Interferon RNA of embryonic origin is expressed transiently during ...

3 downloads 34 Views 2MB Size Report
Results from these pooled sam- .... Godkin, J. D., Bazer, F. W., Thatcher, W. W., and Roberts, R. ... Fincher, K. B., Bazer, F. W., Hansen, P. J., Thatcher, W. W.,.
THEJOURNAL OF BIOLOGICAL CHEMISTRY Vol. 263, No. 26, Issue of September 15, pp. 12801-12804,1988 0 1988 by The American Society for Biochemistry and Molecular Biology, Inc. Printed in U.S.A .

Communication Interferon RNA of Embryonic Origin Is Expressed Transiently during Early Pregnancy in the Ewe* (Received for publication, April 28, 1988)

Thomas R.Hansen, Kazuhiko Imakawa, H. Gregory PolitesS, Keith R. Marotti$, Russell V. Anthony, and R. Michael Roberts5 From the Departments of Animal Science and Biochemistry, University of Missouri, Columbia, Missouri 6521 1 and the $Division of Molecular Biology, The Upjohn Company, Kalamazoo,Michigan 49001

Ovine trophoblast protein-1 (oTP-l),an interferon of embryonic origin, is produced during the peri-implantationperiodof early pregnancy.Secretion of oTP-1 is detectable between days 13 and 21, but not beyond. In this study, the levels of oTP-1 mRNA in embryos were analyzed to determine if they reflected the transient natureof oTP-1 production. Totalcellular RNA (tcRNA) was isolated from embryos representing day 12 (n= 5), 14 (n = 7), 16 (n= 5), 18 (n = 6), 20(n= 4), and 22 (n= 5) of pregnancy and spotted on nylon membranes. ComplementaryRNA was transcribed from a specific oTP-1 cDNA (550 base pairs) templateandapplied (16-1000 pg)tonylon membranes to develop a standard curve. The fixed RNA samples were then allowed to hybridize with the 32Plabeled oTP-1 cDNA. oTP-1 mRNA was not detectable on day 12, increased tohigh levels (3.6 f 1.6 nglpg of embryo DNA) onday 14, decreased about &fold by day 16, 15-fold byday 18, 170-fold by day 20, and 200-fold byday 22 of pregnancy. Atday 14 oTP-1 mRNA comprised 0.060 f 0.019%of the tcRNA and was more abundant thanactin mRNA. Northern analyses of pooled tcRNA representing each day of pregnancy showed that the oTP-1 probe hybridized to a single class of mRNA (-1.1 kilobases) and confirmed the results obtained with dot blots.

essential in preventing a return to ovarian cyclicity. Infusion of oTP-1 into theuteri of nonpregnant ewes between days 12 and 20 of the estrous cycle extended the interestrous interval and was associated with continued secretion of ovarian progesterone (3). Although receptors for oTP-1 are present in the uterine endometrium (4), the mechanisms by which oTP1 functions to elicit maternal recognition of pregnancy are unclear. One effect of oTP-1 seems to be its ability to interfere with either the production or release of prostaglandin F2a (5), a substance which induces regression of the corpus luteum at about day 15 of the estrous cycle (6). Recently several cDNAs corresponding to oTP-1 mRNA were cloned from a day 15-17 sheep conceptus cDNA library (7). Unexpectedly, the inferred amino acid sequence of this cDNA revealed that oTP-1had about 50% sequence identity to human (8-lo), pig ( l l ) , mouse (12), andrat (13) interferons ) about 70% identity to bovine of the aI family ( I F N - ~ Iand I F N - ~ I(10). I This degree of similarity is significant since IFN-as diverge considerably in sequence between species. In addition to sequence identity with IFN-as, oTP-1 possesses antiviral (15)' and antiproliferative properties.' The specific antiviral activity of oTP-1, for example, is about 1.3 x lo" units/mg compared to about 10' units/mg for many recombinant IFNs (16, 17). Interestingly, day 14 sheep conceptuses produce at least 50-100 pg of oTP-1 daily ( l ) , and this protein is the major translation product of day 16 conceptus mRNA (18, 19). Although some normal cells and tissues, including the placenta, have been reported to produce low amounts of IFN, thereis to our knowledge nosystem which produces IFN at this magnitude in the absence of a known viral stimulus. The intent of this study was to examine the transient production of oTP-1 during pregnancy in the ewe. Earlier studies have indicated that secretion of the protein increases from day 13 to 17, declines from day 17 to 21, and is not detectable by day 23 of pregnancy. In this study, we have examined whether the intracellular content of oTP-1 mRNA reflects the protein secretion profile. EXPERIMENTALPROCEDURES

