Studies on thymus products: II. Demonstration and characterization of a circulating thymic hormone

Normal mouse serum confers on bone marrow rosette-forming cells (RFC) from normal mice a high sensitivity to anti-theta serum (AθS) and azathioprine (AZ) which they otherwise lack. Such an effect can also be demonstrated on spleen cells from adult thymectomized mice, `thymus-deprived' and nude mice, but the amount of serum required is higher in the latter mice. This activity of serum on RFC disappears after thymectomy of the serum donor with a half-life of 2.5 hours and reappears in thymectomized mice within 4 days after grafting of a thymus. Serum thymic activity (TA) is present in different amounts in different mouse strains and in ageing mice it progressively disappears. No TA is found in the serum of 4-week-old nude mice. TA is stable after lyophilization but is thermolabile in solution. It passes through UM 10 Amicon membranes, which suggests that its molecular weight (mol. wt) is probably < 10,000. TA is reversibly suppressed in the presence of a serum inhibitor with a mol. wt between 100,000 and 300,000. This inhibitor is not detectable in the serum of thymectomized mice unless the serum is incubated with TA containing serum for 60 minutes at 37°. The biological significance of TA is still a matter of speculation but its role in maturation or expansion of T-cells is suspected.     

Studies on thymus products. VI. The effects of cyclic nucleotides and prostaglandins on rosette-forming cells. Interactions with thymic factor

Cyclic AMP, its dibutyryl derivative and prostaglandins, PGE₁ and PGE₂, do not modify the numbers of rosette-forming cells (RFC), but they do alter these cells' sensitivity to in vitro inhibition by azathioprine (AZ) and anti-theta serum (ATS). In thymectomized mice, cyclic nucleotides and prostaglandins mimic the action of the thymic factor contained in thymic extracts and in normal serum, correcting the low spleen RFC sensitivity to AZ and ATS induced by thymectomy. Moreover, the mixture of infraliminal active doses of cyclic AMP and of thymic extracts (or normal mouse serum) also gives this effect. In normal mice, the nucleotides (but not prostaglandins) lower the AZ and ATS sensitivity of RFC, which remains insensitive to thymic extracts. Theophylline and isoproterenol have effects similar to those of cyclic AMP, a result in accord with their postulated interactions with the adenyl-cyclase system. Noncyclic AMP and cyclic GMP show no effect in the RFC assay system at doses 10–100 times higher than active doses of cyclic AMP. Other vasoactive compounds, including epinephrine, phentolamine, propanolol, and angiotensin, do not show any effect on RFC.     

Studies on thymus products: I. Modification of rosette-forming cells by thymic extracts. Determination of the target RFC sub-population

Thymic extracts confer on normal bone marrow rosette-forming cells (RFC) in vitro a high sensitivity to anti-theta serum (AθS) and azathioprine (AZ) which they usually lack. Thymic extracts can also confer high sensitivity to AθS and AZ to spleen RFC from adult thymectomized, neonatally thymectomized `thymus-deprived' and nude mice. However, the amount of thymic extracts necessary to get the effect is significantly higher on thymus-deprived and nude mouse spleen RFC than on RFC from spleens of adult thymectomized mice or normal mouse bone marrow. Thymic extracts are also active in vivo and there is a good correlation between the in vitro minimum concentration giving AθS and AZ sensitivity to spleen RFC from adult thymectomized mice and the in vivo minimum dose giving such an effect after intravenous injection. The effect of extract in vivo does not appear before the fourth hour after the injection and is transient, disappearing after 48 hours. Injection of thymic extracts induces the appearance of a `thymic activity' (TA) in the serum with a short half-life (2 hours). Injection of thymic extracts into normal mice does not modify sRFC characteristics in spleen, lymph nodes and bone marrow. It is suggested that thymic extracts act reversibly on a population of T-RFC precursors.     

Neonatal thymus grafts. I. Studies on the regulation of the level of circulating thymic factor (FTS).

The level of circulating thymic factor (FTS) tested by its action on spleen rosette-forming cells from adult thymectomized mice, has been shown to be stable in young animals. This stability suggests a regulatory mechanism. An approach of this regulation has been attempted by disrupting the FTS level either by neonatal thymus grafting in adult normal and thymectomized animals, or by injections of synthetic thymic factor into normal, thymectomized mice, grafted or ungrafted. In all cases, after an initial increase over the normal value, FTS levels returned close to the previous range, indicating the existence of some homeostatic mechanism of FTS secretion.     

