Structural Characterization of Recombinant Hepatitis E Virus ORF2 Proteins in Baculovirus-Infected Insect Cells

Structural Characterization of Recombinant Hepatitis E Virus ORF2 Proteins in Baculovirus-Infected Insect Cells

PROTEIN EXPRESSION AND PURIFICATION ARTICLE NO. 12, 75–84 (1998) PT970817 Structural Characterization of Recombinant Hepatitis E Virus ORF2 Protein...

233KB Sizes 21 Downloads 112 Views

PROTEIN EXPRESSION AND PURIFICATION ARTICLE NO.

12, 75–84 (1998)

PT970817

Structural Characterization of Recombinant Hepatitis E Virus ORF2 Proteins in Baculovirus-Infected Insect Cells Robin A. Robinson,*,1 Wilson H. Burgess,† Suzanne U. Emerson,‡ Rande S. Leibowitz,* Svetlana A. Sosnovtseva,* Sergei Tsarev,§ and Robert H. Purcell‡ *Molecular Virology Laboratory, DynCorp, 1 Taft Court, Rockville, Maryland 20850; ‡Holland Laboratory, American Red Cross, 15601 Crabbs Branch Way, Rockville, Maryland 20855; ‡Hepatitis Viruses Section, Laboratory of Infectious Diseases, National Institute of Allergy and Infectious Diseases, National Institutes of Health, 9000 Rockville Pike, Bethesda, Maryland 20892; and §Department of Virus Diseases, Walter Reed Army Institute of Research, Washington, DC 20307

Received July 21, 1997, and in revised form October 2, 1997

The hepatitis E virus (HEV) capsid antigen has been proposed as a candidate subunit vaccine for the prevention of hepatitis E. The full-length HEV ORF2 protein product is predicted to contain 660 amino acids and to weigh 72,000 daltons. Expression of the HEV ORF2 capsid gene from recombinant baculoviruses in insect cells produced multiple immunoreactive proteins ranging in size from 30 to 100 kDa. The most abundant HEV proteins had molecular weights of 72, 63, 56, and 53 kDa. Temporal expression kinetics of these viral polypeptides indicated that the 72- and 63kDa polypeptides were produced abundantly within the initial 36 h. postinfection but were replaced by 56and 53-kDa polypeptides in the cell and medium, respectively, by 48 h postinfection. The 53-kDa protein was secreted as early as 24 h. postinfection, and accumulation in the medium peaked by 72 h postinfection. Purification of the 53-, 56-, and 63-kDa viral polypeptides was accomplished by anion-exchange and subsequent gel filtration chromatography. Sequence analysis of the 53-, 56-, and 63-kDa HEV polypeptides indicated that the amino terminus was amino acid residue 112 of the predicted full-length protein product. The results of carboxy terminal amino acid sequencing indicated that the carboxy terminus of the 53-, 56-, and 63-kDa HEV proteins was located at amino acid residues 578, 607, and 660, respectively. The molecular masses of the 53- and 56-kDa HEV polypeptides were 53,872 and 56,144 as determined by mass spectroscopy. q 1998 Academic Press

Hepatitis E, the major form of acute viral hepatitis in adults in parts of Asia and Africa, is caused by hepa1 To whom correspondence should be addressed. Fax: (301) 7381109. E-mail: [email protected].

titis E virus (HEV) (1–3). HEV infection is usually acquired by consumption of fecally contaminated water. Hepatitis E resembles the clinical disease caused by hepatitis A virus infection but is distinguished by a high mortality during pregnancy (4). The genomes of several HEV strains have been sequenced (5–11). The viral genome contains three open reading frames (ORF) encoding putative nonstructural proteins (ORF1), the viral capsid antigen (ORF2), and a small immunogenic protein of unknown function (ORF3). HEV ORF2 was expressed in transfected mammalian cells as a polyglycosylated 88-kDa protein that was immunoreactive with sera from chimpanzees infected with HEV (12). In contrast, expression of HEV ORF2 from recombinant baculoviruses in insect cells resulted in the intracellular accumulation of a major 55-kDa viral protein and several larger molecular weight species (13–16). Further characterization of the HEV capsid antigens produced in insect cells infected with recombinant baculoviruses containing full-length or truncated forms of HEV ORF2 was undertaken to facilitate vaccine development. Time course infections were performed to determine the kinetics of HEV protein accumulation intracellularly and extracellularly. The recombinant HEV proteins were purified from cell lysates and clarified medium by a three-step purification scheme that included anion-exchange and gel-filtration chromatography. The amino terminal sequences of the HEV proteins were determined by automated micro-Edman degradation. However, the carboxy terminal amino acid sequences of the HEV proteins were determined by three independent methods including carboxypeptidase cleavage analysis, direct amino acid sequencing of Lys C peptides, and amino acid composition analysis of Lys C peptides. The molecular masses of purified HEV proteins were determined by mass spectroscopy. 75

1046-5928/98 $25.00 Copyright q 1998 by Academic Press All rights of reproduction in any form reserved.

AID

PEP 0817

/

6q1a$$$121

01-26-98 19:44:50

pepal

AP: PEP

76

ROBINSON ET AL.

MATERIALS AND METHODS

Cell culture. Spodoptera frugiperda cells, clone 9 (Sf-9), were cultivated as monolayer cultures for plaque assays and transfections and as shaker suspension cultures for production of high-titered virus stocks and recombinant protein. Sf-9 cells were maintained at 287C and 150 rpm in Sf-900 II serum-free medium (SFM) (Life Technologies, Inc., Gaithersburg, MD) in dry-air incubators and were subcultured from a starting density of 0.2 1 106 cells/ml to a final density of 1.0 1 107 cells/ml as suspension cultures up to passage 70. Virus infections. Recombinant Autographa californica multinuclear polyhedrosis baculoviruses (AcMNPV) were passaged in Sf-9 cells (2.0 1 106 cells/ml) at low multiplicity of infection (m.o.i., 0.01). Plaque assays were performed by standard methods with agarose in six-well plates with Sf-9 cell monolayers at 75% confluency. Virus infections for production of recombinant HEV 53- and 56-kDa proteins were initiated at an m.o.i. of 5 pfu/cell and maintained for 4 and 6 days, respectively, when viability decreased to 10%. However, production of the recombinant HEV 63-kDa protein required only a 2-day infection in the presence of protease inhibitors. For production of the recombinant HEV 63-kDa protein, the protease inhibitors—aprotinin (0.2 mg/ml), leupeptin (0.5 mg/ml), and pepstatin A (0.7 mg/ml) purchased from Calbiochem—were added at the time of baculovirus infection and replenished daily. Construction of recombinant baculoviruses. Recombinant baculoviruses (Fig. 1) containing full-length (bHEV ORF2) and a 5*-truncated deletion (bHEV ORF2 5* tr) of HEV ORF2 (Pakistan strain) were constructed previously by standard homologous recombination in Sf-9 insect cells (9, 12). A recombinant baculovirus containing a 5*–3* truncation deletion of HEV ORF2 was constructed by site-specific recombination in Escherichia coli using bacmid vectors (13). Oligonucleotide primers HEV-140 (5*-TTCGGATCCATGGCGGTCGCTCCGGCC-3*) and HEV-141 (5*-TC AAGCTTATCATCATAGCACAGAGTGGGGGGC-3*) were used to amplify a 1512-bp DNA fragment encoding HEV ORF2 amino acids 112 through 607, an ATG translation initiation codon, and multiple stop codons. The PCR fragment was inserted into pCR2.1 (InVitrogen, San Diego, CA) by T/A cloning. The 1514-bp BamHI–EcoRI DNA fragment containing HEV ORF2 (amino acids 112–607) was cloned downstream of the polyhedrin promoter within the polh locus in the baculovirus donor plasmid, pFASTBAC-1 (Life Technologies, Inc.). Recombinant baculoviruses containing the HEV ORF2 DNA were isolated from Sf-9 cells transfected with the recombinant bacmid DNA using the cationic lipid Cellfectin (Life Technologies, Inc.). Plaque-purified virus isolates were screened for the integrity of inserted HEV ORF2 DNA and for HEV pro-

AID

PEP 0817

/

6q1a$$$121

01-26-98 19:44:50

tein expression in insect cells and expanded into a master virus seed bank designated bHEV ORF2 5*–3* tr virus. Infected cell and medium processing. Baculovirusinfected cells and media were separated at indicated times by centrifugation at 500g and 47C for 10 min. Cell pellets (0.5 mg) were resuspended in 1.0 ml of lysis buffer (0.5% NP-40, 50 mM Tris–HCl, pH 8.0, 2 mM EDTA) supplemented with fresh PMSF (final concentration, 0.5 mM). For purification of the recombinant HEV 63-kDa protein from infected cells, a mixture of protease inhibitors, including PMSF (75 mg/ml; Sigma), aprotinin, leupeptin, and pepstatin A (0.2 mg/ml, 0.5 mg/ml, and 0.7 mg/ml, respectively). The resuspended cells were vortexed vigorously for 30 s and incubated for 20 min on ice. Nuclei were pelleted by low-speed centrifugation at 3000g and 47C for 15 min, and the cytoplasmic fraction was collected and used as crude cell lysate. The infected cell lysates and media were clarified by centrifugation in a Sorvall SS34 rotor at 12,000g and 47C for 60 min. Purification of HEV ORF2 protein products. Recombinant HEV ORF2 proteins were purified from clarified baculovirus-infected cell lysates and media separately. The crude cell lysate was diluted 1:10 with loading buffer (50 mM Tris–HCl, pH 8.0, 10 mM NaCl). Diluted crude lysate (1.5 bed volume) was loaded onto a Q Sepharose fast flow (FF) strong anion-exchange column (XK50 column, 5.0 1 7.5 cm, 150 ml; Pharmacia, Piscataway, NJ) at a flow rate of 10.0 ml/min. The column was washed first with 1.0 bed vol of loading buffer at a flow rate of 10 ml/min followed by a second wash with 1.0 bed vol of loading buffer at a flow rate of 20 ml/min. The proteins were eluted with 7.5 bed vol of a continuous linear NaCl gradient (10–300 mM) in loading buffer at a flow rate of 20 ml/min. Additionally, purification of the recombinant HEV 63-kDa protein included an initial step of binding of infected cell lysates to CM Sepharose (XK50 column, 5.0 1 7.5 cm, 150 ml, Pharmacia) at a flow rate of 10 ml/min using loading buffer at pH 4.5. The recombinant HEV 63-kDa protein did not bind to the cation matrix under these conditions and was excluded in the void volume. The void volume fractions were pooled and subjected to anion-exchange chromatography using Q Sepharose FF as presented above. Clarified media were concentrated 10-fold by tangential flow ultrafiltration using a spiral wound cellulosic ultrafiltration cartridge (S1Y10; 1 sq. ft. area; 10,000 MW cutoff; Amicon, Beverly, MA) on an Amicon Proflux M-12 ultrafiltration system at a recirculation rate of 4 L/min. and a transmembrane pressure of 20 psi. The concentrated supernatant was diafiltered against 4 vol of loading buffer. The diafiltrate was loaded directly onto a Q Sepharose fast flow anion-exchange column (150 ml). Aliquots (10 ml) from Q Sepharose column

pepal

AP: PEP

HEV ORF2 PROTEIN PURIFICATION AND CHARACTERIZATION

(Pharmacia) peak protein fractions were subjected to SDS–PAGE analysis to locate fractions containing HEV ORF2 proteins. Pooled fractions were desalted by gel filtration using Sepharose G-25 (Pharmacia) with loading buffer. The peak protein fraction was collected and loaded onto a Source 15 Q high performance (Pharmacia) strong anion-exchange column to resolve and concentrate HEV polypeptides. The column was washed and eluted as described above for Q Sepharose liquid chromatography. Fractions containing HEV proteins were identified by SDS–PAGE and Western blot analyses, pooled, and subjected to size exclusion chromatography using Sephacryl S-200 (Pharmacia) and phosphate-buffered saline (pH 7.2) as a final purification step. All chromatography was performed using a Waters 600E chromatography workstation system (Medford, MA) equipped with Millennium 2010 software for process control and monitoring. Buffer conductivity was determined using an AccuMet 20 conductivity meter. A Corning 220 pH meter was used for determinations of buffer pH. Protein concentrations were determined by the BCA/Pierce microprotein assay at 607C using bovine serum albumin as a protein standard. All buffer components were USP or molecular biology grade raw materials. SDS–PAGE and immunoblot analyses. Proteins were diluted twofold in protein denaturation sample buffer (126 mM Tris–HCl, pH 6.8, 5% b-mercaptoethanol, 20% glycerol, 4% SDS, and 0.005% bromophenol blue) and denatured at 997C for 5 min. Denatured samples were electrophoresed on 8–16% gradient polyacrylamide gels (Novex, San Diego, CA) (19). Proteins were visualized by staining gels with colloidal Coomassie blue stain solution (Novex) as suggested by the manufacturer. Proteins were transferred to nitrocellulose membranes by electroblot techniques (16). HEV ORF2 products were detected chromogenically by binding to primary antiserum (chimp polyclonal a-HEV; 1:500) followed by binding to secondary antiserum (goat ahuman IgG2-conjugated to alkaline phosphatase, 1:5000; Life Technologies, Inc.). NBT/BCIP chromogenic substrate (Life Technologies, Inc.) was used to detect immunoreactive proteins. Amino terminal sequence analysis. Proteins were subjected to polyacrylamide gel electrophoresis in the presence of SDS using the buffer system of Laemmli (19). Proteins were transferred electrophoretically from the gel to a Pro Blot membrane (Applied Biosystems, Foster City, CA) according to the manufacturer’s instructions. Proteins were visualized by Coomassie blue staining, and the HEV proteins were excised for amino terminal sequence analysis using an Applied Biosystems Model 473 gas/pulsed-liquid phase protein sequencer with on-line PTH analyzer.

AID

PEP 0817

/

6q1a$$$121

01-26-98 19:44:50

77

Internal amino acid sequence analysis. Proteins in gels were transferred onto nitrocellulose membranes and visualized with Ponceau S staining. The relevant bands were cut from the membrane and processed for in situ proteolytic digestion with Lys C (Boehringer Mannheim, Indianapolis, IN) according to the procedure of Abersold et al. (20). The Lys C-derived fragments were isolated using a Waters Associates (Medford, MA) high-pressure liquid chromatography system and a Vydac C4 (Hesperia, CA) reversed-phase column. The amino acid sequences of the isolated peptides were determined using an Applied Biosystems Model 477A protein sequencer and Model 120A on-line PTH analyzer. Amino acid analysis. The amino acid compositions of the Lys C-derived fragments described above were determined following vapor phase hydrolysis in 6 N HCl at 1507C for 1 h using a Waters Pico Tag work station. Amino acids were derivatized with phenylisothiocyanate and the resulting PTC amino acids were separated and quantified using a Waters Pico Tag amino acid analysis system. Carboxy terminal sequence analysis. Immobilized carboxypeptidase Y (Pierce, Rockford, IL) was used for the sequential release of amino acids from the carboxy terminus of HEV proteins. Approximately 150 mg of the protein in 800 ml of 0.05 M sodium acetate buffer, pH 5.5, was mixed with a 200 ml suspension of the resin at 377C. Aliquots of the supernatant (100 ml) were taken at 0, 5, 15, 30, 60, 90, and 120 min. A final aliquot was collected at 16 h The samples were dried in vacuo and subjected to amino acid analysis as described above but without the hydrolysis step. Mass spectroscopy. Mass spectrometric detection of purified HEV proteins was performed with A Perkin– Elmer Sciex API-III triple stage quadrupole mass spectrometer (Foster City, CA) equipped with an atmospheric pressure articulated ion spray source. High purity nitrogen served both as the nebulizer gas (operative pressure Å 0.5 MPa) and as curtain gas (flow rate 0.8 I/min.). Argon was used as the target gas at a collision gas mass of 3 1 1015 atoms/cm2. The mass spectra scanning range (m/z) of 100–1500 positive ions was obtained by direct infusion of bovine serum albumin standard solutions (1:200) at 50 ml/min with a Harvard Apparatus Model 11 syringe pump (South Natick, MA). Spectra were collected at 1.0-s intervals. Capillary voltage was maintained at 2 kV and 607C. RESULTS

Kinetics of HEV ORF2 protein expression in insect cells. Temporal expression of the HEV ORF2 gene in baculovirus-infected cells was investigated. Sf-9 insect cells cultivated as shaker suspension cultures in serum-free medium were infected with recombinant baculoviruses encoding the full-length hepatitis E virus

pepal

AP: PEP

78

ROBINSON ET AL.

