With Murine Cytomegalovirus - NCBI - NIH

Lymphoid Cell Necrosis, Thymic Atrophy, and Growth Retardation in Newborn Mice Inoculated With Murine Cytomegalovirus Jared N. Schwartz, MD, Charles A. Daniels, MD, PhD and Gordon K. Klintworth, MD, PhD

During studies on the effect of murine cytomegalovirus on the developing retina, virus was inoculated into the eyes of newborn Swiss mice, and the animals were sacrificed at various times thereafter. Controls consisted of mice inoculated with ultraviolet-inactivated murine cytomegalovius and uninjected mice. Marked lymphoid ceil necrosis, thymic atrophy, pronoumced growth retardation, bacteremia, and death occurred in the animals inoculated with live virus. This virus-induced injury resulted in a marked depletion of lymphocytes in the subcapsular and cotical areas of the thymus as well as in the spleen, lymph nodes, and Peyer's patches. Areas of necrosis with viral inclusions were present at the site of inoculation and in various other organs including the spleen and bone marrow. Since growth retardation has been associated with thymic atrophy due to other causes, the observed abnoma physical development in the present study was interpreted as a sequel to the thymic injury. An implication of this study is that some human infants with concomitant immune deficiency and viral infection may have a primary viral disease with resultant secondary lymphoid tissue alterations, rather than a thymic disorder with a subsequent viral infection. (Am J Pathol 79:509-522, 1975)

SLNCE HTUmAN crTomFGALovnIus (CMV) is pathogenic only in the human host,1 the infection caused in mice by murine CMIV has been used as a model in studying certain aspects of the human infection.2-9 Investigators have primarily studied the pathogenesis of CMV infections in adult mice. Attempts to produce a prototype of the congenital human CMV infection have been tried by inoculating pregnant mice, but these have been unifornly unsuccessful.8'9 Thus a good model for human congenital CMIV infection is not presently available. Chorioretinitis is a common sequela of human CMV infections.1 While studying the effect of the murine virus on the developing eye in newborn mice, we observed marked growth retardation, bacteremia, and thymic atrophy in the infected animals. Considering that the CMV-induced From the Department of Pathology, Duke University Medical Center, Durham, NC. Supported in part by Grants EY-00146-03, EY-CA-00881-03, and 2-TO1-GM-00726-13 and Contract NIH-NIDR-72-2404 from the US Public Health Service; Dr. Klintworth is the recipient of Research Career Development Award EY-44795-04 from the National Eye Institute and an R. P. B. Louis B. Mayer Scholarship. Accepted for publication, February 5, 1975. Address reprint requests to Dr. Jared N. Schwartz, Department of Pathology, Box 3712, Duke University Medical Center, Durham, NC 27710. 509

510

SCHWARTZ ETAL

American Journal of Pathology

thymic atrophy observed might be related to the growth retardation and predisposition to severe infection seen in the congenital human disease, experiments were undertaken to study in greater detail the effect of murine CMV on the developing immune system of the mouse. Materials and Methods Virus and Assay Murine CMV was prepared by injecting ICR/HA (Swiss) mice (Blue Spruce Farms, Altamont, NY) intraperitoneally with 1050 plaque-forming units (PFU)

of the Smith virus strain and harvesting the salivary glands 3 weeks after inoculation.2 These glands were homogenized, sonicated, and centrifuged at 3100g for 10 minutes. The supernatant fluid contained 1076 PFU/ml as assayed on mouse embryo cell cultures.'0'1' A portion of this virus pool was inactivated with ultraviolet light 2 and used as described below. Animal Studies Five microliters of virus (104.3 PFU) were injected into both eyes of 60 Swiss mice (less than 24 hours of age) (Charles River, Wilmington, Mass) with a Hamilton syringe (Whittier, Calif) equipped with a 30-gauge needle. Twenty-five other animals were injected with an iden;tical volume of the inactivated murine CMV, and an equal number of mice were used as uninoculated controls. The animals were observed and weighed daily. After variable time periods (1 to 14 days), 3 to 5 mice from the experimental group and each of the two control groups were randomly selected and sacrificed with ether. Since no evidence of disease was noted in either of the two control groups, for statistical purposes the two control groups were considered as one when compared to the infected animals. Preparation of Tissues

