Abnormalities in the Distribution of Biotin in Certain Tumors and Embryo Tissues Philip M

Home , Biotin deficiency

Abnormalities in the Distribution of Biotin in Certain Tumors and Embryo Tissues Philip M. West, Ph.D., and William H. Woglom, M.D.

(From the Department o~ Cancer Research, College o/ Physicians and S.rgeons, Colt~mbia University, New Yor](, N. Y.) (Received for publication January 27, 1942 )

In a preliminary note (i3) it was pointed out that The basal biotin-free culture medium for the test certain tumors contained much more, others sig- organisms consisted of the following: mannite 5 gin., nificantly less, vitamin H or biotin than did normal dipotassimn phosphate o. 5 gm., potassium nitrate o.2 adult tissues. It was also observed that embryo tissues gin., magnesium sulfate o.2 gm., sodium chloride deviated from the normal biotin levels in the same o.r gin., calcium sulfate o.~ gin., ferric chloride o.o2 manner as the corresponding tumors. Because of the gin., thiamin ioo 7, /3-alanine ioo Y, distilled water indispensability of biotin for the growth of many i liter. The mixture was brought to boiling, filtered microorganisms and higher animals it was considered while hot, and on cooling adjusted if necessary to pH possible that these differences might reflect funda- 6.8. This medium is incapable of supporting growth mental metabolic changes characteristic of rapidly of the organism, but since the only additional factor multiplying cells. Therefore, the earlier observations required is biotin, growth occurs upon the addition have been extended to include n-tony more tumor of tissue extracts in proportion to the amount of types. In addition, various attempts have been made biotin they contain. to alter the biotin content of tumors experimentally and thus influence their growth. Various dilutions of the tissue extract to be assayed were measured in amounts from o.2 ml. to o.ooi ml. MATERIALS AND METHODS into i2 5 ml. Erlenmeyer flasks, and 2 5 ml. of basal medium then added to each. After autoclaving Tissues were assayed for biotin by a new micro- for 3 ~ minutes at 15 pounds' pressure, these mixtures biological method using the indicator organism Rhi- were inoculated with one loopfut of a light suspension zobiurn tri/olii 205, for which biotin is a specific of bacteria in sterile distilled water (approximately 5 ~ growth factor (x2). Only a small proportion of the to ~oo million cells of Rhizobium per hal.). The in- biotin in tissues is in free form (Table I), the re- oculum was grown on slant cultures consisting of mainder being firmly bound and liberated from such basal medium plus agar i. 5 per cent and a small combination only after hydrolysis. Thus the treat- amount of biotin supplied in the form of o.i per cent ment of the tissue to be assayed differed somewhat Difco yeast extract. Insufficient biotin is carried in depending upon whether free or total biotin was to be such an inoculum to permit growth in the basal determined. medium. For total biotin assays, about 0. 3 to o. 5 gm. of tissue was extracted by autoclaving with 5 ml. of These inoculated mixtures were incubated for 5 io per cent sulfuric acid for one hour at ~5 pounds' days at 28 ~ C., after which the opacity resulting from pressure which freed the biotin from its combination bacterial growth was measured with the Evelyn photo- in the tissues. The extract was then filtered into a electric colorimeter, filter 54 o. From standard curves calibrated tube, the solid debris washed, and washings for this organism, readings were converted into num- added to the filtrate. After neutralizing the filtrate to bers of Rhizobium per ml. and the results expressed bromthymol blue with 25 per cent sodium hydroxide graphically. By comparison with crystalline biotin it was diluted to Io ml. or standardized yeast extract curves included in every For free biotin assays, acid hydrolysis was of course assay, the amount of unknown extract inducing half- not necessary. A similar amount of tissue was simply maximum growth was determined, and since this chopped very finely and extracted with 5 nal. of water corresponds to the activity of 0.32 millimicrograms of by placing the tubes in a boiling water bath for 20 biotin in 25 ml. of medium, the biotin content of the minutes. The mixture was then filtered, the residue extract could be readily calculated. All results are ex- washed as before, and the filtrate diluted to 5 or pressed in n>/ (i• -:> gm.) of biotin per gin. of io ml. dried tissue. 324

THE BIOTIN CONTENT OF TUMOR, EMBRYO, AND ful dissection it is possible to obtain rabbit skin rela- NORMAL ADULT TISSUES tively free from the subjacent connective tissue, so In this comparative study of the biotin contents of that by microscopic comparison there is an equally normal and malignant tissues considerable care was high proportion of epithelial tissue in both normal exercised in the selection of material. Only those skin and tumor. Results of biotin assays (Fig. z) tumors for which suitable control tissues could be indicated an average of 36o my of biotin per gm. of obtained were chosen. Samples of each were examined tissue for 5 papillomas, while I6 skin carcinomas microscopically and unless found to be quite healthy averaged 369 my. In contrast to this, four samples and reasonably free from other tissues, the material of normal skin taken from areas adjoining the papil- was discarded. It is obvious that these restrictions must lomas contained only 89 m 7 of biotin per gm. Since have eliminated at once a considerable proportion of certain similarities in the carbohydrate metabolism of all available human material and much animal ma- embryonic and tumor tissues are known to exist, de-

5KIN TUMORS LIVER CARCINOMA LUNG CARCI N OI"IA RABBIT RAT MOUSE: HOU,$E 600 M r 8=onN I~R 61"I. _4OO0

4oo

Do0o ~oo IM /,! LYMPHOID LEUKEMIA CONNECTIVE TI 5SUE TUMORS ISTOMACH CA FIBROl'IYor"t/ .MOUSE RABE5I__r ~ MOUSE MAN MAN .15OO .15oo

F .200o

.1000 .I00o -'~176176 ! .500 -

I ' i a "~176 TUMOR ~ EMBRYO PAPILLOMA I'~ NORMAL ADULT TISSUE I P-C- CARCINOMA- FIG. I.--A comparison of the biotin contents of certain tumors with analogous normal adult and cmbryo tissues. terial as well. For example, many of the comnaonest terminations of biotin in the former were included tumors, such as carcinoma of the breast or uterus, for comparative purposes whenever possible. The skins were unsuitable because a control sample of the cor- of domestic rabbit fetuses approximately 3 days be- responding normal epithelium was not available. Like- fore birth contained an average of 531 my, of biotin wise carcinomas of the stomach or intestine were per gm. From these assays it is apparent that although usually valueless, either because of necrosis in the skin papillomas and carcinomas of the rabbit do not tumor itself or of inability to obtain normal control differ significantly in biotin content, they are both far epithelium free from underlying tissues. The results richer in the vitamin than the corresponding normal which are reported below, thcreforc, represent as adult tissue, while the value for embryo rabbit skin nearly as possible a true comparison of the vitamin H is still higher. contents of certain normal tissues and their malignant Liver tumors.--Primary tumors of the liver, when derivatives. found to be well circumscribed and free from necrosis Skin tumors.--Many specimens of the Shope rabbit and connective tissue, were considered ideal for bio- papilloma and related carcinomas were generously chemical study, since an adequate supply of normal supplied for this work by Dr. John G. Kidd, of the liver was readily available from the same animal for Rockefeller Institute for Medical Research. By care- comparison. Such tumors were produced experi-

mentally in Wistar rats by feeding a diet of unpolished While these analyses of older tumors indicate rice plus o.o6 per cent p-dimethylaminoazobenzene changes in biotin content similar to those observed in (butter yellow) over a period of 3 to 6 months. Sev- carcinoma of the rat liver, the reason for the un- eral suitable tumors were obtained, 6 of which on changed biotin levels of the younger mouse tumors is analysis were found to contain an average of 2o3o my, obscure. This discrepancy between two values for the of biotin per gin. (Fig. i). Some had considerably same tumor strain has also been noted recently with less biotin than this, but were eliminated because the transplanted liver tumors of C3H mice, also sup- histological study showed that they contained excessive plied through the kindness of Doctor Andervont. amounts of connective tissue. Parts of the tumor-bear- A few primary liver cell carcinomas of human ing livers which were shown microscopically to be beings have been assayed for biotin but there was free of cirrhosis as well as of tumor, were used as only one from which ideal material could be obtained controls, and averaged 448o n>/ biotin per gm., a and for which reliable figures may be reported. This value almost identical with those obtained for the appeared as a large, firm, light yellowish-brown mass livers of normal nonpregnant adult Wistar rats on in the left lobe of the liver with several large metastases the standard laboratory ration of equal parts Purina throughout the rest of the organ. The tumor con- dog chow and Rockland mouse or rat diet, with tained very active tissue practically free of necrosis, water ad libitum. Embryo rat liver taken on the I7th while the unaffected liver appeared quite normal, to i8th day of gestation possessed only i28o iny' Of with no evidence of cirrhotic change. The normal biotin per gm. Here again, therefore, embryo and liver assayed 445o m 7 of biotin per gm., and in strik- tumor tissue were found to differ in their biotin con- ing contrast to this the carcinoma was found to content from that of the normal adult tissue, although tain only 746 lTb/ per gin. For purposes of comparison, in this instance the deviation was in the opposite values for human embryo organs obtained from a direction from that of the skin tmnors previously 489 fetus, indicated the biotin level of human discussed. embryo liver to be 4ooo my' per gin., which, unlike The close parallelism between the observed abnor- that for embryo rat livers, is only slightly below inalities in biotin level of embryo and tumor sug- adult values (Fig. i). gested that perhaps the difference in vitamin H con- Lung tumors.--Several strain A mice bearing the tent is related to changes normally occurring in all transplanted lung tumor F were generously supplied rapidly dividing cells rather than to any metabolic by Doctor Andervont. This carcinoma arose spon- peculiarity of those particular tissues. To clarify taneously and its source is thought to have been the this point, rats were partially hepatectomized, and the alveolar epithelium. If so, the normal lung tissue vigorously regenerating livers assayed for biotin 3, 7, selected as control should provide a fair, though ad- I4, and 2i days after the operation. Even during the mittedly imperfect, basis :for comparison. As indi- most active phase of regeneration, which occurred cated in Fig. i, normal adult mouse lung contained in from 3 days to one week, there was no significant an average of 2285 m 7 of biotin per gm., while the change in the biotin levels, I2 regenerating livers carcinomas were all much lower (~i77 my'). Samples averaging 478o my' per gm. (range 43oo to 5720). of mouse embryo lung obtained near term averaged Similar determinations were made on a number of ~33 o m 7 of biotin, a deviation from the adult value subcutaneously transplanted liver carcinomas of mice, which closely approximates that of the tumors. which were generously supplied from the National Connective tissue tumors.--Several types of trans- Cancer Institute by Dr. Howard B. Andervont. The plantable sarcomas were readily available in experi- growth was originally induced by subcutaneous in- mental animals, and as controls normal connective jections of butter yellow. It was surprising to find tissue scraped from beneath the epithelium of adult that in four 28-day tumors of moderate size (o.6 gm. mice was employed. This assayed i28o my of biotin or less) the biotin level was 3832 m 7 per gm. (range per gm., whereas the transplanted mouse sarcomas 313 o to 4216), practically identical with that of the 37 and i8o averaged 556 my, and 455 mT, respectively control livers (from the same animals), which con- (Fig. I). The biotin content of the corresponding tained 3563 my'. The biotin levels of these livers did embryo tissue proved to be 312 m 7 per gin., a find- not differ from those of normal animals of the same ing which again bears out the observation that tumor strain. On assaying the carcinoma at a later stage of and embryo appear to deviate in similar fashion from growth (over i gin.) it was found that while 6 normal the normal adult tissue. livers from tumor animals averaged 36i 3 nay' of Adult rat connective tissue contained an average of biotin per gm., the average biotin level of nine 34-day 246o nay of biotin per gin. While, as in the mouse, tumors was but ,I7o my' (range 555 to 213o ). For samples of connective tissue were taken from sub- comparative purposes, it may be stated that the biotin cutaneous areas apparently free from gross fat de- level of mouse embryo liver has been found to be posits it was nevertheless impossible to obtain the 2909 nay (Fig. I). tissue entirely devoid of fat cells. Fat is very low

