Absence of Uniform Progressive Growth of Long-Term Transplants of Rat Liver Neoplastic Nodules 1,2,3

Transcription

1 Absence of Uniform Progressive Growth of Long-Term Transplants of Rat Liver Neoplastic Nodules 1,2,3 Takaaki Ohmori, 4,5 Kenshi Watanabe, 4,6 and Gary M. Williams 4,7, e, 9 ABSTRACT-Fragments of normal rat liver and neoplastic nodules induced by feeding 0.02% N-2-f1uorenylacetamide for 20 weeks were transplanted into the inguinal mammary fat pads of syngeneic rats. All animals were Charles River inbred male albino rats of the CD F strain derived from the F344 strain. Normal liver transplants and neoplastic nodule transplants were detected histologically in the transplant site up to 24 and 66 weeks of survival, respectively. Of 30 nodule transplants followed up to 121 weeks, only 1 neoplastic nodule developed progressive neoplastic growth. The tumor that arose in the transplant site from this nodule was a well-differentiated adenocarcinoma with mucin production. Thus rat liver neoplastic nodules lack the high frequency of progressive growth previously demonstrated for carcinomas.-jnci 65: , The nodular lesions that arise in the livers of rats exposed to chemical carcinogens have been considered by some investigators to be hyperplastic (1-5) and by others to be neoplastic (6-8). In a study specifically designed to examine this question (8), we reported that these nodules persisted and grew progressively following cessation of carcinogen exposure. Therefore, we have interpreted these lesions as being neoplastic (8, 9), which is in accordance with the recommendations of several recent working groups (10, 11). One of the means for establishment of the neoplastic nature of cellular lesions is the demonstration of the ability to form tumors upon transplantation. Earlier studies on the transplantation of rat liver nodules revealed no tumor development (12, 13). These findings were confirmed when fragments of liver, liver hyperplastic nodules, and hepatocellular carcinomas were transplanted into the inguinal fat pad in which longterm survival of transplanted fragments was obtained (14). Subsequently, we demonstrated that transplanted cells from nodules retained their phenotypic abnormalities but did not grow progressively during 8 months of observation (15). Although transplanted carcinomas formed tumors in the mammary fat pad within 16 weeks (14, 15), 8 months may not have been a sufficient time to allow for development of tumors from liver nodules, inasmuch as well-differentiated mouse liver tumors require up to 545 days for growth in this site (16). Therefore, we transplanted nodules into rats and observed these rats for their lifetimes. Here we show that rat liver neoplastic nodules have little tumor-forming capacity when transplanted into the mammary fat pads of intact syngeneic male rats maintained 121 weeks. MATERIALS AND METHODS The animals were 6- to 8-week-old Charles River inbred male albino rats of the CD F strain derived from the F344 strain (Charles River Breeding Laboratories, North Wilmington, Mass.). The basal diet was Purina Laboratory Chow (Ralston Purina Co., St. Louis, Mo.). To induce liver nodules, we fed 20 donor rats 0.02% 2-FAA in the diet for 20 weeks. The 90 recipient rats were divided into 2 groups. The transplantation procedure was the same as that in previous studies (14, 15). In the first group, 30 rats were given transplants in the inguinal mammary fat pad of one small fragment (=1 mm") of normal liver from 10 healthy donor rats. Three recipients were killed every 4 weeks from 8 to 28 weeks after transplantation. In the second group, 60 rats were given transplants of fragments of macroscopic liver lesions from the 20 donor rats fed 2-FAA. Microscopic examination of tissue from the lesions used for transplantation confirmed 41 neoplastic nodules according to the criteria of the Institute for Laboratory Animal Resources (11). Eleven recipients from group 2 were killed at intervals of 8, 16, 52, 56, and 66 weeks after transplantation. The remaining recipients of both groups were kept for their lifetimes or until a tumor appeared at the transplant site. All recipients were examined at the transplant sites by palpation monthly at the time of ABBREVIATIONS USED, 2-FAA =N -2-f1uorenylacetamide; PAS =periodic acid-schiff. 1 Received January 9, 1980; accepted March 17, Supported by Public Health Service grants CAI2376 and CAI7613 from the National Cancer Institute. 3 Animals were maintained under guidelines set forth by the Naylor Dana Institute. 4 Division of Experimental Pathology and Toxicology, Naylor Dana Institute for Disease Prevention, American Health Foundation, I Dana Rd., Valhalla, N.Y S Visiting Scientist from the Department of Pathology, Nara Medical University, 840 Shijo-cho, Kashihara City, Nara 634, Japan. 6 Present address: The Department of Pathology, Kagoshirna University, School of Medicine, Kagoshima 890, Japan. 7 Department of Pathology, New York Medical College, Valhalla, N.Y Address reprint requests to Dr. Williams at the Division of Experimental Pathology and Toxicology, Naylor Dana Institute for Disease Prevention. 9 We thank Ms. Ellen Sestrom of the Research Animal Facility, Naylor Dana Institute for Disease Prevention, and Ms. Carol Meyer of the Division of Experimental Pathology and Toxicology, Naylor Dana Institute for Disease Prevention, for their excellent technical assistance. We also thank Ms. Bette Meyer for preparing this manuscript. 485 JNCI, VOL. 65, NO.2, AUGUST 1980

