Analytical Techniques for Structural Characterization of Proteins in Solid Pharmaceutical Forms: An Overview

Abstract

Therapeutic proteins as biopharmaceuticals have emerged as a very important class of drugs for the treatment of many diseases. However, they are less stable compared to conventional pharmaceuticals. Their long-term stability in solid forms, which is critical for product performance, depends heavily on the retention of the native protein structure during the lyophilization (freeze-drying) process and, thereafter, in the solid state. Indeed, the biological function of proteins is directly related to the tertiary and secondary structure. Besides physical stability and biological activity, conformational stability (three-dimensional structure) is another important aspect when dealing with protein pharmaceuticals. Moreover, denaturation as loss of higher order structure is often a precursor to aggregation or chemical instability. Careful study of the physical and chemical properties of proteins in the dried state is therefore critical during biopharmaceutical drug development to deliver a final drug product with built-in quality that is safe, high-quality, efficient, and affordable for patients. This review provides an overview of common analytical techniques suitable for characterizing pharmaceutical protein powders, providing structural, and conformational information, as well as insights into dynamics. Such information can be very useful in formulation development, where selecting the best formulation for the drug can be quite a challenge.

Keywords: analytical tools; antibody; excipients; formulation development; lyophilization; protein characterization; protein structure; safe drug; solid pharmaceuticals; stable drug product.

Figure 1

The most common analytical techniques for the structural characterization of proteins in solid pharmaceutical forms are presented with corresponding type of measurements. Changes in secondary/tertiary structure and conformation can be studied on a global and local scale. Protein dynamics can also be traced using some of the above methods. FTIR—fourier transform infrared; NIR—near-infrared; CD—circular dichroism; ss—solid-state, HDX-MS—hydrogen-deuterium exchange mass spectrometry; DSC—differential scanning calorimetry; NMR—nuclear magnetic resonance; DRS—dielectric relaxation spectroscopy.

Figure 2

Second derivative FTIR (fourier transform infrared) spectra of lyophilized lactate dehydrogenase (LDH). Different types of secondary structures are shown, as well as different % of remaining LDH activity after reconstitution. Reproduced with permission from [24], Elsevier, 2013.

Figure 3

Second derivative FTIR spectra of immunoglobulin G1 (IgG1) in solid form in two different formulations with sucrose—sucrose:IgG1 = 4:1 (A) and sucrose:IgG1 = 1:4 (B). Formulations with higher sucrose content show better stability and thus preservation of the native secondary structure. Reproduced with permission from [33], Wiley, 2007.

Figure 4

Second derivative FTIR spectra of lyophilized antibody formulations with different excipients used. The ratio between the sugars and the antibody is 1:1. Native structure spectra were recorded with pure antibody in 5 mM phosphate buffer at pH 7. Reproduced with permission from [41], Elsevier, 2005.

Figure 5

NIR (near-infrared) spectra as second derivatives of structural studies on lyophilized cytochrome c. The influence of sucrose content on the preservation of secondary structure is shown in the left spectra, while the influence of temperature on protein stability can be seen in the right ones. Reproduced with permission from [51], Wiley, 2005.

Figure 6

Raman spectra of native-like and denatured (non-native) freeze-dried LDH. Sample concentration 9 mg/mL, 6.5 mg of material used. Reproduced with permission from [24], Elsevier, 2013.

Figure 7

Second derivative absorbance spectra for UV–Vis analysis of IgG at three different pH values without KCl. The largest shifts were observed for the two peaks at approximately 284 and 292 nm. Reproduced with permission from [76], Elsevier, 2008.

Figure 8

Second derivative UV–Vis spectra for IgG. The effect of different pH is presented for five characteristic peaks. Reproduced with permission from [76], Elsevier, 2008.

Figure 9

Solid-state (ss) fluorescence study of temperature stress effect on lyophilized protein stability. Subfigure (A) presents the fluorescence spectra at various time points during incubation at 60 °C. The decrease in intensity can be clearly observed. Further, the subfigure (B) presents the absolute fluorescence intensity versus time Reproduced with permission from [85], Elsevier, 2008.

Figure 10

Circular dichroism (CD) spectra for reconstituted lyophilized protein incubated at two different temperatures. Changes in tertiary structure are seen in the near-UV CD spectra on the left, while changes in secondary structure are shown in the far-UV CD spectra on the right. Red spectra—lyophilized protein (control), green spectra—incubation at 37 °C, pink spectra—incubation at 60 °C. Reproduced with permission from [85], Elsevier, 2008.

Figure 11

ssNMR study of several lyophilized protein formulations containing different amounts of lysozyme and trehalose. On the left average T_1ρ is plotted against relative humidity (% RH), whereas on the right average T₁ is plotted against % RH. —trehalose; —80% trehalose, 20% lysozyme; □—20% trehalose, 80% lysozyme; —lysozyme. Reproduced with permission from [116], Elsevier, 2002.

Figure 12

Typical dielectric relaxation spectroscopy (DRS) spectra of lyophilized horse myoglobin powder. Real part (permittivity) and imaginary part (dielectric loss) are presented. Reproduced with permission from [140], ACS, 2009.

Figure 13

XRD analysis of lyophilized human serum albumin (HSA) formulations with different amounts of mannitol (M) is depicted in figure (A). Reproduced with permission from [153], Elsevier, 2006. In figure (B) the three forms of mannitol (a—alpha, b—beta and c—delta) are depicted in structural arrangements. Reproduced with permission from [154], Elsevier, 2020.

Figure 14

Time dependence of deuterium exchange for lyophilized monoclonal antibody in several different formulations (exposure to D₂O vapor at 11% RH and 22 °C). Formulations: M3, mannitol 3:1; M1, mannitol 1:1; S6, sucrose 6:1; S3, sucrose 3:1; S2.7, sucrose 2.7:1; S1, sucrose 1:1; T3, trehalose 3:1; T1, trehalose 1:1; H3, histidine 3:1; H2, histidine 2:1; NE (no-excipient), without excipient. Ratios represent excipient/antibody w/w ratio. Reproduced with permission from [137], ACS, 2018.

Figure 15

Size-exclusion chromatography (SEC) analyses of control monoclonal antibodies mAb1 and mAb2, antibody mAb1 after thermal stability studies for 3 months and an in-process antibody mAb2. Chromatograms (A,B) were obtained using UHPLC (ultra high-performance liquid chromatography), whereas chromatograms (C,D) were obtained by HPLC. Reproduced with permission from [185], Elsevier, 2018.

Figure 16

SEC studies on stability of samples containing myoglobin (A), BSA (bovine serum albumine) (B), β-lactoglobulin (C), or lysozyme (D). Formulations are lyophilized (Lyo) or spray-dried (SD) and have three sugar excipients (sucrose—Suc, trehalose—Tre, mannitol—Mann). Reproduced with permission from [175], Elsevier, 2019.

Figure 17

Dynamic light scattering (DLS) analyses for a desorbed quadrivalent human papillomavirus virus-like particles (HPV VLP) vaccine at different pH values: (A) at pH 6.2, (B) at pH 4.0, (C) at pH 8.0. Data are shown as intensity size distribution. Aggregate formation by forced degradation conditions can be observed in Figures (B,C), pH 4.0 and 8.0, respectively. Reproduced with permission from [187], Elsevier, 2017.