Abstract: In humans, common diseases show complex modes of inheritance and, as a result, have been largely refractory to genetic analysis. Rodent systems are more tractable genetically, but the mutations typically represent induced rather than naturally arising alleles, and results are often of limited direct relevance to human disease because of profound differences in physiology. By comparison, the physiology, disease presentation, and clinical response of dogs often mimic human diseases closely. In addition, the modern dog offers key advantages over other animal systems for mapping genes relevant to human disease. In the following discussion, we highlight some of these advantages and provide specific examples where canine genetics is best suited to solve difficult problems in human genetics. In particular, we focus on the strong promise of linkagedisequilibrium (LD) mapping in dogs.

Abstract: Systematics and phylogeny of the horse genetics of colour variation genetics of morphological traits and inherited disorders biochemical genetics and blood groups immunogenetics genetic aspects of disease in horses cytogenetics and physical chromosome maps genetics of behaviour biology of reproduction and modern reproductive technology genetics of peformance traits genetic improvement of the horse. (Part contents).

Abstract: Improved techniques in identifying the chromosome changes and the affected genes that are involved in acute leukemias have led to improved treatments for these diseases. Identification of consistent chromosomal changes has allowed us to target the location of particular genes and has enabled us to focus our treatments more specifically to certain subtypes of leukemia. Translocations, in particular, are common cytogenetic abnormalities in human leukemia, and the prevalence of certain types of translocations varies with age. Cancers, lymphomas and leukemias are now known to be genetic diseases and it is recognized that genotype-specific therapies should be used that take into account the genetic alterations of the particular leukemia.

Abstract: The recent completion of the DNA sequence of human chromosome 21 has provided the first look at the 225 genes that are candidates for involvement in Down syndrome (trisomy 21). A broad functional classification of these genes, their expression data and evolutionary conservation, and comparison with the gene content of the major mouse models of Down syndrome, suggest how the chromosome sequence may help in understanding the complex Down syndrome phenotype.

Abstract: The probe into the science of genetics, in particular human genetics, has in recent years increased by leaps and bounds, not only in the understanding of its intricacy, but also in its possible human application. Yet, to many—including students of medicine—much still remains a mystery if not nebulosity. Furthermore, the understanding and application of human genetics extend well beyond the boundaries of science and medicine. Instead, they border enormously into the realm of legal, moral, social, and psychological implications. The Troubled Helix: Social and Psychological Implications of the New Human Genetics is a collection of many authors’ thoughts on human genetics and aims to provide a comprehensive knowledge of the above areas.

Abstract: *Correspondence to Allen Bale, Department of Genetics, SHM I321, Yale University School of Medicine, 333 Cedar Street,New Haven, CT 06520-8005, USA; Telephone: 203-785-7610; E-mail: bale@biomed.med.yale.edu ; Fax: 203-785-7227.†Jianli Dong is currently at the Department of Human Genetics, Mount Sinai School of Medicine, New York.