Studies of the thymus in mice bearing the Lewis lung carcinoma. III. Possible mechanisms of tumor-induced thymic atrophy

Adrenalectomy prevents thymic atrophy but not splenomegaly in mice implanted with Lewis Lung carcinoma cells. Surprisingly, the presence of the tumor does not lead to increased levels of corticosterone, which argues against an exclusive role of stress in the tumor-induced involution of the thymus. Interestingly, serum from tumor-bearing hosts in vitro displays strong cytolytic activity against normal syngeneic thymocytes. This thymocytotoxicity depends upon the stage of tumor development, i.e., the size of the local tumor, and is concomitant with the severe thymic atrophy. Treatment of donor mice with zinc chloride or excision of the local tumor, both of which have been shown to prevent this involution of the thymus, also abolishes the cytotoxic effect of the serum. The active component of the serum is a nonimmunoglobulin fraction of molecular weight greater than 25,000 Da. The possible mechanisms of tumor-dependent thymic atrophy as well as the in vivo relevance of this serum-mediated thymocytotoxicity are discussed.     

Epithelial proliferation in thymic hyperplasia. An immunohistochemical study and correlation with the developing fetal thymus

Thymic hyperplasia is a B-cell lymphoid proliferation in which an epithelial component has not, to our knowledge, been previously described. We present a case of thymic hyperplasia in which numerous lymphoid follicles with germinal centers were partially surrounded by small sheets of spindle and epithelioid cells. Electron microscopy confirmed the epithelial nature of these cells. Immunostaining was performed using antibodies to keratins, S100 protein, and two B-cell markers, LN1 and MB2. The proliferated epithelium stained only for high-molecular-weight keratin, whereas the lymphoid tissue stained positively for both B-cell markers. To determine the origin of the proliferated epithelium, the staining was compared with that of the developing fetal and normal adult thymus. We have shown that during fetal development, the keratin composition of thymic epithelium changes from staining predominantly with low- to high-molecular-weight keratin. The immunostaining characteristics of the epithelium in this case of thymic hyperplasia suggest an origin from adult-type epithelium. Furthermore, the association of S100-positive interdigitating reticulum cells with the proliferated epithelium suggests that it is of medullary origin. Our results indicate that epithelial proliferation can be an important component of thymic hyperplasia.     

Studies on thymus products. V. Influence of genetic selection based on antibody production on thymus hormone production

Outbred Swiss mice show a stable level of circulating thymic factor (TF) (measured by its action on theta-positive rosette-forming cells (RFC) until the 6th month of life. Afterwards, this level progressively declines. Swiss mice genetically selected as `high' and `low' responders for anti-sheep red blood cell haemagglutinin production show no difference with ordinary Swiss mice with regard to serum TF level. This observation is compatible with previous results showing that the selection mainly induces changes in B-cell reactivity without T-cell modification. Conversely, genetic selection of Swiss mice based on spontaneous antinuclear autoantibody production (tested by immunofluorescence) induces a decrease in TF level, which has already become significant by 2 months of life. The mice thus selected (Swan mice) which present autoimmune manifestations and immune complex disease, like NZB and B/W mice, show the same premature cessation of TF as these other autoimmune mice. Such abnormality is compatible with the T-cell deficiency of NZB, B/W and Swan mice.     

Effect of a human serum thymic factor on hydrocortisone-treated thymocytes.

A human serum thymic factors (SF) stimulated cyclic adenosine-3' ,5'-monophosphate (cAMP) synthesis in normal mouse thymocytes. Such stimulation was no longer observed when thymocytes were depleted of hydrocortisone (HC)-sensitive cells. It was concluded that SF selectively stimulate cAMP in HC-sensitive cells. Furthermore, incubation of thymocytes with SF enlarged the population of HC-resistant thymocytes. These results suggest that SF might act on HC-sensitive thymocytes increasing their cellular cAMP level and inducing their transformation in HC-resistant cells.     

Thymic hormone containing cells. III. Evidence for a feed-back regulation of the secretion of the serum thymic factor (FTS) by thymic epithelial cells.

Cells containing the serum thymic factor (FTS), as measured by indirect immunofluorescence, were studied in mice either FTS depleted by injection of anti-FTS monoclonal antibodies or immunization against FTS coupled to bovine serum albumin (FTS-BSA), or FTS enriched by multiple injections of synthetic thymulin (FTS-Zn). Injections of thymulin did not significantly depress FTS secretion. Conversely, long term elimination of FTS from peripheral blood caused a great increase in the thymic intracellular content of FTS, as evidenced by the higher number of FTS containing cells observed with immunofluorescence. These data could provide the first evidence of a feed-back control of thymic endocrine function.     

