IMMUNODEFICIENCY AND RETICULUM CELL SARCOMA IN MICE SEGREGATING FOR
HRS/J AND SJL/J GENES*
DAVID A. JOrIWSON, LEONARD D. SHULTZ and HENDRICK G. BED1GIAN
The Jackson Laboratory, Bar Harbor, ME 04609, U.S.A.
(Received 13 November 1981. Revised 14 January 1982 and 10 May 1982. Accepted 2 June 1982)
Abstract--Hairless (hr/hr) mice segregating for SJL/J and HRS/J genes (SJL-HRS) were com- pared to their haired counterparts with respect to immune responsiveness, tumour development and ecotropic murine leukemia virus (MuLV) expression. Homozygosity at the hairless locus did not affect expression of MuLV. There was however, a significant depression of the cellular immune response of these mice as characterized by depressed reactions in phytohemagglutinin, concanavalin A and mixed leukocyte assays. Haired and hairless mice did not differ significantly in response to B-cell mitogens or in production of cytotoxic antibody. The depressed cellular immune response in hr/hr mice is associated with a distinctive histologic type of spontaneous reticulum cell sarcomas. The importance of these results in relation to previous studies of HRS/J hairless mice is discussed.
HRS/J hr/hr (HAIRLESS) mice have a significantly higher spontaneous leukemia incidence when compared to their haired (hr/+) counterparts [15, 8] and it has been demonstrated that there are no genotype dependent differences with respect to expression of ecotropic murine leukemia viruses (MuLV) [10, 13] or serum levels of p30, the MuLV group specific antigen [13]. Although it has been suggested that hr/hr mice have higher titres of xenotropic MuLVs, and recombinant MCF-type viruses can be isolated from HRS/J mice [10], a causative relationship with the higher leukemia incidence of hr/hr mice is not yet firmly established.
A number of studies have indicated genotype dependent differential immune respon- siveness may be related to the leukemia incidence differences of HRS/J mice. It has been reported that hr/hr mice are deficient in their ability to produce antibody to tetanus toxoid [9], and to respond in graft-vs-host reactions [11,17]. A disproportion of Ly-l.2,3 + and Ly-1 + cells in hr/hr mice has also been described [19]. In work originat- ing in this laboratory [14] it was found that HRS/J hr/hr mice are more highly suscep- tible to challenge with a syngeneic transplantable leukemia (HTU). They are also defi- cient in their ability to respond to protective immunizations with inactivated HTU cells.
*This research was supported by United States Public Health Service Awards R23CA25944,'CA20408 and CA31102. The authors wish to thank Dr. M. Cherry for carrying out the cytotoxic antibody assays and Dr. D. Coman for his excellent assistance in the histologic evaluations. The Jackson Laboratory is fully accredited by the American Association for Accreditation of Laboratory Animal Care.
Correspondence to: Research Institute of Scripps Clinic. 10466 North Torrey Pines Road, R319, La Jolla, CA 92037. U.S.A.
711
712 DAVID A. JOHNSON e t al.
We demonstrated a parallel defect in their ability to respond to MuLV antigens expressed on HTU cell surfaces [14].
Although it seems clear now that homozygosity at the hr-locus in HRS/J results in partial immunodeficiency and may result in a higher leukemia incidence, it is not estab- lished whether this effect is unique to the HRS/J strain. Recently, Smith et al. [23] reported that 10 stocks of genetically non-haired mice had normal immune function when mitogen responsiveness, antibody responsiveness and numbers of B and T cells were evaluated.
We now describe the effects of homozygosity at the hairless locus in individual mice from a population segregating for genes from two inbred strains, SJL/J and HRS/J. Homozygosity at this locus results in immunodeficiencies and alters tumour type but does not affect ecotropic virus expression or tumour incidence. We conclude that the effects of this gene are not completely dependent on genetic background.
MATERIALS AND METHODS
Mice
In 1977 a spontaneous mutation to 'hairless' was reported in an SJL/J mouse at the Jackson Laboratory. This mutation was subsequently demonstrated to be allelic to the hr gene described in HRS/J mice (D. Corrow, personal communication). Homozygous hairless (hr/hr) mice lose their body hair after about the first two weeks of life in a manner similar to HRS/J hr/hr mice. Heterozygous and + / + mice have normal body hair.
These mice were inadvertently crossed with HRS/J and subsequently sib-mated without selection. Using Southern blotting [24] and hybridization with an ecotropic specific probe, the endogenous ecotropic MuLV sequences present in this population were determined (data not shown). Briefly. Pvu II digestion of liver DNAs demonstrated a random assortment of cell-viral DNA junction fragments which were characteristic of SJL/J endogenous proviruses (4.1 and 4.8 kb) or HRS/J endogenous proviruses (4.3 and 5.4 kb) [25]. Isozyme analysis of randomly selected mice also indicated the presence of a mixture of SJL/J and HRS/J genes. The isozymes evaluated included ldh-I (Chr 1), Pep-3 (Chr 1), Gpd-I (Chr 4) and Es-3 (Chr 11) (B. A. Taylor, personal communication). It is therefore possible to evaluate the effects of homozygosity at the hairless locus indepen- dent of background effects. Any consistently observed effects can be directly attributed to the hr-locus. All mice were housed four to five to a cage and were given food and water ad libitum. Within experimental groups, mice were age (+/-3 days) and sex matched. With the exception of the tumour studies, we made no effort to distinguish + / + and hr/+ mice. These mice are designated ?/+.
Tumour studies Mice were given food and water art libitum. They were observed on a regular basis for signs of illness and,
when moribund, were sacrificed and organ samples were examined histologically.
immune studies
Spleen cell mitogenesis was determined using techniques described [12]. Partially purified phytohemaggluti- nin (PHA) was obtained from Wellcom¢ Reagents Ltd., Beckenham, U.K. Escherichia coil lipopolysaccharide (LPS), pokeweed mitogen (PWM) and concanavalin A (Con-A) were obtained from Sigma Chemical Co.. Saint Louis, Missouri.
Mixed leukocyte reactions (MLR) were carried out using a modification of a procedure described previously [2]. Briefly, spleen cells were suspended at a concentration of 2 x 10 ° per ml of enriched Dulbecco's MEM [4] plus 10% heat-inactivated fetal calf serum. Stimulator cell suspensions were exposed to 5000 R gamma ir- radiation using a cesium source. Replicate tube cultures of 0.25 ml of stimulator cell suspension plus 0.25 ml of reactor cell suspension were incubated for 120 h at 37°C in a water-saturated 5% CO2 atmosphere. [3HlThy- midine (Amersham, Arlington Heights, Illinois) (2/~Ci) were then added to each culture and incubation was continued for an additional 6 h. Trichloroacetic acid-precipitable radioactivity was then determined.
The cytotoxic antibody response to allogeneic lymphoid cells (AKR.L-H-2 b/l Cy spleen cells) was deter- mined using techniques described ['6]. Sera were obtained from individual .9/+ and hr/hr mice which had been injected (i.p.) 14 days earlier with 2 x 106 spleen cells from AKR.I-H2 b/1 Cy (H-2 b) mice.
Enumeration of T and B cells Spleens were gently teased apart in Hank's balanced salt solution (HBSS; Microbiological Associates, Inc..
Bethesda, Maryland) containing I0 mM N-2-hydroxyethyipiperazine-N'-2 ethane sulphonic acid (Hepes) and 5% fetal bovine serum (FBS). Cells were filtered through nylon mesh and washed three times. The lymphocyte fraction was obtained by Ficoll-Hypaqu¢ density centrifugation (Lymphocyte separation medium--Litton Bionetics, Kensington, Maryland). Indirect staining for T cells with monoclonal anti-Thy-l.2 antibody and direct staining for B cells with fluorescein-conjugated F (ab'h fragment of polyvalent goat anti-mouse Ig (Cappeli Inc.) were carried out on single cell suspensions as previously described [22].
Immunodeficiency in hairless mice 713
Virus studies
The presence of ecotropic virus was assessed using the infectious centre assays [16, 20]. Single cell suspen- sions of splenic tissue were prepared by mincing tissue on 60 mesh wire screen platforms, followed by expelling fragments through progressively smaller needles, down to 25 gauge. SC-1 cells seeded 24 h earlier at a density of 2.0 × 105 cells in 60 mm plastic Petri dishes were maintained in Eagle's minimum essential media (EMEM) supplemented with 5°, FBS, 2 mM L-glutamine, antibiotics and 2 mg/ml polybrene (Aldrich Chemical Co.). Twenty-four h later the cells were infected with serial dilutions of splenocytes treated with mytomycin C (25 #g for 30 rain). Four to five days after infection the SC-l cells were u.v.-irradiated and overlaid with l x l0 ~ XC cells. Four days later the cells were fixed, stained and the number of plaques, expressed as plaque forming units per l0 ~ cells, were determined microscopically.
