Cloning and mapping of murine Dgcr2 and its homology to the Sez-12 seizure-related protein

Autor: R Wadey, H O'Donnell, Catherine Taylor, Catherine Roberts, Wendy L. Kimber, Peter J. Scambler, Marie Geneviève Mattei, Anthony Wynshaw-Boris
Rok vydání: 1997
Předmět:
Zdroj: Mammalian genome : official journal of the International Mammalian Genome Society. 8(5)
ISSN: 0938-8990
Popis: Hemizygosity for a region of human Chromosome (Chr) 22q11 has been associated with a wide range of congenital malformation syndromes. The major abnormalities encountered are cardiac defects, dysmorphic facies, T cell dysfunction, clefting, hypocalcemia, and learning or behavioral problems (Wilson et al. 1993). Individually, patients may be diagnosed as DiGeorge syndrome (DGS), velo-cardio-facial syndrome (VCFS), conotruncal anomaly face (CTAF), Cayler syndrome, or Opitz GBBB syndrome. The deletions detected in these conditions are large, encompassing 2 Mb or more of 22q11. Comparisons of terminal and submicroscopic interstitial deletions have been made in order to establish a shortest region of deletion overlap for these disorders. A minimal DiGeorge critical region has recently been proposed (MDGCR, Budarf et al. 1995). A family segregating a 2;22 balanced translocation has been described (Augusseau et al. 1986). The proband, ADU, has DiGeorge syndrome, and his mother, although very mildly affected, has VCFS. Two other family members have the translocation, but there is very little clinical information available, other than they are not severely affected by either condition. The translocation breakpoint (ADUBP) maps within the MDGCR, implying that the translocation disrupts a gene haploinsufficient in DiGeorge syndrome. DGCR2 was isolated during attempts to isolate genes at or adjacent to the ADU breakpoint. The DGCR2 gene encodes a transmembrane protein (Demczuk et al. 1995). The alternative name of IDD was an acronym for Integral membrane protein, Deleted in DiGeorge syndrome (Wadey et al. 1995). The putative extracellular region contains domains with similarity to both the LDL-receptor binding domain and C-type lectins, suggesting a ligand-binding function for this part of the molecule. It has been suggested that this might involve mediation of the interaction of cephalic and cardiac neural crest cells with the substratum or with other cells during their migration. A defective neural crest cell contribution can mimic the DGS in some experimental systems (Kirby et al. 1983). However, no point mutations of DGCR2 have been detected despite extensive searches (Wadey et al. 1995, and unpublished data). Recent data have suggested that the ADUBP exerts a position effect on the gene or genes haploinsufficient in DGS, rather than directly disrupting a protein-encoding locus (Levy et al. 1995; Sutherland et al. 1996). Thus, as the closest structural gene to the ADU breakpoint, DGCR2 remains of interest despite the lack of mutations. To investigate the gene further, we have cloned and mapped the murine homolog of DGCR2. A 10.5 dpc (days post coitum) mouse embryo cDNA library was screened with the human DGCR2 clone, and three positives were obtained. The clone with the largest insert, KT4, was subcloned into M13 and sequenced. An open reading frame of 548 amino acids was detected, and the GCG program bestfit revealed that this ORF was 92% identical to the human DGCR2 sequence, with three gaps (Fig. 1a). The accession number for Dgcr2 is X95480. Database searches were repeated with BLAST and Maspar in order that similarity to recent submissions might be detected. A highly significant match was obtained with GB:D78641 over the entire length of the gene (Fig. 1a), including identity within the 38 UTR (not shown). This gene, Sez-12, is described as encoding a membrane glycoprotein and was isolated from a murine neuronal precursor cDNA library (Kajiwara et al. 1996). Sez-12 is over 99% identical to Dgcr2 at the amino acid level, although there are two gaps in the alignment (Fig. 1a). Differences between the Sez-12 and Dgcr2 sequences were checked and confirmed. The main difference is a gap of three amino acids at Dgcr2 residue 109, in the LDL binding domain of Sez-12. The human and murine Dgcr2 sequences are identical at this point, as are sequences from additional mouse clones, indicating that the sequence presented here is most likely to be correct. No other informative database matches were obtained. Partial sequence of a chick Dgcr2 cDNA was also obtained providing information concerning the aminoterminal 92 amino acids of Cdgcr2 (accession number X95885). This region contains the signal peptide and the LDL-receptor binding domain. This chick sequence was 75% identical and 83% similar to the murine Dgcr2 in this region, with two gaps. A schematic of the Dgcr2 gene showing the position of the conserved motifs is given in Fig. 1b. Northern analysis of mRNA from 10.5 to 15.5 dpc embryos detected a single transcript of 4.4kb (not shown), as demonstrated for the human gene and the Sez-12 transcripts. Comparison with the level of expression of b-actin suggests no great change in the expression of Dgcr2 takes place during this period. Expression of Dgcr2 was studied further by wholemount hybridization of 9.5 dpc embryos, in an initial attempt to identify whether Dgcr2 might be expressed in regions containing neural crest cells and their derivatives. Dgcr2 expression was detected throughout the embryo, but expression was particularly strong in the first and second branchial arches, and in the limb buds (Fig. 2). At 9.5–10 dpc, sections through wholemount embryos detected expression of Dgcr2 in the dorsal half of the mid and hindbrain, and the optic lobes (Fig. 2iii). Dorsal mesenchyme surrounding the neuroepithelium was also positive, particularly around the hindbrain. High levels of expression were found throughout the branchial arches (Fig. 2iv). There Correspondence to: P.J. Scambler
Databáze: OpenAIRE
Externí odkaz: