- Email: [email protected]
VULNERABILITY OF CELL-SURFACE RECEPTORS TO AUTOIMMUNE REACTIONS
684 of 44%). No untoward clinical or biological side-effects of any kind were encountered in any of the patients during these experiments or in the following days. When abnormal involuntary movements were previously the custom after a dose of levodopa, they appeared with the same, but not greater, intensity. All 6 patients during the P.L.G. phase experienced elation over their ability to perform much better even complicated acts. They all reported that their minds were clearer and functioning "as in the old days". This was objectivated by a 53% average reduction in the Hamilton score and a 40% improvement in the Kohs’ block design score. All three cardinal symptoms of akinesia, rigidity, and tremor improved, to variable degrees, in each patients. DISCUSSION
study, carried out as a preliminary to a much detailed protocol with intravenous P.I..G. (Abbott-40509), indicates significant potentiating effects of this tripeptide upon the action of levodopa in man. The improvement obtained in almost all (5/6) cases was of such magnitude that it far surpassed the clinical effect of any of the numerous antiparkinson drugs which we have tested in our laboratory over the past 15 years, including levodopa alone. These observations entirely confirm for man the findings in animals3-g and justify much-expanded research into the mechanism by which M.I.F. acts on the central nervous system. Unpublished observations from our laboratory failed to reveal changes in the level of biogenic amines in the brain of animals given P.L.G. with M.I.F. activity. Similarly we could find no effect on monoamine-oxidase of catechol0-methyl transferase activity. Furthermore P.L.G. does not seem to release the catecholamines in an amphetamine-like manner.4 8 The most likely site of action, if it is synaptic, is the post-synaptic membrane. In favour of this hypothesis is the fact that M.I.F. greatly potentiates apomorphine-induced hyperactivity in intact as well as in hypothalamus-lesioned rats.19 However, many other possibilities must still be considered and tested. It is indeed of interest that serum M.s.H. levels were found elevated in Parkinson’s disease 20 and that administration of M.S.H. aggravated the symptoms of that illr.ess.2’ Could Parkinson’s disease be characterised by deficient hypothalamic M.I.F. production which in turn, through unhampered central M.S.H. activity, could influence the functioning of the melanocyte-like pigmented cells in the substantia nigra and locus coeruleus, resulting in a deranged production of catecholamines? Such hypotheses are being tested in our laboratory. These experiments, while encouraging, do not yet prove that P.L.G. improves motor performance or even intellectual function in Parkinson’s disease. We are still ignorant of the long-term efficacy, or toxicity, of this substance and realise that continuous intravenous therapy, in any disease entity, is no final solution.
Hypothesis VULNERABILITY OF CELL-SURFACE RECEPTORS TO AUTOIMMUNE REACTIONS P. R. CARNEGIE Russell Grimwade School of Biochemistry, University of Melbourne, Parkville, Victoria 3052, Australia
I. R. MACKAY Clinical Research Unit, Walter and Eliza Hall Institute of Medical Research, Post Office, Royal Melbourne Hospital, Victoria 3050, Australia
This
more
We thank Abbott Laboratories for supply of p.L.G. (A-40509); Dr. J. Kastin (New Orleans) for advice and encouragement; Miss D. Bedard, Mtss L. Paul, and Miss C. Larouche for helping in the experiments ; and the United Parkinson Foundation and the W. Garfield Weston Foundation for financial support. Requests for reprints should be addressed to A. B., Department of Neurobiology, Clinical Research Institute of Montreal, 110 Pme Avenue West, Montreal, Quebec, Canada H2W lR7. A.
A previous hypothesis in which myasthenia gravis was explained by an immune response to acetylcholine receptors has been validated, and is here extended to cell receptors in general.
Summary
Receptors on target cells, being accessible to circulating trophic hormones or transmitters, must also be accessible to antibodies which compete with the natural mediator for access to the site. To detect anti-receptor antibodies, physiological assay systems would be more sensitive than conventional immunological assays. Autoimmune responses to receptor sites would require a genetic predisposition to failure of immunological tolerance such as occurs in various autoimmune diseases. This hypothesis is supported by recent findings in hyperthyroidism and a type of insulin-resistant diabetes mellitus, and is applicable to other endocrinopathies, diseases in which dysfunction at receptor sites can be postulated, and regulatory functions within the immune system itself.
DR BARBEAU: REFERENCES A. J., Barbeau, A , Ehrensing, R. H., Plotnikoff, N., Schally, A V. Adv Neurol. 1974, 5, 225. 2. Nair, R. M. G., Kastin, A. J., Schally, A. V. Biochem. Biophys. Res Commun. 1971, 43, 1376. 3 Plotnikoff, N. P., Kastin, A. J., Anderson, M. S , Schally, A. V Life Sci. 1971, 10, 1279. 4. Kastin, A. J., Miller, M. C., Ferrell, L., Schally, A. V. Physiol Behav. 1973, 1.
Kastin,
10, 399 Plotnikoff, N. P., Minard, F. N., Kastin, A. J. Neuroendocrinology, 1974, 14, 271 6. Huibodro-Toro, J. P., Scotti de Carolis, A., Longo, V. G. Pharmac. Biochem. Behav. 1974, 2, 105. 7. Plotnikoff, N P , Kastin, A. J. Archs int. Pharmacodyn. Thér 1974, 211, 5.
211. 8.
Huibodro-Toro, J. P., Scotti de Carolis, A., Longo, V G Pharmac chem. Behav. 1975, 3, 235. Kastin, A. J., Barbeau, A. Can. med. Ass. J 1972, 107, 1079.
Bio-
9 10 Woods, A. C., Chase, T. N. Lancet, 1973, ii, 513. 11. Chase, T. N., Woods, A. C., Lipton, M. A., Morris, C E. Archs Neurol., Chicago, 1974, 31, 55. 12. Fischer, F. A., Schneider, E., Jacobi, P., Maxion, H. Eur Neurol 1974, 12, 360. 13. Barbeau, A., Roy, M. Can. med. Ass. J. (in the press). 14. Yahr, M. D., Duvoisin, R. C., Shear, M. J., Barrett, R. E., Hœhn, M.Archs Neurol., Chicago, 1969, 21, 343. 15. Dery, J. P., De Groot, J. A., Laurin, C., Barbeau, A. Un. méd. Can. 1962,
91, 842. Barbeau, A. Can. med. Ass J 1969, 101, 791. Hamilton, M. J. Neurol. Neurosurg. Psychiat. 1960, 23, 56. Botez, M. I., Barbeau, A. Int. J. Neurol. 1975, 10, 222. Burnett, C., Strother, E., Butterworth, R. F., Barbeau, A Unpublished Shuster, S., Burton, J. L., Thody, A. J., Plummer, N., Goolamali, S. k., Bates, D. Lancet, 1973, i, 463. 21. Cotzias, G C., Van Woert, M. H., Schiffer, L. M. New Engl. J Med. 1967,
16. 17. 18. 19 20.
