Some Factors That Influence the Growth of Neoplastic Cells*

B. E. Kline, M.S., and H. P. Rusch, M.D.

(From the McArdle Memorial Laboratory, Medical School, University o~ Wisconsin, Madison, 6, Wisconsin) (Received for publication July 13, 1944)

It has been suggested that the formation of a cancer appearance of tumors. The details of dosage and time occurs not as a continuous single process but rather relationships are presented in Table I. as a series of biological changes (3, 6, 13, 21). The Croton oil and its resin were tested because of their various steps involved are a matter of conjecture, but pronounced augmenting effect on carcinogenesis as presumably neoplastic cells must first be formed and previously described (2, 3). A solution of naphtho- then they must proliferate to form a grossly recogniz- quinone was used because it is an irritant known to able tumor. The present communication is concerned inactivate certain enzymes (20). Sodium sulfide was with some of the factors that influence tumor forma- applied because it represents a different type of chemi- tion, with special emphasis on the proliferation of the cal irritant, and incisions repeated at regular intervals established neoplastic cell. The agents used included were employed because tumors have been reported to croton oil, naphthoquinone, sodium sulfide, and tissue occur in scars and healing wounds (7, 21, 22, 24). In damage as produced by mechanical wounding. the latter group 3 or 4 parallel skin incisions about The application of certain chemical irritants (2, 3), 2 cm. in length, which penetrated the dermis, were heat (8), cold (1), or physical trauma (7, 13, 21) made with a sharp scalpel. to a region previously treated with a subcarcinogenic The effect of these various forms of irritation on amount of hydrocarbon is known to precipitate the areas previously treated with methylcholanthrene is formation of tumors in a relatively short period of shown in Table I. No visible tumors were present in time, but there is some question whether the secondary any of the animals when the applications of hydrocar- treatment completes the processes involved in the actual bon were discontinued at the end of the preliminary formation of the cancer cell or whether it is in some treatment, and only a few neoplasms eventually devel- manner responsible for the proliferation of cancer cells oped in the controls. However, when a 0.1 per cent already formed. Hence the effect of mechanical wound- solution of naphthoquinone was applied subsequent to ing was also determined on the growth of neoplastic the hydrocarbon treatment many tumors developed. cells whose existence was well established. Thus 14 carcinomas and 4 papillomas arose in this The effect o/ chemical and physical agents on carcino- group (group 8) as compared to 5 epitheliomas and genesis o/ the skin by rnethylcholanthrene.--Two series 7 papillomas observed in the control group. This find- of 100 young adult albino mice were kept on wood ing is of interest not only because it is new but also shavings in metal box cages and were given dog bis- because the effect is comparable to that obtained with cuit and water ad libitum throughout the experiment. croton oil, a known active epicarcinogen (3). In series The mice were painted twice weekly with a solution I, the application of croton oil resin stimulated tumor of methylcholanthrene applied with a camel's hair production to the extent that there were 4 epitheliomas brush to the interscapular area. The length of the and 14 papillomas as compared to only 1 epithelioma initial treatment with the hydrocarbon varied in the in the control group 4 months after discontinuing the two series according to the strength of the solution and hydrocarbon treatment; in series II, after 6 months the susceptibility of the mice. They were then divided there were 14 epitheliomas and 8 papillomas in the into groups of 25 each. In each of the series one group group that received croton oil in contrast to 5 cancers was set aside as a control and received no further treat- and 7 papillomas found in the control group. In fact croton oil was ahnost as effective in developing the ment, while in the remaining groups the area of skin tumors as the continued application of the methyl- that had previously been painted with the hydrocarbon cholanthrene itself (group 6). This parallels the ex- was subjected to various further treatments. The ani- perience of Berenblum (2, 3). Chronic physical trauma mals were carefully examined every 2 weeks for the in the form of repeated incisions of the skin had a * This investigation was aided by a grant from the Jonathan slight but definite augmenting effect on tumor forma- Bowman Fund for Cancer Research. tion. This effect appeared early, and in the first 2 762

