IN VITRO
Vol. 11, No. 4, 1975
EFFECT OF CORTISOL ON THE U L T R A S T R U C T U R E OF NORMAL, LEUKEMIC, A N D C U L T U R E D H U M A N L Y M P H O C Y T E S SEYMOUR WEIITHAMER 1 AND LEONARD AMARAL Departments of Pathology, Dowt~state Medical Ce~der, a~d The Methodist Hospital of Brooklyt~, Brooklytt, New York 11215 SUMMARY
The effect of cortisol on the ultrastructure of normal, leukemic, and cultured human lymphocytes during a 2-hr incubation was investigated. The presence of 10-5 M cortisol in the incubation medimn produced in normal lymphocytes a variety of alterations in cytoplasmic organelles. Mitochondria were most affected and showed evidence of irreversible deterioration (formation of myelin figures). Occasional cells demonstrated an overt rearrangement of their cytoplasmic membranes resulting in a bizarre array of parallel cisternae-like structures. More commonly, the usually underdeveloped Golgi of normal lymphocytes became very pronounced in structure. All of these alterations were produced within 2 hr of incubation, but only in normal human lymphocytes. Under identical conditions, no evidence of ultrastructural changes were produced by cortisol in either lymphoeytes from chronic lymI)hocytic leukemic patients, or those from the R P M [ 1788 cell line. The use of biochemical metimdology in studies in vitro has contributed largely to the understanding of the mechanism by which cortisol depresses the lmmbers of circulating lymphocytes of mammals. Such studies have clearly demonstrate(1 that cortisol inhibits the synthesis of lymphoid I ) N A (1-3), R N A (4-6), and protein (7, 8), enhances the degradation of RNA (5, 8), and decreases the utilization of glucose (9). The sequence of events appears to involve first the uptake of the hormone, an event dependent upon cellular energy and regulated in some manner by a plasma component (10). Following uptake, glucose utilization is affected (9), after which the energy-dependent degradation of newly synthesized R N A occurs (8). Inhibition of protein synthesis is evident long after the above alterations occur, and appears to be independent, during the periods of time studied, of the effects of cortisol on the synthesis of R N A (8). Although the above studies have contributed to the partial understanding of the mechanism in question, the precise effect of cortisol on the components of the lymphocyte has remained
largely mLstudied. Because ultimate understanding nmst eventually relate the biochemical functional effects to structure, we have investigated the effect of cortisol ill vitro oil the structure of human lymphocytcs. The results presented indicate that the presence of 10-'~ M cortisol in the culture medium produces alteration ill mitochondrial structure, induces the development of Golgi and, in occasional cells, a reorganization of the endoplasmic reticulum. These structural alterations are not produced either in lymphocytes of patients with chronic lymphocytic leukemia, or in act established lymphocyte cell line, by concentrations of cortisol as high as 10-~ M. MATERIALS AND METHOI)S
Human tymphoeytes, purified by the method of Rabinowitz (11), were obtained from nine healthy t)ersons, suspended in platelet-frce aut(Jlogous plasma, and incubated for 1 to 2 hr in TC 199 medium (l)ifeo), with and without final concentrations of cortisol (hydroeortisone, Sigma) of 10-7 to 10-~ M. A stock solution of cortisol at 10.4 hi was prepared by dissolving the appropriate 1 Send requests for reprints to Dr. Seymour Werthamer, Department of Pathology, Methodist weight of eortisol in a benzene-ethanol solution (10:1). This stock solution was prepared daily Hospital, 50(i Sixth St., Brooklyn, N. Y. 11215. 212
CORTISOL ON HUMAN LYMPHOCYTES and, before use, appropriate aliqucts were transferred to culture tubes. The solutions were dried under a constant stream of nitrogen, after which appropriate volumes of culture medium were added and the dried cortisol was dissolved. In order to ascertain that indeed all of the cortisol had dissolved in the culture medium, separate aliquots of the cortisol stock solution were added to identical culture tubes, after which 0.1 Ci of [3H]cortisol was added. The solutions were dried as above, medium was added, and a sample of 0.1 ml was then assayed for total radioactivity. The results assured us that all of the cortisol had been dissolved and that the concentrations of cortisol noted were those in the medium. Addi-
213
tional details concerning the culture conditions have been prviously presented (7, 8). Following the period of incubation, the cells were centrifuged at 850 • g, the supernatant fluid was discarded and the cells were suspended in ice-cold 2.5% glutaraldehyde for 2 hr, centrifuged at 850 • g, resuspended in 1% osmium tetroxide for 1 5 hr, dehydrated in ascending concentrations of ethanol (three washes in each), embedded in Epon 812, and sectioned with a Reichert ultramicrotome. The sections were stained with uranyl acetate and lead citrate, and studied with a Hitachi HS-8 electron microscope at a voltage of 50 kv. Other human lymphocytes were obtained from 14 patients diagnosed as having
FIG. 1. Control, normal lymphocyte. N, nucleus; G, Golgi; M, mitochondrion; L, lysosome. X 11,000.
