[CANCER RESEARCH 50, 5112-5118, August 15, 1990]

Quantitative Cytochemical Detection of Malignant and Potentially Malignant Cellsin the Colon1

Alistair James Best,2 Pranab K. Das,3 Hitendra R. H. Patel,2 and Cornelis J. F. Van Noorden4

Laboratory of Cell Biology and Histology, University of Amsterdam, Academic Medical Centre, Meibergdreef 15, 1105 AZ Amsterdam, The Netherlands

ABSTRACT

It was found to be possible to distinguish malignant cells from normalcells by using an oxygen-sensitive tetrazolium salt (neotetrazolium) forthe histochemical demonstration of glucose-6-phosphate dehydrogenaseactivity in cryostat sections of human colon. We have studied 12 cases ofestablished adenocarcinoma of the colon in addition to 4 of ulcerativecolitis and 4 of adenomatous polyposis (polyposis coli). In a nitrogenatmosphere the activities of malignant and normal cells were similar.However, after incubation in an atmosphere of pure oxygen, only malignant cells gave a positive reaction after 5 min. Three of the four cases ofadenomatous polyposis gave a positive reaction for glucose-6-phosphatedehydrogenase activity in oxygen in a manner similar to that of specimenswith severe dysplasia. In general, positive foci were histologically indistinguishable from the neighboring tubuli. However, foci of severelydysplastic epithelium usually showed a positive reaction. All three patients eventually developed clear-cut severe dysplasia. The other patient,who showed a negative reaction in oxygen, was diagnosed after 3 yearsas not suffering from dysplasia. All cases of ulcerative colitis gave areaction in oxygen comparable with that of normal cells. Therefore, theareas with a positive reaction are considered to be either in the processof malignant transformation or malignant. An explanation for the oxygeninsensitivity of cancer cells appeared to be a decrease in the activity ofSuperoxide dismutase (EC 1.15.1.1), as addition of exogenous Superoxidedismutase to malignant cells caused a normal reaction. We wish to suggestthat this test in combination with the routine histology may be exploitedfor the diagnosis of polyps in premalignant conditions.

INTRODUCTION

Because the process of carcinogenesis involves a stepwiseprogression, which often takes over 20 years in humans (1), theprospect of detecting at an early stage those cells that willprogress to malignancy is important. Initiated cells often formstructures considered to be precancerous, such as polyps orpapillomas, some of which will eventually progress to malignancy (2). Therefore a simple test that is able to distinguishsuch cells at an early stage would be a valuable tool in diagnosis.However, morphological changes often follow the events whichcause cells to progress to malignancy (3); thus by the time a cellcan be distinguished as malignant by morphological criteria itscancerous nature is already well established.

Changes in the activities of certain enzymes occur at earlystages of malignancy (4) and hence are detectable before themorphological changes. One such enzyme is glucose-6-phos-phate dehydrogenase (EC 1.1.1.49), the key regulatory enzymeof the pentose shunt pathway that produces both NADPH andthe precursors of nucleic acid synthesis (5) which are needed in

Received 10/18/89; revised 3/28/90.The costs of publication of this article were defrayed in part by the payment

of page charges. This article must therefore be hereby marked advertisement inaccordance with 18 U.S.C. Section 1734 solely to indicate this fact.

1This study has been performed in the Laboratory of Cell Biology andHistology, University of Amsterdam in conjunction with ERASMUS programICP-89-UK-0172.

2 Department of Anatomy and Physiology, University of Dundee, Dundee.

DD24HN, Scotland, United Kingdom.3 Department of Pathology, Academic Medical Centre, Meibergdreef 15. 1105

AZ Amsterdam. The Netherlands.4 Laboratory of Cell Biology and Histology, Academic Medical Centre, Mei

bergdreef 15, 1105 AZ Amsterdam, The Netherlands. To whom requests forreprints should be addressed.

large amounts by malignant cells.Demonstration of this enzyme is possible in tissue sections

by using a tetrazolium salt coupled to an intermediate electroncarrier. Production of an insoluble colored formazan marks theareas of enzyme activity (6). Oxygen present in the immediatereaction area is able to interfere with the production of formazan if either low concentrations of the tetrazolium salts (

CYTOCHEMICAL DETECTION OF MALIGNANCY IN THE COLON

temperature was —¿�25°C;the sections were placed on clean glass slides

and stored in the cryostat cabinet until use (1 to 2 h). Serial sections tothose used for enzyme histochemical studies were also stained withhematoxylin and eosin for reference to all formalin-fixed tissue.

