1963) and examined with a Philips 300 electron microscope (60 Kv). Thymic volume was .... composition of lymph node tissue, spleen and thymus. A quantitative ...
Clin. exp. Immunol. (1977) 27, 516-521.
Epithelial cell proliferation in thymic hyperplasia induced by triiodothyronine J. M. SCHE IFF, A. C. CORD IER* & S. HAUMONT Department Medicine, Brussels, Belgium
oj Histology, University of Louvain, Faculty of
(Received 28 September 1976) SUMMARY
A significant hyperplasia of the thymus was induced in mice, treated with triiodothyronine during the first month of life. Stereological data showed that, in both treated and control mice, mononucleate epithelial cells were four times more numerous in the medulla than in the cortex. After triiodothyronine treatment, their absolute number in both cortex and medulla increased two-fold. The number of thymic epithelial cells could thus be regulated by thyroid hormones. The cortical volume in treated mice was also twice that of controls but medullar volume showed an increase of only fifty percent. Cortical epithelial cells increased at the same rate of the cortex volume but medullary epithelial cells grew more rapidly. In fact the medullary volume enlargement could be explained mainly by the growth of the epithelium. Medullary lymphocytes did thus not proliferate in the same way as cortical lymphocytes after thyroid hormone adminis-
tration. The recently described multinucleate epithelial cells were not modified in number and were thus insensitive to thyroid hormones. INTRODUCTION and thymus has been known since thymic hypertrophy was demonstrated Interaction between thyroid Matti (1912) and Hammar (1929). These old observations were confirmed, a few in Graves' disease by and histological studies (Irvine & Sumerling, 1965; Michie et al., 1967). years ago, by radiological induced experimentally by thyroxine (Marder, 1951, Hdhn, 1959) and has been hypertrophy Thymic by TSH (Gregoire, 1942). Conversely, a reduction of thymic hypertrophy in human thyrotoxicosis was obtained by antithyroid drugs (Van Herle & Chopra, 1971) and thyroidectomy was followed by thymic involution (Marine, Manly & Baumann, 1924). The loss of thymic weight due to thyroidectomy was eliminated by daily administration of thyroxine (Reinhardt & Wainman, 1942). In both experimental hyperthyroidism (Gregoire, 1942, Hohn, 1959) and human thyrotoxicosis (Simpson et al., 1975), the thymic hypertrophy is mainly cortical. This hypertrophy is generally assumed to be due to an increase in lymphocytes and, to our knowledge, no modification of thymic epithelial cells has been reported although they are the permanent part of the thymic parenchyma and are implicated in thymic hormonal secretion (Clark, 1966; Dardenne et al., 1974; Goldstein, 1975). The present report shows that epithelial cell number increases after the administration of triiodothyronine. MATERIALS AND METHODS Newborn NMRI mice (three males and three females) were injected daily the 1st week and every 3rd day later on with 0 5 Pg of 3.3',5 triiodo-L-thyronine (Koch-Light) dissolved in 0 05 ml alkaline physiological solution. They were killed by ether intoxication at the age of one month. NMRI mice of the same age (four males and three females) were used as controls. * Charge de Recherche of the Fonds National de la Recherche Scientifique. Correspondence: Dr J. M. Scheiff, Department of Histology, University of Louv-ain, Faculty of Medicine, B 1200 Brussels, Belgium.
516
Epithelial cell proliferation
517
The thymic region was immediately removed and the thymus dissected free from adjacent connective tissue, blotted and weighed. One lobe was fixed in Bouin's fluid, embedded in paraffin and serial sections were cut for stereological analysis. The other lobe was fragmented partly in pieces of 1 mm3 for thin sectioning and partly in pieces of 4 mm3 for stereological analysis after 1 p-thick sectioning. The small fragments were fixed for 2 hr in 2 5%y glutaraldehyde, rinsed and post-fixed for 1 hr in 1%4 osmium tetroxide. The larger fragments were fixed in 4%4 paraformaldehyde. All solutions were in 0-1 M cacodylate buffer at pH 7-4 containing 0 05%4 CaCl2 and 1% sucrose. The pieces were dehydrated in graded ethanol solutions, treated with uranyl acetate, embedded in Epon according to Luft (1961) and cut with a Reichert OMU 3 microtome. One micrometre thick sections were stained with 1% toluidine blue (Trump, Smuckler & Benditt, 1961). Thin sections were mounted on uncoated copper grids, stained with uranyl acetate (Watson, 1958) followed by lead citrate (Reynolds, 1963) and examined with a Philips 300 electron microscope (60 Kv). Thymic volume was obtained by dividing thymic weight by 1 1-an assumed value for the density of thymic tissue. The stereological analysis was performed on a Wild 510 projection microscope equipped for point counting with a multipurpose test screen (Weibel, Kistler & Scherle, 1966). The percentage of thymic cortex and medulla was calculated in serial paraffin sections by counting the number of points overlaying each area. Absolute volumes of cortex and medulla were deduced from this percentage. The number of mononucleate and of multinucleate epithelial cells per mm3 of both thymic areas were evaluated using two formulae:
(1) Al = Pi . 1.,3 2/ . Z2 (Weibel et al., 1966) where Al represents in mm2 the part of total section area belonging to the tissue component i (i.e. cortex or medulla), pi the number of points overlaying this component and Z the equivalent in mm of the distance between two points of the test screen.
