Gut, 1974, 15, 783-787

An investigation into the enzyme histochemistryof adenocarcinomas of human large intestine andof the transitional epithelium immediately adjacentto themJ. R. MARSDEN1 AND I. M. P. DAWSON

From the Department ofPathology, University ofNottingham

SUMMARY Histochemical enzymatic studies were performed on 30 freshly resected large bowelcarcinomas, 30 samples of normal colonic epithelium, and six samples of the histologically normalepithelium (so-called transitional epithelium) immediately adjacent to a carcinoma. Five enzymeswere studied: nicotine adenine dinucleotide tetrazolium reductase (NADH-TR), glucose-6-phosphatedehydrogenase, succinate dehydrogenase, monoamine oxidase, and acid phosphatase.

Quantitative and qualitative differences in enzyme activity were observed between normal,transitional, and carcinomatous mucosa as follows: monoamine oxidase activity was moderate innormal mucosa, high in transitional mucosa, and low in carcinoma. Succinate dehydrogenaseactivity was high in transitional mucosa and low or moderate in normal and carcinomatous mucosa.Glucose-6-phosphate dehydrogenase activity showed a gradation from low in normal mucosa tohigh in carcinoma while acid phosphatase showed the reverse of this pattern. The tetrazoliumreductase activity was low or moderate in normal and transitional mucosa and high in carcinoma.

These differences in enzyme activity and their possible clinical and metabolic significance arediscussed.

During the course ofan investigation into the enzymehistochemical patterns of normal colon, colonicadenocarcinoma, and carcinomatous metastases tomesenteric lymph nodes, the mucosa immediatelyadjacent to a carcinoma, which is morphologicallynormal, was observed to show characteristicenzymatic changes which are quantitatively and forsome enzymes qualitatively different from those inthe carcinoma and the more distant normal mucosa.Various workers (Wattenberg, 1959a and b; Nachlasand Hannibal, 1961; Mori, Sugimura, Matsumura,and Kawashima, 1963; Cohen, Elizalde, and Miller,1968; McGinty, Delides, and Harrison, 1973) havestudied enzyme patterns in colonic adenocarcinomabut only Filipe (1971) has described changes inenzyme patterns and alterations in mucosubstances(Filipe, 1969) in what she calls 'transitional' epithe-lium. We feel that the findings we describe may haveclinical significance in the assessment ofprecancerouslesions.'The investigations described in this paper formed part of a thesis forthe degree of B. Med. Sci. in the University of Nottingham.Received for publication 20 June 1974.

783

Materials and Methods

Thirty fresh surgically resected adenocarcinomas oflarge bowel, each of which had sufficient attachednormal colon to allow a study of normal mucosa at adistance of more than 10 cm from the carcinoma,were studied. In six of these a special study was madeof the histologically normal 'transitional epithe-lium' immediately adjacent to the carcinoma. Allblocks of tissue were snap frozen in liquid nitrogenimmediately following resection, and stored atminus 30°C for a maximum of two days.The enzymes demonstrated histochemically and

the techniques used were: acid phosphatase (Barkaand Anderson, 1962), monoamine oxidase (Glenner,Burtner, and Brown, 1957), succinate dehydrogenase,(Pearse, 1972), tetrazolium reductase (Pearse, 1972),and glucose-6-phosphate dehydrogenase (Pearse,1972).Frozen serial sections of unfixed tissue were cut

at a thickness of 12 ,u, and were mounted on cover-slips. Those for haematoxylin and eosin (H and E)staiing and for acid phosphatase demonstration

on June 2, 2021 by guest. Protected by copyright.

http://gut.bmj.com

/G

ut: first published as 10.1136/gut.15.10.783 on 1 October 1974. D

ownloaded from

http://gut.bmj.com/

J. R. Marsden and L. M. P. Dawson

Normal Trans

HighM.A.0. Moderate

Low

S. D.

G-6-P.D.

Acid P.

HM

L

HM

L

HM

HNADH-TR M

Fig 1 The relative intensity ofstaining of the finalreaction product ofenzymic activity in normal, transi-tional, and carcinomatous mucosa for the five enzymesstudied.

were postfixed in 10% formol calcium at 4°C. Twosections were used for the substrate control whenthis was necessary, and two were used for the testitself. The first two and the last two sections of eachtissue block were stained with haematoxylin andeosin and examined before commencing any enzymework to ensure that the sections contained therequired tissue and were correctly orientated. Theenzymes were quantitated visually according to theirdepth of staining, using an ordinary light microscopeat a magnification of x 20, and a quantifying scaleof high, moderate, or low. Mean values were calcula-ted for normal, transitional, and carcinomatousmucosa.

Results

Intracellular enzyme activity was localized to thebasal portion of the cytoplasm in normal absorptiveand goblet cells except in the case of glucose-6-phosphate dehydrogenase where the location was

both apical and basal. In transitional mucosa theintracellular location was essentially similar althoughsuccinate dehydrogenase and monoamine oxidaseactivity showed a more diffuse localization in threepatients. By contrast in carcinomatous epithelium allenzymes were either diffusely localized or in the apicalregion of the carcinoma cells.

Details of staining intensity in normal, transitional,and carcinomatous epithelium for each of the fiveenzymes studied are shown in histogram form in fig1 and examples are given pictorially for two enzymesin figures 2- and 3. No further description ,appearsnecessary.

Discussion

In this study we have equated enzyme activity withdepth of staining, which has been estimated on apurely visual basis. Visual assessment can be highlysubjective, and this equation is only justified whentechniques are standardized: one person does theassessing, sections are of uniform thickness, andconcentration of final reaction product is directlyrelated to enzyme concentration. More accuratequantitation, however, is difficult and time consum-ing and does not lend itself to clinical problemsrequiring a rapid solution. It is also inaccurate,though convenient, to talk about enzyme 'staining'.The staining represents a coloured insoluble reactionproduct which is the result of enzymic activity; onecannot stain an enzyme directly. We have notattempted to classify fine distinctions in stainingdensity but we are satisfied that our results are atleast reproducible and our three categories visuallyseparable.An increase in tetrazolium reductase activity in

adenocarcinoma as compared with normal colon hasbeen recorded by Wattenberg (1959a), Nachlas andHannibal (1961), and McGinty and colleagues (1973)as well as by ourselves. Filipe (1971) records uni-formly high levels in normal, transitional, andcarcinomatous mucosa while we found only moder-ate activity in normal and transitional mucosa.The findings of Mori et al (1963), Cohen et al

(1968), and Filipe (1971) of moderate activity ofglucose-6-phosphate dehydrogenase in adenocar-cinoma of the colon, with low activity in normalmucosa, agree with our findings, although McGintyet al (1973) found moderate activity in normalmucosa, with high activity in carcinoma. Filipe(1971) found high activity in transitional mucosawhereas our results showed only a slight increasecompared with normal.

Wattenberg (1959a and b) and McGinty et al(1973) found much less activityfor succinate dehydro-

784

on June 2, 2021 by guest. Protected by copyright.

http://gut.bmj.com

/G

ut: first published as 10.1136/gut.15.10.783 on 1 October 1974. D

ownloaded from

http://gut.bmj.com/

.::

~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~...

Fig 2 Technique for monoamine oxidase. Transitional mucosa containing a high concentration of enzyme as judged byfinal reaction product on the right, carcinomatous epithelium with low enzyme concentration on the left. x 30.

Fig 3 Technique for succinate dehydrogenase. Transitional mucosa showing a high concentration ofenzyme as judgedby final reaction product in the upper part of the field, carcinomatous epithelium with low enzyme concentration in thelower part. x 30.

on June 2, 2021 by guest. Protected by copyright.