The incubation media for glucose-6-phosphate dehydrogenase consisted of: 50 mM glycyl glycine buffer (BDH Chemicals, Poole, Dorset,United Kingdom) (pH 8.0) or 100 mM phosphate buffer (pH 7.45)containing 20% (20 g/100 ml) polyvinyl alcohol (hot water soluble;average molecular weight 40,000; Sigma Chemical Co., St. Louis, MO)(15), 10 mM glucose 6-phosphate (Boehringer, Mannheim, FederalRepublic of Germany), 0.8 mM NADP (Boerhinger), 4.5 mM neotetra-zolium chloride (Polysciences, Northampton, United Kingdom, orSigma; purified with chloroform) and 0.67 mM thionine (Merck,Darmstadt, Federal Republic of Germany) as an alternative electroncarrier to 1-methoxyPMS (16). The incubation medium for the demonstration of NADPH dehydrogenase was identical but 0.67 mM men-adione (sodium salt; Sigma) replaced thionine as the intermediateelectron carrier (16, 17). In the course of the investigation the followingcompounds were added to the basic incubation media either separatelyor in combination: 1 mM dicumarol (Sigma); SOD (Sigma; from bovineliver, 4300 lU/mg; 0.3 mg/ml incubation medium); catalase (Sigma;from bovine liver, 1600 lU/mg; 0.1 mg/ml incubation medium); 10mM sodium azide (Merck).

The reaction was performed in a room maintained at 37°Con the

stage of a Vickers M85a scanning and integrating cytophotometer toobtain kinetic measurements (14). Polyvinyl alcohol-containing buffer(5 ml) was placed in a tonometer (also at 37°C),an instrument used to

equilibrate a liquid with a gas without forming bubbles (18), and wassaturated with the appropriate gas (either nitrogen or oxygen) at a flowrate of 500-800 ml/min for a minimum of 10 min. Other constituentsof the basic incubation medium were then added. This ensured thoroughmixing of the incubation medium in addition to complete saturationwith either nitrogen or oxygen.

The area to be measured was selected and a drop of incubationmedium on a cover slip placed over the section, any readjustmentsneeded were then made. The first measurement was taken at 30 s (timezero; the value measured also was readjusted to zero). Measurementswere then taken every 30 s thereafter for periods of up to 15 min. Allmeasurements were made at 585 nm, the isosbestic wavelength for theformazans of neotetrazolium (19). A Leitz NPL x25 objective lens(numeric aperture 0.50) was used for the measurements. The mask sizeused was Al (10 ¿imdiameter), giving an effective scanning area of78.5 Mm2.A spot diameter of 0.8 Mmwas selected, with a band width

setting of 65. Usually at least three kinetic measurements were takenfrom each tissue sample in serial sections, from areas of both high andlow activity. The same areas were then measured in further serialsections using an incubation medium saturated with the alternative gas.All integrated absorbance values were converted to mean integratedabsorbance (19) and then to nmol of NADPH formed/mm3 of tissue

by the method of Van Noorden and Butcher (20). The mean valuescalculated from the three kinetic reactions performed on each tissuesample were then plotted and the rate of reaction was calculated usingthe linear part of the graph, usually at 5 min. The calculation of residualactivities was achieved by comparing the actual mean activities obtainedafter a time of 5 min (ratio of the activity in oxygen and in nitrogen)rather than the rates; this was found to give more accurate results.

RESULTS

The histochemical method for the detection of glucose-6-phosphate dehydrogenase activity was specific, inasmuch as thecontrol reaction performed by the omission of the coenzymeand substrate from the incubation media (21) reduced theproduction of formazan to unmeasurable levels.

