Improved purification protocol for wild-type and mutant human foamy virus proteases

Protein Expression and PuriWcation 46 (2006) 343–347 www.elsevier.com/locate/yprep

Improved puriWcation protocol for wild-type and mutant human foamy virus proteases Péter Boross, József Tözsér, Péter Bagossi ¤ Department of Biochemistry and Molecular Biology, Research Center for Molecular Medicine, Medical and Health Science Center, University of Debrecen, H-4012 Debrecen, Nagyerdei krt. 98., P.O. Box 6, Hungary Received 5 July 2005, and in revised form 6 September 2005 Available online 5 October 2005

Abstract Wild-type and an active site mutant (S25T) human foamy virus (HFV) proteases were expressed in fusion with maltose binding protein in Escherichia coli. The mutant enzyme contained a Ser to Thr mutation in the -Asp-Ser-Gly- active site triplet of the enzyme, which forms the “Wreman’s grip” between the two subunits of the homodimeric enzyme. The fusion proteins were puriWed by aYnity chromatography on amylose resin, cleaved with factor Xa, and the processed enzymes were puriWed by gel Wltration under denaturing condition. Refolding after puriWcation resulted in active enzymes with comparable yields. Furthermore, both enzymes showed similar catalytic activities in an oligopeptide substrate representing an HFV Gag cleavage site. However, the S25T mutant showed increased stability in urea unfolding experiment, in a good agreement with the suggested role of the Thr residue of Wreman’s grip. © 2005 Elsevier Inc. All rights reserved. Keywords: Human foamy virus protease; Mutagenesis; Urea unfolding; Enzyme kinetics

The aspartyl protease of retroviruses (PR)1 plays a crucial role in the maturation of virus by cleavage of viral polyproteins in the late phase of virus replication [1,2] and may also have a role in the early phase of life cycle [2,3]. The protease of human immunodeWciency virus type 1 (HIV-1) has been a target for chemotherapy of AIDS for 20 years and the protease inhibitors are important components of the anti-retroviral drug cocktails [4,5]. Much less is known about the PR of another human retrovirus, that of the human foamy virus (HFV). Retroviruses of the foamy virus subgroup have several unusual features: the Pro-Pol polyprotein is synthesized independently from Gag, and the viral particles contain almost full-length reverse-transcribed linear cDNA [6,7]. In addition, HFV polyproteins do not appear to be eYciently processed, except at one Gag-site close to the C-terminus and *

Corresponding author. Fax: +36 52 314 989. E-mail address: [email protected] (P. Bagossi). 1 Abbreviations used: PR, protease of retroviruses; HIV-1, human immunodeWciency virus type 1; HFV, human foamy virus; MBP, maltose binding protein; BSA, bovine serum albumin; wt, wild-type. 1046-5928/$ - see front matter © 2005 Elsevier Inc. All rights reserved. doi:10.1016/j.pep.2005.09.004

at one Pro-Pol site between the reverse transcriptase and the integrase [6,8]. Nevertheless the HFV PR is essential for viral infectivity, because mutation of the HFV PR active site Asp resulted in noninfectious virions [9], as previously found for HIV-1 PR [10]. HFV PR was cloned and expressed as a part of fusion protein with thioredoxin [11], His-tag [12–14], and maltose binding protein (MBP) [15], but the activity was lost after trying to eliminate the heterologous protein part [11,15]. Oligopeptides representing the HFV cleavage sites were hydrolyzed with low catalytic constants, but they were similar to those obtained previously with gag-encoded avian retrovirus PR [16]. The -retroviral proteases are produced in an equimolar amount to the structural Gag proteins, while pol-encoded proteases are present in only 10- to 20-fold lower concentration in the virion. The HFV PR contains a Ser in the active site triplet (-DSG-), similar to the proteases of -retroviruses, but most other retroviruses contain Thr in the same position. Substitution of Ser to Thr in the avian sarcoma leukemia virus protease (S38T) substantially increased its activity [17] while the reverse mutation in HIV-1 PR (T26S) decreased the catalytic eYciency [18,19].

344

P. Boross et al. / Protein Expression and PuriWcation 46 (2006) 343–347

Here, we report a modiWed puriWcation protocol for the HFV PR, which allowed us to purify the processed enzyme into an active form. Furthermore, the Ser of the active site triplet of the enzyme was changed to Thr in an attempt to enhance the activity and/or the stability of the enzyme (S25T; the Ser is the 25th residue of the native enzyme, however, due to the two extra N-terminal residues of the cloned enzyme [15] the mutated residue is actually the 27th from the N-terminus of the cloned enzyme). Materials and methods Site directed mutagenesis of the human foamy virus protease of retroviruses Cloning of the HFV PR coding region after the gene coding for MBP, to produce pMBP-HFV PR, was described previously [15]. The wild-type Ser25 was exchanged for Thr using the Quick-Change mutagenesis protocol (Stratagene, La Jolla, CA, USA) with the following oligonucleotides obtained from Sigma–Genosys (The Woodlands, TX, USA). Mutated position is indicated by underlined letters. 5⬘-G TTA GCC CAC TGG GAT ACA GGG GCA ACA ATA AC-3⬘ and 5⬘-GT TAT TGT TGC CCC TGT ATC CCA GTG GGC TAA C-3⬘. Mutation was veriWed by DNA sequencing performed with ABI Prism dye terminator cycle sequencing kit (Applied Biosystem, Foster City, CA, USA) and an ABI Model 373A sequencer (Applied Biosystem, Foster City, CA, USA). PuriWcation of the wild-type and mutant human foamy virus protease of retroviruses Freshly prepared BL21 (DE3) Escherichia coli culture (500 ml) bearing the plasmid construct coding for the wild-type (pMBP-HFV PR) or a mutant enzyme (pMBPHFV PR-S25T) was grown at 37 °C up to an absorbance at 600 nm of 0.7–1.0, in Luria–Bertani medium containing 100 g/ml ampicillin. Then induction with IPTG (1.0 mM) was done for 5 h and cells were harvested by centrifugation at 2000g for 10 min at 4 °C. After removal of the supernatant, 25 ml lysis buVer (50 mM Tris, pH 7.2, 1 mM EDTA, and 100 mM NaCl) was added to the pellet (3–4 g wet weight). Cells were disrupted by freezing–thawing fol-

lowed by sonication on ice. Samples were centrifuged at 9000g for 15 min at 4 °C. The supernatant was loaded on a column containing amylose resin (10 ml), and extensively washed with the lysis buVer. The fusion protein was eluted with lysis buVer containing 20 mM maltose. Protein concentration of the fractions was determined by the Bradford spectrophotometric method [20] using bovine serum albumin (BSA, Sigma–Aldrich, St. Louis, MO, USA) as a standard protein. Cleavage of the isolated fusion protein (having 2–8 mg/ml protein concentration) was performed in Eppendorf tubes with factor Xa (1 mg/ml, New England Biolabs, Beverly, MA, USA) using 12 h incubation at room temperature with constant rotation, at 2500:1 (w/w) protein/factor Xa ratio. After processing, ammonium sulfate was added (2 M Wnal concentration), and the mixture was incubated on ice for 30 min. Precipitate was collected by centrifugation (16,000g, 20 min). The precipitated protein was dissolved in 500 l lysis buVer containing 0.1% -mercaptoethanol and 2 M urea. The cleaved protease was separated from the remaining MBP by gel Wltration chromatography, using a Superdex G-75 HR 10/30 column (Amersham Pharmacia Biotech, Uppsala, Sweden), equilibrated in the same buVer. Collected fractions containing the protease were gel-Wltrated again and the fractions containing the pure enzyme were immediately dialyzed against lysis buVer containing 0.1% mercaptoethanol without urea, to regain activity. Purity of the protease was assessed by SDS–PAGE, using 10– 20% gradient gels and its protein concentration was determined by the Bradford spectrophotometric method using BSA standard. Rabbit antiserum against the conserved active site region of the HFV protease [21] was used for immunoblotting, performed according to Towbin et al. [22]. Rainbow molecular mass markers (Amersham Biosciences, Piscataway, NJ, USA) were used for comparison. Characteristic values (protein amount, activity, speciWc activity, puriWcation fold, and yield) of the puriWcation steps are shown in Table 1 both for wild-type and mutant HFV PR. Proteolytic assay Oligopeptide substrate SRAVN*TVTQS was synthesized as described previously [15]. Kinetic parameters were

