Spectrophotometric enzymatic assay for S-3-hydroxyisobutyrate

184,317-320

ANALYTICALBIOCHEMISTRY

(1990)

Spectrophotometric Enzymatic Assay for S-3-Hydroxyisobutyrate’ Paul M. Rougraff,

Ralph

Paxton,’

Gary W. Goodwin,3

Reid G. Gibson,

and Robert

A. Harris4

Department of Biochemistry Indiana University Schoolof Medicine, 635 Barnhill Drive, Indianapolis, Indiana 46202-5122

Received

September

6,1989

An enzymatic spectrophotometric end-point assay has been developed for determination of S-3-hydroxyisobutyrate in biological fluids. The assay measures NADH production at 340 nm after initiation of the reaction with rabbit liver 3-hydroxyisobutyrate dehydrogenase (EC 1.1.1.31). The assay is not affected by R-3-hydroxyisobutyrate, lactate, malate, S-hydroxybutyrate, 2-methyl-3-hydroxybutyrate, 3-hydroxyisovalerate, 3-hydroxy-n-valerate, 2-methyl-3-hydroxyvalerate, and 3-hydroxypropionate. The assay does measure 2-ethyl-3-hydroxypropionate, a minor metabolite produced by catabolism of alloisoleucine. Application of the method to measure S-3-hydroxyisobutyrate in plasma obtained from normal, 48-h starved, and mildly and severely diabetic rats gave levels of 28,42, 112, and 156 PM, respectively. o 1990AcademiePress.Inc.

S-3-Hydroxyisobutyrate is a normal intermediate in the catabolic pathway of valine. Elevated extracellular levels of 3-hydroxyisobutyrate have been found in ketoacidotic humans (1,2), diabetic rats (l), a patient with methylmalonic acidemia (3), a patient with presumed deficiency of methylmalonic semialdehyde dehydrogenase (4), and a newborn patient with 3-hydroxyisobu’ This work was supported in part by grants from the United States Public Health Service [NIH DK44041 (R.A.H.)], the Indiana University School of Medicine Program for Medical Student Research [T35 HL7584 (P.M.R.)], the Indiana Affiliate of the American Diabetes Association (P.M.R.)], AM20542 (Diabetes Research and Training Center), the Grace M. Showalter Residuary Trust, and the Indiana Heart Association (G.W.G.). * Current address: Department of Physiology and Pharmacology, Auburn University, Auburn, AL 36849. 3 Current address: Department of Biochemistry, University of Washington Howard Hughes Medical Institute, SL 15, Seattle, WA 98195. 4 To whom correspondence and reprint requests should be addressed. 0003~2697/90

$3.00

Copyright Q 1990 by Academic Press, Inc. All righta of reproduction in any form reserved.

tyric aciduria (5). 3-Hydroxyisobutyrate has been suggested to be the most important metabolite from valine catabolism released by the muscle (6) since it can be used for the synthesis of glucose by the liver (7). For these reasons and because of our limited understanding of valine catabolism, it was important to develop a simple enzymatic assay for 3-hydroxyisobutyrate. MATERIALS

AND

METHODS

Sources of materials. Preparation of 3-hydroxyisobutyrate from its methyl ester (Aldrich Chemical Co., Milwaukee, WI) and purification of HIB dehydrogenase5 (9.7 U/mg) from frozen rabbit liver were as described previously (8). 2-Methyl-3-hydroxybutyrate, 3-hydroxyisovalerate, 3-hydroxy-n-valerate, 2-methyl-3-hydroxyvalerate, 3-hydroxypropionate, and 2-ethyl-3-hydroxypropionate were generously provided by Dr. Lawrence Sweetman (University of California, San Diego). Male Wistar rats were from Harlan Industries (Indianapolis, IN). All other reagents were from Sigma Chemical Co. (St. Louis, MO) or Fisher Scientific (Pittsburgh, PA). Sample preparation. All samples were processed at 4°C. Rats were anesthetized with diethyl ether, and blood was removed by cardiac puncture with an heparinized syringe. Blood was centrifuged 2 min in an Eppendorf centrifuge at maximum speed, and the plasma carefully removed. Plasma (1.0 ml, determined gravimetrically) was vigorously mixed with 0.5 ml ice-cold 15% (w/v) perchloric acid. After 15 min on ice, the sample was centrifuged as above for 3 min. Supernatant (0.8 to 0.9 ml) was carefully removed and 0.02 ml universal indicator added. The sample was neutralized with 3 M KOH containing 0.25 M 4-morpholine ethane sulfonic acid and 0.25 M 4-morpholine propane sulfonic acid. The sample was incubated on ice for at least 30 min and cen6Abbreviations dehydrogenase; density, GC-MS,

used: HIB dehydrogenase, 3-hydroxyisobutyrate EDTA, ethylenediaminetetraacetate; OD, optical gas chromatography-mass spectrometry. 317

318

ROUGRAFF

ET

AL.

