June 1991, Volume 41, Issue 6

Original Article


M. Perwaiz Iqbal  ( Departments of Biochemistry, The Aga Khan University, Karachi. )
N. Mehboobali  ( Departments of Biochemistry, The Aga Khan University, Karachi. )
M. Anwar Waqar  ( Department of Biochemistry, The Aga Khan University, Karachi. )
Mushtaq Ahmed  ( Departments of Surgery, The Aga Khan University, Karachi. )


[BH]-methotrexate binding at pH 5.0 and pH 7.2 by the cytosol of tumor tissues and the surrounding normal areas of the gastrointestinal tract of patients suffering from colon or gastric cancer has been used to identify In these cells the presence of a binder of methotrexate having low-affinity for this drug in addition to the enzyme dihydrofolate reductase. Scatchard analysis of the [3H]- methotrexate binding data by a colon tumor sample also reveals that there are two binders of this drug present In the cytosol of these cells. The association constant (Kass) for one binder of methotrexate is = 5.6 x 10 M1 while the Kass for the second binder is = 1.0 x 106 M1. The two binders do not differ very much in their apparent molecular weight. Upon isoelectric focusing, the tumor cell cytosol resolves into 4 major isoproteins each having the ability not only to bind [ H]- methotrexate but also reduce [H]-pteroylglutamic acid to H]- tetrahydropteroylglutamic acid. This suggests that the two binders of methotrexate may be the two forms of dihydrofolate reductase having different affinities for this anticancer drug(JPMA 41:136, 1991).


Dihydrofolate reductase (DHFE., EC. plays a central role in the pathway of pyrimidine and purine biosynthesis and, therefore, is a target enzyme in cancer chemotherapy. The anticancer drug methotrexate (MTX) produces cytocidal effect by binding to DHFR and block­ing its action. There have been quite a few reports on the heterogeneity of the enzyme in terms of its binding to MDC in human as well as murine leukemia cells1-7 and mammalian liver8,9. Colorectal carcinoma is one of the most common solid tumors in adults10. Since MTX as a single agent has never been considered to be of much value in the management of advanced colorectal car­cinoma, we attempted to find out whether these cells contain another binder of MTX having low affinity for this drug and, therefore, would not respond to MDC therapy.


