gestation) was adsorbed with MCMV (1 X 108 pfu/ml), then cultured in 100% rat serum for 3 days. The embryo was fixed in Bouin's solution and embedded in.
American Journal of Pathology, Vol. 127, No. 2, May 1987 Copyright © American Association of Pathologists
Murine Cytomegalovirus Infection of Cultured Mouse Embryos
YOSHIHIRO TSUTSUI, MD and ICHIRO NARUSE, PhD
From Department of Pathology and Department of Embryology, Institutefor Developmental Research, Aichi Prefectural Colony, Kasugai, Aichi,Japan
Isolated mouse whole embryos of 7.5 days' gestation were infected with murine cytomegalovirus (MCMV) and cultured in pure rat serum. Although the MCMV infection had little effect on the survival and development of the embryos during 3 days of cultivation, immunohistochemical analysis of their serial sections using monoclonal antibody showed MCMV-infected cells in various portions ofthe embryos. This monoclonal antibody, when tested with the use ofinfected cultured mouse fibroblasts, reacted with nuclear antigen within 2 hours after infection and also reacted with nuclear inclusions in the late phase of infection. The viral antigen-positive cells detected by the monoclonal antibody were present in almost all ofthe ectoplacental
cone and the yolk sac and in about 82% ofthe embryos. In the embryos, antigen-positive cells were frequently observed in the epithelium ofthe digestive tracts, endothelial cells of the blood vessels, and the mesodermal cells. In some of the embryos, viral antigen-positive cells were clearly observed in a small percentage ofthe blood cells. These findings indicate that blood cells, in addition to cell migration during embryogenesis, may play an important role in transmission of infectious virus into the embryos. Mouse whole embryo culture infected with MCMV can provide a model for the study of cellular tropism related to congenital infection by cytomegalovirus. (Am J Pathol 1987, 127:262-270)
CYTOMEGALOVIRUS (CMV), a member of the herpesvirus group, is a common cause of congenital viral infection in man, resulting in abortion and congenital abnormalities.' It is endemic throughout the world, occurring in about 1% of all newborn infants.2 Little is known about the mechanism and timing of fetal infection, because maternal infections are not usually clinically apparent. As specific animal models for fetal infection with CMV, guinea pigs infected with guinea pig cytomegalovirus (GPCMV) and mice infected with MCMV have been used. Transplacental transmission of CMV has been demonstrated to occur in the guinea pig.3'4 This transmission, however, does not appear to lead to fetal maldevelopment. In the case of MCMV, efforts to establish transplacental infection to fetus have not been successful,5 presumably because ofanatomic differences between trophoblastic membrane of mouse placenta and those of man and guinea pig.6 Another infection route is through the female genital tract by sexual contact with an infected male.7'10 This could lead to the infection of the embryo and to subsequent deleterious effects during fetal development. Recently Baskar et al (1983) demonstrated fetal maldevelopment in mice
whose mothers had been inoculated with MCMV in the endometrial lamina at the time of embryonic implantation. " Animal experiments have been of value in studying the effects of maternal infection on embryonic development.'2 It is, however, difficult to study the direct effect of viruses upon embryonic tissues. Culture techniques of postimplantation rodent embryos have recently been improved.'3 In 1983 Priscott'4 first infected reovirus 3 to cultured rat embryos. In this study we examined the effect of MCMV infection on embryogenesis of cultured mouse embryos and demonstrated the MCMV-infected cells in the embryos immunohistochemically using monoclonal antibody.
262
Materials and Methods Virus and Cell Cultures The Smith strain of MCMV, which had been passaged in mouse embryonic fibroblasts (MEFs) 33 Accepted for publication December 10, 1986. Address reprint requests to Yoshihiro Tsutsui, MD, Institute for Developmental Research, Aichi Prefectural Colony, Kasugai, Aichi 480-03, Japan.
