Elastic, and Collagenic Tissue During Jipiaermai @Arcinogenesis Morphological and Chemical Investigation of Dermal

Home , Elastic fiber

Morphological and Chemical Investigation of Dermal Elastic, and Collagenic Tissue During Jipiaermai1@' .1 1I'@'i @arcinogenesis -

CHUNG K. MAt

(From the Department of Anatomy, Washington Univerrity School of Medicine, St. Louis 10, Missouri)

Before cancer develops in mouse epidermis in three zones. The first is the zone of tumor growth fluenced by methylcholanthrene, generations of devoid of elastic tissue; the second is the zone of epidermal cells live for months in a strangely a! cellular infiltration with decreased elastic tissue; tered kind of existence. Their lives are in a sense while the third is the zone of pressure character askew for they carry on despite many marked de ized by increased elastic tissue. Turning to the ficiencies and some excesses in essential materials. skins of mice, tTlesco-Stroganowa (11) induced can The features of their manner of life have been re cer by applications of coal tar and found in the re cently summarized (1). gion of epithelial hyperplasia a marked increase in Investigation of these epidermal cells is now be elastic tissue and in mast cells while about the ing supplemented by study of accompanying cancer elastic tissue was absent. It is to be noted changes in the underlying dermis. Thus, a new first that in both these tissues of humans and mice method for staining nerve endings and fibers has the investigators reported decrease in elastic tissue been devised (2, 3) and will be used. Sebaceous next to the cancer and a more distant increase, glands andhair follicles (4, 5), lodged in the der second that the conditions described were those mis, undergo interesting modifications. The lipase after cancer development. There is apparently no activity of sebaceous glands and of dermal fatty account in the literature of the sequence of altera tissues, as revealed by the Gomori reaction, is tions in dermal elastic and collagenic tissue during lost (6). Alterations observed in the number and epidermal carcinogenesis except one which was re fluorescence of mast cells (7) suggested operation ported by Bierich (12) who made a study of mouse of the spreading factor (hyaluronidase) in the skin at 5 minute, 6 hour, 24 hour, and 5 day inter breaking down of resistance to invading cancer vals after x-ray irradiation. But some useful infor cells possibly offered by hyaluronic acid in the mation is available concerning elastic and colla dermal tissue fluid. Preliminary experiments in genic tissue on nonmalignant conditions of the volving injections of hyaluronidase in the vicinity skin (13 to 24). I am grateful to Dr. E. V. Cowdry of cancer transplants seem to indicate that metas for suggestions. tasis is thereby promoted (8). On the theory that dermal mast cells may play some part in the pro MATERIAL AND METHODS duction of the hyaluronic acid other experiments histological data.â€”Female Swiss mice 4 weeks old are being made (9). Thus, data are being gathered were employed: (1) Five served as normal controls. on dermal modifications with a view to building up (2) Others received applications of 0.6 per cent methyl a comprehensive picture of what happens. cholanthrene in benzene delivered to their backs by a In this paper morphological and chemical modi No. 4 brush three times a week in exactly the same fications in dermal, elastic, and collagenic tissue manner that is regularly done in work by others in this group. Of these 30 mice, 5 were killed after 10, @0,30, are reported in the hope that they will supply clues 40, 50, and 60 days. (3) Still others received similar ap to the changing physical and chemical properties plications of pure benzene. Of these 10 mice, 5 were of the dermal ground substance. The literature on killed at 10 and @0days. this subject is scanty. It has been reported (10) Excised samples of skin were placed on thick paper that human skin with carcinoma is divisible into and fixed in fresh Helly's fluid. Some tissues were em

@ S Aided by grants from the National Cancer Institute, the bedded in tissuemat and sections, cut @2,5thick, were C. F. Kettering Foundation, and the Women's Advertising stained in alcoholic orcein or in Weigert's resorcin Club of St. Louis. fuchsin. Others were embedded in hard paraffin and f In part fulfillmentfor the degreeof Doctor of Philosophy, sections, cut 8 ,.t thick, were colored by Mallory's stain, Washington University. the quad stain, or by Foot's method. The direction of 481

