- Email: [email protected]
A sensitive, nonradiometric assay for dihydroorotic acid dehydrogenase using anion-exchange high-performance liquid chromatography
ANALYTICAL
BIOCHEMISTRY
161,32-38
A Sensitive, Nonradiometric Using Anion-Exchange
(1987)
Assay for Dihydroorotic Acid Dehydrogenase High-Performance Liquid Chromatography
G. J. PETERS,’ E. LAURENSSE, A. LEYVA, AND H. M. PINEDO Department of Oncology, Free University Hospital, P.O. Box 7057, 1007 MB Amsterdam, The Netherlands Received June 9, 1986 A new method to assay the mitochondrial pyrimidine de novo enzyme, dihydroorotate (DHO) dehydrogenase, which catalyzes the dehydrogenation of DHO, with erotic acid as the product was developed. The assay was optimized using a rat liver mitochondrial preparation. Orotic acid was quantified with high-performance liquid chromatography using an anion-exchange column (Partisil-SAX) with uv detection at 280 nm. Isocratic elution with low phosphate buffer at pH 4.0 was used. The detection limit was 20 pmol per injection, which is comparable to previously described radiometric assays.The HPLC assay was compared with a spectrophotometric assaymeasuring erotic acid formation in a deproteinized reaction mixture. Absorbance was measured at the optimal wavelength for erotic acid, 278.5 nm. This assayis less sensitive and less specific than the HPLC assay, which can also detect UMP which might be formed from erotic acid in whole homogenates. With both assays kinetic parameters of the enzyme were determined. In the high concentration range (80-1000 pM) both K,,, and Vmnx values were comparable. With the HPLC assay the concentration range was extended down to 12 pM and initial rates could be determined. The apparent K,,, was about 12 PM. The HPLC assayis also suitable for use in the study of inhibition of DHO dehydrogenase. o 1987 Academic press, Inc. KEY WORDS:
dihydroorotic acid dehydrogenase; erotic acid; high-performance liquid chromatography; pyrimidine de novo; pyrimidine nucleotides; enzyme assay.
UMP synthase. Here, erotic acid is channeled toward UMP via OMP. For the OPRT reaction PRPP is the cosubstrate, while ODC catalyzes decarboxylation. DHO-DH is different from the first three enzymes and from UMP synthase in various aspects. First, it is not part of an enzyme complex. Second, its location is not cytosolic but mitochondrial(3-6). It is located on the outside of the inner membrane (3,6). Different compounds have been reported to act as electron acceptors (3-5). In bacteria NADf is the hydrogen receptor of DHO-DH, but the reaction in the reverse direction, the reduction of erotic acid, has a higher rate in the presence of NADPH. In mammalian cells DHO-DH is not dependent on NADf (5) but appears to be coupled to the respiratory chain via ubiquinone (3). Oxygen appears to be the final electron acceptor which was
DHO-DH* is the fourth enzyme in pyimidine de nova nucleotide synthesis (I), catalyzing the oxidation of L-DHO to erotic acid (Fig. 1). The other five enzymes of the de novo synthesis are part of two different enzyme complexes located in the cytoplasm. The first three enzymes, carbamoyl phosphate synthetase II, aspartate transcarbamylase, and dihydroorotase form one large multienzyme complex in which the substrates are channeled toward the final product of these three reactions, DHO (1). The fifth and sixth enzymes, OPRT and ODC, also form one enzyme complex (1,2) called ’ To whom correspondence should be addressed. * Abbreviations used: L-DHO, L-dihydroorotic acid; DHO-DH, dihydroorotic acid dehydrogenase; OPRT, orotate phosphoribosyltransferase; ODC, orotidylate decarboxylase; PRPP, 5phosphoribosyl- I -pyrophosphate. 0003-2697187
$3.00
Copyright 0 1987 by Academic Prss, Inc. All rights of reproduction in any form reserved.
32
NONRADIOMETRIC
ASSAY FOR DIHYDROOROTIC
FIG. I. The DHO-DH reaction.
demonstrated by the use of specific inhibitors of the respiratory chain while oxygen deficiency reduced enzyme activity considerably (7). Forman and Kennedy (8) postulated that DHO-DH is capable of superoxide formation. Various indirect and direct assays have been described for DHO-DH. One of the indirect assays is based on the reduction of dichloroindophenol via a superoxide (8). Absorbance of dichloroindophenol at 600 nm decreases after reduction. Another indirect assay involves coupling with the respiratory chain (3). DHO-DH has also been assayed by using the difference in absorbance of L-DHO and erotic acid at 278 nm (4,7-9) or by measuring the disappearance of DHO (4). Several direct radiometric assays for DHO-DH have been described. Conversion of [4-14C]DH0 and [6-14C]DH0 to labeled erotic acid has to be followed by subsequent isolation of the product from the substrate by either electrophoretic (6,lO) or chromatographic procedures (3,ll). DHO-DH can also be assayed by conversion of [curboxyl“C]DH0 to [carboxyZ-‘4C]orotic acid, which must be further metabolized by OPRT and ODC in the presence of PRPP. The liberated 14COZ can be trapped by hyamine and counted (9,12). A tritium release assay, using L-[~,~-~H]DHO as substrate is based on the loss of these labeled hydrogens upon oxidation; in the presence of oxygen the likely product is water (13). The disadvantage of the nonradiochemical assays is that they are more or less indirect and usually relatively
ACID DEHYDROGENASE
33
insensitive. The radiochemical assays are more sensitive, but the radiochemical substrate usually has to be prepared, because of limited commercial availability. High backgrounds might be encountered. In our studies on modulation of pyrimidine de ~OVO synthesis (6,14- 16) we needed an assay for DHO-DH, since this enzyme might be a target for antitumor agents ( 11,17). Since labeled L-DHO was no longer available we developed a new and sensitive assay of DHO-DH, based on the specific determination of erotic acid using anion-exchange high-performance liquid chromatography with isocratic elution. The assay appeared to have comparable sensitivity and accuracy to the radiochemical assay and has the particular advantage of low backgrounds. MATERIALS
AND METHODS
Materials. L-DHO, erotic acid, and orotidine were obtained from Sigma Chemical Co. (St. Louis, MO). RPM1 1640 cell culture medium and fetal bovine serum were obtained from Grand Island Biological Co. (Paisley, Scotland). All other chemicals were of analytical grade. Cell culture. L12 10 cells were cultured in RPM1 1640 medium supplemented with 10% fetal bovine serum and 60 PM 2-mercaptoethanol as described previously ( 14). Enzyme preparation. Livers from male Wistar rats (about 3 months weighing about 200 g) were used for the isolation of the mitochondrial fraction. Rats were killed by decapitation. Livers were removed immediately, rinsed with 0.85% saline, cut in small pieces with a pair of scissors, and suspended in ice-cold homogenization buffer (0.25 M sucrose, 10 mM Tris-HCl, pH 7.4). A 10% homogenate (1 g liver plus 9 ml buffer) was prepared using a hand-operated Potter-Elvehjem apparatus, which was chilled on ice. The homogenate was centrifuged at 600g for 10 min at 4°C. The 600g supernatant was centrifuged for 30 min a’t SOOOgand 4°C in a MSE high-speed centrifuge. The supernatant
34
PETERS ET AL.