Downloaded from cancerres.aacrjournals.org on October 5, 2021. © 1944 American Association for Cancer Research. Kline et al.--lnfluences on Neoplastic Cell Growth 763 months following the discontinuation of hydrocarbon p-dimethylaminoazobenzene were selected, and mid- treatment it seemed to be as effective as the croton line abdominal incisions extending to the sternum were resin applications. However, in spite of continued made under ether anesthesia in such a manner that treatment, a regression of tumors was observed; there the livers could be easily inspected (17). The pres- were 8 papillomas after 2 months and only 6 after ence of large hepatomas having been confirmed, the 4 months. All tumors were localized in or very close wounds were closed with silk sutures. Six of the rats to the area of the scar. Any experiment in which re- were given no further treatment following the first peated incisions are attempted is apt to be complicated incision, and were killed after 40 days. In the remain- by infection of the area, and this could have an adverse ing 34 rats, periodic irritation of the wounded area effect on tumor formation. Furthermore, multiple in- was repeated from 1 to 5 times. This treatment con- cisions are apt to result in the physical destruction or sisted of cutting, perforating, and drawing a threaded elimination of small tumors. Our results, although needle through the healing wound in numerous places limited, suggest that wound healing has at least a at weekly intervals. The animals subjected to repeated

TABLI: I: "I"111.. EFFEf:T OF CHEMICAL AND PHYSICAL AGENTS ON CARCINOGENI-SIS OF TIIE SKIN BY METItYL('tlOLANTtlRENE (The croton oil resin was dissolved in ethyl alcohol; the naphthoquinone and croton oil were applied as solutions in benzyl alcohol) Incidence of tumors ,\ 1 too. 2 mos. 3 mos. 4 or 6 mos. Strain of Initial Subsequent nfice treatment Group t reatment Ca P -N Ca P ~" ca P N Ca P N ABC Stock 0.3'% methylcholan- 1 None 0 0 25 0 0 25 1 0 24 1 0, 24 (Series I) threne in benzene ap- 2 Sat. sol. of Na.~S once 0 0 24 0 0 24 0 0 24 0 2 22 pl led twice weekly for weekly for 7 weeks 8 weeks 3 Incisions of skin twice 0 3 19 1 8 13 1 6 15 1 6 15 a inonth for 4 months 4 0.1% croton oil resin 0 2 23 2 8 15 3 11 11 4 14 7 tx~:ice weekly for 4 months

.Strain C 0.2% lncthylcholan- 5 None 0 1 23 0 3 21 1 4 19 5 7 12 (Series II) threne in benzyl al- 6 Continuation with 0 1 24 1 12 12 4 10 11 22 0 .3 cohol applied twice mcthylcholanthrcnc for weekly for 16 weeks 6 months 7 1% croton oil twice 0 5 20 4 6 15 4 9 12 14 8 3 weekly for 6 months 8 0.1% naphthoquinone 0 6 19 4 4 17 4 5 16 14 4 7 twice weekly for 6 months * Number of months after discontinuing the initial treatment. t The results listed for series i are for 4 months and those in series II for 6 months. Ca -- Carcinoma. P ~ Papilloma. N ---- Negative. transient augmenting effect on the development of irritation were killed at various times during the 5 tumors, and confirm the observations of other investi- week period, close inspection was made of the inner gators (7, 13, 21, 22). surface of the wounds, and pieces of tissue were re- The results of the experiment with sodium sulfide moved for histological examination from representative also demonstrates that it is possible to have irritation sites of the wounded area. without obtaining a cocarcinogenic effect. Weekly ap- Of the 34 rats receiving periodic irritation in the plications of sodium sulfide to the skin in amounts form of wounding, 6 died soon after the operation be- sufficient to produce considerable erythema did not cause of large liver tumors, leaving 28 for subsequent augment tumor formation, although we do not know examination. In 19 of these rats, from 1 to 5 small whether the irritation produced by this chemical was tumor nodules were found growing in the wounds. as intense, prolonged, or penetrating as that obtained These nodules were located in the incision itself or by the methods that proved to be effective. closely adjacent to it, and varied in size from those just The effect oJ tissue damage as produced by wound- visible to others that were about 3 ram. in diameter ing on the growth o[ established tumors.mForty rats (Fig. 1). The wounds were usually bound down to having palpable liver tumors induced by feeding the liver tumors with fresh fibrinous adhesions, and

Downloaded from cancerres.aacrjournals.org on October 5, 2021. © 1944 American Association for Cancer Research. 764 Cancer Research nodules were also seen in these adhesions but no metas- Microscopic study of the wounds also revealed the tases were observed to any other part of the abdominal presence of many nests of similar tumor cells of varying cavity. It is probable, therefore, that the tumor nodules sizes. In fact, some wounds were literally sprinkled

J-re. ] ]'l~;. 2

l:t<.. 3 ]q,;..t

I:l~;. 1. [.c)w pc)wcl of }lcp;.lt(nlla ill '~:()lln(h'd ;.Iica ~)f alltcli()r ;l[)({~)ll]il];|] wall ~)t a rill:. Note .~car tissut, fFIHII I-CCe.'I]~I;.lp;.lr(d.(Hlly incid(,, iust lc) right ~,t: tuln~)r n(,dulc. "l'w~, suture ho!l's containing pieces ,)f silk arc also visible in picture. l:u;. 2.--Iligh powvr ,>f hclmtoma rt'm~)vcd fr~m~ rl_.ccntly w~)undvd area similar to thai sh(,wn in lei v. 1. Fro. 3. This high p()wcr ph~)tomicrograph i!lustratcs dcgcncration and nccrosis that occu,,~ in hcI3atomas after healing Im)c- csscs have bccn completed. lqc. 4.--This shrews pronounced degeneration of hcpatoma found in rcgi,m ~)f an old hcalcd abd-minal incision. Note dense connective tissue that has encompassed the tumor. This is an older wound than that from which the hcpatoma shown in Fig. [3 was obtained. developed from cells that reached the wound by direct with nests of hepatoma cells. These tumors grew most contact. Histological examination demonstrated that vigorously in tile fresh wounds and attained their larg- the nodules were hepatomas identical with the primary est size about 2 to 3 weeks after the injury. After this liver tumor with which they were in contact (Fig. 2). period the nuclei of the tumor cells became pyknotic