FIG. 1. Schematic organization of the hepatitis E virus (HEV) genome and recombinant baculoviruses encoding full-length and truncated HEV ORF2 capsid genes. The HEV ORF2 encodes a primary viral capsid protein with a predicted molecular weight of 72 kDa. Proteolytic cleavage products of HEV ORF2 protein produced in insect cells are 63 and 56 kDa.

ORF2 capsid gene (Fig. 1). Cell lysates and media were harvested from virus infections daily for 4 consecutive days and analyzed by SDS–PAGE and immunoblotting. In addition to the predicted HEV 72-kDa protein, multiple HEV-related proteins appeared in infected cells and, unexpectedly, in media. The most abundant of these proteins had molecular weights of 63, 56, and 53 kDa. The HEV 72-kDa protein was detected as early as 1 day postinfection in infected cell lysates and media and accumulated for several more days in cells but disappeared in media by 4 days postinfection (Fig. 2). Another HEV protein (É63 kDa) appeared in infected cells and media by 1 day postinfection and accumulated over the next 2 days. At 4 days postinfection, the level of 63-kDa protein in cells and media decreased dramatically (Fig. 2). Additionally, an HEV 56-kDa protein appeared in cells by 2 days postinfection and in media by 3 days postinfection (Fig. 2B, lanes 4 and 11). The HEV 56-kDa protein accumulated intracellularly at days 3 and 4 postinfection. Finally, an HEV 53-kDa protein appeared in the media at 3 days postinfection and increased in abundance thereafter (Fig. 2B, lanes 11 and 12). Additionally, HEV proteins with other molecular weights but less abundancy were observed intracellularly and extracellularly. These results suggested that a stochastic cascade of proteolytic cleavages occurred during the infection to render capsid proteins of various sizes. HEV protein purification. Recombinant HEV proteins (both intracellular and extracellular) were purified from Sf-9 insect cell cultures infected with fulllength or truncated forms of recombinant baculovi-

AID

PEP 0817

/

6q1a$$$121

01-26-98 19:44:50

ruses using a purification scheme including anion-exchange and gel exclusion chromatography. Cell lysates were prepared for chromatography by differential lysis of infected cells harvested at 5 days postinfection using

FIG. 2. Temporal protein expression of recombinant baculoviruses encoding HEV ORF2 full-length genes. Sf-9 insect cells were infected at an m.o.i. Å 5 with bHEV ORF2 fl virus. Infected cells (lanes 2– 6) and medium (lanes 8–12) were harvested daily for 4 days. Cell lysates and media were fractionated by SDS–PAGE on 8–16% protein gradient gels and stained with colloidal Coomassie blue dye (A) and immunoblotted from duplicate gels (B). Lane 1, See-Blue prestained protein standards (Novex); lane 2, mock infection; lane 3, 1 day p.i.; lane 4, 2 days p.i.; lane 5, 3 days p.i.; lane 6, 4 days p.i.; lane 7, See-Blue prestained protein standards; lane 8, mock infection; lane 9, 1 day p.i.; lane 10, 2 days p.i.; lane 11, 3 days p.i.; lane 12, 4 days p.i.. Lane assignments are similar for A and B.

pepal

AP: PEP

HEV ORF2 PROTEIN PURIFICATION AND CHARACTERIZATION

a nonionic detergent, Nonidet P-40, followed by dilution with Q loading buffer (50 mM Tris–HCl, pH 8.0, 10 mM NaCl) to reduce the ionic strength. In contrast, media harvested from virus infections were clarified by centrifugation, concentrated 10-fold by tangential flow ultrafiltration and subjected to diafiltration against Q loading buffer to reduce the ionic strength. Purification of the recombinant HEV 63-kDa protein required an additional step, cation-exchange chromatography on CM Sepharose, prior to the initial anion-exchange chromatographic step due to the high abundance of cellular proteins at 2 days postinfection. Proteins in equilibrated cell lysates and diafiltered media were captured separately on Q Sepharose fast flow anion-exchange chromatographic matrix. Fractions were analyzed by SDS–PAGE and immunoblot analyses and pooled. HEV intracellular proteins bound to Q matrix and eluted at 140 mM NaCl (Fig. 3A), and those fractions containing the HEV 56-kDa protein were pooled and desalted by passage through a Sephacryl G-25 column. The desalted proteins were subjected to a second round of anion-exchange chromatography using a Source 15 Q strong-anion high-performance matrix. HEV proteins bound to the matrix and eluted again at 140 mM NaCl (Fig. 3B). Peak fractions containing HEV proteins were pooled and fractionated further by size-exclusion chromatography using Sephacryl S-200 (Fig. 3C). Fractions were analyzed by SDS–PAGE and immunoblotting and then pooled. The purity of the final preparation of recombinant HEV 53- and 56-kDa proteins was estimated to be ú98% by staining of gels with colloidal Coomassie blue dye (Figs. 4A and 4B). The HEV 63-kDa protein was purified only to 90% homogeneity using the above purification scheme (Fig. 4C). The use of CM Sepharose prior to binding of cell lysates to Q Sepharose FF delivered a flow through fraction more enriched for the HEV 63-kDa protein. Also, inclusion of protease inhibitors in the running buffers was necessary to prevent further proteolytic cleavage of the HEV 63-kDa protein. Amino terminal sequence analysis. The amino termini of the HEV 63- and 56-kDa intracellular proteins were determined by automated micro-Edman degradation. Pooled HEV protein fractions were collected from Q Sepharose fast flow columns loaded with diluted cell lysates from Sf-9 insect cells infected with bHEV ORF2 fl virus and harvested at 2 days postinfection. The two HEV intracellular proteins were purified from the peak Q fractions (140 mM NaCl) at a ratio of 1:20 (63:56 kDa). Direct Edman degradation of the HEV 63- and 56-kDa protein bands excised from the ProBlot membrane resulted in an identical N-terminal amino acid sequence through 20 cycles. The sequence corresponded to residues 112 through 131 of the HEV ORF2

