Human TET2 bridges cancer and immunity

In this issue of Blood, Stremenova Spegarova et al report 3 patients with biallelic loss-of-function (LOF) TET2 mutations.¹  These patients suffered from infections, autoimmunity, and lymphoma, demonstrating 3 of the 5 potential phenotypes seen in inborn errors of immunity. TET2 encodes ten-eleven translocation methylcytosine dioxygenase 2 (TET2), 1 of the 3 members of the TET family of epigenetic regulators responsible for converting 5-methylcytosine to 5-hydroxymethylcytosine (5hmC) and subsequent oxidation products in an active DNA demethylation pathway. TET2 is ubiquitous, with particularly strong expression in hematopoietic cells. Somatic LOF TET2 mutations were first reported in patients with myeloproliferative disorders and hematologic cancers over 10 years ago.²  TET2 haploinsufficiency has also recently been reported in families with myeloid or lymphoid cancers. Intriguingly, some of these patients also presented signs of enhanced monocyte- and macrophage-mediated inflammatory responses, together with atherosclerotic plaque development, which has been associated with increases in NLRP3 inflammasome activation.^3,4  Thus, the tumor suppressor role of TET2 has been extensively documented, especially in myeloid lineages.

This elegant report of patients with autosomal-recessive complete TET2 deficiency not only confirms the role of TET2 as a tumor suppressor, but also highlights its function in immunity (see table). The infections observed in the patients included recurrent viral respiratory tract infections and persistent Epstein-Barr virus (EBV) viremia. The principal autoimmune manifestations were thrombocytopenia and anemia, accompanied by hepatosplenomegaly. The leukocyte subsets of these patients were reminiscent of the autoimmune lymphoproliferative syndrome seen in patients with inborn errors of the Fas pathway, with high levels of CD4⁻CD8⁻ double-negative T cells, low levels of T helper 17 (Th17), Th1, and follicular helper T (Tfh) cells, and low levels of class-switched memory B cells. Fas ligand–mediated apoptosis was also found to be impaired in 2 of the 3 patients. Nonhematopoietic clinical features, including developmental delay and hypothyroidism, were also reported.

The spectrum of hemato-immunological and clinical phenotypes in patients bearing myeloid LOF, germline monoallelic LOF, and germline biallelic LOF indicates the existence of a cell-type effect (from myeloid to lymphoid, from hematopoietic to nonhematopoietic) and a gene-dosage effect (from monoallelic to biallelic). The discovery of mono- and biallelic germline mutations also suggests that somatic TET2 mutations act as a “first hit” in some myeloid cancers. Moreover, that the same germline and somatic mutations are associated with the same type of cancer provides strong evidence for a crucial role of the mutated gene in tumorigenesis. A similar situation has been reported for germline NRAS (lymphoproliferative diseases), germline CARD11 (congenital B-cell lymphocytosis), and somatic CARD11 (diffuse large B-cell lymphoma) mutations.^5,6  Thus, these germline mutations define promising drug targets for the corresponding tumors.

That article neatly illustrates the rapid growth of the field of inborn errors of immunity in many unexpected directions, at least partly due to the unpredictable impact of mutations in housekeeping or highly specialized genes.⁷  In recent years, many new genetic defects have been shown to be caused by mutations of genes encoding proteins with basic biochemical functions that are essential to a broad range of cellular processes and are often ubiquitous. Examples include TPPII⁸  for amino acids, PGM3⁹  for glycosylation, and DBR1 for RNA lariat metabolism.¹⁰  Defects in these genes result in surprisingly narrow, and sometimes unique, phenotypes that could not be predicted from mouse models or biochemical studies.

In the case of TET2 deficiency, the preservation of most hematopoietic lineages in the patients, despite the profound increase in DNA methylation in the blood, is also surprising. These 3 patients, together with patients presenting haploinsufficiency for TET2, provide fascinating opportunities to study DNA methylation, particularly for 5hmC, in humans. Studies of the transcriptomes of different cell lineages from these patients will undoubtedly provide insight into methylation-dependent gene regulation. Stremenova Spegarova et al have already differentiated hematopoietic lineages from the patients’ induced pluripotent stem cells and have shown this differentiation to be correlated with DNA hypermethylation. More detailed studies of these cells might also reveal the genomic distribution of 5hmC, related gene activity, and their influence on hematopoietic lineages. They might also differentiate the function of TET2 from those of TET1 and TET3 while also identifying TET2-specific targets and TET2-dependent tissues. It is reasonable to expect that, in the near future, following the identification of a larger number of patients, studies of TET2 deficiency will reveal a number of new DNA methylation targets and important cellular functions associated with them.