214
WEI~TI~AMI,]I( ANI) AMAI,AL
chronic lymphocytic leukemia (CLL, with leukocyte counts in excess of 50,000 per ram3), who at the time were not receiving treatment, a~(l from a human lymphocyte cell line. of G. L. Moore ( R P M I 1788) (purchased fr()m Associate(1 lliomedic ~ystems, Inc., Buffalo, N. Y.), that has been successfully cultured in our laboratory for more than o~le year. RESULTS
The typical ultrastructurc of normal human lymphocytes incubated for either 1 or 2 hr under the control conditions specified in the text is illustrated in Figs. 1 and 2. lCricfly, w(~ll d('fitle(l
mitochon(lria are generally clustered in the vicitfity of tile il~dentation of the nucleus and marginal to the underdeveloped Golgi. The ('ytol)lasm contains a few scattered clefts, vesicles, electron (lel~se bodies (probably lysosomes), all(l scattered ribosomes. The often ilLdente(l mlcleus is usually characterized by dense chromatin (heterochromatin) concentrated e~t it~ periphery. In about half of the se('.tions studied, a well defined ~ucleolus was obacrvcd to be situated slightly off-center. In almost all cases the nucleus was surrom~ded by a ('l~ar peri~uclear spa(~e.
I Jl coHtrast to the above typical ultrastructure,
FIG. 2. Control, normal lymphocyte. X44,000.
CORTISOL ON HUMAN LYMPHOCYTES
215
Fro. 3. Cortisol-treated, normal lymphocyte. AM, altered mitoehondrion; MF, myelin figure; MMF, mitoehondrial myelin figure; G, Golgi. Xll,0C0. only lymphocytes that had been iucubated for a minimum of 2 hr in medium containing l0 -5 M cortisol demonstrated severe alterations in their cytoplasmic organelles (Figs. 3 and 4). Mitochondria appear more angular (Fig. 3), pinched (Fig. 3), or sometimes club-shaped (Fig. 4), and all showed various degrees of disorganized cristae. Myelin figures, which were not observed by us in the control normal lymphocytes, were numerous in the cortisol-treated cells (Fig. 3). 'rhese myelin figures were usually in close proximity to mitochotldria (Fig. 4) or were themselves a part of a migochondrion (Fig. 5). The shape of the myelin figure varied from a sphere (Fig. 6) to a "horse-
shoe" (Fig. 7). Golgi, which is usually not well developed in control normal lymphoeytes, was very pronfinent in about 25% of the cells examined (Figs. 3, 8). In fewer than 5% of the cells examined, the region normally occupied by the Golgi exhibits in its place an array of continuous parallel cytoplasmic membranes (Fig. 4). The electron dense strata between tile stacks of membranes appeared continuous with that of the rest of the cell cytoplasm. In all cases, the cytoplasmic strata of the cortisol-treated cells appeared much more electron dense than those of the control normal lymphoeytes. The above cytoplasmic alterations were less frequent and
216
WERTHAMER AND AMARAL
F~G. 4. Cortisol-treated, normal lymphocyte. C, stacks of cytoplasmic cisternae. X 11,000. much less overt with 1 0 - 6 M of cortisol. At a cortisol concentration of 10-~ M, no significant ultrastructural changes were observed. I n contrast to the above findings, the ultrastructure of the cells of the CLL patient did not appear affected by the presence of eortisol in the medium at concentrations as high as 10-5 M. As is evident from Figs. 9 and 10, no alteration in any cytoplasmic organelle was produced. Lymphoeytes from a total of 14 CLL patients were similarly incubated and all failed to show any of the ultrastructural changes typically produced in normal cells by identical steroid concentrations. Similarly, incubation of human lymphocytes of the cell line R P M I 1788 of Moore with con-
centrations of cortisol as high as 10-5 M failed to yield any noticeable changes in ultrastructure. Fig. 11 depicts the ultrastrueture of control cells and Fig. 12 is that of a R P M I 1788 lymphocyte cultured in the presence of 10-5 M eortisol for 2 hr. DISCUSSION The results obtained in this study suggest that incubating normal human lymphocytes in medium containing 10-~ M eortisol results in gross alterations in the structure of mitoehondria; induces the development of myelin figures, some of which involve mitochondria; increases in some cells the extent of Golgi; and, in a few cells, promotes the reorganization of a F.ortion of the cytoplasmic membranes. These alterations are
CORTISOL ON HUMAN LYMPHOCYTES
217
:FIG. 5. High magnification of a myelin figure of a mitochondrion. X40,260.
