T-Ag Inhibits Implantation by EC Cell Derived ... - Springer Link

Virus Genes 20:3, 195±200, 2000 # 2000 Kluwer Academic Publishers. Manufactured in The Netherlands.

T-Ag Inhibits Implantation by EC Cell Derived Embryoid Bodies ALEX ESPINOSA & LUIS P. VILLARREAL Department of Molecular Biology and Biochemistry, University of California, Irvine, Irvine Research Unit on Animal Viruses Received October 1, 1999; Accepted October 4, 1999

Abstract. When introduced into EC cells of a blastocyst, polyomavirus (Py) T-Ag results in mice mosaic for T-Ag but otherwise essentially normal. It had been reported that SV40 T-Ag does not inhibit differentiation of F9 EC cells, but did inhibit endogenous retrovirus (ERV) production. We therefore sought to determine if Py T-Ag had any affect on EC derived embryoid body implantation onto mouse placenta. F9 EC cells were selected for T-Ag maintenance. Like the SV40 transformed cells, we show that these Py T-Ag selected EC cells no longer express IAP transcripts following differentiation into embryoid bodies. Normal and Py T-Ag selected F9 cells were differentiated into embryoid bodies then implanted into pseudopregnant mice. We observe, that normal F9 derived embryoid bodies underwent the initial stages of implantation whereas the Py T-Ag selected embryoid bodied did not implant. The implications of this observation with respect to trophectoderm and ERV function are discussed. We examine the idea that ERVs may be a required element for normal embryo implantation. Key words:

Introduction Embryonal carcinoma cells have previously been used to study early events in embryo differentiation. These cells can be induced to produce various embryonic tissues, including trophectoderm (1±3); the predecessor of the placental tissue and the syncytiotrophoblast which is the site of much placental full length and and incomplete ERV expression. Like normal trophectoderm, EC differentiated trophectoderm also produces several classes of ERV's (4±6). EC cells can differentiate into embryoid bodies, resembling 3.5 day blastocyst (7) and can be used as stem cells to contribute to postimplantation embryo development, although resulting in abnormal differentiation (8). However, the ability of these EC-embryoid bodies to interact with the uterus or initiate implantation has not been examined. Early in the study of EC cells, it was reported that F9 EC cells that expressed SV40 large T-Ag were suppressed for ERV production (4), but they were still capable of differentiating into normal appearing trophectoderm. Mouse polyomavirus also codes for the production of a large T-Ag. Embryonic

mouse stem cells or embryonal carcinoma cells selected for the expression of episomal mediated Py LT-Ag have been isolated and shown that these stem cells can contribute as a mosaic constituent of the inner blastocyst cells to the development of most mouse tissue, thus indicating LT-Ag expression in ES cells was not inhibitory to subsequent normal tissue development (9,10). It was of interest that SV40 T-Ag selection resulted in suppression of EC ERV expression. In nonembryonic somatic tissues, ERV production is normally highly suppressed via DNA methylation (11±13). During embryogenesis, suppression of ERV production appears to also be via DNA methylation (14) In fact, ERV expression has been used as an assay for genomic DNA methylation patterns (5,13,15±19). The production of endogeneous retroviral (ERV) particles and their gene products in placental tissues of viviparous mammals is a long standing and consistant observation (20±25), involving high level expression (26). This has led some (27,28) to suggest a functional role for ERV-3 in implantation, for reviews see (29,30). This and other observations concerning the local immunosuppressive activity of retroviral env

