Elsevier

Vaccine

Volume 24, Issues 49–50, 30 November 2006, Pages 7214-7225
Vaccine

Physiochemical and functional characterization of antigen proteins eluted from aluminum hydroxide adjuvant

https://doi.org/10.1016/j.vaccine.2006.06.043Get rights and content

Abstract

We have characterized protein antigens after quantitative dissociation from aluminum hydroxide adjuvant. Bovine serum albumin (BSA) and a multi-antigen vaccine for Group A Streptococcus (GrAS Vaccine) were formulated on aluminum hydroxide, stored for ≥10 days then eluted with a 48-h treatment at 4 °C with 0.85% H3PO4 plus 4 M guanidine HCl (GnHCl). BSA is recovered from adjuvant at 92 ± 2%. GrAS antigens are equally recovered from GrAS Vaccine (95 ± 11% of total protein expected using multiple lots stored for up to 12 months). Recovery after elution is similar when determined by RP-HPLC, SEC-HPLC, UV absorbance, or Bradford methods. Eluted antigens are structurally and functionally intact as judged relative to both treated and untreated antigen controls by SDS-PAGE, RP-HPLC, SEC-HPLC, and after desalting by circular dichroism, bis-ANS binding, and antigenicity determined by ELISA. When formulated and stored for a few weeks, BSA has more dimer (31 ± 5%) relative to the elution control (9% dimer) as detected by SEC-HPLC, suggesting that BSA microaggregation is promoted on aluminum. Antigens eluted from very aged GrAS Vaccine (>12 months) show marked changes by RP-HPLC. Structural changes in the antigens under elution conditions were evaluated using bis-ANS, a fluorescent probe of protein structure. Binding of bis-ANS increases fluorescence ∼100-fold and is significantly diminished with increasing GnHCl concentrations indicating a progressive denaturing of the proteins. At 4 M GnHCl (with or without 0.85% H3PO4) the GrAS antigens are fully denatured and BSA is partially denatured. Interestingly, the addition of 0.85% H3PO4 increases bis-ANS binding on GrAS antigens and reduces the denaturing of GrAS antigens and BSA by chaotropes. Desalting or diluting the eluted antigens results in renaturing of the proteins as judged by bis-ANS fluorescence, circular dichroism and antigenicity testing. The elution method provides a novel approach for high recovery and characterization of GrAS Vaccine antigens and may be applicable to the study of many aluminum hydroxide-bound vaccines.

Introduction

The formulation of effective vaccines often requires the use of an adjuvant to generate a protective immune response [1]. The adjuvants most widely used for many commercial vaccines are insoluble salts of aluminum, such as aluminum hydroxide or aluminum phosphate [2], [3]. Vaccine protein antigens bind avidly to aluminum salts as a function of the isoelectric points of the protein and the aluminum salt [4]. Acidic proteins, in general, exhibit the strongest binding to aluminum hydroxide while basic proteins bind best to aluminum phosphate [4], [5], [6], [7].

Much work has been reported on the mechanisms of antigen binding to aluminum salts [5], [6], [7], [8], [9], [10], [11], [12], [13]. Multiple mechanisms for antigen adsorption to aluminum hydroxide have been described, including electrostatic attraction [4], [5], [6], [7], [9], [10], hydrophobic attraction [6], [8], and ligand exchange of antigen phosphate groups [11], [13]. Antigens may be bound through a combination of these forces [7], [11], [12], [13], although electrostatic interactions appear to dominate in most aluminum hydroxide-containing vaccines described [4], [5], [6], [7], [10], [14], [15], [16].

Once the protein is adsorbed to the aluminum salt it is difficult to analyze the antigen for degradation, aggregation, or other structural changes that may occur during formulation or storage. These difficulties are even greater for multi-antigen formulations. Reports on the characterization of antigens after aluminum binding have only recently become available [10], [16], [17], [18], [19]. One goal of such post-formulation studies is to provide guidance for producing more potent and stable vaccines. Much post-formulation analysis is accomplished by animal potency testing, which is expensive and time-consuming. Analysis of antigens still bound to aluminum salts has been accomplished by use of spectrophotometric and calorimetric techniques [17], [18], [19], which may provide correlates of thermal stability with immunogenicity. These techniques, however, are not readily available or always applicable, especially to multi-antigen formulations.

For analysis of post-formulation antigen by conventional physiochemical and functional methods, antigens have been desorbed from the aluminum adjuvant using surfactants [8], pH changes [9], salts [10], or by ligand exchange with highly negative phosphate ions [5], [7], [10], [14] with varying degrees of effectiveness. A recombinant anthrax antigen eluted from aluminum hydroxide with phosphate was shown to be biologically active [10]; but efficient elution was limited to formulations stored for less than 24 h at 4 °C. Other reports on the binding and elution of aluminum-containing vaccines aged for a few weeks or months suggest an increase in antigen binding over time that is refractory to desorption [5], [14], [20]. Characterization of partially eluted product is not particularly useful since the nature of the non-eluting antigen remains unknown, yet potentially significant.

Progress in the development of a broadly effective vaccine for Group A Streptococcus (GrAS Vaccine) has been described [21], [22], [23]. This vaccine is based on protective M-protein antigens produced by recombinant technologies and is the first such vaccine to be tested in a clinical trial setting. The vaccine is composed of multiple recombinant antigens, each containing multiple type-specific M-protein epitopes and formulated onto aluminum hydroxide adjuvant. As many as 26 epitopes have been formulated and tested in human volunteers [22].

We have developed a novel method to elute and subsequently characterize multiple protein antigens in GrAS Vaccine after formulation with aluminum hydroxide (Rehydragel LV) and storage for 10 days to greater than 12 months. The vaccine samples, along with an antigen control (or mixture), are slowly rocked for 48 h at 4–8 °C in elution buffer. This results in complete recovery of antigen that is structurally and functionally intact as compared to reference or control treated antigens. Efficient recovery of GrAS antigens occurs even from aged vaccine stored for months. The primary structures of eluted antigens are analyzed by SDS-PAGE and RP-HPLC. Higher-order structures are analyzed, after desalting, by SEC-HPLC, circular dichroism, bis-ANS binding, and antigenicity by ELISA. This method is useful to characterize antigens after formulation on aluminum hydroxide and to evaluate the effects of formulation variables and product storage on antigen structure.

Section snippets

Vaccine preparations

GrAS Vaccine contains four purified recombinant protein antigens at equal mass formulated on aluminum hydroxide adjuvant (Rehydragel LV, Reheis, Berkeley Heights, NJ) at a total protein to adjuvant ratio of approximately 1:2 (e.g., 800 μg protein and 1600 μg aluminum hydroxide per mL). Each of the four antigens in GrAS Vaccine, designated as Ia, Ib, II, and III, has a theoretical pI of <6 and contains no phosphate groups. Initial experiments (Table 1) used very aged GrAS Vaccine stored for >12

Results

Recovery of protein eluted from aluminum hydroxide was thoroughly evaluated with multiple methods of protein quantification using a preparation of GrAS Vaccine (batch two) stored for 10–14 days. The accuracy and reproducibility of elution was confirmed by the essentially equivalent recoveries of 108 ± 2%, 115 ± 2%, 107 ± 2%, and 109 ± 11% obtained respectively using Bradford, A280, RP-HPLC, and SEC-HPLC assays. The average recovery of GrAS Vaccine (batch two) was 110 ± 3% and for the formulated BSA was 92

Discussion

We have developed and applied a novel elution method for the characterization of aluminum hydroxide formulated vaccines that provides nearly 100% recoveries of intact antigen as determined by multiple analytical methods. High recovery is observed even with vaccine preparations stored for several months at 4 °C. The elution procedure, which involves 48 h incubation with 4 M GnHCl and 0.85% H3PO4 (final pH of ∼1), has no deleterious effect on the antigens as judged by multiple physiochemical,

Acknowledgements

Thanks to Dr. Ronald Ellis for technical review, Linda Jackson for antigenicity ELISA testing, and Dr. Tsutomu Arakawa for generating circular dichroism spectra.

References (25)

Cited by (0)

View full text