276, 374.
685 INTRODUCTION
hypothesis "immunopharmacological block" of Lennon and Carnegie’ has been validated by observations on anti-receptor antibody in clinical and experimental myasthenia gravis. Various other diseases could be accounted for by a pathogenic immune response to cell-surface receptors which, being accessible to circulating hormones or neurotransmitters, should also be accessible to antibody. This type of autoimmune response would be detected best by sensitive physiological systems in which antibody would compete with hormone for access to receptor on responsive cell preparations. The. genetic factors involved in regulation of immune responses would render certain individuals more susceptible to a break in tolerance with formation of anti-receptor antibody-e.g., cross-reaction between a component of an infecting microorganism and an antigenic determinant at the receptor site. Also anti-receptor autoantibodies would need to be of high affinity for accessible components of the receptor site. THE
of
choline receptor site13-16 and a cellular response was also shown.17 Such anti-receptor antibodies could also interfere with function at other acetylcholine receptor sites-e.g., in the central nervous system, provided the blood/brain barrier was breached. The earlier hypothesis implicated genetic factors in failure of tolerance to receptor sites. The histocompatibility antigen HL-A8 is associated with that type of myasthenia gravis in young women who lack the characteristic myoid antibody demonstrable by immunofluorescence,18 whereas such patients do have anti-receptor antibody;’6 HL-A2 is associated with the other group of myasthenic patients who are elderly, and have thymoma and myoid antibody’9 and also anti-receptor antibody.16 Thus, the features of myasthenia gravis are well accounted for by block of receptor sites, and the experimental system shown in the figure provides a model protocol for examining various diseases possibly involving neurotransmitters or endocrine dysfunction. GRAVES’
MYASTHENIA GRAVIS
Simpson2 in 1960 proposed that in myasthenia gravis a blocking antibody competed with acetylcholine for the receptor on the motor end-plate. This proposal was not neglected’but could not be substantiated until the fortuitous recognition by Patrick and Lindstrom45 in 1973 that induction of antibody in rabbits to acetylcholine receptors from Electrophorus electricus resulted in an autoimmune response against the rabbits’ motor endplates and myasthenia. This model of myasthenia gravis (see accompanying figure) was widely confirmed by studies in rabbits, rats, guineapigs, and monkeys.6-12 Not all the anti-receptor antibodies appear to be directed at the precise site binding acetylcholine.11 Most patients with myasthenia gravis have humoral antibodies to acetyl-
DISEASE
Hyperthyroidism in Graves’ disease exemplifies the need for physiological or receptor assay systems to demonstrate functionally important autoantibodies against receptors. Adams and Purves20 in 1956 likewise made a fortuitous discovery of an anti-receptor antibody, the long-acting thyroid stimulator (L.A.T.S.), later found to be an immunoglobulin.21 Serum from patients with Graves’ disease may contain, as well as L.A.T.S. which stimulates mouse thyroid cells, other immunoglobulins (L.A.T.s. protector) which stimulate human but not mouse thyroid cells.22 23 These immunoglobulins both compete with and mimic the action of thyroidstimulating hormone (T.s.H.), but in an uncontrolled way. Immunoglobulins from patients with Graves’ disease inhibit the binding of T.S.H. to thyroid membranes,24-27 and L.A.T.s. and the immunoglobulin which competes with T.s.H. in the receptor assay system are considered to be one and the same. These sensitive physiological assay systems, which have been devised for detecting antibodies against the T.s.H. receptor, could prove to be a simple model for the investigation of anti-receptor activity in other endocrine diseases. DIABETES MELLITUS
Experimental myasthenia gravis in rabbits as an example of a disease by autoantibodies to a receptor site. : tchuhne Ach) receptor was isolated from affinity chromatography and - --J ,o! mmunise rabbits.’ In addition to antibody against eel Ach receptor, ann- -’ developed to the rabbits’ own Ach receptors causing myasthema 4 Pa-
caused
mtasthenta gravis have autoanttbodies to determinants III the human receptor which interfere with neuromuscular transmtssion "’ The ’:.ng agent in man, equivalent to immunisation in the rabbit, is unknown. rrm Carncgie." reproduced by permission of the Editors of Search ,c:, ’.11th
- 1_hultne
Among the endocrine disorders, diabetes mellitus is an obvious candidate for the application of this hypothesis. There are two types of diabetes-the "juvenile-onset" type in which there is a low serum level of insulin, indicative of 3-islet cell dysfunction, and an "adult-onset" type in which there are normal or elevated levels of insulin but an apparent resistance to its action, indicative of a refractory target cell. The juvenile-onset type is associated with autoantibodies to thyroid and gastric antigens28 29 and to pancreatic islet cells3O-32 and insulinoma cells.33 Our hypothesis could be tested, firstly, by in-vitro assay systems developed for assaying the function of isolated 3-islet cells34 and isolated fat cells,35 and secondly, by use of isolated insulin receptors36 in experiments similar to those used in Graves’ disease and mvasthenia gravis. In juvenile-onset diabetes, sera which contained anti 3-islet cell antibodies should contain antibodies which block the response of the isolated 3-islet cells to glucose, in a manner mimicking the pharmacological
686
action of the polypeptide somatostatin;34 these sera should not inhibit binding of iodinated insulin to isolated insulin receptor, nor block the action of insulin using the isolated fat-cell assay. In adult-onset diabetes, the hypothesis would predict the reverse: no influence of sera using the isolated p-islet-cell assay, but a competitive block of the insulin receptor on fat cells. Thus laboratories studying these physiological assay systems could readily test the hypothesis, and indeed Flier et al. 37 found that 3 of 6 patients with extreme insulin resistance had antibodies against insulin receptors on human monocytes. The competitive binding assay which they used did not detect anti-receptor antibody in the more usual type of insulin-resistant diabetic; with such patients a more sensitive assay would be the isolated fat cell. Fat cells respond to insulin by increasing uptake of glucose ;35 antibody to the insulin receptor would be expected to block this response to insulin. POSSIBLE ANTI-RECEPTOR DISEASES
