Preparation of Recombinant HIV-1 Gag Protein and Assembly of Virus-Like Particles In Vitro

Abstract

The mechanism of assembly of retroviruses is not fully understood. Purification of retroviral Gag protein and studying its solution state and assembly properties might provide insights into retroviral assembly mechanisms. Here we describe a rapid method for the purification of Gag and its subsequent assembly into virus-like particles in a defined system in vitro. The purification scheme does not use affinity tags, but purifies the native protein by virtue of its high affinity for phosphocellulose, a property presumably related to the affinity of Gag proteins for nucleic acids.

1. Introduction

Expression of a retroviral Gag protein in mammalian cells leads to the production and release of virus-like particles (VLPs) from the cells, even in the absence of other viral gene products or packageable genomic RNA. Isolation of the protein in pure form allows one to characterize it biochemically, and to assess possible contributions from the intracellular environment to the assembly process. As first shown by Campbell and Vogt (1), retroviral Gag proteins expressed in bacteria can be readily purified in native form; these proteins can then assemble into VLPs under defined conditions. This chapter will describe the purification of an HIV-1-derived Gag protein and its use in an in vitro assembly reaction.

The protein we have used in these experiments is Δ16–99 Δp6 Gag from Gross et al. (2). These investigators termed the protein ΔMA-CA-NC-SP2; it contains sequences from both the NL4–3 and BH10 clones of HIV-1. We find that the deletion of residues 16–99 enhances the efficiency with which purified Gag protein assembles in vitro. The p6 domain at the extreme C-terminus of authentic HIV-1 Gag protein has also been deleted from the protein to avoid cleavage by bacterial pro-teases; to our knowledge, the absence of p6 has no significant effect on the assembly properties of the protein (2). Finally, the protein we describe here lacks the N-terminal myristate modification found on HIV-1 Gag produced in eukaryotic cells. We have found that myristoylated Gag protein has very low solubility, and thus we do not know how the myristyl group might affect the assembly properties of recombinant Gag protein. For convenience, we will refer to the protein described here simply as “Gag”.

We present below a rapid procedure for the purification of HIV-1 Gag protein. This was first described by Campbell and Vogt (3). This protocol will typically yield protein of 80–90% purity; this is sufficient for in vitro assembly experiments and electron microscopy of the VLPs, as well as for many biochemical analyses and for use of the protein in immunoblotting, etc. For applications requiring higher purity, the preparation can be “polished” by other techniques such as size exclusion chromatography. With minor modifications, the same procedure can be used for the preparation of Gag proteins from other retroviruses: the modifications would probably be in the ammonium sulfate concentration used at the initial purification step, and in the NaCl concentrations used in the phosphocellulose chromatography steps. It should be noted that we do not use (His)₆ or other affinity tags, to avoid the possibility that these “extra” residues affect the properties of the protein.

We recommend that small aliquots be saved from multiple steps along the purification protocol; analysis of these aliquots by SDS-PAGE (and/or immunoblotting) will provide invaluable information for trouble-shooting if the ultimate yield is low; if the protein has suffered degradation; or if it is still contaminated by significant amounts of bacterial proteins.

2. Materials

2.1. Protein Expression

1. Plasmid (pET series) (Novagen) (we have used pET3xc).

2. BL21 (DE3)pLysS Escherichia coli

3. Luria–Bertani broth (LB), supplemented in all cases with appropriate antibiotics.

4. Isopropyl-beta-D-thiogalactopyranoside (IPTG).

2.2. Protein Purification

1. A 500-W sonicator with probe (Sonics VCX-500, or similar product).

2. Ammonium sulfate (saturated solution at room temperature and crystals).

3. Fibrous cellulose phosphate (Sigma).

4. Buffer A: 20 mM Tris–HCl (pH 7.4), 1 mM phenylmethylsulfonyl fluoride (PMSF), 10 mM β–mercaptoethanol (or 1 mM Tris[2-carboxyethyl] phosphine [TCEP]),1 μM ZnCl₂ (see Note 1).

5. Buffers B/C/D/E are buffer A with 0.1/0.2/0.5/1.0 M NaCl respectively.

6. Buffer F: 20 mM Tris–HCl pH 8.0, 1 mM DTT.

7. Lysis buffer: Buffer A with 0.75 M NaCl and 10% w/v glycerol. Optional additions include protease-inhibitors of choice (Complete Protease inhibitor (Roche), as recommended by the manufacturer), and 0.05% Triton X-100 (w/v) (Roche).

8. Reagents for SDS-PAGE including 4× loading buffer, Coomassie brilliant blue stain, and MW markers.

9. Reagents for Bradford protein assay with IgG standard.

2.3. In Vitro Assembly

1. Dialysis tubing or Slide-A-Lyzer (Pierce) or equivalent microdialysis apparatus.

2. Yeast tRNA (Ambion), 1 mg/ml stock in 20 mM Tris–HCl, pH 7.4.

2.4. Electron Microscopy

1. Parafilm.

2. Carbon-coated formvar grids (Electron Microscopy Sciences #FCF400-Cu) (glow-discharged just before use).

3. Whatman No.1 paper, cut in thin slivers.

4. 2% aqueous uranyl acetate (Polysciences Inc.) solution filtered through 0. 22-μm filter (see Note 2).

3. Methods

3.1. Preparation of Phosphocellulose (4)

1. Weigh out 2 g of PC resin (fibrous cellulose phosphate) in a 50-mL Falcon tube.

2. Add 0.5 M NaOH to the 50 mL mark and resuspend PC by swirling. Do not use a stir bar or rod as this will generate resin fines. Leave the PC in NaOH for exactly 4 min.

3. Pellet the resin by spinning for 30 s at 1, 000 × g in a swinging bucket rotor. Pour off supernatant. It should have a strong ammonia odor.

4. Suspend the resin in 50 mL of water. Pellet the resin and measure the pH of the water supernatant and then decant it off. Repeat this step till the pH of the water supernatant is <10 (—five to six times), then proceed to step 5. These steps should all be performed rapidly to avoid chemical degradation of the PC at high pH.

5. Add 0.5 M HCl to the 50 mL mark and resuspend the resin by swirling. Leave the PC in HCl for exactly 4 min.

6. Pellet the resin and pour off the supernatant.

7. Suspend the resin in 50 mL of water. Pellet the resin and measure the pH of the water supernatant and then decant it off. Repeat this step till the pH of the water supernatant is >4 (—five to six times), then proceed to step 8. These steps should all be performed rapidly.

8. Suspend the resin in 50 mL of 0.2 M Tris–HCl, pH 7.4 buffer and equilibrate 30 min. Pellet the resin and repeat step 8 twice.

9. Suspend the resin in 50 mL of 20 mM Tris–HCl and store at 4 °C (see Note 3).

3.2. Selection of a High-Producer Bacterial Colony (See Note 4)

1. Transform a culture of BL21 (DE3)pLysS bacteria with the pET expression plasmid. Plate on agarose plates with appro-priate antibiotics and incubate at 37 °C overnight.

2. Pick individual colonies and streak each of them on sectors of two more plates, one without and one supplemented with IPTG (0.4 mM).

3. Select a colony that grows well on the IPTG-free plate but inefficiently on the IPTG-containing plate. Proceed to Section 3.3.

3.3. Expression of Gag in Bacteria

1. Grow 200 mL of overnight culture from the selected colony in LB at 37 °C.

2. Transfer 50 mL of the overnight culture to each of four 2-L baffled Erlenmeyer flasks, each containing 1 L LB. Incubate at 37 °C with shaking at 250 rpm. Monitor the growth of the culture by measuring its O.D. at 600 nm.

3. When the O.D. of the culture has reached 0.8, add IPTG to 0.1 g/L.

4. Continue the incubation with shaking for 4–5 h (see Note 5).

5. Harvest the bacteria by centrifugation at 3, 500 × g for 7 min at 4 °C in preweighed bottles. It may be convenient to centrifuge multiple aliquots of the culture in the same bottle, accumulating all of the bacteria in a single pellet.

6. Drain the bottle by inversion and freeze the bacterial pellet at − 80 °C.

3.4. Purification of Gag Protein (See Note 6)

1. Thaw the bacterial pellet without letting it warm up above ice temperature (see Note 7).

2. Weigh the bottle again to determine the added weight due to the bacterial pellet. A good yield is around 3 g of bacteria per liter of induced culture.

3. Resuspend the pellet in 10 mL of lysis buffer per gram of pellet. Disperse the pelleted material by repeated pipetting. The lysate should have sheen (see Note 8).

4. Sonicate the lysate on ice. Continue the sonication until the lysate is freely pipettable with a micropipette tip (see Note 9).

5. Remove insoluble debris from the sonicated lysate by centrifuging at 12, 000 × g for 15 min. Decant the supernatant into a clean beaker and measure its volume. Place the beaker in an ice bath (see Note 10). Resuspend the pellet in 10 mL buffer A and save for subsequent analysis.

6. Add one-half volume (relative to the supernatant) of a saturated ammonium sulfate solution to the beaker with constant stirring (see Note 11).

7. Turn off the stirrer and leave the mixture on ice for an additional 30 min to allow the protein precipitate to accumulate.

8. Centrifuge at 12, 000 × g for 15 min. Carefully remove the supernatant by decanting (see Note 12) and save it until the fractions have been analyzed by SDS-PAGE.

9. Resuspend the pellet in 15–20 mL buffer D by gentle pipetting. Avoid frothing during the pipetting. If a large fraction of the pellet fails to redissolve, the volume of buffer D can be increased. However, it is likely that a minor portion of the pellet will not be solublized at this step.

10. Remove any undissolved material from the solution by centrifuging at 12, 000 × g for 15 min. The supernatant from this centrifugation should be clear; if not, recentrifuge for an additional 20–30 min.

11. Remove the supernatant to a fresh 50 mL Falcon tube by pipetting. It is important to avoid transferring any material from the pellet at this step.

12. Pour 10 mL settled PC resin (see Note 13) from a 50 ml Falcon tube into the solution. Thoroughly mix the PC slurry with the solution by pouring the mixture back and forth between the two 50 mL Falcon tubes.

13. Divide the slurry equally between the two tubes and add buffer A to reduce the NaCl concentration from the 0.5 M present in buffer D to ~ 0. 1 M. Mix by inverting the tubes.

14. Leave the tubes on ice for 15–30 min to allow the Gag protein to bind to the PC. Gently swirl the contents every few min to re-disperse the PC.

15. Centrifuge at 700 × g for 2 min in a swinging bucket rotor. Pour off and discard the supernatant, retaining an aliquot for subsequent analysis.

16. Fill the tube with buffer B. Mix the contents by inverting the tube, pellet the PC as in step 15, and discard the supernatant, retaining an aliquot for analysis. Repeat.

17. Fill the tube with buffer C and wash the PC twice with this buffer as in step 16.

18. Wash with 20 mL of buffer D. Some protein may start eluting at this stage. This is evidenced by increased tendency of the buffer to form bubbles when the tube is inverted. Remove the supernatant with a Pipetman and save in a Corning glass tube (see Note 14).

19. Elute the Gag protein from the PC by adding 10 mL of bufferE. Centrifuge the PC as in step 15, and save the supernatant in a second tube as in step 18.

20. Repeat step 19 with a fresh addition of 10 mL buffer E to the PC pellet. Save the supernatant from this step in a third tube.

21. Repeat step 19 with 7 mL buffer E. Save the supernatant in a fourth tube.

22. Repeat step 19 with 5 mL buffer E. Save the supernatant in a fifth tube.

23. To each of the tubes collected in steps 18–22, add ammonium sulfate crystals to a final concentration of ~ 60% saturation (0.375 g/mL) and dissolve by gentle mixing. This will precipitate the protein in the tubes; it will be obvious from the visible precipitates which tubes contain significant amounts of protein. Incubate the tubes on ice for 1–2 h (see Note 15).

24. Isolate the precipitated Gag protein by centrifuging those tubes containing visible precipitate at 12, 000 × g for 15 min.

25. Redissolve the precipitates in a small volume of buffer D with 10% w/v glycerol (see Note 16), by gentle pipetting. Centrifuge at 20, 000 × g for 15 min to remove any material that failed to redissolve.

