Further characterization of L -β-hydroxyacid dehydrogenase from Drosophila

Biochimica et Biophysica Acta 841 (1985) 15-21 Elsevier

15

BBA 22084

Further characterization of L-fl-hydroxyacid d e h y d r o g e n a s e from Drosophila M a r i l y n M e n o t t i - R a y m o n d a n d D a v i d T. Sullivan * Department of Biology, Syracuse University, Syracuse, N Y 13210 (U.S.A.) (Received January 2nd, 1985) (Revised manuscript received April 9th, 1985)

Key words: L-fl-Hydroxyacid dehydrogenase; Amino acid analysis; ( Drosophila )

L-fl-Hydroxyacid dehydrogenase (L-fl-hydroxyacid-NAD-oxidoreductase, EC 1.1.1.45) of Drosophila is composed of two, identical subunits with a molecular weight of approx. 33300. The enzyme was purified 938-fold from Drosophila melanogaster. An isoelectric point of 8.6 was determined for L-fl-hydroxyacid dehydrogenase. An amino acid analysis was conducted of the purified enzyme. A single subunit was obtained by SDS-gel electrophoresis of the purified enzyme. Translation of larval and adult mRNA in a mRNAdependent reticulocyte lysate, followed by immune precipitation using anti-L-fl-hydroxyacid dehydrogenase lgG revealed a single L-fl-hydroxyacid dehydrogenase subunit of 33 300. Larval and adult proteins were the same size. The enzyme does not appear to be subjected to substantial post-translational modifications. Introduction

Borack and Sofer [3] have reported a native molecular weight of 63 000 for L-fl-hydroxyacid dehydrogenase based on evidence from density gradient centrifugation and gel filtration. As a result of SDS-polyacrylamide gel electrophoresis of the enzyme, Cannistraro et al. [4] reported that L-fl-hydroxyacid dehydrogenase is composed of two, non-identical subunits with molecular weights of 40000 and 23500. We have a biochemical genetic interest in this enzyme. If L-fl-hydroxyacid dehydrogenase is composed of two dissimilar subunits there could be two genes coding for the enzyme. Simple dehydrogenases have usually been found to be homopolymers. With this in mind we wished to reinvestigate the subunit structure of L-fl-hydroxyacid dehydrogenase and search for any evidence of multiple genes coding for L-fl-hydroxyacid dehydrogenase.

L-fl-Hydroxyacid dehydrogenase (L-fl-hydroxyacid-NAD-oxidoreductase, EC 1.1.1.45) is a soluble enzyme with broad substrate specificity that catalyzes the oxidation of several 3-hydroxyacids [1]. It is stereochemically specific for the 3-OH group in the L-configuration [1]. It has been reported in liver and kidney tissue of several vertebrates [2], and has been partially purified and characterized from Drosophila [3,4] and hog kidney [1]. Electrophoretic variants of L-fl-hydroxyacid dehydrogenase have recently been reported in Anopheles albimanus. [5] In mammals it is believed to be involved in the glucuronic pathway oxidizing L-gulonate to 3keto-L-gulonate [1]. Its metabolic role in Drosophila is currently unknown, though it has been shown to be dispensable to the organism through the study of null mutants [6].

Materials and Methods

* To whom correspondence should be addressed. Abbreviation: Hepes, 4-(2-hydroxyethyl)-l-piperazineethanesulfonic acid.

