Exclusive paternal origin of new mutations in Apert syndrome.

Apert syndrome results from one or other of two specific nucleotide substitutions, both C-->G transversions, in the fibroblast growth factor receptor 2 (FGFR2) gene. The frequency of new mutations, estimated as 1 per 65,000 live births, implies germline transversion rates at these two positions are currently the highest known in the human genome. Using a novel application of the amplification refractory mutation system (ARMS), we have determined the parental origin of the new mutation in 57 Apert families: in every case, the mutation arose from the father. This identifies the biological basis of the paternal age effect for new mutations previously suggested for this disorder.

Organization and chromosomal localization of the human platelet-derived endothelial cell growth factor gene.

Human platelet-derived endothelial cell growth factor (hPD-ECGF) is a novel angiogenic factor which stimulates endothelial cell growth in vitro and promotes angiogenesis in vivo. We report here the cloning and sequencing of the gene for hPD-ECGF and its flanking regions. This gene is composed of 10 exons dispersed over a 4.3-kb region. Its promoter lacks a TATA box and a CCAAT box, structures characteristic of eukaryotic promoters. Instead, six copies of potential Sp1-binding sites (GGGCGG or CCGCCC) were clustered just upstream of the transcription start sites. Southern blot analysis using genomic DNAs from several vertebrates suggested that the gene for PD-ECGF is conserved phylogenetically among vertebrates. The gene for hPD-ECGF was localized to chromosome 22 by analysis of a panel of human-rodent somatic cell hybrid lines.

Structural alterations of the c-mos locus in benign pleomorphic adenomas with chromosome abnormalities of 8q12.

Recent mapping studies have assigned the human c-mos proto-oncogene to chromosome 8, bands q11-12. This region is frequently affected by chromosomal translocations in benign pleomorphic adenomas of the salivary glands. Using Southern blot analysis we report here that the c-mos gene and its flanking sequences are structurally altered in pleomorphic adenomas with chromosomal rearrangements of 8q12. Rearrangements were detected in two out of 23 tumors. Restriction fragment analysis indicated that the rearrangements were due to multiple, subtle mutations involving the c-mos open reading frame and its flanking sequences. There was no direct evidence of translocation of mos in any of the tumors. Control DNAs from the two patients showed a normal restriction pattern for all enzymes tested, indicating that the rearrangements are tumor specific. Collectively, our cytogenetic and molecular data suggest involvement of the c-mos gene in the pathogenesis of pleomorphic adenomas.

Interphase cytogenetic analysis of cultured pleomorphic adenomas with a normal karyotype.

A subgroup of pleomorphic adenomas of the salivary glands is characterized by apparently normal stemline karyotypes. In this investigation we have used FISH to study whether clonal translocations involving chromosomes 8 and 12 might have escaped detection with conventional cytogenetic analysis. Whole chromosome painting probes for chromosomes 8 and 12 were hybridized to interphase nuclei from 22 cytogenetically normal pleomorphic adenomas. In 20 of the 22 cases the majority of the nuclei (75-93%) showed two hybridizing chromosome 8 and chromosome 12 domains of equal size, indicating the presence of two intact copies of each of these chromosomes. Evidence of clonal translocations was only detected in two tumors. These findings show that in the vast majority of adenomas with a normal karyotype we can exclude the possibility of an abnormal non-dividing or very slow growing, and thus undetected, tumor cell population in vitro. Our data provides further evidence to support the existence of a subgroup of pleomorphic adenomas with a normal karyotype.

INT1 and GLI genes are not rearranged or amplified in benign pleomorphic adenomas with chromosome abnormalities of 12q13-15.

A subgroup of pleomorphic adenomas of the salivary glands is characterized by translocations involving chromosome 12, with consistent breakpoints at 12q13-15. Two proto-oncogenes, INT1 and GLI, have been assigned to this region of chromosome 12. We studied the possible involvement of these genes in pleomorphic adenomas with different karyotypic abnormalities, including cases with involvement of 12q13-15. Using detailed restriction fragment analysis of tumor DNAs from 25 cases, we found no evidence of rearrangement or amplification of INT1 or GLI. Because we previously found an adenoma with a del(12)(q13q15), we also analyzed normal and tumor DNAs from the 25 tumors separately to identify possible allelic losses at the GLI locus. Thirteen of the 25 tumors were informative, and none of these showed evidence of allelic losses. Collectively, these findings indicate that neither the INT1 nor the GLI gene appears to be the primary target gene for the translocations and deletions involving the 12q13-15 region in pleomorphic adenomas.

