Berit Olsen Krogh

Optimization of ordered plasmid assembly by gap repair in Saccharomyces cerevisiae

Combinatorial genetic libraries are powerful tools for diversifying and optimizing biomolecules. The process of library assembly is a major limiting factor for library complexity and quality. Gap repair by homologous recombination in Saccharomyces cerevisiae can facilitate in vivo assembly of DNA fragments sharing short patches of sequence homology, thereby supporting generation of high‐complexity libraries without compromising fidelity. In this study, we have optimized the ordered assembly of three DNA fragments into a gapped vector by in vivo homologous recombination. Assembly is achieved by co‐transformation of the DNA fragments and the gapped vector, using a modified lithium acetate protocol. The optimal gap‐repair efficiency is found at a 1:80 molar ratio of gapped vector to each of the three fragments. We measured gap‐repair efficiency in different genetic backgrounds and observed increased efficiency in mutants carrying a deletion of the SGS1 helicase‐encoding gene. Using our experimental conditions, a gap‐repair efficiency of > 106 plasmid‐harbouring colonies/µg gapped vector DNA is obtained in a single transformation, with a recombination fidelity > 90%. Copyright

Investigation of protein selectivity in multimodal chromatography using in silico designed Fab fragment variants

In this study, a unique set of antibody Fab fragments was designed in silico and produced to examine the relationship between protein surface properties and selectivity in multimodal chromatographic systems. We hypothesized that multimodal ligands containing both hydrophobic and charged moieties would interact strongly with protein surface regions where charged groups and hydrophobic patches were in close spatial proximity. Protein surface property characterization tools were employed to identify the potential multimodal ligand binding regions on the Fab fragment of a humanized antibody and to evaluate the impact of mutations on surface charge and hydrophobicity. Twenty Fab variants were generated by site‐directed mutagenesis, recombinant expression, and affinity purification. Column gradient experiments were carried out with the Fab variants in multimodal, cation‐exchange, and hydrophobic interaction chromatographic systems. The results clearly indicated that selectivity in the multimodal system was different from the other chromatographic modes examined. Column retention data for the reduced charge Fab variants identified a binding site comprising light chain CDR1 as the main electrostatic interaction site for the multimodal and cation‐exchange ligands. Furthermore, the multimodal ligand binding was enhanced by additional hydrophobic contributions as evident from the results obtained with hydrophobic Fab variants. The use of in silico protein surface property analyses combined with molecular biology techniques, protein expression, and chromatographic evaluations represents a previously undescribed and powerful approach for investigating multimodal selectivity with complex biomolecules. Biotechnol. Bioeng. 2015;112: 2305–2315.

Evaluation of selectivity in homologous multimodal chromatographic systems using in silico designed antibody fragment libraries.

This study describes the in silico design, surface property analyses, production and chromatographic evaluations of a diverse set of antibody Fab fragment variants. Based on previous findings, we hypothesized that the complementarity-determining regions (CDRs) constitute important binding sites for multimodal chromatographic ligands. Given that antibodies are highly diversified molecules and in particular the CDRs, we set out to examine the generality of this result. For this purpose, four different Fab fragments with different CDRs and/or framework regions of the variable domains were identified and related variants were designed in silico. The four Fab variant libraries were subsequently generated by site-directed mutagenesis and produced by recombinant expression and affinity purification to enable examination of their chromatographic retention behavior. The effects of geometric re-arrangement of the functional moieties on the multimodal resin ligands were also investigated with respect to Fab variant retention profiles by comparing two commercially available multimodal cation-exchange ligands, Capto MMC and Nuvia cPrime, and two novel multimodal ligand prototypes. Interestingly, the chromatographic data demonstrated distinct selectivity trends between the four Fab variant libraries. For three of the Fab libraries, the CDR regions appeared as major binding sites for all multimodal ligands. In contrast, the fourth Fab library displayed a distinctly different chromatographic behavior, where Nuvia cPrime and related multimodal ligand prototypes provided markedly improved selectivity over Capto MMC. Clearly, the results illustrate that the discriminating power of multimodal ligands differs between different Fab fragments. The results are promising indications that multimodal chromatography using the appropriate multimodal ligands can be employed in downstream bioprocessing for challenging selective separation of product related variants.

QSAR models for prediction of chromatographic behavior of homologous Fab variants.

