216659a0Nature2165116196711186596630028-0836196710.1038/216659a0ukNatureNatureNATUREnatureNature is a weekly international journal publishing the finest peer-reviewed research in all fields of science and technology on the basis of its originality, importance, interdisciplinary interest, timeliness, accessibility, elegance and surprising conclusions. Nature also provides rapid, authoritative, insightful and arresting news and interpretation of topical and coming trends affecting science, scientists and the wider public./nature/journal/v216/n5116issueJournal homeArchiveCurrent issueAdvance online publicationPrivacy policySubscribeNature Publishing GroupCurrent issue216659a0The Thymus and the Precursors of Antigen Reactive Cells
AU  - MILLER, J. F. A. P.
AU  - MITCHELL, G. F.The Walter and Eliza Hall Institute of Medical Research, Melbourne, VictoriaThe thymus is important in providing an adequate pool of immunologically competent cells. Thymectomy does not reduce the number of precursor cells, but removes the influence necessary for their differentiation into antigen reactive cells.THE bone marrow is a source of primitive stem cells which are characterized by their capacity for extensive proliferation, self renewal and differentiation into a variety of more mature cells. The colony forming unit is a stem cell which gives rise, in irradiated hosts, to large clones of erythroid and myeloid calls1. In both irradiated and non-irradiated animals there is a steady stream of cells from the bone marrow to the thymus and to secondary lymphoid tissues, such as the lymph nodes and spleen. There are thus, in the bone marrow, cell types capable of repopulating the lymphoid compartment of the thymus and the pool of immunologically competent cells2-4. It is not clear whether there are separate stem cells for the myeloid tissues, the thymus and the immuno-competent cell pool, although recent work suggests that the colony forming unit may also function as a lymphoid stem cell5.
The thymus plays an important part in building up an adequate pool of immunologically competent cells. When thymectomy is performed before this pool has been built up, for example, at birth, the number of cells in it is markedly less than normal and an impairment of those immunological reactions known to be mediated by these cells is immediately evident6'7. When thymectomy is performed in adult life, or after an adequate pool has been built up, no immunological defects become evident until months later8-10, presumably after the pool has been depleted as a consequence of the limited life span of some of its cells and the immunological commitment of others. If, however, mice which have been thymectomized in adult life are subjected to total body irradiation, the anatomical and functional regeneration of the immune system is impaired11-13. The exact relationship between the thymus, the pool of immuno-competent cells and precursors of them is unknown. Very few lymphocytes within the thymus can be detected as antigen-reactive cells14 and very few cells can be shown to have emigrated from the intact thymus15. It is unlikely that the thymus exerts an effect on the population of immuno-competent cells once they have been produced, for antigen-reactive cells respond to antigen by differentiation and proliferation to antibody-producing cells as efficiently in the normal as in the thymectomized animal6-7. Neonatal thymectomy thus does not inhibit the response of antigen-reactive cells to antigen and must therefore impair the production of antigen-reactive cells from some more immature, antigen-independent, precursor cells.
We report here the results of a time course study on the appearance of antigen-reactive cells in normal and thymectomized irradiated mice injected with bone marrow cells, thymus cells or both. Antigen-reactive cells are detected by their ability, when injected into heavily irradiated mice together with sheep erythrocytes14, to produce discrete clusters of haemolysin-producing cells in the spleen14. In this system, antigen-reactive cells are assumed to be the immediate precursors of antibody-producing cells and to belong to the same cell lineage. They are defined as cells which react to antigen, not by the production of detectable amounts of antibody, but by undergoing proliferation and differentiation to give rise to a line of antibody-producing cells.
Precursors of antigen-reactive cells are detected by incubating, in irradiated hosts and for varying periods of time, cell populations which lack antigen-reactive cells (such as bone marrow or thymus) in the absence of the relevant antigen. At the end of the incubation period, the spleen and thoracic duct lymphocyte pool of the irradiated recipients are assayed for their content of antigen-reactive cells. Mice of the highly inbred CBA strain were used in all experiments. Thymectomy of neonatal or adult mice was performed as described earlier16. Bone marrow cells were obtained from the femurs and tibiae of 2 month old normal mice and 2 month old clinically healthy mice thymectomized at birth, and were injected intravenously into 2-3 month old recipient mice of the same strain. Neither thymus cells nor sheep erythrocytes were injected into the irradiated hosts in the first experiment reported here. Some of the recipient mice had been thymectomized when 6 weeks old and all had been subjected to 900 rads of total body irradiation 4 h before marrow injection. They were killed at weekly intervals and the number of cells in their spleens capable of reacting to sheep erythrocytes was determined by the haemolytic focus assay technique of Kennedy et al.1[ast].
