Prevalence and functional analysis of sequence variants in the ATR checkpoint mediator Claspin

Abstract

Mutational inactivation of genes controlling the DNA-damage response contributes to cancer susceptibility within families and within the general population as well as to sporadic tumorigenesis. Claspin (CLSPN) encodes a recently recognized mediator protein essential for the ATR and CHK1-dependent checkpoint elicited by replicative stress or the presence of ssDNA. Here, we describe a study to determine whether mutational disruption of CLSPN contributes to cancer susceptibility and sporadic tumorigenesis. We resequenced CLSPN from the germline of selected cancer families with a history of breast cancer (n = 25) or a multicancer phenotype (n = 46) as well as from a panel of sporadic cancer cell lines (n = 52) derived from a variety of tumor types. Eight nonsynonymous variants, including a recurrent mutation, were identified from the germline of two cancer-prone individuals and five cancer cell lines of breast, ovarian, and hematopoietic origin. None of the variants was present within population controls. In contrast, mutations were rare within genes encoding the CLSPN-interacting protein ATR and its binding partner ATRIP. One variant of CLSPN, encoding the I783S missense mutation, was defective in its ability to mediate CHK1 phosphorylation following DNA damage and was unable to rescue sensitivity to replicative stress in CLSPN-depleted cells. Taken together, these observations raise the possibility that CLSPN may encode a component of the DNA-damage response pathway that is targeted by mutations in human cancers, suggesting the need for larger population-based studies to investigate whether CLSPN variants contribute to cancer susceptibility.

Figure 1. Somatic and germline variants of CLSPN

A, Nucleotide sequence traces depicting missense mutations (arrows) within the breast cancer cell-line MDA-MB-175 (P956L), the leukemia cell-line MOLT4 (R1184W), and the ovarian cancer cell-line ES-2 (I783S). Wildtype sequences from control individuals are shown for comparison (upper panels). B, Schematic representation of the CLSPN protein, indicating the location of variants in relation to known functional domains. βTrCP degron, βTrCP recognition motif DSGxxS; NLS, potential nuclear localization signals; CHK1-BD1, -BD2, -BD3, CHK1 binding domains 1–3.

Figure 2. The CLSPN-I783S variant encodes a functionally defective protein

A, siRNA-mediated knockdown of endogenous CLSPN in U2OS cells is associated with reduced levels of phosphorylated CHK1. U2OS cells transfected with control (con) siRNA, or CLSPN-3'UTR-specific siRNA were untreated or treated with IR (10Gy), UV (10J/m²) or HU (2mM). Cells were harvested 1h post-treatment followed by immunoblotting to measure CLSPN, phospho-CHK1(Ser345), and total CHK1 protein levels. B, In vivo phosphorylation of CLSPN-I783S, following exogenous expression in U2OS cells, is not impaired. U2OS cells infected by adenoviral constructs expressing myc-tagged wildtype (WT) or mutant (I783S) CLSPN were either treated with HU (2mM), or left untreated. Western blots of λ-protein phosphatase (λ-ppase)-treated or -untreated lysates were probed with an anti-CLSPN antibody to measure total levels of CLSPN expression, or with an anti-MYC antibody to measure ectopic CLSPN levels. C, In vivo phosphorylation of CHK1 by CLSPN-I783S, is impaired. U2OS cells were cotransfected with CLSPN-3'UTR-specific siRNAs to knock down the endogenous protein, as well as with constructs exogenously expressing wildtype or I783S constructs lacking a 3’-UTR. Cells were divided into two plates and treated with 10J/M² UV, or left untreated. Cells were harvested 1hr post-treatment, and subjected to immunoblotting with antibodies against CLSPN, phospho-CHK1 (Ser345), or total CHK1. Protein levels were quantified by phosphor imaging and the amount of phosphorylated CHK1, relative to total CHK1, determined (lower panel). D, Wildtype CLSPN, but not CLSPN-I783S, partially rescues the HU-sensitivity of CLSPN-depleted U2OS cells. CLSPN-depleted U2OS cells were co-transfected with exogenous DNA constructs expressing wildtype CLSPN or CLSPN-I783S. Transfected cells were plated at low density and treated with 2 mM HU for 24h. After two weeks, the number of colonies was determined by counting. Experiments were performed in triplicate and representative results, indicating the number of cells surviving HU treatment relative to untreated cells, are shown.