Mechanistic basis for increased human gene targeting by promoterless vectors—roles of homology arms and Rad54 paralogs

Abstract

Gene targeting by homologous recombination provides the definitive tool for analyzing gene function. Promoterless vectors, which do not possess a promoter to drive marker gene expression, confer higher targeting efficiencies than conventional vectors due to the reduced number of drug-resistant clones. We here show that gene-targeting efficiency is typically ≥ 25% with the use of exon-trapping-type promoterless vectors in a human diploid cell line, Nalm-6. The efficiency of exon-trapping gene targeting was correlated with the level of target gene expression when a 2A peptide sequence was linked to the marker gene. Intriguingly, total arm length was not necessarily a determinant of targeting efficiency, as longer arms tend to enhance both homologous (targeted) and nonhomologous (nontargeted) integration of the vector; rather, the presence of an exon in the 5' arm led to a decreased targeting efficiency. Strikingly, loss of Rad54 did not severely affect the targeting efficiency of exon-trap vectors. Moreover, additional deletion of the Rad54 paralog Rad54B had limited impact on the high-efficiency gene targeting. These results indicate that targeted integration occurs in human cells even when both Rad54 and Rad54B are missing. These studies provide additional important insight into the contribution of various DNA repair factors on the targeting mechanics.

Characterization and regulation of cell cycle-independent noncanonical gene targeting

“…Indeed, TI frequency is markedly reduced when Rad51, an essential HR protein 12 , is genetically deleted 10 , 11 , 13 . Similar observations have been made in human and mouse cells that are doubly deficient in Rad54 and Rad54B, which both assist in the role of Rad51 to perform the strand exchange reaction during HR 14 , 15 . Importantly, however, those mutant cells retain the ability to bring about gene targeting although the TI frequency is reduced to less than 5% of that in Rad54/Rad54B-proficient cells 14 , 15 .…”

Complex genetic interactions between DNA polymerase β and the NHEJ ligase

“…The human pre-B-cell line Nalm-6 and its derivatives were cultured at 37°C in Eagle's MEM medium (Nissui Seiyaku, Tokyo, Japan) supplemented with 10% calf serum (Hyclone, Logan, UT, USA), 2 mM MEM non-essential amino acids (Wako Pure Chemical, Osaka, Japan), 1 mM sodium pyruvate (Wako Pure Chemical), 50 lM 2-mercaptoethanol (Wako Pure Chemical), and 0.15 lM Vitamin B12 (Sigma-Aldrich, St. Louis, MO, USA) [43]. LIG4 À/À cells and RAD54 À/À cells were generated as previously described [24,26].…”

Pushing the haystack aside for efficient gene targeting in human cells

“…). In this issue of The FEBS Journal , Adachi and colleagues studied how to increase the ratio of targeted integration to nontargeted integration in a human diploid cell line .…”

“…Adachi and colleagues studied the possible influence of the target gene expression level and length of the homology arms on the efficiency of targeted integration by promoter‐less exon‐trapping vectors. They used human pre‐B leukemia cell line Nalm‐6, which also exhibits high targeting frequency .…”

“…Targeted integration is thought to proceed through HR, whereas nontargeted integration results from NHEJ. Adachi and colleagues investigated the role of RAD54, a SWI/SNF family DNA‐dependent ATPase, and its paralog RAD54B in targeted integration. They found that targeted integration of a promoter‐less exon‐trapping vector takes place at considerably higher efficiency in the absence of RAD54, although that of a promoter‐containing vector was greatly diminished.…”