- Email: [email protected]
c and CBA mice differ in the subtype of T cell involved in moloney murine leukemia virus-induced lymphomas
Leukemia Research Vol. 10, No. 3, pp. 339-344, 1986. Printed in Great Britain.
0145-2126/8653.00 ~ .00 c) 1986 Pergamon Press Ltd.
B A L B / c A N D CBA MICE DIFFER IN THE S U B T Y P E OF T CELL I N V O L V E D IN M O L O N E Y MURINE LEUKEMIA VIRUS-INDUCED LYMPHOMAS* BIRGITTA ,~S.IO and EVA MARIA FENYO Department of Virology, Karolinska Institutet, Stockholm, Sweden (Received 9 July 1985. Revision accepted 10 October 1985)
Abstract--Subsets of T cells can be recognized by presence or absence of terminal deoxynucleotidyl transferase (TdT) or 20ct-hydroxysteriod dehydrogenase (20ctSDH) activity. By use of these enzyme markers Moloney murine leukemia virus (M-MuLV)-induced lymphomas were shown to involve different T-cell subtypes. In the present paper we show that the genotype of the mouse has a strong influence on the subtype of T cell involved in lymphoma. BALB/c mice preferentially develop 20ctSDH-positive lymphomas, whereas CBA lymphomas often have the TdT phenotype. Comparison of the 20ctSDH activity of normal bone marrow cells showed that BALB/c mice have higher enzyme levels than CBA mice. The relative availability of a certain type of cell at a critical early step in leukemogenesis may thus influence lymphoma type. Key words: Moloney lymphoma and T-cell subset, terminal deoxynucleotidyl transferase, 20ct hydroxysteroid dehydrogenase.
INTRODUCTION INOCULATION OF M-MuLV to newborn mice leads to the development of T-cell lymphomas later in life. Two major forms of lymphoma can be distinguished according to the gross pathology. In one form the thymus is involved, mainly or exclusively, whereas in the other form the spleen and lymph nodes are enlarged. Gisselbrecht et al. [1] have shown that localization of a lymphoma to the thymus vs spleen and lymph nodes depends on the genetic background of the mouse. They found no linkage to the major histocompatibility (H-2) complex. In a previous work [2] we have observed differences in the transplantation properties of thymomas and generalized lymphomas. Upon inoculation to syngeneic hosts, thymomas grew as solid tumors at the site of inoculation, whereas cells of generalized lymphomas 'homed' to the spleen and lymph nodes. Based on these differences in transplantation properties we have assumed that the two pathological forms of lym-
*This work was supported by the Swedish Cancer Society, the Swedish Society for Medical Sciences and the National Science Foundation. Abbreviations: M-MuL V, Moloney murine leukemia virus; MCF, mink cell focus inducing virus; TdT, terminal deoxynucleotidyl transferase; 20ctSDH, 20ct-hydroxysteroid dehydrogenase; FMR, Friend, Moloney, Rauscher cell surface antigen; H-2, major histocompatibility complex; Rrnv-1, Rmv-2 and Rmv-3, genes for resistance to M°MuLV. o Correspondence to: Birgitta AsjO, Department of Virology, Karolinska Institutet, c/o SBL, S-105 21, Stockholm, Sweden. 339
phoma originate from different T-cell subsets. This assumption could be substantiated by using enzyme markers to define subsets of T cells [3]. It has been shown previously that in normal mice, immature cortical thymocytes are terminal deoxynucleotidyl transferase (TdT)-positive [4] whereas mature immunocompetent T cells in peripheral lymphoid organs and medullary thymocytes are 20ct hydroxysteroid dehydrogenase (20aSDH)-positive [5, 6]. Precursor cells of both subsets are present in the bone marrow [7, 8]. By determining the enzyme activity in M-MuLV induced T cell lymphomas we could indeed assign distinct T-cell subsets to the two different pathological forms of lymphoma [3]. Thymomas were TdT-positive whereas the majority of generalized lymphomas consisted of 20aSDH-positive cells. Working with the BALB/c and CBA strains we observed that the genetic background of the mouse influences the subtype of T cell involved in the lymphoma [9]. The particular system used was transfer of preleukemic cells from one-month old M-MuLV-inoculated mice to 400 rad irradiated syngeneic recipients. Donor cells were identified by the sex chromosome. Transfer of preleukemic BALB/c bone marrow resulted in both TdT-positive and 20aSDH-positive lymphomas whereas preleukemic spleen cells gave rise to 20ctSDH-positive lymphomas only. In contrast, none of the CBA lyrnphomas was 20ctSDH-positive regardless of which preleukemic organ was transferred. These results prompted us to ask the question whether BALB/c and CBA mice differ in the availability of any particular T-cell subtype. In the present study enzyme activities in thymus, spleen and bone marrow of normal and leukemic mice were measured. Normal CBA and BALB/c mice showed
BtRG[TTA ,~sJO and EVA MARIA FENYO
340
1000" 500"
100"
0
50'
|
Z E3 --t
";
10-
o E el
5-
.>_.
[ 0.s-
•
0 0
W
|
0.1
t
"
1
6
o
o
1
6
I1 1
spleen
6
thymus
1
CBA
6
thymus
spteen
BALBIc
B 50.0 "~
Z E3 O~ 1 e-
O
10.0'
|
8
5.0-
O
•
E
i
o
0 0
•
0 0
1.0-
0.5" E
O
O
N (Z LU
0.1
1 6 spleen
1 6 thymus
CBA
1 6 spree n
1 6 thy mus
BALBIc
FIG. 1. (A) TdT activity and (B) 20c(SDH activity in spleen and thymus of CBA (O) and BALB/c ( 0 ) mice at ! and 6 months of age.
341
Influence of genotype in T-cell subtypes of Moloney lymphomas q u a n t i t a t i v e differences particularly in 2 0 u S D H levels in spleen a n d b o n e m a r r o w . Following n e o n a t a l inoculation o f M - M u L V B A L B / c mice o f t e n developed generalized l y m p h o m a s with a 20ctSDH-positive p h e n o type whereas C B A mice preferentially developed T d T positive l y m p h o m a s .