Animals and Collection of Embryos

Ovine trophoblast protein-1 (oTP-1)' was initially recognized asthe major polypeptide product secreted intothe medium when day 13-21 preimplantation sheep embryos were placed into short-term in vitro culture. oTP-1 has a relative molecular mass of about 18,000 as judged by electrophoresis in the presence of sodium dodecyl sulfate and consists of at leastthree to four isoelectric forms (1). oTP-1 has been strongly implicated in initiating maternal responses to the presence of an embryo within the uterus (2). In particular, secretion of oTP-1 after day 13 of pregnancy is believed to be

* This work was supported by Grant HD21896 from the National Institutes of Health. The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked "advertisement" in accordance with 18 U.S.C. Section 1734 solely to indicate this fact. § To whom correspondence should be addressed. ' The abbreviations used are: oTP-1, ovine trophoblast protein-1; tcRNA, total cellular RNA; IFN, interferon; Mops, 4-morpholinepropanesulfonic acid bp, base pair(s); kb, kilobase(s).

Mature Vi Fin X 'A Dorset X 1 h Rambouillet cross-bred ewes exhibiting normal day 16-17 estrous cycleswere maintainedon pasture and monitored for mating behavior twice daily using intact rams. Day of mating was designated day 0. Embryos were removed by flushing the uterus with sterile medium as previously described (1). Materials Guanidinethiocyanate, acetic acid, 2-mercaptoethanol, cesium chloride, and phenol were obtained from Fisher Biotech (Fair Lawn, NJ). Radiolabeled nucleotides (nick translation: [ o ~ - ~ ' P ] ~ C T800 P, Ci/mmol; end-labeled standards: [o~-~'P]ATP, 3000 Ci/mmol) were obtained from Du Pont-New England Nuclear. The minifluorometer used for determination of DNA concentration was purchased from Hoefer Scientific Instruments (San Francisco, CA). Restriction endonucleases and X HindIII-EcoRI-digested DNA standards were obtained from Promega (Madison, WI) and Bethesda Research Laboratories. The y-actin-containing Okayama-Berg expression vector (20,21) was kindly provided by Dr. L. H. Kedes (Stanford University R.M. Roberts, K. Imakawa, Y. Niwano, M. Kazemi, P. V. Malathy, T. R. Hansen, A. A., Glass, and L. A. Kronenberg, manuscript in preparation.

12801

12802

Interferon RNA of Embryonic Origin

School of Medicine, Palo Alto, CA). Other materials and sources were: Nensorb@column for nucleotide purification (Du Pont-New England Nuclear), Bluescribe M13@transcription vector (Stratagene, San Diego, CA), Biotrans@nylon membranes (ICN, Irvin,CA), nitrocellulose membranes (Micron Separations, Inc.), Kodak intensifying cassettes and XRP film (Eastman Kodak, Rochester, NY), rabbit globin mRNA (Bethesda Research Laboratories), Video densitometer (Bio-Rad), nick translation kit (Amersham Corp.). Reagents were of molecular biology grade.

