Arthritis and Rheumatism O$cid Journal of the qmerican Rheumatismassociation VOL. IV, NO. 1 FEBRUARY, 1961 SPECIAL ARTICLE Vasculitis, Mast Cells and the Collagen Diseases By CHARLEY J. SMYTHAND ORENB. GUM Purpura and necrotizing vasculitis complicate high dose adrenocortical therapy in rheumatoid arthritis. Pathological changes in five cases are summarized. Tissue mast cells are. concentrated around blood vessels and store heparin, histamine, hyaluronic acid and in some species, serotonin. These pleuri-potential cells upon “trauma” yield these potent biological substances which may act as prime movers in the idammatory reaction. Association of these cells with adrenocortical steroids, fibrogenesis and anaphylaxis is further evidence that they play a role in the panangiitic reaction of hypercortisonism. Purpura e vasculitis nectotisante rende complicate le us0 de therapia a alte doses de steroides adrenocortical in arthritis rheumatoide. Es summarisate le alterationes pathologic incontrate in cinque casos. Mastocytos del tissu es concentrate circum le vasos de sanguine e effectua un thesaurisation de heparina, histamina, acido hyaluronic, e-in certe species-de serotonina. Quzndo le mastocytos, que es pluripotential, es “traumatisate”, illos libera le mentionate biologicamente potente substantias que alora pote ager como promotores primari del reaction idammatori. Le association de iste cellulas con steroides adrenocortical, fibrogenese, e anaphylaxe supporta additionalmente le these que illos ha un rolo in le reaction panangiitic de hypercortisonismo. T HE PURPOSE OF THIS PRESENTATION is to propose an hypothesis. By this method it is planned to present certain facts and relationships and from these draw certain inferences which may lead to a better understanding of one of the iinlfying aspects of the so-called collagen vascular diseases. This discussion will center around the pathological changes in and about the blood vessels of patients with rheumatoid arthritis which some observers consider to be the site of the primary pathological Jesion in this illness.lP2 The line of reasoning and steps leading to the development of this new concept will be considered in the same chronological fashion as it has evolved in our group. Some of the evidence which will be used to fabricate this thesis will be factuil; Thb paper was presented aa the Stephen Walter R a n s a Alemoth1 Lecture at Nmthm..vtem Uniuers& Medical School, Chicago, Illinois, November 4, 1959. From the Department of Medicine, Section of Rheumatic Diseases, University of Colorack School of Medicine, Dentm, Colorado. 1 2 S M Y T H AND GUM much is circumstantial and an earnest effort will be made to keep these differences clear and distinct. As is well known, the cause and cure of rheumatoid arthritis is unknown. The need for fresh, new, unorthodox and untried approaches to this problem is clearly evident. No one can possibly predict what new routes are going to lead to the discovery of the ultimate cause of rheumatoid arthritis. It is also cIearly evident that we are going to have to explore some entirely untried avenues if we are to make headway against this stubborn and destructive foe. In this endeavor, evidence will be presented to support a new concept of pathogenesis of the primary vascular lesion of rheumatoid arthritis. Two RECENTCLINICAL OBSERVATIONS Subcutaneous hemorrhages: About two years ago, our attention was first called to the occurrence of ecchymotic areas on the hands and forearms of patients attending the Arthritis Clinics at the Colorado General Hospital. The frequency with which these skin hemorrhages were being encountered appeared to be greater than either of us had been accustomed to seeing in arthritis patients and we began to study the particular group of individuals. It soon became evident that these subcutaneous hemorrhages were developing in patients with rheumatoid arthritis who had been maintained on one or another of the adrenocortical steroids or corticotropin for long periods of time and were in a state of hyperadrenocortisonism. At the time the first patients were studied many of them were receiving 20 to 30 mg. of prednisone or prednisolone daily. Today we would consider these rather excessive doses of the two steroid drugs. Figure 1 contains several illustrations of these subcutaneous hemorrhages which were noted to develop with only minor trauma in patients on long-term steroid therapy. Fig. 1.-Hemorrhagic lesions in the skin of the upper extremity of four patients with rheumatoid arthritis (A,B,C,D). Each patient had iatrogenic hypercortisonism from long-term cortiwteroid therapy. VASCULITIS, MAST CELLS A N D COLLAGEN DISEASES 3 Fig. 2.