(Fig. 3), and regression in the older wounds was rapid types of experimental evidence. First, when methyl- (Fig. 4). No tumors were found after 40 days in the cholanthrene is applied to the skin of mice for a time wounds of the 6 control rats whose midline incisions short of that necessary to induce tumors, the subse- were not subjected to postoperative interference, and quent resumption of the hydrocarbon treatment will with one exception hepatomas were absent in the fully precipitate tumor formation even though 3 to 4 months healed wounds that had been subjected to repeated intervene between the 2 periods of hydrocarbon appli- injury. In one animal the hepatoma continued grow- cation (10). Such observations suggest that the initial ing until death. The conclusion should not be drawn application of hydrocarbon produced alterations in the that all experimental tumors will regress in healed tissues that persisted for at least 4 months, although wounds. Generally the rat hepatoma does not metasta- there is no proof that any neoplastic cells had been size widely, nor is it extremely invasive. It is very completely formed as the result of the initial treatment likely that a more malignant tumor would not be with hydrocarbon. so apt to regress in a healed wound once it had at- Rous and his associates (13, 22) have shown that tained the size achieved by some of the hepatomas ob- wounding rabbit ears caused the appearance of papil- served in this experiment. lomas at sites from which they had previously regressed. The wounding process was therefore thought I)ISCUSSION to stimulate the growth of quiescent neoplastic cells. The growth of hepatoma cells in wounds in the present The present experiments are in harmony with the experiment likewise suggests the stimulating influence general idea that cancer formation occurs following a of irritation on the proliferation of established neo- sequence of biological changes (3, 6, 13). These plastic cells. changes might be classified as follows: I. Induction period. This is the period during which It seems likely that the noncarcinogenic agents that neoplastic cells are formed. favor the growth of neoplastic cells during the critical II. Critical or reversible period. This starts at the period exert their effects on established cancer cells moment the neoplastic cells are formed, and represents rather than on the actual formation of such cells, since that stage in which cellular proliferation is delicately croton oil applied to the skin of mice prior to benz- balanced. Presumably during this phase the cells may pyrene had no effect on tumor formation (3). Further- proliferate for a time, some may die, and others may more, if heat, cold, or trauma was applied during the lie quiescent for varying periods. This is the transi- induction period the process of carcinogenesis was tional period, during which the cells are most sus- actually retarded, presumably by an interference with ceptible to the influence of their environment. It could some of the biochemical changes necessary in the de- vary in time from a few moments to periods of weeks, velopment of the neoplastic cell (23). months, or even years. Whether a tumor cell dies or In an investigation in which the growth rates of ex- proliferates depends upon the balance between the perimental tumors w'ere closely followed, Blum (4) proliferative capacity of the cell and the local tissue demonstrated that the growth rate of the neoplasm did resistance. not remain constant throughout the development nor III. Pcriod of progression. This may be considered did it follow the same pattern in all tumors of the as that period when the neoplastic cells have gained same type. His data suggest that opposing growth ascendancy over the forces that hold them in control. and regressive processes determine the time at which a There is no sharp line between the second and third tumor becomes established, and indicate that tumor periods, but in general the latter is one of relatively cells do not escape and assume their own essential unrestricted, invasive growth, during which regressions proliferation rates, but that the rates of growth are to are infrequent. The exact size that a ncoplasm must a great extent dominated by the controlling influence attain before it reaches this stage varies, no doubt, with of the tissue. different tumors. Indeed the separation of tumor for- The experiments of Fischer also suggest that quies- mation into various periods is only an arbitrary scheme, cent neoplastic cells may be stimulated to activity if intended to assist in the clarification of our views certain favorable conditions are imposed (9). He on carcinogenesis. Actually the different stages are transplanted pieces of apparently normal mammary probably not clearly separable, and within a neo- tissue to other parts of the same mouse. After a few plastic focus, at any period, tumor cells are being days these pieces were removed and reinoculated into formed, others are dividing, and still others are dying. other areas of the same animal and the procedure was The final outcome is the result of a balance of the repeated many times. Several tumors eventually grew various factors involved. in the region of the transplant. These results might The postulation of a second, or critical, period in be explained on the basis that the frequent wounding the development of a tumor is supported by several incident to the inoculations had exerted a favorable in-