Fig. 1. Number of antigen-reactive cells detected at weekly intervals in the spleens of heavily irradiated CBA mice injected with bone marrow cells: O, mice injected with 106 to 107 bone marrow cells from normal donors; A, mice injected with 105 bone marrow cells from normal donors; A, mice injected with 103 bone marrow cells from normal donors; Q, mice injected with 106 to 107 bone marrow cells from neonatally thymectomized donors; o, thymectomized irradiated mice injected with 10fl to 107 bone marrow cells from normal donors. Each point represents the average value of five to forty determinations.
As Fig. 1 shows, bone marrow from either normal or neonatally thymectomized donors contains precursors which, after a period of incubation of between 1 and 2 weeks in irradiated hosts, have developed into sheep erythrocyte-reactive cells. After between 3 and 4 weeks, the numbers of such cells produced in the spleens of the rion-thymectomized irradiated mice inoculated with 106 or 107 bone marrow cells have reached the levels found in normal mice, that is, 1,000-2,000 (refs, 6 and 7). In contrast, the number of antigen-reactive cells detected in the spleens of adult thymectomized irradiated hosts never exceeded 200, even as late as 10 weeks after irradiation. The irradiated mice injected with marrow had their thoracic duct cannulated 2-9 weeks after irradiation using a method modified from that of Boak and Woodruff17, and the number of circulating lymphocytes and antigen-reactive cells was determined as described previously4. Figure 2 shows that the number of cells recoverable in 48 h in sham-thymectomized irradiated mice increased from less than 10 million 2 weeks after irradiation to 40 million 7 weeks later. By contrast, no significant increase in the size of the 48 h pool was recorded from 2 to 9 weeks after irradiation and marrow protection in adult thymectomized mice. The total number of antigen-reactive cells in the 48 h mobilizable pool increased between 4 and 9 weeks after irradiation from 120 to 1,000 in the sham-thymectomized irradiated controls but only from 15 to 66 in the thymectomized irradiated mice (Table 1). The number of antigen-reactive cells per million thoracic duct lymphocytes was virtually the same in mice of both groups 4 weeks after irradiation and then increased significantly only in the controls. By 9 weeks after irradiation the thymectomized irradiated mice, in comparison with controls, showed not only an absolute deficiency of antigen-reactive cells but also a reduced proportion of such cells per million lymphocytes. A similar deficiency has been reported before in neonatally thymectomized mice4.
All these data clearly indicate that the precursor cells, which are antigen-independent and derived from marrow, differentiate and proliferate into antigen-reactive cells largely under the influence of the thymus. Precursor cells of marrow origin cannot, on their own and in the absence of the host thymus, differentiate into antigen-reactive cells. Furthermore, it is evident that the neo-natally thymectomized mouse does not lack precursor cells but simply lacks the influence necessary to drive these cells along the path leading to the production of antigen-reactive cells. Two questions are immediately evident: what is the nature of this thymus influence and why is there a lag period of about one week before antigen -reactive cells become detectable ?
