Department of Biological Science
National University of Singapore
14 Science Drive 4
Zebrafish developmental biology, zebrafish models for liver cancers, fish toxicogenomics and biomonitoring.
Zebrafish models for liver cancers
The zebrafish has become an increasingly popular animal model for high-throughput and low cost studies. Previously, by microarray-based transcriptome analyses, we have demonstrated that the carcinogen-induced zebrafish liver tumors share remarkable similarity with human liver tumors in cancer molecular hall markers, molecular pathways, as well as features in tumor progression, thus validating the zebrafish model for human disease. To generate reproducible zebrafish models for liver cancers, we employed a transgenic approach to over-express selected oncogenes under a liver-specific promoter. So far, we have generated a few inducible and non-inducible transgenic lines by expression of kras, myc or Xmrk; all of them are capable of producing liver cancers when they are highly expressed. In the relatively well studied kras transgenic lines, in which a constitutively active form of kras mutant (G12V) is expressed, abnormal over-growth of liver could be observed within one week of fertilization and the lesion of liver become increasingly severe with time, from hyperplasia to adenoma and to carcinoma. These tumors could be transplanted to other adult fish and survived, propagated and migrated in the host fish. We have applied two inducible transgenic systems using the Tet-on and mifepristone induction. In both inducible systems, we demonstrated that production of liver tumors could be induced at any developmental stages and even in adult fish. More interestingly, the induced tumors show rapid regression upon removal of the inducer. To explore the potential of the transgenic tumor model for future anti-cancer drug screening, we developed a protocol of using transgenic fry for liver tumor assay and found the inhibition of liver hyperplasia by several chemical inhibitors. These observations provide a basis for high-throughput chemical screening for anti-cancer drugs using our transgenic zebrafish models. Currently we are also developing rapid assays in the zebrafish model for testing potential oncogenes and tumor suppressor genes that are capable of driving hepatocarcinogenesis.
Fish toxicogenomics and biomonitoring
We are interested in use of both the zebrafish and medaka to develop tools for monitoring environmental pollution. Previously we have developed several GFP transgenic medaka lines using different inducible promoters and these transgenic medaka show inducible GFP expression in response specifically to several classes of environmental contaminants including estrogenic compounds, polycyclic aromatic hydrocarbons and heavy metals. Thus, these transgenic fish may be used for developing an online water monitoring system. Now with the advent of powerful genomic tools including DNA microarray and next generation sequencing, it is feasible to analyse transcriptome of zebrafish under different chemical insult conditions. We hypothesize that each chemical pollutant would cause some characteristic changes in transcriptomes in exposed fish and these changes may be used to predict the classes of environmental pollutants. To test the hypothesis, we exposed the adult zebrafish with more than a dozen of environment-relevant chemicals and transcriptomic data were generated by DNA microarray. We found that indeed chemical-induced transcriptomic changes could be correctly grouped based on the hierarchical clustering of transcriptomic data. However, for a robust and practical prediction of the presence of certain chemicals, it was necessary to use a small group of selected responsive genes as biomarkers for each chemical. Currently we have generated RNA-seq data from zebrafish exposed to several important environmental chemicals and preliminary bioinformatic analyses indicate that each of these chemicals up-regulates a very distinct set of genes with little overlapping. These up-regulated genes from each of the chemical treatment groups are apparently associated with different biological pathways. These observations provide a basis for identification of specific biomarker genes to be used for determine environmental contaminations through biological effects. Our long-term goal is to develop practical PCR arrays using biomarker genes identified from laboratory zebrafish and to apply universally to all wild fish species. Because of the high content transcriptomic information from these microarray and RNA-seq data, we will also analyze molecular mechanism of toxicity of each pollutant, which is valuable for risk assessment of each pollutant relevant to human health.