Materials. Carboxymethylcellulose (CM52) was purchased from Whatman Chemical Separation. Polybuffer exchanger PBE94 and Polybuffer 96

0304-4165/85/$03.30 © 1985 Elsevier Science Publishers B.V. (Biomedical Division)

16 were purchased from Pharmacia Fine Chemicals. Hydroxylapatite (Bio-Gel HTP) was purchased from Bio-Rad. Protein standards were purchased from Bethesda Research Laboratories and BioRad. ~4C-labeled protein molecular weight standards for gel electrophoresis were purchased from Bethesda Research Laboratories. L-[35S]Methionine (1052.4 C i / m m o l ) and E N 3 H A N C E were purchased from New England Nuclear. D-Gluconate, bovine serum albumin, /3-nicotinamide adenine dinucleotide (NAD +), and dithiothreitol were purchased from Sigma. Heat-killed, formalin-fixed S. aureus was a gift from Dr. Judith Foster. Organisms. The strain Oregon R of D. melanogaster was used as the source of enzyme. Flies were cultured and harvested according to a procedure described by Sullivan et al. [7]. Flies were quickly frozen in liquid nitrogen and stored at - 8 0 ° C prior to use. Preparation of crude extract. All procedures were performed at 4°C. All centrifugations were performed in a Sorvall SS-34 rotor at 20000 x g. All buffers utilized in preparation of the crude extract and in subsequent column chromatography procedures were supplemented with the following components which acted as stabilizers for the enzyme: 1 - 1 0 4 M dithiothreitol; 1-10 . - 4 M EDTA; 110 6 M NAD +. 50 g of flies were ground to a fine powder in liquid nitrogen in a pre-cooled mortar and pestle and suspended in 200 ml 0.1 M sodium phosphate buffer (pH 7.5)/2% (v/v) phenylthiourea. The suspension was slowly stirred for 30 min to melt the slurry prior to filtration through nitex cloth and centrifugation. The supernatant from this step was adjusted to pH 5.1 by dropwise addition of glacial acetic acid, allowed to stand for 20 min and centrifuged for 20 min. The supernatant was adjusted to pH 5.6 and dialyzed for 17 h against 40 vol. of 50 mM sodium acetate buffer (pH 5.6). The dialysate was centrifuged for 20 min prior to applicati¢~ to a carboxymethyl cellulose column. Enzyme assay. Assays were carried out in a Gilford 240 spectrophotometer equipped with recorder, with cuvette chamber maintained at 25°C. L-/~-Hydroxyacid dehydrogenase activity was measured in an assay solution containing 125 mM Tris-HCl (pH 8.6), 50 mM D-gluconate, 25 mM

NAD + in a total volume of 2.5 ml. The reaction was initiated by addition of enzyme and followed by observing the reduction of NAD + at 340 nm. 1 unit of activity is defined as the reduction of 1 tLmol of NAD+ per min. Protein concentrations were determined by using the Bio-Rad protein assay solution and directions supplied by the manufacturer. Bovine serum albumin was used as a standard. Electrophoretic gels. Samples of the L-fl-hydroxyacid dehydrogenase purification procedure were examined by electrophoresis in a 5~15% gradient polyacryiamide vertical slab gel using 237 mM Tris/0.072 N HC1 as separating gel buffer, 39.5 mM Tris/0.064 M phosphoric acid as stacking gel buffer, 37.6 mM T r i s / 4 0 mM glycine as upper tank buffer and 63 mM T r i s / 5 0 mM HC1 as lower tank buffer. A current of 15 mA/gel was applied for 5 h. Proteins in half of the gel were fixed and stained with 25% ( v / v ) i s o p r o p a n o l / 1 0 % ( v / v ) acetic acid/0.025% (w/v) Coomassie blue. The other half of the gel was examined for L-/3-hydroxyacid dehydrogenase activity by immersing the gel in cold 125 mM Tris/HC1 (pH 8.6) for 30 min followed by immersion in 50 ml of the standard reaction mixture for assaying L-/~-hydroxyacid dehydrogenase plus 0.61 mM nitroblue tetrazolium and 0.03 mM phenazine methosulfate. When activity bands had reached the desired intensity, the gel was rinsed in distilled water. SDS-polyacrylamide gel electrophoresis was performed in vertical slabs with a 4% stacking and a 12.5% separating gel using the method and buffers of King and Laemmli [8]. A current of 15 m A / g e l was applied for 5-6 h. Gels of the final product of enzyme purification were fixed and stained for protein with 0.025% Coomassie b l u e / 25% (v/v) isopropanol/10% (v/v) acetic acid. The standard proteins used in this gel for molecular weight determination were bovine serum albumin (66200), ovalbumin (45 000), carbonic anhydrase (31000), soybean trypsin inhibitor (21500) and lysozyme (14400). Gels of translation products and immune precipitation products were fixed in 10% (w/v) trichloroacetic acid/10% (v/v) glacial acetic acid/20% ( v / v ) methanol and prepared for fluorography by soaking in EN~HANCE for 1 h followed by distilled water for 1 h. The gels were dried at 80°C under vacuum and exposed to pre-

