ARCHIVES OF BIOCHEMISTRY AND BIOPHYSICS

Vol. 324, No. 1, December 1, pp. 99104, 1995

Purication and Characterization of Allantoate Amidohydrolase from Bacillus fastidiosus

Zhiwei Xu,1,2 Federico E. de Windt, and Chris van der Drift3
Department of Microbiology and Evolutionary Biology, Faculty of Science, University of Nijmegen, Nijmegen, The Netherlands

Received May 3, 1995, and in revised form September 17, 1995

Allantoate amidohydrolase from Bacillus fastidiosus was puried 170-fold to homogeneity as judged by isoelectric focusing and nondenaturing and sodium dodecyl sulfate polyacrylamide gel electrophoresis. The molecular mass was estimated to be 128 kDa. The enzyme appeared to be a homodimer with a subunit molecular mass of 66 kDa. The enzyme has an isoelectric point of 5.6. Allantoate amidohydrolase is a Mn2/-dependent enzyme exhibiting a pH optimum around 8.8. Its Km value for allantoate was estimated to be 9 mM. Similar to other microbial allantoate amidohydrolases the enzyme can be reversibly activated and inactivated. No indication for the involvement of arginine, lysine, and cysteine residues in the catalytic action of the enzyme was obtained. Diethylpyrocarbonate strongly inhibited the enzyme activity, indicating the involvement of histidine or tyrosine residues in catalytic action. However, no recovery was obtained by treatment with hydroxylamine as would be expected if such residues were modied. The enzyme could be reversibly denatured by urea, guanidine, and sodium dodecyl sulfate. 1995 Academic Press, Inc. Key Words: allantoate amidohydrolase; Bacillus fastidiosus; purication; purine degradation; ureides.

Bacillus fastidiosus is a common soil organism, whose name was given because of its unusual limitation in the utilization of substrates for growth (1). B. fastidiosus can grow on the purine degradation prod1 Supported by a Marie Curie Initiative Fellowship from the European Community. 2 Permanent address: Fujian Institute of Subtropical Botany, Xiamen 361006, Peoples Republic of China. 3 To whom correspondence should be addressed at Department of Microbiology and Evolutionary Biology, Faculty of Science, University of Nijmegen, Toernooiveld 1, NL-6525 ED Nijmegen, The Netherlands. Fax: /31 24 353450.

ucts urate, allantoin, and allantoate as sole carbon and nitrogen source and on glycerol as a carbon source in the presence of the above-mentioned compounds, ureidoglycolate, urea, or ammonia as a nitrogen source (2 6). All attempts to grow the organism on other carbon and nitrogen sources, including tricarboxylic acid cycle intermediates, sugars, alcohols, amino acids, and nitrate, failed (16). Moreover, growth on complex media such as tryptone soy broth, nutrient broth, and brain heart infusion (all tested at 3%, w/v) was very limited or absent (3). All strains of B. fastidiosus studied so far decompose urate along the same catabolic pathway (2, 3). Urate is degraded to allantoate by the successive action of uricase (urate:oxygen oxidoreductase, EC 1.7.3.3) and allantoinase (allantoin amidohydrolase, EC 3.5.2.5). Allantoate amidohydrolase [allantoate amidinohydrolase (decarboxylating), EC 3.5.3.9) catalyzes the conversion of allantoate in 2 mol of ammonia, 1 mol of carbon dioxide, and 1 mol of ureidoglycolate (Fig. 1). Ureidoglycine was shown to be an intermediate in this reaction and most likely this unstable intermediate was also hydrolyzed by allantoate amidohydrolase (7). This enzyme has been found both in bacteria growing anaerobically (e.g., Streptococcus allantoicus, Proteus rettgeri, Escherichia coli) on allantoin and in bacteria growing aerobically (B. fastidiosus, Pseudomonas acidovorans, Alcaligenes eutrophus) on this purine degradation product (7). Also in soybean, yam, and sweet potato allantoate amidohydrolase has been implicated in the breakdown of ureides (8, 9). Another enzyme known to degrade allantoate is allantoicase (allantoate amidinohydrolase, EC 3.5.3.4), which degrades allantoate to equimolar amounts of urea and ureidoglycolate (Fig. 1). This enzyme has been found in animals and in algae, yeasts, fungi, and bacteria growing aerobically on allantoin (7). In B. fastidiosus ureidoglycolate is degraded by ureidoglycolase (ureidoglycolate urea-lyase, EC 4.3.2.3) to glyoxylate and urea. Urea may either be
99

0003-9861/95 $12.00 Copyright 1995 by Academic Press, Inc. All rights of reproduction in any form reserved.

