Characterization of recombinant human IL-15 deamidation and its practical elimination through substitution of asparagine 77.

PURPOSE: The use of recombinant human interleukin (rhIL)-15 as a potential therapeutic immune modulator and anticancer agent requires pure, stable preparations. However, purified rhIL-15 preparations readily accumulated heterogeneities. We sought to improve rhIL-15 stability through process, formulation, and targeted amino acid changes. METHODS: The solution state of rhIL-15 versus buffer composition and temperature was studied using SEC and IEX methods. rhIL-15 deamidation was confirmed using RP-HPLC/ESI-MS, enzymatic labeling, and peptide mapping. Deamidation kinetics were measured versus buffer composition and pH using RP-HPLC. Deamidation-resistant rhIL-15 variants (N77A, N77S, N77Q, G78A, and [N71S/N72A/N77A]) were produced in E. coli, then assayed for T-cell culture expansion potency and deamidation resistance. RESULTS: Adding 20% ethanol to buffers or heating at ≥32°C dispersed rhIL-15 transient pairs, improving purification efficiencies. Asparagine 77 deamidated rapidly at pH 7.4 with activation energy of 22.9 kcal per mol. Deamidation in citrate buffer was 17-fold slower at pH 5.9 than at pH 7.4. Amino acid substitutions at N77 or G78 slowed deamidation ≥23-fold. rhIL-15 variants N77A and (N71S/N72A/N77A) were active in a CTLL-2 proliferation assay equivalent to unsubstituted rhIL-15. CONCLUSIONS: The N77A and (N71S/N72A/N77A) rhIL-15 variants are resistant to deamidation and remain potent, thus providing enhanced drug substances for clinical evaluation.

Preclinical manufacture of anti-HER2 liposome-inserting, scFv-PEG-lipid conjugate. 2. Conjugate micelle identity, purity, stability, and potency analysis.

Analytical methods optimized for micellar F5cys-MP-PEG(2000)-DPSE protein-lipopolymer conjugate are presented. The apparent micelle molecular weight, determined by size exclusion chromatography, ranged from 330 to 960 kDa. The F5cys antibody and conjugate melting points, determined by differential scanning calorimetry, were near 82 degrees C. Traditional methods for characterizing monodisperse protein species were inapplicable to conjugate analysis. The isoelectric point of F5cys (9.2) and the conjugate (8.9) were determined by capillary isoelectric focusing (cIEF) after addition of the zwitterionic detergent CHAPS to the buffer. Conjugate incubation with phospholipase B selectively removed DSPE lipid groups and dispersed the conjugate prior to separation by chromatographic methods. Alternatively, adding 2-propanol (29.4 vol %) and n-butanol (4.5 vol %) to buffers for salt-gradient cation exchange chromatography provided gentler, nonenzymatic dispersion, resulting in well-resolved peaks. This method was used to assess stability, identify contaminants, establish lot-to-lot comparability, and determine the average chromatographic purity (93%) for conjugate lots, described previously. The F5cys amino acid content was confirmed after conjugation. The expected conjugate avidity for immobilized HER-2/neu was measured by bimolecular interaction analysis (BIAcore). Mock therapeutic assemblies were made by conjugate insertion into preformed doxorubicin-encapsulating liposomes for antibody-directed uptake of doxorubicin by HER2-overexpressing cancer cells in vitro. Together these developed assays established that the manufacturing method as described in the first part of this study consistently produced F5cys-MP-PEG(2000)-DSPE having sufficient purity, stability, and functionality for use in preclinical toxicology investigations.

Preclinical manufacture of an anti-HER2 scFv-PEG-DSPE, liposome-inserting conjugate. 1. Gram-scale production and purification.

A GMP-compliant process is described for producing F5cys-PEG-lipid conjugate. This material fuses with preformed, drug-loaded liposomes, to form "immunoliposomes" that bind to HER2/neu overexpressing carcinomas, stimulates drug internalization, and ideally improves the encapsulated drug's therapeutic index. The soluble, single-chain, variable region antibody fragment, designated F5cys, was produced in E. coli strain RV308 using high-density cultures. Affinity adsorption onto horizontally tumbled Streamline rProtein-A resin robustly recovered F5cys from high-pressure-disrupted, whole-cell homogenates. Two product-related impurity classes were identified: F5cys with mid-sequence discontinuities and F5cys with remnants of a pelB leader peptide. Low-pressure cation exchange chromatography, conducted at elevated pH under reducing conditions, enriched target F5cys relative to these impurities and prepared a C-terminal cysteine for conjugation. Site-directed conjugation, conducted at pH 5.9 +/- 0.1 with reaction monitoring and cysteine quenching, yielded F5cys-MP-PEG(2000)-DSPE. Low-pressure size exclusion chromatography separated spontaneously formed, high-molecular-weight conjugate micelles from low-molecular-weight impurities. When formulated at 1-2 mg/mL in 10 mM trisodium citrate, 10% sucrose (w/v), at pH 6.4 (HCl), the conjugate was stable when stored below -70 degrees C. Six scale-up lots were compared. The largest 40-L culture produced enough F5cys to manufacture 2,085 mg of conjugate, enough to support planned preclinical and future clinical trials. The conjugate was 93% pure, as measured by polyacrylamide gel electrophoresis. Impurities were primarily identified as product-related. Residual endotoxin, rProtein A, and genomic DNA, were at acceptable levels. This study successfully addressed a necessary step in the scale-up of immunoliposome-encapsulated therapeutics.

Purification and protective efficacy of monomeric and modified Yersinia pestis capsular F1-V antigen fusion proteins for vaccination against plague.

The F1-V vaccine antigen, protective against Yersinia pestis, exhibits a strong tendency to multimerize that affects larger-scale manufacture and characterization. In this work, the sole F1-V cysteine was replaced with serine by site-directed mutagenesis for characterization of F1-V non-covalent multimer interactions and protective potency without participation by disulfide-linkages. F1-V and F1-V(C424S) proteins were overexpressed in Escherichia coli, recovered using mechanical lysis/pH-modulation and purified from urea-solubilized soft inclusion bodies, using successive ion-exchange, ceramic hydroxyapatite, and size-exclusion chromatography. This purification method resulted in up to 2mg/g of cell paste of 95% pure, mono-disperse protein having < or =0.5 endotoxin units per mg by a kinetic chromogenic limulus amoebocyte lysate reactivity assay. Both F1-V and F1-V(C424S) were monomeric at pH 10.0 and progressively self-associated as pH conditions decreased to pH 6.0. Solution additives were screened for their ability to inhibit F1-V self-association at pH 6.5. An L-arginine buffer provided the greatest stabilizing effect. Conversion to >500-kDa multimers occurred between pH 6.0 and 5.0. Conditions for efficient F1-V adsorption to the cGMP-compatible alhydrogel adjuvant were optimized. Side-by-side evaluation for protective potency against subcutaneous plague infection in mice was conducted for F1-V(C424S) monomer; cysteine-capped F1-V monomer; cysteine-capped F1-V multimer; and a F1-V standard reported previously. After a two-dose vaccination with 2 x 20 microg of F1-V, respectively, 100%, 80%, 80%, and 70% of injected mice survived a subcutaneous lethal plague challenge with 10(8) LD(50)Y. pestis CO92. Thus, vaccination with F1-V monomer and multimeric forms resulted in significant, and essentially equivalent, protection.