26. Dialyze the protein against additional buffer with 10% w/v glycerol.

27. Estimate the concentration and yield of the protein (see Note 17). This estimate can be obtained by inspection of the SDS-PAGE gel; by the optical density of the protein at 280 nm, using a molar absorbance coefficient based on the amino acid composition of the protein; or from a Bradford assay. In the latter case, IgG rather than BSA should be used as the standard in the assay. If assembly experiments will be performed by direct dilution rather than dialysis (see below), it is desirable that Gag be at a concentration of at least 5 mg/mL.

28. Freeze the protein prep in aliquots at −80 °C. Analyze an aliquot of this final preparation, and the aliquots saved over the course of the purification, by SDS-PAGE and Coomassie brilliant blue staining (see Note 18).

3.5. In Vitro Assembly of Virus-Like Particles from Recombinant Gag Protein (See Note 19)

1. Thaw an aliquot of purified Gag protein by gently shaking the tube in a room-temperature water-bath.

2. Remove any precipitated material by centrifugation at 20, 000 × g at 4 °C for 15 min. Estimate the protein concentration in the supernatant by measuring its absorbance. There should not be more than 5% difference in the protein concentration estimates before and after thawing.

3.5.1. Use of Dilution to Achieve Assembly Conditions

1. Dispense 50 μg Gag into one or more microcentrifuge tubes. Add 2 μg yeast tRNA to each tube.

2. Add enough buffer F to bring the NaCl concentration to 0.1 M. The final protein concentration should be around 1 mg/mL. The buffer should be added to the Gag–RNA mixture gradually, with gentle mixing during the stepwise addition.

3. Incubate at 4 °C for 2–4 h.

3.5.2. Use of Dialysis to Achieve Assembly Conditions

1. If the NaCl concentration is to be lowered by dialysis, mix 50 μg Gag protein with 2 μg tRNA as in point 1 of Section 3.5.1. If necessary, dilute the mixture with buffer F containing 0.5 M NaCl to a final Gag concentration of 1 mg/mL.

2. Place the mixture in a microdialysis apparatus and dialyze against buffer F containing 0.1 M NaCl at 4 °C for 2–4 h. The volume of the mixture will increase as the low-salt buffer accumulates inside the dialysis chamber.

3. Collect the material from the dialysis chamber. Be sure to remove any material stuck to the walls, as assembled VLPs are no longer in solution.

4. Analyze the mixture for the presence of VLPs (see Note 20).

3.6. Assessment of Assembly

3.6.1. Assessment of Assembly by Centrifugation

1. Centrifuge the assembly mixture at 20, 000 × g in an Eppendorf centrifuge at 4 °C for 30 min. Remove and save the supernatant in a microfuge tube.

2. Add 17 μL of 4× loading buffer to 50 μL of supernatant.

3. Dissolve the pellet in 67 μL of 1× SDS-PAGE loading buffer.

4. Load 25 μL of the supernatant and pellet fractions onto an SDS-PAGE gel. After electrophoresis, visualize the Gag in the fractions by Coomassie brilliant blue staining.

3.6.2. Assessment of Assembly by Electron Microscopy

1. Place 10 μL of the assembly reaction onto clean Parafilm. Place a carbon-coated Formvar film grid (that has been recently glow-discharged to increase the surface charge density) on the droplet, carbon side down. Allow material from the assembly reaction to adsorb to the grid 2–5 min at room temperature. Wick off excess sample from the grid with thin slivers of What-man no.1 filter paper.

2. Air-dry the grid in a grid box for 5 min at room temperature.

3. Place 10 μL of 2% uranyl acetate on the same side of the grid (this can be accomplished by placing the droplet of uranyl acetate on Parafilm and placing the grid on the droplet, as in 3.6.2 step 1). Allow this staining reaction to proceed for 1–5 min (see Note 21) and again remove excess moisture with Whatman paper wicks. An example of negatively stained VLPs is in Fig. 14.1.

4. Examine the material on the grid in a transmission electron microscope (see Note 22).

Fig. 14.1.