Downloaded from cancerres.aacrjournals.org on October 1, 2021. © 1942 American Association for Cancer Research. West and Woglom--Biotin in Tumors and Embryo Tissues 327 in biotin, however (52 my per gm. for the rat) so microscopic verification of the tissues assayed. The that the small amounts present in the control would values for the tumors in each of these instances differed lower the biotin level somewhat from its true value. widely from the control determinations. Carcinomas The R39 rat sarcoma was found to contain 544 my of of the stomach, sigmoid, and rectum possessed an in- biotin per gin. and rat embryo connective tissue creased biotin content, I5oo, 787, and 444 my per gm., 1280 m 7 (Fig. i). while the respective control values were 833, 284, and Samples of two different transplantable rabbit sar- io2 m 7. In two cases, however, a reverse relationship comas, RS and Kato, were obtained through the was discovered, the biotin levels of carcinomas of courtesy of Doctor Kidd. These assayed 29o and stomach and cecum being 483 and ii4o my while 8i 3 m,/ biotin per gm., respectively. In contrast to corresponding control determinations were I7oo and the results with normal adult rat and mouse connective 2250 my per gin. (Fig. I). As one would expect, tissue, the same tissue in the rabbit assayed only the results of assays on isolated human tumors, each I34 m 7. Embryo rabbit connective tissue from fetuses with its own clinical and pathological peculiarities, do approximately 3 days before birth averaged 133o m 7 not present the uniform picture which is obtained in (Fig. i). The increased biotin content of the rabbit the case of a group of standard animals bearing tumors sarcomas over those of the control adult tissues dupli- of common origin. It is not surprising, in such a cates the findings for epithelial growths in this species, random selection of material from patients, that and the two are contrary to those generally found for tumors of both the biotin-rich and biotin-poor type rat and mouse tumors. In either case, however, the were encountered, for even in the normal tissues a corresponding embryo tissues consistently varied from wide divergence in biotin concentration was found in the normal biotin levels in parallel with the tun-mrs. different patients, probably because of their variable Lymphoid leukemia.--Because of the resemblances states of nutrition. In cases of accidental death in between the leukemias and neoplasia, the biotin levels healthy subjects, the biotin levels of the organs were of normal lymph nodes were compared with the en- remarkably constant. larged nodes of lymphoid leukemia. For this purpose As for benign growths of the human subject, a Dr. Jacob Furth, of Cornell University Medical Col- large lipoma in the posterior cervical region and 4 lege, kindly supplied us with a number of mice that uterine fibromyomas have been examined so far. None had been inoculated with his line AKh lO32. Several contained more biotin than the mother tissue. Thus groups of leukemic nodes and normal nodes from for the lipoma the value was 35 m7 as compared with uninoculated mice of the same and different strains 69 m 7 for normal adult fat. The fibroids, all of which were examined. The results were uniform, the normal were obtained at operation, gave such uniform figures nodes averaging 13o 7 m 7 (range 13oo to I33o), the that they may safely be regarded as characteristic of leukemic nodes i39o m 7 (range I31o to I4o5) biotin this lesion (Fig. I). The average was 67 m 7 and per gin., an insignificant difference. Thus these nodes that for the surrounding normal myometrium 7 ~ m 7 did not exhibit the variation found in most neoplasms, per gm. but as those of other lines might, no conclusions can The uterine fibroid is considered by some to be a yet be drawn. hyperplasia rather than a true neoplasm, and if this Miscellaneous tumors of human origin.--Nearly all view be correct a normal biotin content was perhaps human material obtained either at autopsy or at op- to have been expected since even actively regenerating eration proved to be unsatisfactory in one way or liver showed no increase, and the myometrium in the another. Many bone sarcomas were assayed for biotin fourth month of pregnancy also had a normal biotin value, but here the mass of intercellular material, even content. when inorganic constituents were eliminated, made comparisons of normal tissue and tumor extremely THE TRANSITION FROM EMBRYONIC TO ADULT BIOTIN di~cult to evaluate. Results on several bronchogenic LEVELS IN VARIOUS ORGANS OF THE RAT carcinomas of the lung, although the tumors were ideal In the preceding pages, reference has been made re- in themselves, were discarded because alveolar tissue peatedly to the close parallelism which seems to exist is really not an adequate control and attempts were between the biotin levels of embryonic tissues and not made to assay bronchial epithelium. Most carci- tumors. Certain embryo tissues of the rat; e.g., liver nomas of the gastrointestinal tract were highly un- and connective tissue, contain much less vitamin H satisfactory either because of necrosis or hemorrhage than the same adult tissues, while with other organs or excessive connective tissue in the tumor, or the the situation is just reversed. One immediately won- presence of too much connective tissue or muscle ders whether or not the transition from embryonic to in the sample of normal mucosa selected as a control. adult biotin levels is an abrupt one, whether all A few, however, did prove to be acceptable following organs are involved at once, and whether the time

