Efficient nonviral T-cell engineering: CRISPR takes a giant step towards the clinic

The promise of CRISPR for T-cell engineering

For many years, researchers have explored the possibility of genetically reprogramming T cells for clinical applications such as cancer immunotherapy, the correction of genetic disorders, and conferring resistance to pathogens (Figure 1). While the development of CRISPR/Cas9 genome-editing technology has provided a powerful new method for T-cell engineering, clinical application of this technology has been limited by a reliance on viral delivery of templates for homology-directed repair (HDR) and an associated risk of random genomic integration, due in part to the toxicity associated with electroporation of double-stranded DNA (dsDNA).

In a groundbreaking study recently published in the journal Nature, Prof. Alexander Marson and colleagues describe a highly efficient method for T-cell engineering that circumvents viral delivery and minimizes cellular toxicity by employing coelectroporation of Cas9-gRNA ribonucleoprotein (RNP) complexes in tandem with dsDNA or single-stranded DNA (ssDNA) HDR templates. Using this approach, the authors quickly corrected inherited autoimmune-associated mutations in T cells obtained from siblings suffering from a rare monogenic disorder, and successfully reprogrammed T cells from healthy donors to target cancer cells in in vitro and in vivo models.

Figure 1. Workflow for T-cell-based therapies.

Overcoming the technical limitations

The authors’ approach was based on an established method for obtaining site-specific insertions (gene knockins), in which Cas9-gRNA-mediated formation of double-stranded breaks is followed by homology-directed repair (HDR) involving a DNA template containing the desired insert sequence (Figure 2). Whereas previous efforts at reprogramming T cells in this manner relied on viral delivery of the HDR template due to the toxicity associated with electroporation of dsDNA, Roth et al., discovered that toxicity can be significantly reduced by electroporating HDR templates in combination with Cas9-gRNA RNP complexes. In comparing the performance of dsDNA and ssDNA HDR templates, the authors demonstrated that application of ssDNA resulted in markedly higher cell viability when templates were added at higher doses, and that it reduced the occurrence of off-target and erroneous insertions by 100-fold and 10-fold, respectively, relative to the results observed for dsDNA.

Similar to what has been described with viral HDR templates, we found evidence to suggest that double-stranded templates could integrate independent of target homology, albeit at low rates. These rare events could be reduced almost completely by using single-stranded DNA (ssDNA) templates.”

—Roth et al. 2018

Figure 2. T-cell engineering workflow and applications. Top. The method developed by Roth et al., 2018 involves electroporation of Cas9-gRNA RNPs in tandem with linear dsDNA or ssDNA, which provides a template for homology-directed repair (HDR) following site-specific Cas9-mediated cleavage. Bottom. The authors employed their method to insert fluorescent protein tags, to engineer single-nucleotide substitutions, and to replace entire gene sequences.

Using the insertion of fluorescent protein sequences at various loci to demonstrate performance, the authors achieved insertion efficiencies greater than 50%, as well as biallelic insertions and multiplex insertions at up to three loci simultaneously (Table I and Figure 3). Further analyses indicated that insertion of fluorescent protein sequences in this manner did not alter the localization, function, or regulation of the resulting fusion proteins.

Figure 3. Biallelic and multiplexed knockin of fluorescent protein sequences. Using their method, the authors were able to simultaneously insert GFP and mCherry sequences into separate alleles of RAB11A (Panel A), and to insert fluorescent protein sequences at three different loci in parallel (Panel B).

Making strides for clinical applications

To demonstrate the therapeutic potential of their method, the authors applied it to rapidly correct inherited mutations in two alleles of the gene encoding the IL-2α receptor (IL2RA) in T cells obtained from three siblings afflicted with a rare monogenic autoimmune disease. In each case, the restoration of IL-2α cell-surface expression was observed in a sizable proportion of target cells following electroporation of RNPs combined with corresponding HDR templates, and subsequent analyses confirmed the establishment of IL-2α-mediated signaling. The authors have stated that they ultimately intend to administer the engineered T cells to the children with the aim of curing their disease.

In addition to fine-scale correction of pathogenic mutations, the authors employed their method to generate large-scale insertions relevant to cancer immunotherapy applications. Towards this end, they designed an HDR template that resulted in the replacement of an endogenous TCR-α sequence with paired TCR-β and TCR-α sequences encoding a T-cell receptor (TCR) recognizing the NY-ESO-1 tumor antigen. Application of the engineered T cells to NY-ESO-1⁺ melanoma cells in vitro resulted in robust T-cell activation and clearance of cancer cells at rates comparable to those observed for T cells engineered via retroviral delivery. Importantly, by electroporating millions of T cells from six healthy donors, the researchers could obtain suitable quantities for adoptive cell therapy. The engineered T cells were then transferred to a melanoma mouse model and demonstrated that they performed comparably to lentiviral-engineered T cells in localizing to and proliferating at tumor sites and inhibiting tumor growth.

Figure 4. Reprogramming T-cell antigen specificity to target cancer cells. Using their method, the authors swapped in sequences encoding a TCR specific to the NY-ESO-1 antigen at the TCRa locus. The authors then demonstrated that the engineered T cells successfully targeted NY-ESO-1⁺ cancer cells in vitro and in a mouse model.

Looking forward

In developing a nonviral approach to T-cell reprogramming, the authors have not only devised a solution that is more amenable to clinical application, they have also shortened the timeline for T-cell engineering projects from years or months to a few weeks, dramatically reducing costs, accelerating the pace of discovery, and expanding what is possible. In this regard, the countless hours that were spent optimizing Cas9:HDR template ratios and electroporation conditions will likely benefit the research community for many years to come.

Reference

Roth, T.L et al., Reprogramming human T cell function and specificity with non-viral genome targeting. Nat. Lett. 559, 405–409 (2018).