MATERIALS
AND METHODS
Mice and virus CBA and BALB/c mice from our breeding stock were inoculated with Moloney murine leukemia virus (M-MuLV) when newborn. The source of M-MuLV was a 10070cell-free extract of BALB/c spleen homogenate, harvested 2 1/2 months after virus inoculation. It was obtained from the Division of Cancer Cause and Prevention, National Cancer Institute. Newborn mice were inoculated intraperitoneally with 0.1 ml of virus homogenate diluted 1:10. The inoculum contained 2.6 x 103 XC-PFU.
Thy-I determination The microcytotoxicity assay was carried out as described previously [2]. Anti Thy-1.2 monoclonal antibodies were kindly provided by Dr I. A. Clark, Seattle.
It is k n o w n t h a t physiological involution o f t h y m u s occurs with age a n d primarily effects the cortical TdTpositive thymocytes [13]. Accordingly, T d T activity decreased in t h y m u s a n d also in spleen. T h e decrease is particularly p r o n o u n c e d in B A L B / c mice; all 5 spleens are negative for T d T activity at 6 m o n t h s o f age. O n the o t h e r h a n d , 2 / 5 o f the 6 m o n t h s old C B A mice still have T d T positive spleens. Also 20ctSDH activity shows a n age d e p e n d e n t decrease in b o t h genotypes. Again, the decrease is more p r o n o u n c e d in B A L B / c t h a n in CBA. Since precursor T cells have been s h o w n to be the target ceils for M - M u L V infection in the bone m a r r o w [2, 9] it was of interest to c o m p a r e the enzyme activity of bone m a r r o w cells from C B A a n d B A L B / c mice. Due to the small n u m b e r of b o n e m a r r o w cells o b t a i n e d from one mouse, enzyme d e t e r m i n a t i o n s h a d to be limited to 20ctSDH activity only. Freshly harvested bone m a r r o w cells of 4 - m o n t h old mice were incubated with 3Hprogesterone for 1 h in tissue culture medium. Cells were then disrupted a n d processed for counting radioactivity according to the m e t h o d of Weinstein [5]. Interestingly, CBA bone marrow displayed consistently lower 20ctSDH activities t h a n B A L B / c bone m a r r o w (Fig. 2), the difference being significant at the 3°70 level (p = 0.028).
Biochemical analysis
80
Cell pellets or organ pieces kept at -70°C were used for the assays of enzyme activity. Homogenization was carried out in an Ultra-turrax homogenizer at a setting 7 for 2 x 15 s in 0.25 M KPO4, pH 7.4 containing 5 mM DTT. The homogenate was stirred for 30 rain at 4°C and centrifuged at 84,000 g for 1 h at 4°C. The supernatant was used for both enzyme assays and the pellet for DNA determination according to the method of Burton [10]. The assay for terminal deoxynucleotidyl transferase (TdT) was carried out as described previously [11]. The results are expressed as pmole dGTP polymerized per h and per l-tg DNA and enzyme values > 0.2 are regarded as positive. 20tt-hydroxysteroid dehydrogenase (20ctSDH) activity was determined according to Pepersack et al. [12]. The results are expressed as pmole progesterone converted to 20 OH-pregn-4en-3-one per h and per pg DNA and enzyme values > 0.2 are regarded as positive.
o
70
E ~
60
•
o o o
2
50
.,
40
~
30
c w
CBA
BALB/C
RESULTS
Enzyme activity in lymphoid organs o f normal mice Spleen a n d t h y m u s extracts were p r e p a r e d f r o m 5 individual B A L B / c a n d C B A mice, 1 a n d 6 m o n t h s old. T h e time points were chosen to be representative for the age o f preleukemic cell t r a n s f e r (one m o n t h ) [9] a n d the age o f overt leukemia ( a r o u n d 6 m o n t h s ) . Bone m a r r o w cells were harvested f r o m 4 - m o n t h old mice. A t this age 20ctSDH levels reach as plateau [8]. Figure 1 shows the levels o f T d T a n d 20ct S D H expression in spleen a n d t h y m u s o f B A L B / c a n d C B A mice. Even if there are great variations between individual mice the overall pattern o f enzyme expression is quite clear. In b o t h genotypes thymuses have higher T d T (Fig. 1A) a n d 20ctSDH (Fig. 1B) values t h a n spleens (50-100-fold difference in T d T activity).
FiG. 2. 20ctSDH activity in bone marrow of CBA (O) and BALB/c (O) mice at 4 months of age. In s u m m a r y , T d T and 20c~SDH activity is present in spleen, t h y m u s a n d b o n e m a r r o w of b o t h genotypes. There is however a quantitative difference between CBA a n d B A L B / c mice, particularly in 20ctSDH levels in spleen a n d b o n e marrow. In c o m p a r i s o n to C B A mice, B A L B / c mice have a higher 20ctSDH activity in b o n e m a r r o w ( p = 0 . 0 2 8 ) but a lower activity in spleen (p = 0.001).
Enzyme activity in enlarged lymphoid organs o f leukemic mice M - M u L V was inoculated to n e w b o r n B A L B / c and C B A mice. L y m p h o m a s a p p e a r e d 82-162 days a n d
o BIRGITrA ASJO a n d EVA MARIA FENYO
342
TABLE 1. LATENCY AND GROSS PATHOLOGY OF M-MuLV-INDUCED LYMPHOMAS
Mouse genotype
No. of mice Gross pathology*
Latency (days)
T
G
G/T
Range mean +_ S.E.
Balb/c
17
1
11
5
82-162
100 __+ 5
CBA
12
5
5
2
50-261
153 ± 20
*T: dominance of thymus enlargement. G: dominance of spleen and lymph node enlargement. G/T: thymus, spleen and lymph nodes equally enlarged.
50-~261 days later in B A L B / c and CBA mice, respectively (Table 1). In line with our previous observation the mean latency period was approximately 50 days longer in CBA mice (153 days) than in B A L B / c mice (100 days). The T-cell origin of lymphomas was confirmed by testing for Thy-1 antigen in a microcytotoxicity assay. Sixteen of seventeen (94°70) B A L B / c mice had generalized lymphomas with greatly enlarged spleens and lymph nodes. In 5 of these the thymus was also enlarged (Tables 1 and 2). Only one B A L B / c mouse had
an isolated thymoma. The dominating lymphoma phenotype was 20ctSDH-positive in B A L B / c mice. The only t h y m o m a was double negative. Three mice with equally enlarged spleens and thymuses showed discordant enzyme expression, in that the spleen was 20ctSDHpositive whereas the thymus was double-positive. In a slightly enlarged thymus this may be explained by the T d T activity of remaining normal cells. However, it is difficult to see how this explanation would apply to organs that reach several times the normal size. It is
TABLE 2. M - M u L V INDUCED T-CELL LYMPHOMASIN Balb/c MICE
Type of lymphoma* Organ tested 350 352 353 354 355
G G G G G/T
356 357
G G/T
359 360 390
G G G/T
391 392
G G/T
393 394 493 505 506
G G G T G/T
spleen spleen lymph node spleen spleen thymus spleen spleen thymus spleen spleen spleen thymus spleen spleen thymus spleen spleen spleen thymus spleen thymus
Enzyme activityt TdT 20txSDH Classification 0.12 0.03 0.01 0.07 0.06 0.97 0.07 0.10 0.33 0.06 0.01 0.09 0.16 0.55 0.11 2.70 0.10 0.31 0.13 0.15 0.49 0.27
1.57 6.45 0.11 1.10 0.72 1.50 0.26 0.59 2.04 1.15 0.90 4.91 12.76 J 2.65 0.94 2,09 8.21 0.32 0.89 0.01 0,69 1, 0.55 J
*G dominance of enlarged spleen and lymph nodes. T dominance of enlarged thymus. G/T thymus, spleen and lymph nodes equally enlarged. [Enzyme activities > 0.2 pmole/h/p_g DNA were scored as positive.
20aSDH 20ctSDH -/20aSDH 20ctSDH +/+ 20aSDH 20ctSDH +/+ 20aSDH 20c~SDH 20ctSDH +/+ 20aSDH +/+ 20ctSDH +/+ 20aSDH +/+
Influence of genotype in T-cell subtypes of Moloney lymphomas
343
TABLE3. M-MuLV |NDUCEDT-CELLLYMPHOMASIN CBA MICE
Type of lymphoma* Organ tested Y 185 Y 187
T G/T
Y 188 Y 198 Y 200 Y 209 Y 223 Y 224
G T T T G G
Y 237 Y 238 Y 255
G G G/T
Y 263
T
thymus spleen thymus spleen thymus thymus thymus lymph node lymph node spleen spleen spleen spleen thymus thymus
Enzyme activity1TdT 20ctSDH Classification 0.56 0.37 0.35 0.09 >2.48 3.32 22.47 106.39 1.08 0.27 0.11 0.29 8.44 20.06 1.30
0.01 0.04 0.09} 0.13 0.01 0.01 0.01 0.01 0.01 ~, 0.13J 0.04 3.95 0.42 0.10 0.01
TdT TdT TdT TdT TdT TdT TdT -/-
+/+ +/+ TdT TdT
*t See footnotes to Table 2. more likely that the discordant enzyme pattern represents two separate lymphomas, one 20ctSDHpositive the other double-positive in the same mouse. No lymphoma with TdT expression alone ~as found in neonatally inoculated BALB/c mice. In contrast to BALB/c, nearly half of the CBA mice (5/12) developed thymomas without apparent involvement of spleen and lymph nodes (Tables 1 and 3). The remaining 7 mice had generalized lymphomas with additional pronounced thymus enlargement in two cases. Unlike BALB/c, the dominant lymphoma phenotype (8/12, 66%) in CBA mice was TdT-positive (Table 3). One CBA mouse showed discordant enzyme pattern: the thymus was TdT-positive, whereas the spleen was double positive. No lymphoma with only 20ctSDH expression has been detected in CBA mice. Table 4 summarizes the enzyme phenotypes. BALB/c mice preferentially develop 20ctSDH lymphomas whereas lymphomas arising in CBA mice are predominantly TdT-positive.
TABLE 4. SUBTYPES OF T CELLS IN M-MuLV INDUCED LYMPHOMAS*
TdT
20ctSDH
+/ +
-/-
Balb/c
0t
9
3
2
CBA
8
0
1
2
*Mice with discordant enzyme patterns are not included. tNumber of lymphomas. + / + : positive for both enzymes. - / - : negative for both enzymes.
DISCUSSION In a previous study [9] we have shown that the subtype of T cell from which lymphomas arise after the transfer of preleukemic spleen cells is influenced by the mouse genotype. BALB/c mice preferentially developed 20ctSDH-positive lymphomas, whereas CBA lymphomas were most often TdT-positive. In the present study we have extended this observation and show that the pathological appearance and enzyme phenotype of lymphomas arising after the inoculation of M-MuLV to newborns varies with the mouse genotype. BALB/c mice often develop 20ctSDH-positive generalized lymphomas. In contrast CBA mice develop thymomas and generalized lymphomas in equal proportion with a strong preference for the TdT phenotype. Our study is the first to demonstrate the influence of mouse genotype on the subtype of T cell involved in M-MuLV-induced lymphomas. It was this difference in lymphoma phenotype that led us to study the different T-cell subtypes in spleen, thymus and bone marrow of normal BALB/c and CBA mice. It was quite unexpected to find that the spleen and thymus of CBA mice showed higher enzyme levels than the corresponding organs of BALB/c mice. The age related decrease of TdT activity was particularly pronounced in BALB/c spleens. In contrast, bone marrow showed the reverse situation, 20ctSDH activity was higher in BALB/c than in CBA at 4 months of age. Similarly Weinstein [8] found that the 20ctSDH activity in bone marrow of BALB/c mice is much higher than of C57B1 mice. Conceivably, M-MuLV infects precursor T ceils in the bone marrow and since 20ctSDH-positive cells are more abundant in BALB/c than in either C57B1 or CBA mice, this may influence the subtype of T cells involved in lymphoma. Indeed, generalized lymphomas occur frequently in BALB/c mice [1, 3]. We now know that these are predominantly 20ctSDH-positive. On the other hand, TdT-
o BIRGITTA ASJO a n d EVA MARIA FENYO
344
