Malignant rhabdoid tumours (MRTs) are extremely aggressive cancers of early childhood. They can occur in various locations, mainly the kidney, brain and soft tissues. Cytogenetic and molecular analyses have shown that the deletion of region 11.2 of the long arm of chromosome 22 (22q11.2) is a recurrent genetic characteristic of MRTs, indicating that this locus may encode a tumour suppressor gene. Here we map the most frequently deleted part of chromosome 22q11.2 from a panel of 13 MRT cell lines. We observed six homozygous deletions that delineate the smallest region of overlap between the cell lines. This region is found in the hSNF5/INI1 gene, which encodes a member of the chromatin-remodelling SWI/SNF multiprotein complexes. We analysed the sequence of hSNF5/INI1 and found frameshift or nonsense mutations of this gene in six other cell lines. These truncating mutations of one allele were associated with the loss of the other allele. Identical alterations were observed in corresponding primary tumour DNAs but not in matched constitutional DNAs, indicating that they had been acquired somatically. The observation of bi-allelic alterations of hSNF5/INI1 in MRTs suggests that loss-of-function mutations of hSNF5/INI1 contribute to oncogenesis.

The hSNF5/INI1 gene encodes a member of the SWI/ SNF chromatin remodelling complexes. It was recently identified as a tumour suppressor gene mutated in sporadic and hereditary Malignant Rhabdoid Tumours (MRT). However, the role of hSNF5/INI1 loss-offunction in tumour development is still unknown. Here, we show that the ectopic expression of wild-type hSNF5/ INI1, but not that of truncated versions, leads to a cell cycle arrest by inhibiting the entry into S phase of MRT cells. This G1 arrest is associated with down-regulation of a subset of E2F targets including cyclin A, E2F1 and CDC6. This arrest can be reverted by coexpression of cyclin D1, cyclin E or viral E1A, whereas it cannot be counteracted by pRB-binding deficient E1A mutants. Moreover, hSNF5/INI1 is not able to arrest cells lacking a functional pRB. These observations suggest that the hSNF5/INI1-induced G1 arrest is dependent upon the presence of a functional pRB. However, the observation that a constitutively active pRB can efficiently arrest MRT cells indicates that hSNF5/INI1, at the difference of the ATPase subunits of the SWI/SNF complex, is dispensable for pRB function. Altogether, these data show that hSNF5/INI1 is a potent regulator of the entry into S phase, an effect that may account for its tumour suppressor role.

MethodsIn an observer blind, phase 2 trial, 55 adults were randomized to receive one dose of Ad35.CS.01 vaccine followed by two doses of RTS,S/AS01 (ARR-group) or three doses of RTS,S/AS01 (RRR-group) at months 0, 1, 2 followed by controlled human malaria infection.ResultsARR and RRR vaccine regimens were well tolerated. Efficacy of ARR and RRR groups after controlled human malaria infection was 44% (95% confidence interval 21%-60%) and 52% (25%-70%), respectively. The RRR-group had greater anti-CS specific IgG titers than did the ARR-group. There were higher numbers of CS-specific CD4 T-cells expressing > 2 cytokine/activation markers and more ex vivo IFN-γ enzyme-linked immunospots in the ARR-group than the RRR-group. Protected subjects had higher CS-specific IgG titers than non-protected subjects (geometric mean titer, 120.8 vs 51.8 EU/ml, respectively; P = .001).ConclusionsAn increase in vaccine efficacy of ARR-group over RRR-group was not achieved. Future strategies to improve upon RTS,S-induced protection may need to utilize alternative highly immunogenic prime-boost regimens and/or additional target antigens.Trial RegistrationClinicalTrials.gov NCT01366534

No abstract