-(A) Dry gangrene of right leg and foot of patient (A. M. CoIorado General Hospital #117986) a 55 year old female with rheumatoid arthritis of 14 years’ duration. She had received adrenocortical steroid therapy for eight years and was in a hypercortisone state when this vascular occlusion occurred. (B) Mediumsized artery from the amputated extremity showing an organized thrombus partially occluding the lumen with fibrosis extending through the wall into the adventitia. An analysis was made of 109 patients attending the Arthritic Clinic at the Colorado General Hospital; it was found that this type of vascular skin lesion was present in 40 per cent of the group and formed the basis of a report in 1958.3 At about the same time, other published reports appeared which indicated a similar incidence; Boland4 found 34 per cent among 70 patients and DenkoK reported 20 per cent in 75 patients. Efforts to 4 SMYTH .4ND GUM Fig. 3.-Microscopic section from myocardium. (Patient A. M., Colorado General Hospital, # 117986.) Acute necrotizing rrteritis with extensive infiltration of lymphocytes, mononuclear cells and proliferating fibroblasts involving the adventia. demonstrate the cause of this hemorrhagic tendency revealed no measurable defect of the intravascular clotting factors. By using a negative pressure suction apparatus, no hemorrhages were produced, indicating that there was no defect in the integrity of the capillary wall. At this point our attention was focused upon the possibility that the defect responsible for these purpuric lesions was an abnormality in the extravascular hemostasis. In other words, was the cause for the hemorrhage a failure to quickly close the rent in the capillary continuity, thus permitting a macroscopic hemorrhage to occur? We shall return to this line of thinking a little later. Vasditis in steroid-treated patients: Another event growing out of experiences with adrenocortical steroid therapy in the long-term management of patients with rheumatoid arthritis served to focus attention again to the vascular system. In 1957 one of us ( C . J. S.) was asked to see a female patient, age 67, with rheumatoid arthritis who had been on continuous adrenocortical steroid therapy for three years. She complained bitterly of burning pain and numbness in the toes and feet and tips of the fingers. Soon afterward she was admitted to a Denver hospital with purplish-black fingertips, and the skin over the shins was necrotic. She died after a few weeks in the hospital and at autopsy was found to have a difhse polyarteritis-nodosa-like vasculitis. During the next several months, we observed two other patients both on high dose long-term adrenocortical steroid therapy who died, and at autopsy showed similar diffuse vascular changes which had the pathological characteristics of polyarteritis nodosa except that the lesions were distributed evenly along the length of the vessel. T h e observations in these three cases and an exhaustive review of the literature concerning the vascular lesions in rheumatoid arthritis formed the basis of a recent report by Drs. Johnson, Lubchenco, Valentine, Holt and Smyth.6 Since that report, two additional patients with rheumatoid arthritis on high doses of adrenocortical steroids have been observed. In one patient there was extensive dry Fig. 4A.-Left hand showing gangrene of the fingertips in a 61 year old male. (Initials C. T., Denver Veterans Administration Hospital, #40791) Rheumatoid arthritis for eight years and continuous adrenocortical therapy ranging from 20 to 40 mg. per day for two years. Fig. 4B.--Small arteriole from the submucosa of the intestine of the above patient obtained at autopsy. There is fibrinoid necrosis of the entire thickness of the vessel waII with mononuclear and lymphocytic cellular exudate in the adventitia. 6 SMYTH AND GUM gangrene of the right lower extremity (Fig. 2A). At autopsy there were occlusions of large vessels in this extremity (Fig. 2B). In the liver, kidneys, myocardium and lungs there was a generalized polyvasculitis of the periwteritis nodosa type (Fig. 3). .knother patient developed necrosis of the fingertips (Fig. 4A).The autopsy studies revealed a widespread acute necrotizing arteritis in all visceral organs (Fig. 4B). Many vessels showed residual fibrosis, hyaline degeneration and marked intimal thickening. Doctors Kemper, Baggenstoss and Slocumb? approached this problem by reviewing the frequency of arteritis at the hlayo Clinic and reviewed all cases with the diagnosis of rheumatoid arthritis that came to autopsy through 1954. Of the 52 cases carefully studied, 14 had received cortisone and 38 had not received any. They observed that no cases with polyarteritis were recorded among the 38 who had not received steroids, whereas there were 4 cases with such diffuse vascular lesions among the 14 that had received cortisone. 