a+a.a+s. gibson island research conference on cancer, 1946 · a+a.a+s. gibson island research...

A+A.A+S. Gibson Island Research Conference on Cancer, 1946

Abstracts of Papers Presented On August 12 to 16, 1946, sessions of the summer meet- Pituitary tz~nzors.-AInlost all mice of the C57 strain,

iitg of the Sectiori on Chemistry of the Amierican Asso- but very few mice of other strains, acquirje chromophabe ciation for the Advancement of Slcience, at Gibson Island, adenonlais and/or extellsive hypertrophites (12 to 260 mgnl., Maryland, were devoted to a research cotlierence on whereas the mouse's pituitary glalld normally weighs cancer. Dr. Harold P- Rusd1 Proifessor of Oncology about 2 mgm.) 04 their pituitary glands when given estro- and Director of the Dlepartmenrt of Cancer Research of gells for long periods of time. Although the males are the university of isc cons in was Chairman, and Dr. more susceptible to the tumors than females bsth male allcl u s t i n Brues1 04 the Argollne National I~aboratory. femalte of the C57 strain transmitted the telldeacy f u r University of Cllicago was Vice-chairman. pituitary tumors to their esltrogen-trzat& hybrid youllp there was no single topic of the conbereoce many of the ((-57 x CBA strairl or C ~ H strain). I)apers dealt with 13rob1ems of carcinogenesis. Among the ob,tained by backcrossing the hybrids to the parent stocks Papers there were reWrts On the biological effects showed that the telldency for pituitary tumors among radioactive nitrogen mustards' Abstracts edrogen-treated is genetically transmitted, probably of the papers presented follow. by multiple factors. Pituitary tumors have not appeared

in the untreated mice of these st~rains. STEROID H O R M O N E S I N THE INDUCTION O F Androgens inhibit but do not entirlely prevent the devel-

CANCER." \V. U. GARDNER. (Depalrtment of opment of pituitary tumors in estrogen-treated mice. Anatomy, Yale Un~iversity School of Medicine, New Several pituitary tumors have been transplanted. They Havcn 11, Cutlnecticut.) have grown only in related and estrogen-treated I~osts.

Estrogens are the only lwrmBlly occurring steroid her- The in'cidence of grafts that grow is 'ow, growth is sloW

manes, if not the only known chemically identified sub- and delayed and no transplants could be carried for over stallces, produced within the body, the of 3 generations. The original tumors appear to colltinue results in cancer under specific con,ditions. B~~~~~~ non- growth after discontinuanlce ai estrogen treatment because

steroid, synthetic est,rogen,s have carcinogenic effects sim- mice removed 138 days after cessation of injection still ilar to the norlnally occurring estrogenls, the neoplasia- had large tumors; as large as thos~e of mice continuously inciting action is associated with the biological responses treated.

rather than with an independent, chemical strut- Lynpkoid hcrriors.-Among 7 strains of mice with low

ture. Therefore the collective term, estrogen, is used in i"cidellces of ' ~ m p h o ~ d tumors in control animals, the

the following abstract altllough nine different estrogenic estrogen-treated mice of 3 strains showed an augmellted

chemicals have beell used in most of the incidence of such tunlo's (15 Wr cent) ; of two strains a

enumerated below. slight increase (5 per cent) ; and of two strains no effect. In tumors of the pituitary gland, lymphoid tissue, Large doses of estrogens were more ' Y m ~ ~ o m o g e n ~ c in

testis, mamlnary gland and uterine cervix have appeared the susceptible strains than small dosles. The discontinu-

subsequent to estrogen-treatment, and in animals in which Of estrogell-treatment after weeks did 'lot reduce

th,ey would not have appeared unless given estrogens. the incidence of lymphoid tumors although the tumors did

wiIth few exceptions, however, inherited factors or influ- not appear for several months after cessation d treakment.

ences transmitted from parent or parents to are The leukemogenic effeat of estrogens was inhibited by the of great significance. Estrogen-treated mice one strain simultaneous adm~inistration of tes'tosterone.

give rise to of one or more tissues or glands but The hylbrid offspring of mice of 2 strains susceptible tc llot of others. this the estrogens are not estrogen-induced leukemia acquired a very high incidence

unique; the carcinogenic hydrocarbons also produce dif- of such tumors (4 to 45 Per cen't). Thie incidences of

ferent in nlice with different genetic back- leukemia were intermediate in hybrid offspring of mice oi grounds and differ in effectiveness in strains suvceptible to and not susceptible to estrogen-

It is not improbable that two human induced leukemia, and were equivocal when hybrids with differ from one than mice from one parenlts of two non-responsive strains were studied. All

strain differ from mice of another ,strain. That a lymphoid tumors were transplantaible into untreated and

rapable of eliciting neoplasia in one individual is inade- mice. quate in another is not improbable. Carcinogenic hyd~rocar~bons do not induce increased - incidences of leukemia in mice of thle 'strains used in which

estrogens proved leukemogenic. * The investigations of the author referred to hfere were Testis ~tmovs. . -~umors of ithe interstitial cells develop supported by grants from The Jane Coffin Childs Fund in the testes of estrogen-treated mice of the A and JK for Medical Research and The .4nna Fuller Fund. Sev- ,trains although tumors occur ra,rcly in untreated era1 of the experiments mentioneld (have been undertaken mice. The tumors metasItasize to thie lurnlhar and renal in colla~boration with other investi,oators in the laboratolry, nodes, and occasiolla~~y to the lungs. They are trans- 1larnely, Dr. T. F. Douffherlty. J. H. Dougherty, Dr. plantzble into estrogen-treated hosts of the same strain C. W. Hooker, Dr. C. A. Pfciffer and Dr. 1,. C. Strong. for many transfcr generations. I n this laboratory they

37

Research. on May 22, 2020. © 1947 American Association for Cancercancerres.aacrjournals.org Downloaded from

http://cancerres.aacrjournals.org/

38 Cancer Research

have not grown in untreated mice. Fragments of transplanted tumor tissue may remain dormant ior several months in untreated hosts and start growth when est,ro- gens are administered.

Both male and female mice of susceptible strains trans- mit the tendency for testicular tumors to their hybrid young. Three unt, reated hybrid male mice (A X C3H and AXC57) have also acquired tesficular tumors at advanced ages.

Mammary tumo:rs.--Estrogen-trea~ted male mice from genetically susceptible strains and with the maternally transmitted m a ,n~-nary tumor itmiter acquire mammary adenocarcinomas (Table I ) . The incidence is usually lower than among repeatedly bred females. Among the mice with genetic susceptibility but without the mammary tumor inciter the tumor; appear at advanced age, grow slowly, an.d many are of the squamous cell type. Estrogen-treated male and female mice of the tumor-sus- cepti.b;e C3H strain acquired few mammary tumors when large doses of testosterone propionate were administered simultaneously. Not on.ly was the incidence of mammary tumors reduced but localized overgrowths of the mammary glands seldom appeared and even mammary duct growth was 'subnormal.

TABLE I : INFLUENCE OF SEVERAL FACTORS IN MAMMARY CARCINOGENESIS

Average Percentage chromatin - mammary tumors

Strain o,i origin

CS7 ~ . CBA $ CBA 9 X C57 6, CC2, ~ X CBA 6 CC1 9 X CBA CC2 9 X C57 d~

of suscep- r x �9 tible Mammary "Forced

Hybrid strain,, tumor Estrogen- bred" gro,up (%) inciter treated controls

CC~ 50 ~ 0 43 CC, 50 + 59 85 CC_. 75 + 54 79 CC, 75 - - 14 30 GC~ 25 --I- 1 7

Uterine cervical tumors.--Carcinomas appear in the uterine cervices or uppe,r vaginas o~ estrogen-treated mice of all strains that have ,been studied, but have not been observed among untreated mice except for one strain. The tumors usually appear, as judged by the detection of small infiltrative epithelial lesions, in the lower cervical canal, external cervix or vaginal fornix. Many stages of d.evelopmen, t, rangi.ng f, rom small tumors to large, and m.etastasizing cancers have also been observed. Several untreated mice of one stock, which shows a high incidence of imperforate vagina, have acquired cervical carcinoma; estrogen treatment augments the in, c, idence in mice of this stock.

Simul,taneous administration of testosterone propionate and estrogen did not prevent uterine cervical tumors. The tumors, however, showed less ter~dency to show cornified cells or pearls.

Uterine leiomyosarcomas have been noted in the uteri of several very old mice but not among estrogen-treated ar~imal.s. Some overgrowth of uterine glands, with exten- sion to the serosa (adenomyosis), is commor~ly observed but primarily malignant growths of the uterus proper have not been found.

By proper hybridizaJtion it i's possible to combine transmitted tendencies so that any two or more types of tumors may appear in animals of the same group or even in the same animal within limitations of sex (Table I I I ) . Pitui- tary, lymphoid and mammary; or testi.cular, pituitary and mammary tumors have been found in 'single estrogen- treated hybrid mice. Other than for mammary tumors, there is rio evidence of any tendency for tumors that is transmitted specifically by mice of one sex or the ot~her

TABLE I I I : T H E RELATIVE TENDENCY FOR ESTROGEN- TREATED )/lICE OF SEVERAL STRAINS TO ACQUIRE TUMORS

Testicular Interstitial Uterine

Strain Mammary Lymphoid Ceil Pituitary Cervix In order to determine how the mammary tumor inciter C3H + + + + + + + - - + +

might get into the milk and hence ,be transmitted to the CBA + + + + + + - - + + + offspring, studies were made on the milk which revealed PM -- + + -- 4- + + formed elemems including macrophages. Counts of cells A + + + -- + + + - + + in the blood obtained from t~he tail showed that during C57 - - - + + + + active lactation .the white blood cells were almost 50 per C12I + + • -- -- + + cent below that of the non.-mwsing mouse (Table I I ) . J K - 4- + + - ? Lactating mice given large doses of estrogen had high white blood .cell counts on the 4th day o.f treatmer~t, at which time the weights of their nursing young began to decline.

TABLE II: TIIE AVERAGE TOTAL WHITE BLOOD CELL COUNTS OF LACTATING AND CONTROL MICE UNDER

DIFFERENT CONDITIONS OF LACTATION

Total number white blood cells No. of 27-28

Group mice First count 24 hours hours

1. Lactating control-suckling interrupted and notresumed 10 10,740(66)* 15,610(62') 16,185(62)

2. Lactating- suckling interrupted and resumed 18 7,480(65) 12,970 (72) 6,850(66)

3. Control-non- lactating females or males 13 13,63i3(76) 12,597(71) 12,177(68)'

( ) Percentage lymphocytes.

Testos:terone i.s antagonistic to tumorigenesis of the mammary, pituitary and lymphoid tissues but not to tumors of the uterine cervix, or probably, of the testicular interstitial cells. No evidence of changes in the adrenal glands have ,been noted that .could be associated with tumors M organs other than 'the pituitary gland.

T U M O R S IN I N T R A S P L E N I C O V A R I A N T R A N S - P L A N T S IN C A S T R A T E D MICE. M. H. LI and W. U. G A R D N E R . (Department of Anatomy, Yale University School of Medicine, New t-Iaven 11, Connecticut.)

It has been shown that many s.teroid gonadal ho.r- mones are inactivated when circulated through the hepatic portal system, and that the intrinsic production of gonado- trophins is increased subsequent to castration. This investigation demonstrated the role of the endogenous gona- dotrophic hormones of the pituitary gland in ovarian tumor formation by the tran.splantation of ovary into spleens of castrated mice.

Ir~bred mice of the Slrong A and C3H strair~s, and



A.A.A.S. Gibson Island Research Conference on Cancer, 1946 39

hybrid mice (AC3, ACS) were used. Six groups of experiments were set up: (a) Twenty-one castrated male mice with homotran, splantation of an ovary into the spleen; (b) 52 castrated female mice with autotransplan- tation of an ovary into the spleen; (c) 25 castrated female mice with sub.cutaneous auto.transplantation of an ovary in.to the right axillary region; (d) 12 intact male mice with st~beutan.eous homotr:ansplantation, of an ovary in,to the r ight axillary veg, ion, (e) cas'trated ma!e mice with subcutaneous ho.nmtransplantation of an ovary either into the right axi.llary region or into the abdominal region; and (f) 33 intact male mice wi.th homotrans.plan- tation or heterotransplantation of an ovary into the right testis. Castration and grafting were done when the mice were 1 to 3 months old, except in the group (f) in which a few mice were operated upo,n at 130 days of age. The donors of the ovaries for homotransplantation or hetero- transplantations were u'sually younger than the recipients.

Five granulosa-cell tumors, 2 probably tumors of the same type and 1 mixed tumor consisting of both granulosa and luteo.matous .cells were fou,nd among the 21 castrated male mice with intrasplenie ovarian transplants. Four luteomas and 7 mixed tumors were found ~morLg the 33 castrated female mice that showed no adhesion of the intrasplen.ic ovarian graft to the left uterine horn or to the adjacent peritoneum. Among the 19 castrated female mice with vascularized adhesions of the ovarian transplant 1 lu,teoma was observed in a mouse that had irregular estrous cycles during the latter part o-f the experimental period. Mice of the A strain were used in these 2 groups of experiments and most of the ovarian tumors de,~eloped in castrated mice carrying the intra- splenic ovarian transplants for more than 7 months. No. ovarian tumors were found in other sites of transplantation excet'A for one small granulosa cell tum,or-like growth observed in a subcutaneously transplanted ovary in a castrated female mouse. The mouse was the only one that was operated upo.n, during pregnancy.

T H E M E T A B O L I S M A N D C A R C I N O G E N I C I T Y OF p - D I M E T H Y L A M I N O A Z O B E N Z E N E A N D R E L A T E D C O M P O U N D S IN T H E RAT. J . A . M I L L E R and E. C. M I L L E R . (McArdle Memorial Laboratory, Medical School, University of Wisconsin, Madison 6, Wisconsin.)

