Variations in Histologic Pattern and Functional Effects of a Transplantable Adrenal Cortical Carcinoma in Intact, Hypophysectomized, and Newborn Rats1
KATHARINE C. SNELL and HAROLD L. STEWART, Laboratory of Pathology, National Cancer Institute,2 Bethesda, Maryland
Summary
A transplantable adrenal cortical carcinoma of the rat is described. During 7 generations of transplantation into intact, mature, male and female Osborne-Mendel rats, this tumor line, designated Adrenal cortical carcinoma 494, maintained a consistently similar histologic pattern of cords of uniform, polygonal cells. All tumor-bearing rats developed polyuria, polydipsia, and degenerative changes in the distal convoluted tubules of the kidney. They also showed atrophy of the adrenal glands, sex organs, lymphatic tissue, and pituitary gland. A subline, tumor 494H, developed after transplantation of tumor 494 into hypophysecto- mized rats. It differed in histologic pattern from tumor 494 in that it contained large, bizarre cells in addition to the uniform, polygonal cells of tumor 494. When tumor 494H was returned to intact, adult, male and female rats, it retained its own distinctive histologic pattern, yet it also induced polyuria, polydipsia, and renal tubular degeneration. Rats of both sexes inoculated with tumor 494H developed hyperplastic mammary glands filled with milky secretion. In the females, the uterine horns were enlarged and the vaginal epithelium mucified; in the males, the testes and accessory sex organs were extremely atrophied. A second subline, tumor 494NB, similar in histologic pattern to tumor 494H, developed when tumor 494 was transplanted to newborn rats. This subline was like subline 494H in biologic behavior, and it produced a similar effect on the organs of tumor-bearing male and female rats .- J. Nat. Cancer Inst. 22: 1119-1155, 1959.
Fifty-three of 59 inbred NIH Osborne-Mendel-strain rats 18 months of age or older were found to have primary adenomas or carcinomas of the cortex of the adrenal gland. Seventeen of 26 rats of this strain, 12 to 17 months of age, had hyperplastic nodules or small adenomas of the adrenal cortex. This observation is contrary to the generally accepted opinion that adrenal cortical tumors rarely occur in the rat (1-3). The occurrence of adrenal tumors in our Osborne-Mendel rats may be a manifestation of a general endocrine imbalance, since tumors of the adrenal gland,
1 Received for publication January 13, 1959.
” National Institutes of Health, Public Health Service, U.S. Department of Health, Education, and Welfare.
ovary, mammary gland, or pituitary gland are often found in the same animal. It has not been determined, however, whether all the tumors of the adrenal cortex are hormonally active.
Three adrenal cortical tumors, found at necropsy of Osborne-Mendel rats over 18 months of age and diagnosed histologically as carcinoma, were transplanted. Two of the 3 tumors grew progressively, and 1 of these, Adrenal cortical carcinoma 494, has been maintained and studied in transplantation. The present paper describes the histo- pathology and effects of tumor 494 during 7 generations of transplanta- tion into intact, mature rats, and the transformation both in morphology and physiologic function this tumor underwent when transplanted to hypophysectomized and to newborn rats.
Experiments and Results
Pathologic Findings in Original Tumor-Bearing Animal
Adrenal cortical carcinoma 494 was found in the right adrenal gland of a 25-month-old female inbred Osborne-Mendel rat killed and necrop- sied on February 2, 1955. Twenty-three months previously the rat had been inoculated intraperitoneally with a piece of gastric mucosa excised from a rat fed a diet containing catechol, for 17 months. No evidence of the transplanted gastric mucosa could be found in this rat or in any of 20 rats similarly treated. The incidence of adrenal cortical tumors was the same in the rats that received transplants of gastric mucosa as in 39 untreated rats of similar age. For these reasons, Adrenal cortical carcinoma 494 was assumed to be spontaneous in the female rat concerned here.
The right adrenal gland of the original tumor-bearing rat consisted of a soft, dark-red mass 1.0 cm. in diameter. The mass was divided; the lateral half was used for transplantation and the medial half for histologic study. The medial half was fixed and then divided into 2 parts by cutting it longitudinally in a plane perpendicular to the plane of division made at necropsy, thus supplying 2 quadrants from the original spherical mass. The freshly cut surface of each quadrant was then blocked in paraffin face down so that histologic sections cut from these blocks consisted of adjacent semicircles of tissue from the mid-portion of the medial half of the right adrenal gland. It was recognized that these sections included only a single margin immediately adjacent to the portion of tissue used for transplantation. Therefore, the paraffin blocks were melted, and the 2 quadrants reoriented and reblocked in order to obtain histologic sections from the surface closest to the tissue used for transplantation. Descrip- tion A is based on study of the histologic sections cut from the 2 original paraffin blocks; description B, on study of sections cut from the re- embedded tissue.
Description A: As shown in figure 1, sections revealed 2 histologically different tumors of the adrenal cortex (tumor #1 and tumor #2), a portion of the adrenal medulla compressed between the 2 tumors, and a rim of surviving adrenocortical tissue lying beneath the capsule.
Tumor #1 was roughly triangular, with the apex directed toward the center of the gland and the base toward the capsule. The tumor tissue was separated from the capsule by a thin layer of zona glomerulosa, and, at one point, by cells of the zona fasciculata also. Laterally, thin strands of reticulum separated the tumor from the medulla on one side and from tumor #2 on the other. Within the reticulum were enmeshed large retic- uloendothelial cells containing both iron and periodic acid-Schiff-positive pigment. The cells of tumor #1 (figs. 1 and 2) were arranged in cords, from one to several cells in thickness, that were bounded by delicate strands of reticulum and collagen and were separated by blood sinuses. The cytoplasm of the tumor cells was homogeneous or finely granular and usually contained vacuoles. It was colored with both eosin and hematoxylin so that this tumor appeared darker than tumor #2. Approxi- mately one third of the cells of tumor #1 contained cytoplasmic granules that were purplish red in the periodic acid-Schiff preparation. The nuclei were small and round and contained fine chromatin granules and from 1 to 3 nucleoli. Some cords of tumor cells contained 5 times as many nuclei as would be expected in cell cords of the same size in the normal zona fasciculata. A few tumor cells measured as much as 30 u in diameter; their nuclei were correspondingly large and were hyperchromatic and irregular in outline. One mitotic figure was observed in every 4 or 5 high-power fields.
Tumor #2 comprised approximately two thirds of the section of the right adrenal gland available for examination (fig. 1). Except where it adjoined the medulla or the first tumor, it was separated from the capsule by a small amount of compressed adrenocortical tissue resembling the zona glomerulosa. In a few areas beneath the capsule, the zona fasciculata could be identified also. The deep aspect of tumor #2 contained large areas of hemorrhage and necrosis and widely dilated, thrombosed vessels. Parts of the tumor were composed of cords of polygonal cells that were larger and had more cytoplasm than the cells of the normal zona fasciculata (fig. 3). Occasional cords of tumor cells appeared to blend and intermingle with cells of the zona glomerulosa that were compressed against the capsule of the gland. Other parts of the tumor were composed of exces- sively vacuolated cells with small, deeply colored, peripherally located nuclei (fig. 4). No mitotic figures were observed in this tumor.
Description B: Sections from both re-embedded blocks revealed only tumor #2 and a thin rim of compressed adrenocortical tissue consisting of zona glomerulosa, and, in one area, of zona fasciculata also.
From this study we are certain that tumor #2 was contained in the por- tion of the adrenal gland transplanted. However, we do not know whether any other tumor was contained in that portion of the adrenal gland. Since tumor #1 did not reach the original line of division in any section examined, we are certain that we did not transplant any portion of it.
Other pathologic findings in the original tumor-bearing rat included atrophy of the left adrenal gland, 2 fibroadenomas of the mammary gland, a granulosa-cell tumor of the left ovary, a small chromophobe adenoma of
the pituitary gland, bronchiectasis, bilateral otitis media, and fatty change of the liver. The kidneys showed chronic nephritis and slight degenerative changes in the cells of the distal convoluted tubules.
The left adrenal gland was only half normal size. Cortical atrophy was pronounced (fig. 5). The layers of the cortex were poorly differentiated, although a few cords of cells, probably of the zona fasciculata, could be distinguished. The zona reticularis seemed to be the least atrophied and was the only zone containing an appreciable amount of fat, as demon- strated by Oil Red O. Within it, large reticuloendothelial cells containing both iron and periodic acid-Schiff-positive pigment were intermingled with well-defined strands of reticulum and collagen.
Transplantation to Intact, Adult Rats: Tumor Line 494
Pieces of the original adrenal cortical tumor were transplanted sub- cutaneously into 3 female rats 2 months of age. A total of 29 male and 23 female adult, intact, inbred Osborne-Mendel rats was used in the study of 7 transplant generations of tumor 494. From the outset the tumor grew successfully in all these host rats. Eight male and 8 female non-tumor- bearing rats within the same age range served as controls.
The first-generation transplant required 19 months to reach a di- ameter of 1.0 cm., but beginning with the second generation the trans- planted tumor reached a diameter of 1.0 cm. within 112 to 2 months after inoculation. Tumor-bearing rats from the second generation on, unless killed earlier, died, about 6 months after inoculation, from massive hemor- rhage following ulceration of the tumor through the skin. Most animals were killed from 312 to 512 months after inoculation. The rats gained little or no weight after the transplanted tumor became palpable; terminally most of them lost weight even though their food intake was normal and water was always available. Small metastatic tumor nodules were found in the lungs and liver of a few tumor-bearing rats that survived for 5 months or more. The sex of the host did not appear to affect the growth rate, and the tumor grew as well in castrated as in intact animals. In- jections of ACTH administered to a group of 8 rats during the first 6 weeks following tumor inoculation neither shortened nor lengthened the latent period.
Polydipsia and polyuria developed in all host rats about 10 weeks after inoculation (text-fig. 1). Once these symptoms appeared, increase in their severity roughly paralleled the growth of the transplanted tumor. By the time the tumors measured 3 cm. in diameter, most rats were excreting between 20 and 30 cc. of urine per day, and an occasional rat excreted as much as 60 cc. As polyuria increased, the rats drank more water, often as much as 70 cc. per day. The specific gravity of the urine decreased from an average of 1.037 to 1.012. Untreated control rats of similar age had an average urine excretion of 8.0 cc. and a water intake of approxi- mately 36 cc. per day. Specific gravity on 24-hour urine specimens from 8 control rats varied between 1.043 and 1.031.
One female and 5 male polyuric rats bearing transplanted tumors of
28
AVERAGE 24 HR. URINE VOLUME (CC.)
MALE RATS INOC. Č TUMOR
24
FEMALE RATS INOC. C TUMOR
UNTREATED CONTROLS
20
16
DAY OF INOC.
12
TUMOR PALPABLE
8
4
0
0
4
8
12
16
20
24
28
WEEKS
Broken line: 2 male tumor-bearing rats
Dotted line: 2 female tumor-bearing rats
Solid line: 1 male and 1 female non-tumor-bearing rat
Measurement of daily urine excretion was begun when all rats were 6 weeks old; 2 weeks later the 4 experimental rats were inoculated with tumor 494. Measurement of daily urine excretion was continued until the experimental and the control rats died or were killed.
The onset of polyuria in male rats occurred 2 weeks after the transplanted tumors became palpable. In females, a sharp rise in urine excretion was noted 4 weeks after the tumors became palpable, but the urine volume did not exceed 10 cc. per day until 3 weeks later. Polyuria in tumor-bearing rats increased with growth of the transplanted tumors. Urine excretion of untreated control rats remained fairly constant and did not rise with increasing age and weight of the animals.
the fifth generation were given intramuscular injections of 0.5 cc. of pitres- sin tannate in oil, twice a day for 3 days. No change in the average urine excretion of the rats was noted during or following pitressin treat- ment.
The gross and histologic appearance of all the 52 transplanted tumors of line 494 studied in the course of 7 generations was similar (fig. 6). The tumors were all encapsulated, nodular, and well vascularized. The tumor tissue was reddish brown, soft and friable, and exuded blood from the cut surface. The central portion, especially of large tumors, was usually necrotic. Some tumors contained areas of calcification. The cells of the tumor were arranged in cords bounded by thin strands of
collagen and reticulum and separated by vascular spaces. Sometimes the tumor cells occurred in sheets. The transplanted tumors were composed of cells that were nearly all polygonal and of about the same size. The nuclei were round and basophilic, and the cytoplasm faintly eosinophilic, finely granular, and sometimes vacuolated. One or 2 mitotic figures were found in every other high-power field. Frozen sec- tions stained with Oil Red O revealed that the cytoplasm contained fat droplets. The morphology of the tumor was consistent with a derivation from the cortex of the adrenal gland, and, taking into consideration the differences that may be manifest between a primary tumor and its transplant generations, this tumor could have been derived from tumor #2 found in the original tumor-bearing rat (figs. 3 and 4).
The kidneys of many tumor-bearing rats appeared slightly larger and softer than normal. Histologic sections revealed degenerative changes in both nuclei and cytoplasm of cells lining the distal convoluted tubules (figs. 12 and 13). This lesion was particularly well brought out by the periodic acid-Schiff reaction. In some of the distal convoluted tubules all the epithelial cells appeared degenerated; in others, some cells appeared to be normal and some degenerated, while in a few, all the cells appeared normal.
The cytoplasm of the affected cells was pale, swollen, and vacuolated. No fat, glycogen, or mucus could be demonstrated by special stains. The cytoplasm bordering the lumen appeared frayed or torn, while that near the basement membrane usually stained more deeply and contained numerous small, round granules resembling mitochondria. Linear mitochondria-like structures perpendicular to the basement membrane were noted in the cells of some distal convoluted tubules. None of these tubules was lined by the flattened type of hyperchromatic cell commonly found during regeneration after tubular damage. In some affected cells the nuclei were pyknotic or were absent. The nuclei that remained were located near the lumen in the part of the cell where the cytoplasm was torn or frayed. In some cells fine periodic acid-Schiff- positive granules were found in the cytoplasm at the luminal border. The lumens of many distal convoluted tubules contained desquamated cells and granular, eosinophilic debris.
Most of the other segments of the renal tubule showed no significant pathologic alteration. Occasionally the cells of the ascending limb of Henle’s loop showed degenerative changes like those of the distal tubules. The cells of the proximal convoluted tubules appeared essentially normal, and some of them contained numerous periodic acid-Schiff-positive cytoplasmic granules (fig. 13). The lumens of a few of the proximal tubules contained cellular detritus. Most glomeruli appeared normal but, rarely, there was dilatation of capillaries and thickening of the base- ment membrane.
The average weight of the adrenal glands (weighed singly) of tumor- bearing rats was 6.3 mg., while that of the normal control rats was 24 mg. Both cortex and medulla of the adrenal glands of tumor-bearing rats
showed evidence of atrophy (figs. 10 and 11). In the zona fasciculata, where atrophy was greatest, the usual cordlike arrangement of cells was lost, and the sinusoids were dilated. The zona glomerulosa was somewhat thinner than normal, and its characteristic glomerular pattern was not evident. Frozen sections stained with Oil Red O revealed virtually no fat in the zona fasciculata and zona glomerulosa, and very little fat in the zona reticularis. Numerous large reticuloendothelial cells containing iron, a small amount of fat, and a few periodic acid-Schiff-positive granules were found in the zona reticularis and the inner part of the zona fascicu- lata. Few, or none, of these cells were seen in the adrenal glands of untreated rats.
The adrenal medulla of the tumor-bearing rats showed changes in the cells and varying degrees of congestion. About half the medullary cells appeared normal except for an abnormally large amount of pigment (fig. 11). Other cells were small and polygonal, and groups of them were closely fitted together. Usually the outlines of these abnormal cells were distinct, but occasionally a group of cells resembled a syncytial multi- nucleated cytoplasmic mass. The cytoplasm of the small abnormal cells was free of fat. It was colored purplish brown by the periodic acid-Schiff reaction and a diffuse brown in hematoxylin and eosin prepa- rations. The nuclei, which were colored an even deeper brown, were round or oval and had a thick nuclear membrane that enclosed both fine and coarse chromatin granules and usually 1 nucleolus.
The gonads and accessory sex organs of both male and female tumor- bearing rats consistently showed atrophy of epithelial cells and stroma (figs. 14 to 19). On the average, the testis and uterus weighed half as much, and the ovary, one third as much, as the corresponding organs of non-tumor-bearing rats. In no tumor-bearing rat did the paired seminal vesicles weigh more than 200 mg., while in control rats the average weight was 1275 mg.
Often the size of the testis decreased with the increasing age of the transplanted tumor. Arrest of spermatogensis at the spermatid stage was commonly observed, although histologic alterations varied from diminished spermatogenesis in the least affected tubules to degeneration of spermatocytes and of Sertoli cells in those most affected (figs. 14 to 16). The size and number of interstitial cells were usually decreased. The decreased size of the seminal vesicles was explained by the absence of the usual fluid within the lumen and by the shrinkage of the epithelial lining cells.
The uterine horns of tumor-bearing rats were pale, thin, elongated, and fibrous. The endometrium, endometrial stroma, and myometrium were all atrophied (figs. 18 and 19). The vaginal epithelium resembled that of a spayed rat, being even thinner and showing less secretory activity than that of a normal rat in mid-dioestrus (fig. 22). Many of the ovarian follicles were atretic. Corpora lutea were usually present and appeared normal or only slightly atrophic, but in a few rats inoculated with tumor 494 at 4 weeks of age, corpora lutea were not found. The amount of
stroma in the ovaries was diminished, but stromal fat content was about the same as in untreated controls. Nests of interstitial cells with a radiating spokelike arrangement of the nuclear chromatin were found in many ovaries (fig. 24). These were considered to be the type of “wheel cell” observed in the ovaries of hypophysectomized rats (4) and in the ovaries of old rats with disorders of the estrous cycle (5).
The pituitary glands of rats that had borne the transplanted tumor 494 for 3 months or less did not differ in gross appearance or in average weight from the pituitary glands of matched controls. Histologic sections stained by Mallory’s acid fuchsin-aniline blue method revealed numerous, large, vacuolated, partially degranulated basophils in the anterior lobe. These were considered to be castration cells. No change was noted in the acidophils or chromophobes.
In rats that had borne the transplanted tumor 494 for more than 3 months, the pituitary glands weighed on the average a little less than the glands of the matched controls. Blood spaces occupied more of the anterior lobe, and the cells were more compact and smaller than in the controls. The acidophils, in addition to being smaller and having less cytoplasm than the acidophils in the controls, usually had small hyper- chromatic nuclei; some nuclei showed pyknosis, karyolysis, and kary- orrhexis. Although no cell counts were made, study of serial sections gave the impression that the total number of basophils was decreased and that the vacuolated basophils were absent or could be found less frequently than in rats killed within the first 3 months after inoculation with the tumor.
Histologic sections of the mammary glands revealed normal develop- ment of ducts in the females and of acini in the males. In some of the tumor-bearing rats of both sexes the mammary-gland tissue resembled that occasionally seen in old animals in that the ducts and acini were shrunken and surrounded by thick, hyalinized, connective tissue (figs. 27 and 31). Compare with normal glands (figs. 26 and 30). Mammary ducts were more prominent than normal in several tumor-bearing males that had greatly atrophied testes, while acini were the predominant structure in female rats with the greatest ovarian atrophy, or in which corpora lutea were absent.
Atrophy of the thymus gland (fig. 25) and lymphoid follicles of the intestine was evident in all tumor-bearing rats. In some rats hema- topoiesis and pigment deposition in the splenic pulp was decreased, and the splenic follicles appeared to be smaller than normal; in others, no abnormalities of the spleen could be detected. Leukocyte counts ranged from 2,450 to 4,100 per mm.3 in tumor-bearing rats and from 7,350 to 8,300 per mm.3 in the controls. Erythrocyte counts ranged from 10,530,- 000 to 12,050,000 per mm.3 in tumor-bearing rats and from 7,170,000 to 8,200,000 per mm.3 in the controls. No comparison of eosinophil counts was made. The thyroid gland, salivary glands, pancreas, and other organs of rats bearing the transplanted tumor 494 showed no histopatho- logic alterations attributable to the experimental procedure.
Transplantation to Hypophysectomized Rats and Development of Tumor Line 494H
Growth for 1 generation in a hypophysectomized rat induced an un- expected transformation in tumor 494-a transformation manifested both in the histologic pattern and in the physiologic effect of the tumor. Altered tumor line 494H was derived as follows: A fourth-generation tumor of line 494 grown in a mature, intact, female rat was transplanted to 6 hypo- physectomized rats, in May, 1957. The transplanted tumor reached a diameter of 0.5 cm. in 1 male and in 1 female hypophysectomized rat after a latent period of 7 months; it did not grow at all in the other 4 rats. Intact rats inoculated at the same time developed tumors that measured 1.0 cm. in diameter within 2 months; these tumors in their subsequent growth and in their effect upon the hosts did not differ from other tumors of line 494.
Sections of the transplanted tumors of line 494H that grew in both the hypophysectomized rats contained 2 types of cell (fig. 7). One was the uniform polygonal cell that comprised the entire cell population of un- altered line 494. The other type of cell was much larger, with deeply colored, vacuolated cytoplasm, and irregularly shaped, hyperchromatic nuclei. Mitotic figures were exceedingly numerous. Aside from the alterations secondary to hypophysectomy, no other gross or histologic changes attributable to the effect of the transplanted tumor, 494H, were observed in the host rats. Serial sections of the sella turcica revealed a tiny fragment of anterior pituitary-gland tissue in the male rat, but none in the female rat.
Pieces of tumor 494H that had grown in the hypophysectomized female rat were transplanted subcutaneously into 2 intact male and 2 intact female rats 3 months of age. These 4 animals were not litter mates. In all of them the tumor grew readily and reached a diameter of 1.0 cm. within 2 months. Shortly thereafter, the host rats developed polyuria and poly- dipsia. They were killed from 5 to 6 months after inoculation. All 4 transplanted tumors were composed of small polygonal cells and large bizarre cells, and thus resembled the tumor grown in the hypophysec- tomized rat from which they had been transplanted (fig. 7). A similar combination of cells comprised metastatic tumor tissue in the lungs of all 4 rats and in the adrenal glands of the 2 females (fig. 9). The kidneys revealed the degenerative lesion of the distal convoluted tubules.
Other effects of tumor 494H were decidedly different from the effects of tumor 494 that had always been carried in rats with intact pituitary glands. The adrenal glands of the animals bearing tumor 494H were only moderately atrophied, and the zona glomerulosa was the only layer of the cortex markedly depleted of fat. Mammary glands of both male and female rats were filled with milky secretion (figs. 28 and 32). Atrophy of the testes was extreme, and spermatogenesis was arrested at the stage of the primary spermatocyte (fig. 17). Atretic follicles but no corpora lutea were present in the ovaries. The uterine horns were enlarged (fig.
20). The columnar endometrial cells were unusually tall, and their cytoplasm contained occasional droplets of mucus; the nuclei were basally situated. An extraordinary degree of mucification was present in the intermediate layer of the vaginal epithelium (fig. 23).
Transplantation to Newborn Rats and Development of Tumor Line 494NB
Altering the environment of tumor 494 by transplanting it to newborn rats induced a transformation in its histologic pattern similar to that observed when tumor 494 was grown in hypophysectomized rats. This altered line, designated 494NB, was derived from a fifth-generation tumor of line 494 grown in a mature, intact, male rat. Approximately 1 gm. of tumor tissue was homogenized with 10 cc. of physiologic saline, and 0.1 cc. of this brei was injected subcutaneously into 5 female and 3 male newborn littermate rats less than 24 hours old. Transplanted tumor 494NB required 11 weeks (compared with 8 weeks for tumor 494) to reach a diameter of 1.0 cm., but from then on tumor 494NB grew rapidly. Polyuria and polydipsia were first manifest when the rats were 3 months old. The 8 animals were killed from 4 to 412 months after inoculation.
The transplanted tumors of line 494NB were composed in part of the uniform type of polygonal cell characteristic of tumor 494, but they also contained the large, irregular, bizarre cells with hyperchromatic nuclei and many mitotic figures (fig. 8) as seen in line 494H (fig. 7). Metastatic tumor tissue containing both small polygonal cells and large bizarre cells was found in the liver and lungs of all 8 rats bearing tumor 494NB, in the adrenal glands of 2 males and 3 females, in the spleen of 1 female, and in the mediastinal lymph nodes of 1 male. The mammary glands (figs. 29 and 33), testes, ovary, uterus (fig. 21), and vagina (fig. 23) resembled, in gross appearance and in histology, the corresponding organs of rats bearing transplants of tumor 494H. Degenerative changes in the distal convoluted tubules of the kidney were of the same degree as previously described in rats bearing tumor lines 494 and 494H.
Discussion
Comparison of tumor 494 with other adrenal cortical tumors of the rat .- Primary tumor 494 resembled in some respects the tumors that Houssay and his associates induced in the adrenal glands of rats by castration (1,6). These authors classified the cells of their tumors as A, B, and C. Cells of Type A were small and similar to cells of the zona glomerulosa, and in the early stages of tumor growth they formed triangular sub- capsular nodules. Cells of Type B were polygonal and larger than Type A; the cytoplasm was finely granular and vacuolated, and the cells were arranged in nodules or cords reminiscent of the zona fasciculata. Cells of Type C were large, the nuclei were peripherally placed, and the cyto- plasm was extensively vacuolated and contained both fat and periodic acid-Schiff-positive droplets. The Type C cell of the tumor described by Houssay et al. resembled the cells of tumor #2 (figs. 3 and 4) found in the
adrenal gland of our original tumor-bearing rat. Cell types A and B were similar to some of the smaller, more deeply colored, cells of our tumor #1 (fig. 2). The adrenal cortical tumors of castrated rats described by Houssay and his coworkers were often associated with tumors of the pituitary gland and mammary gland, which has been our experience with the spontaneous tumors of Osborne-Mendel rats.
The transplanted tumor of line 494 is similar in histologic appearance to several other transplantable adrenal cortical tumors of the rat (2, 3, 7)-in its effect on the host it is similar in some respects and different in others. A tumor that originated in an Osborne-Mendel rat fed a low protein diet was successfully transplanted by Mulay and Eyestone (7) to untreated male and female rats. This tumor induced atrophy of the adrenal glands of the host, but no striking changes in any other endocrine gland and no measurements of the urine output of tumor-bearing rats were reported by these investigators. It is not known whether this tumor’s functional capacity was the same as that of tumor 494.
In its physiologic effects, including the induction of polyuria and polydipsia, tumor 494 resembles the adrenal cortical tumor of WR rats described by Cohen, Furth, and Buffett (2). These authors, however, did not mention a renal lesion as being associated with the tumor they studied. When Cohen et al. (8) incubated slices of rat adrenal tumor with and without ACTH, they were able to demonstrate that its secretion consisted mainly of corticosterone. No bioassay has been made on tumor 494, but is seems reasonable to postulate that it, too, may secrete adrenal corticoids.
Tumor 494 grows well in all untreated mature rats and in castrated rats of both sexes. In this respect it differs from the tumor described by Cohen and his associates, which grows better in males than in females and is at least partially dependent upon hormonal imbalance produced in the host by gonadectomy, adrenalectomy, or thyroidectomy. The growth pattern of tumor 494 also differs from that of an adrenal cortical tumor in A X C rats, described by Iglesias and Mardones (3, 9), which grows better in females than in males, requires no induced hormonal imbalance for its growth, and is partially inhibited by ACTH. ACTH neither retarded nor accelerated the growth of our tumor 494, and its functional, biologic, and histologic characteristics were not altered by administering ACTH to the hosts.
Association of polyuria with degenerative changes in the distal convoluted tubules .- Several investigators (10-12) after studying the physiologic effects of various adrenal steroid hormones on the kidney have postulated that these hormones damage the cells of the distal convoluted tubules, thus impeding the normal reabsorption of water and producing polyuria. However, none of these investigators described a histologic lesion of the kidney. In the present study we have demonstrated that the polyuria and polydipsia observed in rats bearing tumor 494 and its sub- lines are associated with a morphologic lesion of the distal convoluted tubules of the kidney. Slight degenerative changes in the cells of the
distal convoluted tubules were observed in a few rats killed before the onset of polyuria and polydipsia, and the most marked degeneration was found in rats that exhibited these symptoms in greatest severity for the longest period of time. It is therefore probable that secretions from the transplanted tumor induced the tubular degeneration. Whether, once initiated, the constant and increased urine flow may also have played a part in the degenerative renal process is not known. The total daily output of urine was not influenced by treatment of the rats with pitressin; hence, it seems probable that the polyuria had a renal origin rather than a hypothalamic or posterior pituitary origin.
Comparison of the renal lesion of rats bearing tumor 494 with the renal lesion of potassium deficiency .- The kidneys of rats bearing the trans- planted adrenal tumor resembled those of rats fed a potassium-deficient diet only in that both manifested increased weight, dilatation of the distal convoluted tubules, and vacuolization of the tubular cells. In other respects the two lesions differed. Neither the cellular regeneration nor the lipoidal droplets and calcareous material in the epithelial cells of the distal convoluted tubules, described by Follis, Orent-Keiles, and McCollum (13) as characteristic of potassium deficiency, were present in the kidneys of our tumor-bearing rats. The renal degeneration associated with tumor 494 was limited almost exclusively to the distal convoluted tubule, whereas the nephron in its entirety was involved in the lesion that Tauxe and his associates (14) induced in rats by a potassium- deficient diet. We did not observe the retrogressive alterations or the prolific regenerative hyperplasia of the cells of both collecting and proximal convoluted tubules described by Oliver and his coworkers (15) during the early stages of potassium depletion. Therefore, the morphology of the renal lesion in our tumor-bearing rats differs from that of rats with a potassium deficiency.
Comparison of tumor 494 with tumor 494H .- In all 7 generations of line 494, the subcutaneous transplants and the few pulmonary and hepatic metastases had the same histologic pattern of cords of uniform polygonal cells (fig. 6). The functional effects of the tumors upon the hosts were also similar. It seems probable that the tumors secreted an excess of certain hormones normally secreted by the cortex of the adrenal gland and that these tumor secretions, by suppressing the pituitary gland, indirectly induced atrophy of the adrenal glands, sex organs, and lymphoid tissue. Renal damage, polyuria, and polydipsia may have been caused by secretions from the transplanted tumor acting directly upon the kidney.
The tumors of subline 494H, developed by sojourn of tumor 494 for 1 generation in a hypophysectomized rat and subsequently returned to intact, mature, male and female rats, were all alike in biologic behavior, histologic pattern, and physiologic effect, but different in some respects from tumor 494. Tumor 494H consisted, in part, of cords of uniform polygonal cells and, in part, of cells that were bizarre in shape, variable in size, and usually contained 1 large hyperchromatic nucleus (fig. 7). A similar combination of cells was found in frequent, large, metastatic
tumors in the lungs and adrenal glands (fig. 9). The appearance of the mammary glands, pituitary glands, testes, ovaries, uteri, vaginas, and adrenal glands of intact, mature rats bearing tumor 494H was similar to that observed after the administration of estrogen and progesterone. Polyuria, polydipsia, and degenerative changes in the renal tubules indicate that the altered tumor 494H retained at least part of the func- tional activity of tumor 494.
Change in environment of the tumor cells during their growth is the most plausible explanation for the difference between tumor 494 and tumor 494H, since the same tumor of line 494 that was used for trans- plantation into the hypophysectomized rats was also transplanted into intact, mature rats. In these latter rats neither the tumor cells nor their effects upon the hosts were altered in any respect. Why transplantation in a hypophysectomized rat should bring about the alterations observed in tumor 494H, however, is not clear. The concomitant appearance in tumor 494H of a type of cell and of a functional effect not hitherto as- sociated with any tumor of line 494 makes it reasonable to assume that the new cell and the new functional activity are related.
The question then arises as to the source of the altered cells. One possibility is that they developed from the uniform polygonal cells and thus represent a change toward anaplasia, a common development in transplantable tumors. However, when the cells of a tumor become anaplastic, the tumor nearly always loses at least some of its functional capacity. We cannot recall any manifestation by tumor cells of a def- initely new functional activity coincident with morphologie anaplasia. Another possible explanation for the altered cells in tumor 494H is that they represent latent cells, which were transplanted from the original tumor and suppressed in the intact animal by some restraining influence of the normal pituitary gland. The absence of the pituitary gland may have allowed some of these tumors cells to grow without restraint and to manifest a hitherto unexpressed hormonal activity. Since the large, bizarre cells were observed for the first time in tumor 494H, they may well be the cells associated with the altered function.
Comparison of tumors 494, 494H, and 494NB .- The tumors of subline 494NB, which developed when the tumor of line 494 was transplanted into newborn rats, contained large, bizarre cells as well as the smaller polygonal cells characteristic of tumor 494. Tumor 494NB thus re- sembled tumor 494H in histologic pattern, since the altered, bizarre cells comprised approximately half the cell population of both tumors. These sublines were also similar in functional effects because both produced the same changes in the sex organs and mammary glands. Tumor 494H and 494NB were alike in that all tumors of both lines metastasized. The metastases occurred earlier and in more sites than in any rat bearing tumor 494. Tumor 494 metastasized infrequently to the liver and lungs of rats that had borne transplants for a long period. Polyuria, polydipsia, and renal tubular degeneration were observed in rats bearing tumor 494 and in rats bearing the sublines 494H and 494NB.
References
(1) HOUSSAY, B. A., HOUSSAY, A. B., CARDEZA, A. F., and PINTO, R. M .: Tumeurs surrénales oestrogéniques et tumeurs hypophysaires chez les animaux castrés. Schweiz. med. Wchnschr. 85: 291-296, 1955.
(2) COHEN, A. I., FURTH, J., and BUFFETT, R. F .: Histologic and physiologic char- acteristics of hormone-secreting transplantable adrenal tumors in mice and rats. Am. J. Path. 33: 631-651, 1957.
(3) IGLESIAS, R., and MARDONES, E .: Spontaneous and transplantable functional tumour of the adrenal cortex in the A X C rat. Brit. J. Cancer 12: 20-27, 1958.
(4) SELYE, H., COLLIP, J. B., and THOMSON, D. L .: On the effect of the anterior pituitary-like hormone on the ovary of the hypophysectomized rat. En- docrinology 17: 494-500, 1933.
(5) WOLFE, J. M., BURACK, E., and WRIGHT, A. W .: The estrous cycle and asso- ciated phenomena in a strain of rats characterized by a high incidence of mammary tumors together with observations on the effects of advancing age on these phenomena. Am. J. Cancer 38: 383-398, 1940.