Vol. 127 * No. 2
MCMV INFECTION OF CULTURED MOUSE EMBRYOS
times, was kindly provided by Dr. Y. Minamishima, Miyazaki Medical College, Miyazaki, Japan."5 The MEFs were prepared from 11-day-old embryos of ICR mice from Shizuoka Agricultural Cooperative Association for Laboratory Animals (Hamamatsu, Shizuoka, Japan) and were grown in Eagle's minimal essential medium (MEM) containing penicillin (100 units) and streptomycin (50,ug/ml) and 10% fetal calf serum (FCS). The titer of the virus stock was 1 X 108 pfu/ml. Embryo Culture and Virus Infection Embryo culture was performed basically according to the technique of Buckley et al'6 with minor modifications.'7 Timed pregnancies were taken as Day 0 when a copulation plug was found following overnight mating. Embryos were collected from ICR mice on Day 7.5 of gestation, corresponding to the headfold stage. For viral infection the embryos were incubated at 37 C in the virus stock solution which was adjusted to about pH 7.4 for 60 minutes with occasional shaking. For mock infection embryos were incubated in MEM containing 10% FCS in the same manner as above. After adsorption the embryos were separated to two or three per glass roller bottle containing 3 ml rat serum, which had been obtained by an immediate centrifugation of whole rat blood, heat-inactivated at 56 C for 30 minutes, plus antibiotics as for the cell culture. The bottles were gassed with a 5% 02, 5% C02, 90% N2 mixture for 1 minute before sealing and placing on a roller apparatus set to 30 rpm at 37 C. After 24 hours the bottles were regassed with 20% 02, 5% C02, 75% N2, and cultured for a further 24 hours. Then the bottles were again regassed with 40% 02,5% C02, 55% N2 and cultured until the termination of culture. The rat serum used for the culture was changed every 24 hours. Embryos were examined for heart beat, blood circulation of the yolk sac, and rotation of embryonic axis at the end of culture. After fixing with Bouin's fluid, the yolk sac was removed, and the embryo proper was examined under a dissection microscope for fusion of the allantois with the ectoplacental cone, somite number, crown-rump length, and gross abnormalities such as closure of the neural tube. Preparation of Monoclonal Antibody to MCMV-Infected Cells Six-week-old BALB/c mice immunized with sonicated nuclei of MCMV-infected cells were prepared as described previously.'8 Spleen cells from the immunized animals were fused with myeloma (NS-1)
263
cells.'9 Hybridomas were selected and cloned by ELISA with the use of nuclei of both the MCMV-infected and mock-infected MEFs.'8 Immunoprecipitation The MEF cells plated in a 25-sq cm flask 24 hours after infection with MCMV (or uninfected cells) were labeled with 35S-methionine (50,uCi/ml, Amersham) for 3 hours in methionine-free MEM supplemented with 10% FCS. Immunoprecipitation and gel electrophoresis were performed as described previously.'8
Immunofluorescence MEF cells plated on coverslips were infected with MCMV or left uninfected. At appropriate times after infection, the monolayers were fixed with 4% paraformaldehyde in 0.1 M phosphate buffer, pH 7.4, and postfixed with cold acetone.'8 The fixed cells were incubated for 30 minutes with cultured fluids from the hybridoma at 37 C, then washed with three changes of phosphate-buffered saline (PBS) and incubated with fluorescein isothiocyanate (FHTC)-conjugated goat anti-mouse IgG (Miles Laboratory) diluted 50-fold. After a final wash the preparations were dehydrated with ethanol, mounted with Entellan (Merck), and viewed through a fluorescence microscope.
Identification of Immediate-Early Proteins Cells were incubated in the medium containing 50 jug/ml cycloheximide (CH) for 30 minutes prior to infection. The cells were infected in the presence of CH and maintained in CH-containing medium for 4 hours after infection. The block of protein synthesized with CH was then released by extensive washings, and the cells were incubated for 2 hours prior to fixation for immunofluorescence. For immunoprecipitation, after release from the CH block, the cells were incubated in the presence of 50 ,uCi/ml of 35Smethionine in methionine-free medium.20 Immunohistochemistry The cultured embryos were fixed in Bouin's solution for a few days at room temperature and embedded in paraffin. Serial 5-,u-thick transverse sections were made from head to tail. The sections were treated with 0.15% periodic acid in distilled water at room temperature for 15 minutes to inactivate endogenous peroxidase,2' then washed in PBS three times. The sections were incubated with the cultured fluids
264
TSUTSUI AND NARUSE
AJP * May 1987
of the hybridoma to MCMV-infected nuclei (D-5) at room temperature for 45 minutes. After washing in PBS, the sections were incubated with peroxidaseconjugated anti-mouse IgG (Cappel Laboratories) diluted 100-fold for 45 minutes at room temperature. After washing in PBS, the sections were reacted at room temperature with 0.025% diaminobenzidine in 0.05 M Tris-HCl, pH 7.7, containing 10 mM hydrogen peroxide and 10 mM sodium azide. The reaction was stopped by washing in distilled water. The preparations were dehydrated with ethanol and mounted with Entellan.