endocrinopathies exhibit features expected in disease due to antibody to receptor sites: failure of the target organ possibly but not necessarily associated with autoantibodies shown by immunofluorescence, and high levels of the trophic hormone but resistance to its action. We suggest that antibody will be found to various receptors : as examples, receptors for A.c.T.H. in "autoimmune" Addison’s disease,38 for parathyroid hormone in pseudohypoparathyroidism,39 and for gastrin in pernicious ansemia. 40 Of relevance is the temporary dysfunction of Other
receptors in newborn children of mothers who have myastheniathyrotoxicosis,41 or diabetes mellitus.41 42 We attribute this to transplacental transfer of maternal anti-receptor antibody, virtually established in the case of neonatal myasthenia and thyrotoxicosis, and now pos-. tulated for transient neonatal diabetes, and for infants of low birth-weight despite high growth-hormone levels.43 In testing the above ideas, human receptors, preferably on viable cells, would probably be more specific and sensitive than animal receptors for the assay of anti-receptor antibodies. Fluctuating symptoms in multiple sclerosis could likewise be explained by the hypothesis. 1 44 In vitro, immunoglobulins from patients block electrical activity in cultured neural tissue;45 and a block of receptors in vivo is suggested by the ability of precursors of neurotransmitters to prevent rapid deterioration in optic-nerve functions induced by hyperthermia.46 47 Recently it was found that some 50% of patients with multiple sclerosis had an immune response to oligodendroglial cells.48 Because of the interaction between the encephalitogenic protein and serotonin49 and the presence of serotonin receptors on oligodendrocytes,50 we postulate an immune response to such receptors in multiple sclerosis. In "biochemical neuropathies" possibly exemplified by schizophrenia, 51 a pharmacological agent used thera-
peutically (phenothiazine) appears to overcome a specific dysfunction of a dopamine receptor, although this dysfunction is not demonstrably due to anti-receptor antibody. However, the earlier data of Heathsz could be reconsidered in that the ability of antibody to interfere with receptor function in the absence of an inflammatory response answers the argument that "the absence of conspicuous tissue reaction in schizophrenia does not seem favourable to an autoimmune origin".51
LYMPHOCYTE RECEPTORS FOR ANTIGEN
Lymphocytes have specific receptors for antigen and increasing evidence that antibody against the specific antigen-binding site in immunoglobulin molecules could function against the corresponding specific antigen receptors on B and possibly T lymphocytes.53 Ramseier’s54 experiments indicate that such anti-idiotypic antibodies can regulate the immune response to histocompatibility antigens, and Strayer et al.’s" experiments on the induction of neonatal tolerance bv antibody to the idiotype reacting with phosphorylcholine support Jerne’s postulate that the immune system could be self-regulatory through a "network"56 57 based in part on idiotypic and anti-idiotypic antibodies. Anti-allotype antibodies could also interfere with the immune response to an antigen, especially if directed against the determinants close to the specific antigen combining site. Dray58 has described how lymphoid cells with a genetic capability of expressing on their surface an immunoglobulin phenotype (allotype) can be blocked by exposure to the appropriate anti-allotype antiserum.
there is
extension of this concept to cancer, antibodies membrane antigens of melanoma could be inagainst volved in localising the tumour; an. anti-idiotypic antibody formed against the antibody to membrane antigens was claimed to result in the blocking effect on tumour In
an
immunity. 59 Returning
to
autoimmune diseases with
fluctuating
clinical symptoms, as exemplified by lupus erythematosus and myasthenia gravis, there could be temporary ascendency of a clone of "suppressor" lymphocytes with production of anti-idiotypic antibody, either to the
pathogenic antibody or to the lymphocyte receptor for auto antigen, thus reducing the pathogenic clones with consequent remission of disease. Finally, there are therapeutic implications, in that diseases which are currently being treated by regular injection of hormone could possibly be treated by manipulation of the immune response to restore immunological tolerance to receptor sites. We thank many colleagues for discussion and data, particularly Dr V. A. Lennon, Dr J. M. Lindstrom, Dr G. F. Mitchell, Dr I. J. Wood. Dr D. D. Adams and Dr M. T. McQuillan. Support by the National Multiple Sclerosis Society, New York, and the National Health and Medical Research Council of Australia is acknowledged.
REFERENCES 1. Lennon, V. A., Carnegie, P. R. Lancet, 1971, i, 630. 2. Simpson, J. A. Scott. med. J. 1960, 5, 419. 3. Mackay, I. R., Burnet, F. M. Autoimmune Diseases. Springfield, Illinois, 1963. 4. Patrick, J., Lindstrom, J. M. Science, 1973, 180, 871. 5. Patrick, J., Lindstrom, J. M., Clup, B., McMillan, J. Proc. natn. Acad. Sci. U.S.A 1973, 70, 3334. 6. Sugiyama, H., Benda, P., Meunier, J-C., Changeux, J-P. FEBS Letters, 1973, 35, 124. 7. Heilbron, E., Mattson, C. J. Neurochem, 1974, 22, 315. 8. Heilbron, E., Mattson, C., Stalberg, E., Hilton-Brown, P. J. neurolog. Sci.
10.
1975, 24, 59. Hazdai, R., Aharonov, A., Silman, I., Fuchs, S., Abramsky, O. Nature, 1975, 256, 128. Lennon, V. A., Lindstrom, J. M., Seybold, M. E. J. exp. Med. 1975, 141,
11.
Lennon, V. A., Lindstrom, J. M., Seybold, M. E. Ann. N Y Acad Sci.
9.
1365. the 12. 13. 14. 15.
(in
press).
Green, D. P. L., Miledi, R., Vincent, A. Proc. R. Soc. B. 1975, 189, 57. Almon, R. R., Andrew, C. G., Appel, S. H. Science, 1974, 186, 55. Almon, R. R., Appel, S. H. Biochem. biophys. Acta, 1975, 393, 66. Bender, A. N., Ringle, S. P., Engel, W. K., Daniels, M. P, Vogel. Z Lancet, 1975, i, 607. 16. Lindstrom, J. M., Lennon, Seybold, M. E., Whittingham, S. Ann. N. Y. Acad. Sci. (in the press).
687
KAPOSI’S SARCOMA: A BYPRODUCT OF TUMOUR REJECTION T. F. C. S. WARNER
progressive growth of the virally transformed mative mesenchyme.