was discarded and the pellet containing the mitochondria was suspended in assay buffer (0.1 M Tris-HCl, pH 8.0). The mitochondrial pellet of 1 g liver was suspended in 5 ml assay buffer. Total cell extracts from L 12 10 cells were prepared after suspension of the cells in 0.1 M Tris-HCl buffer (pH 8.0) followed by sonication as described previously for lymphoid cells ( 15). Protein content was determined using a dye binding assay ( 18). Assay of DHO-DH. DHO-DH was measured in Eppendorf reaction vessels at 37°C in a shaking water bath. The reaction mixture consisted of 0.9 ml mitochondrial preparation (0.05-4 mg protein) and 31.5 ~1 30 mM L-DHO. The enzyme assay was optimized at this saturating DHO concentration. The reaction was terminated after 5-60 min (15-60 min for the spectrophotometric assay) by addition of 180 ~1 16% trichloroacetic acid. The tubes were chilled on ice for 20 min; thereafter denatured protein was removed by centrifugation in a minicentrifuge (5 min at 14,OOOg). Orotic acid present in the supernatant was measured by both the spectrophotometric assay and the HPLC assay. Neutralization of supernatants was performed with alamine-Freon (16). Blanks were processed similarly but without addition of L-DHO. Kinetic parameters were determined with reciprocal Lineweaver-Burk plots and with Eadie-Hofstee plots. Spectrophotometric assay. The absorbance of the supernatant of the deproteinized reaction mixture was measured at the absorbance maximum of erotic acid, 278.5 nm. L-DHO does not interfere at this wavelength (Fig. 2). A solution of erotic acid prepared similarly to the reaction mixture but without protein was always included to serve as a standard. Absorbance was measured in a double-beam Kontron spectrophotometer, Uvikon 722 LC (Kontron, Zurich, Switzerland). HPLC assay. Orotic acid was also determined using an ion-exchange HPLC method. The chromatographic system consisted of a Waters M45 solvent delivery sys-
tern (Waters Associates Inc., Milford, MA), a Kontron MS1 660 autosampler, and a Kontron Uvikon-740 LC fixed wavelength detector (set at 280 nm). Chromatographic data were recorded and quantified using a Perkin-Elmer Sigma 1OB chromatographic data system (Perkin-Elmer, Norwalk, CT). Samples of 20 ~1 were injected onto a Whatman Par-us&SAX column prepacked by Chrompack, Middelburg, The Netherlands (25 X 0.4-cm, length i.d.; particle size, 10 pm). Compounds were separated using isocratic elution (8 mrvt KH2P04, 8 mM KCl, pH 4.0) at a flow of 1.5 ml/min. RESULTS
Spectrophotometric
Assay
The spectrophotometric assay is based on the difference in absorbance between L-DHO and erotic acid (Fig. 2). DHO does not absorb at 280 nm. In isolated mitochondria erotic acid will be the only product of the reaction since the two subsequent enzymes of pyrimidine de lzovo nucleotide synthesis, OPRT and ODC, are cytosolic enzymes. So increase in absorbance will only be due to newly formed erotic acid. Conditions of the assay were essentially as described previously (6), except for the pH. Since the activity of the enzyme is higher at pH 8.0 assays were performed at this pH. The enzyme assay was linear with protein in the tested range of 0.2-2 mg per assay. The assay was linear 1.5
3 0.1 mM erotic
acid
1 mM dihydroerotic acid
1.0
2
VI r= 5 .5 0.5 l!!iilE 2 ‘0 “7 s
1
0
0 200
250
300
350 “ill
200
250
300
I IO “m
FIG. 2. Absorbance spectra of L-DHO and erotic acid.
NONRADIOMETRIC
ASSAY FOR DIHYDROOROTIC
with time up to at least 60 min at optimal substrate concentrations. The detection limit of the assay was about 10 nmol per reaction mixture. Lower concentrations of erotic acid might be detected, but absorbance at 278.5 nm of other cellular compounds resulted in a high background and affected the accuracy of the assay. Therefore for determination of the apparent K, value for L-DHO the range of L-DHO concentrations could not be extended to L-DHO concentrations lower than 80 PM. K,,, and I’,,, values determined in this high concentration range are 74 +_27 PM and 46 1 + 19 nmol/h per mg protein, respectively (means 4 SD from four separate livers).