Results Effect of MCMV Infection on Development of Cultured Mouse Embryos Table 1 summarizes the results of these experiments. The 7.5-day embryos absorbed with MCMV in the virus concentration of 1 X 108 pfu/ml for 60 minutes, or mock-infected for the control experiment were cultured in 100% rat serum. In the control experiment, heart beat and fusion of the allantois with the ectoplacental cone were observed until 24 hours after culture (8.5 days gestation), then axial rotation, blood circulation of the yolk sac, and closure of the neural tube until 48 hours after culture (9.5 days gestation). At termination of culture, the embryos were fixed, and the mean somite number and mean crown-rump length were measured. There was no statistically significant difference in embryogenesis between the embryos infected with MCMV and mock-infected embryos (Table 1). Degenerative and necrotic lesions appeared in both the infected and uninfected embryos when cultured more than 3 days. Characterization of the Monoclonal Antibodies to MCMV-Infected Cells Several hybridoma cell lines which produced antibodies reacting with nuclear lysate of MCMV-in-
fected cells, but not with that of uninfected cells, in ELISA were cloned. Among them we selected one hybridoma cell line (D-5) producing antibodies which recognized nuclear antigen of the MCMV-infected cells in the early phase of the infection. With this antibody, numerous tiny dots of fluorescence were detected in the nucleus of the cells 2 hours after infection with MCMV (Figure 1A). Dotted nuclear fluorescence became larger in size and decreased in number during the course of the infection (Figure l B and C). At the late phase of infection, nuclear inclusions also reacted with this antibody (Figure 1 D). In the cells treated for expression of the immediate-early proteins, as described in Materials and Methods, numerous tiny dots of nuclear fluorescence were observed (Figure 1 E). In uninfected cells, no fluorescence was observed with the monoclonal antibody (Figure IF). In the immunoprecipitation using 35S-methioninelabeled cell lysate from the cells 24 hours after infection, the monoclonal antibody D-5, which showed dotted nuclear fluorescence, precipitated polypeptides with molecular weights (mol wt) of 88,000 (88K), 76K, 50K, 39K, and 37K (Figure 2).
Immunohistochemical Detection of MCMV-Infected Cells in Cultured Mouse Embryos by Monoclonal Antibodies Serial sections of the cultured embryos were stained by the peroxidase-labeled antibody technique with the use of the monoclonal antibody D-5. Figure 3 shows three different sections from the same embryo cultured for 3 days. Viral antigen-positive cells were observed in the wall of the dorsal aorta and the adjacent mesoderm (Figure 3A and B), in the epithelium of the foregut (Figure 3C and D), and in the hindgut (Figure 3C, E, and F). The distribution of the viral antigen-positive cells detected in the cultured embryos by the monoclonal antibody D-5 is shown in Table 2. The experiments 48
Table 1 -Effects of MCMV Infection on Embryogenesis of 7.5-Day Mouse Embryos Cultured for 72 Hours*
Treatment of 7.5-day embryos
Closure of neural
Mean somite numbers
tubet
Mean crownrump length (mm ± SD)
6/6
5/6
2.75 ± 0.43
29.8 ± 0.4
7/8
7/8
3.16 ± 0.79
31.9 ± 6.9
Heart beatt
Fused allantoist
Axial rotationt
Yolk sac circulationt
Adsorption with MCMV Mock
6/6
6/6
6/6
infectiont
8/8
8/8
6/8
(SD)
*The 7.5-day embryos of ICR mouse were collected and incubated with MCMV stock with virus titers of 1 Xi 0 pfu/ml for 60 minutes at 37 C, then cultured for 72 hours in 100% rat serum as described in Materials and Methods. At the end of cultivation, the embryos were fixed in Bouin's fluid and examined under a dissection microscope. tExpressed as the number of positive embryos per number of embryos examined. tFor mock infection, embryos were incubated in MEM containing 10% FCS for 60 minutes prior to the start of cultivation.