17 flashed Kodak X-Omat AR film at - 80°C, 5 days for translation products and 2-3 weeks for immune precipitation products. Amino acid composition. Amino acid content of the purified enzyme was determined on a Beckman Model 3600C amino acid analyzer after 20 h of hydrolysis of sample in 6 M HC1 at 110°C. Preparation of mRNA. Drosophila larval and adult mRNA were prepared according to the procedure by Skuse and Sullivan [9]. Total RNA was stored precipitated at - 2 0 ° C in 2 vol. of 95% ethanol. Prior to translation the RNA was centrifuged at 27000 x g for 30 min at 4°C, dried under vacuum and dissolved in sterile water. Preparation of lysate. Rabbit reticulocyte lysates were made mRNA dependent according to a method by Pelham and Jackson [10]. Translation conditions. Translation of Drosophila RNA was carried out in a reaction volume of 26 ~l which consisted of 10 /~l of nuclease-treated reticulocyte lysate, 8 #l of supplement mixture (65 mM Hepes (pH 7.6), 6.5 mM dithiothreitol, 26 mM creatine phosphate, 1.63 mM spermidine (free base) and 227.5 /zM of each of the 19 essential amino acids, excluding methionine), 2 /~1 of a solution containing 1.82 M potassium acetate and 2.6 mM magnesium acetate, 2/~l of [35S]methionine and 4 /tl of RNA (2-4 /tg). The translation mixture was incubated at 30°C for 75 min. 1 ~1 of the translation reaction was utilized for electrophoretic analysis of the polypeptides synthesized. Immunological techniques, t-fl-Hydroxyacid dehydrogenase antibodies were raised in New Zealand rabbits. Approximately 1 mg of enzyme mixed with Freund's complete adjuvant was injected in a series of intramuscular injections in each thigh and subcutaneous injection in the back of the neck. A second series of injections of 800 ~g of L-fl-hydroxyacid dehydrogenase in Freund's incomplete adjuvant followed in 14 days. The serum titer of L-fl-hydroxyacid dehydrogenase antibodies was monitored by precipitation of L-fl-hydr.oxyacid dehydrogenase activity from crude extracts of adult Drosophila and animals were bled by cardiac puncture. An IgG fraction of rabbit serum was prepared. Heat-inactivated serum was centrifuged for 30 min at 12000 x g, and the resultant supernatant was brought to 40% saturation with ammonium sulfate.

After standing for 1 h at 4°C, it was centrifuged at 12000 x g for 45 min and the resulting pellet resuspended to the original volume of serum with 50 mM Tris-HC1 (pH 7.4)/150 mM sodium chloride. It was dialyzed against 600 vol. of the same buffer for 24 h and then centrifuged for 20 min at 10 000 x g. The serum was stored at - 2 0 ° C . Immunoprecipitation of the translation products followed the procedure of Foster et al. [11] utilizing heat-killed, formalin-fixed Staphylococcus aureus as an immunoadsorbent. In a modification of that procedure antisera was added upon the completion of translation and the reaction was allowed to sit for 2 h at room temperature. Two translations were then combined prior to the series of washes to ensure a detectable amount of L-flhydroxyacid dehydrogenase. Purification of L-fl-hydroxyacid dehydrogenase was accomplished simultaneously with purification for glycerol-3-phosphate dehydrogenase. The first column utilized in this procedure was carboxymethyl cellulose. Crude extract (197.32 g) was applied to a column of carboxymethylcellulose (23 x 2.8 cm) equilibrated with 50 mM sodium acetate buffer (pH 5.6) (flow rate, 60 ml/h). The fraction of sample that did not bind to carboxymethyl cellulose was collected after approx. 1 void vol. of buffer had passed through the column. The sample was washed through the column with 100 ml of buffer. Eluant was monitored for L-fl-hydroxyacid dehydrogenase activity. L-fl-Hydroxyacid dehydrogenase-containing sample (273.56 g) was then dialyzed against 40 vol. of 25 mM ethanolamine acetate buffer (pH 9.4) and then applied to a 46 x 1.8 cm chromatofocusing column of PBE 94 equilibrated with the same buffer. Unbound proteins were removed by washing the column with 100 ml of the same buffer. Sample was eluted from the column with a decreasing pH gradient from 9.0 to 6.0 using Polybuffer 96-acetate (flow rate, 60 m l / h ; 5.9 ml fractions). Activity eluted as a single symmetrical peak. The pH of the fractions containing peak activity was measured and found to be 8.6. This provides an estimate of the isoelectric point of the protein. The active fractions were pooled. Following this step, ampholytes of the polybuffer bound to L-flhydroxyacid dehydrogenase, were removed from the sample by dialysis for 8 h against 150 vol. of

18

Results and Discussion

o

2:

o

"6

- ~-

~0" ~0

2o-

FRACTION

;o-

~0- ;0

NUMBER

Fig. 1. Chromatography of Drosophila L-fl-hydroxyacid dehydrogenase on hydroxylapatite. Elution from the column was accomplished with a linear 500 ml gradient of 10-100 mM sodium phosphate buffer (pH 7.4) (flow rate. 30 ml/h). Fractions of 6.5 ml were collected. Activities were assayed and expressed in p, mol N A D + reduced per rain.

10 mM sodium phosphate/250 mM sodium chloride (pH 7.4), followed by 17 h of dialysis against 300 vol. of sodium phosphate (pH 7.4). The dialysate was applied to a hydroxylapatite column (26 × 1.8 cm) equilibrated with 10 mM sodium phosphate buffer (pH 7.4) (flow rate, 30 m l / h ; 6.5 ml fractions) and washed in with 30 ml of this same buffer. A linear 500 ml gradient of 10-100 mM sodium phosphate buffer (pH 7.4) was then applied. The active fractions were pooled and the final product was analyzed for homogeneity by electrophoresis. Fig. 1 is an elution profile of this column.

The results of a typical L-fl-hydroxyacid dehydrogenase purification is summarized in Table I. In this preparation a final 938-fold purification was attained, and the specific activity of the resultant protein was 63.8 U/mg. The total amount of enzyme isolated from 50 g of Drosophila was 0.329 mg at a final yield of 12.7%. Yields of up to 52% have been obtained from other preparations. The homogeneity of the preparation was tested by electrophoresis of sample in a 5-15% gradient nondenaturing polyacrylamide gel. A single band staining for protein produced on one half of the gel corresponded in position with a single band produced on the other half of the gel, indicating L-fl-hydroxyacid dehydrogenase activity by histochemical staining (Fig. 2). Further evidence of the purity of the L-fl-hydroxyacid dehydrogenase preparation was provided by electrophoresis of sample in 0.1% S D S / 12.5% polyacrylamide slab gels. A single polypeptide was observed (Fig. 3). This also presents evidence that L-fl-hydroxyacid dehydrogenase consists of subunits of equal size. The subunit molecular weight (35000 + 2179) for L-fl-hydroxyacid dehydrogenase was calculated from the mobility of the enzyme against molecular weight standards in this gel. This calculation is an average of three determinations. The distribution of [ 35S]methionine-labeled proteins directed by Drosophila larval and adult mRNA in a messenger-dependent reticulocyte lysate is seen in Fig. 4. A fluorograph of an SDS gel of the immune precipitation of these translation products using anti-L-fl-hydroxyacid dehydro-

TABLE I P U R I F I C A T I O N OF DROSOPHILA L-fl-HYDROXY A C I D D E H Y D R O G E N A S E

Centrifuged crude extract Acid precipitation (pH 5.1) Carboxymethylcellulose column Chromatofocusing column Hydroxylapatite column

Total protein (mg)

Total activity ( p, m o l / m i n )

Specific activity (/~ m o l / m i n / m g )

2443.35 710.35 218.84 4.78 0.33

165 110 53 43 21

0.068 0.154 0.24 9.0 63.8

Purification factor

Yield (%)

2.26 3.6 132.0 938.0

67 32 26 13

19

Fig. 2. Polyacrylamide gel electrophoresis of partially purified L-fl-hydroxyacid dehydrogenase. Electrophoresis of L-fl-hydroxyacid dehydrogenase was carried out in a 5-15% gradient nondenaturing polyacrylamide gel. The gel on the left was stained with Coomassie blue. The gel on the right was stained histochemically for L-fl-hydroxyacid dehydrogenase activity.

genase IgG is also seen in Fig. 4. A single protein band is observed in both larval and adult mRNAdirected translations. Larval and adult proteins are the same size. This further supports our previous

Fig. 3. SDS-polyacrylamide gel electrophoresis of partially purified Drosophila L-~8-hydroxyacid clehydrogenase. Electrophoresis of partially purified L-fl-hydroxyacid dehydrogenase was carried out for 6 h at 15 mA in a 4% stacking and 12.5% separating SDS-polyacrylamide vertical slab gel. The numbers at the left of the figure indicate the positions of molecular weight standards (bovine serum albumin, 66200; ovalbumin, 45000; carbonic anhydrase, 31000; soybean trypsin inhibitor, 21500; lysozyme 14400).

Fig. 4. (A) Fluorograph of an SDS-polyacrylamide gel of

Drosophila larval (Lv) and adult (A) mRNA translation products using a mRNA-dependent reticulocyte lysate with [3SS]methionine. (B) Fluorograph of an SDS-polyacrylamide gel of immunoprecipitates of Drosophila larval (Lv) and Adult (Ad) translation products using anti-L-fl-hydroxyacid dehydrogenase-lgG. The numbers at the right of the figure indicate the positions of molecular weight standards (bovine serum albumin, 68000; ovalbumin, 43000; chymotrypsinogen, 25700; lactoglobulin, 18400).

conclusion that the enzyme is homodimeric. The subunit molecular weight of this in vitro translated product has been determined to be 33 300 _+ 549. This calculation is an average of six determinations. There is no statistically significant difference between the subunit weight of 33 300 determined for the in vitro translated enzyme and the subunit weight of 35 000 determined for the in vivo synthesized enzyme ( P > 0.05 but P < 0.10). We will therefore report a range of 33 300-35000 for the weight of an L-fl-hydroxyacid dehydrogenase subunit. Tobler [6] has previously reported on a dimeric structure for the enzyme. The molecular weight of the native enzyme could therefore approximate between 66600 and 70000. Borack and Sorer [3] has previously reported an enzyme molecular weight of 63000 for L-fl-hydroxyacid dehydrogenase based on results from density gradient centrifugation and gel filtration. Considering the fact that Borack and Sorer utilized a completely

20 TABLE I1 A M I N O ACID COMPOSITION OF L-fl-HYDROXYACID DEHYDROGENASE These are the results from one analysis, n.d., not determined. Amino acid

Residues per 10013

Asx ~ Thr Ser GIx ~ Pro Gly Ala Cys ~ Val Met lie Lev Tyr Phe Lys His Arg Trp

81 50 51 158 99 90 97 14 72 5 35 92 28 20 56 2 4 n.d.