AID

PEP 0817

/

6q1a$$$121

01-26-98 19:44:50

79

FIG. 3. Protein chromatography elution profiles of cell lysates from bHEV ORF2 virus-infected insect cells. (A) Anion-exchange chromatography of crude cell lysates on a Q Sepharose fast flow strong anion-exchange column using a 0–300 mM linear NaCl gradient in Q loading buffer. (B) Protein elution profile of HEV intracellular 56kDa protein from peak Q fractions on Source 15 Q high-performance strong anion-exchange column using a 0–300 mM linear NaCl gradient in Q loading buffer. (C) Pooled fractions containing the HEV intracellular 56-kDa protein from Source 15 Q chromatography were subjected to gel filtration on a Sephacryl S-200 column.

sequence (Pakistani strain). Additionally, the N-terminal sequence of the HEV 53-kDa extracellular protein was shown to be identical to that of the HEV 56- and 63-kDa intracellular proteins (Fig. 7). Since all three proteins shared the same amino terminus but differed in apparent molecular weights, further processing at the carboxy terminus of these proteins must have occurred. Internal amino acid sequence analysis. Since the 63-kDa protein had the N-terminal 111 amino acids removed, this protein was of the appropriate molecular size to be colinear with the 72-kDa HEV protein at the carboxy terminus. However, the apparent size of the 53- and 56-kDa HEV proteins suggested that they were truncated also at their carboxy termini. Therefore, peptidase digestion and fractionation were performed to

pepal

AP: PEP

80

ROBINSON ET AL. TABLE 1

Summary of Amino Acid Composition Analysis of Fraction 24 from Lys C-Digested HEV 56-kDa Protein

FIG. 4. SDS–PAGE analysis of the 53-kDa (A), 56-kDa (B), and 63-kDa (C) HEV proteins during protein purification. Lane 1, SeeBlue protein MW markers; lane 2, crude material (lysate or media); lanes 3, pooled fractions of HEV proteins eluted at 140 mM NaCl from Q Sepharose FF; lane 4, pooled fractions of HEV proteins eluted at 140 mM NaCl from Source 15 Q; and lane 5, pooled fractions of HEV proteins eluted from Sephacryl S-200.

identify the carboxy terminus of the HEV 56-kDa protein. Purified HEV protein was digested with Lys C protease as described in Materials and Methods. The peptide profile of the resulting Lys C digest is shown in Fig. 5. The peptides were subjected to amino acid sequence analysis. The amino acid sequences corresponded to the expected sequence of the HEV ORF2 (Pakistani strain), except that peptides containing amino acids 608 through 660 were not detected. Of particular interest was fraction 24 which yielded 52 cycles of clear

FIG. 5. Lys C digestion peptide profile of the recombinant HEV intracellular 56-kDa protein. The Lys C digestion peptides of purified recombinant HEV ORF2 56-kDa protein were isolated using a Waters Associates (Medford, MA) high-pressure liquid chromatography system and a Vydac C4 (Hesperia, CA) reversed-phase column.

AID

PEP 0817

/

6q1a$$$121

01-26-98 19:44:50

Amino acid

Expected in residues 554–607

Observed in fraction 24

Asn / Asp Gln / Glu Ser Gly His Arg Thr Ala Pro Tyr Val Met Cysa Ile Leu Phe Lys

4 2 6 4 2 1 5 10 3 4 6 0 0 2 6 0 0

4.4 3.2 5.7 6.3 2.1 2.0 5.0 10 3.3 3.5 6.1 0.7 0b 2.7 6.3 0.6 0.9

a b

No derivatization of Cys was performed prior to hydrolysis. Normalized to 10 Ala.

sequence corresponding to amino acid residues 554 through 605 of HEV ORF 2. Increases in PTH leucine at cycles 53 or 55 (residues 606 or 608) were not observed, although an increase in PTH alanine was observed in cycle 54. Since resolution of more than 50 amino acid residues in a single run was not common in our laboratory, it was unclear whether the failure to obtain additional sequence data was caused by a loss of signal due to peptide termination (i.e., the carboxy terminus of the protein) or due to a failure in Edman chemistry. Therefore, determination of the carboxy terminus of the recombinant HEV proteins by several other means was necessary. Amino acid composition analysis. An alternative means to determine the carboxy terminus of the 56kDa HEV protein was by amino acid composition analysis of the Lys C digestion fraction 24 peptide. The amino acid composition of peptide 24 (Table 1) was as predicted for a 54-amino-acid peptide beginning at amino acid 553 and ending at 607 (leucine). These results provided evidence that amino acid 607 was the carboxy terminal amino acid of the 56-kDa protein. Carboxy terminal sequence analysis. Another method of determining the carboxy terminus of the recombinant HEV proteins was by carboxy terminal amino acid analysis of carboxypeptidase Y-digested proteins. Amino acid analysis of free amino acids released from the HEV 56-kDa protein during timed incubations with immobilized carboxypeptidase Y revealed a rapid increase in leucine followed by valine, serine, and histidine (Fig. 6). Significant increases in the amounts of

pepal

AP: PEP

HEV ORF2 PROTEIN PURIFICATION AND CHARACTERIZATION

FIG. 6. Carboxy terminal amino acid analysis of recombinant HEV ORF2 56-kDa intracellular protein. Amino acid analysis of released amino acids following timed incubations of purified recombinant HEV intracellular 56-kDa protein with carboxypeptidase Y.