Cytogenetic and fluorescence in situ hybridization analyses of a microcystic adnexal carcinoma with del(6)(q23q25).

Cytogenetic analysis of a case of microcystic adnexal carcinoma (MAC) revealed three unrelated clones with the karyotypes: 46,XX,del(6)(q23q25)[17]/46,XX,1 approximately 2dmin[4]/46,XX,t(1;3) (p10;q10)[2]. Analysis of the 6q- by fluorescence in situ hybridization (FISH) showed that it had resulted from a long arm deletion. The finding of a clonal 6q deletion as the sole karyotypic change in a MAC is of special interest because we previously have identified 6q deletions in different types of malignant salivary gland tumors. This observation, together with the histopathologic similarities between salivary gland tumors and sweat gland tumors, further emphasize the histogenetic relationships between these tumor types.

Dynamic cranioplasty for brachycephaly in apert syndrome: long-term follow-up study.

OBJECT: Brachycephaly is a characteristic feature of Apert syndrome. Traditional techniques of cranioplasty often fail to produce an acceptable morphological outcome in patients with this condition. In 1996 a new surgical procedure called "dynamic cranioplasty for brachycephaly" (DCB) was reported. The purpose of the present study was to analyze perioperative data and morphological long-term results in patients with the cranial vault deformity of Apert syndrome who were treated with DCB. METHODS: Twelve patients have undergone surgery performed using this technique since its introduction in 1991 (mean duration of follow-up review 60.2 months). Eleven patients had bicoronal synostosis and one had a combined bicoronal-bilambdoid synostosis. Perioperative data and long-term evolution of skull shape visualized on serial cephalometric radiographs were analyzed and compared with normative data. Changes in mean skull proportions were evaluated using a two-tailed paired-samples t-test, with differences being considered significant for probability values less than 0.01. The mean operative blood transfusion was 136% of estimated red cell mass (ERCM) and the mean postoperative transfusion was 48% of ERCM. The mean operative time was 218 minutes. The duration of stay in the intensive care unit averaged 1.7 days and the mean hospital stay was 11.8 days. There were no incidences of mortality and few complications. An improvement in skull shape was achieved in all cases, with a change in the mean cephalic index from a preoperative value of 90 to a postoperative value of 78 (p = 0.000254). CONCLUSIONS: Dynamic cranioplasty for brachycephaly is a safe procedure, yielding high-quality morphological results in the treatment of brachycephaly in patients with Apert syndrome.

Cytogenetics and molecular genetics of human solid tumours.

It is generally accepted that cancer is a genetic disease resulting from the accumulation of multiple genomic rearrangements. These rearrangements involve gross chromosomal abnormalities (e.g. translocations and deletions) as well as submicroscopic mutations affecting both oncogenes and tumour suppressor genes. Recent studies of several tumour specific translocations in sarcomas have shown that the translocations result in so-called fusion genes. In this review we will discuss the specificity and implications of different genetic alterations in both sporadic and hereditary human solid tumours, and provide examples of how these changes can be used as tumour specific markers of both diagnostic and prognostic significance.

Long-term follow-up of dynamic cranioplasty for brachycephaly--non-syndromal bicoronal synostosis.

We followed up 10 patients whose non-syndromal bicoronal synostosis had been operated on with a dynamic cranioplasty technique developed by this craniofacial unit in 1992. With this technique, the growth of the brain is redirected in an anteroposterior direction as wire-mediated compression and restraint are exerted on the transverse and vertical dimensions of the skull. The mean operating time was 160 minutes (range 120-275) and mean stay in the intensive care unit was 36 hours (range 23-58). There was no operative mortality and few complications. The surgical results were assessed objectively by analysis of cephalometric tracings. The mean (SD) cephalic index was 87.6 (4.9) preoperatively and 77.7 (1.8) postoperatively (p = 0.001). The modified Whitaker scale was used as a subjective outcome measurement, and nine patients were classified as Whitaker grade 1 (no additional surgery). One patient required additional intracranial surgery. A questionnaire was sent to all families to obtain an additional subjective measurement of outcome. Parents' satisfaction was high. We conclude that dynamic cranioplasty is a safe and efficient operation for treatment of brachycephaly.

