Physico-chemical characterization of recombinant hepatitis B surface antigen by a multidimensional approach.

Initially, our work was directed to respond to the question: why hepatitis B surface antigen (HBsAg) produces a very broad peak in preparative size-exclusion chromatography (SEC). For this purpose, we used a multidimensional approach based on SEC fractionation of purified HBsAg followed by the individual analysis of SEC fractions by a battery of assays, such as SEC, sodium dodecyl sulfate-polyacrylamide gel electrophoresis, enzyme-linked immunosorbent assay and transmission electron microscopy. As a result, HBsAg particles were shown to be heterogeneous in terms of particle assembly. In order to elucidate the origin of HBsAg heterogeneity, we included here the denaturing SEC into a multidimensional approach. The data from denaturing SEC evidenced the fragmentation of protein monomers within the HBsAg particle that, probably, occurs during fermentation broth, rather than during in vitro HBsAg processing. The fractions isolated from widely separated regions of HBsAg peak differed in the extent of protein fragmentation, suggesting that the variable extent of protein degradation within HBsAg particles may be one of the factors responsible for broadening of the HBsAg peak in SEC.

Aggregation of recombinant hepatitis B surface antigen induced in vitro by oxidative stress.

In order to examine whether oxygen radicals could be responsible for aggregation of recombinant hepatitis B surface antigen (HBsAg) during its assembly in yeast, purified HBsAg was oxidized with ammonium peroxodisulphate (AP) and analyzed by non-denaturing and denaturing size exclusion chromatography, immunoassay and immunoelectron microscopy. As a result, peroxodisulphate radicals induced a reproducible aggregation of HBsAg. At 44 mM AP, the aggregation process took a few hours and the resulting structures were large, branched and non-antigenic. During more gentle oxidation with 9 mM AP and 20-80 microM Cu2+, a continuous structural modification to HBsAg delaying for tens of hours preceded the aggregation event. During this pre-aggregation period, peroxidation of HBsAg lipids and covalent cross-linking of S protein chains occurred that led a complete loss of antigenicity of oxidized particles. In contrast, yeast-derived HBsAg aggregate is decomposed to S monomers under reducing conditions and recognized by anti-HBsAg polyclonal and monoclonal antibodies, suggesting that is has been assembled in vivo from antigenic and reversibly cross-linked particles. Based on these observations, we conclude that oxidation, at least with respect to the specific molecular sites oxidized by AP, is not a primary event in HBsAg aggregate formation in vivo. Since oxidized HBsAg was shown to be irreversibly cross-linked and non-antigenic, there are no suitable techniques for detection HBsAg oxidation in biological samples. Hence, at present, the magnitude of the in-vivo oxidative damage to HBsAg cannot be evaluated and thus, whether the plasma-derived HBsAg undergoes radical-induced oxidation in the course of viral hepatitis remains to be established. If this occurs, this process is expected to contribute to low HBsAg levels in chronic hepatitis B carriers, failure of the currently available immunoassays to identify HBsAg-positive blood donors and inconsistency in the results provided by HBsAg- and anti-HBsAg-based tests in several recent reports.

Aggregation of recombinant hepatitis B surface antigen in Pichia pastoris.

The combination of immunoaffinity and size-exclusion chromatography (SEC) is a powerful tool to analyze multiprotein particle assembly. This approach was used to investigate the source of aggregation of recombinant hepatitis B surface antigen (HBsAg) detected in purified material. As HBsAg aggregation does not originate in the stresses, such as the concentration of HBsAg solutions, temperature and chaotropic agents, it is less probable that the HBsAg aggregate is produced during the process. To test whether aggregation takes place in vivo, crude yeast extract containing the expressed HBsAg was fractioned on a Sephacryl S-400 column just after cell disruption, and each fraction immunopurified individually. As a result, the HBsAg aggregate was isolated from a fraction corresponding to the elution of large particle aggregates only, not native HBsAg particles. It was biologically active, which demonstrates aggregate formation by specific assembly of partially or wholly folded HBsAg intermediates.

Size-exclusion chromatographic study of the reduction of recombinant hepatitis B surface antigen.

The reduction of the P. pastoris-derived hepatitis B surface antigen (HBsAg) has been investigated by size exclusion chromatography performed in a detergent solution containing 0.3% sodium dodecyl sulfate (SDS) and 0.1 M Tris-HCl, pH 7.0. The HBsAg, reduced under different conditions and passed through the TSK G4000 SW column (600x7.5 mm I.D.) at 0.9 ml min(-1), was resolved into two peaks corresponding to the reduced, monomeric, and non-reduced forms, respectively. Under these conditions, the antigen fraction corresponding to the HBsAg dimer can be separated and completely reduced to monomers by repeated reductive treatment with simultaneous lipid removal. The efficiency of reduction was maximal after sample treatment with an equal volume of a solution containing 417 mM dithiothreitol, 4.2% (w/v) SDS and 16% (v/v) 2-mercaptoethanol. In conclusion, complete reduction of recombinant HBsAg to monomer subunits is possible and depends on the efficiency of lipid removal during the reductive treatment.

Sodium dodecylsulfate polyacrylamide gel electrophoresis of recombinant hepatitis B surface antigen particles.

To investigate the factors leading to broadening of the recombinant hepatitis B surface antigen (HBsAg) peak in size-exclusion chromatography, the HBsAg particles eluting in different regions of the peak were subjected here to electrophoretic analysis. In nonreduced samples, the 24-kD band corresponding to the S monomer was detected when excessively large amounts of HBsAg were loaded onto the gel. Hence, some monomers are not disulfide-crosslinked in assembled particles. On the other hand, the results of alkylation experiments indicated the presence of free sulfhydryl group(s) in a little portion of freshly-purified HBsAg which was retarded on the size-exclusion chromatographic column and had significant antigenicity. This fraction of HBsAg was shown to be oligomeric and capable of spontaneous assembly into higher-order structures during aging.

Evidence for the denaturation of recombinant hepatitis B surface antigen on aluminium hydroxide gel.

Despite the complexity of the subject of protein-alum interactions, a valuable information can be obtained by analyzing the adsorbed and desorbed protein by common physico-chemical methods. In the present work, to approach the adsorption of hepatitis B surface antigen (HBsAg) on alum, the experimental data were supported by complementary analyses of the adsorbed protein by immunoelectron microscopy and the desorbed protein by denaturing size-exclusion chromatography and sodium dodecyl sulfate-polyacrylamide gel electrophoresis under reducing conditions. First, the depletion of HBsAg was investigated. The aspects assessed were the conditions, recovery and chromatographic performance of the desorbed protein. The results obtained strongly suggested the loss of particulate structure of HBsAg after adsorption on alum. This conclusion was further reinforced by direct immunoelectron microscopic visualization of HBsAg in the adsorbed state.

Reversed-phase high-performance liquid chromatographic study of thimerosal stability in Cuban recombinant hepatitis B vaccine.

Non-degraded thimerosal was determined in the presence of its decomposition products by directly assaying recombinant hepatitis B vaccine using a reversed-phase liquid chromatographic method. Methanol-water-orthophosphoric acid (65:35:0.9, v/v/v) was used as the eluent. Salicylic acid was employed as an internal standard. The calibration graph was linear (r = 0.99995) up to 2.5 micrograms of thimerosal. Interference from aluminium hydroxide was eliminated by centrifugation. Good stability of thimerosal in the hepatitis B vaccine was demonstrated. The results obtained were in agreement with the recently proposed mechanism of degradation.