UNDERSTANDING THE ROLE OF THE TRANSITION PROTEINS TP1 AND TP2

Submitted by:

Mohinish Shukla Integrated Ph.D. SR# 401095086

This is to certify that Mohinish Shukla, SR# 401095086 has done the work included in this report in my laboratory from August 1997 to March 1998 in partial fulfilment of the requirements of the Integrated Ph.D. course

Prof. M.R.S. Rao, Project Advisor

ii

ACKNOWLEDGEMENTS I would like to thank Prof. M.R.S. Rao for giving me this interesting project and again setting very realistic goals. I wish to thank all my labmates; Geetha, Ruhikanth, Bharath, Debjani, Ashok, Vargheese, Chandrasekhar, Coumarane, Gayathri, Suresh, Savitha and others for an enjoyable time and look forward to greater interactions with them in the lab. Ruhikanth deserves special thanks for the polyclonal anti-TP2 antibodies. Sekhar was very helpful during the screenings.

iii

CONTENTS Introduction

pg. 1

Scope and aim of the project

pg. 5

Materials and methods

pg. 7

Results and discussions

pg. 14

References

pg. 18

iv

INTRODUCTION Spermatogenesis is a complex developmental pathway [Fig.1]. In higher eutherian mammals, spermatogenesis results in the formation of the haploid male gametes, sperms. The entire process involves myriad complex biochemical as well as cytological changes.

Fig.1 Mammalian Spermatogenesis

One of the extremely striking changes occurs in the organisation of DNA. In the stem cells, the chromatin is quite similar to that found in other tissues. It is formed by the wrapping up of the DNA by the histones. Histones are basic proteins which assemble DNA onto them, thus condensing it. In the testes, specifically in the mid-pachytene

spermatocytes the histones that package DNA include the

testis-specific variants of the histones H2A, H2B,and H3 (the H4 variant has not been reported) and also the testis - specific linker histone, H1t. About midway

1

through spermatid development, the nucleus begins to condense, elongate and becomes resistant to mechanical disruption (1). In mature sperm, the DNA is extremely compact and this packaging is mediated through the highly basic proteins, the Protamines (2). One of the interesting developmental problems of spermatogenesis is how the histones are replaced by the protamines during the process of spermiogenesis. The problem is made more interesting by the fact that just before spermiogenesis [Fig.1], when the transition from histones to protamines occurs, transcription is totally shut off (3).

Fig. 2 The stages of Rat spermiogenesis

This transition is accompanied by the presence of proteins which are found only for a short while during spermatid development[Fig.2].Three such ‘Transition Proteins’ have been identified in rat, mouse and human (4). TP1 from rat is a 5KDa, basic protein, with basic residues distributed rather evenly (5). TP2 from rat is a 13 kDa protein(10), also very basic, and has two

2

atoms of zinc bound per molecule(11). TP1 and TP2 are the most abundant of the transition proteins in the developing (round) rat spermatids (6). Using a cDNA probe it had been shown previously that the TP1 and TP2 messages first appears post-meiotically in late round spermatids, just prior to the cessation of transcription, and thus this represents an example of haploid specific gene expression (8). Labelling experiments using rat seminiferous tubules has shown that protamines and transition proteins are deposited into nuclei immediately after their synthesis (7). Recently, it was demonstrated that TP2 synthesis and deposition on chromatin precedes that of TP1. TP2 appears at stage 9 of spermiogenesis, while TP1 appears at stage 12 (9) [Fig.2] Gene expression during spermatogenesis is a very complex process and involves many levels of regulation. It has been shown, for example, that the levels of the RNA Polymerase II components in the rodent spermatids is at least 100-fold higher than in the somatic cells(13). In the round spermatids, transcription of a specific subset of genes occurs and at least in the case of TP1, one of the elements controlling this tissue-specific gene expression is the CRE element upstream of TP1. This element mediates activation by the CREMτ, the activating splice variant of the CREM family of repressors/activators(14). It has also been shown that the functional switch of the CREM family from the repressor forms like CREMα, CREMβ and CREMγ to the activator form, CREMτ, is mediated by the pituitary hormone, FSH(15), thus tying the endocrinology of the system to the molecular basis for tissue-specific transcriptional regulation. However, the precise basis for tissue-specific regulation is far from clear, and the best results have come from transgenic studies in which promoter regions of testis-specific genes have been used to determine what are the elements responsible for testis-specific expression. Thus studies on