196

Espinosa and Villarreal

gene products (31,32) has led us (33) then others (34) to theoretical proposals that genomic retrovirus expression by allogeneic embryos may be required for the suppression of immunological recognition by the viviparous mammalian mother. Several correlative observations seem to support this idea, but others report that ERV-3 env is truncated or not made in some people (35). More recently isolated ERVs, such as MSRV and HERV-K-T47D, are also reported to be expressed as transcripts in placental tissue (36,37) but have not been evaluated for their activity or role in implantation. For human genomes, the ERV3 (28), MSRV (36) and various HERV K10 like ERVs (37), are all expressed in placental tissues. In the Mus musculus genome, although many ERVs are also expressed in placental tissues, of special interest is the IAPE-A ERV. This mouse ERV is low copy, encodes a full env gene and is highly conserved (38). However, this ERV env shows little sequence similarity to those found in the human genome and was not know to express in the mouse embryo. In this report, we establish that the IAPE-A env sequence is highly expressed in a normal mouse post-implantation blastocyst. We then establish that normal differentiated F9 EC cells will also express this env sequence, but this expression is absent in Py T-Ag expressing differentiated F9 EC cells. Using these cells to generate embryoid bodies, we show that while normal F9 EC cell derrived embryo bodies will initiate implantation, those that express Py LT-Ag fail to initiate implantation. Thus, although Py T-Ag does not appear to inhibit normal ES cell development, it inhibits implantation. The report of Reuss characterized the env gene associated with IAPE-A and established that this gene was expressed in NH15-CA2 and AtT20 (39) cells. A BLAST sequence analysis of the Genbank database by us showed that there were no known sequences similar to this in the human genetic database. Thus this sequence is not closely related to any known HERV env, including ERV-3 env. We therefore sought to determine if this env sequence was highly expressed in normal mouse embryos. A probe speci®c for the IAPE-A env sequence was designed for PCR based ampli®cation and successfully ampli®ed the env gene from genomic mouse DNA (Fig. 1). From the IAPE-A sequence an oligo probe was designed and used to make a DNA probe to determine if normal post implantation mouse embryos express IAPE-A env sequence. As shown in Fig. 2, a normal 5

Fig. 1. Successful generation of an env PCR product from IAPEA like env (38,48). Primers for PCR ampli®cation of env from IAPE-A were designed from sequences found in Reuss and Schaller 1991 (48). Env-1 and Env-2 primers amplify a 1761 bp fragment corresponding to the entire env protein coding region. Env-1 corresponds to nucleotides 1655±1675 and Env-2 corresponds to nucleotides 3396±3416 in IAPE-A. Mouse genomic DNA was extracted from liver and kidney of Balb/c ( panel A) and C57/blk ( panel B) mice for PCR ampli®cation. Parameters for PCR were as follows: 94 C 4 min one cycle, 35 cycles of 94 C 1 min, either 60 C±56 C±48 C 1 min, 72 C 1 min, and one cycle of 72 C 5 min. Lanes 1, 2, 3, and 4 in both panels correspond to 4 mM, 3 mM, 2 mM, and 1 mM MgCl2 respectively at 60 C. Lane L corresponds to 1 kb ladder. Annealing temperatures at 56 C and 48 C showed similar results (not shown).

day old mouse embryo showed strong in situ hybridization to a probe containing the IAP sequences and is therefore highly expressing IAP sequence in the trophectoderm of the blastocyst. Therefore, if ERV gene expression is important for implantation, the expression of IAPE-A env sequence would appear to be a good and speci®c marker for such normal expression. As it had been previously established that F9 EC cells expressing SV40 LT-Ag were suppressed for ERV expression in differentiated trophectoderm (4,40), we wanted to determine if Py LT-Ag expressing F9 EC cells were similarly affected. We therefore transfected F9 EC cells with the Py LT-Ag expressing plasmid, pMGD20neo (9,10), and selected for resulting EC cell lines. This cell line (F9LTneo) was then evaluated for the expression of IAPE-A env sequence in relationship to differentiation. As shown in Fig. 3, normal undifferentiated EC cells do not express env sequence. However, when induced to