Aspartate + asparagine. b Glutamate + glutamine. " Cysteic acid.

different experimental technique to arrive at their molecular weight determination for L-j3-hydroxyacid dehydrogenase, we feel that the two determinations are well within range of one another. An analysis of the amino acid composition of L-~8-hydroxyacid dehydrogenase is seen in Table II. The high proportion of basic amino acids accounts for the relatively high isoelectric point of L-j3-hydroxyacid dehydrogenase (pl = 8.6. Cannistraro et al. [4] have reported that L-/3-hyd r o x y a c i d dehydrogenase consists of dissimilarsized subunits of 40000 and 23500.-If L-/3-hydroxyacid dehydrogenase is composed of two different subunits, there could be two genes for this enzyme. However we find that L-/3-hydroxyacid dehydrogenase appears to be composed of subunits of identical size. All of our evidence supports a homodimeric structure for L-/3-hydroxyacid dehydrogenase. This would be expected in that simple dehydrogenases have usually been found to be homopolymers. The nature of the two subunits reported by Cannistraro et al. [4] is not entirely clear to us. If the smaller subunit is equivalent to

the enzyme subunit we have found for L-fi-hydroxyacid dehydrogenase, the larger peptide reported by Cannistraro et al. might be an undissociated dimer. If the larger of the peptides found by Cannistraro et al. is equivalent to the L-fl-hydroxyacid dehydrogenase subunit we have demonstrated, their smaller protein could be an alcohol dehydrogenase subunit, as alcohol dehydrogenase copurifies with L-j3-hydroxyacid dehydrogenase up to the hydroxyapatite column. Our finding of 33 300-35000 for the subunit weight of L-/3-hydroxyacid dehydrogenase falls intermediate between the two proteins said by Cannistraro et al. to have molecular weights of 23 500 and 40000. We have confidence in our subunit molecular weight determination for L-/3-hydroxyacid dehydrogenase as it has been observed as an isolation product from two lines of investigation, both as protein isolated from flies and as an in vitro synthesized protein. We conclude that L-fl-hydroxyacid dehydrogenase is homodimeric. Larval and adult proteins appear to be of identical size. Furthermore, an immune precipitate of L-fl-hydroxyacid dehydrogenase produced in vitro in a reticulocyte lysate where post translational modification does not occur has approximately the same subunit size as the in vivo synthesized L-fl-hydroxyacid dehydrogenase isolated in our purification procedure. This presents evidence that the enzyme is not post-translationally modified in a substantial manner.

Acknowledgments This research was supported by grant GM26830 from the National Institute of Health and by Bio-Medical grant BSRG-S07R077068-17. We thank Dr. Ray Bratcher for his help in raising L-/3-hydroxyacid dehydrogenase antibodies. We also thank Celeste B. Rich for amino acid analysis of L-fl-hydroxyacid dehydrogenase.

References 1 Smiley, J.D. and Ashwell, G. (1961) J. Biol. Chem. 236, 357-364 2 Grollman, A.P. and Lehninger, A.L. (1957) Arch. Biochem. Biophys. 69, 458-467 3 Borack, L.I. and Sofer, W. (1971) J. Biol. Chem. 246, 5345-5350

21 4 Cannistraro, V.J., Borack, L.I. and Chase, T. (1979) Biochim. Biophys. Acta 569, 1-5 5 Narang, S. and Seawright, J.A. (1983) Biochem. Gen. 21, 885-893 6 Tobler, J.E. and Grell, E. (1978) Biochem. Gen. 16, 333-342 7 Sullivan, D., Kitos, R.J. and Sullivan, M.C. (1973) Genetics 75, 651-661

8 King, J. and Laemmli, U.K. (1971) J. Mol. Biol. 62, 465-477 9 Skuse, G. and Sullivan, D. (1985) EMBO J., in the press 10 Pelham, H.R.B. and Jackson, R.J. (1976) Eur. J. Biochem. 67, 247-256 11 Foster, J.A., Rich, C.B., Fletcher, S., Karr, S.R., and Przybyla, A. (1980) Biochem. 19, 857-864