(2) NV = N. (De Hoff & Rhines, 1961 in Weibel, 1973) where N, represents the number of cells per mm3, D the mean equivalent diameter of cells and Na the number of cells per mm2.2 The absolute number of each cell type was calculated by multiplying the number of cells per mm3 by the volume of the
corresponding thymic region. Statistical analysis was performed using the method of balanced incomplete blocks according to Fisher & Yates (1963).
RESULTS
Thymus in control animals Body and thymic weight were different in males and females even before puberty. Females were lighter but their thymus was larger; relative thymic weight was thus significantly higher in females (Table 1). On the other hand, cortico-medullar volume ratio (Table 2) and absolute number of cortical TABLE 1. Effect of triiodothyronine on body and thymus weight
Body weight (gr)
M F
Thymus weight (mg)
Thymus relative weight
Controls
T3
Controls
T3
Controls
T3
22-06+ 1-20 19-47+2-05
18-83+ 1-02 15-39+2-70
67-37+9-83 74-30+4-68
133-86+ 18-90
148-36+ 18-43
0 00305+0 00033 0-00383+000018
0-00972±0-00089
TABLE 2. Percentage of thymic regions Cortex
M F
Medulla
Controls
T3
Controls
T3
67-7+2-5 66.8+ 8
73-4+ 1-6 748+ 2
32-3+2-5 33-2+ 8
26-6+ 1-6 25-2+ 2
0-00712+0.00113
J. M. Scheiff; A. C. Cordier C S. Haumont
518
and medullar epithelial cells (Table 3) were the same in both sexes; thymic lymphocytes were thus more numerous in females. In the medulla, mononucleate epithelial cells were four times more numerous than in the cortex (Fig. 1). The absolute number of multinucleate cells (1%/o of the total epithelial cell population) was nearly identical in both thymic regions (Fig. 2). TABLE 3. Absolute number of mononucleate epithelial cells (x 103)
Medulla
Cortex
M F
Controls
T3
Controls
T3
680+40 676+ 163
1312+ 376 1350+ 124
2596+ 350 3136+ 1290
5248+ 746 5551+833
0
0 x
z
FIG. 1. Effect of triiodothyronine on the absolute number of thymic mononucleate epithelial cells. (P< 0-001). Open columns, cortex; hatched columns, medulla.
Triiodothyronine effect on growth and on thymus weight Growth was slowed down with triiodothyronine: the difference in body weight between treated and control animals was 14-6°/, for males and 209% for females. During this time, the thymic weight increased considerably (98.7°/, in males, 99.6% in females); the increase in thymic relative weight (Table 1) was higher in females (153.8%) than in males (133@4%). The volume increase was greater in the cortex (115%) than in the medulla (54O/%) (Fig. 3). Triiodothyronine efects on epithelial cells After triiodothyronine administration the increase in absolute number of mononucleate epithelial cells was 96°/, in the cortex and 91% in the medulla (Fig. 2). This proliferation was similar to the total thymic growth. The number of mononucleate epithelial cells per mm3 decreased in the cortex (- 12%, nonsignificant) but increased in the medulla (+ 24%/o, significant) (Fig. 4). Proliferation of the non epithelial part, viz the lymphocytes, was thus proportional to the increase in volume and in epithelial cell number in the cortex but was very slight in the medulla, since the rate of
519
Epithelial cell proliferation 1o0 +1
E E
0
0 F50
E E z
x
z
lT3
Contro Is
FIG. 2
FIG. 4
FIG. 3
FIG. 2. Effect of triiodothyronine on the absolute number of thymic multinucleate epithelial cells (n.s.). Open columns, cortex; hatched columns, medulla. FIG. 3. Effect of triiodothyronine on the volume of cortex and medulla (P< 0-001). Open columns, cortex; hatched columns, medulla. FIG. 4. Effect of triiodothyronine on the number of thymic mononucleate epithelial cells per mm3 of cortex (n.s.) and medulla (P< 0-01).
medullar volume enlargement could be mainly explained by the multiplication of epithelial cells alone. The epithelial web was thus not simply extended by thymocyte invasion or proliferation but grew by addition of new epithelial elements. Thymic hypertrophy can thus be interpreted as a real hyperplasia owing to multiplication of epithelial cells. This is confirmed by the fact that, in treated animals, mitoses were easily found in epithelial cells (Fig. 5). The number of multinucleate epithelial cells was the same in both cortex and medulla (Fig. 2). Changes observed were in the range of individual variation. These epithelial cells are thus not influenced by thyroid hormones. In treated animals, macrophages were more numerous and showed lymphophagocytosis figures
(Fig. 6).