other amino acids were not observed in these reactions. These results corroborated the assignment of amino acid 607 (leucine) as the carboxy terminus of the recombinant HEV 56-kDa protein. Amino acid sequencing results of the amino and carboxy termini for the purified HEV 53-, 56-, and 63-kDa proteins are summarized in Figure 7. Amino acids released by carboxypeptidase digestion of the HEV 53-kDa secreted protein were, in succession, arginine, histidine, and glycine (Fig. 7). These results suggested that the carboxy terminus of the HEV 53-kDa protein was located at amino acid 578 of the full-length molecule so the HEV 53-kDa extracellular protein was 29 amino acids shorter at the carboxy terminus than the HEV 56-kDa intracellular protein. Amino acids released by carboxypeptidase digestion of the HEV 63-kDa intracellular protein were, in succession, leucine, glutamic acid, and arginine (Fig. 7). These results suggested that the HEV 63-kDa protein did not undergo proteolytic cleavage at the C-terminus. Mass spectrometric analysis. The molecular weights of the HEV 53- and 56-kDa proteins estimated from the nucleotide sequence of HEV ORF2 (Pakistani

81

FIG. 8. Mass spectroscopic profile of the purified recombinant HEV 56-kDa intracellular protein.

strain) were 50 and 53 kDa, respectively. To determine the absolute mass of these proteins and to discriminate among multiple protein species in purified protein preparations, we performed mass spectroscopy on purified recombinant HEV protein samples. Results from several rounds of MS measurements on the HEV intracellular protein indicated a molecular mass of 56,144 (Fig. 8). Similar MS analysis of the HEV extracellular protein demonstrated a molecular mass of 53,872. Since the degree of accuracy for mass spectroscopy is 0.05%, the proteolytic processing of HEV 53- and 56kDa proteins was confirmed. That the molecular masses were É3000 daltons greater than predicted by the actual amino acid sequences suggested that additional posttranslational modifications may have occurred. Kinetics of HEV ORF2-truncated protein expression in insect cells. Both 5* and 5*–3* truncated deletion mutants of the HEV ORF2 (Fig. 1) were constructed and cloned into baculovirus vectors to study their expression in insect cells. Infections with the 5*-truncated HEV ORF2 virus resulted in the accumulation of 63and 56-kDa proteins primarily in the cells and 63- and 53-kDa proteins extracellularly (Fig. 9). However, the

FIG. 7. Schematic representation of the amino acid sequences predicted from the nucleotide sequence of HEV ORF2 (Pakistani strain) and from the amino acid sequencing results of the purified HEV 53-, 56-, and 63-kDa proteins.

AID

PEP 0817

/

6q1a$$$121

01-26-98 19:44:50

pepal

AP: PEP

82

ROBINSON ET AL.

FIG. 9. Temporal protein expression of recombinant baculoviruses encoding HEV ORF-2-truncated genes. Sf-9 insect cells were infected at an m.o.i. Å 5 with bHEV ORF2 5* tr (A and C) or 5*–3* tr (B and D) viruses. Infected cells and media were harvested daily over 4 days. SDS–PAGE (lanes 1–5) and immunoblot (lanes 6–10) of intracellular proteins (A and B) and extracellular proteins (C and D). Lanes 1 and 6, mock infection; lanes 2 and 7, 1 day p.i.; lanes 3 and 8, 2 days p.i.; lanes 4 and 9, 3 days p.i.; and lanes 5 and 10, 4 days p.i. SeeBlue prestained protein standards were used to determine the molecular weight of indicated proteins.

53-kDa protein was the predominant HEV protein in the medium by three days postinfection. Infections with the bHEV ORF2 5*–3* tr virus produced HEV 56-kDa protein intracellularly and HEV 53-kDa protein extracellularly (Fig. 9). Comparison of the quantities of HEV 56- and 53-kDa proteins revealed that the yield of intracellular HEV proteins was greater in cells infected with 5* tr virus. Conversely, the yield of extracellular HEV proteins was greater from cells infected with 5*–3* tr virus. DISCUSSION

It has not been possible to purify significant quantities of hepatitis E virions because the virus reaches only low titers in animal models and an efficient cell culture system has yet to be discovered. However, the capsid antigen of HEV has been expressed in several eukaryotic systems including insect, mammalian, and yeast cells. In addition to the full-length 72-kDa protein predicted from the sequence of the entire ORF2 gene, abundant ORF 2-related polypeptides with molecular weights of 53, 56, and 63 kDa were detected during infection of insect cells with recombinant baculoviruses. The 53-kDa polypeptide was secreted from insect cells as early as 1 day postinfection and reached maximal levels in the medium by 3 days postinfection. Sequence data suggested that the 63- and 56-kDa polypeptides located intracellularly and the 53-kDa polypeptide located extracellularly may be proteolytic cleavage products of the ORF 2 72-kDa protein.