Table 1 Characteristic values of puriWcation steps for wild-type (wt) and S25T mutant HFV proteases Enzyme

wtHFV PR

Step

Cell lysis AYnity chromatography Gel Wltrations and refolding

S25T PR

Cell lysis AYnity chromatography Gel Wltrations and refolding

Total protein

Activity ¡1

SpeciWc activity (mmol min¡1 mg¡1)

PuriWcation (fold)

(mg)

(%)

(mmol min )

(%)

929 30.0

100 3.2

32.5 20.0

100 62

0.035 0.67

— 19

1.8

0.2

5.9

18

3.26

93

643 19.3

100 3.0

8.7 7.6

100 87

0.014 0.39

— 28

2.4

0.4

3.4

39

1.42

101

P. Boross et al. / Protein Expression and PuriWcation 46 (2006) 343–347

determined in 50 mM MES, 100 mM Tris, 50 mM acetate, 1 M NaCl, pH 6.6 buVer (META), in the presence of 3–33 M puriWed enzyme and 0–2 mM substrate at 37 °C for 1 h in volume of 20 l. The reaction was stopped by the addition of 180 l of 1% triXuoroacetic acid, and an aliquot was analyzed by reversed-phase HPLC as described previously [15]. Cleavage products of PR-catalyzed hydrolysis were previously identiWed by amino acid analysis for wild type HFV protease [15] and the mutant enzyme produced the same cleavage fragments characterized by identical retention time. Kinetic parameters were determined by Wtting the data obtained at less than 20% substrate hydrolysis to the Michaelis–Menten equation by using the Fig. P program (Fig. P Software, Durham, NC, USA). The urea denaturation curves were determined at META buVer in the presence 0–2.5 M urea. The urea concentration leading to 50% loss in enzymatic activity (D1/2) was read from the best-Wtted sigmoidal curves.

kDa

1

2

3

4

5

6

7

97 66

MBP - HFV PR

46

MBP

30 21 14

HFV PR

Fig. 1. SDS–PAGE after various puriWcation steps of wild-type HFV protease. Lane 1: molecular weight marker; lane 2: fraction obtained with amylose aYnity chromatography; lane 3: fraction after cleavage with factor Xa; lanes 4 and 5: supernatant and pellet after salting out with ammonium sulfate, respectively; lanes 6 and 7: fraction after Wrst and second gel Wltration, respectively.

A kDa

Results and discussion Recent studies suggested that MBP is very eVective in promoting the solubility of polypeptides to which it is fused, compared to other commonly used proteins, like glutathione S-transferase and thioredoxin [23,24]. Nevertheless, we previously observed that a substantial part of MBP–HFV PR fusion protein already forms aggregates, and puriWcation of active processed enzyme was unsuccessful in the absence of chaotropic agents, due to the increased aggregation (our unpublished results). When we attempted to purify the protease from the MBP by gel Wltration in the presence of 4 M guanidine–HCl (or urea), only residual activity was recovered [15]. The loss of activity after processing from another type of HFV PR fusion protein (containing thioredoxin) has also been reported in the literature [11]. Here, we modiWed the expression/puriWcation protocol, which allowed us to produce pure, processed, active enzyme. Ammonium sulfate precipitation after factor Xa cleavage removed most of the MBP (Fig. 1) and concentrated the processed PR, while reduction of the concentration of chaotropic agent to 2 M and immediate dialysis after gel Wltration were the necessary modiWcations to yield active enzyme (Fig. 2). By using these modiWcations 20–40% of the total activity was regained from the factor Xa cleaved HFV wt PR, MBP mixture. The same puriWcation protocol was successfully applied for S25T mutant enzyme (Fig. 2). Based on our results, salting out with a moderate concentration (1.5–2 M) of ammonium sulfate is very useful in puriWcation of at least some MBP– retroviral protease construct, as Wrst reported for the puriWcation of murine leukemia virus protease [25]. After salting out of the reaction mixture of factor Xa, the majority of MBP remains in the supernatant and the pellet contains the processed PR, the low amount of unprocessed MBP– PR fusion protein, and the minority of MBP. The