Abstract: SIR,-YOU published an article (13 April, p. 113) on new hospital development under the title " Two for the Price of One." Some daily papers also derived this inference from a vague statement made by the Minister of Health at a press conference (1 April 1968) and from a booklet' concerning two new hospital projects, in which it was stated that " the overall saving may amount to 40%,h or more." It can be affirmed categorically that, using the ordinary meaning of the term " hospital capital cost," it is impossible by any architectural method to cut the cost to that extent. The Ministry of Health has not achieved the impossible ; but, starting with both the laudable aim of reasonable economy and by the exploitation of some more advanced thinking current in our profession on the future role of the district hospital, the Ministry has succeeded in altering the distribution of capital costs, some of which are transferred to other sectors to the benefit of the hospital's costing. A careful scrutiny of this wordy booklet is revealing. In the assessment of needs of acute beds for a population of 170,000 on the traditional figure of three beds per thousand people some 540 beds would be required. The project team reckoned that by using a policy of high efficiency of administration and of admissions and discharges, by the integration of community care and aftercare within the hospital, by the extended use of " day surgery," and by the fuller use of outpatient investigation (often a questionable policy) it would be feasible to reduce the acute beds to 340-that is, at the rate of two beds per 1,000 population. Thus with one blow a 33% cut is made in the major inpatient provision. The increased cost of the community services so involved is of course on another account, and the extra cost of staff (doctors and nurses) required when the degree of " nursing dependency " is heightened goes to annual expenditure. The suggestion, too, that supplementary peripheral outpatient clinics will be required postpones allocation of capital to a later date. Provision is made for only 50 geriatric beds (acute and assessment), and the rest are to be housed in upgraded existing accommodation, and the cost of upgrading will appear in another balance-sheet, while modern views of geriatric care are ignored. Other similar operations reduce the capital costs: firstly by limiting residential building to trainee nurses and essential medical and other staff, relying on the local housing authority's allocation of available accommodation, and secondly by providing some of the industrial zone functions such as laundry, workshops, and the central sterilizing supply department off the site. The description of the architectural features designed to cut costs also tends to mislead. Such flat sites with adjacent wellserviced roads do not represent average conditions. There is no scale on the sketch plans, and the reader must estimate the dimensions ; the Architects' 7ournal2 filled in some helpful details. It is clear that standards have been lowered. The wards have only 11% of single beds, which are unfunc-tionally distributed; the other patients are in six and eight open bays (wmith small day spaces) of approximately 550 sq. ft. (51 sq. jn.). Clerestory lighting is used, which can produce cold downdraugnts in the wards beneath, treatment rooms are very distant from the wards, and the traffic in lengthy communication corridors is mixed with motorized floats. The claim of flexibility in ward use reveals the planners' ignorance of the significance of the Falkirk ward.3 The hospitals occupy over five acres each, and are built on two floors with only two lifts and a service ramp. The departments most likely to grow, pathology and radiology, are closely hemmed in and have only a derisory possibility of expansion-the hallmark of poor planning. It is even suggested that if the pathology department gets overloaded some of the work could be moved elsewhere, and the idea of changing the user of this mostly expensively serviced department is viewed with equanimity. The entrance concourse, with its snack bar for everybody, shop, bank, and registration office, carries the outpatient traffic and all the two-way traffic of the dispensary, x-ray, and pathology departments and breaks every traffic rule. The records office is separated from the outpatients department by the emergency and accident department, and the approach from the hospital to the physiotherapy area looks unsatisfactory in the sketch. These low-rise buildings absolutely squander land-the Ministry ignores this. The majority of urban sites are not so extensible, and the Ministry's policy on on-costs is as shortsighted as it is wasteful in its inevitably excessive site-coverage. A notable omis'ion is any mention of the increase of porters required for such hospitals. Whatever may be one's personal views of the relative importance of the priorities of public expenditure, it is regrettable that the Ministry should describe so much false economy in such misleading terms.-I am, etc., London N.W.8. HUGH GAINSBOROUGH.

Abstract: Since its inception in 1996, Pfam has aimed to be a comprehensive database of protein families defined by the presence of shared domains.

Abstract: Chemokines and their receptors have critical roles in inflammatory and immunological responses, and thus their genetic contribution to various human disorders needs investigation. In this study, systematic variation screening of the entire coding regions of CXCR1 (IL8RA), CXCR2 (IL8RB) and CXCR3 was carried out, using genomic DNA from a large number of Japanese healthy individuals and patients with rheumatic diseases. In addition to the previously reported variations in CXCR1 and in CXCR2, two non-synonymous, two synonymous substitutions and one nonsense mutation of CXCR1, one non-synonymous and two synonymous substitutions of CXCR2, two non-synonymous substitutions of CXCR3 were newly identified. The common single nucleotide polymorphisms (SNPs) at CXCR1 codon 827 and CXCR2 codon 786 were in strong linkage disequilibrium. In addition, familial analysis indicated that human CXCR3 is located on chromosome X. No significant association was observed between the variations and the tested rheumatic diseases. However, CXCR variations identified in this study will provide valuable information for the future studies in medical sciences as well as in human genetics.

Abstract: Cleft palate is a common disease of humans, occurring in 1 in 2000 live births with heritable and sporadic forms. The complex biological events required for palate formation in the human embryo mean that disruption of various developmental steps may result in clefting (Ferguson 1987). Consequently, acquiring sufficient data to identify genetic loci involved in cleft palate formation has proven difficult. However, study of X-linked cases found in relatively inbred kindreds (Moore et al. 1987, 1988; Gorski et al. 1992) has mapped an X-linked cleft palate locus (CPX) to a 2-Mb interval on the human genome (Forbes et al. 1996). A potential mouse model for X-linked cleft palate was generated in experiments aimed at expressing the Epstein-Barr virus (EBV) LMP1 gene in transgenic mice (Wilson et al. 1990). The disrupted locus ( DXRib1) was subsequently cloned and mapped to a position close toA-raf on the mouse X Chromosome (Chr) (Wilson et al. 1993). In humans, A-raf maps to Xp11, and the CPX locus to Xq21.3 (Forbes et al. 1996), suggesting either that DXRib1 and CPX are distinct or have been rearranged relative to each other during divergence of the human and mouse X Chrs. Analysis of the site of insertion in the mouse model allowed the cloning and sequencing of the disrupted gene (Laverty and Wilson 1998). This gene exhibited tremendous sequence similarity to rat CASK (Hata et al. 1996) and was termed murine CASK (mCASK). The predicted amino acid sequences of the CASK genes indicate similarity to Ca /calmodulin-dependent protein kinase II (CaM kinase II) and membrane-associated guanylate kinase (MAGUK). In this paper, we describe the use of the mCASK clone in the isolation of its human counterpart, comparison of the human and rodent genes, mapping of the hCASK gene, and analysis of hCASK expression in human fetal and adult tissues. Human CASK was cloned by screening a lZAP cDNA library generated by poly[dT] priming of RNA from human fetal tissue at age 6 weeks (the time of palatal fusion), followed by insertion of cDNAs into theEcoRI site of pBluescript (Stratagene). The phage library was plated on the ER1647 strain of Eschericia coli;filter lifts were taken and hybridized with a radiolabeled probe generated from a 1.2-kb 3 8 RACE cloned fragment of mCASK (Laverty and Wilson 1998). Nine positive plaques were identified, purified by four cycles of screening, and recombinant phage isolated. Plasmids were excised by using the Stratagene ExAssist InterferenceResistant Helper Phage with SOLRTM Strain protocol. Restriction enzyme mapping of the recovered plasmids yielded two digestion patterns (data not shown) indicating that the purified phage most likely represented two populations. Plasmids f7 andf9 were chosen as representative of each and sequenced. (Genbank accession numbers are AF262404 and AF262405 respectively). Analysis of both hCASK sequences indicated tremendous similarity to rodent CASK; however, both clones carried completely