positive t h y m o m a is the dominating lymphoma type in C57B1 mice [1, 14]. Judged on the basis of 20ctSDH expression in bone marrow CBA mice assume an intermediary position between B A L B / c and C57B1 [8]. In line with this, both thymomas and generalized lymphomas occur in CBA. Whereas thymomas are uniformly TdT-positive, generalized lymphomas show a variety of phenotypes, TdT-positive, double-negative, doublepositive even, but never 20ctSDH-positive alone. The genetic background of the mouse may exert its effect on leukemogenesis through the control of virus replication. Rapidly developing viremia has been observed in B A L B / c mice whereas viremia increased more gradually in CBA mice [1]. In contrast, C57B1 mice showed low frequency of viremia [1]. Three immune response genes (Rmv-1, Rmv-2 and Rmv-3) linked to the major histocompatibility (H-2)complex have been identified that control M-MuLV viremia [15]. Rmv-1 and 2 seem to account for the major control of antibody response while Rmv-3 probably controls production of T killer lymphocytes specific for M-MuLV-induced cell surface F M R antigen [16]. It is conceivable that the rapidly developing and high level viremia favors the induction of 20ctSDH-positive lymphomas through some kind of chronic immune stimulation [17, 18]. An immune response capable to slow down virus replication would lead to the development of TdTpositive lymphomas. The fact that the TdT-positive CBA lymphomas arise about 50 days later than the 20ffSDH-positive B A L B / c lymphomas seems to corroborate this notion. Preliminary results suggest that indeed viremia is much delayed and of lower level in CBA than in B A L B / c mice. The source of the M-MuLV used in the present work is a tissue homogenate from leukemic spleens that has not been subjected to biological or molecular cloning. It is therefore likely to contain virus particles with different properties, some with a preference for 20ctSDHpositive cells others with specificity for the T d T lineage. If so, the phenotype of the target cell involved in leukemogenesis would be determined by the virus. Cional isolates of M - M u L V should give only one or the other type of lymphoma in mouse strains that allow development of both types of lymphoma. This type of experiment is presently under way. Acknowledgements--The expert technical assistance of Mrs Anita BergstrOm, Mrs Britt Samuelsson and Mrs Barbara ,~kerblom is gratefully acknowledged.
REFERENCES 1. Gisselbrecht S., Pozo F., Debr6 P., Hurot M. A., Lacombe M. J. & Levy J. P. (1978) Genetic control of sensitivity to Moloney-virus-induced leukemias in mice. I. Demonstration of multigenic control. Int. J. Cancer 21, 626. o 2. Spira J., Asj0 B., Cochran A., Shen F. W., Wiener F. & Klein G. (1981) Chromosomal, histopathological and cell surface marker studies on Moloney virus induced lymphomas. Leukemia Res. 5, 113.
3. Asj6 B., Skoog L., FenyO E. M. & Klein G. (1982) Different T-cell subtypes are associated with pathologically distinct forms of Moloney leukemia virus (M-MuLV)induced lymphoma. Int. J. Cancer 29, 163. 4. Kung P. C., Silverstone A. E., McCaffrey R. P. & Baltimore D. (1975) Murine terminal deoxynucleotidyl transferase: cellular distribution and response to cortisone. J. exp. Med. 141,855. 5. Weinstein Y. (1977) 200t-hydroxysteriod dehydrogenase: A T lymphocyte-associated enzyme. J. Immun. 119, 1223. 6. Weinstein Y. & Bercovich, Z. (1981)Testosterone effect on bone marrow, thymus and suppressor T cells in the (NZB × NZW)Fl mice: its relevance to autoimmunity. J. Immun. 126, 998. 7. Silverstone A. E., Cantor H., Goldstein G. & Baltimore D. 0976) Terminal deoxynucleotidyl transferase is found in prothymocytes. J. exp. Med. 144, 543. 8. Weinstein Y~ (1981) Expression of 20ct-hydroxysteroid dehydrogenase in the mouse marrow cells: strain differences, thymic effect on enzymatic activity, and possible localization in pre T lymphocytes. Thymus 2, 305. 9. Asj6 B., Skoog L., Palminger l., Wiener F., Isaak D., Cerny J. & FenyO E. M. (1985) Influence of genotype and the organ of origin on the subtype of T-cell in Moloney lymphomas induced by transfer of preleukemic cells from athymic and thymus-bearing mice. Cancer Res. 45, 1040. 10. Burton K. A. (1956) A study of conditions and mechanism of the diphenylamine reaction for the colorimetric estimation of deoxyribonucleic acid. Biochem. J. 62, 315. I I. Skoog L., NordenskjOld B., Ost ,~, Andersson B., Hast R., Giannoulis N., Humla S., H~igerstrt)m T. & Reizenstein P. (1981) Glucocorticoid receptor concentration and terminal transferase activity as indicators of prognosis in acute nonlymphocytic leukemia. Br. reed. J. 282, 1826. 12. Pepersack L., Lee J. C., McEwan R. & Ihle J. N. (1980) Phenotypic heterogeneity of Moloney leukemia virusinduced T cell lymphomas. J. Immun. 124, 279. 13. Pazmino N. H. & Ihle J. N. (1976) Strain-, and age-, and tumor-dependent distribution of terminal deoxynucleotidyl transferase in thymocytes of mice. J. lmmun. 117, 620. 14. Pazmino N. H., McEwan R. N. & lhle J. N. (1978) Radiation leukemia in C57B1/6 mice. III. Correlation of altered expression of terminal deoxynucleotidyl transferase to induction of leukemia. J. exp. Med. 148, 1338. 15. Debr6 P., Gisselbrecht S., Pozo F & L6vy J. P. (1979) Genetic control of sensitivity to Moloney leukemia virus in mice. I1. Mapping of three resistance genes within the H-2 complex. J. Immun. 123, 1806. 16. Debr6 P., Boyer B., Gisselbrecht S., Bismuth A. & L6vy J. P. (1980) Genetic control of sensitivity to Moloney leukemia virus in mice. III. The three H-2-1inked Rmv genes are immune response genes controlling the antiviral antibody response. Fur. J. Irnmun. 10, 914. 17. lhle J. N. & Lee J. C. (1982) Possible immunological mechanisms in C-type viral leukemogenesis in mice. Curr. Topics Microbiol. Immun. 91t, 85. 18. Lee J. C., Horak I. & lhle .1. N. (1981) Mechanisms in T cell leukemogeneisis. I1. T cell responses of preleukemic Balb/c mice to Moloney leukemia virus antigens. J. lmmun. 126, 715.