While quantitative structure activity relationship (QSAR) models have been employed successfully for the prediction of small model protein chromatographic behavior, there have been few reports to date on the use of this methodology for larger, more complex proteins. Recently our group generated focused libraries of antibody Fab fragment variants with different combinations of surface hydrophobicities and electrostatic potentials, and demonstrated that the unique selectivities of multimodal resins can be exploited to separate these Fab variants. In this work, results from linear salt gradient experiments with these Fabs were employed to develop QSAR models for six chromatographic systems, including multimodal (Capto MMC, Nuvia cPrime, and two novel ligand prototypes), hydrophobic interaction chromatography (HIC; Capto Phenyl), and cation exchange (CEX; CM Sepharose FF) resins. The models utilized newly developed “local descriptors” to quantify changes around point mutations in the Fab libraries as well as novel cluster descriptors recently introduced by our group. Subsequent rounds of feature selection and linearized machine learning algorithms were used to generate robust, well‐validated models with high training set correlations (R2 > 0.70) that were well suited for predicting elution salt concentrations in the various systems. The developed models then were used to predict the retention of a deamidated Fab and isotype variants, with varying success. The results represent the first successful utilization of QSAR for the prediction of chromatographic behavior of complex proteins such as Fab fragments in multimodal chromatographic systems. The framework presented here can be employed to facilitate process development for the purification of biological products from product‐related impurities by in silico screening of resin alternatives. Biotechnol. Bioeng. 2017;114: 1231–1240.

Purification and characterization of bioactive his6-tagged recombinant human tissue inhibitor of metalloproteinases-1 (TIMP-1) protein expressed at high yields in mammalian cells

Tissue inhibitor of metalloproteinases-1 (TIMP-1) is an endogenous inhibitor of matrix metalloproteinases (MMPs) with reported tumor promoting, as well as inhibitory, effects. These paradoxical properties are presumably mediated by different biological functions, MMP-dependent as well as -independent, and probably related to TIMP-1 levels of protein expression, post-translational modifications, and cellular localization. TIMP-1 is an N-glycosylated protein that folds into two functional domains, a C- and an N-terminal domain, with six disulfide bonds. Furthermore, TIMP-1 is processed in the N-terminal sequence. These three biochemical properties make TIMP-1 difficult to produce in conventional bacterial, insect, or yeast expression systems. We describe here a HEK293 cell-based strategy for production and purification of secreted and N-glycosylated recombinant his6-tagged human TIMP-1 (his6-rTIMP-1), which resulted in large amounts of highly purified and bioactive protein. Matrix-assisted laser desorption ionization mass spectrometry confirmed the N- and C-termini of his6-rTIMP-1, and N-glycosylation profiling showed a match to the N-glycosylation of human plasma TIMP-1. The his6-rTIMP-1 was bioactive as shown by its proper inhibitory effect on MMP-2 activity, and its stimulatory effect on cell growth when added to the growth medium of four different breast cancer cell lines. This study provides an easy set-up for large scale production and purification of bioactive, tagged recombinant human TIMP-1, which structurally and functionally is similar to endogenous human TIMP-1, while using an expression system that is adaptable to most biochemical and biomedical laboratories including those that do not perform protein purifications routinely.

Sensitive and specific in situ hybridization for early drug discovery.

High-throughput analyses of gene expression such as microarrays and RNA-sequencing are widely used in early drug discovery to identify disease-associated genes. To further characterize the expression of selected genes, in situ hybridization (ISH) using RNA probes (riboprobes) is a powerful tool to localize mRNA expression at the cellular level in normal and diseased tissues, especially for novel drug targets, where research tools like specific antibodies are often lacking.We describe a sensitive ISH protocol using radiolabelled riboprobes suitable for both paraffin-embedded and cryo-preserved tissue. The riboprobes are generated by in vitro transcription using PCR products as templates, which is less time consuming compared to traditional transcription from linearized plasmids, and offers a relatively simple way to generate several probes per gene, e.g., for splice variant analyses. To ensure reliable ISH results, we have incorporated a number of specificity controls in our standard experimental setup. We design antisense probes to cover two non-overlapping parts of the gene of interest, and use the corresponding sense probes as controls for unspecific binding. Probes are furthermore tested on sections of paraffin-embedded or cryo-preserved positive and negative control cells with known gene expression. Our protocol thus provides a method for sensitive and specific ISH, which is suitable for target validation and characterization in early drug discovery.