Vol. 11, No. 5 Printed in U.S.A.
INFECTION AND IMMUNITY, May 1975, p. 1024-1030 Copyright 0 1975 American Society for Microbiology
Immunodepression by Rowson-Parr Virus in Mice: Effect of Rowson-Parr Virus and Friend Leukemia Complex Infections on Background Antibody-Forming Cells to Various Erythrocytes M. BENDINELLI,* A. TONIOLO, AND M. CAMPA Institute of Microbiology, University of Pisa, 56100 Pisa, Italy Received for publication 30 December 1974
The numbers of background antibody-forming cells (BPFC) toward erythrocytes of various species present in the lymphoid organs of unimmunized susceptible BALB/c and resistant C57BL/6 mice were investigated at various times after infection with Friend leukemia complex (FLC) or Rowson-Parr virus (RPV). Both virus preparations induced an increase of BPFC numbers in both animal strains, but the rate and magnitude of the enhancements produced by RPV were much lower. The degree of potentiation varied with the specificity of the BPFC populations and was more pronounced in the spleen than in the lymph nodes and in BALB/c than in C57BL/6 mice. In the late stage of FLC infection, the numbers of splenic BPFC to some erythrocytes underwent a dramatic fall, which was not observed in RPV-infected mice. BPFC present in BALB/c splenocytes cultured in diffusion chambers implanted in the peritoneal cavity of isogeneic normal mice were not affected by viral infection of the chambers.
Recently it has been recognized that Friend leukemia virus is a complex of various entities (19, 28, 31), one of which is the Rowson-Parr virus (RPV). RPV, which can be easily isolated from Friend leukemia complex (FLC) preparations (6, 28) because of the high titers it attains in FLC-infected mice (8), induces an early depression of the antibody response in BALB/c mice (9, 13). This finding has suggested that RPV plays an important role in the immunodepressive effects of FLC preparations and consequently cooperates in the ability of FLC to induce leukemogenesis (5). Since it has not yet been possible to free FLC from RPV, the only available experimental approach to the above hypothesis is to compare the immunodepressive properties of RPV with those of the complex. In a previous paper, in which the effect of RPV and FLC infections on the responsiveness of adult BALB/c mice to sheep erythrocytes (SRC) was studied, we found that the number of natural background plaqueforming cells (BPFC) to SRC present in the spleen of nonimmunized animals is greatly increased by FLC but not by RPV (9). This aspect is further investigated here by extending the range of target erythrocytes. It is shown that although both RPV and FLC may increase the numbers of BPFC, there are considerable differences between the effects exerted by the two
virus preparations. In addition, the results are presented of experiments performed to elucidate the mechanism whereby BPFC numbers are increased after infection. MATERIALS AND METHODS Mice. Male inbred BALB/c and random-bred C57BL/6 mice grown in this Institute and aged 8 to 12 weeks were used. Viruses. The strains of RPV and FLC used were free of the lactic dehydrogenase virus and were prepared as previously described (8). Their passage histories (4) indicate that they are NB tropic. FLC titer was 103-5 mean infectious doses per 0.1 ml. RPV, obtained from newborn mice, titrated 105 ° mean infectious doses per 0.1 ml. Infection was performed by retro-orbital injection of 0.1 ml of virus preparation. Assay for BPFC. Hemolytic plaque-forming cells were assayed by the method of Jerne et al. (24). Unless otherwise specified, the cells were suspended in Eagle minimum essential medium and separately tested against the desired erythrocytes. Sheep (SRC), chicken (CRC), horse (HRC), donkey (DRC), rabbit (RRC) and human group 0 erythrocytes (HuRC) were used; they were maintained in Alsever solution and used between 1 and 5 weeks after collection. A 0.1-ml amount of a 20% suspension of washed erythrocytes in saline was used in each plate. For each antigen at least two plates were prepared. All plates were inspected for false plaques before addition of complement. The anti-mouse immunoglobulin rabbit serum (Miles Laboratories, Kankakee, Ill.) used in some
1024
VOL. 11, 1975
ANTIBODY-FORMING CELLS IN RPV AND FLC INFECTIONS
experiments was diluted 1:10 before use. For each group the geometric mean of BPFC was calculated together with its 95% confidence limits, and the statistical significance between groups was assessed by Student's t test. Membrane filter diffusion chamber cultures. Spleen cell cultures in diffusion chambers were established by the method described by Capalbo et al. (12). Once removed from the hosts, the chambers were immersed in 0.5% Pronase (Sigma, St. Louis, Mo.) in minimum essential medium for 1 h at 28 C to improve cell recovery. Pronase was diluted out by keeping the chambers in minimum essential medium for 30 min before collecting the cultured cells.
RESULTS Effect of FLC on absolute numbers of BPFC in the spleens of BALB/c mice. Groups of normal mice and of mice infected with FLC on day - 20, - 15, -10, or -5 were tested for splenic BPFC to SRC, CRC, HRC, DRC, RRC, and HuRC (Fig. 1). In the controls, BPFC averaged 341 to CRC, 84 to SRC, 68 to HRC, and 28 to DRC, and no plaques were detected when RRC or HuRC were used as target. After FLC infection there was a progressive enhancement of BPFC to CRC, with a maximum of 6.7fold at 20 days. BPFC to SRC and HRC were
1025
increased in the first 15 days (with a maximum of 4.3-fold on day 10 for the former and 3.3-fold on day 15 for the latter) and then rapidly retumed to normal values on day 20. BPFC to DRC were moderately potentiated at days 5 and 10 and thereafter declined to normal levels. BPFC numbers to RRC and HuRC remained below detectability. Effect of RPV on the absolute numbers of BPFC in the spleens of BALB/c mice. The BPFC that had been most affected by FLC infection (i.e., BPFC to SRC, CRC, and HRC) were investigated (Fig. 2). BPFC levels in normal animals were comparable with those found in the controls of the FLC experiments. In contrast, the consequences of RPV infection markedly differed from those of FLC infection. The numbers of BPFC of all specificities underwent a slow increase that was evident 10 to 15 days after infection and did not significantly decline in the subsequent days. The magnitude of the increase was generally lower than in FLC-infected mice; the BPFC that had been most affected by FLC were the least affected by RPV, and vice versa. The peak increases in BPFC number were 3.5-fold for HRC, 2.8-fold for SRC, and 1.8-fold for CRC. Splenic BPFC numbers versus spleen weight in infected BALB/c mice. During RPV infection and in the early stages of FLC infection (at least for a period of 2 weeks with the virus titers used in these experiments), as well as in normal mice, the weight of the spleen is directly proportional to the number of nucleated cells recoverable from the organ (Bendinelli, unpublished data). To estimate the behavior of BPFC as compared with that of the
z w a.
g
.ol a.
0
5 10 DAYS AFTER FLC
15
20
INFECTION
FIG. 1. Absolute numbers of BPFC to chicken (0), sheep (V), ho,rse (0), donkey (v), rabbit (-), and human (0) erythrocytes in the spleens of BALBIc mice at various times after infection with FLC. Numbers of mice are indicated near each symbol.
x
0
5
10
DAYS AFTER RPV
15 20 NFECTION
FIG. 2. Absolute numbers of BPFC to chicken (0), sheep (V), and horse (0) erythrocytes in the spleens of BALBIc mice at various times after infection with RPV. Numbers of mice are indicated near each symbol.
1026
INFECT. IMMUN.