the regulation of

expression of the mouse protamine gene mP1 have characterised a sequence which 3

causes testis-specific expression (16). Besides, in vitro studies have identified certain protein complexes in testicular extracts which can bind to oligonucleotides bearing the regions of mP1 known to be important for testis-specific expression(17). Such complexes have also been identified for the testis-specific histone variant H1t(18) promoter, for which it has been shown that 1kb of upstream sequences of the rat H1t are sufficient to direct the expression of a downstream gene to the pachytene spermatocytes of transgenic mice(19). Protein expression in the testis is also controlled at the translational level. Thus, though transcription is shut off in late round spermatids, there is still a need for temporal control over the expression of the various RNA that are already transcribed. For example , the TP1 transcripts are not translated for nearly three days in the mouse spermatids (20), while the transcripts of the protamines are kept translationally repressed for almost a week(21). This translational regulation appears to be achieved through the 3’UTR of the transcripts. For example, transgenic

studies have shown that the 3’UTR of the mouse mP1 gene for

protamine 1 is sufficient for the correct temporal expression of a linked reporter gene(22). Spermatogenesis is thus an excellent system for understanding haploid specific gene expression in mammals as well as studying a complex developmental program, controlled both spatially and temporally.

SCOPE AND AIM OF THE PROJECT Our lab has been trying to understand the role of transition in spermiogenesis. Towards this end, we have undertaken two different approaches to study the function in vivo.

4

The first approach is to express these proteins transiently in cell lines and see their effect on the cells in culture. To do this, we needed to generate the following reagents: a) Shuttle vectors with TP1 and TP2 under a constitutive promoter for expression in a cell culture system. b) Specific antibodies against TP1 and TP2 to detect and localise the proteins inside the cells. In my first project the TP1 shuttle vector had been constructed. The TP2 shuttle vector has also been generated in the lab. It had previously been tried to raise antibodies to TP1, but the attempt was not successful.. Antibodies to TP2 has already been generated in the laboratory. In order to make antibodies to TP1, we needed a steady supply of the TP1 protein. TP1 has many cysteines and histidines whose codons are very poorly utilised in bacteria. and which could have resulted in low levels of expression. Therefore in order to maximise expression and also for ease of purification, the aim was: 1. To make a synthetic gene for the overexpression of TP1 having E.coli -preferred codons and a 6X Histidine tag. As mentioned above antibodies to TP2 had already been raised in rabbits using recombinant TP2. The second aim was: 2. To purify the polyclonal antibodies by immunoaffinity chromatography to obtain monospecific, anti-TP2 antibodies. . The other approach that has been initiated in the lab to understand the in vivo functions of TP1 and TP2

is by using Transgenic Mouse technology.

Towards this end, we desired to obtain the mouse genomic clones of TP1 and TP2. So, the aim was: 3. To isolate the mouse genomic TP1 clone from the genomic DNA library and characterise it by restriction map. 5

The cloning of the rat TP1 genomic fragment was reported in my earlier project and its restriction map was characterised and is presented in this report.

MATERIALS AND METHODS Bacterial strains (E. coli): DH5α supE44 ΔlacU169(φ80lacZΔM15) hsdR17 recA1 endA1 gyrA96 thi-1relA1 BL21 hsdS gal (λcIts857 ind1 Sam7 nin5 lac UV5-T7 gene 1) XL1-Blue supE44 hsdR17 recA1 endA1 gyrA46 thi relA1 lacF’[proAB+ lacIq lacZΔM15 Tn10(tetr)] Plasmid

preparations,

restriction

digestion,

cloning

and

agarose

gel

electrophoresis The standard protocols were used according to Sambrook et al.(23). However, for preparing plasmid for sequencing using the automated sequencer, the method recommended by Perkin-Elmer Corp. was used which is as follows: cells bearing the plasmid were grown in LB broth (1% tryptone, 0.5% yeast extract, 1% NaCl) with 100µg/ml ampicillin overnight. Cells were spun down (upto 4.5ml per microfuge tube). Cells were suspended in 200µl of GTE (50mM glucose, 25mM Tris HCl(pH 8.0), 10mM EDTA(pH 8.0)) by gentle pipetting and kept on ice for 5min. 300µl of solution II(0.2N NaOH, 1%SDS) was added, the tube inverted a few times and incubated on ice for 5min. 200µl solution III(3M potassium, 5M acetate) was added, the tube inverted twice and incubated on ice for 5min. The microfuge tubes were spun at 12,000 rpm for 8min at room temperature. The supernatant was carefully removed and to it RNase A was added to a final concentration of 20µg/ml and incubation was carried out for 1½hrs. The supernatant was extracted twice with an equal volume of chloroform. The DNA