B 1

2

3

345

C

kDa

46

46

30

30

1

2

3 S

P1 P2 20

20

14

14

a

b 6

6

Fig. 2. SDS–PAGE (A), immunoblot (B), and activity (C) of puriWed wild-type and S25T mutant HFV proteases. The wild-type (lane 2) and the S25T mutant (lane 3) HFV proteases were puriWed as described in Materials and methods, then subjected to SDS–PAGE (A) and immunoblotting using a speciWc anti-HFV PR antibody (B). Rainbow molecular weight markers were used for comparison (lane 1). Same amount of puriWed wild-type (chromatogram a) and mutant (chromatogram b) enzymes were assayed with the oligopeptide substrate SRAVN*TVTQS to demonstrate their similar speciWc activity (C). P1: SRAVN; P2: TVTQS, S: uncleaved substrate.

346

P. Boross et al. / Protein Expression and PuriWcation 46 (2006) 343–347

Table 2 Kinetic parameters determined for the wild-type (wt) and S25T mutant HFV proteases for substrate SRAVN*TVTQS Enzyme

Km (mM)

kcat (s¡1)

kcat/Km (mM¡1 s¡1)

wtMBP-HFV PR wtHFV PR MBP-S25T PR S25T PR

0.56 § 0.10 0.32 § 0.03 0.66 § 0.14 0.27 § 0.04

0.006 § 0.0005 0.003 § 0.0001 0.009 § 0.0007 0.003 § 0.0001

0.011 § 0.002a 0.010 § 0.001 0.013 § 0.003 0.010 § 0.002

a

These values were determined previously [15].

diVerence between the molecular weights of a retroviral protease monomer (10–17 kDa) and the MBP (43 kDa) and a MBP–PR fusion protein (53–60 kDa) gives a convenient way for further puriWcation by gel Wltration. The activity of puriWed wild-type HFV PR was compared to the previously determined parameters obtained for the wild-type fusion protein (Table 2). While both the Km and kcat values were lower for the processed enzyme, the speciWcity constant was found to be identical within the experimental error of the measurements. The S25T mutant enzyme showed similar characteristics: the Km and kcat values were changed, but the kcat/Km value was the same for the MBP-S25T fusion protein and for the pure S25T mutant enzyme. These results are in line with our previous Wndings, where the fusion protein and the processed but unpuriWed protein showed identical relative activity, pH optimum, and urea denaturation proWle [15]. The conserved network of hydrogen bonds (“Wreman’s grip”) involving the two -D-T/S-G- active site triplets is an important structural feature of aspartic proteases [26], including the homodimeric retroviral proteases [27]. Its main function was considered to stabilize the active site geometry and the dimeric structure of the enzyme, but it is not absolutely required for activity [28,29]. Our result showed that the speciWcity constant was not altered if the Ser25 was changed to Thr in HFV PR, therefore the presence of Ser in the active site triplet does not appear to be an important determinant of the low catalytic eYciency of the enzyme. This result was independent from the studied

relative activity (%)

100

80

60

40

20

0 0.00

0.25

0.50

0.75

1.00

1.25

1.50

1.75

2.00

2.25

2.50

urea (M)

Fig. 3. Urea denaturation curves for processed and puriWed wild-type (open triangles) and S25T (open circles) mutant HFV PRs. Continuous lines are the best-Wtted sigmoidal curves.