These studies co.ver some of our efforts to define the carcinogen operative in carcinogenesis by the am~noazo dyes through investigations on their metabolism and the effect of structure on their activity.

Usii~g a sensitive quantitative method for the deter- ruination .of basic aminoazo dyes in tissue we have demonstrated that one of the initial pathways in the metabolism of p-dimethylam, inoazobenzer~e ( D A B ) is it, s reversible demethylation to p-monomethylaminoazoben.zene ( M A B ) followed by an irreversible demethylation to p-aminoazo.- benzel~ ( A B ) (3, 4). These basic aminoazo dyes were in.itially characterized by their absorption spectra and adsorption characteristics. ' W e have further characterized the dyes by reduction to their component amines which were then determined by an independent method described below. DAB and MAB have been found to be metabolically and .carcinogenically interchangeable (6). In aeimals fed these .compounds the liver is the only organ that contains appreciable amounts of all 3 dyes. The tumors arising in livers .containing the basic dyes contain similar amounts of these compounds. AB is

temporarily stored in the red 'blood cells and is also excreted in the bile. The same amount of AB appears in the blood after any o~ the 3 dyes are fed; this seems to indicate that the metabolism of DAB and MAB pro.- ceedls largely through a,n i n, iti~al d, emeth?clar to, AB. The amount of AB ir~ .the 51ood varies directly with the cormentration of dye in the diet. The dye levels in the tissues reach a maximum in about a week of feeding. When the dye is withheld from the diet the dye contents of the tissues drop about 50 to 90 per cent in 3 days and in a week very little is left. The basic aminoazo dyes found in the body at ar~y one time could account for about 8 per cent of the dye ingested within the previous day. None of the various diets known either to increase or decrease the rate of hepatic tumor formation with DAB have been found to alter greatly the levels of dye in the liver and blood.

A study of the over-all metabolism of DAB, particularly the dyes and their constituent amines that may be found in the urine has also been made. In general our observations ,on the over-all metabolism of DAB agree with those of Stevenson and her gr.ot~p (7) in that p-phenylene d, iamine and p-aminophenol appear to be quantitatively the majo.r final metabolites of the dye. The application of a sensitive and relatively specific method for the determinatior~, of six monophenyl amines in urine has discl.osed the presence of minor quantities of other metabolites. In addition a new class of am.inoazo dye,s appears ,t.o ,be present. The t~rine of a rat fed DAB contains much less basic aminoazo dye than can account for the red color given with acid. The urine becomes orange when made alkal.ine but the co.lo.r cannot be extracted with petroleum ether or benzene. These properties indicated that one or more ampheteric aminoazo dyes were present. No hydrolysis to simpler azo dyes has proven possible and a means has not been found for their removal from urine uncontaminated with mo1~ophenyl- amine. Instead, we have had to rely on an approximate means for their estimation. This consists in reducing the az.o dyes in the urine in the cold at an acid pH with Na._.S~O~ and then determining the free amines formed. This procedure determines only free amptmteric dyes and the free I~.rtions of conjugated dyes. It was necessary to release the conjugated monophenyl amines present in the urine for analysis by hot acid hydrolysis under reducing aonditions. The 3 basic amlnoazo dyes and their p ' -hydroxy derivatives, which are representati.ve amphoteric aminoazo dyes, were found to be reduced quantitatively to their constituent amines by refluxing in dilute HC1 in the presence of tin. Hen.ce the m,ormphenyl amines were determined %y the difference between the total amines found after reductive hydro.lys~is in acid and the amines found after redtzction in the cold.

Four aromatic amines could arise in the reduction of the basic aminoazo dyes, v/z., dimethyl-p-phenylene diamine ( D M P D ) , monomethyl-p-l~henylene diamine ( M M P D ) , p-phenyl.ene diamine ( P D ) , and aniline ( A ) . p-Aminophenol (p -AP) was already known as a metab- olite of DAB. After a method had been devised for the estimation of these amines, it became evident that still another aminophenol, o-aminophenol ( o - A P ) , was present in the urin.e. The m.e~hod eml~loye.d allows tbe separation and estimation of these 6 amines in microgram quantkies as colored complexes through an application of the B6nige.r reaction. Pr imary aromatic amines were found to react instantly in the cold at p.H 7 with excess s.odlum-3-naphthoquinone-4-slllfonate to form highly col-



40 Cancer Research

ored phenolic Schiff bases. Benzene will extract the Schiff bases formed witch D M P D , MMPD, o-AP, and A. The first two amines form blue and purple solutions respectively and the M M P D base may be selectively destroyed with acetic anhydride, o-AP and A form orange and yellow solutions respectively. Hence these four Schiff bases it~ the benzene extract can be separately determined by measurements of light absorption at differ- eI~t wave lengths before and after the addition of acetic anhydride. PD forms a h.ighly insoluble Sehiff base which can be quantitatively filtered on asbestos and then dissolved therefrom by dilute base to form an orange- red ,solution.. The extracted and filtered aqueous solution still contains the p -AP base, which may then. be extracted with amyl acetate in which it forms an orange solution. The interference by normal urine constituents is low and satisfactory recoveries of added amines were obtained.

Each of the 3 basic aminoazo dyes, the ,three p'-hydroxy aminoazo dyes, and the six monophenyl amines have been fed to rats and the urines examined for these compounds and their derivatives. When free amines were excreted the urines were collected through funnels coated with hydroquinone and caught in dilute HC1 in t,he presence of .either tir~ or hydroquinone and benzene. Approxi- mately 50 per cent of each of the azo dyes cou.ld be accounted for in the urine in terms of the azo dyes and amines found. While r major metabolites in each case were PD and p-AP, small amounts of MMPD, A, and o-AP were noted. Very small quantities of DMPD, close to the limit of the method, appeared to be present in the urines of rats fed the dimethyl- and monomethyl- aminoazo dyes b u t were not noted in ttae urines of rats fed either of the comgletely demethylated dyes. Basic azo dyes were found only in the urines of rats fed the basic dyes. Aml)tmteric azo dyes were found in the urines of rats fed' any of the 6 azo dyes. The probable compasition of the amphoteric azo dyes in the urine is indicated by the observation that from the urines of rats fed DAB and MAB the amines M M P D , PD, p-AP, and o-AP were formed after cold acid reduction. The urine of rats fed AB yielded PD, p-AP, and o-AP under such conditions. After reduct.ior~, the urine of rats fed the / /-hydroxy derivatives contained similar compounds except that no o-AP was noted. The am photeric aminoazo dyes appear to account for only a few per cent of the azo dye fed. No free monophenyl amines were found in the fres.h urines of rats fed any of the 6 azo dyes. When the urine of azo dye-fed rats was allowed to stand at room temperature without a preservative the amphoteric aminoazo .dyes disappeared and free PD and p-AP appeared. A different s'ituation was found when the monophenyl amines were fed since cor~sideraNe free amine was excreted in each case although the majority was in the co.1~jugated form. Seventy-three or more per cent of each of the diamlnes was recovered in the urine. AI- rhough most of the D M P D and' M M P D fed wa's: demethylated to PD appreciable amounts of D M P D and M M P D were found in the urine after feeding DMP'D. M M P D yielded only M M P D and P D and only PD was found in the urines of rats fed PD. When A was fed it was metabolized to p-AP although some o-AP and A were excreted. Each aminophenoI when, fed yielded only that aminophenol in, the urine.

The above data indicate that a considerable portion of the metabolism of DAB probably proceeds through the basic aminoazo dyes to. the amphoteric aminoazo dyes and thence to the mormphenyl amines. A minor pathway from

AB direct to A and PD probably exists also. There is as yet 11o evidence for deamination at any step, rearrange- ment of hydroazo compounds, or for the direct reduction of the methylated basic aminoazo dyes since the M M P D and the small amomlt of apparent D M P D in the urine could also come from the reduction of hydroxy aminoazo dyes. Preliminary data indicate that the over-all metabolisnl of DAB is not greatly al;tered by diets which significantly alter its carcinogeni.city.

Ou~r studies .on the .effect of structure on the carcino- genicity of the aminoazo dyes have been designed to define more clearly the structural features necessary for stroJzg carcinogenic activity and to test the activity of possible metabolites of DAB. Each compound was fed to at least 10 rats at a molar level equivalent to 0.06 per cent of DAB, the reference compound, for 4 to 8 months in a diet which allows a high incidence of tumors wi,th DAB. The compounds were rated on a rough scale of activity in which DAB is assigned a value of 6.. A.p- proximately 40 azo dyes and amines were tested and only a few of the azo dyes were found to possess strong activity. Alterations in the azo linkage, the r'ing subslituents, a1~d the su'bsl'itu.en~s of t,he ami.n.o grot~ps have been tested. When the azo lin,kage in DAB was replaced by ~ N = C H ~ , ~ C H = N ~ , or by ~ C O ~ N H ~ (benzamide of D M P D ) the activity was abolished. Whev~ the methyl groups in DAB were .rep,laced by 2 H ~ , t t and CH~CH_~--, or 2 CHiCHi--- compcmr~ds N zero activity resulted. On the other hand if the amino group carried H - - and C H a ~ or CHa'--- and C H a C H = ~ the activities were equal to that .of DAB. In con.tr.ast to the latter case the azo dye with C H a ~ and' t - IOCHaCHz~ on the amino group was inactive. Ir~activity was also produced when the amino grot~p substituents were C H a - - and pheny l~CH.~- - o,r H ~ and C I t O - - . Ring aub- stituents produced great variations =in acti%ty. When the ring hydrogens were substituted one at a time with methyl groups the activities of the compound's ranged from 0 to 10-12 (1, 5), in con:trust to the prediction of Kensler and his co-workers (2). Although - - N O 2 grou.ps in the o ' ~ and m ' ~ positions reduced the activity o.f DAB, the m'--NO_~ compound, like the m ' ~ CH3 dye, was more active than the corresponding o '~der iva t ive . None of the possible metabolites of DAB other than MAB :have shown any activity. The p'-hydroxy derivatives of DAB, MAB, and AB were ir~active. D M P D and A fed together were inacti~e ev, en at 3. X tlhe equi.valer~t to 0.06 per cent DAB. MMPD, PD, A, p-AP, and o-AP were 1.ik.ewise inactive, 2,4'-Diamino-5-Oimettaylamino- diphenyl, 3-dimethylaminocarbazole, and 4-hydroxyazoben- zene also proved to be inactive.

The conclusion drawn from the above studies is that the real carcinogen operative in carcinogenesis by DAB is either one or both of the two dyes, DAB and MAB, or some azo dye which is very close in its structure to these coml~m)ds.

Recently, we have found a tight combination, presumably chemical, between an unidentified aminoazo dye and a cellular oonstituent, probably protein, in the livers of rats fed DAB. T,he dye could not. be extracted from isolated crude liver protein by boiling organic solvents, bu.t it could be ex:tracted with ethyl ether-ethan,ol after the protein had been degraded with either hot alkali or trypsin. For an approximate estimation, of the "bound dye," tissue protein was digested with boiling alcoholic K O H and the digest was extracted with ethyl ether ; after removing the solvent the residue was dissolved in alco-



A.A.A.S. Gibson Island Research Con]erence on Cancer, 1946 41

holic HC1 and t'he color inter~si:ty was read at 520 mFz. In rats fed DAB the .b~.u'nd dye was found only ia. the liver and not in the blood, lungs, spleen,, heart., skeletal mu.scle, in, testine, or kidneys. The livers o.f rabbits, guinea pigs, and cottort rats .did not contain t~he bound (lye while too,use livers had only about a fourth a,s much as rat livers. The livers of rats fed DAB and MAI" appeared to. contain similar quantities of t.tae same bound dye(s) while the livers of rats fed AB had very little of this dye. These findings correlate well with publislaed data on the specificity of DAB in producing tumors of the liver, on the inability of other workers to induce tumors in rabbits or guinea pigs, the lesser suscepedbility of rn,ice as: comp,a~red with r~ts, and the extremely low car.- cinogenicity of AB. About a month was required for the level of bound dye to reach a maximum after feeding with DAB was initiated, but the bound dye content dropped approximately t,o zero within a few days after the rats were red dye-free diets. The tumors arising irt livers oontaining the bound dye contained none of this compound. The data suggest that the bou,nd dye may have an important place in carcinogenesis by the aminoazo dyes.

R E F E R E N C E S

1. GIESE, J. E., MILLER, J. A., and BAUMANN, C. A. The Carcenogenicity of m3-Methyl-p-Dimethyl- amir~oazober~z,erLe and of p-Mo.n, omethylamintoazo- benzene. Cromer Research, 5:337-340. 1945.

2. KENSLER, C. J., DZXTER, S. O., and RItOADS, C. P. The Inhibition of a Diphosphopyridine Nucleotide System by Split Products of Dimethylaminoazoben- zene. Cancer Research, 2:1-10. 1942.

3. MILLER, J. A., and BAUMANN, C . A . The Determina- tion of p-Dimethylaminoazobenzen,e, p-Mon~methyl- am.i,n.oazobe~zene, and p-Amin, oazobenzene i.ll Ti.ssu,:'. Cancer Research, 5:157-161. 1945.

4. ~{ILLER, J. A., MILLER, E. C., and BAUMANN, C. A. On the Methylation and Demethylation of Certain Carcinogenic Azo Dyes in the Rat.. Cancer Re- search, 5:162-168. 1945.

5. MILLER, J. A., and BAUMANN, C. A. The Ca~rcin(> gel~icit.y of Certain Azo Dyes Related to p-Di- methylaminoazobenzene. Cancer Research, 5:227- 234. 1945.