(6) HOUSSAY, B. A., CARDEZA, A. F., FOGLIA, V. G., HOUSSAY, A. B., and PINTO, R. M .: Evolution anatomique et fonctionnelle des tumeurs surrénales chez les rats castrés. Compt. rend. Soc. biol. 148: 918-919, 1954.
(7) MULAY, A. S., and EYESTONE, W. H .: Transplantable adrenocortical adeno- carcinomas in Osborne-Mendel rats fed a carcinogenic diet. J. Nat. Cancer Inst. 16: 723-739, 1955.
(8) COHEN, A. I., BLOCH, E., and CELOZZI, E .: In vitro response of functional experimental adrenal tumors to corticotropin (ACTH). Proc. Soc. Exper. Biol. & Med. 95: 304-309, 1957.
(9) IGLESIAS, R., and MARDONES, E .: Influence of hormones on the growth of a transplantable suprarenal tumour. Brit. J. Cancer 12: 28-31, 1958.
(10) GAUNT, R .: The interrelationship between the adrenal cortex, posterior pituitary and anterior pituitary in water metabolism. In Ciba Foundation Colloquia on Endocrinology, vol. IV: Anterior Pituitary Secretion and Hormonal In- fluences in Water Metabolism (Wolstenholme, G.E.W., and Cameron, M.P., eds.). New York, Blakiston Co., 1952, pp. 455-462.
(11) PITTS, R. F .: Effects of adrenal cortical hormones on renal function. In Adrenal Cortex: Trans. of 3rd Conf., Josiah Macy, Jr. Foundation (Ralli, E. P., ed.). New York, 1952, pp. 11-53.
(12) WINTER, C. A .: Comparison of the effect of cortisone acetate and of desoxy- corticosterone acetate upon water balance. In Ciba Foundation Colloquia on Endocrinology, vol. IV: Anterior Pituitary Secretion and Hormonal In- fluences in Water Metabolism (Wolstenholme, G.E.W., and Cameron, M.P., eds.). New York, Blakiston Co., 1952, pp. 499-514.
(13) FOLLIS, R. H., Jr., ORENT-KEILES, E., and MCCOLLUM, E. V .: The production of cardiac and renal lesions in rats by diets extremely deficient in potassium. Am. J. Path. 18: 29-39, 1942.
(14) TAUXE, W. N., WAKIM, K. G., and BAGGENSTOSS, A. H .: The renal lesions in experimental deficiency of potassium. Am. J. Clin. Path. 28: 221-232, 1957.
(15) OLIVER, J., MACDOWELL, M., WELT, L. G., HOLLIDAY, M. A., HOLLANDER, W., WINTERS, R. W., WILLIAMS, T. F., and SEGAR, W. E .: The renal lesions of electrolyte imbalance. I. The structural alterations in potassium-depleted rats. J. Exper. Med. 106: 563-574, 1957.
PLATES
Downloaded from https://academic.oup.com/jnci/article-abstract/22/6/1119/933444 by Universite Paris Sud user on 15 May 2019
PLATE 101
FIGURE 1 .- Half of the right adrenal gland of original tumor-bearing rat. Line of incision dividing the adrenal gland into 2 parts at necropsy is at bottom of illustra- tion. Elsewhere the gland is surrounded by the capsule, beneath which lies a thin rim of compressed adrenocortical tissue. Arrow points to tumor designated #1 in the text; the compressed medulla is just to right. All the rest of the section con- tains tumor #2, which extends to the line of incision and was therefore included in the tissue used for transplantation. Note the dilated, thrombosed vessel along the line of incision. A portion of the kidney is at right margin of the illustration. Hematoxylin and eosin. X 22
FIGURE 2 .- Tumor #1 of the right adrenal gland of original tumor-bearing rat. Hematoxylin and eosin. × 360
2
1
2
PLATE 102
FIGURE 3 .- Tumor #2 of the right adrenal gland of original tumor-bearing rat. Hematoxylin and eosin. × 1,050
FIGURE 4 .- Tumor #2 of the right adrenal gland of original tumor-bearing rat: Cells show excessively vacuolated cytoplasm and peripherally located nuclei in contrast to the cells of the portion of the tumor illustrated in figure 3. Hema- toxylin and eosin. × 1,050
3
4
PLATE 103
FIGURE 5 .- Left adrenal gland of original tumor-bearing rat, showing pronounced atrophy of the cortex. Hematoxylin and eosin. X 50 ☒
FIGURE 6 .- Tumor 494. A fifth-generation transplant grown in the subcutaneous tissues of a 6-month-old intact female rat inoculated 412 months previously. Note uniform polygonal cells in cordlike arrangement and large, thin-walled vascular channels. Hematoxylin and eosin. X 210 ☒
5
6
PLATE 104
FIGURE 7 .- Tumor 494H. Subcutaneous transplant tumor from an 8-month-old intact female rat inoculated 5 months previously with tumor 494H that had been grown in a hypophysectomized female rat. Contrast histologic appearance with that of the corresponding generation of tumor 494 shown in figure 6. The large, bizarre cells that first appeared in tumor 494H can be correlated with a change in the functional effects of the transplanted tumor. Hematoxylin and eosin. X 210
FIGURE 8 .- Tumor 494NB. Subcutaneous transplant tumor from a 4-month-old female rat inoculated, when less than 24 hours old, with the fifth-generation trans- plant of tumor 494 grown in an intact mature male rat. The histologie pattern of line 494NB is like that of line 494H (fig. 7) and different from that of tumor 494 (fig. 6). Hematoxylin and eosin. X 210
7
8
PLATE 105
FIGURE 9 .- Metastasis of tumor 494H in the adrenal gland of an intact female rat, 8 months of age, inoculated 5 months previously with the tumor grown in the hypo- physectomized female rat. Large, bizarre cells are like those illustrated in figure 7. The metastatic tumor has distorted this portion of the adrenal gland so that the cortical atrophy cannot be appreciated. Hematoxylin and eosin. X 135