2 of these 4 showed other features of chronic hypercortisonism. Widespread necrotizing arteritis is well known in cases of rheumatoid arthritis in which the patients had not received corticotropin. However, the occurrence of this nsmotizing angiitis or panmesmhymal reaction is much more pronounced and appears with a greater frequency and severity in patients with chronic hypercortisonism. Less severe forms of this entity than those illustrated above are now frequently recognized. The presence of peripheral neuritis and the skin lesions are the most common clinical manifestations of this condition. A gradual reduction of the dose of adrenocortical steroids is, in our experience, usually accompanied by a gradual disappearance of the symptoms due to this reaction. MASTCELL - ]JLOOD\'ESSEL hLATIONSIIIP Our interest in this presentation is to suggest an explanation of the mechanism for the production of this diftuse panangiitic or polyarteritis noclosa-like lesion. The observations of our own group and those of others point to the production of or the marked accentuahon of a lesion which is both perivascular or adventitial as well as mural and intramural in location. Let us now turn back to the first cases which were described, those with subcutaneous hemorrhages which developed upon minor trauma. What are the structures immediately surrounding the small blood vessels which might be involved in the extravascular clotting of blood? Our attention has been drawn to a group of cells located immediately outside the capillaries, venules and arterioles. They were first described by Paul Ehrlich in 11177 when he was a medical student.8 Their demonstration depends upon special metachromatic stains. He proposed the name "Mastzellen" or mast cells which means "wellfed cells." They are not seen in the usual hematoxylin and eosin stained and formalin-fixed tissues because the granules have an avidity for basic dyes, and it is this reaction which characterizes these cells. They, therefore, escape attention in the usual study of tissue sections. Numerous studies of these cells in various species have shown that they have a particular tendency to cluster fairly close to blood vessels. Riley, in a comprehensive study of the mast cell in the rat and cattle, demonstrated their close relationship to the great and small blood vessels and pointed out that their number was determined largely by the extent to which the blood vessels are equipped with an adventitial coat of mesenchymnl cells (Fig. 5).4" The perivascular location of the mast cells is illustrated in the book on Heparin by Jorpes (1946, p. 64).b4 A similar arrangement of mast cells in the aclventia of blood vessels was also depicted by Ehrlich.8 Staernmler,"B QuenseP and VASCULITIS, MAST CELLS AND COLLAGEN DISEASES Fig. 5.-In most tissues the mast cells lie predominately around arterioles and capillaries. Tissue spread of uterine fringe from a rat. Toluidine blue (X-108). Reprinted by permission, from Riley, J. F.: The Mast CeUs, E. and S. Livingstone Ltd., Edinburgh, 1959. S ~ n d b e r ghave ~ ~ each made detailed studies of the mast cells in man and all have noted the tendency for these cells to aggregate into groups around the blood vessels. Distribution of the n u t cell: The mast cells are interesting from several other viewpoints. They are described as being most commonly found in the following tissues in the higher vertebrates including man: kin,^^*^^ ciliary body,-’zp4isynovial membrane,22.*1*42.4* gastro-intestinal t r a ~ t , 4 ~cardiovas+~~,~~ cular pleu~a,40*~~ pericardium and lung^.^".^^ It was this distribution of the mast cells in the loose connective tissues of certain structures and organs, especially about blood vessels, which suggested a possible relationship between these ubiquitous cells and the so-called “collagen diseases.” If one may be permitted to draw an inference - it is these tissues with the high mast cell content that are the site of many of the changes one encounters in the connective tissue diseases. MASTCELLSAND CORTISONE One of the particularly provocative observations concerning mast cells especially in view of the skin hemorrhages and vascular changes which ‘are a part of hypercortisonism is the action of cortisone upon the mast cell. AsboeHansen of Copenhagen in 19521° presented evidence that in human patients the mast cells of dermal connective tissue decrease in numbers and become degranulated when exposed to the action of ACTH and cortisone ( Fig. 6 ) . These observations led this long-time student of the mast cell to consider 8 SMYTH AND G U N Fig. 6.- (A) Normal mast cell. Granules evenly distributed throughout the cytoplasm. (B) Mast cell from human skin influenced by cortisone. Vacuolation and conglomeration of granules. (C) Mast cell from human skin influenced by cortisone. Granules distributed in major and minor clusters and lumps, staining orthochromatically or metachromatically. Magnification approximately 3000 x. Reprinted by permission from Asboe-Hansen, G.: The Mast Cell. Cortisone Action on Connective Tissue. Proc.Soc.Exper.Biol.