- Email: [email protected]
A simplified assay for methotrexate in biological fluids using a crude yeast lysate
207
Clinica Chimica Acta, 88 (1978) 207-213 @ Elsevier/North-Holland Biomedical Press
CCA 9355
A SIMPLIFIED ASSAY FOR METHQTREXATE USING A CRUDE YEAST LYSATE
E. EARL WEIR a, HERBERT KAIZER and DAVID B. LUDLUM b***
IN BIOLOGICAL
*-*, JACOB ROSENBERG
FLUIDS
b
a Oncology Center and Department of Pediatrics, The Johns Hopkins University School of Medicine, Baltimore, Md. 21205 and b Department of Pharmacology and Experimental Therapeutics, University of Maryland Hospital, Baltimore, Md. 21201 (U.S.A.) (Received
May 4th, 1978)
Summary Routine pharmacokinetic monitoring of methotrexate levels in biological fluids appears to be important for clinical applications involving large doses of methotrexate with leucovorin rescue. Recently several laboratories have independently described a competitive radiobinding assay that has the necessary sensitivity, precision and simplicity for this purpose. This study describes a of this assay method using a lysate of dry, active yeast as a source _ modification of binding activity. The assay is sensitive at blood concentrations of about 1 nM. No significant interference can be shown with naturally occurring folates. Specificity has also been shown by recovery experiments from serum and urine. Thus this modified assay matches the previously described radiobinding assays but with the advantage of using a methotrexate binding reagent that is readily available and quite inexpensive.
Introduction The clinical use of high doses of methotrexate (MTX) coupled with folinic acid rescue requires the routine monitoring of drug levels in biological fluids [ 11. Several laboratories have recently described a competitive radiobinding
* To whom requests for reprints should be addressed at: The Johns Hopkins Hospital. Oncology 3-121. 601 North Broadway. Baltimore, Md. 21205, U.S.A. ** Present address: Department of Pharmacology and Experimental Therapeutics. The Albany Medical C0llege.Albany.N.Y. 12208. U.S.A. The abbreviations used are: MTX, methotrexate (4-amino-N1 O-methylpteroylglutamic acid); 13HlMTX. methotrexate nominally tritiated in positions 3’. 5’ and 9; DHFR, dihydrofolate reductase (= tetrahydrofolate dehydrogenase; 5.6.7.8-tetrahydrofolate:NADP+ oxidoreductase. EC 1.5.1.3).
208
assay that serves as a specific rapid and sensitive assay for MTX in biological fluids [ 2-41. The present study describes a modification of that assay that uses a lysate of yeast as the binding reagent. The advantage of this modified assay is the ready availability and low cost of the binding reagent. Materials and methods The following reagents were obtained from Sigma Chemical Co., St. Louis, MO.: NADPH (Type I), NADH (Grade III), NADP (Sigma grade), NAD (Grade V), folic acid (Sigma grade), dihydrofolate (Grade 0), tetrahydrofolate (Grade III), 5methyletrahydrofolic acid (Grade II), and 2-mercaptoethanol (Type I). Leucovorin (U.S.P.) was a product of Lederle Laboratories, Pearl River, N.Y. Radioactive MTX (Amersham Searle, Arlington Heights, Ill., 5.3 Ci/mmol) was diluted to a concentration of 0.13 &i/ml (25 nmol) in 0.05 M potassium phosphate buffer, pH 8.5. Unlabeled MTX (Drug Development Branch, National Cancer Institute, Bethesda, Md.) was prepared in 0.05 M potassium phosphate buffer, pH 8.5. The charcoal suspension contained 5 g of activated charcoal (untreated powder, Sigma Chemical Co.) and 2.5 g of dextran (Clinical grade, Sigma) per liter of 0.05 M potassium phosphate buffer, pH 5.9. Yeast lysate is prepared from Fleischman’s or Red Star brand active, dry yeast with all steps carried out at 4°C. The yeast (2.25 oz) is suspended in 200 ml of deionized H,O and lysed by stirring for 1.5-2 h on a Kraft nonaerating stirrer (Kraft Apparatus, Inc., Mineola, N.Y.) with a motor setting of 9.5. Equivalent preparations can be obtained by 8-16 strokes of a motor driven Teflon pestle glass homogenizer (“Thomas Tissue Grinder”, A.H. Thomas, Inc., Philadelphia, Pa.). A Waring-type blender is ineffective. The defatted supematant obtained after centrifugation at 48 000 X g for 2 h is the source of the binding activity and is stable on storage at 4°C for about 2 weeks. Assay procedure
