Abstract
Background and hypothesis: Childhood pre-B acute lymphoblastic leukemia (ALL) can frequently be retraced to a pre-leukemic clone carrying a prenatally acquired genetic lesion (e.g. ETV6-RUNX1gene rearrangement). After birth, pre-leukemic clones can acquire secondary mutations and, hence, evolve towards overt leukemia. While this concept is well established, the mechanism(s) driving clonal evolution are not known. Epidemiological findings hint to a role of delayed childhood infections and chronic inflammation as etiologic factors of childhood ALL, but do not illuminate mechanism of clonal evolution of pre-leukemic cells. In this study, we demonstrate that cooperation between the AID cytosine deaminase and the RAG1/RAG2 V(D)J recombinase promotes acquisition of secondary genetic lesions that promote progress of pre-leukemic B cell precursors towards full-blown leukemia.
Results: The enzymatic activity of RAG1/RAG2 (VDJ recombination) and AID (somatic hypermutation, class-switch recombination) are strictly segregated to early and late stages of B cell development, respectively. While RAG1 and RAG2 are actively expressed at stages of early B cell development (bone marrow and fetal liver) that give rise to pre-B ALL, little is known about the function of AID in early B-lymphopoesis. As the involvement of AID in pre-B leukemic clonal evolution is incumbent on its expression during early stages of B-lymphopoesis, we tested CD19+ pre-B cells isolated from human bone marrow (BM) for indicators of AID activity, namely, somatic hypermutation (SHM) and class switch recombination (CSR). Interestingly, most pre-B cell clones carry rearranged Ig VH region genes that are mutated at low levels (average mutation frequency 26 x 10-3 bp). Likewise, pre-B cells isolated from fetal liver tissues (three donors; 10-19 weeks of gestation) carried Ig VH region genes mutated at low levels (average mutation frequency 14 x 10-3 bp). In addition, about one third of fetal liver pre-B cells had undergone CSR to Cγ3, Cγ1 and Cα regions. These findings highlight the previously unknown function of AID in two important sites of early human B-lymphopoesis. Based on these results, we hypothesized that a specific B cell subset during early pro- and pre-B cell differentiation can concomitantly express both AID and the RAGs and, hence, would be particularly susceptible to clonal evolution of cells that carry a pre-leukemic lesion.
Our subsequent studies identified late pre-B cells (Fraction D) as a natural subset of increased genetic vulnerability. Late pre-B cells downregulate IL7 receptor/Stat5 signaling, which enables expression of RAG1 and RAG2 and immunoglobulin light chain gene rearrangement. Loss of IL7 receptor/Stat5 signaling also removes an important safeguard against premature expression of AID. Therefore, late pre-B cells are poised to express AID at high levels in response to inflammatory stimuli (e.g. LPS) in concurrence with RAG1 and RAG2. Studying clonal evolution of patient-derived pre-B ALL cells, we found evidence for concomitant AID and RAG1/RAG2 activity. Further studying a genetic mouse model for pre-leukemic pre-B cells carrying ETV6-RUNX1, we found that repeated exposure to LPS can cause overt leukemia but not in the absence of either AID or Rag1. Additionally, whole exome sequencing of human B cell clones that were engineered to express AID, RAG1/RAG2 alone or in combination revealed that concurrent expression of AID with RAG1/RAG2 dramatically increased the frequency of structural chromosomal lesions.
Conclusion: Consistent with epidemiological findings on the etiology of childhood ALL, we conclude that reduced cytokine signaling (here, IL7R) in late pre-B cells renders pre-leukemic clones distinctively vulnerable to genetic lesions that can be acquired in the context of repeated exposure to inflammatory stimuli (e.g. chronic and recurrent infections during childhood). Our results support a role for AID and RAGs cooperation for the generation of secondary lesions in leukemia subgroups that require additional leukemogenic events, and therefore, provide the genetic and molecular basis to support the Delayed Infections Hypothesis for leukemia progression in children.
Disclosures
No relevant conflicts of interest to declare.
Driver mutations in Janus kinases in a mouse model of B-cell leukemia induced by deletion of PU.1 and Spi-B