Analysis of Insulin Allostery in Solution and Solid State With FTIR

Exploring the Higher Order Structure and Conformational Transitions in Insulin Microcrystalline Biopharmaceuticals by Proton-Detected Solid-State Nuclear Magnetic Resonance at Natural Abundance

The integrity of a higher order structure (HOS) is an essential requirement to ensure the efficacy, stability, and safety of protein therapeutics. Solution-state nuclear magnetic resonance (NMR) occupies a unique niche as one of the most promising methods to access atomic-level structural information on soluble biopharmaceutical formulations. Another major class of drugs is poorly soluble, such as microcrystalline suspensions, which poses significant challenges for the characterization of the active ingredient in its native state. Here, we have demonstrated a solid-state NMR method for HOS characterization of biopharmaceutical suspensions employing a selective excitation scheme under fast magic angle spinning (MAS). The applicability of the method is shown on commercial insulin suspensions at natural isotopic abundance. Selective excitation aided with proton detection and non-uniform sampling (NUS) provides improved sensitivity and resolution. The enhanced resolution enabled us to demonstrate the first experimental evidence of a phenol-escaping pathway in insulin, leading to conformational transitions to different hexameric states. This approach has the potential to serve as a valuable means for meticulously examining microcrystalline biopharmaceutical suspensions, which was previously not attainable in their native formulation states and can be seamlessly extended to other classes of biopharmaceuticals such as mAbs and other microcrystalline proteins.

A new paradigm for antiangiogenic therapy through controlled release of bevacizumab from PLGA nanoparticles

Monoclonal antibodies have deserved a remarkable interest for more than 40 years as a vital tool for the treatment of various diseases. Still, there is a raising interest to develop advanced monoclonal antibody delivery systems able to tailor pharmacokinetics. Bevacizumab is a humanized immunoglobulin IgG1 used in antiangiogenic therapies due to its capacity to inhibit the interaction between vascular endothelial growth factor and its receptor. However, bevacizumab-based antiangiogenic therapy is not always effective due to poor treatment compliance associated to multiples administrations and drug resistance. In this work, we show a promising strategy of encapsulating bevacizumab to protect and deliver it, in a controlled manner, increasing the time between administrations and formulation shelf-life. Nanoencapsulation of bevacizumab represents a significant advance for selective antiangiogenic therapies since extracellular, cell surface and intracellular targets can be reached. The present study shows that bevacizumab-loaded poly (lactic-co-glycolic acid) (PLGA) nanoparticles does not impair its native-like structure after encapsulation and fully retain the bioactivity, making this nanosystem a new paradigm for the improvement of angiogenic therapy.

Effect of the Freezing Step in the Stability and Bioactivity of Protein-Loaded PLGA Nanoparticles Upon Lyophilization

PURPOSE

The freezing step in lyophilization is the most determinant for the quality of biopharmaceutics. Using insulin as model of therapeutic protein, our aim was to evaluate the freezing effect in the stability and bioactivity of insulin-loaded PLGA nanoparticles. The performance of trehalose, sucrose and sorbitol as cryoprotectants was evaluated.

METHODS

Cryoprotectants were co-encapsulated with insulin into PLGA nanoparticles and lyophilized using an optimized cycle with freezing at -80°C, in liquid nitrogen, or ramped cooling at -40°C. Upon lyophilization, the stability of protein structure and in vivo bioactivity were assessed.

RESULTS

Insulin was co-encapsulated with cryoprotectants resulting in particles of 243-394 nm, zeta potential of -32 to -35 mV, and an association efficiency above 90%. The cryoprotectants were crucial to mitigate the freezing stresses and better stabilize the protein. The insulin structure maintenance was evident and close to 90%. Trehalose co-encapsulated insulin-loaded PLGA nanoparticles demonstrated enhanced hypoglycemic effect, comparatively to nanoparticles without cryoprotectant and added with trehalose, due to a superior insulin stabilization and bioactivity.

CONCLUSIONS

The freezing process may be detrimental to the structure of protein loaded into nanoparticles, with negative consequences to bioactivity. The co-encapsulation of cryoprotectants mitigated the freezing stresses with benefits to protein bioactivity.

Multivariate analysis of phenol in freeze-dried and spray-dried insulin formulations by NIR and FTIR

Dehydration is a commonly used method to stabilise protein formulations. Upon dehydration, there is a significant risk the composition of the formulation will change especially if the protein formulation contains volatile compounds. Phenol is often used as excipient in insulin formulations, stabilising the insulin hexamer by changing the secondary structure. We have previously shown that it is possible to maintain this structural change after drying. The aim of this study was to evaluate the residual phenol content in spray-dried and freeze-dried insulin formulations by Fourier transform infrared (FTIR) spectroscopy and near infrared (NIR) spectroscopy using multivariate data analysis. A principal component analysis (PCA) and partial least squares (PLS) projections were used to analyse spectral data. After drying, there was a difference between the two drying methods in the phenol/insulin ratio and the water content of the dried samples. The spray-dried samples contained more water and less phenol compared with the freeze-dried samples. For the FTIR spectra, the best model used one PLS component to describe the phenol/insulin ratio in the powders, and was based on the second derivative pre-treated spectra in the 850-650 cm(-1) region. The best PLS model based on the NIR spectra utilised three PLS components to describe the phenol/insulin ratio and was based on the standard normal variate transformed spectra in the 6,200-5,800 cm(-1) region. The root mean square error of cross validation was 0.69% and 0.60% (w/w) for the models based on the FTIR and NIR spectra, respectively. In general, both methods were suitable for phenol quantification in dried phenol/insulin samples.