6. MILLER, E. C., a,nd BAUMANN, C .A. The Ca~reinlogen- icity of p-M,onomethylaminoazobenzene in V~,rions Diets and ~the Activity of this Dye Relative to p-Dimethylaminoazobenzene. Cancer Research. 6:289--o95. 1946.

7. STEVENSON, E. S., DOBRINER, K., an.d RItOAI)ES, C. P. The Metabolism of Dimethylaminoazobenzene (But- ter Yellow) in Rats. Cancer Research, 2:160- 167. 1942.

N U C L E A R D I F F E R E N T I A T I O N AND' T H E ORI- GIN O F T U M O R S . J A C K S C H U L T Z . (The Institute for Carmer Research, The Lankenau Hos- pital, Philadelphia 30, Pennsylvania.)

This is :a review in, which the ,basis, for a concept of the differentiation of cell nuclei during development is discussed. I.~ is shown that a new theory of the origin of tumors, based o.n~ this concept, can be developed withou+ any additional ad koc assumptions.

Exa, mples of the correlative differentiation of the nucleus and cytopla'sm are given under four categories. (a) Lo.sses of parts of chromosomes during .early development

(Ascaris). (b) Increase in chromosome n.umber wi,thout nuclear division,.: "endomitosis." (c) Differential repro- duction of gmles, during endomitosis. (d) Changes in state of chromosomes; increase in nucleo.protein content in py.cr~o.ti.c nuclei, i-elatively hig:h protein c.onten:t of "dit- [use" nuclei. This last category ot ntue,ear differen, tiation involves visible changes i1~ the chrona.osomes, h.en.cc in the ntlmectlate vicinity ot the genes, and possibly within the genes themselves. The differen.ccs between the same ch,romos,onm regio.n in the two types of lmclei discussed are analogous to those found between the heterochromatic regions (rich in desoxyribosenucleic acid) and tSe eu.chro- matte r.cgion=s (relatively poor in this substance) within a single nucleus. The analysis of the genetic function ot the two types of chromosome region provides a ,basis for a working hypothesis regarding nuclear differentiation. In particular, the combination of cytochemical and genetical studies in Drosophila gives evidence for the connection of the heter.ochromatic region, s with t~e overall nucleoprotein metabolism of the cell: ribosenucleoprotein in the cyto- presto, desoxyribose in the chromosome. The euchromatic regions seem by and large, in this at pre'sent obviously oversimplified picture, to be concerned with the diversified syntheses that are diagnostic of different genes. In Drosophila there are m a w cases in w,hich euc:hromatic regions placed Imxt to heterochromatin change their prop- t rues ; individuals carrying such chromosome rearrange- ments show somatic "mutation" for the genes in the abnormally placed euchromatin. These changes are effec- tively losses of the specific activities of the .genes, and are correlated cytologically with the assumption of the properties of heterochromatin by the regions, of ,ohromosome involved, in the differentiation o,f the nuclei ~ the different tissues. There is thus a basis f.or "the postulate that the changes in "pycl~os, is" of r diff.eren~t types of nuclei arc reflections of changes in gene activity, and in extreme cases, of the genes themselves.

The exten~sion of this postulate leads to ethe point of view that during develop.m, ent a process of genic "adaptation" takes place. This "adaptationT' occurs concurrent with the establishment of a senti-autonomous system of elements in the cytoplasm, associated with cytoplasmic ribose-nucleoproteins. The genes whose specific activiti(s are essential for the maintenance of the cytoplasmic system differ for the differene~ .cell types, and probably relatively few remain in the "active" (euchromatic) state in any one nucleus. The much larger remaining num,ber o~ genes is adapted to a s.tate of activity compa, raJble to tha~ of the heterochromatic regions., a state of generalized function concerned with gene reproductior~ and the mitotic process. The further the process of nuclear differentiation has pro- ceeded, ~he more these adaptations are irreversible; and in the germ cell all genes are in the "adaptable" 'state almost by defil~ition. It should be stated thaq: this. di.scussion i~ in.tended especially for the last of ~he fo~u,r ca'.tegories of nuclear differentiati.on; for the other types, where differ- e.ntiali,on i,s, obviou,s,, ,it:s: con, sequen,ces for the d~i~scu~s,sion to follow are t'he same in principle.

On the assumption that an irreversible adaptation of genes for generalized and. for specific function occurs during devel.opment, the nuclei of differentiated .cells .have lost an adap'mbility present in. their em.bryo.ni.c for'bears. O~ the total number of gene's present, the majori ty are conceived to be adapted for the generalized Sunction; but since in the ordinary life o{ a d'ifferentiated cell, the stimuli for mitosis are ei't,her rare or highly specific and mediated by an already functio.n..ing cytoplasmic system.



42 Cancer Research

th.e reactions in which they participate are conceived rarely to occur. Shored, how, ever, the estabnsned cytoplasmic system be broken aown, and the cell not me, a new system must be regenerated, l l n s regeneration ot the cytoplasmic system takes place un4er the mtluence ot me 6,rrerent~ated nu.cleus, wma me preponderance ot genes adapted for gen- ~ranzed tuncuon ("heterochromatm-adap.ed"). Such genes are maxxma,ly functional in embryonic tissue; consequently the type ot cytoplasmic system regenerated wii1 resemble mat ot the embryonic ceil. Mitosxs wi,1 occur; but since the nucleus is a differentia'ted one, without the "'adapta~ble" genes oi an embryor~ic cell, the daughter cells, instead of being responsive .to "organizing" influences about tlmm, wall continue to produce the embryonic type of cytoplasm. Thus a cell type, committed under most cir- cumstances to mitosis, and relatively unresponsive to the ~m,cls oi d~tl.erentiation about it, will arise. Its characteristics are those of the malignant cell.

v;omparison of this hypothetical picture with the actual data strengthens a belief in i,ts val.idity. (a) Th,e multi- plicity of stimuli evocative of tum,~s calls for an effect on some constituent of the cell. Recen, t cytological work (see especially Opie) shows a stage in tumor development to be the breakdown of the nucl,eoprotein granules of the cytoplasm. Viruses, on this view, act by the cytoplasmic oamage they cause, and propagate in the tumor cell at the expense ot its embryonic cytoplasm. The precancerous lesions have the characteristics of embryonic tissue. (b) T h e n u c l e i of t u m o r cel ls r e scuab le m o s t the m o r e d i f -

f,erentiated of the nuclei in the tissue of origin (Biesele) ; their high nucleop.rot.ein, content (Stowell) indicates pos- s,ible direct eviden,ce for heterochromatic adaptation. (c) The direct chemical analyses of the constitution, the enzymatic activity, and the metabolism of tumor cells accord more .closely with those of embryonic cells than with any other group. It seems likely, and in some of these cases it is a l r e a d y c l ea r , t h a t th.ese t n e a s u r e n ~ e n t s re f lec t s,imi-

larities in the cytoplasm, s. (d) Other groups of data ( a n t i g e n i c p r o p e r t i e s , b e h a v i o r o n transplan.tation, relation of malignancy to tissue type of origin, differen.tiation i:n the tumor itself), ins4ar as available, are consistent with the view. Obviously many kinds of experiment are neces- sa,ry, to test the basic assumpti,ons, and to make precise the properties of the tumor cell iz~ usable terms. The working hypothesis is tenable, that the malignant cell is the result of the combination of an, "old" nucleus with "young" cytoplasm.

C A R C I N O G E N E S I S AND T H E M E C H A N I S M OF G E N E ACTION. S. S P I E G E L M A N . (Depart- ment of Bacteriology, Was,hington University School of Medicine, St. Louis 10, Missouri.)

The basic problem of ca, rcinogenesi's involves explaining the appearance of a sudden heritable change :in somatic cells which implies l~he involvement of the hereditary mechanism at some stage in the process. On this assumption i,t becomes imperative to examine the question f, rom the point of view of what is known abou~ the mechanism of gene action. Here one must emphasize the distinction between two quite different aspects of the gene problem. One involves simply their transmission from one generation to the next and t.he other the mechanism whereby genes effect their control over c.ellula.r phys.i.ology. It is with the latter problem that the pres,en,t paper is primarily concerned.

Recent work employing micro-organisms as genetic ma-

terials has led to some ratt~er well defined concepts on the nature of gene action. The basic assumption underlying much of this work is that since the enzymes of a cell deternlme its physiological cagabilities, geues must control tile phenotype of a cell by virtue of their ability to control enzymatic consti,tution. Adaptive enzyme formation and its inheritance in yeast was studied in an attempt to learn more about the nature and extent of this cor~trol.

We may summarize the important findings and co.nclu- sions relating to genes and enzyme formation by the following statements.

1. Usually, the transmission of characteristic enzymes and the products of their activities follows the classical Mendelian laws derived from the assumption that the controlling units are self-duplicating entities, get~es, located on chromosomes in the nucleus.

2. The existence of a particular gene in the nucleus of a cell does not guarantee that the correspondir~g enzyme will be found in the cytoplasm, as evidenced by such phenomena as cellular differemiation and enzymatic adaptation. Genes, therefore, have as their primary function the indefinite .retention for the cell of the potentiality for enzyme formation.

3. The actual formation of an enzyme in the cytopla0sm is mediated directly by a cytoplasmic unit (plasmagene) which possesses the capacity for self-duplication in the presence or absence of the corresponding gene.

4. T h e p r e s e n c e of the h o m o l o g o u s s u b s t r a t e s a c c e n -

t u a t e s the capacity of these self-duplicating plasmagenes to produce enzyme.

5. Competitive interactions exist anmngst the cyto.pla, s- mic enzyme-forming units.

6. Nucleo.proteins are involved in the synthesis of enzymes.

T h e s e f i n d i n g s a n d c o n c l u s i o n s h a v e b e e n syn, t h e s i z e d

into a theory of gene action whick assun~es that genes continually produce at various rates more or less complete replicas of themselves which enter the cytoplasm. These replicas or plasmagenes are presumed to be nucleo.protein in nature in view of ,their origin and to possess to varying, degrees the capacity of self-duplication. Their presence in the cytoplasm controls the types and am, ounts of-proteins and enzymes synthesized. These plas.magenes, like all se~f-duplicating entities, would compete with each other for protein and .energy, and the outcome of su, ch competitive interactior~s would then determine the enzymatic con- sti4ution of the cytoplasm. Inherent in. this con.cept i.s the possibility of changing the ultimate result of this competi- tion by va,rying the eoiKlitions (e.g., substrates available) under which it takes place. The various reaetior~s and the r61e of substrate in the process are detailed in Fig. 1 in terms of one germ G~ and its "corresponding enzyme E~.

All the double arrows denote self-duplicating reactions. Gene G~ continually produces its plasmagenes (Ply) at a ra,te denoted by k. The plasmagene b.y its very nature must possess het.erocatalytic potentialities, i.e., in addition to being autosyntheti.c it must po.sses,s the capacity of catalyzing the synthesis of un, its (enzymes) other than itself. Consequently, once in the cytoplasm:, several things may happen to plasmagene (PI~). If it is sucoes'sful in obtaining the proper material (M) in the cytoplasm it will duplicate itself. It may, on the other hand, combine wi,th precursor protein (Pr ) and convert it to E,, resulting in the formation, of the PI~E~ complex.

Since little enzyme is found experimentally in the ab- serme of substrate, one must as~sume that this complex is




highly unstable and quickly breaks n,p into i.ts two com- ponents. A plasmagene on.co formed cannot of course exist indefinitely, particularly in a po,pulaticn of other such units activ.ely competing for the mate,rial of which it is composed. The reaction leading from Pl~ to I P (inac- t;ve protein) in Fig. 1 descri'bes this fact. Enzyme may

CYTOPLASM NUCLEU~X~ L P,,,, - - " ' PI', ~ ~ 2PI=

§

+S=

E,S~ +Pr+M+S~ ~Pii El S,

FIG. 1 . - - D i a g r a m of t h e o r y of gene a n d p l a s m a gone ac t ion . G l ~ g e n e ; P l l ~ p l a s m a g e n e ; P r z e n z y m e p recu r so r ; E l ~ - e n z y m e ; S l ~ s u b s l . r a t e ; a n d M z c y t o p l a s m i c m a t e r i a l used in d u p l i c a t i o n of p lasmagene . T h e c o n s t a n t (k) is a react.ion veloci ty c o n s t a n t m e a s - ur ing ra t e of p r o d u c t i o n of p l a smagenes f ro in genes. All doub le arr( ,ws i~ld~ca.e a se, f -du l : l i ca t ing pro. tess ; bro,ken-lme a r r o w s d e n o t e decay to i n a c t i v e p r o t e i n ( I P ) of ~.he e l emen t f rom which t h e ~(rrow poinLs. W h e r e f o r w a r d a n d b a c k w a r d r eac t ions a re deno ted , "rite longer of ~he two a r rows ind ica t e ~he m a j o r t e n d e n c y of the reac t ion .

aiso break down to inactive p.rotein as i:s indicated by lo~s ol activity or removal of su,bstrate. By I P or inactive protein we mean m,ereG tha~t the plasmage~e has broken ~own to a protein unit which .has lost the capacities for self-duplication and enzyme formatioil, and in tile case of enzyme, into a protein whi.ch no longer possesses enzyme activity.

t 'hus far we have described the reactions which tak:~ place in the a,bsen, ce of su~bstrate, i t is clear that little enzyme would be found in the cell, unless the rate of Pl~ p,roductioll by G~ were extremely hig-,h or tt~e stabili~ty of the enzymes or enzyme-plasmagene complex wery great. Neither condition is apparently satisfied in the cases of the enzym.es reported on here. W.hen substrate S~ is added, however, it will combine with E~ arid two things m,ay result. It is well kn, own that the addition of substrate to enzyme stabilizes it against inactivation, Hel~ce the presence o,f substrate would decrease the rate at which wee E~ is cm~verted to inactive protein. More critical however is the po.s,si.bility that S would combine with E~ while t~he latter i's still un, ited with Ph, thus resulting in a PI~E~S~ complex. The substra'te would now not only stabilize the enzyme, but could als<) stabilize the unstable plasmagel~e-enzynle ( P h E , ) .combination. Such sta.biliza- tions of unstable complexes :b.y the addition o,f a third component arc quite common, in organic chemistry.

