A data science roadmap for open science organizations engaged in early-stage drug discovery.

The Structural Genomics Consortium is an international open science research organization with a focus on accelerating early-stage drug discovery, namely hit discovery and optimization. We, as many others, believe that artificial intelligence (AI) is poised to be a main accelerator in the field. The question is then how to best benefit from recent advances in AI and how to generate, format and disseminate data to enable future breakthroughs in AI-guided drug discovery. We present here the recommendations of a working group composed of experts from both the public and private sectors. Robust data management requires precise ontologies and standardized vocabulary while a centralized database architecture across laboratories facilitates data integration into high-value datasets. Lab automation and opening electronic lab notebooks to data mining push the boundaries of data sharing and data modeling. Important considerations for building robust machine-learning models include transparent and reproducible data processing, choosing the most relevant data representation, defining the right training and test sets, and estimating prediction uncertainty. Beyond data-sharing, cloud-based computing can be harnessed to build and disseminate machine-learning models. Important vectors of acceleration for hit and chemical probe discovery will be (1) the real-time integration of experimental data generation and modeling workflows within design-make-test-analyze (DMTA) cycles openly, and at scale and (2) the adoption of a mindset where data scientists and experimentalists work as a unified team, and where data science is incorporated into the experimental design.

Choice of fusion proteins, expression host, and analytics solves difficult-to-produce protein challenges in discovery research.

High quality biological reagents are a prerequisite for pharmacological research. Herein a protein production screening approach, including quality assessment methods, for protein-based discovery research is presented. Trends from 2895 expression constructs representing 253 proteins screened in mammalian and bacterial hosts-91% of which are successfully expressed and purified-are discussed. Mammalian expression combined with the use of solubility-promoting fusion proteins is deemed suitable for most targets. Furthermore, cases utilizing stable cell line generation and choice of fusion protein for higher yield and quality of difficult-to-produce proteins (Leucine-rich repeat-containing G-protein coupled receptor 4 (LGR4) and Neurturin) are presented and discussed. In the case of Neurturin, choice of fusion protein impacted the target binding 80-fold. These results highlight the need for exploration of construct designs and careful Quality Control (QC) of difficult-to-produce protein reagents.

Engineering Xaa-Pro dipeptidyl aminopeptidase for specific cleavage of glucagon and glucagon-like peptide 1 from fusion proteins.

N-terminal extensions ("tags") have proven valuable for producing peptides using high throughput recombinant expression technologies. However, the applicability is hampered by the limited options for specific and efficient proteases to release the fully native sequence without additional amino acids in the N-terminal. Here we describe the Escherichia coli (E. coli) expression, purification and characterization of engineered variants of Xaa-Pro dipeptidyl aminopeptidase (Xaa-Pro-DAP) derived from Lactococcus lactis for cleavage of Gly-Pro dipeptide extension in the N-terminal of glucagon and glucagon-like peptide 1 (GLP-1(7-37)). By single amino acid substitution in the Xaa-Pro-DAP protease, significantly higher product yields were achieved. The combination of HRV14 3C protease and engineered Xaa-Pro-DAP is suggested for obtaining native N-terminal of peptides.

The EndoC-ßH1 cell line is a valid model of human beta cells and applicable for screenings to identify novel drug target candidates.