When animals died or were sacrificed, complete necropsies were performed with the aid of a Zeiss stereo-operating microscope (6 x magnification). All tissues were examined, but only those pertinent to this study will be discussed. The thymus, spleen, heart, kidneys, and eyes were removed, and their weights were individually recorded. Representative sections of thymus, spleen, intestine, mesenteric lymph nodes, and bone marrow were submitted for light microscopic examination. The organs were immersed in Zenker's fixative (4 hours, 27 C), embedded in paraffin, sectioned at 6 [t, and stained with hematoxylin and eosin. Employing sterile technics, heart blood was obtained from the sacrificed animals and those found dead. Approximately 0.1 ml of blood from each mouse was inoculated on petri dishes with blood agar medium and in screw-capped tubes containing thioglycolate broth.

Results Mortality

Of those mice inoculated with infectious virus that died, the cumulative mortality throughout the 14-day observation period was 94%. Ap-

MURINECYTOMEGALOVIRUS

Vol. 79, No. 3 June 1975

511

proximatelv three-quarters of these animals succumbed at 3 days, and 84% of the fatalities were recorded bv the fourth experimental dav. None of the 50 controls died during the studv. General Appearance

No difference between the animal groups was noted until 2 days after birth. At this time, periorbital edema was present at the inoculation sites of the mice injected with infectious virus. The infected animals were less active, with sluggish movements; however, thev suckled and their mothers attended them in a normal fashion. As the animals aged, phvsical differences between the groups became more apparent (Figures 1 and 2). In particular, after the sixth dav the skin of infected animals was dry and 'wrinkled; hair was short in length, sparse in distribution; and body growth was stunted (Text-figure 1). Gross Anatomic Findings

Beginning at 2 davs, the infected animals possessed small soft thymuses that had a gray tinge instead of their usual pearly-white color. Throughout the experiment, mice that died had smaller thvmuses than infected animals which were sacrified. Both of these infected groups, however, had significantly smaller thvmuses than the controls wvhen the organ size was expressed either as the chronologic age or as a percentage of the animal's body weight (Text-figure 2). Beginning at 4 days, the spleens of mice inoculated with virus were conspicuouslN paler than normal and small when considered as a percentage of bodv weight. By 2 weeks, however, the organ color and size was not markedly different from controls of the same weight (Text-

84 TEXT-FIG 1-Body weights of control and infected mice expressed as the mean weight + standard error of the mean. 2

8 , o

l

BIRTH

2

4

6 AGE (days)

8 v

IA la


SCHWARTZ ETAL

512

50-

Control-

I z

Infected-_---

F

E 3

'

10-

TEXT-FIG 2-The mean absolute and relative weights of the spleen and thymus of infected and control animals expressed as standard error of the

IRTH.

1.0 08 L6 0

ZU.

BIRTH

2

4

6

8

14

AGE (days)

figure 2). Although the eyes, heart, and kidneys were smaller in infected than in uninfected animals, the weights of these structures were at all times proportional to the size of the mouse. Light Microscopic Findings Thymus

One day after birth, the thymuses of normal animals possessed a welldefined cortex and medulla. The cortex was tightly packed with numerous lymphocytes, and the medulla contained a vascular stroma with many epithelial cells and a few small lymphocytes. During the next week, the cortex increased in size, and by 14 days it contained a gradation of large to small lymphocytes in various stages of development (Figures 3, 5, 7, 9, and 11). The medulla at 2 weeks had few Hassall's

corpuscles. Pronounced histologic abnormalities in infected animals were first detected at 2 days (Figure 4). At this time a marked decrease in the cellularity of the thymic cortex, and lymphoid cell necrosis with pyknosis, karyolysis, and karyorrhexis were conspicuous. Hyperchromatic lymphocytes were also present in the medulla. Many cortical and a few medullary epithelial cells were large and contained enlarged nucleoli and eosinophilic globular concretions within the nucleoplasm. Margination of the nuclear chromatin was prominent. Polymorphonuclear leukocytes were rarely found in the necrotic areas of the thymic