(fusion or processed and puriWed) forms of the enzyme. Our data are in good agreement with the Wndings that the same substitution in the foamy viral genome did not cause a change in virion infectivity [9]. We have also compared the dimer stability of the wild-type and mutant enzymes by measuring their urea denaturation curves (Fig. 3). The urea concentration leading to 50% loss in enzymatic activity (D1/2) of the wild-type HFV PR was 0.5 M and the Ser25 to Thr mutation increased this value to 0.8 M, in good agreement with the suggested role of this residue in dimerization [28–30]. D1/2 value of HIV-1 PR is 1.85 M [31], which suggests that other residues of HFV PR structure may also be responsible for the substantially lower dimer stability of HFV PR as compared to HIV-1 PR. Acknowledgments We thank János Kádas for help in mutagenesis, Dr. István Balogh for DNA sequencing, Mónika Kónya for help in protein puriWcation and protease assays, and Szilvia Petö for technical assistance. This research was sponsored by the Hungarian Science and Research Fund (OTKA F25807, F34479, F35191, and T43482) and the János Bolyai Fellowship of the Hungarian Academy of Sciences (to Peter Bagossi). References [1] S. Oroszlan, R.B. Luftig, Retroviral proteinases, Curr. Top. Microbiol. Immunol. 157 (1990) 153–185. [2] J. Tözsér, S. Oroszlan, Proteolytic events of HIV-1 replication as targets for therapeutic intervention, Curr. Pharm. Des. 9 (2003) 1803– 1815. [3] M. Rumlova, T. Ruml, J. Pohl, I. Pichova, SpeciWc in vitro cleavage of Mason-PWzer monkey virus capsid protein: evidence for a potential role of retroviral protease in early stages of infection, Virology 310 (2003) 310–318. [4] J.J. Eron, HIV-1 protease inhibitors, Clin. Infect. Dis. 30 (Suppl. 2) (2000) S160–S170. [5] J. Tözsér, Stages of HIV replication and targets for therapeutic intervention, Curr. Top. Med. Chem. 3 (2003) 1447–1457. [6] M. Linial, Why aren’t foamy viruses pathogenic? Trends Microbiol. 8 (2000) 284–289. [7] A. Rethwilm, The replication strategy of foamy viruses, Curr. Top. Microbiol. Immunol. 277 (2003) 1–26. [8] R.M. Flügel, K.I. Pfrepper, Proteolytic processing of foamy virus Gag and Pol proteins, Curr. Top. Microbiol. Immunol. 277 (2003) 63–88. [9] J. Konvalinka, M. Löchelt, H. Zentgraf, R.M. Flügel, H.G. Krausslich, Active foamy virus proteinase is essential for virus infectivity but not for formation of a Pol polyprotein, J. Virol. 69 (1995) 7264–7268. [10] N.E. Kohl, E.A. Emini, W.A. Schleif, L.J. Davis, J.C. Heimbach, R.A. Dixon, E.M. Scolnick, I.S. Sigal, Active human immunodeWciency virus protease is required for viral infectivity, Proc. Natl. Acad. Sci. USA 85 (1988) 4686–4690. [11] K.I. Pfrepper, M. Löchelt, M. Schnolzer, R.M. Flügel, Expression and molecular characterization of an enzymatically active recombinant human spumaretrovirus protease, Biochem. Biophys. Res. Commun. 237 (1997) 548–553. [12] K.I. Pfrepper, H.R. Rackwitz, M. Schnolzer, H. Heid, M. Löchelt, R.M. Flügel, Molecular characterization of proteolytic processing of the Pol proteins of human foamy virus reveals novel features of the viral protease, J. Virol. 72 (1998) 7648–7652.