sections was transverse to the long axis of the body in to the slope of the hair, that is longitudinally to the some instances and longitudinal in others. long axis of the body and likewise to the hair fol Attempts were made quantitatively to measure licles. Part of the erector muscle is located to the alterations observed in the elastic tissue (25). A disc, right of the hair follicle in the obtuse angle formed ruled in large and small squares, inserted in the ocular between it and the epidermis. Elastic fibers en was calibrated by reference to a micrometer slide. At the magnification employed one of the small squares circle the distal portions of the hair follicles, @ covers square microns of tissue. Twenty-five of these stretch between them and along the erector small squares were included in each square. muscles. Some fibers are concentrated just be Chemical data.â€”Female Swiss mice of the same age neath the epidermis, especially at the attachments (4 weeks)wereusedandthemethylcholanthrenewas of these muscles, and seem, as so-catted Herx applied in the same manner. For elastin, chemical heimer's fibers, to be attached to the epidermis. analyses were made of 90 mice killed in lots of 18 on 0, Individual fibers are approximately 1.2js in thick 10, @0,30, and 60 days; for collagen, analyses were ness. made of 108 mice killed in lots of 18 on 0, 10, @0,30, 40, But the actual orientation and arrangement of and @0daysfrom the initiation of application. Six deter elastic tissue in living skin is undoubtedly some minations were made at each stage and the averages are given in the tables. what different. When still living skin is freshly cx For normal mice, the skin was removed from almost cised, it can be seen to shrink noticeably so that the entire back, while from the treated mice it was only when flattened out on a piece of stiff paper for fixa taken from the area painted with carcinogen. Pannicu tion it occupies a smaller area than it did originally lus carnosus was carefully removed from both with a in the animal. The extent of this shrinkage has scalpel. Epidermis was separated from dermis by the been measured in biopsy specimens of antecubital heat method of Baumberger, Suntzeff, and Cowdry skin by Evans, Cowdry, and Nielson (29). Skins (@6). of young persons in their thirties lost 38 to .50 per The elastin content of dermis for normal mice and mice of 10, @0,30, and 60 days, and the collagen con cent of their original areas. In mice this shrinkage tent of dermis for normal mice and mice of 10, @0,30, is more difficult to determine accurately because of 40, and 50 days after the first application of carcinogen extreme hairiness. There is also a further shrinkage were determined exactly as described by Lowry, @il1i- of at least 10 per cent owing to fixation and other gan, and Katersky (@7). Since the water content in the steps in the preparation of sections. Consequently dermis was markedly affected by carcinogen, all dermis the elastic tissue of the living animal is more samples were frozen and dried for 48 hours before ex spread out, made up of thinner fibers, and under traction with alkali in order that the results could be greater tension than Figure 1 indicates. calculated on a uniform dry weight basis. It is also to be noted that if the sections repre The percentages thus obtained sufficed to show the relative amounts of elastin and collagen. To compare sented in Figures 1 to 4 were of the thickness the elastin and collagen content in various stages of usually employed in histopathological examina carcinogenesis, an absolute measurement of elastin and tions, say 10 ;s, instead of 25 @,the amount of collagen is necessary. Hence, further work was done by elastic tissue contained in them would be much taking unit areas of dermis and weighing both before less. The advantage of these thick sections is that and after being dried at a temperature of 110Â°C.(23). with a binocular microscope the arrangement of Unit area of dermis was taken by means of a quadri fibers in depth can be better studied. Individual lateral razor-blade device which covers an area of @275 fibers can be traced through longer distances and square mm. (1 1 mm. X@25 mm.). The skin sample was alterations in their girth can be detected. The obtained directly from the carcass of the shaved mouse photomicrographs give only an artificial, flattened by using this instrument in as uniform a manner as pos sible to minimize distortion. The dry weight of the out picture in one plane of the fibers. dermis of a unit area of dermis was thus determined, By counting the elastic fibers in many 25 square and by multiplying it by the percentage of either elastin micra areas a rough measure was obtained of the or collagen of the corresponding stage, the absolute number of elastic fibers. The dermis selected for weight of elastin or collagen of that special stage was such determinations waslocated between hair fol obtained. licles and at some distance from them so as to To separate the dermal part from the epithelial part avoid the concentrations of elastic tissue in their of the cancer is impossible; consequently, the elastin and immediate vicinity. Also, in making the counts the collagen content of the cancer were compared with that upper margin of the large square in the ocular as of fresh normal skin, that is,of dermis plus epidermis and the results were calculated on a wet weight basis. nearly as possible was superposed on the distal cx tremities of the layer of basal epidermal cells so OBSERVATIONS that in each specimen fibers in an equal depth of Figure 1 illustrates the appearance of elastic tis dermis covered by the small squares were counted. sue in a section of normal mouse skin cut parallel These counts showed a gradual and progressive

