home > epc > summer 2014 > new potential
European Pharmaceutical Contractor

New Potential

Many of the world’s best-known vaccines require the use of an adjuvant to improve the efficacy of the antigen. Adjuvants broadly fall into one of two categories: they are either immunopotentiators, factors that mimic microbial infection; or particulate agents that enhance the delivery and uptake of the antigen by the antigen presenting cells. Aluminium salts, such as aluminium oxyhydroxide and aluminium phosphate (often erroneously referred to as alum), are the most established particulate adjuvant type (1,2). Although they were discovered early last century, they remain the only licensed vaccine adjuvant in the US, and today they are used in some of the top-selling vaccines, such as Prevnar 13® (Pfizer), PENTAct-HIB® (Sanofi) and Gardasil® (Merck).

Structural Integrity

Binding antigen proteins on the surface of insoluble aluminium nanoparticles, while being crucial for promoting a strong immune response, has implications for the structure of the antigen itself. In these cases, the antigen is no longer free in solution and surrounded predominantly by water molecules – it now has to contend with the constraining ionic interactions binding it to the aluminium salt, as well as the interfacial forces resultant from adhering to a solid surface.

Furthermore, immediately surrounding the surface of the aluminium nanoparticles is a microenvironment that is substantially different from that of the bulk solution. The effect of this microenvironment can produce acid/base conditions up to two pH units different from the bulk solution (3). Consequently, being bound to an aluminium adjuvant can have profound effects upon the structural integrity of the antigen, both in the short term following formulation, and in the long term for stability during storage (4-6).

Traditional Techniques

Despite this, little is known about the effect of aluminium salt binding on protein antigens. This is because protein structural analysis largely uses light-based biophysical techniques. Simply looking at an aluminium adjuvant suspension shows that it is turbid and, in some cases, almost opaque. This does not lend itself to light-based analysis – rapid particle sedimentation readily restricts any time-based analysis of the suspension. Equally problematic is light scattering by the suspension particles, which dissipates any emission signal. To complicate matters further, as the antigen chromophores are not homogenously distributed within the sample, this produces Duysen’s absorption flattening – a wavelength-dependant deviation from the Lambert-Beer law (7). This is the result of the particle absorbing the incident light, and essentially means that the concentration of the antigen behind the particle does not contribute to the effective chromophore concentration.

Traditionally, due to the issues described above, such in situ antigen analysis has not been performed. Instead, traditional solution-based analysis has been employed following antigen desorption. Generally, however, the interaction between the antigen and adjuvant is strong to ensure a high and consistent level of complexation. This means that removing the antigen, without incurring any structural damage, is very difficult – if not impractical. Furthermore, unless quantitative desorption has been achieved, there is the potential for selective desorption of pure antigen, leaving antigen-related impurities still bound to the adjuvant. Moreover, even if successful desorption can be achieved without structural damage, the antigens will have been removed from the interfacial and microenvironment influences. Effectively, this means that antigen integrity can only be inferred from expensive in vivo animal studies, in order to ensure that vaccine safety and efficacy have been maintained.

Circular Dichroism

The single, major biophysical technique for assessing the secondary and tertiary structures of proteins is circular dichroism (CD) (8). CD spectroscopy measures differences in the absorption of left- and right-handed polarised light, which arise due to structural asymmetry. In the far-ultraviolet (UV) spectral region (190-250nm) the chromophore is the peptide bond, and the asymmetric signal arises due to a regular, folded environment such as with the secondary structural features of alpha-helix, beta-sheet and random coil. Well-established algorithms are available to deconvolute the far-UV spectrum to provide quantitative data on the secondary structure components.

With the near-UV spectral region (250-350nm), the chromophores are the aromatic amino acids and disulfide bonds, and the CD signals they produce are sensitive to the overall tertiary structure of the protein. More specifically, signals in the region from 250-270nm are attributable to phenylalanine residues; 270-290nm are attributable to tyrosine; and 280-300nm are attributable to tryptophan. Disulphide bonds give rise to broad, weak signals throughout the near-UV spectrum. This end of the spectrum can be sensitive to small changes in tertiary structure due to protein-protein interactions and/or changes in solvent conditions.

CD is capable of providing detailed secondary and tertiary structural data for the analysis of vaccine antigens; however, as a light-based technique, with colloidal samples, traditional CD suffers from the issues discussed earlier. Consequently, the challenge becomes how to apply CD to determine antigen structure, in situ, in aluminium and emulsion adjuvanted vaccines.

New Approach

To address this challenge of determining protein structure in colloidal vaccines and biopharmaceuticals, solid state CD (ssCD) has been developed (9). This technique incorporates a specialised rotating cell to prevent sample sedimentation and reduce anisotropic effects. Additionally, light gathering is optimised to reduce light-scatter effects, and a specific algorithm is applied to correct the data stream for wavelengthdependent absorption flattening. In combination, the ssCD technological approach enables the collection of both far- and near-UV spectra of colloidal vaccine protein antigens, effectively eliminating the artefacts associated with colloidal aluminium salts and emulsions.

Spectrum Comparison (see the pdf version for all figures and tables)

To compare the benefits of ssCD over classical CD for the secondary structure analysis of an aluminium-based sub-unit vaccine, Antigen Y – both in solution and bound to aluminium hydroxide (Alhydrogel®) – was analysed using far-UV CD. Figure 1, part A demonstrates the classical CD approach in which the bound antigen produced a typical far-UV CD spectrum for a colloidal vaccine: the spectrum demonstrated the classical features of wavelength-dependent flattening, loss of spectral band features, and a poor signal-to-noise ratio compared to the soluble antigen.

In contrast, with ssCD, a uniform dispersion of the sample was achieved, eliminating anisotropic effects and maximising light signal capture. Accordingly, all of the artefacts associated with the analysis of the particulate sample were effectively eliminated, and the free and bound antigen produced similar spectra in shape and signal strength (see B in Figure 1).

Interestingly, Antigen Y retained the secondary structure of the soluble protein when bound to Alhydrogel. As with conventional CD, it is possible to estimate protein secondary structure using algorithms that reference a dataset of protein structures. We have applied one of these algorithms, CDSSTR, to deconvolute the far-UV spectra and estimate the relative secondary structure components (10). From the data, it was evident that there was little difference in the percentage of alpha-helix and beta-sheet components between bound and free Antigen Y (see Table 1). Moreover, these values were similar to those obtained for the protein from X-ray crystallography studies.

In contrast to Antigen Y, the far-UV ssCD spectrum of a different antigen, Antigen Z, revealed structural differences between the free protein and Alhydrogel-bound protein. Increased negative signal was detected with Antigen Z-Alhydrogel, particularly in the 190-215nm region (see Figure 2). This is indicative of increased chirality, which may result from the structural constraints of being anchored to the surface of the adjuvant. Certainly the CDSSTR analysis indicated that there was a marked reduction in beta-sheet component.

Overall, the data from the two antigens indicates that antigen-Alhydrogel binding might have different effects upon the protein secondary structure. It is highly likely that the absence or presence of such effects on secondary structure are dependent on the antigen involved.

Tertiary Structure

ssCD can also be used in the near-UV mode to evaluate antigen tertiary structural features. With the near-UV spectral region (250-320nm), the chromophores are the aromatic amino acids and disulphide bonds, and the CD signals they produce are sensitive to the overall tertiary structure of the protein.

Near-UV ssCD scans of free and Alhydrogel-bound Antigen Y were capable of identifying tertiary structural differences between the two antigen states (see Figure 3). Free Antigen Y produced a characteristic near-UV CD spectrum with two tryptophan maxima at 291nm and 284nm, a tyrosine/phenylalanine minima at 269nm, and a phenylalanine maxima at 253nm. Upon Alhydrogel binding, the tryptophan maxima were reduced in signal strength, as was the minima at 269nm. In the tyrosine region, two new minima were identified at 278nm and 273nm, and the phenylalanine maxima had a reduced signal strength and had sharpened.

From the ssCD spectra, it can therefore be speculated that the immobilisation of Antigen Y on the Alhydrogel particle leads to constraints upon the protein tertiary structure, which results in increased chirality.

Vaccine Formulation

At a basic level, ssCD can provide characterisation data for vaccine development studies. An additional use of the technique is vaccine pre-formulation and real-time stability studies. As with classical CD, the ssCD method can utilise a Peltier temperature-controlled cell, allowing for thermal degradation studies.

To demonstrate the application of ssCD in a forced thermal degradation study, Antigen Z-Alhydrogel was subjected to increasing temperatures and far-UV ssCD was used to compare the secondary structure. With increasing temperature, it was found that both the far-UV spectrum of Antigen Z-Alhydrogel and free Antigen Z revealed changes in secondary structure (see A and B in Figure 4). Most notable was the increase in the minima at 208nm of Antigen Z, particularly as the protein underwent melting around 85ºC. Antigen Z-Alhydrogel revealed a similar rise in the 208nm minima with increasing temperature; however, this was not as pronounced as in the free protein, as the 20ºC baseline value was set at a more negative minima (see C in Figure 4). Interestingly, the 208nm values for both bound and free Antigen Z converged when the protein melted.

Another indicator of structural change is the wavelength crossover point. Both free and bound Antigen Z revealed a red-shift in the wavelength crossover point as the temperature increased – although this was much more pronounced in free Antigen Z (see D in Figure 4). Taken together, this thermal denaturation data indicates that anchoring the Antigen Z protein to the Alhydrogel surface restrained the secondary structural changes associated with free protein denaturation, but ultimately did not prevent the protein melt to occur around 85ºC.

Stability Studies

The results of a real-time stability study are exemplified in Figure 5, where two prototypes comprising Antigen Y bound to Alhydrogel were produced; Antigen Y-Alhydrogel was formulated with the same antigen and adjuvant content, but the vaccines differed in their bulk pH values (pH6 and pH7). Immediately following formulation, ssCD revealed that the bound Antigen Y structure was largely unperturbed, similar to that of an antigen free in solution (data for which is not shown). Aliquots of both prototypes were then stored at 2-8ºC for up to 30 months.

During this time period, far- and near-UV ssCD analysis was performed and the antigen structures were compared. With far-UV ssCD, the Antigen Y-Alhydrogel pH6 vaccine had a significantly reduced signal compared to Antigen Y-Alhydrogel pH7 (see A in Figure 5), indicative of loss of secondary structure. This was particularly evident in the loss of the positive signal in the 190-200nm wavelength range, as well as a marked reduction in the minima signals at the 208nm and 215nm wavelengths.

Antigen Potency

Near-UV ssCD also detected differences in tertiary structure between Antigen Y-Alhydrogel pH6 and Antigen Y-Alhydrogel pH7 (see B in Figure 5). The Antigen Y-Alhydrogel pH7 sample presented spectral features characteristic of Antigen Y tertiary structure; namely, two tryptophan maxima at 291nm and 284nm, a minima at 269nm and a broad positive phenylalanine band at 250-260nm. In contrast, the Antigen Y-Alhydrogel pH6 spectrum had lost most of the structural features, most notably the 284nm and 291nm tryptophan maxima and the broad positive 250-260nm region which had become negative; overall, the data was indicative of a loss of native tertiary structure.

The structural differences observed with the ssCD technique would be expected to have consequences for vaccine potency. Using an animal model to determine potency, it was shown that Antigen Y-Alhydrogel pH6 was significantly less potent than Antigen Y-Alhydrogel pH7. From this it can be deduced that perturbations to the secondary and tertiary structure of Antigen Y result in a less effective vaccine.


As a variant on classical CD, ssCD – for the first time – allows protein structural data to be directly determined in colloidal vaccine formulations, even if they are highly turbid or practically opaque. Technical solutions to the problematic issues of particle sedimentation, light scattering, anisotropy and absorption flattening mean that CD spectra can be produced effectively, eliminating the artefacts which result from having a particulate sample. Accordingly, ssCD can analyse protein containing colloidal formulations, producing spectra equivalent to those of proteins in solution. The technique is directly applicable to understanding antigen structure in situ for aluminium salt vaccines.

Introducing the ssCD technique into the field of vaccine formulation development and manufacture opens up a range of new analytical possibilities. At its most basic, ssCD can be used for characterisation purposes – providing data on the antigen secondary and tertiary structures in the particulate vaccine. Such information can show to what extent, if any, the interaction with the adjuvant affects the antigen secondary and tertiary structure. Additionally in vaccine manufacture, such characterisation data can be used to demonstrate consistent batch reproducibility of antigen structure.

The capabilities of ssCD also extend beyond such analysis to include vaccine formulation and stability studies. Previously, when contemplating formulating an antigen with a particulate adjuvant, the evaluation of excipients was generally based on how they affected the soluble antigen, and these results were extrapolated to the final formulation. Now, such pre-formulation studies can be easily performed in situ on the bound antigen, allowing rapid and direct screening of excipients. Similarly, ssCD has been shown to be stability-indicating and can readily be applied to forced, accelerated, or even real-time stability studies, to determine vaccine antigen consistency and rapidly evaluate stability-enhancing excipients.

Finally, ssCD is a useful tool for monitoring manufacturing and/or stability issues. It can be used to evaluate antigen structure features to ascertain whether issues with the antigen component are responsible for production of an inconsistent batch or a vaccine with reduced stability. ssCD can also be applied at the line stage of the manufacturing process, to monitor batch quality and ensure that the target attributes of the bound antigen are maintained.

Overall, ssCD provides a very powerful tool in a vaccine scientist’s armoury. It opens up the possibility for gaining a deeper understanding of the complex interaction of protein antigens with adjuvant, thus enhancing the potential to provide more potent and stable vaccine formulations to the global market.

1. Lindblad EB, Aluminium compounds for use in vaccines, Immunology and Cell Biology 82: pp497-505, 2004
2. Hem SL and Hogenesch H, Relationship between physical and chemical properties of aluminum-containing adjuvants and immunopotentiation, Expert Rev Vaccines 6: pp685-698, 2007
3. Wittayanukulluk A, Jiang D, Regnier FE and Hem SL, Effect of microenvironment pH of aluminum hydroxide adjuvant on the chemical stability of adsorbed antigen, Vaccine 22: pp1,172-1,176, 2004
4. Jones LS et al, Effects of adsorption to aluminum salt adjuvants on the structure and stability of model protein adjuvants, Journal of Biological Chemistry 280: pp13,406- 13,414, 2005
5. Soliakov A, Kelly IF, Lakey JH and Watkinson A, Anthrax sub-unit vaccine: The structural consequences of binding rPA83 to Alhydrogel®, Eur J Pharm Biopharm 80: pp25-32, 2012
6. Watkinson A et al, Increasing the potency of an alhydrogel-formulated anthrax vaccine by minimizing antigen-adjuvant interactions, Clin Vaccine Immunol 20: pp1,659-1,668, 2013
7. Duysens LNM, The flattening of the absorption spectrum of suspensions, as compared to that of solutions, Biochimica et Biophysica Acta 1: pp1-12, 1956
8. Kelly SM, Jess TJ and Price NC, How to study proteins by circular dichroism, Biochim Biophys Acta 1751: pp119-139, 2005
9. Ganesan A et al, Circular dichroism studies of subtilisin Carlsberg immobilised on micron sized silica particles, Biochimica Biophysica Acta 1764: pp1,119-1,125, 2006
10. Compton LA and Johnson WC, Analysis of protein circular dichroism spectra for secondary structure using a simple matrix multiplication, Anal Biochem 155: pp155-167, 1986

Read full article from PDF >>

Rate this article You must be a member of the site to make a vote.  
Average rating:

There are no comments in regards to this article.

Dr Allan Watkinson has a degree in Biochemistry from Leeds University, and a PhD in Physiological Biochemistry from Imperial College, London. He has extensive experience in R&D gained from over 10 years in academic research, followed by 20 years in industrial research. Over the last decade, Allan has been involved in vaccine development, specialising in formulation, analytics, good manufacturing process and stability. He is an author on multiple peer-reviewed scientific papers, including recent publications on vaccine analysis and formulation. Allan joined XstalBioAnalytics as Principal Scientist (Vaccines), in order to provide expertise in advanced analytical techniques and vaccine formulation.

Dr Jan Vos has a degree in Forensic and Analytical Chemistry, and a PhD in Protein Formulation Chemistry from the University of Strathclyde, Glasgow. He is Project Director at XstalBio, where he manages a range of client and internal R&D projects, addressing the formulation and delivery of biological molecules, including monoclonal antibodies, peptides and vaccines. Jan is a named inventor on several XstalBio patent families and is a senior member of the XstalBio team. He has more than 10 years’ experience in biologics formulation, and is responsible for the design, implementation and data review of all ssCD projects.
Dr Allan Watkinson
Dr Jan Vos
Print this page
Send to a friend
Privacy statement
News and Press Releases

Mr Turhan Ozen Chief Cargo Officer at Turkish Airlines elected to the International Air Cargo Association (TIACA) Committee

Turhan Ozen, serving as Chief Cargo Officer at the global brand Turkish Airlines, was elected to the International Air Cargo Association (TIACA) Committee, the joint voice of the industry that aims to enhance the air cargo profile with intensive efforts towards the global standards.
More info >>

White Papers

e-tools enhance clinical trial performance


In today’s ever-evolving medical business, technological changes require innovation to meet growing market demands and to bring about advances in the industry. One such development currently taking place in the field of clinical research is how e-tools are initiating a boost in the productivity of clinical trials.
More info >>

Industry Events

DIA 2019 Global Annual Meeting

23-27 June 2019, San Diego Convention Center

DIA 2019 is more than just a meeting: it’s where brilliant minds come together to create solutions. Thousands of global innovators will convene to engage in discussions on today’s hottest topics in the life sciences field, propose ways to combat daily challenges, and network to create lasting connections.
More info >>



©2000-2011 Samedan Ltd.
Add to favourites

Print this page

Send to a friend
Privacy statement