BENDINELLI, TONIOLO, AND CAMPA
total sple en cellularity, the results of the previous two g roups of experiments are presented in Fig. 3. Dluring the first 5 to 10 days of FLC infection, the BPFC of all specificities increased at a rate aipproximating that of the entire spleen cell popuilation. After this time, although in some cas4 es the absolute BPFC numbers were still incre asing, their ratios rapidly declined. In contrast, in RPV-infected mice BPFC numbers maintain(ed at a fairly constant ratio with spleen we ight. It shouild be noticed, however, that RPV caused a much lower splenomegaly than did FLC (Fig 1 and 2). Specifi[city and immunological nature of BPFC fri Dm normal and infected spleens. The correlatio In coefficient was determined between all the po *ssible combinations of the numbers of BPFC of different specificities detected both in normal aynd infected individuals. In no case was
z
the probability of linear correlation significant. Moreover, in some experiments cell suspensions from normal and FLC- or RPV-infected spleens were assayed against a mixture of two types of erythrocytes; plaques with complete lysis were rarely observed, and their number never exceeded 1% of total plaques. These results clearly demonstrate that both in normal and infected spleens at least the great majority of BPFC directed towards different erythrocytes belong to different populations. To ascertain the immunological nature of the plaques, experiments were performed in which normal or infected spleen cell suspensions containing equal numbers of cells were plated against various erythrocytes and the plates were incubated for 1 h with anti-mouse immunoglobulin rabbit serum or with normal rabbit serum before addition of complement. The results of one of these experiments are given in Table 1 and show that BPFC (to SRC, CRC, and HRC) from normal and infected spleens were equally susceptible to partial inhibition by normal rabbit serum and to almost complete suppression by the anti-immunoglobulin serum. These data, together with the fact that
w
1.0 _
J
CLIXn
no
plaques developed
in the absence of complement, demonstrate the
immunological nature of BPFC present in
_
mal and infected
UL
Effect of FLC and RPV infections on BPFC numbers in the lymph nodes of BALB/c mice. In one experiment groups of mice were infected with FLC on day -20 or -10 and the superficial lymph nodes (axillary, inguinal, and cervical) from each individual mouse were collected, pooled, weighed, and tested for BPFC to SRC and CRC on day 0. The mesenteric lymph nodes
0
U. IL CD
FLC
R PV I O
II
I
10
II
I
I
20
DAYS AFTER
10
0
nor-
spleens.
20
INFECTION
FIG. 3. 1Numbers of BPFC to chicken (0), sheep (V), horse (0), and donkey (V) erythrocytes per milligram of spleen weight at various times after infection w 'ith RPV and FLC in BALBIc mice.
were separately processed in the same way. Ten
days postinfection both superficial and
mesen-
teric lymph nodes contained numbers of BPFC similar to the controls, whereas 20 days after
infection superficial lymph nodes still had normal numbers of BPFC, but mesenteric lymph nodes showed increased BPFC counts (Table 2,
TABLE 1. Inhibition of background plaques to SRC, CRC, and HRC in normal, RPV-infected or FLC-infected BALBIc spleens by normal rabbit serum or anti-mouse immunoglobulin rabbit serum Background plaques to:
Normal mice .............. RPV-infected mice ......... FLC-infected mice .........
NRSa
Anti-IgRSb
NRS
Anti-IgRS
NRS
Anti-IgRS
71.7c 76.5 64.4
89.6 91.5 89.1
87.2 84.4 89.7
100 97.5 98.3
77.2 75.6 74.4
84.9 96.4 95.3
Normal rabbit serum. b Anti-mouse immunoglobulin rabbit serum. c Percentage of inhibition relative to untreated plates.
a
HRC
CRC
SRC
Cell donor
1027
ANTIBODY-FORMING CELLS IN RPV AND FLC INFECTIONS
VOL. 11, 1975
experiment 1). The difference between mean numbers of BPFC to CRC in the controls and in the mice infected at day -20 was significant (P < 0.01). The mean weight of mesenteric lymph nodes was also considerably augmented by FLC infection, whereas that of superficial lymph nodes showed little, if any, change. In a second experiment (Table 2, experiment 2), the potentiation of BPFC in mesenteric lymph nodes 20 days after FLC infection was confirmed and RPV was found only to produce minor changes on mesenteric lymph node BPFC. Effect of RPV and FLC infections on BPFC in spleen cells cultured in diffusion chambers. Viable spleen cells (3 x 107) from normal BALB/c mice (containing 14 BPFC to SRC and 42 BPFC to CRC per 107 cells) were placed into 50-nm membrane filter (Millipore Corp.) diffusion chambers and mixed with normal mouse plasma or with RPV or FLC, taking care not to infect the extemal of the chambers. The sealed chambers were then implanted into the peritoneal cavity of normal isogeneic hosts. Fourteen days later the chambers were removed from the mice, and the cells were counted and assayed for BPFC to SRC and CRC. The recovery of
cells and of BPFC from the chambers was low: from six uninfected chambers 1.2 x 107 cells were obtained, and these contained 5 BPFC to SRC (4/107 cells) and 9 BPFC to CRC (7/107 cells). Infection slightly increased cell recovery but did not significantly modify the recovery of BPFC. From five RPV-infected chambers 1.5 x 107 cells were recovered, which produced 4 plaques to SRC (3/107 cells) and 4 to CRC (3/107 cells). From six FLC-infected chambers 2.0 x 107 cells were recovered, and BPFC were 7 to SRC (3/107 cells) and 9 to CRC (5/107 cells). Effect of FLC and RPV infections on BPFC in the spleens of C57BV6 mice. C57BL/6 mice are known to be resistant to Friend leukemia. After FLC infection they show a very slight and transient splenomegaly, and after RPV infection their spleen weights remain in the normal range. Groups of mice, either normal or infected with FLC or RPV on day 20 or 10, were tested for splenic BPFC to SRC, CRC, and HRC. Mice infected in either way 10 days earlier had slightly increased numbers of BPFC to all the erythrocytes tested (Table 3). The differences were significant in the FLC-infected group for BPFC to SRC (P < 0.05) and in the RPVinfected group for BPFC to CRC (P < 0.01). By -
-
TABLE 2. Effect of RPV and FLC infection on the number of BPFC to SRC and CRC present in the lymph nodes of BALB/c mice Superficial lymph nodes Expt no.
a
Pretreatment of mice
No. miceof
Mean wt (mg)
32 39 49
1
None FLC at day - 10 FLC at day -20
7 8 10
2
None RPV at day -20 FlC at day -20
10 8 8
BPFC to SRC
BPFC to CR(
14a (5-35) 12 (7-19) 12 (6-24) 10 (5-21) 16 (7-40) 11 (5-29)
Mesenteric lymph nodes BPFC BPFC Mean wt (mg) to SRC to CRC
69 93 136
19 (6-60) 15 (8-29) 12 (8-18) 11 (6-17) 39 (21-75) 83 (34-201)
69 62 116
5 (3-10) 21 (8-58) 10 (3-36) 32 (8-133) 14 (5-36) 45 (25-87)
Geometric mean (in parentheses, the 95% confidence limits).
TABLE 3. Effect of RPV and FLC infection on the number of BPFC to SRC, CRC, and HRC present in the spleens of C57 BL/6 mice Pretreatment of mice
Spleen wt (mg)
None
129 (88-163)b
RPV at day -10 RPV at day -20 FLC at day -10 FLC at day -20
128(97-153) 150(102-156) 184(121-269) 175(149-221)
aIn parentheses, numbers of mice tested. b Mean i range.
BPFC to SRC 95%
BPFC to CRC
BPFC to HRC
95% 95% Geometric Confidence Geometric Confidence Geometric Confidence limits mean mean mean limits limits
11 (19)a 23(6) 8 (8) 40(15) 13(8)
5-24 2-261 2-30 18-90 3-54
27 (19) 141 (6) 55(8)
60(15) 74(8)
13-53 46-662 14-220 36-102 32-169
4 (8) 21(6) 6 (8) 15 (4) 11 (8)
0-15 3-164 1-27 8-28 3-42
1028
BENDINELLI, TONIOLO, AND CAMPA
INFECT. IMMUN.