Electron micrograph of VLPs assembled from Δ16–99 p6 Gag by dilution method, as described. Samples were negatively stained with 2% uranyl acetate. Note the heterogeneity in morphology, including some incompletely assembled VLPs.

4. Notes

1. Unless stated, all solutions should be prepared in water with a resistivity of at least 18. 2 MΩand total organic content of less than five parts per billion. Where used, PMSF and β-mercaptoethanol (or TCEP) should be added to buffers just prior to use.

2. Store the filtered uranyl acetate in the dark as exposure to light causes precipitate to accumulate.

3. Suspensions of cellulose phosphate may be stored at 2–8 °C in the presence of a bacteriostat (0.02% w/v sodium azide) for up to 1 year. However, if Gag fails to bind to the PC in 0.1 M NaCl (step 3.4.15), it may be necessary to make a fresh PC preparation, or even obtain a fresh batch of dry PC.

4. For reasons we do not understand, sometimes individual colonies of bacteria transformed with identical plasmid molecules vary markedly in their expression of protein following induction. Colonies with high expression can be recognized by their inefficient growth on agar plates supplemented with the inducer.

5. Shorter inductions are also acceptable. The protein is generally visible by SDS-PAGE of the total bacterial lysate after only ~2. 5 h.

6. All steps from here on are at 4 °C unless otherwise specified. Buffers used should be prechilled to 4 °C before the purification.

7. The safest way to thaw the pellet is on ice, but one can also thaw it in a room-temperature water bath with frequent agitation. In this case, it should be placed on ice as soon as it has thawed. The surface of the thawed pellet should look glassy. This is an indication that the bacteria have been successfully lysed. Another indication of lysis is if the pellet begins to run or slide as the supernatant is decanted. If neither of these signs is observed, it is likely that many of the bacteria did not lyse. In this case, the bottle should be re-frozen and re-thawed.

8. If the lysate looks “chalky”, the lysis was probably incomplete. If this is a persistent problem, the lysis buffer could be supplemented with detergents such as Triton X-100 at 0.05% w/v or with lysozyme at 0.5–2.0 mg/mL. While adding detergents, it is important to consider whether these might interfere with subsequent experiments; in this case, further steps may be necessary to eliminate traces of detergent.

9. In sonicating the bacterial lysate, it is of critical importance to keep the lysate chilled at all times. The vibrating probe of the sonicator produces heat within the lysate. Thus the lysate should be surrounded by ice, but further steps should also be taken to avoid localized heating within the lysate. The sonicator should be on pulse-mode and the lysate should be allowed to re-cool between the sonication pulses. Here, use of a sonicator probe with a temperature sensor helps monitor actual temperature in the lysate. If necessary, the lysate should be sonicated in smaller 50 ml aliquots rather than all at once, so that the entire lysate has a high surface: volume ratio and thus good contact with the ice-bath. The wattage of pulses used, number of pulses delivered, volume of each batch of lysate sonicated, and time allowed to cool for maximum recovery of protein must be determined empirically. We use 12–15 pulses of 0.5 s, each interspersed with 0.7 s gaps, using a 13-mm probe with maximum power in a 500-W ultrasonicator. This regimen is repeated with 1-min intervals in between to cool, until the sample is freely pipettable.

10. It is important that the supernatant be poured off carefully, so that very little pelleted material is brought along with it. The pellet will not be firm at this point.

11. The lysate should be stirred constantly as the ammonium sul-fate is added, and the ammonium sulfate should be added gradually, in a dropwise fashion, at no more than 0.5–1.0 mL/min. These are precautions to ensure that local concentrations of ammonium sulfate never rise above the target concentration, as other proteins from the bacteria will precipitate at higher ammonium sulfate concentrations. We indicate here that a volume of saturated ammonium sulfate equal to half the volume of the lysate should be added, bringing the final ammonium sulfate concentration to 33% saturation. It should be noted that this concentration can be varied if necessary. For example, 40% saturation is required for precipitation of Moloney murine leukemia virus Gag protein. If the yield of HIV-1 Gag protein is particularly high, it may be advantageous to reduce the final ammonium sulfate concentration somewhat, e.g., to 28–30% saturation.

