HOS Comparisons FAQs
What is Higher Order Structure?
- Usually, Higher Order Structure (HOS) refers to protein structure beyond the amino acid sequence (primary structure). In the context of CD spectroscopy, the protein subunit arrangement is excluded, and HOS refers specifically to secondary and tertiary structure.
- Secondary structure refers to local structural elements that are stabilised mainly by hydrogen bonds. The most common secondary structure elements are α-helices and β-strands, which can make up β-sheets if arranged in larger folds, as well as turns and random coils.
- Tertiary structure refers to the overall fold of a single protein molecule, which is stabilised by interactions between amino acid side chains that can be located far apart in the amino acid sequence. These interactions include disulfide bonds and non-covalent interactions such as ionic bonds, hydrophobic interactions, and Van der Waals forces between nonpolar residues.
Generally, any CD spectrum is the sum of contributions from all chiral chromophores in the sample, and individual spectral bands usually reflect overlapping contributions. As such, spectral features in protein CD spectra can normally not be assigned to specific motifs or domains, let alone individual amino acid residues.
- Far-UV CD data, typically obtained for wavelengths up to 260 nm, informs about secondary structure. In addition to secondary structure elements such as α-helices and β-sheets, disulfide bonds can contribute to the signal around 230 nm.
- Two different proteins will show very similar spectral profiles in the far-UV if they have similar contents of secondary structure, even if they are largely different in size or tertiary structure.
- Near-UV CD data, typically obtained in the range of 250 nm to 350 nm, informs about tertiary structure. The signal is influenced by the nature of the surrounding environment of aromatic side chains of the amino acids tryptophan, phenylalanine, and tyrosine and may also be affected by spectral contributions from disulfide bonds.
- Near-UV CD spectra of two different proteins will virtually never look similar, even if the two molecules are comparable in size or secondary structure. This is because the numbers of the different aromatic amino acids in proteins normally differ significantly. In consequence, near-UV CD spectra can be considered a fingerprint property.
- The goal of any HOS comparison is the pairwise assessment of samples to determine whether there are statistically significant differences between the HOS of two samples or not.
- In the context of CD spectroscopy, this means that spectral data for any two samples must be obtained at identical conditions and then be boiled down to a single value that informs about the similarity between the samples.
- For a full HOS comparison, separate comparisons based on far- and near-UV CD data are being carried out.
Generally, there are different strategies of data handling and analysis, and approaches to HOS comparisons can differ in, e.g., how much they are affected by noisy data, and whether they require additional data processing, and how they are interpreted. Some approaches include:
- Spectral Difference (SD)
- Weighted Spectral Difference (WSD)
- Correlation coefficients
- Derivative correlation
- Area of Overlap
- Root Mean Square Deviation (RMSD)
- Normalized Root Mean Square Deviation (NRMSD)
HOS comparisons can be helpful throughout the entire development pipeline of a biotherapeutic. This can include:
- Production & Manufacture: Comparison of different batches or lots
- Biotherapeutic Screening: Comparison of different candidates
- Method Validation: Comparison between different days or sites
- Scale-Up and Bridging Studies: Comparison between early stage and late stage
- Biosimilar Development: Comparison of innovator and biosimilar
- Formulation Development: Comparison of different buffers, excipients, pH, etc.
- Forced Degradation Studies: Comparison of different stress conditions
You can find some exemplary use cases of HOS comparisons here.
- High-sensitivity instrumentation: The power of HOS comparison analysis lies in the fact that, as opposed to mere visual data inspection, it allows evaluating the statistical significance of minute structural differences, which reveal themselves in minor spectral differences. In order to be able of identifying these, high-quality data with a good signal-to-noise ratio is necessary, which can be obtained with systems using solid state detectors like large-area avalanche photodiodes (LAAPDs).
- High reproducibility: Any additional manual steps in the experimental setup can introduce uncertainty. When obtaining CD data for HOS comparison analysis, an automated system with liquid handling robotics that cleans the sample cell according to a defined protocol should be used to eliminate any user-dependent variation.
- Multiple technical replicates: Confidence in objective statistical analysis requires a sufficient number of technical replicates for each sample. Whereas n = 3 samples is the absolute minimum required for statistical analysis and should be considered the bare minimum for proof-of-principle HOS comparisons, a number of n = 5 or more replicates is required for routine HOS comparisons. If data reproducibility is low, e.g., because a manual instrument is used for data acquisition, a larger number of replicates is advisable.