The use of phosphate buffer resulted in a lag phase of 4 minon the average in normal tissue when the incubation was performed in oxygen. On the other hand, glycyl glycine buffer gavea lower rate of reaction and a longer lag phase in oxygen innormal tissue (approximately 8 min). Therefore it was a distinct

advantage in showing the differences between malignant andnormal cells to use glycyl glycine buffer. In human colonietissue formazan formation was selectively prevented in cancercells in oxygen for approximately 1.5 ±1.1 min (SD) (Table 1)and in normal cells in oxygen for approximately 7.5 ±3.4 min(Table 2; Fig. 1). Of even greater importance was the purity ofthe neotetrazolium used. Neotetrazolium from Sigma was heavily contaminated with impurities (22) that caused formazanformation in normal tissue in oxygen, hence making it impossible to locate areas of malignant cells. Sigma neotetrazoliumtherefore had to be purified with chloroform prior to use (22).The batches obtained from Polysciences were found to be pureenough to be used without chloroform treatment and were usedfor all studies thereafter.

Carcinoma Cells. Carcinoma cells in human colon displayedvarying activities of glucose-6-phosphate dehydrogenase (Table

1). The lag phase before any formazan formation was observedin carcinoma cells was typically below 1 min when the reactionwas performed in nitrogen (mean, 0.4 min). In oxygen the lagwas slightly increased to 1 or 2 min (mean, 1.5 min).

The rate of formazan formation was not seriously affected bythe presence of oxygen in the majority of cases (Table 1; Fig.2a). The residual activity observed in malignant cells was between 30 and over 100% ofthat in nitrogen (mean, 59%; Table1; Fig. 1) after 5 min incubation. No significant difference wasobserved between the reaction rates in malignant cells in nitrogen compared to oxygen. In nitrogen, a linear response wasobserved up to 7 to 11 min, in oxygen the linear response wasusually for a longer period of time (up to 15 min of incubation).No relationship was observed between the degree of differentiation of malignant cells and the lag phase or rate of reaction.

Normal Cells. In nitrogen, the lag phase exhibited by normalcells was below 3 min (mean, 1.0 min), in oxygen this wasextended over a range with a mean value of 7.8 min (Table 2;Fig. 2b). The difference between the lag phase for normal cellsin nitrogen and in oxygen was significant (P < 0.001). Anyreaction observed in normal cells before 5 min had elapsed wasalways lower than 10% when compared to the reaction innitrogen (mean, 1.2%; P < 0.001 compared to malignant cells;c.f. Tables 1 and 2). Hence normal cells could still easily bedistinguished from malignant cells by comparison of the reaction after 5 min in oxygen with that obtained in nitrogen (Fig.1). The reaction rate at 5 min was always considerably lower inoxygen than nitrogen (P < 0.001; Table 2; Fig. 2b).

Adenomatous Polyps. All cases of adenomatous polyposiswere classified independently in serial sections stained withhematoxylin and eosin and were of varying grades. Resultsobtained with the glucose-6-phosphate dehydrogenase reactionare shown in Table 3 and Fig. 3. Those areas classified asnormal colonie tissue all showed no formazan formation inoxygen (negative reaction). Some areas of cells showing anegative reaction were classified by the pathologist as dysplas-tic, whereas other areas showed a high activity in oxygen(positive reaction; Table 3). These latter areas were classifiedby the pathologist as severely dysplastic but were morphologically difficult to discriminate from the areas of dysplasia showing a negative reaction on frozen sections (Fig. 3). On the otherhand, grading of dysplasia is often subjective when using hematoxylin and eosin-stained sections of formalin-fixed tissue.

Positive reactions were typically seen either in areas composedof many glands showing a high reaction and separated fromsurrounding structures by connective tissue, or in singularglands surrounded by areas displaying very low reaction or no

5113

on June 26, 2021. © 1990 American Association for Cancer Research. cancerres.aacrjournals.org Downloaded from

http://cancerres.aacrjournals.org/

CYTOCHEMICAL DETECTION OF MALIGNANCY IN THE COLON

Table l Glucose-6-phosphale dehydrogenase activity in malignant epithelial cells in cases of adenocarcinoma of the colonThe incubation was performed with neotetrazolium in nitrogen (N2) and oxygen (O2) at 37°Cfor 15 min. Residual activity is the ratio of mean activity from three

kinetic reactions in N2 and in ( )- after 5 min.