RESULTS AND DISCUSSION

10 min

FIG. 1.

Spectrophotometric recording of the reaction catalyzed by HIB dehydrogenase. Assay of 6.9 nmol of 3-hydroxyisobutyrate in 0.3 ml processed rat plasma was performed as described under Materials and Methods. HIBDH refers to 5 ~1 addition of 0.6 U/ml rabbit liver HIB dehydrogenase.

trifuged rate.

as above for 30 s to remove potassium

perchlo-

A typical spectrophotometric recording of the formation of NADH by the reaction catalyzed by HIB dehydrogenase (3-hydroxyisobutyrate + NAD+ + methylmalonate semialdehyde + NADH + Hf) is shown in Fig. 1. The second addition of HIB dehydrogenase did not result in any further absorbance changes, indicating the reaction was complete. The reaction time to reach completion varies with 3-hydroxyisobutyrate concentration but is usually complete within 30 min. The assay was Iinear with 0.1 to 0.4 ml plasma extract added to the cuvette (r = 0.991, Fig. 2). Recovery of S-3-hydroxyisobutyrate from the plasma treatment procedure was determined by adding 100 nmol of the compound to 1.0 ml rat plasma. The perchloric acid extraction was performed as described for plasma with and without added 3-hydroxyisobutyrate so that percentage recoveries could be corrected for endogenous 3-hydroxyisobutyrate. Percentage recovery of S3-hydroxyisobutyrate was 99 + 2% (means f SEM) for plasma samples from four different rats. The R-isomer of 3-hydroxyisobutyrate is not measured by the assay described here, as determined by lack of detection of this compound when the extraction/assay was conducted with 1.0 ml rat plasma with 100 nmol of added R-3-hydroxyisobutyrate. This is explained by the 350-fold greater specificity constant (K,,/K,) of rabbit liver HIB dehydrogenase for the S-isomer compared

Assay of 3-hydroxyisobutyrate. Reactions were conducted at 30°C in glass cuvettes (path length 1.0 cm) in a 1.0 final volume containing 0.67 M Tris (hydroxymethyl) aminomethane, 3.3 mM MgS04, 1.7 mM EDTA, 0.13 M hydrazine sulfate, 1 mM NAD+, 3 mU HIB dehydrogenase, and 0.1 to 0.4 ml neutralized plasma extract. After establishment of a baseline absorbance (2.5 min), reactions were initiated by addition of 5 ~1 of 0.6 U/ml HIB dehydrogenase. The reactions are typically complete in 30 min. NADH production was observed at 340 nm with the chart recorder set at a full scale of 0.1 OD unit and a period of 2.5 s on a Cary 219 automatic recording dualbeam spectrophotometer. Absorbance changes due to enzyme addition were corrected by the addition of water in place of plasma extract (0.002 OD units for a typical enzyme preparation). Sample dilution was determined volumetrically. Animal studies. Male Wistar rats (200-250 g) were fasted for 30 h prior to injection with either 65 or 150 mg streptozotocin/kg body wt. Rats were maintained on normal laboratory chow before the fasting period and after injection. Rats were sacrificed by decapitation 48 h after induction of diabetes. In diabetic rats, plasma glucose values measured greater than 350 mg/dl, assayed spectrophotometrically with the hexokinase/glucose-6phosphate dehydrogenase coupled enzyme system (9).

0.0

0.1

FIG. 2.

0.3

0.2 Volume

0.4

(ml)

Effect of plasma extract volume on the assay of 3-hydroxyisobutyrate. Indicated volumes of processed rat plasma were assayed as described under Materials and Methods. Optical density data points were corrected for addition of HIB dehydrogenase. Each point is the mean + SD for three independent determinations. Coefficient of variance was 12%. 3-Hydroxyisobutyrate concentration in the original sample was 40 nmol/ml for a 48-h starved rats.