Downloaded from cancerres.aacrjournals.org on September 30, 2021. © 1949 American Association for Cancer Research. Fio. 1.â€”Skinof normal mouse, cut parallel to the hair cur shown. X 350. rent to show the elastic fibers and their relation to the hair FIG. 8.â€”Skin of mouse after 60 days. Islands of elastic tis follicles and erector pill. Mag. X 850. sue, (the so-called â€œelasticamimicaâ€•). X 350. Fm. 2.â€”Skinof mouse 40 days after initial application of Fm. 4.â€”Skin of mouse likewise after 60 days. Masses of carcinogen. Increase of elastic fibers cut longitudinally is cancer cells with no elastic fibers. X 350.

increase of elastic fibers in specimens examined 10, found in the stroma of the resulting cancer is @0,and 30 days after the beginning of applications shown by Figure 4. of the skin of 0.6 per cent methyicholanthrene in Figure 5 shows the dermis of normal mouse. It benzene. After 10 days individual fibers became is composed of densely arranged collagenic fibers thickened to about 1.5 @sand retained this thick and a few connective cells, the cytoplasm of which ness to the stage shown in Figure 3. The peak in is not easily seen and the identification of which is number seemed to be at 30 days for at 40 days only marked by smaller, darker nuclei. there was a noticeable decrease which became very Figures 6 and 7 are pictures from mice 30 days much greater at 50 days. The fibers became so after treatment with carcinogen. Though there closely bound together at 60 days that counting seems a little difference between these two one was impossible. thing is common, that is they become more cellular. This increase in elastic fibers to a maximum at The connective tissue cells are larger, irregular in 30 days in the specimens examined is definite but shape, with abundant granular cytoplasm and some individual variations may exist. The vertical large nuclei. The collagenic fibers appear to be thickness of the dermis in preparations of the fragmented in comparison with the long wavy skins of normal untreated mice ranged from 60 to course of that of normal dermis. @ 90 This thickness rapidly increased with the The stroma of cancer is composed of loosely ar repeated applications of the carcinogen. In the 30 ranged collagenic fibers and is much diminished in and 40 day specimens, thicknesses of @00and 400 bulk in comparison with the invading epithelial @zwere measured. If the dermis had, on the con trary, decreased in thickness, the increase in elas cells. This is shown in Figure 8. tic fibers might have been attributed to crowding The results of chemical analysis together with of approximately the same number into a thinner the standard errors of the mean are shown in the sheet of dermis. following tables:

TABLE 1

PERCENTAGE OF ELASTIN AND COLLAGEN IN FROZEN AND DRIED DERMIs o days 10days 10 days 30 days 40 days 30 days 60 days Elastin 1.75Â±0.22 1.57Â±0.06 1.88Â±0.14 3.06Â±0.16 2. 23Â±0. 11 Collagen 31.2 Â±1.17 17.5 Â±0.57 12.8 Â±0.98 24.5 Â±0.72 16.9Â±1.03 15.2Â±0.51