Rate at 5 min (fromlinear reaction(MMAgePatient

(yr)1

852753534605716627638759

831071118412

62Lag

period(min)SexFemaleMaleFemaleFemaleMaleMaleFemaleMaleFemaleFemaleMaleMaleN200000000.511110,0.50.50.50.50.51.520.522.533.5mm3N29.033.011.114.027.112.08.011.09.35.56.35.8min')0211.330.212.37.218.67.06.210.05.44.86.05.0Residualactivity

(%)1127872675133438757493029

Mean valuesLag phase

In N2: 0.4 ±0.5 minIn O2: 1.5 ±1.1 min

Reaction rateIn N2: 12.7 ±8.6 nM mm"3 min~'In O2: 10.3 ±7.4 nM mm"3 min"'

Residual activity 59.0 ±25.3%

Table 2 Glucose-6-phosphate dehydrogenase activity of normal colonie epithelial cells

Experimental conditions as for Table

Rate at 5 min (fromlinear reaction(nMAgePatient

(yr)1

852753534605716627638759

831071118412

62Lag

period(min)SexFemaleMaleFemaleFemaleMaleMaleFemaleMaleFemaleFemaleMaleMaleN2110.51.52000.5111.52.5024.594111056.53.57.58159mm' min')N28.311.78.93.74.58.07.75.56.86.76.35.3020.700.600011.50000Residualactivity

(%)4.80100018.30000

Mean valuesLag phase

In N¡:1.0±0.8 minIn O2: 7.8 ±3.4 min

Reaction rateIn N2: 6.9 ±2.2 nM mm"3 min"In O3: 0.2 ±0.5 nM mm"3 min"

Residual activity: 1.2 ±2.6%

reaction at all in oxygen (Fig. 3¿>).In nitrogen these activitieswere very similar.

Cases of Ulcerative Colitis. Table 3 also shows the meanresidual activity in epithelial cells affected by ulcerative colitis.None of the cells exhibited any morphological signs of malignancy and reacted in a manner which would classify them asnonmalignant according to the present oxygen sensitivity test.No reaction was seen in the connective tissue or, more importantly, in the cellular infiltrate if thionine was used in theincubation medium. On the other hand, when 1-methoxyPMSwas used instead of thionine, a very high reaction was found ininflammatory cells in oxygen and therefore could not be usedto indicate areas of malignancy.

Elucidation of the Mechanism of the Oxygen InsensitivityPhenomenon. By replacing 1-methoxyPMS or thionine in theincubation media with menadione, the activity of the enzymeNADPH dehydrogenase is revealed (Fig. 4a). The reaction wasspecific inasmuch as addition of dicumarol, a specific inhibitor

of this enzyme (23), reduced formazan production to low levels.The effect of oxygen upon the residual activity of this enzymein malignant cells was far greater than the corresponding reaction for glucose-6-phosphate dehydrogenase activity (14% com

pared with 31%). Addition of dicumarol to the incubationmedium for glucose-6-phosphate dehydrogenase resulted in a

small but significant increase in the reaction (Fig. 4a).The effect of exogenous SOD and catalase upon the residual

activity of malignant cells in oxygen is shown in Fig. 4b. Adecrease from 31% activity to 5% activity was noted; i.e., themalignant cells showed the characteristic residual activity displayed by normal cells in oxygen when exogenous SOD wasadded to the medium. Addition of azide, an inhibitor of catalase(12), increased the reaction in normal cells in oxygen (notshown). The reaction was dependent upon an intermediatecarrier in the medium as it was reduced to low levels when 1-

methoxyPMS or thionine was omitted (Fig. 5).5114

on June 26, 2021. © 1990 American Association for Cancer Research. cancerres.aacrjournals.org Downloaded from

http://cancerres.aacrjournals.org/

CYTOCHEMICAL DETECTION OF MALIGNANCY IN THE COLON

Fig. 1. Serial cryostat sections (8 /an thick) of adenocarcinoma of the colon,a, hematoxylin and eosin staining of normal (A/) and malignant (CA) tissue; b,glucose-6-phosphate dehydrogenase activity after 5 min of incubation usingneotetrazolium in an atmosphere of nitrogen. Formazan is deposited in bothnormal (N) and malignant (CA) tissue; c, as for b, but reaction was performed inan atmosphere of oxygen, showing formazan deposition only in malignant cells(CA).Bar, lOO^m.