at which this transformation takes place coincides biotin contents of the organs of a 4~-month human with any known changes in metabolism of the tissues fetus and the organs of normal adult persons autopsied concerned. after accidental death, present the same picture with Rat fetuses were removed at varying stages of de- only minor modifications. These changes in vitamin velopment, the age being estimated from the age- H activity might well reflect a fundamental alteration weight tables of Donaldson (4). Sufficient embryo in cellular metabolism occurring at the time of transb livers could be obtained as early as 5 days before tion from intra-uterine to extra-uterine life. Particu- birth, kidney at 31/2 days, and other organs at 11/2 days. larly interesting in this respect is the observation of Further samples of these organs were taken for assay Norris, Blanchard, and Povolny (io) and of Burk (r) at the time of birth and at intervals thereafter in that the high anaerobic glycolysis of embryo liver order to follow closely the changes in biotin activity falls to normal adult values very shortly after birth. over the period just before and after delivery. These This is correlated by them with the decrease in hema- data are summarized in Fig. 2. The rapid shift in the topoietic elements in the liver which takes place at this time. Since malignant change is said by some

MF" BIOTIN investigators to be accompanied by increased glycolysis, PER GM LiV[~ 350C both aerobic and anaerobic (,), and at the same time ~(iI) NEV MUSCLE ? the biotin levels of the tumors shift back toward those of the corresponding embryo tissues, a close relation- 300C I II ship between biotin activity and the unusual carbo~ hydrate metabolism of embryo and tumor appears to II be an interesting possibility. 2500

6ooo II II GROWTH OF TRANSPLANTED TUMORS IN BmTIN- DEPLETED MtCE iiI ',,. I1 Various attempts were made either to raise or lower

4000 1300 the biotin contents of the tissues of experimental animals with the idea of observing the course ot trans- x', I planted tumors under such altered conditions. It soon

I000 became apparent that mice are capable of tolerating

}'000 enormous doses of biotin extract without any change

9~ " - - --~d~ z 5-~ TABLE l: BIOTIN* CONTF.'gT @ OF ORGANS OF MICE TREATED --.| ...... AS S HOX.VN 5KIN "'" - " --e Injections of 4 3 B Z I I Z v~3 4. 5 6 biotin con- Egg white oars BEFORE tRrH BIRTH C'A AfTer, ,Rr. ADULT Stock diet eentrate diet Per cent ~a__~_ .~ _z~, de- Fro. 2.--The biotin levels of various rat organs before and Organ ~rotal Free ;Fotal Free Total Freer pletion after birth. Skin ..... 618 78 633 56 76 ~ 5 88 Muscle ... 905 65 IOlO 47 IO2 < 5 89 Lung ..... 1681 123 I42O 75 230 22 87 distribution of biotin among the representative organs Liver ..... 312o lO6 533 o 65 58o 12 82 shown is most striking. Beginning a day or so before Spleen .... 585 . . . 51o 9 " II4 ~ 5 8I birth, those organs characterized by a higher biotin Kidney ... 359 ~ 128 13,33o I00 lO75 9 7 ~ activity in the embryonic stage lose vitamin H very Brain ..... 2210 88 2000 54 845 25 63 quickly, and by the third post-natal day the normal * my biotin per gm. of tissue. t Free values in group on egg white diet obtained from a group adult values are reached. At exactly the same time of deficient animals other than those used in total biotin assays. that this precipitous decline in biotin activity is occurring in some organs, liver, which is character- in the free or total biotin content of the tissues. As istically low in the embryo as compared with the illustrated in Table I, mice which had received 2oo rat adult, shows a sudden rise, and also within 3 days units of SMA vitamin H concentrate per day (0.2 ml. after birth attains normal adult levels. As indicated intravenous, o.2 ml. intraperitoneal), did not show any in Fig. 2, the situation in the case of kidney is some- increase in free biotin whatever but even a slight what more complex but the same general trends are drop below normal in some cases. Total biotin values evident. also remained constant, the only notable exception As to the meaning of these drastic readjustments of being the kidney, where they reached more than three biotin value in the various organs, one can at present times the normal. Since such results were obtained by only hazard a guess. Comparisons made between the giving maximal doses of biotin concentrate, attempts