S. O’LOUGHLIN
Mayo Clinic, Rochester, Minnesota 55901, U.S.A. It is suggested that Kaposi’s sarcoma is the result of a chronic immunological reaction between antigenically altered or transformed lymphoid cells and normal lymphocytes. In the course of this local graft-versus-host type activity, an angiogenesis factor is liberated and intense proliferation of mesenchymal and endothelial cells ensues. During the G.V.H.-like activity, an oncogenic virus is either transferred to or induced in the cells responsive to the angiogenesis factor. Thus, the stage is set for neoplastic transformation of these cells in an environment which is conducive to the
Summary
vasofor-
INTRODUCTION sarcoma is an interesting and important in tumour many respects. Like Burkitt’s lymphoma, it is relatively rare outside the African continent where in certain regions it accounts for nearly 10% of all tumours.In non-African patients an unusually high incidence of lymphomas has been observed2 and the disease has also been associated with immunosuppression in homograft recipients.3 The demonstration of herpestype viruses in explants of Kaposi’s sarcoma in tissueculture may establish this tumour as another useful model for the study of viral involvement in human
Kaposi’s
malignancy.44
There has been much controversy about the exact tiscell of origin of Kaposi’s sarcoma. Despite this, most would now concede that the neoplasm arises in cells of reticuloendothelial origin or, more specifically, vasoformative mesenchymal cells, and Hashimoto and Lever56 are of the opinion that it is a neoplasm of periS. Clin. Immun. O., Fuchs, 17. Abramshy, Aharonov, A., Webb, C., exp. 1975, cytes and endothelial cells. However, we propose that 19, 11. the initial event in the histogenesis of Kaposi’s sarcoma 18. Feltkamp, T. E. W., Van den Berg-Loonen, P. M., Nijenhuis, L. E., Engleis not direct transformation of vasoformative cells by an friet, C. P., Van Rossum, A. L., Van Loghem, J. J., Oosterhuis, H. J. G. H Br. med. J. 1974, i, 131. oncogenic agent (? virus) but that the neoplasm is a 19. Fritze, D., Herrman, C., Naeium, F., Smith, G. S., Walford, R. L. Lancet, byproduct of tumour rejection. 1974, i, 240. 20 Adams, D. D., Purves, H. D. Proc. Univ. Otago medœ. Sch. 1956, 34, 11. Dorffel7 has recognised several stages during tumour 21 Kriss, J. P., Pleshakov, V., Chain, J. R. J. clin. Endocr. 1964, 24, 1005. development. Briefly these are: (1) an inflammatory 22. Adams, D. D., Kennedy, T. H. ibid. 1967, 27, 173. 23. Shishiba, Y., Shimizu, T., Yoshimura, S., Shizume, K. ibid. 1973, 36, 517. lesion, (2) a phase of capillary proliferation, (3) an 24. Manley, S. W., Bourke, J. R., Hawker, R. W. J. Endocr. 1974, 61, 419. angiomatous phase, and (4) an angiosarcomatous stage 25 Manley, S. W., Bourke, J. R., Hawker, R. W. ibid. 437. in which malignant spindle cells proliferate. There is 26. Mehdi, S. Q., Nussey, S. S., Gibbons, C. P., El Kabir, D. J. Biochem. Soc. Trans. 1973, 1, 1005. much overlap and a multiplicity of histological changes 27. Smith, B. R., Hall, R. Lancet, 1974, ii, 427. is necessary for a definite diagnosis.8 Montgomery9 has 28. Ungar, B., Stocks, A. E., Martin, F. I. R., Whittingham, S., Mackay, I. R. ibid. 1968, ii, 415. emphasised the mononuclear inflammatory component 29. Irvine, W. J., Clarke, B. F., Scarth, L., Cullen, D. R., Duncan, L. J. P. ibid. which consists of lymphocytoid cells, lymphocytes, his1970, ii, 163. 30. Bottazo, G. F., Florin-Christensen, A., Doniach, D. ibid. 1974, ii, 1279. tiocytes, and plasma cells. The lymphocytoid cells of 31. Lendrum, R., Walker, G., Gamble, D. R. ibid. 1975, i, 880. Marchand7 probably represent lymphocytes which are 32 MacCuish, A. C., Barnes, E. W., Irvine, W. J., Duncan, L. J. P. ibid. 1974, responding to a specific mitogenic stimulus. Thus, this ii, 1529. 33 Maclaren, N. K., Huang, S-W., Fogh, J. ibid. 1975, i, 997. early lesion could represent a local reaction on the part 34 Okamoto, H., Noto, Y., Miyamoto, S., Mabuchi, H., Takeda, R. FEBS Letof the host to transformed autochthonous lymphocytes ters, 1975, 54, 103. 35 Lockwood, D. H., Livingston, J. N., Amatruda, J. M. Fedn. Proc. 1975, 34, -i.e., a host-versus "graft" or a "graft" versus-host 1564. response. Indeed as tumours of the reticuloendothelial 36. Cuatrecasas, P., Parikh, I. Meth. Enzym. 1974, 34, 653. 37 Flier, J. S., Kahn, C. R., Roth, J., Bar, R. S. Science (in the press). system, Kaposi’s sarcoma and Burkitt’s lymphoma are 38. Bresser, G. M., Cullen, D. R., Irvine, W. J., Ratcliffe, J. G., Landon, J. Br. peculiar in that primary tumours are commonly found med. J. 1971, i, 374. in sites which possess a sparse lymphoid-tissue com39 Chase, L. R., Melson, L., Aurbach, G. D. J. clin. Invest. 1969, 48, 1832. 40 Strickland, R. G., Mackay, I. R. Am. J. dig. Dis. 1973, 18, 426. ponent. Furthermore, lymphoma-like changes have been 41 Scott, J. S. Br. med. J. 1966, i, 1559. described in lymph-nodes of patients with Kaposi’s sar42. Cornblath, M., Schwartz, R. Disorders of Carbohydrate Metabolism in Incoma 10 and in a few cases sarcomatous changes merged fancy; p. 57. Philadelphia, 1966. 43 Humbert, J. R., Gotlin, R. W. Pediatrics, Springfield, 1971, 48, 190. with those of lymphoma.11-14 A high incidence of lym44 Carnegie, P. R. Nature, 1971, 229, 25. 45 Bornstein, M., Crain, S. M. Science, 1965, 148, 1242. phoma, especially Hodgkin’s disease, has been found in 46 Harrer, G. Germ. Med. 1973, 3, 116. non-African patients.2 Such an impressive association 47 Harrer, G., Fishbach, R. J. neural Trans. 1973, 34, 205. has not been encountered in African cases, and this may 48. Myers, L. W., Ellison, G. W., Fewster, M. E., Wolfgram, F. Archs. Neurol. 1975, 32, 354. be due to involvement of a younger age-group and a 49 Carnegie, P. R., Smythies, J. R., Caspary, E. A. Field, E. J. Nature, 1972, greater number of fatal cases.15 Burkittl6 has cited addi240, 561. 50 Murray, M. R. In Biology of Neuroglia (edited by W. F. Windle); p. 176. tional similarities between Kaposi’s sarcoma and BurSpringfield, Illinois, 1958. kitt’s lymphoma including long-term survival, recurrence 51 Matthysse, S. , Lipinski, J. A. Rev. Med. 1975, 26, 551. at a different site, and the poor prognosis when lymph52 Heath, R. G., Krupp, I. M. Archs gen. Psychiat. 1967, 16, 1. 53 Eichmann,K., Eur.J. Immun. 1975, 5, 511. nodes are involved. In addition, the coexistence of a pro54 Ramseier, H. Curr. Top. Microbiol. Immun. 1973, 60, 31. 55 Straver, D. S, Cosenza, H., Lee, W. M. F., Rowley, D. A., Kohler, H. gressive growth in one site with a regressive tumour elseScience, 1974, 186, 640. where has been recorded in both diseases but has not sue or
56 Jerne, N. K. Scient. Am. 1973, 229, no. 1, p. 52. 57 Richter, P J Eur J. Immun. 1975, 5, 350. 58. Dray, S. in Ontogeny of Acquired Immunity (Ciba Foundation Symposium); p 87 Amsterdam, 1972. 59. Hartman, D., Lewis, M. G., Proctor, J. W., Lyons, H. Lancet, 1974, ii, 1481. Carnegie, P. R. Search, 1974, 5, 472.