OBJECTIVE: To characterize the EndoC-ßH1 cell line as a model for human beta cells and evaluate its beta cell functionality, focusing on insulin secretion, proliferation, apoptosis and ER stress, with the objective to assess its potential as a screening platform for identification of novel anti-diabetic drug candidates. METHODS: EndoC-ßH1 was transplanted into mice for validation of in vivo functionality. Insulin secretion was evaluated in cells cultured as monolayer and as pseudoislets, as well as in diabetic mice. Cytokine induced apoptosis, glucolipotoxicity, and ER stress responses were assessed. Beta cell relevant mRNA and protein expression were investigated by qPCR and antibody staining. Hundreds of proteins or peptides were tested for their effect on insulin secretion and proliferation. RESULTS: Transplantation of EndoC-ßH1 cells restored normoglycemia in streptozotocin induced diabetic mice. Both in vitro and in vivo, we observed a clear insulin response to glucose, and, in vitro, we found a significant increase in insulin secretion from EndoC-ßH1 pseudoislets compared to monolayer cultures for both glucose and incretins. Apoptosis and ER stress were inducible in the cells and caspase 3/7 activity was elevated in response to cytokines, but not affected by the saturated fatty acid palmitate. By screening of various proteins and peptides, we found Bombesin (BB) receptor agonists and Pituitary Adenylate Cyclase-Activating Polypeptides (PACAP) to significantly induce insulin secretion and the proteins SerpinA6, STC1, and APOH to significantly stimulate proliferation. ER stress was readily induced by Tunicamycin and resulted in a reduction of insulin mRNA. Somatostatin (SST) was found to be expressed by 1% of the cells and manipulation of the SST receptors was found to significantly affect insulin secretion. CONCLUSIONS: Overall, the EndoC-ßH1 cells strongly resemble human islet beta cells in terms of glucose and incretin stimulated insulin secretion capabilities. The cell line has an active cytokine induced caspase 3/7 apoptotic pathway and is responsive to ER stress initiation factors. The cells' ability to proliferate can be further increased by already known compounds as well as by novel peptides and proteins. Based on its robust performance during the functionality assessment assays, the EndoC-ßH1 cell line was successfully used as a screening platform for identification of novel anti-diabetic drug candidates.

Large-Scale Biophysical Evaluation of Protein PEGylation Effects: In Vitro Properties of 61 Protein Entities.

PEGylation is the most widely used method to chemically modify protein biopharmaceuticals, but surprisingly limited public data is available on the biophysical effects of protein PEGylation. Here we report the first large-scale study, with site-specific mono-PEGylation of 15 different proteins and characterization of 61 entities in total using a common set of analytical methods. Predictions of molecular size were typically accurate in comparison with actual size determined by size-exclusion chromatography (SEC) or dynamic light scattering (DLS). In contrast, there was no universal trend regarding the effect of PEGylation on the thermal stability of a protein based on data generated by circular dichroism (CD), differential scanning calorimetry (DSC), or differential scanning fluorimetry (DSF). In addition, DSF was validated as a fast and inexpensive screening method for thermal unfolding studies of PEGylated proteins. Multivariate data analysis revealed clear trends in biophysical properties upon PEGylation for a subset of proteins, although no universal trends were found. Taken together, these findings are important in the consideration of biophysical methods and evaluation of second-generation biopharmaceutical drug candidates.

Leukemia inhibitory factor increases glucose uptake in mouse skeletal muscle.

Members of the IL-6 family, IL-6 and ciliary neurotrophic factor (CNTF), have been shown to increase glucose uptake and fatty acid oxidation in skeletal muscle. However, the metabolic effects of another family member, leukemia inhibitory factor (LIF), are not well characterized. Effects of LIF on skeletal muscle glucose uptake and palmitate oxidation and signaling were investigated in ex vivo incubated mouse soleus and EDL muscles from muscle-specific AMPKα2 kinase-dead, muscle-specific SOCS3 knockout, and lean and high-fat-fed mice. Inhibitors were used to investigate involvement of specific signaling pathways. LIF increased muscle glucose uptake in dose (50-5,000 pM/l) and time-dependent manners with maximal effects at the 30-min time point. LIF increased Akt Ser(473) phosphorylation (P) in soleus and EDL, whereas AMPK Thr(172) P was unaffected. Incubation with parthenolide abolished LIF-induced glucose uptake and STAT3 Tyr(705) P, whereas incubation with LY-294002 and wortmannin suppressed both basal and LIF-induced glucose uptake and Akt Ser(473) P, indicating that JAK and PI 3-kinase signaling is required for LIF-stimulated glucose uptake. Incubation with rapamycin and AZD8055 indicated that mammalian target of rapamycin complex (mTORC)2, but not mTORC1, also is required for LIF-stimulated glucose uptake. In contrast to CNTF, LIF stimulation did not alter palmitate oxidation. LIF-stimulated glucose uptake was maintained in EDL from obese insulin-resistant mice, whereas soleus developed LIF resistance. Lack of SOCS3 and AMPKα2 did not affect LIF-stimulated glucose uptake. In conclusion, LIF acutely increased muscle glucose uptake by a mechanism potentially involving the PI 3-kinase/mTORC2/Akt pathway and is not impaired in EDL muscle from obese insulin-resistant mice.

Functional proteomics defines the molecular switch underlying FGF receptor trafficking and cellular outputs.

