Hemp (Cannabis sativa L), as a potentially profitable, renewable crop, has long been cultivated as a source of durable fibre used in textile, pulp and paper industries. Separation of fibre from nonfibre tissues known as retting is the first and crucial step in successful fibre production. Enzyme retting has been developed due to many disadvantages of the traditional retting methods. Pectinases including polygalacturonases and pectate lyases have potential to effectively ret bast fibres such as those of hemp and flax. In order to screen for enzymes useful for retting, we obtained 5 pectate lyase and 2 polygalacturonase-encoding sequences by data mining against the genomes of Streptomyces coelicolor A3(2) and Xanthomonas campestris pv. campestris. We have cloned and expressed all the 7 target pectinase genes. However, only 3 clones were found to have pectate lyase activity, and one exhibited polygalacturonase activity. The most active pectate lyase and polygalacturonase clones were further characterized. Importantly, these enzymes are active in the treatment of the natural hemp fibre, evidenced by the release of significant amount of pectin-degraded products. Our results suggested that genome data mining is an useful approach for new gene discovery with potential applications in industry.