Fla. 6. High magnification of a myelin figure. X91,500.
produced within a 2-hr period of incubation al~d are significantly evident only with a minimum concentration of 10-~ M cf corti~ol. In contrast to these effects, CLL lymphocytes, previously proven to be more sensiti~ e to depression of RNA and protein synthesis by cortisol than normal cells (8), fail to exhibit any noticeable changes in their ultrastructure. Lymphocytes of the R P M I 1788 cell line are also apparently not affected. These findings suggest that, although both the normal and the CLL lymphocyte are biochemically affected by cortisol, the latter being more sensitive, the mechanisms whereby metabolic synthesis are affected might in part be different for the two types of cells. The alterations observed in the normal lymphocyte, particularly those of the mitochondria, have been previously observed by others in mouse thymic lymphoeytes (12) and in rat thymic epithelial cells (13). In all cases the alterations were consistent with various stages of cell degeneration and consonant with processes denoting an ongoing autophagic process (14). Whether degenerative or autophagic processes in lymphoid cells are specifically induced by cortisol,
F I 6 . 7 . High nlagnification of a horseshoe-like myelin figure. X50,325.
2]8
WERTHAMER AND AMARAL
FI(~. 8. Cortisol-treated, normal lymphocyte. Note extensive Golgi. X ll,0C0. or whether other steroids can similarly effect changes, is a question which has not yet been answered. However, other steroids have been shown to affect the metabolic processes of lymphocytes (7, 15, 16). Furthermore, the administration of cortisol to rats has been observed to induce the formation of myelin figures in mitochondria (17), suggesting that perhaps all cells affected negatively in their biosynthetic processes by cortisol may similarly exhibit structural alterations of the mitochondria. Although others have noted that severe alterations in the morphology of the nucleus of cul-
tured thymic cells of the rat (18) are produced by similar concentrations of cortisol during the first 2 hr of culture, we have not observed any significant alterations in the nuclear morphology of huma~l lymphocytes, either of the normal or CLL origin. The finding that the mitochondrion is adversely affected by the presence of cortisol in the medium corrdates well with those results of others which illustrate that glucose transport and utilization are decreased by cortisol within 15 rain of incubation. We have noted that, within this period of exposure to a similar concentration
CORTISOL ON HUMAN LYMPHOCYTES
219
FIG. 9. Control CLL lymphocyte. X10,000. of cortisol, severe mitochondrial changes are induced (19). Such alterations as noted in the text would be expected to alter the physical association of those units of the inner mitochondrial membrane involved in oxidative phosphorylation. If this is indeed the case, then the possibility that a shortage of ATP might act as a trigger for the cell to obtain substrate by selfingestion might be real. Self-ingestion might then involve the isolation of a portion of the cytoplasm via the formation of some such vesiclelike structure as was found in the cell in Fig. 4, the structure eventually undergoing constriction.
This constricted structure, when sectioned per pendicularly to its longitudinal axis, might have the appearance of a horseshoe-like myelin figure as seen in Fig. 7. During the formation of this myelin figure, the Golgi might be involved in the formation of vesicles containing packaged hydrolytic enzymes required for the digestion of the segregated cytoplasmic body; hence the reason for its unusually well developed state as depicted in Fig. 3, and especially ill Fig. 8. The relationship of the above ultrastructurat changes promoted by cortisol to the proven presence of intracellular cortisol receptors is not
220
WERTHAMER AND AMARAL
FIG. 10. Cortisol-treated, CLL lymphocyte. X10,000. clear. I t is gelmrally accepted that the effect of cortisol on a lymphoid cell depends upon the existence of specific cortisol receptors within the cytoplasm. Therefore, lymphoid cells sensitive to cortisol may exhibit such receptors, whereas those that did not contain such receptors would be insensitive to the hormone. However, recent studies indicate that certain cell lines containing such receptors are resistant to the steroid (20). Hence, although the effects of the steroid imply the presence of receptors, the presence of receptors is not the only factor necessary for the expression
of sensitivity to the steroid. Inasmuch as we have been able to illustrate that the normal human lymphocyte contains a glucocorticoid receptor practically identical to plasma transcortin (21), and since both this receptor and the level of plasma transcortin (8) are much decreased in the CLL state, the structural alterations produced ill the normal lymphocyte and not in the CLL cell might be comoletely dependent upon the existence of intracellular transcortin. Further studies, some of which are in progress, may determine the validity of the above speculations.
C O R T I S O L ON H U M A N L Y M P H O C Y T E S
FIG. 11. Control R P M I 1788 lymphocyte. X l0,000.
