Reversible cold inactivation of microsomal 3-hydroxy-3-methylglutaryl coenzyme A reductase from rat liver.

Abstract Soluble 3-hydroxy-3-methylglutaryl coenzyme A reductase (HMG-CoA reductase, EC 1.1.1.34) from rat liver microsomes has been shown to be inactivated reversibly by cold temperature whereas microsomal enzyme did not show any cold sensitivity. At 0° the soluble enzyme lost 80 % of its activity within 30 min and warming at 37° for 20 min restored all the lost activity. Active enzyme preparations did not lose activity if quick frozen at -80°. The simplest explanation of these observations is the hypothesis that the enzyme can exist in two forms, i.e. a cold-inactivated and a heat-activated form. The enzyme had minimum activity between 0 and 19° and the specific activity was less than 0.1 nmole per min per mg. A 10-fold increase in activity was obtained by incubation at 37° for 20 min, but activity was destroyed at 42°. The original assay method employed for this solubilized enzyme involved incubation at 37° which does not differentiate between the two reactions (a) inactive enzyme → active enzyme and (b) substrate → product. However, more readily interpretable results were obtained either by activating the enzyme at 37° prior to assay or by conducting the assay at a temperature at which activation does not occur, i.e. between 16 and 19°. The Arrhenius plot of the activation of the enzyme by heat (log Vmax versus 1/T) showed apparent changes in slope at 19 and 28°. The rate of activation was found to be strongly protein concentration-dependent indicating an intermolecular reaction. The enzyme was stable in 40 to 120 mm buffer concentrations for at least 5 hours, but addition of 4 m KCl caused an irreversible loss in activity of cold-inactivated enzyme and did not prevent the cold inactivation of activated enzyme. Km for substrate d-3-hydroxy-3-methylglutaryl-CoA of the active soluble enzyme was estimated to be 2.2 x 10-5 m, similar to that of the microsome-bound enzyme. HMG-CoA reductase is the first example of a microsomal enzyme that is reversibly cold-sensitive.

It is the rate-limiting enzyme of the sterol biosynthetic pathway and the site of action for the feedback inhibition by dietary cholesterol and certain of its derivatives (l-5). The enzyme is associated with the endoplasmic reticulum and exhibits a very short half-life of 2 to 4 hours (6-S). This is markedly different from the turnover rate of 2 to 3 days for the endoplasmic reticulum membranes (9) and for most of the microsomal enzymes of rat liver that have been studied (10 Purification of 3-Hydroxy-3-methylglutaryl CoenzymeA Reductase Using a solubilization method which requires sodium deoxycholate, they obtained a homogeneous protein with a specific activity of 3 to 5 nmol/min/mg.
The purification procedure reported here is short and relatively simple and yields a preparation with a specific activity of 407 to 480 nmol/min/mg.
We also report some intriguing properties of the enzyme as well as the production of antibody and certain immunochemical studies. The 30 to 50% (NH,),SO,fraction was incubated at 37" for at least 30 min before it was applied to the column. Any turbidity developing during this period was removed by centrifugation at 34,000 x g for 10 min and only the clear supernatant was applied to the column. Fractions of 2 ml each were collected and aliquots from these assayed for HMG-CoA reductase activity. The enzyme appeared in an elution volume (V,) of 330 ml to 390 ml and preceded the salt peak from this column. Fractions containing activities greater than 5 nmol/min/ml were pooled, and the volume was reduced to 5 to 6 ml in an Amicon ultrafiltration cell by using a PM 10 filter. The concentrate was collected and incubated at 37" for 45 to 60 min, and again any turbidity that developed was removed by centrifugation. The supernatant was fractionated on a Sephadex G-200 column (2.8 x 40 cm) equilibrated with Buffer II at room temperature. Fractions following the void volume (V,) and corresponding to an elution volume of 110 to 150 ml contained enzyme activity. These were pooled and concentrated again in an Amicon cell, and this preparation is hereafter referred to as purified enzyme.
The concentrate was checked for purity by using polyacrylamide gel electrophoresis as well as gel electrophoresis in the presence of sodium dodecyl sulfate.
One unit of enzyme activity is defined as the amount which produces 1 nmol of mevalonate in 1 min at 37" under the assay conditions used. Protein content was determined by the biuret (21) and Lowry (22) methods. Interference by dithiothreitol with the reaction was prevented by precipitation of the protein with 5% trichloroacetic acid and redissolving it in 0.2 N NaOH.

