- Email: [email protected]
Simultaneous demonstration of lens regeneration from dorsal iris and tumour production from ventral iris in the same newt eye after carcinogen administration
Differentiation (1997) 61:285–292
© Springer-Verlag 1997
O R I G I NA L A RT I C L E
&roles:Mitsumasa Okamoto
Simultaneous demonstration of lens regeneration from dorsal iris and tumour production from ventral iris in the same newt eye after carcinogen administration &misc:Accepted in revised form: 17 February 1997
&p.1:Abstract It is well known that urodeles have the most powerful regenerative capacities among vertebrates, but there is little realisation that they are extremely resistant to spontaneous or chemically induced tumours. Regeneration and carcinogenesis have been considered to be two sides of the same mechanism. Since antagonism between regeneration and carcinogenesis was expected in previous studies, the present study was intended to clarify this relationship in greater detail by changing the amounts of carcinogen stepwise. When 1 µl nickel subsulfide solution was administered in various amounts (1 µg/µl ~40 µg/µl) into lentectomized newt eyes, the delay of initiation in lens regeneration for 6 months and an increased inhibition rate of lens regeneration at 6 months were observed in proportion to the increase in carcinogen dosage. The tumour production rate increased in accordance with the increase in the amounts of carcinogen. The conspicuous result obtained in the present study was that lens regeneration from dorsal iris and tumour induction from ventral iris occurred simultaneously in the same eye after administration of moderate amounts (10 µg/µl) of carcinogen. These data clearly indicated that the regenerating dorsal iris is persistently resistant to carcinogen, whereas the ventral iris, which cannot regenerate lens, is susceptible to tumour induction. This strongly suggests that the lens regeneration system in the newt has special advantages for research on the relationship between regeneration and carcinogenesis.&bdy:
Introduction Amphibians, especially urodeles, have remarkable regenerative capacities. The newts can regenerate limbs, tail and even neural tissue throughout life. In addition, special mention should be made of the remarkable resistance of the newt to most carcinogens. For example, no M. Okamoto Division of Biological Science, Graduate School of Science, Nagoya University, Chikusa-ku, Nagoya 464-01, Japan&/fn-block:
tumour production was found when carcinogens were given to the newt at a concentration about 100 times higher than that required to induce 100% tumour formation in rats [2, 7]. Embryologists have long been fascinated with the relationship between regeneration and carcinogenesis. According to Waddington [20] and Needham [8], the fundamental fact about cancerous tissue is that it has escaped from the normal growth-controlling agent of the body, called the “individuation field”. The main characteristic of this field is that all tissue lying within it tends to be built up into a complete embryo, and in any one part of the field all tissue tends to be organized into an organ corresponding to that part. Usually, this field has gradually become weak or has vanished during development. However, the field has been considered to be persistent and potent until adulthood in animals that are capable of regeneration. One of the key ways to verify these ideas is to observe whether a tumour can be induced in regeneration-competent tissues by applying chemical carcinogens. Several investigations have thus been made with regard to limb and tail regeneration in amphibia. In 1983, the effects of chemical carcinogens on regenerating and non-regenerating tissues in amphibia were reviewed by Tsonis [18]. Non-regenerating tissue in his report meant that the regenerating process is not induced in the regenerationcompetent limbs or tails. According to Tsonis, there is a considerable degree of controversy in the literature concerning the effects of carcinogens on regenerating and non-regenerating limbs. But, he suggested that in both instances regenerative capacity may be a deterrent to neoplasia in limb tissues. There is a genuine concern that such a problem cannot be solved by attempts using limb or tail regeneration systems, because neighboring tissue such as the trunk is difficult to use as a strict control for the regeneration of competent tissue. On the other hand, the lens regeneration system in the newt may provide a valuable tool for research on such a problem, because regenerating dorsal and non-regenerative ventral iris are located opposite
286
each other on either side of the pupil, and they belong to the same tissue that is defined as an “iris” in anatomical criteria. Therefore, the dorsal and ventral iris can be compared to each other under the same conditions in relation to a carcinogen placed in the pupil. There are few reports on the effects of carcinogens on lens regeneration in the newt except for those of Stone and Vultee [15] and Eguchi and Watanabe [5]. Stone and Vultee were able to prevent lens regeneration for 4–5 months by placing a crystal of a carcinogen in the pupil after lens extirpation. However, there was no description of tumour formation. Eguchi and Watanabe observed abnormal lens regeneration with supernumerary lenses from non-regenerative ventral and lateral irises, but found no tumour production. In previous studies [9, 10], the author examined the effects of a high dosage of a potent carcinogen, nickel subsulfide (αNi3S2), on lens regeneration and tumour production in the adult newt eye. The results demonstrated that lens regeneration was inhibited and ocular tumours were induced at a high incidence at 9 months. In the previous report, a population consisting of disarranged and aberrantly proliferating pigmented cells, which occupied the entire globe of the eye at 9 months, was identified as melanoma-like malignant tumours by light and electron microscopic observations. They were poorly cohesive cells with irregular or small round nuclei, prominent nucleoli and abundant cytoplasm with numerous pigment granules. They showed characteristics similar to the human epithelioid melanoma-like cell. Although the tumour cells invaded surrounding tissues, no metastasis to organs outside the eye was observed. In electron microscopic observations, crystalline αNi3S2 was occasionally found in some melanoma-like cells in samples fixed at 11 months. Acid-fast bacteria, the presence of which is reported in infectious granuloma cells, were not found in the cells. This was the first report on the induction of ocular melanoma in the amphibia class. The tumour was assumed to originate from the iris, because the aberrantly proliferating cell population was found in the iris in treated eyes at 3 months. Although the demonstration of antagonism between regeneration and carcinogenesis was expected in the previous studies, it was unclear whether the induced tumour is truly derived from regenerative dorsal iris, because the newt iris is composed of both regenerative dorsal and non-regenerative ventral iris. Moreover, since the dosage of carcinogen in the previous study was excessive and tumour cells occupied the entire globe of the eyeballs, it was not possible to identify the original point of an induced tumour. In the present study, to clarify the origin of an induced tumour and the relationship between regeneration and carcinogenesis, various amounts of αNi3S2 and amorphous nickel monosulfide (am-NiS) were injected into lentectomized newt eyes, and their effects on the incidence and origin of tumour production and lens regeneration were examined in detail. Nickel compounds are known to have a carcinogenic property. The carcinogenicity of αNi3S2 in man and ex-
perimental animals was reviewed by Sunderman Jr. [16], who mentioned that αNi3S2 is the most carcinogenic among numerous nickel compounds. The administration of this drug to animals results in malignant tumours at the sites of deposition in almost all cases [17]. It was reported that αNi3S2 and am-NiS possess extremely different carcinogenicities, despite their similar physical and chemical properties [17]. According to the study, the incidence of sarcoma in rats receiving a single injection of αNi3S2 was almost 100%, compared to none from am-NiS. The present results demonstrated inhibition of lens regeneration from high-dose administration of αNi3S2, delayed regeneration initiation from a medium dose, and normal regeneration from a low dose. In addition, the following were found in the same eye: dorsal iris regenerated a lens without tumour production, whereas ventral iris produced a tumour when moderate amounts of carcinogen were injected into the pupillary region. These data also implied that regenerative tissue, that is, the dorsal iris, is persistently resistant to carcinogens, while the ventral iris, considered non-regenerative tissue, is susceptible to tumour production.