P. Boross et al. / Protein Expression and PuriWcation 46 (2006) 343–347 [13] K.I. Pfrepper, M. Löchelt, H.R. Rackwitz, M. Schnolzer, H. Heid, R.M. Flügel, Molecular characterization of proteolytic processing of the Gag proteins of human spumavirus, J. Virol. 73 (1999) 7907–7911. [14] K.I. Pfrepper, J. Reed, H.R. Rackwitz, M. Schnolzer, R.M. Flügel, Characterization of peptide substrates and viral enzyme that aVect the cleavage site speciWcity of the human spumaretrovirus proteinase, Virus Genes 22 (2001) 61–72. [15] G. Fenyöfalvi, P. Bagossi, T.D. Copeland, S. Oroszlan, P. Boross, J. Tözsér, Expression and characterization of human foamy virus proteinase, FEBS Lett. 462 (1999) 397–401. [16] J. Tözsér, P. Bagossi, I.T. Weber, T.D. Copeland, S. Oroszlan, Comparative studies on the substrate speciWcity of avian myeloblastosis virus proteinase and lentiviral proteinases, J. Biol. Chem. 271 (1996) 6781–6788. [17] G. Arad, M. Chorev, A. Shtorch, A. Goldblum, M. Kotler, Point mutation in avian sarcoma leukaemia virus protease which increases its activity but impairs infectious virus production, J. Gen. Virol. 76 (1995) 1917–1925. [18] J.R. Rose, L.M. Babe, C.S. Craik, DeWning the level of human immunodeWciency virus type 1 (HIV-1) protease activity required for HIV-1 particle maturation and infectivity, J. Virol. 69 (1995) 2751–2758. [19] J. Konvalinka, M.A. Litterst, R. Welker, H. Kottler, F. Rippmann, A.M. Heuser, H.G. Krausslich, An active-site mutation in the human immunodeWciency virus type 1 proteinase (PR) causes reduced PR activity and loss of PR-mediated cytotoxicity without apparent eVect on virus maturation and infectivity, J. Virol. 69 (1995) 7180–7186. [20] M.M. Bradford, A rapid and sensitive method for the quantitation of microgram quantities of protein utilizing the principle of protein–dye binding, Anal. Biochem. 72 (1976) 248–254. [21] V.A. Morozov, T.D. Copeland, K. Nagashima, M.A. Gonda, S. Oroszlan, Protein composition and morphology of human foamy virus intracellular cores and extracellular particles, Virology 228 (1997) 307–317.

347

[22] H. Towbin, T. Staehelin, J. Gordon, Electrophoretic transfer of proteins from polyacrylamide gels to nitrocellulose sheets: procedure and some applications, Proc. Natl. Acad. Sci. U.S.A. 76 (1979) 4350–4354. [23] R.B. Kapust, D.S. Waugh, Escherichia coli maltose binding protein is uncommonly eVective at promoting the solubility of polypeptides to which it is fused, Protein Sci. 8 (1999) 1668–1674. [24] C. Wang, A.F. Castro, D.M. Wilkes, G.A. Altenberg, Expression and puriWcation of the Wrst nucleotide-binding domain and linker region of human multidrug resistance gene product: comparison of fusions to glutathione S-transferase, thioredoxin and maltose binding protein, Biochem. J. 338 (1999) 77–81. [25] A. Fehér, P. Boross, T. Sperka, S. Oroszlan, J. Tözsér, Expression of the murine leukemia virus protease in fusion with maltose binding protein in Escherichia coli, Protein Expr. Purif. 35 (2004) 62–68. [26] L. Pearl, T. Blundell, The active site of aspartic proteinases, FEBS Lett. 174 (1984) 96–101. [27] L.H. Pearl, W.R. Taylor, A structural model for the retroviral proteases, Nature 329 (1987) 351–354. [28] K. Strisovsky, U. Tessmer, J. Langner, J. Konvalinka, H.G. Krausslich, Systematic mutational analysis of the active-site threonine of HIV-1 proteinase: rethinking the “Wreman’s grip” hypothesis, Protein Sci. 9 (2000) 1631–1641. [29] M. Ingr, T. Uhlikova, K. Strisovsky, E. Majerova, J. Konvalinka, Kinetics of the dimerization of retroviral proteases: the “Wreman’s grip” and dimerization, Protein Sci. 12 (2003) 2173–2182. [30] P. Bagossi, Y.S. Cheng, S. Oroszlan, J. Tözsér, Activity of linked HIV1 proteinase dimers containing mutations in the active site region, Protein Eng. 9 (1996) 997–1003. [31] E.M. Wondrak, N.T. Nashed, M.T. Haber, D.M. Jerina, J.M. Louis, A transient precursor of the HIV-1 protease. Isolation, characterization, and kinetics of maturation, J. Biol. Chem. 271 (1996) 4477– 4481.