For decades, biologists using model organisms such as mice and fruit flies have faced concerns about the relevance of their findings to humans. Using a model that is more evolutionarily similar to humans, such as another primate, could potentially close this frustrating gap. In a paper online in Nature, Zhang et al.1 use CRISPR–Cas9 gene-editing techniques to generate macaque monkeys lacking the gene SIRT6. Strikingly, they show that the SIRT6 protein has a role in embryonic development in macaques that was not previously uncovered in mice.
Mammalian SIRT6 removes acetyl groups from histone proteins. DNA is packaged around histones in the nucleus, and this deacetylation condenses the packaged DNA, suppressing gene expression2. In mice, SIRT6 is known to be a longevity protein that regulates many factors that alter during ageing, including genome stability, inflammation and metabolism2. Indeed, overexpression of SIRT6 in male mice leads to health improvements and extends lifespan3, whereas SIRT6-deficient mice die a few weeks after birth, displaying features of premature ageing4.
It is unknown whether SIRT6 is involved in longevity in humans. However, recent data show5 that an inactivating mutation in human SIRT6 causes overexpression of embryonic stem-cell genes, which leads to abnormal development and severe brain defects, resulting in embryonic death. These findings suggest a previously unappreciated role for SIRT6 in embryonic development, which should be considered separately from its role in ageing.
Zhang and colleagues used CRISPR–Cas9 to create one male and three female macaque embryos that did not express SIRT6. The females died shortly after birth and the male died in the middle of gestation. The absence of SIRT6 caused severe, whole-body developmental delays. Compared with wild-type newborns, the mutants showed lower bone density, lower levels of subcutaneous fat and immature intestines and skeletal muscle. The authors also found that SIRT6-deficient monkeys had smaller brains owing to delayed neuronal maturation and an increase in the number of immature neural progenitor cells. Overall, the SIRT6-mutant animals were born much smaller than controls and showed gene-expression and morphological profiles closer to those of a typical three-month-old fetus than a full-term animal born after six months of gestation (Fig. 1).
Because of the known role of SIRT6 in suppressing gene expression2, Zhang et al. examined changes in gene expression in the mutants. Among the most upregulated genes was H19, which encodes a long non-coding RNA that is known to regulate fetal growth6. H19 expression levels were increased in all tissues examined, with the highest expression in the brain.
Next, the authors used a different gene-editing approach to generate human neural progenitor cells lacking SIRT6 in vitro, and showed that the differentiation of these cells into neurons was delayed when compared with wild-type cells. This defect was accompanied by higher levels of H19 RNA. Finally, the group found that SIRT6 removes acetyl groups associated with H19 transcription, and showed that reducing H19 expression in human cells lacking SIRT6 resolved their defects in neuronal differentiation. Thus, SIRT6 inhibits H19 expression to modulate neuronal development in human cells, as in monkeys.
Several avenues for further work arise from these results. For instance, the absence of SIRT6 altered the expression of thousands of genes in various tissues, and it is unlikely that H19 is the only gene responsible for the defects observed. Indeed, a human developmental disorder called Silver–Russell syndrome can be caused by increased H19 levels but, in contrast to SIRT6-deficient monkeys, people who have this disorder have normal lifespans and less-severe developmental changes6. This discrepancy suggests that SIRT6-modulated genes other than H19 also contribute to the severe effects seen in the authors’ mutant monkeys. It will be hard to pinpoint the precise genes that cause developmental defects in SIRT6-deficient animals, but this should be investigated in the future.
From an evolutionary point of view, SIRT6 is fascinating. In all mammals studied, the gene’s deletion causes premature death, and the protein has the same enzymatic activity and involvement in glucose metabolism and stem-cell differentiation7. However, as we climb the evolutionary ladder from mice to monkeys to humans, some of the traits caused by SIRT6 deletion become progressively more severe. SIRT6-deficient mice die a few weeks to months after birth8, whereas monkeys die within hours, and humans harbouring a SIRT6-inactivating mutation are not even born. This increasing severity could be explained by the acquisition of regulatory roles for SIRT6 over the course of evolution. In support of this idea, the severe brain defects seen in SIRT6-deficient primates have not been reported in mice, and this change correlates well with differences in brain complexity in these species. It will be extremely interesting to further explore the source of this trait enhancement across evolution.
What can we learn about the role of SIRT6 in human ageing from this primate model? At first glance, there is not an obvious connection between the developmental defects seen in the monkeys and ageing, as they are at opposite ends of life’s timeline. However, key pathways regulated by SIRT6 are conserved between these species, and genome-wide association studies have found a correlation between SIRT6 and increased lifespan in humans9. These facts, together with data indicating that SIRT6 helps to protect the brain against ageing-related disorders such as Alzheimer’s disease10, strongly suggest that the versatile SIRT6 protein might promote healthy longevity in humans. In the future, developments in CRISPR engineering might enable gene editing in specific tissues, and at chosen time points; if the latter were achieved, it would be fascinating to characterize the role of SIRT6 in primate lifespan.
More generally, genome editing is an exciting future strategy for human therapy. However, the challenge is to induce the desired edits without creating non-specific mutations or producing mosaic embryos in which only some cells express the edited gene. Promisingly, Zhang and colleagues found no mosaicism or detectable off-target mutations in their mutant animals, and another group that have used CRISPR in monkeys also report no off-target effects11. Although there are still many ethical and technical caveats to be considered, the authors’ achievement — along with a similar success in human embryos12 — gives hope that human genetic therapies using CRISPR engineering will be possible in the future.
Nature 560, 559-560 (2018)