T-Ag Inhibits Implantation

Fig. 2. Normal implanted embryos express IAP sequences. In situ hybridization was done using digoxigenin labeled plasmid pMIA10 (49) as probe, which contains the IAP sequences in a pBR322 backbone. DNA probe was chemically labeled with digoxigenin following the GeniusRM kit (Boehringer Mannheim) protocol. The tissue used was from a 5 day old embryo which was sectioned at 10 um. Sections were air dried before ®xation with ethanol:acetic acid (3:1) for 15 min and ethanol for another 5 min. The frozen embryo sections were rehydrated in decreasing concentrations of ethanol with a last wash in dH2 O to remove the OCT. Endogenous alkaline phosphatase activity was inhibited by dipping the sections in 0.2 M HCl for 8 min followed by dehydration in increasing concentrations of ethanol. Slides were incubated in denaturing solution for 5 min at 100 C and incubated with the appropriate dilution of the probe in hybridization solution at 37 C for 2 h. The hybridized probe was detected by incubation with a dilution of Fab fragments of an anti-digoxigenin alkaline phosphatase conjugate in 1% BSA in PBS (Boehringer Mannheim) for 2 h at 37 C. Substrate NTB/BCIP (both from Boehringer Mannheim) in the presence of levamisol was added and the product was detected after 2 h at 4 C when slides where counterstained with nuclear fast (Digene), dehydrated, mounted with Accu Mount mounting medium (Baxter) and coverslipped.

differentiate into trophectoderm, these cells highly express env. Thus, in this regard, these EC cells are similar to normal post-implantation mouse embryos. However, the Py LT-Ag expressing EC cells did not express env following differentiation. This established that both SV40 and Py LT-Ag expression are able to suppress differentiation induced ERV expression. Normal F9 EC cells are able to undergo morphological differentiation and form embryoid bodies, that resemble 3.5 day blastocyst, following 15 days of differentiation. As shown in Fig. 4, the F9LTneo cells were also able to undergo morpho-

197

Fig. 3. Establishment of env expression in F9 embryoid bodies containing a stably transfected Py large T plasmid. Dot blot with RNA extracted from F9 cells that were either differentiated with retinoic acid in the absence of LIF or maintained undifferentiated in the presence of LIF. RNA was hybridized to the env speci®c probe (38) designed from sequences published by Reuss and Schaller 1991 (48) and as used in Reuss et al. (38) for detection of env sequences in IAP. The probe represents the antisense strand of a conserved region in the IAPE env gene (nucleotides 2871 to 2890) that is found in all IAPE cDNAs where the env part of the genome has been sequenced. The probe was 32P end labeled using the protocol found in Maniatis (50) and hybridized to a dot blot containing RNA from samples as mentioned above.

Fig. 4. Development of embryoid bodies in vitro from F9 embryonal carcinoma cells. F9 embryonal carcinoma cells stably transfected with the plasmid pMGD20 neo (9,10) were differentiated by treatment with 10 ÿ 6 M retinoic acid and without LIF (leukemia inhibitory factor). The undifferentiated cells ( panel A) are grown in Eagle's minimal essential medium with 10% fetal bovine serum, L-glutamine, non-essential amino acids, penicillin/streptomycin, 10 ÿ 6 M b-mercaptoethanol, and 2000 U/ mL LIF to inhibit spontaneous differentiation into trophectoderm. Differentiated cultures form rounded clumps of cells after three days and will form hollow spherical structures (embryoid bodies panel B) resembling 3.5 day embryos, after 15 days of culture in retenoic acid (4,7,51,52).