6

was recovered by adding an equal volume of isopropanol followed immediately by centrifugation at 12,000rpm for 10min and decanting the supernatant. The DNA pellet was washed with 500µl of 70% ethanol, and dried in the vacuum drier. The DNA pellet was resuspended in 32µl of autoclaved, double-distilled deionised water. To this was added 8µl of 4M NaCl and 40µl of 13% PEG8000 and the solution was kept on ice for 1hr. The DNA was pelleted by centrifugation at 12,000rpm for 15min. at 4ºC. The supernatant was carefully aspirated and the DNA pellet rinsed with 500µl of 70% ethanol. The pellet was vacuum dried and resuspended in 10-20µl of autoclaved, double-distilled deionised water. Screening the lambda phage mouse genomic library The λFIX mouse genomic library was screened as described below. The library contains Sau3AI partially digested mouse genomic DNA, cloned into two BamHI sites which are the inner edges of two oppositely oriented multi-cloning sites. The inserts are 9-22 Kb in length. The host strain used was E.coli, XL1-Blue MRA(P2). Titres were estimated in all the cases by doing a standard plaque count on 9cm petriplates. After plating onto large (15cm.) plates about half a million phage (corresponding to approximately 0.5 genome equivalents) plaques were lifted onto nylon filters (HYBOND, Amersham). The filters were

soaked in denaturing

solution which consisted of 0.5N NaOH and 1.5M NaCl. After neutralising in 0.5M Tris HCl (pH 8.0), the filters were rinsed in 0.2M Tris HCl (pH 7.5), 2X SSC and were air-dried on blotting paper. The DNA was crosslinked to the filters in the UV crosslinker (Amersham) at the 70,000 µJ.sec-1 setting. The filters were prewashed with 5X SSC, 0.5% SDS, 1mM EDTA, pH 8.0 for 2 hours at 42oC and were then scrubbed. Prehybridization was done using 6X

7

SSC, 5X Denhardt’s reagent, 0.5% SDS, 100 µg/ml denatured salmon sperm DNA, for 2 hours at 65oC. Hybridisation was done in the same buffer, along with the radiolabelled probe. Radiolabelled probe was made by using α32P dATP and random primers, extended by Klenow according to the manufacturers recommendations, using a commercial kit(Amersham). In a typical reaction using 100ng of DNA (rat TP1 cDNA fragment excised from the vector), specific activity of around 8 X 108 cpm/µg DNA was routinely obtained. Hybridisation was done overnight at 65oC in the hybridisation oven (Amersham) with gentle agitation. Filters were

washed as follows: (a) Two washes in 2X SSC at room

temperature (b) One wash in 2X SSC at 68oC (c) One wash (highest stringency) in 0.2X SSC at 68oC. 0.1% SDS was used in all the cases. The filters were autoradiographed. For the λ screening, filters were generally exposed for 48-72 hours at -70oC. For the Southern analysis, filters were exposed for 12-36 hours at 20oC. Phage stocks were prepared by the plate lysate method. Phage DNA was prepared by the plate lysate method as follows: Host bacteria are grown in fresh medium (NZYM) for 2-3 hours, spun down and made to OD600 = 0.5 in 10mM MgSO4. 300ul of the host was inoculated with approx. 60000 pfu and incubated for o

15' at 37 C. The host and phage were then plated onto NZY agar along with 6.5ml top agarose. The plates were incubated till a webbing, indicating extensive lysis, was seen. 10ml TM(50mM Tris Hcl, pH 7.5, 0.2% MgSO4.7H2O) was poured onto the plates and they were shaken for two hours at RT. The TM was collected, the plates were rinsed with 1ml of TM again and all the TM was collected into centrifuge tubes. The bacterial