100

XU, DE WINDT, AND VAN DER DRIFT

FIG. 1. Degradation routes of allantoate in microorganisms.

an end product of the degradation or is further split into ammonia and carbon dioxide by urease (urea amidohydrolase, EC 3.5.1.5)-positive strains (3). Glyoxylate is incorporated into cell carbon (4, 10). Allantoate amidohydrolases from bacterial origin have one unique property in common: they contain two different binding sites for Mn2/ ions in the catalytic center of the enzyme. Binding at one site results in an inactive enzyme, while binding at the other site is essential for catalytic activity (1115). When the ions are bound to the noncatalytic site, binding to the catalytic site is prevented. Between pH 5 and 8 Mn2/ ions are exclusively bound to the noncatalytic site in the absence of substrate, resulting in an inactive enzyme. This enzyme can be reactivated by treatment at pH values below 5 or at pH values around 6 in the presence of complexing anions such as oxalate or citrate. A number of other anions, e.g., phosphate, are also able to activate the enzyme, but only in the presence of EDTA (1114). These treatments remove Mn2/ ions from the noncatalytic site and the metal-free enzyme can be brought into the active form by addition of substrate and Mn2/ ions. This reversible activation/inactivation phenomenon also explains why the activity of allantoate amidohydrolase in cell-free extracts of bacterial cells grown on allantoin usually is low, but can be enhanced considerably by the above-mentioned pH pretreatments. Thus far (1115, 18, 20, 21) all extracts

were prepared at pH values around 7.2 in the absence of metal-chelating anions, and apparently sufcient Mn2/ ions came out of the cells to inactivate the enzyme. Only the enzyme from S. allantoicus has been partly puried and characterized (14). No detailed studies on allantoate amidohydrolase from other microorganisms and plants were reported. In this communication we describe for the rst time the complete purication of allantoate amidohydrolase. B. fastidiosus was chosen as a source for this enzyme since the organism grows rapidly on allantoin and the activity of allantoate amidohydrolase in cell extracts is at least 6 times higher compared to other microorganisms.
MATERIALS AND METHODS
Organism and cultivation. B. fastidiosus (DSM 83) was grown in batch cultures at 37C with rotary shaking (200 rpm) in a mineral salts medium (5) supplemented with 30 mM allantoin as the sole carbon and nitrogen source. The pH after sterilization was 7.2. Preparation of cell-free extracts. Cells were harvested at the end of the exponential phase and washed twice with 50 mM TrisHCl buffer (pH 7.5) containing 0.17 mM EDTA (Buffer A). Cells were broken by two successive passages of a cell suspension in Buffer A through a French pressure cell at 138 MPa. The supernatant obtained after centrifugation at 4C for 30 min at 43,000g was used as cell-free extract. The protein content was determined with the BioRad protein assay kit (Bio-Rad Laboratories, Inc., Richmond, CA) and bovine serum albumin as the standard.

ALLANTOATE AMIDOHYDROLASE FROM B. fastidiosus Allantoate amidohydrolase assay. The enzyme was assayed by measurement of the amount of ureidoglycolate formed from allantoate. Ureidoglycolate was determined with the differential analyses of glyoxylate derivatives (16). The standard deviation of this method was less than 2%. All assays were done in triplicate. The incubation mixture contained, unless stated otherwise, per milliliter 50 mmol sodium allantoate, 0.15 mmol MnSO4 , 1.06 mmol glutathione, 150 mmol diethanolamineHCl buffer (pH 8.8), and an appropriate amount of enzyme (between 5 and 15 mg crude protein). Incubation was at 37C. One unit (U) of enzyme activity is dened as the amount of enzyme which catalyzes the formation of 1 mmol ureidoglycolate per minute under the incubation conditions used. Purication. All purication procedures were performed at 4C. Crude cell-free extract was adjusted to 40, 55, and 75% (NH4)2SO4 saturation by dropwise addition of a neutralized aqueous saturated (NH4)2SO4 solution. After gently stirring for 30 min the precipitated proteins were collected by centrifugation for 30 min at 40,000g and taken up in Buffer A containing 1 M (NH4)2SO4 . The 5575% fraction, which contained almost all activity, was applied to a Fractogel TSKbutyl 650(S) column (1.6 1 11 cm; ow rate 1 ml min01; E. Merck AG, Darmstadt, Germany) equilibrated with Buffer A containing 1 M (NH4)2SO4 . Bound proteins were eluted by 80 ml Buffer A containing 0.9 M (NH4)2SO4 followed by a 160-ml linear gradient of 0.9 to 0 M (NH4)2SO4 in Buffer A and 160 ml Buffer A. Active fractions were pooled and freed of (NH4)2SO4 by extensive washing with Buffer A in an Amicon ultraltration unit equipped with a PM 30 lter. The desalted pool was loaded on a Q-Sepharose Fast Flow column (2.2 1 11 cm; ow rate 2 ml min01; Pharmacia, Uppsala, Sweden) equilibrated with Buffer A. The column was washed with 40 ml Buffer A, 80 ml Buffer A with 0.2 M NaCl, a 80-ml linear gradient of 0.20.3 M NaCl in Buffer A, and nally 80 ml Buffer A with 0.3 M NaCl. The active fractions were collected and freed of NaCl by extensive washing with Buffer A in an ultraltration unit equipped with an Amicon PM 30 lter. This pool was applied to a Fractogel EMD DEAE-650(M) column (1.5 1 21 cm; ow rate 1 ml min01; E. Merck AG). The column was washed successively with 20 ml Buffer A, 60 ml Buffer A with 0.22 M NaCl, a 60-ml linear gradient of 0.22 0.27 M NaCl in Buffer A, and 60 ml Buffer A with 0.27 M NaCl. Active fractions were collected and concentrated on an Amicon PM 30 lter. The concentrated enzyme pool (2.9 ml) was divided into three parts which were loaded on a Superose 12 prep-grade column (1.6 1 50 cm; ow rate 0.5 ml min01; Pharmacia). The proteins were eluted with Buffer A and the active fractions of all three columns were pooled. As a nal step this pool was applied to a Protein Pak DEAE-5PW column (0.75 1 7.5 cm; ow rate 1 ml min01; TosoHaas, Montgomeryville, PA). The column was washed successively with 20 ml Buffer A, 30 ml Buffer A with 0.1 M NaCl, and a 60-ml linear gradient of 0.10.3 M NaCl in Buffer A. The active fractions were pooled and freed of NaCl by washing with Buffer A on an Amicon PM 30 lter. The puried enzyme was stored at 020C. Electrophoretic methods. Gel electrophoresis was performed on precast slabgels using Pharmacia Phast-System equipment (Pharmacia). Native PAGE4 was performed either on gradient gels (8 25%) or on homogeneous gels (20%). Urease (Mr 545,000 and 272,000), bovine serum albumin (Mr 132,000 and 66,000), chicken egg albumin (Mr 45,000), and a-lactalbumin (Mr 14,200) were used for calibration. For SDSPAGE a 1015% gradient gel was used. A mixture of rabbit muscle phosphorylase B (Mr 97,400), bovine serum albumin (Mr 66,000), chicken egg albumin (Mr 45,000), bovine carbonic anhydrase (Mr 31,000), soybean trypsin inhibitor (Mr 21,500), and hen egg white lysozyme (Mr 14,400) was used for calibration. Isoelectric focusing was performed on IEF gels with a pH range 3 9. Markers used were phycocyanin (pI 4.65), b-lactoglobulin (pI 5.1),

101

bovine carbonic anhydrase (pI 6.0), human carbonic anhydrase (pI 6.5), equine myoglobin (pI 7.0), human hemoglobin C (pI 7.5), lentil lectin (three bands; pIs 7.8, 8.0, and 8.2), and cytochrome c (pI 9.6). PAGE gels were stained with Coomassie brilliant blue (PhastSystem Development Technique File No. 200; sensitivity about 20 ng protein/ band), and IEF gels were stained with silver (PhastSystem Development Technique File No. 210; sensitivity about 15 ng protein/band). Chemical modication. The involvement of certain amino acid residues in the catalytic action of the enzyme was studied by means of chemical modication. For lysine, arginine, and cysteine the specic reagents 2,3-butanedione and glyoxal, pyridoxal 5-phosphate and 2,4,6-trinitrobenzene sulfonic acid, and p-chloromercuribenzoate, Nethylmaleimide, and iodoacetamide were used, respectively. Diethylpyrocarbonate was used as modifying reagent for histidine and tyrosine residues. Reactivation of diethylpyrocarbonate-inactivated enzyme was attempted with hydroxylamine (17). Since hydroxylamine at concentrations above 5 mM in the incubation mixture interfered with the ureidoglycolate analysis, the compound was removed by extensive washing of the treated enzyme preparation with Buffer A in an ultraltration unit. Chemicals. Sodium allantoate was prepared by alkaline hydrolysis of allantoin (18, 19). Sodium ureidoglycolate was prepared according to Trijbels and Vogels (20). Allantoin was purchased from E. Merck AG. All other chemicals were of the highest purity commercially available.

RESULTS

Activation and Inactivation of Allantoate Amidohydrolase The enzyme present in extracts of B. fastidiosus cells prepared in Buffer A was for the greater part in its active form. Pretreatment of the extracts with phosphate buffer, pH 6, containing 0.2 mM EDTA for various times at 4C resulted maximally in an increase of 1.6-fold. Previously cell extracts were prepared in phosphate buffer (pH 7.2) and in that case the activity of the untreated enzyme preparation amounted only to 1.5% of that of the fully activated enzyme. However, the specic activities of activated enzyme were found to be about the same irrespective of the nature of the buffer used for extract preparation: with phosphate buffer (pH 7.2) and TrisHCl buffer (pH 7.5) containing 0.17 mM EDTA values of 22.4 (3) and 44.2 mmol/ min/mg protein were found, respectively. A 60-fold dilution of the extract prepared in Buffer A (protein concentration 106 mg) with 50 mM TrisHCl buffer (pH 7.5) followed by a 2-h incubation at 4C resulted in a 43% decrease of the activity. After this treatment the enzyme could be reactivated by incubation for 5 min at 4C with 50 mM phosphate buffer (pH 6.0) containing 0.2 mM EDTA to the same level as untreated enzyme. Reactivation of the inactivated enzyme could also be achieved, although much slower, at pH 7.5 in TrisHCl buffer by addition of 0.2 mM EDTA. A 60-fold dilution of the enzyme with Buffer A, which contains EDTA, or dilution with phosphate buffer (pH 6.0) containing 0.2 mM EDTA followed by incubation of the diluted enzyme at 4C had no effect on the activity. However, at 30C activity was gradually lost on incubation: after 20 min

Abbreviations used: PAGE, polyacrylamide gel electrophoresis; SDS, sodium dodecyl sulfate; IEF, isoelectric focusing.

102

XU, DE WINDT, AND VAN DER DRIFT TABLE I

Purication of Allantoate Amidohydrolase from Bacillus fastidiosus DSM83

Total protein (mg) 921 391 171 29.2 5.26 1.51 0.07 Total activity (units) 39,438 37,238 36,651 20,383 15,589 5,791 509 Specic activity (U/mg) 42.8 95.2 214.3 698 2964 3835 7271 Purication factor 1.0 2.2 5.0 16.3 69.2 89.6 170

Step Crude extract (NH4)2SO4 precipitation Fractogel TSK-butyl Q-Sepharose Fractogel EMD DEAE Superose 12 Protein Pak DEAE

Yield (%) 100 94.4 92.9 51.7 39.5 14.7 1.3

the activity in both cases was decreased for about 50% and thereafter remained at this level. To prevent inactivation at pH 7.5 and 4C at least 80 mM EDTA had to be present in the Tris buffer. Optimal Reaction Conditions of the Allantoate Amidohydrolase Reaction The optimal conditions for the allantoate amidohydrolase activity were determined mainly with a crude enzyme preparation. The enzyme showed a broad pH optimum: maximal activity was observed at pH values between 8.5 and 10, while at pH 7.2 half-maximal activity was found. Borate buffers were very inhibitory: activity was inhibited 80% in the presence of 10 mM borate. The optimal temperature for enzymic action was between 40 and 45C. Above 50C activity was lost rapidly; after a heating period of 5 min half-maximal activity was found at 54C. At 25C about 75% activity was observed. The reaction was stimulated by the addition of glutathione: at the optimal glutathione concentration (1 mM) the activity was stimulated 35%. Mn2/ ions were essential for activity. The optimal concentration was found to be 0.1 mM. Addition of 10 mM EDTA to the incubation mixture at pH 8.8 completely inhibited the reaction; activity could be restored by addition of Mn2/ ions. Purication of Allantoate Amidohydrolase The enzyme was puried 170-fold to apparent homogeneity in a six-step procedure as outlined in Table I. The most active fractions obtained from Protein Pak DEAE-5PW were pooled and subjected to native PAGE. Only one band was observed after staining for protein both on a 20% gel (Fig. 2) and on a 825% gel. The puried enzyme had an apparent molecular mass of 128 kDa. SDSPAGE of the puried enzyme showed a single band with an apparent molecular mass of 66 kDa. These results indicate that allantoate amidohy-