All operations are carried out at 22°C and involve the sequential addition of the following reagents: 15~1 of [3H]MTX (2.0 nCi, 0.376 pmol), 100 ~1 of MTX standard or unknown and 280 ~1 of the lysate mix. The lysate mix must be prepared fresh for each assay and contains: lysate, diluted to a level that will bind approximately 50-70% of the [3H]MTX in the absence of unlabeled MTX: and NADPH and 2-mercaptoethanol at final concentrations of 2.1 mM and 0.3 M, respectively. Buffering capacity is provided by the yeast lysate and the pH of the final solution has always been 5.9-6.0 without additional buffer. Following addition of the above reagents, the mixture is incubated for 90 min. The bound MTX is determined by the addition of 300 ~1 of the charcoal suspension followed by a 10 min incubation. The mixture is centrifuged at 3000 X g for 10 min and an aliquot of the supernatant counted in Scintiverse Scintillation Cocktail (Fisher Chemical Co.) with sufficient counts accummulated to give a 2% probable error. Background counts are determined by replacing the lysate mix with 0.1 M potassium phosphate buffer pH 6.0.
209
Results Determination of standard curve and calculation of MTX in unknowns A representative standard curve is shown in Fig. 1A. An unweighted least squares analysis of this data provides a linear regression equation with a correlation coefficient of >0.99. Calculation of the concentration of MTX in the unknowns is then determined by using the reciprocal of the fraction bound in the regression equation. Fig. 1B shows this standard curve replotted by the method of Scatchard [ 51. The linearity of this latter curve with a correlation coefficient of >0.97 indicates ‘that, despite the use of a crude extract as a source of binding protein, the assay is detecting a relatively uniform class of binding sites with an association constant of 5.5 X lo* M-l. The slope and intercept of the standard curve is a function of pH (optimum 5.5-6.0) and concentration of NADPH (optimum 2.1 mM before addition of 100 ~1 standard or unknown MTX solution). Changes in salt concentration or the presence of EDTA do not change the standard curve appreciably. The specificity of the assay was tested by addition of the following naturallyoccurring folate compounds: folic acid, 5-methyltetrahydrofolic acid, dihydrofolic acid, leucovorin, tetrahydrofolic acid. NAD and NADP were also tested for possible interference. In no case were significant changes in radiobinding observed. The largest effect was found in the case of 5-methyltetrahydrofolic acid with 300 pmol registering as 0.1 pmol MTX in the assay.
0.030
-
0.028
-
/ 7,
/
P
‘: i
0026-
.s -
0.024-
0.022
0016’
/
-
’ 000
f
f
020
040 pmoles
060 MTX/
080 tube
00
\
0.0’ 0
I
2 MTX
3 bo,,nd
4
(nM)
Fig. 1. Standard curve. (A) The reciprocal of the 8 of 13HlMTX bound is plotted against the concentration of unlabeled MTX. Determinations were done in triplicate and the range is shown. (B) The data from (A) are replotted according to the method of Scatchard [51. In this method, the ratio of the total MTXbound to MTXeee is plotted against the concentration of MTXbound.
5
-4-
-10
/
-9
-S
-7
-6
-5
-4
-3
-2
Fig. 2. Recovery experiments. Known amounts of unlabeled MTX are added to control samples of seem (circles) or urine (squares). All serum samples in the range of 10 mM to 10 nM were diluted at least fivefold. Serum samples with concentrations of 0.5-5.0 nM were deproteinized and assayed undiluted. Urine samples were similarlv treated except that urine samples at the lowest concentration range were assayed without dilution or deproteinization. The concentration found by the radiobinding assay is plotted against the concentration calculated by the amount of MTX added. The assays were repeated twelve times and the standard error of determination for each point is shown.