debris was spun down. The supernatant was

transferred to clean conical flask. 1µl of 10mg/ml DNaseI and 10µl of 10mg/ml

8

RNase A were added and the flask was kept for 30min at 37°C. 4.7gms of NaCl and 8gms of PEG8000 was added to the mixture and the volume was made upto 80ml. The mixture was incubated on ice for 1hr.. The precipitated phage were recovered by centrifugation. The supernatant was drained off and the phage pellet was resuspended in 0.5ml of TE(10mM Tris Hcl, 1mM EDTA). The debris was removed by brief centrifugation. 5µl of 10% SDS was added to the phage and incubated for 5min at 68°C. 10ul of 5M NaCl was added and the proteins etc. were removed by a phenol:chloroform step followed by a chloroform step. Equal volume of isopropanol was added and the mix was stored at -70°C for 15min. The DNA was recovered by centrifugation and washed with 70% ethanol and resuspended in autoclaved, double-distilled water. Purification of 6X His-tagged TP2 using a nickel-agarose column a)Induction of the expression vector and preparation of cell lysate The expression vector was a modified pTrc99A containing the partially synthetic TP2 with a 6X His tag at the C-terminus (10). The vector was transformed into fresh CaCl2 competent BL21 cells. 3 ml of preinoculum was used to inoculate 330ml of LB-Amp in a 1litre flask. After 2½ hours of growth at 37°C, at an OD600 of approx. 0.6-0.8, 150µl of 1M IPTG was added to the flask, and growth was continued at 37oC for a further 4 hrs. Bacteria were pelleted and resuspended in 10ml of buffer containing 50 mM Tris HCl, pH 8.0; 0.8M NaCl; 20mM imidazole; 10mM β mercaptoethanol and 0.2mM phenylmethylsulfonyl fluoride(PMSF). The suspension was made upto 0.4 mg/ml of fresh lysozyme and incubated at RT for 10min. The volume was made up to 20ml and the suspension was kept at -70°C overnight. The next day the suspension was thawed and sonicated for 30min. at the maximum output. The suspension was then subjected to ultracentrifugation for 1hr at 40,000rpm. 9

The supernatant, the cell lysate was taken for further processing. b) Nickel-Agarose Purification Ni-NTA agarose beads from Qiagen were packed into a column. Approximately 2ml of beads were used. All the packing, loading and washing were done in the cold-room. After equilibrating the column with the buffer used for making the bacterial lysate, the lysate was loaded onto the column and allowed to flow at a rate of approximately 20ml/hour. The column was then washed with 5 volumes of the column buffer, followed by the column buffer containing 50mM imidazole. Finally, the protein was eluted from the column with the same buffer, but now containing 250mM imidazole, at pH 7.5. 10µl of the eluate was TCAA precipitated and checked by SDS-PAGE. The rest of the eluate was dialysed against PBS and concentrated in the vacuum drier. Purification of monospecific anti-TP2 antibodies by immuno-affinity technique a) Conjugating TP2 to the CNBr beads CNBr-sepharose (Sigma) was reswelled in 1mM HCl (200ml/gm). The protein was dialysed against coupling buffer (0.1M NaHCO3, pH 8.3; 0.5M NaCl). Protein solution was mixed with 200µl of beads (5-10mg of protein per ml of beads) overnight in the cold-room in an end-over-end tumbler. The remaining sites on the beads were blocked with the same NaHCO3 buffer with 0.2M glycine, pH 8.0, for 16hrs at 4°C. The beads were washed with coupling buffer followed by acetate buffer (0.1M Na-acetate, pH 4; 0.5M NaCl). Column was then washed with the coupling buffer. The column was washed several times with PBS. The polyclonal antibody was also in PBS, and it was mixed with the beads and kept overnight on an endover-end tumbler at 4°C. The supernatant was removed, and the bound antibodies were eluted with 3.5M MgCl2 in PBS for 1.5hrs at 4°C. The eluate was dialysed 10

extensively against PBS. The dialysed eluate was checked for specificity by western blotting at 1:500 dilution. Restriction mapping the genomic TP1 clones To restriction map the genomic clones, the lambda DNA was digested with various enzymes singly or in combination, and Southern hybridisation with the rat TP1 cDNA was done. From the size of the hybridising fragments, a rough map of the restriction sites with respect to the gene was made. Overlapping, step-PCR for making the synthetic TP1 The schematic representation for making the synthetic TP1 with the 6X His tag at the C-terminus is shown in Fig. 3. Four oligonucleotides were synthesised, which correspond to the entire TP1 coding region with all E.coli.-preferred codons, an NcoI site at the ATG of TP1, a (CAT CAC)3 corresponding to the 6-histidine tag at the N terminus and a BamHI site after the stop codon. All the oligonucleotides had a Tm of approx. 56ºC in the overlapping region. The overlapping PCR was done in three steps. In the first step, primers 1 and 2 alone were used. The product obtained was used for the second cycle using primers 1 and 3. Finally, this product was used for the third cycle, wherein primers 1 and 4 were used. The PCR reactions were all done using Pfu polymerase(Stratagene) to minimise PCR-mediated errors. It was observed that the overlapping PCR always gave a smear, so step-PCR was resorted to. In this, PCR for the II and III cycles were done in two steps. In the first step, only the downstream primer is used (i.e., 3 or 4) at an annealing temperature of 48°C, for 5 cycles. In the second step, primer 1 is added and PCR is continued for 25 more cycles at 61°C. The overlapping-step PCR gave sharp bands.

