The application of Cas9 for genetic and epigenetic therapies in humans raises concerns over immunogenicity of this foreign protein. We report pre-existing human CD8+ T cell immunity to Streptococcus pyogenes Cas9 in the majority of healthy individuals screened. In a proof-ofprinciple study, we demonstrate that Cas9 protein can be modified to eliminate immunodominant epitopes through targeted mutation while preserving its function and specificity.Of the Cas9 orthologs derived from bacterial species, the SpCas9 is the best characterized. S. pyogenes is a ubiquitous pathogen, with an annual incidence of 700 million worldwide 26 , but immunity to SpCas9 in humans has not been reported. Here, we sought to characterize the pre-existing immune response to SpCas9 in healthy individuals and to identify the immunodominant T cell epitopes with the aim of developing SpCas9 proteins that have diminished capacity to invoke human adaptive response.CRISPR application for human therapies will span its use both for gene editing (through DNA double-strand breaks) or epigenetic therapies (without DNA double-strand breaks). In fact, recent reports shed light on CRISPR's ability to activate or repress gene expression in mice [27][28][29] , which opens the door to a variety of new therapeutic applications such as activating silent genes, compensating for disrupted genes, cell fate reprogramming, or silencing disrupted genes, without the concern over permanent change in DNA sequence. However, unlike the use of Cas9 for gene editing, which may only require Cas9 presence in cells for a few hours, current techniques for CRISPR-based epigenetic therapies require longer term expression of Cas9 in vivo, possibly for weeks and months 28,29 , which poses the challenge of combating pre-existing immune response towards Cas9. This challenge will need to be addressed before CRISPR application for human therapies, especially for epigenetic therapies, can be fully implemented. Delivery of CRISPR in vivo by incorporating its expression cassette in adeno-associated virus (AAV), will most likely shape many of the initial clinical trials as AAV-based gene delivery is one of the safest and most prevalent forms of gene therapies in human. AAV will enable longer term expression of Cas9, desirable for epigenetic therapies. Therefore, unlike Cas9 delivery in the form of ribonucleoprotein complexes (which are short term), it is highly likely that CRISPR delivery through AAV and its expression within target cells will engage CD8+ T cell immunity.