[3H] MTX with a specific activity 11.8 Ci/mmole and [3H] pteroylglutamic acid ([3H]PGA) with a specific activity 34 Ci/mmole were purchased from Amer­sham/Searle. [3 H]MTX was purified by chromatography on a Bio Gel P-30 column. NADPH, pteroylglutamic acid (PGA), MDC, pH ampholytes (Ampholines), phenyl­methylsulfonylfluoride (PMSF), Norit A Neutral char­coal, dextran (molecular weight 10,000), bovine serum albumin, dextran blue, cytochrome C, horse raddish peroxidase, ethanolamine were purchased from Sigma Chemical Company. Trasylol was obtained from Mobay (New York). Preparation of tissue extracts Human tumor tissues (colon cancer) and the normal tissues (surrounding the tumor), excised at the time of surgery were homogenized as described previously11, but with a slight modification that the homogenates were prepared using 3 ml of 0.6 M sodium citrate buffer, pH 7.2, for everygm of tissue. Since these tissue cells contain lysosomal enzymes, trasylol (1000 KIU/1), PMSF (3.5 mg/i) and 0.02% sodium azide were included in the buffer used for making homogenates. These homogenates were then centrifuged at 105,000 x g for 1 hr and the cytosol obtained was analyzed for [3H]MTX binding at pH 5.0 and pH 7.2.
[3H]MTX binding studies
Binding of [3HJMTX by the tissue extracts was car­ried out by a procedure as described by Rothenberg et al12. A total reaction volume of 0.5 ml in 0.06 M citrate, pH 4.8 or pH 7.4, containing 48 uM NADPH, 5.8 mM 2-mercaptoethanol, 2.2 nM [3H]MTX, 0-22 nM un­labelled MDC and 50 ul cell cytosol were incubated at room temperature for 30 minutes. The reactions were then stopped by the addition of 0.4 ml of one percent Norit. A neutral charcoal in 0.5 percent dextran (molecular weight 10,000). After centrifugation, radioac­tivity in 0.5 ml of supernatant solution representing the enzyme bound [3H]MTX was counted in an LS-3801 Spectrometer (Beckman Instruments, Palo Alto, CA) using 5 ml of 3a70 scintillation fluor (Research Product International, USA). A blank containing all constituents of the reaction mixture except the cytosol was run with each assay series to determine the radioactivity not removed by charcoal in the absence of the cytosol preparation. This radioactivity was subtracted from the experimental samples to determine the net counts per minute of bound [3H]MTX. Preliminary experiments in­dicated that the coefficient of variation for separating bound and free MDC by dextran-coated charcoal was between 2.1% and 3.8% when the binding activity was greater than 5% of the total MDC.
Gel Column Chromatography
Sephadex G-75 was equilibrated with 0.05 M Tris­HCI buffer, pH 7.4, containing 0.02% sodium azide and packed in a column of size 0.7 cm x 50 cm. Half ml of colon tumor cytosol was applied to the column and eluted at a flow rate of 4.8 mI/hr. Fractions in a volume of 0.5 ml were collected and assayed for [3H]MDC bind­ing at pH 7.2. Dextran blue, pure goat liver DHFR (Mr 20,000), horse raddish peroxidase (Mr 40,000), cytochrome C (Mr 12,400) and3H2O were used as column markers.
Isoelectric Focusing
A colon tumor cytosol sample was analyzed by isoelectric focusing in a standard LKB 1 10-mI column packed with 1% (W/V) ampholytes at a pH range of 3.5  10 in a zero to 50% sucrose gradient. The sample (1 ml) was dialyzed against distilled water for 4 hrs to remove electrolytes prior to application to the column. The dialyzed sample and the 75% of total ampholytes used were added to the solution which was adjusted to con­tain 50% sucrose (“dense solution”). The remaining am­pholytes (25%) were mixed with “light solution” comprising of 5% sucrose. The electrode solution for the cathode (25 ml) was 0.262 M. ethanolantine in 60% sucrose. The electrolyte solution for the anode (20 ml) was 0.173 M H3P04 prepared in distilled water. A frac­tion of this solution, enough to completely cover the respective electrode was placed on top of the column. The column was run for 20 hrs at a maximum power of 5W with the maximum voltage set at 1.6 KV. Two mgs. of pure hemoglobin was added alongwith the sample as a marker for focusing. The column was eluted at a constant flow rate of 100 mI/hr and 0.9 ml fractions were col­lected. Every other fraction was dialyzed against 0.025 M potassium phosphate buffer, pH 7.4 containing 0.025 M potassium chloride for 20 hrs to remove Ampholines and then analyzed for [3H]MTX binding12 and for the reduction of [3H]PGA to [3H]-tetrahydro-PGAB.


This work was supported by a research grant (No. S-AKU/Chem-191) awarded to M. Perwaiz Iqbal by the Pakistan Science Foundation.