Inherent in the very definition, of a self-duplicating en- tity is the concept that should such a unit undergo a modification at any given, moment, all subsequent replicas would bear this modifi.cation. Thus wher~ Pll exists alon.e it duplicates only Ply. However , when substr:ate i's added and the stab.le P1,E,S, combi.natio~ results this is now duplicat.ed, which is indicated in the diagram ,by the double arrow. This 'scheme p,rovides therefore a concrete mechan-

is,m whereby su, b.strate can modify competitive interactions between p.lasmagenes. Wha t i's esser~tially accomplished b.y st~bstrate is the creation of a new self-d,uplicating unit which du,plicates not o.nly the plasmagene but also the enzyme corresponding to the st~bstrate added.

F r o m a general point of view, the un,ique feature of the theory outlined above of gone action is that wkile st~p- plying a l.ink between thee gone and the enzyme, it at the same time predi.cts that cells with identical genomes need not possess identical enzymatic constitutions. Whe the r a part icular enzyme will ,be t ransmit ted f rom one cell generation to the next in a Mendelian fashion, will thus d~epen, d on the relative r~te.s cff d,upl, i,c~tio,l~ of the con.trolling cytopl:a.sm,~c units as compared with their ra,t~e of p,ro- duction xrom the genome. If the latter is quantitatively determining, Mendelian inheritance will be oh.served. If the former is determining, the Mendelian picture will be obscured to varying degrees, depending upon the self- duplicating capacity of the plasmagenes. I t is clear also how subst.rate could so il~te'nsify cytoplasmic ir~heritan,ce of a part icular enzyme as to completely obscure the segre- gation of the correspondi~g gene.

As a tentative working hypothesis this theory has the advantage of furnishing a unified point of view from which such diverse and apparently contradictory phenomena as classical Mendelian genetics, cytoplasmic inh.eri- tahoe, cellular differentiation, and enzyma,tic adaptat ion may be analyzed. Its impl.ications for the cancer problem are evident. F r o m the viewpoint of the plasmagene mechanism of gone action, heritable changes su,ch as c~r- cirLoger~esis need not nec:ess~rily be referred ba~ck to the ger~es. The pla'sm~gene by being self-dupli,cating prov,ides us with another level at which a nmtat ion can 'cake place and be subsequently t ransmit ted via the cytoplasm from on.e cell generat ion to the next. This concept may also furnish a basis f.or reoonciling the somatic muta t ion ver- su, s the virus theories of cancer, sin, ce, if the muta t ion does occur at the plasmagene level, tim diffe,renoe between these two opposing views becomes tenuous indeed.

P A T H S O F G E N E A C T I O N IN M A M M A R Y T U M O R D E V E L O P M E N T IN MICE. W . E . H E S T O N . (Nati,o,i~al Cm~cer Resea, r.ch Institute, Na- tiona, l Insti,tute of Health, Uni ted States Publ,ic Heal th Servic'e, Bet.hes.da 14, Marylal~d.)

Now that genetic differences in the development o.f mammary tu,mors in mice have .been demonstrated, the present problem is that of determining the physiologic mechanisms through which these variations in genie complex become manifest. F r o m recent w,ork on gene chemistry such genic differences ,ca.n. be expected t.o affect enzymatic differences controll ing specific meta,bolic re.ac- tions related to the subsequent development of the tumor.

At present three paths through which differences in genie action affect the probability that a ma mma ry tumor will appear have been identified. There is evidence suggest ing that genic differences can influence the response of the m a m m a r y .tissue cell to the hormonal st imulation and to the milk .agen,t stimulation. In either case the gone action must occur in the m a m m a r y tissue cell. There is evidence that genic differences can influence the deve!opment of mamma:ry tm~ors th rough the hormonal mechanism, possibly in controll ing the product ion of the ho,rmonal stimurlation. In this path the gene action can be expected to be in the phys,i.ology of the endocrine system. Genie differen,ces can influence the propagat ion and



44 Cancer Research

transmission o.f the milk agent. Since it has not been dem~3nstrated that the milk agent is limited to any particular ~tissue, one cannot predict just where the gene action occurs in this case. Through the effect on the transmission of t,he agent, gene actio~a in one female may influence the probability that a tumor will appear in her daughter or in a foster-nursed female.

Possible bearings that these findings in the mouse may have on the investigation, s of breast cancer in the lauman being were discussed.

T H E I N D U C T I O N OF G E R M I N A L M U T A T I O N S BY A C A R C I N O G E N I C C H E M I C A L ( M E T H - Y L C I - I O L A N T H R E N E ) . L E O N E L L C. STRONG. (Department of Anatomy, Yale University School of Medi,cine, New Raven 11, Connect,icu,t.)

Fifty-two germinal mutations affecting intensity and dist,ri~bulion of pigment and spo't,ting patterns of the hair of mice have been obtained in the descendants of mice, whose ancestry has beer~ ir~jected with methylcholanthrene over a number of generations. None of these mutants have occurred in mice of the untreated control series. In the series treated earlier (4 generations), no germinal mu.tations were obtained and the mice, as a rule, died early following the injection of methylcholantl'rrene, with the progressive growth of local tumors appearing at the site of injection. However, selection toward resistance to tumors induced by methylcholanechrene has resulted in the suppression of local tumors, as well as other types of internal tumors whose latent periods are considerably longer than the ones obtained with local tumors at the site of injection. When ethe effect of selection toward resistance to tumors had been great enough to permit the appearance of abdomilaal tumors (with long latent periods) such as primary carcinoma of the liver, adenocarcinoma o.f the gastric mucosa, leiomyosarcoma of the uterus, and granulosa cell ,carcinoma of .~he ovary, the descendants of these selected mice began to show many germinal nmtations.

At present, it is intended to discuss only three groups of these mutants. (A) There has apparently appeared a series of alleles at the brown-,black locus. Not all genetic tests have been made on all these mutants but apparently each type ,breeds true and segregates following an out- cross. Any mouse may have one or two of these genes but no more. A cross between two o~ these new mutants gives .one or the other of the new mutants, not the wild type. This series of alleles (new mutatioils) classified according to intensity of pigmerttation by visual inspec- tion is (1) black (es~a~blished d, o m i n a n t = B B ) , (2) grap,l-ti.'te, (3) dark gunme'oal, (4) gunmetal, (5) da~-k bronze, (6) bronze, (7) darker sepia, (8) dark sepia, (9) sepia, (10) chestnut, (11) brown (established reces- s i v e - - b b ) , (12) light brown, (13) lighter brown., (14) velvet, and (15) buff. (B) Another series of mutations involving pigmentation produces a lighter ventral area than occurs on the dorsum. This series has occurred in mice whose ancestry had shown nothing fo.r many genera- ti.ons but non-agouti, where there is no differen~dal pigmentation on the d, orsum and ventrum. Thus there has been obtained a series of mutants classified according to pigmentation on the ven,trum of (1) bla,ck or non-agouti. (2) light black, (3) rusty, (4) red, (5) chestnut, (6) tan, (7) sandy, (8) derris and (9) buff. (C) The third series of new mutants involves the formation of different colored spots on either the forehead or ventrum.

The colors of spots have been, (1) white, (2) gray, and (3) yellow.

In addition to these clear-cut germinal mulations there has been obtained many somatic mosaics between any two of the above color mutants; such as between buff and brown, chestnu'c and brown and ru.sty and graphite, etc. Many of the new color mutations are permanent from the growth of the first hair until old age, while others shift more and more toward the recessive condition with advancing age--rem, iniscent of the shift of black toward brown with advancing age particularly when a black mouse is heterozygous for brown.

In several cases, the appearance of a germinal mutation affecting color of the hair has also suddenly brought aJbout physiological changes that alter suscepedbility or resistance to tumors induced by methylcholanthrene.

It is concluded that there must be some correlation between the induction, of germinal mutations by methylc'ho- lanthrene affecting .body characteristics and the unknown changes in sc, matic tissue which leads to cancer induced by the same carcinogen.

T H E A C T I O N OF D I A M I D I N E S A N D R E L A T E D C O M P O U N D S ON N U C L E O P R O T E I N S . M . J . KOPAC. (Department of Biology, Washington Square College of Arts and Sciences, New York University, New York 3, N. Y.)

Cellular structures, in, particular those cor~ta'~n!ng nucleoproteins, can be modified .by a number of chemical agents. Various oncotytic agents (bacteriM polys,accharides, qua- ternary ammonium halides, diphenylethylamines, stilbamidine, and x-radiation) may initiate abnormalities ranging from minor disturbances in nuclear configurations to the complete destruction of the n.ucleus.

The changes induced in cells by the action of nitrogen mustards frequently cannot be recognized excep.t ,by care- ful cytological study or by gen, etic analysis. These sub- stances, at sub-lethal con centratiorrs, may irmrease ~the rate of incidence, in Drosophila, of sex-linked lethals--an effect previously obtained only by the applica~tion of high frequency radiation.

The so-called mitotic pois,ons cause still different effects. Colchicine, in most instances, produces a metaphase block. Some of the dip,he:mthylamine derivatives a, re also mitotic poisons of the colchicine type. Several nitroph,enots induce a prophase block in cleaving sea-ur,chin eggs. Nitro- gen mustards have been reported to distut~b the interkinetic phase in a variety of cells.

The virus nucleoproteins are also susceptible to chemical agents, especially to those compounds .capable of denatur- ating proteins. Accordingly, detergents, pyridines, pico- lines, guanidines, and. urea have ,been frequently employed ei~ther as destroyers of virus activity, or as 1ools for the elucidation of virus structure.

The experiments described in this report were designed for two purposes: (1) to measure the effect of selected organic compounds on the structure of various types of nucleoproteins, and (2) to develop and tesec methods that might be employed later for determining the action of such compounds on living cells.

E.rperi.mental.--Changes in. protein s.true*nre can be followed by measuring denatu,ration rates. Urffortu,nately, classical methods were no,t sufficiently sensitive to measure the denaturation of proteins induced ,by compeunds at low concentrations. A procedure was developed, however, so that protein molecules could be exposed to the simultaneous




action o.f imerffacial forces and one of the compounds maintained at physiologically effective concentratio.ns. In this way, the denatu,rating effect of one ag.enCc could, be used to augment a similar action inid.ated .by another agent. For example, with these surface chemical me~'hods, it was previously shown that stil,ba,midine, at physiologic concentrations, enhanced the imeriacial denatu.ration o.f variou's pro- t eir~ molecules. Several other diamidines, however, depressed interracial denaturation,.

In general, three-dimen,sior~al protein molecules when exposed to appropriate surface o,r interracial fo.rces will unfold and.become two-<timens, ional platelets. Such changes in architecture r.equ{re the 'splitt.ing of several crkical side- chain, linkag, es. App'arenlly, some of these lin,kages can be split by surface forces alone, while others may he split orfly `by the action of chemical agents. When. both forces are in operation,, t'be effect can be demon.strated by m; increase i'n 'su,rface denaturation of the proteins tested.

On the other band, other compounds which rein.force side-chain linkages will indicate the effect by a decreased surface denaturation. Several compound,s, includling fatty acids and certain amino acid's, when added to protein solu- ti.ons can protect such proteins against heat and other den,atu.ra.~ing agents. The'se p,rotecti,n,g compounds also ap.precla,bly reduce the surface denaturation of protein molecules.

The nucl,eop,rot.eins (liver nueleoproteins, tohacco mosaic virus nucle.oproeein, p , rotamine~ nucleate complexes, and a cytoplasmic fraction, rich in. mitochondrla) were exposed to the simultaneou~s action of in,terfaclaI ferces (oil-water inlerfaces) aM one of the chemical agents (at con,centra- tion.s of 0.001 M', or Iower). The in:te,rfacial denaturation of the nucl.eoproteins was measured w{th the semi-auto- matic, drop-retraction apparatus. The meth,od .is funda- men'tally a m{cro-ada~em,tior~ of the procedure develol~ed by D,evaux fo,r measuring protein m,onolayers at oil-water h~.terfaces. The same ~roeedu,rcs and a~.p~aratu's may he used to study oil-cytoplasm interfaces. In this way, tlw acti,on of chemical agents on cytoplasmic proteins can he determined.

L~7~er nucleoprote~ns.--The action of the various com- n,oun,d's already investigated on liver nucl:eoproteins may be classlfied into 4 group,s:

I. Compounds that augment surface forces .by weaken- in~ c,erta,ln s,id, e-c,hain li,nkage,s in the n,ror mole,-ule. Increased interfa,cial denaturation (stilbamldine. methvl- bis-. ethyl-bls-, and tris,-2-ehIo,roethylami'n,e, ben,zylis,oth{ou- rea'~.

II. Comr)ounds that cou,ntera.ct su.rface forces by strengthening critical side cba.ln linkages'. De,creased in- torfacial denaturation (r~ror)amidine. n,entami.dine, nhenami- din,,. 1.2-di-/~-anis.vlethvl'amine. n,henethylpyridinlum bro-. mld~ bi.~-amldinomethyldihenzyl).

III. Coml~ou.nd's that neither augment nor eo.unteract s,rface fo,rces. Unch.an~ed interf, r-;al denatu,ration (1.2- d:,~h.onyl ethvlamine, benzvlimlnoethv~e~'b e.r ~.