Methods Animals Adult Japanese newts, Cynops pyrrhogaster, were used in the present experiment. The animals were purchased from Hamamatsu Seibutsu Kyouzai (Shizuoka, Japan), kept in the laboratory and fed with pork meal chips. Test compounds Crystalline nickel subsulfide (αNi3S2 in crystalline low temperature form) and information related to it were kindly provided by INCO Ltd., Toronto, Ontario, Canada. Amorphous nickel monosulfide (am-NiS hereinafter) should be used as soon as possible after formation, since it tends to oxidize easily to sulfate when stored. Thus, we made it ourselves according to a memo by D.F. Colton, kindly provided by J.S. Warner (INCO Ltd.). Preparation of am-NiS was as follows: 0.3 l 200 g/l Ni2+ solution was placed in a 1.5-l beaker and stirred at a rate just below the point of vortexing; 36.4 g/l S2− solution was then added from a fine-tipped burette at 10 ml/min until 0.45 l had been added (the amount of S 2− required to precipitate only 50% of the Ni 2+ in solution as NiS). The vortex rate was increased gradually as the volume of the slurry increased, but always kept just below the point of vortexing. After addition of the S2− solution (0.45 l in 45 min), the slurry was digested for 10 min. The final slurry was filtered on a 15-cm buchner funnel using no. 41 filter paper, and washed with 4×200 ml distilled water. The filter cake was repulped in 1 l distilled water, slurried for 10 min, and filtered (as above) and washed with 4×200 ml distilled water. A quantitative nickel mesurement was made in a representative dried portion from the wet filter cake according to the dimethylglyoxime method in the textbook of inorganic chemistry. Standard nickel solution (1.00 mg Ni/ml) was purchased from Nakarai Chemicals (Kyoto, Japan). The amorphous nature of the am-NiS made in the present experiment was checked by X-ray diffraction pattern, which was kindly measured by Dr. Hitoshi Oosato, Nagoya Industrial University. NaCl and NiCl 2 were purchased from Nakarai Chemicals. The am-NiS content (1, 10, and 40 µg/µl described in the present experiment means the same Ni
287 content as 1, 10, and 40 µg/µl αNi3S2, respectively. Since the solubility of NiCl2 is up to 105 times that of am-NiS and 106 times that of αNi3S2, respectively, and in addition, the localized concentrations of administered am-NiS and αNi3S2 seemed to be very high, we used 0.1 µg/µl as NiCl2 contents. The NaCl content was decided by the same principle as for NiCl2.
and embedded in Paraplast x-tra (Sigma Chemical Company, St. Louis, USA). Serially cut 10-µm-thick sections were stained with haematoxylin and eosin by the routine method.
Lentectomy and intraocular injection of carcinogen and non-carcinogenic compounds
Effects of αNi3S2 on lens regeneration
The method used was basically the same as in the previous studies [9, 10]. Animals were anesthetized with 0.1%–0.2% MS222 and transferred to an operation dish with Hanks’ balanced salt solution modified to 80% ionic strength for the newt. Both lenses were removed with a pair of forceps through a slit in the cornea. Injection into the eye chamber was performed with a glass micropipette, in this case a micro-type Labopettor (Hirschman Laborgerate, Germany). It was superior to the hand-made glass micropipette used in the previous studies, because it guaranteed a precise injection volume and smooth injection by piston movement. The suspension of αNi3S2 and am-NiS in sterile 0.6% NaCl solution was continuously agitated with a magnetic stirring apparatus to ensure that constant amounts of chemicals were aspirated into the micropipette set up to a volume of 1 µl.
Histological examinations Experimental animals were decapitated at 1, 3, and 6 months after injection. The isolated heads were fixed with Bouin’s fluid, decalcified in solution A (Wako Pure Chemical Industries Ltd, Osaka, Japan) by the Plank Rychlo method overnight, then dehydrated
Fig. 1a–c The reaction of dorsal iris to the carcinogen nickel subsulfide (αNi3S2, 10 µg/µl) administered into lentectomized newt eyes. a Disorganized cell mass found in dorsal iris at 1 month. b Mitotic figure (arrowhead) observed at the region shown in a. c Normally appearing dorsal iris with regenerating lens at 3 months. Bar a and c 300 µm, b 30 µm&ig.c:/f
Results
Lens regeneration from dorsal iris was examined at 1, 3 and 6 months after lentectomy following administration of αNi3S2 at various concentrations (Table 1). In all cases of injection of 1 µg/µl, which is the minimum concentration of αNi3S2 selected in the present experiment, regenerating lenses of stage 9–11 of Sato’s regeneration stages [12] were observed at 1 month and those of stage 13 at 3 months. At more than 1 µg/µl αNi3S2 concentration, lens regeneration was inhibited at a high rate at 1 month. But as time progressed, the incidence of regeneration rose to about 70% except at 40 µg/µl administration. Since severe destruction of eye structure was found in many cases at 40 µg/µl administration, lens regeneration could not be observed within 6 months except in one case. Simple lentectomy without drug administration invariably regenerates a lens from dorsal marginal iris as described previously [9, 10]. Effects of αNi3S2 on tumour production The incidence of tumour production was examined 6 months after the administration of various concentrations of αNi3S2 into lentectomized newt eyes (Table 2). No tumour production was found at the minimum concentration of αNi3S2 (1 µg/µl) in all cases, but tumour incidence
288
from the ventral iris rose in proportion to the increase in amounts of carcinogen. None of the tumours was induced from the dorsal iris in the present experiment. In several cases at 1 month after 10 µg/µl αNi3S2treatment, the two-layer structures of outer and inner epithelium in the marginal region of dorsal and ventral irises were disorganized and changed to clumps of aberrantly arranged cells (Fig. 1a). These regions were composed of a heterogeneous cell population with various cell shapes and nuclear sizes. Mitotic figures were found in these cells (Fig. 1b). It was noteworthy that not only the
cell mass of the ventral iris but also that of the dorsal iris at 1 month was similar to those found in tumour cells described later. However, the dorsal tumour-like cell mass vanished at 3 months and seemed to have recovered to the normal arrangement (Fig. 1c). In 10 µg/µl αNi3S2-treated eyes at 6 months, valuable features from the standpoint of regeneration and carcino-
Table 1 Lens regeneration following intraocular injection of nickel subsulfide (αNi3S2) after lentectomy in Cynops pyrrhogaster&/tbl.c:& Compound
Fig. 2a–c Typical example of the eye showing simultaneous appearance of lens regeneration from dorsal iris and tumour production from ventral iris in the same eye after intraocular injection of 10 µg/µl αNi3S2 at 6 months. a Overall view of the eye at low magnification. DI and VI dorsal and ventral iris, respectively. T tumour induced from ventral iris. b Higher magnification photomicrograph of the tumour region designated as T in a. Arrowheads indicate mitotic figures found frequently in the tumour region. c High-magnification photograph of one of mitotic figures indicated by arrowheads in b. Bar a 700 µm; b and c 60 and 25 µm, respectively&ig.c:/f
αNi3S2
Dosage µg/µl/eye
1 10 20a 40
Lens regeneration 1 month (%)
3 months (%)
6 months (%)
100 (6/6) 0 (0/6) 10 (2/20) 0 (0/6)
100 (6/6) 50 (3/6) 50 (11(22) 0 (0/6)
100 (6/6) 71 (5/7) 73 (11/15) 17 (1/6)
Numbers in parentheses, No. of regenerates/No. of available cases a Separate experiment from other series&/tbl.:
289 Table 2 Tumor incidence at 6 months after intraocular injection of potent carcinogen, nickel subsulfide (αNi3S2) into lentectomized newt eye&/tbl.c:&
Table 3 Lens regeneration following intraocular injection of noncarcinogenic compounds (amorphous nickel sulfide: am-NiS, NaCl and NiCl2) after lentectomy in Cynops pyrrhogaster&/tbl.c:&
Compound
Compound
αNi3S2
Dosage µg/µl/eye 1 10 20a 40
Tumor incidence at 6 months D-iris (%)
V-iris (%)
0 (0/6) 0 (0/6) 0 (0/15) 0 (0/6)
0 (0/6) 57 (4/7) 60 (9/15) 83 (5/6)
Numbers in parentheses, No. of tumors/No. of available cases; D, dorsal; V, ventral a Separate experiment from other series&/tbl.:
am-NiS
0.6% NaCl NiCl2
Dosage µg/µl/eye
1 10 40 0.1a
Lens regeneration 1 month %
3 months %
6 months %
17 (1/6) 0 (0/6) 0 (0/6) 100 (6/6) 100 (6/6)
100 (6/6) 100 (6/6) 33 (2/6) 100 (6/6) 100 (6/6)
100 (6/6) 100 (6/6) 83 (5/6) 100 (6/6) 100 (6/6)
Numbers in parentheses, No. of regenerates/No. of available cases a NiCl concentration is 100 times that of αNi S considering the 2 3 2 solubility of each&/tbl.: Table 4 Tumour incidence at 6 months after intraocular injection of non-carcinogenic compounds (amorphous nickel sulfide: amNiS, NaCl and NiCl2) into lectectomized newt eye&/tbl.c:& Compound
am-NiS
Dosage (µg/µl/eye) 1 10 40
0.6% NaCl NiCl2
0.1a
Tumor incidence at 6 months D-iris (%)
V-iris (%)
0 (0/6) 0 (0/6) 0 (0/6)
0 (0/6) 0 (0/6) 0 (0/6)
0 (0/6)
0 (0/6)
0 (0/6)
0 (0/6)
Numbers in parentheses: No. of tumors/No. of available cases a NiCl concentration is 100 times that of αNi S considering the 2 3 2 solubility of each &/tbl.: Fig. 3a, b Non-carcinogenic, amorphous nickel monosulfidetreated eyes. a Inhibition of lens regeneration at lowest dose (1 µg/µl) at 1 month. b Lens regeneration at 6 months after 10 µg/µl injection. Bar 200 µm&ig.c:/f
genesis were observed. Normal lens regeneration from the dorsal iris and tumour production from the ventral iris were induced simultaneously in the same eye (Fig. 2a). The tumour induced at 6 months described above had a characteristic feature similar to those at 9 months reported in the previous study except for the frequent appearance of mitotic figures (Fig. 2b, c). The features of tumorous tissue in the ventral iris differed from the inflammatory reaction in the am-NiS series described later. No tumour production from dorsal iris was observed in this series of experiments. Effects of non-carcinogenic compounds on lens regeneration and tumour production
Fig. 4a, b Lens regeneration in NaCl (a) and NiCl2 (b) treated eyes at 6 months. Bar 200 µm&ig.c:/f
Am-NiS inhibited lens regeneration at a high frequency (83% inhibition), even at the low concentration (1 µg/µl) at 1 month (Table 3 and Fig. 3a). Many neutrophils and macrophages were found in the pupillary region in some am-NiS-treated eyes, suggesting that am-NiS might have caused inflammation by its toxicity. But, 3 or 6 months later as shown in Fig. 3b, the inflammatory features found in the previous stage disappeared, and a high per-
290
centage of lens regeneration was obtained (Table 3). AmNiS did not induce any tumours from either dorsal or ventral irises even at the highest dose (Table 4). Using excess concentrations of NaCl and NiCl2, control experiments were done a test a toxic effect of vehicle and Ni ions on lens regeneration and tumour production (Tables 3, 4). No effect on lens regeneration and tumour production was observed in any series of the drug administrations (Fig. 4a, b).
Discussion Whereas the origin of the induced tumour and the relationship between regeneration and carcinogenesis were unclear in previous studies [9, 10] because of excessive drug administration, the present study clearly demonstrated that αNi3S2 inhibited lens regeneration from dorsal iris with a high incidence following injection of more than 10 µg/µl carcinogen at 1 month, and delayed the initiation of lens regeneration in proportion to the increase in the amounts of carcinogen. The tumour incidence from ventral iris at 6 months rose according to the increased amounts of carcinogen. Another important finding was that αNi3S2 did not disturb lens regeneration from dorsal iris and at the same time induced tumour formation from ventral iris in the same eye with the administration of a moderate dose (10 µg/µl/eye). This is the first demonstration of both lens regeneration from regenerative dorsal iris and tumour production from non-regenerative ventral iris in the same eye. The present result clearly indicated that the lens regeneration system in the newt has special advantages for research on regeneration and carcinogenesis, since regenerative dorsal iris and non-regenerative ventral iris are situated opposite each other on either side of the pupil, enabling dorsal and ventral iris to receive the same stimulus of carcinogen. In fact, in the present study, both irises showed different reactions to the same carcinogen: dorsal iris was resistant to carcinogen whereas ventral iris was susceptible to that drug. The antagonistic reaction to carcinogen between regenerating dorsal iris and nonregenerative ventral iris has considerable implications for the relationship between regeneration and carcinogenesis. It is important to know what happens when αNisS2 is injected into normal eyes without lentectomy. In a limb regeneration system, a supernumerary forearm developed in one case in the forearm region from the administration of N-methyl-N’-nitro-N-nitrosoguanidine (MNNG) into intact forearm [19]. Thus, the administration of αNi3S2 into normal eyes was attempted preliminarily in this study. Although stabilized results have not yet been obtained because of the technical difficulties involved, no tumour production was found at 6 months after administration in available cases (unpublished data). There is a possibility that the drug placed on the stromal surface of the iris could not reach the pigmented iris epithelial cells.
In this study, the regenerative dorsal iris proved to be resistant to carcinogens, whereas the ventral iris as nonregenerative tissue was susceptible to tumour production. However, will the regenerative dorsal iris resist carcinogens in any case? Or will the non-regenerative ventral iris always be susceptible to tumour production? The cell population that appeared in the ventral iris after αNi3S2 administration at 6 months was probably not a true tumorous tissue, because the exact identification of tumour cells in the newt is rather difficult due to the rare occurrence of spontaneous or chemically induced tumours. It is known that normal iris epithelium of the newt lacks cell replicative ability, but this begins on day 4 after lentectomy [21–24]. The percentages of the number of labeled cells and mitotic cells per iris epithelium increase from day 4 to 7. It is well known that a ventral iris equally contributes to this increase. However, in the normal condition, the partially depigmented ventral iris returns to the original level of pigmentation after the passage of three to four cell cycles. In the dorsal iris, completely depigmented cells transform into lens cells after the passage of more than two to three cell cycles. The cell population appearing in the ventral iris in the present study may have resulted only from the continuation and/or enhancement of growth caused by drug administration during the initial phase of lens regeneration. But, as shown in Fig. 2, the presence of an aberrantly proliferating cell population at 6 months, when normal regeneration has completely finished and the iris has regained its orginal state, is sufficiently abnormal in itself. Thus, it seems reasonable to regard such a cell population as tumorous tissue. The disorganized cell mass found in the dorsal marginal iris in αNi3S2-treated eye at 1 month (Fig. 1a) is suggestive. The cell mass appearing in the dorsal iris showed some similarities to tumour cells induced in the ventral iris. Nevertheless, regenerative tissue may be made to revert to normal conditions by its individual control mechanism, although it tentatively proceeded toward the early stages of carcinogenesis. Seilern-Aspang and Kratochwil [13] demonstrated that tumours induced in mucous glands in newt epithelium by the administration of chemical carcinogen eventually redifferentiated into normal tissues. The same authors reported that the regenerative area of tails in the newt, Triturus cristatus, developed tumours at a low incidence after subcutaneous injection of carcinogens. The tumors on the tails grew without invasion and differentiated into non-malignant normal tissues [14]. In two reports [3, 11], tumours were induced in regeneration-competent tissues by the administration of carcinogens. However, these works concerned non-regenerating tissues. Among regenerationcompetent tissues that have resistance to carcinogens, regenerating tissues seem more resistant than non-regenerating ones [18]. Therefore, the fact that in this study the regenerating dorsal iris failed to induce tumors is consistent with previous works indicating that regenerating tissues in limb and tail regeneration failed to induce any tu-
291
mour even with a high dose of carcinogen [18]. Zilakos et al. induced epidermal squamous cell carcinomas by implanting 20-methylcholanthrene s.c. into the scapular region of the newt, a tissue that cannot regenerate [25]. They evaluated these tumors observed in the primitive amphibian as phylogenetic conservation of caricature of tissue renewall, an idea that is derived from the study of rodent tumors. Since the ventral iris of the eye and the scapular region are considered to be the same in that they are non-regenerative tissue, it may be considered that tumors in the ventral iris in the present study are the caricatures of tissue renewal by which the dorsal marginal cells which transformed into lens cells are compensated after the cell loss during the process of lens regeneration [4]. Herreno-Saenz et al. recently revealed that the antitumour drug 3-nitrobenzothiazolo [3,2,-a] quinolium chloride (NBQ) stimulates the in vivo lens regeneration in the adult newt Notophthalmus viridescens [6]. They also showed that another antitumour drug, doxorubicin, which is an intercalating agent just like actinomycin D, strongly inhibited lens regeneration after long-term treatment. It is suggested that NBQ acts at a different level of cellular/molecular control than doxorubicin and/or actinomycin D. Although it remains unclear whether doxorubicin truly behaves as an antitumour drug in newt iris tissue, this report suggests a close relationship between a high capacity for regeneration and remarkable resistance to carcinogens in urodeles. Dedifferentiation following active cell growth is common in both regeneration and carcinogenesis. However, in regeneration, the individual control mechanism reorganizes the dedifferentiated cells to complete the tissue regeneration. In carcinogenesis, dedifferentiated cells escape from this control mechanism such as the individuation field and become tumours. Thus, it seems that urodeles having high regenerative capacities are resistant to developing tumours. Analysis of the control mechanisms of the newt may provide important information about ways to prevent the development of human cancer. In the present study, the tumour was induced by the aministration of αNi3S2, but am-NiS did not induce any tumours. It is known that the carcinogenic potencies of nickel compounds are generally inversely related to their solubilities in aqueous media. According to Sunderman Jr. [16], the nickel compounds that appear to be most carcinogenic in rodents are insoluble or sparingly soluble in water (less than 1×10−3 M), and the nickel compounds that appear to be non-carcinogenic such as NiCl2 are readily soluble in water. There is a notable exception to this rule since am-NiS (less than 1×10−5 M) was not found to be carcinogenic. In the present experiment, NiCl2 showed no carcinogenicity, suggesting a possible application of the general rule mentioned above to the newt lens regeneration system. Abbracchio et al. [1] observed that crystalline NiS particles were actively phagocytosed by all cultured cells while am-NiS was not significantly phagocytosed. They revealed that crystalline
NiS has a net negative surface charge, whereas the surface charge of the amorphous one appears to be positive. It has been considered that active uptake of carcinogen into the cell is dependent upon the interaction of the negative surface charge of carcinogen and the positive surface charge of cells. αNi3S2 showed the same remarkable phagocytotic activities in a cell culture system as did crystalline NiS. In the present study, though phagocytotic activities were not measured, it is likely that a similar phenomenon occurred in newt cells. &p.2:Acknowledgements The author wishes to express his gratitude to Tokindo S. Okada, Head of the Biohistory Research Hall, for his warm encouragement of this study. Gratitude is also expressed to Ms. Mayumi Ito of Nagoya University for her valuable discussion in preparing the manuscript and to Drs. Masaharu Kawada and Katsushi Owaribe of Nagoya University for their kind assistance and suggestions.
References 1. Abbracchio MP, Heck JD, Caprioli RM, Costa M (1981) Differences in surface properties of amorphous and crystalline metal sulfides may explain their toxicological potency. Chemosphere 10:897–908 2. Argus MF, Hoch-Ligeti C (1961) Comparative study of the carcinogenic activity of nitrosamines. J Natl Cancer Inst 27: 695–709 3. Breedis C (1952) Induction of accessory limbs and of sarcoma in the newt (Triturus viridescens) with carcinogenic substances. Cancer Res 12:861–866 4. Eguchi G, Shingai R (1971) Cellular analysis on localization of lens forming potency in the newt iris epithelium. Dev Growth Differ 13:337–349 5. Eguchi G, Watanabe K (1973) Elicitation of lens formation from the “ventral iris” epithelium of the newt by a carcinogen, N-methyl-N’-nitro-N-nitrosoguanidine. J Embryol Exp Morphol 30:63–71 6. Herreno-Saenz D, Ortiz JR, Baez A (1994) Effects of 3-nitrobenzothiazolo [3,2,-a] quinolium chloride (NBQ) and doxorubicin on lens regeneration in the adult newt: a morphological study. Differentiation 55:169–174 7. Ingram AJ (1972) The lethal and hepatocarcinogenic effects of dimethylnitrosamine injection in the newt. Triturus helveticus. Br J Cancer 26:206–215 8. Needham J (1936) New advances in the chemistry and biology of organized growth. Proc Roy Soc Med 29:1577–1626 9. Okamoto M (1987) Induction of ocular tumor by nickel subsulfide in the Japanese common newt, Cynops pyrrhogaster. Cancer Res 47:5213–5217 10. Okamoto M (1988) Inhibition of lens regeneration by nickel subsulfide in the Japanese common newt, Cynops pyrrhogaster. Dev Growth Differ 30:75–80 11. Outzen HC, Custer RP, Prehn RT (1976) Influence of regenerative capacity and innervation on oncogenesis in the adult frog (Rana pipiens). J Natl Cancer Inst 57:79–84 12. Sato T (1940) Vergleichende Studien über die Geschwindigkeit der Wolffschen Linsenregeneration bei Triton taeniatus und bei Diemyctylus pyrrhogaster. Wilhelm Roux Arch Entw Mech Org 140:570–613 13. Seilern-Aspang F, Kratochwil K (1962) Induction and differentiation of an epithelial tumour in the newt (Triturus cristatus). J Embryol Exp Morphol 10:337–356 14. Seilern-Aspang F, Kratochwil K (1965) Relation between regeneration and tumor growth. In: Kiortsis V,Trampusch HAL (eds) Regeneration in animals and related problems. NorthHolland Publishing Comp, Amsterdam, pp 452–473
292 15. Stone LS, Vultee JH (1949) Inhibition and release of lens regeneration in the dorsal iris of Triturus v viridescens. Anat Rec 103:560–561 16. Sunderman Jr FW (1973) The current status of nickel carcinogenesis. An Clin Lab Sci 3:156–180 17. Sunderman Jr FW, Maenza RM (1976) Comparisons of carcinogenicities of nickel compounds in rats. Res Commun Chem Pathol Pharmacol 14:319–330 18. Tsonis PA (1983) Effects of carcinogens on regenerating and non-regenerating limbs in amphibia (Review). Anticancer Res 3:195–202 19. Tsonis PA, Eguchi G (1981) Carcinogens on regeneration. Differentiation 20:52–60 20. Waddington CH (1935) Cancer and the theory of organizers. Nature 135:606–608
21. Yamada T (1977) Control mechanisms in cell-type conversion in newt lens regeneration. Monographs in developmental biology 13 S. Karger, Basel München Paris London New York Sydney 22. Yamada T, Roesel M (1969) Activation of DNA replication in the iris epithelium by lens removal. J Exp Zool 171:425–432 23. Yamada T, Roesel M (1971) Control of mitotic activity in Wolffian lens regeneration. J Exp Zool 177:119–128 24. Yamada T, Roesel M, Beauchamp JJ (1975) Cell cycle parameters in dedifferentiating iris epithelial cells. J Embryol Exp Morphol 34:497–510 25. Zilakos NP, Tsonis PA, Rio-Tsonis KD, Parchment RE (1992) Newt squamous carcinoma proves phylogenetic conservation of tumors as caricatures of tissue renewal. Cancer Res 52: 4858–4865
© Springer-Verlag 1997
O R I G I NA L A RT I C L E
&roles:Mitsumasa Okamoto
Simultaneous demonstration of lens regeneration from dorsal iris and tumour production from ventral iris in the same newt eye after carcinogen administration &misc:Accepted in revised form: 17 February 1997