Table 1. NUMBER OF SHEEP-ERYTHROCYTE-ANTIGEN-REACTIVE CELLS IN
THE 48 h MOBILE LYMPHOCYTE POOL OF SHAM-THYMECTOMIZED AND THYMECTOMIZED, MARROW PROTECTED, IRRADIATED MICE
No. cf antigen- No. of antigen-reactive cells/ reactive cells/ million thoracic 48 h mobilizable duct cells pool
5 120
25 1,000
3 15
6 66
Group
Weeks after irradiation
Sham-thymectomized
irradiated Thymectomized
irradiated
The simplest relationship between the thymus and the pool of antigen-reactive cells would be one in which lym-phoid precursor cells of marrow origin are transformed in the thymus to lymphocytes, some of which migrate out and mature further to become antigen-reactive cells. If this is true, the one week lag period already noted could be explained by the failure of bone marrow cells to re-populate the thymus of irradiated hosts during the first week after irradiation18. Furthermore, antigen-reactive cells ought to be detected in populations of thymus lymphocytes incubated for prolonged periods of time in irradiated hosts. Accordingly, experiments were set up to check these possibilities. One hundred million marrow cells or 100 million thymus cells were given by slow intravenous injection to two sets of irradiated mice, and spleen cells from these were transferred after one week into two further sets of irradiated mice. Cell suspensions from the spleens were passaged serially at weekly intervals into further irradiated recipients, one spleen equivalent being passaged each time. The controls received no initial inoculum and only irradiated spleen was serially transferred. At each passage, aliquots of the cells pooled from the spleens of one set of mice were assayed for their content of antigen-reactive cells by the haemolytic focus assay method14. There was no histological evidence of thymus lymphocyte repopulation in irradiated mice killed at weekly intervals in this experiment. It can be seen from Fig. 3 that incubation of the original thymus cell suspension in successive generations of irradiated hosts for up to 3 weeks yielded no more than fifty antigen-reactive cells. Similarly, incubation of the original bone marrow cell suspension for a total period of 3 weeks gave values not significantly different from those obtained when only irradiated spleen was passaged. This last Fesult is in contrast to that obtained when the bone marrow inoculum was incubated for 3 weeks in a single irradiated host. It can be argued that, because the original suspension is diluted at each transfer, the number of marrow cells incubated for the entire period of time must be less than 108. But, even if a ten-fold dilution at each passage is assumed, so that only the equivalent of 105 cells was incubated for 3 weeks, at least 350 antigen-reactive cells would be expected in each spleen in the final host (Fig. 1).
Fig. 2. Total number of lymphocytes drained in 48 h from the thoracic duct of adult CBA mice cannulated at intervals from 2 to 9 weeks after sham-thymeetomy (O) or tliymectomy (o) at 2 months of age, 900 rads and injection of 107 syngeneic marrow cells. Each point represents the average of determinations made on three to four mice.
Fig. 3. Number of antigen-reactive cells detected in the spleens of heavily irradiated mice after the injection of bone marrow or thymus cells. O, Number detected in mice receiving one initial inoculum of 105 bone marrow cells from normal adult donors (from Fig. 1); o, in mice incubating for only 1 week spleen cells transferred at weekly intervals from successive sets of irradiated mice initially inoculated with 108 bone marrow cells; A"in mice incubating for only 1 week spleen cells transferred at weekly intervals from successive sets of irradiated mice initially inoculated with 108 thymus cells; A, in mice incubating for only 1 week spleen cells transferred at weekly intervals from successive sets of irradiated mice not initially inoculated. Each point represents the average of five to ten determinations.
These results, taken at their face value, suggest that the thymus lymphocyte population lacks cells capable, on their own, of giving rise to cells producing haemolysin, even after 3 weeks in successive generations of irradiated hosts. This does not support the hypothesis that bone marrow precursors become antigen-reactive cells only after differentiating within the thymus. The failure of bone marrow precursors to become antigen-reactive cells in the absence of thymus repopulation (Fig. 3) does, however, suggest the possibility that thymus lymphocytes might be essential for the rapid differentiation of precursors derived from marrow into haemolysin-producing cells. Experiments were therefore set up in which either 108 thymus cells or 108 bone marrow cells were incubated with or without sheep erythrocytes for one week in heavily irradiated hosts; cells from the spleens of these mice were transferred to a second group of irradiated mice together with either sheep erythrocytes only or with 107 bone marrow cells and sheep erythrocytes. The capacity of the second host to produce haemolysin-plaque-forming cells was assayed 4, 6 and 8 days after transfer according to the technique of Jerne19. Figure 4 shows the results obtained when spleens were transferred from thymus-incubating donors and Fig. 5 shows results obtained when spleens from bone marrow incubators or uninoculated donors were transferred. It is evident that a significant plaque forming cell response occurred within 6 days in mice which received bone marrow cells, sheep erythrocytes and spleen transferred from those irradiated donors inoculated with thymus cells and sheep erythrocytes one week before. There was no significant response if thymus cells were incubated without sheep erythrocytes in the first irradiated host, and a similar result was obtained if bone marrow cells were not given to the second irradiated host. We do not think that the response in the latter host can be ascribed to the transfer of lymphocytes of the type found in the circulating pool contaminating the thymus cell population and stimulated by the antigen given to the first host. If this had been so, one could have expected, in the second host, a response which was not dependent on the simultaneous presence of bone marrow cells.