RESULTS

Dot blots (not shown)in which the 32P-labeledoTP-1 cDNA probe was hybridized to a series of RNA samples indicated that thisprobe failed to bind to 2 pg of tcRNA extracted from porcine tissue andto rabbit globin mRNA. However,the oTP1probe bound strongly to RNA that had been extracted from day 16 ovine embryos and then purified by centrifugation through CsC1. The oTP-1 cDNA probe also hybridized with cRNA standards prepared by transcription from the oTP-1 Methods cDNA. In a parallel experiment with identical dot blots, the Isolation of RNA-Total cellular RNA was obtained from ovine embryos representing day 12 ( n = 5), 14 ( n = 7), 16 ( n = 5), 18 ( n = RNA samples were allowed to hybridize with a 32P-labeled 51,20 ( n = 4), and 22 ( n = 5) of pregnancy by a modification of actin cDNA. This probe gave a strong signal with the CsC1procedures described by Davis et al. (22). Briefly, embryos were purified ovine embryo RNA and porcine ovary RNA, but, as homogenized in 4 M guanidine thiocyanate, and nucleic acids were expected, failed to bind the globin mRNA and the oTP-1 precipitated with ethanol in the presence of25mM acetic acid and cRNA standards. 0.1 M 2-mercaptoethanol. The pellets were resuspended in buffer (50 RNA isolated from individual embryos on each day of mM potassium citrate, 1mM EDTA, 1%(w/v) sodium dodecyl sulfate, pregnancy was spotted on nylon membranes and hybridized pH 7.6), and an aliquot was removed for fluorometric (Hoechst 33258 with either the oTP-1 or actin probe. Resulting autoradidye) determination of DNA (23). After extraction with phenol/chloographs were scanned with a densitometer to obtain quantiroform and precipitation with ethanol, the resulting nucleotide pellet was resuspended in sterile distilled water and stored at -80 "C. RNA tative results (Table I). Optical density values of the signals concentrations were determined spectrophotometrically. Ratios of for each day of pregnancy were combined to provide average A z o / A ~averaged 1.86 0.09. To provide material for negative and values. No hybridization signal with the oTP-1cDNA probe positive controls for dot blot and Northern analysis total cellular was obtained with day 12 conceptus RNA. However, a strong RNA was extracted from porcine ovarian tissue and pooled (n = 5) signal was found on day 14. Thereafter, the strength of the day 16 ovine embryos, and theRNA was purified on a cesium chloride hybridization signal declined 200-fold byday 22. At this stage gradient (24). a faintpositive signal was detected in only two of five concepcDNA Probes-A specific oTP-1 cDNA (kaz 2 clone) was isolated tuses. In contrast to results with oTP-1, actin mRNA was from our ovine embryo X g t l l cDNA expression library (7) by using an oTP-1 antibody (4), subcloned into pBR322, and sequenced (7). high on day 14 and declined only 10-fold by day 22 when a The kaz 2 clone represented nucleotides 528-972 of the full-length positive signal was evident in all conceptuses. However, the (972 bp) cDNA. This clone contained 54% of the coding region and hybridization signal for actin was belowthe levels of detection the entire 3' end of the oTP-1 cDNA. A full-length cDNA (2.1 kb) on day 12. From the specific activities of the actin cDNA for human fibroblast cytoplasmic y-actin was isolated by digestion probe (length -2.1 kb; 1.0 X lo7 cpm/pg) and oTP-1 cDNA with BamHI from an Okayama-Berg expression vector (20,21). This probe (length 550 bp; 0.7 x lo7 cpmlpg) and assuming that clone recognizes all actin sequences from higher eukaryotes. oTP-1 the efficiency of hybridization of the probes to their respective and actin cDNA inserts were purified on a Nensorb@column and RNAs was about the same, we calculate that on day 14, there nick-translated (24) in the presence of [cP~*P]~CTP specific to activwere about four times as many