Preparative electrophoresis: an improved method for the isolation of human recombinant apolipoprotein A-I.

Protein expression is a powerful technique to study the functional role of proteins both in vivo and in vitro. Various systems have been established to produce proteins in milligram quantities; however, the purification of many expressed proteins commonly results in trace amounts. We report on an improved method for the isolation of apolipoprotein A-I (apoA-I) using the Model 491 Prep Cell (Bio-Rad, Munich, Germany). This system has been previously used for similar protein isolation problems (3,7). We expressed apoA-I using a baculovirus system in Chinese hamster ovary (CHO) cells and in E. coli in excess (8). Apolipoproteins such as apoA-I are amphiphilic, and their purification is known to be difficult because of their amphipathic properties. Several purification methods have been used for isolating apoA-I (1,2,8,9); however, none of these isolation procedures has succeeded in separating proteins with the difference of 2 kDa in their molecular weight. Our method combines the advantages of easy handling, high efficacy and rapid purification. Human recombinant apoA-I was expressed in transfected CHO cells containing the human metallothionein II promotor (4,5). Briefly, cells were cultured in Ham’s F-12, grown to 80% confluency, washed extensively with phosphate-buffered saline (PBS) and then cultured in serum-free medium (CHO-S-SFM II; Life Technologies, Berlin, Germany) for 70 h. ApoA-I expression was induced using 70 μM ZnSO4. Medium was harvested, centrifuged and concentrated 20-fold by ultrafiltration with Ultrafree-15 (Millipore, Bedford, MA, USA). We used an improved procedure for large-scale preparation of recombinant proteins, the Model 491 Prep Cell, which enables preparative sodium dodecyl sulfate polyacrylamide gel electrophoresis (SDS-PAGE). A 5-cm-high 12% (30:0.8 acrylamide/bisacrylamide) separating gel was poured into the 28mm-diameter tube of the used gel apparatus. A 0.5-cm-high 3.5% stacking gel was loaded on the top of the separating gel. The concentrated expression medium containing 150 μg protein was diluted 2:1 with sample buffer (252 mM Tris-HCl, pH 6.8, 40% glycerol, 8% SDS, 0.01% bromophenol blue [BPB] and 15% β-mercaptoethanol) and incubated for 5 min at 95°C. Sample (2.0 mL) was loaded. Running time was 6 h at constant power (12 W). The elution buffer (25 mM Tris-HCl, 200 mM glycine, 0.1% SDS) was pumped at 0.8 mL/min. By the time the BPB reached the bottom of the gel, 3-mL fractions were collected. Samples were analyzed by 15% SDS-PAGE and visualized by silver staining (Novex, San Diego, CA, USA) and by immunoblot analysis, using a polyclonal affinity-purified goat anti-human apoA-I primary antibody. We used a biotin-labeled, affinity-purified rabbit anti-goat IgG (Kirkegaard & Perry, Rockville, MD, USA) as a secondary antibody. The blot was stained with 4-chloro-1-naphthol after incubation with a streptavidin-biotinylated horseradish peroxidase complex (Amersham International plc, Bucks, England, UK). Figure 1 illustrates the silver-stained SDS polyacrylamide gel of collected fractions. The apoA-I appears in the fractions 15–25 (28 kDa). This was confirmed by immunoblot analysis (not shown). Isolated fractions were dialyzed against 50 mM NH4HCO3 extensively, lyophylized and stored at -80°C for further analysis. This isolation procedure was very reproducible. The efficiency of the apoAI yield was 50%–70%; therefore, this method is able to purify apoA-I in a single step from crude extract. As shown in Figure 1, this preparative electrophoresis method shows a high-resolution capacity in separating proteins. Proteins with a difference as low as 2 kDa in their molecular weight can be separated by this method. Under optimized conditions, we were able to separate the commonly seen 26-kDa apoA-I fragment from the 28-kDa intact apoA-I (Figure 2). To separate the fragments of apoA-I, a degraded apoA-