Drug Toxicity Screening Platform
We have developed new tools to interrogate the effects of small molecules on biological function (Figure DD-1). These include cell- and protein-based high-throughput screening, on chip immunoassays and high-throughput biocatalysis on a microarray platform (Lee et al. Proc. Natl. Acad. Sci. USA 102, 983-987 (2005); Sukumaran et al. J. Biomol. Screen. 14, 668-678 (2009)]. With respect to the cell-based screening platform, we have developed (in collaboration with Doug Clark’s group at UC Berkeley) a miniaturized three-dimensional (3D) cellular array chip (the DataChip, Figure DD-2) that has been used for in vitro toxicological assessment of drug candidates and other chemicals in high-throughput [Lee et al. Proc. Natl. Acad. Sci. USA 105, 59-63 (2008)]. Although recent emphasis has been placed on understanding cellular metabolism in 3D, relatively little effort has been spent using 3D cultures to study cytotoxicity, particularly at very small volumes consistent with high-throughput screening. Furthermore, we have recently begun to investigate the toxicity of compounds on human stem and progenitor cells in comparison to their differentiated progeny.
Figure DD-1. New high-throughput tools being developed. Upper left is the DataChip (Data Analysis Toxicology Assay Chip) for high throughput 3D cell culture screens. Upper right is the MetaChip (Metabolizing Enzyme Assay Chip for high-throughput interrogation of enzyme function. In this case, CYP450 inhibition assays7. Lower left is a modification of the DataChip to enable on-chip, in-cell immunofluorescence assays and quantitative analysis of protein expression levels within cells6. Lower right is a schematic of an "Artificial Golgi", which includes an image of a digital microfluidic device.
Figure DD-2. The DataChip[Lee et al. Proc. Natl. Acad. Sci. USA 105, 59-63 (2008)]. Upper left represents the hemispherical spots generated by microarray spotting of cells in alginate solution, which upon contact with poly-L-lysine and BaCl2 results in near immediate gelation. The cells grow well up to 5 days in culture (upper right). Lower left shows an actual image of a stained DataChip (live-dead assay) and representative dose response curves for drug candidates and chemicals in the presence or absence of metabolizing enzymes contained within a MetaChip and stamped on top of the DataChip. Lower right is the output of the on-chip, in-cell immunofluorescence assay [Fernandes et al. Anal. Chem. 80, 6633-6639 (2008)]. The dose response to an activator of HIF-1α in a pancreatic tumor cell line (triangles represent standard Western blotting analysis on mL scale and open squares represent the on-chip immunofluorescence assay in volumes as low as 30-nL).
More recently, we have developed a novel chip design that enables interrogation of differential expression of various drug-metabolizing enzymes (DMEs) in the human liver. Such information is relevant to inter-individual variability in drug pharmacokinetic profiles, drug efficacy and toxicity. Specifically, we developed the “Transfected Enzyme and Metabolism Chip” (TeamChip), which predicts potential metabolism-induced drug or drug-candidate toxicity [Kwon et al. Nat. Commun. 5, 3739 (2014)]. The TeamChip is prepared by delivering genes into miniaturized three-dimensional cellular microarrays on a micropillar chip using recombinant adenoviruses in a complementary microwell chip (Figure DD-3). The device enables users to manipulate the expression of individual and multiple human metabolizing-enzyme genes (such as CYP3A4, CYP2D6, CYP2C9, CP1A2, CYP2E1, and UGT1A4) in THLE-2 cell microarrays. To identify specific enzymes involved in drug detoxification, we created 84 combinations of metabolic-gene expressions in a combinatorial fashion on a single microarray (Figure DD-4). Thus, the TeamChip platform can provide critical information necessary for evaluating metabolism-induced toxicity in a high-throughput manner.
Figure DD-3. TeamChip schematics and photographs. (a) Micropillar/microwell chip components in relation to a standard glass microscope slide. (b) The micropillar chip containing THLE-2 cells encapsulated in matrigel droplets. (c) The microwell chip containing recombinant adenoviruses carrying genes for drug-metabolizing enzymes (the red color indicates the no-virus control and the two colors represent different viruses). (d) Stamping of the micropillar/microwell chips for drug-metabolizing gene expression. (e) Experimental procedure for use of the TeamChip. [Kwon et al. Nat. Commun. 5, 3739 (2014)].
Figure DD-4. Effects of combinatorial expression of human drug-metabolizing enzymes on the toxicity of tamoxifen. (a) Layout of the microwell chip containing 84 combinations of multiple recombinant adenoviruses (three sets of recombinant adenoviruses dispensed sequentially) to prepare the TeamChip for high-throughput gene transduction, and an additional microwell chip containing 200 mM tamoxifen for metabolism-induced toxicity screening. (b) Scanned image of THLE-2 cells expressing 84 combinations of multiple drug-metabolizing enzymes on the chip exposed to 200 mM tamoxifen for 48 h (top) and normalized THLE-2 cell viability at different drug-metabolizing enzyme expression levels (bottom). The viability of multiple drug-metabolizing enzyme-expressing THLE-2 cells exposed to tamoxifen was normalized by the fluorescent intensity of THLE-2 cells incubated in the absence of compound. The least toxic region in the scanned image is highlighted in a yellow box, and the red circle in the graph designates normalized THLE-2 cell viability calculated from the least toxic region that involved the combination of CYP1A2 and UGT1A4.