a
0.0051
react ion
b 0.005
1
HPLC Assay The main advantage of a HPLC assay is the specific determination of erotic acid. Figure 3a gives an example of a typical chromatographic determination of erotic acid in a deproteinized reaction mixture. It appears that in extracts of liver mitochondria no interfering peaks are detected in the region of erotic acid (Fig. 3b). So blanks are very low and only dependent on the stability of the detector. Orotidine, which might be formed out of erotic acid ( 19) was eluted just before erotic acid at 6.8 min. However, in none of the assays performed with mitochondria was orotidine formed. The detection limit of the assay was 20 pmol per injection (Fig. 3~). The time required for one sample was 15 min. Thus, using the autosampler 80 samples might be analyzed overnight. A relatively low concentration of phosphate and KC1 were sufficient for separation and no gradient elution was required. These might be the reasons that the Partisil-SAX column did not show a loss of separation characteristics after a considerable number of analyses. Up to now at least 1000 samples could be analyzed with the same column. The sensitivity of the assay might be increased by injecting a higher volume of the reaction mixture onto the column as was performed oc-
35
ACID DEHYDROGENASE
0
5
10
15
C 1
standard
2
FIG. 3. HPLC assay of DHO-DH. A typical chromatogram of a reaction mixture (a) and a blank (b) are shown. Twenty microliters of a deproteinized sample was injected. (c) The chromatogram of a standard containing 20 pmol erotic acid. Orotic acid was eluted at 7.8 min.
casionally in samples with a low amount of erotic acid. The assay was linear with protein in the range 0.05-2 mg protein/ml (Fig. 4). Assays were also linear with time up to at least 60 min at optimal substrate concentrations. Estimation of kinetic parameters for L-DHO in the same samples used for the spectrophotometric assay and in the same concentration range gave comparable results. The kinetic parameters from these and some additional rats were: K,,,, 56 f 14 PM and If,,,,,, 396 +- 64 nmol/h per mg protein (means + SD of eight rat livers). Consumption of substrate at 80 pM L-DHO did not exceed 15%. Exten-
PETERS ET AL.
36
0
0.4
0.8 1.2 mg protein
lation. It appeared that DHO-DH showed biphasic kinetics in the range of 12- 1000 PM (Fig. 5). Eadie-Hofstee plots gave similar results. The HPLC assay was also used to measure the activity in other fractions obtained after differential centrifugation of a rat liver homogenate. No activity could be detected in the SOOOg supernatant, which contains all cytosolic enzymes. Of the enzyme activity present in the total liver homogenate, a considerable part was also present in the 600g pellet. This is probably due to contamination of the 6008 pellet with mitochondria. However, the major part of the activity was present in the 600g supematant (60 nmol/h per mg protein), which contains the major part of the mitochondria. Although OPRT and ODC are present in the 600g supematant no nucleotides were formed. With the HPLC system UMP which might be formed out of erotic acid (9) can be detected. The retention time of UMP is 11.7 min. However, no UMP could be detected. This is probably due to the absence of PRPP, which is an essential cosubstrate for OPRT. In tissues PRPP is usually present at suboptimal to optimal concentrations (1) but due to di-
1.6
FIG. 4. Standard curve for protein against erotic acid formed in the DHO-DH assay, showing linearity of the assaywith protein. Incubation time was 15 min for each sample. In other experiments at higher amounts of protein than indicated in the figure, linearity was observed.
sion of the concentration range of L-DHO was possible because of the lower detection limit of the HPLC assay. However, at these lower concentrations (down to 12 PM), substrate consumption after 30 min in the presence of high amounts of protein was between 20 and 80%, leading to high erotic acid concentrations, which will inhibit DHO-DH (3,4). At a low concentration of L-DHO with low amounts of protein the reaction was only linear up to 10 min (substrate consumption less than lo%), which allowed initial rate determination, essential for accurate Km calcu-
12
/’ I
I
-160
I
I
- 120
I
I
- 80
I
I
- 40
/
1
I
I
I
l/s
40 (mM -'I
0
I
I
I
SO
FIG. 5. Double reciprocal Lineweaver-Burk plot of DHO-DH. The plot is of one representative experiment out of five. The enzyme activities were determined with the HPLC assayafter IO-min incubations. For each assay 0.1 mg protein was used. The apparent K,,, in the low concentration range was 12. I + 2.4 pM and the apparent V,, was 276 f 45 nmol/h/mg protein (means ? SD of five rat livers). In separate experiments (not shown) more points in the high concentration range (> 100 GM) were included, in order to determine the high K,,,. u = nmol/h/mg protein.
NONRADIOMETRIC
ASSAY
FOR
DIHYDROOROTIC
lution and likely degradation during homogenization the concentration of PRPP will be too low in the reaction mixture for conversion of erotic acid to OMP. Comparison of the Spectrophotometric and HPLC Assays All assays of DHO-DH analyzed by measurement of uv absorbance at 278.5 nm were also analyzed using the HPLC assay. Comparable results were obtained. Assays of DHODH were usually performed in reaction mixtures of about 1 ml since this volume is required for the spectrophotometric assay. However, for the HPLC assay the volume might be lowered since only 20 ~1 is required for injection into a HPLC system. For assay of DHO-DH in the murine leukemia cell line L12 10 the volume was lowered to 200 ~1, since some extra volume is required when using the autosampler. Preparation of mitochondria from these cells by standard procedures such as Potter-Elvehjem homogenization was not suitable because of the small size of the cells and the flexibility of the cellular membranes. Sonification or freezing of the cells would also disrupt cellular organelles. Therefore, DHO-DH was assayed in total cell extracts. In some cases a large number of cells was cultured and DHO-DH assayed with the spectrophotometric assay amounted to 17.6 f 4.0 nmol/h per lo6 cells (mean + SD of four separate experiments). In these samples the DHO-DH as measured with the HPLC assay was 16.0 + 2.1 nmol/h per lo6 cells. The apparent K, for L-DHO in extracts from L12 10 cells was about 11 PM. No biphasic kinetics were observed. DISCUSSION
Using anion-exchange HPLC a highly sensitive method for determination of DHODH activity has been developed. The method compares very favorably to other methods in terms of sensitivity and specificity. The assay can detect picomoles of orotic acid and has the particular advantage of
ACID
DEHYDROGENASE
37
low backgrounds, which permits the reliable estimation of low activities of DHO-DH, observed at low substrate concentrations or after inhibition of the enzyme. The results of this HPLC assay performed at pH 7.4 (data not shown) are comparable to those obtained previously with the radiometric assay (6). Labeled [6-14C]DH0 for this assay has been prepared from [6-‘4C]orotic acid using a commercially available preparation of DHO-DH from Zymobacterium oroticum. However, [6-‘4C]orotic acid had to be purified before use, because of considerable contamination. Other radiochemical assays (3,8- 12) may suffer from the same disadvantage, leading to high backgrounds. Although a reaction volume of 0.9 ml was used in our assays, it is also possible to use smaller volumes, since only 20 ~1 of sample are required for chromatographic analysis. This permits the use of the same small volumes as might be used for the radiochemical analyses. Data obtained with the HPLC assay agree well with those reported by others (3,9), who reported enzyme activities of DHO-DH in the same range as we do. The enzyme activity at pH 7.4 also agrees with data reported by others (3). Also the subcellular distribution reported previously (3-6,13) agrees with the present observations with the HPLC assay indicating the validity of the assay. The apparent K, value of about 12 PM obtained with our HPLC assay is in the same range as the values reported for fractionated rat liver mitochondria, 5.2 PM (3), intact murine liver mitochondria, 6.7 and 10 PM (11,13), and human spleen mitochondria, 5.3 PM (12). A higher K, (180 PM) was reported by Kennedy (4), but it was determined under conditions which did not allow initial rate determination. In a subsequent paper the same authors (20) reported a lower K, (2 PM) for the purified enzyme. These authors also reported that detergents and chaotropic agents usually employed for the isolation of this enzyme from mitochondria, affected enzyme kinetics (20). Biphasic kinetics for DHO-DH have not been reported previously. However, K,,,
38
PETERS ET AL.