The effects of RPV infection on BPFC in the day 20 postinfection BPFC were again at normal levels except those directed to CRC in the spleens of BALB/c mice are substantially differFLC-infected group, which were still, though ent from those exerted by FLC. Even RPV augments the numbers of BPFC, but not as not significantly, potentiated. rapidly and markedly as the entire complex. DISCUSSION Significant enhancements are first detectable Background plaques are known to result from 10 to 15 days after infection, and in the subsethe activity of single cells releasing 19S anti- quent 5 to 10 days there are few further modifibody (21), but the origin and significance of cations. The kinetics of increase of different these cells are still uncertain. Being absent or BPFC is not completely asynchronous as after very few in young animals (4, 21), they are FLC infection; in fact, the extent of BPFC thought to arise as a consequence of continuous potentiation parallels the slight spleen enlargestimulation by antigens, such as microorga- ment caused by RPV. Moreover, the various nisms, which share determinants with erythro- BPFC are affected differently by the two virus cytes, but definite proof is still lacking (25, 32). preparations: the highest increase is achieved Because the formation of BPFC is thymus by BPFC to CRC in the FLC-infected spleens independent, and for other reasons (1, 21), it is and by BPFC to HRC in the RPV-infected generally agreed that they are not precursors spleens. These findings explain how in earlier of-nor are in any way connected with-the experiments performed with SRC alone the antibody-forming cells elicited by inoculation potentiating activity of RPV on BPFC escaped with the corresponding antigen (21). detection (9). Also, the counts of BPFC in Several treatments have been shown to in- RPV-infected mesenteric lymph nodes were crease the number of BPFC to SRC in the little changed. mouse spleen. These include administration of The increase in BPFC numbers produced by phytohemagglutinin, bacterial lipopolysaccha- FLC and RPV infections may have several rides, and unrelated antigens (11, 21). The explanations. One possibility is that BPFC ability of FLC to enhance the number of BPFC production is stimulated. This might occur to SRC in adult mice was originally described because host defenses are lowered so that invadby Hirano et al. (22, 23) and confirmed by us (9) ing microorganisms stimulate BPFC formation and others (17). In the present study FLC and through the activity of cell wall lipopolysacchaRPV are shown to affect all the BPFC studied rides (11, 25, 33). This explanation, however, that are normally present in consistent numbers contrasts with the augmentation of the reticuloin the spleens of BALB/c mice, i.e., those endothelial system activity (26) and the normal directed against SRC, CRC, HRC, and DRC. levels of complement (10) of FLC-infected mice. The modifications observed during FLC in- Alternatively, FLC and RPV might increase fection on these BPFC can be divided into three BPFC production acting as mitogens. Mitogens, distinct phases. Initially, their absolute num- as well as many thymus-independent antigens bers increase and the ratios of BPFC to spleen capable of unspecific stimulation of lymphocyte cellularity remain unchanged. In the second division, may cause a burst of policlonal antiphase, the numbers per spleen remain steady or body synthesis in vivo (11) and in vitro (2, 18, still increase, but the ratios of BPFC to cellular- 30). A variant of the last explanation is that ity decrease. This is usually followed by a third immunologically committed cells are preferenphase, during which the absolute BPFC num- tially stimulated. It has been observed that bers increase and the ratios of BPFC to spleen antibody-forming cell numbers may increase in phases are not synchronous for BPFC of differ- the spleens of mice primed several weeks before ent specificities. For instance, 20 days after and then infected with FLC (16). Both these infection the absolute numbers of BPFC to CRC explanations, if maintained in the case of RPV, are still increasing, whereas those of BPFC are difficult to reconcile with the differential directed against SRC, HRC, and DRC have activity of FLC on the BPFC of various specalready markedly declined. Moreover, there is a ificities. As a further alternative, BPFC producremarkable lack of homogeneity in the magni- tion might be increased due to a transient lowertude of the enhancement of different BPFC. A ing of the threshold of immunocytes to antigenic similar divergent behavior after FLC infection triggering. In this case BPFC potentiation would has been observed between BPFC to SRC and also depend on the availability of cross-reacting to Escherichia coli (23) and is in keeping with antigens, explaining why BPFC that are absent the concept that different BPFC belong to or very few in the normal spleen (such as those different cell populations, as shown by the to RRC and HuRC) are not induced by infecabsence of linear correlation between various tion. This mode of action, however, sharply contrasts with the depression of immunological BPFC in this as well as other studies (21).
VOL. 11, 1975
ANTIBODY-FORMING CELLS IN RPV AND FLC INFECTIONS
reactivity that characterizes FLC and RPV infections (15, 20, 29). Cross-reactivity between the viral antigens and the antigens used as targets to detect BPFC is ruled out by previous observations (23) and by the wide diversity of erythrocytes with which cross-reaction should occur to justify the present data. It is also possible that BPFC in infected mice are formed at a normal rate and that the increase in the spleen is due to modifications of their behavior. For instance, their decay might be slower in infected than in normal animals. Nothing is known about the life span of BPFC, and preliminary findings showing that normal and infected BPFC have similar survival times in vitro do not rule out this possibility. Also, since there is evidence that the secretion of antibody by BPFC is a pulsating phenomenon (27), FLC and RPV might stimulate preexisting low-producer antibody-forming cells to release quantities of antibody large enough to permit detection. This explanation, however, does not account for the lack of parallelism in the potentiation of different BPFC by FLC infection and contrasts with the finding that BPFC of infected spleens are inhibited by anti-immunoglobulin serum as effectively as those of normal mice. Alternatively, BPFC produced elsewhere (34) might accumulate in the spleen as a result of an alteration in their circulation. Lymph nodes are less susceptible to the immunodepressive effect of FLC and RPV (6) and might therefore continue to function as a source of BPFC. In FLC-infected mice the spleen accumulates lymphoid cells, whereas superficial lymph nodes show neither gross defects in lymphocyte circulation (3) nor change in BPFC numbers. The lack of synchrony and homogeneity between the effects of FLC on various BPFC might be due to differences in their rate of production and circulation. This hypothesis is in keeping with the failure of FLC and RPV to affect BPFC numbers in diffusion chambers and with the previous observation that FLC suppresses the appearance of BPFC in the spleens of immunologically immature mice (4), but attempts to test such a possibility directly have been hampered by failure to cause a lasting abatement of splenic BPFC by irradiation or other means (1, 21; unpublished data). No gross modifications of lymphoid cell circulation have, however, been observed in RPVinfected spleens (3). The decrease of BPFC-to-spleen cellularity ratios that occurs in the late stages of FLC infection but not in RPV infection is most likely due to dilution of BPFC by the transformed cells that proliferate quickly in the spleens of FLC-infected mice. However, a reduction of
1029
BPFC production, arrival, and/or survival can also contribute, as shown by the frequent decline in the absolute numbers of BPFC at this time. The other discrepancies observed in the effects of the two virus preparations cannot be accounted for on the basis of the virus titers used for infection, since the batch of RPV used was obtained from neonatally infected mice in which viral titers approached those present in FLC preparations (8). However, in adult mice infected with RPV alone, this virus replicates less effectively than in mice infected with the entire complex (8), and a substantial replication might be a prerequisite for maximal BPFC potentiation. In C57BL/6 mice, RPV and FLC undergo limited replication (14) and only cause moderate and transient increases of BPFC numbers. However, in the light of previous results obtained in neonatally infected BALB/c mice, in which RPV did not modify the gradual appearance with aging of splenic BPFC to SRC though replicating at high titer (4), it is felt that the differences between the effects of RPV and FLC in adults do not simply depend on the extent of viral replication. The present results emphasize that, in spite of considerable similarities between the effects of RPV and FLC infections on the antibody response to SRC (6, 9), other immunological parameters may be differently affected by the two virus preparations. Results showing a differential effect on another aspect of immunological reactivity (namely, the response to a skin sensitizer) are presented in the accompanying paper (7). Whether these discrepancies underlie basic differences in the mechanisms of RPVand FLC-induced immunodepressions is not possible to decide on the basis of presently available data. The question is important because it is directly related to the problem of the role played by RPV in the immunodepressive activity of FLC. However, at least after RPV infection, potentiation of BPFC and immunodepression do not seem to be strictly correlated phenomena since the former is more pronounced at a time when the latter begins to wane (9). ACKNOWLEDGMENTS This investigation was supported by the Italian National Research Council (grant no. CT 73.00462.04). We thank G. Falcone for helpful advice and suggestions, L. Montagnani for skillful technical assistance, and G. Greco for typing the manuscript. LITERATURE CITED 1. Aisenberg, A. C., and C. Davis. 1968. The thymus and recovery of the sheep erythrocyte response in irradiated mice. J. Exp. Med. 128:1327-1338.