&Med. 80:677, 1952. them to be the target organ of the adrenocortical steroids. and he has designated the mast cells as the peripheral transmitters of hormonal influences at the connectizje tissue F. Bloom" reported a marked reduction in the number of mast cells in a mastocytcma in dogs following the administration of cortisone. At the same time he found that the hyaluronic acid of the ground substance was reduced. G. Bloom'' treated a child 10 months old with urticaria pigmentosa, a rare dermatological condition characterized by luge numbers of mast cells, with subcutaneous injections of cortisone in doses of 12.5 mg. daily for nine days. Biopsy specimens obtained from this child after cortisone treatment showed vacuolization and disruption of mast cells with a release of the cellular granules. KelCnyi15administered ACTH to rats, and 24 hours following the injection of 60 units he observed swelling and degranulation of the mast cells. He considered that the morphological chants corresponded to the initial changes in the mast cells seen immediately following lccal X-ray irradiation. Several other authors (Baker and Whitaker, 1950; Taubenhaus, Taylor and Morton, 1952; Zachariae and Moltke, 1954) have pointed out that cortisone as well as hydrocortisone act directly upon the mast cells. Other substances known to degranulate or disintegrate mast cells are colchicine.'J nitrogen mustard,'> roentgen therapy*4*1f' and the potent pharmacolcgical substance which is also a histamine liberator, compound 48/80.'17 An interesting avenue of future investigation would be to study the effect of other drugs ( gold, salicylates, chloroquine, iproniazid, reserpine, hydralazine, 'Compound 48/80 is a potent histamine liberator and a condensation prodiict of pmethoxyphenethylmethylnlnine with formaldehyde. 9 VASCULITIS, MAST CELLS AND COLLAGEN DISEASES phenylbutazone) and physical agents ( heat, massage, and ultra-sound ) commonly used in or possibly related to the rheumatic disorders to determine to what extent they alter mast cell morphology and function. PLTJRIPOTEN'I.IAL FIJNCTION OF MAST CELLS There is conflicting infcrmation pertaining to some of the functions of the mast cells (Fig. 7). From the standpoint of function they first received special attention when Jorpes9 demonstrated that they were ;he source of heparin. There is general agreement today that they contain heparin. Also it has been that firmly established by Riley and his co-WC)rkers16,%"'2a~~o and by 0thers*9~2'~~5 mast cells are rich in histamine. Asboe-HanserP proposed that they secrete hyaluronic acid. Benditt et a1.l') produced evidence that mast cells contain serotonin ( 5-hydroxy-tryptamine or 5 HT) . By direct analysis of mast cells isolated from rat peritoneal fluid it was shown that they contain approximately 0.7 mg. of serotonin per cu. mm. of cells. Studies by Parratt and Westz1 have shown that much of the serotonin as well as histamine in the skin of the rat is contained in tissue mast cells. The finding of serotonin in the mast cells of some species is a point which may have far-reaching implications in the pathophysiology of connective tissue and in the pathogenesis of collagen diseases. Sjoerdsma and associates"* were unable to demonstrale serotonin in the skin PLURIPOTENTIAL FUNCTIONS OF MAST CELL a . . a@)-. ... - . .. . * * Controverr i a l Serotonin I955(1 9 ) R at*+ I957 e t a 1 , 9 5 7 m h um a n G 0 Benditt. -7 and W e s t . Siaerdrmo . Fig. 7.-The potential role of mast cells in this dynamics of connective tissue is related to the chemical substances (heparin, histamine, hyaluronic acid and possibly serotonin) which they either synthesized or store. 10 SMYTH AND G U M from a patient with urticaria pigmentosa, a tumor rich in mast cells, and in two normal individuals. They did find large amounts of serotonin irr three human carcinoid tumors. These tumors arise from the chromafEn cells of the gastro-intestinal tract and patients with metastatic carcinoid tumors excrete excessive amounts of 5-hydroxyindoleacetic acid, a metabolic product of serotonin. Lembeck has previously demonstrated the presence of large amounts of serotonin in a carcinoid tumor.2‘;Some of the additional possible links between tissue mast cells and connective tissue diseases were recently summarized in an editorial in Arthritis and Rheumatism.“+ SEROLUNIN A N D COKNECTIVJC TISSUEDISEASE$ There are several bits of circumstantial evidence which have stimulated our interest in the possible relationship of serotonin to some of the changes seen in the connective tissue diseases. The reasons for believing that serotonin or a related derivative of tryptophane may be implicated in the collagen diseases are as follows: (1) In the malignant carcinoid syndrome there is frequently a marked fibroblastic proliferation of the subendocardium and valves of the right side of the In these instances there are metastases to the liver and ii high concentration of serotonin reaches the heart through the hepatic vein and bathes the interior. The maximum content of serotonin is present in the blood only during this brief passage from the liver to tlie right side of the hcwt to the lungs. This has been suggested as the reason for the subendocardial, and the tricuspid and pulmonary valvular fibrosis being localized in the right side of the heart.29Changes in the left side of the heart are not seen in the carcinoid syndrome and it is postulated that the high concentration of monoamine oxidase in the lungs inactivates the serotonin before it reaches the left side of the heart. In a recently reported case with malignant carcinoid in whom there was a patent foramen ovale, there was fibrosis of the valves on the left as well as the right side of the heart?” This observation lends support to the concept that the valvular lesions are the result of the high serotonin content of the blood. (2) The number of cases of connective tissue diseases occurring in patients with the carcinoid syndrome seems higher than would be expected from chance alone. Olson and Gray3‘ commented that the incidence of arthritis i s Table 1.