been
satisfactorily explained. 16 GRAFT-VERSUS-HOST DISEASE
Experimentally induced graft-versus-host (G.v.H.)
dis-
study, carried out as a preliminary to a much detailed protocol with intravenous P.I..G. (Abbott-40509), indicates significant potentiating effects of this tripeptide upon the action of levodopa in man. The improvement obtained in almost all (5/6) cases was of such magnitude that it far surpassed the clinical effect of any of the numerous antiparkinson drugs which we have tested in our laboratory over the past 15 years, including levodopa alone. These observations entirely confirm for man the findings in animals3-g and justify much-expanded research into the mechanism by which M.I.F. acts on the central nervous system. Unpublished observations from our laboratory failed to reveal changes in the level of biogenic amines in the brain of animals given P.L.G. with M.I.F. activity. Similarly we could find no effect on monoamine-oxidase of catechol0-methyl transferase activity. Furthermore P.L.G. does not seem to release the catecholamines in an amphetamine-like manner.4 8 The most likely site of action, if it is synaptic, is the post-synaptic membrane. In favour of this hypothesis is the fact that M.I.F. greatly potentiates apomorphine-induced hyperactivity in intact as well as in hypothalamus-lesioned rats.19 However, many other possibilities must still be considered and tested. It is indeed of interest that serum M.s.H. levels were found elevated in Parkinson’s disease 20 and that administration of M.S.H. aggravated the symptoms of that illr.ess.2’ Could Parkinson’s disease be characterised by deficient hypothalamic M.I.F. production which in turn, through unhampered central M.S.H. activity, could influence the functioning of the melanocyte-like pigmented cells in the substantia nigra and locus coeruleus, resulting in a deranged production of catecholamines? Such hypotheses are being tested in our laboratory. These experiments, while encouraging, do not yet prove that P.L.G. improves motor performance or even intellectual function in Parkinson’s disease. We are still ignorant of the long-term efficacy, or toxicity, of this substance and realise that continuous intravenous therapy, in any disease entity, is no final solution.
Hypothesis VULNERABILITY OF CELL-SURFACE RECEPTORS TO AUTOIMMUNE REACTIONS P. R. CARNEGIE Russell Grimwade School of Biochemistry, University of Melbourne, Parkville, Victoria 3052, Australia
I. R. MACKAY Clinical Research Unit, Walter and Eliza Hall Institute of Medical Research, Post Office, Royal Melbourne Hospital, Victoria 3050, Australia
This
more
We thank Abbott Laboratories for supply of p.L.G. (A-40509); Dr. J. Kastin (New Orleans) for advice and encouragement; Miss D. Bedard, Mtss L. Paul, and Miss C. Larouche for helping in the experiments ; and the United Parkinson Foundation and the W. Garfield Weston Foundation for financial support. Requests for reprints should be addressed to A. B., Department of Neurobiology, Clinical Research Institute of Montreal, 110 Pme Avenue West, Montreal, Quebec, Canada H2W lR7. A.
A previous hypothesis in which myasthenia gravis was explained by an immune response to acetylcholine receptors has been validated, and is here extended to cell receptors in general.
Summary
Receptors on target cells, being accessible to circulating trophic hormones or transmitters, must also be accessible to antibodies which compete with the natural mediator for access to the site. To detect anti-receptor antibodies, physiological assay systems would be more sensitive than conventional immunological assays. Autoimmune responses to receptor sites would require a genetic predisposition to failure of immunological tolerance such as occurs in various autoimmune diseases. This hypothesis is supported by recent findings in hyperthyroidism and a type of insulin-resistant diabetes mellitus, and is applicable to other endocrinopathies, diseases in which dysfunction at receptor sites can be postulated, and regulatory functions within the immune system itself.
DR BARBEAU: REFERENCES A. J., Barbeau, A , Ehrensing, R. H., Plotnikoff, N., Schally, A V. Adv Neurol. 1974, 5, 225. 2. Nair, R. M. G., Kastin, A. J., Schally, A. V. Biochem. Biophys. Res Commun. 1971, 43, 1376. 3 Plotnikoff, N. P., Kastin, A. J., Anderson, M. S , Schally, A. V Life Sci. 1971, 10, 1279. 4. Kastin, A. J., Miller, M. C., Ferrell, L., Schally, A. V. Physiol Behav. 1973, 1.
Kastin,
10, 399 Plotnikoff, N. P., Minard, F. N., Kastin, A. J. Neuroendocrinology, 1974, 14, 271 6. Huibodro-Toro, J. P., Scotti de Carolis, A., Longo, V. G. Pharmac. Biochem. Behav. 1974, 2, 105. 7. Plotnikoff, N P , Kastin, A. J. Archs int. Pharmacodyn. Thér 1974, 211, 5.
211. 8.
Huibodro-Toro, J. P., Scotti de Carolis, A., Longo, V G Pharmac chem. Behav. 1975, 3, 235. Kastin, A. J., Barbeau, A. Can. med. Ass. J 1972, 107, 1079.
Bio-
9 10 Woods, A. C., Chase, T. N. Lancet, 1973, ii, 513. 11. Chase, T. N., Woods, A. C., Lipton, M. A., Morris, C E. Archs Neurol., Chicago, 1974, 31, 55. 12. Fischer, F. A., Schneider, E., Jacobi, P., Maxion, H. Eur Neurol 1974, 12, 360. 13. Barbeau, A., Roy, M. Can. med. Ass. J. (in the press). 14. Yahr, M. D., Duvoisin, R. C., Shear, M. J., Barrett, R. E., Hœhn, M.Archs Neurol., Chicago, 1969, 21, 343. 15. Dery, J. P., De Groot, J. A., Laurin, C., Barbeau, A. Un. méd. Can. 1962,
91, 842. Barbeau, A. Can. med. Ass J 1969, 101, 791. Hamilton, M. J. Neurol. Neurosurg. Psychiat. 1960, 23, 56. Botez, M. I., Barbeau, A. Int. J. Neurol. 1975, 10, 222. Burnett, C., Strother, E., Butterworth, R. F., Barbeau, A Unpublished Shuster, S., Burton, J. L., Thody, A. J., Plummer, N., Goolamali, S. k., Bates, D. Lancet, 1973, i, 463. 21. Cotzias, G C., Van Woert, M. H., Schiffer, L. M. New Engl. J Med. 1967,
16. 17. 18. 19 20.
276, 374.