Reconstitution
test for the assay method
In order to determine if biological fluids would contain other substances that might interfere with the assay, recovery experiments were carried out. The results of a representative recovery experiment are shown in Fig. 2 in which the concentration calculated from adding a known amount of MTX to control urine or serum is plotted against the concentration as calculated by assay. The coefficient of correlation for a least square line assuming a recovery of 100% is 0.99, showing excellent recovery for both serum and urine studies. The lower limit of detection for serum concentrations as shown by these recovery experiments is 10 nM if the serum is diluted 5 fold or more. Undiluted serum gives a falsely high background. However, serum concentrations as low as 0.5 nM can be detected by using undiluted serum, deproteinized by boiling for 5 min. Comparison
of the radiobinding
assay to the enzyme
inhibition
assay
In order to compare the current assay with the most widely used assay of presumed equivalent sensitivity, serum samples from a patient receiving 9 g MTX in a 6-h infusion were obtained at various times after the start of infusion. MTX concentration was then determined by the radiobinding assay and by the enzyme inhibition assay as described by Bertino et al. [6]. Fig. 3 shows the concentration determined by each of these assays at various time points and shows good general agreement.
211
I
0-e
I
I
I
I
I
60 12 24 36 40 WS. Post Beginntng MTX lnfuston
72
04
Fig. 3. Comparison of radiobinding and enzyme inhibition assay. Serum was obtained at various times after the start of an infusion of a patient receiving a 6-h infusion of 9 g of MTX. The MTX concentration of these sera was determined by the radiobinding assay (squares, with the range of triplicate determinations shown) and by the enzyme inhibition assay (crosses).
Discussion Numerous methods have been proposed to measure the level of MTX in body fluids. They include bioassays [7], radioimmunoassays [S], assay methods based on the inhibition of dihydrofolate reductase [6], and a fluorometric assay [ 91. Studies of the cytotoxicity of MTX indicate a threshold level of 5 nM for the inhibition of DNA synthesis [lo]. The bioassays and fluorometric assay do not reach this level of sensitivity and therefore may not be useful for clinical monitoring. The enzyme inhibition assay has the requisite sensitivity but is complex, time consuming and requires some enzyme purification. The radioimmunoassay requires specially prepared antisera; quality control of antisera with regard to specificity and binding constant may be difficult to achieve. The radiobinding assay independently described by Meyers et al. [ 31, Kamen et al. [ 41, and Arons et al. [ 21 appears to fulfill the needs for a rapid, sensitive and relatively simple assay that could be adapted for routine clinical use. The assay is based on competition between [ 3H]MTX and unlabeled MTX for
212
binding to DHFR with subsequent removal of unbound drug by charcoal adsorption. The technique does not require measurement of enzyme activity and may not even require enzymatically active preparations; although, it does require a binding activity of relatively high affinity and specificity. The previously published descriptions of this assay use as a source of binding activity partially purified bacterial DHFR [ 31, partially purified guinea pig DHFR [4], or clarified but otherwise unpurified lysate of L1210 mouse leukemia cells [2]. All of these sources of binding reagents show the requisite specificity and uniformity of binding. Since any living system should contain folate binding activity, the current study utilized a crude lysate of yeast, selected for its ready availability and low cost, as a source of binding activity. The results show uniformity and specificity at least equivalent to the published reports [2-41. The sensitivity of the radiobinding assays, including the one described here, has a lower limit of detection of at least 1.0 nM. The specificity of the assay has been validated both by testing other interfering substances as competitors for [3H]MTX and by recovery experiments (Fig. 2). When the radiobinding assay is compared to the enzyme inhibition assay (Fig. 3), good correlations are observed except for disparities in later time points when serum levels fall below 100 nM. Although this disparity remains unexplained, the excellence of the recovery data and the good correspondence between our radiobinding data and other data on the pharmacokinetics of six hour infusions of MTX [ll-131, support our confidence in the radiobinding method. One note of caution regarding the use of this assay is related to the observation that control serum must be diluted five-fold or it will give a falsely high background. This effect is abrogated by deproteinization of serum. This general type of interference has been observed previously [2,14]. In addition to the potential clinical utility of the radiobinding assay, the methodology could be used to compare the relative affinity of various antifolate drugs for dihydrofolate reductase from various sources. For example, preliminary experiments in our laboratory suggest that aminopterin competes more successfully for binding sites than does MTX: a finding which correlates with the known higher potency of aminopterin. If the clinical application of antifolate drugs is related to their relative affinity for dihydrofolate reductase from different sources (i.e. bacterial, protozoan and mammalian), then this methodology could be used as a method for screening the potential application of various antifols. Acknowledgements This investigation was supported 20292, CA 06973-13.
by USPHS Grant Numbers
CA 20129, CA
References 1
Frei.
