Purification and Properties of a Soluble Protein Activator of Rat Liver Squalene Epoxidase*

A soluble rat liver protein, termed "supernatant protein factor" (SPF), that stimulates microsomal squalene epoxidase has been purified approximately 11,000-fold. The most highly purified preparation obtained by isoelectric focusing shows a single coincident peak for activity and protein (the isoelectric point, pI, was 6.74). SPF is about 95% pure, judged by sodium dodecyl sulfate-polyacrylamide gel electrophoresis, and it migrates to a position corresponding to an apparent molecular weight of 47,000. An amino acid analysis of SPF is presented, and the properties of SPF and of the various soluble protein activators of microsomal sterol biosynthesis described by other laboratories are compared.

A soluble rat liver protein, termed "supernatant protein factor" (SPF), that stimulates microsomal squalene epoxidase has been purified approximately ll,OOO-fold. The most highly purified preparation obtained by isoelectric focusing shows a single coincident peak for activity and protein (the isoelectric point, p1, was 6.74). SPF is about 95% pure, judged by sodium dodecyl sulfate-polyacrylamide gel electrophoresis, and it migrates to a position corresponding to an apparent molecular weight of 47,000. An amino acid analysis of SPF is presented, and the properties of SPF and of the various soluble protein activators of microsomal sterol biosynthesis described by other laboratories are compared.
It has been known for some time that the NADPH-and O,dependent conversion of squalene to lanosterol by liver microsomes is enhanced by the addition of 105,000 x g supernatant (1). After squalene-2,3-oxide was demonstrated to be an intermediate in the squalene-lanosterol conversion (2,3), supernatant dependence was found to be associated with squalene epoxidase, a microsomal activity distinct from squalene-2,3oxide cyclase (4). One of the soluble components that stimulated squalene epoxidase was a heat-labile supernatant protein, termed "supernatant protein factor;" it was purified 84fold and had an apparent molecular weight of 44,000 (5). We here report the purification of SPF' to apparent homogeneity and show that it is a single polypeptide chain with a molecular weight of 47,000. Assays -SPF activity was assayed under conditions similar to those described previously (5). Amo-1618, a plant growth inhibitor (7), was added to the assay mixtures to inhibit squalene-2,3-oxide cyclase (8). [ (9); incubations were performed as described except that l-ml assay volumes and 40 nmol of squalene were used. Products were isolated and assayed as described in epoxidase assays above. One unit of SCP, activity is defined as 1 incremental nmol of squalene-2,3-oxide plus sterols produced in 120 min under the described conditions in the presence of SCP, over controls without SCP,. Protein Assays -Protein concentrations were estimated either by the microbiuret method (IO) or by the spectrophotometric quantitation of trichloroacetic acid precipitates stained with Amido black 10B (11) on Millipore HAWP 04700 filters. Both assays were standardized to a solution of bovine serum albumin, assuming an extinction coefficient at 280 nm of 0.660 liters g-l cm-' for the standard protein (12 with 300ml batches of S,,. The S,,, stirred at 100 'pm, was chilled to 0". As ice crystals began to form, 246 ml of reagent grade acetone chilled to -70" were added, and the bath temperature was lowered to -10". After 20 min at -lo", the 45% (v/v) acetone/S,, mixture was centrifuged for 10 min at -10" (14,000 x g). An additional 654 ml of -70" acetone were added to the clear supernatant to raise the acetone concentration to 75% (v/v). The bath was maintained at -10" for 10 min, after which the mixture was centrifuged for 5 min as before.
The pellet was resuspended in 15 ml of chilled 5 mM potassium phosphate, 0.5 mM dithiothreitol buffer, pH 7.5, with the aid of a Teflon pestle and then centrifuged at 4" for 10 min (12,100 x g) to remove particulate matter.
Acetone was removed by immediately applying the supernatant to a column of Sephadex G-25 (medium) (5.8 x 38 cm) eouilibrated with the 5 mM phosphate, 0.5 mM dithiothreitol buffer and eluting at 400 ml/h. Void volume fractions were pooled and frozen at -70" ("acetone fraction"). All subsequent steps were performed at 4", and buffer pH values were measured at 4". The acetone fraction (866 ml) was Dumped at 400 ml/h through a bed of DEAE-cellulose (9.0 x 8.6 cm)#and abed of phosphocellulose (9.0 x 5.7 cm). effective separation of SPF from inactive protein than pH 3 to 10 systems; a single major peak of absorbance at 280 nm appeared in the pH 6 to 8 region (p1 = 6.74) coincident with SPF activity (Fig. 2). The focusing produced a 2-fold purification (Table I)  in the effluent of the high resolution isoelectric focusing step (Fig. 2) indicated homogeneity with respect to isoelectric point. SDS-polyacrylamide gel electrophoresis of 4.6 kg of the focused fraction (Fig. 3A ) showed a major polypeptide band at RF 0.46, with minor bands at RF 0.57 and 0.70. According to a scan of the gel at 550 nm (Fig. 3B), the major band comprised 95% of the total. A similar electrophoretic analysis of 9.1 pg (not shown) indicated that the minor bands constituted 2.9 and 1.7% of the total absorbance, respectively.
