Anti-Cas9 immunity: A formidable challenge for muscle genome editing

Hakim et al.1Hakim C.H. Kumar S.R.P. Perez-Lopez D.O. Wasala N.B. Zhang D. Yue Y. Teixeira J. Pan X. Zhang K. Million E.D. et al.Cas9-specific immune responses compromise local and systemic AAV CRISPR therapy in multiple dystrophic canine models.Nat. Commun. 2021; 12: 6769Crossref PubMed Scopus (23) Google Scholar recently published a paper in Nature Communications describing immune responses to the Cas9 nuclease in several dog models of Duchenne muscular dystrophy (DMD) undergoing gene editing therapy. Treatment via adeno-associated virus serotype 8 (AAV8)-mediated delivery of CRISPR/Cas9 (AAV-CRISPR) was initially effective at restoring dystrophin protein levels. However, strong muscle inflammation and Cas9-specific immune responses were observed by 6 weeks, corresponding with loss of AAV genomes and disappearance of dystrophin-positive fibers. It has long been suspected that persistent expression of the bacterially derived Cas9 nuclease could pose safety and efficacy concerns for gene therapy. This study is the first to demonstrate such a response in skeletal muscle in a larger mammalian model. DMD is one of the more severe of all human genetic diseases, and remains at the forefront of gene therapy and gene editing efforts. Due to the large 11.5 kb size of the dystrophin coding sequence, it cannot be delivered with single or even dual AAV vectors. There is a compelling case to be made for correcting the patient’s own DNA in situ to permanently restore dystrophin expression. Three landmark papers published in Science in 2016 showed that AAV delivery of CRISPR/Cas9 could delete exon 23 in the mdx mouse model—restoring expression of a truncated, but partially functional dystrophin protein.2Tabebordbar M. Zhu K. Cheng J.K.W. Chew W.L. Widrick J.J. Yan W.X. Maesner C. Wu E.Y. Xiao R. Ran F.A. et al.In vivo gene editing in dystrophic mouse muscle and muscle stem cells.Science. 2016; 351: 407-411Crossref PubMed Scopus (730) Google Scholar, 3Nelson C.E. Hakim C.H. Ousterout D.G. Thakore P.I. Moreb E.A. Castellanos Rivera R.M. Madhavan S. Pan X. Ran F.A. Yan W.X. et al.In vivo genome editing improves muscle function in a mouse model of Duchenne muscular dystrophy.Science. 2016; 351: 403-407Crossref PubMed Scopus (799) Google Scholar, 4Long C. Amoasii L. Mireault A.A. McAnally J.R. Li H. Sanchez-Ortiz E. Bhattacharyya S. Shelton J.M. Bassel-Duby R. Olson E.N. Postnatal genome editing partially restores dystrophin expression in a mouse model of muscular dystrophy.Science. 2016; 351: 400-403Crossref PubMed Scopus (666) Google Scholar Since this time, DMD mutations have been corrected in pig5Moretti A. Fonteyne L. Giesert F. Hoppmann P. Meier A.B. Bozoglu T. Baehr A. Schneider C.M. Sinnecker D. Klett K. et al.Somatic gene editing ameliorates skeletal and cardiac muscle failure in pig and human models of Duchenne muscular dystrophy.Nat. Med. 2020; 26: 207-214Crossref PubMed Scopus (103) Google Scholar and canine6Amoasii L. Hildyard J.C.W. Li H. Sanchez-Ortiz E. Mireault A. Caballero D. Harron R. Stathopoulou T.R. Massey C. Shelton J.M. et al.Gene editing restores dystrophin expression in a canine model of Duchenne muscular dystrophy.Science. 2018; 362: 86-91Crossref PubMed Scopus (298) Google Scholar models. However, the ultimate success of AAV-CRISPR as a therapy may depend upon how myofibers expressing the Cas9 nuclease are treated by the patient’s immune system. Cas9 is the protein component of the CRISPR/Cas9 editing system. The most commonly used forms of Cas9 are derived from Streptococcus pyogenes (SpCas9) and Staphylococcus aureus (SaCas9). These bacteria are common pathogens we encounter on a daily basis, which can also cause life-threatening illness. Interestingly, due to this pervasive environmental exposure, most of us have pre-existing immunity to Cas9.7Charlesworth C.T. Deshpande P.S. Dever D.P. Camarena J. Lemgart V.T. Cromer M.K. Vakulskas C.A. Collingwood M.A. Zhang L. Bode N.M. et al.Identification of preexisting adaptive immunity to Cas9 proteins in humans.Nat. Med. 2019; 25: 249-254Crossref PubMed Scopus (421) Google Scholar,8Wagner D.L. Amini L. Wendering D.J. Burkhardt L.M. Akyuz L. Reinke P. Volk H.D. Schmueck-Henneresse M. High prevalence of Streptococcus pyogenes Cas9-reactive T cells within the adult human population.Nat. Med. 2019; 25: 242-248Crossref PubMed Scopus (195) Google Scholar The precise prevalence of anti-Cas9 immunity in humans is a matter of debate, as studies have used assays with different sensitivities, methods, reagents, and populations. Nonetheless, the problem is expected to be significant—rivalling that of pre-existing immunity to the AAV capsid. Even less is known about the consequences of Cas9 immunity in the setting of gene editing therapeutics. Studies in mouse models have shown that AAV-CRISPR delivery to skeletal muscle elicits both humoral and cellular immune responses to Cas9.9Chew W.L. Tabebordbar M. Cheng J.K. Mali P. Wu E.Y. Ng A.H. Zhu K. Wagers A.J. Church G.M. A multifunctional AAV-CRISPR-Cas9 and its host response.Nat. Methods. 