0145-2126/8653.00 ~ .00 c) 1986 Pergamon Press Ltd.
B A L B / c A N D CBA MICE DIFFER IN THE S U B T Y P E OF T CELL I N V O L V E D IN M O L O N E Y MURINE LEUKEMIA VIRUS-INDUCED LYMPHOMAS* BIRGITTA ,~S.IO and EVA MARIA FENYO Department of Virology, Karolinska Institutet, Stockholm, Sweden (Received 9 July 1985. Revision accepted 10 October 1985)
Abstract--Subsets of T cells can be recognized by presence or absence of terminal deoxynucleotidyl transferase (TdT) or 20ct-hydroxysteriod dehydrogenase (20ctSDH) activity. By use of these enzyme markers Moloney murine leukemia virus (M-MuLV)-induced lymphomas were shown to involve different T-cell subtypes. In the present paper we show that the genotype of the mouse has a strong influence on the subtype of T cell involved in lymphoma. BALB/c mice preferentially develop 20ctSDH-positive lymphomas, whereas CBA lymphomas often have the TdT phenotype. Comparison of the 20ctSDH activity of normal bone marrow cells showed that BALB/c mice have higher enzyme levels than CBA mice. The relative availability of a certain type of cell at a critical early step in leukemogenesis may thus influence lymphoma type. Key words: Moloney lymphoma and T-cell subset, terminal deoxynucleotidyl transferase, 20ct hydroxysteroid dehydrogenase.
INTRODUCTION INOCULATION OF M-MuLV to newborn mice leads to the development of T-cell lymphomas later in life. Two major forms of lymphoma can be distinguished according to the gross pathology. In one form the thymus is involved, mainly or exclusively, whereas in the other form the spleen and lymph nodes are enlarged. Gisselbrecht et al. [1] have shown that localization of a lymphoma to the thymus vs spleen and lymph nodes depends on the genetic background of the mouse. They found no linkage to the major histocompatibility (H-2) complex. In a previous work [2] we have observed differences in the transplantation properties of thymomas and generalized lymphomas. Upon inoculation to syngeneic hosts, thymomas grew as solid tumors at the site of inoculation, whereas cells of generalized lymphomas 'homed' to the spleen and lymph nodes. Based on these differences in transplantation properties we have assumed that the two pathological forms of lym-
*This work was supported by the Swedish Cancer Society, the Swedish Society for Medical Sciences and the National Science Foundation. Abbreviations: M-MuL V, Moloney murine leukemia virus; MCF, mink cell focus inducing virus; TdT, terminal deoxynucleotidyl transferase; 20ctSDH, 20ct-hydroxysteroid dehydrogenase; FMR, Friend, Moloney, Rauscher cell surface antigen; H-2, major histocompatibility complex; Rrnv-1, Rmv-2 and Rmv-3, genes for resistance to M°MuLV. o Correspondence to: Birgitta AsjO, Department of Virology, Karolinska Institutet, c/o SBL, S-105 21, Stockholm, Sweden. 339
phoma originate from different T-cell subsets. This assumption could be substantiated by using enzyme markers to define subsets of T cells [3]. It has been shown previously that in normal mice, immature cortical thymocytes are terminal deoxynucleotidyl transferase (TdT)-positive [4] whereas mature immunocompetent T cells in peripheral lymphoid organs and medullary thymocytes are 20ct hydroxysteroid dehydrogenase (20aSDH)-positive [5, 6]. Precursor cells of both subsets are present in the bone marrow [7, 8]. By determining the enzyme activity in M-MuLV induced T cell lymphomas we could indeed assign distinct T-cell subsets to the two different pathological forms of lymphoma [3]. Thymomas were TdT-positive whereas the majority of generalized lymphomas consisted of 20aSDH-positive cells. Working with the BALB/c and CBA strains we observed that the genetic background of the mouse influences the subtype of T cell involved in the lymphoma [9]. The particular system used was transfer of preleukemic cells from one-month old M-MuLV-inoculated mice to 400 rad irradiated syngeneic recipients. Donor cells were identified by the sex chromosome. Transfer of preleukemic BALB/c bone marrow resulted in both TdT-positive and 20aSDH-positive lymphomas whereas preleukemic spleen cells gave rise to 20ctSDH-positive lymphomas only. In contrast, none of the CBA lyrnphomas was 20ctSDH-positive regardless of which preleukemic organ was transferred. These results prompted us to ask the question whether BALB/c and CBA mice differ in the availability of any particular T-cell subtype. In the present study enzyme activities in thymus, spleen and bone marrow of normal and leukemic mice were measured. Normal CBA and BALB/c mice showed
BtRG[TTA ,~sJO and EVA MARIA FENYO
340
1000" 500"
100"
0
50'
|
Z E3 --t
";
10-
o E el
5-
.>_.
[ 0.s-
•
0 0
W
|
0.1
t
"
1
6
o
o
1
6
I1 1
spleen
6
thymus
1
CBA
6
thymus
spteen
BALBIc
B 50.0 "~
Z E3 O~ 1 e-
O
10.0'
|
8
5.0-
O
•
E
i
o
0 0
•
0 0
1.0-
0.5" E
O
O
N (Z LU
0.1
1 6 spleen
1 6 thymus
CBA
1 6 spree n
1 6 thy mus
BALBIc
FIG. 1. (A) TdT activity and (B) 20c(SDH activity in spleen and thymus of CBA (O) and BALB/c ( 0 ) mice at ! and 6 months of age.
341
Influence of genotype in T-cell subtypes of Moloney lymphomas q u a n t i t a t i v e differences particularly in 2 0 u S D H levels in spleen a n d b o n e m a r r o w . Following n e o n a t a l inoculation o f M - M u L V B A L B / c mice o f t e n developed generalized l y m p h o m a s with a 20ctSDH-positive p h e n o type whereas C B A mice preferentially developed T d T positive l y m p h o m a s .