198


logical differentiation and form normal appearing embryoid bodies. Because these embryoid bodies resemble normal blastocyst in many respects, including the differentiation of trophecotderm, and because we have shown that we can prevent IAPE-A env expression in the T-Ag expressing trophectoderm, it seemed possible that such cells could be used to generally examine the effects of T-Ag on implantation. However, few studies have previously examined the implantation activity of these EC derived blastocysts-like structures. Yet it has been reported that normal trophectoderm is not readilly rejected when implanted into the uterus and provides active protection from macrophage attack (41). Thus there was good reason to suspect that trophectoderm containing EC derrived blastocyst would conserve some implantation function, although it is also clear that F9 EC cells are not totipotent and are thus not expected to generate viable embryos. To determine this, we attempted to implant EC cell derrived embryoid bodies from T-Ag (and ERV) expressing and non-expressing EC cells in psudeopregnant female mice. The results of these experiments are shown in Fig. 5. With normal F9 EC blastocysts, there was clear evidence of at least a partially successful implantation event. As can be seen, these embryoid bodies do establish tight, although uneven contact with the uterine tissues and were retained for at least 12 days. Although it is also clear that development is abberent (with necrotic and infarcted areas), some regions of blastocyst-uterus interaction seem relatively normal. Clearly, these blastocysts have retained some implantation functions, although embryo development is clearly abnormal. However, when we attempted to implant embryoid bodies derrived from the blastocysts that did not produce IAPE-A env, no implantation was observed to occur. Only normal uterine tissues was observed with no reminants of the embryoid body. As the implantation studies were done asymetrically in that normal embryos were successfully implanted into the right uterine horn of each psudeopregnent mouse whereas the EC derrived embryoid bodies were implanted into the left uterine horn, it seems clear that the non-env expressing EC derrived blastocyst are defective for implantation. The speci®c effect of Py T-Ag on embryo implantation is dif®cult to acertain. Large T-Ag is a multifunctional protein, whose established activities include interactions with the viral DNA at the

Fig. 5. Histology of implantation in normal embryos, wt F9 embryoid bodies, and T-Ag F9 embryoid bodies. Implantation studies require the induction of superovulation. We used CB6F1 mice which were ®rst injected with pregnant mare serum gonadotropin (PMSG) intraperitoneally at a dose of 5 IU or 0.1 ml. hCG is the second gonadotropin that is administered to induce superovulation also at a dose of 5 IU or 0.1 mL and is injected intraperitoneally. The times that the PMSG and hCG are administered relative to each other is 48 h apart. These mice are then mated with stud males and the following day checked for plugs and separated. Preimplantation blastocysts are collected postcoitus at day 3.5. The uterine horn is identi®ed and cut to allow for cervical exposure which is used to harvest the blastocyst using a 20 gauge syringe. The isolated blastocysts (8/mouse) are then transferred to the uterus of an anesthetized plugged pseudopregnant ICR mouse 3.5 d post coitus. ICR mice are prepared by mating females in estrus with vasectomized or genetically sterile males (53). At 12 days post transfer into ICR mice, the uterine horn is dissected, sectioned and analyzed by histology (H&E stain) for signs of implantation or local in¯ammation. To study the possible activity of EC embryoid bodies for implantation, F9 EC cell embryoid bodies are prepared in culture as described previously and the embryoid bodies present at 15 days of culture in retenoic acid used in substitution for the 3.5 day blastocyst. In panel A, we observed that F9 pMGD20 neo derived embryo bodies (non-IAP expressing), were implanted into the right uterine horns of ICR mice and showed no evidence of in¯ammation or placentation and had apparently completely eliminated the embryoid body, while in panel B the normal F9 (IAP expressing) derived embryoid bodies showed signs of an aborted implantation in that infarcted, degenerated, and in¯amed placental tissue was present.

origin of replication, binding cellular PCNA-primase complex, ATPase and helicase activity, interactions with Rb and other cellular proteins. These complex activities could in principle affect many different host functions. However, that said, it is also clear that Py large T-Ag does not interfere with the post implantation development of most mouse tissues, even when expressed from the same promoter used

T-Ag Inhibits Implantation

by us. This suggest that T-Ag has had a rather speci®c effect on the early embryo not observed in ES cells of a blastocyst. As the early embryo is also unusual for its high level ERV production, which was inhibited by T-Ag, it therefore seems possible that T-Ag could be affecting embryo implantation by indirectly affecting methylation and ERV production. The number and complexity of ERV sequences present in the genomes of all placental mammals makes it very dif®cult to evaluate the combined role of these agents in the implantation of the embryo. It seems clear, however, that many such different sequences are expressed in the placenta and trophectoderm so determining which of these might be important for implantation is dif®cult. The Y chromosome is particularly abundant with ERV sequences (42,43). As stem cells do not form trophectoderm they cannot be used to alter trophectoderm expression. Altering trophectoderm expression patterns must be accomplished prior to blastocyst formation. Because EC cells will form both trophectoderm and some omnipotent stem cells, they allowed us to select for T-Ag expression while still allowing the formation of new trophectoderm. Thus, using Py T-Ag, we were able to generate trophectoderm cells that did not express IAPE-A env sequences, which we showed to be expressed in normal blastocysts. We have previously proposed that ERV expression in viviparous mammalian embryos is required for implantation and avoiding maternal immunorecognition. The rapidity of the embryoid rejection we now observe suggest that nonadaptive or innate events might be involved. Im¯ammatory reactions frequently appear to be involved in failed implantation (for references see (44)). We have suggested that ERV env gene product may regulate such rapid and cytokine mediated events, resulting in Th1 response (31), which is consistant with suggestion that normal pregnancy operates with a biased Th1 response (45). Thus the general Th1 bias observed in normal pregnency would not be unexpected. Others have recently reported that trophectoderm may induce immuno supression by affecting tryptophane levels in nearby immune cells (46). However, the recent identi®cation of the RD114 receptor for the env protein as the neutral amino acid transporter and the report that affecting amino acid transport induces immunosupression suggest that ERV expression may also be involved in this process (47). We suggest our

199

results are consistent with ERV production being required by normal trophectoderm for implantation. References 1. Strickland S. and Mahdavi V., Cell 15, 393±403, 1978. 2. Strickland S., Smith K.K., and Marotti K.R., Cell 21, 347±355, 1980. 3. Solter D., Shevinsky L., Knowles B.B., and Strickland S., Dev Biol 70, 515±521, 1979. 4. Howe C.C. and Overton G.C., Mol Cell Biol 6, 150±157, 1986. 5. Hojman-Montes de Oca F., Dianoux L., Peries J., and EmanoilRavicovitch R., J Virol 46, 307±310, 1983. 6. Tonjes R.R., Limbach C., Lower R., and Kurth R., J Virol 71, 2747±2756, 1997. 7. Martin G.R., Wiley L.M., and Damjanov I., Dev Biol 61, 230± 244, 1977. 8. Waters B.K. and Rossant J., J Embryol Exp Morphol 98, 99± 110, 1986. 9. Gassmann M., Donoho G., and Berg P., Proc Natl Acad Sci USA 92, 1292±1296, 1995. 10. Camenisch G., Gruber M., Donoho G., Van Sloun P., Wenger R.H., and Gassmann M., Nucleic Acids Res 24, 3707±3713, 1996. 11. Jaenisch R., Harbers K., Jahner D., Stewart C., and Stuhlmann H., J Cell Biochem 20, 331±336, 1982. 12. Jahner D., Stuhlmann H., Stewart C.L., Harbers K., Lohler J., Jaenish R. Nature 298, 623±628, 1982. 13. Harbers K., Schnieke A., Stuhlmann H., Jahner D., and Jaenisch R., Proc Natl Acad Sci USA 78, 7609±7613, 1981. 14. Lamb B.T., Satyamoorthy K., Li L., Solter D., and Howe C.C., Gene Expr 1, 185±196, 1991. 15. Doer¯er W., Biol Chem Hoppe Seyler 372, 557±564, 1991. 16. Heller H., Kammer C., Wilgenbus P., and Doer¯er W., Proc Natl Acad Sci USA 92, 5515±5519, 1995. 17. Zeschnigk M., Schmitz B., Dittrich B., Buiting K., Horsthemke B., and Doer¯er W., Hum Mol Genet 6, 387±395, 1997. 18. Chuang L.S., Ian H.I., Koh T.W., Ng H.H., Xu G., and Li B.F., Science 277, 1996±2000, 1997. 19. Baylin S.B., Science 277, 1948±1949, 1997. 20. Nelson J., Leong J.A., and Levy J.A., Proc Natl Acad Sci USA 75, 6263±6267, 1978. 21. Dirksen E.R. and Levy J.A., J Natl Cancer Inst 59, 1187±1192, 1977. 22. Wivel N.A. and Smith G.H., Int J Cancer 7, 167±175, 1971. 23. Stromberg K. and Benveniste R., Virology 128, 518±523, 1983. 24. Ueno H., Imamura M., and Kikuchi K., Virchows Arch A Pathol Anat Histopathol 400, 31±41, 1983. 25. Suni J., Wahlstrom T., and Vaheri A., Int J Cancer 28, 559±566, 1981. 26. Piko L., Hammons M.D., and Taylor K.D., Proc Natl Acad Sci USA 81, 488±492, 1984. 27. Venables P.J., Brookes S.M., Grif®ths D., Weiss R.A., and Boyd M.T., Virology 211, 589±592, 1995. 28. Larsson E., Andersson A.C., and Nilsson B.O., Ups J Med Sci 99, 113±120, 1994. 29. Lower R., Lower J., and Kurth R., Proc Natl Acad Sci USA 93, 5177±5184, 1996.

200


30. Urnovitz H.B. and Murphy W.H., Clin Microbiol Rev 9, 72±99, 1996. 31. Haraguchi S., Good R.A., and Day N.K., Immunol Today 16, 595±603, 1995. 32. Mangeney M. and Heidmann T., Proc Natl Acad Sci USA 95, 14920±14925, 1998. 33. Villarreal L.P., J Virol 71, 859±865, 1997. 34. Harris J.R., Bioessays 20, 307±316, 1998. 35. de Parseval N. and Heidmann T., J Virol 72, 3442±3445, 1998. 36. Blond J.L., Beseme F., and Duret L., et al. J Virol 73, 1175± 1185, 1999. 37. Seifarth W., Baust C., and Murr A., et al. J Virol 72, 8384±8391, 1998. 38. Reuss F.U., Frankel W.N., Moriwaki K., Shiroishi T., and Cof®n J.M., J Virol 70, 6450±6454, 1996. 39. Reuss F.U., J Virol 66, 1915±1923, 1992. 40. Linnenbach A., Huebner K., and Croce C.M., Proc Natl Acad Sci USA 77, 4875±4879, 1980. 41. Sionov R.V., Yagel S., Har-Nir R., and Gallily R., Biol Reprod 49, 588±595, 1993. 42. Eicher E.M., Hutchison K.W., Phillips S.J., Tucker P.K., and Lee B.K., Genetics 122, 181±192, 1989.

43. Fennelly J., Harper K., Laval S., Wright E., and Plumb M., Mamm Genome 7, 31±36, 1996. 44. Raghupathy R., Immunol Today 18, 478±482, 1997. 45. Lin H., Mosmann T.R., Guilbert L., Tuntipopipat S., and Wegmann T.G., J Immunol 151, 4562±4573, 1993. 46. Munn D.H., Zhou M., and Attwood J.T., et al. Science 281, 1191±1193, 1998. 47. Rasko J.E., Battini J.L., Gottschalk R.J., Mazo I. and Miller A.D., Proc Natl Acad Sci U S A 96, 2129±2134, 1999. 48. Reuss F.U. and Schaller H.C., J Virol 65, 5702±5709, 1991. 49. Mietz J.A., Grossman Z., Lueders K.K., and Kuff E.L., J Virol 61, 3020±3029, 1987. 50. Sambrook J., Fritsch E.F., and Maniatis T., Molecular Cloning, A Laboratory Manual, Cold Spring Harbor Laboratory Press, Cold Spring Harbor, 1989, Ed. 2. 51. Howe C.C., Lugg D.K., and Overton G.C., Mol Cell Biol 4, 2428±2436, 1984. 52. Doetschman T.C., Eistetter H., Katz M., Schmidt W., and Kemler R., J Embryol Exp Morphol 87, 27±45, 1985. 53. Hogan B., Manipulating The Mouse Embryo: A Laboratory Manual, Cold Spring Harbor Press, Plainview, New York, 1994, Ed. 2, pp. 1±332.