MCMV INFECrION OF CULTURED MOUSE EMBRYOS
Vol. 127 * No. 2
265
Figure 1-Fluorescence micrographs of MEF cells adsorbed with MCMV by the use of monoclonal antibody D-5. After the adsorption, the cells were incubated for 2 (A), 4 (B), 6 (C), and 24 hours (D). The cells treated with cycloheximide (CH) before, during, and after the adsorption in order to accumulate mRNAs of the immediate-early proteins as described in Materials and Methods. After incubation for 4 hours in the presence of CH, the cells were washed extensively and incubated without CH for 2 hours (E). Mock infected cells were reacted with the antibody (F). (X720)
I
U
88K76K-
50K
39K 37K-
Figure 2-Immunoprecipitation of polypeptides of MCMV-infected cells with the monoclonal antibodies D-5. At 22 hours after the infection or mock infection, the MEF cells were labeled with 50 pCi 35S-methionine /ml for 2 hours. Cell extracts of infected cells (i) or uninfected cells (u) were immunoprecipitated with the monoclonal antibody D-5, and the labeled proteins were analyzed by polyacrylamide gel electrophoresis and autoradiography.
hours after infection and 72 hours after infection were done independently with the use of different litters of the mouse embryos. Viral antigen-positive cells were found in about 82% of the infected embryos. Although the ratio of the embryos which had antigenpositive cells in each location to all the infected embryos was almost the same in both experiments, the number of antigen-positive cells in each location was increased in the embryos 72 hours after infection. The antigen-positive cells were frequently observed in the walls of the vessels, the mesodermal regions, and the epithelium of the digestive tracts (Table 2). In contrast, the viral antigen-positive cells were not observed in the neural tubes and the other ectodermal regions, like epidermis, which cover the embryos at this stage of embryogenesis. In some embryos, the viral antigen-positive cells were also observed in some of the blood cells (Figure 4A and B) and the wall of the heart (Figure 4C). Although the type of antigen-positive cells in the blood cells could not be identified, the ratio ofthe nucleus to the cytoplasm was large, as in lymphoid cells. There was usually a continuity of the viral antigenpositive cells between the wall of the vessels and of the digestive tract via the positive cells in the mesodermal region when traced along the serial sections of the embryos (Figure 3D). Viral antigen-positive cells were also observed in the regions beneath the covering cells of the ectoplacental cones (Figure 4D) and the yolk sac (Figure 4E). Although the ectoplacental
266
TSUTSUI AND NARUSE
AJP * May 1987
_
Al
I
*
F
Figure 3-Transverse sections of a cultured mouse embryo infected with MCMV, stained with the monoclonal antibody D-5. The mouse embryo (7.5 days' gestation) was adsorbed with MCMV (1 X 108 pfu/ml), then cultured in 100% rat serum for 3 days. The embryo was fixed in Bouin's solution and embedded in paraffin. Then serial transverse sections were made. The sections were stained by the peroxidase-labeled antibody technique using the monoclonal antibody. Three photographs (A, C, and E) were different transverse sections from the same embryo, and parts of them were magnified to make clear the viral A and B-A section across myelencephalon (mc), auditory vesicles (av), and diencephalon (dc). Viral antigen-positive antigen-positive cells (B, D, and F). cells were observed in the wall of the dorsal aorta (da) and the mesoderm. C and D-A section across the pharynx (ph) and foregut (fg). The viral antigen-positive cells were observed in the epithelium of the foregut, the mesoderm, and the wall of the dorsal aorta. The positive cells were also seen in the E and F-A section across the hindgut. Viral antigen-positive cells were observed in the epithelium of the hindgut and the mesoderm but not in hindgut (hg). the neural tube (nt). (A, C, and E, X45; B, D, and F, X31 0)
MCMV INFECllON OF CULTURED MOUSE EMBRYOS
Vol. 127 * No. 2
267
Table 2-Distribution of MCMV Antigen Positive Cells in the Cultured Mouse Embryos* Hours after Adsorption with MCMV 48 hours 72 hours
Number of Positive Embryost
Vessel Wall and Heartf
Digestive
Neural
Blood CelIt
Mesodermt
Tractt
Tubet
Placentat
9/11 5/6
6/11 5/6
4/11 2/6
9/11 5/6
8/11 5/6
0/11 0/6
3/3 6/6
*The 7.5-day embryos were adsorbed with MCMV and cultured in the same manner as in Table 1. After infection, the embryos were fixed and embedded in paraffin and serial sections of the embryos were stained immunohistochemically with the use of monoclonal antibody D-5. tNumber of embryos that had viral antigen-positive cells per total embryos infected. fNumber of embryos that had positive cells in each location per total embryos examined.
