A novel rapid analysis using mass spectrometry to evaluate downstream refolding of recombinant human insulin-like growth factor-1 (mecasermin).

RATIONALE: Mecasermin is used to treat elevated blood sugar as well as growth-hormone-resistant Laron-type dwarfism. Mecasermin isolated from inclusion bodies in extracts of E. coli must be refolded to acquire sufficient activity. However, there is no rapid analytical method for monitoring refolding during the purification process. METHODS: We prepared mecasermin drug product, in-process samples during the oxidation of mecasermin, forced-reduced mecasermin, and aerially oxidized mecasermin after forced reduction. Desalted mecasermin samples were analyzed using MALDI-ISD. The peak intensity ratio of product to precursor ion was determined. The charge-state distribution (CSD) of mecasermin ions was evaluated using ESI-MS coupled with SEC-mode HPLC. The drift time and collision cross-sectional area (CCS) of mecasermin ions were evaluated using ESI-IMS-MS coupled with SEC-mode HPLC. RESULTS: MALDI-ISD data, CSD values determined using ESI-MS, and the CCS acquired using ESI-IMS-MS revealed the relationship between the folded and unfolded proteoforms of forced-reduced mecasermin and aerially oxidized mecasermin with the free-SH:protein ratio of mecasermin drug product. The CCS area, which is determined using ESI-IMS-MS, provided proteoform information through rapid monitoring (<2 min) of in-process samples during the manufacture of mecasermin. CONCLUSIONS: ESI-IMS-MS coupled with SEC-mode HPLC is a rapid and robust method for analyzing the free-SH:protein ratio of mecasermin that allows proteoform changes to be evaluated and monitored during the oxidation of mecasermin. ESI-IMS-MS is applicable as a process analytical technology tool for identifying the "critical quality attributes" and implementing "quality by design" for manufacturing mecasermin.

Characterization of cysteine-linked conjugation profiles of immunoglobulin G1 and immunoglobulin G2 antibody-drug conjugates.

Two US FDA-approved antibody-drug conjugates (ADCs; Kadcyla(®) and Adcetris(®) ) have accelerated clinical interest in the potential of targeted cancer therapeutics as the next generation of oncology drugs that are designed to increase efficacy while reducing overall toxicity. Thiol conjugates are produced by partial reduction of the interchain disulfides, followed by conjugation with a drug-linker, resulting in a heterogeneous mix of molecules that differ with respect to the site of conjugation and the number of drugs per antibody. ADCs that have been characterized in this class have an immunoglobulin G1 (IgG1) framework and there is little information available on IgG2 ADCs. As IgG1s and IgG2s differ in the number of disulfides and molecular conformations, each subclass could lead to unique combinations of possible conjugation sites. We conducted in-depth characterization of two ADCs, an IgG1 and an IgG2 conjugated to monomethyl auristatin E. The results demonstrate that the IgG1 monoclonal antibodies favor conjugation to the cysteines between the light and heavy chains, whereas IgG2s demonstrate preference for the hinge region cysteines. The drug-loading distribution and conjugation sites of ADCs have been reported to influence pharmacokinetics, toxicity, and clearance. Therefore, an understanding of the conjugation profiles is important for the selection and engineering of ADCs.

A comparative study of monosaccharide composition analysis as a carbohydrate test for biopharmaceuticals.

The various monosaccharide composition analysis methods were evaluated as monosaccharide test for glycoprotein-based pharmaceuticals. Neutral and amino sugars were released by hydrolysis with 4-7N trifluoroacetic acid. The monosaccharides were N-acetylated if necessary, and analyzed by high-performance liquid chromatography (HPLC) with fluorometric or UV detection after derivatization with 2-aminopyridine, ethyl 4-aminobenzoate, 2-aminobenzoic acid or 1-phenyl-3-methyl-5-pyrazolone, or high pH anion exchange chromatography with pulsed amperometric detection (HPAEC-PAD). Sialic acids were released by mild acid hydrolysis or sialidase digestion, and analyzed by HPLC with fluorometric detection after derivatization with 1,2-diamino-4,5-methylenedioxybenzene, or HPAEC-PAD. These methods were verified for resolution, linearity, repeatability, and accuracy using a monosaccharide standard solution, a mixture of epoetin alfa and beta, and alteplase as models. It was confirmed that those methods were useful for ensuring the consistency of glycosylation. It is considered essential that the analytical conditions including desalting, selection of internal standards, release of monosaccharides, and gradient time course should be determined carefully to eliminate interference of sample matrix. Various HPLC-based monosaccharide analysis methods were evaluated as a carbohydrate test for glycoprotein pharmaceuticals by an inter-laboratory study.