oTP-1 mRNA molecules as ities of approximately 0.7 X IO7 and 1.0 X lo7cpm/pg, respectively. Filter Hybridization-RNA (2 pg) was spotted on duplicate nylon actin mRNA molecules/cell. oTP-1 optical density values were converted to picograms membranes with a dotblot apparatus. Membranes were prehybridized of mRNA by comparing values obtained from oTP-1 cRNA in 75 mM sodium citrate (pH 7.0), 0.75 M sodium chloride, 50 mM sodium phosphate (pH 7.0), 0.1% (w/v) sodium dodecyl sulfate, 0.1% standards (16-1000 pg) (Table I). Care was taken to ensure (w/v) Ficoll, 0.1% (v/v) polyvinylpyrrolidine, 0.1% (w/v) bovine that each densitometric value fell within the linear (R2= 1.0) serum albumin, and 150 pgof herring sperm DNA/ml for 4 hat 42 "C. region of the oTP-1cRNA standard curve. oTP-1 mRNA/pg Membranes were then hybridized in the same buffer with a 32P- tcRNA was not detectable at day 12 but reached its highest labeled cDNA probe for 18 h at 42 "C. Membranes were washed in 15 value 2 days later at day 14. Thereafter, values declined over mM sodium citrate (pH 7.0), 0.15 M sodium chloride, and 0.1% (w/v) sodium dodecyl sulfate three times at 42"C (5 min/wash) or until 2-fold by day 16, 4.5-fold by day 18, and 14-fold by days 20 background counts/min werelow, and placed on XRP film and and 22. By comparison of the autoradiographic signals given by the conceptus RNA with those given bythe standard oTPexposed in intensifying cassettes overnight at -80 "C. Transcription Vector-The oTP-1 cDNA insert was also subcloned 1 cRNA samples, it can be calculated that at day 14, oTP-1 into the Lac Z region of the pBS M13 transcription vector. This mRNA comprised about 0.06% of the totalcellular RNA and vector transcribed cRNA specified by the oTP-1cDNA insert. Typi- at day 20 approximately 0.003%. Fig. 1 shows the calculated cally each molecule of cDNA (550 bp) template yielded about 15 amounts of oTP-1 mRNA present per pg of conceptus DNA. cRNA molecules. Varying concentrations (16-1000 pg) of cRNA were To confirm that the two labeled probes were each hybridspotted on nylon membranes to develop a standard curve. izing to only a single size class of mRNA and that the mRNA Northern Transfer-A2-pg sample of total cellular RNA was pooled from each embryo for each day of pregnancy for Northern was not partiallydegraded, Northern analyses were performed blots (22). Two pg of each pool of total cellular RNA was denatured on 2-pg samples of pooled RNA representing each day of in 2.2 M formaldehyde, loaded on a 1.5% (w/v) agarose gel (1.1 M pregnancy (Fig. 2, A and B ) . The oTP-1 cDNA bound to an cDNA formaldehyde, 20 mM Mops, 5 mM sodium acetate, pH 7.0), and mRNA of approximately 1.1kb in length, and the actin separated electrophoretically. Transfer of the RNA from the gel to to one of2.1 kb in length. These size values are in good nitrocellulose filters (22) was conducted overnight in 0.15 M sodium agreement with the calculated length of the mRNAs (0.97 and citrate (pH 7.0) and 1.5 M sodium chloride. After baking the filters 2.1 kb, respectively). The results also confirmed that oTP-1 (2 h, 80 "C), hybridization was conducted as aforementioned. HindIIImRNA was very low or absent a t day 12, was most abundant and EcoRI-digested X DNA was end-labeled with 32Pand used for at days 14 and 16, and declined to undetectable levels as a base pair markers for determination of the size of message. Statistical Analysis-The general linear mode1:PM option ( t test) proportion of total RNA by day 22 of pregnancy. The actin of the Statistical Analysis System (14) was used to analyze the mRNA was again detectable at all stages of conceptus develsignificance of day of pregnancy effects on embryo oTP-1 and actin opment tested except day 12. Results from these pooled samples also suggested that actin mRNA levels were higher at RNA levels.