Downloaded from cancerres.aacrjournals.org on October 1, 2021. © 1942 American Association for Cancer Research. West and Woglom~Biotin in Tumors and Embryo Tissues 329 to raise the biotin content of the tissues had to be tissues of the host, which cannot function in the discontinued, although it appears possible that if pure absence of an adequate supply of biotin. It was at biotin were available sufficient could be administered first thought that a flourishing neoplasm might rob without toxic effects to accomplish this end. the host of its already depleted biotin store, but on The withdrawal of biotin from the tissues of mice analysis it was found that the tumors contained only proved to be much simpler. Feeding a diet rich in the same small fraction of their normal vitamin H the biotin inhibitor, avidin (5), produced early signs contents as the muscle, lung, and liver of the ani- of deficiency after 6 weeks, severe signs after i2 weeks, mal in which they were growing--in this case 2o and death usually at about the I6th week. For the to 24 per cent. last few weeks of life all animals appeared cachectic, Further evidence that these two tumors possess the most of the fur was lost, and the skin was thin, in- power of carrying on their normal metabolic processes flamed, and in places scaly. Most of the animals unimpaired by lack of a biotin supply may perhaps were blind, and many had developed a peculiar be furnished by in vitro studies. With the addition spastic gait. Others seemed to be in constant pain, as of avidin to a protein-free culture medium, in excess though the cutaneous sensory nerves were irritated, of the amount required to bind all of the available because they continually licked themselves. Often the biotin (as demonstrated by inability of the biotin- lightest touch with the tip of a pencil was sufficient requiring test organism, RMzobiurn, to grow in it), to induce convulsive seizures. The mice seemed to the cells of sarcomas 37 and i8o continued to multiply continue in about the same condition for the last few actively. It is to be emphasized, however, that this weeks of life, and then died suddenly. Histologically investigation, which later will be made the subject of the internal organs showed nothing remarkable; studies a separate paper, is mentioned only in passing. The on the central nervous system are in progress. growth of normal tissues in vitro in the absence of Data given in Table I indicate the extent to which biotin has not yet been examined, as of course it must the various tissues were depleted of biotin at the time be before any significance can be attached to the pre- of death. All the organs assayed except kidney and liminary observation just cited. brain were found to have lost 80 to 9 ~ per cent of their biotin activity. The retention of biotin by kidney DISCUSSION and brain is emphasized by other figures obtained at The possibility that regulatory substances present in an earlier stage of biotin depletion, at a time when cells in only catalytic amounts might be concerned in the other organs had already lost approximately 65 per the malignant process led to this investigation. Biotin, cent of their normal vitamin supply, while kidney or vitamin H, the most physiologically active vitamin and brain had not even begun to be drawn upon and known, was selected because it has been thought to be showed normal values. essential in the energy-yielding reactions of every As to the fate of tun-mrs transplanted into biotin- normal cell. deficient hosts, two examples will serve to illustrate. Results have shown that if carefully selected malig- When it was estimated that a very deficient lot of nant tissues be compared with their normal counter- animals had approximately 2 to 3 weeks to live, frag- parts, striking differences in the biotin levels are ments of the rapidly growing sarcomas 37 or I8o were found. For the most part, the tumors contained only introduced subcutaneously into these and into normal a fraction of that amount of biotin present in the control animals on the standard laboratory ration. normal controls. A few types, however, were defi- These tumors, it will be recalled, are of the type nitely higher in the vitamin. It is interesting to com- which are biotin-poor in contrast to the corresponding pare these findings with other recent investigations normal tissue. Contrary to expectation all transplants and to speculate upon the possible interrelationships. grew rapidly, producing large healthy growths even For example, the biotin level of liver carcinoma was though the ,nice during this time were in an extreme found to be far below that of normal or vigorously stage of biotin depletion. After 17 days the average regenerating liver, and Burk (I) has found the respira- weight of 8 sarcomas 37 from deficient animals was tion of such tumors also to be deficient, while re- 0.26 gm., the average of i i control tumors o.36. In generating liver remained unchanged. Greenstein and another group the average weight of 9 sarcomas 18o others (6, 7) reported that the arginase and thymonu- was o.99 gin. after i o days' growth in biotin-deficient cleodepolymerase activity in such tumors was reduced mice in comparison with I.O6 gm. for a corresponding while simple byperplastic liver did not differ from the set of controls. normal. In a recent paper (8) these investigators re- It appears therefore, that these two sarcomas, the port similar results concerning the xanthine dehydro- growth of which is essentially unimpaired by extreme genase, catalase, and amylase contents of hepatic tissues. biotin deficiency, differ fundamentally from the normal Furthermore, they too observed a striking parallelism