Quantitative evaluation of protein conformation in pharmaceuticals using cross-linking reactions coupled with LC-MS/MS analysis.

The need for a simple and high-throughput method for identifying the tertiary structure of protein pharmaceuticals has increased. In this study, a simple method for mapping the protein fold is proposed for use as a complementary quality test. This method is based on cross-linking a protein using a [bis(sulfosuccinimidyl)suberate (BS(3))], followed by peptide mapping by LC-MS. Consensus interferon (CIFN) was used as the model protein. The tryptic map obtained via liquid chromatography tandem mass spectroscopy (LC-MS/MS) and the mass mapping obtained via matrix-assisted laser desorption/ionization time-of-flight mass spectroscopy were used to identify cross-linked peptides. While LC-MS/MS analyses found that BS(3) formed cross-links in the loop region of the protein, which was regarded as the biologically active site, sodium dodecyl-sulfate polyacrylamide gel electrophoresis demonstrated that cross-linking occurred within a protein molecule, but not between protein molecules. The occurrence of cross-links at the active site depends greatly on the conformation of the protein, which is determined by the denaturing conditions. Quantitative evaluation of the tertiary structure of CIFN was thus possible by monitoring the amounts of cross-linked peptides generated. Assuming that background information is available at the development stage, this method may be applicable to process development as a complementary test for quality control.

Enhanced pharmacological activity of recombinant human interleukin-11 (rhIL11) by chemical modification with polyethylene glycol.

In order to improve the pharmacological efficacy of recombinant human interleukin-11 (rhIL11) and to reduce the frequency of administration, we examined the feasibility of chemical modification of rhIL11 by polyethylene glycol. The rhIL11 was chemically modified by using branched type (PEG2), or linear type (PEG) polyethylene glycol-N-hydroxysuccinimide with various molecular weights. Plasma profiles of immunoreactive rhIL11 after i.v. injection of the 20 kDa PEG2 conjugated rhIL11 (PEG2 (20 K)-rhIL11) were determined by ELISA. Peripheral platelet counts after the administration of the various conjugates were measured. Pharmacokinetic analysis revealed that the mean residence time of PEG2 (20 K)-rhIL11 after i.v. injection extensively increased by a factor of ca 60 compared with the native rhIL11. Maximum peripheral platelet increase of 67% for PEG2 (20 K)-rhIL11 and that of 50% for PEG (20 K)-rhIL11 over the control was observed whereas no significant change was associated with the bolus i.v. injection of native rhIL11. On the other hand, the remaining biological activity of these PEGylated-rhIL11s was 14-16% of native rhIL11, suggesting that retention of rhIL11 in plasma is much effective in order to potentiate the pharmacological efficacy of the cytokine. Chemical modification of rhIL11 by PEG is a promising approach for improving the clinical efficacy of rhIL11 by prolonged retention in plasma.

Detection of lot-to-lot variations in the amorphous microstructure of lyophilized protein formulations.

Reconstitution of lyophilized protein formulations sometimes results in a cloudy solution, depending on the compositions and manufacturing conditions, which causes quality concerns. In this study, the lyophilized protein formulation of recombinant human Interleukin-11 (rhIL-11) was investigated using different lots with varying dissolution behaviors upon reconstitution due to differing processing conditions. In an attempt to distinguish the solid structures in the different lots, relatively new techniques such as inverse gas chromatography (IGC) and thermally stimulated depolarized current (TSDC) as well as powder X-ray diffraction (PXRD) and differential scanning calorimetry (DSC) were adopted for analysis. PXRD, DSC, and IGC all failed to distinguish between the solid structures, but TSDC was able to discern the differences. Interestingly, TSDC suggested that the variations in dissolution behavior were attributable to the differences in molecular mobility and the micro heterogeneity of amorphous components in the solid structures. Since even the cloudiest reconstituted solutions became transparent in several minutes, it was likely that the differences in the solid structures of the different lots of lyophilized cakes were slight. This study demonstrates the usefulness of TSDC in the analysis of lot-to-lot variations in amorphous pharmaceuticals.