2 486 Ohmori, Watanabe, and Williams weighing. At necropsy if a transplant was not identified by gross observation, the entire inguinal fat pad at the transplant site was examined microscopically by subserial sections. To evaluate the stability of resistance to iron accumulation of neoplastic nodules (17, 18) after transplantation, as was done in (15), we iron loaded all recipients by feeding them 0.8% 8-hydroxyquinoline and 2.19% ferrous gluconate (17) in the diet from 1 week to 30 weeks in the normal liver transplant group and for 13 weeks in the nodule transplant group. The histologically identified transplants were examined histochemically by the Prussian blue reaction for iron or PAS staining for glycogen. RESULTS TABLE Of 18 normal liver transplants, 8 were histologically detected as clusters or nests of cells up to 24 weeks after transplantation. The iron-loading regimen produced iron deposition in macrophages in the mammary fat pad, and surviving liver cells contained different degrees of cytoplasmic iron granules (fig. 1) as well as PAS-positive granules (table 1). No mitotic figures were found. Of the 60 carcinogen-induced lesions used for transplantation, 41 were histologically confirmed to be neoplastic nodules by examination of the remaining tissue. Of 11 transplants examined at 8, 16, 52, 56, and 66 weeks after transplantation, only 2 were histologically identified, 1 at week 16 and the other at week 66 after transplantation. Compared with the original tissue used for transplantation, these 2 surviving nodule transplants showed the persistence of phenotypic characteristics and also some degree of increased abnormality. Mitotic figures were seen in both transplants. The transplanted nodule examined at 16 weeks showed some acinar arrangements of slightly basophilic cells (fig. 2). The other transplanted nodule found at 66 weeks represented the longest documented period of survival at the transplant site. It also showed increased morphologic abnormalities. The nodule was encapsulated by a rim of connective tissue (fig. 3). The cells within the nodule were arranged in acinar structures composed of highly basophilic cells that occupied more than two-thirds of the nodule and compressed slightly the other less atypical cells. The original nodule used for this transplant contained an area of atypical basophilic cells (fig. 4). Thirty transplanted neoplastic nodules were followed for more than 66 weeks, but only 1 developed a tumor at approximately 71 weeks after transplantation. When the rat bearing this tumor was killed at 75 weeks after receiving the transplant, the tumor was found to be 6.5X4.5X6.0 cm in size and showed central necrosis and metastasis to neighboring lymph nodes. It was a welldifferentiated adenocarcinoma consisting of tubular and/or transitional to trabecular patterns with mucin production and a variable amount of fibrous stroma (fig. 5). The residual portion of the original lesion used for transplantation had a focus of cells arranged in an acinar pattern (fig. 6). All remaining rats in both the normal liver and the neoplastic nodule transplant groups were observed for their lifetimes and killed when moribund. The last animal was killed at 121 weeks after receiving the transplant. All tumors that developed at or around the transplant sites were examined histologically. In a normal liver recipient, 1 mammary adenocarcinoma and 1 fibroadenoma of the mammary gland were found. No other hepatocellular tumors were observed. I.-Behavior and phenotypic characteristics of transplants of normal liver and neoplastic nodulesa Transplants Period of No. of No. of transplants No. of Iron Ini- No. of transplants transplan- transplants Status Mitoses followed tumors examined accumu- Donor tial tation, wk identified lation No. at interim Normal liver Nest of cells Cluster of cells Cluster of cells Cluster of cells Cluster of cells NA NA NA NA NA NA NA Neoplastic nodule NA NA NA Nests b of cells + -/ NA NA NA NA NA NA Nodule" of cells d 1 1 Adenocarcinoma" +++ a -=Negative; +=found in a rare cell; ++=strongly positive; NA=not applicable; -/+=weakly positive; +++=found frequently. b Glandular structures present. C Tubular structures present. d Original transplant was a neoplastic nodule with atypical features. e Well-differentiated adenocarcinoma with tubular, cystic, and papillary patterns. JNCI, VOL. 65. NO.2, AUGUST 1980