When SPF was run on SDS-polyacrylamide gels along with known protein standards (Fig. 41, migration of the protein corresponded to a molecular weight of approximately 47,000 for the polypeptide chain. Amino acid analyses of pure SPF are shown in Table II. Pure SPF was active in stimulating the two-step conversion of squalene to lanosterol, i.e. it had the activity ascribed to SCP, (9). The data presented in Table III indicate that in this assay SPF possessed maximum specific activities ranging from 250 to 400 units/mg. DISCUSSION The procedure described here affords a homogeneous SPF preparation alter approximately ll,OOO-fold purification. Judged from SDS-acrylamide electrophoresis, the squalene epoxidase-stimulating protein consists of a single 47,000-dalton polypeptide chain. In a previous report the native molecu- lar weight of SPF determined by Sephadex chromatography was estimated to be 44,000 (5). SPF is therefore a monomeric protein. On the basis of these properties alone, it appears that SPF is not identical with any of the soluble proteins that have been implicated in the promotion of various steps in the microsome-associated synthesis of cholesterol from squalene. The distinction is very clear with respect to two proteins that stimulate biosynthetic steps beyond lanosterol. The "squalene and sterol carrier protein" isolated by Ritter and Dempsey that stimulates both A?-sterol-A5-dehydrogenase (17) and As*'sterol-A'reductase (18) has the following properties. It exhibits a single band on SDS-polyacrylamide gel electrophoresis corresponding to a molecular weight of 16,000 and a native molecular weight of 17,000 (gel filtration) (19). Also, the amino acid composition differs significantly from that of SPF (Table  II), and squalene and sterol carrier protein is reported to comprise as much as 1% of the total supematant protein (19). Thus, SPF and the Ritter-Dempsey protein are clearly different entities. Similarly, SPF cannot& identical with the small protein (10,300 on SDS gels; 10,000 to 20,000 native molecular weight by gel filtration) isolated by Gaylor and Delwiche (20) that stimulates Cmethyl sterol oxidase.
Unlikely, but perhaps not entirely ruled out, is the possible identity of SPF with the protein activator designated "sterol carrier protein," (SCP,) by Scallen and colleagues. In their most recent report (9), 575-fold purified SCP, is stated to have molecular weights of 26,000 for the native protein (sedimenta-s : ( 4. Determination of the molecular weight of SPF by SDSpolyacrylamide gel electrophoresis. Electrophoretic conditions were the same as those for Fig. 3; proteins of known subunit molecular weight used to calibrate the gel system included bovine serum albumin (68,000), catalase (58,000), glutamic dehydrogenase (53,000), fumarase (49,000), ovalbumin (43,000), aldolase (40,000), glyceraldehyde-3-phosphate dehydrogenase (36,000), and carbonic anhydrase (29,000). The positions of the standard proteins presented are the results of two identical experiments, and the RF of SPF indicated is the average of four determinations. ' Seventy-hour value. q Calculated from 20-and 70-h data according to Moore and Stein (15).
tion equilibrium) and 13,000 for the subunit (sedimentation equilibrium in 6 M guanidine HCl). Although these molecular sizes and the dimeric subunit structure would seem to differentiate SCP, from SPF, under the conditions for SCP, assay (squalene to epoxide and sterols) we find SPF to be active with a specific activity 2 to 3 times that of SCP,. However, SCP, activity is defined as an entity that stimulates a two-step process, squalene epoxidation and squalene epoxide-lanosterol cyclization. Therefore, SCP, may affect either or both of the two steps. We have demonstrated that SPF stimulates squalene epoxidase, the first of the two events. If we assume that the two steps are activated by separate proteins, our results would be consistent with the view that SCP, affects the subsequent step, i.e. the squalene epoxide-lanosterol cyclization. But we cannot rule out the possibility that SCP, promotes squalene epoxidation as well. One further argument against the identity of SPF and SCP, is the degree of enrichment needed for obtaining the respective proteins in homogeneous form. The reported purification factor for SCP, is 575 (9), compared to 11,000 for SPF.
We have previously offered some experimental support for the view that SPF acts internally within the membrane, facilitating the access of substrate to specific enzyme sites (21). A similar conclusion has been reached by Srikantaiah et al. (9). The isolation of three and possibly four distinct protein activators is consistent with the view that the transformation of the various intermediates in sterol biosynthesis is controlled by several distinct noncatalytic proteins. Whether this involves substrate transport between cellular compartments or, as seems to us more likely, intramembrane events remains to be established.