150-

IOD

50-

IO 15

time

CYTOCHEMICAL DETECTION OF MALIGNANCY IN THE COLON

4*'^

^ :•>.C;••¿�*

Fig. 3. Serial cryostat sections (8 ¿2Lc20-ËaT3

CYTOCHEMICAL DETECTION OF MALIGNANCY IN THE COLON

15-

E

I10-

«PMS

•¿�oIE5Õ

5-o.OzsCn»PMS-PMS-PMS1

1

t umour/Anormal '

Fig. 5. Glucose-6-phosphate dehydrogenase activity (4- PMS) and the proportion of NADPH utilized in cellular detoxification reactions (—PMS; type Ireaction) in samples of adenocarcinoma of the colon (tumor) and normal colonieepithelial cells (normal). The data are expressed as nin NADPH formed/mm3 of

tissue/min. Type I activity in malignant tissue is not increased in proportion tototal glucose 6-phosphate activity.

oxygen upon the formation of formazan from triphenyl tetra-zolium has already been documented (26, 27). For nitrobluetetrazolium (NBT), which has a lower electron potential, areversible reaction occurs with atmospheric oxygen, again withthe production of oxygen radicals (Reaction B) (7, 11, 28).Neotetrazolium does not show the same reversible reaction,increasing the concentration of neotetrazolium in the incubation medium does not decrease the effect of oxygen (28, 29).Only when all the oxygen has been removed will the reactionproceed ( 18).

With respect to the initial formation of formazan exhibitedby carcinoma cells in an atmosphere of oxygen (the oxygeninsensitivity phenomenon), a factor is responsible for the increased reaction in these malignant cells. A comparison ofTables 1 and 2 shows that carcinoma cells may display a rateof reaction similar to that of normal cells in an atmosphere ofnitrogen but exhibit a significantly higher rate of reaction thannormal cells in oxygen (P < 0.001). Without the addition of 1-methoxyPMS to the incubation medium the proportion ofNADPH utilized for detoxification reactions is revealed (6, 29).These reactions involve the enzymes NADPH cytochrome P-450 reducÃ-ase and NADPH dehydrogenase. Since NADPHcytochrome P-450 reducÃ-asehas been indicated as the factorresponsible for the oxygen insensitivity phenomenon (30) theactivity of this pathway was investigated. Fig. 5 shows that thisenzyme accounts for only a small proportion of the overallreaction in both normal and malignant cells (reaction in theabsence of PMS). Moreover it is not known as a transformation-linked enzyme (31) and has also been shown to be oxygensensitive (32). We therefore conclude that this enzyme is notinvolved in the oxygen insensitivity phenomenon in malignantcells. On the other hand NADPH dehydrogenase is transformation linked (17) but Fig. 4 shows that the reaction for thisenzyme is oxygen insensitive only to a limited extent in malignant cells but not to the extent that the reaction for glucose-6-phosphate dehydrogenase activity appears to be. Furthermorethe reaction for glucose-6-phosphate dehydrogenase is not di-cumarol sensitive. The slightly increased reaction is probably

due to the inhibition of endogenous electron transport pathways. This evidence suggests that oxygen insensitivity is notmediated by NADPH dehydrogenase but occurs at the level ofthe neotetrazolium radical. This is also indicated by the factthat oxygen sensitivity in normal cells has been observed whilestudying succinate dehydrogenase activity in normal rat liver(18), suggesting that oxygen insensitivity in malignant cells isindependent of NADPH. Within certain limitations the enzymeused to demonstrate oxygen insensitivity is in itself unimportant.

SOD is involved directly in the inhibition of formazan production, inasmuch as its addition to the incubation mediumcaused a decreased formation of formazan in malignant cells inoxygen (Fig. 4b). It is known that the activity of the manganese-SOD is absent in carcinoma cells, whereas the activity ofcopper/zinc-SOD is sometimes lowered (12, 33). The effect ofSOD on neotetrazolium-formazan formation cannot be explained in the same way as for nitroblue tetrazolium, where theequilibrium reaction (Reaction B) is affected. Increasing theconcentration of neotetrazolium did not affect the oxygen insensitivity of normal cells and therefore Reaction A is not anequilibrium. Thus it is not the removal of oxygen radicals thatis important here but the reformation of molecular oxygen inthe reaction area. This reformed oxygen can then react repeatedly with neotetrazolium radicals. In the normal cell, the dis-mutation of oxygen radicals is extremely rapid due to thepresence of SOD. This accelerates the normal dismutationreaction by a factor of xlO4 at physiological pH (7.45) (12). In

malignant cells the dismutation reaction is much slower due todecreased SOD activity and is in competition with other reactions undergone by the Superoxide radical that do not producemolecular oxygen (12, 34). Thus in a malignant cell oxygen isremoved from the reaction area within approximately 1 min,depending on the residual levels of SOD present. Formazanformation then occurs at a rate similar to that seen in nitrogen(Table 1).

Catalase (CAT) is able to enhance the reaction of SOD intwo ways: (a) under the influence of catatase three-fourths ofthe Superoxide radicals are converted into molecular oxygeninstead of only one-half by SOD alone (Reactions C and D);(b) catatase removes hydrogen peroxide formed by SOD; it isknown that hydrogen peroxide inhibits copper/zinc-SOD (35).

2Or + 2H+ H2O2 + 02

CAT2H2O2 »2H2O + O2

(C)

(D)

Addition of azide to the incubation media reduces the lag phasein normal cells by inhibiting catalase, thereby affecting theefficient functioning of SOD. Catalase and SOD have beenshown to be mutually supportive to each other (36). Once theinitial oxygen has been removed the reaction proceeds unhindered with the formation of formazan as shown in Reaction E.

NT—FT + NT—H' -> NT + NT—H2(formazan) (E)

If oxygen is recycled in the manner described here, a relativelysmall proportion of oxygen can exert an effect much greaterthan the absolute amount alone could account for. This hasbeen documented with regard to fatty acid oxidation, the exaggerated effect of oxygen being explained in a manner similarto that proposed here (37).

In conclusion it is stated that the oxygen insensitivity ofmalignant cells is caused by a decrease in cellular activity of

5117

on June 26, 2021. © 1990 American Association for Cancer Research. cancerres.aacrjournals.org Downloaded from

http://cancerres.aacrjournals.org/

CYTOCHEMICAL DETECTION OF MALIGNANCY IN THE COLON

SOD. This prevents the recycling of molecular oxygen, thusallowing the deposition of formazan within the first 5 min ofincubation. Because levels of SOD are decreased in the earlystages of malignancy and remain depressed (13), the detectionof cancerous cells before morphological parameters becomeapparent is a possibility as was shown with the four cases ofcolonie polyps studied here. Therefore, the oxygen insensitivitytest could be a useful additional tool for grading of dysplasia.

The present study is based on cytophotometric analysis ofthe reactions to firmly establish the validity of the test. However, cytophotometry is not an essential tool for the test andmicroscopic inspection of serial sections after 5 min of incubation in nitrogen and oxygen is sufficient to detect malignantor potentially malignant cells. False positive reactions have notbeen found thus far in the present study when thionine wasused as exogenous electron carrier. Inflammatory cells showedsome positive reaction in the presence of oxygen when thioninewas replaced by 1-methoxyPMS. Other studies on the basis ofthe oxygen insensitivity phenomenon (3, 8, 9, 10, 30) did notreport false positive findings either.

ACKNOWLEDGMENTS

The authors would like to express their gratitude to Dr. NormanWalfoord for his help in diagnosis of specimens. They would also liketo thank the Pathology Department, Academical Medical Centre, forsupply of the specimens. Thanks also to J. Peeterse for photographicwork and to Professors J. James and P. J. Stoward for making thiswork possible.

REFERENCES

1. Farber, E. Pre-cancerous steps in carcinogenesis their physiological adaptivenature. Biochim. Biophys. Acta, 738: 171-180, 1984.

2. Farber, E. Chemical carcinogenesis: a current biological perspective. Carcinogenesis (Lond.), 5: 1-5, 1984.

3. Heyden, G. Histochemical investigation of malignant cells. Histochemistry,59:327-334, 1974.

4. Bannasch, P., Moore, M. A.. Klimek, F., and Zerban, H. Biological markersof preneoplastic foci and neoplastic nodules in rodent liver. Toxicol. Pathol-,10: 19-34, 1982.