221
222
WERTHAMER AND AMARAL
FIG. 12. Cortisol-treated, RPMI 1788 lymphocyte. X10,000. REFERENCES 1. Nowell, P. C. 1961. Inhibition of human leukocyte mitosis by prednisolone in vitro. Cancer Res. 21: 1518-1521. 2. Hofert, J. F., and A. White. 1968. Effect of a single injection of cortisol on the incorporation of H~-thymidine and H~-deoxycytidine into lymphatic tissue DNA of adrenalectomized rats. Endocrinology 82: 767-776. 3. Rosen, J. M., J. J. Tina, R. J. Milholland, and F. Rosen. 1970. Inhibition of glucose uptake in lymphosarcoms P 1798 by cortisol and its relationship to biosynthesis of deoxyribonucleic acid. J. Biol. Chem. 245: 2074-2080. 4. Makman, M. H., S. Nakagawa, B. Dvorkin, and A. White. 1970. Inhibitory effects of cortisol and antibiotics on substrate entry
5.
6.
7.
8.
and ribonucteic acid synthesis in rat thymocytes in vitro. J. Biol. Chem. 245: 1485-1497. AmarM, L., and S. Werthamer. 1970. Protein and RNA synthesis in human leukocytes and lymphocytes: 3. Cortisol enhanced destruction of newly synthesized RNA of cultured lymphocytes. Life Sci. 9: 661-666. Werthamer, S., B. Pachter, and L. Amaral. 1971. Protein synthesis in human leukocytes and lymphocytes: 2. The effect of cortisol on RNA and protein synthesis in lymphocytes. Life Sci. 10: 1039-1044. Werthamer, S., C. Hicks, and L. Amaral. 1969. Protein synthesis in human leukocytes and lymphocytes: 1. Effect of steroids and sterols. Blood 34: 348-356. Werthamer, S., and L. Amaral. 1971. The re-
CORTISOL ON HUMAN LYMPHOCYTES
9.
10.
11.
12.
13.
14.
sponses of normal and chronic lymphocytic leukemic lymphocytes to cortisol. The possible role of transcortin. Blood 37: 463-472. Munck, A. 1968. Metabolic site and time course of cortisol action on glucose uptake, lactic acid output and glucose 6-phosphate level of rat thymus cells in vitro. J. Biol. Chem. 243 : 1039-1046. Amaral, L., A. J. Samuels, and S. Werthamer. 1971. The transport of cortisol in human lymphocytes normal and leukemic. Experientia 27: 511-512. Rabinowitz, Y. 1964. Separation of lymphocytes, polymorphonuclear leukocytes and monocytes on glass bead columns, including tissue culture observations. Blood 23:811 822. Kodama, T., and M. Kodama. 1972. Ultrastructural alterations in the liver parenchymal cells and thymus lymphocytes following the administration of hydrocortisone. Cancer Res. 32: 208-214. Lundin, P. M., and U. Schelin. 1969. The effect of steroids on the histology and ultrastructure of lymphoid tissue. Pathol. Eur. 4: 58-68. Novikoff, A. B., and E. Essner. 1962. Cytolysosomes and mitochondrial degeneration. J. Cell Biol. 15: 140-155.
223
15. Spangler, A. S., H. N. Antoniades, S. L. Sotman, and T. M. Inderbitizin. 1969. Enhancemerit of the anti-inflammatory action of hydrocortisone by estrogen. J. Clin. Endocrinol. Metab. 29: 650-655. 16. Bodel, P., M. Dillard, S. S. Kaplan, and S. E. Malawista. 1972. Anti-inflammatory effects of estradiol on human leukocytes. J. Lab. Clin. Med. 80:373 384. 17. Bullock, G. R., R. F. Peters, and A. M. White. 1969. Changes in mitochondrial structure and ribosomal activity in muscle as a consequence of the interaction between a glucocorticoid and some anabolic steroids. Biochem. J. 115: 47P. 18. Burton, A. F., J. M. Storr, and W. L. Dunn. 1967. Cytolytic action of corticosteroids on thymus and lymphoma cells in vitro. Can. J. Biochem. 45: 289-297. 19. Unpublished observations. 20. Gehring, U., B. Mohit, and G. M. Tomkins. 1973. Glucocorticoid action on hybrid clones derived from cultured myeloma and lymphoma cell lines. Proc. Natl. Acad. Sci. U.S.A. 69:3124 3127. 21. Werthamer, S., A. J. Samuels, and L. Amaral. 1973. Identification and partial purification of "transcortin-like" protein within human lymphocytes. J. Biol. Chem. 248:6398 6407.
We wish to thank Mr. Alex Fulop and Miss Joanne Gallagher for their excellent assistance, Miss Anita Aschettino for the preparation of the manuscript, and a special thanks to Dr. Charity Waymouth for improving the manuscript by judicious editing. This work was in part supported by Grant ET-47 from the American Cancer Society.
