Preparative Isoelectric Focusing of Poliovirus Polypeptides in Urea-Sucrose Gradients

Transcription

1 J. gen. ViroL (I98O), 47, Printed in Great Britain 423 Preparative Isoelectric Focusing of Poliovirus Polypeptides in Urea-Sucrose Gradients By K. J. WIEGERS AND R. DRZENIEK Heinrich-Pette-Institut fiir Experimentelle Virologie und Immunologie an der Universitiit Hamburg, Martinistrasse 52, 20o0 Hamburg 20, Federal Republic of Germany (Accepted 9 November I979) SUMMARY Isoelectric focusing in urea using small density gradient columns for dissociated poliovirus resulted in a complete separation of the four virus polypeptides. Identification and purity of VPI, 2, 3 and 4 was shown by SDS-polyacrylamide gel electrophoresis. The isoelectric points were determined and compared with the values obtained in gels. The recovery of the individual polypeptides was about 8o %. Up to 5oo #g of poliovirus per tube could thus be separated in a one-step procedure giving pure and SDS-free poliovirus polypeptides. INTRODUCTION The protein of poliovirus particles was separated by SDS-PAGE into four polypeptides (Maizel, t963). The use of markers allowed the determination of their mol. wt. (Maizel & Summers, 1968). This method is still the most extensively applied procedure for the analysis and characterization of virus proteins (Maizel, I97I). It has also been used for preparative work, in which it has some disadvantages. The recovery of picornavirus polypeptides from polyacrylamide gels is relatively low, as is the amount of protein which can be separated (Stoltzfus & Rueckert, i9721. Due to their very similar reel. wt., the peaks are often only poorly resolved, especially to the extent which would allow their harvest from preparative runs (Vande Woude et al. 1972; Talbot et al. 1973; Milstien et al. I977). In addition, the polypeptides obtained are contaminated by free or bound SDS. The low recovery of polypeptides prompted the search for other methods, which led to the separation of SDS polypeptides on hydroxylapatite columns (Moss & Rosenblum, I972) used successfully for Mengo virus (Ziola & Scraba, 1975). This method, however, gave poor results for poliovirus (K. J. Wiegers & R. Drzeniek, unpublished results). Other procedures, such as gel chromatography in guanidine hydrochloride and specific precipitation reactions, were used in only a few cases for analytical or preparative work in the field of picornaviruses (Stoltzfus & Rueckert, 1972). The successful application of isoelectric focusing to poliovirus polypeptides in ureacontaining polyacrylamide gels (Hamann et al. I977, I978; Hamann & Drzeniek, t978) provides a new possibility both for the analytical separation and for the isolation of pure poiiovirus polypeptides. Isoelectric focusing in gels has a number of advantages, but it is rather difficult to extract the polypeptides quantitatively. We have therefore developed the isoelectric focusing of poliovirus polypeptides in urea-sucrose gradients, which overcomes this obstacle. With the method described here it is possible to obtain pure, SDS-free, soluble poliovirus polypeptides in high yield in a simple one-step procedure, even in the presence of reducing agents such as fl-mercaptoethanol. OO22-i317/80/OOOO--3899/$O2.00 (~) 1980 SGM

2 424 K. I. WIEGERS AND R. DRZENIEK METHODS Materials. Ampholytes (Servalyte), SDS and bovine pancreatic ribonuclease A (RNase 1, EC ) were obtained from Serva (Heidelberg, Federal Republic of Germany). Acrylamide (Merck, Darmstadt, Federal Republic of Germany) was re-crystallized before use. Nonidet P4o (NP-4o) was from Shell (Hamburg, Federal Republic of Germany), sucrose (ultra pure, density gradient grade) from Schwarz-Mann (Orangeburg, N.J., U.S.A.). Other chenlicals were obtained from either Serva or Merck and were analytical grade. L-a~S-methionine (sp. act. 500 Ci/mmol) was obtained from the Radiochemical Centre (Amersham, U.K.) and Buchler (Braunschweig, Federal Republic of Germany). Virus. Labelled and unlabelled poliovirus type J, strain Mahoney, was grown in HeLa Ss cells and purified as described by Yamaguchi-Koll et al. 0975). Virus was stored in 3 M-CsC1 at -2o C. Before use it was dialysed for 6 to I2 h against o'i5 M-NaC1 (isotonic saline). Dissociation. Dissociation of the virus was achieved by adding solid urea to the virus solution to give a 9 M-concentration, i.e. 93 mg urea were added to IOO #1 of virus. Two #1 RNase (5 mg/ml) and, if necessary, I o/a of a 1"4 M-solution of fl-mercaptoethanol were also added. The mixture was incubated for I h at 25 C. Isoelectricfocusing. Isoelectric focusing was performed in glass tubes, 18 cm long, with an inner diam. of o'75 cm, containing a urea-sucrose gradient and ampholytes. The tubes were prepared as follows. One end was covered with a dialysis membrane and fastened with a rubber ring. A small plug of polyacrylamide (o'5 cm long) was placed on the membrane by polymerization of 0"4 ml of To ~o acrylamide containing 7 M-urea and 30 % sucrose (w/v). The tubes were placed in the apparatus described for gel rod isoelectric focusing (Hamann & Drzeniek, 2978) and inserted in the cold (4 C) lower electrode solution. They were then filled with 7 ml sucrose (o to 30 %, w/v) gradients containing 7 M-urea and a 2 % mixture of ampholytes &the ph ranges 5 to 7, 7 to 9 and 2 to II (2:1 :o'5), An overlay consisting of 7 M-urea and 2 % ampholytes was carefully added on top of the gradients. The gradients for the separation of poliovirus polypeptides always contained o'o5 M-fl-mercaptoethanol. The upper reservoir was filled with o. I M-phosphoric acid; the lower electrode solution consisted of o'l M-ethanolamine. The gradients were pre-focused for 4 to 6 h at o.o6 W per tube at an initial potential of 25o V 1o generate a ph gradient. The concentration of urea in the dissociated virus sample was lowered to 8 M in order to introduce the sample between the 7 M-urea overlay and the 7 M-urea-sucrose gradient after pre-focusing was terminated. The sample was electrophoresed for 20 to 24 h at 4 C at constant power intensity using a commercial power supply (ISCO model 492, Lincoln, Neb., U.S.A.). Finally, the voltage was increased to 3oo V. Fractionation. After focusing was complete, the gradient tube was removed and transferred to a fractionating device (Buchler Inst., Fort Lee, N.J., U.S.A.). Four-drop fractions (c. Ioo #1) were collected by inserting a canula through the acrylamide plug and replacing the gradient from the top with paraffin oil by a syringe. Measurement of ph. The fractions were collected into microtitre plates, where each fraction could be measured directly without dilution with a half-micro combined glass electrode (Radiometer, Copenhagen, Denmark). Measurements were done immediately after fractionation at 4 C under precautions to avoid CO2-absorption and evaporation of water from the sample. The radioactivity of the fractions was measured with a sample as previously described (Wiegers et al I977). SDS-polyacrylarnide gel electrophoresis. Spacer-gel buffer was added to the appropriate fractions of the gradient and the samples were made 1% SDS and 1% mercaptoethanol and