*

Interferon RNA of Embryonic Origin

12803

TABLE I Embryo oTP-1 and actin mRNA levels during early pregnancy in the ewe Two-pg samples of tcRNA were isolated from individual embryos and spottedon nylon membranes and hybridized as described under "Experimental Procedures" with either the oTP-1 or y-actin cDNA probe. Optical density (OD) of the signals from resulting autoradiographswas determined with a densitometer; and for the oTP1 probe, converted to picograms of mRNA by comparing values obtained from the oTP-1 cRNA standards (161000 pg). Results are expressed as the mean f S.E. of values for individual embryos on each day of pregnancy. Means with different superscripts differ ( p < 0.05). Day of Dremancv

Embryo DNA

Number Of embrvos

Pg

12 14 16 18 4 20 22

Embryo RNA recovered

oTP-1 mRNA' DNA OD/pg OD/Pg DNA

Pi?

Actin mRNA'

oTP-1 mRNA PdPg tcRNA

ND2 ND 10.0 7 10.73 & 9.13 4.77" f 4.78" 589 f 193" 5 k3.31 0.68' f 1.91".' 267 f 91"*' 118.66 f1.24 0.45'.' f 0.90' 129 f 40'*' &0.91 0.05' f24 0.82' f 7d 5 0.05 f0.76 0.02' f42 0.22' f 15c*d Note that the oTP-1probe was only 550 bp in length and had a specific activity of 7 X lo6 cpmlpg or about 2.8 X lo6 cpm/pmol, while the actin probe was essentially full length (2.1 kb) and had a specific activity of lo7 cpmlpg or about 1.34 X lo7cpm/pmol. ND, not detectable.

5

3.6

3.7 f 2.0" f 1.9" 42.3 f 66.1 15.1' f 39.6"' 79.8' 159.8 f72.9 516.6 f 112.7d 338.9

f 0.4" 47.8 f 16.1' f2.05 12.0' 159.4 f 0.6 107.7' f0.06 28.1' f 101.0'

12

14 1 6 18 20 2 2 DAY OF PREGNANCY

FIG. 1. Amount of oTP-1 mRNA relative to total cellular RNA in embryos during early pregnancy. Densitometric scanning of autoradiographs provided optical density values for signals representing sach day of pregnancy. Optical density was converted to picograms of oTP-1 mRNA through the use of the oTP-1 cRNA standards. Values (oTP-1 mRNA/pg of embryo DNA) represent the mean f S.E. for individual embryos on each day of pregnancy.

days 14-18 than at earlier or later stages of pregnancy. The apparent discrepancies in the relative strengths of the signals from the Northern hybridizations (Fig. 2) with those from the dot blots (Table I) most probably relate to the fact that the former experiment was carried out with pooled RNA samples. DISCUSSION

The sheep embryo exists as a small 2-mm spherical blastocyst at day 12 of pregnancy. Generally by day 13, the time in pregnancy at which the embryo is first believed to signal its presence (2, 6), the blastocyst has expanded to reach a diameter of about 8 mm.However,by day 14 a dramatic change in morphology is evident. At this stage the embryo has elongated to a longer threadlike form comprised mainly of trophectoderm and underlying extraembryonic endoderm. As development proceedsin a non-twin pregnancy the trophoblast (preplacenta) extends into both uterine horns. Attachment is initiated during this period of elongation but actual implantation does not begin until after day 30 when placentomes form. Previous studies have shown that oTP-1 production was detectable by day 13 and continued until day 21 but not

FIG.2. Northern blot analysis ofembryo total cellular RNA with "P-labeled cDNAs for oTP-1 ( A ) and yactin ( B ) .Two pg of tcRNA from each embryo in Table I was pooled foreach day of pregnancy. Two pgof the pooled embryo RNA was denatured in formaldehyde, loaded on 1.5% (w/v) agarose gels, and separated electrophoretically. Also loaded and separated on gels wasthe positive control (ovine embryo tcRNA ( O E ) )and 32P-labeledDNA standards. Transfer to nitrocellulose membranes and hybridization was conducted as described under "Experimental Procedures." Exposure of the membranes to Kodak XRP film was for 19 h. significantly beyond that time (1).In a more quantitative study, involving immunoprecipitationof metabolically labeled oTP-1 from the culture medium, Hansen et al. (18) showed that oTP-1production increased over 25-fold between day13 and 17 and then declined by over 50% by day 21. Although only three stages of conceptus development (days 13,17, and 21)were studied by Hansen et al. (18), the results were consistent with the present study. Here it is shown that oTP1mRNA/pg of tcRNA was very lowor absent at day 12, rose markedly by day 14, and then declined to low levels by day 20. Byday 22 oTP-1 mRNA wasnot detectable in the majority of samples. The results suggest that the changing rate of synthesis of oTP-1 during the critical day 12-22 period of pregnancy is at least partly a reflection of the amount of mRNA present. Since immunostaining has revealed oTP-1 to be confined to the trophectoderm (4) which comprises the outer epithelium of the elongating trophoblast, it can be assumed that oTP-1 mRNA is also localizedto those cells. If the amount of oTP-1 mRNA was expressed per cell (on the basis of total conceptus DNA) rather than in relation to tcRNA, a more dramatic picture emerges. Onthis basis, oTP-1 mRNA peaked

