Regulatory Factors in Hoi*monally Induced Polyarteritis Nodosa in Rats By JAMES C. PERRYAND NANCYB. PERRY Diffuse polyarteritis was induced in rats by the long-term administration of estrogen and follicle-stimulating hormones. Tumorous changes also appeared in the anterior pituitary gland which were associated with cortical adenoma of the adrenal. It was postulated that mineralocorticoid secretions of the latter played a role in the pathogenesis of the vasculitis. Polyarteritis nodose esseva inducite in rattos per le administration a longe vista de estrogeno e hormones folliculo-stimulante. Alterationes tumorose appareva etiam in le glandula pituitari anterior, associate con adenoma cortical del glandula adrenal. Esseva postulate que secretiones mineralocorticoide del glandula adrenal haheva un rolo in le pathogenese del vasculitis. T HE HOHMONAL INDUCTION of periarteritis nodosa, or as it is now more preferably termed, polyarteritis nodosa, in rats has been described . ~ early work indicated that treatment of male rats by the senior a ~ t h o r This with estrogen at three-day intervals for 30 days followed by treatment with follicle-stimulating hormone (FSH) at the same intervals for another 30 days eventually resulted in widespread polyarteritis. Definite indications of the disease commenced approximately four months after treatment and became progressively more severe up to 12 months thereafter if death of the animals did not intervene. This investigation revealed that the testes, pituitary and adrenal glands were involved as regulatory factors in the causation of the resultant polyarteritis. The present work was undertaken in an attempt to investigate the respective roles of these organs in bringing about the malady. PROCEDURES Male rats of both Sprague-Ihwley and Holtzman strains approsinlately 6 months of age were used in this investigation. Each group studied was made up of animals from each strain. Because of the large numbers of animals studied, no attempt to use litter mates was made. Animals used to repeat the original investigations as well as all the respective controls were treated at the same time. All animals used in this study with the exception of hypophysectomized groups were in the same weight range, approximately 375 to 400 grams. The rats used to investigate the role of the testes in the development of polyarteritis were injected at the same age, whether surgery preceded or followerl iniections. Similarly, hypophysectomized animals were injectcd at the same i ~ g c I(wel, w h ~ t h e r siirgrry was From the Department of Biology, iuarqrrette Ilni~ersihj,Milwaukee, W i s . This investigation was supported bv research grants A 1134 and A 7134 ( c ) from the National Institute of Arthritis and Metabolic Diseuses of the Nutwnal Institictes of Health, United States Public Heulth SertSce. In addition to the aid aflcwded by the Lbaited States Public Health Grants the authors &end their gratitude to Dr. Robert Gaunt of Ciba Pharmaceuticals for generous s u p p h of Ovocylin and D C A and to Dr. lrby Bunding of Amour Laboratories for F S H . Gratitude is also extended to Dr. J . F . Kuzma, Director of the Department of Pathology, Marquette University, for his interest in und interpretation of pathologic results. W e arc further indebted to Mr. Stmaislaus Rutajczak for his aid in in-epciring the nhotogrnphs. 244 REGULATION OF PERIARTERITIS NODOSA 245 antecedent or subsequent to hormonal treatment. The average weight of these animals prior to surgery was approximately 260 grams. (Hypophysectomy is rather difficult in older and heavier animals ) . Dosages of hormones were the same as in the original experiments unless otherwise indicated. Injections of 0.1 ml. estradiol proprionate (Ovocylin-Ciba, 1 mg. per ml.) were given subcutaneously every third day until 10 had been given. These were followed immediately with similar injections of 0.1 ml. FSH ( A r m o u r 3 0 0 gammn per ml.) every third day until 10 had been given. Criteria for the diagnosis of polyarteritis were as follows. (1.) The presence of grossly visible nodules along the courses of greatly thickened arteries, particularly mesenteric and splenic arteries, was taken as positive proof of polyarteritis, provided that histologic sections revealed the characteristic pathology of the malady. (2. ) In the absence of grossly visible appearances, skip serial microscopic sections stained with hematoxylin and eosin were made from the testes, pancreas, thymus and likely intestinal regions. The presence of numerous typically polyarteritic arterioles in these organs was considered as a positive response. In addition to the above histologic studies, serial sections of pituitary glands stained with hematoxylin and eosin were routinely made and checked for the presence of neoplastic alterations. Serial sections of one adrenal gland of each animal were likewise prepared and checked for the presence of adenomata. T h e alternate adrenal glands were chromated and checked to determine whether their tumors were of cortical or medullary origin. Finally, hematoxylin and eosin sections were made from kidneys of animals in which these organs appeared cystic or otherwise abnormal. Specialized staining procedures were employed in the study of various organs where such procedures were indicated. Pertinent data gathered from the above studies are summarized in table 1 (page 250). RESULTS Repetition experiments.