FIG. 5
FIG. 6
FIG. 5. Mitosis in a thymic medullary epithelial cell. Bundles of microfilaments (arrows). (Magnification x 5200.)
FIG. 6. Macrophage in thymic cortex. Phagocytosis of lymphocytes (arrows) (Magnification x 4800.)
520
J. M.
Scheiff, A. C. Cordier L
S. Haumont
DISCUSSION Thymic hypertrophy in thyrotoxicosis was frequently ascribed to the effect of thyroid hormone on lymphocytes. It seemed, since Hammar's description (1929), not different from the general hypertrophy of lymphoid tissue. This theory is sustained by several arguments. The number of pyroninophilic cells in the thymic cortex (Ernstrom, 1963), of lymphocytes in thymic veins (Ernstrom & Larsson, 1965) and of mitochondria in blood lymphocytes (Ernstrom & Larsson, 1961) rise after thyroxine administration. Moreover fixation and desiodination of thyroid hormones by lymphocytes are enhanced in hyperthyroidism (Holm et al., 1975) and nuclear receptors for T3 are found in human lymphocytes (Tsai & Samuels, 1974). As shown by our results, thymic hypertrophy is not only due to proliferation of lymphocytes. Mononucleate epithelial cells increase in proportion to the growth of the thymus and their multiplication is thus regulated by hormonal secretion. These cells are relatively scarce compared to lymphocytes and their contribution to the total thymic growth is negligible. Their response to thyroid hormone is nevertheless of great significance: they are the functional part of the thymus and thymocyte proliferation is induced by their secretion (Metcalf, 1956; Goldstein, Slater & White, 1966). Increase of thymic lymphocytes in thyrotoxicosis, could thus be, at least partially, consequent to the multiplication of epithelial cells. Hypertrophy of the thymic cortex is known in thyrotoxicosis, but growth of the medulla is, in our experiments, also evident. The lesser importance of this increase is due to the fact that medullary lymphocytes do not proliferate. This suggests the presence of two populations of thymic lymphocytes with different sensitivities to thyroid hormones. The cortical population is directly or indirectly stimulated while the medullar population is not. Poorly differentiated multinucleate epithelial cells, whose number increases with age, have been recently described in both thymic regions (Scheiff, 1976). Their absolute number is not affected after triiodothyronine administration. This insensitivity to thyroid hormone is another difference between these multinucleate cells and common epithelial cells. Germinal centers, plasma cell infiltrations or proliferation of Hassall bodies are common in Graves' disease (Hammar, 1929; Gunn, Michie & Irvine, 1964; Michie et al., 1967; Van Herle & Chopra, 1971). They were not observed after triiodothyronine administration and must be attributed to autoimmune processes. Research for this paper was supported by grant 10-013 of the Fonds de la Recherche Fondamentale Collective. The authors wish to thank Dr A. Leek for his critical reading of the manuscript. The authors also extend their gratitude to Miss L. Vander Straaten for technical assistance, to Mrs E. Claes and Mr J. Goossens for photographic work and to Mrs J. Vander Perre for her secretarial assistance.