5. Evans, A. W.. Johnson, N. W., and Butcher. R. G. A quantitative cytochem-ical study of three oxidative enzymes during experimental oral carcinogenesisin the hamster. Br. J. Oral Surg., IS: 3-16, 1980.

6. Van Noorden, C. J. F. Histochemistry and cytochemistry' of glucose-6-phosphate dehydrogenase. Prog. Histochem. Cytochem., /5:1-84, 1984.

7. Van Noorden, C. J. F., and Butcher, R. G. The involvement of Superoxideanions in the nitro blue tetrazolium chloride reduction mediated by NADHand phenazine methosulphate. Anal. Biochem., 176: 170-174, 1989.

8. Ibrahim. K. S., Husain, O., Bitensky, L., and Chayen, J. A modified tetrazolium reaction for identifying malignant cells from gastric and coloniecancer. J. Clin. Pathol.. 36: 133-136, 1983.

9. Petersen, O. W., Hoyer, P. E., Hilgers. J.. Briand. P., and Van Deurs. B.Characterization of epithelial cell islets in primary' monolayer cultures ofhuman breast carcinomas by the tetrazolium reaction for glucose-6-phosphatedehydrogenase. Virchows Arch. B Zellpathol.. SO:27-42, 1985.

10. Butcher, R. G. The oxygen insensitivity phenomenon as a diagnostic aid incarcinoma of the bronchus. In: J. R. Pattison. L. Bitensky, and J. Chayen(eds.), Quantitative Cytochemistry and Its Applications, pp. 241-251. NewYork: Academic Press, 1979.

11. Ponti, V., Dianzani. M. U., Cheeseman, K., and Slater, T. F. Studies on the

reduction of nitro blue tetrazolium chloride mediated through the action ofNADH and phenazine methosulphate. Chem.-Biol. Interact., 23: 281-291,1978.

12. Halliwell, B., and Gutteridge, J.M.C. Free Radicals in Biology and Medicine.Oxford, United Kingdom: Oxford University Press, 1985.

13. Oberely, L. W., and Buettner, G. R. Role of Superoxide dismutase in cancer:a review. Cancer Res., 39: 1141-1149, 1979.

14. Van Noorden, C. J. F. Enzyme reaction rate studies in tissue sections. Proc.R. Microsc. Soc., 23: 93-97, 1988.

15. Van Noorden, C. J. F., and Vogels, I. M. C. Polyvinyl alcohol and othertissue protectants in enzyme histochemistry: a consumer's guide. Histochem.J., 21: 373-380, 1989.

16. Butcher, R. G., and Evans, A. W. Diffusion during dehydrogenase reactions:the effects of intermediate electron acceptors. Histochem. J., 16: 885-895,1984.

17. Schor, N. A., and Cornelisse, C. J. Biochemical and quantitative histochem-ical study of reduced pyridine nucleotide dehydrogenation by human coloniecarcinomas. Cancer Res., 43: 4850-4855, 1983.

18. Butcher, R. G. Oxygen and the production of formazan from neotetrazoliumchloride. Histochemistry. 56: 329-340, 1978.

19. Butcher, R. G. Precise cytochemical measurement of neotetrazolium formazan by scanning and integrating microdensitometry. Histochemistry, 32:171-190, 1972.

20. Van Noorden, C. J. F., and Butcher, R. G. The out-of-range error inmicrodensitometry. Histochem. J., IS: 397-398, 1986.

21. Butcher, R. G., and Van Noorden, C. J. F. Reaction rate studies of glucose-6-phosphate dehydrogenase in sections of rat liver using four differenttetrazolium salts. Histochem. J., 17: 993-1008, 1985.

22. Altman, F. P. Tetrazolium salts—a consumer's guide. Histochem. J., 8: 471-

485, 1976.23. Dixon, M., and Webb, E. Enzymes, Ed. 2, London: Longmans, 1964.24. Bodmer, W. F., Bailey, C. J., Bodmer, J., Bussey, H. J. R., Ellis, A., Gorman,

P., et ai. Localization of the gene for familial adenomatous polyposis onchromosome 5. Nature (Lond.), 328:614-616, 1987.

25. Solomon, E., Voss, R., Hall, V., Boomer, W. F., Jass, J. R., Jeffreys, A. J.,et al. Chromosome 5 alÃ-eleloss in human colorectal carcinomas. Nature(Lond.), 328: 616-619, 1987.

26. Sato. S. Free radicals in NADPH-microsomes—Triphenyl tetrazolium chloride system as evidence by initiation of sulphite oxidation. Biochim. Biophys.Acta, 143: 554-561, 1967.

27. Sato, S., and Iwaizumi, M. Free radical mechanism by which triphenyltetrazolium chloride stimulates aerobic oxidation of NADPH by microsomes.Biochim. Biophys. Acta, 172: 30-36, 1969.

28. Van Noorden, C. J. F. On the role of oxygen in dehydrogenase reactionsusing tetrazolium salts. Histochem. J., 20: 587-593. 1988.

29. Altman, F. P. Quantitative dehydrogenase histochemistry with special reference to the pentose shunt dehydrogenases. Prog. Histochem. Cytochem., 4:225-273, 1972.

30. Petersen, O. W., Hoyer, P. E., and Van Deurs, B. Effect of oxygen on thetetrazolium reaction for glucose-6-phosphate dehydrogenase in cryosectionsof human breast carcinoma, fibrocystic disease and normal breast tissue.Virchows Arch. B Zellpathol., 50: 13-25, 1985.

31. Buchmann, A., Kuhlmann. W.. Schwarz. M.. Kunz, W., Wolf, C. R., Moll,E., et al. Regulation and expression of four cytochrome P-450 isoenzymes,NADPH cytochrome P-450 reducÃ-ase,the glutathione transferases B and Cand microsomal epoxide hydrolase in preneoplastic and neoplastic lesions inrat liver. Carcinogenesis (Lond.), 6: 513-521, 1985.

32. Altman, F. P., Bitensky, L., Butcher, R. G.. and Chayen. J. Integrated cellularchemistry applied to malignant cells. In: D. M. D. Evans (ed.), CytologyAutomation, pp. 82-97. Edinburgh: E. & S. Livingstone, 1970.

33. Dionisi, O., Galeotti, T., Terranova, T., and Azzi, A. Superoxide radicalsand hydrogen peroxide formation in mitochondria from normal and neoplastic tissues. Biochim. Biophys. Acta, 403: 292-300, 1975.

34. Freeman, B. A., and Crapo, J. D. Biology of a disease. Free radicals andtissue injury. Lab. Invest., 47: 412-426, 1982.

35. Hodgeson, E. K., and Fridovich, I. The interaction of bovine erythrocyteSuperoxide dismutase with hydrogen peroxide: inactivation of the enzyme.Biochemistry. 14: 5294-5299, 1975.

36. Geerts, A., and Roels, F. In vivo co-operation between hepatic catalase andSuperoxide dismutase demonstrated by diethyldithiocarbamate. FEBS Lett.,140:245-247, 1982.

37. Fried, R., Fried, L. W., and Babin, D. R. Biological role of xanthine oxidaseand tetrazolium-reductase inhibitor. Eur. J. Biochem., 33: 439-445, 1973.

5118

on June 26, 2021. © 1990 American Association for Cancer Research. cancerres.aacrjournals.org Downloaded from

http://cancerres.aacrjournals.org/

1990;50:5112-5118. Cancer Res Alistair James Best, Pranab K. Das, Hitendra R. H. Patel, et al. Potentially Malignant Cells in the ColonQuantitative Cytochemical Detection of Malignant and

Updated version

http://cancerres.aacrjournals.org/content/50/16/5112

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

.pubs@aacr.orgDepartment 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/50/16/5112To request permission to re-use all or part of this article, use this link

on June 26, 2021. © 1990 American Association for Cancer Research. cancerres.aacrjournals.org Downloaded from

http://cancerres.aacrjournals.org/content/50/16/5112http://cancerres.aacrjournals.org/cgi/alertsmailto:pubs@aacr.orghttp://cancerres.aacrjournals.org/content/50/16/5112http://cancerres.aacrjournals.org/