cones and yolk sac were not examined in all cases, the viral antigen-positive cells were observed in all cases examined (Table 2). In one case cultured for 3 days, a cluster of the viral antigen-positive cells became a relatively large lesion (Figure 4F). This lesion expanded into the mesoderm with continuity to the wall of the vessel and consisted of cells that had nuclear inclusions and multinucleated giant cells. In the periphery of the lesion, dotted nuclear antigen-positive cells were observed that were similar to those observed in the early phase ofthe infection with the use of the MCMV-infected MEFs. These cells were present discontinuously from the main lesion, presumably because of rapid growth and migration ofthe embryonic cells.
Discussion In the present study, cultured mouse embryos were analyzed after infection with MCMV. Although maternal physiology and immunologic condition were neglected, it was possible to study direct effects of the virus upon the embryos in controlled experimental conditions. Direct viral infection on the embryos presented in this study seems to be significant in considering ascending infection by the genital tract.8-'0 In this study we used 7.5-day embryos corresponding to the headfold stage. Embryos in this stage have proved to be the most amenable to growth in culture during the period of organogenesis.'3 The time at which the virus gains access to the fetus is important in the prognosis of intrauterine infection with rubella and Toxoplasma. In the case of CMV, however, primary infection at any stage of pregnancy appears to present a risk,22'23 and data are not adequate to ascribe a more precise risk to each particular trimester.24 It was an unexpected finding that, even after exposure to a high concentration ofvirus (1 X 108 pfu/ml), the embryos not only survived but also developed in almost the same way as the controls. In the case of Reovirus 3, Priscott14 reported that the viral infection induced dose-dependent retardation of the develop-
ment of the cultured rat embryos. As MCMV infection had no clear effect on embryogenesis of the cultured mouse embryos during the 3 days of cultivation, it came into question whether the infectious virus had been able to gain access to the inside of the embryos. Immunohistochemical studies on serial sections of the embryos showed viral antigen-positive cells. Therefore, there is no complete barrier which blocks virus transmission into the embryos. However, virus antigen-positive cells were not detected in all the embryos, but in about 82% of the infected embryos. There might be some defense mechanism in the embryos against virus transmission. With the present culture techniques, it is not possible to observe embryogenesis of the 7.5-day embryos for more than 3 days, because necrotic lesions appear even in the embryos ofthe control experiments. Ifit were possible to culture for a longer period, MCMV infection might cause some effect on embryogenesis. Embryo sites which are directly exposed to the virus during virus adsorption should be the surface layers of the primitive endoderm and the ectoplacental cones, because these cells are located on the surface at this stage.25'26 The primitive endoderm may develop into parietal endoderm which covers the yolk sac,25 or may become visceral embryonic endoderm like the lining cells ofthe digestive tract by reversing the germ layers, ie, the endoderm on the inside, the ectoderm on the outside, and mesoderm between them.25 The fact that virus antigen-positive cells were frequently observed in lining cells of the digestive tracts could be explained by the reversion of the germ layer. It is noteworthy that in several embryos some blood cells had virus nuclear antigen. Furthermore, antigen-positive cells were also frequently observed in the walls of the vessels, and continuity of the viral antigen positive cells was often seen between the wall of the vessels and the epithelium of the digestive tract (Figure 3D). Because blood cells proliferate in the blood islands of the yolk sac26 and the blood circulation of the embryos begins about 24 hours after the viral adsorption, it is possible that the virus-infected cells are
268
AJP * May 1987
TSUTSUI AND NARUSE
............ '1l
.:.E ** ..