MATERIALS
AND METHODS
Mice and virus CBA and BALB/c mice from our breeding stock were inoculated with Moloney murine leukemia virus (M-MuLV) when newborn. The source of M-MuLV was a 10070cell-free extract of BALB/c spleen homogenate, harvested 2 1/2 months after virus inoculation. It was obtained from the Division of Cancer Cause and Prevention, National Cancer Institute. Newborn mice were inoculated intraperitoneally with 0.1 ml of virus homogenate diluted 1:10. The inoculum contained 2.6 x 103 XC-PFU.
Thy-I determination The microcytotoxicity assay was carried out as described previously [2]. Anti Thy-1.2 monoclonal antibodies were kindly provided by Dr I. A. Clark, Seattle.
It is k n o w n t h a t physiological involution o f t h y m u s occurs with age a n d primarily effects the cortical TdTpositive thymocytes [13]. Accordingly, T d T activity decreased in t h y m u s a n d also in spleen. T h e decrease is particularly p r o n o u n c e d in B A L B / c mice; all 5 spleens are negative for T d T activity at 6 m o n t h s o f age. O n the o t h e r h a n d , 2 / 5 o f the 6 m o n t h s old C B A mice still have T d T positive spleens. Also 20ctSDH activity shows a n age d e p e n d e n t decrease in b o t h genotypes. Again, the decrease is more p r o n o u n c e d in B A L B / c t h a n in CBA. Since precursor T cells have been s h o w n to be the target ceils for M - M u L V infection in the bone m a r r o w [2, 9] it was of interest to c o m p a r e the enzyme activity of bone m a r r o w cells from C B A a n d B A L B / c mice. Due to the small n u m b e r of b o n e m a r r o w cells o b t a i n e d from one mouse, enzyme d e t e r m i n a t i o n s h a d to be limited to 20ctSDH activity only. Freshly harvested bone m a r r o w cells of 4 - m o n t h old mice were incubated with 3Hprogesterone for 1 h in tissue culture medium. Cells were then disrupted a n d processed for counting radioactivity according to the m e t h o d of Weinstein [5]. Interestingly, CBA bone marrow displayed consistently lower 20ctSDH activities t h a n B A L B / c bone m a r r o w (Fig. 2), the difference being significant at the 3°70 level (p = 0.028).
Biochemical analysis
80
Cell pellets or organ pieces kept at -70°C were used for the assays of enzyme activity. Homogenization was carried out in an Ultra-turrax homogenizer at a setting 7 for 2 x 15 s in 0.25 M KPO4, pH 7.4 containing 5 mM DTT. The homogenate was stirred for 30 rain at 4°C and centrifuged at 84,000 g for 1 h at 4°C. The supernatant was used for both enzyme assays and the pellet for DNA determination according to the method of Burton [10]. The assay for terminal deoxynucleotidyl transferase (TdT) was carried out as described previously [11]. The results are expressed as pmole dGTP polymerized per h and per l-tg DNA and enzyme values > 0.2 are regarded as positive. 20tt-hydroxysteroid dehydrogenase (20ctSDH) activity was determined according to Pepersack et al. [12]. The results are expressed as pmole progesterone converted to 20 OH-pregn-4en-3-one per h and per pg DNA and enzyme values > 0.2 are regarded as positive.
o
70
E ~
60
•
o o o
2
50
.,
40
~
30
c w
CBA
BALB/C
RESULTS
Enzyme activity in lymphoid organs o f normal mice Spleen a n d t h y m u s extracts were p r e p a r e d f r o m 5 individual B A L B / c a n d C B A mice, 1 a n d 6 m o n t h s old. T h e time points were chosen to be representative for the age o f preleukemic cell t r a n s f e r (one m o n t h ) [9] a n d the age o f overt leukemia ( a r o u n d 6 m o n t h s ) . Bone m a r r o w cells were harvested f r o m 4 - m o n t h old mice. A t this age 20ctSDH levels reach as plateau [8]. Figure 1 shows the levels o f T d T a n d 20ct S D H expression in spleen a n d t h y m u s o f B A L B / c a n d C B A mice. Even if there are great variations between individual mice the overall pattern o f enzyme expression is quite clear. In b o t h genotypes thymuses have higher T d T (Fig. 1A) a n d 20ctSDH (Fig. 1B) values t h a n spleens (50-100-fold difference in T d T activity).
FiG. 2. 20ctSDH activity in bone marrow of CBA (O) and BALB/c (O) mice at 4 months of age. In s u m m a r y , T d T and 20c~SDH activity is present in spleen, t h y m u s a n d b o n e m a r r o w of b o t h genotypes. There is however a quantitative difference between CBA a n d B A L B / c mice, particularly in 20ctSDH levels in spleen a n d b o n e marrow. In c o m p a r i s o n to C B A mice, B A L B / c mice have a higher 20ctSDH activity in b o n e m a r r o w ( p = 0 . 0 2 8 ) but a lower activity in spleen (p = 0.001).
Enzyme activity in enlarged lymphoid organs o f leukemic mice M - M u L V was inoculated to n e w b o r n B A L B / c and C B A mice. L y m p h o m a s a p p e a r e d 82-162 days a n d
o BIRGITrA ASJO a n d EVA MARIA FENYO
342
TABLE 1. LATENCY AND GROSS PATHOLOGY OF M-MuLV-INDUCED LYMPHOMAS
Mouse genotype
No. of mice Gross pathology*
Latency (days)
T
G
G/T
Range mean +_ S.E.
Balb/c
17
1
11
5
82-162
100 __+ 5
CBA
12
5
5
2
50-261
153 ± 20
*T: dominance of thymus enlargement. G: dominance of spleen and lymph node enlargement. G/T: thymus, spleen and lymph nodes equally enlarged.
