Researchers publishing in Cell Reports Medicine have taken a look at what causes CRISPR/Cas9 gene-editing technology to drive cells senescent and investigated a potential method of preventing it.
Breaking and repairing DNA
Genetic engineering of living cells through CRISPR/Cas9 technology requires three things to occur. The DNA must be disconnected in a double-strand break, new DNA must be inserted, and the cell must then close the DNA with the inserted gene inside [1].
Beyond any concerns with the DNA being inserted, the very process of breakage and repair raises issues. Breaking apart DNA leads to a DNA damage response, part of which involves p53, a factor that encourages cellular senescence [2]. The viral vectors that insert DNA, such as lentiviruses and the commonly used AAV6, lead to increased p53 as well [3]. Previous work has found that temporarily inhibiting p53 may allow these cells to proliferate long enough to be effective [2].
Successful engineering leads to increased senescence
For their first experiment, the researchers electroporated and targeted human hematopoietic stem and progenitor cells (HSPCs) with a variety of edits: a guide RNA, a single-strand oligodeoxynucleotide, and AAV6-driven gene modifications adding green fluorescent protein (GFP) at two separate loci, AAVS1 and IL-2RG. There was also an electroporation-only control group.
Compared to the other groups, the two AAV6 gene modifications had much stronger inflammatory reactions, including upregulation of interleukins, p16, CDKN1A, and the crucial senescence biomarker SA-β-gal. AAV6 was found to be more effective in genetically engineering cells when targeting the IL-2RG locus compared to the AAVS1 locus; however, targeting this locus led to more inflammation in most metrics as well. Even four days after the modification, the engineered cells were still showing persistent signs of DNA damage and inflammation.
Higher doses of AAV6 led to more efficient engineering, as measured by comparisons of homology-directed repair (HDR) versus non-homologous end joining (NHEJ). The former is less prone to genetic errors than the latter.
However, cells that were successfully engineered, as measured by the presence of GFP, were much more likely to become senescent than control groups. These effects continued even as the cells clonally expanded, with some effects being seen even in cells that were exposed to AAV6 but did not express GFP.
These findings were recapitulated in mice. These human cells were grafted into immunodeficient animals and allowed to proliferate. Cells that were exposed to higher doses of AAV6 grew more slowly than cells exposed to lower doses, which were statistically no different from the control group. These effects were directly related to cellular senescence, and cells that had successfully expressed the GFP protein were more likely to become senescent. These altered stem cells also produced different results: engineered HSPCs produced more B cells and fewer T cells than the control group.
The researchers made sure that these results were not due to GFP in particular; similar results were found when an entirely different nerve growth factor reporter was used.
A potential solution
The researchers decided to combat this inflammatory and senescence-promoting effect by administering anakinra, a direct antagonist of the IL-1 cytokine, alongside AAV6. This significantly reduced the number of successfully engineered cells that exhibited signs of senescence. Targeting NF-κB, another inflammatory factor, with SC514 yielded similar results, as did targeting p53 with GSE56. Critically, none of these treatments, administered separately or in combination, seemed to affect the efficiency of the genetic engineering.
These findings, too, were recapitulated in mice. Directly targeting these sources of inflammation and senescence led to more cellular proliferation without affecting engineering efficiency, thus leading to significantly more GFP cells being found in the mice when the graft was placed in the spleen. There was a trend in this direction in the bone marrow, with treated engineered cells being able to form significantly more colonies in this area than untreated engineered cells.
Interestingly, however, these treatments had different effects on the mutation rate. SC514 and GSE56 both increased mutation rate, raising the possibility of genotoxicity; Anakinra, on the other hand, lowered it, decreasing both unwanted deletions and the number of micronuclei in addition to less mutational burden overall, including mutations of cancer-related genes.
While clinical trials involving genetically engineered cells pretreated with anakinra are required to ensure the efficacy of this approach in people, these findings strongly suggest that such pretreatment may quickly become standard procedure in generating modified cells destined for engraftment. The DNA damage response and related senescence are unwanted side effects of genetic engineering, but this study demonstrates that they are side effects that can be mitigated.
Literature
[1] Salsman, J., & Dellaire, G. (2017). Precision genome editing in the CRISPR era. Biochemistry and cell biology, 95(2), 187-201.
[2] Schiroli, G., Conti, A., Ferrari, S., Della Volpe, L., Jacob, A., Albano, L., … & Di Micco, R. (2019). Precise gene editing preserves hematopoietic stem cell function following transient p53-mediated DNA damage response. Cell stem cell, 24(4), 551-565.
[3] Piras, F., Riba, M., Petrillo, C., Lazarevic, D., Cuccovillo, I., Bartolaccini, S., … & Kajaste‐Rudnitski, A. (2017). Lentiviral vectors escape innate sensing but trigger p53 in human hematopoietic stem and progenitor cells. EMBO molecular medicine, 9(9), 1198-1211.
