Multiple endocrine neoplasia syndromes associated with mutation of p27.

Multiple endocrine neoplasias (MEN) are autosomal dominant disorders characterized by the occurrence of tumors in at least two endocrine glands. Until recently, two MEN syndromes were known, i.e. the MEN type 1 (MEN1) and type 2 (MEN2), which are caused by germline mutations in the MEN1 and RET genes, respectively. These two syndromes are characterized by a different tumor spectrum. A few years ago we described a variant of the MEN syndromes, which spontaneously developed in a rat colony and was named MENX. Affected animals consistently develop multiple endocrine tumors, with a spectrum that shares features with both MEN1 and MEN2 human syndromes. Genetic studies identified a germline mutation in the Cdkn1b gene, encoding the p27 cell cycle inhibitor, as the causative mutation for MENX. Capitalizing on these findings, germline mutations in the human homologue, CDKN1B, were searched for and identified in patients with multiple endocrine tumors. As a consequence of this discovery, a novel human MEN syndrome, named MEN4, was recognized, which is caused by heterozygous mutations in p27. These studies identified Cdkn1b/CDKN1B as a novel tumor susceptibility gene for multiple endocrine tumors in both rats and humans. Here we review the characteristics of the MENX and MEN4 syndromes and we briefly address the main function of p27 and how it is affected by MENX- or MEN4-associated mutations.

Characterization of MENX-associated pituitary tumours.

AIMS: The aim of this study is to evaluate the pathological features, serum hormone levels and ex vivo cultures of pituitary adenomas that occur in rats affected by MENX syndrome. MENX is multiple endocrine neoplasia syndrome caused by a germline mutation in the cell cycle inhibitor p27. Characterization of MENX adenomas is a prerequisite to exploit this animal model for molecular and translational studies of pituitary adenomas. METHODS: We investigated MENX pituitary adenomas with immunohistochemistry, double immunofluorescence, electron microscopy, reverse transcription polymerase chain reaction (RT-PCR), measurement of serum hormone levels and ex vivo cultures. RESULTS: Adenomas in MENX rats belong to the gonadotroph lineage. They start from 4 months of age as multiple neoplastic nodules and progress to become large lesions that efface the gland. Adenomas are composed of chromophobic cells predominantly expressing the glycoprotein alpha-subunit (αGSU). They show mitotic activity and high Ki67 labelling. A few neoplastic cells co-express gonadotropins and the transcription factor steroidogenic factor 1, together with growth hormone or prolactin and Pit-1, suggesting that they are not fully committed to one cell lineage. Ex vivo cultures show features similar to the primary tumour. CONCLUSIONS: Our results suggest that p27 function is critical to regulate gonadotroph cells growth. The MENX syndrome represents a unique model to elucidate the physiological and molecular mechanisms mediating the pathogenesis of gonadotroph adenomas.

Hashimoto's thyroiditis with papillary thyroid carcinoma (PTC)-like nuclear alterations express molecular markers of PTC.

AIMS: Focal papillary thyroid carcinoma (PTC)-like nuclear alterations have been documented in Hashimoto's thyroiditis; however, the molecular association between PTC and Hashimoto's thyroiditis is poorly understood. The aim of this study was to determine whether molecular expression patterns of PTC are present in association with PTC-like nuclear alterations in Hashimoto's thyroiditis. METHODS AND RESULTS: The expression of four genes known to be up-regulated in PTC [LGALS3 (galectin3), CITED1, KRT19 (cytokeratin 19) and FN1 (fibronectin-1)] and the human mesothelial cell protein identified by monoclonal antibody HBME1 was evaluated. Immunohistochemistry was performed on 23 cases of Hashimoto's thyroiditis with focal or diffuse Hürthle cell change and PTC-like nuclear alterations, 37 PTC and 18 normal thyroids. Focal expression of galectin3 (GAL3), CITED1, cytokeratin 19 (CK19), HBME1 and fibronectin-1 (FN1) was seen in 87%, 65%, 43%, 26% and 17% of Hashimoto's thyroiditis, respectively, only in thyrocytes showing PTC-like nuclear alterations. In contrast, diffuse expression of GAL3, CITED1, CK19, HBME1 and FN1 was seen in 100%, 95%, 70%, 87% and 89% of PTC, respectively. Normal thyroid tissues did not express any of these proteins. Following immunohistochemistry, four Hashimoto's thyroiditis cases were found to contain foci of PTC. These foci were highlighted by the diffuse and strong expression of PTC-associated proteins, which prompted additional retrospective scrutiny of the haematoxylin and eosin-stained sections leading to appreciation of complete PTC-type nuclear atypia. CONCLUSIONS: Focal PTC-like immunophenotypic changes in Hashimoto's thyroiditis suggest the possibility of early, focal premalignant transformation in some cases of Hashimoto's thyroiditis.