Polyacrylamide
Gel Electrophoresis-Disc gel electrophoresis of the enzyme was carried out according to the system described by Nelson et al. (23) in 5% acrylamide using a Tris-glycine buffer system. After 6 weeks 20 to 30 ml of blood were withdrawn, the blood clot removed by centrifugation, and the antibody titer of the serum determined.
For later bleedings, doses of 200 pg of enzyme were given subcutaneously at 1. to 2-week intervals and rabbits bled 1 week later. Control serum was obtained from uninjected rabbits. Both control and antisera were stored at 20". Antiserum was evaluated by immunodiffusion (33) and by quantitative titration with solubilized enzyme (34). The enzyme and antisera were mixed and incubated at 37" for 1 hour and kept at 4' for 3 hours. After centrifugation of this reaction mixture, the supernatant was assayed for enzyme activity.
The pellet was washed twice with 0.9% NaCl, dissolved in 0.2 N NaOH, and the protein concentration determined.
To check whether the antibody could also inhibit HMG-CoA reductase activity in the microsomes, the antiserum was incubated with the microsomal homogenate at 37" for 60 min and the mixture assayed for enzyme activity. with Buffer II, and kept at room temperature (A). Numbers on the abscissa mark the milliliters of effluent.
Fractions 72 to 96, indicated by the botched area in B, were pooled and ammonium sulfate was added to 50% saturation at 4". The precipitate was dissolved in Buffer II and split in two aliquots. One aliquot was kept at 37" for 60 min and applied to the Bio-Gel A-0.5m column kept at room temperature (B). The other aliquot was fractionated through the same Bio-Gel A-0.5m but equilibrated and kept at 6" (C). The molecular weight of the enzyme in A and C was estimated to be approximately 200,000, determined by calibrating the column with standard proteins as described in Fig. 1. In B the enzyme appeared in the void volume (determined with blue dextran 2,000 and horse apoferritin, molecular weight 480,000) and corresponded to a molecular weight of 480,000 or more. of the chromatogram seen upon exposure to iodine vapor mobility relative to proteins of known molecular weight elecsuggested the presence of phospholipid.
Calorimetric determi-trophoresed in the same gel slab (Fig. 6). nations of the cholesterol and cholesterol ester content have Amino Acid Analyses-The amino acid composition of the given values of 15 pg of cholesterol and 3 pg of cholesterol purified enzyme is presented in Table III. Mean values from ester/mg of purified protein.
duplicate analyses are presented. The values for tryptophan Criteria of Purity-Purity of the enzyme was established by and cysteic acid were not determined. (a) disc gel electrophoresis; the purified enzyme preparation Quantitative Precipitin Reactions of HMG-CoA Reductase was dialyzed for 48 hours against a lo-fold diluted Buffer I, and - Fig. 7 shows results of quantitative precipitin reactions with aliquots from this fraction were analyzed for purity by, gel antisera obtained from two immunized rabbits and partially electrophoresis.
Fractions of enzyme from the solubilized purified enzyme from Fraction I. All of the activity in an Fraction I were similarly dialyzed and electrophoresed. Using aliquot containing 0.4 to 0.5 unit was precipitated by 100 ~1 of 5% polyacrylamide gels and the Tris-glycine buffer system serum obtained from immunized rabbits, whereas no enzyme described by Nelson et al. (23), the purified preparations gave was precipitated by control serum. Addition of enzyme greater only one protein staining band with an RF of 0.81 (Fig. 4). than 2.3 units gave significant HMG-CoA reductase activity in When the gel system No. 6 described by Maurer (24) was used, the supernatant. The immunodiffusion pattern between antithe enzyme migrated to an RF of 0.2 and also gave a single serum prepared against purified HMG-CoA reductase and the band. (b) Electrophoresis in sodium dodecyl sulfate-polyac-solubilized Fraction I is shown in Fig. 8, inset. Only a single rylamide gel slabs; electrophoresis of the enzyme .sodium precipitin line could be detected indicating the presence of a dodecyl sulfate complex in the presence of urea and 2-mercap-specific antibody to HMG-CoA reductase. toethanol gave a single component (Fig. 5). Its molecular Inhibition of HMG-CoA Reductase Activity of Intact Miweight was estimated to be around 120,000 based on its crosomes by Antibody-Complete inhibition of enzyme activ-ABCDEF FIG. 4 (left). Disc gel electrophoresis in 5% acrylamide using the Tris-glycine buffer system described by Nelson (23). Gel A