perinatal death. Cell 71: 369 382. 84. Floss T, Arnold HH, Braun T (1996) Myf 5(m1)/Myf 6(m1) compound heterozygous mouse mutants down regulate Myf 5 expression and exert rib defects: evidence for long range cis effects on Myf 5 transcription. Dev Biol 174: 140 147. 85. Held T, Paprotta I, Khulan J, Hemmerlein B, Binder L, et al. (2006) Hspa4l deficient mice display increased incidence o f male infertility and hydronephrosis development. Mol Cell Biol 26: 8099 8108. 86. Yadav RP, Kotaja N (2013) Small RNAs in spermatogenesis. Mol Cell Endocrinol. 87. Bjork JK, Sandqvist A, Elsing AN, Kotaja N, Sistonen L (2010) miR 18, a member of Oncomir 1, targets heat shock transcription factor 2 in spermatogenesis. Development 137: 3177 3184. 88. Grinchuk OV, Jenjaroenpun P, Orlov YL, Zhou J, Kuznetsov VA (2010) Integrative analysis of the human cis antisense gene pairs, miRNAs and their transcription r egulation patterns. Nucleic Acids Res 38: 534 547. 89. Greenlee AR, Shiao MS, Snyder E, Buaas FW, Gu T, et al. (2012) Deregulated sex chromosome gene expression with male germ cell specific loss of Dicer1. PLoS One 7: e46359. 90. Westholm JO, Lai EC (2011) Mirtrons: microRNA biogenesis via splicing. Biochimie 93: 1897 1904. 91. Saito K, Nishida KM, Mori T, Kawamura Y, Miyoshi K, et al. (2006) Specific association of Piwi with rasiRNAs derived from retrotransposon and heterochromatic regions in the Drosophil a genome. Genes Dev 20: 2214 2222. 92. Deng W, Lin H (2002) miwi, a murine homolog of piwi, encodes a cytoplasmic protein essential for spermatogenesis. Dev Cell 2: 819 830. 93. Grivna ST, Beyret E, Wang Z, Lin H (2006) A novel class of small RNAs in mouse spermatogenic cells. Genes Dev 20: 1709 1714. 94. Wang X (2009) Cre transgenic mouse lines. Methods Mol Biol 561: 265 273.
100 95. Chandler KJ, Chandler RL, Broeckelmann EM, Hou Y, Southard Smith EM, et al. (2007) Relevance of BAC transgene copy number in mic e: transgene copy number variation across multiple transgenic lines and correlations with transgene integrity and expression. Mamm Genome 18: 693 708. 96. Vidal F, Sage J, Cuzin F, Rassoulzadegan M (1998) Cre expression in primary spermatocytes: a tool for genetic engineering of the germ line. Mol Reprod Dev 51: 274 280. 97. Rassoulzadegan M, Magliano M, Cuzin F (2002) Transvection effects involving DNA methylation during meiosis in the mouse. Embo j 21: 440 450. 98. Yoshioka H, Geyer CB, Hornecker JL, Pate l KT, McCarrey JR (2007) In vivo analysis of developmentally and evolutionarily dynamic protein DNA interactions regulating transcription of the Pgk2 gene during mammalian spermatogenesis. Mol Cell Biol 27: 7871 7885. 99. Ando H, Haruna Y, Miyazaki J, Okab e M, Nakanishi Y (2000) Spermatocyte specific gene excision by targeted expression of Cre recombinase. Biochem Biophys Res Commun 272: 125 128. 100. Bhullar B, Schmidt JV, Truong T, Rancourt D, van der Hoorn FA (2001) Germ cell specific promoter drives ect opic transgene expression during embryogenesis. Mol Reprod Dev 59: 25 32. 101. Inselman AL, Nakamura N, Brown PR, Willis WD, Goulding EH, et al. (2010) Heat shock protein 2 promoter drives Cre expression in spermatocytes of transgenic mice. Genesis 48: 114 120. 102. Bergqvist I, Eriksson B, Eriksson M, Holmberg D (1998) Transgenic Cre recombinase expression in germ cells and early embryogenesis directs homogeneous and ubiquitous deletion of loxP flanked gene segments. FEBS Lett 438: 76 80. 103. Rempe D, Van geison G, Hamilton J, Li Y, Jepson M, et al. (2006) Synapsin I Cre transgene expression in male mice produces germline recombination in progeny. Genesis 44: 44 49. 104. Nagy A, Gertsenstein M, Vintersten K, Behringer R (2003) Manipulating the mouse embryo: A laboratory manual. New York: Cold spring harbor laboratory press. 105. Pepling ME (2006) From primordial germ cell to primordial follicle: mammalian female germ cell development. Genesis 44: 622 632. 106. Hartshorne GM, Lyrakou S, Hamoda H, Oloto E, Ghaf ari F (2009) Oogenesis and cell death in human prenatal ovaries: what are the criteria for oocyte selection? Mol Hum Reprod 15: 805 819.
