Business & Finance

Large-scale production of recombinant hepatitis B surface antigen from Pichia pastoris

Large-scale production of recombinant hepatitis B surface antigen from Pichia pastoris
of 11
All materials on our website are shared by users. If you have any questions about copyright issues, please report us to resolve them. We are always happy to assist you.
Related Documents
  Journal of Biotechnology 77 (2000) 157–167 Large-scale production of recombinant hepatitis B surfaceantigen from  Pichia pastoris Eugenio Hardy *, Eduardo Martı´nez, David Diago, Rau´l Dı´az,Daniel Gonza´lez, Luis Herrera Center for Genetic Engineering and Biotechnology ,  P . O .  Box  6162  ,  Ha  ana  10600  ,  Cuba Received 25 January 1999; received in revised form 2 August 1999; accepted 29 September 1999 Abstract The ability of the  Pichia pastoris -based technology for large-scale production of recombinant hepatitis B virussurface antigen (HBsAg) and both reproducibly purify HBsAg and remove most of the relevant contaminants wasascertained by evaluating ten industrial production batches, five in 1993 and five in 1998. At an early stage, theclarification of mechanically disrupted yeast cells by acid precipitation renders HBsAg with a purity as low as3.8  0.6%. However, by adsorption / desorption from diatomaceous earth matrix, the purity of HBsAg rapidlyincreases to 18.8  5%, which is suitable for chromatographic processing. This step also eliminates non-particulatedforms of HBsAg, significantly lowers the amount of carbohydrates and lipids, and concentrates the HBsAg 4.8-fold.Finally, a sequential purification procedure that includes large-scale immunoaffinity, ion-exchange, and size-exclusionchromatographies further purifies the preparation, resulting in a product (HBsAg at a concentration of 1.3  0.2 gl − 1 ) with a purity of 95% or more. Furthermore, each of the other contaminants measured reaches the following lowlevels per 20   g HBsAg: host deoxyribonucleic acid (  10 pg), carbohydrates (1.2  0.02   g), lipids (14  0.28   g),immunopurification-released immunoglobulin G (less than 100 ppm), and endotoxins (106.7  19.3 pg). These valuesare below those specified for recombinant DNA hepatitis B vaccines according to World Health Organization (WHO)guidelines. © 2000 Published by Elsevier Science B.V. All rights reserved. Keywords :   Production; Recombinant; Hepatitis B; Vaccine;  Pichia pastoris / locate /  jbiotec 1. Introduction Infection with hepatitis B virus leads to theproduction of large virus particles of about 42-nmtogether with surface-antigen particles of 22-nmaverage size (termed HBsAg) that are stabilizedby disulfide bonds and contain carbohydrates andphospholipids (Dane et al., 1970; Robinson et al.,1974; Tiollais et al., 1985). Vaccination of several Abbre  iations :   DNA, deoxyribonucleic acid; ELISA, en-zyme-linked immunosorbent assay; HBsAg, recombinant hep-atitis B virus surface antigen; Ig, immunoglobulin; WHO,World Health Organization.* Corresponding author. Fax:  + 53-7-336008. E  - mail address : (E. Hardy)0168-1656 / 00 / $ - see front matter © 2000 Published by Elsevier Science B.V. All rights reserved.PII: S0168-1656(99)00201-1  E  .  Hardy et al  .  /   Journal of Biotechnology  77 (2000) 157–167  158 million individuals world-wide with HBsAg parti-cles purified from the plasma of asymptomatichuman carriers of hepatitis B virus infection hasbeen used for almost two decades as an effectivemeans of preventing hepatitis B-associated healthproblems such as liver failure, cirrhosis, or hep-ato-carcinoma. Unfortunately, although plasma-derived vaccines have repeatedly been shown tobe safe and effective (Stephenne, 1988, 1990), theyhave several drawbacks: (i) the manufacturingprocess must involve tedious, stringent and time-consuming procedures to inactivate infectioushepatitis B virus and often living pathogens thatmight be present in plasma and are responsiblefor blood-transmitted diseases such as acquiredimmuno-deficiency syndrome; (ii) the cost of pro-duction is relatively high, which is against theimplementation of mass immunization programsby health services of poor and even medium-levelcountries; (iii) the availability of suitable amountsof human plasma is limited; and (iv) a lengthy(about 6 months) innocuity test in chimpanzees isrequired.To overcome the above limitations there isobviously a need for new technologies to producelarge quantities of HBsAg particles for use inhepatitis B vaccines, obviating isolation andpurification of natural HBsAg from humanserum. Suitable technologies should be