Submicroscopic deletions of 3p sequences in pleomorphic adenomas with t(3;8)(p21;q12).

A subgroup of benign pleomorphic adenomas of the salivary glands is characterized by translocations, or on rare occasions deletions, with breakpoints at 3p21. We have applied restriction fragment length polymorphism (RFLP) analysis to assess the frequency of allelic losses at four different loci located within 3p21-->p25 in 35 pleomorphic adenomas, 18 of which were also karyotyped. Parallel analysis of constitutional and tumor DNAs in informative tumors revealed that all patients retained heterozygosity in their tumor DNA at the D3S2 and RAF1 loci. Among the 29 tumors informative for THRB three showed loss of heterozygosity (LOH). All three tumors had a t(3;8)(p21;q12). Of the 23 tumors informative for D3F15S2, one showed LOH. This tumor also had a t(3;8)(p21;q12). To further map the deletions in relation to the 3p21 translocation breakpoint, we also sublocalized the THRB locus. Using in situ hybridization we assigned the gene to 3p24.1-3. The fact that none of the tumors with loss of 3p alleles showed cytogenetic evidence of deletions indicates that the losses are submicroscopic, probably interstitial, and in most cases distal to the 3p21 breakpoint. This was confirmed in one case with loss of a THRB allele where both proximal (D3F15S2) and distal (RAF1) markers retained heterozygosity. Our results suggest that deletion of 3p sequences might be of progressional importance in a subset of pleomorphic adenomas with t(3;8)(p21;q12).

Detection of hidden structural rearrangements by FISH in pleomorphic adenomas.

Previous cytogenetic analysis of pleomorphic adenomas of the salivary glands has revealed a subgroup of tumors with interstitial deletions of 8q with breakpoints at q12 and q21-23. In this paper we have re-examined four adenomas with confirmed or suspected deletions of 8q by FISH. Painting of interphase nuclei and metaphase chromosomes with whole chromosome libraries revealed that in each of the four cases the deleted chromosome 8 material was found inserted or translocated onto other chromosomes. In two of the cases we were also able to resolve several other complex rearrangements. Our FISH data thus suggest that gross deletions of 8q do not occur in pleomorphic adenomas. These observations are of crucial importance for the molecular interpretations of the 8q12 rearrangements. The fact that all rearrangements with breakpoints at 8q12 seem to be translocations or insertions strongly indicates that they involve a similar molecular mechanism. Our findings illustrate that chromosome painting is a valuable and useful adjunct to conventional cytogenetic analysis of solid tumors.

Expression of the ERBB2 protein in benign and malignant salivary gland tumors.

Fifty-two primary human salivary gland tumors were analyzed for expression of the p185ERBB2 protein using immunohistochemical and immunoblotting techniques. About 63% (33/52) of the tumors expressed the ERBB2 protein. The highest expression levels were detected among the carcinomas, where 32% of the tumors showed intense membrane staining in 25-100% of the tumor cells. In benign pleomorphic adenomas, the corresponding figure was only 12%. Clinical follow-up data available for 18 of the 19 patients with carcinomas suggested an association between high ERBB2 protein levels and poor prognosis as measured by recurrence of disease and/or the appearance of metastases. These results indicate that ERBB2 activation and overexpression could be an important genetic event with possible prognostic implications in a subset of malignant salivary gland tumors.

The 12q13-q15 translocation breakpoints in pleomorphic adenoma and clear-cell sarcoma of tendons and aponeuroses are different from that in myxoid liposarcoma.

The transcription factor gene CHOP was recently shown to be rearranged in myxoid liposarcoma with t(12;16)(q13;p11). We have analyzed whether the CHOP gene is the target of rearrangements in pleomorphic adenoma and clear-cell sarcoma of tendons and aponeuroses with chromosome abnormalities of 12q13-q15. Restriction fragment analysis showed that the CHOP gene and its flanking sequences were not rearranged in any of these tumor types, indicating that the 12q translocation breakpoints in pleomorphic adenoma and clear-cell sarcoma are different from that in myxoid liposarcoma.

Multiple symmetric lipomas with high levels of mtDNA with the tRNA(Lys) A-->G(8344) mutation as the only manifestation of disease in a carrier of myoclonus epilepsy and ragged-red fibers (MERRF) syndrome.