Vol. 79, No. 3 June 1975

MURINE CYTOMEGALOVIRUS

513

cortex although neutrophils were present in the inoculated eyes and in the peripheral blood. Cortical necrosis was maximal at 4 days (Figures 6 and 8), but by 2 weeks there was little evidence of necrotic debris within the thymus. When compared to controls at this time, however, the thymuses from the infected mice had markedly fewer lymphoid cells in both the cortex and medulla (Figures 10 and 12). In addition, the medulla contained more Hassall's corpuscles and larger-than-normal epithelial cells. Spleen

The spleens of newborn mice contained many immature cells, with a sparse scattering of small lymphocytes. The lymphocytes increased in number and by 4 days were found clustered around small splenic central arteries and arterioles (T cell areas) (Figure 13).1214 These perivascular lymphocytic collections at 8 days consisted of many large and a few small lvmphocytes. The number of cells increased dramatically, and by 2 weeks a well-defined red and white pulp could be identified (Figure 15). Coincident with the thymic changes, abnormalities also existed in the spleen of infected animals. After 48 hours, scattered diffusely throughout the tissue were markedly enlarged cells with large intranuclear eosinophilic inclusions surrounded by prominent halos. The spleen contained fewer lymphocytes than normal, and many had pyknotic nuclei. By 4 days the number of large inclusion-bearing cells (Figure 14) and the amount of lymphocytic necrosis had increased. In contrast to the controls, there were relatively few collections of lymphoid cells in the T-cell areas. Over the next week, the necrosis and inclusions became less conspicuous, and at 14 days the only abnormalities noted were a sparsity of small lymphocytes and an increased number of histiocytes with enlarged nucleoli (Figure 16). Lymph Nodes and Peyer's Patches

In normal newborn mice, the lymph nodes and Peyer's patches were immature and consisted predominantly of rare focal collections of small lymphocytes and a few large cells with abundant cytoplasm and vesicular nuclei. Over the following 14 days, the lymph nodes and Peyer's patches increased in size, became heavily populated by small lymphocytes, and attained their usual morphologic appearance (Figure 17). On the other hand, the lymph nodes and Peyer's patches of the infected animals remained immature. This lack of morphogenesis was most prominent in the 4- and 8-day-old-animals. After 2 weeks, the

514

SCHWARTZ ETAL


mesenteric nodes were populated by histiocytes with prominent nucleoli, large lymphocytes, and a scattering of small lymphocytes (Figures 18). Bone Marrow

At birth, the bone marrow of uninfected animals contained megakaryocytes, as well as erythroid and myeloid elements. These cells steadily increased in number over the 2-week period (Figure 19). Multiple focal areas of necrosis were prominent in the bone marrow of infected animals of 2 days. Necrotic megakaryocytes were present, but due to the immaturity of the marrow, it was not possible to determine what other cell types were involved. Inclusions were first seen in enlarged cells 4 days after inoculation of virus in a sparsely populated and markedly necrotic marrow (Figure 20). By 8 days, necrosis and viral inclusions were absent and evidence of marrow recovery was manifested by an increased cellularity. After 2 weeks, the bone marrow of the infected animals did not differ from the controls. Bacteriologic Findings

Organisms were not isolated from the blood of control animals, however, after the second day a mixed population of bacteria was cultured from animals inoculated with murine CMV. Staphylococci, anaerobic streptococci, and gram-negative bacilli were prominent isolates. Discussion