Vol. 11, No. 4, 1975
EFFECT OF CORTISOL ON THE U L T R A S T R U C T U R E OF NORMAL, LEUKEMIC, A N D C U L T U R E D H U M A N L Y M P H O C Y T E S SEYMOUR WEIITHAMER 1 AND LEONARD AMARAL Departments of Pathology, Dowt~state Medical Ce~der, a~d The Methodist Hospital of Brooklyt~, Brooklytt, New York 11215 SUMMARY
The effect of cortisol on the ultrastructure of normal, leukemic, and cultured human lymphocytes during a 2-hr incubation was investigated. The presence of 10-5 M cortisol in the incubation medimn produced in normal lymphocytes a variety of alterations in cytoplasmic organelles. Mitochondria were most affected and showed evidence of irreversible deterioration (formation of myelin figures). Occasional cells demonstrated an overt rearrangement of their cytoplasmic membranes resulting in a bizarre array of parallel cisternae-like structures. More commonly, the usually underdeveloped Golgi of normal lymphocytes became very pronounced in structure. All of these alterations were produced within 2 hr of incubation, but only in normal human lymphocytes. Under identical conditions, no evidence of ultrastructural changes were produced by cortisol in either lymphoeytes from chronic lymI)hocytic leukemic patients, or those from the R P M [ 1788 cell line. The use of biochemical metimdology in studies in vitro has contributed largely to the understanding of the mechanism by which cortisol depresses the lmmbers of circulating lymphocytes of mammals. Such studies have clearly demonstrate(1 that cortisol inhibits the synthesis of lymphoid I ) N A (1-3), R N A (4-6), and protein (7, 8), enhances the degradation of RNA (5, 8), and decreases the utilization of glucose (9). The sequence of events appears to involve first the uptake of the hormone, an event dependent upon cellular energy and regulated in some manner by a plasma component (10). Following uptake, glucose utilization is affected (9), after which the energy-dependent degradation of newly synthesized R N A occurs (8). Inhibition of protein synthesis is evident long after the above alterations occur, and appears to be independent, during the periods of time studied, of the effects of cortisol on the synthesis of R N A (8). Although the above studies have contributed to the partial understanding of the mechanism in question, the precise effect of cortisol on the components of the lymphocyte has remained
largely mLstudied. Because ultimate understanding nmst eventually relate the biochemical functional effects to structure, we have investigated the effect of cortisol ill vitro oil the structure of human lymphocytcs. The results presented indicate that the presence of 10-'~ M cortisol in the culture medium produces alteration ill mitochondrial structure, induces the development of Golgi and, in occasional cells, a reorganization of the endoplasmic reticulum. These structural alterations are not produced either in lymphocytes of patients with chronic lymphocytic leukemia, or in act established lymphocyte cell line, by concentrations of cortisol as high as 10-~ M. MATERIALS AND METHOI)S
Human tymphoeytes, purified by the method of Rabinowitz (11), were obtained from nine healthy t)ersons, suspended in platelet-frce aut(Jlogous plasma, and incubated for 1 to 2 hr in TC 199 medium (l)ifeo), with and without final concentrations of cortisol (hydroeortisone, Sigma) of 10-7 to 10-~ M. A stock solution of cortisol at 10.4 hi was prepared by dissolving the appropriate 1 Send requests for reprints to Dr. Seymour Werthamer, Department of Pathology, Methodist weight of eortisol in a benzene-ethanol solution (10:1). This stock solution was prepared daily Hospital, 50(i Sixth St., Brooklyn, N. Y. 11215. 212
CORTISOL ON HUMAN LYMPHOCYTES and, before use, appropriate aliqucts were transferred to culture tubes. The solutions were dried under a constant stream of nitrogen, after which appropriate volumes of culture medium were added and the dried cortisol was dissolved. In order to ascertain that indeed all of the cortisol had dissolved in the culture medium, separate aliquots of the cortisol stock solution were added to identical culture tubes, after which 0.1 Ci of [3H]cortisol was added. The solutions were dried as above, medium was added, and a sample of 0.1 ml was then assayed for total radioactivity. The results assured us that all of the cortisol had been dissolved and that the concentrations of cortisol noted were those in the medium. Addi-
213
tional details concerning the culture conditions have been prviously presented (7, 8). Following the period of incubation, the cells were centrifuged at 850 • g, the supernatant fluid was discarded and the cells were suspended in ice-cold 2.5% glutaraldehyde for 2 hr, centrifuged at 850 • g, resuspended in 1% osmium tetroxide for 1 5 hr, dehydrated in ascending concentrations of ethanol (three washes in each), embedded in Epon 812, and sectioned with a Reichert ultramicrotome. The sections were stained with uranyl acetate and lead citrate, and studied with a Hitachi HS-8 electron microscope at a voltage of 50 kv. Other human lymphocytes were obtained from 14 patients diagnosed as having
FIG. 1. Control, normal lymphocyte. N, nucleus; G, Golgi; M, mitochondrion; L, lysosome. X 11,000.