IV. Comgound's that simultaneously au~me.nt and counteract surface fo.rce.s hy weabening some linkages and stren'gtheni'n~ others within the molecule. The net effec~ depends on the ~,r.edomlnan.ce of one action, over t~hat of the other. A h,ala,n,ce 5etwe,en the ~,vo w,o.uld p,r,o&,co v~ chan~e ;n interfadal denaturat;on. Under '~ueh condition o, vrour)s I I I and IV can. be d!iffet,entiated hy testln~ s~ch compounds in camhination with o+ilbnmidln,e (coIchicine. p,henethyl-4-n,icolon,lum bromide, 2-/)-dimethylaminostvryl- 4-methyl,pyridine-methobromlde, 2-/>dim, ethylamin,o,styryl. pvr{din.e-meth o,brom,ide, benzyIami.dine).

Tobacco rnosMc v~rus nucleoprotefn.--The surface dena-

tu~ration of molecul.es of the virus type requi,re.s: (1) dis- soci,ati.or~ of the large molecules into smaller molecules, and (21) unfolding of the smaller molecules at the interface. Interracial fo:rces are sufficiently large in` some in~stanc.es to. produce both dissociation and unfolding. Agents that only dissociate the large molecules, will er~hance interracial denaturation. If these agent's also faci, litate unfolding, in,terfacial denaturation would be enhanced to an even hig.her degree.

Stil`bamidilae, propam,idine, or peniamidine increased the interracial demitu'ration of the virus nu,cleop.roiein. The increased denaturation. ,by these compounds was, in all pro`bability, the resul't o.f di,ssociating the giant virus molecule. The experimer~tal data indicate that the suhsequent unfoIding of the small molecule!s was induced primarily by surface fo,vces. In no. i r~s*arme has propamidine or pen- tamidine enhanced in.terfacial unfolding of protein molecules to any appreciable extent.

On' the other hand, phenamidir~e or bis-amidinomeehyl- d'ibenzyl appeared to strengthen the viru, s molecule so that interfaci, al forces were unable to dissocia'te the large mole- cute. Acoordingly, interfacial denaturation of the vi, rus nucleopro,tein" was dept~es,sed 'by these two co,mpounds. Neither bis-amidinonmthyldihenzyl r~o.r phenamidine were able to antagonize the actior~ o.f stilbamidine.

Cytoplasmic proteins.--Stilbamidine enhanoed the interfacial denaturation of cytoplasm, ic proteins, ,buffered at p i t 7.5, while the o.ther diamidines d, epressed. Prop- amidine depre'ssed considerably more than. pbenamidine or bis-ami.dinom, ethyldibenzyl. Thus., the latter two diamidines produced similar effects with ,both virus nucl.eopro- teil~s and cy, t.oplasm, ic woteins.

Either stilhamidine or propamidinc were equally effective in enhancing the in.terfacial denaturation of cytoplasmic proteins, buffered at pH 7.9. Coldaici.ne depressed interracial dcnatut~ation, thus producing effects on cy,to- plasmi, c proteins s'imilar to those p,rcviously described fvr liver nucleoprotcins. Tris-2-chloroethylamine depressed interracial denaturation` of th,e cytoplasmic proteins, in comrast to its action on liver Imcleoproteir~s.

Combination of chemical agents with stilbamidine.--The st,viking d.enatu,rating action by stilbanaidine was not ob- tattled o,n adding t'h.is agent to liver n.ucleoprotein,s previously treated with propamidine, pen.tamidir~e, or p,henam;... dine. The s,tilbamidine effect was thereby an,tagonized and in proportion to the depressing action of the other diamidines tested alm~e, e.g. ,pentami, dine ~ propamidine )- phen, amidine. The interracial der~aturation of liver nucleoprotein treated with pbenethylpyridinium Br + stilham,idine was laearly identical to that measured with phenamidine + stilbamidine. These data indicate that 9benamidine and phenethylpyridinium Br have a similar action.

The interracial denaturation o~ liver nu, cleopro~eins in the presence of colchicine + sr was higher than in the presence of stil`bamidine alone. Colchicine, at the same concentration,, depressed interracial denatu,ration. i~henethyl-4-picolonium Br amagon, ized stilh.amidine sl.ightly, however, s.u,bsequel~t analyses of the denaturation curves suggest that this compo,und is similar to co lchicin.e.

Some linkages are weabened while others are strength- ened in the liver nucleoprotein molecules by either colchicine or pheneth2fl-4-pico,lon,ium Br. If stil,bamidine was added wk;h these corn,pounds, interracial denaturation was en,harmed since st'il,bamid'ine tends to weaken sid, e-ehain linkages. Phenethyl-4-picolonium Br is a stronger de- pressor than colchicine and accordingly this agent antag- onized stilbam, idin, e to a greater degree,



46 Cancer Research

Significance of chemical struct~res.--Relatively slight chang~es in structure of an agent may produce con, sider- ably different effects on the inte.rfacial denaturation of nu.cleoproteins. The striking differen~ces between phenethylpyridinium Br and phen, ethyl-4-pi.eolonium Br show that a single p-methyl group can modify the action of a compound on liv.er nucleoproteins.

All aromatic diami.dines, with the exception of stilbamidine, depressed int.erfacial 6enaturation of liver nucleoproteins. Liver nucleoprotein molecu!es were fortified against interracial forces by bis-amidinemethyld:ibenzyl and weakened by stilbamidin,e. When the two agents were present simultaneously, the weakening action o,f stilbamidine was almost entirely ane~agonized. The other diamidines "a!so antagon,ized the action, o.f stilhamidine aM in proportion to thei,r abil4ty to antagonize interracial forces.

Many oneNyti.c compounds iMuce morph.ologieal or fune- tion,al changes in tumor cells especially in those s~tructures that contain nucleoproteins. These agents also produced clear-cut .effects on the int.erfacial denaturation o~ liver nucleo.peotelns, virus nuc!eoprotein, protamine = nucleate complexes, or of cytoplasmic protein.s.

Any change in imerfacial denaturation implies some fundamental alteration in the structure of such nucleoproteins and these alterations may, in many instances, be of sufficient magnitude to affect the fun~:ti,onet activity of those cells in wh.i.ch corresoonding changes in nucleopm- t.ein structure might occur. T,he ability of certain compounds to bind side-chain linkages and at the same time weaken others may explain ,, in, part, the action of mitotic poison's.

Nucleoprot,eins may even be dissociated by a chemical agent, providirrg the protein, moiety is, at least partially denatured. In complex nu.cleoproteins, denaturation of the protein moiety is a prerequisite for ,the separation o.f nucleic acid from the protein, molecule.

Thus, n.ucleoprotein,s car~ ,be in~act~ivated in two ways: (1) the nucleic add may be irrev.ers~ibly d,issociated from the protein moiety, or (2) critical intramotecula.r structure of the nuc!.eoprotein may be blocked by strengthening side- chain linkages. In either event, such molecules must be- com, e les,s capa,ble of performing essential function,s.

There is 'surface chemical evidence to indicate that specific,ity of action might ,b,e fur"thor enhanced by employing a combination of agents, in which the ingividual properties are mutually modified. For examp!e, the combination of stilbamidin, e4-colchicine was different in~ its aclion on nucleop,roteins than that obtained by the individual action of either agent. S,imila~rly, the effects prc,du.~_ed by stilbamidine 4-phenethyl-4-rJ,ieolonium Br or by s,tilbamidine 4- d,iphenylethylam,ine differed str, ikingly from those produc.e.d by any on`e of the agents tested individually. Similar combina'don's of ch.em,ical agents should, therefore, be tested on living cells.

SOME C H E M I C A L C H A N G E S I N D U C E D BY M E T H Y L C H O L A N T H R E N E IN T H E TRANS- F O R M A T I O N OF M O U S E E P I D E R M I S TO S(31IAMOIJS CELL CARCINOMA. C. CAR- R U T H E R S and V. SUNTZEFF. (Research De- partment. The Barnard Free Skin and Cancer Hos- pital, St. Louis 3, and the Department of Anatomy, Washington University Medi.cal School, St. Louis 10, Missouri.)

The research program at this hospital on some of the chemical, histological, and physical changes induced in the epidermis of mice vndergoing carcinogenesis by methylcholanthrene has led to the accumulation of con-

siderable data on the chemical comoositicm of normal, methylcholanthrene-trea`ted (hyperpla.stic) epidermis, and transplantable squamous cell carcinoma of mice, and also some information on the rain.era! composition of human epidermis.

'For tile various analyses, the epidermis was separated from the dermis at 50 ~ C. on a warm plate by the procedure of Baumberger, Suntzeff, and Cowdry (1), or at room temperature by a recently devised method. Since only small amounts ,of epidermis were obtained from normal mice, it was necessary to devise microchemical met.hvds for the quantitative determination of the min- erals, vitamins, and lipids.

Using nucleoprotein phosphorus as a basis of reference, tI:e potassium, calcium, magnesium, i.ron, copper, and zir~c contents of normal epidermis, hyperplastic epidermis, and of a transplantable squamou's nell carcinoma were determined. Wh,ile s,ignificant de,creases in the calcium, iron, copper, and zinc nucleoprotein .phosphorus ratios were observed as a result of the application of 1 to 3 paintings of the carcinogen, the potassium, sodium, magnesium, and ascorbic acid ratios showed no perceptible change. Moreover, the decrease in the heavy metal and calcium ratios so manifestly eviden:t throughout hyper- plasia was even more spectacular with the onset o.f car- ci~mma. O~ the other hand, the ascorbic acid nucleoprotein phosI)horus ratio remained unchanged through oust the hyperplastic and carcinomatous stages, while the sodium, nmgnesium, and potassium ratios decreased slightly in the tissues of squamous cell carcinoma.

In this connection, analyses were also made of the mineral composition of human epidermis (for purposes of comparison) and it was discovered that decreases occurred in the calcium, copper, and zinc contents in, squamous cell carcinoma strikingly similar to those detected in the comparable stage of the mouse. These experiments also revealed that normal human and hyperplastic epidermis of the mouse anroe well with respect to their mineral composition, which is in, con{o,rmity width thei.r 'somewhat similar morphology.

Determination of the water content of the epidermis undera-oinr carcinogenesis established the fact that the amount of water increased from 60 per cent in the normal to an average value of 66 per c enl in the hyperplastle stage, and rose to nearly 82 per cent in the trat,_:plantable carcinoma.

Investigations on the epidermal content of choline. biotin, p-amin~benzo,ic acid, in,osir and pyri,dt~xine by Tatum and his group (4) u,sing mutant strdns of Ne,rospora demon'strated that only biotin wa,s sign,ifi- camly altered during ca.rcino,g, enesis. This vitamin, decreased to 64 per cent of the normal.

Wicks and Suntzeff (5, 6) also showed that the total lipid- and total cholesterol-protein-nitro,yen ratios were about 50 r~er cent less than normal in byperplastic epidermis. The decrease in the total li.pld was found to svnchronize with the disappearance of the sebaceous ,~lands. The relationshlp of the latter structures lo ear- cln.oeenesis has been investi~a'ed by Simpson and Cramer (3~. and is briefly discussed.

Experiments were also conducted with a view to elicit inforwation concernin,~ the influence of age on the resr~onse of carclno~en in relation to epidermal calcium and total lipid content. Youn~ and old mice of the Swiss and New Buffalo strains were used for the lyurpose. The resul's showed conclusively the absence .of any co rrela- tion between the &crease in the content of these sub- stances and the response to the carcinogen.




The activities of succinic dehydrogenase and cytochrome oxidase in epidermal carcinogenesis were followed by employing the procedu~re of Schneider and Potter (2). Our study showed that the activ;ty of succlnic dehydrogenase remained the same both in the normal untreated epidermis and in the hyperplastic epidermis, but increased significantly in the carcinoma. On the other hand, the activity of cytochrome oxidase in the late hyperplastic epidermis was twice that of the normal, but the value for its activity in the carcinoma was less than that of hyperplastic epidermis, but greater than that of the normal.

Finally, an attempt was made to correlate the chemical changes induced by the carcinogen with some of the morphological alterations.

R E F E R E N C E S .

1 BAUMBERGER, J. P., SUNTZEFF, V., and COWDRY. E . V . Methods for the Separation of Er~idermis from Dermis and, Som,e Phys~iologic and Chemical Properties of Isolated Epidermis. J. Nat. Cancer Inst., 2:413-423. 1942.

2. SCHNEIDER, W. C., and Porr~R, V. R. The Assay m Animal Tissues for Respiratory Enzymes II. Succinic Dehydrogenase and Cytochrome Oxidase. J. Biol. Chem., 149:217-227. 1943.

3. SimPsoN, W. L., and CRAMER, W. Sebaceous Glands and Experimental Skin Careinogenesi; in Mice. Cancer Research. 3:515-518. 19'43.

4. TATUM, E L., RITCHEY, M. G., COWDRY, E. V., and Wick:s, L. F. Vitamin Content of Mouse Epider- mis ~Du~ring Methylcholal~ttgrene Careinoaenesis. I. Biotin, Choline, Inositol, p-Aminobenzoic Acid. and Pyridoxine. J. Biol. Chem., 163:675-682. 1946.

5. WICKS, L. F., and SUNTZEFF, V. Reduction of Total Lipid-Protein-Nitrogen Ratio. of Mouse Epidermis by a Single Atyplication of 1V[ethylchoIanthrene. I. Nat. Cancer Inst,, 3:221-226. 1942.

6. WICKS, L. F., and SmVTZEFF, V. Chan~es in Epi- dermal Cholesterol During Methylcholanthrene Carcinogenesis in Mice. Cancer Research, 5:464- 468. 1945.

T H E S I G N I F I C A N C E OF I N D U S T R I A L CANCER IN T H E CANCER PROBLEM. W. C. H U E P E R . (Warner Institute for Therapeut, ic Research, New York, N. Y.)

In view of the known differences existing in the su.s- ceptibility and reactivity of va~rlou.s svecies to the numer- ou,s ,recognized physical and ch,emicall carcinogenic agents. indu'strial cancers posse's,s the great advan4:ag~e over ex!~eri- mental cancers in anlmMs in, that they permit the direct study of many of the most important and fundamental aspeets of cancer in the human organism. The considerable rise in +..he absohlte ~-m.ber and in different types of occupational cancers pa,ralIelin~ the enormous growth o, ~ modem, industrial develop.recurs reflects an imgo.rtant biological maladjus~ment of civilized man to his new artificial industrial envi.ronment and represents a striking i!lustra- tion of the momentous r61e which exo~ennus environ-. mental agents apvar,ently r~lay in the causatiort of cancer in ,~oneral. There , t o ~qod reasons for ass-mlne that 'he ma]orlty of oemmatlon.al carclno~enle a.o-en~cs is still unknown, and thi~ nnnl.les esneclallv to those occupational carcinoeon.s wh,ich have a low r~ten,cv and therefnre r~rn.. d,,.ce a low incid, ene.e of cancer am,on~ *he indixdduMs exgosed. This conclu'sion is SUpl~nrfed hv evidence obtained durin~ recent years wb,lch makes it likety that new

industrial cavcin,ogenic agents and heretofore unknown types of occupational cancers will be discovered in the next few years in American industries. When these facts and their implications are properly a.nd fully apgreciated, occupational cancers are bound to become a public health problem of the first order, which deserves serious and extensive investigation.

The existence of wi.d,e differences in the phys, ical and chemical properties of the various established or suspected occupational carcinogens suggests that these agents do not exert their specific p.athogen.ic ac'tion through the 'same mechanism. Some of them seem to cause a carmerous cellul~ar transformation by acting directly either in their original form or as m,eta~bolites or conjugates upon the cellular su,bstrate. Others may elicit such an effect by changing some normal chemical constituent of the cells or tissue fluids in such a way that it becomes endowed with carcinogenic properties. A third group of occupational carcinogenic agents may display such an action by producing functional disturbances in certa,ln organs, s~ch a's the liver, adrenal gIands, pltu, itary gIand and, gona6s Ieading to the endogenous generation of carcinogenic su.bs~tances.

Inasmuch as the great prevalence of industrial cancers in males is the result of an expo.sure factor and is evi- dentJy not sex-conditioned, it is suggested that environ- mental carcinogenic factors may be responsible for the orea'er fren-ency of several cryptogenetic cancers (lung, larynx, gastrointestinal tract, skin) among ma~es than .,mong females.

Observations made in connection with occupa.tlonal cancers s.how clea,rIy that a senescence factor is not involved in the production of these tumors, which have been seen in individuals between the ages o~ 10 to. 80' years. The age at onset of exoosnre together with its durati,on and inten,- sity conr 'the man,lfe,st'ati.on a~e of o,ccupatlonM cancers.

There is no concrete and valid evid,ence indicating that a genetic hereditary predispositi,on plays any im:portant r(,le in the ~roductlon and incidence of occupational cancers. The intensity, type and duration of expo,su,re to, and the potency of the carcinogenic aeents involved appear to be the main factors in determ.in, ing the susceptibility of the individual and the length of the laten,t r~eriod.

The r61e of the various !aroea.rcino~enic, antlcareino- oenic, nr~warcinotic and anticarcinqtic factors is evidently not a fixed one, but varies with the type and contact of tt~e carcinegen, ic agent invoI,~,ed and the in'een, s,lty of its action.

Although the United States: has been for several decade's the lead!ng industrialized .country, only very incomplete data as to the actual m~m~ber and variety of occupat,lonaI cancers, their causes, and the agents and o~.rat, ion, s restmn.. sib~e for their development a.re available, beeau'se these industrial diseases are not reportab!e in most of the states a ,d becm~se American indus'ries have been reluctant l'o nlace their cases of occurmtion,al cancer on ~e,cord. By the end of 1948 only 32 states had vas.sed some form of compensa'tion law coverln~: industrial diseases in, ~eneral. Since the maiority of such laws either contain time limi- Latlon clans.es as to the filing of com~yensatlon claims o*" ar~ply on,Iv to a limited and w'ell defined number of hazard- ous agent's or o.r~e,rations, they n,rovid,e inMeouate o,r defee- tire cove.ra~e against industrial cancer hazards. No com- r~etent and extensive investi~ation:s exict in rea, ard to the nnssi~!.e r61e that industrial or related exo~enou's a~ents might rJlay in the causation of some of the lar.~e numbers of cancers of th,e crvn.r tvee. A c omr~.rehenslve and thorough survev of rill ind,r n.roducin,~ or hand- Iing known or suspected carcino~enic agents, including



48 Cancer Research

those occurring in the form o~ .con, tam.inants, by-p,roducts or waste products, would fu, rn, i's~h a sound basds for the development of effective measures controlling the industrial cancer hazards of this country.

C A R C I N O G E N I C A C T I O N OF S O M E SUB- S T A N C E S W H I C H MAY BE A P R O B L E M IN C E R T A I N F I J T U R E I N D U S T R I E S . A U S T I N M. BRUES, H E R M A N N LISCO, and M I R I A M P. F I N K E L . (Argonne National Laboratory, Univer- sity o.f Chicago., Chicago 80, Illinois.)

In. the course of manufa.cturing pluton.ium in a chain- reacting pile, p,roducts cff u, ranium fiss.ion a,re evolved. These atomic split products are largely radioactive and must be separated from the pile uranium as part of the pluton, ium separation process.

W e have investigated, the late or ch,ronic effects ef exposure to certain, fission products a.r~d pluton,ium. T,he fission products, since they occupy the middle of the peri- odic table, include most o.f the rare earths, also iodine. barium and strontium. Experiment's were set up. to ob- serve the effects of ra<tioact,ive Sr 8~, C e ~ - P r TM, Y"~, and plutonium. Radium was also 'stud~ed for comparison with earlier human and animal work. W e used Sprague- Dawley rats, Carwo,rt,h CF-1 and ABC hybrid mice and a few C58 mice, as well as rabbits and .dogs. All an,finals received complete autopsy, terminal 'blood study, and roen,t- geno~rams. This is a progress report.

S,r ~~ with a 5,5.-day ,half-life and a ~cenden.cy to. co,ncen- tra,te in bone, has no vis,ible effect .on mammary tumor incidence and a slight po,si"dve effect on lymphoma in.cl- dence. It .is a producer par excellence of bone tumo.rs. Analysis of bone tumor inciden,ce ~n, a series of over 3,000 mice, 400 to 500 days after treatment, 'shows that expect- ante of tumo,r development is approximat,ely proportional to dose an.d to time after the on.set o~ a "latent pe.riod," but that the laten.t period it.self .i.s a fun.ct4on of dose which increases only gradually (perhaps logarlthmi.cally) with de.creashlg dose. Such a scheme is 'shown to fit, as well as can. be determined, the exi.st,ing human radium data. It appears that age of the animal at the time tpeatment ~s administered has little effect on latent period or tumor expectancy, even when the isotope i.s absorbed by the growing fetus.

Radium, in addition to bone tumor induct.ion, produces heavy cal.cification in the media of the larger arteries and in eer~a.in other s,ites.

Radioactive cerium--presyodym,ium, which gives an initial high liver concen~:ration, causes atrophy of this organ exeeI~t in the marginal areas, where ionization is less and regeneration can r~roceed. Bone sarcoma is also produced.

Ytt.rium ~*, which was fed by sto~nach tube. is virtually vna.bsorbed. One-tenlh of the a.eut,e lethal dose can be fed daily f.or 3 months with no measu,rable detriment. An,finals surviving this treatment, as well as tho'se surviving a single acute dose of 20 to 3(l me. per k~m., show a~ter several months a variety of intes'tinal lesions with o,betruction.

Plutonium ( P u ~ ) , when iniected in,travenously, has its highest immediate con,centrat{.on, in liver and srJleen, and {s later transloca~ed to h~ne. Acute l)luton{sm is similar to acute total body rad{a~tion sickness, with the addition,s of splenic .atrophy and gross liver changes. Ch,ronic plulon- {sin involves greying of hair, p,rogr,e.ssive liver damage, and bone sarcoma. Following subcutaneous injection of 1 ,gin. of plut~on, ium. local .fibrosarcomas may a,pgear

,Mthin a year. Local ep,ilatiol~s, ulcera~tion, keratosis, and spontaneous amputation a.re also seen.

T H E C A R C I N O G E N I C E F F E C T OF P I L E R A D I A - TION,S. P . S . H E N S H A W , E. F. R I L E Y and G. E. S T A P L E T O N . (Clinton Labo.rator~es, Oak Ridge, Tennessee.)

A variety of high energy radiations (fast and slow neutrons, gamma rays, beta rays, etc.) are associated with the uranium chain reaction in a pile reactor. All of these radiations are known to cause deleterious biological effeet.s and hence con.stitute health hazard:s.

In an attempt to understand more fully the types of injury induced, and a,t the 'same time obtain i.n.forma.tion on threshold and tolerance levels, numerous ~nimal experiments were carri'ed out. Four strair~s of mi.ce (CF~. ABC, A, C58) and, one strain of rats (Sprague-Dawley) were the main types o.f test material. The different radiations we,re obtained by mean,s of sele,e~ive fitt,r,aJt,ion, second,ary ,em~t.ters and rad~ioactive isotopes. Sin~gl,e massive dose treatments as well as small daily trea,tments were administered. The radiations used, with the exceplion of beta rays, were of the penetrating type, and for the most part were distributed throughout the. animal bodies.

Resu!ts revealed that the length of life var,ied inversely with the amount of exposure, an,d further that lhe amount of effect varied directly with the .concent, ration of ions in space and time. The termlrml eff.eet,s obtained when. penetrating radiations were used were mainly of two ty,pes: gen.eralized atrophy, and neoplasria of the ,hemopoietie organs. The atrophy wa's .attended~ by lowered leukocyte level, emaciation, loss of weight, and p,rem~atu,re death. The neopIasms seen consisqt.ed mainly o,f metiia..stln.al ]ym- phomatoses. The incidence of the latter was rMsed from less than 15 per cent in the controls to more than'. 60 per cent in certain, of the experimental groups.

When non-penetrating rad'iations x~er, e used t.he changes were limited mainly to the skin. Several mon.ths after single mass,lye d.o.ses skin. abnormalities of every type were l~resent. In an~m,al.s that ra,riely show skin les,i,ons of an3' type, the inc'.'dence of car.clnema was ,raised to 100 per cone and locl per animal were as gr.ea~ .as 50 to 1'00.

Th.e relative effectiveness: of fast n.eut.ron.s when compared width gamma rays appea'red to be clefin,{t,ely greater when late modifications were eon~si,dered rather t'han acute. Threshold levels of daily expo.sure, so far as shortening of life ~s concerned, were found, to ,be in t,he range of I r of gamma rays, and something well less than. 0.1 r.

E T I O L O G I C S T U D I E S I N H O D G K I N ' S D I S E A S E . H E R M A N A. HO'STER. (The Divi's.ion of Cancer Research, The Oh,io State University, Columbus. Ohio.)

An introductory reference yeas made to, the di.str~bution of Hodgkin's disease; the significance of tubercle bacilli and Bruce!la in Hodgkin.'s tissue and body fluids: the age dis'tributlon of patients with Hodgkln's all.sea,so; the ,prog- nostic importance of the sedimen,tation rate. ]ym,'0,hopenla, and anemia: the absence of pa.in ~n early Hod~kin's d,is - ca.so ; and t,he selective preference of the d~se,ase for specific tissue systems.

Obsera,ation:s following inoculation of tissue culture cells with cell-free material from Hodgkin's disease, lym.p,homa- *osis, and control tis'sue extracts and body fluids were descr:i,bed. Embryo spleen cells in tissue culture, stained with Seller's stain, were used routinely to demonstrate the presence or absence of fuchslnophilic c.-r s,mic inclu-



A.A.A.S. Gibson Island Researc/z ConCerence on Cancer, 1946 49

s,ions. Som"ces of embryo spleen were the chicken, duck, and guinea pig. All p,reparations of ~resh, fro.zen, o.r lyop,hilized I-Iodgki~'s ti, ssue, lymp:homatosis, and control tissues were min, ced and eem'ri~uged for 30 minutes at 4,000 r.p.m. The supernata~t fluid obtained ' was used as the sou.tee o{ inoculum befo,re arid, after d, ifferential low gravity and ultracentr,ifugation, ultrafiltr.at~ion, and dilu- tion to or~e part in 2 b,ill, io,n. Cyto-stimulating and, cy~o,ly- tic changes were notfed following inoculatio.n o( tissue cultures wit,h cell-free material of H.odgkin's dise.a'se and lymphoma~o,sis. T'hese changes were l~ot o;bserved ' i,i~ cultures inoculatedt wiecb control tissues and .body fluids. Inocula composed of the resuspen.,ded sediment following high gravi ty centrifugation produced: in tissue cultu,res an iner.ease in cyto,stimulation and the number of in, clusion bodies.

Fuchsinoph, ilic intracytoplasnl.ic irmlusion, s were found not only ir~ ti,ssue culture's inocula.~ed with m a~terial from Hodgkin's disease and lymphomatosis, but also were noted in some control preparations containing either chi.cken or duck oon.stituents, o.nly. It~es,tigat,ion, of this p,henomenon resul'ted in' the fin,d,ing that chickens and d'u,cks, from which the p la.sm,a and embryos had been, obtained, were infected wi th lymphomatos,is. Because of this potential s,ource .o.f con,taminatio,n, adult and emrbryo chicken and duck m.ater'ials were considered unsuita61e for fur• use in tissue cultures. Therefore homologou's gu'inea pig sub- s'trate and' nutrient medium we.re ado.pred.

Inact,i,~atio.n of the ag~ents o.f Hodgkin,'s, dis:ealse and lymphonl, atosis by ultraviol,e,t 1.1ght was described.

The thi.rd phase of the work d,eals with clinical observation.s made by the author and h,i's associates. LTnder the direction of Dr. Perk Lee Davis of the Women's Medical College of P.eimsylvania, two patients with Hodgkin 's d'is- ease rece,ived intranmscular injections of 10 cc. of pleural fluid, obtain,ed from a terminal ca;se of Hodgkhfs disease of the thora, clc type. The donor aI'so reoeived an iden'tical inject,ion of his own fluid. At the end of 48 to 72 hours all 3 recipient, s developed a generaI,ized m, aculopapular eruption with s,evere itch,ing. Most of ~the lesions soon dis,appeared', but a few still remained; these enlarged, and some showed a central necr.os,i's between the fourth and seventh days following the injections. These les.iofls varied in size from 0.2 to, 2.0 cm. and did not respond to t.reatm,er~t with chemot,herapeutc agents applied locally. Examinat ion of biopsy material obtained fr,om the skin lesions revealed ~th,e typical Sternh,erg-Reed cells (H.odgkin's cells) in all ca'ses. Following the'raI~eutic doses of x-rays, these I.esior~s disappeared leaving brown pigmented ~reas, but the downh, ill course of the pa.tients appeared to be aeeelerated folio.wing the in.iectio.ns.

Lesions simil,ar to, ,but less, a.cu• ,than, those just described have been ob,served by the author in several patients wi th I-Iodgki,n's disease:

(a) Multiple sk.in lesions occurring immediately after a seri,e's of five daily x-ray treatments to a greatly enlarged s,pleen were seen in ease of H .O . , a 37 yea,r old white male wi~h acute Hodgkirfs d.i,sease.

(b) M. H., a 34 year old white female, developed muI- tiple skin lesions of the back 48 hours follow.ing the subcutaneous ino.culat,ior~ of an .extract of H.odgkin's ti's,sue that had been i'na,ctivat, ed by ultravi,ol,et i,rra,diatl.on.

(c) The au~thor has observed # cases of H,odgkin's disease tha t ex.h,ibit,ed skin. lesion, s of spontaneous origin similar to those des~cribed above by Dr. Davi,s. One of the aueh.o,r's ass,ociates, Dr. Anthony Rot t{no, of St. Vin- cent's Hospital in New York City, has invest, iga~ted and made biopsies of 2 s,imilar cases.

T I S S U E C U L T U R E S T U D I E S O F L Y M P H N O D E S OF H O D G K I N ' S D I S E A S E . C. G. GRAND. (New York University Washington Square College, New York 3, and, St. Vincent's Hospital, New York, N. Y.)

This report is based on, the study, including tissue culture, of lymph nodes from 75 cases d early and late stages of t-I,odgkin's disease. As controls, normal lymph nodes were used, also lymphosarcomas, leukemias, meta- static carcinomas and lymphadeniti.s.

The tissue cultures we're maintained for periods varying from a few days to several weeks. After 72 hours of incubation, the .cultures of the Hodgkin's disease showed, on the ,periphery of the explant, large nmltinu- cleated giant cells of the endothelial type which resem- bled Reed-Sternberg cells. They were absent in cu.l.tu,res of other lymphomas and of normal nodes.

Brilliant cresyl blue, as a vital stain for virus cell inclusions, stained the granules of the central body of these cells from red to pur~ple within 10 to 15 minutes. Similarly stained inclusions also were found in macrophages, lymphocytes, and in many cases fibrocytes. Seller's stain was also. used. This colored the in,clusions a bright red, while the nucleus and regular cell granules were deep blue.

The outgrowing cells of the Hodgkin's disease cul.tures were seen to disintegrate in such a manner as t,o indicate some sort of infection. This is not seen in euI.tu.res of true neoplasms. The time for complete disintegration of a culture depends on the number of Reed-Sternberg cells and the number of cells containing inclusion bodies. The time may be from 1 to 49 days before all the cells i11 t~le cul,ture have cy'stolysed and the medium liquefied.

Supernatant fluid was removed from cultures of Hodg- kin's tissues grown in Carrel flasks and was centrifuged at 2.000 r.p.m, for half an hour. When this cell-free fluid was adcled to cultures of normal human lymph node, chick spleen, mouse spleen and lymph node, the lymphocytes and macrophages showed specific cell inclusions within vacu- o r s after 3 days. Serial passages in tissue culture o{ the supernatant fluid showed, an increase in potency so that at the seventh passage, the time for complete disintegration o.f the cultures was cut from I0 or I5 to 3 days. There was also an increase in the number of cell inclusions with the successive passages. Sttpernatant fluid from cultures of normal human lymph nodes and lymphomas other than Hodgkin's disease gave negative results.

Concentration of the agent was done by fractionation centrifugation and was found to be very active: when added to tissue cultures of chick and mouse tissues when diluted 1:1,000.

This material will not pass bacterial filters except under certain conditions. It was found to keep its activity for 2 yea,rs a~t 4 ~ C. if tl~e pt,rified ma~e,rial is SUSl)encRd in .05 per cent serum hydrolysat.e at pH 7.0 or if suspended in inactivated rabbit serum.

T H E E F F E C T OF N I T R O G E N M U S T A R D S ON P R O L I F E R A T I N G E M B R Y O N I C T I S S U E S . D. B O D E N S T E I N . (Medical Division, Edgewood Ar- senal, Maryland.)

Em,bryos of .dmblystoma, p unctatu~m (tail-bud-stages) were exposed for 45 minutes to a 0.001 per cent solution of the nitrogen mustard compound m,ethyl-bis (betaehl,oroethyl) amine hyd,rochloride.

The agent affects selectivity centers of proliferation and leaves differentiating regions unaffected. With in a short



50 Cancer Research

time of the exposure (about 2 days) all mitotic activity is abolished. The affected mitotic cells exhibit clearly two distinct reactions. They either break down rather rapidly into chromophilic fragments or increase strikingly in size. The nuclei of these large cells may fragment in later development, while some of them seem to recover. In organs where the processes of proliferation and differentiation occur in different cells, as in the nervous system for exam,pie, enlargement and early breakdown of the mitotic cells occur side ,by side. In the embryonic gut, however, where differentiation and proliferation occur within the same cell, only cell enlargement occurs, while at the same time cellula,r d, iffer,entia~ion continues un- hindered. This results in the formation of giant cells which apparently function normally. It was found that the ,cells showing the enlargement reaction are character- istically cells which in later development make up pro- liferating regions of the embryo, while the cells exhibiting the early breakdown reaction without enlargement are those whose mitotic activity is limited to early embryonic development. Consequently fewer and fewer broken cells of this type are found as the age of the animals at exposure increases, because more and more of these mitotic cells have entered their differentiation phase and are hence not affected.

By its selective action in showing up reMons of proliferation the agent is an exceedingly useful tool for the study of growth patterns. The agent's specific ability to arrest mitotic activity without interfering with differentiation allows us to control experimentally the size of organ, s and gives us a new approach to the problems of growth control.

E X P E R I M E N T A L O B S E R V A T I O N S ON T H E E F F E C T S OF T H E N I T R O G E N M U S T A R D S ON N E O P L A S T I C T I S S U E S D . A . K A R N O F - SKY. J. H. B U R C H E N A L , R. A. O R M S B E E , I. C O R N M A N . and C. P. R H O A D S . (Sloan- Kettering Institute, New York 21, N. Y.)

The effects of the nitrogen mustards N(C.:H,C1):: (called H N 3 ) , CH:~N(C~H,C1)_~ (called H N 2 ) , C._,H:,N

(C_oH~C1)~ (called H N 1 ) a n d ~ : ~ C H N ( C ~ H , C 1 ) . _ , , ~,_, 1 • l / :

used in all cases as hydrochloride salts, have been stvAied on normal animals and tissues, and then assayed for chemotherapeutic activity against various types of neoplastic tissues under several conditions: I.eukemia in mice, mouse sarcoma 180 growing on the chick chorioallantoic membrane and sarcoma 180. adenocarcinoma 1025 and lung carcinoma MA 387 in roller-tube cultures.

In, n,ormal animals it is found tha;t certain t iss.ucs, including the hematopoietic and lymphatic tissues and the intestinal epithelium, are unusually sensitive to the action of these compounds. It may be significant that the sensitive tissues are those which possess relatively hlgh rates of oell divi,s.ion. H N 2 and HN3, the on, Iy one's tested, on intravenous injection in.to animals have a very rarfid and presumably direct action on the affected tissues. In marly respects the sulfur and nitrogen mustards are found to resemble x-rays in their action; for examgle, in their toxi- cological effects on animals, their inhibiting action on cell division, atgd in their ability to produce mutations in Drosophila (2) and Neuroshora (3).

The ~dditive lethal effec~ in mic.: ~ of H N 2 and x-ray~ vary with ~he sequence of admin, istration. If 80 per cent of a lethal dose of x-ray is followed an hour later by 80 per cent of a lethal dose of HN2, the lethal action

of the tw,o agents is only slightly additive. If the sequence is reversed, and the same dose of I-IN2 is given ] hour before the x-rays, an additive, much greater let.hal effect results.

Four compounds, the methyl-bis, ethyl-bis, iso.propyl-bis and the tris-(fi-chloroethyl)amine were tested against 1 myeloid and 3 lymphoid leukemias in mice. The mice used were of a highly inbred stock Afb in which transmitted leukemias of the 4 strains used take in apwoxi- mately 100 per cent. The experimental procedure was as follows: a leukemic mouse with enlarged spleen and nodes was weighed and injected intravenously with a given mgm./kgm, dose of the compound to be tested dissolved in saline. One hour later (less in: the higher doses where toxic effects often killed the mouse in from 5 to 50 nlhmtes) the mouse was killed, the spleen removed, and the sgleMc br.ci injected subcutaneously in,to 4 young mice of the Afb stock as a bioassay. Sections of liver, spleen, and kidney of the donor mouse were made. All mice used for bioassay were then watched for development of leukemia, and in all questionable cases sections were made. It has been our experience that large doses of the nitrogen mustards, even when not completely cytocidal to leukemic cells, so affect them that the incubation period in the bioassay animal may be much prolonged over the normal time.

The problem of how long the nitrogen mustard should stay in the live animal in order to have a maximal effect on the leukemic cells of the spleen is unsolved. Fre- quently the animal died rapidly and we arbitrarily set 7 minutes survival time for the donor animal as necessary for a valid bioassay.

From the preliminary data, it would appear that for the 4 strains of leukemia tested the cytocidal d.ose of tr%-(3- chhwoethyl)-amine varies from 7 to 22 times the LDS0 or 10 to 33 mgm./kgm. The methyl-bis compound inacti- vates the leukemic cells at 8 to 12 times the LD50 or 33 to 75 mgm./kgm. With the et.hyl-bis compound, only in the Af9417 leukemia were there any consistent results and there the cytocidal effect was manifest at 22 times the I.DS0 or 33 mgm./kgm. With the isopropyl-bis(3-chloroethyl)-amine, a dosage of 37 times LDS0 or 50 mgm./ kgm. inactivated the cells of 3 strains of leukemia in some experiments.

It appears from this experiment that with the tris- and the methyl-bis compounds there is a definite cytocidal dose for leukemic cells. With the ethyl and isopropyl analogs, the evidence of cytocidal activity is not clear even at much higher multiples of the LDS0. If this is borne out in the finished experiments, it would seem to indicate that the former compounds are more likely to be of value clinically than are the latter.

The lethal dose of HN3 in fertile chick eggs is almosr directly proportional to the mass of the embryo; the LDS0 by injection into the yolk sac is about 0.015 mgm./egg in the 3-day chick and about 0.5 mgm/egg in the 12-day chick. Anomalies of the beak and legs have been produced in 3-day-old chicks surviving LDS0 doses of HN3, but this has not occurred in the older eggs. A similar effect has been produced with the sulfonamides (1). In 12 days eggs with actively growing sarcoma 180 tumors on their chorioallantoic membrane, a dose of H N 3 which is not lethal to the egg will stop the growth of the sarcoma 180, render it incapable of growth when transplanted into mice and I~r.'.Muce. as shown bv histological examination, dissolution of a majority of the sarcoma cells and the appearance of remarkable giant cells containing large ab-




normal nuclei, occasionally multilobed or fragmented. The appearance a n d location of these giant, cells s trongly sug- g.esCc tha~ they stem from the sarcoma cell,s. No significant change in the chick membranes or vascular tissues was observed. The LD50 of H N 3 for the t u mor in the egg is in 'the range of 0.05 to 0'.10 mgm. /egg , when, injected at 12 days, about 1/5 to 1/10 of the LD50 for the chick embryo. P re l iminary experiments designed to show the speed of the lethal action of HN3 in such egg- tumor preparmions indicate that the tumor can: be fatally dam- aged within less than 30 minutes ( the shortest period studied) by 0.2 mgm. / egg of HN3.

Using the roller tube technic, cultures of sarcoma 180, lung carc inoma Ma387, and normal mouse epithelium were exposed to various con,centrations o.f N(C2H~C1):; dissolved in the nutrient medium. The results of these experiments u p to the present time indicate that certain tumors appear to be more susceptible than others to t.he action of this compound. Sarcoma 180 cells have consistently been killed ~,t concentrations of 20 mgm./1 after 24 hours exposure as indicated by their failure ~o cause tumors when tran.s~plar~ted i~r~to su's,cepti,ble m,i,ce. MA387, however, has in some cases survived concentrations as high as 80 mgm. /1 .

There was sugges'tive evidence that the tumor ceils were somewhat mere sensitive to concentrations of H N 3 than were the corresponding normal cells. At concentrations of up to 80 mgm./1, the deleterious effects on. nor- real cells appeared less distinct than on tumor cells. At ~0 mere./1, Ma387 and sarcoma 180 = cells were visibly affected alt'ho.ugh, not dis,i~tegmted, whevea, s n~,rm.aI cells appeared ur~affected. At 20 mgm./1, both normal and tumor cells lo.oked normal cytologically, but those of sarcoma 180 were apparently rendered nonviable as evidenced by negative bioassay results.

While these lethal concentrations will have to be more precisely defined by further exwriments , it is interesting to note the difference in susceptibility of the tumors employed and that tumor cells could be consistently rendered nonviable by concentrations which caused little visible damage to the cells involved.

The clinical evahmtion o'f the methyl-his and t r is - (~- chloroethyl)anfine hydro,chlorides is in progress. In addition t.o vary ing the dosage level, the effect of these drugs rm various neoplasms not connected with the blood-forming organs is being evaluated. Neuro,blas~toma, Ewing 's tumor, glioMastoma, and malignant melanoma are under treatment at the present time, but it is still too early to n.ote any definite results.

It is to be re-emphasized here that this is in the nature of a pre l iminary report and that much work remains to be done ,before these data can be considered definitive.

R E F E R E N C E S

1. ANCEL, P., and I,ALLE~{ANI), S. Sur une malfor- mation du bec et des memgr.es obtenue chez l 'em- bryon de Poulet, fi l'aide de sulfamides. Compt. rend. S,oc. de Biol., 136:255-256. 1942.

2. AUmeBACm C., and RoBsoN, J. M. Chemical Pro- duction of Mu,'ations. Nature, 157:302'. 1946.

3. HoRowt'rz, N. H., HOI_ILAttAN, M. B., HUNGATE, M. G., and WRIGHT, B. Mustard Gas Mutat ions in Ncurospora. Science, 104:233-234. 1946.

T H E C L I N I C A L A P P I , I C A T I O N O F A N I T R O G E N M U S T A R D C O M P O U N D M E T H Y L BIS ((.3 C H L O R O E T H Y L ) A M I N E T O T H E T R E A T - M E N T O F N E O P L A S T I C D I S O R D E R S O F

T H E H E M O P O I E T I C S Y S T E M . * C H A R L E S L. S P U R R , L E O N O. JACO,BrSON, T A Y L O R R. S M I T H , and E. S. G U Z M A N B A R R O N . (Dept. of Medicine, Un, ivers,ity of Chicago, Ch, icago, Ill~n.ois.)

Gilman, Goodman and their co-workers (1) were t~he first to study the effects, of a ni t rogen mus ta rd compound on human neoplasia. They studied the p.harm, acologiea! and clinical effects of t r is-(f l -chloroethyl) amine administered intravenously in 7 terminal cases. Independent ly in

.March, 1943, Jaeobson and his group (2, 3), un.dertgok a similar cl,i.n'ical study in an at tempt to evaluate another nitrogen, mu.stard; namely, methyl-bi,s (fl-chI,oroethyl) amine as a t~h:erapeutic agent. A n un~selected group of 59 cases of neoplastic and allied 6i's,orders o.f the hem,opo,ietic sys,tem have been studied.

Methyl-his (fl-chloroethyl) amine wh, ich i,s highly cyto- toxic is adm,ini.stered intra~,enously in a dose o~ 0.1 mgm. per kgm. of body weight diluted in approximately 100 cc. of normal saline on consecmive days in series of 2 to 6 injections. The toxic reactions observed after the administration of this compound have been nausea and vomiting, phlebothrombosi.s, varying degrees of destruction of hemopoietic .ti.ssues and consequent d,epression of the hema,to- logical constituents of the peripheral blood. Nausea and vo,mi~ing Madch. are partial,ly cemr'~l a, nd local in, origin usuMly follo.w ea, cb dose wlth,in 3 hours and p.e.r'si,s~c for a similar period. Phlebo<hrombos,is occurs only if the drug is not sufficiently dilute when injected or delayed by c;bstructing masses in the veins used for the admin,istration.

Evaluat ion of the reaction of the hemopoietic system to, the therapeutic application of the methyl-h.ls (B-chloroethyl) amine demonstrated tha% following .each serie's, a c.haraotevistlc lym.phopenia became apparent within 48 hours. Recovery of the lymph,o.ctyes usually begins in the second week following treatment.

A n,eutropen.ia develops in the third week and is usually of about one week's duration. There is also a promp~ but variable th, rombocyto,pen, ia during the fir'st 3 weeks. This is occasiona~lly sufficient to produce an increase in the bleeding time but only on 3 o.ccasions have petechi.ae appeared.

A series of t r i -weekly sternal mar row aspiration, s were made on. 5 case's over a per, iod of (5 weeks after t reatment . Tb.ese show a depression of the total nucleated cell count from an average of 100,000 per cu. ram. to 6,400 per cu. mm. on the 2'ls.~ day and a sharp recovery to 65,000 per cu. ram. in the fourth week. There is a slight depression in 'the bta(st forms and pro.myeleeytes during the first week and a moderate increase in the second week. The myelocytes and metamyelocytes decline through the first 2 weeks and regenerate to a peak in the third and fourth weeks respectively. The percentage o,f polymorph.onuclea~r cells increase.s in the firs~ 2 weeks and decl'ines sharply in the third week, regenerat ing to a peak in the fif~th week. There {'s also. a depression, i.n the e.rythro~id seri,e,s for 2 weeks with ord, erly regenerat ive peaks of p.ro.erythroblast and b.asov'hilic erythroblasts in the third week and of the orthochrom:ati.c crythr0blast in the sixth week. Many giant form,s of the granular series are seen in the peripheral blood esp,ecially du,ring the first week after adminis~ration of the drug. Infections and other clinical man'i- fes*ations usually attributed to, se'vere bukooenias were conspicuou.sly rare. In~ the .one tristan, co in whi,ch an infec--

* This projee.t hafs been supported in par t by a grant from the Commit tee on Growth of the Amer ican Cancer Society.



52 Cancer Research

t, ior~ occut~red it was succes,sfttlly controlled by pen, iciIlin therapy.

The reaction of the peri:p,heral blood and the clinical re'sponse served as the principal guides in evalua~ting the most efficacious total dose per course of treatment. In a series of 79 courses the severity of the depression of the peripheral blood increased with the number of con'secutive injection, s per course. There was a fall of hemoglobin and erythrocyte values from 0.7 gm. per cent and 140,000 per cu. ram. respectively following 2 injections (average total dose 12 mgm.) to 1.6 gin. per cent and 460,000 pe~ cu. m m. respectively after 6 ,injections (average dose 36 mgm.). An irmreasingly severe decline in leukocy*es wa,s also noted. The average percemage of fall in the total l, eukocy,te count increased from 3.5, per cent to a plateau at 65 per cem and 67 per cent as the courses were increased from 2 to 4 and 6 injections. An irregular increase in the incidence of leulmpenias of les.s than 3,000 cell,s per cu. ram. developed aJs the courses were prolonged.; however, the.re was a sha,rp increase in, the severity and duration of leukopen'ias with tota,1 doses above a 25 to 30 mgm. level or with 4 or more injections.

The pos, sibil,ity that t~his h.emotoxi,e material may have a cumu, lative effect with repeated do,ses and everaually lead to a permanent hypopla,stic reaction in the bone mar- row has been assessed by grouping 58 courses according to the sum total dose administered in 50 mgm. steps from 50 to 250 mgm. The reaction of the peripheral blood in each group as judged by the averag, e pre'treatment and post-treatment leuk.ocytic counts shows no rema, rka~ble t, ven`d. However, the percen,tag, e irmid, en, ee of leukopen,ias increases from 40 per cent in the 0 to 50 mgm. group to 86 per cent in the group w.i~h a cumulated dose of 200 to 250 mgm. This is the only indication that the hemoIsyietic system may become more sen,sitive to the compound as the co,urses of 15reatmenq: are repea'ted. Two factors tend to contradict this': (a) the I.eukocytes return to ~mrma! range following each course, and (b) the cases in this group are of longer duration and several had leukopenia prior to initiating treatment. It is suggested that a cumulated dose of 600 mgm. or more will be necessary before an unequivocal answer may be made to this question.

The clinical response to methyl-bis (r amine hydrochtoride in 29 cases of Hod,gkin's disea'se observed over a period o.f 3 to 36 mentbs shows-ins~armes in which the various symptoms of the disease have been controlled. In this g.roup of causes 94 per cent of 1_29 courses of treal- ment produced clinically significant remission,s. Among the 8 cases that failed to respend to treatment, 5 were in a terminal phase of their disease, 2 had advanced di.sease and were no Ion`get responsive to roentgen therapy and 1 had a remission from fever of only 3 weeks. Of particular interest are 4 cases which had been considered "x-ray resistant." Thi, s g, roup has responded repeatedly to courses of n,i,trogen muslard. The remis.sion's induced in this group of patients with Hodgkin's disease vary from 0 to 10 months, although t'he length of remission increa'ses with the total dose per course. A course of four injection,s of 25 to 30 mgm. was selected as opt,imum because .of the plateau in the severity of leukopen'ia at this level. There is a slight decrease in, the average remi.ss,io.n with repeated courses; ho,wever, the number of cases cumulating do,ms a,bove 150 mgm. is too small to evaluate this question. Two cases have appeared to develop a tolerance or resistance to the material.

A group of 9 cases of chron,i,c lymphatic leukemia has shown, remissions of 2 to 24 months. This variation

depends largely upon !he activity o.f the d{,sease. Although each case responded to the ir~itial course, the group, survival i,s poorer th, arL in Hodgkirfs di,sease. Five patients have died with, in 18 months after the initial treatmenl from advancing disease or it,s complicatior~s.

Six cases of lymp,hosa, rco.ma have beel~ treated with m.ethyl-bis (fl-chloroethyl) amine. Two cases, .both af- flicted with an extremely undifferentiated type of tumor, failed to respond to the initial eou,rse of treatm,er~t. The remaining cases had remi:ssion, s of 2 to 6 months and have been followed for 2 years. A case of giant follicular lymphoma responded in a manner simila~r to those of lymphosarcoma.

Seven cases of chronic myeloid leubemia have responded to treatmen't with methy'l-,bis (fl-chloroethyl) amine with tran, sitory symptomatic and hematological improvement. We do not believe that the remi~ssions induced were clinically sign.ificant.

Two cases of ~cute leukemia, 1 lymphati, c a.nd, 1 myelog- enous respectively have beert treated wi~thout sy,mpto- matic relief, severe leukopen`ia developed in the latter case.

Seven cases of polycythem,ia ru:bra vera have responded favorably following cou'rses o,{ methyl-bi's (~8-chlo,roethyt) amine with remiss:ion, s of from 3 to 17 m<~mh,s,. Six continue in thei,r initial remission. The hematological picture in each reverted to normal values wil,hin 1 month and remain within normal values. One case althougk free of symptoms had a moderate increase in th,e hemoglobin and erythrocyte levels and was treated 6 months after his first course.

The remissions produced in Hodgkin's disease and lymphosarcoma and in some cases of chrors 13rmpha~ic leukemia are suffieien.t to wa,rran~ extended study with a view to using methyl-bis (5-ch,loroethyl) amine or a related compound as an adjunct to roentgen therapy. The remissions produced in. cases of polycythemia rubra vera are comparable to th~yse following treatment wi, h' radioactive phosphorus and poss,i,bly superior to roentgen therapy. The n`itrogen mustafds have not been, proven superior to roentgen therapy in the treatment of any of the lym.phomas.

R E F E R E N C E S

1. GILMAN, A., and PIIILIPS, F. S. The Biological Ac- tions and Therapeutic App;lications of the fl-Chtoro- ethyl A.min, es and. Sulfides. S,eience, 103:409-415. 1946.

This article refers to the work done by A. Gil- man, L. Goodman, C. E. Lindskog, and J. Dough- erty at the New Haven Hospital.

2. GOODMAN, L. S., WI>ITROBE, M. M., DA,-Vt~SI-IEK, W., GOODMAN, M. J., GILMAN, A., and M c L E ~ A x , M . T . Nitrogen Mustard Therapy. Use of Me~h- yl-B is (Beta-Chloroethy,l) Amine Hyd.rochloride' and Tr?s (Beta-Ch,loroethyl) Am~ine Hydrochlor,ide for Hodgkin's Disea.se, I.ymphos'.a,rcoma, Leukemia and Certain Allied and Mis.cellatmous Disorders. J.A.M.A., 132:126-132. 1946.

.3. Jacousmv, L. O., SPURR, C. L., BARROX, E. S. G., S~t~TH, T., LVSHBAUG~, C., and D~cK, G. F. Nitrogen Mustard Tbera,py. Studies' on the Effect of MethyI-Bis (Beta-Cbloroethyl) Amine Hyd,rochloride on Neoplastic Diseases and Allied Dis- orders of the Hemopoietic System. J.A.M.A., 132 : 263-271. 1946.

These authors have another article with sim.ila,r title in Iyres's for pt,blication, in the Journal of Clin- ira1 Investigation.



1947;7:37-52. Cancer Res Abstracts of Papers PresentedA.A.A.S. Gibson Island Research Conference on Cancer, 1946.

Updated version

http://cancerres.aacrjournals.org/content/7/1/37.citation

Access the most recent version of this article at:

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

Subscriptions

Reprints and

[email protected] at

To order reprints of this article or to subscribe to the journal, contact the AACR Publications

Permissions

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

.http://cancerres.aacrjournals.org/content/7/1/37.citationTo request permission to re-use all or part of this article, use this link



http://cancerres.aacrjournals.org/cgi/alerts

mailto:[email protected]



a+a.a+s. gibson island research conference on cancer, 1946 · a+a.a+s. gibson island research...

Documents