Impairment of DNA mismatch repair (MMR) function leads to the development and progression of certain cancers. Many environmental contaminants can target DNA MMR system. Currently, measurement of MMR activity is limited to in vitro or in vivo methods at the cell line level, and reports on measurement of MMR activity at the live organism level are lacking. Here, we report an efficient method to measure DNA MMR activity in zebrafish embryos. A G-T mismatch was introduced into enhanced green fluorescent protein (EGFP) gene. Repair of the G-T mismatch to G-C in the heteroduplex plasmid generates a functional EGFP expression. The heteroduplex plasmid and a similarly constructed homoduplex plasmid were injected in parallel into the same batch of embryos at 1-cell stage and EGFP expression in EGFP positive embryos was quantified at 24h after injection. MMR efficiency was calculated as the total fluorescence intensity of embryos injected with the heteroduplex construct divided by that of embryos injected with the homoduplex construct. Our results showed 73% reduction of MMR activity in embryos derived from MMR-deficient mlh1 mutant fish (positive control) when compared with embryos from MMR-competent wild type AB line fish, indicating feasibility of in vivo MMR activity measurement in zebrafish embryos. We further applied this novel assay for measurement of MMR efficiency in embryos exposed to environmental chemicals such as cadmium chloride (CdCl2), benzo[a]pyrene (BaP), and perfluorooctanesulphonic acid (PFOS) from 6 hpf to 24 hpf. We observed significant reductions of MMR efficiency in embryos exposed to 0.1 mu M CdCl2 (52%) and 0.5 mu M BaP (34%), but no effect in embryos exposed to PFOS. Our study for the first time provides a model system for in vivo measurement of DNA MMR activity at the organism level, which has important implications in risk assessment of various environmental carcinogens. (C) 2015 Elsevier B.V. All rights reserved.