1. Rothenberg, S.P. and Iqbal, M.P. Human leukemia and normal leukocytes contain a species of immunoreactive but nonfunctional dihydrofolate reductase. Proc. NatL Acad. Sci. USA., 1982; 79: 645.
2. Rothenberg. S.P. and Iqbal, M.P. Identification of a high hmolecular-weight nonfunc­tional protein in L1210 leukemia cells with common antigenic determinants to dihydrofolate reductase. Cancer Rca., 1983; 43: 529.
3. Duffy, T.H., Beckman, S.B. and Huennekens, F.M. Multiple forms of L.1210 dibydrofo. late reductase differing in affinity for methotrexate. Biochern. Biophys. Res. Commun., 1984; 119. 352.
4. Dedhar, S.and Goldie, J.H. Overproduction of two antigenically distinct forms of dihydrofolate reductase in a highly methotrexate-resistant mouse leukemia cell line. Cancer Res., 1983; 43: 4863.
5. Dedhar, S. and Goldie, J.H. Methotrexate-resistant human Promyelocytic leukemia (HL-60) cells express dihydrofolate reductase with altered propertiesassocisted with in­creased enzyme activity. Biochem. Biophys. Res. Commun., 1985; 129: 536.
6. Dedhar, S., Hartley, D., Gibbons, D.F., Phillips, G. and Goldie, .J.H. Heterogeneity in specific activity and methotrexate sensitivity of dihydrofolate reductase from blast cells of acute myelogenous leukemia patients. J. Clin. Oncol., 1985; 3: 1545.
7. Iqbal, S.F., Waqar, M.A., Mehboobali, N. and Malik, I. A low- affinity binder of methotrexate in human leukemia cells. Biochem. Soc. Trans., 1990; 18: 633.
8. Kyner, D. and Rothenberg. S.F. Age-dependent expression of a novel protein in mouse liver immunologically and functionally homologous with dihydrofolate reductase. Biochim. Biopbys. Acts, 1989; 993: 56.
9. lqbal, M.P., Mehboobali, N. and Waqar, MA. Multiple forms of dibydrofolate reduc­tase in goat liver. Biochem. Soc. Trans., 1989; 17: 561.
10. Rabinovich, MG., Perez, I.E., Macchiaveli, M., Romero, A., Kremer, A., Leone, B.A. and Strauss, B. Sequential combination chemotherapywith methotrexate and 5-fluorouracil in advanced colrirectal carcinoma. Tumori, 1984; 70: 549.
11. Rothenberg. S.P., Iqbal, M.P. and daCosta, M. Effect of folate compounds on the ac­cumulation of methotrexate and the activity of dihydrofolate reductase in liver, kidney and small intestine of the mouse. J. Pharmacol. Exp. Ther., 1982; 223: 631.
12. Rothenberg, S.F., daCoata, M. and lqbal, M.P. Ligand binding radioassay for the An. tifolate compounds. Application in patients receiving methotrexate. CancerTreat. Rep., 1977; 61:575.
13. Rothenberg, S.F. A rapid radioassay for folic acid reductase and amethopterin. Anal. Biochem., 1966; 16: 176.
14. Bertino, J.R., Booth, BA., Bieber, AL., Cashmore, A. and Sartorelli, AC. Studies on the induction of dihydrofolate reductase by the folate antagonists. 1. Biol. Chem., 1964;239: 479.
15. Scatchard, C. The attraction of proteins for small molecules and ions. Ann. N.Y. Acad. Sci., 1949; 51: 660.
16. Baccanari, D.P., Stone, D. and Kuyper, L. Effect of a single amino acid substitution on Escherichia coli dihydrofolate reductase catalysis and ligand bindin& J. Biol. Chem.,1981; 256: 1738.
17. Haber, D.A., Beverley, SM., Kiely, M.L. and Schimke, R.T. Propertiea of an altered dihydrofolate reductase encoded by amplified genes in cultured mouse fibroblasts. J. Biol. Chem., 1981; 256: 9501.
18. Jackson, R.C. and Harrap, K.R. Studies with a mathematical model of folate metabo­lism. Arch. Biochem. Biophys., 1973; 158:820.
19. White, J.C. and Goldman, I.D. Mechanism of action of methotrexate IV Free intracel­lular methotrexate required to suppress dihydrofolate reduction to tetrahydrofolate by Ehrlich aacites tumor cells in vitro. Mol. Phsrmacol., 1976; 12: 711.
20. Houghton, J.A., Tice, A.J. and Houghton, P.J. The selectivity of action of methotrexate in combination with 5-fluorouracil in xenografts on human colon adenocarcinomas. Mol. Pharmacol., 1982; 22: 771.
21. Cantrell, I.E. Jr., Brunet, R., Lagarde, C., Schein, P.S. and Smith, F.P. Phase II study of sequential methotrexate-5-FU therapy in advanced measurable colorectal cancer. Can­cer Treat. Rep., 1982; 66: 1563.
22. Heim, M.E., Flechtner, H., Edler, I., and QueiBer, W. Sequential high-dose methotrexate and 5-fluorouracil in the treatment of advanced colorectal cancer. Tumor Diagnostik & Therapie, 1980; 7: 197.
23. Bertino, JR., Sawicki, W.L., Lindqust, C. and Gupta, VS. Schedule-dependent an­titumor effects of methotrexate and 5- fluorouracil. Cancer Res., 1977; 37: 327.
24. Bertino, JR., Mini, B. and Fernandes, DJ. Sequential methotrexate and 5-fluorouracil. mechanisms of synergy. Semin. Oncol., 1983; 10 (2 suppl.2): 2.
25. Benz, C., Tattleman, E. and Cadman, B. Phase I pilot study using 24 hour sequential methotrexate and 5-fluorouracil. Proc. Am. Soc. Clin. Oncol., 1982; 1: 13.
26. Gewirtz, AM. and Cadman, B. Preliminary report on the efficacy of sequential methotreaxate and 5-fluorouracil in advanced breast cancer. Cancer, 1981; 47:2552.