685 INTRODUCTION
hypothesis "immunopharmacological block" of Lennon and Carnegie’ has been validated by observations on anti-receptor antibody in clinical and experimental myasthenia gravis. Various other diseases could be accounted for by a pathogenic immune response to cell-surface receptors which, being accessible to circulating hormones or neurotransmitters, should also be accessible to antibody. This type of autoimmune response would be detected best by sensitive physiological systems in which antibody would compete with hormone for access to receptor on responsive cell preparations. The. genetic factors involved in regulation of immune responses would render certain individuals more susceptible to a break in tolerance with formation of anti-receptor antibody-e.g., cross-reaction between a component of an infecting microorganism and an antigenic determinant at the receptor site. Also anti-receptor autoantibodies would need to be of high affinity for accessible components of the receptor site. THE
of
choline receptor site13-16 and a cellular response was also shown.17 Such anti-receptor antibodies could also interfere with function at other acetylcholine receptor sites-e.g., in the central nervous system, provided the blood/brain barrier was breached. The earlier hypothesis implicated genetic factors in failure of tolerance to receptor sites. The histocompatibility antigen HL-A8 is associated with that type of myasthenia gravis in young women who lack the characteristic myoid antibody demonstrable by immunofluorescence,18 whereas such patients do have anti-receptor antibody;’6 HL-A2 is associated with the other group of myasthenic patients who are elderly, and have thymoma and myoid antibody’9 and also anti-receptor antibody.16 Thus, the features of myasthenia gravis are well accounted for by block of receptor sites, and the experimental system shown in the figure provides a model protocol for examining various diseases possibly involving neurotransmitters or endocrine dysfunction. GRAVES’
MYASTHENIA GRAVIS
Simpson2 in 1960 proposed that in myasthenia gravis a blocking antibody competed with acetylcholine for the receptor on the motor end-plate. This proposal was not neglected’but could not be substantiated until the fortuitous recognition by Patrick and Lindstrom45 in 1973 that induction of antibody in rabbits to acetylcholine receptors from Electrophorus electricus resulted in an autoimmune response against the rabbits’ motor endplates and myasthenia. This model of myasthenia gravis (see accompanying figure) was widely confirmed by studies in rabbits, rats, guineapigs, and monkeys.6-12 Not all the anti-receptor antibodies appear to be directed at the precise site binding acetylcholine.11 Most patients with myasthenia gravis have humoral antibodies to acetyl-
DISEASE
Hyperthyroidism in Graves’ disease exemplifies the need for physiological or receptor assay systems to demonstrate functionally important autoantibodies against receptors. Adams and Purves20 in 1956 likewise made a fortuitous discovery of an anti-receptor antibody, the long-acting thyroid stimulator (L.A.T.S.), later found to be an immunoglobulin.21 Serum from patients with Graves’ disease may contain, as well as L.A.T.S. which stimulates mouse thyroid cells, other immunoglobulins (L.A.T.s. protector) which stimulate human but not mouse thyroid cells.22 23 These immunoglobulins both compete with and mimic the action of thyroidstimulating hormone (T.s.H.), but in an uncontrolled way. Immunoglobulins from patients with Graves’ disease inhibit the binding of T.S.H. to thyroid membranes,24-27 and L.A.T.s. and the immunoglobulin which competes with T.s.H. in the receptor assay system are considered to be one and the same. These sensitive physiological assay systems, which have been devised for detecting antibodies against the T.s.H. receptor, could prove to be a simple model for the investigation of anti-receptor activity in other endocrine diseases. DIABETES MELLITUS
Experimental myasthenia gravis in rabbits as an example of a disease by autoantibodies to a receptor site. : tchuhne Ach) receptor was isolated from affinity chromatography and - --J ,o! mmunise rabbits.’ In addition to antibody against eel Ach receptor, ann- -’ developed to the rabbits’ own Ach receptors causing myasthema 4 Pa-
caused
mtasthenta gravis have autoanttbodies to determinants III the human receptor which interfere with neuromuscular transmtssion "’ The ’:.ng agent in man, equivalent to immunisation in the rabbit, is unknown. rrm Carncgie." reproduced by permission of the Editors of Search ,c:, ’.11th
- 1_hultne
Among the endocrine disorders, diabetes mellitus is an obvious candidate for the application of this hypothesis. There are two types of diabetes-the "juvenile-onset" type in which there is a low serum level of insulin, indicative of 3-islet cell dysfunction, and an "adult-onset" type in which there are normal or elevated levels of insulin but an apparent resistance to its action, indicative of a refractory target cell. The juvenile-onset type is associated with autoantibodies to thyroid and gastric antigens28 29 and to pancreatic islet cells3O-32 and insulinoma cells.33 Our hypothesis could be tested, firstly, by in-vitro assay systems developed for assaying the function of isolated 3-islet cells34 and isolated fat cells,35 and secondly, by use of isolated insulin receptors36 in experiments similar to those used in Graves’ disease and mvasthenia gravis. In juvenile-onset diabetes, sera which contained anti 3-islet cell antibodies should contain antibodies which block the response of the isolated 3-islet cells to glucose, in a manner mimicking the pharmacological
686
action of the polypeptide somatostatin;34 these sera should not inhibit binding of iodinated insulin to isolated insulin receptor, nor block the action of insulin using the isolated fat-cell assay. In adult-onset diabetes, the hypothesis would predict the reverse: no influence of sera using the isolated p-islet-cell assay, but a competitive block of the insulin receptor on fat cells. Thus laboratories studying these physiological assay systems could readily test the hypothesis, and indeed Flier et al. 37 found that 3 of 6 patients with extreme insulin resistance had antibodies against insulin receptors on human monocytes. The competitive binding assay which they used did not detect anti-receptor antibody in the more usual type of insulin-resistant diabetic; with such patients a more sensitive assay would be the isolated fat cell. Fat cells respond to insulin by increasing uptake of glucose ;35 antibody to the insulin receptor would be expected to block this response to insulin. POSSIBLE ANTI-RECEPTOR DISEASES
endocrinopathies exhibit features expected in disease due to antibody to receptor sites: failure of the target organ possibly but not necessarily associated with autoantibodies shown by immunofluorescence, and high levels of the trophic hormone but resistance to its action. We suggest that antibody will be found to various receptors : as examples, receptors for A.c.T.H. in "autoimmune" Addison’s disease,38 for parathyroid hormone in pseudohypoparathyroidism,39 and for gastrin in pernicious ansemia. 40 Of relevance is the temporary dysfunction of Other
receptors in newborn children of mothers who have myastheniathyrotoxicosis,41 or diabetes mellitus.41 42 We attribute this to transplacental transfer of maternal anti-receptor antibody, virtually established in the case of neonatal myasthenia and thyrotoxicosis, and now pos-. tulated for transient neonatal diabetes, and for infants of low birth-weight despite high growth-hormone levels.43 In testing the above ideas, human receptors, preferably on viable cells, would probably be more specific and sensitive than animal receptors for the assay of anti-receptor antibodies. Fluctuating symptoms in multiple sclerosis could likewise be explained by the hypothesis. 1 44 In vitro, immunoglobulins from patients block electrical activity in cultured neural tissue;45 and a block of receptors in vivo is suggested by the ability of precursors of neurotransmitters to prevent rapid deterioration in optic-nerve functions induced by hyperthermia.46 47 Recently it was found that some 50% of patients with multiple sclerosis had an immune response to oligodendroglial cells.48 Because of the interaction between the encephalitogenic protein and serotonin49 and the presence of serotonin receptors on oligodendrocytes,50 we postulate an immune response to such receptors in multiple sclerosis. In "biochemical neuropathies" possibly exemplified by schizophrenia, 51 a pharmacological agent used thera-
peutically (phenothiazine) appears to overcome a specific dysfunction of a dopamine receptor, although this dysfunction is not demonstrably due to anti-receptor antibody. However, the earlier data of Heathsz could be reconsidered in that the ability of antibody to interfere with receptor function in the absence of an inflammatory response answers the argument that "the absence of conspicuous tissue reaction in schizophrenia does not seem favourable to an autoimmune origin".51
LYMPHOCYTE RECEPTORS FOR ANTIGEN
Lymphocytes have specific receptors for antigen and increasing evidence that antibody against the specific antigen-binding site in immunoglobulin molecules could function against the corresponding specific antigen receptors on B and possibly T lymphocytes.53 Ramseier’s54 experiments indicate that such anti-idiotypic antibodies can regulate the immune response to histocompatibility antigens, and Strayer et al.’s" experiments on the induction of neonatal tolerance bv antibody to the idiotype reacting with phosphorylcholine support Jerne’s postulate that the immune system could be self-regulatory through a "network"56 57 based in part on idiotypic and anti-idiotypic antibodies. Anti-allotype antibodies could also interfere with the immune response to an antigen, especially if directed against the determinants close to the specific antigen combining site. Dray58 has described how lymphoid cells with a genetic capability of expressing on their surface an immunoglobulin phenotype (allotype) can be blocked by exposure to the appropriate anti-allotype antiserum.