214
WEI~TI~AMI,]I( ANI) AMAI,AL
chronic lymphocytic leukemia (CLL, with leukocyte counts in excess of 50,000 per ram3), who at the time were not receiving treatment, a~(l from a human lymphocyte cell line. of G. L. Moore ( R P M I 1788) (purchased fr()m Associate(1 lliomedic ~ystems, Inc., Buffalo, N. Y.), that has been successfully cultured in our laboratory for more than o~le year. RESULTS
The typical ultrastructurc of normal human lymphocytes incubated for either 1 or 2 hr under the control conditions specified in the text is illustrated in Figs. 1 and 2. lCricfly, w(~ll d('fitle(l
mitochon(lria are generally clustered in the vicitfity of tile il~dentation of the nucleus and marginal to the underdeveloped Golgi. The ('ytol)lasm contains a few scattered clefts, vesicles, electron (lel~se bodies (probably lysosomes), all(l scattered ribosomes. The often ilLdente(l mlcleus is usually characterized by dense chromatin (heterochromatin) concentrated e~t it~ periphery. In about half of the se('.tions studied, a well defined ~ucleolus was obacrvcd to be situated slightly off-center. In almost all cases the nucleus was surrom~ded by a ('l~ar peri~uclear spa(~e.
I Jl coHtrast to the above typical ultrastructure,
FIG. 2. Control, normal lymphocyte. X44,000.
CORTISOL ON HUMAN LYMPHOCYTES
215
Fro. 3. Cortisol-treated, normal lymphocyte. AM, altered mitoehondrion; MF, myelin figure; MMF, mitoehondrial myelin figure; G, Golgi. Xll,0C0. only lymphocytes that had been iucubated for a minimum of 2 hr in medium containing l0 -5 M cortisol demonstrated severe alterations in their cytoplasmic organelles (Figs. 3 and 4). Mitochondria appear more angular (Fig. 3), pinched (Fig. 3), or sometimes club-shaped (Fig. 4), and all showed various degrees of disorganized cristae. Myelin figures, which were not observed by us in the control normal lymphocytes, were numerous in the cortisol-treated cells (Fig. 3). 'rhese myelin figures were usually in close proximity to mitochotldria (Fig. 4) or were themselves a part of a migochondrion (Fig. 5). The shape of the myelin figure varied from a sphere (Fig. 6) to a "horse-
shoe" (Fig. 7). Golgi, which is usually not well developed in control normal lymphoeytes, was very pronfinent in about 25% of the cells examined (Figs. 3, 8). In fewer than 5% of the cells examined, the region normally occupied by the Golgi exhibits in its place an array of continuous parallel cytoplasmic membranes (Fig. 4). The electron dense strata between tile stacks of membranes appeared continuous with that of the rest of the cell cytoplasm. In all cases, the cytoplasmic strata of the cortisol-treated cells appeared much more electron dense than those of the control normal lymphoeytes. The above cytoplasmic alterations were less frequent and
216
WERTHAMER AND AMARAL
F~G. 4. Cortisol-treated, normal lymphocyte. C, stacks of cytoplasmic cisternae. X 11,000. much less overt with 1 0 - 6 M of cortisol. At a cortisol concentration of 10-~ M, no significant ultrastructural changes were observed. I n contrast to the above findings, the ultrastructure of the cells of the CLL patient did not appear affected by the presence of eortisol in the medium at concentrations as high as 10-5 M. As is evident from Figs. 9 and 10, no alteration in any cytoplasmic organelle was produced. Lymphoeytes from a total of 14 CLL patients were similarly incubated and all failed to show any of the ultrastructural changes typically produced in normal cells by identical steroid concentrations. Similarly, incubation of human lymphocytes of the cell line R P M I 1788 of Moore with con-
centrations of cortisol as high as 10-5 M failed to yield any noticeable changes in ultrastructure. Fig. 11 depicts the ultrastrueture of control cells and Fig. 12 is that of a R P M I 1788 lymphocyte cultured in the presence of 10-5 M eortisol for 2 hr. DISCUSSION The results obtained in this study suggest that incubating normal human lymphocytes in medium containing 10-~ M eortisol results in gross alterations in the structure of mitoehondria; induces the development of myelin figures, some of which involve mitochondria; increases in some cells the extent of Golgi; and, in a few cells, promotes the reorganization of a F.ortion of the cytoplasmic membranes. These alterations are
CORTISOL ON HUMAN LYMPHOCYTES
217
:FIG. 5. High magnification of a myelin figure of a mitochondrion. X40,260.
