NBN: protein instability, ALL susceptibility

In this issue of Blood, Escherich et al report that deleterious heterozygous germ line variants in the nibrin (NBN) gene result in a loss of protein stability and are associated with an increased risk of B-cell acute lymphoblastic leukemia (ALL) in children.¹ 

The role of germ line genetic variation in the development of childhood ALL is well established. Sequencing studies in familial ALL and in large numbers of sporadic cases have revealed the contribution of rare and pathogenic/likely pathogenic germ line variants in predisposition genes, including in the lymphoid transcription factor genes IKZF1, ETV6, and PAX5, and in the tumor suppressor gene TP53.^2-5 In addition, several autosomal recessive chromosome instability disorders, including Fanconi anemia, ataxia telangiectasia, Bloom syndrome, and Nijmegen breakage syndrome (NBS), are characterized by varying clinical manifestations as well as an increased risk of childhood ALL.⁶ 

NBS is caused by biallelic loss-of-function germ line variants in the DNA damage repair gene NBN, with most patients with NBS being of Eastern European origin and harboring the p.K219fs founder mutation. NBS is associated with a high risk of lymphoid malignancies, especially lymphomas, but also for ALL,^7,8 whereas heterozygous carriers of the p.K219fs variant, although clinically asymptomatic, are reported to have an increased risk of adult-onset cancers.⁹ Although putatively deleterious heterozygous germ line variants have been identified in a handful of childhood patients with ALL,¹⁰ the role of NBN variants in ALL susceptibility has not been comprehensively studied.

To address this knowledge gap, Escherich et al describe the detection and characterization of germ line NBN variants in a cohort of ≈4300 patients with B-cell ALL (see figure). They identified rare variants that were computationally predicted to be deleterious in 1.2% of patients with B-cell ALL, a frequency significantly higher than that found in population-based controls. Employing the consistent summary counts–based rare variant burden test (CoCoRV), which to some extent helps to account for ancestry differences between cases and controls (more on this later), they estimated that rare and predicted deleterious NBN variants confer a 1.8-fold risk of B-cell ALL. Several variants, including the NBS founder mutation p.K219fs, were individually enriched in patients with ALL compared with controls. No biallelic NBN variants were detected among the patients, albeit this is not entirely surprising given the rare nature of NBS.

Importantly, the authors went on to explore the functional effects of the 25 rare and predicted deleterious NBN variants identified in their patient cohort (see figure). First, they examined NBN protein stability in HEK293T cells, by knocking out the endogenous NBN gene and then introducing each variant of interest into the cell line model. Three protein truncating variants, including p.K219fs, led to a complete loss of protein stability, as expected. Of 22 missense variants identified, 4 resulted in complete loss of NBN protein expression and 6 led to partial loss of protein stability, with the remaining 12 variants showing mild or no loss of stability. Second, they investigated the sensitivity of cells harboring NBN variants to the radiomimetic drug mitomycin C (MMC). MMC sensitivity is an established clinical assay for the diagnosis of NBS because of the hypersensitivity of patient cells to irradiation. All 7 of the variants that resulted in a loss of protein stability also demonstrated a high sensitivity to MMC and were therefore classified as “nonfunctional.” On the basis of the results of both functional assays, an additional 7 missense variants were classified as “partially functional,” with the remaining 11 variants designated as “functional.” Of note, 13 of the 14 nonfunctional or partially functional variants were located in the N-terminal forkhead-associated and BRCA1 C-terminal (FHA-BRCT)-repeat domain of NBN, suggesting that variation in this region has a significant impact on protein function.

In examining the clinical implications of NBN variants, Escherich et al found no associations with patient minimal residual disease, nor with overall or event-free survival. Hyperdiploidy was enriched among patients with B-cell ALL with nonfunctional NBN variants, similar to the molecular subtype patterns found in patients with rare germ line variants in IKZF1 or ETV6.^2,3 And when considering only the 33 patients harboring nonfunctional or partially functional NBN variants, this corresponds to ≈0.8% of their patient cohort, a frequency similar to those reported for predisposition genes IKZF1, ETV6, and TP53 in patients with sporadic ALL.^2-4 

Altogether, this study provides strong evidence for the role of heterozygous deleterious germ line NBN variants in the development of B-cell ALL. It is important to highlight that the analyses were limited to B-cell ALL, and the association of germ line NBN variants with risk of T-cell ALL and other hematological malignancies should be assessed. Furthermore, results from their functional assays in NBN^–/– cells represent the effects of homozygous NBN variants, and additional studies are required to understand the effects of heterozygous variants and to decipher the precise mechanisms through which loss of NBN protein function may lead to ALL development. Finally, although a significant enrichment of NBN variants was found in patients with B-cell ALL compared with population-based controls using summary counts data, there are limitations to the approach used. For one, it does not account for variation in genetic ancestry that exists within population groups, which is of particular importance for admixed populations, such as African American/Black and Hispanic/Latino individuals. More accurate estimates of the effects of NBN variants on ALL risk would be obtained by sequencing large numbers of patients and controls from diverse populations, along with adjustment for genetic ancestry in association tests. Sequencing in larger sample sets may also reveal the impact of individual NBN variants on ALL risk, as these are likely to differ based on their functional effects (eg, nonfunctional vs partially functional variants).

Results to date suggest that each childhood ALL predisposition gene individually contributes a small piece of the susceptibility puzzle, and collectively account for only 5% to 10% of patients with sporadic ALL. There are likely additional predisposition genes waiting to be discovered, and investigation of germ line variants in the genes underlying other inherited chromosome instability disorders besides NBS is warranted.

Conflict-of-interest disclosure: A.J.d.S. declares no competing financial interests.