Gene expression in papillary thyroid carcinoma reveals highly consistent profiles.

Papillary thyroid carcinoma (PTC) is clinically heterogeneous. Apart from an association with ionizing radiation, the etiology and molecular biology of PTC is poorly understood. We used oligo-based DNA arrays to study the expression profiles of eight matched pairs of normal thyroid and PTC tissues. Additional PTC tumors and other tissues were studied by reverse transcriptase-PCR and immunohistochemistry. The PTCs showed concordant expression of many genes and distinct clustered profiles. Genes with increased expression in PTC included many encoding adhesion and extracellular matrix proteins. Expression was increased in 8/8 tumors for 24 genes and in 7/8 tumors for 22 genes. Among these genes were several previously known to be overexpressed in PTC, such as MET, LGALS3, KRT19, DPP4, MDK, TIMP1, and FN1. The numerous additional genes include CITED1, CHI3L1, ODZ1, N33, SFTPB, and SCEL. Reverse transcriptase-PCR showed high expression of CITED1, CHI3L1, ODZ1, and SCEL in 6/6 additional PTCs. Immunohistochemical analysis detected CITED1 and SFTPB in 49/52 and 39/52 PTCs, respectively, but not in follicular thyroid carcinoma and normal thyroid tissue. Genes underexpressed in PTC included tumor suppressors, thyroid function-related proteins, and fatty acid binding proteins. Expression was decreased in 7/8 tumors for eight genes and decreased in 6/8 tumors for 19 genes. We conclude that, despite its clinical heterogeneity, PTC is characterized by consistent and specific molecular changes. These findings reveal clues to the molecular pathways involved in PTC and may provide biomarkers for clinical use.

Effects of topoisomerase II inhibitors on gastric cancer cells characterized by different genetic lesions.

Gastric cancer is poorly-responsive to widely used antitumour drugs, the efficacy of which is thought to be related to the capacity of triggering apoptosis. This process requires a series of gene products including a functional p53 protein. We tested the effects of two DNA topoisomerase II poisons, etoposide and doxorubicin, on gastric cancer cell lines with different genetic lesions. We characterised MKN74 and MKN28 cells for p53 gene status and for the expression of p53 and p21 proteins, as well as of topoisomerase II alpha and beta isoforms. After drug treatments, the cells were analysed for drug cytotoxicity, colony forming ability, cell cycle distribution and presence of apoptotic features. Our findings demonstrated that both etoposide and doxorubicin have a potent anti-proliferative effect on gastric cancer cells. Cell death kinetics was different in the two cell lines, MKN74 cells being more sensitive than MKN28 to the drugs. MKN74 cells, although harboring a wt p53 gene, were unable to undergo a massive apoptosis following etoposide treatment. The response of this cell line might be related to the topoisomerase II beta isozyme, the expression of which proved to be undetectable.

p53-mediated negative regulation of stathmin/Op18 expression is associated with G(2)/M cell-cycle arrest.

Utilizing the technique of differential display of mRNA, we have identified p53-responsive genes that are transcriptionally up- or down-regulated as cells enter growth arrest. One gene that was down-regulated, pong16, was found to be identical to stathmin/Op18, a protein involved in the regulation of microtubule dynamics. Evidence that p53 is directly or indirectly involved in negative regulation of stathmin/Op18 expression includes the following: (i) p53-mediated growth inhibition is associated with repression of stathmin/Op18 expression following serum stimulation, (ii) reporter gene assays revealed p53-mediated repression of stathmin/Op18 promoter activity and (iii) constitutive over-expression of stathmin/Op18 bypasses a p53-mediated G(2)/M arrest in the cell cycle. These results suggest that p53-mediated negative regulation of stathmin/Op18 plays an important role in cell-cycle control.

Mutations in KERA, encoding keratocan, cause cornea plana.