101 107. Lee JT (2005) Sex chromosome inactivation: the importance of pairing. Curr Biol 15: R249 252. 108. Disteche CM (20 12) Dosage compensation of the sex chromosomes. Annu Rev Genet 46: 537 560. 109. McCarrey JR, Thomas K (1987) Human testis specific PGK gene lacks introns and possesses characteristics of a processed gene. Nature 326: 501 505. 110. Suarez SS (2008) Control of hyperactivation in sperm. Hum Reprod Update 14: 647 657. 111. Suarez SS, Dai XB, DeMott RP, Redfern K, Mirando MA (1992) Movement characteristics of boar sperm obtained from the oviduct or hyperactivated in vitro. J Androl 13: 75 80. 112. Jin JL, O'Doherty AM, Wang S, Zheng H, Sanders KM, et al. (2005) Catsper3 and catsper4 encode two cation channel like proteins exclusively expressed in the testis. Biol Reprod 73: 1235 1242. 113. Ren D, Navarro B, Perez G, Jackson AC, Hsu S et al. (2001) A sperm ion channel required for sperm motility and male fertility. Nature 413: 603 609. 114. Quill TA, Sugden SA, Rossi KL, Doolittle LK, Hammer RE, et al. (2003) Hyperactivated sperm motility driven by CatSper2 is required for fertilizati on. Proc Natl Acad Sci U S A 100: 14869 14874. 115. Qi H, Moran MM, Navarro B, Chong JA, Krapivinsky G, et al. (2007) All four CatSper ion channel proteins are required for male fertility and sperm cell hyperactivated motility. Proc Natl Acad Sci U S A 104 : 1219 1223. 116. Ho HC, Granish KA, Suarez SS (2002) Hyperactivated motility of bull sperm is triggered at the axoneme by Ca2+ and not cAMP. Dev Biol 250: 208 217. 117. Stendardi A, Focarelli R, Piomboni P, Palumberi D, Serafini F, et al. (2011) Evaluatio n of mitochondrial respiratory efficiency during in vitro capacitation of human spermatozoa. Int J Androl 34: 247 255. 118. Hung PH, Miller MG, Meyers SA, VandeVoort CA (2008) Sperm mitochondrial integrity is not required for hyperactivated motility, zona binding, or acrosome reaction in the rhesus macaque. Biol Reprod 79: 367 375. 119. Ovadi J, Saks V (2004) On the origin of intracellular compartmentation and organized metabolic systems. Mol Cell Biochem 256 257: 5 12. 120. Graham JW, Williams TC, Morgan M Fernie AR, Ratcliffe RG, et al. (2007) Glycolytic enzymes associate dynamically with mitochondria in response to respiratory demand and support substrate channeling. Plant Cell 19: 3723 3738.
102 121. Haggie PM, Verkman AS (2002) Diffusion of tricarboxylic a cid cycle enzymes in the mitochondrial matrix in vivo. Evidence for restricted mobility of a multienzyme complex. J Biol Chem 277: 40782 40788. 122. Odet F, Gabel SA, Williams J, London RE, Goldberg E, et al. (2011) Lactate dehydrogenase C and energy metab olism in mouse sperm. Biol Reprod 85: 556 564.
103 BIOGRAPHICAL SKETCH s s degree as a chemical engineer from Ko rea University in 2001 and Korea Advanced Institute of Science and Technology (KAIST) in 2003 respectively During this time, he was inspired by in vitro generation of organs using tissue engineering After graduation, he worked in a biotechnology company from 200 3 to 2007 to develop an artificial cornea u sing corneal epithelial stem cells In 2007, he joined the Interdisciplinary Program in Biomedical Sciences (IDP) at the University of Florida. His research focused on the origin and function s of Ant4 in male germ cells After completion of his doctorate, he plans to continue to study stem cells and regeneration for his post doctoral work.