It has recently been shown that efficient and stable expres- sion of transgenes from the A2UCOE is at least in part due to its resistance to DNA methylation- mediated silencing. Ana- lysis of a deletion series from the CBX3 end of a fully func- tional 2.2kb A2UCOE where expression of an eGFP reporter gene is directly driven off the HNRPA2B1 promoter, revealed a 1.2kb and 1.7kb truncation that retained full UCOE activity following transduction of P19 and F9 cells in both undiffer- entiated and differentiated states. These 1.2–1.7 and a further deletion 0.5kb A2UCOEs, were also able to retain stable ex- pression in murine embryonic stem cells and now in human iPS cells during differentiation into embryoid bodies and tissue specific cell types. In addition, from a practical perspective our finding that the 0.5-1.2-1.7A2UCOEs retain the same stability of expression as the larger 2.2A2UCOE patent suggests that it can effectively replace the latter within therapeutic LV constructs allowing a greater capacity and also more safety for the gene of interest by cutting off the enhancer elements which carries a po- tential mutation risk. A further project now on human mAb pro- duction with recombinant CHO cells by using our new UCOE models has been started and the early results show that they are more productive than the control A2UCOEs and able to produce human recombinant mAbs in mg level even under two months of cultivation. The results already encouraging to provide a new better UCOE model for recombinant protein production in mammalian cells.