Epithelial cell heterogeneity in the guinea pig thymus: immunohistochemical characterization of four thymic epithelial subsets defined by monoclonal anti-keratin antibodies

Keratins are a family of related polypeptides constitutive of the cytoskeleton of epithelial cells and are never found in nonepithelial tissues. Thymic epithelial cells (TEC), known to induce T cell differentiation, are the keratin-containing cells within the thymus. Using four monoclonal anti-keratin antibodies (KL1, KL4, AE2, AE3) directed against keratins of different molecular weight, we have investigated the guinea pig thymic epithelium. The immunohistochemical analysis of thymic cryostatic sections revealed that the keratin expression of TEC varied according to their location in the thymic lobula; the thymic cortex was specifically stained by AE3 whereas the thymic medulla and the subcapsular cortex were recognized by KL4. In addition, KL1 and AE2 exclusively labeled Hassall's corpuscles. The biochemical analysis of keratins extracted from the thymus showed that each TEC subset was characterized by an unique pattern of keratin polypeptides. This study extends the concept of thymic epithelium heterogeneity and suggests that anti-keratin antibodies which allow the typing of TEC subsets may be valuable tools for studying the differentiation of thymic epithelium and its in vitro function on T lymphocytes.     

Studies on the thymus in Chagas' disease. I. Changes in the thymic microenvironment in mice acutely infected with Trypanosoma cruzi

Previous observations demonstrated severe thymocyte depletion in mice undergoing acute Chagas' disease. These data led us to investigate the status of the thymic microenvironment in these animals. Young adult C57BL/6 and C3H/HeJ mice were infected i.p. with 10(5) blood-derived trypomastigote forms of Trypanosoma cruzi (CL strain) and killed 7-14 days after infection. Sera were then analyzed for thymic hormone (thymulin) levels, and frozen thymus sections were studied by immunohistochemistry for the expression of functional antigens (thymulin and Ia), the distribution of distinct thymic epithelial cell subsets and extracellular matrix components. Infected mice exhibited a transient decrease in thymulin production and those with severe thymic atrophy showed a denser Ia-bearing cellular network. In addition, an abnormal localization of the TR5 and CK18 antigens restricted to the medullary and cortical TEC subsets, respectively, was observed. Furthermore, an increase in the basement membrane proteins was detected within thymic lobules. We suggest that the thymic microenvironment is also affected during T. cruzi infection, extending the concept that the thymus should be regarded as a target in Chagas' disease.     

Lymphocyte markers, serum thymic factor and IgE in IgA deficiency.

Eleven patients having a selective IgA deficiency had very low T-cell percentages assessed by using E-rosettes, active E-rosetts and an anti-human anti-T-lymphocyte antigen serum. B-cells were normal or slightly decreased. IgE concentrations were often low. Patients with serum IgA under 50 IU/ml often had elevated serum IgE values and low T-cell percentages. Serum thymic factor dosages were correlated with the E-rosette assay.     

Antigenic specificity of two antibodies directed against the thymic hormone serum thymic factor (FTS)

Serum thymic factor (facteur thymique serique, FTS) induces in vitro differentiation of T-cell precursors into more mature cells with T-cell characteristics. As isolated from porcine serum, FTS is the nonapeptide GIp-Ala-Lys-Ser-Gln-Gly-Gly-Ser-Asn. We have described two radioimmunoassays that detect FTS but not other thymic hormones [Ohga et al., (1982) J. Immun. Methods, in press]. One assay is based on a monoclonal antibody from a hybridoma derived by fusion of mouse myeloma cells and spleen cells from a mouse immunized with an FTS-mouse IgG conjugate. The second assay is based on the antiserum from a rabbit immunized with FTS bound to F(ab')₂ fragments of rabbit IgG. The detailed antigenic specificity of these anti-FTS antibodies was determined by measuring the ability of FTS and 12 synthetic FTS peptide analogues to compete with a radioiodinated FTS analogue in these radioimmunoassays. The mouse monoclonal antibody and the rabbit antiserum showed similar structural requirements for binding of the FTS peptides. Since FTS had been attached to the carrier proteins through the ε-amino group of Lys-3, both antibodies were relatively insensitive to omission of 1 or 2 N-terminal residues, replacement of Glp-1 with either Ala or Tyr-Ala, or substitution of Lys-3 with Ala. In contrast, binding of FTS to the antibodies was substantially decreased by omission of 3 or 4 N-terminal residues, omission of 2 or 3 C-terminal residues, or replacement of Gly-7 or Asn-9 with Ala. Relative to the mouse monoclonal antibody, the rabbit antiserum was more sensitive to the omission of 4 N-terminal residues and much more sensitive to replacement of Gln-5, Gly-6, or Asn-9 by Ala. Both antibodies were relatively specific for molecules ending in -Xxx-Xxx-Gly-Gly-Ser-Asn-OH, which corresponds to the biologically active region of FTS.     