&p.1:Abstract It is well known that urodeles have the most powerful regenerative capacities among vertebrates, but there is little realisation that they are extremely resistant to spontaneous or chemically induced tumours. Regeneration and carcinogenesis have been considered to be two sides of the same mechanism. Since antagonism between regeneration and carcinogenesis was expected in previous studies, the present study was intended to clarify this relationship in greater detail by changing the amounts of carcinogen stepwise. When 1 µl nickel subsulfide solution was administered in various amounts (1 µg/µl ~40 µg/µl) into lentectomized newt eyes, the delay of initiation in lens regeneration for 6 months and an increased inhibition rate of lens regeneration at 6 months were observed in proportion to the increase in carcinogen dosage. The tumour production rate increased in accordance with the increase in the amounts of carcinogen. The conspicuous result obtained in the present study was that lens regeneration from dorsal iris and tumour induction from ventral iris occurred simultaneously in the same eye after administration of moderate amounts (10 µg/µl) of carcinogen. These data clearly indicated that the regenerating dorsal iris is persistently resistant to carcinogen, whereas the ventral iris, which cannot regenerate lens, is susceptible to tumour induction. This strongly suggests that the lens regeneration system in the newt has special advantages for research on the relationship between regeneration and carcinogenesis.&bdy:
Introduction Amphibians, especially urodeles, have remarkable regenerative capacities. The newts can regenerate limbs, tail and even neural tissue throughout life. In addition, special mention should be made of the remarkable resistance of the newt to most carcinogens. For example, no M. Okamoto Division of Biological Science, Graduate School of Science, Nagoya University, Chikusa-ku, Nagoya 464-01, Japan&/fn-block:
tumour production was found when carcinogens were given to the newt at a concentration about 100 times higher than that required to induce 100% tumour formation in rats [2, 7]. Embryologists have long been fascinated with the relationship between regeneration and carcinogenesis. According to Waddington [20] and Needham [8], the fundamental fact about cancerous tissue is that it has escaped from the normal growth-controlling agent of the body, called the “individuation field”. The main characteristic of this field is that all tissue lying within it tends to be built up into a complete embryo, and in any one part of the field all tissue tends to be organized into an organ corresponding to that part. Usually, this field has gradually become weak or has vanished during development. However, the field has been considered to be persistent and potent until adulthood in animals that are capable of regeneration. One of the key ways to verify these ideas is to observe whether a tumour can be induced in regeneration-competent tissues by applying chemical carcinogens. Several investigations have thus been made with regard to limb and tail regeneration in amphibia. In 1983, the effects of chemical carcinogens on regenerating and non-regenerating tissues in amphibia were reviewed by Tsonis [18]. Non-regenerating tissue in his report meant that the regenerating process is not induced in the regenerationcompetent limbs or tails. According to Tsonis, there is a considerable degree of controversy in the literature concerning the effects of carcinogens on regenerating and non-regenerating limbs. But, he suggested that in both instances regenerative capacity may be a deterrent to neoplasia in limb tissues. There is a genuine concern that such a problem cannot be solved by attempts using limb or tail regeneration systems, because neighboring tissue such as the trunk is difficult to use as a strict control for the regeneration of competent tissue. On the other hand, the lens regeneration system in the newt may provide a valuable tool for research on such a problem, because regenerating dorsal and non-regenerative ventral iris are located opposite
286
each other on either side of the pupil, and they belong to the same tissue that is defined as an “iris” in anatomical criteria. Therefore, the dorsal and ventral iris can be compared to each other under the same conditions in relation to a carcinogen placed in the pupil. There are few reports on the effects of carcinogens on lens regeneration in the newt except for those of Stone and Vultee [15] and Eguchi and Watanabe [5]. Stone and Vultee were able to prevent lens regeneration for 4–5 months by placing a crystal of a carcinogen in the pupil after lens extirpation. However, there was no description of tumour formation. Eguchi and Watanabe observed abnormal lens regeneration with supernumerary lenses from non-regenerative ventral and lateral irises, but found no tumour production. In previous studies [9, 10], the author examined the effects of a high dosage of a potent carcinogen, nickel subsulfide (αNi3S2), on lens regeneration and tumour production in the adult newt eye. The results demonstrated that lens regeneration was inhibited and ocular tumours were induced at a high incidence at 9 months. In the previous report, a population consisting of disarranged and aberrantly proliferating pigmented cells, which occupied the entire globe of the eye at 9 months, was identified as melanoma-like malignant tumours by light and electron microscopic observations. They were poorly cohesive cells with irregular or small round nuclei, prominent nucleoli and abundant cytoplasm with numerous pigment granules. They showed characteristics similar to the human epithelioid melanoma-like cell. Although the tumour cells invaded surrounding tissues, no metastasis to organs outside the eye was observed. In electron microscopic observations, crystalline αNi3S2 was occasionally found in some melanoma-like cells in samples fixed at 11 months. Acid-fast bacteria, the presence of which is reported in infectious granuloma cells, were not found in the cells. This was the first report on the induction of ocular melanoma in the amphibia class. The tumour was assumed to originate from the iris, because the aberrantly proliferating cell population was found in the iris in treated eyes at 3 months. Although the demonstration of antagonism between regeneration and carcinogenesis was expected in the previous studies, it was unclear whether the induced tumour is truly derived from regenerative dorsal iris, because the newt iris is composed of both regenerative dorsal and non-regenerative ventral iris. Moreover, since the dosage of carcinogen in the previous study was excessive and tumour cells occupied the entire globe of the eyeballs, it was not possible to identify the original point of an induced tumour. In the present study, to clarify the origin of an induced tumour and the relationship between regeneration and carcinogenesis, various amounts of αNi3S2 and amorphous nickel monosulfide (am-NiS) were injected into lentectomized newt eyes, and their effects on the incidence and origin of tumour production and lens regeneration were examined in detail. Nickel compounds are known to have a carcinogenic property. The carcinogenicity of αNi3S2 in man and ex-
perimental animals was reviewed by Sunderman Jr. [16], who mentioned that αNi3S2 is the most carcinogenic among numerous nickel compounds. The administration of this drug to animals results in malignant tumours at the sites of deposition in almost all cases [17]. It was reported that αNi3S2 and am-NiS possess extremely different carcinogenicities, despite their similar physical and chemical properties [17]. According to the study, the incidence of sarcoma in rats receiving a single injection of αNi3S2 was almost 100%, compared to none from am-NiS. The present results demonstrated inhibition of lens regeneration from high-dose administration of αNi3S2, delayed regeneration initiation from a medium dose, and normal regeneration from a low dose. In addition, the following were found in the same eye: dorsal iris regenerated a lens without tumour production, whereas ventral iris produced a tumour when moderate amounts of carcinogen were injected into the pupillary region. These data also implied that regenerative tissue, that is, the dorsal iris, is persistently resistant to carcinogens, while the ventral iris, considered non-regenerative tissue, is susceptible to tumour production.