there is
extension of this concept to cancer, antibodies membrane antigens of melanoma could be inagainst volved in localising the tumour; an. anti-idiotypic antibody formed against the antibody to membrane antigens was claimed to result in the blocking effect on tumour In
an
immunity. 59 Returning
to
autoimmune diseases with
fluctuating
clinical symptoms, as exemplified by lupus erythematosus and myasthenia gravis, there could be temporary ascendency of a clone of "suppressor" lymphocytes with production of anti-idiotypic antibody, either to the
pathogenic antibody or to the lymphocyte receptor for auto antigen, thus reducing the pathogenic clones with consequent remission of disease. Finally, there are therapeutic implications, in that diseases which are currently being treated by regular injection of hormone could possibly be treated by manipulation of the immune response to restore immunological tolerance to receptor sites. We thank many colleagues for discussion and data, particularly Dr V. A. Lennon, Dr J. M. Lindstrom, Dr G. F. Mitchell, Dr I. J. Wood. Dr D. D. Adams and Dr M. T. McQuillan. Support by the National Multiple Sclerosis Society, New York, and the National Health and Medical Research Council of Australia is acknowledged.
REFERENCES 1. Lennon, V. A., Carnegie, P. R. Lancet, 1971, i, 630. 2. Simpson, J. A. Scott. med. J. 1960, 5, 419. 3. Mackay, I. R., Burnet, F. M. Autoimmune Diseases. Springfield, Illinois, 1963. 4. Patrick, J., Lindstrom, J. M. Science, 1973, 180, 871. 5. Patrick, J., Lindstrom, J. M., Clup, B., McMillan, J. Proc. natn. Acad. Sci. U.S.A 1973, 70, 3334. 6. Sugiyama, H., Benda, P., Meunier, J-C., Changeux, J-P. FEBS Letters, 1973, 35, 124. 7. Heilbron, E., Mattson, C. J. Neurochem, 1974, 22, 315. 8. Heilbron, E., Mattson, C., Stalberg, E., Hilton-Brown, P. J. neurolog. Sci.
10.
1975, 24, 59. Hazdai, R., Aharonov, A., Silman, I., Fuchs, S., Abramsky, O. Nature, 1975, 256, 128. Lennon, V. A., Lindstrom, J. M., Seybold, M. E. J. exp. Med. 1975, 141,
11.
Lennon, V. A., Lindstrom, J. M., Seybold, M. E. Ann. N Y Acad Sci.
9.
1365. the 12. 13. 14. 15.
(in
press).
Green, D. P. L., Miledi, R., Vincent, A. Proc. R. Soc. B. 1975, 189, 57. Almon, R. R., Andrew, C. G., Appel, S. H. Science, 1974, 186, 55. Almon, R. R., Appel, S. H. Biochem. biophys. Acta, 1975, 393, 66. Bender, A. N., Ringle, S. P., Engel, W. K., Daniels, M. P, Vogel. Z Lancet, 1975, i, 607. 16. Lindstrom, J. M., Lennon, Seybold, M. E., Whittingham, S. Ann. N. Y. Acad. Sci. (in the press).
687
KAPOSI’S SARCOMA: A BYPRODUCT OF TUMOUR REJECTION T. F. C. S. WARNER
progressive growth of the virally transformed mative mesenchyme.