12804

Interferon RNA of Embryonic Origin

on day 14 and declined over 200-fold by day 22. However, it should be emphasized that trophectoderm most likely comprises a far higher percentage of the total number of cells at day 14 than at day 22. It is, therefore, difficult to assess by what proportion mRNA levels had changed in the synthetically active cells themselves. This problem is presently being investigated by in situ hybridization procedures. The question also arises as to whether actin should be regarded asan appropriate “housekeeping” gene in these studies. Dot blots revealed that levels of actin mRNA were high during days 14-22. There were also indications from the Northern blots of higher levels of mRNA/pg of tcRNA at days 14-18 than at days 20 and 22. However, actin mRNA levels were clearly very low at day 12. The low actin mRNA levels in day 12 conceptuses could be due to a lower recovery of intact RNA than on subsequent days of pregnancy. An alternative and more likely explanation for low actin mRNA levels in day 12 embryos relates to the morphological transition that occurs after day 12. During the analogous period in the pig, there is a massive reorganization of the cytoskeletal system (25), and very high rates of incorporation of radioactive amino acids into cellular actin can be observed as the embryos e l ~ n g a t e . ~ seems It likely that similar biochemical changes occur in the sheep when the spherical blastocyst begins to elongate at around day 13. We speculate that the levels of actin mRNA reflect this development transition. The amount of oTP-1 thatis synthesized by conceptuses is remarkable when it is considered that this proteinis an interferon. A t least 50-100 pg/day is being synthesized by each conceptus in the day 15-17 period (l),and itis the major translation product of the RNA at this time (18, 19). Its message constitutes approximately 0.06% by weight of the total RNA of day 14 conceptuses, and thismRNA, for reasons discussed above, isprobably confined to justone cell type, the trophectoderm. By contrast, actin mRNA which is probably ubiquitous to all cells, is present in lower amounts (at least 4fold less abundant on a molar basis) than the oTP-1mRNA in embryos at day 14 of pregnancy. It is not presently clear whether these amountsof oTP-1 mRNA are entirelythe result of high rates of transcription of the oTP-1 genes or are due to other factors such as increased message stability. It is also possible that high rates of oTP-1 production result from efficient translation of the message. Curiously, IFN-Ps are characterized by having unstable mRNAs with short halflives (26). Expression is usually transient andunder transcriptional control (27). Virus induction, for example, appears to involve a series of cis elements upstream from the mRNA transcription start site (28). It remains to be seen whether the genes for oTP-1 possess these viral control sequences and whether oTP-1 mRNA is unusually stable. Acknowledgments-We thank V. I. Burke, M. G. Burke, H. Francis, C. K. Neubauer, Dr. D. H. Keisler, Dr. P. V. Malathy, Dr. M. Kazemi, 3T. W. Raub, R.V. data.