3 Transplanted Liver Neoplastic Nodules 487 DISCUSSION In the present study we demonstrated that transplants of carcinogen-induced liver nodules persisted for long periods in the transplant site and retained their phenotypic abnormalities but did not often give rise to progressive neoplastic growth. Of the 41 nodules transplanted, II were examined at intervals during the first 66 weeks of the experiment. Of these, 2 were identified histologically and both displayed the morphologic abnormalities of nodule cells that we have described in a previous study (15). In addition, the first nodule transplant examined during the period of iron administration remained resistant to iron accumulation, characteristic of neoplastic nodules in situ (17, 18) and as previously described for transplants (15). Thus 30 nodule transplants were maintained for weeks and, among these, only I transplant gave rise to a hepatocellular adenocarcinoma in the transplant site. The results of this experiment can be interpreted in several ways. First, neoplastic nodules, as shown in previous studies (14, 15), do not give rise to rapid tumor development in high incidence as do hepatocellular carcinomas. Thus neoplastic nodules are clearly a biologically different lesion, as we have previously suggested (9, 15). Another issue addressed by this experiment is whether these nodules are neoplastic. Our results provide only indirect evidence in support of this interpretation. The nodule transplants that were recovered and examined histologically had retained their phenotypic abnormalities and had even greater abnormalities not seen in transplants of the normal liver. Thus as we reported previously (8, 15), the phenotypic abnormalities in nodules appear to be highly stable. This observation is consistent with nodules' being a permanently altered neoplastic population. However, with the exception of I transplant, the nodules did not show progressive growth in the transplant site. This negative finding must be interpreted cautiously. In a similar study of the transplantability of mouse liver nodules (16), we observed progressive growth of both type A and type B nodules, the neoplastic nature of which had been disputed. Thus this transplantation system can be considered to be reasonably supportive for the growth of neoplastic lesions. Therefore, the failure of rat liver nodules to display a uniform progressive growth suggests that these nodules may depend on some factors in the liver that are not available or operative in the transplant site. The final point of interest in this study is the development after 71 weeks of transplantation of I adenocarcinoma from a transplanted nodule. The first consideration in the interpretation of this finding is whether the original lesion used for transplantation was already a carcinoma. The residual tissue at the periphery of the primary lesion did not display the morphologic aspects of a carcinoma, although it did have some minor atypical features. Regardless, transplantation of carcinomas in this system has generally resulted in tumor growth in the transplant site within much shorter periods (14, 15). Therefore, we believe, as the histology shows, that the primary lesion in this instance was not a carcinoma. Thus the development of a tumor may be due to a very minor component of neoplastic cells in the transplant or may reflect the evolution of a carcinoma from a transplanted neoplastic nodule. The association of carcinoma with nodules in vivo has been illustrated (2, 5, 6, 19). However, several lines of evidence indicate that transition of nodules to carcinoma under most circumstances must occur only at low frequency (9). The present finding, if indicative of transition at all, conforms with this concept. The tumor that developed in the transplant site had the histologic characteristics usually considered to suggest origin from bile ducts. The transplanted fragment may have contained a minor cell type not apparent in the residual tissue available for histologic study. Nevertheless, we (15) previously found that such tumors can arise from transplanted hepatocellular adenocarcinomas or cultured liver epithelial cells with hepatocellular properties. Therefore, we favor the possibility that this tumor arose from neoplastic hepatocytes that, as discussed, may have been present in the transplant fragment or may have evolved from nodule cells. This experiment provides evidence that liver nodules have the phenotypic stability of neoplasms but lack a major capability for independent progressive growth in a transplant site. Furthermore, if nodules give rise to carcinomas, they do so only in low frequency. REFERENCES (l) FIRM INGER HI. Histopathology of carcinogenesis and tumors of liver in rats. J Natl Cancer Inst 1955;15: (2) POPPER H, STERNBERG SS, OSER BL, OSER M. The carcinogenic effect of aramite in rats: A study of hepatic nodules. Cancer 1960;13: (3) REUBER MD. Development of preneoplastic and neoplastic lesions of the liver in male rats given percent N-2-fluorenyldiacetamide. J Nat! Cancer Inst 1965;34: (4) NEWBERNE PM, WOGAN GN. Sequential morphologic changes in aflatoxin B 1 carcinogenesis in the rat. Cancer Res 1968;28: (5) FARBER E. Hyperplastic liver nodules. Methods Cancer Res 1973; 7: (6) SASAKI T, YOSHIDA T. Experimentelle erzeugung des Lebercarcinoms durch Fiitterung met o-amidoazotoluol. Virchows Arch [pathol Anat] 1935;295: (7) BANNASCH P. Cytology and cytogenesis of neoplastic (hyperplastic) hepatic nodules. Cancer Res 1976;36: (8) HIROTA N, WILLIAMS GM. Persistence and growth of rat liver neoplastic nodules following cessation of carcinogen exposure. JNCI 1979;63: (9) WILLIAMS GM. The pathogenesis of rat liver cancer caused by chemical carcinogens. Biochim Biophys Acta 1980;605: (10) SQUIRE RA, LEVITT MH. Report of a workshop on classification of specific hepatocellular lesions in rats. Cancer Res 1975;35: (11) Institute of Laboratory Animal Resources, National Research Council, National Academy of Sciences. Histologic typing of liver tumors of the rat. JNCI 1980;64: (12) REUBER MD, FIRMINGER HI. Morphologic and biologic correla- JNCI. VOL. 65, NO.2, AUGUST 1980