3 Focusing of polio virus polypeptides 425 heated for 2 min at loo C. They were then analysed on a 14% acrylamide gel as previously described (Hamann et al. I977; Hamann & Drzeniek, I978). RESULTS Preliminary assessment of method with haemoglobin To develop a method for the analytical and preparative separation of poliovirus polypeptides by isoelectric focusing in solution, we have used haemoglobin labelled with fluorescein isothyocyanate in preliminary experiments. Labelled haemoglobin gives a large number of visible bands both in polyacrylamide gels (Hamann & Drzeniek, I978) and in urea-sucrose gradients. This permitted a fast exploration of different methods and conditions and finally led to the procedure described (see Methods). Sealing of the tubes with a dialysis membrane and a polyacrylamide plug prevented mixing of the alkaline catholyte with the urea-sucrose gradient. This had the advantage that urea and sucrose could be omitted from the cathode solution, as observed in preliminary experiments performed in gradients lacking the plug or having a plug which contained am-pholyte without urea and/or sucrose. Due to the solubility of urea and sucrose at 4 C, a compromise was reached between a urea concentration high enough to prevent reaggregation of virus polypeptides and sucrose solutions giving a stable gradient. Since it had been shown in polyacrylamide gels that a concentration of 7 M-urea was sufficient to prevent reaggregation of poliovirus polypeptides (Hamann & Drzeniek, 1978), this urea concentration and a gradient ranging from o to 3o % (w/v) sucrose were chosen. Separation of poliovirus polypeptides Separation of poliovirus polypeptides obtained by dissociation of radioactive virus particles by urea and the ph gradient generated in the density gradient column by isoelectric focusing are shown in Fig. I. Four clearly separated peaks of radioactivity were obtained having isoelectric points of 8.I, 7"2, 6"5 and 5"8, corresponding to polypeptides VPr, VP4, VP2 and VP3, respectively. The peaks and polypeptides were correlated by SDS-polyacrylamide gel electrophoresis (Fig. 2). The separation of the four peaks by isoelectric focusing and the presence of only one polypeptide in each peak demonstrate the high resolution of the method employed. However, the major polypeptides occur as doublets. In our hands they are found when the virus has been treated with urea and are due to differences in conformation and SDS-binding. The separation is readily reproducible. The peaks are found almost in the same fractions of the gradient and there is good agreement among the ph values obtained in gels (Hamann et al. I977) despite the differences observed in the isoelectric points for VP2 and VP3, which are explained in the Discussion. Recovery of virus polypeptides Total recovery of the four polypeptides was about 80%, measured as the ratio of radioactivity of all peaks to the radioactivity of the sample. The recovery of the individual polypeptides varied, being highest for VP2 and lowest for VPI (Table I). Since the virus preparation was labelled with zss-methionine, the higher amount of radioactivity in the VP3 (Fig. I) peak reflects the higher methionine content of this polypeptide (Wouters & Vandekerckhove, I976). The content of methionine in the individual polypeptides was taken into account in our calculations. The recovery of 8o to Ioo% for VPz, VP3 and VP4 (Table 1) is good for a preparative procedure and acceptable for an analytical method. The high recotery of VP4 makes this

4 ? O X 426 K. J. WIEGERS AND R. DRZENIEK I I I I I I VPI VP4 VP2 VP2L VP3 1 6 ff.o d~ Fraction number Fig. I. Isoelectric focusing of poliovirus type I, strain Mahoney, in a sucrose density gradient column: Ioo #g of 35S-methionine-labelled poliovirus were dissociated in 9 M-urea for I h at 25 C in the presence of pancreatic ribonuclease A. The sample was diluted to 8.o M-urea by the addition of distilled water and loaded to the column after the gradient was pre-focused for 6 h. Running time was 2o h; initial potential was 15o V. Fractions were collected into microtitre plates and the ph of each fraction was measured. A sample was taken for the measurement of radioactivity. --, Radioactivity; A-- A, ph. Fraction no Ref.... ::: VPI VP~ VP VP 4 Fig. 2. SDS-PAGE analysis of the peak fractions of Fig. I. The peak fractions of Fig. I as indicated by the numbers on top of the gel were heated for 2 min at Ioo C in the presence of 1% SDS and 1%,8-mercaptoethanol and analysed by SDS-disc electrophoresis in a 14 % slab gel. 35S-methioninelabelled poliovirus was co-electrophoresed as a reference (Ref.).