http://gut.bmj.com

/G

ut: first published as 10.1136/gut.15.10.783 on 1 October 1974. D

ownloaded from

http://gut.bmj.com/

J. R. Marsden and I. M. P. Dawson

genase in carcinoma than in normal mucosa, whichwas found to show moderate activity. Czernobilskyand Tsou (1968) found a generally higher activity inadenocarcinoma. Filipe (1971) recorded low tomoderate activity in the normal mucosa, with aslight increase in transitional mucosa and moderateactivity in the tumour. These results do not agreewith our finding of low to moderate activity in thenormal and cancerous crypt cells, with markedactivity in transitional mucosa.Low to moderate monoamine oxidase activities in

normal crypt cells, similar to our results, wererecorded by Wattenberg (1959a). However, thisauthor also found a relative increase in activity inthe well differentiated carcinomas and a relativedecrease in the poorly differentiated ones. McGintyet al (1973) recorded strong activity in normal andmoderate activity in adenocarcinoma, qualitativelysimilar to our results of moderate activity in normaland low activity in the tumour. Filipe's work did notinclude monoamine oxidase, and therefore ourfinding of marked activity in transitional mucosa isimpossible to compare.Our observation that adenocarcinoma of colon

only has slight acid phosphatase activity when com-pared with the moderate activity found in normalcrypt cells corresponds well with the work of Filipe(1969) and of Czernobilsky and Tsou (1968).McGinty et al (1973) found marked acid phospha-tase activity in normal mucosa, with a marked reduc-tion in adenocarcinoma.The clinical significance of these findings lies in

the fact that histochemical enzyme techniques candemonstrate distinctive intracellular changes beforethese become observable histologically. Wattenberg(1959a and b), in his studies on proliferative lesionsof the large intestine, recorded qualitative andquantitative changes in succinate dehydrogenase andcytochrome oxidase in atypical glands found nearmalignant polyps and other benign proliferativelesions. The author thought that these changes mightbe an early indication of carcinogenesis. Further un-published work by Filipe (1969) has documented adisruption of intracellular enzyme localization forglucuose-6-phosphate dehydrogenase, tetrazoliumreductase, and succinate dehydrogenase in crypt cellsadjacent to tumours. Enzyme changes of this naturecould be responsible for the failure of sulphation ofmucopolysaccharides that Filipe noticed in tran-sitional mucosa (Filipe, 1969).We feel that these enzyme changes are likely to be

consequent on, rather than the cause of, carcino-genesis. Their basis may lie in an intracellular meta-bolic and/or genetic change in enzyme control, maydepend on extracellular environmental factors, ormay merely represent a normal cellular adaptive

response. The high levels of succinate dehydro-genase exemplify this last idea. A reduction in bloodsupply due to mechanical compression by a tumourmass, or due to changes in connective tissue andblood vessel wall mucopolysaccharides, as describednear tumours by Majewski, Tkaczyk, and Majewski(1966), would decrease the oxygen and substratesupply to the transitional mucosa. A compensatoryenzyme 'hyperplasia' of the Kreb's cycle enzymesmight enable the cell to use at least as much sub-strate or oxygen as it received. The fact that thelevels of these enzymes subsequently fall to thoselevels found in adenocarcinoma of the colon can beexplained on the basis of the cells having acquiredalternative pathways of metabolism more suited totheir requirements than a long-term massive increasein enzyme protein.The increase in succinate dehydrogenase implies

relatively normal functioning of the electron trans-port chain, an implication consistent with the findingof identical levels of activity for tetrazolium reductasein both normal and transitional mucosa. The cells inthe carcinomas are thought, on the basis of theirhigh tetrazolium reductase levels, to be dependent ondiaphorase activity as a 'sink' for metabolic hydro-gen that would normally go through the electrontransport chain to oxygen. The cells of the tran-sitional mucosa appear not to have developeddependence on the diaphorases in this way. Theelevation in glucose-6-phosphate dehydrogenasecould be explained on the basis of the cell requiringmore riboses for nucleic acid synthesis if it is begin-ning to develop the increased mitotic capacity of acancer cell.The changes observed for monoamine oxidase and

acid phosphatase in transitional mucosa are difficultto interpret metabolically. Monoamine oxidase andsuccinate dehydrogenase, the two enzymes whichshow a marked rise, are also the two enzymes forwhich a change of localization in transitional cellswas observed.

Certain conclusions can be drawn from this studyThere are well marked quantitativedifferences, in thefive enzymes tested, between normal and carcinoma-tous colon. These differences are variable in theirextent. Qualitative changes are limited to smalldifferences in localization of the enzyme betweennormal and carcinomatous mucosa. All the enzymes,except tetrazolium reductase, show differences of'staining' intensity in transitional mucosa from thepattern obtained for normal or cancerous mucosa.This could be a manifestation of developing cancercell metabolism or of normal cellular adaptation to achanged environment. We feel that a combination ofenzyme techniques and techniques for mucosub-stances as recommended by Filipe (1969) could be of

786

on June 2, 2021 by guest. Protected by copyright.

http://gut.bmj.com

/G

ut: first published as 10.1136/gut.15.10.783 on 1 October 1974. D

ownloaded from

http://gut.bmj.com/

An investigation into the enzyme histochemistry ofadenocarcinomas ofhuman large intestine 787

value in detecting malignant transformation inbenign lesions, or in detecting precancerous changein histologically normal mucosa in a preselected,vulnerable population. Their use should be exploredin familial adenomatosis, chronic ulcerative colitis,and histologically benign adenomas and papillomasof large intestine.

Our thanks are due to those surgeons in the GeneralHospital in Nottingham and especially Mr J.Burke who allowed us to use material from theircases, to Dr Peter James who helped us with theon-the-spot collection and snap freezing of material,to Mr W. Brackenbury for the photographs, andMiss Colleen Peel for secretarial assistance. Part ofthe costs of the investigation were defrayed by agrant from the Cancer Research Campaign.

ReferencesBarka, T., and Anderson, P. J. (1962). Histochemical methods for acid

phosphatase using hexazonium pararosanilin as coupler. J.Histochem. Cytochem., 10, 741-753.

Cohen, H. J., Elizalde, A., and Miller, S. P. (1968). Cytologic studiesof glucose-6-phosphate dehydrogenase in malignancy. Cancer(Philad.), 21, 1055-1060.

Czernobilsky, B., and Tsou, K. C. (1968). Adenocarcinoma, adeno-mas, and polyps of colon: the histochemical study. Cancer(Philad.), 21, 165-177.

Filipe, M. I. (1969). Value of histochemical reactions for muco-substances in the diagnosis of certain pathological conditions ofthe colon and rectum. Gut, 10, 577-586.

Filipe, M. I. (1971). Some aspects of the histochemical enzymedistribution in colon and rectum: normal and pathological.Acta histochem., Suppl. 9, pp. 333-337.

Glenner, G. C., Burtner, H. J., and Brown, G. W., Jr. (1957). Thehistochemical demonstration of monoamine oxidase activityby tetrazolium salts. J. Histochem. Cytochem., 5, 591-600.

Majewski, C., Tkaczyk, J., and Majewski, W. (1966). Acid muco-polysaccharide activity in cancer of the colon and rectum.(Polish). Pat. pol., 17, 313-319. Also Abstract in ExcerptaMed. (Amst.), Sect. V, 20, no. 1491.

McGinty, F., Delides, G., and Harrison, D. (1973). The significanceof enzyme histochemical patterns in carcinomas of the largeintestine in man. Gut, 14, 502-505.

Mori, M., Sugimura, M., Matsumura, T., and Kawashima, H. (1963).Histochemical study on the localisation of glucose-6-phosphatedehydrogenase in human tumors. Gann, 54, 433-442.

Nachlas, M. M., and Hannibal, M. J. (1961). Histochemical observa-tions on the polyp-carcinoma sequence. Surg. Gynec. Obstet.,112, 534-542

Pearse, A. G. E. (1972). Histochemistry, Theoretical and Applied,Vol. 2. Churchill, London.

Wattenberg, L. W. (1959a). A histochemical study of five oxidativeenzymes in carcinoma of the large intestine in man. Amer. J.Path., 35, 113-137.

Wattenberg, L. W. (1959b). A thistochemical study of succinic de-hydrogenase and cytochrome oxidase in proliferative lesions ofthe large intestine. Cancer Res., 19, 1118-1123.

on June 2, 2021 by guest. Protected by copyright.

http://gut.bmj.com

/G

ut: first published as 10.1136/gut.15.10.783 on 1 October 1974. D

ownloaded from

http://gut.bmj.com/