Fig. 4. Production of haemolysin-plaque-fonning cells in the spleens of heavily irradiated mice (second hosts) injected with cells (sheep erythrocytes (SRBC) and/or bone marrow cells (BM)) and spleen transferred from a first irradiated host injected with thymus cells (T) and/or SRBC.
Group Cells injected into first host
10" T + 108SRBC 108 T + 108 SRBC
10"T
Cells injected into second host
Spleen from first host
Spleen from first host
Spleen from first host
+ 107BM +108 SRBC + 10" SRBC
+ 107BM +108 SRBC
Each point represents the average of three to twelve determinations.
Fig. 5. Production of haemolysin-plaque-forming cells in the spleens of heavily irradiated mice (second hosts) injected with cells (sheep erythrocytes (SRBC) and/or bone marrow cells (BM)) and spleen transferred from a first irradiated host injected with BM and/or SRBC.
Group
Cells injected into first host
Cells injected into second host
108 BM + 108 SRBC Spleen from +107 BM
first host +10" SRBC 108 BM + 108 SRBC Spleen from +108 SRBC
first host 108 SRBC Spleen from +107 BM
first host + 108SRBC 108 SRBC Spleen from +108 SRBC
first host
O
D
Each point represents the average value of four determinations. The graph is on the same scale as that in Fig. 4.
Our results corroborate those obtained by Clamari et a/.20, who showed that suspensions containing a mixture of adult marrow and thymus cells were far more active in producing haemolysins against sheep erythrocytes when transferred to irradiated syngeneic recipients than could be accounted for by summating the activities of each cell population alone. They also show that the thymus lymphocyte population has had to react with antigen in some wa}^ before interaction with bone marrow cells could be expected to result in significant haemolysin production. The nature of the reaction between thymus cells and antigen is obscure, but it is presumably the same as that which has been observed when cells derived from a thymus graft responded vigorously to antigenic stimulation by mitosis21'22, a response which did not, by itself, lead to antibody production23.
All the experimental evidence thus indicates that there is some sort of interaction between thymus cells, bone marrow cells and antigen. Two alternative interpretations of the results will be offered here. One possibility is that antigen-reactive cells are derived from the thymus but fail to react to sheep erythrocytes by producing plaque forming cells in heavily irradiated hosts, because of the destruction or disruption of an essential mechanism for trapping antigens. It has been shown that the splenic follicles and marginal zone cells play a pait in trapping antigens24, and that they are destroyed by irradiation25 and rapidly reconstituted by cells derived from bone marrow18. In these experiments, cells reactive to sheep erythrocytes could be derived directly from some of the thymus lymphocytes, the bone marrow population providing cells essential for the repair of the antigen trapping apparatus.
Our failure to detect precursors of antigen-reactive cells in populations of thymus lymphocytes would suggest an alternative interpretation; that the precursors of the haemolysin-producing cells are derived from marrow. The thymus might then provide cells necessary for trapping antigen or, on the other hand, thymus cells may be essential for the rapid differentiation of marrow precursor cells into antigen-reactive cells. It is difficult, however, to envisage the mechanism of action of these thymus cells.
In summary, sheep erythrocyte-antigen-reactive cells are the progeny of antigen-independent precursor cells, the differentiation of which is dependent on the thymus. The thymectomized animal does not lack precursor cells but lacks the influence necessary for their differentiation into antigen-reactive cells. Such reactive cells have not been detected after the incubation of thymus lymphocytes for up to 3 weeks in irradiated hosts. When, however, marrow cells were introduced, haemolytie, plaque forming cells appeared within a week in response to a challenge of sheep erythrocytes. One interpretation of these results is that the thymus does indeed provide antigen-reactive cells but that these cannot be detected in irradiated hosts unless marrow cells are made available presumably to repair an essential antigen-trapping apparatus disrupted by the irradiation. Another interpretation is that thymus lymphocytes lack antigen-reactive cells or their precursors but act to provide a stimulus to ensure the rapid differentiation of precursor cells derived from marrow.