designedtaking into account the World Health Organiza-tion (WHO) guidelines for hepatitis B (WHO,1989), and then, the HBsAg obtained must resem-ble the natural antigen. To this end, an intensivevalidation work should be done including: (a) adetailed description of the main manufacturingsteps and their rationality; (b) an evaluation of the effectiveness of each step; and (c) a demon-stration of the potency, efficacy, and safety of thevaccine manufactured.By the early 1980s, advances in genetic engi-neering and biotechnology allowed the first hep-atitis B vaccine to be obtained by formulation of HBsAg produced in recombinant strains of theyeast  Sacharomyces cere  isiae , thus overcomingthe above-limitations (Stephenne, 1990; Elliot etal., 1994). Following cultivation of the  S  .  cere -  isiae  containing the gene encoding HBsAg, theantigenic protein was released from the yeasts bycell disruption and purified by several extractionand chromatographic steps.In line with the  S  .  cere  isiae -based approach,the technology for the large-scale obtainment of HBsAg was developed, based on the expression of the HBsAg gene under control of the  Pichia pas - toris  alcohol oxidase I enzyme gene promoter anda transcription terminator of the glyceraldehyde 3  phosphate dehydrogenase gene of   S  .  cere  isiae .The HBsAg gene was integrated by homologousrecombination to the genome of the  P .  pastoris strain MP-36 (his 3).Given that this genetic construction lacks thesecretion signal required for the traveling of theprotein through the glycosylation path, the yeast-expressed HBsAg was found to be not glycosy-lated (Cremata, J., 1995. Personal commun-ication). More than 85% of the HBsAg sequencewas corroborated by studies based on mass spec-trometry peptide mapping and Edman degrada-tion (Padro´n et al., 1989). These studies alsoproved that both N- and C-terminal ends of theprotein closely match that of the natural protein.The downstream process for the purification of HBsAg was established first at laboratory scale(Pento´n, 1992; Agraz et al., 1993; Pa´ez et al.,1993; Agraz et al., 1994; Pe´rez et al., 1994) andfurther scaled-up from about 250- to 500-fold(Pe´rez et al., 1994). However, the effectiveness of the key steps for the large-scale production of   P .  pastoris -derived HBsAg had not been describedyet.The performance of the production technologyin terms of its ability to both render a highly pureHBsAg and remove most of the intrinsic (yeasttotal proteins, nucleic acids, carbohydrates, lipids)and extrinsic [immunopurification-released im-munoglobulin (Ig) G, endotoxin] contaminants isanlayzed here. The results obtained verified thatthis technology satisfies most of the WHO re-quirements for the safe purification of yeast-derived, biologically-active HBsAg particles.Consequently, the vaccine (HEBERBIOVAC HB,Heberbiotec S. A, Cuba) which is formulated with P .  pastoris -derived HBsAg has proven to be safeand efficacious, providing protection against hep-atitis B infection (Pento´n et al., 1994).  E  .  Hardy et al  .  /   Journal of Biotechnology  77 (2000) 157–167   159 2. Materials and methods 2  . 1 .  Materials High-quality reagents for yeast cultivation,chromatography and other purposes were pur-chased from international suppliers. The im-munoaffinity chromatographic matrix with animmobilized CB HEP 1 anti-HBsAg murinemonoclonal antibody was supplied by the Im-munotechnology and Physical-Chemical Divisions(CIGB, La Habana, Cuba). All bioreactors werefrom CHEMAP, Switzerland. 2  . 2  .  Methods 2  . 2  . 1 .  Large - scale production of HBsAg  Ten independent industrial batches of HBsAg,five made in 1993 and five in 1998, were obtainedas previously described (Pe´rez et al., 1994).Briefly, the recombinant  P .  pastoris  yeast strainwas kept as a master seed lot at  − 70°C toguarantee that each bioreaction run was startedfrom the same srcinal preparation. Under care-fully controlled multiplication conditions, the HB-sAg gene-containing yeast cells were passed fromshake flasks into bioreactors of 75, 300 l, andfinally into a bioreaction unit of 3000 l. Themedium used for yeast growth in the bioreactorscontained mainly common inorganic salts [e.g.