Methods Animals Adult Japanese newts, Cynops pyrrhogaster, were used in the present experiment. The animals were purchased from Hamamatsu Seibutsu Kyouzai (Shizuoka, Japan), kept in the laboratory and fed with pork meal chips. Test compounds Crystalline nickel subsulfide (αNi3S2 in crystalline low temperature form) and information related to it were kindly provided by INCO Ltd., Toronto, Ontario, Canada. Amorphous nickel monosulfide (am-NiS hereinafter) should be used as soon as possible after formation, since it tends to oxidize easily to sulfate when stored. Thus, we made it ourselves according to a memo by D.F. Colton, kindly provided by J.S. Warner (INCO Ltd.). Preparation of am-NiS was as follows: 0.3 l 200 g/l Ni2+ solution was placed in a 1.5-l beaker and stirred at a rate just below the point of vortexing; 36.4 g/l S2− solution was then added from a fine-tipped burette at 10 ml/min until 0.45 l had been added (the amount of S 2− required to precipitate only 50% of the Ni 2+ in solution as NiS). The vortex rate was increased gradually as the volume of the slurry increased, but always kept just below the point of vortexing. After addition of the S2− solution (0.45 l in 45 min), the slurry was digested for 10 min. The final slurry was filtered on a 15-cm buchner funnel using no. 41 filter paper, and washed with 4×200 ml distilled water. The filter cake was repulped in 1 l distilled water, slurried for 10 min, and filtered (as above) and washed with 4×200 ml distilled water. A quantitative nickel mesurement was made in a representative dried portion from the wet filter cake according to the dimethylglyoxime method in the textbook of inorganic chemistry. Standard nickel solution (1.00 mg Ni/ml) was purchased from Nakarai Chemicals (Kyoto, Japan). The amorphous nature of the am-NiS made in the present experiment was checked by X-ray diffraction pattern, which was kindly measured by Dr. Hitoshi Oosato, Nagoya Industrial University. NaCl and NiCl 2 were purchased from Nakarai Chemicals. The am-NiS content (1, 10, and 40 µg/µl described in the present experiment means the same Ni
287 content as 1, 10, and 40 µg/µl αNi3S2, respectively. Since the solubility of NiCl2 is up to 105 times that of am-NiS and 106 times that of αNi3S2, respectively, and in addition, the localized concentrations of administered am-NiS and αNi3S2 seemed to be very high, we used 0.1 µg/µl as NiCl2 contents. The NaCl content was decided by the same principle as for NiCl2.
and embedded in Paraplast x-tra (Sigma Chemical Company, St. Louis, USA). Serially cut 10-µm-thick sections were stained with haematoxylin and eosin by the routine method.
Lentectomy and intraocular injection of carcinogen and non-carcinogenic compounds
Effects of αNi3S2 on lens regeneration
The method used was basically the same as in the previous studies [9, 10]. Animals were anesthetized with 0.1%–0.2% MS222 and transferred to an operation dish with Hanks’ balanced salt solution modified to 80% ionic strength for the newt. Both lenses were removed with a pair of forceps through a slit in the cornea. Injection into the eye chamber was performed with a glass micropipette, in this case a micro-type Labopettor (Hirschman Laborgerate, Germany). It was superior to the hand-made glass micropipette used in the previous studies, because it guaranteed a precise injection volume and smooth injection by piston movement. The suspension of αNi3S2 and am-NiS in sterile 0.6% NaCl solution was continuously agitated with a magnetic stirring apparatus to ensure that constant amounts of chemicals were aspirated into the micropipette set up to a volume of 1 µl.
Histological examinations Experimental animals were decapitated at 1, 3, and 6 months after injection. The isolated heads were fixed with Bouin’s fluid, decalcified in solution A (Wako Pure Chemical Industries Ltd, Osaka, Japan) by the Plank Rychlo method overnight, then dehydrated
Fig. 1a–c The reaction of dorsal iris to the carcinogen nickel subsulfide (αNi3S2, 10 µg/µl) administered into lentectomized newt eyes. a Disorganized cell mass found in dorsal iris at 1 month. b Mitotic figure (arrowhead) observed at the region shown in a. c Normally appearing dorsal iris with regenerating lens at 3 months. Bar a and c 300 µm, b 30 µm&ig.c:/f
Results
Lens regeneration from dorsal iris was examined at 1, 3 and 6 months after lentectomy following administration of αNi3S2 at various concentrations (Table 1). In all cases of injection of 1 µg/µl, which is the minimum concentration of αNi3S2 selected in the present experiment, regenerating lenses of stage 9–11 of Sato’s regeneration stages [12] were observed at 1 month and those of stage 13 at 3 months. At more than 1 µg/µl αNi3S2 concentration, lens regeneration was inhibited at a high rate at 1 month. But as time progressed, the incidence of regeneration rose to about 70% except at 40 µg/µl administration. Since severe destruction of eye structure was found in many cases at 40 µg/µl administration, lens regeneration could not be observed within 6 months except in one case. Simple lentectomy without drug administration invariably regenerates a lens from dorsal marginal iris as described previously [9, 10]. Effects of αNi3S2 on tumour production The incidence of tumour production was examined 6 months after the administration of various concentrations of αNi3S2 into lentectomized newt eyes (Table 2). No tumour production was found at the minimum concentration of αNi3S2 (1 µg/µl) in all cases, but tumour incidence
288
from the ventral iris rose in proportion to the increase in amounts of carcinogen. None of the tumours was induced from the dorsal iris in the present experiment. In several cases at 1 month after 10 µg/µl αNi3S2treatment, the two-layer structures of outer and inner epithelium in the marginal region of dorsal and ventral irises were disorganized and changed to clumps of aberrantly arranged cells (Fig. 1a). These regions were composed of a heterogeneous cell population with various cell shapes and nuclear sizes. Mitotic figures were found in these cells (Fig. 1b). It was noteworthy that not only the
cell mass of the ventral iris but also that of the dorsal iris at 1 month was similar to those found in tumour cells described later. However, the dorsal tumour-like cell mass vanished at 3 months and seemed to have recovered to the normal arrangement (Fig. 1c). In 10 µg/µl αNi3S2-treated eyes at 6 months, valuable features from the standpoint of regeneration and carcino-
Table 1 Lens regeneration following intraocular injection of nickel subsulfide (αNi3S2) after lentectomy in Cynops pyrrhogaster&/tbl.c:& Compound
Fig. 2a–c Typical example of the eye showing simultaneous appearance of lens regeneration from dorsal iris and tumour production from ventral iris in the same eye after intraocular injection of 10 µg/µl αNi3S2 at 6 months. a Overall view of the eye at low magnification. DI and VI dorsal and ventral iris, respectively. T tumour induced from ventral iris. b Higher magnification photomicrograph of the tumour region designated as T in a. Arrowheads indicate mitotic figures found frequently in the tumour region. c High-magnification photograph of one of mitotic figures indicated by arrowheads in b. Bar a 700 µm; b and c 60 and 25 µm, respectively&ig.c:/f
αNi3S2
Dosage µg/µl/eye
1 10 20a 40
Lens regeneration 1 month (%)
3 months (%)
6 months (%)
100 (6/6) 0 (0/6) 10 (2/20) 0 (0/6)
100 (6/6) 50 (3/6) 50 (11(22) 0 (0/6)
100 (6/6) 71 (5/7) 73 (11/15) 17 (1/6)
Numbers in parentheses, No. of regenerates/No. of available cases a Separate experiment from other series&/tbl.:
289 Table 2 Tumor incidence at 6 months after intraocular injection of potent carcinogen, nickel subsulfide (αNi3S2) into lentectomized newt eye&/tbl.c:&
Table 3 Lens regeneration following intraocular injection of noncarcinogenic compounds (amorphous nickel sulfide: am-NiS, NaCl and NiCl2) after lentectomy in Cynops pyrrhogaster&/tbl.c:&
Compound
Compound
αNi3S2
Dosage µg/µl/eye 1 10 20a 40
Tumor incidence at 6 months D-iris (%)
V-iris (%)
0 (0/6) 0 (0/6) 0 (0/15) 0 (0/6)
0 (0/6) 57 (4/7) 60 (9/15) 83 (5/6)
Numbers in parentheses, No. of tumors/No. of available cases; D, dorsal; V, ventral a Separate experiment from other series&/tbl.:
am-NiS
0.6% NaCl NiCl2
Dosage µg/µl/eye
1 10 40 0.1a
Lens regeneration 1 month %
3 months %
6 months %
17 (1/6) 0 (0/6) 0 (0/6) 100 (6/6) 100 (6/6)
100 (6/6) 100 (6/6) 33 (2/6) 100 (6/6) 100 (6/6)
100 (6/6) 100 (6/6) 83 (5/6) 100 (6/6) 100 (6/6)
Numbers in parentheses, No. of regenerates/No. of available cases a NiCl concentration is 100 times that of αNi S considering the 2 3 2 solubility of each&/tbl.: Table 4 Tumour incidence at 6 months after intraocular injection of non-carcinogenic compounds (amorphous nickel sulfide: amNiS, NaCl and NiCl2) into lectectomized newt eye&/tbl.c:& Compound
am-NiS
Dosage (µg/µl/eye) 1 10 40
0.6% NaCl NiCl2
0.1a
Tumor incidence at 6 months D-iris (%)
V-iris (%)
0 (0/6) 0 (0/6) 0 (0/6)
0 (0/6) 0 (0/6) 0 (0/6)
0 (0/6)
0 (0/6)
0 (0/6)
0 (0/6)
Numbers in parentheses: No. of tumors/No. of available cases a NiCl concentration is 100 times that of αNi S considering the 2 3 2 solubility of each &/tbl.: Fig. 3a, b Non-carcinogenic, amorphous nickel monosulfidetreated eyes. a Inhibition of lens regeneration at lowest dose (1 µg/µl) at 1 month. b Lens regeneration at 6 months after 10 µg/µl injection. Bar 200 µm&ig.c:/f
genesis were observed. Normal lens regeneration from the dorsal iris and tumour production from the ventral iris were induced simultaneously in the same eye (Fig. 2a). The tumour induced at 6 months described above had a characteristic feature similar to those at 9 months reported in the previous study except for the frequent appearance of mitotic figures (Fig. 2b, c). The features of tumorous tissue in the ventral iris differed from the inflammatory reaction in the am-NiS series described later. No tumour production from dorsal iris was observed in this series of experiments. Effects of non-carcinogenic compounds on lens regeneration and tumour production
Fig. 4a, b Lens regeneration in NaCl (a) and NiCl2 (b) treated eyes at 6 months. Bar 200 µm&ig.c:/f
Am-NiS inhibited lens regeneration at a high frequency (83% inhibition), even at the low concentration (1 µg/µl) at 1 month (Table 3 and Fig. 3a). Many neutrophils and macrophages were found in the pupillary region in some am-NiS-treated eyes, suggesting that am-NiS might have caused inflammation by its toxicity. But, 3 or 6 months later as shown in Fig. 3b, the inflammatory features found in the previous stage disappeared, and a high per-
290
centage of lens regeneration was obtained (Table 3). AmNiS did not induce any tumours from either dorsal or ventral irises even at the highest dose (Table 4). Using excess concentrations of NaCl and NiCl2, control experiments were done a test a toxic effect of vehicle and Ni ions on lens regeneration and tumour production (Tables 3, 4). No effect on lens regeneration and tumour production was observed in any series of the drug administrations (Fig. 4a, b).
Discussion Whereas the origin of the induced tumour and the relationship between regeneration and carcinogenesis were unclear in previous studies [9, 10] because of excessive drug administration, the present study clearly demonstrated that αNi3S2 inhibited lens regeneration from dorsal iris with a high incidence following injection of more than 10 µg/µl carcinogen at 1 month, and delayed the initiation of lens regeneration in proportion to the increase in the amounts of carcinogen. The tumour incidence from ventral iris at 6 months rose according to the increased amounts of carcinogen. Another important finding was that αNi3S2 did not disturb lens regeneration from dorsal iris and at the same time induced tumour formation from ventral iris in the same eye with the administration of a moderate dose (10 µg/µl/eye). This is the first demonstration of both lens regeneration from regenerative dorsal iris and tumour production from non-regenerative ventral iris in the same eye. The present result clearly indicated that the lens regeneration system in the newt has special advantages for research on regeneration and carcinogenesis, since regenerative dorsal iris and non-regenerative ventral iris are situated opposite each other on either side of the pupil, enabling dorsal and ventral iris to receive the same stimulus of carcinogen. In fact, in the present study, both irises showed different reactions to the same carcinogen: dorsal iris was resistant to carcinogen whereas ventral iris was susceptible to that drug. The antagonistic reaction to carcinogen between regenerating dorsal iris and nonregenerative ventral iris has considerable implications for the relationship between regeneration and carcinogenesis. It is important to know what happens when αNisS2 is injected into normal eyes without lentectomy. In a limb regeneration system, a supernumerary forearm developed in one case in the forearm region from the administration of N-methyl-N’-nitro-N-nitrosoguanidine (MNNG) into intact forearm [19]. Thus, the administration of αNi3S2 into normal eyes was attempted preliminarily in this study. Although stabilized results have not yet been obtained because of the technical difficulties involved, no tumour production was found at 6 months after administration in available cases (unpublished data). There is a possibility that the drug placed on the stromal surface of the iris could not reach the pigmented iris epithelial cells.
In this study, the regenerative dorsal iris proved to be resistant to carcinogens, whereas the ventral iris as nonregenerative tissue was susceptible to tumour production. However, will the regenerative dorsal iris resist carcinogens in any case? Or will the non-regenerative ventral iris always be susceptible to tumour production? The cell population that appeared in the ventral iris after αNi3S2 administration at 6 months was probably not a true tumorous tissue, because the exact identification of tumour cells in the newt is rather difficult due to the rare occurrence of spontaneous or chemically induced tumours. It is known that normal iris epithelium of the newt lacks cell replicative ability, but this begins on day 4 after lentectomy [21–24]. The percentages of the number of labeled cells and mitotic cells per iris epithelium increase from day 4 to 7. It is well known that a ventral iris equally contributes to this increase. However, in the normal condition, the partially depigmented ventral iris returns to the original level of pigmentation after the passage of three to four cell cycles. In the dorsal iris, completely depigmented cells transform into lens cells after the passage of more than two to three cell cycles. The cell population appearing in the ventral iris in the present study may have resulted only from the continuation and/or enhancement of growth caused by drug administration during the initial phase of lens regeneration. But, as shown in Fig. 2, the presence of an aberrantly proliferating cell population at 6 months, when normal regeneration has completely finished and the iris has regained its orginal state, is sufficiently abnormal in itself. Thus, it seems reasonable to regard such a cell population as tumorous tissue. The disorganized cell mass found in the dorsal marginal iris in αNi3S2-treated eye at 1 month (Fig. 1a) is suggestive. The cell mass appearing in the dorsal iris showed some similarities to tumour cells induced in the ventral iris. Nevertheless, regenerative tissue may be made to revert to normal conditions by its individual control mechanism, although it tentatively proceeded toward the early stages of carcinogenesis. Seilern-Aspang and Kratochwil [13] demonstrated that tumours induced in mucous glands in newt epithelium by the administration of chemical carcinogen eventually redifferentiated into normal tissues. The same authors reported that the regenerative area of tails in the newt, Triturus cristatus, developed tumours at a low incidence after subcutaneous injection of carcinogens. The tumors on the tails grew without invasion and differentiated into non-malignant normal tissues [14]. In two reports [3, 11], tumours were induced in regeneration-competent tissues by the administration of carcinogens. However, these works concerned non-regenerating tissues. Among regenerationcompetent tissues that have resistance to carcinogens, regenerating tissues seem more resistant than non-regenerating ones [18]. Therefore, the fact that in this study the regenerating dorsal iris failed to induce tumors is consistent with previous works indicating that regenerating tissues in limb and tail regeneration failed to induce any tu-