We have investigated the morphology, cytogenetics, and the fraction of mtDNA with the tRNA(Lys) A-->G(8344) mutation in three lipomas in a carrier of this mutation. The son of the patient had myoclonus epilepsy and ragged-red fibers syndrome. The fraction of mtDNA with the tRNA(Lys) mutation varied between 62% and 80% in cultured skin fibroblasts, lymphocytes, normal adipose tissue, and muscle. In the three lipomas the mean fraction of mutated mtDNA was 90%, 94%, and 94%. Ultrastructural examination of the lipomas revealed numerous mitochondria with changes such as electron-dense inclusions in some adipocytes. When considered cytogenetically, the lipomas were characterized by a mixture of karyotypically abnormal and normal cells. An identical del(6)(q24) was found in two tumors. The fraction of mutated mtDNA in cultured lipoma cells was the same as in the lipoma in situ, indicating that the cultured cells were representative of the primary tumor. These findings indicate that the lipomas have originated with a grossly normal stem line and subsequently have developed the 6q deletion. We conclude that the lipomas represent clonal growth of adipocytes with a high content of mtDNA with the tRNA(Lys) mutation. The tRNA(Lys) mutation may be either the direct or the indirect cause of pertubation of the maturation process of the adipocytes, leading to an increased risk of lipoma formation.

Genomic organization, sequence analysis, and chromosomal localization of the human carboxyl ester lipase (CEL) gene and a CEL-like (CELL) gene.

The gene encoding human carboxyl ester lipase (CEL), including 1628 bp of the 5'-flanking region, has been isolated and characterized from two overlapping lambda phage clones. The gene spans 9832 bp and contains 11 exons interrupted by 10 introns. The exons range in size from 88 to 204 bp, except for the last exon, which is 841 bp. A major and a minor transcription initiation site were determined 13 and 7 bp, respectively, upstream of the initiator methionine. The nucleotide sequence is identical with that of the previously reported cDNA, except for the third nucleotide in the 5'-untranslated sequence, a C, which in the cDNA is a T. A TAAATA sequence is present 26 nt upstream from the major CAP site, and within the 5'-flanking region there are several putative transcription factor binding sites. Seven Alu repetitive sequence elements are present in the region analyzed. The organization of the human CEL gene is similar to that of the recently reported rat pancreatic cholesterol esterase gene. The CEL gene was assigned to chromosome 9q34-qter, which confirms the recently reported results of Tayler et al. (1991, Genomics 10: 425-431). A previously unknown gene with a striking homology to the human CEL gene, here called the CEL-like gene (CELL), has also been isolated and characterized, including 1724 bp of the 5'-flanking region. The CELL gene, which most likely is a psuedogene, spans 4846 bp, and due to the absence of a 4.8-kb segment, the CEL gene exons 2-7 are not present in the CELL gene. Despite these differences, the CELL gene is transcribed. We have also assigned the CELL gene to a separate locus at chromosome 9q34-qter.

Assignment of the gene encoding the latent TGF-beta 1-binding protein (LTBP1) to human chromosome 2, region p12-->q22.

Latent transforming growth factor beta 1-binding protein (LTBP1) is an important component of the large latent TGF-beta 1 complex. It plays a role in the assembly and secretion of the latent TGF-beta 1. In this paper we have used a cDNA probe for LTBP1 to determine the chromosomal localization of the human gene. Using a panel of well-defined human x rodent somatic cell hybrid lines, LTBP1 could be assigned to chromosome 2. Further sublocalization of the gene to 2p12-->q22 was achieved using three hybrid lines which contain partially over-lapping fragments of chromosome 2.

Reuse of tumorous calvarial bone after gamma irradiation.

Reconstruction of calvarium after tumor resection may present several technical difficulties. The authors reused the resected calvarial bone in four patients after submitting the bone to a lethal dose of gamma radiation. The authors conclude that resected, irradiated, tumorous bone can be reused for the reconstruction of its own defect. This provides a simple method of reconstruction. Partial bone resorption should be anticipated but further reconstruction, if needed, will be facilitated.

Human 4-hydroxyphenylpyruvate dioxygenase. Primary structure and chromosomal localization of the gene.