Humans congenitally infected with CMV manifest a spectrum of pathologic alterations, including low birthweight, an increased susceptibility to bacterial sepsis, and the persistent excretion of virus, despite the presence of circulating antibodies against the virus.1'15"s6 The severity of the lesions has been related to the age of the fetus at the time of virus exposure.'5"6 If infection occurs early in pregnancy (3 to 5 months), severe alterations usually result.'6 In the past, emphasis has been placed on the anomalies of the heart, brain, and eye. Little attention has been directed towards alterations produced by the virus in the structures associated with immune function despite the fact that these tissues are also undergoing differentiation during early ontogenesis. After 2 months of gestation, lymphocytes appear in the human thymus, but not until 4 months is a well-defined cortex present.'7 Exposure of the fetus to CMV during this important developmental period might be expected to alter and/or destroy the maturing thymic lymphoid cells, resulting in structural and functional anomalies of the infant's immune system. Indeed, hypoplastic thymuses have been described in some human fetuses exposed to CMV.'5

Vol. 79, No. 3 June 1975


515

The neonatal mouse, like the 3- to 5-month-old human fetus, is immunologically immature, and cells in the thvrnus are activelv differentiating into T lrmphocvtes.'92° As reported here, inoculation of murine CNINM into newborn mice results in extensive morphologic alterations of the developing lvmphoid tissues. The results of this study suggest the possibility that some infants with a concomitant immune deficiencv and

viral disease may have developed the severe lvNmphoid tissue abnormalities as the result of a primarv viral disease. Similar to the human fetus, the newborn mouse infected with CMVN" manifests a marked retardation of growth and an increased susceptibility to microbial invasion. Growth retardation in association with thymic atrophy is not unique to CMIV infections in the mouse. These pathologic alterations are features of congenital human infection with CMV''.15o6 or rubella.'1 The relationship of deficient thvmic tissue with growth retardation has been demonstrated under several circumstances.22-27 Retarded physical development occurs in mice after thvrnectomv,2 as well as after the eradication of thxmic tissue by irradiation 27 or cortisone.26 Impaired growth is also characteristic of a genetic mutant mouse (nude mouse) in which the thvmus is congenitally- absent.23 Although the role the thvmus plays in the attainment of normal body growth is presently unclear, there is evidence that this organ is important in immune surveillance and keeps intestinal flora from infecting the host.28 Thvmectomized animals housed in a germ-free environment or treated prophvlacticallv with broad spectrum antibiotics do not develop the runting svndrome.2 92 Thus, several investigators have suggested that the runting of thvmicallv deprived animals is in some fashion caused by the bacterial infection.2930 This view is supported by the present study, as CMV-infected mice with the most severe growth retardation and thvmic atrophy died xvith bacterial sepsis. -Moreover, CfMV-infected newborn mice whose mothers receive tetracycline are not as runted as and survive longer than animals whose mothers are not given anti-

biotics.31 Another herpesvirus, the thymic virus of mice, produces extensive medullarv necrosis of the thymus,32.3' but unlike murine CMV has not been reported to cause growth retardation, alterations in the health of the animals, or a disseminated infection. In addition, a number of nonherpesvirus viruses, including lvmphocytic choriomeningitis virus,34'35 21 3 738 ~36 canine and feline enteritis virus, rubeolla, virus, distemper rubella,2' have been reported to cause cellular alterations in the thymus. Interestinglv, these viruses which affect the thvmus and other lymphoid tissues have been sho'wn to produce persistent infections in their native

516

SCHWARTZ ETAL


hosts, Although the mechanisms by which viruses are able to evade the surveillance of the immune system and persist in the tissues are unknown, the phenomenon may be related to the ability of these agents to alter the host's lymphocytes. As previously stated, murine CMV in newborn mice causes necrosis and a marked depletion of lymphocytes in the developing lymphoid tissues. Likewise, in studies on adult Swiss mice with CMV, Howard et al 39,40 reported extensive lymphocyte necrosis and depletion of the T-cell areas of the spleen during the acute infection, but morphologic alterations in the thymus were not found.39 The reason for the different distribution of lesions in the lymphoid tissues of the newborn and adult mouse is not known; however, it might be related to the different mitotic activities of the T cells in the lymphoid tissues of these variably aged animals. According to current concepts, for a herpesvirus to replicate in a lymphocyte, the lymphocyte must be undergoing cell division.4' In the neonatal mouse and the human fetus, a high percentage of thymic lymphoblasts are undergoing mitosis, and exposure of these dividing cells to the virus might be expected to infect these leukocytes. In the adult mouse (and the newborn human), lymphoid cells of the thymus have a low mitotic index, and the organ presumably would not be as susceptible to the virus. T cells in the peripheral lymphoid tissues, however, are still undergoing blastogenesis at this age and would theoretically support the virus infection. Like the mice in the present study, humans with congenital CMV infection may have low thymic weights; however, thymic necrosis has not been described in man. The paucity of reported abnormalities in the thymus in human congenital CMV infection may in part be related to the age of the lesion. In our studies, necrotic debris was not evident in the thymus after 2 weeks. Thus, thymic injury of the human fetus infected early in utero might have resolved morphologically by the time of birth. The present study demonstrates that murine CMV may be more useful for investigating the analagous human congenital infection than previously appreciated. Instead of injecting newborn or adult mice intraperitoneally and obtaining a lethal hepatitis, the inoculation of virus into the vascular eye of these immature animals resulted in a disease that is similar in several respects to the human congenital infection. References 1. Weller TH: The cytomegaloviruses: Ubiquitous agents with protean clinical manifestations. N Engl J Med 285:203-214, 267-273, 1971

Vol. 79, No. 3 June 1975


517

2. Schwartz JN, Daniels CA, Shivers JC, Klintworth GK: Experimental cytomegaloviral ophthalmitis. Am J Pathol 77:477-492, 1974 3. McCordock HA, Smith, MG: The visceral lesions produced in mice by the salivary gland virus of mice. J Exp Med 63:303-310, 1936 4. Henson D, Strano AJ: Mouse cytomegalovirus: Necrosis of infected and morphologically normal submaxillary gland acinar cells during termination of chronic infection. Am J Pathol 68:183-20-2, 1972 5. Brodsky I, Rowe WP: Chronic subclinical infection with mouse salivary gland virus. Proc Soc Exp Biol Med 99:654-655, 1958 6. Raubner BH, Miyai K, Slusser RJ, Wedemeyer P, Medearis DN: Mouse cytomegalovirus infection: An electron microscopic studv of hepatic parenchymal cells. Am J Pathol 44:799-821, 1964 7. Osborn JE, Shahidi NT: Thrombocytopenia in murine cytomegalovirus infection. J Lab Clin Med 81:53-63, 1973 8. Medearis DN: Mouse cytomegalovirus infection. III. Attempts to produce intrauterine infections. Am J Hyg 80:113-120, 1964 9. Johnson KP: Mouse cytomegalovirus: Placental infection. J Inf Dis 120: 445-450, 1969 10. Youngner JS: Mlonolayer tissue cultures. I. Preparation and standardization of suspensions of trypsin-dispersed monkey kidney cells. Proc Soc Exp Biol Med 85:202-205, 1954 11. Plummer G, Benyesh-Melmick MI: A plaque reduction neutralization test for human cytomegalovirus. Proc Soc Exp Biol Med 117:145-150, 1964 12. Joel DD, Hess MW, Cottier H: Magnitude and pattern of thymic lymphocyte migration in neonatal mice. J Exp Med 135:907-923, 1972 13. Parrot DMV, DeSousa MAB, East J: Thymus-dependent areas in the lymphoid organs of neonatally thymectomized mice. J Exp Med 123:191-204, 1966 14. Raff MC, Wortis HH: Thymus dependence of 0-bearing cells in the peripheral lymphoid tissues of mice. Immunology 18:931-942, 1970 15. Naeye RL: Cytomegalic inclusion disease: The fetal disorder. Am J Clin Pathol 47:738-744, 1967 16. Benirschke K, Mendoza GR, Bazeley PL: Placental and fetal manifestations of cytomegalovirus infections. Virchows Arch [Cell Pathol] 16:121-139, 1974 17. Harr JL: Light and electron microscopy of the human fetal thy-mus. Anat Rec 179:463-476, 1974 18. Mandel T: Differentiation of epithelial cells in the mouse thymus. Z Zellforsch mikrosk Anat 106:498-515, 1970 19. Moore MAS, Owen JJT: Experimental studies on the development of the thymus. J Exp Med 126:715-726, 1967 20. Owen JJT, Ritter MA: Tissue interaction in the development of thymus lymphocytes. J Exp Med 129:431-442, 1969 21. Garcia AGP, Olinto F, Fortes TGO: Thymic hvpoplasia due to congenital rubella. Arch Dis Child 49:181-185, 1974 22. Miller JFAP: Some similarities between the neonatal thymectomy syndrome and graft-versus-host disease. J Reticuloendothel Soc 1:369-392, 1964 23. Wortis HH: Immunologic responses of "nude" mice. Clin Exp Immunol 8:

305-317, 1971

24. DeSousa MAB, Parrott DMV, Pantelouris EMN: The lymphoid tissues in mice with congenital aplasia of the thymus. Clin Exp Immunol 4:637-644, 1969

518

SCHWARTZ ETAL


25. Pierpaoke W, Sorkin E: Hormones and immunological capacity. I. Effect of heterologous anti-growth hormone (ASTH) antiserum on thymus and peripheral lymphatic tissue in mice. Induction of a wasting syndrome. J Immunol 101:1036-1043, 1968 26. Jutila JW: Wasting disease induced with cortisol acetate. III. Immunologic studies. J Immunol 102:963-969, 1969 27. Good RA, Gabrielsin AE: The Thymus in Immunobiology: Structure, Function and Role in Disease. New York, Harper and Row, 1964 28. Asar HA: Bacterial infection and wasting in neonatally thymectomized rats. Proc Soc Exp Biol Med 116:817-823, 1964 29. Wilson R, Sjodin K, Bealmear M: The absence of wasting in thymectomized germfree (axenic) mice. Proc Soc Exp Biol Med 117:237-239, 1964 30. Brooke MS: Experimental runt disease in mice caused by Salmonella typhimurium, var. Copenhagen. J Exp Med 120:375-387, 1964 31. Schwartz JN, Daniels CA, Klintworth GK: Unpublished observations 32. Rowe WP, Capps WI: A new mouse virus causing necrosis of the thymus in newborn mice. J Exp Med 113:831-844, 1960 33. Parker JC, Vernon ML, Cross SS: Classification of mouse thymic virus as a herpesvirus. Infect Immun 7:305-308, 1973 34. Lillie RD, Armstrong C: Pathology of lymphocytic choriomeningitis in mice. Arch Pathol 40:141-152, 1945 35. Hanaoka M, Suzuki S, Hotchin J: Thymus-dependent lymphocyte destruction by lymphocytic choriomeningitis virus. Science 163:1216-1219, 1969 36. White RG, Boyd JF: The effect of measles on the thymus and other lymphoid tissues. Clin Exp Immunol 13:343-357, 1973 37. McCullough B, Krakowka S, Koestner A: Experimental canine distemper virus-induced lymphoid depletion. Am J Pathol 74:155-170, 1974 38. Rohovsky MW, Griesemer RA: Experimental feline infectious enteritis in the germfree cat. Pathol Vet 4:391-410, 1967 39. Howard RJ, Najarian JS: Cytomegalovirus-induced immune suppression. I. Humoral immunity. Clin Exp Immunol 18:109-118, 1974 40. Howard RJ, Miller J, Najarian JS: Cytomegalovirus induced immune suppression. II. Cell-mediated Immunity. Clin Exp Immunol 18:119-126, 1974 41. Wheelock EF, Toy ST: Participation of lymphocytes in viral infections. Adv Immunol 16:123-184, 1973

Acknowledgments Excellent technical assistance was provided by Carl Bishop, Jessie Calder, and Betty Schlitz.

