Statistics from Altmetric.com
We read with interest the work by Yan et al published in Gut.1 The serrated (hyperplastic) polyposis syndrome (SPS) is a heterogeneous disease defined by the presence of multiple serrated polyps throughout the colon,2 causing an increased risk (16%) of colorectal cancer (CRC).3 By performing whole-exome sequencing in 20 SPS families, Gala et al identified a germline mutation, c.338C>A (p.R113*), in the RING-type E3 ubiquitin ligase RNF43, an inhibitor of the Wnt pathway, in two independent families.4 The c.394C>T (p.R132*) mutation was subsequently identified in two SPS-affected members of one family, supporting a causal role in SPS.5 In the study by Yan et al whole-exome sequence analysis of four SPS families identified a deleterious germline mutation, c.953–1G>A (p.E318fs), in six members of one family, five fulfilling the WHO criteria for SPS.1 Buchanan et al assessed the mutation status of RNF43 in 74 selected SPS families, identifying two rare missense variants, c.443C>G (p.A148G) and c.640C>G (p.L214V), predicted deleterious by in silico algorithms, in two families. No carriers of p.R113* or p.R132* were detected in 221 additional patients with SPS.6
Mutation screening of RNF43 was carried out by Sanger sequencing in 96 unrelated patients with serrated polyposis, all of them recruited at the Hereditary Cancer Program of the Catalan Institute of Oncology (Spain) from 1998 to 2017. Informed consent was obtained from all subjects (IDIBELL Ethics Committee PR193/15). Two rare (population minor allele frequency (MAF) <1%) variants, the reported c.394C>T (p.R132*) and c.1821G>A (p.S607=), were identified. The c.1821G>A variant (MAFExome Aggregation Consortium (ExAc)=0.03%) was identified in a man diagnosed with CRC, three serrated polyps and one adenoma at age 24, and in his sister aged 30 years, who recently underwent polypectomy of 17 hyperplastic polyps. However, no defects in RNA processing (exon 7–9 amplification) were identified in the proband’s lymphocytes cultured in presence and absence of puromycin, suggesting no deleterious effect. The c.394C>T (p.R132*) mutation was detected in a woman diagnosed with CRC and >50 polyps (including serrated lesions) at age 55, but not in the proband’s children or her sister aged 69 years, who underwent removal of six hyperplastic polyps and one tubular adenoma at age 59. This mutation was previously identified in an SPS family5 and not reported as germline variant in aggregation databases (ExAC and Genome Aggregation Database). A somatic splice-site mutation in RNF43, c.2309–1G>A, was identified in the CRC tested, in line with the somatic inactivation of the RNF43 wildtype allele previously reported (table 1). The proband’s tumour had the CpG island methylator phenotype (CIMP), a common feature of the serrated pathway, in a mismatch repair proficient context. To date, CIMP has been observed in >75% of the RNF43 syndrome-associated lesions (table 1).
The c.394C>T (p.R132*) mutation and the two previously reported missense variants, c.443C>G (p.A148G) and c.640C>G (p.L214V),6 were transfected into the HCT116 CRC cell line, which does not express RNF43, in order to analyse their effect on RNF43 protein expression and Wnt activity.7 As expected, no RNF43 was detected in the p.R132*-transfected cell line (figure 1A). Moreover, p.A148G and p.L214V diminished the inhibitory effect of RNF3 on Wnt signalling (figure 1B). These results support the functional impact of all three variants. Taking into consideration the two largest SPS series with the complete coding region of the gene studied (Buchanan et al 6 and our study), the frequency of RNF43 mutations is 1.76% (3/170).
Of the 13 mutation carriers reported so far, 12 developed serrated polyps and/or CRC (mean age at diagnosis: 44 years; range: 18–65). Also, all 23 colonic lesions analysed showed RNF43 somatic loss or mutation (table 1); evidence in favour of a causal role of the tumour suppressor gene RNF43 in SPS. Nevertheless, additional supporting segregation data are required to definitely confirm RNF43 causal involvement in a dominantly inherited form of SPS. Until then, caution should be taken when interpreting RNF43 findings in the clinical setting.
The authors thank Andreas Wanisch and Gemma Aiza for technical assistance, the participating patients and families and the Units of Genetic Counseling and Genetic Diagnostic of the Hereditary Cancer Program of the Catalan Institute of Oncology in Hospitalet de Llobregat, Girona and Badalona.
IQ, RM-L and MT contributed equally.
Contributors LV conceived the study. LV, RM-L, MG and GC contributed to the study design. IQ, RM-L, MT, SB and PM performed the experiments. MN, VP, EG, ED, AS, JB and GC were responsible for the selection and recruitment of patients and for the accrual samples. LV wrote the manuscript with the assistance and final approval of all authors.
Funding This work was supported by the Spanish Ministry of Economy and Competitiveness, cofunded by FEDER funds – a way to build Europe – (grants SAF2016-80888-R and SAF2015-68016-R, and Juan de la Cierva postdoctoral fellowship (PM)); Instituto de Salud Carlos III (CIBERONC CB16/12/00234) and the Government of Catalonia (Pla estratègic de recerca i innovació en salut (PERIS) and 2014SGR338).
Competing interests None declared.
Ethics approval Ethics Committee of Institut d’Investigació Biomèdica de Bellvitge (IDIBELL).
Provenance and peer review Not commissioned; externally peer reviewed.
Data sharing statement Experimental conditions, primer sequences and descriptive data of the studied patients are available upon request (firstname.lastname@example.org).
If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.