Anthony, and R. M. Roberts, unpublished

Dr. J. Cross, and Dr. N. H. Ing for help with the surgeries and conceptus RNA isolation and Gail Foristal for typing the manuscript. We are grateful to Dr. L. Kedes for supplying the plasmid containing the actin cDNA. REFERENCES 1. Godkin, J. D., Bazer, F. W., Moffatt, R. J., Sessions, F., and Roberts, R. M. (1982) J. Reprod. Fertil. 6 5 , 141-150 2. Bazer, F. W., Vallet, J. L., Roberts, R. M., Sharp, D. C., and Thatcher, W. W. (1986) J. Reprod. Fertil. 7 6 , 8 4 - 8 5 0 3. Godkin, J. D., Bazer, F. W., Thatcher, W. W., and Roberts, R. M. (1984) J . Reprod. Fertil. 7 1 , 57-64 4. Godkin, J. D., Bazer, F. W., and Roberts, R. M. (1984) Endocrinology 114,120-130 5. Fincher, K. B., Bazer, F. W., Hansen, P. J., Thatcher, W. W., and Roberts, R. M. (1986) J. Reprod. Fertil. 7 6 , 425-433 6. Bazer, F. W., and First, N.L. (1983) J . Anim. Sci. 57,425-460 7. Imakawa, K., Anthony, R. V., Kazemi, M., Marotti, K.R., Polites, H. G., and Roberts, R. M. (1987) Nature 330,377-379 8. Goeddell, D. V., Leung, D. W., Dull, T. J., Gross, M., Lawn, R.

M., McCandliss, R., Seeburg, P. H., Ullrich, A., Yelverton, E., and Gray, P. W. (1981) Nature 290,20-26 9. Lawn, R.M., Gross, M., Houck, C. M., Franke, A.E., Gray, P. V., and Goeddel, D.V. (1981) Proc. Natl. Acad. Sci. U. S. A . 78,5435-5439 10. Capon, D. J., Shepard, H. M., and Goeddel, D. V. (1985) Mol. Cell. Biol. 5,768-779 11. Lefevre, F., and LaBonnardiere, C. J. (1986) J. Interferon Res. 6 , 349-360 12. Shaw, G. D., Boll,W., Taira, H., Mantei, N., Lengyel, P., and Weissmann, C. (1983) Nucleic Acids Res. 1 1 , 555-573 13. Dijkema, R., Pouwels, P., de-Reus, A., and Schellekens, H. (1984) Nucleic Acids Res. 1 2 , 1227-1243 (1985) Statistics, SAS 14. Statistical Analysis System User’s Guide Institute, Inc., Cary, NC 15. Pontzer, C. H., Torres, B. A., Vallet, J. L., Bazer, F.W., and Johnson, H. M. (1988) Bwchem Bwphys. Res. Cornrnun. 1 6 2 , 801-807 16. Pestka, S., Langer, J. A., Zoon, K. C., and Samuel, C. E. (1987) Annu. Rev. Biochem. 5 6 , 727-777 17. Rehberg, E., Kelder, B., Hoal, E. G., and Pestka, S. (1982) J. Biol. Chem. 257,11497-11502 18. Hansen, P. J., Anthony, R. V., Bazer, F. W., Baumbach, G. A., and Roberts, R. M. (1985) Endocrinology 1 1 7 , 1424-1430 19. Anthony, R.V., Helmer, S. D., Roberts, R.M., Hansen, P. J., Thatcher, W.W., and Bazer, F. W. (1988) Endocrinology, in press 20. Okayama, H., and Berg, P. (1983) Mol. Cell. Biol. 3,280-289 21. Gunning, P., Ponte, P., Okayama, H., Engel, J., Blau, H., and Kedes, L. (1983) Mol. Cell. Biol. 3 , 787-795 22. Davis, L. G., Dibner, M.D., and Battey, J. F. (1986) in Basic Methods in Molecular Biology, pp. 129-136, Elsevier Scientific Publishing Co., Inc., New York 23. Labarca, C., and Paigen, K. (1980) Anal. Biochern. 102,344-352 24. Maniatis, T., Fritsch, E. F., and Sanbrook, J. (1982) Molecular p. 545, Cold SpringHarbor Cloning, ALaboratoryManual, Laboratory, Cold Spring Harbor, NY 25. Geisert, R. D., Brookbank, J. W., Roberts, R. M., and Bazer, F. W. (1982) Biol. Reprod. 27,941-955 26. Raj, N. B. K., and Pitha, P. M. (1981) Proc.Natl. Acad. Sci. U. S. A . 78, 7426-7430 27. Raj, N. B. K., Kellum, M., Kelly, K. A. Antrobus, S., and Pitha, P. M. (1985) J. Interferon Res. 6,493-510 28. Goodbourn, S., Zinn, K., and Mantiatis, T. (1985) Cell 41,509520