:: .!..jjj
....
.:
-w l'
-.
:..
....
.; ..
n.ti. l
.I.
Figure 4-Sections of cultured mouse embryos infected with MCMV, stained with the monoclonal antibody D-5. The mouse embryos (7.5 days' gestation) were A and B-Viral antigen-positive cells in the blood vessels. adsorbed with MCMV and cultured for 3 days in the same manner as described in Figure 3. D-Viral antigen-positive cells beneath the covering cells of the ectoplacental cone. C-Viral antigen-positive blood cell in the heart. (Xl 50) (X31 0) F-A rather large cluster of viral antigen-positive cells in the mesoderm of the embryo. (X31 0) E-Yolk sac. (X31 0) (X31 0)
Vol. 127 * No. 2
MCMV INFECTION OF CULTURED MOUSE EMBRYOS
carried through the blood circulation to the inside of the embryos. It has been reported that a small percentage of lymphocytes and monocytes could be infected with HCMV in vitro, as judged by expression of immediate-early27 and early viral protein detected by monoclonal antibodies28 and also detected by in situ hybridization using cloned HCMV DNA probe.29 In the murine experiments using MCMV, latent MCMV infection has been found in a small subset of lymphocytes and in macrophages.'3031 In the present study we directly demonstrated, for the first time, MCMV-infected blood cells in embryos. These results suggest that blood cells may serve as a vehicle for transmission ofinfectious virus to embryos, although it still remains a possibility that the blood cells that expressed immediate early antigen are the result of nonpermissive infection. In the present study, the viral antigen detected by the monoclonal antibody D-5 was located in the nucleus and detected within 2 hours after infection. This viral antigen was also expressed in the cells which were treated for expression of immediate-early antigens. By radioimmunoprecipitation and polyacrylamide gel electrophoresis this monoclonal antibody precipitated multiple polypeptides, including polypeptides of molecular weight similar to the immediate-early antigens already reported.32 It is still undetermined why this monoclonal antibody precipitates so many polypeptides. Because the immunohistochemical procedures using the monoclonal antibody D-5 in this study are so sensitive in detecting cells which express the viral antigens in the embryos, these procedures, in combination with in situ hybridization, would be effective in studying cellular tropism related to congenital infection by MCMV.
References 1. Weller TH: The cytomegalovirus: Ubiquitous agents with protein clinical manifestations. N Engl J Med 1971, 285:203-214. 2. Stagno S, Pass RF, Dworsky ME, Britt WJ, Alford CA: Congenital and perinatal cytomegalovirus infections: Clinical characteristics and pathogenic factors, CMV: Pathogenesis and Prevention of Human Infection. Edited by SA Plotkin, S Michelson, JS Pagano, F Rapp. New York, Alan R. Liss, 1984, pp 65-85 3. Choi YC, Hsiung GD: Cytomegalovirus infection in guinea pig: II. Transplacental and horizontal transmission. J Infect Dis 1978, 138:187-202 4. Kumar ML, Nankervis GA: Experimental congenital infection with cytomegalovirus: A guinea pig model. J Infect Dis 1978, 138:650-654 5. Johnson KP: Mouse cytomegalovirus: Placental infection. J Infect Dis 1969, 120:445-450 6. Enders AC: A comparative study ofthe fine structure of trophoblast in several hemochorial placenta. Am J Anat 1965, 116:29-67
269
7. Baskar JF, Stanat SC, Huang ES: Murine cytomegalovirus infection of mouse testes. J Virol 1986, 57:11491154 8. Chretien JH, McGinnis CG, Miller A: Venereal causes of cytomegalovirus mononucleosis. J Am Med Assoc 1977, 238:1644-1645 9. Jordan MC, Rousseau WE, Noble GR, Stewart JA, Chin TDY: Association of cervical cytomegaloviruses with venereal disease. N Engl J Med 1973, 288:932-