1030
BENDINELLI, TONIOLO, AND CAMPA
INFECT. IMMUN.
2. Andersson, J., 0. Sjoberg, and G. Moller. 1972. Induction 18. Countinho, A., and G. Moller. 1973. B cell mitogenic of immunoglobulin and antibody synthesis in vitro by properties of thymus-independent antigens. Nature lipopolysaccharides. Eur. J. Immunol. 2:349-353. (London) 245:12-14. 3. Bainbridge, D. R., and M. Bendinelli. 1972. Circulation 19. Dawson, R. J., R. B. Tacke, and A. H. Fieldsteel. 1968. of lymphoid cells in mice infected with Friend leukemia Relationship between Friend virus and an associated virus. J. Natl. Cancer Inst. 49:773-781. lymphatic leukemia virus. Br. J. Cancer 22:569-576. 4. Bendinelli, M. 1971. Effect of Friend leukemia virus and 20. Dent, P. B. 1972. Immunodepression by oncogenic viRowson-Parr virus on immunological maturation of ruses. Prog. Med. Virol. 14:1-35. mice. Infect. Immun. 4:1-5. 21. Hege, J. S., and L. J. Cole. 1967. Antibody plaque-form5. Bendinelli, M. 1971. Immunodepression by Friend virus, ing cells in unsensitized mice. Specificity and response to neonatal thymectomy, X-irradiation and PHA. J. p. 314. In J. L. Melnick (ed.), Proc. 2nd Int. Congr. Virol. S. Karger, Basel. Immunol. 99:61-70. 6. Bendinelli, M. 1973. Immunodepression by Rowson-Parr 22. Hirano, S., and H. Friedman. 1969. Effect of murine virus, p. 181-221. In W. S. Ceglowski and H. Friedman leukemia virus on "background" antibody plaque(ed.), Virus tumorigenesis and immunogenesis. Acaforming cells to sheep erythrocytes and E. coli antigen. demic Press Inc., New York. Nature (London) 224:1316-1318. 7. Bendinelli, M., M. Campa, and A. Toniolo. Im- 23. Hirano, S., H. Friedman, and W. S. Ceglowski. 1971. munodepression by Rowson-Parr virus in mice: effect Immunosuppression by leukemia viruses. VII. Stimulaof Rowson-Parr virus and Friend leukemia complex tory effects of Friend leukemia virus on pre-existing infections on contact sensitivity in susceptible and antibody-forming cells to sheep erythrocytes and EschImmun. resistant mice. Infect. erichia coli in non-immunized mice. J. Immunol. 11:1031-1037. 8. Bendinelli, M., and L. Nardini. 1973. Immunodepression 107:1400-1409. by Rowson-Parr virus in mice. I. Growth curves of 24. Jerne, N. K., A. A. Nordin, and C. Henry. 1963. The agar Rowson-Parr virus and immunological relationships plaque technique for recognising antibody-producing with Friend virus. Infect. Immun. 7:152-159. cells, p. 109-122. In B. Amos and H. Koprowski (ed.), 9. Bendinelli, M., and L. Nardini. 1973. Immunodepression Cell-bound antibodies. Wistar Institute Press, Philaby Rowson-Parr virus in mice. II. Effect of Rowson-Parr delphia. virus infection on the antibody response to sheep red 25. Nordin, A. A. 1968. The occurrence of plaque forming cells in vivo and in vitro. Infect. Immun. 7:160-166. cells in normal and immunized conventional and 10. Bendinelli, M., L. Nardini, and M. Campa. 1974. Neugermfree mice. Proc. Soc. Exp. Biol. Med. 129:57-62. tralization of Friend leukemia virus by sera of unimmu- 26. Old, L. J., D. A. Clarke, B. Benaceraff, and M. Goldnized animals. J. Gen. Virol. 22:207-214. smith. 1960. The reticuloendothelial system and the 11. Braun, W. 1965. Influence of nucleic acid degradation neoplastic process. Ann. N.Y. Acad. Sci. 88:264-280.. products on antibody synthesis, p. 525-534. In J. Sterzl 27. Rappaport, I. 1973. A replica plating technique for the (ed.), Molecular and cellular basis of antibody formastudy of plaque-forming centers. Cell. Immunol. tion. Academic Press Inc., New York. 6:473-478. 12. Capalbo, E. E., J. F. Albright, and W. E. Bennett. 1964. 28. Rowson, K. E. K., and I. Parr. 1970. A new virus of Evaluation of the diffusion chamber culture technique minimal pathogenicity associated with Friend virus. I. for study of the morphological and functional characIsolation by end-point dilution. Int. J. Cancer teristics of lymphoid cells during antibody production. 5:96-102. 29. Salaman, M. H., and N. Wedderburn. 1966. The imJ. Immunol. 92:243-251. 13. Carter, R. L., F. C. Chesterman, K. E. K. Rowson, M. H. munodepressive effect of Friend virus. Immunol. 10:445-458. Salaman, and N. Wedderburn. 1970. A new virus of minimal pathogenicity associated with Friend virus. II. 30. Schrader, J. W. 1974. Evidence for the presence in unimmunized mice of two populations of bone marrow Histological changes and immunodepressive effect. Int. derived B lymphocytes, defined by differences in adJ. Cancer 5:103-110. 14. Ceglowski, W. S., B. P. Campbell, R. F. Mortensen, and herence properties. Cell. Immunol. 10:380-393. H. Friedman. 1974. Humoral and cellular immune 31. Steeves, R. A., R. J. Eckner, M. Bennett, E. A. Mirand, and P. J. Trudel. 1971. Isolation and characterization responses in susceptible and resistant strains of mice of a lymphatic leukemia virus in the Friend virus infected with Friend leukemia virus. Proc. Soc. Exp. Biol. Med. 146:619-624. complex. J. Natl. Cancer Inst. 46:1209-1217. 15. Ceglowski, W. S., and H. Friedman. 1967. Suppression of 32. gterzl, J., J. Vesely, M. Jilek, and L. Mandel. 1965. The the primary antibody plaque response of mice following inductive phase of antibody formation studied with isolated cells, p. 463-475. In J. §terzl (ed.), Molecular infection with Friend disease virus. Proc. Soc. Exp. Biol. Med. 126:662-666. and cellular basis of antibody formation. Academic 16. Ceglowski, W. S., and H. Friedman. 1969. ImmunosupPress Inc., New York. pression by leukemia viruses. II. Cytokinetics of ap- 33. Trentin, J. J. 1967. Intervention in discussion, p. 45-46. In J. J. Trentin (ed.), Cross-reacting antigens and pearance of hemolysin-forming cells in infected mice neoantigens. The Williams & Wilkins Co., Baltimore. during the anamnestic response to sheep erythrocytes. 34. Wigzell, H. 1966. Antibody synthesis at the cellular level: J. Immunol. 102:338-346. some studies on natural anti-sheep red cell antibodies 17. Cerny, J., and J. Halasa. 1973. Antigen binding rosette in the mouse. J. Immunol. 97:608-611. forming cells in a Friend virus-induced leukemia. Experientia 29:101-102.