11

P1 P2 P3 P4

SYNTHETIC TP1: 201bp NcoI

6X His Tag

BamHI

Fig. 3. Overlapping - PCR strategy for making synthetic TP1

During the creation of the synthetic gene, an extra NcoI site had been created starting at position 25 with respect to the correct NcoI at the first ATG. In order to clone the NcoI-BamHI fragment containing the entire gene, complete digestion with BamHI was followed by a partial digestion with NcoI. The products of the digestion were separated on an 8% polyacrylamide gel with TAE as the running buffer. The slowest moving bands obtained in the partial digestion lanes were geleluted by shaking the gel pieces overnight in TE buffer. This DNA was precipitated and used for ligation into the vector pTrc99A, digested with NcoI and BamHI The ligation mix was transformed into DH5α, colonies were screened and a positive clone was sequenced in the ABI Prism 377 sequencer.

RESULTS AND DISCUSSION Restriction mapping the rat TP1 in λDASH Restriction mapping of the rat TP1 showed a pattern similar to the published sequence. Certain sites in the regions outside the previously published sequences were also identified.

12

23.13 9.416 6.577 4.361 2.322 2.027 .125

E

!

B E E X B " S

E E E E X E P K X # H rat

Figure 4: A representative Southern blot used to determine restriction maps λ=λ-HIndIII marker; E=EcoRI; B=BamHI; ο=XhoI; X=XbaI; P=PstI; K=KpnI; β=BglII; H=HindIII;

Map of Genomic Clone of Rat TP1 in !DASH II

20 Kb arm

Kb->

XbaI SalI

1

Pst I

2

BglII

approx. 20 Kb insert

3

4

BamHI

9 Kb arm

5

6

7

EcoRI

XbaI

EcoRI

8

9

10

NotI EcoRI

ExonI

Intron

ExonII

Figure 5. Genomic Rat TP1 clone

Cloning and restriction mapping the mouse genomic TP1 The mouse genomic TP1 was cloned in the λFIX vector as described in the materials and methods. The screening was done using the rat TP1 cDNA as a probe, as there is 97% similarity between the two sequences at the nucleotide level.

13

The restriction map matched the published sequence.

Mouse TP1 gene in Lambda FIX vector coarse map of an internal 10 kb fragment approx. 20kb genomic insert (as a XhoI fragment)

(exact position within the 20 kb insert is unknown)

Kb

0

1

2

EcorI BamHI

3

4

5

6

7

EcoRI PstI

KpnI SalI

8

9

XbaI

HindIII

10

SalI

BamHI

Figure 6. Genomic mouse TP1 clone

Immuno-affinity purification of mono-specific anti-TP2 antibodies The recombinant TP2 was purified to near-homogeneity as seen by SDSPAGE (data not shown). Immuno-affinity

purified

mono-specific

anti-TP2

antibodies, purified as described in materials and methods, were checked by Western blotting purified recombinant TP2 (which had been prepared and made available from the lab), sonication-resistant testicular nuclear extract (SRN, which is enriched for the transition proteins) and the total testicular extract. As can be seen, the antibody is specific to TP2, and picks up only a single band in the SRN lane, which corresponds to the band in the purified TP2 lane.

14

1

2

3

Figure 7. Western to show specificity of mono-specific anti-TP2 antibodies (1) SRN (2) Total testicular extract (3) Purified TP2

Making the synthetic TP1 gene The synthetic gene encoding TP1 was generated as described in materials and methods. Overlapping step PCR gave the required 200bp fragment. Clones were screened by NcoI-BamHI digestion, and the 200 bp fragment was confirmed by sequencing[Fig. 8]. A small discrepancy in the sequence (the presence of a gap at the stop codon, marked as ‘N’) is most probably a sequencing artefact.