AID

PEP 0817

/

6q1a$$$121

01-26-98 19:44:50

The recombinant HEV proteins expressed from full-length or truncated ORF2 gene constructs were purified to near homogeneity (98%). The purification scheme entailed gentle nonionic detergent cell solubilization, tangential flow ultrafiltration of media, preparative strong anion-exchange chromatography to capture HEV proteins, high-resolution strong anion-exchange chromatography to resolve HEV proteins, and size exclusion to remove trace impurities from HEV proteins. The yield and purity of HEV proteins from these infections were dependent on several factors, including precursor forms of the ORF-2 gene (full-length versus truncated HEV ORF2), cell lines, and time of harvesting. Typical yields of purified HEV 56-kDa intracellular protein from cells infected with bHEV 5*truncated baculovirus were 15 mg/L. Yields of purified HEV 53-kDa extracellular protein from cells infected with bHEV 5*–3*- truncated baculovirus were 25 mg/ L. The yield of recombinant HEV 63-kDa protein from Sf-9 insect cells infected with the bHEV 5*-truncated baculovirus was considerably less at 1– 2 mg/L of infected cells. McAtee and colleagues reported the purification of baculovirus-expressed HEV proteins (Burmese strain) with molecular weights of 56.5 and 63 kDa (14, 15). The amino terminus of the 56.5- and 63-kDa proteins was amino acid 112, similar to our findings for the HEV 53-, 56-, and 63-kDa proteins. However, the carboxy terminus of their 56.5-kDa protein was reported as amino acid 636, and the carboxy termini of the doublet 63-kDa proteins were reported as amino acids 637 and

pepal

AP: PEP

HEV ORF2 PROTEIN PURIFICATION AND CHARACTERIZATION

652. These results do not correlate with the differences in the molecular weights observed. In contrast, we determined the carboxy terminus of the 56-kDa intracellular protein to be amino acid 607. Although different strains of HEV (Burmese and Pakistani) were used to construct recombinant baculoviruses, reported carboxy termini of the HEV proteins are different, even though the sequences at the putative cleavage sites are identical in both virus strains. Further, the Pakistani 72-kDa protein, in addition to the 53-, 56-, and 63-kDa proteins, was soluble in cell lysates in this study, whereas the Burmese 72-kDa protein was insoluble. Direct comparison of the HEV proteins produced from these two sets of recombinant baculoviruses in the same insect cell lines may provide an explanation for the differences. Another significant finding of this study was that HEV capsid proteins were secreted from baculovirusinfected cells. Secretion had not been reported for baculovirus-expressed HEV proteins previously. Since these proteins were detected as early as 1 and 2 days postinfection when cell viability as determined by trypan blue exclusion was ú95%, the accumulation of these proteins in medium was a result of protein export and not cell lysis. Accumulation of secreted HEV 53-kDa protein to levels ú50 mg/ml of culture media reflected the robust expression and processing of HEV ORF2 gene products with these recombinant baculoviruses and Sf-9 cell line, but not with any Trichoplusia ni cell lines including High 5 cells. Truncation of the ORF2 gene apparently allowed more efficient processing and transport of secretory proteins. The Sf-9 cells used in this study had been adapted for large-scale growth in suspension cultures and selected for high-level recombinant protein secretion. Indeed, other recombinant proteins, such as malaria surface receptors, have been secreted efficiently with this cell line but not with other insect cell lines (R. Robinson, personal communication). That the molecular masses for the HEV 53- and 56kDa proteins were approximately 3 kDa larger than expected from the predicted amino acid sequence suggested posttranslational modification of these proteins. Two sites for N-linked glycosylation and seven sites for myristylation are present in these HEV proteins. Whether these modifications occur and whether they are necessary for immunogenicity will be of great interest. The HEV 56-kDa protein was shown previously to be immunogenic and protective against hepatitis E (but not necessarily HEV infection) in cynomolgus and rhesus monkeys after challenge with a high dose of homologous or heterologous virus (21, 22). The HEV 53- and 56-kDa proteins were shown to be antigenically similar by quantitative ELISA assays using reference antisera from humans infected with HEV (unpublished results). Efforts to purify all four HEV polypeptides, 53, 56, 63,

AID

PEP 0817

/

6q1a$$$121

01-26-98 19:44:50

83

and 72 kDa, and to evaluate their immunogenic and protective potentials as vaccine candidates in primates are under way at present. ACKNOWLEDGMENTS This work was supported by USPHS Contracts N01-AI-35154 and N01-AI-05069 from the National Institute of Allergy and Infectious Diseases and from the World Health Organization Programme for Vaccine Development. We are grateful to Tina Morris and Mario Di Paola for technical assistance with mass spectroscopy measurements and Ewa Szylobryt with amino acid sequence analysis.