III,
E.,
Jaffe,
N.,
TattersaIl,
M.H.N.,
Pitman,
S.
and
Parker,
L.
(1975)
N.
EngI.
J. Med.
292.
846-851 2
Arons,
E.,
Rothenberg,
S.P.,
da
Costa,
M.,
Fischer,
C. and
IqbaI.
M.P.
(1975)
Cancer
Res.
35,
2033-
2038 3
Meyers, 3683-3686
C.E.,
Lippman,
M.E.,
Eliot.
H.M.
and
Chabner,
B.A.
(1975)
Proc.
Natl.
Acad.
Sci.
U.S.A.
72.
213 4 Kamen,
B.A., Takach,
P.L.. Vatev.
R. and Caston.
J.D. (1976)
Anal. Biochem.
5 Scatchard, G. (1949) Ann. N.Y. Acad. Sci. 51.660-672 6 Bertino. J.R.. Perkins, J.P. and Johns, D.G. (1965) Biochemistry 4. 839-846 7 Burchenal, J.H., Waring, G.B.. Ellison. R.R. and Reilly, H.C. (1951) Proc. 8 9 10 11 12 13 14
70. 54-63
Sot.
EXP. Biol.
Med. 78,
603-605 Raso, V. and Schrieber, R. (1975) Cancer Res. 35. 1407-1410 Freeman, M.V. (1957) J. Pharmacol. Exp. Ther. 120, l-8 Chabner. B.A. and Young, R.C. (1973) J. Clin. Invest. 52,1804-1811 Pratt, C.B., Roberts, D., Shanks, E. and Warmath, E.L. (1975) Cancer Chemother. Rep., Part 3, 6, 13-18 Tattersall, M.H.N., Parker, L.M.. Pitman, S.W. and Frei. III. E. (1975) Cancer Chemother. Rep., Part 3, 6, 25-29 Staller. R.G.. Jacobs, S.A., Drake, J.C.. Lutz, R.J. and Chabner, B.A. (1975) Cancer Chemother. Rep. Part 3, 6,19-24 Bertino. J.R. and Fischer, G.A. (1964) Methods Med. Res. 10, 297-307
Clinica Chimica Acta, 88 (1978) 207-213 @ Elsevier/North-Holland Biomedical Press
CCA 9355
A SIMPLIFIED ASSAY FOR METHQTREXATE USING A CRUDE YEAST LYSATE
E. EARL WEIR a, HERBERT KAIZER and DAVID B. LUDLUM b***
IN BIOLOGICAL
*-*, JACOB ROSENBERG
FLUIDS
b
a Oncology Center and Department of Pediatrics, The Johns Hopkins University School of Medicine, Baltimore, Md. 21205 and b Department of Pharmacology and Experimental Therapeutics, University of Maryland Hospital, Baltimore, Md. 21201 (U.S.A.) (Received
May 4th, 1978)
Summary Routine pharmacokinetic monitoring of methotrexate levels in biological fluids appears to be important for clinical applications involving large doses of methotrexate with leucovorin rescue. Recently several laboratories have independently described a competitive radiobinding assay that has the necessary sensitivity, precision and simplicity for this purpose. This study describes a of this assay method using a lysate of dry, active yeast as a source _ modification of binding activity. The assay is sensitive at blood concentrations of about 1 nM. No significant interference can be shown with naturally occurring folates. Specificity has also been shown by recovery experiments from serum and urine. Thus this modified assay matches the previously described radiobinding assays but with the advantage of using a methotrexate binding reagent that is readily available and quite inexpensive.
Introduction The clinical use of high doses of methotrexate (MTX) coupled with folinic acid rescue requires the routine monitoring of drug levels in biological fluids [ 11. Several laboratories have recently described a competitive radiobinding
* To whom requests for reprints should be addressed at: The Johns Hopkins Hospital. Oncology 3-121. 601 North Broadway. Baltimore, Md. 21205, U.S.A. ** Present address: Department of Pharmacology and Experimental Therapeutics. The Albany Medical C0llege.Albany.N.Y. 12208. U.S.A. The abbreviations used are: MTX, methotrexate (4-amino-N1 O-methylpteroylglutamic acid); 13HlMTX. methotrexate nominally tritiated in positions 3’. 5’ and 9; DHFR, dihydrofolate reductase (= tetrahydrofolate dehydrogenase; 5.6.7.8-tetrahydrofolate:NADP+ oxidoreductase. EC 1.5.1.3).