drolase from B. fastidiosus is a homodimer. The puried enzyme was also found to be homogeneous when analyzed by IEF. The isoelectric point was estimated to be 5.6. Partial Characterization of the Pure Enzyme The enzymic reaction followed simple Michaelis Menten kinetics. An apparent Km value for allantoate of 9 mM was calculated and a Vmax of 16,667 U/mg protein which corresponds to a kcat of 32,778 s01. No activity loss occurred when the enzyme was preincubated in the dark for 30 min at 4C with 2,4,6-trinitrobenzene sulfonate or pyridoxal phosphate, both at 10 mM. Also, preincubation with 2,3-butanedione and glyoxal, both at 25 mM, for 3 h in the dark at 25C did not affect the enzymic activity. p-Chloromercuribenzoate, N-ethylmaleimide, and iodoacetamide, present at 1 mM in incubation mixtures without glutathione, had no ef-

FIG. 2. PAGE (20%) analysis of allantoate amidohydrolase after TSK DEAE-5PW chromatography. Lane 1 represents calibration proteins (see bars); lane 2 is the puried enzyme (40 ng of protein).

ALLANTOATE AMIDOHYDROLASE FROM B. fastidiosus

103

fect on the activity. Preincubation of the enzyme with diethylpyrocarbonate for 10 min at 20C resulted in a residual activity of 4%. Treatment with 250 mM hydroxylamine did not restore the activity as would be expected if histidine or tyrosine residues were involved in the catalytic action of the enzyme (17). A similar treatment of the enzyme with hydroxylamine, but in the absence of diethylpyrocarbonate, did not affect the enzyme activity. The enzyme was very resistant to urea and guanidine. Urea (4 M) present during the incubation did not affect the activity. Inhibitions of 33 and 68% were found in the presence of 5 and 6 M urea, respectively. Pretreatment with 5 M urea for 10 min at 4C reduced the activity by 30%. A 50-fold dilution of the urea-treated enzyme resulted in a recovery of the activity to 95% of that of the untreated enzyme. Also inhibition by guanidine and SDS was reversible. Pretreatment of the enzyme for 30 min at 4C with 1 and 5 M guanidine resulted in an activity loss of 60 and 97%, respectively. When the 5 M guanidine-inactivated enzyme was diluted 50-fold and incubated for 2 min at 4C a recovery of activity to 86% of that of the untreated enzyme was observed. Pretreatment with 2.5 mM SDS for 10 min at 4C resulted in an activity loss of 90%. A 50-fold dilution of the SDS-treated enzyme followed by an incubation for 5 min at 4C restored the activity to 82% of the untreated enzyme. The enzyme was not inhibited by NaN3 and KCN, both at 1 mM, but 1 mM NaF resulted in an inhibition of 20% of the activity.
DISCUSSION

Remarkable at rst sight was the very high activity of the enzyme in cell-free extracts of B. fastidiosus. Previously (3) it was found that cell extracts showed only a very low activity and that this activity could be enhanced considerably, about 60- to 70-fold, by pretreatment at pH 6 with both phosphate and EDTA. However, such cell extracts were prepared in EDTAfree phosphate buffer, while we prepared the extracts in Tris buffer containing EDTA. Dilution of the extracts with EDTA-free Tris buffer resulted in an inactivation of the enzyme. Inactivated enzyme could be reactivated by incubation with EDTA-containing phosphate buffer (pH 6) or, albeit slower, at pH 7.5 with EDTA-containing Tris buffer. These results are in accordance with the previously formulated hypothesis (1315) that inactivation is due to binding of Mn2/ ions to a noncatalytic site, whereby binding of these ions to the catalytic site is prevented. Consequently, no enzyme activity is found. The incorrectly bound ions can be removed either by treatment at low pH values or around pH 6 in the presence of complexing agents (11, 1315). All microbial allantoate amidohydrolases studied so far show this reversible activation/inactivation behavior (7). For the plant enzyme no studies were reported concerning this phenomenon (8, 9).

