Novel CPG-Depleted and Codon-Optimised CFTR CDNAs Maintain the Structure and Function of CFTR Protein.

Varathalingam A, Lawton AE, Munkonge FM, Chan M, Pringle IA, Greisenbach U, Alton EW, Gill DR, Hyde SC

Pediatric Pulmonology, 40 S28 275

The North American Cystic Fibrosis Conference, Baltimore, 2005

Pre-clinical and clinical studies have identified two limitations of non-viral mediated CFTR gene transfer for CF gene therapy. Firstly, there is a need to enhance the level and duration of CFTR gene expression, and secondly there is a need to minimise gene transfer-induced host inflammation. CpG dinucleotides present in the plasmid DNA (pDNA) component of non-viral gene transfer agents are known to contribute to the induction of an acute inflammatory response when delivered to the lungs of experimental animal models and CF patients. Reduction in the number of CpGs within pDNA expression vectors has been shown to reduce the host inflammatory response to non-viral gene transfer. The standard CFTR cDNA contains 59 CpG motifs. Thus we have generated two new forms of the CFTR cDNA in which all 59 CpGs have been eliminated without altering the amino acid sequence of the encoded CFTR protein. In the simplest of the two new forms (termed sCFTR) we introduced 96 base changes - the minimum number of changes required to eliminate all CpGs along with the cryptic bacterial promoter in exon 6B (previously shown to destabilise the cDNA during propagation in bacterial hosts). The sCFTR cDNA is 97.8% identical to the standard CFTR cDNA. In addition, we noted that the standard CFTR cDNA utilises a high frequency of codons that are poorly translated. Therefore a second CFTR cDNA was generated (termed soCFTR) in which 1010 base changes were made resulting in a cDNA which in addition to being CpG depleted, was also codon optimised; and was only 77.4% identical to the standard CFTR cDNA. To study any impact of these base changes we inserted these three CFTR cDNAs into a common pDNA backbone under the control of the CMV immediate-early enhancer promoter and investigated the expressed CFTR protein in HEK293T cells. Western blotting showed that all three forms of CFTR cDNA directed the expression of mature, fully glycosylated band C, and immature, core glycosylated band B CFTR protein. Furthermore, there were no changes in the steady state ratio of band C to band B (ANOVA, p=0.58) suggesting that CpG depletion with or without codon optimisation had no deleterious effect on the glycosylation status or the efficiency of CFTR trafficking in this mammalian cell line. CFTR function was investigated using an Iodide125 efflux assay. Forskolin/IBMX-dependent Iodide125 efflux generated by sCFTR and soCFTR was not significantly different from standard CFTR (p>0.05), but all three were significantly different from cells transfected with an irrelevant pDNA (ANOVA, p=0.013, p=0.019 and p=0.001, respectively). These data confirm that CpG depletion with or without codon optimisation to the CFTR cDNA does not alter the CFTR structure and function in cell culture. Non-viral mediated CFTR gene transfer containing these two new forms of CFTR cDNA is under investigation in animal models and could produce functional CFTR with minimal inflammatory effects in vivo.