In the past decades, significant progresses have been achieved in genetic engineering of nucleases. Among the genetically engineered nucleases, zinc finger nucleases, transcription activator-like (TAL) effector nucleases, and CRIPSPR/Cas9 system form a new field of gene editing. The gene editing efficiency or targeting effect and the off-target effect are the two major determinant factors in evaluating the usefulness of a new enzyme. Engineering strategies in improving these gene editing enzymes, particularly in minimizing their off-target effects, are the focus of this paper. Examples of using these genetically engineered enzymes in genome modification are discussed in order to better understand the requirement of engineering efforts in obtaining more powerful and useful gene editing enzymes. In addition, the identification of naturally existed anti-Cas proteins has been employed in minimizing off-target effects. Considering the future application in human gene therapy, optimization of these well recognized gene editing enzymes and exploration of more novel enzymes are both required. Before people find an ideal gene editing system having virtually no off-target effect, technologies used to screen and identify off-target effects are of importance in clinical trials employing gene therapy.

DNA polymerases without the 3' exonuclease function (exo(-) pol) have been widely used in sequencing and SNP genotyping. As a major player that expedited the coming of the postgenomic era, exo(-) polymerases worked remarkably well in the Human Genome Sequencing Project. However, it has become a challenge for this class of polymerases to efficiently screen the large number of SNPs that are found in the human genome. For more than three decades it has been recognized that polymerase fidelity varied according to the presence of proofreading activity that is mediated by its internal 3' exonuclease. Polymerases with proofreading function are famous for their high fidelity in DNA replication both in vivo and in vitro, but this well-known class of polymerases has been almost completely neglected in genetic analysis in the postgenomic era. We speculate that exo(+) polymerases may exhibit higher nucleotide identification ability when compared to exo- polymerases for an in vitro genetic analysis. With the application of exo(+) polymerases in SNP assays, a novel mechanism for the maintenance of DNA replication, the on/off switch, was discovered. Two new SNP assays have been developed to carry out genome-wide genotyping, taking advantage of the enzymatic properties of exo(+) polymerases. Furthermore, the on/off switch mechanism embodies a powerful nucleotide identification ability, which can be used to discriminate the bases that are upstream of the 3' terminus, and thus defines a new concept in de novo sequencing technology. Application of exo(+) polymerases to genetic analysis, and especially SNP assays, will greatly accelerate the pace to personalized medicine.

A primerless amplification suitable for enrichment of particular genotype cfDNA which is a one-dimensional material has been developed. This primerless amplification coordinated by two thermostable enzymes of endonuclease and proofreading polymerase, functions as a genotype switch in analyzing cfDNA. The endonuclease digests the wild-typed fragments into mega-primer and discriminately destroys the wild-type DNA alleles. The DNA polymerase proofreads the megaprimer and then extends the mega-primer using the mutant DNA as the template. The prototypes of this technology were applied to two hotspot mutations of APC and EGFR with confirmed by DNA sequencing analysis. Genotype switch was then employed to clinical cfDNA assay targeting PIK3CA. Data from the clinical application suggest its potential in early cancer diagnosis.

The target lengths of gene-editing enzymes vary from approximately 20 to 80 nucleotides. The ideal reporter genes should be able to tolerate the insertion of target sequences to differentiate the phenotypic changes between in-frame and frame shift mutations for evaluating the enzymatic activities and off-target effects of gene-editing enzymes such as zinc finger nucleases, transcription activator-like effector nucleases, clustered regularly interspaced short palindromic repeats–associated system. This study demonstrated that the zeocin resistance gene (ZeoR) tolerated the insertion of long targets for a variety of tests using a strategy of inserting tandem repeats of NNT. Four genes such as CCR5, AR, HPV 16E7 early gene, and EGFR applications of ZeoR in quantitatively analyzing the off-targets of CRISPR were successful. The data indicated that evaluating the enzymatic activities and off-target effects with assays employing ZeoR had great potential in facilitating the application of gene editing to human gene therapy.