between the relative activity of fetal to adult liver almost, if not quite, independent of biotin. Should tissue and that of hepatic tumor to normal liver. this prove true of other tumors in various species it The classic respiration studies of Warburg (~ i) and would suggest that biotin is of little or no mort-rant in the more recent investigations of Burk and his as- the life of neoplastic tissues, possibly because of their sociates (i, 2) have shown that embryo possesses cer- altered carbohydrate metabolism, and that any shifts tain metabolic features in common with tumors. For observed in the level of this vitamin are a conse- this reason, the biotin level of embryo tissues cor- quence, and not a cause, of malignancy. Nor is other responding to the neoplasms selected was determined evidence lacking to support such a suggestion. For whenever possible, and here again, as with the tumors, example, the biotin content of the Shope papilloma, it was found that all differed widely from the normal already high, does not rise with the advent of the adult values. Furthermore, the deviation of both carcinomatous transformation. Furthermore, young tumor and embryo was closely parallel. In other transplanted hepatic carcinomas of the mouse have a words, where the value for an embryonic organ was biotin level similar to that of normal liver from this higher than that for its adult counterpart, the content species, and not until they have approximately doubled of the tumor was high in comparison with the adult in size do they show the low value that has been found mother tissues; and where the value for an embryonic in butter yellow carcinoma of the rat's liver. Yet at a organ was lower than that for the corresponding adult time when they weigh half a gram they are just as structure, the tumor contained less biotin than the truly carcinomas as they are when they have come to adult organ. weigh one gram. A further link between the respiratory studies of Burk (I) and the present observations is to be found SUMMARY in a comparison of changes occurring during the I. With but one exception all malignant tumors transition from embryo to adult. Immediately after studied, whether of animal or human origin, differed birth the biotin content of the various organs of the widely in their biotin content from the corresponding rat undergoes rapid change and reaches the normal normal tissues. adult values within 3 days. Likewise, the embryo 2. The biotin levels of embryo tissues also diverged type of glucose disassimilation shifts to that charac- widely from those of normal adult tissues, the major teristic of the adult very quickly after birth. shift from embryonic to adult biotin values occurring It also appears of significance that certain trans- in the rat within 3 days after birth. plantable sarcomas were found to be capable of 3. In all cases, the biotin content of both tumor abundant growth seen-tingly independent of a supply and embryo tissues deviated in the same direction of biotin, even though they apparently have no power from the corresponding normal adult levels. to synthesize the substance. During the last week of 4. Attempts to raise the general biotin content of life in animals dying from biotin deficiency these animals by injection of biotin concentrate were un- tumors were quite unimpaired by lack of this vitamin satisfactory because of rapid excretion of the excess although in its relative absence the normal host vitamin. tissues were unable to live. 5- Both tumors and normal tissues could be greatly Hitherto, the various organisms capable of growth depleted of biotin by maintaining the animals on a in the absence of biotin have been shown to possess ration containing avidin, and although the animals the power of synthesizing this substance themselves suffered severe deficiency symptoms as a result, the (9). So far as is known biotin is necessary to all living growth of certain transplanted tumors was unimpaired. tissues and if, as is the case with certain bacteria, Thus, while biotin is known to be an essential factor yeasts, fungi, and mammals, the ability to synthesize in the metabolism of normal mammalian tissues, these biotin has been lost, they become dependent upon an tumors, at least, are capable of maintaining their usual outside source for their existence. However, the two activity in its relative absence. sarcomas mentioned appear to fall in a different The authors gratefully acknowledge the able technical as- category, for it seems that they neither synthesize this sistance of Mrs. Ada-Mae Stein, without whose help the early vitamin, nor depend upon an extracellular source of completion of this work would not have been possible. supply for their well-being. Perhaps this is in some way connected with the fact that even in a normal REFERENCES animal these tumors are biotin-poor. I. BuRK,D. On the Specificity of Glycolysis in Malignant Liver The most interesting fact to emerge from the whole Tumors as Compared with Homologous Adult or Grow- investigation, and perhaps the most significant, is this ing Liver Tissues. Collected Reports of the University of Chicago. University of Wisconsin Symposium on Respira- demonstration that the cells of the two sarcomas so tion Enzymes and the Biological Action of the Vitamins, far examined, unlike any other cells yet known, are September, 1941. In press.

