Multiple aberrant splicing of the p53 transcript without genomic mutations around exon-intron junctions in a case of chronic myelogenous leukaemia in blast crisis: a possible novel mechanism of p53 inactivation

Analysis of 3760 hematologic malignancies reveals rare transcriptomic aberrations of driver genes

BACKGROUND

Rare oncogenic driver events, particularly affecting the expression or splicing of driver genes, are suspected to substantially contribute to the large heterogeneity of hematologic malignancies. However, their identification remains challenging.

METHODS

To address this issue, we generated the largest dataset to date of matched whole genome sequencing and total RNA sequencing of hematologic malignancies from 3760 patients spanning 24 disease entities. Taking advantage of our dataset size, we focused on discovering rare regulatory aberrations. Therefore, we called expression and splicing outliers using an extension of the workflow DROP (Detection of RNA Outliers Pipeline) and AbSplice, a variant effect predictor that identifies genetic variants causing aberrant splicing. We next trained a machine learning model integrating these results to prioritize new candidate disease-specific driver genes.

RESULTS

We found a median of seven expression outlier genes, two splicing outlier genes, and two rare splice-affecting variants per sample. Each category showed significant enrichment for already well-characterized driver genes, with odds ratios exceeding three among genes called in more than five samples. On held-out data, our integrative modeling significantly outperformed modeling based solely on genomic data and revealed promising novel candidate driver genes. Remarkably, we found a truncated form of the low density lipoprotein receptor LRP1B transcript to be aberrantly overexpressed in about half of hairy cell leukemia variant (HCL-V) samples and, to a lesser extent, in closely related B-cell neoplasms. This observation, which was confirmed in an independent cohort, suggests LRP1B as a novel marker for a HCL-V subclass and a yet unreported functional role of LRP1B within these rare entities.

CONCLUSIONS

Altogether, our census of expression and splicing outliers for 24 hematologic malignancy entities and the companion computational workflow constitute unique resources to deepen our understanding of rare oncogenic events in hematologic cancers.

Regional analysis of p53 mutations in rheumatoid arthritis synovium

The p53 tumor suppressor protein plays a central role in cell cycle regulation, DNA repair, and apoptosis. Recent studies indicate that DNA damage and somatic mutations in the p53 gene can occur because of genotoxic stress in many tissues, including the skin, colon, and synovium. Although somatic mutations in the p53 gene have been demonstrated in rheumatoid arthritis (RA) synovial tissue and synoviocytes, no information is available on the location or extent of p53 mutations. Using microdissected RA synovial tissue sections, we observed abundant p53 transition mutations, which are characteristic DNA damage caused by oxidative stress. p53 mutations, as well as p53 mRNA expression, were located mainly in the synovial intimal lining rather than the sublining (P < 0.01). Clusters of p53 mutant subclones were observed in some microdissected regions, suggesting oligoclonal expansion. Because IL-6 gene expression is regulated by wild-type p53, IL-6 mRNA expression in microdissected tissues was quantified by using real-time PCR. The regions with high rates of p53 mutations contained significantly greater amounts of IL-6 mRNA compared with the low mutation samples (P < 0.02). The microdissection findings suggest that p53 mutations are induced in RA synovial tissues by inflammatory oxidative stress. This process, as in sun-exposed skin and inflamed colonic epithelium, provides some of the mutant clones with a selective growth advantage. A relatively low percentage of cells containing p53 mutations can potentially affect neighboring cells and enhance inflammation through the elaboration of proinflammatory cytokines.

Splicing mutations in TP53 in human squamous cell carcinoma lines influence immunohistochemical detection

The mutational status of the tumor suppressor gene TP53 is often examined by immunohistochemistry. We compared the incidence of TP53 mutations in 12 permanent squamous cell carcinoma lines of the head and neck with the immunohistochemical staining obtained with two different antibodies. The mutational status of the TP53 gene was assessed by sequencing the complete coding frame of the TP53 mRNA. All 12 tumor cell lines had TP53 mutations. Six of them showed missense mutations and five had premature stop codons caused either by splicing mutations or nonsense mutations or by exon skipping. One tumor cell line was heterozygous, with a truncating splicing mutation and an additional missense mutation located on different alleles. In one case, an in-frame insertion of 23 extra codons was found. All missense mutations were positive in immunhistochemistry and Western blotting. The truncated p53 was not immunohistochemically detected in three cases with the DO-7 antibody and in five cases with the G59-12 antibody, giving false-negative results in 25% or 40%, respectively, of all tumor cell lines examined. We conclude that splicing mutations are common in squamous cell carcinoma lines and that the incidence of p53 inactiviation by erroneous splicing is higher than yet reported. Sequencing of only the exons of TP53 may miss intronic mutations leading to missplicing and may therefore systematically underestimate the TP53 mutation frequency.

p53 website and analysis of p53 gene mutations in human cancer: forging a link between epidemiology and carcinogenesis

The p53 tumor suppressor gene has proven to be one of the genes most often mutated in human cancers. It involves mainly point mutations leading to amino acid substitutions in the central region of the protein which impairs normal functions. Analysis of the mutational events that target the p53 gene has revealed evidence for both exogenous and endogenous mutational mechanisms. For example, the p53 mutational spectrum reveals evidence for a direct causal effect of ultraviolet radiation in skin cancer, of aflatoxin B1 in liver cancer, and of tobacco smoke in lung cancer. This novel field, molecular epidemiology of human cancer risk, has added a new dimension to classical associative epidemiology by providing a direct link between human cancer and carcinogen exposure. For such analysis, we devised a generic software called UMD (Universal Mutation Database). It was developed as a generic software to create locus-specific databases (LSDBs) with the 4(th) Dimension(R) package from ACI. This software includes an optimized structure to assist and secure data entry and to allow the input of various clinical data.

