Restriction for gene insertion within the Lactococcus lactis Ll.LtrB group II intron

The LI.LtrB intron, from the low G+C gram-positive bacterium Lactococcus lactis, was the first bacterial group II intron shown to splice and mobilize in vivo. The detailed retrohoming and retrotransposition pathways of LI.LtrB were studied in both L. lactis and Escherichia coli. This bacterial retroelement has many features that would make it a good gene delivery vector. Here we report that the mobility efficiency of LI.LtrB expressing LtrA in trans is only slightly affected by the insertion of fragments < 100 nucleotides within the loop region of domain IV. In contrast, LI.LtrB mobility efficiency is drastically decreased by the insertion of foreign sequences > 1 kb. We demonstrate that the inhibitory effect caused by the addition of expression cassettes on LI.LtrB mobility efficiency is not sequence specific, and not due to the expression, or the toxicity, of the cargo genes. Using genetic screens, we demonstrate that in order to maintain intron mobility, the loop region of domain IV, more specifically domain IVb, is by far the best region to insert foreign sequences within LI.LtrB. Poisoned primer extension and Northern blot analyses reveal that Ll.LtrB constructs harboring cargo sequences splice less efficiently, and show a significant reduction in lariat accumulation in L. lactis. This suggests that cargo-containing LI.LtrB variants are less stable. These results reveal the potential, yet limitations, of the LI.LtrB group II intron to be used as a gene delivery vector, and validate the random insertion approach described in this study to create cargo-containing LI.LtrB variants that are mobile.