12. After decanting, remove as much of the remaining super-natant as possible by pipetting and drain the pellet by inverting the tube. This is to remove as much of the ammonium sulfate as possible and will help resolublize the pellet.

13. Keep the PC equilibrated in buffer A and temperature-equilibrated on ice, prior to adding it to the protein.

14. All steps after step 18 should use Corning glass tubes: the protein is now quite dilute (and relatively purer, as contaminating bacterial proteins have been largely removed) and will tend to stick to plastic, resulting in significant losses.

15. If the protein yield is low, the tubes can be incubated on ice overnight to improve the recovery.

16. The pellets at this stage should look white. This is usually a sign of good purity with relatively less E. coli contaminating proteins. Add sequential small amounts of buffer D with 10% glycerol (e.g., 100 μL) to the pellets and try dissolving the pellets in the smallest volumes possible.

17. The yield of protein and the relative purity, for different Gag constructs varies. Yields of Gag are usually in the range of 2–5 mg/L of induced bacterial culture with purity of 75–90%, as determined by SDS-PAGE and Coomassie staining.

18. Evaluation of the purity and yield from the purification will help optimize the purification scheme for different Gag constructs. See Fig. 14.2 for an example of such an analysis. It shows the progressive purification of Gag at different stages of the scheme. Note that if a substantial fraction of the Gag protein being expressed is not solubilized in the initial lysis (i.e., is in the pellet saved at step 3.4.5), this may be due to inefficient lysis of the bacteria (see Note 8), but it may also mean that the protein itself has low solubility. In the latter case, temperatures of 30 °C or below should be tried for growth and induction of the bacteria. Addition of detergent or lysozyme can also help increase the recovery of proteins from bacteria; see Note 8. If it is found that a substantial fraction of the Gag remained in the supernatant after ammonium sulfate precipitation (step 3.4.8), then the ammonium sulfate concentration used for the initial precipitation (step 3.4.6) may need to be adjusted.

19. Assembly of the Gag protein into VLPs is induced by addition of nucleic acid, and simultaneously reducing the NaCl concentration from 0.5 to 0.1 M. At this reduced ionic strength, the protein has a much higher affinity for the nucleic acid than it does at 0.5 M NaCl. The dilution to reduce the NaCl concentration in the presence of nucleic acid can either be performed by direct dilution, or more gradually by dialysis. We describe both methods below: direct dilution as procedure 3.5.1 and dialysis as procedure 3.5.2. If the starting protein concentration is low, we suggest using the dialysis procedure.

20. In turn, assembly of the Gag protein into VLPs can be assessed either by a pelleting assay or by electron microscopy. We describe the pelleting as procedure 3.6.1 and the electron microscopy as procedure 3.6.2 here.

21. The length of time that stain is applied to the grid can be varied. Grids can be stained for different times and examined to decide on the optimal staining protocol.

22. If there is too much stain on the grid, so that everything is very dark, then excess stain can be removed by applying a droplet of water to the grid for 15 s. It should then be removed with a Whatman paper wick and the grid can be re-examined.

Fig. 14.2.

SDS-PAGE analysis of different stages of the purification procedure. The arrow on the left shows the position of Gag on the gel. Lanes 1: supernatant from step 3.4.5 after sonication and centrifugation (8 μL); 2: step 3.4.9 before centrifugation (5 μL); 3: step 3.4.10 after centrifugation (5 μL); 4: supernatant from step 3.4.15 – unbound material from PC (25 μL); 5: step 3.4.16 −0. 1 M NaCl wash of PC (25 μL); 6: step 3.4.17 −0. 2 M NaCl wash of PC (25 μL); 7: step 3.4.18 0. 5 M NaCl wash of PC (20 μL); 8 (4 μL) and 9 (20 μL) of pooled fractions from step 3.4.19 to 22 −1 M NaCl washes; 10: molecular weight markers. M.W. indicated next to bands in kDa. Note that some Gag failed to bind the PC at step 3.4.15, but this loss is offset by the fact that a very large fraction of the contaminating proteins has been eliminated here (lane 4).