We report the primary structure of 4-hydroxyphenylpyruvate dioxygenase [4-hydroxyphenyl-pyruvate:oxygen oxidoreductase (hydroxylating, decarboxylating)]. The work is based on the isolation of cDNA clones from human liver lambda gt11 libraries. Several overlapping clones covering the coding sequence were characterized. In parallel, peptides from four different digests of the purified protein were analysed for their amino-acid sequence. These peptide sequences covered 86% of the cDNA-derived amino-acid sequence. This gives the sequence for a polypeptide of 392 amino acids with a calculated molecular mass of 44.8 kDa. There is more than 80% identity between the human and the pig enzymes and also between these enzymes and the F antigen from rat and the two allelic forms of this antigen from mouse. The enzyme has 53% conserved amino acids and 27% identical amino acids in common with 4-hydroxyphenylpyruvate dioxygenase from Pseudomonas sp. P.J. 874 and 52% conserved and 28% identical residues, with a protein from Shewanella colwelliana. At the C-terminus there is 61% identity between the seven proteins. These results indicate that these proteins are all 4-hydroxyphenylpyruvate dioxygenases. The identity of the C-terminus makes this part of the molecule a candidate for a functional role in the catalytic process. At conserved positions in all seven enzymes, there are two tyrosine residues and three histidine residues, i.e. amino acids which have been implicated as ligands for iron in 2-oxoacid-dependent dioxygenases. The gene encoding the enzyme was localized to chromosome 12q14-->qter by Southern-blot analysis of human-rodent somatic-cell hybrids.

Mental development after modified pi procedure: dynamic cranioplasty for sagittal synostosis.

A prospective developmental assessment was performed on 26 patients operated on with dynamic cranioplasty for sagittal synostosis. Because this technique entails the application of compressive force, it was of great concern to assess the effect of surgery on development and mental status. The surgical technique used was a modified pi procedure. Perioperative variables were recorded. Six patients underwent preoperative intracranial pressure (ICP) measurements. To evaluate objectively the developmental outcome, the Griffiths' Mental Development Scales was used for analysis before and after surgery. A parental questionnaire was used for subjective outcome measurement. Preoperative ICP recordings during sleep ranged from 12.8 to 22.8 mmHg (mean, 16.1 mmHg). The mean age at the time for surgery was 6.9 months (range, 4-16 months; standard deviation [SD], 2.32 months). The surgical technique included shortening of the anteroposterior diameter of the skull by a mean of 16.6 mm. The mean global development quotient (GDQ) preoperatively was 104.5 (range, 82-144; SD, 12.4) and the mean GDQ postoperatively was 101.4 (range, 62-129; SD, 13.6). Mean age at follow-up was 16.3 months (range, 9-40 months; SD, 4.04 months). There was no significant correlation between the amount of intraoperative shortening and mental development. In comparison of means, the GDQ preoperatively did not differ significantly from the GDQ postoperatively. The modified pi procedure is safe and efficient. When surgery was performed before 1 year of age, no significant (p = 0.33) effect on mental development-either detrimental or beneficial-was demonstrated.

Identification and characterization of LTBP-2, a novel latent transforming growth factor-beta-binding protein.

Latent transforming growth factor-beta (TGF-beta)-binding protein (LTBP) is a component of the latent TGF-beta complex in human platelets. LTBP is composed of two different cysteine-rich repeat sequences, i.e. epidermal growth factor (EGF)-like repeats and a repeat containing 8 cysteine residues. The overall structure of LTBP is similar to those of the microfibrillar proteins fibrillin-1 and fibrillin-2. Here we report the identification of a novel protein termed LTBP-2, which is structurally related to LTBP. cDNA for LTBP-2 was obtained from human foreskin fibroblast cDNA libraries using a fragment of the LTBP cDNA as a probe. LTBP-2 is composed of 20 EGF-like repeats and four copies of the 8-cysteine repeat. The amino acid sequence of LTBP-2 is 41% identical to that of LTBP and 25% identical to that of fibrillin-1. LTBP-2 is synthesized as a 240-kDa protein by human foreskin fibroblasts and also by COS cells transfected with the isolated LTBP-2 cDNA. Similar to LTBP, a considerable part of LTBP-2 was found to be associated with extracellular matrix. Co-transfection of cDNAs for LTBP-2 and TGF-beta 1 revealed that LTBP-2 forms a high molecular weight complex with the TGF-beta 1 precursor. The LTBP-2 gene was assigned to chromosome 14q24. These results indicate that different forms of latent TGF-beta complexes occur and suggest that the different associated proteins may function to target the complexes to specific sites.