Epigenetic inheritance of DNA methylation in mammals requires a multifunctional protein UHRF1, which is believed to recruit DNMT1 to DNA replication forks through a unique hemi-methylated CpG-binding activity. Here we demonstrate that the UHRF1 mutants deficient in binding either hemi-methylated CpG or H3K9me2/3, but not both, are able to associate with pericentric heterochromatin, recruit Dnmt1 and partially rescue DNA methylation defects in mouse Uhrf1 null ES cells. Furthermore, we present evidence that the flip out of the methylated cytosine induced by UHRF1 binding is unlikely essential for subsequent DNA methylation by DNMT1. Together, our study demonstrates that UHRF1 can target DNMT1 for DNA maintenance methylation through binding either H3K9me2/3 or hemi-methylated CpG, and that the presence of both binding activities ensures high fidelity DNA maintenance methylation. In addition, our study indicates that UHRF1 mediates cross-talk between H3K9 methylation and DNA methylation at the level of DNA methylation maintenance.

Recent studies demonstrate that UHRF1 is required for DNA methylation maintenance by targeting DNMT1 to DNA replication foci, presumably through its unique hemi-methylated DNA-binding activity and interaction with DNMT1. UHRF2, another member of the UHRF family proteins, is highly similar to UHRF1 in both sequence and structure, raising questions about its role in DNA methylation. In this study, we demonstrate that, like UHRF1, UHRF2 also binds preferentially to methylated histone H3 lysine 9 (H3K9) through its conserved tudor domain and hemi-methylated DNA through the SET and Ring associated domain. Like UHRF1, UHRF2 is enriched in pericentric heterochromatin. The heterochromatin localization depends to large extent on its methylated H3K9-binding activity and to less extent on its methylated DNA-binding activity. Coimmunoprecipitation experiments demonstrate that both UHRF1 and UHRF2 interact with DNMT1, DNMT3a, DNMT3b and G9a. Despite all these conserved functions, we find that UHRF2 is not able to rescue the DNA methylation defect in Uhrf1 null mouse embryonic stem cells. This can be attributed to the inability for UHRF2 to recruit DNMT1 to replication foci during S phase of the cell cycle. Indeed, we find that while UHRF1 interacts with DNMT1 in an S phase-dependent manner in cells, UHRF2 does not. Thus, our study demonstrates that UHRF2 and UHRF1 are not functionally redundant in DNA methylation maintenance and reveals the cell-cycle-dependent interaction between UHRF1 and DNMT1 as a key regulatory mechanism targeting DNMT1 for DNA methylation.

Recently, chimeric antigen receptor (CAR)-T cell therapy has shown great promise in treating haematological malignancies1–7. However, CAR-T cell therapy currently has several limitations8–12. Here we successfully developed a two-in-one approach to generate non-viral, gene-specific targeted CAR-T cells through CRISPR–Cas9. Using the optimized protocol, we demonstrated feasibility in a preclinical study by inserting an anti-CD19 CAR cassette into the AAVS1 safe-harbour locus. Furthermore, an innovative type of anti-CD19 CAR-T cell with PD1 integration was developed and showed superior ability to eradicate tumour cells in xenograft models. In adoptive therapy for relapsed/refractory aggressive B cell non-Hodgkin lymphoma (ClinicalTrials.gov, NCT04213469), we observed a high rate (87.5%) of complete remission and durable responses without serious adverse events in eight patients. Notably, these enhanced CAR-T cells were effective even at a low infusion dose and with a low percentage of CAR+ cells. Single-cell analysis showed that the electroporation method resulted in a high percentage of memory T cells in infusion products, and PD1 interference enhanced anti-tumour immune functions, further validating the advantages of non-viral, PD1-integrated CAR-T cells. Collectively, our results demonstrate the high safety and efficacy of non-viral, gene-specific integrated CAR-T cells, thus providing an innovative technology for CAR-T cell therapy.

The generation of induced pluripotent stem cells (iPSCs) from differentiated somatic cells by over-expression of several transcription factors has the potential to cure many genetic and degenerative diseases currently recalcitrant to traditional clinical approaches. One such genetic disease is β-thalassemia major (Cooley's anemia). This disease is caused by either a point mutation or the deletion of several nucleotides in the β-globin gene, and it threatens the lives of millions of people in China. In the present study, we successfully generated iPSCs from fibroblasts collected from a 2-year-old patient who was diagnosed with a homozygous 41/42 deletion in his β-globin gene. More importantly, we successfully corrected this genetic mutation in the β-thalassemia iPSCs by homologous recombination. Furthermore, transplantation of the genetically corrected iPSCs-derived hematopoietic progenitors into sub-lethally irradiated immune deficient SCID mice showed improved hemoglobin production compared with the uncorrected iPSCs. Moreover, the generation of human β-globin could be detected in the mice transplanted with corrected iPSCs-derived hematopietic progenitors. Our study provides strong evidence that iPSCs generated from a patient with a genetic disease can be corrected by homologous recombination and that the corrected iPSCs have potential clinical uses.

In mammals it is unclear if UHRF1-mediated DNA maintenance methylation by DNMT1 is strictly dependent on histone H3K9 methylation. Here we have generated an Uhrf1 knockin (KI) mouse model that specifically abolishes the H3K9me2/3-binding activity of Uhrf1. The homozygous Uhrf1 KI mice are viable and fertile, and exhibit ∼10% reduction of DNA methylation in various tissues. The reduced DNA methylation occurs globally in the genome and does not restrict only to the H3K9me2/3 enriched repetitive sequences. In vitro UHRF1 binds with higher affinity to reconstituted nucleosome with hemi-methylated CpGs than that with H3K9me2/3, although it binds cooperatively to nucleosome with both modifications. We also show that the nucleosome positioning affects the binding of methylated DNA by UHRF1. Thus, while our study supports a role for H3K9 methylation in promoting DNA methylation, it demonstrates for the first time that DNA maintenance methylation in mammals is largely independent of H3K9 methylation.