INFECTION AND IMMUNITY, May 1975, p. 1024-1030 Copyright 0 1975 American Society for Microbiology
Immunodepression by Rowson-Parr Virus in Mice: Effect of Rowson-Parr Virus and Friend Leukemia Complex Infections on Background Antibody-Forming Cells to Various Erythrocytes M. BENDINELLI,* A. TONIOLO, AND M. CAMPA Institute of Microbiology, University of Pisa, 56100 Pisa, Italy Received for publication 30 December 1974
The numbers of background antibody-forming cells (BPFC) toward erythrocytes of various species present in the lymphoid organs of unimmunized susceptible BALB/c and resistant C57BL/6 mice were investigated at various times after infection with Friend leukemia complex (FLC) or Rowson-Parr virus (RPV). Both virus preparations induced an increase of BPFC numbers in both animal strains, but the rate and magnitude of the enhancements produced by RPV were much lower. The degree of potentiation varied with the specificity of the BPFC populations and was more pronounced in the spleen than in the lymph nodes and in BALB/c than in C57BL/6 mice. In the late stage of FLC infection, the numbers of splenic BPFC to some erythrocytes underwent a dramatic fall, which was not observed in RPV-infected mice. BPFC present in BALB/c splenocytes cultured in diffusion chambers implanted in the peritoneal cavity of isogeneic normal mice were not affected by viral infection of the chambers.
Recently it has been recognized that Friend leukemia virus is a complex of various entities (19, 28, 31), one of which is the Rowson-Parr virus (RPV). RPV, which can be easily isolated from Friend leukemia complex (FLC) preparations (6, 28) because of the high titers it attains in FLC-infected mice (8), induces an early depression of the antibody response in BALB/c mice (9, 13). This finding has suggested that RPV plays an important role in the immunodepressive effects of FLC preparations and consequently cooperates in the ability of FLC to induce leukemogenesis (5). Since it has not yet been possible to free FLC from RPV, the only available experimental approach to the above hypothesis is to compare the immunodepressive properties of RPV with those of the complex. In a previous paper, in which the effect of RPV and FLC infections on the responsiveness of adult BALB/c mice to sheep erythrocytes (SRC) was studied, we found that the number of natural background plaqueforming cells (BPFC) to SRC present in the spleen of nonimmunized animals is greatly increased by FLC but not by RPV (9). This aspect is further investigated here by extending the range of target erythrocytes. It is shown that although both RPV and FLC may increase the numbers of BPFC, there are considerable differences between the effects exerted by the two
virus preparations. In addition, the results are presented of experiments performed to elucidate the mechanism whereby BPFC numbers are increased after infection. MATERIALS AND METHODS Mice. Male inbred BALB/c and random-bred C57BL/6 mice grown in this Institute and aged 8 to 12 weeks were used. Viruses. The strains of RPV and FLC used were free of the lactic dehydrogenase virus and were prepared as previously described (8). Their passage histories (4) indicate that they are NB tropic. FLC titer was 103-5 mean infectious doses per 0.1 ml. RPV, obtained from newborn mice, titrated 105 ° mean infectious doses per 0.1 ml. Infection was performed by retro-orbital injection of 0.1 ml of virus preparation. Assay for BPFC. Hemolytic plaque-forming cells were assayed by the method of Jerne et al. (24). Unless otherwise specified, the cells were suspended in Eagle minimum essential medium and separately tested against the desired erythrocytes. Sheep (SRC), chicken (CRC), horse (HRC), donkey (DRC), rabbit (RRC) and human group 0 erythrocytes (HuRC) were used; they were maintained in Alsever solution and used between 1 and 5 weeks after collection. A 0.1-ml amount of a 20% suspension of washed erythrocytes in saline was used in each plate. For each antigen at least two plates were prepared. All plates were inspected for false plaques before addition of complement. The anti-mouse immunoglobulin rabbit serum (Miles Laboratories, Kankakee, Ill.) used in some
1024
VOL. 11, 1975
ANTIBODY-FORMING CELLS IN RPV AND FLC INFECTIONS
experiments was diluted 1:10 before use. For each group the geometric mean of BPFC was calculated together with its 95% confidence limits, and the statistical significance between groups was assessed by Student's t test. Membrane filter diffusion chamber cultures. Spleen cell cultures in diffusion chambers were established by the method described by Capalbo et al. (12). Once removed from the hosts, the chambers were immersed in 0.5% Pronase (Sigma, St. Louis, Mo.) in minimum essential medium for 1 h at 28 C to improve cell recovery. Pronase was diluted out by keeping the chambers in minimum essential medium for 30 min before collecting the cultured cells.
RESULTS Effect of FLC on absolute numbers of BPFC in the spleens of BALB/c mice. Groups of normal mice and of mice infected with FLC on day - 20, - 15, -10, or -5 were tested for splenic BPFC to SRC, CRC, HRC, DRC, RRC, and HuRC (Fig. 1). In the controls, BPFC averaged 341 to CRC, 84 to SRC, 68 to HRC, and 28 to DRC, and no plaques were detected when RRC or HuRC were used as target. After FLC infection there was a progressive enhancement of BPFC to CRC, with a maximum of 6.7fold at 20 days. BPFC to SRC and HRC were
1025
increased in the first 15 days (with a maximum of 4.3-fold on day 10 for the former and 3.3-fold on day 15 for the latter) and then rapidly retumed to normal values on day 20. BPFC to DRC were moderately potentiated at days 5 and 10 and thereafter declined to normal levels. BPFC numbers to RRC and HuRC remained below detectability. Effect of RPV on the absolute numbers of BPFC in the spleens of BALB/c mice. The BPFC that had been most affected by FLC infection (i.e., BPFC to SRC, CRC, and HRC) were investigated (Fig. 2). BPFC levels in normal animals were comparable with those found in the controls of the FLC experiments. In contrast, the consequences of RPV infection markedly differed from those of FLC infection. The numbers of BPFC of all specificities underwent a slow increase that was evident 10 to 15 days after infection and did not significantly decline in the subsequent days. The magnitude of the increase was generally lower than in FLC-infected mice; the BPFC that had been most affected by FLC were the least affected by RPV, and vice versa. The peak increases in BPFC number were 3.5-fold for HRC, 2.8-fold for SRC, and 1.8-fold for CRC. Splenic BPFC numbers versus spleen weight in infected BALB/c mice. During RPV infection and in the early stages of FLC infection (at least for a period of 2 weeks with the virus titers used in these experiments), as well as in normal mice, the weight of the spleen is directly proportional to the number of nucleated cells recoverable from the organ (Bendinelli, unpublished data). To estimate the behavior of BPFC as compared with that of the
z w a.
g
.ol a.
0
5 10 DAYS AFTER FLC
15
20
INFECTION
FIG. 1. Absolute numbers of BPFC to chicken (0), sheep (V), ho,rse (0), donkey (v), rabbit (-), and human (0) erythrocytes in the spleens of BALBIc mice at various times after infection with FLC. Numbers of mice are indicated near each symbol.
x
0
5
10
DAYS AFTER RPV
15 20 NFECTION
FIG. 2. Absolute numbers of BPFC to chicken (0), sheep (V), and horse (0) erythrocytes in the spleens of BALBIc mice at various times after infection with RPV. Numbers of mice are indicated near each symbol.
1026
INFECT. IMMUN.