The enormous pool of chemical diversity found in nature serves as an excellent inventory for accessing biologically active compounds. This chemical inventory, primarily found in microorganisms and plants, is generated by a broad range of enzymatic pathways under precise genetic and protein-level control. In vitro pathway reconstruction can be used to characterize individual pathway enzymes, identify pathway intermediates, and gain an increased understanding of how pathways can be manipulated to generate natural product analogs. Moreover, through in vitro approaches, it is possible to achieve a diversification that is not restricted by toxicity, limited availability of intracellular precursors, or preconceived (by nature) regulatory controls. Additionally, combinatorial biosynthesis and high-throughput techniques can be used to generate both known natural products and analogs that would not likely be generated naturally.
We have exploited advances in DNA sequencing technology, genome mining, and microbial metagenomics to identify natural product biosynthetic pathways and open up a vast portion of the unexploited natural product space. In vitro reconstruction of natural product biosynthetic pathways (Figure MB-1) can be used to [Kwon et al. Curr. Opin. Chem. Biol. 16, 186-195 (2012)]: 1) generate highly pure biosynthetic intermediates and final products; 2) limit competing side reactions and cell metabolites (including some that are potentially toxic) that often occur in vivo, thereby complicating pathway understanding; 3) overcome unknown or poorly controlled regulatory constraints, including posttranslational limitations such as protein–protein interactions; and 4) enable unnatural synthetic building blocks to be used. An example of our approach is in the diversification of type III polyketide synthase (PKS) reactions performed in high throughput.
Figure MB-1. Schematic of in vitro reconstitution and biosynthetic pathway engineering for generating (unnatural) terpenoids, polyketides and nonribosomally generated peptides. Enzyme notation: terpenoid synthase (TS); halogenase (Hal); ligase (L); thioesterase (TE); aromatase (ARO); cyclase (CYC); methyltransferase (MT); oxygenase/oxidase (Ox); reductase (Re); glycosyltransferase (GT); epimerase (E); peroxidase (PO); prenyltransferase (PT) and carbamoyltransferase (CT) Kwon et al. Curr. Opin. Chem. Biol. 16, 186-195 (2012)].
The generation of biological diversity by engineering the biosynthetic gene assembly of metabolic pathway enzymes has led to a wide range of “unnatural” variants of natural products. However, current biosynthetic techniques do not allow the rapid manipulation of pathway components and are often fundamentally limited by the compatibility of new pathways, their gene expression, and the resulting biosynthetic products and pathway intermediates with cell growth and function. To overcome these limitations, we have developed an entirely in vitro approach to synthesize analogs of natural products in high throughput. For example, Figure MB-2 highlights the central location of the type III PKS in aromatic metabolism, and in particular the pathway connections among polyphenols, lignin biosynthesis, and oligophenol biosynthesis.
Figure MB-2. Central metabolic pathways involving phenolic compounds, and branch point to type III PKS pathway.
Using several type III polyketide synthases (PKS) together with oxidative post-PKS tailoring enzymes, we performed 192 individual and multienzymatic reactions on a single glass microarray [Kwon et al. ACS Chem. Biol. 2, 419-425 (2007); Kim et al. Org. Lett. 11, 3806-3809 (2009)]. Subsequent array-based inhibition screening with a human tyrosine kinase led to the identification of three compounds that acted as modest inhibitors in the low micromolar range. Receptor tyrosine kinases are critical targets for the regulation of cell survival. Cancer patients with abnormal receptor tyrosine kinases (RTK) tend to have more aggressive disease with poorer clinical outcomes. As a result, human epidermal growth factor receptor kinases, such as EGFR (HER1), HER2, and HER3, represent important therapeutic targets (Figure MB-3). We have developed an in vitro route to the synthesis and subsequent screening of unnatural polyketide analogues with N-acetylcysteamine (SNAc) starter substrates and malonyl-coenzyme A (CoA) and methylmalonyl-CoA as extender substrates. The resulting polyketide analogues possessed a similar structural polyketide backbone (aromatic-2-pyrone) with variable side chains. Screening chalcone synthase (CHS) reaction products against BT-474 cells resulted in identification of several trifluoromethylcinnamoyl-based polyketides that showed strong suppression of the HER2-associated PI3K/AKT signaling pathway yet did not inhibit the growth of non-transformed MCF-10A breast cells. This approach, therefore, enables the rapid construction of analogs of natural products as potential pharmaceutical lead compounds.
Figure MB-3. Type III PKS generation of highly potent anticancer compounds. Upper left is a cartoon depicting the potential interaction of a trifluormethylcinnamoyl-based pyrone against the HER receptor tyrosine kinase isoforms. Right panels depict very low IC50 values (~30 nM) against BT-474 (HER-overexpressing cell line) vs. non-HER overexpressing cell line (MCF-7) and a non-transformed cell line (MCF-10A) [Kwon et al. ACS Chem. Biol. 2, 419-425 (2007)].
Douglas Clark - U.C. Berkeley
MooYeal Lee - Cleveland State University