The appearance of elastic tissue at 40 days is in TABLE 2 dicated in Figure 3. The elastic tissue immediately PERCENTAGE OF ELASTIN AND beneath the epidermis is reduced but the amount COLLAGEN IN FRESH SAMPLES stretching in a band between remnants of hair fol Fresh dermis Cancer licles is certainly increased. Some of the fibers are Elastin 1.46Â±0.09 0.38Â±0.03 also of larger girth. Careful study showed a little Collagen 10.4 Â±0.55 1.47Â±0.20 irregularity in thickness of individual fibers. A few detached or broken ends were slightly swollen. No TABLE S change could be detected in their affinity for spe DIFFERENCE IN WEIGHT OF A UNIT AREA (275 MM.) cific stains. OF MOUSE DERMIS, BEFORE AND AFTER ThEAT At 60 days entangled masses not unlike the so MENT WITH METHYLCHOLANTHRENE called elastica mimica in facial skin (15, 16, 17, 30) dryMeanweightMean of wetMean of were found and are shown in Figure @.A large dermiswaterinof dermisweight of gramscontentUntreated gramsin mass made of deeply staining degenerated fibers is skin to be seen a little to the left of center. A larger and MCtreatedskin0.0496Â±0.00890.171@Â±0.0320.0246Â±0.00400.0508Â±0.001751.4%71.3% less dense mass, in which individual fibers can more clearly be distinguished, is further to the left. Dry weight of unit area dermis was determined In the course of epidermal carcinogenesis, the on the thirtieth day. Table S shows a 40 per cent elastic fibers lose their attachments to the epider relative increase in water cofitent. The weight of mis as this tissue becomes hyperplastic. As the hair formed elements is doubled. Therefore in absolute follicles degenerate and disappear this process of weight the dermis on the thirtieth day contains a! detachment is advanced so that these islands of most four times as much elastin and over 60 per elastic tissue appear to lose form and become cent more collagen than that of normal dermis. If granular, fibrous vestiges. a curve were plotted, the elastin content would be That elastic tissue fades out and can hardly be similar though it would not quite coincide with

FIG. 5.â€”Dermis of normal, uiitreated mouse. There are Fm. 7.â€”Mousedermis SOdays after application of carcino numerous densely arranged collagenous fibers but few con gen, to illustrate the dominant connective tissue cells. nective tissue cells. X 350. Fia. 6.â€”Dermisof mouse 30 days after application of car FIG. 8.â€”Illustrating that the connective tissue stroma is cinogen. Connectivetissue cellsare muchincreasedin sizeand diminished in bulk in comparison with the epithelial cells in in number. X 350. cancer.

that obtained by plotting the fiber counts. In both hyaluronic acid and chondroitin sulphuric acid there would be a rise followed by a fall. Fiber . content amount to @4.5Â±5.7 and @6;@Â±4.7 counts are impossible in collagen fibers. mgm. per 100 gui. of fresh tissue, respectively. In jection of hylauronidase seems to facilitate the DISCUSSION spreading of cancer (8). Some change may there Study of methyicholanthrene-induced epider fore occur in dermal hyaluronic acid after applica mal carcinogenesis by this group has largely tion with carcinogen. But data are still lacking. stressed the chemical and morphological altera The thickening of dermis referred to is probably tions of the dermis. This report has considered accompanied by edema, the possible relationship quantitative shifts in some of the connective tissue of which to changes in elastic fibers has been dis elements of the dermis under the standardized con cussed by Dick (14) ; but it is not clarified by these ditions employed here. experiments. An initial increase in amount of dermal elastic The chemical procedures of Lowry (27) for and collagenic fibers after exposure to methyl preparation of elastin and collagen paved the way cholanthrene has been described; this is followed for analysis of connective tissue fiber changes in by a