(NH 4 ) 2 SO 4 , KH 2 PO 4 , MgSO 4 ] supplemented withglycerol, vitamins (e.g. biotin, riboflavin), andtrace elements (e.g. KI, CuSO 4 × 5H 2 O, FeCl 2 ).To guarantee optimal growth, bioreactors wereoperated at 30°C, pH about 5, under continuousagitation and aeration. When the glycerol in the3000-l bioreactor was depleted, methanol wascontinuously added at a flow of 2.9–10.9 g l − 1 h − 1 to induce the synthesis of HBsAg that accu-mulates intracellularly. While maintaining an ade-quate balance between biomass increase andmethanol supply, vitamin and trace elements wereperiodically added to the bioreactor. At the end of the bioreaction, the culture reached cell densitiesof 75–88 g dry-cell weight l − 1 . Samples wereremoved regularly from the different bioreactorsfor in-process control which included a determi-nation of cell density, HBsAg expression, or mi-crobial purity. After harvesting, the yeast cellswere disrupted (Pa´ez et al., 1993) to recover andpurify HBsAg by a series of well-established steps.These included acid precipitation (Pa´ez et al.,1993), adsorption / desorption from diatomaceousearth matrix (Agraz et al., 1993), and finally,successive purification through immunoaffinity,ion-exchange and gel-filtration chromatographicprocedures (Agraz et al., 1994; Pe´rez et al., 1994).Each step was currently checked to be withinspecifications of the WHO guidelines for qualityand / or in-process control procedures. Also, thequality of the water and all buffer solutions weremonitored for conductivity, pH and microbiologi-cal or pyrogenic contaminants, and were strictlycontrolled. 2  . 2  . 2  .  Analytical methods The HBsAg concentration was measured by anenzyme-linked immunosorbent assay (ELISA)system which uses sheep anti-HBsAg polyclonalantibodies for plate coating, followed by incuba-tion with anti-sheep IgG antibodies-horse radishperoxidase conjugate (Gonza´lez et al., 1993).The HBsAg working standard preparation wascalibrated against the Paul Erlich Institute(Frankfurt, Germany) standard. Total proteincontent was determined by the method of Brad-ford (1976) using bovine serum albumin asstandard. Carbohydrates were determined accord-ing to Carney (1986); lipids according to Wood-man and Price (1972). Deoxyribonucleic acid(DNA) was determined by a dot blot hybridiza-tion procedure according to Sambrook et al.(1989), using probes of repetitive regions of the  32 P-marked  P .  pastoris  yeast chromosome. The con-centration of mouse IgG was determined by asandwich ELISA using a commercial (Sigma, St.Louis, MO) goat anti-mouse IgG (wholemolecule), affinity-isolated antibody, adsorbedwith human serum proteins. Endotoxin concen-tration was determined by limulus amebocytelysate assay according to the manufacturer’s in-structions (Chromogenix AB, Sweden) using lipo-polysaccharides from  Escherichia coli   0111:B4 asstandard and the chromogenic substrate Acetyl-Ile-Gly-Gly-Arg-Pna·HCl. Electrophoresis on12.5% sodium dodecyl sulfate (SDS)-polyacryl-  E  .  Hardy et al  .  /   Journal of Biotechnology  77 (2000) 157–167  160 amide gels was performed as described byLaemmli (1970); gel-separated proteins werestained with Coomassie blue R-250 as usual andthen analyzed by gel densitometry. 3. Results and discussion 3  . 1 .  Disruption of HBsAg  - containing P .  pastoriscells and assembly of mature HBsAg particles The large-scale HBsAg production started bymultiplication of an srcinal recombinant  P .  pas - toris  strain (Pento´n, 1992) in Petri dishes followedby multistep passage of the yeast cells undercarefully controlled multiplication conditionsfrom shake flasks through intermediate bioreac-tors to a large-scale bioreactor unit. Once the cellgrowth finished in the bioreactor, the culture wasrapidly cooled to 4°C to minimize proteolysis.Then, the yeast cell wall was quantitatively dis-rupted with the aid of a bead-mill disrupter torelease intracellular HBsAg protein (Pa´ez et al.,1993). The disruption buffer (20 mmol Tris–HCl,pH 8, containing 5 mmol l − 1 EDTA, 10 g l − 1 sucrose) was supplemented with concentratedpotassium thiocyanate.Observing the velocity sedimentation through asucrose gradient, it was found that a large frac-tion of the cell-released HBsAg was aggregated(Fig. 1). Wampler et al. (1985) evidenced by su-crose gradient centrifugation and by SDS-PAGEanalysis under reducing or non-reducing condi-tions that the HBsAg that is initially liberatedfrom yeast cells is a non-disulfide-bonded aggre-gate of monomeric subunits. This aggregate canbe converted into fully disulfide-bonded particlesthat resemble the natural HBsAg by treatmentwith 3 mol l − 1 thiocyanate, which is suggested tofacilitate exchange (intrachain to interchain)within already oxidized HBsAg polypeptides(Wampler et al., 1985). Intrachain and interchaindisulfide linkages between dimers and higher mul-timers are known to be responsible for stabilizing Fig. 1. Distribution of HBsAg particle size during the downstream process. Samples (1 ml) corresponding to the main steps of theprocess were subjected in parallel to a sedimentation speed of 25000 rpm [sucrose (from 20 to 60%) gradient for 16 h at 4°C in aBeckman RPS 40T rotor]. After collection of fractions (from 1 to 13, 0.5 ml each) from the bottom of the tube, ultracentrifugation-separated samples were assayed for HBsAg content by specific ELISA as described under Section 2.  