Statistical analysis
Differences between means were determined using the Student's t-test with a minimum level of confidence of 95" ,,,
RESULTS
Immune studies These studies were carried out using spleen cells from individual 2-3-month-old mice.
In all cases, spleens were normal in appearance. In the mitogen and MLR studies, the data from separate experiments, each depicting the means of the values obtained from eight animals are indicated.
Preliminary T-cell mitogenesis studies using PHA [2, 3] indicated that maximum stimulation occurred when spleen cells were incubated in the presence of 2/~l of PHA for 24 h followed by an additional 6 h incubation in the presence of [3H]thymidine. All subsequent PHA experiments (indicated in Fig. l) were carried out using this regimen. In four separate experiments the spleen cells of haired 9./+ mice responded more strongly than did spleen cells from hr/hr mice. The differences between each group were highly significant (p values < 0.01).
In order to examine further T-cell mitogenesis, spleen cells were stimulated with a second T-cell mitogen, Con-A [1]. Optimal stimulation occurred when cultures were incubated for 36 h followed by 6 h in the presence of [3H]thymidine. The results are indicated on Fig. 2. Once again ?/+ spleen cells were stimulated more strongly in each experiment (p values <0.01) than spleen cells from hr/hr mice.
Next, studies were carried out to determine whether the decreased response to T-cell mitogens was paralleled by a decreased responsiveness to the B-cell mitogen LPS Ill .
2 0
'o x 12
o .
8
PHA
C~ hrJ/hrJ ? / +
A B C
Y/,
D
FIG. 1. PHA stimulation of spleen cells from haired (?/+1 and hairless (hr/hr) mice. A, B, C and D represent separate experiments carried out on different days. A total of 32 two- to three-month-
old mice were assayed.
714 DAVID A. JOHNSON et al.
20
16i
912 x
G.
CON-A hrJ/hr J
A 8 C
FIG. 2. Con-A stimulation of spleen cells from haired (?/+ ) and hairless (hr/hr) mice. A, B and C represent separate experiments carried out on different days. A total of 24 two- to three-month-
old mice were assayed.
16
9 z 2 N
LPS
r - ~ h r J / h r J ~ a ? 1 +
A B C O
FIG. 3. LPS stimulation of spleen cells from haired (?/+ ) and hairless (hr/hr) mice. A, B, C and D represent separate experiments carried out on different days. A total of 32 two- to three-month-
old mice were assayed.
PWM
r - ~ hr-//hr v 24. EZ~ ?/.t-
20
±
4
0 A
o~6
.L
8 C
FIG. 4. PWM stimulation of spleen cells from haired (':/+) and hairless (hr/hr) mice. A, B and C represent separate experiments carried out on different days. A total of 24 two- to three-month-
old mice were assayed.
Immunodeficiency in hairless mice 715
20
16
x 12
o.
u 8
MLR
/~ J/h r J
1223 ?/+
//,,
A B
I t
I
FIG. 5. MLR reactivity of spleen cells from haired (?/+) and hairless (hr/hr) mice. Target cells were derived from allogenic AKR.L-H-2 b/l Cy spleens. A, B and C represent separate experi-
ments carried out on separate days. A total of 24 two- to three-month-old mice were assayed.
The results of these studies are shown in Fig. 3. Optimal stimulation occurred with a 36 h incubation period followed by an additional 6h incubation in the presence of [3H]thymidine. There was no significant difference in response to LPS in any of the four experiments indicated.
Figure 4 shows the results of stimulation experiments with PWM, a B- and T-cell mitogen 1-21]. Optimal stimulation occurred with a 48 h incubation time followed by 6 h in the presence of [3H]thymidine. Once again, no significant differences in stimulation were observed between ?/+ and hr/hr spleen cells.