-Reports of Carcinoid Syndrome and Connective Tissue Diseases - Author Year Sjoerdsma et 31.82 Jaiiies and MacI>onald4’ 1956 1956 Heilmeyer et al.5fi Saw34 Thorsonafi 1956 1958 1958 Zarafonetis et al.“! Asboe-Hansen69 1958 1959 Carcinoid Cases 4 8 (autopsied) 21 (reviewed) 1 8 12 1 1 Associated Connective Tissue Chanses 2 Rheumatoid arthritis 4 “Joint involvement” 6 “Joint involvement” 1 Rheumatic symptoms 1 Rheumatoid arthritis 1 Finger swelling around and between joints I Scleroderma 1 Scleroderma 11 VASCXLITIS, MAST CELLS AND COLLAGEN DISEASES increased in this disease. In a report of four cases with malignant carcinoid by Sjoerdsma et a1.,s2 two were thought to have rheumatoid arthritis. MacDonald33 has reported “joint involvement” in four of eight carcinoid patients autopsied at the Mallory Institute of Pathology. One of his cases showed marked thickening of the skin of the fingers, toes and face. These changes suggested those occurring in scleroderma or dermatomyositis. Sauer and his associatess4 studied 8 patients with the carcinoid syndrome at the Mayo Clinic and one had rheumatoid arthritis. An analysis of the joint findings among 12 patients personally studied and 67 from literature reviewed by Thorsonq5 indicated that articular and periarticular changes are not an uncommon feature of this syndrome. He found 13 cases with swelling, stiffness and pain referred to joints and/or sudden localized edema of the hands or face. Three of these patients had only joint manifestations and three had joint manifestations as well as localized edema. Seven had localized edema but no joint symptoms. One of his own cases ( E . Pa.) had finger swelling related to severe flush episodes and located not only around the joints but also between the joints. The association of scleroderma with malignant carcinoid was reported in 195836and 1959.0D ( 3 ) Dermal fibrosis in rats has been produced by the repeated injection of serotonin creatinine sulfate.37The daily injection of 8 mg. for 72 consecutive days produced marked thickening of the dermis, a decrease of hair follicles and thickening of the epidermis. These changes simulate the proliferation of collagenous and fibrous tissues within the dermis seen in human scleroderma. ( 4 ) The similarity in chemical structure of 5 HT and its metabolic product, 5-hydroxy, 3-indole acetic acid to the primary growth hormone in plants, 3-indole acetic acid, may be pertinent. The closely related structures of serotonin, 5-hydroxyindoleacetic acid and indoleacetic acid are shown in Figure 8. It does not seem unlikely that these substances might participate in similar enzymatic reactions with the basic aniine serotonin better adaDted for the alkaline pH of animal fluids and indoleacetic acid for the acidic juices of plants .6’J (5) The striking clinical effect produced after iniecting serotonin into and around the joints of patients with rheumatoid arthritis and svstemic lupus erythematosus renorted by Scherbel and Harrisonae and ScherbeP is additional evidence of the possible role of serotonin in this disease. In normal hdividuals, they demonstrated that serotonin produced a transient swelling and COMPOUNDS RELATED TO SEROTONIN SEROTONIN I5.HYDROXYTRYPTAMINEI 5-HYDROXYINDOLE ACETICACID INDOLEACETIC ACID (PLANT GROWTH H O R M O N E ) Fig. 8.-The similarity of structural formulae of serotonin, 5-hydroxyindole acetic acid and indole acetic acid the plant growth hormone is immediately apparent. 12 S M Y T H AND G U M erythema which disappeared within 30 minutes, while in the rheumatoid patients the response was exaggerated and lasted from two to eight hours. The reaction was characterized by diffuse p a h and erythema with swelling over the dorsum of the hand, which spread rapidly into the forearm and to all of the fingers. About five minutes after the injection, cyanosis occurred in all of the fingers and became progressively more severe during the next ten minutes and persisted for periods ranging from one to two hours. This exaggerated reactivity to the extravascular injection of serotonin could be blocked by the intravenous injection of two of the serotonin antagonists ( a ) 2-bromo-d-lysergic acid diethylamide and (b) Hydergine.