Figs 1 and 2-A control (upper animal) and infected mouse (lower animal) are shown at 4 (1) and 14 (2) days. Compared to the controls, mice inoculated with virus are smaller in size and have a decreased amount of subcutaneous tissue as demonstrated by the wrinkled skin, flaired rib cages, and thinned torso.

4

3

PVj6QJ

5

7

1I_) 4 ; > Figs 3 and 4-A lobule of a normal thymus at 2 days contains a prominent cortex populated by many small lymphocytes and a centrally placed medulla (3). The thymus of an animal 48 hours after infection reveals a thin cortex that is indistinctly demarcated from the medulla (4). (H&E, x 62) Figs 5 and 6-The outermost, subcapsular area of the normal thymus at 4 days contains large cells with vesicular nuclei blending into the densely populated cortex containing'small lymphocytes (5). The thymus of an infected mouse of the same age shows the subcapsular region filled with pyknotic and fragmented cells (6). (H&E, x 215) Figsi 7 and 8-The subcapsular cortex of a normal (7) and an infected (8) 4-day-old mouse shown at a higher magnification. Note the large immature lymphocytes in the control animal and the marked karyorrhexis, karyolysis, and absence of neutrophils in the infected animal. (H&E, X 660)

9

41

z

10

1%~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~

Figs 9 and l0-Corticomedullarly area of a thymic lobule at 14 days in an uninfected mouse (9). The cortex and medulla are not well defined in the infected animal at this time (10). (H&E, x 264) Figs 11 and 12-The cortex of the thymus at 14 days is shown at a higher magnification. Many small lymphocytes are present in the cortex of the control animal (11); however, this part of the thymus of the infected animal is depleted of small lymphocytes (12). Note the uniform large cells in the stroma with abundant cytoplasm and centrally located nucleoli. A few lymphocytes are present in the medulla, and necrotic debris is no longer evident in the infected animals (12). (H&E, x 660) Figs 13 and 14 At 4 days, the layer of lymphocytes around the central arteriole of the control spleen is well demarcated. In the infected animal, accumulations of necrotic lymphoid cells are present adjacent to the central vessel, the cellularity is diminished, and numerous cytomegalicinclusion-bearing cells are observed in the spleen (See inset). (H&E, x 264; inset, x 600)

15

.A,0, 16 5~~~~~~

6

VI 18l

17

.~ ~ ~ ~ ~ ~ ~ ~ ~ ..hI'A.~~~~~~~~~~~~~~~~~~~~. a 0 191

Figs 15 and 16By 2 weeks the arteriolar cuff of lymphocytes in the control animals is well developed (15). The infected animals have prominent histiocytes and fewer small lymphocytes. As in the thymus at this time, necrotic debris is no longer present (16). (H&E, Figs 17 and 18A section through a mesenteric lymph node in a normal (17) x 165) and infected mouse (18) at 14 days. Many small lymphocytes are diffusely scattered throughout the stroma of the normal animal. However, in the infected animals there is a relative absence of small lymphocytes and a predominance of large lymphocytes and Figs 19 and 20-The bone marrow of a control mouse is histiocytes. (H&E, x 264) very cellular at 4 days (19). In contrast, tissue in an infected mouse at 4 days (20) has considerably fewer cells, many of which are pyknotic. Also, occasional enlarged cells with intranuclear inclusions are present (See inset).