Neonatal thymectomy of mice, when no ectopic thymus existed, constantly resulted in developmental arrest of the ovary but not of the testis; it also caused sterility in the female. The ovaries of thymectomized mice were extremely small and were characterized by absence of follicles and corpora lutea. Such an ovarian dysgenesia was observed when the mice were thymectomized at 3 days of age, but not at 7 days or later; it was prevented by thymus grafting.

Various organ-specific autoimmune diseases can be readily induced in some strains of mice by thymectomy during the critical neonatal period (NTx), 1 on day 2-4 after birth, without any exogenous sensitization with autoantigen (I-4). ~t is assumed in this model that NTx may eliminate T cells that would otherwis6 ontogenically peripherize via the thymus after the day of thymectomy (Tx), and may consequently allow the residual population of T cells that have already peripherized before Tx to respond to self-antigen(s) and eventually to cause autoimmune diseases in adulthood. In other words, NTx may deplete a subpopulation of T cells that in the normal state should suppress or regulate self-reactive clones. If this assumption is correct, then it should be possible to prevent the development of autoimmune disease by reconstituting NTx mice with the appropriate lymphoid cells. In fact, it was demonstrated that autoimmune diseases induced by NTx could be prevented by inoculation of splenic T ceils or thymocytes from normal syngeneic adult mice (4-6).In this study we have attempted to characterize the regulatory T cell population that is capable of preventing autoimmune oophoritis by cell surface antigens, particularly Lyt antigens, and other immunobiological features. The cells responsible for preventing the development of oophoritis in the spleens were found to be Thy-1 +, Lyt-l+,23 -cells, and were not eliminated by adult thymectomy (ATx). Thymocytes having such a preventive capacity were also found to be Lyt-l+,23 -. Thus, the depletion of this Lyt-1 regulatory subpopulation by NTx may result in the development of oophoritis as well as other post-thymectomy autoimmune diseases. Materials and MethodsMice. A/J mice were kindly provided by Dr. K. Moriwaki, Institute for Genetics, Misima, Japan; and both BI0.A(3R) and B10.A(5R) mice were kindly provided by Dr. M. Taniguchi, *

Isografts of heterotypic recombinants of embryonic mammary epithelium with salivary mesenchyme undergo development morphogenetically resembling that of salivary gland. However, cytodifferentiation of the epithelium is like that of mammary gland. In lactating hosts these isografts respond to endogenous hormonal stimulation and synthesize a milk protein, alpha-lactalbumin.

Neonatal thymectomy during the critical period, 2-4 d after birth, can induce various organ-specific autoimmune diseases including oophoritis in A/J mice. The oophoritis thus induced was passively transferred into neonatal mice by injection of spleen cells obtained from syngeneic donors with the disease. Recipient ovaries were rapidly damaged with remarkable mononuclear cell infiltration and destruction of follicular structures. The phenotype of effector cells responsible for successful adoptive transfer was found to be Thy-1+, Lyt-1+,23-, Ia-, Qa-1-, and was sensitive to antithymocyte serum treatment but resistant to cyclophosphamide treatment or in vitro X-ray irradiation. The compatibility between donor and recipient at the major histocompatibility complex was not required for the effector phase of transfer. The oophoritis induced in BALB/c (nu/+ or +/+) was also shown to be transferred into athymic BALB/c nude mice with resulting ovarian lesion and circulating autoantibodies against oocytes. In this transfer system, the effector cells were also demonstrated to be T cells with the Lyt-1+,23- phenotype. Adoptive transfer experiments in both systems revealed that the destruction of ovaries in postthymectomy autoimmune oophoritis was mediated by Lyt-1 T cells. Whether these T cells can be distinguished from other Lyt-1 cells, such as T helper cells and effector T cells in delayed-type hypersensitivity (DTH), is not clear at present, but the results suggest that the effector mechanisms may be closely related to a DTH reaction.

Autoimmune diseases appeared frequently in adults in the prostate and stomach of C3.129 mice after thymectomy on day 3 (Tx-3) without any additional treatment. Lesions of both organs could be completely prevented by a single i.p. injection of spleen cells from syngeneic adult mouse on day 4. For prevention of prostatis, the most effective cell source was normal males (4 X 10(6); normal females or Orx-0 males were less effective as the cell source, and higher doses of cells (4 X 10(7)) were needed. In contrast, spleen cells (4 X 10(6)) from these three donors had equivalent capacity for the prevention of gastritis. Similar autoimmune prostatis developed at very high frequency when spleen cells (4 X 10(6)) from normal females or Orx-0 males, but not from normal males, were injected i.p. into C3.129 nu/nu mice at 4 d. However, no sign of prostatis was found in nu/+ recipients. Injection of a larger dose (4 X 10(7)) from the same donors was not effective for induction of prostatis. Gastritis could not be induced in nu/nu mice by this procedure. Injection of spleen cells from Tx-3 males or females was effective for induction of both prostatis and gastritis in nu/nu recipients. It was also shown that a T cell population (Thy-1.2+, Ig-) had the capacity to prevent and to induce autoimmune diseases. These results together strongly suggest a role for active tissue-specific suppressor T cells in self tolerance, and elimination of such T cell populations causes autoimmunity.