Note added in proof. During the revision of this manuscript for publication, a report of extracellular HEV ORF2 proteins from baculovirus-infected High 5 insect cells appeared in the following paper: Li et al,. J. Virol. 71, 7207–7213 (1997). REFERENCES 1. Bradley, D. W., Krawczynski, K., Cook, E. H., Jr., McCaustland, K. A., Humphrey, C. D., Spelbring, J. E., Myint, H., and Maynard, J. E. (1987) Enterically transmitted non-A, non-B hepatitis: Serial passage of disease in cynomolgus monkeys and tamarins and recovery of disease-associated 27 to 34-nm virus-like particles. Proc. Natl. Acad. Sci. USA 84, 6277–6281. 2. Purcell, R. H., and Ticehurst, J. R. (1988) Enterically transmitted non-A, non-B hepatitis: Epidemiology and clinical characteristics, in ‘‘Viral Hepatitis and Liver Disease’’ (Zuckerman, A. J., Ed.), pp. 131–137, A. R. Liss, New York. 3. Bradley, D. W. (1990) Enterically-transmitted non-A, non-B hepatitis. Br. Med. Bull. 46, 442–461. 4. Ticehurst, J. R. (1991) Identification and characterization of hepatitis E virus, in ‘‘Viral Hepatitis and Liver Disease’’ (Hollinger, F. R., Lemon, S. M., and Margolis, H. S., Eds.), pp. 501–513, Williams & Wilkins, Baltimore. 5. Tam, A. W., Smith, M. M., Guerra, M. E., Huang, C.-C., Bradley, D. W., Fry, K. E., and Reyes, G. R. (1991) Hepatitis E virus (HEV): Molecular cloning and sequencing of the full-length viral genomes. Virology 185, 120–131. 6. Tsarev, S. A., Emerson, S. U., Reyes, G. R., Tsareva, T. S., Legters, L. J., Malik, I. A., Iqbal, M., and Purcell, R. H. (1992) Characterization of a prototype strain of hepatitis E virus. Proc. Natl. Acad. Sci. USA 89, 559–563. 7. Huang, C.-C., Nguyen, D., Fernandez, J., Yun, K. Y., Fry, K. E., Bradley, D. W., Tam, A. W., and Reyes, G. R. (1992) Molecular cloning and sequencing of the Mexico isolate of hepatitis E virus (HEV). Virology 191, 550–558. 8. Aye, T. T., Uchida, T., Ma, X.-Z., Iida, F., Shikata, T., Zhuang, H., and Win, K. M. (1992) Complete nucleotide sequence of a hepatitis E virus isolated from the Xinjiang epidemic (1986– 1988) of China. Nucleic Acids Res. 20, 3512. 9. Bi, S.-L., Purdy, M. A., McCaustland, K. A., Margolis, H. S., and Bradley, D. W. (1993) The sequence of hepatitis E virus isolated directly from a single source during an outbreak in China. Virus Res. 28, 233–247. 10. Aye, T. T., Uchida, T., Ma, X.-Z., Iida, F., Shikata, Ichikawa, M., Rikihisa, T., and Win, K. M. (1993) Sequence and gene structure of the hepatitis E virus isolated from Myanmar. Virus Genes 7, 95–110. 11. Yin, S, Purcell, R. H., and Emerson, S. U. (1994) A new Chinese isolate of hepatitis E virus: Comparison with stains recovered from different geographical regions. Virus Genes 9, 23–32. 12. Jameel, S., Zafrullah, M., Ozdener, M. H., and Panda, S. K.

pepal

AP: PEP

84

ROBINSON ET AL. (1996) Expression in animal cells and characterization of the hepatitis E virus structural elements. J. Virol. 70, 207–216.

13. Ho, J., Tam, A. W., Yarbough, P. O., Reyes, G. R., and Carl, M. (1993) Expression and diagnostic utility of hepatitis E virus putative structural proteins expressed in insect cells. J. Clin. Microbiol. 31, 2167–2174. 14. McAtee, C. P., Zhang, Y., Yarbough, P., Fuerst, T. R., Stone, K. L., Samander, S., and Williams, K. (1996) Purification and characterization of a hepatitis E vaccine protein candidate by liquid chromatography-mass spectrometry. J. Chromatogr. B 685, 91–104. 15. McAtee, C. P., Zhang, Y., Yarbough, P., Bird, T., and Fuerst, T. R. (1996) Purification of a soluble hepatitis E open reading frame 1-derived protein with unique antigenic properties. Protein Express. Purif. 8, 262–270. 16. Tsarev, S. A., Tsareva, T. S., Emerson, S. U., Kapikian, A. Z., Ticehurst, J., London, W., and Purcell, R. H. (1993) ELISA for antibody to hepatitis E virus (HEV) based on complete open reading frame-2 protein expressed in insect cells: Identification of HEV infection in primates. J. Infect. Dis. 168, 369–378. 17. Tsarev, S. A., Tsareva, T. S., Emerson, S. U., Govindarajan, S., Shapiro, M., Gerin, J. L., Robinson, R., Gorbalenya, A. E., and Purcell, R. H. (1996) Prospects for prevention of hepatitis E, in

AID

PEP 0817

/

6q1a$$$121

01-26-98 19:44:50

18.

19.

20.

21.

22.

‘‘Enterically Transmitted Hepatitis Viruses’’ (Buisson, Y., Coursaget, P., and Kane, M., Eds.), pp. 373–383, La Simarre, Joueles Tours, France. Luckow, V. A., Lee, S. L., Barry, G. F., and Olin, P. O. (1993) Efficient generation of infectious recombinant baculoviruses by site-specific transposon-mediated insertion of foreign genes into a baculovirus genome propagated in Escherichia coli. J. Virol. 67, 4566–4579. Laemmli, U. K. (1970) Cleavage of structural proteins during the assembly of the head of bacteriophage T4. Nature 227, 680–685. Abersold, R. H., Leavitt, J., Saavedra, R. A., Hood, L. E., and Kent, S. B. H.. (1987) Internal amino acid sequence analysis of proteins separated by one-or two-dimensional gel electrophoresis after in situ protease digestion on nitrocellulose. Proc. Natl. Acad. Sci. USA 84, 6970–6974. Tsarev, S. A., Tsareva, T. S., Emerson, S. U., Govindarajan, S., Shapiro, M., Gerin, J. L., and Purcell, R. H. (1994) Successful passive and active immunization of cynomolgus monkeys against hepatitis E. Proc. Natl. Acad. Sci. USA 91, 10198–10202. Tsarev, S. A., Tsareva, T. S., Emerson, S. U., Yarbough, P. O., Legters, L. J., Moskal, T., and Purcell, R. H. (1994) Infectivity titration of a prototype strain of hepatitis E virus in cynomolgus monkeys. J. Med. Virol. 43, 135–142.

pepal

AP: PEP