loss of fibers. Little significance is attached to carcinogenesis. The method for elastin is simple the time scale of alterations in fiber content since and depends upon the relative insolubility of this samples for analysis were taken at 10-day inter material in dilute alkali. It is to be expected that vals. Thus, the increase in elastic fibers first noted any other materials insoluble in dilute alkali will in the 10-day sample may actually reflect a situa be concentrated with elastin. As yet we have de tion existing several days earlier, and the peak in vised no procedure for separation of the mucopro elastic fiber content of the dermis described at 30 tein of elastic tissue from the alkali resistant days may occur some time between 21 and 39 residue . days. The essential point is that application of SUMMARY methyicholanthrene to the skin of the mouse re Marked changes take place in dermal elastic tis sults in a transient increase in both elastic and col sue during epidermal carcinogenesis induced by lagenic fibers followed by a pronounced drop in the cutaneous applications of 0.6 per cent methyl content of these fibers. cholanthrene in benzene. The number of elastic In view of the fact that so little is known of the fibers is increased 10 days after the first applica physiology of connective tissue fibers, it is difficult tion. Through %0 day specimens this increase is to attempt to evaluate the significance of the progressive and attains a maximum at 30 days. changes here reported. Three clear possibilities After 10 days individual fibers thicken to a diame seem to exist. Variations in connective tissue fibers ter of about 1.5 s. Some become rather irregular reflect alterations in the behavior of the ubiquitous in.girth at about 40 days but in general they retain fibroblasts, and in surface adsorptive capacities of this diameter to approximately 60 days. From 30 the fibers. Third, and perhaps most significant, days onward the number of fibers decreases. Fol loss of a network of fibers from the dermis may lowing loss of their epithelial anchorages a few fi facilitate metastatic activities of the squamous hers exhibit slightly swollen ends and at 60 days cancer cells. all elastic tissue is reduced to fibrous islands of Some information is available on the interrela material in dermis otherwise free of elastic fibers. tions of fibroblasts and connective tissue fibers. In primary cancers 60 days and more after the Morrione (31) induced liver cirrhosis with carbon first application of methylcholanthrene the dermis tetrachloride and p-dimethylaminoazobenzene in is almost devoid of elastic fibers. This primary in albino rats. In the course of fibrosis he noted a pro crease and later decrease in elastic fibers was sup liferation of fibroblasts. After stopping application plemented by direct chemical analysis of elastin by of hepatotoxic substances, the fibrous tissue under Lowry's method which revealed a similar primary went resorption. He attributed the latter to the ac increase and later decrease. tivity of liver cells, the function of which is in turn Other dermal changes are those of collagenic dependent upon diet. Recovery from cirrhosis is fibers and connective tissue cells. As carcinogenesis impaired by a high fat, low protein diet. is in progress, the dermis tends to become more Hass and MacDonald (3@) have demonstrated cellular. With chemical analysis of collagen by in tissue cultures that depression of pH restricts Lowry's method, first a slight increase and then a collagen fiber deposition and enhances resorption. decrease were found. Factors possibly effecting the This has been discussed by Cowdry (33). increase and decrease of elastic and collagenic fi In 1949, Pearce and Watson (34) analyzed hu bers were discussed but for elucidation of the real man skin and concluded that, in fresh samples, the mechanism involved, further study is necessary.