4 488 Ohmori, Watanabe, and Williams tion of lesions obtained in hepatic carcinogenesis in A X C rats given percent N-2-fluorenyldiacetamide. J Natl Cancer Inst 1963;31: (13) REUBER MD, ODASHIMA S. Further studies on the transplantation of lesions obtained in hepatic carcinogenesis in rats given 2-(diacetamidofluoren). Gan 1967;58: (14) WILLIAMS GM, KLAIBER M, FARBER E. Differences in growth of transplants of liver, liver hyperplastic nodules and hepatocellular carcinomas in the mammary fat pad. Am J Pathol 1977;89: (15) WILLIAMS GM, OHMORI T, WATANABE K. Persistence and phenotypic stability of transplanted rat liver neoplastic nodules. Am J Pathol 1980;99:1-12. (16) WILLIAMS GM, HIROTA N, RICE JM. The resistance of spontaneous mouse hepatocellular neoplasms to iron accumulation during rapid iron loading by parenteral administration and their transplantability. Am J Pathol 1979;94: (17) WILLIAMS GM, YAMAMOTO RS. Absence of stainable iron from preneoplastic and neoplastic lesions in rat liver with 8-hydroxyquinoline-induced siderosis. J Natl Cancer Inst 1972;49: (18) WILLIAMS GM. Functional markers and growth behavior of preneoplastic hepatocytes. Cancer Res 1976;36: (19) WILLIAMS GM, WATANABE K. Quantitative kinetics of development of N-2-fluorenylacetamide-induced, altered (hyperplastic) hepatocellular foci resistant to iron accumulation and of their reversion or persistence following removal of carcinogen. JNCI 1978;61: FIGURE I.-Normal liver transplant at 16 wk after transplantation. A cluster of hepatocytes situated in dense connective tissue contains fine PAS-positive material and also some brown ferritin granules (arrow). PAS reaction and hematoxylin. Bar = 50!Lm. X 240 FIGURE 2.-Neoplastic nodule transplant at 16 wk after transplantation. Three different-sized clusters of nodule cells display persistence of the characteristic morphology of neoplastic nodule cells and resistance to iron accumulation. Acinar arrangements of slightly basophilic cells can be seen (arrows). Prussian blue and hematoxylin. Bar=50!Lm. X 240 FIGURE 3.-Neoplastic nodule transplant tissue at 66 wk after transplantation. Highly basophilic neoplastic cells occupy the greater part of the cell mass and form acinar structures (arrows) that slightly compress adjacent cells. Hematoxylin and eosin. Bar =100!Lm. X 120 FIGURE 4.-Residual liver tissue surrounding the tissue used for the transplant shown in fig. 3. Cells surrounding the vacant space (lower two-thirds) include a region of basophilic cells. 'Hematoxylin and eosin. Bar = 100!Lm. X 120 JNCI, VOL. 65, NO.2, AUGUST 1980

5 Transplanted Liver Neoplastic Nodules 489 JNCI, VOL. 65, NO.2, AUG UST 1980

6 490 Ohmori, Watanabe, and Williams FIGURE 5.-Adenocarcinoma that appeared 71 wk after transplantation of a neoplastic nodule. Tumor cells form cystic, tubular, and papillary structures within a fibrous stroma. Occasional transition to a trabecular pattern is apparent. Hematoxylin and eosin. Bar= 100 }.'m. X 120 FIGURE 6.-Residual liver tissue surrounding the tissue used for the transplant shown in fig. 5. Upper left part delineated by a fibrous band is the remaining portion of the neoplastic nodule used lor transplantation. Acinar arrangement of cells is present (arrow). Hematoxylin and eosin. Bar= 100 }.'m. X 120 JNCI, VOL. 65, NO.2, AUGUST 1980