S. O’LOUGHLIN
Mayo Clinic, Rochester, Minnesota 55901, U.S.A. It is suggested that Kaposi’s sarcoma is the result of a chronic immunological reaction between antigenically altered or transformed lymphoid cells and normal lymphocytes. In the course of this local graft-versus-host type activity, an angiogenesis factor is liberated and intense proliferation of mesenchymal and endothelial cells ensues. During the G.V.H.-like activity, an oncogenic virus is either transferred to or induced in the cells responsive to the angiogenesis factor. Thus, the stage is set for neoplastic transformation of these cells in an environment which is conducive to the
Summary
vasofor-
INTRODUCTION sarcoma is an interesting and important in tumour many respects. Like Burkitt’s lymphoma, it is relatively rare outside the African continent where in certain regions it accounts for nearly 10% of all tumours.In non-African patients an unusually high incidence of lymphomas has been observed2 and the disease has also been associated with immunosuppression in homograft recipients.3 The demonstration of herpestype viruses in explants of Kaposi’s sarcoma in tissueculture may establish this tumour as another useful model for the study of viral involvement in human
Kaposi’s
malignancy.44
There has been much controversy about the exact tiscell of origin of Kaposi’s sarcoma. Despite this, most would now concede that the neoplasm arises in cells of reticuloendothelial origin or, more specifically, vasoformative mesenchymal cells, and Hashimoto and Lever56 are of the opinion that it is a neoplasm of periS. Clin. Immun. O., Fuchs, 17. Abramshy, Aharonov, A., Webb, C., exp. 1975, cytes and endothelial cells. However, we propose that 19, 11. the initial event in the histogenesis of Kaposi’s sarcoma 18. Feltkamp, T. E. W., Van den Berg-Loonen, P. M., Nijenhuis, L. E., Engleis not direct transformation of vasoformative cells by an friet, C. P., Van Rossum, A. L., Van Loghem, J. J., Oosterhuis, H. J. G. H Br. med. J. 1974, i, 131. oncogenic agent (? virus) but that the neoplasm is a 19. Fritze, D., Herrman, C., Naeium, F., Smith, G. S., Walford, R. L. Lancet, byproduct of tumour rejection. 1974, i, 240. 20 Adams, D. D., Purves, H. D. Proc. Univ. Otago medœ. Sch. 1956, 34, 11. Dorffel7 has recognised several stages during tumour 21 Kriss, J. P., Pleshakov, V., Chain, J. R. J. clin. Endocr. 1964, 24, 1005. development. Briefly these are: (1) an inflammatory 22. Adams, D. D., Kennedy, T. H. ibid. 1967, 27, 173. 23. Shishiba, Y., Shimizu, T., Yoshimura, S., Shizume, K. ibid. 1973, 36, 517. lesion, (2) a phase of capillary proliferation, (3) an 24. Manley, S. W., Bourke, J. R., Hawker, R. W. J. Endocr. 1974, 61, 419. angiomatous phase, and (4) an angiosarcomatous stage 25 Manley, S. W., Bourke, J. R., Hawker, R. W. ibid. 437. in which malignant spindle cells proliferate. There is 26. Mehdi, S. Q., Nussey, S. S., Gibbons, C. P., El Kabir, D. J. Biochem. Soc. Trans. 1973, 1, 1005. much overlap and a multiplicity of histological changes 27. Smith, B. R., Hall, R. Lancet, 1974, ii, 427. is necessary for a definite diagnosis.8 Montgomery9 has 28. Ungar, B., Stocks, A. E., Martin, F. I. R., Whittingham, S., Mackay, I. R. ibid. 1968, ii, 415. emphasised the mononuclear inflammatory component 29. Irvine, W. J., Clarke, B. F., Scarth, L., Cullen, D. R., Duncan, L. J. P. ibid. which consists of lymphocytoid cells, lymphocytes, his1970, ii, 163. 30. Bottazo, G. F., Florin-Christensen, A., Doniach, D. ibid. 1974, ii, 1279. tiocytes, and plasma cells. The lymphocytoid cells of 31. Lendrum, R., Walker, G., Gamble, D. R. ibid. 1975, i, 880. Marchand7 probably represent lymphocytes which are 32 MacCuish, A. C., Barnes, E. W., Irvine, W. J., Duncan, L. J. P. ibid. 1974, responding to a specific mitogenic stimulus. Thus, this ii, 1529. 33 Maclaren, N. K., Huang, S-W., Fogh, J. ibid. 1975, i, 997. early lesion could represent a local reaction on the part 34 Okamoto, H., Noto, Y., Miyamoto, S., Mabuchi, H., Takeda, R. FEBS Letof the host to transformed autochthonous lymphocytes ters, 1975, 54, 103. 35 Lockwood, D. H., Livingston, J. N., Amatruda, J. M. Fedn. Proc. 1975, 34, -i.e., a host-versus "graft" or a "graft" versus-host 1564. response. Indeed as tumours of the reticuloendothelial 36. Cuatrecasas, P., Parikh, I. Meth. Enzym. 1974, 34, 653. 37 Flier, J. S., Kahn, C. R., Roth, J., Bar, R. S. Science (in the press). system, Kaposi’s sarcoma and Burkitt’s lymphoma are 38. Bresser, G. M., Cullen, D. R., Irvine, W. J., Ratcliffe, J. G., Landon, J. Br. peculiar in that primary tumours are commonly found med. J. 1971, i, 374. in sites which possess a sparse lymphoid-tissue com39 Chase, L. R., Melson, L., Aurbach, G. D. J. clin. Invest. 1969, 48, 1832. 40 Strickland, R. G., Mackay, I. R. Am. J. dig. Dis. 1973, 18, 426. ponent. Furthermore, lymphoma-like changes have been 41 Scott, J. S. Br. med. J. 1966, i, 1559. described in lymph-nodes of patients with Kaposi’s sar42. Cornblath, M., Schwartz, R. Disorders of Carbohydrate Metabolism in Incoma 10 and in a few cases sarcomatous changes merged fancy; p. 57. Philadelphia, 1966. 43 Humbert, J. R., Gotlin, R. W. Pediatrics, Springfield, 1971, 48, 190. with those of lymphoma.11-14 A high incidence of lym44 Carnegie, P. R. Nature, 1971, 229, 25. 45 Bornstein, M., Crain, S. M. Science, 1965, 148, 1242. phoma, especially Hodgkin’s disease, has been found in 46 Harrer, G. Germ. Med. 1973, 3, 116. non-African patients.2 Such an impressive association 47 Harrer, G., Fishbach, R. J. neural Trans. 1973, 34, 205. has not been encountered in African cases, and this may 48. Myers, L. W., Ellison, G. W., Fewster, M. E., Wolfgram, F. Archs. Neurol. 1975, 32, 354. be due to involvement of a younger age-group and a 49 Carnegie, P. R., Smythies, J. R., Caspary, E. A. Field, E. J. Nature, 1972, greater number of fatal cases.15 Burkittl6 has cited addi240, 561. 50 Murray, M. R. In Biology of Neuroglia (edited by W. F. Windle); p. 176. tional similarities between Kaposi’s sarcoma and BurSpringfield, Illinois, 1958. kitt’s lymphoma including long-term survival, recurrence 51 Matthysse, S. , Lipinski, J. A. Rev. Med. 1975, 26, 551. at a different site, and the poor prognosis when lymph52 Heath, R. G., Krupp, I. M. Archs gen. Psychiat. 1967, 16, 1. 53 Eichmann,K., Eur.J. Immun. 1975, 5, 511. nodes are involved. In addition, the coexistence of a pro54 Ramseier, H. Curr. Top. Microbiol. Immun. 1973, 60, 31. 55 Straver, D. S, Cosenza, H., Lee, W. M. F., Rowley, D. A., Kohler, H. gressive growth in one site with a regressive tumour elseScience, 1974, 186, 640. where has been recorded in both diseases but has not sue or
56 Jerne, N. K. Scient. Am. 1973, 229, no. 1, p. 52. 57 Richter, P J Eur J. Immun. 1975, 5, 350. 58. Dray, S. in Ontogeny of Acquired Immunity (Ciba Foundation Symposium); p 87 Amsterdam, 1972. 59. Hartman, D., Lewis, M. G., Proctor, J. W., Lyons, H. Lancet, 1974, ii, 1481. Carnegie, P. R. Search, 1974, 5, 472.
been
satisfactorily explained. 16 GRAFT-VERSUS-HOST DISEASE
Experimentally induced graft-versus-host (G.v.H.)
dis-