1046 Biochemical Society Transactions (2010) Volume 38, part 4 High-content screening of small compounds on human embryonic stem cells Ivana Barbaric, Paul J. Gokhale and Peter W. Andrews 1 Centre for Stem Cell Biology, University of Sheffield, Western Bank, Sheffield S10 2TN, U.K. Abstract Human ES (embryonic stem) cells and iPS (induced pluripotent stem) cells have been heralded as a source of differentiated cells that could be used in the treatment of degenerative diseases, such as Parkinson’s disease or diabetes. Despite the great potential for their use in regenerative therapy, the challenge remains to understand the basic biology of these remarkable cells, in order to differentiate them into any functional cell type. Given the scale of the task, high-throughput screening of agents and culture conditions offers one way to accelerate these studies. The screening of small-compound libraries is particularly amenable to such high-throughput methods. Coupled with high-content screening technology that enables simultaneous assessment of multiple cellular features in an automated and quantitative way, this approach is proving powerful in identifying both small molecules as tools for manipulating stem cell fates and novel mechanisms of differentiation not previously associated with stem cell biology. Such screens performed on human ES cells also demonstrate the usefulness of human ES/iPS cells as cellular models for pharmacological testing of drug efficacy and toxicity, possibly a more imminent use of these cells than in regenerative medicine. Introduction ES (embryonic stem) cells, obtained from a blastocyst stage embryo [1], and iPS (induced pluripotent stem) cells, derived by somatic cell reprogramming [2,3], have the ability to self- renew in culture while remaining pluripotent. These features make human ES/iPS cells a potential renewable source of all somatic cell types for use in regenerative therapies. However, the clinical potential of ES cells is hampered by many obstacles, such as the difficulties in achieving directed differentiation of ES cells in an effective and efficient manner [4]. Indeed, during embryonic development, stem cells reside in complex microenvironments that present them with a plethora of signals to instruct them on their developmental pathways. Deciphering these signals and mimicking them in vitro to drive the differentiation of human ES cells to a desired cell type presents an imposing challenge. Classical genetic strategies to identify signalling pathways governing biological processes in ES cells involve identifying potentially relevant genes, and then overexpressing or inactivating the gene or altering the protein’s function by genetic modification [5–7]. As an alternative in recent years, bioactive small compounds have emerged as powerful tools for probing signalling pathways that specify stem cell fates. Key words: drug discovery, embryonic stem cell, high-content screening, high-throughput screening, induced pluripotent stem cell. Abbreviations used: ES, embryonic stem; iPS, induced pluripotent stem; Oct4, octamer-binding protein 4; SSEA3, stage-specific embryonic antigen 3. 1 To whom correspondence should be addressed (email P.W.Andrews@sheffield.ac.uk). Chemical genetics approach in studying stem cell biology The complementary approach to classical genetics, where small molecules are exploited to probe biological functions, is termed ‘chemical genetics’ [8,9]. Whereas classical genetics establishes genotype–phenotype relationships through means of genetic manipulation, chemical genetics uses small molecules that intervene in signalling pathways through direct interaction with proteins to unravel relationships between proteins and phenotypes. By analogy to classical genetics, small compounds in chemical genetics are equivalent to mutations in classical genetics. In fact, given that chemicals can be removed from cells by simple washing, they are analogous to conditional mutations. Although small molecules allow reversible temporal and dose-dependent control of protein function, one of their major limitations is that a single compound often affects multiple proteins and, ipso facto, multiple pathways simultaneously [10,11]. The promiscuous binding of a compound to a range of proteins seriously impedes efforts to ascertain its true molecular targets and mechanism of action. However, as long as the binding properties of compounds are known, their pleiotropic nature can also be beneficial, for example, if a compound affects a combination of proteins such that the overall effect results in a phenotype of interest. In fact, genetic studies in various models have shown that inactivation of a single gene is often not enough to lead to a discernable phenotype, due to mechanisms of genetic robustness [12,13]. Thus treatment with a single small molecule can be analogous to simultaneous inactivating mutations in several genes. C  The Authors Journal compilation C  2010 Biochemical Society Biochem. Soc. Trans. (2010) 38, 1046–1050; doi:10.1042/BST0381046