CRISPR-Cas9 Genome Editing: A New Era in Characterizing and Treating Hematologic Disease

Prokaryotic bacteria and archaea have evolved an adaptive immune defense mechanism whereby clustered regularly interspaced short palindromic repeats (CRISPR) associate with a Cas9 nuclease to degrade foreign nucleic acids under the guidance of short RNA (gRNA) sequences. In two independent seminal papers^1,2  the research groups of Dr. Feng Zhang and Dr. George Church demonstrated that the CRISPR/Cas9 system can be engineered to alter the human genome and that multiple gRNAs can be introduced to enable simultaneous editing of several target sites. A 20–base pair sequence at the 5′ end of the gRNA directs the CRISPR/Cas9 complex to a specific DNA sequence, where the Cas9 nuclease cleaves the target DNA and creates a double-strand break, which is repaired by endogenous eukaryotic mechanisms. Repair through nonhomologous end-joining (NHEJ) produces local insertions or deletions (indels) and disrupts the target gene, whereas homology-directed DNA repair (HDR) corrects the target gene by using a customized donor DNA template. Unlike other designer nucleases such as homing meganucleases, zinc finger nucleases (ZFNs), and transcription activator–like effector nucleases (TALENs), the CRISPR/Cas9 system is much easier to engineer since only the short gRNA has to be changed to detect a new target sequence. It is also more efficient, inexpensive, and provides a robust, remarkable tool that has revolutionized genome manipulation.

Sickle cell disease (SCD) and β thalassemia are monogenic disorders resulting from mutations in the β globin gene and represent prime targets for therapeutic genome editing. β globin gene correction through HDR in CD34+ hematopoietic stem and progenitor cells (HSPCs) using CRISPR/Cas9 and a wild-type gene donor sequence is relatively inefficient since these cells are quiescent, and the HDR pathway is restricted to the S and G2 phases of the cell cycle when sister chromatids are available as repair templates. NHEJ is the preferred repair pathway in HSPCs, but this may have a negative impact since the high rate of indel formation may disrupt β globin production or affect other regions of the gene. To overcome these hurdles in repairing the coding sequence of the faulty gene, an alternative strategy is to target a modifier of disease severity such as fetal hemoglobin (HbF), since clinical symptoms of SCD or β thalassemia are alleviated in patients who have co-inherited hereditary persistence of fetal hemoglobin (HPFH). This is a nonmalignant genetic condition caused by mutations that attenuate the switch from fetal γ globin to adult β globin, resulting in increased expression of HbF. Disrupting repressors of HbF, including the transcription factors LRF and BCL11A, or the BCL11A enhancer, by CRISPR/Cas9-mediated NHEJ, resulted in the induction of HbF. An elegant example³  of the therapeutic potential of CRISPR/Cas9 technology is provided by the ex vivo recapitulation of one of the HPFH mutations, a 13-nucleotide deletion from -102 to -114 within the gene promoter, in peripheral blood HSPCs from two healthy adults, leading to elevated levels of HbF. This positive response prompted the researchers to edit HSPCs from three patients with SCD, which increased HbF-expressing cells to 90 percent and markedly inhibited polymerization of HbS and the pathologic sickling of these cells under hypoxic conditions. To characterize potential additional mutations induced by the error-prone NHEJ process, deep sequencing of the region encompassing the predicted cleavage sites revealed a predominance of the targeted 13-nucleotide deletion, as well as 30 other on-target smaller indels. No indels in predicted off-target sites were observed. The therapeutic relevance of this exciting study of gene disruption is highlighted by a clinical trial⁴  in which the CCR5 gene in CD34+ T cells from HIV-infected patients was edited ex vivo by ZFN-mediated NHEJ, followed by reinfusion of these cells. This was the first clinical trial of an editing technology, and it demonstrated genetic efficacy and safety of the procedure.

Numerous new developments using the CRISPR/Cas9 system were reported at the 2016 ASH Annual Meeting, reflecting the importance of this technology and the rapid expansion of the field. Treatment of β hemoglobinopathies continued to focus on the reactivation of HbF synthesis by the targeted deletion of a 13.6-kb genomic region containing putative intergenic HbF silencers.⁵  Other hematologic diseases investigated included pyruvate kinase deficiency,⁶  and bleeding disorders such as deficiencies of Factor VIII or Factor IX. There is no single predominant mutation in these disorders, and thus, an innovative universal CRISPR/Cas9 gene-targeting approach was developed in a mouse model of hemophilia B.⁷  Proof of principle was demonstrated, and this paves the way for further development to apply this vector system to a majority of the patients with hemophilia B.

An interesting development in the field of acute myeloid leukemia (AML) was to remove CD33, an AML-associated antigen, from normal HSPCs by CRISPR/Cas9 gene editing, thus rendering them resistant to CAR T-cell CD33-targeted therapy.⁸  This represents a novel system that maximizes efficacy and minimizes toxicity. Future clinical implementation of this approach would involve the administration of CD33 knockout HSPCs as an allogeneic stem cell transplant in combination with anti-CD33 CAR T-cells in patients with AML.

Frameshift mutations are easier to repair than point mutations using template-free CRISPR/Cas9-mediated NHEJ.⁹  This was demonstrated by transducing hematopoietic cells with integrase-deficient lentiviral particles carrying endonucleases with gRNAs directed against specific mutations, which resulted in repair of up to 10 percent of cells with frameshift mutations.

In addition to therapeutic applications, the simplicity of programming the CRISPR/Cas9 system provides a new way to interrogate gene function on a genome-wide scale. Lentiviral delivery of a genome-scale CRISPR/Cas9 knockout library enables the identification of genes essential for cell viability, as well as those required for resistance (e.g., imatinib-sensitizing genes in chronic myeloid leukemia).¹⁰  The catalytic function of the Cas9 protein can be deactivated whilst retaining binding to DNA. This “dCas9” may be fused to repetitive peptide epitopes (SunTag) recruiting multiple copies of antibody-fused de novo DNA methyltranferase 3A that enables site-specific manipulation of DNA methylation to study the epigenome.¹¹ The speed and ease of CRISPR/Cas9 genome manipulation has also been used to generate cell models such as natural killer cell lymphoma models as well as transgenic animal models,¹²  where engineered human HSPCs were transplanted into immunodeficient mice to generate models of clonal hematopoiesis and malignancy.¹³ 

Despite these rapid and exciting advances, several critical aspects need to be addressed and optimized before this technology can be implemented in a clinical setting to treat patients. These include 1) harvesting of target cells, such as HSPCs, T cells, or induced pluripotent stem cells; 2) delivery of editing reagents; 3) choice of target genes; 4) efficacy of editing; 5) detection and monitoring of off-target effects to avoid oncogenic events and to ensure safety; and 6) ethical implications of this new technology.