Studies on the effect of a thymic humoral factor on differentiation of thymus-derived lymphocytes

The effect of a thymic humoral factor (THF) on differentiation of thymus-derived (T) lymphocytes was studied in various graft-versus-host models. It was demonstrated that THF, which is known to be capable of restoring immunocompetence of cells from neonatally thymectomized mice, does not act on lymphoid populations devoid of T cells. Immunologically defective lymphoid cells such as spleen cells obtained from mice treated with anti-lymphocyte serum could be activated by THF by in vivo or in vitro interaction. These experiments indicate that the target cell acted upon by THF is a T cell with a low level of immunocompetence. It is suggested that THF acts on young T cells and transforms them into mature T cells active in cell-mediated immune reactions, presumably without substantial multiplication.     

Studies of the thymus in Chagas' disease: III. Colonization of the thymus and other lymphoid organs of adult and newborn mice by Trypanosoma cruzi

Among newborn and adult euthymic and athymic mice infected intraperitoneally with 10(2) or 10(4) trypomastigotes, neonates and adult athymic animals exhibited greater susceptibility with heavier colonization of central and peripheral lymphoid organs. Contiguity was also found to be a significant factor in colonization: after similar inocula, the parasitic load in the thymus was comparable in adults and neonates when the former were infected by subcutaneous injection in the anterior neck. In the nude athymic mice, the Y (reticulotropic) strain and the CL (myotropic) strain exhibited similar invasiveness, suggesting that the host immune response modulated the tissular tropism of these strains seen in the heterozygous littermates (Nu/+). The bone marrow was heavily infected; parasites were found in chondrocytes and cells of the mononucleate phagocytic system. Ultrastructural studies of infected thymus specimens showed that the corticomedullary architecture was unaltered and that both macrophages and epithelial cells were infected: both these cell types were infected in vitro.     

Ultrastructural studies of thymic reticulum: I. Epithelial [corrected] component

In studying the epithelial component of the mouse thymic reticulum, we identified by their ultrastructural aspect three types of epithelial cells: type I, present in the cortical and medullary zones; and type II and type III, present only in the medulla. These cells form a network entrapping lymphocytes. In addition, some cells are regrouped and form two types of associations that are found in the medulla: cystic cavities and Hassall's corpuscles [corrected]. Comparisons are made between our observations and those described in older publications. The morphologies and roles of these cells as well as the terminologies used to describe them are discussed in light of the new knowledge acquired concerning the function of the thymus.     

Epidermal growth factor can replace thymic mesenchyme in induction of embryonic thymus morphogenesis in vitro

The thymus is surrounded by a thin layer of mesenchyme and the epithelial-mesenchymal interaction is known to be essential for the thymus development. To clarify the roles of mesenchyme in the thymus lobule formation that occurs around embryonic days 14–15 in vivo, we set up a three-dimensional organ culture system. The epithelium of embryonic day 13 thymic primordium was separated from the mesenchyme and cultured in Matrigel (reconstituted basement membrane). Addition of the mesenchyme to a chamber separated by a membrane filter induced the lobule formation of the thymic epithelium in vitro. We found that epidermal growth factor (EGF) can replace the mesenchyme for lobulation of the embryonic thymus in vitro. Among other growth factors tested, only transforming growth factor (TGF)-α was as effective as EGF, in agreement with the fact that EGF and TGF-α bind to the same receptor. These results suggest that EGF or its family members may be involved in morphogenesis and differentiation of the thymus gland epithelium, although we cannot exclude the possibility that other unknown factors are required in vivo.     

Low-grade metaplastic carcinomas of the thymus: biphasic thymic epithelial tumors with mesenchymal metaplasia--an update.

A group of biphasic low-grade thymic epithelial tumors is presented that we suggest calling low-grade metaplastic carcinoma of the thymus [37]. Four of the patients were men, their age ranging from 44 to 71 years. Three tumors invaded mediastinal fat or pleura. No metastases were present. Histologically, the tumors showed a biphasic pattern with solid carcinomatous areas merging with a spindle cell component. Only few lymphocytes were present. Cytological atypia and mitotic activity were variable in the solid areas, but low in the spindle cell component. The tumors showed expression of cytokeratin, vimentin and/or epithelial membrane antigen (EMA), both in the carcinomatous and in spindle cell components. In two cases, actin expression was also present in both components. In one case, chromogranin, S100 protein, glial fibrillary acidic protein, and neuron specific enolase were expressed in a minority of cells of both components. None of the patients had myasthenia gravis. All patients are alive without distant metastasis, but meanwhile one patient suffers from local recurrence. We conclude that metaplastic carcinoma of the thymus is a clinicopathological entity that is probably distinct from the recently described "thymoma with pseudosarcomatous stroma", and should be distinguished from the usually benign medullary thymomas and the highly aggressive carcinosarcomas and sarcomatoid carcinomas.