50-~261 days later in B A L B / c and CBA mice, respectively (Table 1). In line with our previous observation the mean latency period was approximately 50 days longer in CBA mice (153 days) than in B A L B / c mice (100 days). The T-cell origin of lymphomas was confirmed by testing for Thy-1 antigen in a microcytotoxicity assay. Sixteen of seventeen (94°70) B A L B / c mice had generalized lymphomas with greatly enlarged spleens and lymph nodes. In 5 of these the thymus was also enlarged (Tables 1 and 2). Only one B A L B / c mouse had
an isolated thymoma. The dominating lymphoma phenotype was 20ctSDH-positive in B A L B / c mice. The only t h y m o m a was double negative. Three mice with equally enlarged spleens and thymuses showed discordant enzyme expression, in that the spleen was 20ctSDHpositive whereas the thymus was double-positive. In a slightly enlarged thymus this may be explained by the T d T activity of remaining normal cells. However, it is difficult to see how this explanation would apply to organs that reach several times the normal size. It is
TABLE 2. M - M u L V INDUCED T-CELL LYMPHOMASIN Balb/c MICE
Type of lymphoma* Organ tested 350 352 353 354 355
G G G G G/T
356 357
G G/T
359 360 390
G G G/T
391 392
G G/T
393 394 493 505 506
G G G T G/T
spleen spleen lymph node spleen spleen thymus spleen spleen thymus spleen spleen spleen thymus spleen spleen thymus spleen spleen spleen thymus spleen thymus
Enzyme activityt TdT 20txSDH Classification 0.12 0.03 0.01 0.07 0.06 0.97 0.07 0.10 0.33 0.06 0.01 0.09 0.16 0.55 0.11 2.70 0.10 0.31 0.13 0.15 0.49 0.27
1.57 6.45 0.11 1.10 0.72 1.50 0.26 0.59 2.04 1.15 0.90 4.91 12.76 J 2.65 0.94 2,09 8.21 0.32 0.89 0.01 0,69 1, 0.55 J
*G dominance of enlarged spleen and lymph nodes. T dominance of enlarged thymus. G/T thymus, spleen and lymph nodes equally enlarged. [Enzyme activities > 0.2 pmole/h/p_g DNA were scored as positive.
20aSDH 20ctSDH -/20aSDH 20ctSDH +/+ 20aSDH 20ctSDH +/+ 20aSDH 20c~SDH 20ctSDH +/+ 20aSDH +/+ 20ctSDH +/+ 20aSDH +/+
Influence of genotype in T-cell subtypes of Moloney lymphomas
343
TABLE3. M-MuLV |NDUCEDT-CELLLYMPHOMASIN CBA MICE
Type of lymphoma* Organ tested Y 185 Y 187
T G/T
Y 188 Y 198 Y 200 Y 209 Y 223 Y 224
G T T T G G
Y 237 Y 238 Y 255
G G G/T
Y 263
T
thymus spleen thymus spleen thymus thymus thymus lymph node lymph node spleen spleen spleen spleen thymus thymus
Enzyme activity1TdT 20ctSDH Classification 0.56 0.37 0.35 0.09 >2.48 3.32 22.47 106.39 1.08 0.27 0.11 0.29 8.44 20.06 1.30
0.01 0.04 0.09} 0.13 0.01 0.01 0.01 0.01 0.01 ~, 0.13J 0.04 3.95 0.42 0.10 0.01
TdT TdT TdT TdT TdT TdT TdT -/-
+/+ +/+ TdT TdT
*t See footnotes to Table 2. more likely that the discordant enzyme pattern represents two separate lymphomas, one 20ctSDHpositive the other double-positive in the same mouse. No lymphoma with TdT expression alone ~as found in neonatally inoculated BALB/c mice. In contrast to BALB/c, nearly half of the CBA mice (5/12) developed thymomas without apparent involvement of spleen and lymph nodes (Tables 1 and 3). The remaining 7 mice had generalized lymphomas with additional pronounced thymus enlargement in two cases. Unlike BALB/c, the dominant lymphoma phenotype (8/12, 66%) in CBA mice was TdT-positive (Table 3). One CBA mouse showed discordant enzyme pattern: the thymus was TdT-positive, whereas the spleen was double positive. No lymphoma with only 20ctSDH expression has been detected in CBA mice. Table 4 summarizes the enzyme phenotypes. BALB/c mice preferentially develop 20ctSDH lymphomas whereas lymphomas arising in CBA mice are predominantly TdT-positive.
TABLE 4. SUBTYPES OF T CELLS IN M-MuLV INDUCED LYMPHOMAS*
TdT
20ctSDH
+/ +
-/-
Balb/c
0t
9
3
2
CBA
8
0
1
2
*Mice with discordant enzyme patterns are not included. tNumber of lymphomas. + / + : positive for both enzymes. - / - : negative for both enzymes.
DISCUSSION In a previous study [9] we have shown that the subtype of T cell from which lymphomas arise after the transfer of preleukemic spleen cells is influenced by the mouse genotype. BALB/c mice preferentially developed 20ctSDH-positive lymphomas, whereas CBA lymphomas were most often TdT-positive. In the present study we have extended this observation and show that the pathological appearance and enzyme phenotype of lymphomas arising after the inoculation of M-MuLV to newborns varies with the mouse genotype. BALB/c mice often develop 20ctSDH-positive generalized lymphomas. In contrast CBA mice develop thymomas and generalized lymphomas in equal proportion with a strong preference for the TdT phenotype. Our study is the first to demonstrate the influence of mouse genotype on the subtype of T cell involved in M-MuLV-induced lymphomas. It was this difference in lymphoma phenotype that led us to study the different T-cell subtypes in spleen, thymus and bone marrow of normal BALB/c and CBA mice. It was quite unexpected to find that the spleen and thymus of CBA mice showed higher enzyme levels than the corresponding organs of BALB/c mice. The age related decrease of TdT activity was particularly pronounced in BALB/c spleens. In contrast, bone marrow showed the reverse situation, 20ctSDH activity was higher in BALB/c than in CBA at 4 months of age. Similarly Weinstein [8] found that the 20ctSDH activity in bone marrow of BALB/c mice is much higher than of C57B1 mice. Conceivably, M-MuLV infects precursor T ceils in the bone marrow and since 20ctSDH-positive cells are more abundant in BALB/c than in either C57B1 or CBA mice, this may influence the subtype of T cells involved in lymphoma. Indeed, generalized lymphomas occur frequently in BALB/c mice [1, 3]. We now know that these are predominantly 20ctSDH-positive. On the other hand, TdT-
o BIRGITTA ASJO a n d EVA MARIA FENYO
344