In an effort to determine whether the T-cell hyporesponsiveness of hr/hr spleen cells was paralleled by decreased responses in antigen specific reactions, classical MLR assays were carried out using allogeneic spleen cell targets (from AKR.L-H-2 b/1 Cy mice). The results are shown on Fig. 5 and indicate, once again, that hr/hr spleen cells respond less well (p values <0.01) than do spleen cells from ?/+ mice. Studies to determine the cytotoxic antibody response to AKR.L-H-2 b/1 Cy spleen cells resulted in no significant difference, the 50~o cytotoxic titres for hr/hr and ?/+ mice being 67.0 + 28.5 and 60.4 + 24.6 respectively.
Immunofluorescence studies were carried out to determine the relative numbers of lymphoid cells bearing the T-cell marker Thy-l.2 or surface immunoglobulin. No signifi- cant genotype dependent differences were observed (see Table 1).
Turnout studies All of the 32 SJL-HRS hairless mice observed from weaning developed neoplasms at a
mean age of 13.3 + 1.22 months. All 22 hr/+ controls developed RCN type B at a mean
TABLE 1. LYMPHOCYTE SUBPOPULATIONS IN SPLEENS OF
HAIRLESS MICE
Genotype* ~o S-Ig ÷ cells p~ ° o Thy-l.2 ÷ cells p
*Three pairs of hr/hr and +/hr male mice were tested at 15 weeks of age.
"]'Significance determined by Student's t-test. N S = p > 0 . 0 5 .
716 DAVID A. JOHNSON et al.
TABLE 2. EXPRESSION OF ECOTROPIC MuLV FROM SPLEEN SUSPENSIONS OF
SJL-HRS HAIRED AND HAIRLESS MICE
Virus infectious Isolation centres/10" cells
Genotype Age in months (°o) (log I0)*
+/? 3 4/5 (80) 2.2 7 4/4 (100) 2.3
hr/hr 3 4/5 (80) 2.2 7 4/4(100) 2.3
*Values are the mean log of the number of plaque-forming cells/107 cells.
age of 12.7 4- 0.92 months, as is typical of SJL/J mice [22]. This difference was not statistically significant. Of particular interest was the finding that the predominant neo- plasm in hr/hr mice consisted of large, pale reticulum cells with an absence of small lymphocytes (Fig. 6). This was distinctly different from the typical florid host response seen with hr/+ mice.
Virus studies Sixteen of 18 .9/+ (2-8 months old) and hr/hr mice were tested for the presence of
MuLV using the infectious centre assay. Although ecotropic virus was readily detected, we did not find a significant difference in titre between the two phenotypes (Table 2).
DISCUSSION
The experiments described here were designed to determine whether homozygous hairless SJL-HRS hybrid mice have differences in immune responsiveness and tumouri- genesis as has been described in the HRS/J strain [14, 15]. First, however, we wished to examine the expression of ecotropic murine leukemia viruses. These viruses are known to be associated with leukemia in mice and may be involved in recombinational events leading to the production of highly leukemic MCF-type viruses ~7]. No significant genotype dependent variations were found in the present study. It is evident that the hr gene does not affect ecotropic virus expression, in agreement with earlier studies of HRS/J hairless mice [10, 13].
With this information in hand, we turned to an analysis of the immune responsiveness of SJL-HRS hybrid mice. The stimulation studies using the T-cell mitogens demon° strated that, relative to haired (.9/+) mice, hairless (hr/hr) mice have a significantly lower response. This is distinctly different from the situation with HRS/J hr/hr mice in which a slightly enhanced PHA responsiveness was reported by Heiniger et al. [9] and was subsequently confirmed by Morrissey et al. [17]. The depressed response in hybrid mice was not the result of a general inability of lymphoid cells to proliferate, since there were no significant genotype dependent differences in response to the B-cell mitogen LPS 1.1], or to PWM which stimulates both B and T cells E21]. Hence, these results are strongly indicative of a T-cell specific defect.
We next wished to determine whether this defect was also apparent in antigen specific immune responses. The classically T-cell related MLR assay 12] was chosen, using spleen cell targets from AKR.L-H-2 b/1 Cy mice. Once again the response of hr/hr spleen cells was significantly lower than that of .9/+ mice. When mice of both genotypes were immunized with AKR.L-H-2 b/1 Cy spleen cells and subsequently analysed for their ability to produce cytotoxic antibody, no significant differences were observed.