BENDINELLI, TONIOLO, AND CAMPA
total sple en cellularity, the results of the previous two g roups of experiments are presented in Fig. 3. Dluring the first 5 to 10 days of FLC infection, the BPFC of all specificities increased at a rate aipproximating that of the entire spleen cell popuilation. After this time, although in some cas4 es the absolute BPFC numbers were still incre asing, their ratios rapidly declined. In contrast, in RPV-infected mice BPFC numbers maintain(ed at a fairly constant ratio with spleen we ight. It shouild be noticed, however, that RPV caused a much lower splenomegaly than did FLC (Fig 1 and 2). Specifi[city and immunological nature of BPFC fri Dm normal and infected spleens. The correlatio In coefficient was determined between all the po *ssible combinations of the numbers of BPFC of different specificities detected both in normal aynd infected individuals. In no case was
z
the probability of linear correlation significant. Moreover, in some experiments cell suspensions from normal and FLC- or RPV-infected spleens were assayed against a mixture of two types of erythrocytes; plaques with complete lysis were rarely observed, and their number never exceeded 1% of total plaques. These results clearly demonstrate that both in normal and infected spleens at least the great majority of BPFC directed towards different erythrocytes belong to different populations. To ascertain the immunological nature of the plaques, experiments were performed in which normal or infected spleen cell suspensions containing equal numbers of cells were plated against various erythrocytes and the plates were incubated for 1 h with anti-mouse immunoglobulin rabbit serum or with normal rabbit serum before addition of complement. The results of one of these experiments are given in Table 1 and show that BPFC (to SRC, CRC, and HRC) from normal and infected spleens were equally susceptible to partial inhibition by normal rabbit serum and to almost complete suppression by the anti-immunoglobulin serum. These data, together with the fact that
w
1.0 _
J
CLIXn
no
plaques developed
in the absence of complement, demonstrate the
immunological nature of BPFC present in
_
mal and infected
UL
Effect of FLC and RPV infections on BPFC numbers in the lymph nodes of BALB/c mice. In one experiment groups of mice were infected with FLC on day -20 or -10 and the superficial lymph nodes (axillary, inguinal, and cervical) from each individual mouse were collected, pooled, weighed, and tested for BPFC to SRC and CRC on day 0. The mesenteric lymph nodes
0
U. IL CD
FLC
R PV I O
II
I
10
II
I
I
20
DAYS AFTER
10
0
nor-
spleens.
20
INFECTION
FIG. 3. 1Numbers of BPFC to chicken (0), sheep (V), horse (0), and donkey (V) erythrocytes per milligram of spleen weight at various times after infection w 'ith RPV and FLC in BALBIc mice.
were separately processed in the same way. Ten
days postinfection both superficial and
mesen-
teric lymph nodes contained numbers of BPFC similar to the controls, whereas 20 days after
infection superficial lymph nodes still had normal numbers of BPFC, but mesenteric lymph nodes showed increased BPFC counts (Table 2,
TABLE 1. Inhibition of background plaques to SRC, CRC, and HRC in normal, RPV-infected or FLC-infected BALBIc spleens by normal rabbit serum or anti-mouse immunoglobulin rabbit serum Background plaques to:
Normal mice .............. RPV-infected mice ......... FLC-infected mice .........
NRSa
Anti-IgRSb
NRS
Anti-IgRS
NRS
Anti-IgRS
71.7c 76.5 64.4
89.6 91.5 89.1
87.2 84.4 89.7
100 97.5 98.3
77.2 75.6 74.4
84.9 96.4 95.3
Normal rabbit serum. b Anti-mouse immunoglobulin rabbit serum. c Percentage of inhibition relative to untreated plates.
a
HRC
CRC
SRC
Cell donor
1027
ANTIBODY-FORMING CELLS IN RPV AND FLC INFECTIONS
VOL. 11, 1975
experiment 1). The difference between mean numbers of BPFC to CRC in the controls and in the mice infected at day -20 was significant (P < 0.01). The mean weight of mesenteric lymph nodes was also considerably augmented by FLC infection, whereas that of superficial lymph nodes showed little, if any, change. In a second experiment (Table 2, experiment 2), the potentiation of BPFC in mesenteric lymph nodes 20 days after FLC infection was confirmed and RPV was found only to produce minor changes on mesenteric lymph node BPFC. Effect of RPV and FLC infections on BPFC in spleen cells cultured in diffusion chambers. Viable spleen cells (3 x 107) from normal BALB/c mice (containing 14 BPFC to SRC and 42 BPFC to CRC per 107 cells) were placed into 50-nm membrane filter (Millipore Corp.) diffusion chambers and mixed with normal mouse plasma or with RPV or FLC, taking care not to infect the extemal of the chambers. The sealed chambers were then implanted into the peritoneal cavity of normal isogeneic hosts. Fourteen days later the chambers were removed from the mice, and the cells were counted and assayed for BPFC to SRC and CRC. The recovery of
cells and of BPFC from the chambers was low: from six uninfected chambers 1.2 x 107 cells were obtained, and these contained 5 BPFC to SRC (4/107 cells) and 9 BPFC to CRC (7/107 cells). Infection slightly increased cell recovery but did not significantly modify the recovery of BPFC. From five RPV-infected chambers 1.5 x 107 cells were recovered, which produced 4 plaques to SRC (3/107 cells) and 4 to CRC (3/107 cells). From six FLC-infected chambers 2.0 x 107 cells were recovered, and BPFC were 7 to SRC (3/107 cells) and 9 to CRC (5/107 cells). Effect of FLC and RPV infections on BPFC in the spleens of C57BV6 mice. C57BL/6 mice are known to be resistant to Friend leukemia. After FLC infection they show a very slight and transient splenomegaly, and after RPV infection their spleen weights remain in the normal range. Groups of mice, either normal or infected with FLC or RPV on day 20 or 10, were tested for splenic BPFC to SRC, CRC, and HRC. Mice infected in either way 10 days earlier had slightly increased numbers of BPFC to all the erythrocytes tested (Table 3). The differences were significant in the FLC-infected group for BPFC to SRC (P < 0.05) and in the RPVinfected group for BPFC to CRC (P < 0.01). By -
-
TABLE 2. Effect of RPV and FLC infection on the number of BPFC to SRC and CRC present in the lymph nodes of BALB/c mice Superficial lymph nodes Expt no.
a
Pretreatment of mice
No. miceof
Mean wt (mg)
32 39 49
1
None FLC at day - 10 FLC at day -20
7 8 10
2
None RPV at day -20 FlC at day -20
10 8 8
BPFC to SRC
BPFC to CR(
14a (5-35) 12 (7-19) 12 (6-24) 10 (5-21) 16 (7-40) 11 (5-29)
Mesenteric lymph nodes BPFC BPFC Mean wt (mg) to SRC to CRC
69 93 136
19 (6-60) 15 (8-29) 12 (8-18) 11 (6-17) 39 (21-75) 83 (34-201)
69 62 116
5 (3-10) 21 (8-58) 10 (3-36) 32 (8-133) 14 (5-36) 45 (25-87)
Geometric mean (in parentheses, the 95% confidence limits).
TABLE 3. Effect of RPV and FLC infection on the number of BPFC to SRC, CRC, and HRC present in the spleens of C57 BL/6 mice Pretreatment of mice
Spleen wt (mg)
None
129 (88-163)b
RPV at day -10 RPV at day -20 FLC at day -10 FLC at day -20
128(97-153) 150(102-156) 184(121-269) 175(149-221)
aIn parentheses, numbers of mice tested. b Mean i range.
BPFC to SRC 95%
BPFC to CRC
BPFC to HRC
95% 95% Geometric Confidence Geometric Confidence Geometric Confidence limits mean mean mean limits limits
11 (19)a 23(6) 8 (8) 40(15) 13(8)
5-24 2-261 2-30 18-90 3-54
27 (19) 141 (6) 55(8)
60(15) 74(8)
13-53 46-662 14-220 36-102 32-169
4 (8) 21(6) 6 (8) 15 (4) 11 (8)
0-15 3-164 1-27 8-28 3-42
1028
BENDINELLI, TONIOLO, AND CAMPA
INFECT. IMMUN.