FIGURE 10 .- Adrenal gland of an untreated, non-tumor-bearing female rat 412 months of age. Hematoxylin and eosin. X 135
FIGURE 11 .- Adrenal gland of a 41/2-month-old female rat inoculated 3 months pre- viously with Adrenal cortical carcinoma 494. The cortex is atrophied, and the medulla contains small, deeply colored cells. Hematoxylin and eosin. X 135
9
10
11
PLATE 106
FIGURE 12 .- Kidney of a 712-month-old female rat inoculated 412 months previously with the fourth-generation transplant of tumor 494. The subcutaneous transplant tumor weighed 10.5 gm. Degenerative changes are present in the cells of the distal convoluted tubules. In the week before necropsy the rat’s average daily water intake was 57 cc. and its average daily urine excretion was 24 cc. Periodic acid- Schiff. × 260
FIGURE 13 .- Kidney of a 9-month-old male rat inoculated 512 months previously with the fourth-generation transplant of tumor 494, showing a higher-power view of the degenerated cells of the distal convoluted tubules. Many cells of the normal proximal convoluted tubules contain periodic acid-Schiff-positive granules. In the week before necropsy this rat had an average daily urine excretion of 60 cc., the largest daily output of any rat in our series. Periodic acid-Schiff. X 470
12
13
PLATE 107
FIGURE 14 .- Testis (weight 1332 mg.) from a normal untreated rat 8 months of age. Hematoxylin and eosin. × 165
FIGURE 15 .- Testis (weight 675 mg.), showing moderate atrophy, from a 9-month-old rat inoculated 512 months previously with tumor 494. Hematoxylin and eosin. × 135
FIGURE 16 .- Testis (weight 381 mg.), showing pronounced atrophy, from an 81/2- month-old rat inoculated 5% months previously with tumor 494. There is no evidence of spermatogenesis, and only Sertoli cells remain in the tubules. Hema- toxylin and eosin. × 135
FIGURE 17 .- Testis (weight 215 mg.), showing pronounced atrophy, from an 8-month- old rat inoculated 5 months previously with tumor 494H. There is interstitial edema and pronounced atrophy of tubular cells. Spermatogenesis is arrested at the stage of the primary spermatocyte. Contrast with lesion illustrated in figure 16. Hema- toxylin and cosin. × 135
14
15
16
17
PLATE 108
FIGURE 18 .- Uterus (weight 323 mg.) from a normal rat 6 months of age. Hema- toxylin and eosin. × 95
FIGURE 19 .- Atrophic uterus (weight 109 mg.) from a 4}2-month-old rat inoculated 3 months previously with a third-generation transplant of tumor 494. Hematoxylin and eosin. × 95 ☒
FIGURE 20 .- Enlarged uterus (weight 502 mg.) from a 9-month-old rat inoculated 6 months previously with tumor 494H. Compared with the normal uterus (fig. 18) this uterus shows hypertrophy and hyperplasia of endometrium, endometrial stroma, and myometrium. Hematoxylin and eosin. × 95 ☒
FIGURE 21 .- Enlarged uterus (not weighed) from a rat bearing tumor 494NB. The tall hyperplastic epithelial cells are like those illustrated in figure 20. Hematoxylin and eosin. × 200
18
19
20
8
21
PLATE 109
FIGURE 22 .- Vagina from a 416-month-old rat inoculated 3 months previously with tumor 494, showing atrophy of the mucosa and fibrosis of the wall. Hematoxylin and eosin. × 170
FIGURE 23 .- Vagina from a 4-month-old rat bearing tumor 494NB. The epithelium is thickened and shows mucification. The underlying muscular wall is not included in this illustration because it is separated from the subepithelial connective tissue by a zone of edema shown at bottom of illustration. Hematoxylin and eosin. X 170
FIGURE 24 .- Wheel cells in the ovary of a 412-month-old rat inoculated 3 months previously with the sixth-generation transplant of tumor 494. Periodic acid-Schiff. × 1,050
FIGURE 25 .- Atrophic thymus gland of an 8-month-old female rat inoculated 412 months previously with the fifth-generation transplant of tumor 494. Hematoxylin and eosin. × 125
22
23
24
25
PLATE 110
FIGURE 26 .- Mammary gland of a normal female rat 8 months of age. Hematoxylin and eosin. × 125
FIGURE 27 .- Mammary gland of an 8-month-old female rat inoculated 412 months previously with tumor 494. There is an excessive amount of hyalinized stroma around the ducts. Hematoxylin and eosin. X 125
FIGURE 28 .- Mammary gland of a 9-month-old female rat inoculated 6 months previously with tumor 494H. The ducts and acini are hyperplastic and contain a milklike secretion. Hematoxylin and eosin. × 125
FIGURE 29 .- Mammary gland of a 4-month-old female rat bearing tumor 494NB. The ducts and acini are hyperplastic and contain a milklike secretion. Hema- toxylin and eosin. × 125
26
27
28
29
PLATE 111
FIGURE 30 .- Mammary gland of a normal male rat 6 months of age. Hematoxylin and eosin. × 125
FIGURE 31 .- Mammary gland of a 6-month-old male rat inoculated 3% months previously with tumor 494. There is an excessive amount of hyalinized stroma around the acini. Hematoxylin and eosin. X 125
FIGURE 32 .- Mammary gland of an 8-month-old male rat inoculated 5 months pre- viously with tumor 494H that had been grown in a female hypophysectomized rat. The ducts and acini are hyperplastic and contain a milklike secretion. Hematoxylin and eosin. × 125
FIGURE 33 .- Mammary gland of a 412-month-old male rat bearing tumor 494NB. The ducts and acini are hyperplastic and contain a milklike secretion. Hematoxylin and eosin. × 125
30
31
32
33