Specialized collagens and small leucine-rich proteoglycans (SLRPs) interact to produce the transparent corneal structure. In cornea plana, the forward convex curvature is flattened, leading to a decrease in refraction. A more severe, recessively inherited form (CNA2; MIM 217300) and a milder, dominantly inherited form (CNA1; MIM 121400) exist. CNA2 is a rare disorder with a worldwide distribution, but a high prevalence in the Finnish population. The gene mutated in CNA2 was assigned by linkage analysis to 12q (refs 4, 5), where there is a cluster of several SLRP genes. We cloned two additional SLRP genes highly expressed in cornea: KERA (encoding keratocan) in 12q and OGN (encoding osteoglycin) in 9q. Here we report mutations in KERA in 47 CNA2 patients: 46 Finnish patients are homozygous for a founder missense mutation, leading to the substitution of a highly conserved amino acid; and one American patient is homozygous for a mutation leading to a premature stop codon that truncates the KERA protein. Our data establish that mutations in KERA cause CNA2. CNA1 patients had no mutations in these proteoglycan genes.

Gene amplification and proliferative kinetics in relation to prognosis of patients with gastric carcinoma.

BACKGROUND: The differences in survival of gastric carcinoma patients who have identical clinical or pathologic stages prompted the authors to investigate the prognostic significance of biologic features that are known to affect the clinical aggressiveness of other tumor types. METHODS: One hundred twenty-four tumor samples from patients who had received radical or palliative surgery were analyzed for c-myc, c-K-ras, hst, and c-erb B-2 gene amplification by means of the Southern blot technique. Of these tumors, 70 were also examined for cell kinetics by means of the thymidine labeling index (TLI). RESULTS: The analysis of associations between gene amplification and the anatomicopathologic variables (TNM classification, site of tumor, and histology) showed that amplification represents a late event in the natural history of gastric carcinoma. Gene amplification showed a slight, statistically insignificant, negative impact on overall survival (OS) (P = 0.09). Amplification of c-erb B-2 correlated in a statistically significant way with reduced OS (P = 0.03). Cox multiple regression analysis revealed that neither c-erb B-2 amplification nor TLI had prognostic significance in relation to OS. CONCLUSIONS: These data indicate that amplification of the examined oncogenes did not reveal a new independent prognostic factor for patients with gastric carcinoma. However, the authors' results did show a strong correlation between gene amplification and tumor progression, which warrants further study involving larger series of patients. At the same time, the TLI results underlined the need to identify the most suitable biologic material for use in the estimation of proliferative indexes in gastric carcinoma.

DNA damage and p53-mediated cell cycle arrest: a reevaluation.

Most mammalian cells exhibit transient delays in the G1 and G2 phases of the cell cycle after treatment with radiation or radiomimetic compounds. p53 is required for the arrest in G1, which provides time for DNA repair. Recently, a role of p53 in the G2/M transition has also been suggested. However, it has been reported that the presence of functional p53 does not always correlate with the induction of these checkpoints. To precisely assess the role of p53 in activating cell cycle checkpoints and in cell survival after radiation, we studied the response of two isogenic human fibrosarcoma cell lines differing in their p53 status (wild type or mutant). We found that when irradiated cells undergo a wild-type p53-dependent G1 arrest, they do not subsequently arrest in G2. Moreover, wild-type p53 cells irradiated past the G1 checkpoint arrest in G2 but do not delay in the subsequent G1 phase. Furthermore, in these cell lines, which do not undergo radiation-induced apoptosis, the wild-type p53 cell line exhibited a greater radioresistance in terms of clonogenic survival. These results suggest that the two checkpoints may be interrelated, perhaps through a control system that determines, depending on the extent of the damage, whether the cell needs to arrest cell cycle progression at the subsequent checkpoint for further repair. p53 could be a crucial component of this control system.

Study on aneuploidy and p53 mutations in astrocytomas.

To determine whether a correlation exists between aneuploidy and p53 status in astrocytic tumors we analyzed 48 astrocytomas with different grades of malignancy for the presence of p53 mutations and aneuploidy of chromosomes 10 and 17 (Ch10, Ch17), known to be particularly involved with this type of tumor. We used polymerase chain reaction (PCR)-based denaturing gradient gel electrophoresis (DGGE) analysis on exons 5-8 of the p53 gene, and fluorescence in situ hybridization (FISH) analysis on interphase nuclei using chromosome specific pericentromeric probes, respectively. Our results showed that Ch10/Ch17 aneuploidy is a common early event in astrocytomas (90% of low grade tumors are aneuploid). p53 mutations and Ch17 aneuploidy are early events, but their incidence is not dependent on tumor grade. Loss of Ch10 is the only alteration that significantly correlates with tumor progression. No significant correlation between the presence of Ch10/Ch17 aneuploidy and p53 mutations was found. However, the coexistence of p53 mutations and aneuploidy, was observed in a subset of cases. The presence of p53 mutations appeared to be a significant predictor of a poor prognosis. In conclusion, genomic instability may or may not be associated with p53 mutations in astrocytomas, thus suggesting that other cellular determinants can also be responsible for the aneuploidy observed.