distinct (from each other and rCASK) 5 8 sequences juxtaposed to the CASK sequences (Fig. 1C). BLAST database searching revealed similarity between the 5 8 end off7 and human Chr 5 pac clone 117n3 (Genbank accession: AC006959.1) and the 5 8 end of f9 to inversely oriented human ferritin light chain mRNA (Genbank accession No. M10119). To map hCASK, nylon filters comprising a yeast artificial chromosome (YAC) library of human genomic DNAs were probed with sequences representing an internal region of hCASK, or the 5 8 ends off7 or f9 (probes 2, 4, or 5 respectively; Fig. 1C). The position of positively hybridizing YAC clones on the human genome was identified, and adjacent markers were noted by retrieving information from the HGMP database. Probe 2 (hCASK) hybridized with YAC clones 949_A_7, 781_F_4, 796_D_5 and 705_G_6. The former two (of size 1410and 520-kb respectively) map to the X Chr and have unambiguous hits for marker DXS993 at map position 73 cM. This region is close to A-raf (Xp11) and indicates that the CASK locus is distinct from the human X-linked cleft palate locus, CPX (Xq21.3). During preparation of this report, the hCASK gene has been mapped to a location consistent with our observations (position Xp11.4), by using human–hamster radiation hybrid mapping panels (Dimitratos et al. 1998). Our independent mapping of the CASK locus is strong evidence for the veracity of this observation and provides confirmation that the CASK locus is distinct from that of CPX. Clone 796_D_5 mapped ambiguously to Chr 10, and 705_G_6 was unmapped. It is possible that 796_D_5 may represent a pseudogene or gene similar to CASK on Chr 10 or that the ambiguous mapping data are inaccurate. Probe 4 (5 8 end of f7) identified 20 positives, 9 of which were from the same area of Chr 5 (consistent with BLAST data). Probe 5 (5 8 end of f9) resulted in five positives, two of which (753_G_9 and 763_E_2) provided unambiguous hits for map position 11 cM on Chr 20, close to one of the mapped positions for the ferritin light chain. The other clones were either ambiguous hits or unmapped. These observations imply that the juxtaposition of 5 8 sequences of 7 andf9 with CASK is due to aberrant ligation in generation of the cDNA library. Comparison of the CASK-related f7 andf9 cDNA sequences shows differences at regions 1 and 2 (see Fig. 1 and Table 1). Similar variation at region 1 has been noted between variant mCASK-A and -B sequences and identified as an alternatively spliced exon (Laverty and Wilson 1998; Table 1). Comparison of f7 andf9 cDNAs with rodent CASK sequences and two independently generated hCASK sequences (Genbank accession No. AF032119, Cohen et al. 1998; Genbank accession No. AF035582, Zha and Hu, unpublished) highlights a number of other sites at which sequence variation occurs (indicated in Fig. 1B and re* Present address:CID School of Biological Sciences, The Stopford Building, University of Manchester, Oxford Road, Manchester M13 9PT, UK.

Abstract: Preface. Preface to the First Edition. Chapter 1. Understanding Human Disease. Chapter 2. the Genetic System: Chromosomes. Chapter 3. the Genetic System: Mendel's Laws of Inheritance and Genetic Linkage. Chapter 4. The Molecular Biology of the Gene. Chapter 5. Recombinant DNA Technology. Chapter 6. Genetic and Physical Mapping of the Human Genome. Chapter 7. Discovering Human Disease Genes. Chapter 8. Bioinformatics: Genomics, Functional Genomics, and Proteomics. Chapter 9. Human Population Genetics. Chapter 10. Molecular Genetics of Complex Disorders. Chapter 11. Genomic Imprinting: An Epigenetic Modification. Chapter 12. Molecular Genetics of Mitochondrial Disorders. Chapter 13. Molecular Genetics of Muscle Disorders. Chapter 14. Molecular Genetics of Neurological Disorders. Chapter 15. Molecular Genetics of the Eye. Chapter 16. Molecular Genetics of Cancer Syndromes. Chapter 17. Counseling, Diagnostic Testing, and Management of Genetic Disorders. Glossary. Index.

Abstract: This is an impressive work ^ a substantial two-volume book with high-quality illustrations of over 800 pages. Perhaps what is most remarkable is that it is written by only two authors, in contrast to other texts of metabolic disease. This has the advantage of a higher degree of consistency in format and style. The authors state that the book aims to provide ``in a comprehensive and concise text the clinical symptomatology, cardinal manifestations, laboratory diagnostic features, molecular aspects and pathology diagnostic features of metabolic disease'', with an emphasis on the clinical and pathological expression of disease. Is it possible for only two authors to have a su¤ciently in-depth knowledge of all metabolic disease to write a comprehensive text? For the more recent discoveries the text is not always up-to-date. For example, it is stated that the location of the occipital horn syndrome (OHS) gene on the X chromosome is not known, whereas it has been known from 1994 that both OHS and Menkes disease are due to mutations within the same gene (ATP7A). There are other areas where I would judge the book to be not completely accurate. However, I do not wish to detract from the achievement of the authors. For the major pathological features and clinical presentations the book appears to do well and the overall presentation is excellent.

Abstract: The PROCRUSTES server provides a method for determining protein-coding sequences in genomic DNA.

Abstract: For those interested in genome-wide or more restricted comparisons of proteins across species, the Clusters of Orthologous Groups (COGs) website provides the kind of information many of us want.

Abstract: channels (VDCCs) are found in the plasma membranes of all excitable cells and are hetero-oligomeric complexes composed of up to five subunits:a1, a2, d, b and g. The a1 subunit acts as a channel pore, voltage sensor, and receptor for many drugs. The other accessory subunits modify channel properties such as current magnitude and kinetics (Walker and De Waard 1998). Thea2 andd chains are derived by proteolytic cleavage of the same gene product, the d subunit acting as a membrane anchor for the extracellulara2 subunit. Three genes encoding a2d subunits have been described: human, mouse, rat, and rabbit CACNA2D1 (a2d1), human CACNA2D2 ( a2d2), and mouseCacna2d3(a2d3) (Klugbauer et al. 1999). Genes encoding VDCCs have been implicated in the aetiology of a wide range of mammalian phenotypes (Fletcher et al. 1998). The genes encoding the a1A, b4, and g2 subunits have been shown to underlie the mouse epilepsy phenotypes tottering, lethargic, and stargazer (Fletcher and Frankel 1999). Recently, we have demonstrated that the Cacna2d2gene is also associated with a phenotype of ataxia and epilepsy in ducky mice (unpublished). TheCacna2d2gene is located on mouse Chromosome (Chr) 9 approximately 59–60 cM from the centromere, and the human ortholog CACNA2D2 is located on human Chr 3p21.3 (Gao et al. 2000). Here, the genomic structures of the mouse Cacna2d2and human CACNA2D2 genes are described and compared for the first time. The two regions of alternative splicing identified in mouse brain RNA are discussed together with a consideration of the associated non-consensus splice sites. The 5.5-kb human CACNA2D2 cDNA (GenBank AF042792) sequence has been previously described, and we have isolated the mouseCacna2d2cDNA sequence (GenBank AF247139). Genomic clones containing the Cacna2d2gene were identified from the WI/MIT YAC library (y203E7, y257D12, y465F1) and the RPCI21 mouse PAC library (p428C5, p432G2, p524O8, and p524G24). Most of the intron/exon boundaries were determined by using the ExpandTM Long Template PCR system (Roche Diagnostics, UK) to amplify mouse genomic and YAC DNA with primers contained within the cDNA sequence. PCR products were sequenced to determine the positions of the intron/exon boundaries. Smaller introns were sequenced in their entirety and exact sizes determined. Sizes of larger introns (1–8 kb) were estimated by comparison of PCR products to size standards. Introns 1 and 2 could not be amplified by PCR; therefore, intron/exon boundaries were determined by direct sequencing of the PAC clones by using Big-Dye technology (Applied Biosystems, UK). These intron sizes were estimated by Southern blot hybridization of digested PAC clones.Cacna2d2is organized into 39 exons (Table 1 and Fig. 1A) that are distributed over 85 kb of genomic DNA. Exon 1 contains the start codon, and exon 39 contains the stop signal. The mouseCacna2d2cDNA sequence was aligned with the genomic sequence derived from a cosmid contig of human Chr 3p21.3 (GenBank Z84493, Z84494, Z84495, Z75743, Z75742, and Z84492). The positions of the mouse intron/exon boundaries were compared with regions of divergence between the mouse cDNA sequence and the human genomic DNA sequence. The positions of all the intron/exon boundaries and exon sizes are conserved between human and mouse, so CACNA2D2 is also organized into 39 exons over a genomic distance of at least 118 kb (Table 1 and Fig. 1B). Overall, the genomic organizations of the two genes are highly similar. Two regions of alternative splicing were identified in mouse brain RNA (Fig. 2). Exon 23 (Fig. 2A) is present in 13% of subclones of an RTPCR product spanning exons 22–24. At the protein level, it results in the sequence change KYF to KLPISKLKDF. A splice variant at the same position has been described in human cDNA (Hobom et al. 2000). The 5 8 splice donor does not conform to the consensus gt, but contains a gc in both the mouse and human sequences (Table 1). Such a variation at the 58 splice site has been previously described (Shapiro and Senapathy 1987) and is not expected to significantly impair splicing. Presumably, this represents a mechanism whereby the recognition of the splice site is slightly reduced, consistent with the relatively low abundance of RNA species containing exon 23. The second region of alternative splicing involves the 3 8 splice acceptor of intron 38. The 6 bp indicated (Fig. 2B) was identified in 25% of subclones from an RTPCR product spanning exons 37–39 and results in the insertion of two amino acids (CP). Thus, two 38 splice donor sites are associated with intron 37. The more favored more 3 8 splice site contains the consensus ag motif. The second splice site contains a non-consensus at motif, again suggesting that divergence from the consensus produces alternatively spliced products of relatively lower abundance. This nonconsensus splice site is conserved in humans, and the same splice variant has been described in human medulla thyroid carcinoma cells (Hobom et al. 2000). Sequence analysis of full-length Cacna2d2cDNA clones (n4 9) and in silico analysis of cDNA sequences have not identified transcripts that contain exon 23 and the 6-bp variant of exon 38. Tissue-specific splicing of human CACNA2D2 has been described, although functional expression of splice variants did not reveal differences at the electrophysiological level (Hobom et al. 2000). Analysis at the protein level is required to provide definitive proof that the alternatively spliced transcripts represent native proCorrespondence to: J. Barclay; E-mail: jane.barclay@pharma.novartis.com

Abstract: The genetic mapping of various mammalian species, including human, over the last 10 years has demonstrated the existence of conserved syntenies between these species. The technique of chromosome painting has largely confirmed this fact (Wienberg and Stanyon 1997; reviewed by Chowdhary et al. 1998). It has, therefore, become important to carry out comparative mapping of human and domestic animal species, not only to use maps with a high gene density, such as in human or mice, to improve those of domestic animals, but also to use our knowledge of the mammalian genome gained from animal mapping to improve the human map (O’Brien et al. 1997). Our previous reciprocal chromosome painting study of humans and pigs (Goureau et al. 1996 and the Internet site 〈h tp://www. toulouse.inra.fr/lgc/pig/compare/compare.htm 〉) allowed us to make fragment-by-fragment matches between human and porcine chromosomes over most of the porcine geneome. However, in many cases we were unable, when using the chromosome painting technique, to establish the precise localization of the synteny breakpoints either on the human or the porcine chromosomes. Thus, some chromosome bands of one species may show matches with one, two, three, or even four chromosome regions of the other species: for example, Sscr 7q13 and 7q14 bands match both human Chromosomes Hsap 6 and 15 (Fig. 1a). Now comparative mapping consists of deducing the position of a gene in one species from the one that it occupies in the other species, and this lack of precision could compromise the quality of results when selecting possible candidate genes. We therefore decided to improve this aspect in order to be able to make the best use of human mapping when trying to identify porcine genes of economic interest. We selected Chr 7 as it carries genes that may be of use in pig breeding. In particular, several QTL (qualitative trait loci) for growth rate and carcass qualities have been located there (Rohrer and Keele 1998a, 1998b; Wang et al. 1998). Monodirectional studies with chromosome painting (humans over pigs, Rettenberger et al. 1995; Fro ̈nicke et al. 1996) or bidirectional ones (Goureau et al. 1996) have demonstrated two facts. First, porcine Chr 7 carries conserved syntenies on human Chrs 6, 14, 15 (Fig. 1a). Second, the chromosome segments of human Chrs 6, 14, and 15 that do not correspond to porcine Chr 7 do correspond to porcine Chr 1, namely, the p arm match with Hsap 6 and the q arm match in two contiguous conserved syntenies with Hsap 14 and 15 (Fig. 1b). All these results were in line with the comparative mapping data available at the time (Yerle et al. 1995). The work of these three teams therefore allows us to identify synteny breakpoints on Sscr 7 denoted 7.1 to 7.4 from pter to qter (Fig. 1c). Nevertheless, Fig. 1c shows that there is no agreement concerning the localization of these synteny breakpoints (the same is true for the synteny breakpoints on Sscr 1 and Hsap 14 and 15). Moreover, each experiment found overlapping of conserved syntenies on Sscr 7. Thus, there is great imprecision in the matching of the chromosome bands Sscr 7q and Sscr 1q14-q25 with the human chromos me bands. The gene IGF1R located on human Chr 15 in q25-q26 is located on porcine Chr 1, whereas these two human chromosome bands were demonstrated by chromosome painting to match Sscr 7. Finally, results published after our painting study (Goureau et al. 1996) was completed concerning the localization of the genes IGA and MOK2 (Frönicke et al. 1996; Jørgensen et al. 1997; Thomsen et al. 1998) point to the existence of possible matches between Sscr 7 and Hsap 19 that painting has not detected. Nevertheless, the gene named IGA by the authors (Thomsen et al. 1998) is in fact IGHA, mapped on Hsap 14q32-33. Concerning MOK2 (Kruppel zinc finger protein 2), it belongs to the wide gene family which codes for zinc finger proteins. The gene mapped by Jørgensen (1997) is probably one member of this family, but is not orthologous to the one mapped on Hsap 19. Consequently, the correspondence between Sscr 7 and Hsap 19 has not been demonstrated and will have to be confirmed by mapping additional markers. In order to specify the localization of the synteny breakpoints on porcine Chrs 1 and 7 on one hand, and on human Chrs 6, 14, and 15 on the other hand, we have selected human YACs (Chumakov et al. 1995 and the internet site of the CEPH 〈http:// www.cephb.fr/infoclone.html 〉) specific to one or two human chromosome bands and used them as probes in heterologous FISH experiments. As we thought that in heterologous hybridization the signals would be very weak, we first used pools of YACs (2 to 4) specific to the same human chromosome band. In each pool, at least one YAC had already been located by FISH on human chromosomes (Bray-Ward et al. 1996) and was used to identify the pool. The others were selected from the same stack or from an adjacent one (Chumakov et al. 1995). Subsequently, some individual YACs were successfully used. The YACs we selected and their human cytogenetic localizations are listed in Table 1. The probes were mainly produced by random priming labeling of 300 ng of DNA extracted from yeasts containing the YACs. In two cases, our probes were human inserts separated from the vector by PFGE (Kingsley et al. 1997), which were most kindly provided by T. Haaf (Max-Planck-Institute of Molecular Genetics, Berlin, Germany). The amplification of the inserts and their labeling were carried out by using DOP-PCR (Telenius et al. 1992). All the probes used in heterologous hybridizations were pretested on human chromosomes to verify their localization and the absence of chimerism. The heterologous hybridization conditions were developed from those previoulsy used for reciprocal chromosome painting (Goureau et al. 1996), but were modified as follows: hybridization buffer with 30% formamide, hybridization for 96 h at 37°C, washes at 37°C with 30% formamide in the first three baths. The signal quality was good or very good for both homologous and heterologous conditions (see Table 1). For each probe, meaCorrespondence to: A. Goureau; E-mail: andre.goureau@educagri.fr Mammalian Genome 11, 796–799 (2000). DOI: 10.1007/s003350010158

Abstract: Sirs, Approximately 30% of human malignant gliomas display allelic loss of 22q, with deletion mapping suggesting the presence of multiple tumor suppressor candidate regions [1]. hCHK2 is a human homologue of the yeast Rad53 gene, which plays a role in control of the G2-M cell cycle checkpoint. The hCHK2 gene maps to 22q11, and is mutated in the germline of some Li-Fraumeni syndrome families, including patients with malignant gliomas [2]. hCHK2 therefore represents a particularly attractive candidate for the 22q glioma suppressor gene since: (1) it maps to a region of 22q relevant to malignant gliomas; (2) it is altered in the germline of patients who are predisposed to malignant glioma formation; and (3) it encodes a cell cycle control checkpoint protein, and other cell cycle control genes are known to be altered in malignant gliomas [3]. To address this possibility, we screened the hCHK2 gene for mutations in 51 sporadic malignant gliomas (45 glioblastomas, 3 anaplastic astrocytomas, and 3 anaplastic oligoastrocytomas). We selected 26 tumors that had allelic loss of 22q at the D22S275 locus, which maps within hCHK2, but also studied tumors that had not lost this region of 22q, since the only sporadic tumor cell line noted to have an hCHK2 mutation retained the wild type allele [2]. Single-strand conformation polymorphism (SSCP) analysis was performed as published, with some modifications [4]. Amplicons for exons 1 through 9 used primer sequences employed previously for mRNA-based screening [2], with restriction digestion to obtain suitable fragment lengths for SSCP analysis (Table 1). For exons 10 through 14, new primers were designed to facilitate analysis, with primers placed within flanking intron sequences (Table 1). Exons showing shifts were reamplified and directly sequenced. Exon 1 had a silent change of adenosine to guanosine at nucleotide 252 in 10% of cases, as reported previously [2], and a one-base pair deletion in a run of five adenosines in intron 1, 43 base pairs downstream of the exon-intron border in 29% of cases; these changes were also present in corresponding blood DNA. No aberrant shifts were detected in exons 2 through 11. Exons 12, 13, and 14 displayed recurrent aberrant band patterns that were identical in both tumor and germline specimens in over 30% of cases. These presumably represent polymorphisms, which are likely to be intronic since they were not detected on prior mRNA-based screening [2]. We conclude that hCHK2 is not the target of somatic inactivation in malignant gliomas. Some tumor suppressor genes predispose to tumor formation when altered in the germline, but are not frequent targets of somatic mutation in the corresponding sporadic tumors. For instance, in malignant gliomas, defects in the mismatch repair enzyme genes may lead to Turcot syndrome, in which colorectal carcinoma and malignant glioma both occur, but somatic mismatch repair gene defects are rare [3]. hCHK2, which encodes a cell cycle checkpoint protein activated by DNA damage and aberrant DNA replication, appears to fall into this class of glioma tumor suppressor gene.

Abstract: We have isolated the human homolog of a novel rodent gene that may be involved in the regulation of pituitary gene transcription. The human PREB gene encodes a predicted protein of 417 amino acids, exhibiting several sequences characteristic of the WD-motif protein family. PREB transcripts were detected in every human fetal and adult tissue examined, although a great variation in levels of expression was observed. PREB was mapped to human Chromosome 2p23, a region of the genome associated with partial trisomy 2p syndrome. Although variable, the common duplication phenotype includes facial abnormalities, skeletal defects, growth and mental retardation, congenital heart and neural tube defects, and abnormalities of the genitalia. We propose that PREB has a role during human development and that abnormal dosage of this transcription factor may be involved in some of the developmental abnormalities observed in patients with partial trisomy 2p.