2016; 13: 868-874Crossref PubMed Scopus (366) Google Scholar In the liver, pre-immunization of mice with Cas9 protein does not impede AAV-CRISPR editing, but results in an increase in CD8+ T cells and elimination of Cas9-expressing cells over time.10Li A. Tanner M.R. Lee C.M. Hurley A.E. De Giorgi M. Jarrett K.E. Davis T.H. Doerfler A.M. Bao G. Beeton C. Lagor W.R. AAV-CRISPR gene editing is negated by pre-existing immunity to Cas9.Mol. Ther. 2020; 28: 1432-1441Abstract Full Text Full Text PDF PubMed Scopus (83) Google Scholar In this new paper, Hakim et al. performed AAV-CRISPR gene editing in three different canine models of DMD: golden retriever, Welsh Corgi, and Labrador retriever. In contrast to mouse models, adult dogs had high levels of anti-Cas9 IgG at baseline, closely mimicking the human situation. Interestingly, newborn puppies lose maternally derived Cas9 antibodies between 2 and 6 weeks of age, prior to acquisition of their own immunity. To minimize the potential for an immune reaction to the therapy, the authors used the muscle-specific creatine kinase 8 (CK8) promoter to restrict expression, as well as high-dose prednisolone. Following intramuscular injection of AAV-CRISPR into the Labrador and golden retrievers, robust dystrophin rescue was achieved in the short term. However, the authors also saw abundant infiltration of CD4+ and CD8+ T cells, muscle cytokine elevation, increased anti-Cas9 serum antibodies, and Cas9-specific T cell responses indicated by interferon gamma (IFN-γ) ELI-spot assay on peripheral blood mononuclear cells. Further studies were performed to determine whether this immune response could eliminate dystrophin-positive muscle fibers, using co-injection of AAV encoding an alkaline phosphatase reporter. In Welsh Corgi dogs with confirmed pre-existing immunity, robust dystrophin correction was seen at 3 weeks, but was substantially lower at 6 weeks post-injection. The loss of dystrophin-positive fibers was associated with CD4+ and CD8+ T cell infiltration, reduced AAV genome copy numbers, and Cas9-specific antibody and T cell responses. Experiments using SERCA2a and microdystrophin transgenes convincingly ruled out responses to the AAV capsid or dystrophin epitopes. Hakim et al. went on to show that this immune response also occurs in numerous muscle depots of animals treated systemically with AAV-CRISPR through intravenous injection. The response was not unique to DMD, as it was also observed in normal dogs. Furthermore, the effects were specific to expression of Cas9, since it occurred in the absence of gRNA delivery and associated gene editing activity. Granzyme B+ T cells were found around dying muscle fibers, providing compelling evidence of T cell-mediated killing. Overall, this is an impressive publication addressing a problem of critical importance to DMD as well as other CRISPR-based gene therapies. Varied numbers of animals, ages, time points, and routes of administration between experiments require some patience from the reader. Although animal numbers are low, the experiments were comprehensive in assessing dystrophin correction, T cell infiltration, and Cas9-specific immune responses. Three different canine models of DMD were used as well as normal dogs, providing confidence the effects are not model specific. Rigor was ensured through serial sections of muscle near the injection site, blinded assays for anti-Cas9 antibodies and ELIspot, and reporting antibody titers in absolute units of μg/mL. Accordingly, this paper will serve as a valuable benchmark for future work. This study is significant because it is the first demonstration of a robust anti-Cas9 immune response following a CRISPR therapy in a large mammal. Although there have been hints from mice that immunity to Cas9 may limit the efficacy of genome editing, this concern had been mostly theoretical. Amoasii et al. previously reported efficient correction of dystrophin expression in the delta 50 canine model of DMD.6Amoasii L. Hildyard J.C.W. Li H. Sanchez-Ortiz E. Mireault A. Caballero D. Harron R. Stathopoulou T.R. Massey C. Shelton J.M. et al.Gene editing restores dystrophin expression in a canine model of Duchenne muscular dystrophy.Science. 