determinations for DHO-DH were usually determined in a concentration range lower than 75 PM (1 I- 13) but not at higher concentrations, probably because this DHO concentration is above physiological levels and will not affect regulation of pyrimidine de nova nucleotide synthesis in vivo. An accumulation of DHO might only be caused by a lack of oxygen or its equivalent (7,21) or when inhibition of OPRT leads to an accumulation of erotic acid, which will inhibit DHO-DH (2 1). The HPLC assay has now been applied to study the mechanism of action of a new quinoline carboxylic acid derivative, DUP 785 (NSC 368390), which has shown antitumor activity against a variety of human xenografts of different histological origin and against murine tumors and L12 10 leukemia (22). Preliminary studies demonstrated that this compound has antipyrimidine effects by inhibition of DHO-DH from various sources (23,24). With our newly developed HPLC assay we could analyze the mode of inhibition of DHO-DH (24). In conclusion, we developed a new and sensitive assay for DHO-DH with the particular advantage of low backgrounds. The amount of erotic acid formed was measured using HPLC with an ion-exchange column. The method produced enzyme activities and kinetic values comparable to those obtained with previously described spectrophotometric and radiometric assays.
3. Chen, J.-J., and Jones, M. E. (1976) Arch. Biochem. Biophys. 176, 82-90. 4. Kennedy, J. (1973) Arch. B&hem. Biophys. 157, 369-373.
5. Miller, R. W., Kerr, C. T., and Curry, J. R. (1968) Canad. J. Biochem. 46, 1099-I 106. 6. Peters, G. J., and Veerkamp, J. H. (I 984) Adv. Exp. Med. Biol. 165A, 531-534. 7. Lijffler, M. (1980) Eur. J. Biochem. 107,207-215. 8. Forman, H. J., and Kennedy, J. (1976) Arch. Eiothem. Biophys. 173,219-224. 9. Dileepan, K. N., and Kennedy, J. (1983) FEBS Left 153, 1-5.
IO. Westwick, W. J., Allsop, J., and Watts, R. W. E. (1972) Biochem. Pharmacol. 21, 1955-1966. I I. Bennett, L. L.. Smithers, D., Rose, L. M.. Adamson, D. J.. and Thomas, H. J. (1979) Cancer Rex 39, 4868-4874. 12.
13.
14.
15. 16. 17.
18. 19.
Smithers, G. W.. Gero, A. M.. and O’Sullivan, W. J. A. (1978) Anal. Biochem. 88,93-103. Kensler. T. W., Cooney, D. A., Jayaram, H. N., Schaeffer, C., and Choie, D. D. (1981) Anal. Biothem. 117, 3 15-3 19. Leyva, A., Appel, H., Smith, P., Lankelma, J., and Pinedo, H. M. (1981) Cancer Lett. 12, 169-173. Peters, G. J., Oosterhof, A., and Veerkamp, J. H. (1983) Int. J. B&hem. 15, 51-55. Peters, G. J., Laurensse, E., Lankelma, J. Leyva. A., and Pinedo, H. M. ( 1984) Eur. J. Cancer C/in. Oncof. 20, 1425-1431. De Frees, S. A., Sawick, D. P., Cunningham, B. A., Morre, D. J., Cassidy, J. M., and Heinstein, P. F. (1984) Proc. Amer. Assoc. Cancer Res. 25, 19 (Abstract 73). Sedmak, J. J., and Grossberg, S. E. (1977) Anal. Biochem. 79, 334-352. Janeway, L. M., and Cha, S. (197 I) Cancer Res. 37, 4382-4388.
Forman, H. J.. and Kennedy, J. (1977) J. Biol. Chem. 252, 3379-338 1. 21. Chen, J.-J., and Jones, M. E. (1979) J. Biol. Chem. 20.
254,4908-49
ACKNOWLEDGMENTS This work was supported by the Dutch Cancer Foundation. “Koningin Wilhelmina Fends,” Grant IKA 83- 16. The authors thank S. L. Sharma for his contributions to this work.
REFERENCES 1. Jones, M. E. (1980) Annu. Rev. Biochem. 49, 253-279. 2. Traut, T. W. (1982) Trends Biochem. Sci. 7, 255-257.
22.
14.
Dexter, D. L., Hesson, D. P., Ardecky, R. J., Rao, G. V., Tippett, D. L., Dusak, B. A., Paull, K. D., Plowman, J., De Larco, B. M., Narayanan, V. L., and Forbes, M. (1985) Cancer Rex 45, 5563-5568.
Chen, S. F., Ruben, R. L., and Dexter, D. L. (1986) Proc. Amer. Assoc. Cancer Res. 27,298 (Abstract 1183). 24. Peters, G. J., Laurensse, E. L., Sharma, S. L., Leyva, A., and Pinedo, H. M. (1986) in Proceedings, 14th International Cancer Congress, Budapest, Vol. 1, p. 163 (Abstract 615). Karger, Basel; Akademia Kiadb, Budapest. 23.
BIOCHEMISTRY
161,32-38
A Sensitive, Nonradiometric Using Anion-Exchange
(1987)
Assay for Dihydroorotic Acid Dehydrogenase High-Performance Liquid Chromatography
G. J. PETERS,’ E. LAURENSSE, A. LEYVA, AND H. M. PINEDO Department of Oncology, Free University Hospital, P.O. Box 7057, 1007 MB Amsterdam, The Netherlands Received June 9, 1986 A new method to assay the mitochondrial pyrimidine de novo enzyme, dihydroorotate (DHO) dehydrogenase, which catalyzes the dehydrogenation of DHO, with erotic acid as the product was developed. The assay was optimized using a rat liver mitochondrial preparation. Orotic acid was quantified with high-performance liquid chromatography using an anion-exchange column (Partisil-SAX) with uv detection at 280 nm. Isocratic elution with low phosphate buffer at pH 4.0 was used. The detection limit was 20 pmol per injection, which is comparable to previously described radiometric assays.The HPLC assay was compared with a spectrophotometric assaymeasuring erotic acid formation in a deproteinized reaction mixture. Absorbance was measured at the optimal wavelength for erotic acid, 278.5 nm. This assayis less sensitive and less specific than the HPLC assay, which can also detect UMP which might be formed from erotic acid in whole homogenates. With both assays kinetic parameters of the enzyme were determined. In the high concentration range (80-1000 pM) both K,,, and Vmnx values were comparable. With the HPLC assay the concentration range was extended down to 12 pM and initial rates could be determined. The apparent K,,, was about 12 PM. The HPLC assayis also suitable for use in the study of inhibition of DHO dehydrogenase. o 1987 Academic press, Inc. KEY WORDS:
dihydroorotic acid dehydrogenase; erotic acid; high-performance liquid chromatography; pyrimidine de novo; pyrimidine nucleotides; enzyme assay.
UMP synthase. Here, erotic acid is channeled toward UMP via OMP. For the OPRT reaction PRPP is the cosubstrate, while ODC catalyzes decarboxylation. DHO-DH is different from the first three enzymes and from UMP synthase in various aspects. First, it is not part of an enzyme complex. Second, its location is not cytosolic but mitochondrial(3-6). It is located on the outside of the inner membrane (3,6). Different compounds have been reported to act as electron acceptors (3-5). In bacteria NADf is the hydrogen receptor of DHO-DH, but the reaction in the reverse direction, the reduction of erotic acid, has a higher rate in the presence of NADPH. In mammalian cells DHO-DH is not dependent on NADf (5) but appears to be coupled to the respiratory chain via ubiquinone (3). Oxygen appears to be the final electron acceptor which was
DHO-DH* is the fourth enzyme in pyimidine de nova nucleotide synthesis (I), catalyzing the oxidation of L-DHO to erotic acid (Fig. 1). The other five enzymes of the de novo synthesis are part of two different enzyme complexes located in the cytoplasm. The first three enzymes, carbamoyl phosphate synthetase II, aspartate transcarbamylase, and dihydroorotase form one large multienzyme complex in which the substrates are channeled toward the final product of these three reactions, DHO (1). The fifth and sixth enzymes, OPRT and ODC, also form one enzyme complex (1,2) called ’ To whom correspondence should be addressed. * Abbreviations used: L-DHO, L-dihydroorotic acid; DHO-DH, dihydroorotic acid dehydrogenase; OPRT, orotate phosphoribosyltransferase; ODC, orotidylate decarboxylase; PRPP, 5phosphoribosyl- I -pyrophosphate. 0003-2697187
$3.00
Copyright 0 1987 by Academic Prss, Inc. All rights of reproduction in any form reserved.
32
NONRADIOMETRIC
ASSAY FOR DIHYDROOROTIC
FIG. I. The DHO-DH reaction.
demonstrated by the use of specific inhibitors of the respiratory chain while oxygen deficiency reduced enzyme activity considerably (7). Forman and Kennedy (8) postulated that DHO-DH is capable of superoxide formation. Various indirect and direct assays have been described for DHO-DH. One of the indirect assays is based on the reduction of dichloroindophenol via a superoxide (8). Absorbance of dichloroindophenol at 600 nm decreases after reduction. Another indirect assay involves coupling with the respiratory chain (3). DHO-DH has also been assayed by using the difference in absorbance of L-DHO and erotic acid at 278 nm (4,7-9) or by measuring the disappearance of DHO (4). Several direct radiometric assays for DHO-DH have been described. Conversion of [4-14C]DH0 and [6-14C]DH0 to labeled erotic acid has to be followed by subsequent isolation of the product from the substrate by either electrophoretic (6,lO) or chromatographic procedures (3,ll). DHO-DH can also be assayed by conversion of [curboxyl“C]DH0 to [carboxyZ-‘4C]orotic acid, which must be further metabolized by OPRT and ODC in the presence of PRPP. The liberated 14COZ can be trapped by hyamine and counted (9,12). A tritium release assay, using L-[~,~-~H]DHO as substrate is based on the loss of these labeled hydrogens upon oxidation; in the presence of oxygen the likely product is water (13). The disadvantage of the nonradiochemical assays is that they are more or less indirect and usually relatively
ACID DEHYDROGENASE
33
insensitive. The radiochemical assays are more sensitive, but the radiochemical substrate usually has to be prepared, because of limited commercial availability. High backgrounds might be encountered. In our studies on modulation of pyrimidine de ~OVO synthesis (6,14- 16) we needed an assay for DHO-DH, since this enzyme might be a target for antitumor agents ( 11,17). Since labeled L-DHO was no longer available we developed a new and sensitive assay of DHO-DH, based on the specific determination of erotic acid using anion-exchange high-performance liquid chromatography with isocratic elution. The assay appeared to have comparable sensitivity and accuracy to the radiochemical assay and has the particular advantage of low backgrounds. MATERIALS
AND METHODS
Materials. L-DHO, erotic acid, and orotidine were obtained from Sigma Chemical Co. (St. Louis, MO). RPM1 1640 cell culture medium and fetal bovine serum were obtained from Grand Island Biological Co. (Paisley, Scotland). All other chemicals were of analytical grade. Cell culture. L12 10 cells were cultured in RPM1 1640 medium supplemented with 10% fetal bovine serum and 60 PM 2-mercaptoethanol as described previously ( 14). Enzyme preparation. Livers from male Wistar rats (about 3 months weighing about 200 g) were used for the isolation of the mitochondrial fraction. Rats were killed by decapitation. Livers were removed immediately, rinsed with 0.85% saline, cut in small pieces with a pair of scissors, and suspended in ice-cold homogenization buffer (0.25 M sucrose, 10 mM Tris-HCl, pH 7.4). A 10% homogenate (1 g liver plus 9 ml buffer) was prepared using a hand-operated Potter-Elvehjem apparatus, which was chilled on ice. The homogenate was centrifuged at 600g for 10 min at 4°C. The 600g supernatant was centrifuged for 30 min a’t SOOOgand 4°C in a MSE high-speed centrifuge. The supernatant
34
PETERS ET AL.
was discarded and the pellet containing the mitochondria was suspended in assay buffer (0.1 M Tris-HCl, pH 8.0). The mitochondrial pellet of 1 g liver was suspended in 5 ml assay buffer. Total cell extracts from L 12 10 cells were prepared after suspension of the cells in 0.1 M Tris-HCl buffer (pH 8.0) followed by sonication as described previously for lymphoid cells ( 15). Protein content was determined using a dye binding assay ( 18). Assay of DHO-DH. DHO-DH was measured in Eppendorf reaction vessels at 37°C in a shaking water bath. The reaction mixture consisted of 0.9 ml mitochondrial preparation (0.05-4 mg protein) and 31.5 ~1 30 mM L-DHO. The enzyme assay was optimized at this saturating DHO concentration. The reaction was terminated after 5-60 min (15-60 min for the spectrophotometric assay) by addition of 180 ~1 16% trichloroacetic acid. The tubes were chilled on ice for 20 min; thereafter denatured protein was removed by centrifugation in a minicentrifuge (5 min at 14,OOOg). Orotic acid present in the supernatant was measured by both the spectrophotometric assay and the HPLC assay. Neutralization of supernatants was performed with alamine-Freon (16). Blanks were processed similarly but without addition of L-DHO. Kinetic parameters were determined with reciprocal Lineweaver-Burk plots and with Eadie-Hofstee plots. Spectrophotometric assay. The absorbance of the supernatant of the deproteinized reaction mixture was measured at the absorbance maximum of erotic acid, 278.5 nm. L-DHO does not interfere at this wavelength (Fig. 2). A solution of erotic acid prepared similarly to the reaction mixture but without protein was always included to serve as a standard. Absorbance was measured in a double-beam Kontron spectrophotometer, Uvikon 722 LC (Kontron, Zurich, Switzerland). HPLC assay. Orotic acid was also determined using an ion-exchange HPLC method. The chromatographic system consisted of a Waters M45 solvent delivery sys-
tern (Waters Associates Inc., Milford, MA), a Kontron MS1 660 autosampler, and a Kontron Uvikon-740 LC fixed wavelength detector (set at 280 nm). Chromatographic data were recorded and quantified using a Perkin-Elmer Sigma 1OB chromatographic data system (Perkin-Elmer, Norwalk, CT). Samples of 20 ~1 were injected onto a Whatman Par-us&SAX column prepacked by Chrompack, Middelburg, The Netherlands (25 X 0.4-cm, length i.d.; particle size, 10 pm). Compounds were separated using isocratic elution (8 mrvt KH2P04, 8 mM KCl, pH 4.0) at a flow of 1.5 ml/min. RESULTS
Spectrophotometric
Assay
The spectrophotometric assay is based on the difference in absorbance between L-DHO and erotic acid (Fig. 2). DHO does not absorb at 280 nm. In isolated mitochondria erotic acid will be the only product of the reaction since the two subsequent enzymes of pyrimidine de lzovo nucleotide synthesis, OPRT and ODC, are cytosolic enzymes. So increase in absorbance will only be due to newly formed erotic acid. Conditions of the assay were essentially as described previously (6), except for the pH. Since the activity of the enzyme is higher at pH 8.0 assays were performed at this pH. The enzyme assay was linear with protein in the tested range of 0.2-2 mg per assay. The assay was linear 1.5
3 0.1 mM erotic
acid
1 mM dihydroerotic acid
1.0
2
VI r= 5 .5 0.5 l!!iilE 2 ‘0 “7 s
1
0
0 200
250
300
350 “ill
200
250
300
I IO “m
FIG. 2. Absorbance spectra of L-DHO and erotic acid.
NONRADIOMETRIC
ASSAY FOR DIHYDROOROTIC
with time up to at least 60 min at optimal substrate concentrations. The detection limit of the assay was about 10 nmol per reaction mixture. Lower concentrations of erotic acid might be detected, but absorbance at 278.5 nm of other cellular compounds resulted in a high background and affected the accuracy of the assay. Therefore for determination of the apparent K, value for L-DHO the range of L-DHO concentrations could not be extended to L-DHO concentrations lower than 80 PM. K,,, and I’,,, values determined in this high concentration range are 74 +_27 PM and 46 1 + 19 nmol/h per mg protein, respectively (means 4 SD from four separate livers).
a
0.0051
react ion
b 0.005
1
HPLC Assay The main advantage of a HPLC assay is the specific determination of erotic acid. Figure 3a gives an example of a typical chromatographic determination of erotic acid in a deproteinized reaction mixture. It appears that in extracts of liver mitochondria no interfering peaks are detected in the region of erotic acid (Fig. 3b). So blanks are very low and only dependent on the stability of the detector. Orotidine, which might be formed out of erotic acid ( 19) was eluted just before erotic acid at 6.8 min. However, in none of the assays performed with mitochondria was orotidine formed. The detection limit of the assay was 20 pmol per injection (Fig. 3~). The time required for one sample was 15 min. Thus, using the autosampler 80 samples might be analyzed overnight. A relatively low concentration of phosphate and KC1 were sufficient for separation and no gradient elution was required. These might be the reasons that the Partisil-SAX column did not show a loss of separation characteristics after a considerable number of analyses. Up to now at least 1000 samples could be analyzed with the same column. The sensitivity of the assay might be increased by injecting a higher volume of the reaction mixture onto the column as was performed oc-
35
ACID DEHYDROGENASE
0
5
10
15
C 1
standard
2
FIG. 3. HPLC assay of DHO-DH. A typical chromatogram of a reaction mixture (a) and a blank (b) are shown. Twenty microliters of a deproteinized sample was injected. (c) The chromatogram of a standard containing 20 pmol erotic acid. Orotic acid was eluted at 7.8 min.
casionally in samples with a low amount of erotic acid. The assay was linear with protein in the range 0.05-2 mg protein/ml (Fig. 4). Assays were also linear with time up to at least 60 min at optimal substrate concentrations. Estimation of kinetic parameters for L-DHO in the same samples used for the spectrophotometric assay and in the same concentration range gave comparable results. The kinetic parameters from these and some additional rats were: K,,,, 56 f 14 PM and If,,,,,, 396 +- 64 nmol/h per mg protein (means + SD of eight rat livers). Consumption of substrate at 80 pM L-DHO did not exceed 15%. Exten-
PETERS ET AL.
36
0
0.4
0.8 1.2 mg protein
lation. It appeared that DHO-DH showed biphasic kinetics in the range of 12- 1000 PM (Fig. 5). Eadie-Hofstee plots gave similar results. The HPLC assay was also used to measure the activity in other fractions obtained after differential centrifugation of a rat liver homogenate. No activity could be detected in the SOOOg supernatant, which contains all cytosolic enzymes. Of the enzyme activity present in the total liver homogenate, a considerable part was also present in the 600g pellet. This is probably due to contamination of the 6008 pellet with mitochondria. However, the major part of the activity was present in the 600g supematant (60 nmol/h per mg protein), which contains the major part of the mitochondria. Although OPRT and ODC are present in the 600g supematant no nucleotides were formed. With the HPLC system UMP which might be formed out of erotic acid (9) can be detected. The retention time of UMP is 11.7 min. However, no UMP could be detected. This is probably due to the absence of PRPP, which is an essential cosubstrate for OPRT. In tissues PRPP is usually present at suboptimal to optimal concentrations (1) but due to di-
1.6
FIG. 4. Standard curve for protein against erotic acid formed in the DHO-DH assay, showing linearity of the assaywith protein. Incubation time was 15 min for each sample. In other experiments at higher amounts of protein than indicated in the figure, linearity was observed.
sion of the concentration range of L-DHO was possible because of the lower detection limit of the HPLC assay. However, at these lower concentrations (down to 12 PM), substrate consumption after 30 min in the presence of high amounts of protein was between 20 and 80%, leading to high erotic acid concentrations, which will inhibit DHO-DH (3,4). At a low concentration of L-DHO with low amounts of protein the reaction was only linear up to 10 min (substrate consumption less than lo%), which allowed initial rate determination, essential for accurate Km calcu-
12
/’ I
I
-160
I
I
- 120
I
I
- 80
I
I
- 40
/
1
I
I
I
l/s
40 (mM -'I
0
I
I
I
SO
FIG. 5. Double reciprocal Lineweaver-Burk plot of DHO-DH. The plot is of one representative experiment out of five. The enzyme activities were determined with the HPLC assayafter IO-min incubations. For each assay 0.1 mg protein was used. The apparent K,,, in the low concentration range was 12. I + 2.4 pM and the apparent V,, was 276 f 45 nmol/h/mg protein (means ? SD of five rat livers). In separate experiments (not shown) more points in the high concentration range (> 100 GM) were included, in order to determine the high K,,,. u = nmol/h/mg protein.
NONRADIOMETRIC
ASSAY
FOR
DIHYDROOROTIC
lution and likely degradation during homogenization the concentration of PRPP will be too low in the reaction mixture for conversion of erotic acid to OMP. Comparison of the Spectrophotometric and HPLC Assays All assays of DHO-DH analyzed by measurement of uv absorbance at 278.5 nm were also analyzed using the HPLC assay. Comparable results were obtained. Assays of DHODH were usually performed in reaction mixtures of about 1 ml since this volume is required for the spectrophotometric assay. However, for the HPLC assay the volume might be lowered since only 20 ~1 is required for injection into a HPLC system. For assay of DHO-DH in the murine leukemia cell line L12 10 the volume was lowered to 200 ~1, since some extra volume is required when using the autosampler. Preparation of mitochondria from these cells by standard procedures such as Potter-Elvehjem homogenization was not suitable because of the small size of the cells and the flexibility of the cellular membranes. Sonification or freezing of the cells would also disrupt cellular organelles. Therefore, DHO-DH was assayed in total cell extracts. In some cases a large number of cells was cultured and DHO-DH assayed with the spectrophotometric assay amounted to 17.6 f 4.0 nmol/h per lo6 cells (mean + SD of four separate experiments). In these samples the DHO-DH as measured with the HPLC assay was 16.0 + 2.1 nmol/h per lo6 cells. The apparent K, for L-DHO in extracts from L12 10 cells was about 11 PM. No biphasic kinetics were observed. DISCUSSION
Using anion-exchange HPLC a highly sensitive method for determination of DHODH activity has been developed. The method compares very favorably to other methods in terms of sensitivity and specificity. The assay can detect picomoles of orotic acid and has the particular advantage of
ACID
DEHYDROGENASE
37
low backgrounds, which permits the reliable estimation of low activities of DHO-DH, observed at low substrate concentrations or after inhibition of the enzyme. The results of this HPLC assay performed at pH 7.4 (data not shown) are comparable to those obtained previously with the radiometric assay (6). Labeled [6-14C]DH0 for this assay has been prepared from [6-‘4C]orotic acid using a commercially available preparation of DHO-DH from Zymobacterium oroticum. However, [6-‘4C]orotic acid had to be purified before use, because of considerable contamination. Other radiochemical assays (3,8- 12) may suffer from the same disadvantage, leading to high backgrounds. Although a reaction volume of 0.9 ml was used in our assays, it is also possible to use smaller volumes, since only 20 ~1 of sample are required for chromatographic analysis. This permits the use of the same small volumes as might be used for the radiochemical analyses. Data obtained with the HPLC assay agree well with those reported by others (3,9), who reported enzyme activities of DHO-DH in the same range as we do. The enzyme activity at pH 7.4 also agrees with data reported by others (3). Also the subcellular distribution reported previously (3-6,13) agrees with the present observations with the HPLC assay indicating the validity of the assay. The apparent K, value of about 12 PM obtained with our HPLC assay is in the same range as the values reported for fractionated rat liver mitochondria, 5.2 PM (3), intact murine liver mitochondria, 6.7 and 10 PM (11,13), and human spleen mitochondria, 5.3 PM (12). A higher K, (180 PM) was reported by Kennedy (4), but it was determined under conditions which did not allow initial rate determination. In a subsequent paper the same authors (20) reported a lower K, (2 PM) for the purified enzyme. These authors also reported that detergents and chaotropic agents usually employed for the isolation of this enzyme from mitochondria, affected enzyme kinetics (20). Biphasic kinetics for DHO-DH have not been reported previously. However, K,,,
38
PETERS ET AL.