Since the above data indicated the presence of a defect in hr/hr mice, specific for the cellular arm of the immune system, we wished to determine whether they differed in
FIG. 6. Reticulum cell neoplasms from 12-month-old hr/+ and hr/hr mice. (a) Reticulum cell tumour type B from hr/+ mouse. Lymph node replaced by cells of several types including reticulum cells, lymphocytes and plasma cells. Mitotic figures indicated by arrows. (b) Reticulum cell neoplasm from hr/hr mouse. Lymph node largely replaced with reticulum cells, some of which
are in mitosis. Invasion through capsule indicated b.y arrow,
717
Immunodeficiency in hairless mice 719
respect to relative numbers of T and B lymphocytes. Immunofluorescent antibody assays to detect cell surface immunoglobulins or Thy-l.2 antigens failed to detect any significant differences, indicating that the defect we now report is not due to differences in absolute numbers of T cells.
Our data, therefore, indicate that homozygosity at the hr locus results in an immuno- deficiency which is related to T-cell proliferation independent of the genetic background of the host. Since a similar situation has been described in the highly leukemic HRS/J hr/hr mice 1-15, 14] we were interested in examining tumour incidence in hairless SJL-HRS hybrids. Haired SJL/J mice spontaneously develop reticulum cell neoplasms (RCN) type B at a 70-909~o incidence by 13 months of age [18]. Designation of these tumours as appropriate models for Hodgkins Disease in man, is supported by the pres- ence of Reed-Sternburg type cells and by the florid host response [18]. Our studies showed that the tumour incidence of hr/hr mice was identical to the incidence reported for SJL/J mice. There was, however, a distinct difference with respect to histologic type of tumour, the predominant neoplasm of hybrid hr/hr mice consisting of large pale reticulum cells with an absence of small lymphocytes. The histology of the hybrid ?/+ tumours was identical to those found in inbred SJL/J mice, indicating the lack of a florid response was dependent on homozygosity at the hairless-locus, and is therefore correlated to the cellular immunodeficiency in these mice. The failure of the immune defect of hr/hr mice to alter the latency of tumours is suggestive that the immune system does not directly affect this neoplasm in contrast to the situation with HRS/J mice where there is a close association between immunodeficiency and leukemia incidence [14]. This may, there- fore, be indicative of differences in susceptibility of reticulum cell sarcomas and lym- phomas to immune modulation. It is of interest to note that no thymic leukemias characteristic of HRS/J mice were observed in this study, suggesting the possible pres- ence of SJL/J genes which suppress the development of such tumours. Alternatively, this may be due to a difference in tumour latency.
In conclusion, the present studies indicate that homozygosity at the hairless-allele results in altered tumour histologic type and a cellular immunodeficiency in mice of varied genetic background. This immunodeficiency can be distinguished from that which occurs in inbred HRS/J mice by the depressed response to T-cell mitogens, which is not present in HRS/J hr/hr mice. The failure of the defective cellular response to alter the tumour incidence in the hybrid mice indicates the lack of a close link between cellular immune function and susceptibility to reticulum cell sarcomas which develop in these animals.
REFERENCES
1. ANDERSON V., SJOBERG O. & MOLLER G. (1972) Selective induction of DNA synthesis in T and B lympho- cytes. Cell. lmmun. 4, 381.
2. BRADLEY L. M. (1980) Mixed lymphocyte responses. In Selected Methods in Cellular lmmunoloay (MISHELL B. B. & SHI1GI S. M.. Eds.I, p. 162. W. H. Freeman and Co., Reading.
3. BYRD W., voNBOEHMEO H. & ROUSE B. (1973) The role of the thymus in maturational development of phytohemagglutinin and pokeweed mitogen responsiveness. Cell. Iramun. 6, 12.
4. CEROTTINI J.. ENGERS H., McDONALD H. & BRENNER K. (1974) Generation of cytotoxic T lymphocytes in vitro. I. Response of normal and immune mouse spleen cells in mixed leukocyte cultures, d. exp. Med. 140, 703.
5. CHATTOPADHYAY S. K.. LANDER M. R., RANDS E. & LOWY D. L. (1980) Structure of endogenous murine leukemia virus DNA in mouse genomes. Prec. nam. Acod. Sci. U.S.A. 77, 5774.
6. CHERRY M. & SNELL G. D. (1969) A description ofmu: a non-H-2 RUe-antigen in C3H.SN mice. Transplan- tation 8, 319.
7. ELDER J. H.. GAUTSCH J. W.. JE~SEN F. C., LERNER R. A., H~d~TLEY J. W. & ROWE W. P. (1977) Biochemical evidence that MCF murine leukemia viruses are envelope (env) gene recombinants. Prec. natn. Acod. Sci. U.S.A. 74, 4676.