5 Focusing of poliovirus polypeptides Table I. Recoveries of poliovirus polypeptides after isoelectric focusing in urea-sucrose gradient columns r Recovery in ~* ~ ' a Polypeptide Mean value s.d. VPI 6I +5 VP2 9o + 7 VP3 VP4 9o I I * The recoveries were obtained by assuming the following methionine contents (residues of methionine per molecule): VPI = 6"5, VP2 = 5'8, VP3 = 8.8, VP4 = 2.1 (Wouters & Vandekerckhove, I976). 427 method a valuable tool for the analysis and preparation of this polypeptide since earlier methods have always revealed problems with VP 4 (see Discussion). The low recovery of VP~ is due to aggregation and a co-aggregation of VP3. These aggregates are found on top of the gradient (not shown). There are a number of conditions which have to be avoided because of their negative effect on the separation of poliovirus polypeptides. When isoelectric focusing was performed at room temperature instead of 4 C, the amount of aggregate at the top of the gradient increased while the amount of VPI decreased drastically, and also some VP3 was lost. The neutral detergent NP-4o, recommended as a solubilizing aid for isoelectric focusing, had the reverse effect in our system because VPI and VP 3 did not focus in its presence. Thus our procedure achieves optimal separation and recovery. Other conditions, e.g. the use of larger columns or inversion of the ph gradient, did not improve the method; they decreased the recovery or the resolution. Re-focusing of VP2 and VP2L Isoelectric focusing of poliovirus in urea-sucrose gradients gives four peaks containing the four virus polypeptides. However, in a number of experiments two further peaks were observed in different amounts, one at the top as mentioned above and the other between VP2 and VP3. The latter peak (Fig. 3 a) was due to an isoelectric variant of VP2 named VP2z because of its lower isoelectric point (Hamann et al. 1977)- The easy and high recovery of VP2 and VP2~ from electrofocusing gradients enabled us to demonstrate that these two isoelectric variants of VP2 were not interconvertible during isoelectric focusing (Fig. 3b). This finding rules out the possibility that VP2~ is an artefact of this procedure. We have observed that VPzL occurred in virus preparations stored for weeks or months at -zo C (Hamann & Drzeniek, I978). The use of fresh virus material abolished the occurrence of VP2~, and gave a better separation of VP2 and VP3 (Fig. I). Loading capacity of the columns The separation of all four poliovirus polypeptides leading to pure components in a onestep procedure prompted us to determine the maximum amount of poliovirus which could be separated in one column. Up to 7oo #g of poliovirus could be separated with no overlapping of the peaks. However, above 5oo #g the peaks contained small amounts of contaminant polypeptides. Since up to eight columns can be operated in a simple electrophoretic device (Hamann & Drzeniek, 1978), this accounts for 4 mg of poliovirus separated in one run.

6 428 K, 5. W1EGERS AND R. DRZENIEK (a) I I 1 I I I VP1 VP4 VP2 VP2 L VP3? 6-0 X e~ E (b) 15 VP2 VP2L X Fraction number Fig. 3. Isoelectric focusing of'aged' poliovirus and re-focusing of VP2 and VP2~. (a) ~ss-methioninelabelled poliovirus (O--O), which had been stored at -20 C for several months, was dissociated and focused in a urea-containing sucrose density gradient. (b) Fractions 42 (O--O) and 47 (,~--A) from (a) were re-focused in parallel urea-containing sucrose density gradients. Focusing conditions were as described in Fig. I.

7 Focusing of poliovirus polypeptides 429 DISCUSSION A prerequisite for the study of structure and function of virus polypeptides is a method which allows a preparative separation of #g to mg quantities of virus protein with good resolution and good recoveries. The study of specific functions of virus polypeptides especially requires conditions which avoid an irreversible denaturation of the protein. As shown previously, poliovirus is completely dissociated in high concentrations of urea, but reconstituted infectious virus can be recovered upon careful removal of the urea (Drzeniek & Bilello, 1972). On the other hand, it has been shown that by isoelectric focusing in acrylamide gels in the presence of 9 M-urea the four potiovirus polypeptides can be separated and their isoelectric points dctermined (Hamann et al. 1977, 1978; Hamann & Drzeniek, I978). In this paper, the successful adoption of this method of isoelectric focusing in urea to small sucrose density gradients is described. The reproducibility of this method is illustrated by the appearance of the individual polypeptides almost in the same fraction (Fig. I, 3a, 3 b). The ph gradients were reproducible and easy to measure. The apparent pi values are in good agreement with the values found in polyacrylamidc gels (Hamann et al. 1977). The small differences for VP2 and VP3 may be due to the ph range of the ampholytes which wcre used in the density gradients since the main goal of this study was to obtain a distinct separation of the four virus polypeptides in one run rather than to measure exact pi values. Furthermore, the separation was done under slightly different conditions, i.e. in 7 M instead of 9 M-urea and at 4 C instead of room temperature. It should be noted that with our method about 7o % of VP 4 was recovered, which is good since, to our knowledge, the recovery of VP 4 or the corresponding polypeptides of other picornaviruses has been difficult (Stoltzfus & Rueckert, I972). VP4 is also hard to detect after isoelectric focusing in polyacrylamide gels because of its rapid diffusion into the staining solution (Hamann & Drzeniek, I978). Many proteins, when analysed by isoelectric focusing, occur as isoelectric variants (Righetti & Drysdale, 1974). The reason for this microheterogeneity has been discussed in great detail. In some virus preparations of poliovirus type I often an additional band of VP2 is found. This isoelectric variant of VPa, with an isoelectric point of 6.1 instead of 6"4, was called VP2~ (Hamann et al. I977). The fact that VP2 L is also detected in density gradient columns (Fig. 3 a) indicates that it is not an artefact due to the presence of acrylamide during isoelectric focusing in gels (Hamann & Drzeniek, ~978). On the other hand, upon re-focusing no conversion of VP2 to VP2L is seen, indicating that the reason for its appearance is not the isoelectric focusing process itself, e.g. by binding of ampholytes. So far, the chemical basis for this modification of VP2 remains unclear, the most favourable explanation at the moment being deamidation of asparagine or glutamine during storage of the virus. Further investigations, such as amino acid analysis, are under way to clarify this point. The good reproducibility and the high recoveries obtained with this simple technique are especially useful for large-scale preparation of the four poliovirus type t polypeptides for biological and chemical analysis. The Heinrich-Pette-Institut is financially supported by Freie und Hansestadt Hamburg and Bundesministerium ftir Jugend, Familie und Gesundheit, Bonn. We thank Miss M. Hilbrig for the preparation of poliovirus.