We thank Sue Hughes and Winnie House for excellent technical assistance. This work was supported by grants from the National Health and Medical Research Council, the Australian Research Grants Committee, the Damon Runyon Memorial Fund for Cancer Research, Inc., the Jane Coffin Childs Memorial Fund for Medical Research and the Anna Fuller Fund.
Note added in proof (November 6, 1967). Since this paper was submitted, we have demonstrated that the precursors of the haemolysin forming cells are derived not from either the thymus or thoracic duct lymphocytes but from bone marrow26.Till, , J. E., McCulloch, , E. A., and Siminovitch, , L., Proc. US Nat. Acad. Sci., 51, 29 (1964).ChemPortMicklem, , H. S., Ford, , C. E., Evans, , E. P., and Gray, , J., Proc. Roy. Soc., B. 165, 78 (1966).ISIChemPortFord, , C. E., Ciba Foundation Symp., Thymus: Experimental and Clinical Studies (edit. by Wolstenholme, G. E. W., and Porter, R.), 131 (Churchil, London, 1966).Goldschneider, , I., and McGregor, , D. D., Nature, 212, 1433 (1966).ArticleISITill, , J. E., McCulloch, , E. A., Phillips, , R. A., and Siminovitch, , L., Cold Spring Harbor Symp. Quant. Biol., 23 (1968).Miller, , J. F. A. P., Mitchell, , G. F., and Weiss, , N. S., Nature, 214, 992 (1967).ArticlePubMedISIChemPortMiller, , J. F. A. P., and Mitchell, , G. F., Proc. First Intern. Cong. Transpl. Soc. (edit. by Dausset, J.) (Munksgaard, Copenhagen, 1967).Miller, , J. F. A. P., Nature, 208, 1337 (1965).ArticlePubMedISIChemPortMetcalf, , D., Nature, 208, 1336 (1965).ArticlePubMedISIChemPortTaylor, , R. B., Nature, 208, 1334 (1965).ArticlePubMedISIChemPortMiller, , J. F. A. P., Nature, 195, 1318 (1962).ArticleISIMiller, , J. F. A. P., Leuchars, , E., Cross, , A. M., and Dukor, , P., Ann. NY Acad. Sci., 120, 205 (1964).PubMedChemPortCross, , A. M., Leuchars, , E., and Miller, , J. F. A. P., J. Exp. Med., 119, 837 (1964).ArticlePubMedISIChemPortKennedy, , J. C., Siminovitch, , L., Till, , J. E., and McCulloch, , E. A., Pro. Soc. Exp. Biol. and Med., 120, 868 (1965).ISIChemPortNossal, , G. J. V., Ann. NY Acad. Sci., 120, 171 (1964).PubMedChemPortMiller, , J. F. A. P., Brit. J. Cancer, 14, 93 (1960).PubMedISIChemPortBoak, , J. L., and Woodruff, , M. F. A., Nature, 205, 396 (1965).ArticlePubMedISIChemPortBalner, , H., and Dersjant, , H., J. Nat. Cancer Inst., 36, 513 (1966).ISIChemPortJerne, , N. K., Nordin, , A. A., and Henry, , C., in Cell Bound Antibodies, 109 (Wistar Institute Press, Philadelphia, 1963).Claman, , H. N., Chaperon, , E. A., and Triplett, , R. F., Proc. Soc. Exp. Biol. and Med., 122, 1167 (1966).ISIChemPortDavies, , A. J. S., Leuchars, , E., Wallis, , V., and Koller, , P. C., in Transplantation, 4, 438 (1966).PubMedChemPortMiller, , J. F. A. P., de Burgh, , P. M., Dukor, , P., Grant, , G., Altman, , V., and House, , W., Clin. Exp. Immunol., 1, 61 (1966).PubMedChemPortDavies, , A. J. S., Leuchars, , E., Wallis, , V., Marchant, , R., and Elliott, , E. V., Transplantation, 5, 222 (1967).Nossal, , G. J. V., Austin, , C. M., Pye, , J., and Mitchell, , J., Intern. Arch. Allergy Appl. Immunol., 29, 368 (1966).ISIChemPortKeuning, , F. J., ,van der Meer, , J., Nieuwenhuis, , P., and Oudendijk, , P., Lab. Invest., 12, 156 (1963).PubMedISIChemPortMitchell, , G. F., and Miller, , J. F. A. P., Proc. US Nat. Acad. Sci. (in the press).