fluencc on the growth of resting or latent neoplastic other wounds is more frequent than could be accounted cells present in the original transplants. The opposite for by the mere laws of chance (24). These observa- effect, a regression of tumor growth in an unfavorable tions could be explained on the basis of the suggestions environment, was recently demonstrated by Briggs and just offered. Grant (5). They observed that a tumor grew well in the tail of a tadpole for several weeks but that it SUMMARY stopped growing even before the metamorphic regres- Methylcholanthrene was applied to the skin of mice sion of the tail commenced. There appeared to be some for a period just short of the formation of visible relationship between metamorphosis and the prolifera- tumors. The skin was then treated with naphtho- tion of neoplastic cells. quinone, croton oil, croton oil resin, or sodium sulfide, It is a well known fact that injury to tissues is fol- or it was subjected to frequent incisions. Many tumors lowed by an improved blood supply, an elevated tem- developed following exposure to naphthoquinone, cro- perature, and an increase of nutritive substances to the ton oil, croton oil resin, or to the processes of wound region. Conditions favorable for cellular proliferation repair. The sodium sulfide was ineffective. probably persist until the injury has been repaired. Frequent incisions made through the abdominal Loofbourow has described growth-promoting nucleo- wall of rats with primary hepatomas resulted in nests tides produced by injured cells (11, 12), while Menkin of proliferating hepatoma cells in the regions of the has demonstrated stimulating agents in infammatory wound. These neoplasms regressed during the later exudates (14). The latter investigator injected an stages of the healing process. It is suggested that such exudate from inflamed pleura into the ear of a rabbit transplanted cancer cells found a favorable environ- and observed striking proliferation, characterized by ment for growth in the region of the healing wounds, hyperplasia and metaplasia of the normal epithelial and that the defense mechanism of tissues that inhibit layer and by loci of keratinization. Small nodules in such growth normally is decreased in areas where re- the cartilage were also observed. Among other things, parative processes are active. the exudate contained lactic acid and glucose (15, 16). These results are correlated with the suggestion that These findings support the suggestion by Potter that tumor formation may be divided into three phases: injury and irritation result in the breakdown of (a) period of induction, during which the neoplastic adenosinetriphosphate (18, 19). The adenosine and cell is formed; (b) critical or reversible period, in inorganic phosphate resulting from this primary reac- which the growth of the neoplastic cell is in delicate tion could lead to an increased glycolysis and a balance and depends upon the local tissue environment; localized vasodilatation (18), and thus would favor and (c) the final period of progression, during which the proliferation of fibroblasts as well as any neoplastic growth is relatively unchecked. cells that might be present in the involved area. As the healing process nears completion the breakdown of The authors are deeply indebted to Professors V. R. Potter and adenosinetriphosphate would diminish until, finally, C. A. Baumann for valuable suggestions, especially as concerns tissue oxidations would be able to esterify phosphate the Discussion, on the division of cancer formation into three phases. and adenosine at concentrations that are below the growth threshold. On this basis the continuation of REFERENCES any factor, be it gross irritation or some more subtle 1. Br.REr~'ZLUM, I. Further Investigations on the Induction of agent, that would accelerate the breakdown of adeno- Tumours with Carbon Dioxide Snow. Brit. J. Exper. sinctriphosphate or hamper its resynthesis would lead Path., 11:208-211. 1930. to a milieu favorable for proliferation. The subsequent 2. BERENI3LUM, I. The Cocarcinogenic Action of Croton Resin. Cancer Research, 1:44-48. 1941. growth of neoplastic cells in a healed tissue would then 3. BERENI3Lt3,Xi,I. The Mechanism of Carcinogenesis. A Study depend upon the inherent malignancy of the cells and of the Significance of Cocarcinogcnic Action and Related the size the tumor had attained before the healing Phenomena. Cancer Research, 1:807-814. 1941. process was completed. A short period of irritation 4. BI_UM,H. F. Growth Rate and Development of Tumors In- with the subsequent increase in connective tissue and duced with Ultraviolet Radiation. J. Nat. Cancer Inst., decrease in blood supply would restrain the continued 4:559-569. 1944. 5. BRIc,Gs, R., and GRAr~-T, R. Growth and Regression of Frog proliferation of the cells, whereas a longer period of Kidney Carcinoma Transplanted into the Tails of Perma- healing would aid the tumor to attain a size that would nent and Normal Tadpoles. Cancer Research, 3:613-620. be less dependent upon the local tissue environment. 1943. There has been much uncertainty as to the connec- 6. CR^.xrv.R, W. On Experimental Carcinogenesis: The Local Resistance of the Skin to the I)cvclopment of Malignancy. tion between irritation or injury and cancer formation. Brit. 1. Expcr. Path., 10:335-346. 1929. Apparently irritation per se is not carcinogenic (23), 7. D~rr.r,tAx, H. T. The Part Played by Injury and Repair in and yet the development of tumors in old burns or the Development of Cancer, with Some Remarks on