* Less marked but similar responses follow the injection of histamine. They interpreted these exaggerated vascular reactions to the decreased activity in the tissues of one of the tissue enzymes, umine oxiidase, one of the enzymes responsible for the conversion of serotonin to 5-hydroxyindoleacetic acid. MASTCELLS AND ANTIGEN-ANTIBODY REACTION There are good grounds for believing that hypersensitivity has much to do with the mechanism of tissue damage in systemic lupus erythematosus and perhaps rheumatoid arthritis.lVs0 The mast cells may be a link in the chain of immunological events which results in the vascular lesion of these and allied connective tissue diseases. That the mast cells were directly influenced by an antigen-antibody reaction was demonstrated by Stuarts1 in 1952 who observed the disruption and degranulation of mast cells in rabbits, guinea pigs, dogs and mice subjected to anaphylactic shock induced by hen's egg albumin. Another study which related the tissue mast cells to the local hypersensitivity reaction was reported by Carter, Higginbothan and Dougherty.62 They injected homologous and heterologous antigens in previously sensitized mice and found that within five minutes after the injection of horse serum there was an explosive degranulation of mast cells. These results were considered to clearly demonstrate the hyper-reactivity of sensitized tissues to an otherwise relatively innocuous agent. Additional reports by Mota and Virgman,BSHumphrey and M ~ t a and , ~ ~Motaea have shown that mast cell alterations are produced by several antigen-antibody systems. These phenomena in experimental animals have not been reproduced in man. but similar studies in humans may prove a fruitful avenue of approach to exnlain the immediate vascular reactions in serum sickness and perhaps the delayed vascular responses of the other rheiimatic diseases. THEMASTCELLCYCLE If we return again to the two clinical observations with which this discussion was started, how do the known facts explain the skin hemorrhages and increased frequency of vasculitis observed in patients with hypercortisonism? The proposed scheme for the participation of the mast cells in the dynamics of the connective tissue is shown in Figure 9 which has been adapted from 'A mixture of three hydrogenated ergot alkaloids-Sandoz Pharmaceuticals. VASCULITIS, MAST CELLS AND COLLAGEN DISEASES 13 THE MAST-CELL CYCLE Adoptad from Ri1.y. I V S V (4 0 ) Fig. 9.--Riley has proposed a scheme for the participation of the mast cells in the local perivascular events of the connective tissues. Components of the ground substance, derived originally from the fibroblast, are stored in altered form and released following “trauma” to re-enter the cycle via mesenchymal cells. The relationship to collagen fibrogenesis and scar formation is shown. Hiley.4O To illustrate the part played by the mast cell in the local events of the connective tissue, let it be assumed that there is some trauma (mechanical, chemical, thermal or biological) at the point of the arrow on this chart. It will be noted that these events are illustrated as taking place about a capillary shown as an endothelially lined space extending horizontally across the diagram. The mast cells are situated in the pen-capillary area and, when stimulated promptly, release their metachromatic granules into the ground substance. In the case of long-term steroid therapy, these cells may have been constantly stimulated to produce, among other things, heparin. The arthritic patient with hypercortisonism who receives a minor skin injury may have an ineffective extravascular clotting mechanism because the pericapillary area of the ground substance is constantly heparinized. The products of the mast cell or their degradation products or some combination of these may set off a chain of events which we recognize as chronic inflammation. One of the first events is the production of acute water-rich edema and the presence of the histamine will cause capillary dilatation and further accumulation of fluid. These free granules form the free chemotrophic substance and round cell invasion soon follows. Fibroplasia and recognizable collagen begins to take shape to regenerate and repair. In chronic inhnmation the mast cell population further increases and an end point is reached with the formation of an avascular scar tissue composed largely of collagen fibers. Serotonin may enter the picture as the stimulating substance for the pro- 14 SMYTH AND GUM duction of fibrils. The presence of heparin in the ground substance may also play a role in the formation of collagen fibers. Collagen fibers have been shown by Morrione"2 to form in vitro by the action of heparin on solutions of collagen and may be effective in inducing collagen fiber formation when present in concentrations as low as 1:80,000.If the stimulus for production of the perivascular inflammatory reaction is continued over a long period, the ultimate end result is the chronic perivascrilar inflammation and scar. SUMMARY More facts are needed for a final evaluation of the role of the mast cell in the connective tissues, but at present there appears to be some evidence that it plays a part in the deposition of fibrils and scar formation in the connective tissue diseases. Mast cells are known to be in high concentration in certain tissues. They have been shown to either store or synthesize potent biological materials. They can be looked upon as prime movers in the chain of events we know as the inflammatory reaction. Beyond this much remains to be explained. Many of these observations must, at present, remain speculative, but it is hoped that these ideas will stimulate others to study the relationship of the mast cells and their products to the connective tissue diseases. REFERENCES 1. Racdn, C.: Heberden Oration, 1958. Role of hypersensitivity in the patho- matoid arthritis. Ann.Int.Med. genesis of rheumatoid arthritis. Ann. rheumat.Dis. 18:l-7, 1959. 