positive t h y m o m a is the dominating lymphoma type in C57B1 mice [1, 14]. Judged on the basis of 20ctSDH expression in bone marrow CBA mice assume an intermediary position between B A L B / c and C57B1 [8]. In line with this, both thymomas and generalized lymphomas occur in CBA. Whereas thymomas are uniformly TdT-positive, generalized lymphomas show a variety of phenotypes, TdT-positive, double-negative, doublepositive even, but never 20ctSDH-positive alone. The genetic background of the mouse may exert its effect on leukemogenesis through the control of virus replication. Rapidly developing viremia has been observed in B A L B / c mice whereas viremia increased more gradually in CBA mice [1]. In contrast, C57B1 mice showed low frequency of viremia [1]. Three immune response genes (Rmv-1, Rmv-2 and Rmv-3) linked to the major histocompatibility (H-2)complex have been identified that control M-MuLV viremia [15]. Rmv-1 and 2 seem to account for the major control of antibody response while Rmv-3 probably controls production of T killer lymphocytes specific for M-MuLV-induced cell surface F M R antigen [16]. It is conceivable that the rapidly developing and high level viremia favors the induction of 20ctSDH-positive lymphomas through some kind of chronic immune stimulation [17, 18]. An immune response capable to slow down virus replication would lead to the development of TdTpositive lymphomas. The fact that the TdT-positive CBA lymphomas arise about 50 days later than the 20ffSDH-positive B A L B / c lymphomas seems to corroborate this notion. Preliminary results suggest that indeed viremia is much delayed and of lower level in CBA than in B A L B / c mice. The source of the M-MuLV used in the present work is a tissue homogenate from leukemic spleens that has not been subjected to biological or molecular cloning. It is therefore likely to contain virus particles with different properties, some with a preference for 20ctSDHpositive cells others with specificity for the T d T lineage. If so, the phenotype of the target cell involved in leukemogenesis would be determined by the virus. Cional isolates of M - M u L V should give only one or the other type of lymphoma in mouse strains that allow development of both types of lymphoma. This type of experiment is presently under way. Acknowledgements--The expert technical assistance of Mrs Anita BergstrOm, Mrs Britt Samuelsson and Mrs Barbara ,~kerblom is gratefully acknowledged.
REFERENCES 1. Gisselbrecht S., Pozo F., Debr6 P., Hurot M. A., Lacombe M. J. & Levy J. P. (1978) Genetic control of sensitivity to Moloney-virus-induced leukemias in mice. I. Demonstration of multigenic control. Int. J. Cancer 21, 626. o 2. Spira J., Asj0 B., Cochran A., Shen F. W., Wiener F. & Klein G. (1981) Chromosomal, histopathological and cell surface marker studies on Moloney virus induced lymphomas. Leukemia Res. 5, 113.
3. Asj6 B., Skoog L., FenyO E. M. & Klein G. (1982) Different T-cell subtypes are associated with pathologically distinct forms of Moloney leukemia virus (M-MuLV)induced lymphoma. Int. J. Cancer 29, 163. 4. Kung P. C., Silverstone A. E., McCaffrey R. P. & Baltimore D. (1975) Murine terminal deoxynucleotidyl transferase: cellular distribution and response to cortisone. J. exp. Med. 141,855. 5. Weinstein Y. (1977) 200t-hydroxysteriod dehydrogenase: A T lymphocyte-associated enzyme. J. Immun. 119, 1223. 6. Weinstein Y. & Bercovich, Z. (1981)Testosterone effect on bone marrow, thymus and suppressor T cells in the (NZB × NZW)Fl mice: its relevance to autoimmunity. J. Immun. 126, 998. 7. Silverstone A. E., Cantor H., Goldstein G. & Baltimore D. 0976) Terminal deoxynucleotidyl transferase is found in prothymocytes. J. exp. Med. 144, 543. 8. Weinstein Y~ (1981) Expression of 20ct-hydroxysteroid dehydrogenase in the mouse marrow cells: strain differences, thymic effect on enzymatic activity, and possible localization in pre T lymphocytes. Thymus 2, 305. 9. Asj6 B., Skoog L., Palminger l., Wiener F., Isaak D., Cerny J. & FenyO E. M. (1985) Influence of genotype and the organ of origin on the subtype of T-cell in Moloney lymphomas induced by transfer of preleukemic cells from athymic and thymus-bearing mice. Cancer Res. 45, 1040. 10. Burton K. A. (1956) A study of conditions and mechanism of the diphenylamine reaction for the colorimetric estimation of deoxyribonucleic acid. Biochem. J. 62, 315. I I. Skoog L., NordenskjOld B., Ost ,~, Andersson B., Hast R., Giannoulis N., Humla S., H~igerstrt)m T. & Reizenstein P. (1981) Glucocorticoid receptor concentration and terminal transferase activity as indicators of prognosis in acute nonlymphocytic leukemia. Br. reed. J. 282, 1826. 12. Pepersack L., Lee J. C., McEwan R. & Ihle J. N. (1980) Phenotypic heterogeneity of Moloney leukemia virusinduced T cell lymphomas. J. Immun. 124, 279. 13. Pazmino N. H. & Ihle J. N. (1976) Strain-, and age-, and tumor-dependent distribution of terminal deoxynucleotidyl transferase in thymocytes of mice. J. lmmun. 117, 620. 14. Pazmino N. H., McEwan R. N. & lhle J. N. (1978) Radiation leukemia in C57B1/6 mice. III. Correlation of altered expression of terminal deoxynucleotidyl transferase to induction of leukemia. J. exp. Med. 148, 1338. 15. Debr6 P., Gisselbrecht S., Pozo F & L6vy J. P. (1979) Genetic control of sensitivity to Moloney leukemia virus in mice. I1. Mapping of three resistance genes within the H-2 complex. J. Immun. 123, 1806. 16. Debr6 P., Boyer B., Gisselbrecht S., Bismuth A. & L6vy J. P. (1980) Genetic control of sensitivity to Moloney leukemia virus in mice. III. The three H-2-1inked Rmv genes are immune response genes controlling the antiviral antibody response. Fur. J. Irnmun. 10, 914. 17. lhle J. N. & Lee J. C. (1982) Possible immunological mechanisms in C-type viral leukemogenesis in mice. Curr. Topics Microbiol. Immun. 91t, 85. 18. Lee J. C., Horak I. & lhle .1. N. (1981) Mechanisms in T cell leukemogeneisis. I1. T cell responses of preleukemic Balb/c mice to Moloney leukemia virus antigens. J. lmmun. 126, 715.