No abstract

Recent progress in developing programmable nucleases, such as zinc-finger nucleases, transcription activator-like effector nucleases (TALENs) and clustered regularly interspaced short palindromic repeat (CRISPR)-Cas nucleases, have paved the way for gene editing to enter clinical practice. This translation is a result of combining high nuclease activity with high specificity and successfully applying this technology in various preclinical disease models, including infectious disease, primary immunodeficiencies, hemoglobinopathies, hemophilia and muscular dystrophy. Several clinical gene-editing trials, both ex vivo and in vivo, have been initiated in the past 2 years, including studies that aim to knockout genes as well as to add therapeutic transgenes. Here we discuss the advances made in the gene-editing field in recent years, and specify priorities that need to be addressed to expand therapeutic genome editing to further disease entities.

IntroductionStandard gene-therapy approaches to cancer treatment, such as transfer of suicide genes that confer sensitivity to prodrugs, have limitations as cytoreductive strategies owing to insufficient bystander effects of the therapeutic gene combined with suboptimal transduction efficiency of currently available gene delivery vectors. A more compelling approach in this situation is the use of a vector or virus that is able to replicate within the tumor tissue, resulting in direct cell death through cytolysis or toxicity of viral proteins. Ideally, such an agent should also be capable of stimulating a potent immune response to the tumor within which it can replicate.Studies throughout the twentieth century have documented the lytic effects of various viruses on many types of human cancer, 1 and systematic study of candidate oncolytic viruses is intensifying. Viruses under investigation as oncolytic agents include human adenoviruses, ONYX-015, 2,3 reovirus, 4 herpes viruses 5,6 and vesicular stomatitis virus. 7 All of these viruses have shown promise in preclinical studies, and clinical studies of some of the agents are now in progress. 8 Viruses of the Paramyxoviridae family are also oncolytic. Almost 30 years ago, the human paramyxovirus, mumps, was administered to 90 patients with advanced malignancy, 9 resulting in significant (although mostly short-lived) responses. Toxicity was minimal. More recently, Newcastle disease virus, an avian paramyxovirus, has also shown promising results in preclinical studies, [10][11][12] and clinical trials in human subjects have begun.In this study, we have investigated another human paramyxovirus, measles, as a potential antitumor agent for lymphoid malignancies. Measles virus (MV) may be particularly promising as an oncolytic virus for the treatment of lymphoid malignancy for a number of reasons. First, a nonpathogenic strain of MV is available, well characterized, and safe. Live attenuated MV vaccines, derived from the Edmonston-B strain (MV-Ed), 13 have been used worldwide for more than 30 years, and in excess of 160 million doses have been administered in the United States alone with an excellent safety record. Second, although many human cell types are permissive for MV infection in vitro, in the presence of an intact immune system, virus replication after natural infection is limited to a few cell types in vivo. Lymphoid organs are prominent sites of MV replication; indeed, multinucleated giant cells develop during infection in lymph nodes as a result of gross cell-cell fusion. 14 Third, we have recently shown that expression of virally derived fusogenic membrane glycoproteins in tumor cells, including MV fusion (F) and hemagglutinin (H) glycoproteins, 15-17 results in a potent cytopathic effect mediated by massive cell-cell fusion. The considerable local bystander effect implies that transduction of all tumor cells would not be necessary to achieve significant tumor cell kill. However, the use of MV as a replicating vector with which to deliver the F and H glycopro...

The engineering of proteins to manipulate cellular genomes has developed into a promising technology for biomedical research, including gene therapy. In particular, zinc-finger nucleases (ZFNs), which consist of a nonspecific endonuclease domain tethered to a tailored zinc-finger (ZF) DNA-binding domain, have proven invaluable for stimulating homology-directed gene repair in a variety of cell types. However, previous studies demonstrated that ZFNs could be associated with significant cytotoxicity due to cleavage at off-target sites. Here, we compared the in vitro affinities and specificities of nine ZF DNA-binding domains with their performance as ZFNs in human cells. The results of our cell-based assays reveal that the DNA-binding specificity--in addition to the affinity--is a major determinant of ZFN activity and is inversely correlated with ZFN-associated toxicity. In addition, our data provide the first evidence that engineering strategies, which account for context-dependent DNA-binding effects, yield ZFs that function as highly efficient ZFNs in human cells.