E  .  Hardy et al  .  /   Journal of Biotechnology  77 (2000) 157–167   161Fig. 2. Electrophoresis profile of total proteins after of eachstep of the downstream process. Samples were electrophoresedin a glycine SDS-PAGE system (Laemmli, 1970) using adual-slab gel (7.5 × 10 × 0.1 cm) device from Sigma. Thissystem used a 3.5% stacking gel and a 12.5% separating gel.The gel was photographed on a white background to contrastthe Coomassie blue-stained protein bands. Samples per lanes:(1) supernatant of cellular disruption; (2) supernatant of acidprecipitation; (3) eluant of semipurification by diatomaceousearth matrix; (4) eluant of immunoaffinity chromatography;(5) eluant of ion-exchange chromatography; (6) eluant of size-exclusion chromatography. The gel-separated bandsmarked with an arrow correspond to the HBsAg monomer(  24 kDa) and dimer. Batches and date of evaluations aredescribed under Table 1. similar properties (molecular weight, isoelectricpoint, hydrophobicity). Following disruption, thecell debris was diluted with potassium thio-cyanate-containing disruption buffer and the pHadjusted from 8 to an optimum value of 4.5(Pento´n, 1992; Pa´ez et al., 1993). Consequently,many of the unwanted yeast proteins of isoelectricpoints around 4.5 and the non-protein contami-nants precipitated simultaneously. Acid precipita-tion was followed by centrifugation to remove theprecipitated proteins, carbohydrates, and lipids(Table 1, AP). With respect to the previous step,the purity of HBsAg increased slightly to 3.8  0.6% (Fig. 4, AP), though the content of both,carbohydrates (Fig. 3, AP) and lipids (Fig. 4, AP)per 20   g of HBsAg, did not change.Some investigators have described the yield of HBsAg after acid precipitation in a small (mg)-scale to be in the range of 80 (Craig and Siegel,1989) to 97% (Sugahara et al., 1986). In theauthors’ experience, when working in the large(g)-scale, the acid-precipitated HBsAg is recov-ered in a smaller amount ranging from 43 to 60%.That, however, is still acceptable as a yield-effec-tive downstream purification. 3  . 3  .  Diatomaceous earth matrix - assisted semipurification Following acid precipitation, the clarified HB-sAg preparation was subjected to batch adsorp-tion at acid pH onto diatomaceous earth matrix(Hyflo Super Cell) followed by elution from thematrix at low ionic strength and basic pH (Agrazet al., 1993). At pH 4 and a loading concentrationranging from 100 to 150   g HBsAg / g Hyflo SuperCell, the loading capacity is described to be ashigh as 90–95% (Agraz et al., 1993). Althoughnot known, the mechanism underlying adsorptionis most likely governed by electrostatic and hydro-phobic interactions between the matrix and theproteins. While contributing a yield of 29–45%and a still relatively low rate of purification (Table2, SP), this step was found to offer the followingbenefits: (i) the subpopulation (fractions from 1 to3 in Fig. 1) that contains non-particulated formsof HBsAg such as oligomers was eliminated, and,at the same time, the mature disulfide-bondedthe correct three-dimensional structure of highlyimmunogenic HBsAg particles.The purity of HBsAg (hereafter defined as theamount of HBsAg in a 1-ml sample, relative tothat of total yeast proteins) was found as low as1.5  0.3%. At this early stage, the HBsAg prepa-ration was found associated with a large amountof intracellular cytoplasmatic proteins (SCD un-der either Table 1 or Fig. 2) together with othercontaminants such as carbohydrates (SCD undereither Table 1 or Fig. 3), lipids (SCD under eitherTable 1 or Fig. 4), and cell debris, which must beremoved to meet the specifications for the finalHBsAg product. 3  . 2  .  Acid precipitation Proteins are probably the most difficult con-taminants to remove, since those remaining withthe product at the end of purification may have

Comfort women

May 17, 2018


May 17, 2018
Related Search
We Need Your Support
Thank you for visiting our website and your interest in our free products and services. We are nonprofit website to share and download documents. To the running of this website, we need your help to support us.

Thanks to everyone for your continued support.

No, Thanks