291
mour even with a high dose of carcinogen [18]. Zilakos et al. induced epidermal squamous cell carcinomas by implanting 20-methylcholanthrene s.c. into the scapular region of the newt, a tissue that cannot regenerate [25]. They evaluated these tumors observed in the primitive amphibian as phylogenetic conservation of caricature of tissue renewall, an idea that is derived from the study of rodent tumors. Since the ventral iris of the eye and the scapular region are considered to be the same in that they are non-regenerative tissue, it may be considered that tumors in the ventral iris in the present study are the caricatures of tissue renewal by which the dorsal marginal cells which transformed into lens cells are compensated after the cell loss during the process of lens regeneration [4]. Herreno-Saenz et al. recently revealed that the antitumour drug 3-nitrobenzothiazolo [3,2,-a] quinolium chloride (NBQ) stimulates the in vivo lens regeneration in the adult newt Notophthalmus viridescens [6]. They also showed that another antitumour drug, doxorubicin, which is an intercalating agent just like actinomycin D, strongly inhibited lens regeneration after long-term treatment. It is suggested that NBQ acts at a different level of cellular/molecular control than doxorubicin and/or actinomycin D. Although it remains unclear whether doxorubicin truly behaves as an antitumour drug in newt iris tissue, this report suggests a close relationship between a high capacity for regeneration and remarkable resistance to carcinogens in urodeles. Dedifferentiation following active cell growth is common in both regeneration and carcinogenesis. However, in regeneration, the individual control mechanism reorganizes the dedifferentiated cells to complete the tissue regeneration. In carcinogenesis, dedifferentiated cells escape from this control mechanism such as the individuation field and become tumours. Thus, it seems that urodeles having high regenerative capacities are resistant to developing tumours. Analysis of the control mechanisms of the newt may provide important information about ways to prevent the development of human cancer. In the present study, the tumour was induced by the aministration of αNi3S2, but am-NiS did not induce any tumours. It is known that the carcinogenic potencies of nickel compounds are generally inversely related to their solubilities in aqueous media. According to Sunderman Jr. [16], the nickel compounds that appear to be most carcinogenic in rodents are insoluble or sparingly soluble in water (less than 1×10−3 M), and the nickel compounds that appear to be non-carcinogenic such as NiCl2 are readily soluble in water. There is a notable exception to this rule since am-NiS (less than 1×10−5 M) was not found to be carcinogenic. In the present experiment, NiCl2 showed no carcinogenicity, suggesting a possible application of the general rule mentioned above to the newt lens regeneration system. Abbracchio et al. [1] observed that crystalline NiS particles were actively phagocytosed by all cultured cells while am-NiS was not significantly phagocytosed. They revealed that crystalline
NiS has a net negative surface charge, whereas the surface charge of the amorphous one appears to be positive. It has been considered that active uptake of carcinogen into the cell is dependent upon the interaction of the negative surface charge of carcinogen and the positive surface charge of cells. αNi3S2 showed the same remarkable phagocytotic activities in a cell culture system as did crystalline NiS. In the present study, though phagocytotic activities were not measured, it is likely that a similar phenomenon occurred in newt cells. &p.2:Acknowledgements The author wishes to express his gratitude to Tokindo S. Okada, Head of the Biohistory Research Hall, for his warm encouragement of this study. Gratitude is also expressed to Ms. Mayumi Ito of Nagoya University for her valuable discussion in preparing the manuscript and to Drs. Masaharu Kawada and Katsushi Owaribe of Nagoya University for their kind assistance and suggestions.
References 1. Abbracchio MP, Heck JD, Caprioli RM, Costa M (1981) Differences in surface properties of amorphous and crystalline metal sulfides may explain their toxicological potency. Chemosphere 10:897–908 2. Argus MF, Hoch-Ligeti C (1961) Comparative study of the carcinogenic activity of nitrosamines. J Natl Cancer Inst 27: 695–709 3. Breedis C (1952) Induction of accessory limbs and of sarcoma in the newt (Triturus viridescens) with carcinogenic substances. Cancer Res 12:861–866 4. Eguchi G, Shingai R (1971) Cellular analysis on localization of lens forming potency in the newt iris epithelium. Dev Growth Differ 13:337–349 5. Eguchi G, Watanabe K (1973) Elicitation of lens formation from the “ventral iris” epithelium of the newt by a carcinogen, N-methyl-N’-nitro-N-nitrosoguanidine. J Embryol Exp Morphol 30:63–71 6. Herreno-Saenz D, Ortiz JR, Baez A (1994) Effects of 3-nitrobenzothiazolo [3,2,-a] quinolium chloride (NBQ) and doxorubicin on lens regeneration in the adult newt: a morphological study. Differentiation 55:169–174 7. Ingram AJ (1972) The lethal and hepatocarcinogenic effects of dimethylnitrosamine injection in the newt. Triturus helveticus. Br J Cancer 26:206–215 8. Needham J (1936) New advances in the chemistry and biology of organized growth. Proc Roy Soc Med 29:1577–1626 9. Okamoto M (1987) Induction of ocular tumor by nickel subsulfide in the Japanese common newt, Cynops pyrrhogaster. Cancer Res 47:5213–5217 10. Okamoto M (1988) Inhibition of lens regeneration by nickel subsulfide in the Japanese common newt, Cynops pyrrhogaster. Dev Growth Differ 30:75–80 11. Outzen HC, Custer RP, Prehn RT (1976) Influence of regenerative capacity and innervation on oncogenesis in the adult frog (Rana pipiens). J Natl Cancer Inst 57:79–84 12. Sato T (1940) Vergleichende Studien über die Geschwindigkeit der Wolffschen Linsenregeneration bei Triton taeniatus und bei Diemyctylus pyrrhogaster. Wilhelm Roux Arch Entw Mech Org 140:570–613 13. Seilern-Aspang F, Kratochwil K (1962) Induction and differentiation of an epithelial tumour in the newt (Triturus cristatus). J Embryol Exp Morphol 10:337–356 14. Seilern-Aspang F, Kratochwil K (1965) Relation between regeneration and tumor growth. In: Kiortsis V,Trampusch HAL (eds) Regeneration in animals and related problems. NorthHolland Publishing Comp, Amsterdam, pp 452–473
292 15. Stone LS, Vultee JH (1949) Inhibition and release of lens regeneration in the dorsal iris of Triturus v viridescens. Anat Rec 103:560–561 16. Sunderman Jr FW (1973) The current status of nickel carcinogenesis. An Clin Lab Sci 3:156–180 17. Sunderman Jr FW, Maenza RM (1976) Comparisons of carcinogenicities of nickel compounds in rats. Res Commun Chem Pathol Pharmacol 14:319–330 18. Tsonis PA (1983) Effects of carcinogens on regenerating and non-regenerating limbs in amphibia (Review). Anticancer Res 3:195–202 19. Tsonis PA, Eguchi G (1981) Carcinogens on regeneration. Differentiation 20:52–60 20. Waddington CH (1935) Cancer and the theory of organizers. Nature 135:606–608
21. Yamada T (1977) Control mechanisms in cell-type conversion in newt lens regeneration. Monographs in developmental biology 13 S. Karger, Basel München Paris London New York Sydney 22. Yamada T, Roesel M (1969) Activation of DNA replication in the iris epithelium by lens removal. J Exp Zool 171:425–432 23. Yamada T, Roesel M (1971) Control of mitotic activity in Wolffian lens regeneration. J Exp Zool 177:119–128 24. Yamada T, Roesel M, Beauchamp JJ (1975) Cell cycle parameters in dedifferentiating iris epithelial cells. J Embryol Exp Morphol 34:497–510 25. Zilakos NP, Tsonis PA, Rio-Tsonis KD, Parchment RE (1992) Newt squamous carcinoma proves phylogenetic conservation of tumors as caricatures of tissue renewal. Cancer Res 52: 4858–4865