-Repetition of the original investigations4 resulted in confirmation of the findings of polyarteritis in 25 out of 30 animals employed. Although some of these animals succumbed during the early months following treatment, they showed signs of incipient polyarteritis. The majority survived beyond 8 months and exhibited severe and extensive involvement of the arteries of many organs. Four types of control animals given respectively estrogen, FSH, estrogen plus normal saline, and sesame oil did not exhibit polyarteritis six or more months after treatment. Since controls were treated either with only one of the single hormones employed or with solvents for the hormones as indicated, there was reasonable assurance that estrogen plus FSH was the effective stimulus of the disease. In all ensuing experiments at least 10 animals were used in each group. Experiments on the role of the testes.-Investigations on the role of the testes were studied in vasoligated-vasectomized and in castrate animals. Ten animals were vasoligated and vasectomized and thereafter given the hormonal treatment to determine whether this procedure would alter the original findings. Polyarteritis developed with the same regularity and severity as in the above intact animals. Another group of animals was castrated and after recovery was given the standard hormonal treatment. Polyarteritis developed, perhaps somewhat more slowly, but with severity at least equal to that found in intact rats. An additional group of rats was castrated immediately after completing the hormonal treatment. An equal number was castrated four months after hormonal treatment. In both instances polyarteritis developed with the same sev- 246 JAMES C. PERRY AND NANCY B. PERRY erity and extent as in intact animals. The development of the disease, whether castration was performed either before or after treatment, is evidence that the testes do not play a direct role in its causation. Although testicular tubular atrophy is the usually observed response to estrogen-FSH treatment in intact animals, it appears to be only coincidental to the development of polyarteritis. It is noteworthy that some castrated animals seemed less able to survive than their intact counterparts, once polyarteritis had become severe. Extensive studies on animals employed to confirm original findings and on vasectomized and castrated rats shed new light on the onset and progress of the disease. The presence of the disease is most easily and usually detected first in the arterioles of the testes (fig. 1).It may appear elsewhere whether the testicular arterioles are involved or not. It has been found in the arterioles of the alimentary canal, pancreas and thymus as early as four months after treatment. With the lapse of time the larger arteries of these organs become involved. The thickened walls and nodular swellings of these vessels present a striking appearance (fig. 2 ) . Not infrequently larger arteries may be affected with little or no indication of testicular xteriolar involvement. Many of the afflicted animals experienced great difficulty in movement as the disease became severe. The hind legs and feet wre particularly afflicted. The animals could only drag these swollen arthritic-like members as they slowly moved about their cages. Grimstvedt and Johansenii have reported similar difficulties in man resulting from periarteritis involving the spinal cord. That the locomotor difficulties observed in our rats are caused by a similar condition has not been determined to date. FIG.1 . S e c t i o n of testis of rat, autopsied eight months after treatment, showing extensive mass of polyarteritic arterioles and atrophied seminiferous tubules. x 100; H. & E. stain. REGULATION OF PERIARTERITIS NODOSA 247 FIG.2.-Abdominal organs of rat, autopsied nine months after treatment, showing extensive and massive polyarterities nodosa in inesenteric and other arteries. Histochemical tests for the presence of iron-containing pigment were carried out on sections of the spleen, lymph nodes, and even hypertrophied hemolymph nodes in experimental animals. In sections of these structures stained with hematoxylin and eosin large numbers of macrophages laden with brownstaining granular pigment were present. The tests for iron revealed that the pigment contained iron; most likely it was hemosiderin. Presumably lymph node hyperplasia, so commonly observed in polyarteritic animals, is due in part to the proliferation of macrophages formed in response to extensive cellular necrosis, particularly of erythrocytes, associated with the progress of polyarteritis. The pituitary glands of animals developing even incipient polyarteritis become hypertrophied. Frequently they may exlibit a tenfold increase in weight and volume compared with nonnal glands. Their cells lose their typical vesicular arrangement, and fibrinoid material appears in expanded sinusoids of the gland. Tumorous cells, often aggregated into small nodular areas, are commonly present. In approximately 50 per cent of the animals with polyarteritis the tumorous cells have many of the morphologic characteristics of malignancy. Many nuclei are greatly enlarged and exhibit irregular distribution and amounts of chromatin material and contain one or more prominent nucleoli. Many cells and nuclei are polymorphic and hyperchromatic in addition to exhibiting variable size and shape. Enlarged niultinucleated cells are numerous, and many of these nuclei are incompletely divided. Both typical and bizarre mitotic figures are frequently observed (fig. 3). In certain instances tumorous cells have been observed in both intermediate and posterior lobes (fig. 4 ) . 248 JA\IKS (:. PERHY AND NANCY R. PERRY FIG.3.-Srnall area of massivc! neoplasm in pitiiitiiry !if rat having polyarteritis. Nett: large nuc!leoli, hyperchromatic staining and irrc-giiliir distrilmtion of chromatin, incoinpleti~ly divided multinucleatcd cells and initotic tigiires. Iron hc~iiintoxylinstain; x 430. Metastasis of the tumorous cells to othcr organs has not lieell o b s c w c d to date. Neoplastic changes in the anterior lobe have bccn observcd in animds autopsied during the first few weeks after hormonal treatment. Constant and readily detectable indicatioiis of the onset of the syntlromc are found in the appearance of adrenal cortical adenomata. These tumors do not exhibit the chromaffin reaction and therefore are considered of cortical origin. These adenomata are usually located at the boundary of the medulla and cortex. Occasionally they may appear at the periphery of the gland. Two or more tumors are often present in the same gland. Usually, but riot always, adenomata are found in both adrenals of an animal. On occasion d y small groups of scattered tumorous cells may be preseent in many areas of the cortex. As post-treatment time increases, the tumors become very extensive and account for a consistently observed enlargement of the glands to two or more times their normal volume. The kidneys in approximately onc half thc animals with poly'trteritis are visibly enlarged and frequently cystic. Histologically many tubules are seen to be dilated and filled with hyaline casts. The glomeriilar arterioles art' frtquently necrotic, but glomeruli themselves 'ire not noticeably abnormal. Extensive intertubular fibrotic areas are normally present. Since the pituitary normally regnlates the :id rend glands and since it w a s greatly altered in experimental animals, the following invcstigations were undertaken. Experiments on h!/pophysecfoniizetl crtiii~irils.-Twelve rats were hypophysectomized and, aftcr recovc~r\rfrom surgery. \ w r c giwn tlw risrial Iior- REGULATION OF PERIAR'IERITIS NODOSA 249 FIG.4.--Section of pituitary from polyarteritic rat showing portions of intemiediate and posterior lobes. Note prominent mass of tumor cells in posterior lobe. Main tumor mass was was found in anterior lobe. x 430; H. & E. stain. monal treatment. Twelve animals were hypophysectomized one week after the completion of the hormonal treatment. PolyarteIitis did not develop in either group up to eight months following treatment. Both groups of animals failed to survive beyond this time except for one rat that died after nine months. Untreated hypophysectomized controls died in like periods of time. The adrenal glands of both control and experimental hypophysectomized animals were about one fourth the size and weight of those of iintreated animals. Adenomata were uniformly absent. Testicular and other arterioles gave no indication of polyarteritis. It was therefore concluded that an intact pituitary was necessary for the development of the adrenal cortical adenomata and hence of poiyarteritis under the conditions of these experiments. Experiments with I)CA.-Since Selye5v6had observed the development a€ periarteritis ( polyarteritis ) after the administration of DCA ( desoxy-corticosterone-acetate ) to unilaterally nephrectomized rats given high salt and protein diets, this hormone was administered to rats following the usual estrogenFSH treatment. It seemed that DCA might augment this treatment even in the absence of the conditions of Selye's experiment. DCA did not hasten the onset nor enhance the polyarteritis to a greater extent than w'ts obtained without its use. This result would indicate that internal regulating factors were operating to bring aboiit the disease independently of the extrinsically mpplied mineralcorticoid hormone which under diverse experimental conditions is also capable of inducing polyarteritis. 250 JAMES C. PERRY AND NANCY B. PERRY TABLEl.-Summosy of Pertinent Responses of Erperimetrtul und Control Animals Total Numbers Total with Macroscopic Pobarteritia Total with Microscopi? Polyartentla Only Total witb Tumorous Pituitaries Total with Adrenal Adenomata Total witb No Changes Detected. 17 8 21 23 5 6 3 9 9 1 7 2 7 8 1 6 1 7 7 3 8 0 7 8 2 5 3 6 6 2 0 0 0 12 0 0 0 12 0 0 0 0 10 0 0 0 0 10 0 0 0 0 10 0 0 0 0 10 and Treatment Per Group 30: Repetition estrogen and FSH 10: Vasect. plus both hormones 10: Castrated before receiving both hormones 10: Castrated after receiving both hormones 10: Castrated 4 Mos. after receiving both hormones 10: Given DCA plus both hormones 12: Hypophysect. before homiones given 12: Hypophysect. after hormones given 10: Controls estrogen only 10: Controls FSH only 10: Controls estrogen and saline 10: Controls sesame oil *Some experimental animals died prematurely. Above responses are from animals autopsied 6 to 9 months after treatment. DISCUSSION The development of polyarteritis nodosa following the administration of estrogen and FSH has been regularly and consistently confirmed in well over 100 animals including those in which it was originally discovered and others still under investigation. It is, at the present writing, appearing in female rats given a like treatment. The fact that it does not become severe and grossly evident until several months after the cessation of treatment is evidence that the mechanisms involved in its causation are endogenous and not merely the direct actions of the administered hormones. Certain conclusions as to the mechanisms involved may be advanced. Although the testes of afflicted animals seem to bc preferentially affected, they do not play a direct role in the causation of the disease. It is significant that polyarteritis did not develop in the absence of the pituitary in animals given estrogen-FSH treatment. Chi the other hand alteration of the pituitary glands in intact animals which were given this treatment seems to play a primary role in the etiology of this type of polyarteritis. The presence of adrenal cortical adenomata in animals with tumorous pituitaries suggests a specialized adrenocorticotropic action of the nature of a REGULATION OF PERIARTERITIS NODOSA 251 tumorogenic humoral agent. The development of polyarteritis which increases in severity with the growth of the adenomata points to them as the immediate causative agents of the disease. Since Selyea has shown that the mineralocorticoid DCA, when given exogenously under diverse conditions, can provoke the disease, it would seem logical to infer that these adenomata are endogenously producing a substance capable of inducing polyarteritis. Endogenous production of the mineralocorticoid, aldosterone, as a natural product of cortical activity is proof of the ability of the gland to form high titers of this hormone.2 Moreover, cortical adenomata have been found in man to be the source of hyperaldosteronism.' Glucocorticoids, on the other hand, are known to alleviate polyarteritis. It seems likely that the cortical adenomata observed in these investigations are the sources of polyarteritic-stimulating mineralocorticoids which in these rats may be \very similar to if not identical with aldosterone. Final proof of this conclusion must await the availability of increased supplies of this hormone or its extraction and identification as a product of the adenomata present in rat adrenals as observed under similar experimental conditions. The foregoing interpretation does not preclude the possibility that polyarteritis may often be the result of factors causiiig renal hypertension. Selye6 reports that periarteritis nodosa develops in animals with an endocrine kidney. In this and in other instances the kidney liberates a renal pressor substance (RPS) which is apparently capable of producing polyarteritis even in the absence of increased mineralocorticoids. Selye".s has also shown that overdosage with DCA under previously mentioned conditions results in periarteritis (polyarteritis) nodosa. In both instances there is an alteration of renal function. It may be presumed that polyarteritis may develop concurrently with renal hypertension, but it may likewise appear in the absence of renal hypertension due to the effects of mineralocorticoids on renal function. Aldosterone and other mineralocorticoids characteristically exert a sodiumretaining, potassium-eliminating action on kidney function.2 The pathologic changes observed in the kidneys of animals studied in this report do not completely duplicate those observed in cases of reiial hypertension. It may be concluded that the polyarteritis observed in this investigation is more likely due to mineralocorticoidal action on the kidney rather than to renal hypertensive factors. S U M U Y Polyarteritis nodosa has been shown to develop in male rats several months after treatment with estrogen followed by treatment with follicle-stimulating hormone of the pituitary. Evidence is presented which indicates that tumorous alterations of the pituitary that occur following this hormonal treatment are responsible for the appearance of adrenal cortical adenomata. The experimental production of polyarteritis following exogenous administration of the mineralocorticoid ( desoxy-corticosterone-acetate adds plausibility to the conclusion that these adenomata may liberate increased titers of mineralocorticoid -most likely aldosterone-which are responsible for the development of the polyarteritis nodosa observed in this investigation. .- 252 JAMES C . PERRY AND NANCY B. P E R H r REFERENCES 1. Conn, J. W. and Loui-, L. H.: Primary aldosteronism: A ncw clinical entity. Tr. A. Am. Physicians. 63:215-233, 1956. 8. Gaunt, R., Renzi, A. 4 and &-Art, J. J.: Aldosterone-A review. J. Clin. Endocrinol. 153321-846, 1955. 3. Grimstvtdt, M. and Johansen, A.: Periarteritis nodosa with predominant symptoms from the spinal cord. Act. Path. et Microbiol. Scand. 40:81-88, 1957. 4. Perry, J. C.: Experiincntal indwtion of periarteritis nodosa in white rats. Proc. Soc. Expl. Biol. & hled. 89:200-004, 1955. 5 . Selye, H.: The general adaptation syridrome and the diseases of adaptation. J. Clin. Endocrinol. 6:117-230, 1946. 6. -: Stress (The Physiology and Pathology of Exposure to Streis) Montrea!, Ada, Inc., 1950. James C. Perry, Ph.D., Professor of Zoologil. College of Liberal Arts and Graclunte School, Mnrqziette Ziniuersity, Milwciukee, Wis. Nancy B. Perry, Personcil Technicim, Alilwtzrikee, M’is.