208
assay that serves as a specific rapid and sensitive assay for MTX in biological fluids [ 2-41. The present study describes a modification of that assay that uses a lysate of yeast as the binding reagent. The advantage of this modified assay is the ready availability and low cost of the binding reagent. Materials and methods The following reagents were obtained from Sigma Chemical Co., St. Louis, MO.: NADPH (Type I), NADH (Grade III), NADP (Sigma grade), NAD (Grade V), folic acid (Sigma grade), dihydrofolate (Grade 0), tetrahydrofolate (Grade III), 5methyletrahydrofolic acid (Grade II), and 2-mercaptoethanol (Type I). Leucovorin (U.S.P.) was a product of Lederle Laboratories, Pearl River, N.Y. Radioactive MTX (Amersham Searle, Arlington Heights, Ill., 5.3 Ci/mmol) was diluted to a concentration of 0.13 &i/ml (25 nmol) in 0.05 M potassium phosphate buffer, pH 8.5. Unlabeled MTX (Drug Development Branch, National Cancer Institute, Bethesda, Md.) was prepared in 0.05 M potassium phosphate buffer, pH 8.5. The charcoal suspension contained 5 g of activated charcoal (untreated powder, Sigma Chemical Co.) and 2.5 g of dextran (Clinical grade, Sigma) per liter of 0.05 M potassium phosphate buffer, pH 5.9. Yeast lysate is prepared from Fleischman’s or Red Star brand active, dry yeast with all steps carried out at 4°C. The yeast (2.25 oz) is suspended in 200 ml of deionized H,O and lysed by stirring for 1.5-2 h on a Kraft nonaerating stirrer (Kraft Apparatus, Inc., Mineola, N.Y.) with a motor setting of 9.5. Equivalent preparations can be obtained by 8-16 strokes of a motor driven Teflon pestle glass homogenizer (“Thomas Tissue Grinder”, A.H. Thomas, Inc., Philadelphia, Pa.). A Waring-type blender is ineffective. The defatted supematant obtained after centrifugation at 48 000 X g for 2 h is the source of the binding activity and is stable on storage at 4°C for about 2 weeks. Assay procedure
All operations are carried out at 22°C and involve the sequential addition of the following reagents: 15~1 of [3H]MTX (2.0 nCi, 0.376 pmol), 100 ~1 of MTX standard or unknown and 280 ~1 of the lysate mix. The lysate mix must be prepared fresh for each assay and contains: lysate, diluted to a level that will bind approximately 50-70% of the [3H]MTX in the absence of unlabeled MTX: and NADPH and 2-mercaptoethanol at final concentrations of 2.1 mM and 0.3 M, respectively. Buffering capacity is provided by the yeast lysate and the pH of the final solution has always been 5.9-6.0 without additional buffer. Following addition of the above reagents, the mixture is incubated for 90 min. The bound MTX is determined by the addition of 300 ~1 of the charcoal suspension followed by a 10 min incubation. The mixture is centrifuged at 3000 X g for 10 min and an aliquot of the supernatant counted in Scintiverse Scintillation Cocktail (Fisher Chemical Co.) with sufficient counts accummulated to give a 2% probable error. Background counts are determined by replacing the lysate mix with 0.1 M potassium phosphate buffer pH 6.0.
209
Results Determination of standard curve and calculation of MTX in unknowns A representative standard curve is shown in Fig. 1A. An unweighted least squares analysis of this data provides a linear regression equation with a correlation coefficient of >0.99. Calculation of the concentration of MTX in the unknowns is then determined by using the reciprocal of the fraction bound in the regression equation. Fig. 1B shows this standard curve replotted by the method of Scatchard [ 51. The linearity of this latter curve with a correlation coefficient of >0.97 indicates ‘that, despite the use of a crude extract as a source of binding protein, the assay is detecting a relatively uniform class of binding sites with an association constant of 5.5 X lo* M-l. The slope and intercept of the standard curve is a function of pH (optimum 5.5-6.0) and concentration of NADPH (optimum 2.1 mM before addition of 100 ~1 standard or unknown MTX solution). Changes in salt concentration or the presence of EDTA do not change the standard curve appreciably. The specificity of the assay was tested by addition of the following naturallyoccurring folate compounds: folic acid, 5-methyltetrahydrofolic acid, dihydrofolic acid, leucovorin, tetrahydrofolic acid. NAD and NADP were also tested for possible interference. In no case were significant changes in radiobinding observed. The largest effect was found in the case of 5-methyltetrahydrofolic acid with 300 pmol registering as 0.1 pmol MTX in the assay.