B. fastidiosus is the rst organism from which allantoate amidohydrolase has been puried to apparent homogeneity. The enzyme appeared to consist of two identical subunits and has a molecular mass of 128 kDa. The pI was found to be 5.6. The values of kcat and kcat/ Km place the enzyme among a small group of enzymes that are diffusion-controlled. The data obtained in experiments with amino acid-modifying reagents indicate that the enzyme does not contain arginine, lysine, and cysteine residues that are reactive with these reagents; most likely these residues are not involved in the catalytic action of the enzyme. Whether histidine and tyrosine residues are involved is unclear. Although diethylpyrocarbonate strongly inhibited the activity, this activity was not recovered upon treatment with hydroxylamine as would be expected (17). Hydroxylamine did not affect the activity of the enzyme as was previously also demonstrated for the S. allantoicus enzyme (van der Drift, unpublished results). The enzyme has a high pH optimum, around pH 8.8, and Mn2/ ions are essential for activity as was observed for the other microbial allantoate amidohydrolases (7). Glutathione is stimulatory, but not required for activity. A similar observation was made for the enzyme from S. allantoicus (21). The strong inhibition of enzyme activity by borate ions was also observed for the enzyme from S. allantoicus (21) and soy bean (22). The reversible inhibition by urea and the inhibition by uoride ions, even in the absence of phosphate, was also noted with the S. allantoicus enzyme (21). Asparagine (10 mM), which is an inhibitor of the soy bean enzyme (22), had no effect on the activity of the B. fastidiosus enzyme (data not shown).
REFERENCES
1. Den Dooren de Jong, L. E. (1929) Zentralbl. Bakteriol. Abt 2 79, 344358. 2. Kaltwasser, H. (1971) J. Bacteriol. 107, 780786. 3. Bongaerts, G. P. A., and Vogels, G. D. (1976) J. Bacteriol. 125, 689697. 4. Braun, W., and Kaltwasser, H. (1979) Arch. Microbiol. 121, 129 134. 5. Van der Drift, C., Smits, R. A. M. M., Michiels, G. A. M., and Op den Camp, H. J. M. (1986) Arch. Microbiol. 146, 292294. 6. Muruke, M. S. H., Op den Camp, H. J. M., and van der Drift, C. (1993) FEMS Microbiol. Lett. 114, 195200. 7. Vogels, G. D., and van der Drift, C. (1976) Bacteriol. Rev. 40, 403468. 8. Winkler, R. G., Blevins, D. G., Polacco, J. C., and Randall, D. D. (1987) Plant Physiol. 83, 585591. 9. Osuji, G. O., and Ory, R. L. (1987) J. Agric. Food Chem. 35, 219 223. 10. De Windt, F. E., and van der Drift, C. (1980) Biochim. Biophys. Acta 613, 556562. 11. Vogels, G. D. (1966) Biochim. Biophys. Acta 113, 277291. 12. Trijbels, F., and Vogels, G. D. (1966) Biochim. Biophys. Acta 113, 292301.

104

XU, DE WINDT, AND VAN DER DRIFT 18. Vogels, G. D. (1963) Ph.D. Thesis, Technological University, Delft, The Netherlands. 19. Hermanowicz, W. (1948) Rocz. Chem. 22, 159180. 20. Trijbels, F., and Vogels, G. D. (1966) Biochim. Biophys. Acta 118, 387395. 21. Van der Drift, C. (1968) Ph.D. Thesis, University of Nijmegen, Nijmegen, The Netherlands. 22. Lukazewski, K. M., Blevins, D. G., and Randall, D. D. (1992) Plant Physiol. 99, 16701676.

13. Van der Drift, C., and Vogels, G. D. (1969) Enzymologia 36, 269 277. 14. Van der Drift, C., and Vogels, G. D. (1967) Biochim. Biophys. Acta 139, 162168. 15. Van der Drift, C., and Vogels, G. D. (1969) Enzymologia 36, 278 286. 16. Vogels, G. D., and van der Drift, C. (1970) Anal. Biochem. 33, 143157. 17. Pelton, P. D., and Ganzhorn, A. J. (1992) J. Biol. Chem. 267, 59165920.