ESHG Posters 7

Flanking microsatellite markers and their association with instability of the FMR-1 CGG repeat, involved in the fragile X syndrome, were analyzed in DNA samples from Belarus population. Comparison of DXS548-FRAXAC1-FRAXAC2 haplotype frequencies in the 189 normal, 22 intermediate and 9 fragile X chromosomes suggested a strong linkage disequilibrium between normal alleles and haplotype 7-3-4+ as well as between expanded alleles and haplotype 2-1-3. Haplotype 2-1-3 was reported to be significantly enriched among high-end normal alleles. In our study we also revealed a number of closely-allied haplotypes (1-1-3, 2-1-2, 3-1-3) which are almost exclusively linked with “gray zone” CGG repeats. We suggest that this group could arise from the founder chromosome with haplotype 2-1-3 and intermediate size FRAXA allele, by the series of single mutational events, by mechanism similar to replication slippage. It means that such chromosome must have a higher rate of replication mistakes in this region or inability of repair system to correct changes naturally occurring in repetitive sequences. Remarkably, that the same mechanism was proposed for gradual progression of intermediate alleles towards instability threshold. Instability in both FMR-1 and the flanking markers in Xq27 can be influenced by a common trans-acting factor promoting general instability in microsatellite sequences. It can also explain the existence of the haplotypes which are rare or absent in the control samples but are well represented in intermediate and fragile X chromosomes.

Possible involvement of chi-like and minisatellite sequences in the formation of two chimeric CYP21P/CYP21 genes in steroid 21-hydroxylase deficiency

H. Lee¹, D. Niu², C. Lin³;
¹Yuan-Shan Research Institute, I-Lan, TAIWAN REPUBLIC OF CHINA, ²Department of Pediatrics, Veterans General Hospital-Taipei,, Taipei, TAIWAN REPUBLIC OF CHINA, ³Kingcar Food Industrial Co., Ltd. Yuan-Shan Research Institute, I-Lan, TAIWAN REPUBLIC OF CHINA.

Congenital adrenal hyperplasia (CAH) is a common autosomal recessive disorder mainly caused by defects in the steroid 21-hydroxylase (CYP21) gene. More than 90% of CAH cases are caused by mutations of the CYP21 gene. Approximately 75% of the defective CYP21 genes are generated through intergenic recombination termed “apparent gene conversion” from the neighboring CYP21P pseudogene. A chimeric CYP21P/CYP21 gene with its 5’ end corresponding to CYP21P and 3’ end corresponding to CYP21 has been identified. This kind of gene is nonfunctional because it produces a truncated protein. We have found that there are two distinct chimeric genes in CAH patients. Both had a sequence with -300 nucleotides of the 5’ head as the CYP21P gene. Otherwise, the coding region consisted of a fusion molecule with the CYP21P gene in two different regions. One of the junctions was located in chi-like sequence of GCTGGGC in the third intron and the other was in the minisatillite consensus TGGCAGGAGG of the exon 5 of the CYP21P gene. In addition, the analysis of restriction fragments length polymorphism for these two 3.3-kb chimeric molecules showed these sequences arose as a consequence of unequal crossover between the CYP21P and CYP21 genes. It is plausible that both consensus sequences may be responsible for the gene conversion of these two chimeric genes.

Sertoli cell-germ cell interaction: Functional analysis of murine calgizzarin gene.

A. U. Mannan, G. Nica, K. Nayernia, W. Engel;
Institute of Human Genetics, Göttingen, GERMANY.

In the present study we undertook functional analysis of a differentially regulated gene, named calgizzarin, which shows a decreased expression in murine Sertoli cell-germ cell cocultures compared to cultured Sertoli cells alone. Calgizzarin is expressed in low level in all adult tissues examined; however, a high mRNA level for calgizzarin in mouse testis is maintained until stage P15 and then declines dramatically, whereas the expression pattern in the ovary remains constantly high throughout development. Therefore its expression in Sertoli cells might be repressed by some unknown factor/s originated from spermatocytes/spermatids. The gene is approximately 6 kb pairs in size and contains three exons. The ORF encodes for a 98 amino acid polypeptide containing two putative EF-hand calcium-binding domains. To elucidate the potential role of calgizzarin gene, we decided to delete the gene in mouse by targeted disruption (knock-out). A replacement-targeting vector was designed to delete the two exons encoding calgizzarin protein and replaced them with the neomycin phosphotransferase gene. The ES cell were transfected with the targeting vector and selected for homologous recombination event. Five recombinant clones were identified by genomic Southern. Two clones produced germ line transmitting chimeras after injecting them into blastocyst derived from C57B/J6 female. These chimeras were bred with C57B/J6 female to generate F1 animals, which were heterozygous for calgizzarin deleted allele. Generation of homozygous animal is awaited. The phenotypic and molecular analysis of these mice might give us insights, about the potential role of calgizzarin in spermatogenesis.

Novel mutations in the RFXANK gene. Promotor binding of mutant RFX without MHC II transactivation.

W. Wiszniewski¹, M. Fondaneche², P. Louise-Plence³, C. Picard⁴, J. Bal⁵, A. Fischer², B. Lisowska-Grospierre²;
¹Institute of Mother and Child, Warsaw, POLAND, ²INSERM U 429, Paris, FRANCE, ³INSERM U 475, Montpellier, FRANCE, ⁴INSERM U 550, Paris, FRANCE, ⁵Institute Mother and Child, Warsaw, POLAND.

MHC class II deficiency or bare lymphocyte syndrome is a combined immunodeficiency caused by defects in MHC specific regulatory factors that control MHC II expression at transcriptional level. MHC II expression is controlled at least by four trans-acting genes: CIITA, RFXANK, RFX5 and RFXAP. RFXANK encodes a subunit of the tripartide RFX transcription complex that functions in the assembly of multiple transcription factors on MHC II promoters. It has four ankyrin repeats important for interaction with RFXAP and CIITA. So far seven different RFXANK mutations have been reported in 26 unrelated patients. The most frequent mutation - deletion of 26 bp (752delG-25) was identified in 17 patients. The other mutations are nonsense or splice site mutations leading to proteins lacking all or part of the ankyrin repeat region. Here we report two novel point mutations of RFXANK gene: D121V and R212X found in two families with BLS and additional studies on tyrosine residues 224Y and 235 Y located in fourth ankyrin domain. We found D121V allele is expressed in vitro but is unable to form, with other factors, stable RFX complex. The experimental mutant Y224A is able to form RFX complex which binds to the MHC II promoter in vitro but cannot reverse BLS phenotype. These data the importance of the D121 and Y224 residues deduced from modelling studies.

I. M. Hanson, A. G. Brown;
University of Edinburgh, Edinburgh, UNITED KINGDOM.

Members of the SIX gene family are involved in a variety of developmental processes in vertebrates and invertebrates. All SIX genes encode two highly conserved motifs, a homeodomain that mediates DNA binding, and a SIX domain that mediates protein-protein interactions. It is now clear that Drosophila has three SIX genes (sine oculis, Optix and D-Six4) while humans have six (SIX1-6). Phylogenetic analysis of Drosophila and mammalian amino acid sequences shows that the proteins fall into three clear subgroups, suggesting that a common ancestral organism possessed three SIX genes. An intriguing feature of the human SIX genes is their genomic organisation, as revealed by the Human Genome Sequencing Project; five of the six genes are clustered. We have shown that the mouse SIX genes appear to have a similar genomic arrangement. Despite extensive amino acid homology within the SIX-domain, SIX proteins differ dramatically in their interactions with transcriptional co-factors encoded by the EYA gene family. We report work in progress which shows that while SIX1 interacts strongly with EYA2 and EYA4 in a yeast two-hybrid assay, SIX3 interacts with neither. This implies that there are clear functional differences between the subgroups of SIX proteins.

P. Greiser¹, S. Schulz¹, U. Schagdarsurengin¹, T. Süß¹, D. Rehfeld¹, A. Nordwig¹, A. Kabisch², U. Müller-Werdan³, K. Werdan³, C. Gläser¹;
¹MLU, Inst. of Human Genetics, Halle, GERMANY, ²MLU, Bloodbank, Halle, GERMANY, ³MLU, Dep. of Internal Med., Halle, GERMANY.

The potent cytokine FGF2 is involved in the proliferative response to vascular injury of many cell types and is therefore suggested to be an important candidate gene for CAD. Materials and methods: We investigated the individual FGF2 protein- (ELISA; R&D-System) and mRNA-expression (competitive RT-PCR) in 99 patients with CAD (49.9y, SD 8.7, 85 males). Moreover we examined the potential role of the exon1-polymorphism C223T of FGF2 in regulating the mRNA- and protein-expression pattern. This polymorphism, located in the 5’-UTR in the functionally important ribosome entry site, modifies the calculated mRNA folding-structure and may therefore influence the expression of FGF2. Results: Investigating the FGF2 expression on transcriptional and translational level we determined an mRNA expression of 13.23 ag/cell and a protein expression of 49.19 pg/ml serum in our patient group. Furthermore a significant positive correlation of FGF2 mRNA and protein expression could be detected (p<0.001). The study of the genotype distribution of the C223T-polymorphism resulted in the following frequencies: 0.84 (CC); 0.14 (CT); 0.02 (TT). An analysis of the influence of the C223T-polymorphism on the FGF2 expression revealed significant correlations on transcriptional (p<0.01) as well as on translational level (p<0.001). The highest values of both mRNA- and protein-expression were determined among CC-carriers (14.3ag/cell; 54.4pg/ml serum), whereas CT-carriers showed intermediated values (8.3ag/cell; 23.2pg/ml serum). The lowest values were found among TT-carriers: 3.3ag/cell; 14.5pg/ml serum. Conclusions: The FGF2 mRNA- and protein-expression was shown to be significant dependent on the C223T-polymorphism. Further investigations regarding the FGF2 expression pattern should bear this association in mind.

J. Mykkänen, M. Toivonen, M. Kleemola, A. Peippo, K. Rantanen, M. Savontaus, P. Aula, O. Simell, K. Huoponen;
University of Turku, Turku, FINLAND.

The SLC7A7 gene encodes a cationic amino acid transporter y+LAT1 expressed mainly in basolateral compartment of epithelial cells in intestine and renal tubules. Mutations of SLC7A7 cause a rare but severe aminoaciduria, lysinuric protein intolerance (LPI). We have investigated putative promoter regions of SLC7A7 using luciferase reporter gene and electrophoretic mobility shift assays (EMSA). The expression level of the gene in LPI patients and controls was also studied.
The 5' region of the first untranslated exon of SLC7A7 contains no classical promoter elements like TATA-box or Sp1 binding sites. However, it includes several other putative transcription factor binding sequences, like CACATG and CCCCTGGC, sufficient to promote luciferase expression in human embryonic kidney cells (HEK-293). The rise of the luciferase activity was 10-fold in comparison with the negative control; no activity was observed in fibroblasts. Accordingly, by EMSA we demonstrated that the sequence elements were able to bind protein in HEK-293 cells and adult kidney tissue extract, but not in fibroblasts. Northern blot analysis showed very low and equal SLC7A7 mRNA levels in control, carrier and LPI patient fibroblasts.
In conclusion, it is likely that identified transcription factor binding sites contribute to the low expression level and the epithelial specific regulation of the SLC7A7 gene.

D. Zink¹, S. Haas², E. Coward², M. Vingron², B. Korn³;
¹German Cancer Research Center, Heidelberg, GERMANY, ²Max-Planck Institute for Molecular Genetics, Berlin, GERMANY, ³RZPD, German Resource Center, Heidelberg, GERMANY.

Public EST databases currently contain more than 3 million human EST sequences, representing probably 40-50.000 human genes. Within these data exists a large redundancy. We take advantage of this redundancy by analysing the differences of sequences belonging to the same gene. The EST sequences are clustered and assembled to a consensus sequence. However, many clusters cannot be assembled into a single consensus sequence. The sequences then fall into multiple consensus sequences (contigs) within one clusters. The differences might be due to imperfect sequence data (e.g. partially unspliced sequence templates) or due to alternative splicing. Instead of one gene coding for one mRNA leading to one protein, alternative splicing of transcripts may lead to potentially different proteins.
Splice variants are often due to alternative exon usage, which we verify by RT-PCR.
We have set up a medium throughput strategy that does allow us to screen expression of genes in 31 different human tissues. We initiated this project by analysing genes on Chromosome 21 and 22.
Our results indicate, that the theoretical data represented in EST databases can be verified in many cases by our experimental design. We re-sequence PCR products in question, to confirm their origin and nature. In more than 35% of the cases, we cannot experimentally support EST data by RT-PCR.
In future we want to extend splice variant analysis to other chromosomes and gene families. We intend to automate RT-PCR and ultimately design a chip to discriminate a large number of different splice forms of medically relevant genes.

RET upstream sequences modulate gene expression via cell-line specific chromatin acetylation.

F. I. Puppo¹, P. Griseri¹, M. Fanelli², F. Schena¹, G. Romeo³, I. Ceccherini¹, P. Pelicci⁴, R. Ravazzolo¹, G. Patrone¹;
¹G. Gaslini Institute, Genoa, ITALY, ²University of Camerino, Camerino (MC), ITALY, ³International Agency for Research on Cancer, Lyon Cedex, FRANCE, ⁴European Institute of Oncology, Milan, ITALY.

Disregulation of the RET proto-oncogene transcription may play a role in both inherited cancer syndromes and Hirschsprung disease. Understanding the gene regulation might provide new clues to clarify pathogenesis. Recently, we reported that RET transcription is highly cell-line specific, while the promoter region is equally active in different cell-lines.
Here we show that RET upstream sequences can modulate gene expression via cell-line specific chromatin acetylation level contributing to promoter function. Acetylation and deacetylation activities, working on specific lysines of histonic tails, alter the accessibility of transcription factors to DNA thus modulating gene expression. Histone deacetylase inhibitors, such as sodium butyrate, up regulate both RET mRNA levels and transcription rate in a RET expressing cell-line. The same treatment allows transcript detection in RET negative cell lines (lymphoblasts), while no enhancement is seen in cells expressing RET already at high level. Sensitivity to sodium butyrate appears to be sequence specific in transient transfection assays. Chromatin immunoprecipitation experiments, using antibody against H4 tetra-acetylated hystones, confirm the direct role of histone acetylation level within RET upstream region, in regulating gene expression.
Sodium butyrate derepressive effect allowed us to overcome the lack of this gene expression in the majority of adult cells and analyse RET mRNA from Hirschsprung patients, by treating lymphoblastoid cell lines. Anomalous transcripts, defective gene expression, as well as nucleotide substitutions in the coding region were detected.

N. M. Raevskaya, L. V. Dergunova, I. P. Vladychenskaya, S. A. Limborska;
Institute of Molecular Genetics RAS, Moscow, RUSSIAN FEDERATION.

We have investigated the structure of the genomic clone Ghfb hybridizing with the brainspecific cDNA clone Hfb1 and obtained from the cosmid chromosome V library. We have shown Hfb1 to be a fragment of the 3'-terminal exon of the human synaptic protein complexin 2. This exon encodes the extended 3'-untranslated mRNA region. Sequencing of the genomic clone Ghfb let us determine the complete structure of three exons (3'-terminal and two neighboring) and two corresponding introns of the complexin 2 gene. The nucleotide sequence of three named exons includes the partial 5'- untranslated region, the ORF and 3'-untranslated region of the corresponding mRNA. Nucleotide sequence of the clone from the human genomic sequences database (AC010241) showing the highest similarity to Ghfb allowed us to determine the sequences of two another exons and two corresponding introns. These exons represent the 5'-end of the 5'-untranslated region of the complexin 2 mRNA. Computer analysis of the human EST database proposed the existence of an additional exon within the intron 2 sequence at the distance of about 7 kb from the exon III. The resulting alternative transcript is thought to differ with its 5'-end from that for the previously described transcript. Using RT-PCR technique we confirmed the existence of the predicted transcript.

The comparison of substrate recognizing regions of Cytochrome P450 2C subfamily enzymes suggests the diversity-enhancing evolution of these regions during the evolution.

The enzymes belonging to cytochrome P450 2C subfamily metabolize numerous drugs in liver. Four enzymes (2C8, 2C9, 2C18 and 2C19) members of this subfamily have been identified in humans. These enzymes play roles in metabolizing various toxic compounds. Considering the diverse nature of the chemical structure of these compounds, it is reasonable to assume that the substrate recognizing regions of these enzymes undergo diversity-enhancing evolution just as the antigen-binding regions of major histocompatibility complex. The test for the diversity-enhancing evolution is performed by the comparison of synonymous (silent) nucleotide changes versus nonsynonymous (amino acid altering) nucleotide changes during the molecular evolution. If nonsynonymous substitution is estimated to have occurred more frequently than synonymous substitution, then we suspect that diversity-enhancing evolution might have occurred. Cytochrome P450 2C subfamily has 6 substrate recognizing regions. By the analysis using a window with 30 codons, nonsynonymous substitutions were found to augment near the substrate recognizing regions. When the comparison was made between synonymous and nonsynonymous changes in cytochrome P450 2C subfamily enzymes, the diversity-enhancing substitutions were observed in different substrate recognizing regions. The hypothesis of diversity-enhancing evolution was tested by the simulation experiments and it was supported in almost cases with significance. Similar results were obtained in rabbit, rat and golden hamster. These results suggest that the substrate recognizing regions of cytochrome P450 2C subfamily enzymes have undergone diversity-enhancing evolution due to the selection pressure favoring the diversity adapting to the environment in which a variety of possibly toxic compounds attack the species.

A. Reigo¹^,2, A. Metspalu¹^,2;
¹University of Tartu, Institute of Molecular and Cell Biology, Tartu, ESTONIA, ²Estonian Biocentre, Tartu, ESTONIA.

Genes encoding for C2H2 zinc finger proteins are known to regulate normal cell proliferation and differentiation and are often involved in tumor growth regulation. Therefore we chose this protein motif consensus as a virtual probe for in silico analysis of EST databases and genomic sequences from 9p22, a chromosomal region of our interest. Assembling physically clustered positive sequence fragments we detected some putative exons of an uncharacterised gene, which were then used for primer design to screen transcript pools from 12 tissues and 8 cell lines generated by RT-PCR. The corresponding full-length cDNAs from human peripheral blood leukocytes were sequenced and the exon-intron structure of the gene was determined. Two main transcripts of the novel gene were visualized by Northern blot hybridization.
The human gene designated as lymphocyclin contains 8 exons spanning approximately 450kb of genomic DNA in 9p22.2-22.3 between markers D9S156(tel) and D9S1218(cen). It has two main transcripts of 3.82kb and 0.84kb and also a splice variant lacking exon 2 (126nt). The longest transcript encodes a 1099aa protein with three pairs of adjacent C2H2 zinc fingers close to its C-terminus and has three putative nuclear localization signals. Other significant features of lymphocyclin are a 22aa leucine zipper motif and an alpha-helix forming a serine stripe.
Our experimental data so far indicate quite restricted expression of this new member of C2H2 zinc finger protein gene superfamily and refer to its possible role as lymphocyte differentiation and/or proliferation regulator.

J. Cocquet¹, E. DeBaere², F. Jaubert³, X. Xuhua⁴, L. Messiaen², E. Pailhoux⁵, C. Cotinot⁵, M. Fellous¹, R. Veitia¹;
¹Département d'immunologie Institut Pasteur, Paris, FRANCE, ²Department of Medical Genetics, Ghent University Hospital., Ghent, BELGIUM, ³Service d’Anatomopathologie, Hôpital Necker-Enfants Malades, Paris, FRANCE, ⁴Bioinformatics Laboratory, HKU-Pasteur Research Centre, Hong Kong, HONG KONG SPECIAL ADMINISTRATIVE REGION OF CHINA, ⁵Laboratoire de Biologie du Developement et Biotechnologies, INRA, Jouy en josas, FRANCE.

Mutations in the FOXL2 gene have recently been shown to cause the blepharophimosis-ptosis-epicanthus inversus syndrome (BPES), a rare genetic disorder. There are two types of BPES. In BPES type I, a complex eyelid malformation is associated with premature ovarian failure (POF) while in BPES type II, the eyelid malformation occurs as an isolated entity. The FOXL2 protein belongs to the family of forkhead transcription factors. These proteins have a highly conserved 100 amino-acid DNA-binding domain (the forkhead domain) and are involved in different developmental processes. The FOXL2 protein also contains a polyalanine tract that might have a role in transcriptional repression.
In order to study the FOXL2 locus (ORF and promoter), we have performed a comparative analysis of the sequence in several species including human, goat and mouse. The amino-acid sequences are highly similar and conservation is further demonstrated by low Ka/Ks ratios. From the alignment of the 5’upstream sequences in these species, we have deduced the sequence of a "core" promoter.
In addition, we have produced and characterized two polyclonal antibodies directed against the N- and C-terminal regions of the human FOXL2 protein which are also capable of recognizing the other orthologs. Immunohistochemistry shows that FOXL2 is a very early marker of ovarian development as it is expressed in follicular cells. The domain of expression of FOXL2 seems to restrict as development proceeds in the goat ovary.

B. Marzec, T. Kubiatowski, D. Szczesniak, M. Gawlowicz, J. Wojcierowski;
Department of Human Genetics,University School of Medicine, Lublin, POLAND.

Infection of B-lymphocytes with Epstein-Barr Virus in vitro induces a G0 to G1 transition followed by DNA synthesis and cell division. Infected cells undergo blast transformation resulting in the outgrowth of immortal lymphoblastoid cell lines. Numerous cellular proteins are switched on in the infected cells, including p53. DNA viruses encode oncoproteins that modulate the level of p53 and stimulate host cell DNA synthesis. It was reported earlier that p53 can be bounded by EBV gene products: EBNA5 nad BZLF-1 leading to its accumulations. We have investigated expression of p53 and bcl-2 genes on the level of its mRNA and protein in LCL human cell lines established from the blasts of EBV infected patients. For estimations of mRNA we have used RN-ase Protection Assay (h-CC2 multiprobe template set), the protein level was studied immunocytochemically. In our study we have observed increased level of p53 mRNA and protein expression, similary to observed in mitogene stimulated peripherial blood lymphocytes from healthy donors.
The increased level of p53 and bcl-2 expression in EBV infected LCL may suggest that transcription of p53 might be activated by EBV infection. As long as it is physiologically tolerable level it doesn't lead to growth arrest or apoptosis but promote proliferation.

Characterization, expression pattern, and chromosomal assignment of the novel human RAB22A gene belonging to Rab small GTPases

D. Schlote, S. Kussmann, S. Worch, I. Hansmann;
Institut für Humangenetik und Medizinische Biologie, Halle / Saale, GERMANY.

The mouse chromosome 2 segment (MMU2) corresponding to human chromosome 20 (HSA20) is known to be involved in both, maternal as well as paternal noncomplementation (genomic imprinting). Uniparental disomies for distinct regions of MMU2 result in different neonatal lethalities with opposite anomalous phenotypes, strongly suggesting the presence of imprinted genes in this region. These chromosomal regions show a conserved synteny of gene loci to human 20q13 segment, predicting the presence of imprinted genes in this syntenic human chromosomal region. We have identified a new gene in this region of interest which is located on a BAC RPCIB753L051096 proximal to GNAS1 on HSA 20q13. Sequencing the full-length cDNA revealed a novel isoform of the human RAB22 subfamily of small GTP-binding proteins which are located in distinct intracellular compartments and plays an important role in the regulation of vesicular trafficking. Based on the EST WI-12997 this new isoform was isolated containing 2242 nucleotides and is designated RAB22A. Structurally, the RAB22A encodes a polypeptide of 194 amino acids which has 97% identity to the canine rab22. Northern blot analysis revealed ubiquitous expression slightly increased in heart. The genomic structure was completed by database analysis and sequencing of the isolated BAC clone RPCIB753L051096. The gene consists of 7 exons spanning about 50 kb of genomic sequence. Physical and FISH mapping revealed that RAB22A is located proximal to GNAS1 but downstream of PCK1 on human chromosome 20q13. Supported by DHGP.

Phylogeny of the alternative NF1 exon 10a-2 reveals that intron sequences surrounding alternatively spliced exons are highly conserved

B. Bartelt, O. Kenner, R. Müller, W. Vogel, D. Kaufmann;
University of Ulm, Ulm, GERMANY.

We recently found an additional splice product of the Neurofibromatosis type 1 gene (NF1) with the new 45 bp exon 10a-2 inserted between exon 10a and 10b. Phylogenetic analysis of gDNA revealed the origin as well as the demise of exon 10a-2 in evolution. 10a-2 homologous sequences are not present in Drosophila melanogaster but in Fugu rubripes. It is expressed and highly conserved in birds and mammals but independently lost again in mouse, rat, sheep, cow and dog. Intronic sequences surrounding exon 10a-2 are highly conserved over more than 600 bp among mammals expressing 10a-2. We therefore suggest a specific function of exon 10a-2 and the surrounding intronic sequences lost again in some mammalian species. Extended intron homology, uncommon for NF1, was also found around three other alternatively spliced NF1 exons, one of which is rodent-specific (23b). Investigation of two other genes, the cystic fibrosis transmembrane conductance regulator (CFTR) and the Wilms tumor 1 (WT1) gene, also revealed high intron sequence conservation around the alternative CFTR exon 10b and WT1 exons 5 and 9. We therefore speculate that high conservation of the intronic sequences surrounding exons is related to alternative splicing.

S. Jacolot¹^,2, G. Le Gac¹^,2^,3, C. Mura¹^,2;
¹Laboratoire de génétique moléculaire et génétique épidémiologique INSERM 0115, Brest, FRANCE, ²Université de Bretagne Occidentale, Brest, FRANCE, ³Etablissement Français du Sang, Brest, FRANCE.

The human HFE gene is clearly involved in hereditary hemochromatosis a common genetic disorder in caucasian populations which is characterized by duodenal iron hyperabsorption. However, the precise function of the HFE protein and its role in the pathogenesis of the disease are still unknown. A possible role in homeostatic iron regulation is suggested by the co-localization and the interaction of the HFE protein with the transferrin receptor. We recently examined the functional organization of the HFE gene promoter that did not revealed any iron responsive element. Here, we studied the HFE gene expression in Caco-2 cells along with differentiation and holo-transferrin treatment. We used Caco-2 cells that differentiate with morphological and biochemical features of mature intestine enterocytes as a model to study the HFE mRNA expression. We examined the HFE mRNA level using real-time PCR in Caco-2 human intestinal cells grown on polycarbonate membrane inserts under different transferrin-saturated iron conditions. Preliminary studies indicate that HFE mRNA expression increases untill cells reach confluence then the level decreases after confluence. HFE mRNA is up-regulated when holo-transferrin was added to the basolateral compartment during 48h before RNA extraction, however the up-regulation is observed only till cell confluence. In conclusion, the iron status seems to play a role in HFE expression before differentiation of intestinal cells that may be involved in the iron sensing to regulate iron absorption.

The interrelationship between type of the copper nutrition and activity of the genes supporting copper balance in newborn rats during development

N. Platonova¹, R. Povalikhin¹, B. Mischenko², T. Zhivul'ko¹, E. Zhiguleva¹, N. Tsymbalenko¹, L. Puchkova¹;
¹Research institute of experimental medicine, St.-Petersburg, RUSSIAN FEDERATION, ²St.Petersburg State Technical University, St.-Petersburg, RUSSIAN FEDERATION.

The effect of copper (Cu) nutrition on genes expression controlled the copper turnover during mammal development has been studying. The methods of Northern- and Western-blot hybridization, immunoelectrophoresis, different enzymatic assays and atomic absorption spectrometry were used. A group of newborn rats were fed with baby formula (Cu contains as an inorganic salt) during 8 days from the 1st day after birth (experimental group). The Cu and ceruloplasmin (Cp) levels in their serum were increased 3 times and the liver Cu concentration was decreased 2 times when compared with rats which were fed by rat Dams (Cu packed into milk Cp, control group). The liver Cp-mRNA content was increased and Wilson ATPase gene expression was appeared in liver of the experimental group. The effect of baby formula was distinctly appeared in 48 hours independent of starting fed of newborn rats (from first or eighth day of life). The brain Cu concentration was not changed, but both Cp and Cu levels in cerebrospinal fluid of experimental rats were increased 7 times. Cu was progressively accumulated during first 12 days of life and it was found in nuclei and lysosomes, but its level was already decreased to 5th day of experiment. The Menkes and Wilson ATPases as well as Cp genes expression were mainly expressed in hypophysis and choriodic plexus but not in cerebral cortex and cerebellum of 5-days old rats of both groups.
The roles of milk Cp and Cp gene expression in mammary glad in newborn is discussed.

E. Makrinou¹, M. Antoniou², J. Collinge¹, J. Collinge¹;
¹University College London, London, UNITED KINGDOM, ²King's College, London, UNITED KINGDOM.

Prion protein (PrP) is intimately linked with a class of neurodegenerative diseases known as transmissible spongiform encephalopathies, which in humans includes Creutzfeldt-Jakob disease (CJD), Gerstman-Sträussler-Scheinker disease (GSS), Kuru and the recently recognised form of variant CJD (vCJD). Employing bioinformatics and direct molecular analysis we demonstrate that the human PrP gene (PRNP) locus, which is situated at chromosome position 20p12-ter, consists of a functional domain of approximately 55 kb containing three genes; PRNP, DOPPEL or PRND located 20 kb 3’ of PRNP and a novel gene, designated HSM8 that maps 3 kb 3’ to PRND and which is transcribed to generate at least three alternatively spliced mRNAs. All three genes of this locus show a similar two-exon structure with the protein coding region restricted to the larger exon 2, but low sequence homology implying that although they may be evolutionarily related they are functionally distinct. Analysis of both adult and foetal human tissues confirmed the ubiquitous but variable expression profile of PRNP with highest levels observed in the CNS and testis. Contrastingly, although PRND shows a wide tissue expression pattern in foetal tissues, it is exclusively expressed in adult testis whereas all three HSM8 isoforms were only detected in adult testis implying that PRND is developmentally regulated. An investigation of the regulatory mechanisms underlying this complex gene expression pattern from the PRNP locus should provide insight into the function of these genes and their involvement in prion protein diseases.

N. Bogdanova¹, W. Schempp², A. Markoff³, B. Dworniczak¹, J. Horst¹;
¹Institute of Human Genetics, Münster, GERMANY, ²Institute of Human Genetics and Anthropology, Freiburg, GERMANY, ³Intitute of Medical Biochemistry,ZMBE, Münster, GERMANY.

PKD1 is the first gene responsible for the condition of autosomal dominant polycystic kidney disease (ADPKD) and its gene product is called polycystin 1. There are several homologous genes (HG) to PKD1 which are located proximally to the master gene on the same chromosome. The 3' regions of the HG contain the previously described chromosome 16-specific low-copy number repeat element, mapped to chromosome 16 p13, p12, and q22. They share a high degree of homology (95% - 97%) to PKD1 over a large genomic segment. We recently showed that these genes are pseudogenes, duplicated in the course of molecular evolution.
It is known that homologues to PKD1 are not present in mouse and lower species. We therefore sought to perform searches for PKD1 homologous sequences in primates in order to reveal at which stage of evolution HG appeared. For this purpose we performed FISH analysis on chromosome spreads from human, chimp, gorilla and orang-utan using BACs containing human PKD1 homologues genes. The expected pattern of signals was obtained by human and gorilla whereas chimp and orang-utan gave additional signals on different orthologs of the human chromosome 7. The orang-utan lacks in addition the signal on the long arm of human chromosome 16 ortholog.
Sequencing analysis of chimp-PKD1 gene revealed more than 99% identity between the master gene and its homologue(s) in the analysed regions. The implication of these results on our knowledge about the evolution of the PKD1 homologues and their possible function in primates will be discussed.

Simultaneous expression of two murine clustered ADAM family genes, Testase 2a and Testase 2b, during testicular germ cell development .

E. Bolcun-Filas¹, K. Nayernia¹, P. Grzmil², T. Rzymski¹, W. Engel¹;
¹Institut of Human Genetics, University of Goettingen, GERMANY, ²Department of Genetics and Evolution,Jagiellonian University, Krakow, POLAND.

The ADAM (A Disintegrin and A Metaloprotease domain) family presents the best characterized candidates for mediating gamete interaction and membrane fusion in mammals. In this study we describe the expression patterns of two clustered ADAM family genes for testase 2a and testase 2b. Previous studies showed that murine testase 2 (ADAMA 25) gene is expressed specifically in testis. However, we found two different restriction patterns of subcloned fragment of the gene, suggesting presence of two novel testase 2 transcripts. Further experiments and Celera database search revealed that these two transcripts are the products of two separate clustered genes, which show high similarity to the published testase 2 ( 87.8% and 95.4%, respectively) and in 87,4% to each other. Both genes are located on chromosome 8 (as well as ADAM 3, ADAM 5, ADAM 9) in close distance of 24 kb. Genomic structure of testase 2a and b is different from other ADAM family members, like cyritestin or fertilin which possess around 20 short exons. They are composed of only two exons. The first exon is about 85 bp while the second is over 2.5 kb long. Both genes demonstrate the same temporal and spatial expression pattern during testicular germ cell development with onset of expression in haploid stages. Further functional analysis of these molecules will contribute not only to a better understanding of the molecular mechanisms underlying mammalian sperm-egg fusion but also to the development of new methods for both, fertility regulation and diagnosis and treatment of human infertility.

J. P. Jakupciak¹, A. Bacolla², A. Jaworski², J. E. Larson², C. D. O'Connell¹, R. D. Wells²;
¹National Institute of Standards & Technology, Gaithersburg, MD, ²Institute of Biosciences and Technology, Houston, TX.

The genetic defects in autosomal dominant polycystic kidney disease (ADPKD) cases are mutations in PKD1, a gene that encodes a transcript of 14 kilobases from 46 exons spanning 50 kbp on chromosome 16p13.3. The 2.5kbp poly(purine-pyrimidine)(R-Y) tract in intron 21 of the PKD1 gene may contribute to its high mutation frequency. The rate of somatic mutations must be high given the frequent occurrence of ADPKD and the very large number of cysts observed, suggesting the existence of a local hot spot for mutagenesis. The poly(R-Y) tract is one of the ten longest sequences of this kind; it is 66% G-C-rich with 95% C+T in the sense strand and is partly repeated in introns 1 and 22. The tract contains 23 mirror repeat sequences, which would be expected to adopt three-stranded DNA structures with stems of at least 10 bp. In addition, 163 direct repeat sequences were identified, which may adopt slipped, mispaired conformations. These DNA conformations were recognized as lesions and were cleaved by the NER system. Current work focuses on evaluating the frequency and types of mutations induced by the R-Y tract.

Cloning of a novel gene on 3q25-26 with homology to nonsense mediated mRNA decay (NMD) proteins via a RA-induction gene trap approach.

O. Batista¹, W. Wurst², I. Hansmann¹, M. Schlicker¹;
¹Institut für Humangenetik und Medizinische Biologie, Halle (Saale), GERMANY, ²Max-Planck Institut, München, GERMANY.

For better understanding retinoic acid (RA)/retinoid (RX) induced teratogenesis we set up an RA-induced gene trap approach using mouse ES clones (Forrester et al. 1996). One of these trapped clones showed repressed RA-induced LacZ reporter gene expression. By 5´ RACE-PCR a fusion transcript was identified which after cloning revealed 1081bp upstream sequences of the integration site. A 1.754 kb cDNA was established from clones identified by screening mouse cDNA libraries. This cDNA was used to isolate a human cDNA by screening a fetal brain Marathon cDNA library. Sequence homology between both cDNAs with an ORF of 1515 nucleotides is 90%. The putative protein contains 503 aas which are 94% identical between both species. This protein shows high homology to NMD3 proteins from yeast species and to proteins conserved in drosophila, caenorhabditis, arabidopsis and leishmania. We isolated a mouse P1 clone and a human BAC clone and established genomic structure of the gene and characterized 5´ and 3´ sequences. The human approx. 33 kb gene contains 15 exons and 14 introns. Northern analysis revealed a 2.5 kb transcript in mouse embryos from day 8-15 pc in heart and craniofacial tissue of newborns as well as in heart and brain of adult mice. Northern analysis of human RNA revealed 2 transcripts of 2.7kb and 3.0kb in heart, liver, pancreas, skeletal muscle and kidney and weaker transcripts in placenta, brain and lung. The novel and potentially RA-repressed gene was FISH mapped to chromosome 3q25q26. (O.B. is a recipient of a DAAD fellowship.)

A T-box containing transcription factor on human chromosome 15q14 identified through a retinoic acid-induction gene trap approach.

A. Voigt¹, M. Weickert¹, W. Wurst², I. Hansmann¹, M. Schlicker¹;
¹Institut für Humangenetik und Medizinische Biologie, Halle (Saale), GERMANY, ²Mack-Planck Institut, München, GERMANY.

Prenatal exposure to retinoic acid (RA) and retinoids (RX) is related with specific congenital anomalies. The molecular mechanism for this teratogenesis and the genes involved are not well known. To identify such RA/RX downstream genes we set up an RA-induced gene trap approach based on mouse ES clones (Forrester et al. 1996). From a sample of trapped ES clones with RA-induced reporter-gene repression 5´-RACE-PCR identified a fusion transcript of 1.135 kb in clone EScD1. Screening of cDNA libraries and in silico analysis revealed an approx. 9 kb mouse cDNA. This gene represents a transcription factor containing a T-box and a bHLH Zip-domain. Northern analysis revealed tissue specific expression of this single locus gene with transcripts of varying size. A prominent 9.5 kb transcript is present in spleen, kidney, stomach, lung, brain, testis, ovary, placenta, heart and skeletal muscle. Weaker and smaller transcripts of 7.2 kb were observed in placenta and testis and of approx. 4 kb in testis indicating alternative splicing. A genomic P1 clone isolated contains the complete cDNA. Screening of human cDNA libraries and in silico analysis identified the corresponding human gene with sequence homology >82% to mouse cDNA. BACs were isolated for establishing genomic structure and mapping. The gene contains 24 exons, the same domains as the mouse gene and maps to 15q14, according to FISH. The gene was found to be triplicated in a young proband with mental as well as growth retardation and malformations due to partial trisomy 15q14-q15.

L. E. D'Urbano¹, S. Guida¹, E. Mantuano², L. Veneziano², M. Frontali², C. Jodice¹;
¹Tor Vergata University of Rome, Rome, ITALY, ²INMM,CNR, Rome, ITALY.

The CACNA1A gene, coding for the a1A-voltage-dependent calcium channel subunit type P/Q, is responsible for Episodic Ataxia type 2 (EA2), Familial Hemiplegic Migraine (FHM) and Spinocerebellar Ataxia type 6 (SCA6).
Several mutations causing these diseases have been described, and they account for aminoacid substitutions, protein truncations and small polyglutamine expansions. In addition, families segregating the disease with the CACNA1A, but not showing any mutations in the coding region, have also been found. For this reason we started the characterization of the regulating regions of this gene as possible sites for mutations.
Through bioinformatic analysis of sequences upstream the first coding exon of the gene, two candidate regions have been identified. Primer extension experiments have shown a band of about 300 bp suggesting a transcription starting site at 250-300 bp upstream to exon 1. The characterization of this region is ongoing.
The CACNA1A gene shows a considerable complexity of ribotypes. Different isoforms may be associated with different channel activity, as demonstrated in the rabbit. The characterization of different splice variants and of splice enhancer sites will increase the chances to detect non coding mutations in FHM or EA2 patients.
Comparison of different cDNA clones selected with CACNA1A specific probes from an adult cerebellar and a total fetal brain cDNA libraries, revealed the presence of new isoforms. Characterization of isoforms at the 5’ end has been performed.
This work was supported by Telethon grant E847.

S. Yakut¹, S. B. Karauzum¹, F. C. Sargin¹, O. Taskin², S. Tacoy³, G. Luleci¹;
¹Akdeniz University, Faculty of Medicine, Department of Medical Biology and Genetics, Antalya, TURKEY, ²Akdeniz University, Faculty of Medicine, Department of Obstetrics and Gynecology, Antalya, TURKEY, ³Akdeniz University, Faculty of Medicine, Department of Pediatrics, Antalya, TURKEY.

Ribozomal protein S4 (RPS4X) gene is located on chromosome X and escapes from X inactivation. Thus, insufficient expression of RPS4X may play a role in development of Turner syndrome, the complex human phenotype associated with monosomy X. In this study we have investigated RPS4X gene expression levels in peripheral blood of 21 patients with 46,XX karyotype and Turner phenotype or its predominant feature primary amenorrhea by RT-PCR, while two 45,X individuals were studied as control subjects to be able to compare the expression levels as we expect lower levels of RPS4X gene expression in 45,X patients due to a single X chromosome.We found different expression levels in 7 of our 21 patients, 6 of which were diagnosed to have primary amenorrhea while 1 had a Turner phenotype, yet observed an increased level of expression in one of the 45,X control patients as well. Based on our results, we debate whether haploinsufficieny of RPS4X is the cause of Turner syndrome or not, and plan to perform sequence analysis to investigate any possible sequence defects which could be responsible for the change in the level of expression in RPS4X gene in our cases. Our results were discussed comparatively with the previously reported data.

Disruption of KCC2 reveals an essential role of K-Cl-cotransport in synaptic inhibition.

C. A. Hübner¹^,2, V. Stein¹, I. Hermans-Borgmeyer¹, T. J. Jentsch¹;
¹ZMNH, Hamburg, GERMANY, ²Institut für Humangenetik, Hamburg, GERMANY.

Synaptic inhibition is crucial for the control and modulation of neuronal activity. Disturbing the interplay between excitation and inhibition causes various neurological disorders. GABA and glycine are the main inhibitory neurotransmitters of the adult central nervous system. Synaptic inhibition by GABAA and glycine receptors, which are ligand-gated Cl- channels, depends on the intracellular chloride concentration ([Cl-]i). High [Cl-]i can lead to excitatory GABA responses that are deemed to be important during development. Several potassium-chloride cotransporters can lower [Cl-]i, including the neuronal isoform KCC2, which was substantiated by antisense experiments in vitro. Analysis of the expression pattern of KCC2 during murine embryonic and postnatal development by in situ hybridization and Western blot analysis, shows that KCC2 parallels neuronal differentiation and precedes the functional GABA switch. Neonate KCC2 knockout (Kcc2-/-) mice die due to severe motor deficits including loss of respiration. Sciatic nerve recordings reveal abnormal spontaneous electrical activity indicating a spastic disorder. Spinal cord responses to peripheral electrical stimuli are altered in Kcc2-/-mice as observed in the mouse mutant spastic. In wild-type animals, immunofluorescence and electron microscopy demonstrated KCC2 expression close to inhibitory synapses. Patch-clamp measurements of spinal cord motoneurons demonstrated an excitatory GABA and glycine action in the absence, but not in the presence of KCC2. This shows that the functional GABA/glycine switch in the spinal cord occurs earlier than in the hippocampus. It depends crucially on the expression of KCC2, and is indispensable for the normal function of motor circuits already at birth.

N. E. Gyulikhandanova¹, O. V. Voronina¹, V. S. Babich¹, A. V. Vasin², L. V. Puchkova¹;
¹Research Institute for Experimental Medicine, St. Petersburg, RUSSIAN FEDERATION, ²St. Petersburg State Technical University, St. Petersburg, RUSSIAN FEDERATION.

The numerous of tissue specific extracellular molecular forms of ceruloplasmin (Cp, multicopper ferroxidase) and its membrane isoforms: glycosylphosphatidylinositol-anchor Cp, receptor Cp and intracellular membrane Cp-like ferroxidase, play a central role in copper transport through body as well as iron metabolism. Perhaps they are the products of the single Cp gene copy in haploid chromosome number whose mutations lead to various neurodegeneration disorders. In this work the structure analysis of Cp gene promote region (4000 bp upstream transcription start point) was carried out by the original programs for identification and mapping cis-elements potentially taking part in gene regulation. The sequence specific sites for the nuclear receptors of 9-cis-retinoic acid and thyroid hormone and sites for specific protein expression in liver, lung and mammary gland were found and these localization were mapped. The structure and localization of these sites are very similar in rat and human. In vivo experiments shown that estradiol stimulated Cp transcription and translation in the rat liver to 3 folds. The selective interaction of the hormone responsible elements with soluble nuclear proteins isolated from liver and brain newborn and adult rats was demonstrated by gel-shift assays. The similar results were obtained for transcription factors isolated from rat mammary gland during lactation.
The role of Cp gene tissue specific activity in the safe turnover of the copper and iron in mammals is discussed.

Investigation of chromosomal DNA loop organization within a region of human chromosome 16q22.1

S. Shaposhnikov¹, E. Frengen², P. Thomsen³, V. Zverev¹, H. Prydz²;
¹Research Institute for Viral Preparations, Russian Academy of Medical Sciences, Moscow, RUSSIAN FEDERATION, ²Biotechnology Centre of Oslo, University of Oslo, Oslo, NORWAY, ³The Royal Veterinary and Agricultural University, Copenhagen, DENMARK.

We have previously generated a 2.8 Mb high-resolution map surrounding the LCAT gene cluster on human chromosome 16q22.1 (Frengen, Rocca-Serra, Shaposhnikov, et al., Genomics 70:273-285, 2000). We suggest that the tight organization of the LCAT gene cluster has biological significance. The domain organization of this chromosome region is currently analyzed by COMET-FISH and by the topoisomerase II-mediated DNA loop excision protocol. COMET-FISH is gel-electrophoresis of DNA from a single nucleus, which is lysed on microscope slide to produce nucleoids. The nucleoids are electrophoresed extending the loops into tails. Subsequent FISH allows mapping of specific DNA sequences/genes from the LCAT region relative to the DNA loops. We are also using a complementary approach, which is based on the ability of cellular topoisomerase II to cleave DNA in the presence of several anticancer agents. The cleavage sites are associated with regions of DNA attachment to the nuclear skeleton, thus indicating DNA loop anchorage sites. Analysis of the topoisomerase II cleavage sites within the region therefore permits mapping of the domain organization. Human cells have been exposed to topoisomerase II inhibitor, the DNA from the cells was cleaved with restriction enzymes, separated by pulsed field gel electrophoresis, blotted, and hybridized with probes from our map. This dual approach allows exploration of the topological organization of the region of human chromosome 16q22.1 and permits correlation of the structural architecture to the functional organization of the DNA.

A. Andre'¹, A. D'Angelo¹, E. Marra¹, V. Aglio¹, S. Olivieri², G. Arrigoni², A. Ballabio¹, M. Zollo¹;
¹Telethon Institute of Genetics and Medicine (TIGEM), Naples, ITALY, ²HSR San Raffaele, Laboratorio di Anatomia Patologica, Milan, ITALY.

Transgenic mice have proven to be a powerful system to study normal and pathological functions of genes. We have previously reported the isolation and characterization of the human PRUNE gene. Furthermore we have found by “in vitro” experiments that overexpression of prune increases proliferation in NIH3T3 cells. The human PRUNE gene is located in the 1q21.3 chromosomal region, in the epidermis cluster locus where two skin disorders have been mapped: (AD) atopic dermatitis and psoriasis. Prune is expressed in the adult human skin and, in particular, in the basal and granular layers. For these reasons we have cloned the human PRUNE cDNA under the control of a peculiar loricrin promoter, which drive the expression of the transgene both in the undifferentiated and differentiated layers of the skin. Four founders lines have been obtained and morphological studies are in progress by Immunohistochemistry (IHC) analysis with anti-prune Ab, keratinocytes markers (K14, K1, K6) and proliferative markers (example Ki67). BrdU “in vivo” uptake analysis and apoptosis TUNEL assays will be performed to define if prune may interfere with apoptosis. Furthermore previous unpublished results have shown a defect in the skin of SCID mice by the use of retrovirus technology. A dominant negative mutation of prune protein (PRUNE-D) affecting its PDE activity, results in a specific alteration of the cell cycle proliferation and differentiation of the SCID hair follicles. For this reason the mutated prune gene has been cloned under the same promoter and used for the generation of a dominant negative transgenic mouse.