2. Sokoloff, L., McCluskey, R. T. and Bunim, J. J.: Vascularity of the early subcutaneous nodules in rheumatoid arthritis. A.M.A.Arch.Path. 55475, 8. Ehrlick, P.: Beitrage zur kenntnis der 1953. 3 Nashelsky, G. and Smyth, C. J.: Subcutaneous hemorrhages in rhcumatoid patients treated with prednisone. Ann. Rheumat.Dis. 17:7882, 1958. 4. Boland, E. W.:Prednisone and prednisolone therapy in rheumatoid arthritis. J.A.M.A. 160:613-621,1956. 5. Denko, C. M7. and Schroeder, L. R.: Ecchymotic skin lesions in patients receiving prednisone. J.A.M.A. 164: 41-43, 1957. 0. Johnson, R. L., Smyth, C. J., Holt, G. W.,Lubchenco, A. and Valentine, E.: Steroid therapy and vascular lesions in rheumatoid arthritis. Arth.&Rheum. 2:224-249, 1959. 7. Kemper, J. W., Baggenstoss, A. H., and Sloclimb, C. H.: The relationship of therapy with cortisone to the incidence of vascular lesions in rheu- 46: 831, 1957. aniling arbunge and ihrer vorwendung in der microskopischen technik. Arch. mikr.Anat. 13:263-277, 1877. 9. Jorpes, J. E.: On heparin: its chemical nature and properties. Acta Med. Scandinav. 88:427-433,1936. 10. Asboe-Hansen, G.:The mast cell. Cortisone action on connective tissue. Proc.Soc.Exper.Biol.& Med. 80:677- 679, 1952. 11. Bloom, F.:Effect of cortisone on mast cell tumors (mastocytoma) of the dog, Proc.Soc.Exper.Bio1.R Med. 79: 651, 1952. 12. Bloom, G.: Cytological changes in human tissue mast cells after cortisone treatment. Acta Morphol. NeerdlandsScandinavica 1:331-336,1957. 13. Asboe-Hansen, G.: A survey of the normal and pathological occurrence of mucinous substances and mast cells in the dermal connective tissues. Acta Dermat.-venereol. 30:338, 1950. 14. Smith, D. E. and Lewis, Y. S.: Effects of total-body X-ray irradiation on tissue mast cells. Proc.Soc.Exper. VASCULITIS, M A W CELLS AND COLLAGEN DISEASES IS cuspid regurgitation without septa1 defects ), peripheral vasomotor symptoms, bronchoconstriction, and an on tissue mast cells. Acta Morphol. uniisual type of cyanosis. A clinical 4:345-351, 1954. and pathologic syndrome. Am.Hcart 18. KelBnyi, G.: Changes in the mast cells J . 47:795,1954. following X-ray irradiation. Acta 28. Mattingly, ‘r. W.:Functioning carcihlovhol. 3:377-381, 1953. noid tumor - new clinical entity; 17. Gustafsson, B. E. and Cronberg, S.: review of clinical features of nonComparison of the effects of comfunctioning and functioning carcinoid pound 48/80, protomine, and turincluding 38 cases from the literature. pentine oil on mast cell degrandaMed.Ann.District of Columbia. 25: tion. Acta Rheum. Scand. 3:189-202, 304-312, 355-356, 1956. 1957. 18. Riley, J. F.: The effects of histamine- %!. Goble, A. J., Hay, D. R., Hudson, R. and Sandler, M.: Acquired heart disliberators on the mast cells of the ease with argentaffin carcinoma. Brit. rat. J.Path.& Bact. 65471-479,1953. Heart J. 18544, 1956. 19. Benditt, E. R., Wong, P. L., Arase, M. and Heper, E.: Fj-hydroxytryptamine 30. \4cKusick, V. A,: Carcinoid cardiovascular disease. Bull.Johns Hopkins Hosp. in iliast cells. Proc.Soc.Exper.Biol.& 98:13-36,1956. hfed. 90:303-304,1955. 20. Riley, J. F., and West, G. B.: Tissue 31. Olson, T. E. and Gray, S. J.: Serotonin and gastroenterology. Am.J.Gastromast cells. Studies with a histamineenterology 29:280, 1958. liberator of low toxicity (compound 48/80). J.Path.& Bact. 69:269-282, 32. Sjoerdsma, A,, Weissbach, H. and Undenfriend, S.: A clinical physiological 1955. and biochemical study of patients 21. Parratt, J. R. and West, G. B:. 5-hywith malignant carcinoid. Am.J.Med. droxytryptamine and the maphylac20:520, 1956. toid reaction in the rat. J.Physio1. 33. MacDonald, R. A.: A study of 356 139:27-41,1957. carcinoids of the gastrointestinal 22. Asboe-Hansen, G.: The origin of sytract. Report of four cases of the novial niucin. Ann.Rheumat.lXs. 9: carcinoid syndrome. Am.J.Med. 21: 149, 1950. 867, 1956. 23. Cass, R., Riley, J. P., West, G. B., 34. Sauer, W. G., Dearing, W. H. and Head, K. W. and Stroud, S. W.: Flock, E. V.: Diagnosis and clinical Heparin and histamine in mast cell management of functioning carcitumors from dogs. Nature. 174:318noids. J.A.M.A. 166:139-147, 1958. 319, 1954. 24. Benditt, E. P., Arase, M. and Roeper, 35. Thorson, A. H.: Studies on carcinoid disease. Acta Med.Srandinav.Supp. hi. E.: Histamine and heparin in iso334, 1958. lated rat mast cells. J.Histochem. 