NcoI

NcoI

AAA CAG ACC ATG GCG ACC AGC CGC AAA CTG AAA ACC CAT GGC ATG CGT CGT GGC AAA AAC CGT GCG CCG CAC AAA GGC GTG AAA CGT GGC GGC AGC AAA CGT AAA TAC CGT AAA AGC AGC CTG AAA AGC CGT AAA CGT GGC GAT AAT GCG AGC CGT AAC TAC CGT AGC CAC CTG CAT CAC CAT CAC CAT CAC TNA GGA TCC TCT AGA BamHI

6X His Tag

Fig.8: Sequence of the synthetic TP1. The relevant restriction sites as well as the position of the C-terminal 6X-His tag is marked

15

This project has generated some of the reagents required for the study of the function of the transition proteins, TP1 and TP2. The availability of the genomic clones of TP1(this project) and TP2 (generated in the lab) of mouse will be useful in performing knockouts of these genes in mice. The genomic clones

are useful as they can also be used for ectopic

expression of these genes in transgenic mice. Some work has already been initiated in this direction, and we have decided to express the genes at an earlier stage of spermiogenesis in mice to see their effects. Also, the antibodies to TP2 and the expression vector for TP1 which will be used to prepare protein to raise antibodies against TP1 will be vital in visualising the co-localisation of these two proteins when co-transfected into cells in culture.

16

REFERENCES (1) Grimes,S.R., Jr., Meistrich, M.L., Platz, R.D., Hnilica, L.S. (1977) Exp Cell Res. 110:31-39 (2) Kleene, K.C., Distel, R.J., Hecht, N.B. (1985) Biochemistry 24:719-722 (3) Kierszenbaum, A.L.,Tres, L.L. (1974) J Cell Biol. 60:39-53 (4) Meistrich, M.L., Brock, W.A., Grimes, S.R. Jr.,Platz, R.D., Hnilica, L.S. (1978) Fred. Proc. 37:2522-2525 (5) Heidaran, M.A., Kozak, C.A., Kistler, W.S. (1989) Gene 75:39-46 (6) Meistrich, M.L. in Histones and Other Basic Nuclear Proteins (Hnilica, L.S., Stein, G.S., Stein, J.L., Eds.), pp165-182, CRC Press, Boca Raton (1989) (7) Green, G.R., Balhorn, R., Poccia, D.L., Hecht, N.B. (1994) Mol. Reprod. Dev. 37:255-263 (8) Heidaran, M.A., Kistler, W.S. (1987) J. Biol. Chem. 262:13309-13315 (9) Kistler, W.S., Henriksén, K., Mali, P., Parvinen, M. (1996) Exp. Cell Res. 225:374-381 10) Meteei, A.R., Rao, M.R.S. (1996) Protein Expression Purif. 8:409-415 11) Baskaran, R., Rao, M.R.S. 1499 (1991) Biophys. Res. Commun. 179:149212) Unni, E., Meistrich, M.L. (1992) J. Biol. Chem. 267:25359-25363 13) Schmidt, E.E., and Schibler, U. (1995) Development 121:2373-2383 14) Kistler, M.K., Sassone-Corsi, P. and Kistler, W.S. Biol Reprod 51:1322-1329 15) Foulkes, N.S., Schlotter, F., Pévet, P. and Sassone-Corsi, P. (1993) Nature 362:254-267 16) Zambrowicz B.P., Harendza, C.J., Zimmermann, J.W., Brinster, R.L. and Palmiter, R.D. (1993) Proc Natl Acad Sci USA 90:5071-5075 17

17) Zambrowicz B.P., and Palmiter, R.D. (1994) Biol Reprod 50:65-72 18) Clare, S.E., Hatfield, W.R., Fantz, D.A., Kistler, W.S. and Kistler, M.K. (1997) Biol Reprod 56:73-82 19) Bartell, J.G., Davis, T., Kremer, E.J., Dewey, M.J and Kistler, W.S. (1996) J Biol Chem 271:4046-4054 20) Yelick, P.C., Kwon, Y.K., Flynn, J.F., Borzorgzadeh, A., Kleene, K.C. and Hecht, N.B. (1989) Mol Reprod Dev 1:193-200 21) Kleene, K.C., Distel, R.J. and Hecht, N.B. (1984) Dev Biol 105:71-79 22) Braun, R.E., Peschon, J.J., Behringer, R.R., Brinster, R.L and Palmiter, R.D. (1989) Genes Dev 3:793-802 23) Sambrook, J., Fritsch, E.F. and Maniatis, T. Molecular Cloning: A Laboratory Manual CSHL Press, Cold Spring Harbor, NY (1989)

18

19