The stimulation of fibroblast growth factor receptors (FGFRs) with distinct FGF ligands generates specific cellular responses. However, the mechanisms underlying this paradigm have remained elusive. Here, we show that FGF-7 stimulation leads to FGFR2b degradation and, ultimately, cell proliferation, whereas FGF-10 promotes receptor recycling and cell migration. By combining mass-spectrometry-based quantitative proteomics with fluorescence microscopy and biochemical methods, we find that FGF-10 specifically induces the rapid phosphorylation of tyrosine (Y) 734 on FGFR2b, which leads to PI3K and SH3BP4 recruitment. This complex is crucial for FGFR2b recycling and responses, given that FGF-10 stimulation of either FGFR2b_Y734F mutant- or SH3BP4-depleted cells switches the receptor endocytic route to degradation, resulting in decreased breast cancer cell migration and the inhibition of epithelial branching in mouse lung explants. Altogether, these results identify an intriguing ligand-dependent mechanism for the control of receptor fate and cellular outputs that may explain the pathogenic role of deregulated FGFR2b, thus offering therapeutic opportunities.

Predictive mutagenesis of ligation-independent cloning (LIC) vectors for protein expression and site-specific chemical conjugation.

Ligation-independent cloning (LIC) allows for cloning of DNA constructs independent of insert restriction sites and ligases. However, any required mutations are typically introduced by additional, time-consuming steps. We present a rapid, inexpensive method for mutagenesis in the 5' LIC site of expression constructs and report on the construction of expression vectors with N-terminal serine, cysteine, threonine, or tyrosine residues after tobacco etch virus (TEV) protease cleavage. In a practical application, the N-terminal serine was oxidized to an aldehyde, subsequently reacted with an amino-oxy functionalized polyethylene glycol (PEG) ligand under aniline catalysis to provide a protein selectively modified at the N-terminus.

A secretory system for bacterial production of high-profile protein targets.

Escherichia coli represents a robust, inexpensive expression host for the production of recombinant proteins. However, one major limitation is that certain protein classes do not express well in a biologically relevant form using standard expression approaches in the cytoplasm of E. coli. To improve the usefulness of the E. coli expression platform we have investigated combinations of promoters and selected N-terminal fusion tags for the extracellular expression of human target proteins. A comparative study was conducted on 24 target proteins fused to outer membrane protein A (OmpA), outer membrane protein F (OmpF) and osmotically inducible protein Y (OsmY). Based on the results of this initial study, we carried out an extended expression screen employing the OsmY fusion and multiple constructs of a more diverse set of human proteins. Using this high-throughput compatible system, we clearly demonstrate that secreted biomedically relevant human proteins can be efficiently retrieved and purified from the growth medium.

Screening of genetic parameters for soluble protein expression in Escherichia coli.

Soluble expression of proteins in a relevant form for functional and structural investigations still often remains a challenge. Although many biochemical factors are known to affect solubility, a thorough investigation of yield-limiting factors is normally not feasible in high-throughput efforts. Here we present a screening strategy for expression of biomedically relevant proteins in Escherichia coli using a panel of six different genetic variations. These include engineered strains for rare codon supplementation, increased disulfide bond formation in the cytoplasm and novel vectors for secretion to the periplasm or culture medium. Combining these variants with expression construct truncations design, we report on parallel cloning and expression of more than 300 constructs representing 24 selected proteins; including full-length variants of human growth factors, interleukins and growth factor binding proteins. This rapid screening approach appears highly suitable for high-throughput efforts targeting either large sets of proteins or more focused investigations regarding individual high-profile targets.

Selection and characterisation of affibody molecules inhibiting the interaction between Ras and Raf in vitro.

Development of molecules with the ability to selectively inhibit particular protein-protein interactions is important in providing tools for understanding cell biology. In this work, we describe efforts to select small Ras- and Raf-specific three-helix bundle affibody binding proteins capable of inhibiting the interaction between H-Ras and Raf-1, from a combinatorial library displayed on bacteriophage. Target-specific variants with typically high nanomolar or low micromolar affinities (K(D)) could be selected successfully against both proteins, as shown by dot blot, ELISA and real-time biospecific interaction analyses. Affibody molecule variants selected against H-Ras were shown to bind epitopes overlapping each other at a site that differed from that at which H-Ras interacts with Raf-1. In contrast, an affibody molecule isolated during selection against Raf-1 was shown to effectively inhibit the interaction between H-Ras and Raf-1 in a dose-dependent manner. Possible intracellular applications of the selected affibody molecules are discussed.