4: 38. Zarafonetis, C. J. D., Lorber, S. H. and 419-420, 1956. Hanbon, S. M.: Association of func25. Mota, L., Beraldo, W. T., Ferri, A. G. tioning carcinoid syndrome and and Junqueira, L. C. U.: Intracellusclerodemia. Am.J.M.Sc. 261:1, 1958. lar distribution of histamine. Nature. 37. hlacDonald, R. A., Robbins, S. L. and 174:698, 1954. Mallory, G. K.: Dermal fibrosis follow26. Lembeck, F.: Shydroxytryptamine in a ing subcutaneous injections of serocarcinoid tumor. Nature. 172:910, tonin creatinine sulfate. ProcSoc. 1953. Exper.Biol.& Med. 97:334-337, 1958. 27. Thorson, A., Biorck, G., Bjorlanan, A., and Waldenstrom, J.: Malignant car- 38. Sjoerdsma, A., Waalkes, T. P., and Weisbach, H.: Serotonin and histacinoid of the small intestine and mine in mast cells, Science 125-1202metastases to the liver, valvular 1203, 1957. heart disease of the right side of the heart (pulmonary stenosis and tri- 39. Scherbel, A. L. and Harrison, J. W.: Biol.& Med. 82:208-212,1953. 15. KelBnyi, G.: Effect of cytotoxic agents 16 40. 41. 42. 43. 44. 45. 46. 47. 48. 49. 50. 51. SMYTH AND CUM Response to serotonin and its antago1084-1089, 1957. nists in patients with rheumatoid arth- 52. Morrione, T. G.: The formation of colritis and related diseases. A n g i ~ l ~ g y lagen fibers by the action of heparin 10:29-33, 1959. on soluble collagen. An electron miRiley, J. F.: The Mast Cell. Edinburgh croscope study. J.Exper.Med. 96:107, & London, E. and S . Livingstone, 1952. Ltd, 1959. 53. ?myth, C. J. and Gum, 0. B.: Mast James, J. and McDonald, J. R.: Mast cells in connective tissue diseases. Cells. Their distribution in various Arth.& Rheum. 1:178-180, 1958. human tissues. A.M.A.Arch.Path. 45: 54. Jorpes, J. E.: Heparin in the Treatment 622-634, 1948. of Thrombosis. ed. 2. London, Oxford Ilavies, D. V.: The staining reaction University Press, 1946. of normal synovial membrane with 55. Quensel, U.: Studien iiber die gewebsspecial reference to the origin of mastzellen. Acta path.et microbiol. synovial mucin. J.Anat. 77:160, 1942Scandinav.supp1. 16358-375, 1933. 43. 56. Heilmeyer, L., Kiihn, H. A,, Clotten, R. Mota, R., Ferri, -4. G. and Yoneda, S.: and Lipp, A.: Metastasierendps DiinnThe distribution of mast cells in the darmcarcinoid mit nachwels von digestive tract of laboratory animals; oxyindolesigsaure in blut und ham its bearings on the problem of the durch hochspannungsele ktrophorese. Duetsche med.Wchnschr . 81 :501localization of histanline in tissues. Quart.J.Micr.Sa. 97:231-256, 1956. 508, 1956. Wislocki, G.B., Bunting, H. and Demp57. Scherbel, A. L.:The effect of Marsilid sey, E. W.: Further observations on in patients having rheumatoid arththe chemical cytology of megakaryoritis; the theoretical causal role of certain amine oxidases. J.Clin.& Excytes and other cells of hemopoietic per.Psychopath. 19:118, 1958. tissues, Anat.Rec. 98:527, 1947. Michels, N. A.: T h e mast cells. In H. 58. Wooley, D. W.: Probable evolutionary relationship of serotonin and indoleDowney Handbook of Haematology. acetic acid and some practical conNew York, Hoeber, 1958. sequences therefrom. Nature 180:630, Kelsall, M. A. and Crabb, E. D.: Lym1957. phocytes and Mast Cells. Williams 59. Asboe-Hansen, G.:Sclerodenna in cnr& Wilkins, Baltimore, 1959. cinoid syndrome. Acta dermat. Jorpes, J. E., Holmgren, H. and WilVenereol. 39:270-273, 1959. ander, 0.: Uber das vorkommen von 60.Deicher, H. R. G., Holman, H. R. and heparin in den gufasswanden und in Kunkel, H. G.: Anti-cytoplasmic facden angen. Ztschr.mikr.-anat.Forsch. tors in the sera of patients with sys42:279-300, 1937. temic lupus erytheniatosus and cerSundberg, M.: On the mast cells in the tain other diseases. Arth.& Rheumat. human vascular wall; a quantitative study on changes at different ages. 31-15. 1960. Acta path.& microhiol.scandinnv.supp1. 61. Stuart, E. G.: Mast cell response to anaphylaxis. Anat. Rec. 112:394, 107, 1955. 1952. Staemmler, M.: Untersuchung iiber 62. Carter, P. B.,Higginbotham, R. D. and vorkommen und bedentung der histoDougherty, R. D.: The local response genen mastzellen in menschlicken of tissue mast cell to antigen in korper unter normalen und pathosensitized mice. J.Immuno1. 79~259logischen verthaltnissen. Frankfurt 264, 1957. Ztschr.Psth. 25391-435, 1921. Riley, J. F. and West, G. B.: The 63. Mota, I., and Vugman, I.: Effect of anaphylactic shock and compound presence of histamine in tissue mast 48/80 on the mast cells of the cells. J.PhySi01. 120:528-537, 1953. guinea pig lung. Nature 177:427, nollak, 0. J.: Mast cells in the circula1956. tory system of man. Circulation 16: VASCULITIS, M A S T CULS AND COLLAGEN DISEASES H. and Mota, I.: The mechanism of anphylaxis: specificity of antigen-induced mast cell damage in anaphylaxis in the guinea pig. Immunology 2:31, 1959. 84. Humphrey, J. 17 65. Mota, I.: Effect of antigen and OWlamine on mast cells and histamine content of sensitized guinea pig tissues. J.Physio1. 1.#7:425-438, 1959. Charley 1. Smyth, A.B., M.S., M.D., Associate Professor of Medicine, Director, Arthritis Clinic, Uniiiersity of Colorado School of Medicine. Oren B. Gum,Ph.D., M.D., Assistant Professor of Medicine, Unioersity of Colmudo School of Medicine.