0.030
-
0.028
-
/ 7,
/
P
‘: i
0026-
.s -
0.024-
0.022
0016’
/
-
’ 000
f
f
020
040 pmoles
060 MTX/
080 tube
00
\
0.0’ 0
I
2 MTX
3 bo,,nd
4
(nM)
Fig. 1. Standard curve. (A) The reciprocal of the 8 of 13HlMTX bound is plotted against the concentration of unlabeled MTX. Determinations were done in triplicate and the range is shown. (B) The data from (A) are replotted according to the method of Scatchard [51. In this method, the ratio of the total MTXbound to MTXeee is plotted against the concentration of MTXbound.
5
-4-
-10
/
-9
-S
-7
-6
-5
-4
-3
-2
Fig. 2. Recovery experiments. Known amounts of unlabeled MTX are added to control samples of seem (circles) or urine (squares). All serum samples in the range of 10 mM to 10 nM were diluted at least fivefold. Serum samples with concentrations of 0.5-5.0 nM were deproteinized and assayed undiluted. Urine samples were similarlv treated except that urine samples at the lowest concentration range were assayed without dilution or deproteinization. The concentration found by the radiobinding assay is plotted against the concentration calculated by the amount of MTX added. The assays were repeated twelve times and the standard error of determination for each point is shown.
Reconstitution
test for the assay method
In order to determine if biological fluids would contain other substances that might interfere with the assay, recovery experiments were carried out. The results of a representative recovery experiment are shown in Fig. 2 in which the concentration calculated from adding a known amount of MTX to control urine or serum is plotted against the concentration as calculated by assay. The coefficient of correlation for a least square line assuming a recovery of 100% is 0.99, showing excellent recovery for both serum and urine studies. The lower limit of detection for serum concentrations as shown by these recovery experiments is 10 nM if the serum is diluted 5 fold or more. Undiluted serum gives a falsely high background. However, serum concentrations as low as 0.5 nM can be detected by using undiluted serum, deproteinized by boiling for 5 min. Comparison
of the radiobinding
assay to the enzyme
inhibition
assay
In order to compare the current assay with the most widely used assay of presumed equivalent sensitivity, serum samples from a patient receiving 9 g MTX in a 6-h infusion were obtained at various times after the start of infusion. MTX concentration was then determined by the radiobinding assay and by the enzyme inhibition assay as described by Bertino et al. [6]. Fig. 3 shows the concentration determined by each of these assays at various time points and shows good general agreement.
211
I
0-e
I
I
I
I
I
60 12 24 36 40 WS. Post Beginntng MTX lnfuston
72
04
Fig. 3. Comparison of radiobinding and enzyme inhibition assay. Serum was obtained at various times after the start of an infusion of a patient receiving a 6-h infusion of 9 g of MTX. The MTX concentration of these sera was determined by the radiobinding assay (squares, with the range of triplicate determinations shown) and by the enzyme inhibition assay (crosses).