Selection of affibody molecules to the ligand-binding site of the insulin-like growth factor-1 receptor.

Affibody molecules binding to the site of hormone interaction in IGF-1R (insulin-like growth factor-1 receptor) were successfully selected by phage-display technology employing a competitive-elution strategy during biopanning, whereby release of receptor-bound phagemids was accomplished by competition with IGF-1 (insulin-like growth factor-1). In non-competitive selections, the elution of receptor-bound phagemids was performed by imidazole or low-pH incubation, which also resulted in the isolation of affibody molecules that could bind to the receptor. An ELISA-based assay showed that the affibody molecules generated by IGF-1 competition during elution, in addition to affibody molecules generated in the non-competitive selections, could compete with IGF-1 for binding to the receptor. The affinities of the isolated variants to IGF-1R-overexpressing MCF-7 cells were determined and ranged from high nanomolar to 2.3 nM. The most promising variant, Z4:40, was shown to recognize IGF-1R efficiently in several different contexts: in analyses based on flow cytometry, fluorescence microscopy and receptor pull-down from cell extracts. In addition, when Z4:40 was added to the medium of MCF-7 cells that were dependent on IGF-1 for efficient growth, it was found to have a dose-dependent growth-inhibitory effect on the cells. Applications of affibody-based reagents for quantitative and qualitative analyses of IGF-1R status, as well as applications of affibody-based reagents for therapy, are discussed.

Affibody-mediated retention of the epidermal growth factor receptor in the secretory compartments leads to inhibition of phosphorylation in the kinase domain.

Abnormal activity of the epidermal growth factor receptor (EGFR) is associated with various cancer-related processes and motivates the search for strategies that can selectively block EGFR signalling. In this study, functional knockdown of EGFR was achieved through expression of an affibody construct, (ZEGFR:1907)(2-)KDEL, with high affinity for EGFR and extended with the amino acids KDEL to make it resident in the secretory compartments. Expression of (ZEGFR:1907)(2-)KDEL resulted in 80% reduction ofthe cell surface level of EGFR, and fluorescent staining for EGFR and the (ZEGFR:1907)(2-)KDEL construct showed overlapping intracellular localisation. Immunocapture of EGFR from cell lysates showed that an intracellular complex between EGFR and the affibody construct had been formed, further indicating aspecific interaction between the affibody construct and EGFR. Surface depletion of EGFR led to a dramatic decrease in the amount of kinase domain phosphorylated EGFR, coincident with a significant decrease in the proliferation rate.

Affinity-based entrapment of the HER2 receptor in the endoplasmic reticulum using an affibody molecule.

Interference with the export of cell surface receptors can be performed through co-expression of specific affinity molecules designed for entrapment in the endoplasmic reticulum during the export process. We describe the investigation of a small (6 kDa) non-immunoglobulin-based HER2 receptor binding affibody molecule (Z(HER2:00477)), for use in affinity mediated entrapment of the HER2 receptor in the ER. Constructs encoding Z(HER2:00477) or a control affibody protein, with or without ER-retention peptide extensions (KDEL), were expressed in the HER2 over-expressing cell line SKOV-3. Intracellular expression of the full-length affibody constructs could be confirmed by probing cell extracts by Western blotting. Confocal immunofluorescence microscopy experiments showed extensive co-localization of the HER2 receptor and Z(HER2:00477)-KDEL in the ER, whereas the use of a KDEL-extended control affibody molecule resulted in distinct and separate signals from cell surface-localized HER2 receptor and ER-localized affibody protein. This indicated a capability of the Z(HER2:00477)-KDEL fusion protein to functionally interfere with the export process of HER2 receptor in a specific manner. Using flow cytometry and cell proliferation analyses, it could be shown that expression of the Z(HER2:00477)-KDEL fusion construct in the SKOV-3 cell line resulted both in a marked reduction in cell surface level of HER2 receptors and that the cell population doubling time was significantly increased. Expression of the Z(HER2:00477)-KDEL fusion protein in additional cell lines of different origin and with different expression levels of endogenous HER2 receptor compared to SKOV-3, also resulted in depletion of the cell surface levels of HER2 receptor. This indicated upon a general ability of the Z(HER2:00477)-KDEL fusion protein to functionally interfere with the export process of HER2.