The effects of RPV infection on BPFC in the day 20 postinfection BPFC were again at normal levels except those directed to CRC in the spleens of BALB/c mice are substantially differFLC-infected group, which were still, though ent from those exerted by FLC. Even RPV augments the numbers of BPFC, but not as not significantly, potentiated. rapidly and markedly as the entire complex. DISCUSSION Significant enhancements are first detectable Background plaques are known to result from 10 to 15 days after infection, and in the subsethe activity of single cells releasing 19S anti- quent 5 to 10 days there are few further modifibody (21), but the origin and significance of cations. The kinetics of increase of different these cells are still uncertain. Being absent or BPFC is not completely asynchronous as after very few in young animals (4, 21), they are FLC infection; in fact, the extent of BPFC thought to arise as a consequence of continuous potentiation parallels the slight spleen enlargestimulation by antigens, such as microorga- ment caused by RPV. Moreover, the various nisms, which share determinants with erythro- BPFC are affected differently by the two virus cytes, but definite proof is still lacking (25, 32). preparations: the highest increase is achieved Because the formation of BPFC is thymus by BPFC to CRC in the FLC-infected spleens independent, and for other reasons (1, 21), it is and by BPFC to HRC in the RPV-infected generally agreed that they are not precursors spleens. These findings explain how in earlier of-nor are in any way connected with-the experiments performed with SRC alone the antibody-forming cells elicited by inoculation potentiating activity of RPV on BPFC escaped with the corresponding antigen (21). detection (9). Also, the counts of BPFC in Several treatments have been shown to in- RPV-infected mesenteric lymph nodes were crease the number of BPFC to SRC in the little changed. mouse spleen. These include administration of The increase in BPFC numbers produced by phytohemagglutinin, bacterial lipopolysaccha- FLC and RPV infections may have several rides, and unrelated antigens (11, 21). The explanations. One possibility is that BPFC ability of FLC to enhance the number of BPFC production is stimulated. This might occur to SRC in adult mice was originally described because host defenses are lowered so that invadby Hirano et al. (22, 23) and confirmed by us (9) ing microorganisms stimulate BPFC formation and others (17). In the present study FLC and through the activity of cell wall lipopolysacchaRPV are shown to affect all the BPFC studied rides (11, 25, 33). This explanation, however, that are normally present in consistent numbers contrasts with the augmentation of the reticuloin the spleens of BALB/c mice, i.e., those endothelial system activity (26) and the normal directed against SRC, CRC, HRC, and DRC. levels of complement (10) of FLC-infected mice. The modifications observed during FLC in- Alternatively, FLC and RPV might increase fection on these BPFC can be divided into three BPFC production acting as mitogens. Mitogens, distinct phases. Initially, their absolute num- as well as many thymus-independent antigens bers increase and the ratios of BPFC to spleen capable of unspecific stimulation of lymphocyte cellularity remain unchanged. In the second division, may cause a burst of policlonal antiphase, the numbers per spleen remain steady or body synthesis in vivo (11) and in vitro (2, 18, still increase, but the ratios of BPFC to cellular- 30). A variant of the last explanation is that ity decrease. This is usually followed by a third immunologically committed cells are preferenphase, during which the absolute BPFC num- tially stimulated. It has been observed that bers increase and the ratios of BPFC to spleen antibody-forming cell numbers may increase in phases are not synchronous for BPFC of differ- the spleens of mice primed several weeks before ent specificities. For instance, 20 days after and then infected with FLC (16). Both these infection the absolute numbers of BPFC to CRC explanations, if maintained in the case of RPV, are still increasing, whereas those of BPFC are difficult to reconcile with the differential directed against SRC, HRC, and DRC have activity of FLC on the BPFC of various specalready markedly declined. Moreover, there is a ificities. As a further alternative, BPFC producremarkable lack of homogeneity in the magni- tion might be increased due to a transient lowertude of the enhancement of different BPFC. A ing of the threshold of immunocytes to antigenic similar divergent behavior after FLC infection triggering. In this case BPFC potentiation would has been observed between BPFC to SRC and also depend on the availability of cross-reacting to Escherichia coli (23) and is in keeping with antigens, explaining why BPFC that are absent the concept that different BPFC belong to or very few in the normal spleen (such as those different cell populations, as shown by the to RRC and HuRC) are not induced by infecabsence of linear correlation between various tion. This mode of action, however, sharply contrasts with the depression of immunological BPFC in this as well as other studies (21).
VOL. 11, 1975
ANTIBODY-FORMING CELLS IN RPV AND FLC INFECTIONS
reactivity that characterizes FLC and RPV infections (15, 20, 29). Cross-reactivity between the viral antigens and the antigens used as targets to detect BPFC is ruled out by previous observations (23) and by the wide diversity of erythrocytes with which cross-reaction should occur to justify the present data. It is also possible that BPFC in infected mice are formed at a normal rate and that the increase in the spleen is due to modifications of their behavior. For instance, their decay might be slower in infected than in normal animals. Nothing is known about the life span of BPFC, and preliminary findings showing that normal and infected BPFC have similar survival times in vitro do not rule out this possibility. Also, since there is evidence that the secretion of antibody by BPFC is a pulsating phenomenon (27), FLC and RPV might stimulate preexisting low-producer antibody-forming cells to release quantities of antibody large enough to permit detection. This explanation, however, does not account for the lack of parallelism in the potentiation of different BPFC by FLC infection and contrasts with the finding that BPFC of infected spleens are inhibited by anti-immunoglobulin serum as effectively as those of normal mice. Alternatively, BPFC produced elsewhere (34) might accumulate in the spleen as a result of an alteration in their circulation. Lymph nodes are less susceptible to the immunodepressive effect of FLC and RPV (6) and might therefore continue to function as a source of BPFC. In FLC-infected mice the spleen accumulates lymphoid cells, whereas superficial lymph nodes show neither gross defects in lymphocyte circulation (3) nor change in BPFC numbers. The lack of synchrony and homogeneity between the effects of FLC on various BPFC might be due to differences in their rate of production and circulation. This hypothesis is in keeping with the failure of FLC and RPV to affect BPFC numbers in diffusion chambers and with the previous observation that FLC suppresses the appearance of BPFC in the spleens of immunologically immature mice (4), but attempts to test such a possibility directly have been hampered by failure to cause a lasting abatement of splenic BPFC by irradiation or other means (1, 21; unpublished data). No gross modifications of lymphoid cell circulation have, however, been observed in RPVinfected spleens (3). The decrease of BPFC-to-spleen cellularity ratios that occurs in the late stages of FLC infection but not in RPV infection is most likely due to dilution of BPFC by the transformed cells that proliferate quickly in the spleens of FLC-infected mice. However, a reduction of
1029
BPFC production, arrival, and/or survival can also contribute, as shown by the frequent decline in the absolute numbers of BPFC at this time. The other discrepancies observed in the effects of the two virus preparations cannot be accounted for on the basis of the virus titers used for infection, since the batch of RPV used was obtained from neonatally infected mice in which viral titers approached those present in FLC preparations (8). However, in adult mice infected with RPV alone, this virus replicates less effectively than in mice infected with the entire complex (8), and a substantial replication might be a prerequisite for maximal BPFC potentiation. In C57BL/6 mice, RPV and FLC undergo limited replication (14) and only cause moderate and transient increases of BPFC numbers. However, in the light of previous results obtained in neonatally infected BALB/c mice, in which RPV did not modify the gradual appearance with aging of splenic BPFC to SRC though replicating at high titer (4), it is felt that the differences between the effects of RPV and FLC in adults do not simply depend on the extent of viral replication. The present results emphasize that, in spite of considerable similarities between the effects of RPV and FLC infections on the antibody response to SRC (6, 9), other immunological parameters may be differently affected by the two virus preparations. Results showing a differential effect on another aspect of immunological reactivity (namely, the response to a skin sensitizer) are presented in the accompanying paper (7). Whether these discrepancies underlie basic differences in the mechanisms of RPVand FLC-induced immunodepressions is not possible to decide on the basis of presently available data. The question is important because it is directly related to the problem of the role played by RPV in the immunodepressive activity of FLC. However, at least after RPV infection, potentiation of BPFC and immunodepression do not seem to be strictly correlated phenomena since the former is more pronounced at a time when the latter begins to wane (9). ACKNOWLEDGMENTS This investigation was supported by the Italian National Research Council (grant no. CT 73.00462.04). We thank G. Falcone for helpful advice and suggestions, L. Montagnani for skillful technical assistance, and G. Greco for typing the manuscript. LITERATURE CITED 1. Aisenberg, A. C., and C. Davis. 1968. The thymus and recovery of the sheep erythrocyte response in irradiated mice. J. Exp. Med. 128:1327-1338.
1030
BENDINELLI, TONIOLO, AND CAMPA
INFECT. IMMUN.
2. Andersson, J., 0. Sjoberg, and G. Moller. 1972. Induction 18. Countinho, A., and G. Moller. 1973. B cell mitogenic of immunoglobulin and antibody synthesis in vitro by properties of thymus-independent antigens. Nature lipopolysaccharides. Eur. J. Immunol. 2:349-353. (London) 245:12-14. 3. Bainbridge, D. R., and M. Bendinelli. 1972. Circulation 19. Dawson, R. J., R. B. Tacke, and A. H. Fieldsteel. 1968. of lymphoid cells in mice infected with Friend leukemia Relationship between Friend virus and an associated virus. J. Natl. Cancer Inst. 49:773-781. lymphatic leukemia virus. Br. J. Cancer 22:569-576. 4. Bendinelli, M. 1971. Effect of Friend leukemia virus and 20. Dent, P. B. 1972. Immunodepression by oncogenic viRowson-Parr virus on immunological maturation of ruses. Prog. Med. Virol. 14:1-35. mice. Infect. Immun. 4:1-5. 21. Hege, J. S., and L. J. Cole. 1967. Antibody plaque-form5. Bendinelli, M. 1971. Immunodepression by Friend virus, ing cells in unsensitized mice. Specificity and response to neonatal thymectomy, X-irradiation and PHA. J. p. 314. In J. L. Melnick (ed.), Proc. 2nd Int. Congr. Virol. S. Karger, Basel. Immunol. 99:61-70. 6. Bendinelli, M. 1973. Immunodepression by Rowson-Parr 22. Hirano, S., and H. Friedman. 1969. Effect of murine virus, p. 181-221. In W. S. Ceglowski and H. Friedman leukemia virus on "background" antibody plaque(ed.), Virus tumorigenesis and immunogenesis. Acaforming cells to sheep erythrocytes and E. coli antigen. demic Press Inc., New York. Nature (London) 224:1316-1318. 7. Bendinelli, M., M. Campa, and A. Toniolo. Im- 23. Hirano, S., H. Friedman, and W. S. Ceglowski. 1971. munodepression by Rowson-Parr virus in mice: effect Immunosuppression by leukemia viruses. VII. Stimulaof Rowson-Parr virus and Friend leukemia complex tory effects of Friend leukemia virus on pre-existing infections on contact sensitivity in susceptible and antibody-forming cells to sheep erythrocytes and EschImmun. resistant mice. Infect. erichia coli in non-immunized mice. J. Immunol. 11:1031-1037. 8. Bendinelli, M., and L. Nardini. 1973. Immunodepression 107:1400-1409. by Rowson-Parr virus in mice. I. Growth curves of 24. Jerne, N. K., A. A. Nordin, and C. Henry. 1963. The agar Rowson-Parr virus and immunological relationships plaque technique for recognising antibody-producing with Friend virus. Infect. Immun. 7:152-159. cells, p. 109-122. In B. Amos and H. Koprowski (ed.), 9. Bendinelli, M., and L. Nardini. 1973. Immunodepression Cell-bound antibodies. Wistar Institute Press, Philaby Rowson-Parr virus in mice. II. Effect of Rowson-Parr delphia. virus infection on the antibody response to sheep red 25. Nordin, A. A. 1968. The occurrence of plaque forming cells in vivo and in vitro. Infect. Immun. 7:160-166. cells in normal and immunized conventional and 10. Bendinelli, M., L. Nardini, and M. Campa. 1974. Neugermfree mice. Proc. Soc. Exp. Biol. Med. 129:57-62. tralization of Friend leukemia virus by sera of unimmu- 26. Old, L. J., D. A. Clarke, B. Benaceraff, and M. Goldnized animals. J. Gen. Virol. 22:207-214. smith. 1960. The reticuloendothelial system and the 11. Braun, W. 1965. Influence of nucleic acid degradation neoplastic process. Ann. N.Y. Acad. Sci. 88:264-280.. products on antibody synthesis, p. 525-534. In J. Sterzl 27. Rappaport, I. 1973. A replica plating technique for the (ed.), Molecular and cellular basis of antibody formastudy of plaque-forming centers. Cell. Immunol. tion. Academic Press Inc., New York. 6:473-478. 12. Capalbo, E. E., J. F. Albright, and W. E. Bennett. 1964. 28. Rowson, K. E. K., and I. Parr. 1970. A new virus of Evaluation of the diffusion chamber culture technique minimal pathogenicity associated with Friend virus. I. for study of the morphological and functional characIsolation by end-point dilution. Int. J. Cancer teristics of lymphoid cells during antibody production. 5:96-102. 29. Salaman, M. H., and N. Wedderburn. 1966. The imJ. Immunol. 92:243-251. 13. Carter, R. L., F. C. Chesterman, K. E. K. Rowson, M. H. munodepressive effect of Friend virus. Immunol. 10:445-458. Salaman, and N. Wedderburn. 1970. A new virus of minimal pathogenicity associated with Friend virus. II. 30. Schrader, J. W. 1974. Evidence for the presence in unimmunized mice of two populations of bone marrow Histological changes and immunodepressive effect. Int. derived B lymphocytes, defined by differences in adJ. Cancer 5:103-110. 14. Ceglowski, W. S., B. P. Campbell, R. F. Mortensen, and herence properties. Cell. Immunol. 10:380-393. H. Friedman. 1974. Humoral and cellular immune 31. Steeves, R. A., R. J. Eckner, M. Bennett, E. A. Mirand, and P. J. Trudel. 1971. Isolation and characterization responses in susceptible and resistant strains of mice of a lymphatic leukemia virus in the Friend virus infected with Friend leukemia virus. Proc. Soc. Exp. Biol. Med. 146:619-624. complex. J. Natl. Cancer Inst. 46:1209-1217. 15. Ceglowski, W. S., and H. Friedman. 1967. Suppression of 32. gterzl, J., J. Vesely, M. Jilek, and L. Mandel. 1965. The the primary antibody plaque response of mice following inductive phase of antibody formation studied with isolated cells, p. 463-475. In J. §terzl (ed.), Molecular infection with Friend disease virus. Proc. Soc. Exp. Biol. Med. 126:662-666. and cellular basis of antibody formation. Academic 16. Ceglowski, W. S., and H. Friedman. 1969. ImmunosupPress Inc., New York. pression by leukemia viruses. II. Cytokinetics of ap- 33. Trentin, J. J. 1967. Intervention in discussion, p. 45-46. In J. J. Trentin (ed.), Cross-reacting antigens and pearance of hemolysin-forming cells in infected mice neoantigens. The Williams & Wilkins Co., Baltimore. during the anamnestic response to sheep erythrocytes. 34. Wigzell, H. 1966. Antibody synthesis at the cellular level: J. Immunol. 102:338-346. some studies on natural anti-sheep red cell antibodies 17. Cerny, J., and J. Halasa. 1973. Antigen binding rosette in the mouse. J. Immunol. 97:608-611. forming cells in a Friend virus-induced leukemia. Experientia 29:101-102.