ENZYMATIC

319

ASSAY FOR S-3-HYDROXYISOBUTYRATE

with the R-isomer (8). The high K, for the R-isomer (1.4 mM) compared with the S-isomer (0.061 mM) explains the lack of interference at normal physiological concentrations of the former compounds (8). Thus R-3-hydroxyisobutyrate could influence the assay in a pathologic state where the plasma concentration approached the K, of this compound. No such pathologic state is known at this time. The enzymatic determination of 3-hydroxyisobutyrate was applied to plasma samples obtained from rats at various nutritional and hormonal states (Table 1). After rats were starved for 48 h 3-hydroxyisobutyrate plasma concentration increased l-s-fold compared with the fed state. Presumably this results from the release of 3-hydroxyisobutyrate from peripheral tissues as a consequence of increased proteolysis/valine catabolism (6,10,11). The 5-fold increase in plasma 3-hydroxyisobutyrate in severely diabetic rats is consistent with the previous observation that urinary 3-hydroxyisobutyrate is increased in patients with ketoacidosis of various etiologies (1). The equilibrium constant of 3-hydroxyisobutyrate oxidation to methylmalonate semialdehyde has been determined to be pH dependent with a value of 0.036 at pH 9.0 (12). Therefore, at pH 9.0 with 1.0 mM NAD+ and 0.1 mM 3-hydroxyisobutyrate present, the reaction would proceed to only 44% completion. To drive the reaction to completion, hydrazine is added to react irreversibly with methylmalonate semialdehyde in a manner similar to that used for the enzymatic determination of lactate (13), malate (14), and 3-hydroxybutyrate (15). Complete oxidation in the presence of hydrazine was confirmed by a stoichiometric relationship between quantity of 3-hydroxyisobutyrate added and NADH produced (data not

eating that it was free of contamination by dehydrogenases which could utilize these substrates. A commercial preparation of Rhodopseudomonus spheroides 3-hydroxybutyrate dehydrogenase (EC 1.1.1.30) has been shown to be contaminated by HIB dehydrogenase (16). Because both 3-hydroxyisobutyrate and 3-hydroxybutyrate are normally found in the blood and both are enzymatically assayed in a similar manner, 3-hydroxybutyrate concentrations may have been significantly overestimated in numerous previous studies (16). Other short-chain hydroxy-carboxylic acids were examined with respect to possible oxidation by HIB dehydrogenase or inhibition of HIB dehydrogenase catalyzed oxidation of 3-hydroxyisobutyrate. The following compounds at 1 mM showed no activity (or inhibition with 1 mM 3-hydroxyisobutyrate present): 2-methyl-3-hydroxybutyrate, 3-hydroxyisovalerate, 3-hydroxy-n-valerate, 2-methyl-3-hydroxyvalerate, and 3-hydroxypropionate. However, 2-ethyl-3-hydroxypropionate (2-ethylhydracrylic acid) was a good substrate for the NAD+-dependent reaction catalyzed by HIB dehydrogenase. Because of the structural similarity to 3-hydroxyisobutyrate, it is not surprising that the latter compound, a normal metabolite in the catabolic pathway of alloisoleucine, is oxidized by HIB dehydrogenase. Alloisoleucine is a minor, non-protein amino acid that can be produced following nonenzymatic racemization of the keto acid resulting from isoleucine transamination (17). The normal concentration of 2-ethyl-3-hydroxypropionate in biological fluids or its role in interorgan exchange of amino acid carbon skeletons has not been investigated. 3-Hydroxyisobutyrate has been previously measured in biological fluids by gas chromatography-mass spectrometry (GC-MS). 3-Hydroxyisobutyrate interferes with the measurement of 3-hydroxybutyrate by GC-MS, since the t-butyldimethylsilyl and trimethylsilyl derivatives of R-3-hydroxybutyrate and 3-hydroxyisobutyrate cannot be resolved on many columns commonly employed for GC-MS analysis of ketone bodies (18-20). Recently, a new procedure for analyzing 3-hydroxybutyrate by GC-MS has been developed with little interference by 3-hydroxyisobutyrate (21). We have described here a new means of measuring 3-hydroxyisobutyrate that should prove simpler and more convenient for most laboratories. In addition, the enzyme from rabbit liver shows considerable stereospecificity toward S3-hydroxyisobutyrate over R-3-hydroxyisobutyrate so that the assay can enzymatically measure the concentration of the S-isomer. Although it is possible to resolve the latter compounds by GC-MS (4), the more conventional method of GC-MS measures the sum of the two compounds.

shown).

ACKNOWLEDGMENT

TABLE 3-Hydroxyisobutyrate

State Chow fed 48-h starved Mildly diabetic* Severely diabetic’

1

Levels

in Rat

Plasma

3-Hydroxyisobutyrate concentration” (nmol/ml) 28k 1 42 + 3** 112f 4** 155 f 14+*

’ Values are the means + S.E.M. for plasma samples from five or six rats. * Mild diabetes produced by injection of streptozotocin (65 mg/lOO g body ti) ’ Severe diabetes produced by injection of streptozotocin (150 mg/ 100gbodywt). ** P < 0.001 versus chow fed by unpaired Student’s t test.

The purified HIB dehydrogenase used in this work did not oxidize lactate, malate, and 3-hydroxybutyrate, indi-

We are very eral short-chain

grateful to Dr. Lawrence Sweetman carboxylic acids not commercially

for providing available.

sev-

320

ROUGRAFF

REFERENCES 1. Landaas,

2. Liebich, 3. Kolvraa,

S. (1975) Clin. Chim. Acta 64,143-154. H. M., and For&, C. (1984) J. Chromutogr. S., Gregersen, E., Christensen, E., and

309,225-242. Rasmussen,

K.

(1980) J. Inherited Met& Dis. 3,63-66. 4. Pollit, R. J., Green, A., and Smith, R. (1985) J. Inherited Metab. Dis. 8,75-79. 5. Congdon, P. J., Haigh, D., Smith, R., Greene, A., and Pollit, R. J. (1981) J. Inherited Metab. Dis. 4,79-80. 6. Lee, S-H. C., and Davis, E. J. (1986) Biochem. J. 233,621-630. 7. Letto, J., Brosnan, M. E., and Brosnan, J. T. (1986) Biochem. J. 240,909-912. 8. Rougratf, P. M., Paxton, R., Kuntz, M. J., Crabb, D. W., and Harris, R. A. (1988) J. Biol. Chem. 263,327-331. H. U.. Bernt. E.. Schmidt. F.. and Stork. H. (1974) in 9. Berzmever. Methods of Enzymatic Analysis (Bergmeyer, H. U.; Ed.), Voi. 3, pp. 1196-1201, Academic Press, New York. 10. Wagenmakers, (1985) Znt. 11. Spydevold, 12. Robinson, 521.

A. J. M.,

Salden,

H. J. M.,

and Veerkamp,

J. H.

J. Biochem. 9,957-965. 0. (1979) Eur. J. Biochem. 97,389-394. W. G., and Coon, M. J. (1957) J. Biol. Chem. 225,511-

ET

AL.

in Methods

13. Gawehn, K., and Bergmeyer, matic Analysis (Bergmeyer, ademic Press, New York.

H. U. (1974) H. U., Ed.), Vol.

of Enzy3, pp. 1492-1495, Ac-

14. Gutmann, I., and Wahlefeld, matic Analysis (Bergmeyer, ademic Press, New York.

A. W. (1974) in Methods of EnzyH. II., Ed.), Vol. 3, pp. 1585-1589, Ac-

15. Williamson, D. H., and Mellanby, J. (1974) in Methods of Enzymatic Analysis (Bergmeyer, H. U., Ed.), Vol. 4, pp. 1836-1839, Academic Press, New York. 16. WorraIl, E. B., Gassain, T. N. (1987) Biochem. 17. Mamer,

0. A., Tjoa,

S., Cox,

D. J., Sugden,

M. C., and Palmer,

J. 241,297-300.

S. S., Striver,

C. R., and Klassen,

G. A. (1976)

Biochem. J. 160,417-426. 18. Miles, J. M., M. W. (1984)

Anal. B&hem.

19. Miles, J. M., M. W. (1986)

Amer. J. Physiol. 251,

20. Bougneres,

Schwenk, Schwenk,

P. F., BaIasse,

W. F., McClean,

K. L., and

Haymond,

W. F., McClean, K. L., and E185-E191.

Haymond,

141,110-115.

E. O., Ferre,

P., and Bier,

D. M. (1986)

J. Lipid Res. 27,215-220. 21. DesRosiers, M., David,

C. D., Montgomery, J. A., Desrochers, S., Garneau, F., Mamer, 0. A., and Brunengraber, H. (1988) Anal.

Biochem. 173.96-105.