REFERENCES CLARK, S.L., Jr (1966) Cytological evidences of secretion in FISHER, R.A. & YATES, F. (1963) Statistical Tables for biological, agricultural and medical research. Oliver and the thymus. The Thymus: Experimental and Clinical Boyd, London. Studies. (ed. by G. E. W. Wolstenholme and R. Porter), GoLDsTEIN, G. (1975) The isolation of thymopoietin p. 3. Churchill, London. (thymin). Ann. N.Y. Acad. Sci. 249, 177. DAiWENNE, M., PAPIERNIK, M., BACH, J.F. and STUTMAN, 0. (1974) Studies on thymus products. III. Epithelial GOLDSTEIN, A., SLATER, F.D. & WHITE, A. (1966) Preparation, assay, and partial purification of a thymic lymphoorigin of the serum thymic factor. Immunology, 27, 299. poietic factor (thymosin). Proc. nat. Acad. Sci. (Wash.) ERNSTROM, U. (1963) Thyroxin-induced changes in cell USA 56, 1010-1017. composition of lymph node tissue, spleen and thymus. A quantitative examination on young Guinea pigs. GREGOIRE, CH. (1942) Recherches sur les facteurs de l'hyperplasie du thymus et du tissu lymphoide dans la Acta path. microbial. scand. 59, 145. maladie de Basedow. Arch. Internat. Pharmacodyn. ERNSTROM, U. & LARSSON, B. (1961) The influence of The'rapie, 68, 263. thyroxin on blood lymphocytes. A quantitative study of the mitochondrial contents of blood lymphocytes in GuNN, A., MICHIE, W. & IRVINE, W.J. (1964) The thymus in thyroid disease. Lancet, ii, 776. Guinea pigs. Acta path. microbial. scand. 51, 17. ERNSTROM, U. & LARSSON, B. (1965) Thyroxin-stimulated HAMMAR, J.A. (1929) Kasuistischer Beitrag zur Frage nacl dem Einfluss endokriner Erkrankungen auf die Thymusvenous output of small lymphocytes from the thymus. druse. Acta med. Scand. 70, 449. Acta Path. Microbiol. Scand. 65, 203.
Epithelial cell proliferation HOHN, E.O. (1959) Action of certain hormones on the thymus of the domestic hen... Endocrinol. 19, 282. HOLM, A.CH., LEMARcHAND-BERAUD, TH., SCAZZIGA, B.R. & CUTTELOD, S. (1975) Human lymphocyte binding and deiodination of thyroid hormones in relation to thyroid function. Acta Endocr. 80, 642. IRVINE, W.J. & SUMERLING, M.D. (1965) Radiological assessment of the thymus in thyroid and other diseases. Lancet, i, 996. LUFT, J.H. (1961) Improvements in epoxy resin embedding methods. 7. Biophys. Biochem. Cytol. 9, 409. MARDER, S.N. (1951) Effect of thyroxine on lymphoid tissue mass of immature female mice. J. nat. Cancer Inst. 11,
1153. MARINE, D., MANLEY, O.T. & BAUMANN, E.J. (1924) The influence of thyroidectomy, gonadectomy, suprarenalectomy, and splenectomy on the thymus gland of rabbits. J. exp. Med. 40, 429. MATTI, H. (1912) Uber die Kombination von Morbus Basedowii mit Thymushyperplasie. Cor. B!. Schweiz.
Aerzte. 42, 1345. METCALF, D. (1956) The thymic origin of the plasma lymphocytosis stimulating factor. Brit. J. Cancer, 10, 442. MICHIE, W., BECK, J.S., MAHAFFY, R.G., HONEIN, E.F. & FowLER, G.B. (1967) Qpantitative radiological and histological studies of the thymus in thyroid disease. Lancet, i, 691. REINHARDT, W.O. & WAINMAN, P. (1942) Effect of thyroidectomy, castration and replacement therapy on thymus,
K
521
lymph nodes, spleen in male rats. Proc. Soc. exp. Biol. (N.Y.) 49, 257. REYNOLDS, E.S. (1963) The use of lead citrate at high pH as an electron-opaque stain in electron microscopy. J. Cell. Biol. 17, 208. SCHEIFF, J-M. (1976) Multinucleate cells in thymus of C3H mice. Cell. Tiss. Res. 170, 305. SIMPSON, J.G., GRAY, E.S., MICHIE, W. & BECK, J.S. (1975) The influence of preoperative drug treatment on the extent of hyperplasia of the thymus in primary thyrotoxicosis. Clin. exp. Immunol. 22, 249. TRUMP, B.F., SMUCKLER, E.A. & BENDITT, E.P. (1961) A method for staining epoxy sections for light microscopy. J. Ultrastruct. Res. 5, 343. TSAI, J.S. & SAMUELS, H.H. (1974) Thyroid hormone action: Demonstration of putative nuclear receptors in human lymphocytes. J. cdin. Endocrinol. 38, 919. VAN HERLE, A.J. & CHOPRA, I.J. (1971) Thymic hyperplasia in Graves' disease. 7. cdin. Endocrinol. Metab. 32, 140. WATSON, M.L. (1958) Staining of tissue sections for electron microscopy with heavy metals. J. Biophys. Biochem. Cytol. 4, 475. WEIBEL, E.R. (1973) Stereological techniques for electron microscopic morphometry. Principles and techniques of electron microscopy. Biological applications. (ed. by M. A. Hayat), volume 3, p. 237. Van Nostrand Reinhold Company, New York and London. WEIBEL, E.R., KISTLER, G.S. & SCHERLE, W.R. (1966) Practical stereological methods for morphometric cytology. 7. Cell Biol. 30, 23.