Growth of l".xperimental Cancers. Proc. Roy. Soc. Med. Inquiry into the Mechanism Involved in Diabetes En- (Sect. Pathol.), 20:19-20. 1927; also Brit. M. J., hanced by Inflammation. Am. J. Physiol., 138:396-407. 1:872. 1927. 1943. 8. D~s LmX'ERIS, M. J. A. Precancer and Carcinogenesis. Am. 17. MILLtR, J. A., MINER, D. L., Rvs('H, H. P., and B^t:M^~.,4, J. Cancer, 40:1-46. 1940. C.A. Diet and Hepatic Tumor Formation. Cancer Re- 9. FISCHER, A. The Theory of the Developmental Physiology search, 1:699-708. 1941. of Malignant Tumors. Am. J. Cancer, 31:1-20. 1937. 18. POVXER, V. R. Biological Energy Transformations and the 10. LAwK, P. S., MooR~, P. R., Rusc~1, H. P., and BAUM^.~S, Cancer Problem. Advances in Enzymology, 4:201-256. C. A. Some Additive Effects of Carcinogenic Hydro- 1944. carbons. Cancer Research, 2:189-192. 1942. 19. POTrER, V. R. Unpublished data. 11. Lool~BouRow, J. R., DWVER, C. M., and L^XE, M. M. Pro- 20. I'OTVE~, V. R., and Du Bols, K. P. Studies on the Mecha- liferation-13romoting Intercellular Hormones. I. Quantita nism of Hydrogen Transport in Animal Tissues..VI. In- tive Studies of Factors Produced by Injured Animal hibitor Studies with Succinic Dehydrogenase. J. Gen. Tissue Cells. Biochem. J., 34:432-441. 1940. Physiol., 26:391-404. 1943. 12. LooF~ot:Row, J. R., WEBB, A. M., Loovuotmow, D. G., and 21. PULLINGER, B. D. The Localisation of Experimental Tumors L1sco, H. Further Observations on the Increased Yield of in Scars and Healing Wounds. J. Path. & Bact., 55:301- Nucleic Acid from Irradiated Yeast. Nature, London, 309. 1943. 149:328. 1942. 22. Rous, P., and KIND, J. G. Conditional Neoplasms and Sub- 13. M^cKE~,'ZIE, I., and Rows, P. The Experimental Dis- threshold Neoplastic States; A Study of Tar Tumors of closure of Latent Neoplastic Changes in Tarred Skin. Rabbits. J. Exper. Med., 73:365-390. 1941. I. Exper. Med., 73:391-416. 1941. 23. RuscH, H. P. Extrinsic Factors that Influence Carcinogenesis. 14. MEYKI.,,,, V. Cellular Injury in Relation to Proliferative and Physiol. Rev., 24:177-204. 1944. Neoplastic Response. Cancer Research, 1:548-556. 1941. 24. SLVE, M., HOLMES, H. F., and WELLS, H. (;. Primary Spon- 15. MENrls, V. Biochemical Factors in Inflammation and taneous Squamous (;ell Carcinomas in Mice. Studies on Diabetes Mellitus. Arch. Path., 34: 182-195. 1942. the Incidence and Inheritability of Spontaneous Tumors 16. MENKIN, V. Glyconeogenesis and Cellular Injury. A Further in Mice. J. Cancer Research, 6:57-86. 1921.

B. E. Kline and H. P. Rusch

Cancer Res 1944;4:762-767.

Updated version Access the most recent version of this article at: http://cancerres.aacrjournals.org/content/4/12/762.citation

E-mail alerts Sign up to receive free email-alerts related to this article or journal.

Reprints and To order reprints of this article or to subscribe to the journal, contact the AACR Publications Subscriptions Department at [email protected].

Permissions To request permission to re-use all or part of this article, use this link http://cancerres.aacrjournals.org/content/4/12/762.citation. Click on "Request Permissions" which will take you to the Copyright Clearance Center's (CCC) Rightslink site.