Coordinated effects of insulin-like growth factor I on inhibitory pathways of cell cycle progression in cultured cardiac muscle cells.

Insulin-like growth factor I (IGF I) plays a key role in the regulation of cell proliferation. Progression of the cell cycle is regulated by stimulatory and inhibitory pathways. In order to understand the mechanisms through which IGF I regulates cardiac muscle growth, we have studied the effects of IGF I on inhibitory pathways involving p53 and WAF1 in cultured cardiac muscle cell line H9C2. The onset of DNA synthesis in response to IGF I stimulation was preceded by activation of p53 expression. In addition, IGF I increased p53-dependent and p53-independent induction of WAF1 in H9C2 cells. Dose-response studies showed that IGF I effects on p53-dependent and p53-independent induction of WAF1 occur at physiological concentrations of IGF I. These data indicate that IGF I coordinately regulates inhibitory pathways of cell cycle progression, and that p53-dependent and p53-independent induction of WAF1 may provide negative control mechanisms to regulate stimulatory pathways of cell cycle progression activated by IGF I.

The basic carboxy-terminal domain of human p53 is dispensable for both transcriptional regulation and inhibition of tumor cell growth.

To investigate the relevance of the C-terminal domains of the human p53 tumor suppressor gene to its growth suppressive and transcriptional regulatory properties deletion mutants were generated which eliminated 30 (p53 delta 363), 60 (p53 delta 333) and 87 (p53 delta 306) amino acids from the C-terminus of the p53 protein. p53 delta 363 has lost the highly basic tail of the protein (residues 360-386). p53 delta 333 and p53 delta 306 lack the oligomerization domain (residues 320-360); p53 delta 306 has also lost the major nuclear localization signal of p53 (NLSI, residues 316-325). These mutants were assayed for transactivation from two p53 consensus binding sites and for transcriptional repression of two promoter systems in Calu6 lung cancer cells (p53 null). Moreover, their ability to inhibit cell growth in tumor cell lines with a defined p53 status was analysed. Deletion of the oligomerization domain correlated with significant loss of: (a) transactivation from a genomic sequence; (b) transcriptional repression; (c) the ability to inhibit colony formation. An intact NLSI was not a prerequisite for transactivation. p53 delta 363 behaved similarly to wt p53 in all the assays. We established an inducible expression system for p53 delta 363 in a human fibrosarcoma cell line known to be growth-suppressed by wt p53. The induction of p53 delta 363 expression also inhibited cell proliferation albeit to a lesser extent than wt p53. However, p53 delta 363 could upregulate WAF1/CIP1, GADD45 and MDM2 genes. Thus, the basis tail of p53 appears not to be required for the biological functions of the protein assayed.

Analysis and clinical implications of K-ras gene mutations and infection with human papillomavirus types 16 and 18 in primary adenocarcinoma of the uterine cervix.

Experimental models indicate that activated ras genes and HPV oncogenic sequences may cooperate in inducing a completely transformed phenotype in epithelial cells. We searched for K-ras gene mutations and HPV type-16 and -18 sequences in 67 primary adenocarcinomas of the uterine cervix by analyzing DNAs from formalin-fixed, paraffin-embedded tissue samples. Target sequences were amplified by PCR and analyzed by denaturing gradient gel electrophoresis (DGGE) and sequencing for the detection of K-ras gene mutations and by Southern blotting for the detection of HPV infection. We found 16 mutations in 15 cases; 14 were at codon 12 and 2 at codon 13; 11 were base transitions and 5 were transversions. Mutations were more frequent in mucin-secreting than in non-mucinous tumors. HPV oncogenic sequences were detected in 58 cases with no significant difference between K-ras-mutated and wild-type tumors. HPV oncogenic sequences were also more frequent in mucin-secreting than in non-mucinous tumors. Both molecular events were present simultaneously in 13 out of 58 cases, all of which had histologically grade-2 and grade-3 tumors. Clinico-pathological parameters of the disease and the overall survival, however, were independent of K-ras mutations and of HPV-16 and -18 infection, as shown by univariate and multivariate analysis. In contrast, stage of disease, lymph-node metastases, deep infiltration, clear-cell histology and low grade of differentiation were risk factors for tumor-related death.