Discussion Numerous methods have been proposed to measure the level of MTX in body fluids. They include bioassays [7], radioimmunoassays [S], assay methods based on the inhibition of dihydrofolate reductase [6], and a fluorometric assay [ 91. Studies of the cytotoxicity of MTX indicate a threshold level of 5 nM for the inhibition of DNA synthesis [lo]. The bioassays and fluorometric assay do not reach this level of sensitivity and therefore may not be useful for clinical monitoring. The enzyme inhibition assay has the requisite sensitivity but is complex, time consuming and requires some enzyme purification. The radioimmunoassay requires specially prepared antisera; quality control of antisera with regard to specificity and binding constant may be difficult to achieve. The radiobinding assay independently described by Meyers et al. [ 31, Kamen et al. [ 41, and Arons et al. [ 21 appears to fulfill the needs for a rapid, sensitive and relatively simple assay that could be adapted for routine clinical use. The assay is based on competition between [ 3H]MTX and unlabeled MTX for
212
binding to DHFR with subsequent removal of unbound drug by charcoal adsorption. The technique does not require measurement of enzyme activity and may not even require enzymatically active preparations; although, it does require a binding activity of relatively high affinity and specificity. The previously published descriptions of this assay use as a source of binding activity partially purified bacterial DHFR [ 31, partially purified guinea pig DHFR [4], or clarified but otherwise unpurified lysate of L1210 mouse leukemia cells [2]. All of these sources of binding reagents show the requisite specificity and uniformity of binding. Since any living system should contain folate binding activity, the current study utilized a crude lysate of yeast, selected for its ready availability and low cost, as a source of binding activity. The results show uniformity and specificity at least equivalent to the published reports [2-41. The sensitivity of the radiobinding assays, including the one described here, has a lower limit of detection of at least 1.0 nM. The specificity of the assay has been validated both by testing other interfering substances as competitors for [3H]MTX and by recovery experiments (Fig. 2). When the radiobinding assay is compared to the enzyme inhibition assay (Fig. 3), good correlations are observed except for disparities in later time points when serum levels fall below 100 nM. Although this disparity remains unexplained, the excellence of the recovery data and the good correspondence between our radiobinding data and other data on the pharmacokinetics of six hour infusions of MTX [ll-131, support our confidence in the radiobinding method. One note of caution regarding the use of this assay is related to the observation that control serum must be diluted five-fold or it will give a falsely high background. This effect is abrogated by deproteinization of serum. This general type of interference has been observed previously [2,14]. In addition to the potential clinical utility of the radiobinding assay, the methodology could be used to compare the relative affinity of various antifolate drugs for dihydrofolate reductase from various sources. For example, preliminary experiments in our laboratory suggest that aminopterin competes more successfully for binding sites than does MTX: a finding which correlates with the known higher potency of aminopterin. If the clinical application of antifolate drugs is related to their relative affinity for dihydrofolate reductase from different sources (i.e. bacterial, protozoan and mammalian), then this methodology could be used as a method for screening the potential application of various antifols. Acknowledgements This investigation was supported 20292, CA 06973-13.
by USPHS Grant Numbers
CA 20129, CA
References 1
Frei.
III,
E.,
Jaffe,
N.,
TattersaIl,
M.H.N.,
Pitman,
S.
and
Parker,
L.
(1975)
N.
EngI.
J. Med.
292.
846-851 2
Arons,
E.,
Rothenberg,
S.P.,
da
Costa,
M.,
Fischer,
C. and
IqbaI.
M.P.
(1975)
Cancer
Res.
35,
2033-
2038 3
Meyers, 3683-3686
C.E.,
Lippman,
M.E.,
Eliot.
H.M.
and
Chabner,
B.A.
(1975)
Proc.
Natl.
Acad.
Sci.
U.S.A.
72.
213 4 Kamen,
B.A., Takach,
P.L.. Vatev.
R. and Caston.
J.D. (1976)
Anal. Biochem.
5 Scatchard, G. (1949) Ann. N.Y. Acad. Sci. 51.660-672 6 Bertino. J.R.. Perkins, J.P. and Johns, D.G. (1965) Biochemistry 4. 839-846 7 Burchenal, J.H., Waring, G.B.. Ellison. R.R. and Reilly, H.C. (1951) Proc. 8 9 10 11 12 13 14
70. 54-63
Sot.
EXP. Biol.
Med. 78,
603-605 Raso, V. and Schrieber, R. (1975) Cancer Res. 35. 1407-1410 Freeman, M.V. (1957) J. Pharmacol. Exp. Ther. 120, l-8 Chabner. B.A. and Young, R.C. (1973) J. Clin. Invest. 52,1804-1811 Pratt, C.B., Roberts, D., Shanks, E. and Warmath, E.L. (1975) Cancer Chemother. Rep., Part 3, 6, 13-18 Tattersall, M.H.N., Parker, L.M.. Pitman, S.W. and Frei. III. E. (1975) Cancer Chemother. Rep., Part 3, 6, 25-29 Staller. R.G.. Jacobs, S.A., Drake, J.C.. Lutz, R.J. and Chabner, B.A. (1975) Cancer Chemother. Rep. Part 3, 6,19-24 Bertino. J.R. and Fischer, G.A. (1964) Methods Med. Res. 10, 297-307