2. BURK, D., O. K. BEHRENS, and K. SUGIURA. Metabolism of Normal and Cancerous Hepatic and Mammary Tissues Butter Yellow Rat Liver Cancers. Cancer Research, I: and in Milk and Sera of Several Species. J. Nat. Cancer 733-734- 1941. Inst., I:845-863. 1941. 3. BURK, D., H. SPRINCE, E. A. KABAT, and J. FORTH. Metab- 8. GREENSTEIN, J. P., and W. V. JENRETTE. The Relative olism of Chicken Tumors and Leukoses. Cancer Research, Activity of Xanthine Dehydrogenase, Catalase, and Amy- I:732-733. 1941 . lase in Normal and Cancerous Hepatic Tissues of the 4. DONALDSON,H. H. The Rat. Memoirs of the Wistar Insti- Rat. I. Nat. Cancer Inst., ~:I7-22. 1941. tute of Anatomy and Biology. Philadelphia, 1924 . 9. LANDY, M., and D. M. DICKEN. Biotin Synthesis by Micro- 5. EAKIN, R. E., E. E. SNELL, and R. ]. WILLIAMS. A Con- organisms. Proc. Soc. Exper. Biol. & Med., 46:449-452. ,tituent of Egg White Capable of Inactivating Biotin in 1941. vitro. J. Biol. Chem., I36:8oI-8O2. 194 o. IO. NORRIS, J. L., J. BLANCHARn, and C. POVOLNY. Am. J. 6. GREENSTEIN, J. P., W. v. JENRETTE, G. B. MINER, and J. Path. Opie Dedication Number. In press. WroTE. Chemical Studies on the Components of Normal II. WARBUWG, O. The Metabolism of Tumors. Constable, and Ncoplastic Tissues. V. The Relative Arginasc Activ- London, 193o. ity of Certain Tumors and Normal Control Tissues. J. 12. W~sT, P. M., and P. W. WILSON. Biotin as a Stimulant for Nat. Cancer Inst., 1:687-7o6. I941 . Nodule Bacteria. Enzymologia, 8:152-162. 194o. 7. GREENSTEIN,J. P., and W. V. JENRETTE. The Depolymeriza- 13. WEST, P. M., and W. H. WOGLO~. The Biotin Content of tion of Thymonucleic Acid by an Enzyme System in Tumors and Other Tissues. Science, 93:525-527 . 1941.

Philip M. West and William H. Woglom

Cancer Res 1942;2:324-331.

Updated version Access the most recent version of this article at: http://cancerres.aacrjournals.org/content/2/5/324.citation

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

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

Permissions To request permission to re-use all or part of this article, use this link http://cancerres.aacrjournals.org/content/2/5/324.citation. Click on "Request Permissions" which will take you to the Copyright Clearance Center's (CCC) Rightslink site.