The nationwide Danish polyposis register includes all known Danish cases of familial adenomatous polyposis (FAP) and their relatives. By identifying all FAP patients born between 1920 and 1949, we found the frequency of the disease to be 1 in 13,528. By comparing the number of affected and nonaffected offspring born to affected parents during the same period we found the penetrance of the disease for inherited cases to be close to 100% at the age of 40 years. The mutation rate found by the direct method was 9 mutations per million gametes per generation and the proportion of new mutants was estimated to 25%. Fitness for patients between 15 and 29 years was found close to one, while for patients older than 30 the fitness was reduced, but increasing during the three decades (from 0.44 to 0.71) probably because treatment became more widespread and efficient. As we have used the overall fitness in the period, 0.87, to estimate the mutation rate by the indirect method, we found a lower value than by the direct method, namely 5 mutations per million gametes per generation.

A breast carcinoma biopsy showed cytochemical evidence of epithelial mesenchymal transition and an ␣-smooth muscle actin-positive stromal reaction. To study the lineage, and the nature of the cells in the stromal reaction, we derived a novel cell line, HBFL-1, from the explanted biopsy. HBFL-1 cells are immortal and exhibit a shared non-random X-chromosome inactivation pattern with the epithelial tumor of origin. Yet they closely resemble normal, finite-life-span fibroblasts by morphology, lack of tumor formation in nude mice, marker expression profile, protein pattern using two-dimensional gel electrophoresis and the ability to undergo myofibroblast conversion. HBFL-1 interacts reciprocally with tumor cells in collagen gel to induce activation of MMP2, leading to tumor-like behavior of epithelial colonies. In vivo, HBFL-1 cells resembled normal-derived myofibroblasts and conferred a significant 3.5-to 7-fold increase in MCF-7 tumor size in nude mice. However, that they were indeed not normal fibroblasts was revealed by residual keratin expression and formation of epithelial microfoci in a reconstituted basement membrane and in nude mice. We conclude that breast cancer can generate its own nonmalignant stroma and that one function for this is that of a reciprocal interac- Epithelial mesenchymal transition (EMT) was originally described as a normal developmental process. 1,2 Later, it was adopted as an explanation for mesenchymal conversion in a number of cultured epithelial cells. 3,4 In cancer, EMT generally depicts a more aggressive behavior of the tumor cells. 5,6 In breast cancer, EMT has been estimated to occur in as much as 18% of tumors in vivo. [7][8][9] Under these conditions EMT is defined as the occurrence of a variable proportion of tumor cells that express mesenchymal markers such as vimentin, tenascin and stromelysin-3. 7,10 In its most elaborate form, EMT-derived cells of mixed epithelial-mesenchymal breast tumors may be difficult to distinguish from resident normal stromal cells. 11 These tumors which are also referred to as carcinosarcomas or metaplastic carcinomas in particular offer an excellent opportunity to study the nature and the consequence of this subset of EMT. 5,6 Such studies, however, have been hampered by lack of representative cell lines most likely due to low frequency of overtly metaplastic carcinomas but also to difficulties in culturing breast cancer cells in general. In the present study we succeeded in isolating a mesenchymal-like cell line from a metaplastic human breast carcinoma. Clonality analysis revealed that the cell line and the epithelial tumor cells of origin had a common ancestor. Even though the cells were immortal and severely aneuploid, and exhibited a rudimentary epithelial phenotype in terms of keratin expression and formation of microfoci in Matrigel and in vivo, they nevertheless behaved remarkably like normal resident fibroblasts. In particular they responded to transforming growth factor-␤ (TGF-␤) by having ␣-smooth muscle actin (␣-sm actin) induced and t...

We have used array comparative genomic hybridization to map DNA copy-number changes in 94 patients with cri du chat syndrome who had been carefully evaluated for the presence of the characteristic cry, speech delay, facial dysmorphology, and level of mental retardation (MR). Most subjects had simple deletions involving 5p (67 terminal and 12 interstitial). Genotype-phenotype correlations localized the region associated with the cry to 1.5 Mb in distal 5p15.31, between bacterial artificial chromosomes (BACs) containing markers D5S2054 and D5S676; speech delay to 3.2 Mb in 5p15.32-15.33, between BACs containing D5S417 and D5S635; and the region associated with facial dysmorphology to 2.4 Mb in 5p15.2-15.31, between BACs containing D5S208 and D5S2887. These results overlap and refine those reported in previous publications. MR depended approximately on the 5p deletion size and location, but there were many cases in which the retardation was disproportionately severe, given the 5p deletion. All 15 of these cases, approximately two-thirds of the severely retarded patients, were found to have copy-number aberrations in addition to the 5p deletion. Restriction of consideration to patients with only 5p deletions clarified the effect of such deletions and suggested the presence of three regions, MRI-III, with differing effect on retardation. Deletions including MRI, a 1.2-Mb region overlapping the previously defined cri du chat critical region but not including MRII and MRIII, produced a moderate level of retardation. Deletions restricted to MRII, located just proximal to MRI, produced a milder level of retardation, whereas deletions restricted to the still-more proximal MRIII produced no discernible phenotype. However, MR increased as deletions that included MRI extended progressively into MRII and MRIII, and MR became profound when all three regions were deleted.

We report on three unrelated mentally disabled patients, each carrying a de novo balanced translocation that truncates the autism susceptibility candidate 2 (AUTS2) gene at 7q11.2. One of our patients shows relatively mild mental retardation; the other two display more profound disorders. One patient is also physically disabled, exhibiting urogenital and limb malformations in addition to severe mental retardation. The function of AUTS2 is presently unknown, but it has been shown to be disrupted in monozygotic twins with autism and mental retardation, both carrying a translocation t(7;20)(q11.2;p11.2) (de la Barra et al. in Rev Chil Pediatr 57:549-554, 1986; Sultana et al. in Genomics 80:129-134, 2002). Given the overlap of this autism/mental retardation (MR) phenotype and the MR-associated disorders in our patients, together with the fact that mapping of the additional autosomal breakpoints involved did not disclose obvious candidate disease genes, we ascertain with this study that AUTS2 mutations are clearly linked to autosomal dominant mental retardation.

Tooth development is under strict genetic control. Oligodontia is defined as the congenital absence of 6 or more permanent teeth, excluding the third molar. The occurrence of non-syndromic oligodontia is poorly understood, but in recent years several cases have been described where a single gene mutation is associated with oligodontia. Several studies have shown that MSX1 and PAX9 play a role in early tooth development. We screened one family with non-syndromic oligodontia for mutations in MSX1 and PAX9. The pedigree showed an autosomal-dominant pattern of inheritance. Direct sequencing and restriction enzyme analysis revealed a novel heterozygous A to G transition mutation in the AUG initiation codon of PAX9 in exon 1 in the affected members of the family. This is the first mutation found in the initiation codon of PAX9, and we suggest that it causes haploinsufficiency.