K-ras and p53 gene mutations in pancreatic cancer: ductal and nonductal tumors progress through different genetic lesions.

We studied K-ras and p53 gene mutations in a panel of 57 primary pancreatic cancers including ductal and nonductal tumors. DNAs were obtained from formalin-fixed, paraffin-embedded material. Target sequences were amplified by polymerase chain reaction and analyzed by denaturing gradient gel electrophoresis and sequencing. Both K-ras and p53 genes were frequently mutated in ductal cancers (25 of 35, 71.4%; 18 of 35, 51.4%, respectively). K-ras mutations were confined to the second position of codon 12 where base transitions and transversions were equally observed. p53 changes were mainly missense mutations. Transitions and transversions were found equally with a prevalence of G:C-->A:T changes among transitions. No gene alterations were present in the 6 exocrine nonductal tumors and (with one exception) in the 12 endocrine tumors analyzed. Our results indicate that mutated K-ras and p53 genes can cooperate in the establishment of ductal pancreatic cancers, whereas other genetic events have to be present in nonductal tumors. Moreover, K-ras alterations may represent an early event in ductal tumorigenesis, as suggested both by the high gene mutation frequency and by the presence of mutations in low-grade tumors. On the contrary, p53 gene changes seem to represent an event required for the malignancy progression of ductal tumors from lower to higher grades.

Intraductal papillary-mucinous tumours represent a distinct group of pancreatic neoplasms: an investigation of tumour cell differentiation and K-ras, p53 and c-erbB-2 abnormalities in 26 patients.

Intraductal papillary growth of mucin producing hypersecreting, columnar cells characterizes a group of rare pancreatic exocrine neoplasms which we propose to call intraductal papillary-mucinous tumors (IPMT). We analysed the histopathology of 26 IPMT in relation to gastro-enteropancreatic marker expression, genetic changes and biology. Four IPMT showing only mild dysplasia were considered to be adenomas. Nine tumours displayed moderate dysplasia and were regarded as borderline. Severe dysplasia-carcinoma in situ changes were found in 13 IPMT which were therefore classified as intraductal carcinomas. Six of these carcinomas were frankly invasive and two of these had lymph node metastases. The invasive component resembled mucinous non-cystic carcinoma in all but one tumour which showed a ductal invasion pattern. Immunohistochemically, an intestinal marker type was found in most carcinomas, while gastric type differentiation prevailed among adenomas or borderline tumours. K-ras mutations (seven at codon 12 and one at codon 13) were found in 31% of IPMT (2 adenomas, 1 borderline, 5 carcinomas). Nuclear p53 overexpression was detected in 31% of IPMT (6 carcinomas and 2 borderline IPMT) and correlated with p53 mutations (one at exon 8 and the other at exon 5) in two carcinomas. p53 abnormalities were unrelated to K-ras mutation. c-erbB-2 overexpression was observed in 65% of IPMT, with various grades of dysplasia. Twenty-two of 24 patients are alive and well after a mean post-operative follow-up of 41 months. Only two patients, both with invasive cancer at the time of surgery, died of tumour disease. It is concluded that pancreatic IPMT encompass neoplasms which, in general, have a favorable prognosis, but are heterogeneous in regard to grade of dysplasia and marker expression. Adenoma, borderline tumour, intraductal carcinoma and invasive carcinoma can be differentiated. p53 changes but not K-ras mutation or c-erbB-2 overexpression are related to the grade of malignancy. Most IPMT differ in histological structure, marker expression and behaviour from ductal adenocarcinoma.

Primary retroperitoneal mucinous cystoadenocarcinomas: an immunohistochemical and molecular study.