8 430 K. 3". WIEGERS AND R. DRZENIEK REFERENCES DRZENIEK, R. & BILELLO, P. (I972). Dissociation and reassociation of infectious poliovirus particles. Nature New Biology 24o, I i HAMANN, A. & DRZENIEK, R. (1978). Isoelectric focusing of viral polypeptides in urea. A methodological study on poliovirus. Journal of Chromatography 147, HAMANN, A., WIEGERS, K. J. & DRZENIEK, R. (I977). Isoelectric focusing and 2D-analysis of poliovirus proteins. Virology 78, HAMANN, A., REICHEL, C., WIEGERS, K. J. & DRZENIEK, R. (I978). Isoelectric points of polypeptides of standard poliovirus particles of different serological types and of empty capsids and dense particles of poliovirus type I. Journal of General Virology 38, o. MAIZEL, J. V. (1962). Evidence for multiple components in the structural protein of type I poliovirus. Biochemical and Biophysical Research Communications 13, MAIZEL, J. V. (I971). Polyaclylamide gel electrophoresis of viral proteins. In Methods in Virology, vol. V, PP Edited by K. Maramorosch & H. Koprowski. New York and London: Academic Press. MAIZEL, S. V. & SUMMERS, D. V. (I968). Evidence for differences in size and composition of the poliovirusspecific polypeptides in infected HeLa cells. Virology 36, MILSTIEN, J. B., WALKER, J. R. & ERON, L. J. 0977). Correlation of virus polypeptide structure with attenuation of poliovirus type I. Journal of Virology 23, 8I MOSS, B. & ROSENBLUM, E. N. (1972). Hydroxylapatite chromatography of protein-sodium dodecylsulfate complexes. A new method for the separation of polypeptide subunits. Journal of Biological Chemistry 247, RIGHETTL P. G. & DRYSDALE, J. W. (I974). Isoelectric focusing in gels. Journal of Chromatography 98, I. STOLTZFUS, C. M. & RUECKERT, R. (1972). Capsid polypeptides of mouse Elberfeld virus. I. Amino acid composition and molar ratios in the virion. Journal of Virology xo, TALBOT, P., ROWLANDS, D. J., BURROUGHS, J. N., SANGAR, D. V. & BROWN, F. (I973). Evidence for a group protein in foot-and-mouth disease virus particles. Journal of General Virology 19, o. VANDE WOUDE, G. r., SWAYEY, J. B. & BACrmACH, ri. L. (I972). Chemical and physical properties of foot'and" mouth disease virus: a comparison with Maus Elberfeld virus. Biochemical and Biophysical Research Communications 48, 1222-I 229. WIEGERS, K. J., YAMAGUCHI-KOLL, U. & DRZENtEK, R. (I977)- Differences in the physical properties of dense and standard poliovirus particles. Journal of General Virology 34, WOUTERS, M. & VANDEKERCKHOVE, J. (I976). Amino acid composition of the poliovirus capsid polypeptides isolated as fluorescamine conjugates. Journal of General Virology 33, YAMAGUCHI-KOLL, U., WIEGERS, K. J. & DRZENIEK, R. (1975). Isolation and characterization of 'dense particies' from poliovirus infected HeLa-cells. Journal of General Virology 26, 3o ZIOLA, B. R. & SCRABA, O. G. 0975). Structure of the Mengo virion. III. Purification and amino acid compositions of the major capsid polypeptides. Virology 64, (Received 1o August I979)