REFERENCES Mimicaâ€• bei verschiedenen R.assen. Dermat. Ztschr., 62: 95â€”104,1931. 1. COWDRY,E. V. Epidermal Carcinogenesis. J.A.M.A., 135: 18. Wouzits, C. Die Querstreifung der kollagenen Binde 408â€”411,1947. gewebsfibrille. Virchow's Arch. f. path. Anat., 312: 292â€” 2. Cnu, C. H. U. Staining of Nerve Endingsin Mouse Epider 302, 1944. mis by Feulgen's Nucleal Reaction. Science, 106:70, 1947. 19. Ci@nx, G. L., and SCHAAD,J. A. X-ray Diffraction of 8. LIANG, H. M. A New Method for Staining Nerves and Tendon and Intestinal Wall Collagen. Radiology, 27:339â€” Their Endings. Anat. Rec., 99:511â€”522, 1947. 356,1936. 4. StrnTzxs'r, V., CARRUTHERS,C., and COWDRY,E. V. Role 20. Bnaa, R. S. Long X-ray Diffraction Spacings of Collagen. of Sebaceous Glands and Hair Follicles in Epidermal Car J. Am. Chem. Soc., 64:727, 1942. cinogenesis. Cancer Research, 7:439â€”443, 1947. 21. DAY, T. D. Connective Tissue Cementing Substance. Na @ 5. LIANG, H. M. Localized Changes in Methyicholanthrene ture,159:100â€”101,1947. Treated Epidermis. Cancer Research, 8:211â€”219, 1948. 22. Lv@cn, F. W. Elastic Tissue in Fetal Skin. Arch. Dermat. 6. KUNG, S. K. Lipase Activity during Experimental Epider and Syph., 29:57, 1934. mal Carcinogenesis. J. Nat. Cancer Inst., 1949. In press. 23. STEARNS, M. L. Studies of the Development of Connective 7. CRAMER,W., and Szawsow, W. L. Mast Cells in Experi Tissue in Transparent Chamber in the Rabbit Ear. Am. J. mental Skin Carcinogenesis. Cancer Research, 4: 601-616, Anat.,66:133â€”175,1940. 1944. 24. STEARNS, M. L. Studies of the Development of Connective 8. Snn'soN, W. L., and GOPAL-AYENGAR,A.R. Influence of Tissue in Transparent Chamber in the Rabbit Ear. Am. J. Injections of Hyaluronidase on Transplants of Squamous Anat., 67:55â€”97, 1940. Cell Carcinomata. (Presented at the Chicago Meeting of 25. LINDHOLM, E. tYber die Schwankungen in der Verteilung the American Association for Cancer Research). 1949. In der elastischen Fasern in der menschlichen Haut, als press. Beitrag zur Konstitutionspathologie. Frankfurt. Ztschr. f. 9. MA, S. C. Effect of Hyaluronidase Injections on Dermal Path., 42:394â€”414, 1981. Mast Cells. Cancer Research. In press. 26. BAtTMBERGER, J. P., SUNTZEFP, V., and COWDRY, E. V. 10. Oineo, T. Pathologisch-histologische Studien liber die Methods for the Separation of Epidermis from Dermis and elastischen Fasern der menschlichen Haut. Arch. f. Der some Physiologic and Chemical Properties of Isolated mat. u. Syph., 149:486â€”492, 1925. Epidermis. J. Nat. Cancer Inst., 2:413â€”423, 1942. 11. ULESCO-STROOANOVA,K.tYber die Bedeutung des Bind 27. LOWRY, 0. H., GILLIGAN,D. R., and KATERSKY,E. M. gewebes bei der Krebsentwicklung. Ztschr. f. Krebsforsch, The Determination of Collagen and Elastin in Tissue with 25:394â€”406, 1927. Results Obtained in Various Normal Tissue from Different 12. BimucH, R. flber die Beteiligung des Bindegewebes bei der Species. J. Biol. Chem., 139:795â€”804, 1941. experimentaller Krebsbildung. Virchow's Arch. f. path. 28. STJNTZEFF,V., and CARRIJYHERS,C. The Water Content in Anat., 239:1â€”19,1922. the Epidermis of Mice Undergoing Carcinogenesis by 13. JALOWY, B. Kollagen, Elastin und Reticulln der Haut. Methylcholanthrene. Cancer Research, 6:574â€”577, 1946. Ztschr. f. Zellforsch. u. mikr. Anat., 27:667â€”690, 1937. 29. Evxns, R., COWDRY, E. V., and NIELSON, P. E. Ageing of 14. DICK, J. C. Observation on the Elastic Tissue of the Skin Human Skin. Anat. Rec., 86: 545â€”565,1948. with a Note on the Reticular Layer at the Junction of the SO. UNNA, P. G. Uberythema Acneforme. Internat. Selt. Dermis and Epidermis. J. Anat., 81:201â€”211, 1947. Hautkrankheiten. Hamburg-Leipzig: Leopold Voss, 1891. 15. SCHIEFERDECKER,P. t)ber das Auftreten der elastischen 81. MORRIONE, T. C. Factors Influencing Collagen Content in Fasern in der Tierreihe, Uber das Verhalten derselben in Experimental Cirrhosis. Am. J. Path., 25: 273â€”285,1949. der Wangenhaut bei verschiedenen Menschenrassen mid 82. HAm, G., and MCDONALD, F. The production of Collagen tiber Bindegewebe und Sprache. Arch. f. mikr. Anat., 95: In Vitro under Variable Experimental Conditions. Am. J. 134â€”185,1921. Path., 16:525â€”548, 1940. 16. SCHIEFERDECKER,P.Der histologische mid mikroscopisch 33. COWDRY,E. V. A Textbook of Histology. pp. 268â€”270. topographische Bau der Wangenhaut des Menschen. Arch. Philadelphia: Lea & Febiger, 1946. f. Anat. u. Entwickelungsgesch., 191â€”224,1913. 34. PEARCE, R. H., and WATSON, E. M. The MucopÃ³lysac 17. VoHwnex1r.z@,K.H. flber die Alterserscheinungen des charides of Human Skin. Canad. J. Research, 27: 43â€”57, Hautbindegewebes und liber die sogenannte â€œElastica 1949.

Downloaded from cancerres.aacrjournals.org on September 30, 2021. © 1949 American Association for Cancer Research. Morphological and Chemical Investigation of Dermal Elastic, and Collagenic Tissue During Epidermal Carcinogenesis

Chung K. Ma

Cancer Res 1949;9:481-487.

Updated version Access the most recent version of this article at: http://cancerres.aacrjournals.org/content/9/8/481

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/9/8/481. Click on "Request Permissions" which will take you to the Copyright Clearance Center's (CCC) Rightslink site.