2018; 362: 86-91Crossref PubMed Scopus (298) Google Scholar However, those studies were performed in puppies treated at 1 month of age and were not subjected to such a detailed long-term follow up. Hakim et al. clearly indicates that anti-Cas9 immunity poses a risk in a species where Cas9 immunity is acquired through environmental exposure. In addition, the authors present the most convincing data yet of T cell-mediated removal of Cas9-expressing fibers. Notably, the use of a tissue-specific promoter and broad immunosuppression with prednisolone did not prevent the cytotoxic T cell response. The effects of high-dose AAV-CRISPR in human skeletal muscle are hard to predict. While it is true that immunity in dogs and mice is very different from humans, problems identified in these models should not be ignored. There is a need for replication of this work, as well as experiments in other large animal models that might be more predictive, such as in non-human primates. Such studies must carefully survey and control for pre-existing immunity to Cas9 as well as the AAV capsid. In addition, the impact of the anti-Cas9 immune response on heart function is of tremendous importance to DMD, and requires special attention. In the near term, it seems likely that pre-screening for anti-Cas9 antibodies may be needed for inclusion in clinical trials with systemic delivery of AAV-CRISPR. This would also apply to therapies involving viral delivery of base editors and prime editors, which currently rely on the Cas9 protein for their DNA binding activity. Unfortunately, this could greatly narrow the pool of patients eligible for therapy who lack pre-existing immunity to both AAV and Cas9. Immunosuppression with prednisolone continues to be standard practice for gene therapy with AAV, but these results suggest that it may be insufficient to prevent loss of Cas9-expressing fibers. A key question is whether the temporal window of Cas9 protein expression can be managed to prevent this cytotoxic T cell response against the edited myofibers. Achieving this will likely require major advances in non-viral delivery technologies, and/or highly efficient self-inactivating systems. Dose is also an important consideration, which is already a daunting problem for muscle-directed AAV gene therapy. More efficient and specific vectors will certainly be required. Finally, delivery systems capable of efficient genome editing in satellite cells, the resident muscle stem cell population, are an important unmet need. An open question is whether immunomodulation can effectively manage this cytotoxic T cell response, to ensure the longevity of dystrophin-positive gene edited myofibers. This might be achieved through more targeted immunosuppression approaches, strategies to prevent antigen presentation, or tolerance promotion by regulatory T cells. While there have been some successful proof-of-concept studies in these areas, it should be noted that the underlying issue is environmental exposure to Cas9 and AAV. Although some epitopes may be more immunogenic than others, the antibody and T cell repertoire will vary dramatically between individuals. Preserving editing activity while mutating multiple Cas9 epitopes may prove too difficult. Ultimately, the best solution to these problems may not be Cas9 at all, but different proteins with similar activities. These could be Cas orthologs derived from bacteria or phages that humans do not normally encounter, or perhaps entirely synthetic proteins inspired by CRISPR/Cas9. The current work should not be viewed as a setback for gene editing of DMD and other muscle diseases. Rather, we should thank Hakim and colleagues for the valuable lesson, reminding us that such experiments in model organisms provide unexpected insights that improve our ability to treat disease and ultimately save patients’ lives. This paper should refocus our efforts on anti-Cas9 immunity and accelerate progress toward a cure. To accomplish this, we will need to understand the nature and pathophysiological consequences of these immune responses, and develop creative strategies to manage or circumvent them.

Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.