934 10. Lang DJ, Kummer JF: Cytomegalovirus in semen. Observation in selected population. J Infect Dis 1975, 132:472-473 11. Baskar JF, Stanat SC, Sulik KK, Huang ES: Murine cytomegalovirus-induced congenital defects and fetal maldevelopment. J Infect Dis 1983, 148:836-843 12. Lansdown ABG, Brown JD: Pathological observations on experimental cytomegalovirus infections in pregnancy. J Pathol 1978, 125:1-9 13. New DAT: Whole-embryo culture and the study of mammalian embryos during organogenesis. Biol Rev 1978, 53:81-122 14. Priscott PK: The growth of retrovirus 3 in cultured rat embryos and implications for human reproductive failure. Br J Exp Pathol 1983, 64:467-473 15. Ebihara K, Minamishima Y: Protective effect of biological response modifiers on murine cytomegalovirus infection. J Virol 1984, 51:117-122 16. Buckley SKL, Steele CE, New DAT: In vitro development of early postimplantation rat embryos. Dev Biol 1978, 65:396-403 17. Naruse I, Shoji R: Whole embryo culture as a primary screening system for teratogens. Proc Jpn Acad 1986, 62:31-34 18. Tsutsui Y, Yamazaki Y, Kashiwai A, Mizutani A, Furukawa T: Monoclonal antibodies to guinea pig cytomegalovirus: An immunoelectron microscopic study. J Gen Virol 1986, 67:107-118 19. Kohler G, Milstein C: Continuous culture of fused cells secreting antibodies of predefined specificity. Nature 1975, 256:495-497 20. Blanton RA, Tevethia MJ: Immunoprecipitation of virus-specific immediate-early and early polypeptides from cells lytically infected with human cytomegalovirus strain AD 169. Virology 1981, 112:262-273 21. Nakane PK: Recent progress in the peroxidase-labeled antibody method. Ann NY Acad Sci 1975, 254:203211 22. Monif GRG, Egan II EA, Held B, Eftzman DV: The correlation of maternal cytomegalovirus infection during varying stages in gestation with neonatal involvement. J Pediatr 1972, 80:17-20 23. Stagno S, Rass RF, Dworsky ME, Henderson RE, Moore EG, Walton PD, Alford CA: Congenital cytomegalovirus infection. The relative importance of primary and recurrent maternal infection. N Engl J Med 1982, 306:945-949 24. Ho M: Congenital and perinatal human cytomegalovirus infection, Cytomegalovirus: biology and infection. New York, Plenum Medical Book Co., 1982, pp 131-149 25. Gardner RL, Papaioannou VE: Differentiation in the trophectoderm and inner cell mass, The early development of mammals. Cambridge University Press, 1975, pp 107-132 26. Rugh R: Normal development of the mouse, The mouse: its reproduction and development. Mineapolis, Burgess Publishing Co., 1967, pp 44-101 27. Rice GPA, Schrier RD, Oldstone MBA: Cytomegalovirus infects human lymphocytes and monocytes: Virus expression is restricted to immediated-early gene products. Proc Natl Acad Sci USA 1984, 81:6134-6138
270
TSUTSUI AND NARUSE
28. Einhorn L, Ost A: Cytomegalovirus infection of human blood cells. J Infect Dis 1984, 149:207-214 29. Schrier RD, Nelson JA, Oldstone MBA: Detection of human cytomegalovirus in peripheral blood lymphocytes in a natural infection. Science 1985, 230:10481051 30. Brautigam AR, Dutko FJ, Olding LB, Oldstone MBA: Pathogenesis of murine cytomegalovirus infection: The macrophage as a permissive cell for cytomegalovirus infection, replication and latency. J Gen Virol 1979, 44:349-359 31. Olding LB, Jensen FC, Oldstone MBA: Pathogenesis of cytomegalovirus infection: I. Activation of virus from bone marrow derived lymphocytes by in vitro allogenic reaction. J Exp Med 1975, 141:561-572
AJP . May 1987
32. Keil GM, Fibi MR, Koszinowski UH: Characterization of the major immediate-early polypeptides encoded by murine cytomegalovirus. J Virol 1985, 54:422-428
Acknowledgments We thank Drs. Y. Minamishima and Y. Eizuru, Miyazaki Medical College, for their generous gift of MCMV and their helpful advice, Mrs. N. Kawamura and A. Kashiwai for excellent assistance, and also Mr. J. Aoki, Nagoya University School of Medicine, for excellent photographic preparation.