Special immunohistochemical stains for the identification of gastroenteropancreatic antigens in two cases of primary retroperitoneal mucinous cystoadenocarcinomas (PRMC) show that these tumours have patterns similar to ovarian mucinous tumours. Markers of pyloric type gastric mucosa differentiation (M1, cathepsin E, concavavalin A, pepsinogen II) are mostly positive in benign and borderline areas with endocervical type differentiation, while immunoreactivity for intestinal cell markers (M3SI and CAR-5) and for DU-PAN-2 is present mainly in frankly malignant areas, regardless of differentiation type. DNA analysis shows a point mutation of K-ras oncogene at codon 12 (GGT to CGT) in one case. The immunohistochemical and genotypic similarity of PRMC and ovarian mucinous tumours may indicate similar mechanisms in their histogenesis.

K-ras gene mutations: an unfavorable prognostic marker in stage I lung adenocarcinoma.

Activation of K-ras gene by point mutations, a common finding in lung adenocarcinomas, has been suggested to decrease patient survival. We investigated 109 lung adenocarcinomas, mostly small, peripheral, stage I tumours (81/109) for presence of K-ras gene mutations at codons 12 and 13. Mutations were detected by denaturing gradient gel electrophoresis analysis of specific sequences amplified by polymerase chain reaction from DNA extracted from archival pathological material. Thirty-three of 109 (30.3%) tumours showed mutations at codon 12 (28/33, 84.8%) or 13 (5/33, 15.2%) of the gene. Mutations and type of nucleotide substitutions were differently distributed among cytological subtypes, being more prevalent among less differentiated (G2 and G3) tumours and among bronchial than bronchiolo-alveolar type adenocarcinomas. Survival analysis showed an adverse effect of K-ras mutation on survival, restricted to stage I tumours. Median survival for 81 stage I patients was 30 months for non-mutated tumours versus 20 months for mutated tumours (p = 0.016). Multivariate analysis showed that age of patient (p = 0.001) and K-ras mutation status (p = 0.04) were the only independent factors influencing survival significantly. These data strengthen the hypothesis that K-ras gene mutations may be useful in identifying a subgroup of patients with poor outcome.

Loss of heterozygosity and K-ras gene mutations in gastric cancer.

In order to identify relevant genetic lesions in gastric carcinoma, we searched for tumor suppressor gene inactivation and K-ras gene mutations by analyzing tumor and control DNAs from 34 patients. These were from an epidemiologically defined area of Italy characterized by one of the world's highest incidences of stomach cancer. Allele losses were investigated by the Southern blotting procedure at 16 polymorphic loci on 11 different chromosomes. Our data demonstrate that chromosomal regions 5q, 11p, 17p and 18q are frequently deleted, and that 7q and 13q chromosome arms are also involved, although at a lower frequency. Loss of heterozygosity (LOH) at region 11p was not found during other surveys carried out on patients of different geographic origins. No specific combination of allelic losses could be recognized in the samples analyzed, the only exception being that tumors with 17p allelic loss also showed LOH on the 18q region. When matching frequent LOH events and the stage of progression of the tumors, we observed a trend of association between advanced stages and allelic losses on 17p and 18q chromosome arms. The analysis of K-ras, carried out by the polymerase chain reaction and denaturing gradient gel electrophoresis, demonstrated transforming mutations in only 3 out of 32 cases. Colorectal tumorigenesis proceeds by the accumulation of genetic alterations, including K-ras mutations and inactivation of tumor suppressor genes on the 5q, 17p and 18q regions. Our data indicate that, although gastric and colorectal neoplasias share common genetic alterations, they probably progress through different pathways.

Base transitions are the most frequent genetic changes at P53 in gastric cancer.

We searched for P53 mutations in gastric carcinoma by analyzing tumor DNAs from 29 patients. We detected 13 different somatic mutations in 15 patients (52%) and a biallelic polymorphism in exon 6 (5 heterozygous subjects). The somatic mutations were mainly localized in the sequences corresponding to the highly conserved domains of the protein. Twelve samples showed a single base change: 11 missense and 1 nonsense mutations. Three samples showed deletions leading to a frame shift, to the in-frame loss of 2 amino acids, and to the deletion of a splicing site. All point mutations, except one, were transitions, and 91% of them were G:C-->A:T changes. We previously analyzed this panel of tumors for allelic loss at the 17p13 chromosomal region, where the P53 gene had previously been located: the results showed an increasing incidence of allelic loss in late-stage tumors. On the contrary, in the present study no trend between P53 mutations and tumor stages was found. This observation indicates that mutation events precede allelic loss in gastric cancer. Half (54%) of the mutations occurred in samples without allelic loss, suggesting that specific mutated alleles, acting in a dominant negative fashion, can alter in vivo the P53 protein function.