determinations for DHO-DH were usually determined in a concentration range lower than 75 PM (1 I- 13) but not at higher concentrations, probably because this DHO concentration is above physiological levels and will not affect regulation of pyrimidine de nova nucleotide synthesis in vivo. An accumulation of DHO might only be caused by a lack of oxygen or its equivalent (7,21) or when inhibition of OPRT leads to an accumulation of erotic acid, which will inhibit DHO-DH (2 1). The HPLC assay has now been applied to study the mechanism of action of a new quinoline carboxylic acid derivative, DUP 785 (NSC 368390), which has shown antitumor activity against a variety of human xenografts of different histological origin and against murine tumors and L12 10 leukemia (22). Preliminary studies demonstrated that this compound has antipyrimidine effects by inhibition of DHO-DH from various sources (23,24). With our newly developed HPLC assay we could analyze the mode of inhibition of DHO-DH (24). In conclusion, we developed a new and sensitive assay for DHO-DH with the particular advantage of low backgrounds. The amount of erotic acid formed was measured using HPLC with an ion-exchange column. The method produced enzyme activities and kinetic values comparable to those obtained with previously described spectrophotometric and radiometric assays.
3. Chen, J.-J., and Jones, M. E. (1976) Arch. Biochem. Biophys. 176, 82-90. 4. Kennedy, J. (1973) Arch. B&hem. Biophys. 157, 369-373.
5. Miller, R. W., Kerr, C. T., and Curry, J. R. (1968) Canad. J. Biochem. 46, 1099-I 106. 6. Peters, G. J., and Veerkamp, J. H. (I 984) Adv. Exp. Med. Biol. 165A, 531-534. 7. Lijffler, M. (1980) Eur. J. Biochem. 107,207-215. 8. Forman, H. J., and Kennedy, J. (1976) Arch. Eiothem. Biophys. 173,219-224. 9. Dileepan, K. N., and Kennedy, J. (1983) FEBS Left 153, 1-5.
IO. Westwick, W. J., Allsop, J., and Watts, R. W. E. (1972) Biochem. Pharmacol. 21, 1955-1966. I I. Bennett, L. L.. Smithers, D., Rose, L. M.. Adamson, D. J.. and Thomas, H. J. (1979) Cancer Rex 39, 4868-4874. 12.
13.
14.
15. 16. 17.
18. 19.
Smithers, G. W.. Gero, A. M.. and O’Sullivan, W. J. A. (1978) Anal. Biochem. 88,93-103. Kensler. T. W., Cooney, D. A., Jayaram, H. N., Schaeffer, C., and Choie, D. D. (1981) Anal. Biothem. 117, 3 15-3 19. Leyva, A., Appel, H., Smith, P., Lankelma, J., and Pinedo, H. M. (1981) Cancer Lett. 12, 169-173. Peters, G. J., Oosterhof, A., and Veerkamp, J. H. (1983) Int. J. B&hem. 15, 51-55. Peters, G. J., Laurensse, E., Lankelma, J. Leyva. A., and Pinedo, H. M. ( 1984) Eur. J. Cancer C/in. Oncof. 20, 1425-1431. De Frees, S. A., Sawick, D. P., Cunningham, B. A., Morre, D. J., Cassidy, J. M., and Heinstein, P. F. (1984) Proc. Amer. Assoc. Cancer Res. 25, 19 (Abstract 73). Sedmak, J. J., and Grossberg, S. E. (1977) Anal. Biochem. 79, 334-352. Janeway, L. M., and Cha, S. (197 I) Cancer Res. 37, 4382-4388.
Forman, H. J.. and Kennedy, J. (1977) J. Biol. Chem. 252, 3379-338 1. 21. Chen, J.-J., and Jones, M. E. (1979) J. Biol. Chem. 20.
254,4908-49
ACKNOWLEDGMENTS This work was supported by the Dutch Cancer Foundation. “Koningin Wilhelmina Fends,” Grant IKA 83- 16. The authors thank S. L. Sharma for his contributions to this work.
REFERENCES 1. Jones, M. E. (1980) Annu. Rev. Biochem. 49, 253-279. 2. Traut, T. W. (1982) Trends Biochem. Sci. 7, 255-257.
22.
14.
Dexter, D. L., Hesson, D. P., Ardecky, R. J., Rao, G. V., Tippett, D. L., Dusak, B. A., Paull, K. D., Plowman, J., De Larco, B. M., Narayanan, V. L., and Forbes, M. (1985) Cancer Rex 45, 5563-5568.
Chen, S. F., Ruben, R. L., and Dexter, D. L. (1986) Proc. Amer. Assoc. Cancer Res. 27,298 (Abstract 1183). 24. Peters, G. J., Laurensse, E. L., Sharma, S. L., Leyva, A., and Pinedo, H. M. (1986) in Proceedings, 14th International Cancer Congress, Budapest, Vol. 1, p. 163 (Abstract 615). Karger, Basel; Akademia Kiadb, Budapest. 23.