Figure 1 shows the binding of [3H]MTX at pH 5.0 and pH 7.2 by the colon tumor cytosol. At pH 5.0, the maximum concentration of [3H]MTX bound was 0.73 nM as against 1.06 nM at pH 7.2 when the total MDC con­centration in the reaction was 24.2 nM. There was nearly 45% increase in the [3H]MTX binding at pH 7.2. Since DHFR has been shown to have a higher affinity for MDC at acidic pH and a lower affinity for the inhibitor at pH 7.214, increased MDC bindingat pH 7.2 as compared to at pH 5.0 suggests that in addition to DHFR there is an additional species of binding sites participating in the reaction. Had there been only one form of the binder of this drug in these cells, the binding of [3H]MTX at saturating concentration of total MDC in the reaction would have been the same at pH 5.0 and pH 7.2. A number of tumor tissues of gastrointestinal tract (GIT) and the surrounding normal areas of GIT were analyzed for [3H]MTX bind­ing at pH 5.0 and pH 7.2. As shown in Table,

it is evi­dent that in most of the tissues ex­amined, there is more binding of [3H]MTX at pFI 7.2 as compared to pH 5.0. The difference in the total binding of [3H]MTX at the of drug bound to by one of the colon tumor cytosol samples at in­creasing con­centrations of the unlabelled MDC in the reactions.

Figure 2 shows that one of the binders of MDC had an associa­tion constant (Kass) = 5.6x 107 M-1, while the other one had a Kass = 1.06x 106 M-1. These values are close to the two Kass values reported for the two binders of MTX in human leukemia cells7. We could not separate the two binders of MTX by gel-filtration chromatography because a single peak of [3H]MTX binding vas ob­tained when colon tumor cytosol was ap­plied to Sephadex G-75 column suggest­ing that the two binders may not be differing much in their ap­parent molecular weight which ap­pears to around 20,000 (Figure 3).

Previous multiple forms of DHFR in various types of cells, i.e., mouse fibroblasts, Escherichia coli, L1210 and L5178 Y leukemia cells and goat liver cells, indicate that their molecular weights are either identical16,17 or very close­ly similar3,4,9. In our study the major limitation in terms of separating and characterizing these two binders of MDC has been the small amount of enzyme in these human tissues. At the present moment we cannot be absolutely certain that the low affinity binder of MDC is in fact a form of DHFR having weaker affinity for this antifolate. However, the isoelectric focusing profile of the colon tumor cytosol (Figure.4)

shows 4 major isoproteins of dihydrofolate reductase activity with their pIs 7.3, 6.5, 5.7 and 4.7, respectively. These isoproteins were obtained by analyzing the post-focusing fractions not only for [3HJMTX binding at pH 7.2 but also for [3H]PGA reduction to [3H]-tetrahydro-PGA. The fact that the four enzyme activity peaks co-elute exactly with the four [3H]MTX binding peaks suggests that the low-af­finity binder of [3H]MTX in colon tumor cells is probably another form of DHFR with altered affinity for this anti-cancer drug. Since these cancer patients had never been treated with MTX, the presence of a low affinity form of DHFR in their tumor cells indicates that insen­sitivity to MDC is an intrinsic phenomenon in certain types of cells. Similar findings have been reported by Dedhar et al6 for blast cells of acute myelogenous leukemia patients. MDC as a single agent has never been considered effective in the management of colorectal or gastric carcinoma and that may have been due to the fact that such cells might be containing a low affinity form of DI IFR. It has been reported by Jackson and Harrap18 and White and Goldman19 that no more than 5% of the folate reductase activity is required to generate sufficient tetrahydrofolate cofactors to maintain cell viability. Thus. if these cancer cells contain a small amount of low affinity form of DHFR, it would be sufficient to maintain this de novo DNA synthesis, even if all of the high affinity form of the enzyme has been inactivated by MTX. There­fore, in these cells unless very high doses of MTX are used, MDC in conventional or intermediate doses would be unlikely to produce any good results. There have been quite a few reports of sequential combination chemotherapy with MTX and 5-fluorouracil in the management of colorectal cancer10,20,22, but in these cases the role of MDC is more in terms of enhancing the binding of 5-fluorodeoxy-uridylate (FdUMP) to thymidy­late synthetase rather than completely knocking off the activity of DHFR23,24. In other words, MDC in these cells has a synergistic effect on the intracellular utilization of 5-fluorouracil25,26. Thus, heterogeneity of MDC binding in the colon tumor provides us with a plausible explanation about the ineffectiveness of MDC in the treatment of such tumors. However, very high doses of MTX which have rarely been attempted in the management of colorectal carcinoma, in theory at least, may be of value in obtaining the desired therapeutic responses.

Journal of the Pakistan Medical Association has agreed to receive and publish manuscripts in accordance with the principles of the following committees: