As part of the requirements for full structural characterization of a biopharmaceutical, we are required to investigate the higher order structure (HOS) of the product to get information on the 3-dimensional shape of the molecule. This therefore begs the question: what is the best way to do this and how can we interrogate structure to obtain as full a knowledge as possible?
First of all, it’s important to recognize that the higher order structures of molecules can be composed of several different types of spatial features. These form as a direct result of the primary amino acid sequence and how the constituent amino acid side-chains interact with each other in the solution environment to produce the final 3-dimensional shape.
At the secondary structural level, ordered structural features such as alpha helices or beta sheet structures can be produced, depending on the amino acid sequence, but regions of more random-type structure can also exist. How these secondary structural units interact with one another and assemble in three dimensions gives rise to the tertiary level of molecular structure. Our investigations are therefore not based on the assessment of a single higher order structural feature, but rather on the presence of multiple features that will be present in varying relative levels depending on the nature of the primary amino acid sequence, as well as how the molecule folds to achieve the precise energy state it ultimately adopts. We will only be considering secondary and tertiary structure in this blog. The orthogonal assessment of multimeric assemblies and aggregates has been considered in this previous blog.
How do we investigate a system where a range of HOS features are present and abundances can vary? Not surprisingly the approach that should be taken is one of orthogonality, where different analytical techniques are applied and data brought together in a comparative sense (when considering biosimilarity, for example). The concept of building orthogonality into structural characterization has been covered in a separate blog (Part 1 here and Part 2 here) but it is worth expanding on how this can be specifically applied to HOS analysis.
The idea and application of orthogonality in structural characterization is very highly regarded and encouraged by the regulatory agencies. Thus, it naturally follows that orthogonality in HOS investigations is important for the same reasons. The application of an orthogonal approach to HOS analysis means that data will be derived from various techniques, each of which has its own relative strengths for assessing specific structural features. So, in order to get the best and most meaningful data, different structural techniques need to be employed, thus playing to the strengths of each technique.
At BioPharmaSpec, we employ the techniques of Circular Dichroism (CD), Fourier Transform Infra-Red spectroscopy (FT-IR), fluorescence analysis (both intrinsic and extrinsic) and Nuclear Magnetic Resonance (NMR, 1D and 2D) for HOS analysis at both the secondary and tertiary levels. This allows us the use of techniques with particular sensitivities for the secondary level structures of alpha helices (i.e. CD) or beta sheet (i.e. FT-IR) as well as the ability to assess, in fine detail, the spatial distributions and chemical environments of amino acids within the structure as a whole, through the use of NMR. Intrinsic and extrinsic fluorescence are also often used to investigate surface profiles of proteins through inherent fluorescence and fluorophore interactions.
Figure 1: An expansion of the 1D 1H NMR analysis of two batches of a biopharmaceutical Reference Medicinal Product (RMP) against a single batch of Biosimilar to show the response from the biopharmaceutical component.
CD is also important since it can be performed in the far-UV for assessment of secondary structures and near-UV for tertiary structure analyses. Outputs, in terms of tabulated abundances of the secondary structures detected by CD and FT-IR, are produced through computer processing of raw data and comparison with internal database relative abundances of secondary structures in specified proteins. This means that these techniques are most useful for protein rather than peptide work, where HOS features will be more limited, if present at all, due to the small size of the molecule.
Figure 2: CD Analysis of batches of Simponi (golimumab). Stacked far UV data (left) and stacked near UV data (right).
| Sample | Relative % of α-helix | Relative % of other helix | Relative % of β-sheets | Relative % of turns | Relative % of poly(Pro) | Relative % of other structures |
| 1 | 1 | 2 | 32 | 13 | 10 | 43 |
| 2 | 1 | 2 | 32 | 12 | 10 | 43 |
| 3 | 1 | 2 | 32 | 13 | 9 | 42 |
Table 1: Far UV CD secondary structure fitting data
Since different techniques will measure different aspects of HOS, it is important to remember that the data generated from these techniques will not necessarily match, since different techniques have different sensitivities to the presence of the different higher order structures in the sample. For example, the relative abundances of the various secondary structures will differ based on CD and FT-IR data interpretation for the same product. What matters is that the data between samples matches for each of the techniques used and this can then be taken as a measure of the comparability and/or biosimilarity between samples.
| Sample | Relative % of α-helix | Relative % of β-sheets | Relative % of bends | Relative % of turns | Relative % of random structure |
| 1 | 2 | 41 | 18 | 10 | 30 |
| 2 | 2 | 41 | 18 | 10 | 30 |
| 3 | 3 | 43 | 16 | 11 | 28 |
Table 2: FT-IR secondary structure fitting data based on Amide I wavenumber region, from the same Simponi batches in Table 1 above
It is also worth pointing out that HOS will not be the same between samples if any disulfide bridges present are not generated exactly the same, since these bridges help to support the HOS and hold the 3-dimensional structure. Thus, disulfide bridge analysis, which forms a key part of the primary structure investigations, can also be used as orthogonal evidence that the HOS is the same between different samples. Disulfide bridge analysis is discussed in these case studies.
So, the application of orthogonality to your HOS investigations, both through direct HOS measurements using a variety of instrumentation and a mass spectrometric assessment of the disulfide bridge profile if appropriate, will produce robust, well considered data for this particular aspect of your molecular characterization.
BioPharmaSpec has a full range of methods to interrogate the higher order structure of your molecule. Contact us today to find our what methods are applicable to your project and stage of development.
