Inflammatory cytokines promote clonal hematopoiesis with specific mutations in ulcerative colitis patients

Abstract

Epidemiological sequencing studies have revealed that somatic mutations characteristic of myeloid neoplasms can be detected in the blood of asymptomatic individuals decades prior to presentation of any clinical symptoms. This premalignant condition is known as clonal hematopoiesis of indeterminate potential (CHIP). Despite the fact these mutant clones become readily detectable in the blood of elderly individuals (∼10% of people over the age of 65), the overall rate of disease progression remains relatively low. Thus, in addition to genetic mutations, there are likely environmental factors that contribute to clonal evolution in people with CHIP. One environmental stress that increases with age is inflammation. Although chronic inflammation is detrimental to the long-term function of normal hematopoietic stem cells, several recent studies in animal models have indicated hematopoietic stem cells with CHIP mutations may be resistant to these deleterious effects. However, direct evidence indicating a correlation between increased inflammation and accelerated CHIP in humans is currently lacking. In this study, we sequenced the peripheral blood cells of a cohort of patients with ulcerative colitis, an autoimmune disease characterized by increased levels of pro-inflammatory cytokines. This analysis revealed that the inflammatory environment of ulcerative colitis promoted CHIP with a distinct mutational spectrum, notably positive selection of clones with DNMT3A and PPM1D mutations. We also show a specific association between elevated levels of serum interferon gamma and DNMT3A mutations. These data add to our understanding of how cell extrinsic factors select for clones with specific mutations to promote clonal hematopoiesis.

Figure 1:. Clonal hematopoiesis in UC patients.

(A) Number of CHIP-c variants identified in UC patients (left), and categories of those SNVs (right). (B) Number of CHIP-hs variants identified in UC patients (left), and categories of those SNVs (right). (C) Number of CHIP-c mutations identified per gene. (D) Number of CHIP-hs mutations identified per gene. (E) CHIP incidence by age in referenced studies. Error bars represent the 95% binomial confidence interval of the point estimate in UC patients. The conference interval bands for referenced studies represent the 95% confidence interval around the point estimate. (F) CHIP-hs incidence by gender in UC patients over age brackets. Grey bars denote the CHIP-hs incidence in given age bracket. (G) Incidence of CHIP-c mutations by gene across indicated studies. Significance is indicated for incidence of CHIP-c mutations in given gene in UC patients compared to aggregate incidence across other reference studies.

Figure 2:. CHIP Dynamics in UC Patients.

(A) Correlation of clonal hematopoiesis in UC patients with clinical variables and treatment history. (B) Tukey plot showing ages of CH− and CH+ UC patients. *p<0.05, two-tailed t-test. (C) Tukey plot showing time since diagnosis of CH− and CH+ UC patients. (D) Tukey plots showing blood count parameters of CH− and CH+ UC patients. (E) VAF of CH variants in UC patients. Each dot represents a single mutation, line represents mean VAF. (F) Mutational profile of CH variants in UC patients. (G) Mutational spectrum of somatic variants identified in DNMT3A and PPM1D in UC patients. (H) TNFα and IFNγ levels in serum of UC samples, and clone size (by VAF) in patients surveyed in this experiment. Each dot indicates the levels of a given cytokine in the serum of one patient. Graphs represent mean ± S.E.M. *p<0.05 one-way ANOVA with Bonferroni multiple test correction.