720 DAVID A. JOHNSON et al.
8. GREEN N., HIAI H.. ELDER J. H., SCHWARTZ R. S., KHIROYA R.. THOMAS C. Y.. TSICHLIS P. & COFFIN J. (1980) Expression of leukemogenic recombinant viruses associated with a recessive gene in HRS/J mice. J. exp. Med. 152, 249.
9. HEINIGER H. J., MEIER H., KALISS N.. CHERRY M., CHEN H. & STONER R. (1974) Hereditary immunodefi- ciency and leukogenesis in HRS/J mice. Cancer Res. 34, 201.
10, HIAI H., MORRISSEY R., KHIROYA R. & SCHWARTZ R. S. (1977) Selective expression of xenotropic virus in congenic HRS/J (hairless) mice. Nature, LoRd. 270, 247.
11. I'ANSON V. & GASSER D. (1973) Relationship between graft-vs-host reactivity and possession of the high leukemia genotype hr/hr. J. lmmun. 3, 1604.
12. JOHNSON D. A., BEDIGIAN H., CHERRY M. & MEIER H. (1980) Leukemogenesis immune responsiveness, and murine leukemia virus expression in congenic AKA/J mice differing at H-2. Dlfect. Immun. 29, 1007.
13. JOHNSON D. A., BEDIGIAN H. G. & MEIER H. (1980) Differential tumor susceptibility and immune respon- siveness in HRS/J mice. In Genetic Control of Natural Resistance to Infection and Malignancy (SKAMENE E.. KONGSHARN P. A. & LANDRY M., Eds.), p. 461. Academic Press. New York.
14. JOHNSON D. A. & MEIER H. (1981) Immune responsiveness of HRS/J mice to syngeneic lymphoma cells. J. lmmun. 127, 461.
15. MEIER H., MEYERS D. & HUEBNER R. J. (1969) Genetic control by the hr-locus of susceptibility and resistance to leukemia. Proc. natn. Acad. Sci. U.S.A. 63, 759.
16. MELIEF C. V. M., DATTA S. K., LOUIE S., JOHNSON S., MELIEF W. & SCHWARTZ R. S. (1975) Splenocytes plaque assay for the detection of murine leukemia virus. Proc. Soc. exp. Biol. Med. 149, 1015.
17. MORRISSEY P. J., PARKINSON D. R.. SCHWARTZ R. S. & WAKSAL S. D. (1980) Immunological abnormalities in HRS/J mice. I. Specific deficit in T lymphocyte helper function in a mutant mouse. J. Immun. 125, 1558.
18. MURPHY E. D. (1963) SJL/J, a new inbred strain of mouse with a high early incidence of reticulum cell sarcoma. Proc. Am. Ass. Cancer Res. 4, 46.
19. RESKE-KUNZ A. B.. SCHEID M. & BOYSE E. A. (1979) Disproportion in T-cell subpopulations in immunode- ficient mutant hr/hr mice. J. exp. Med. 149, 228.
20. ROWE W. P., PUGH W. & HARTLEY J. (1970) Plaque assay techniques for murine leukemia viruses. Viroloey 42, 1136.
21. SHORTMAN D., BYRD W., CEROTTINI J. & BRUN~ER K. (1973) Characterization and separation of mouse lymphocyte subpopulation responding to phytohemagglutinin and pokeweed mitogen. Cell. lmmun. 6, 25.
22. SIDMAN C. L., SHULTZ L. D. & UNANUE E. R. (1978) The mutant motheaten. I. Development of lymphocyte populations. J. Immun. 212, 2392.
23. SMITH S. M., FORBES P. D. & L1NNA T. J. (1982) Immune-responses in non-haired mice. lnt. Arch s Allergy appl. lmmun. 67, 254.
24. SOUTHERN E. M. (1975) Detection of specific sequences among DNA fragments separated by gel electro- phoresis. J. molec. Biol. 98, 503.
25. JENKINS N. A., COPEL^ND N. G., TAYLOR B. A. & LEE B. K. (1982) Organization, distribution, and stability of endogenous ectotropic murine leukemia virus DNA sequences in chromosomes of Mus musculus, J. Virol. 43, 26.