Fla. 6. High magnification of a myelin figure. X91,500.
produced within a 2-hr period of incubation al~d are significantly evident only with a minimum concentration of 10-~ M cf corti~ol. In contrast to these effects, CLL lymphocytes, previously proven to be more sensiti~ e to depression of RNA and protein synthesis by cortisol than normal cells (8), fail to exhibit any noticeable changes in their ultrastructure. Lymphocytes of the R P M I 1788 cell line are also apparently not affected. These findings suggest that, although both the normal and the CLL lymphocyte are biochemically affected by cortisol, the latter being more sensitive, the mechanisms whereby metabolic synthesis are affected might in part be different for the two types of cells. The alterations observed in the normal lymphocyte, particularly those of the mitochondria, have been previously observed by others in mouse thymic lymphoeytes (12) and in rat thymic epithelial cells (13). In all cases the alterations were consistent with various stages of cell degeneration and consonant with processes denoting an ongoing autophagic process (14). Whether degenerative or autophagic processes in lymphoid cells are specifically induced by cortisol,
F I 6 . 7 . High nlagnification of a horseshoe-like myelin figure. X50,325.
2]8
WERTHAMER AND AMARAL
FI(~. 8. Cortisol-treated, normal lymphocyte. Note extensive Golgi. X ll,0C0. or whether other steroids can similarly effect changes, is a question which has not yet been answered. However, other steroids have been shown to affect the metabolic processes of lymphocytes (7, 15, 16). Furthermore, the administration of cortisol to rats has been observed to induce the formation of myelin figures in mitochondria (17), suggesting that perhaps all cells affected negatively in their biosynthetic processes by cortisol may similarly exhibit structural alterations of the mitochondria. Although others have noted that severe alterations in the morphology of the nucleus of cul-
tured thymic cells of the rat (18) are produced by similar concentrations of cortisol during the first 2 hr of culture, we have not observed any significant alterations in the nuclear morphology of huma~l lymphocytes, either of the normal or CLL origin. The finding that the mitochondrion is adversely affected by the presence of cortisol in the medium corrdates well with those results of others which illustrate that glucose transport and utilization are decreased by cortisol within 15 rain of incubation. We have noted that, within this period of exposure to a similar concentration
CORTISOL ON HUMAN LYMPHOCYTES
219
FIG. 9. Control CLL lymphocyte. X10,000. of cortisol, severe mitochondrial changes are induced (19). Such alterations as noted in the text would be expected to alter the physical association of those units of the inner mitochondrial membrane involved in oxidative phosphorylation. If this is indeed the case, then the possibility that a shortage of ATP might act as a trigger for the cell to obtain substrate by selfingestion might be real. Self-ingestion might then involve the isolation of a portion of the cytoplasm via the formation of some such vesiclelike structure as was found in the cell in Fig. 4, the structure eventually undergoing constriction.
This constricted structure, when sectioned per pendicularly to its longitudinal axis, might have the appearance of a horseshoe-like myelin figure as seen in Fig. 7. During the formation of this myelin figure, the Golgi might be involved in the formation of vesicles containing packaged hydrolytic enzymes required for the digestion of the segregated cytoplasmic body; hence the reason for its unusually well developed state as depicted in Fig. 3, and especially ill Fig. 8. The relationship of the above ultrastructurat changes promoted by cortisol to the proven presence of intracellular cortisol receptors is not
220
WERTHAMER AND AMARAL
FIG. 10. Cortisol-treated, CLL lymphocyte. X10,000. clear. I t is gelmrally accepted that the effect of cortisol on a lymphoid cell depends upon the existence of specific cortisol receptors within the cytoplasm. Therefore, lymphoid cells sensitive to cortisol may exhibit such receptors, whereas those that did not contain such receptors would be insensitive to the hormone. However, recent studies indicate that certain cell lines containing such receptors are resistant to the steroid (20). Hence, although the effects of the steroid imply the presence of receptors, the presence of receptors is not the only factor necessary for the expression
of sensitivity to the steroid. Inasmuch as we have been able to illustrate that the normal human lymphocyte contains a glucocorticoid receptor practically identical to plasma transcortin (21), and since both this receptor and the level of plasma transcortin (8) are much decreased in the CLL state, the structural alterations produced ill the normal lymphocyte and not in the CLL cell might be comoletely dependent upon the existence of intracellular transcortin. Further studies, some of which are in progress, may determine the validity of the above speculations.
C O R T I S O L ON H U M A N L Y M P H O C Y T E S
FIG. 11. Control R P M I 1788 lymphocyte. X l0,000.
221
222
WERTHAMER AND AMARAL
FIG. 12. Cortisol-treated, RPMI 1788 lymphocyte. X10,000. REFERENCES 1. Nowell, P. C. 1961. Inhibition of human leukocyte mitosis by prednisolone in vitro. Cancer Res. 21: 1518-1521. 2. Hofert, J. F., and A. White. 1968. Effect of a single injection of cortisol on the incorporation of H~-thymidine and H~-deoxycytidine into lymphatic tissue DNA of adrenalectomized rats. Endocrinology 82: 767-776. 3. Rosen, J. M., J. J. Tina, R. J. Milholland, and F. Rosen. 1970. Inhibition of glucose uptake in lymphosarcoms P 1798 by cortisol and its relationship to biosynthesis of deoxyribonucleic acid. J. Biol. Chem. 245: 2074-2080. 4. Makman, M. H., S. Nakagawa, B. Dvorkin, and A. White. 1970. Inhibitory effects of cortisol and antibiotics on substrate entry
5.
6.
7.
8.
and ribonucteic acid synthesis in rat thymocytes in vitro. J. Biol. Chem. 245: 1485-1497. AmarM, L., and S. Werthamer. 1970. Protein and RNA synthesis in human leukocytes and lymphocytes: 3. Cortisol enhanced destruction of newly synthesized RNA of cultured lymphocytes. Life Sci. 9: 661-666. Werthamer, S., B. Pachter, and L. Amaral. 1971. Protein synthesis in human leukocytes and lymphocytes: 2. The effect of cortisol on RNA and protein synthesis in lymphocytes. Life Sci. 10: 1039-1044. Werthamer, S., C. Hicks, and L. Amaral. 1969. Protein synthesis in human leukocytes and lymphocytes: 1. Effect of steroids and sterols. Blood 34: 348-356. Werthamer, S., and L. Amaral. 1971. The re-
CORTISOL ON HUMAN LYMPHOCYTES
9.
10.
11.
12.
13.
14.
sponses of normal and chronic lymphocytic leukemic lymphocytes to cortisol. The possible role of transcortin. Blood 37: 463-472. Munck, A. 1968. Metabolic site and time course of cortisol action on glucose uptake, lactic acid output and glucose 6-phosphate level of rat thymus cells in vitro. J. Biol. Chem. 243 : 1039-1046. Amaral, L., A. J. Samuels, and S. Werthamer. 1971. The transport of cortisol in human lymphocytes normal and leukemic. Experientia 27: 511-512. Rabinowitz, Y. 1964. Separation of lymphocytes, polymorphonuclear leukocytes and monocytes on glass bead columns, including tissue culture observations. Blood 23:811 822. Kodama, T., and M. Kodama. 1972. Ultrastructural alterations in the liver parenchymal cells and thymus lymphocytes following the administration of hydrocortisone. Cancer Res. 32: 208-214. Lundin, P. M., and U. Schelin. 1969. The effect of steroids on the histology and ultrastructure of lymphoid tissue. Pathol. Eur. 4: 58-68. Novikoff, A. B., and E. Essner. 1962. Cytolysosomes and mitochondrial degeneration. J. Cell Biol. 15: 140-155.
223
15. Spangler, A. S., H. N. Antoniades, S. L. Sotman, and T. M. Inderbitizin. 1969. Enhancemerit of the anti-inflammatory action of hydrocortisone by estrogen. J. Clin. Endocrinol. Metab. 29: 650-655. 16. Bodel, P., M. Dillard, S. S. Kaplan, and S. E. Malawista. 1972. Anti-inflammatory effects of estradiol on human leukocytes. J. Lab. Clin. Med. 80:373 384. 17. Bullock, G. R., R. F. Peters, and A. M. White. 1969. Changes in mitochondrial structure and ribosomal activity in muscle as a consequence of the interaction between a glucocorticoid and some anabolic steroids. Biochem. J. 115: 47P. 18. Burton, A. F., J. M. Storr, and W. L. Dunn. 1967. Cytolytic action of corticosteroids on thymus and lymphoma cells in vitro. Can. J. Biochem. 45: 289-297. 19. Unpublished observations. 20. Gehring, U., B. Mohit, and G. M. Tomkins. 1973. Glucocorticoid action on hybrid clones derived from cultured myeloma and lymphoma cell lines. Proc. Natl. Acad. Sci. U.S.A. 69:3124 3127. 21. Werthamer, S., A. J. Samuels, and L. Amaral. 1973. Identification and partial purification of "transcortin-like" protein within human lymphocytes. J. Biol. Chem. 248:6398 6407.
We wish to thank Mr. Alex Fulop and Miss Joanne Gallagher for their excellent assistance, Miss Anita Aschettino for the preparation of the manuscript, and a special thanks to Dr. Charity Waymouth for improving the manuscript by judicious editing. This work was in part supported by Grant ET-47 from the American Cancer Society.