p53 gene mutations and p21 protein expression induced independently of p53, by TGF-beta and gamma-rays in squamous cell carcinoma cells

p53 gene mutation and the influence of TGF-beta and gamma-rays on p21 promoter activity, p21 mRNA and protein expression were investigated in nine cell lines (OSC-1 to -9) established from metastatic squamous cell carcinomas (SCC) of the cervical lymph nodes. The direct DNA sequence analysis of exons 2 to 11 of the p53 gene revealed 16 point mutations in all cell lines, but neither deletions nor additions were observed. TGF-beta upregulated p21 promoter activity by approximately 2-fold of the control and concurrently increased p21 mRNA expression, except in OSC-8 and -9. However, gamma-rays suppressed p21 promoter activity, although p21 mRNA expression in irradiated cells was increased except for OSC-8 and -9. In parallel with the messenger expression, p21 protein expression was strongly increased by TGF-beta, but only weakly increased by gamma-rays. These results indicate that point mutation of the p53 gene is frequent in metastatic SCC cells and p21 mRNA and its protein expression is p53-independently induced by both TGF-beta and gamma-rays, although the mechanism of induction by TGF-beta and gamma-rays is different.

Microsatellite instability in human hepatocellular carcinoma: relationship to p53 abnormalities

AIMS/BACKGROUND

Microsatellite instability was sought in 10 human hepatocellular carcinomas (HCCs) to determine whether defective DNA mismatch repair might be implicated in the multiple genetic alterations observed in the p53 tumor suppressor gene in some of these patients' tumors.

METHODS

Genomic DNA from HCCs and adjacent nontumorous livers was subjected to PCR with primers for nine microsatellites, and PCR products were resolved in a denaturing gel. Microsatellite instability was defined as the presence of band shifts or additional bands for at least two microsatellite sequences in an HCC compared to the nontumorous liver tissue from the same patient.

RESULTS

Microsatellite instability was detected in four of ten HCCs. Three of these four HCCs did not have p53 exon mutations. However, one HCC had microsatellite instability as well as multiple p53 exon mutations and multiple intron alterations. Four other patients with multiple p53 intron alterations in HCC (compared to their own nontumorous liver), three of whom also had a mutation in the exons, had no microsatellite instability.

CONCLUSIONS

Defective DNA mismatch repair, as indicated by microsatellite instability, might have played a role in hepatocarcinogenesis in four of the ten patients, but in general it was not associated with p53 alterations. In one of the ten patients, defective DNA mismatch repair might have been the cause of multiple mutations in both the coding and intron sequences of the p53 gene.

Splicing of exon 5 in the WT1 gene is disrupted in Wilms' tumour

Using a reverse transcriptase polymerase chain reaction to examine alternate splicing at site I (exon 5) and site II (exon 9) in the Wilms' tumour suppressor gene, WT1, we found that in seven of the 10 Wilms' tumours examined, splicing at site I was disrupted. This is predicted to result in isoform imbalance in Wilms' tumours, with an increase in isoforms in which the 17 amino acids encoded by exon 5 are missing. These observations could not be explained by mutations or rearrangements in flanking introns. Disrupted alternate splicing of exon 5 may play a role in the aetiology of Wilms' tumour.

Chromosome 17 abnormalities and inactivation of the p53 gene in chronic myeloid leukemia and their prognostic significance

We have reviewed all the relevant studies on the loss of the short arm of chromosome 17 (17p) and inactivation of the p53 gene in chronic myelogenous leukemia (CML) in an attempt to clarify their roles in the progression of CML. Loss of a 17p (hemizygous 17p) and p53 inactivation emerged as the disease progressed and were closely associated with each other. About half of the cases with loss of a 17p, however, did not show p53 inactivation. In these cases loss of a 17p preceded p53 inactivation, which suggested that either reduction of the p53 gene dosage or inactivation of another tumor-suppressor gene on 17p might contribute to the disease progression. Both loss of a 17p and p53 inactivation may serve as poor prognostic factors but the prognostic significance of the former only emerged when metaphase cells with loss of a 17p were dominant amongst the total cell population analyzed.

Hemizygous expression of the wild-type p53 allele may confer a selective growth advantage before complete inactivation of the p53 gene in the progression of chronic myelogenous leukaemia

We analysed the expression of the p53 gene in various clinical phases of chronic myelogenous leukaemia (CML) by quantitative reverse transcriptase (RT) polymerase chain reaction (PCR). Hemizygous expression of the wild-type p53 allele was observed in two samples showing the loss of one p53 allele, and affected p53 expression was associated with p53 allelic loss rather than the clinical phase of CML. In four patients with CML showing p53 inactivation, we performed a detailed sequential analysis of p53 allelic loss, p53 mutation and expression from the onset of the disease to the patients' death. Consequently, we demonstrated that the loss of a wild-type p53 allele preceded mutation of the remaining allele, and that cells hemizygous for the wild-type p53 allele dominated those with both wild-type alleles, then were replaced by cells with complete p53 inactivation. These observations indicate that not only complete p53 inactivation but also hemizygous expression of the wild-type p53 allele may confer a selective growth advantage, and that the former is implicated in a more malignant phase than the latter. Alternatively, the inactivation of another undefined anti-oncogene on chromosome 17p may allow selective growth before the p53 mutation occurs.