| Human Molecular Genetics | Pages |
©1999 Oxford University Press |
MID2, a homologue of the Opitz syndrome gene MID1: similarities in subcellular localization and differences in expression during development
Introduction
Results
Identification of the human MID2 cDNA
Sequence analysis of MID2
MID2 maps to Xq22 and has a similar genomic structure to MID1
Identification of the murine Mid2 cDNA
Mid2 maps to the mouse syntenic region
Expression studies
MID2 localizes to microtubules
Discussion
Materials And Methods
cDNA identification and genomic structure
cDNA sequence analysis
Expression studies
Genetic mapping in the mouse
Cell culture, transfection and immunofluorescence
Acknowledgements
References
MID2, a homologue of the Opitz syndrome gene MID1: similarities in subcellular localization and differences in expression during development
Received April 27, 1999; Revised and Accepted June 8, 1999
The B-box family is an expanding new family of genes encoding proteins involved in diverse cellular functions such as developmental patterning and oncogenesis. A member of this protein family, MID1, is the gene responsible for the X-linked form of Opitz G/BBB syndrome, a developmental disorder characterized by defects of the midline structures. We now report the identification of MID2, a new transcript closely related to MID1. MID2 maps to Xq22 in human and to the syntenic region on the mouse X chromosome. The two X-linked genes share the same domains, the same exon-intron organization, a high degree of similarity at the protein level and the same subcellular localization, both being confined to the cytoplasm in association to microtubular structures. The expression pattern studied by RNA in situ hybridization in mouse revealed that Mid2 is expressed early in development and the highest level of expression is detected in the heart, unlike Mid1 for which no expression was detected in the developing heart. Together, these data suggest that midin and MID2 have a similar biochemical function but a different physiological role during development.
INTRODUCTION
The B-box protein family is characterized by the presence of a tripartite motif that includes two zinc fingers (the RING and the B-box) and a coiled-coil domain. In addition, a conserved region which was first described in the Ret finger protein and hence named RFP-like domain, is often associated to the tripartite motif. These domains are found in different combinations; however; their order from the N- to the C-terminus of the proteins is conserved in all cases. Genes belonging to this protein family are thought to act as transcriptional regulators and have been implicated in a variety of processes. In particular, the Xenopus laevis nuclear factor 7 (Xnf7) is involved in the dorsal-ventral patterning of the Xenopus embryo (1), while another member of the family, the PwA33 gene, appears to have a role in transcription during oogenesis (2). Three genes, PML, RFP and Tif1, have oncogenic potential in man and mouse when involved in a chromosomal translocation resulting in the fusion of the tripartite motifs with other proteins (3-6).
The B-box gene family has also been implicated in the pathogenesis of inherited disorders such as familial Mediterranean fever (FMF; MIM 249100). FMF is characterized by recurrent attacks of inflammation in the peritoneum, synovium or pleura. The gene responsible for this disorder, MARENOSTRIN/PYRIN, was isolated by positional cloning by two different groups and encodes a B-box protein (7,8).
Recently, a positional cloning effort led our group to the identification of MID1 (midline defect 1), the gene responsible for the Xp22-linked form of Opitz G/BBB syndrome (OS; MIM 145410), and showed it to be a member of the B-box family (9). The same gene was also isolated by two different groups (10,11).
OS is genetically heterogeneous, with one locus mapped on the distal part of the short arm of the X chromosome and one form linked to the D22S345 locus on chromosome 22q11.2 (12), and the two forms are not distinguishable on a clinical basis (13).
Several members of the B-box gene family, including PML (14-16), Tif1 (6) and RFP (17), were detected within subnuclear structures which for PML were named PML nuclear bodies, while midin recently has been shown to be localized to the cytoplasm and to be associated with microtubules (18,19).
Here we report the identification and functional characterization of MID2, a new gene closely related to MID1. In addition to sequence similarity, midin and MID2 also share the same subcellular localization, but they have a different expression pattern during development.
RESULTS
Identification of the human MID2 cDNA
We exploited the availability of a public expressed sequence tag database (dbEST) to identify human homologues of the MID1 gene. The full-length MID1 cDNA sequence (GenBank accession no. Y13637) was used as a query to search the dbEST database.
One EST (AA016125; ze31b07.r1 Soares retina N2b4HR Homo sapiens cDNA clone) which was homologous, but not identical, to MID1 was identified. The IMAGE cDNA clone 360565 corresponding to EST AA016125 was entirely sequenced and used to screen both a fetal brain and a teratocarcinoma/neuron cDNA library. Three different cDNA clones were isolated and characterized by end sequencing, restriction mapping and PCR, using specific primers in combination with vector primers. Characterization of each of the cDNA clones allowed us to establish a cDNA contig of 2524 bp (GenBank accession no. Y18880) (Fig. 1). The authenticity of the 5[prime] end of the cDNA was validated by sequencing the corresponding genomic region. The putative initiation codon was identified at position 258 and is located within a nucleotide sequence that fulfils Kozak's criteria for an initiation codon (20). The first in-frame stop codon (TAA) was identified at nucleotide 2403, predicting a protein product of 715 amino acids (Fig. 1).
Figure 1. Schematic representation of the human MID2 cDNA contig. The four cDNA clones analysed in detail are shown as thick black bars. The size of the coding region (dashed rectangle) and of the 5[prime]- and 3[prime]-UTRs (thin lines) are indicated. Also indicated are the ESTs corresponding to the MID2 transcript.
Sequence analysis of MID2
Bestfit analysis revealed that the newly identified transcript is highly homologous to the MID1 gene and was thus named MID2 (HGMW-approved gene symbol and gene name). Overall MID1 and MID2 display 84% similarity and 77% identity at the protein level, and 70% identity at the nucleotide level. Sequence analysis of the predicted amino acid sequence using BLAST2 software (21) showed that, like midin, MID2 displays a RING finger, two B-boxes, a coiled-coil and a conserved RFP-like C-terminal domain (Fig. 2).
Figure 2. Bestfit analysis between the MID2 (top) and the midin (GenBank accession no. Y13667) predicted protein products (bottom). The RING finger domain is indicated by a solid line, the B-box domain by a dashed line, the coiled-coil domain by a thick line, and the RFP-like domain is boxed at the C-terminal portion of the protein. The cysteine and histidine residues in the RING finger and in the B-box domains are in bold.
The RING finger is a conserved cysteine-rich domain of 40-60 residues (called C3HC4 zinc finger) (22,23) probably involved in mediating protein-protein interactions. The spacing of the cysteines in this domain is CXXCX(9-39)CX(1-3)HX(2-3)CXXCX(4-48)CXXC, where X may be any amino acid. As shown in Figure 2, the presence and the spacing of the cysteine and histidine residues is conserved between midin and MID2 in this domain (amino acids 10-59). Immediately following the RING is a cysteine/histidine (CH)-rich motif termed the B-box domain (Fig. 2). This motif defines a new class of CH structures, and can be present in one or two copies (B1 and B2). The spacing of the CH residues in this domain is CXXHX(7)CX(7)CXXCX(5)HXXH (24,25). Both the midin and MID2 proteins display two copies of this domain (Fig. 3A and B). Figure 3A shows the alignment of a subset of different proteins in the B-box domain. As shown in this figure, other amino acids besides the C and H residues are conserved in the different proteins, suggesting a possible role in the proper functioning of this domain. In vitro site-directed mutagenesis experiments showed that mutations of specific residues within the B-box affect the ability of some of these proteins to multimerize to a varying degree (26). It should be noted that several tripartite domain-containing proteins, including midin and MID2, display a conserved aspartic acid (D) residue in the B2 domain, while in the same position a subset of proteins displays an extra potential cysteine metal ligand (C, underlined in Fig. 3A). The coiled-coil domain (amino acids 176-308) is believed to play a role in homomultimerization processes (26). Analysis of the MID2 protein using the COILS (27) program reveals the presence of coiled-coils from amino acid 210 to 318. An RFP-like domain is recognized at the C-terminus of the MID2 protein (amino acids 479-631). This domain can be found in association with the RING finger, B-box and coiled-coil domains, or by itself, as in the milk fat globule membrane protein butyrophilin (Fig. 3B). The RFP-like domain is particularly rich in tryptophans and tyrosines and its functional significance is still unknown. Figure 3B shows a schematic representation of the proteins displaying RING, B-boxes, coiled-coil and RFP-like domains present to date in the non-redundant protein database. As illustrated in the figure, the four domains can be found in different combinations, but their order from the N- to the C-terminus of the proteins is always conserved, with the RING domain always located at the N-terminus and the RFP-like domain always located at the C-terminal end.
A
 |
B
 |
Figure 3. (A) Alignment of a subset of proteins in the B-box domain. The conserved cysteine and histidine residues characterizing the domain are shown in bold. Underlined in bold is the aspartic acid residue substituted by an extra potential cysteine metal ligand in a subset of proteins including midin and MID2. (B) Domain structure of the B-box proteins present in the public databases to date. R (black rectangle), RING finger domain; B1 and B2 (stippled and cross-hatched ovals), B-box domains; C-C (striped rectangle), coiled-coil domain; RPF (light grey rectangle), RFP-like domain. Midin (GenBank accession no. Y13667); MID2 (GenBank accession no. Y18880); EFP, oestrogen-responsive finger protein (GenBank accession no. A49656); Marenostrin/Pyrin (GenBank accession no. Y14443/O15553); EFP-like (GenBank accession no. D87458); TIF1 (GenBank accession no. AF009353); Xnf7 (GenBank accession no. AAB35876); ATDC (GenBank accession no. L24203); RFP, Ret finger protein (GenBank accession no. J03407); Ro/SSA (U90547); PML (GenBank accession no. P29590); PWA33 (GenBank accession no. Q02084); butyrophilin (GenBank accession no. U90548).
MID2 maps to Xq22 and has a similar genomic structure to MID1
The RPCI-5 arrayed total PAC library was screened by PCR analysis using a sequence-tagged site (STS) generated within the MID2 coding region. A PAC clone (dJ974N21) was identified and used for fluorescence in situ hybridization (FISH) experiments. Hybridization signals appeared on the long arm of the X chromosome within the q22 region. No significant signals were found on any other chromosome (data not shown). To establish the exact mapping assignment, the human MID2 cDNA was hybridized to filters representing the X chromosome YAC collection. Four different YAC clones, cX-39A7 (alias yWXD8382 and ICRFy900c125), cX-43G9 (alias yWXD8261 and ICRFy900c1111), cX-43G11 (alias yWXD8383 and ICRFy900d04107) and cX-105F7 (alias yWXD7276 and CEPH 813B10), were found to be positive. These results were confirmed by hybridization of the full-length cDNA on digested plugs prepared from the above-mentioned YAC clones (data not shown). YAC 813B10 is known to contain the entire COL4A6 gene and part of COL4A5 which is the gene responsible for Alport syndrome, while the YAC clone ICRFy900c125 partially overlaps with 813B10 and is located centromeric to COL4A6 (28). This finding places the MID2 gene proximal and close to the COL4A6 locus in Xq22.3. Figure 4A shows a schematic representation of the mapping information obtained for MID1 and MID2.
Figure 4. Schematic representation of the mapping assignment of MID1 and MID2. (A) Comparison of the mapping assignment of MID1 and MID2 in human. MID1 is located in the Xp22 region, while MID2 is located on the Xq22 region very close to the COL4A6 locus. Also indicated are the other genes mapped in the close proximity of both loci. (B) Top: the haplotype figure from the Jackson BSB backcross showing part of the X chromosome with loci linked to Mid2. Loci are listed in order, with the most proximal at the top. The black boxes represent the C57BL6/JEi allele and the white boxes show the SPRET/Ei allele. The number of animals with each haplotype is given at the bottom of each column of boxes. The recombination (R) percentage between adjacent loci is given to the right of the figure, with the standard error (SE) for each R. Missing typings were inferred from surrounding data where assignment was unambiguous. Raw data from The Jackson Laboratory were obtained from the http://www.jax.org/resources/documents/cmdata . Bottom: a schematic representation of the gene order on the mouse X chromosome of all loci reported in (A). The gene order (not in scale) does not correlate with the actual physical distances. The map is depicted with the centromere toward the top.
Direct sequencing of the PAC dJ974N21 DNA with primers designed on the full-length cDNA sequence allowed us to establish the MID2 exon-intron boundaries. Nine different exons were identified and the sequences of all intron-exon boundaries were determined. Exon sizes and splice junction sequences are shown in Table 1. Junction sequences are all in agreement with the 5[prime] and 3[prime] splice site consensus motifs, with the exception of the exon 6 5[prime] splice site which shows the sequence GCAAGT instead of the consensus 5[prime] splice site sequence. Differences in the GT dinucleotide have been reported in 0.13% of splice site sequences (29). Comparison of the MID1 and MID2 genomic structures revealed that the two genes have a conserved genomic structure, with the splice site occurring at the same position in the two genes, with the exception of MID2 exon 6 which has 30 additional amino acids as compared with MID1 (Fig. 2). These amino acids are also absent in the murine homologue (see below) and fall in the region between the coiled-coil and the RFP-like domains.
Table 1. Splice junction sequences of the MID2 gene
| Exon no. | Splice junctions | Exon size (bp) | |
| 3[prime] Splice site | 5[prime] Splice site | ||
| 1 | ....................... | GAGAAACTCAAG/gtaagggatct | |
| 2 | tgcgatgacag/CAAACTCTGGA | CAGCAGGTTGAG/gtatgtaacag | 96 |
| 3 | tttcttggcag/GTGAATACTGC | AAAGAGACAAAG/gtaaagcgcag | 108 |
| 4 | attccttaaag/GTTATGAAACT | TATTGCTGAGAG/gtcagttcctt | 149 |
| 5 | ctctttttcag/GGTCGCTATGG | ATTATTTAACAG/gtgtgaaaata | 128 |
| 6 | tttccttacag/CCCCAAACCCA | GGGCGCCACGAG/gcaagtgtttg | 234 |
| 7 | ccatgttacag/GCCTGTATAAT | TAAAAACAAACA/gtacgttgtgg | 162 |
| 8 | ttatttttcag/GCCAACCCTTT | TTCCTCAACATG/gtgagtggatc | 208 |
| 9 | ctgtctcacag/GTATGCAATTG | ........................ |
Identification of the murine Mid2 cDNA
To identify the MID2 mouse homologue, the human full-length cDNA was used as a query against the dbEST. Although no mouse ESTs were detected, one rat EST (AA925587) was identified. Primers were designed on the rat EST sequence in a region corresponding to MID2 exon 1 outside the RING finger domain and used to amplify mouse genomic DNA. A PCR product of the expected size was obtained, sequenced and subsequently used as a probe against a mouse embryonic carcinoma cDNA library. Three different cDNA clones were identified and entirely sequenced. Characterization of each of the cDNA clones allowed us to establish a cDNA contig of 2524 bp (GenBank accession no. Y18881). The first in-frame stop codon (TAA) was identified at nucleotide 2435, predicting a protein product of 685 amino acids. Best-fit analysis revealed that the mouse Mid2 gene is highly homologous to its human counterpart (92.5% identity at the nucleotide level and 99.3% identity at the protein level) (data not shown).
Mid2 maps to the mouse syntenic region
A polymorphism between C57BL/6Jei and Mus spretus in the Mid2 3[prime]-untranslated region (3[prime]-UTR) was exploited to map Mid2 using the Jackson BSB backcross panel (30). Haplotype analysis of the segregating M.spretus allele placed Mid2 between DXMit4 and DXBir17/DXMit34 in the central portion of the mouse X chromosome (Fig. 4B, top panel). This region of the mouse X chromosome also contains other genes including Btx, Ags, Plp, Prps1 and Col4a5, which, like MID2, map to human Xq21.33-q22 (Fig. 4A and B, bottom panel). Thus, this interval appears to be part of a conserved linkage group between mouse and man. The Mid2 haplotype data have been submitted to the Jackson Laboratory World Wide Web address http://www.jax.org/resources/documents/cmdata . Figure 4B (bottom panel) shows a scheme with the gene order in mouse of the loci reported in Figure 4A. The position of Mid1 which spans the murine pseudoautosomal boundary (PAB) (31,32) is indicated, as is the position of Sts, which is the only gene isolated so far from the murine pseudoautosomal region (PAR) (33).
Expression studies
The human MID2 cDNA was hybridized against commercially available northern blots containing poly(A)+ RNA extracted from a variety of fetal and adult human tissues. A low abundant ~7 kb and a fainter ~3.5 kb transcript were observed in several tissues. In fetal organs, both MID2 transcripts were observed in kidney and lung at low levels, while in adult tissues the ~7 kb transcript was observed at low levels in adult prostate, ovary and small intestine (data not shown). Indirect information on the low abundance of the MID2 transcript can also be derived from the paucity of cDNA clones obtained during the cDNA library screening and by the fact that only three ESTs, one from fetal retina (AA016125), one from thyroid (AA385245) and one from pooled brain tumour tissues (AI241748), were identified through database searching (Fig. 1). The same search performed using the MID1 sequence as a query (GenBank accession no. Y13637) reveals the presence of 47 different human ESTs.
To begin investigating the spatio-temporal expression pattern of Mid2 during development, a partial murine cDNA was used to perform RNA in situ hybridization studies on mouse embryonic tissue sections at embryonic day 10.5 (E10.5) and 14.5 (E14.5). At E10.5, the level of expression of Mid2 is very low and mostly confined to the central nervous system (CNS) and the developing heart, while at later stages, when organogenesis is advanced, other organ systems show Mid2 expression above the basal level. On the contrary, Mid1 was ubiquitously transcribed at early stages (E9-E10.5) in all tissues examined except the developing heart (32).
In the cephalic region of the E10.5 embryo, expression is seen in the neuroepithelium, in the trigeminal ganglia, which at this stage are very large, and in the heart. Caudally, expression in the nervous system is evident in the differentiated dorsal root ganglia (Fig. 5A). Figure 5B shows a coronal section of an E14.5 mouse embryo. At this stage, Mid2 expression is ubiquitous but particularly high in the heart (in the walls of the right and the left ventricle chambers, as well as in the septum), in the kidneys and in the CNS.
Figure 5. Expression pattern of Mid2 during mouse development. In (A) and (B), the white colour represents the signal. (A) Sagittal section of the whole E10.5 mouse embryo. Expression is detected in the trigeminal ganglia (V), in the heart (h) and in the dorsal root ganglia (drg). (B) Coronal section of an E14.5 mouse embryo. The expression is high in the heart and in the kidney (k). (C-H) Micrographs showing sections after double exposure: red represents the in situ hybridization signal and blue shows the nuclei stained with Hoechst 33258 dye. (C and D) Sagittal sections of the developing heart of an E10.5 embryo. The arrows indicate the pericardium (C) and the atrioventricular canal (D). bc, bulbus cordis; I, first branchial arch; v, common ventricular chamber; ac, atrial cavity. (E-H) Coronal sections of an E14.5 embryo. (E) Mid2 is expressed in the mucosa of the stomach (st). (F) Section showing Mid2 expression in the thyroid (ty) and the thymus (th). (G) Coronal section of the developing ear showing expression in the pinna (p). (H) Strong Mid2 expression is observed in the kidneys (k) and in the dorsal root ganglia (drg).
Figure 5C and D shows the Mid2 signal in different E10.5 sagittal sections at the level of the developing heart. At this stage of development, the heart is the most prominent organ system of the embryo and the first to differentiate and function. Figure 5C shows expression of Mid2 in the bulbus cordis before its absorption into the right ventricle; this structure will later develop into the future conotruncus and right ventricle. Lack of expression was observed in the wall overlying the pericardial cavity. The same panel shows a Mid2 signal in the mandibular component of the first branchial arch. In a more medial section (Fig. 5D), the signal is observed in the trabeculated wall of the common ventricular chamber of the heart. The micrograph shows that the endocardial cushion tissue associated with the wall of the atrioventricular canal is devoid of Mid2 expression. The endocardial cushion is involved in the formation of the atrioventricular channels and its formation is due to cells coming from the adjacent endocardium under the influence of factors produced by the myocardium.
At E14.5, a high level of expression is also seen in the mucosal lining of the stomach (Fig. 5E), and a strong signal is detected in the two lobes of the thymus (Fig. 5F). At this developmental stage, the thymus is not yet differentiated into medullary and cortical regions, confirmed by the homogeneity of its histological morphology. Consistent with this, Mid2 is expressed in the whole structure. In the same panel, expression is observed in the two lobes of the thyroid gland which consist of large numbers of buds. At the same developmental stage, a peculiar signal is evident in the pinnae of the ears which at this stage have expanded and are turned forward covering about half of the external acoustic meatus (Fig. 5G). A signal is also detected in the nasal and oral cavity epithelia, and in the eye (data not shown).
At E14.5, strong expression of Mid2 is observed in the developing kidney and in particular in the primitive glomeruli that are first seen and dispersed throughout the kidney at this stage (Fig. 5H). Although the subdivision into cortical and medullary regions of the kidney at this stage is not well defined, the most peripheral part of the cortex, which consists mainly of poorly differentiated metanephric cap tissue, appears to be devoid of Mid2 expression (Fig. 5H). At E16.5, Mid2 expression is still detectable in the heart, but is not distinguishable from the ubiquitous basal signal (data not shown).
Overall, the level of expression of Mid2 is lower than that of Mid1 in the same developmental stages (32). Qualitatively, there is partial overlap in the distribution of the two transcripts: both are expressed in kidney, in the CNS, in the nasal and oral cavity epithelia and in the eye. Additionally, Mid2 expression is also detected in the developing heart, which was one of the few organs clearly devoid of Mid1 signal, and a significant level of expression is found at the level of the thymus and thyroid.
MID2 localizes to microtubules
A characteristic feature of some of the B-box family members is a peculiar subcellular distribution. PML, Tif1 and RFP were found to be confined within subnuclear structures which, for PML, were named PML nuclear bodies (14-16), and it is known that for PML a correct localization is crucial for its function. On the other hand, midin has been found within the cytoplasm in association with the microtubule apparatus (18,19).
To assess MID2 subcellular localization, we fused its coding region with a Myc-green fluorescent protein (GFP) tag and transiently transfected this expression construct in Cos7 cells. Direct observation of the cells to detect the GFP signal shows a cytoplasmic filamentous distribution (green signal of Fig. 6A). The same pattern of distribution was observed by performing indirect immunofluorescence using an anti-Myc antibody (data not shown). As shown in Figure 6B, the perfect coincidence of the signal obtained after double staining with an anti-[beta] tubulin antibody demonstrated that the MID2 signal corresponds to microtubular structures.
Figure 6. Subcellular localization of MID2. (A) Direct fluorescence of Cos7 cells transfected with MID2-GFP (green signal). (B) The same cell after staining with anti-[beta]-tubulin antibody followed by a rhodamine anti-mouse antibody (red signal).
DISCUSSION
MID1 was isolated from the Xp22 region by our group and found to be the gene responsible for the Opitz G/BBB syndrome (9). By using a bioinformatic approach, we have isolated MID2, a human paralogue of MID1. MID1 and MID2 share many similarities and display a few differences, which are summarized in Table 2. MID2 was mapped on the long arm of the human X chromosome in the q22 band close to and telomeric of the COL4A6 locus. The Xq22 band represents one of the most conserved regions between the human and murine X chromosomes. Several human transcripts, including BTK, GLA, PLP, PRPS1 and COL4A5, which map to the human Xq22 region, have been shown to map to the mouse X chromosome within a single conserved linkage group (34). As expected, Mid2 was mapped in the mouse in the same syntenic region close to the Plp locus (Fig. 4B). On the contrary, in mouse, Mid1 has been shown to span the PAB, which is the point where the PAR diverges into the X- and Y-specific sequences (31,32).
Table 2. MID1/MID2 differences and similarities
| MID1 | MID2 | |
| Human map location | Xp22 | Xq22 |
| Mouse map location | spanning the PAB | F1-F2 |
| Domains | RING | RING |
| two B-boxes | two B-boxes | |
| coiled-coil | coiled-coil | |
| rfp-like | rfp-like | |
| Subcellular localization | cytoplasm in association with microtubules | cytoplam in association with microtubules |
| Expression in fetal tissue | high level, ubiquitousno heart | low level, ubiquitoushigher in heart, thymus and thyroid |
| Associated disorder | X-linked OS | ? |
Interestingly, MID1 is not the only example of an Xp22 gene with a closely related homologue in the Xq22 region. PRPS1, located 2 Mb centromeric to MID1, has a closely related homologue, PRPS2, in the Xq22 region (35). Furthermore, we have recently identified a serine-threonine phosphatase gene (PPEF-1) from the Xp22 region, and a serine-threonine phosphatase gene (PPP6C) with homology to PPEF-1 has been identified and mapped to the Xq22.3 region (36). These data indicate the presence of a region of homology between Xp22 and Xq22 and suggest the possibility that an intrachromosomal duplication involving the short and the long arms of the X chromosome may have occurred early in evolution. Other examples of intrachromosomal and interchromosomal duplication events are known (37), and recently a 26.5 kb gene-rich duplication region between Xq28 and 16p11.1 was identified (38). It is conceivable that the ongoing large-scale human genome sequencing project will help in clarifying whether an intrachromosomal duplication event also occurred in the Xp22-Xq22 regions.
OS is a genetically heterogeneous disorder and, to date, only nine mutations have been identified on a total of 22 familial cases and 18 sporadic individuals analysed (9,39). Considering the genetic heterogeneity of OS and the low percentage of mutations identified so far, we decided to evaluate the possible role of MID2 in OS. A preliminary mutation study was performed by single-strand conformation polymorphism (SSCP) analysis in 15 OS patients with no mutation in MID1 (B. Franco, unpublished results), and no causative mutations were detected; however, further studies will be needed to exclude a role for MID2 in the pathogenesis of OS. A number of human genetic disorders have been mapped to the Xq22 region, including Arts syndrome (MIM 301835) (40), X-linked megalocornea (MIM 309300) (41), a locus for congenital deafness (DFN2, MIM 304500) (42), a syndrome characterized by epilepsy and mental retardation with an X-linked dominant inheritance pattern and sparing of males (EFMR, MIM 300088) (43), and a locus for non-specific mental retardation (MRX23, MIM 300046) (44). MID2 could be considered a positional candidate gene for all these disorders. A locus for FG syndrome (MIM 305450), which can also be considered a defect of the midline structures, has been mapped to the long arm of the X chromosome. FG shares many clinical features with OS syndrome, such as imperforate anus and congenital heart defects, but recent linkage data demonstrated genetic heterogeneity for the disorder and placed one of the loci in the Xq12-q21.31 region (45), thus excluding the region where MID2 has been mapped. However, in view of the genetic heterogeneity, mutation analysis would be needed to exclude an involvement of MID2 in the pathogenesis of FG syndrome.
Midin and MID2 are highly homologous (84% similarity and 77% identity at the protein level), have a conserved genomic structure and display a RING finger, two B-boxes, a coiled-coil and a conserved RFP-like domain. These motifs characterize the B-box family of genes which have been shown to be involved in different cellular functions such as cellular growth and differentiation. The subcellular localization is a critical key to the understanding of a protein function, and midin expression was found confined to the cytoplasm in association with microtubular structures (18,19). To assess the subcellular localization, transfection of MID2-GFP fusion constructs in Cos7 cells was performed, and this experiment showed that MID2, similar to midin, is associated with microtubular structures. These data suggest that the analogy between MID1 and MID2 extends beyond the simple sequence similarity and that the two genes may share a similar function. Several lines of evidences indicate that the RING finger, the B-boxes and the coiled-coil are protein-protein interaction domains involved in the formation of multiprotein complexes (46-48). Some of the B-box proteins have also been shown to form homodimers through interactions of their domains (49,50). Co-immunoprecipitation studies indicate that these domains do not mediate midin-MID2 interaction (19), although further experiments will be required to assess whether MID2, like midin and other B-box proteins, is able to homodimerize in vivo.
The expression pattern of Mid2 was evaluated by RNA in situ hybridization studies on mouse embryonic tissue sections at E10.5 and E14.5. Overall, Mid2 expression is lower than that of Mid1 at the same developmental stages. The most striking differences in the expression pattern of Mid1 and Mid2 are in the developing heart. Mid1 expression was studied by RNA in situ hybridization from E10.5 to E16.5 and was never detected in the developing heart (9,32; E. Rugarli, unpublished data). On the contrary, a striking expression of Mid2 is present in the heart starting from E10.5 through E16.5. These data suggest that Mid2 may be involved in the diverse mechanisms that underlie a correct morphogenesis and development of the cardiovascular apparatus in the early stages. To test this hypothesis as well as to gain deeper insight into the function of this gene, the generation of mice with targeted mutation in Mid2 will be necessary. These animals will also be valuable in studying the pathways in which both Mid1 and Mid2 may be involved and in verifying the possible interaction between MID2 and other members of the B-box proteins.
MATERIALS AND METHODS
cDNA identification and genomic structure
In order to identify the MID2 full-length transcript, two human cDNA libraries were screened: a teratocarcinoma/neuron cDNA library (mature hNT neuron, Stratagene 937233) and a fetal brain cDNA library (Clontech HL3003a). For the isolation of Mid2 full-length cDNA, a mouse embryonic carcinoma cDNA library was used (Stratagene 937317). Plating, hybridization and washing conditions were as described previously (51). To establish the genomic structure, pools for the total PAC genomic library RPCI-5 (52), available at the YAC and PAC Screening Centre at the San Raffaele Biomedical Science Park (Milan, Italy) were screened. The primers used for the screening were TO4466 (5[prime]-GCCTGTATAATTCAGTAGAC-3[prime]) and TO4467 (5[prime]-TGTTTGTTTTTAGTCGGGTAG-3[prime]) (Fig. 1); the PCR was performed at an annealing temperature of 56°C. Direct PAC DNA sequencing was performed using 2 mg of purified DNA (Qiagen, Chatsworth, CA) and 50 pmol of cDNA-derived primers with the ABI Prism Big Dye Terminator Cycle Sequencing Kit (Perkin Elmer, Foster City, CA). Details on the X chromosome YAC collection can be found at http://www.mpimg-berlin-dahlem.mpg.de/~xteam/yacollexpl.html
cDNA sequence analysis
cDNA sequence analysis and nucleotide and protein database searches were performed as described previously (53). Data on similarity/identity were obtained using the Best-fit program of the GCG software package, v.8.1. The multiple alignment analyses were generated using the PileUp program of the Wisconsin GCG software package, v.8.1. The coil scan analysis was carried out using the COILS program of the Wisconsin GCG software package, v.8.1.
Expression studies
Commercial northern blots (Clontech, Palo Alto, CA) containing human RNA from fetal and adult tissues were hybridized and washed using the conditions recommended by the manufacturer. The probe selected for RNA in situ hybridization experiments showed 66% identity with the murine homologue of MID1. Sense and antisense 35S-labelled riboprobes were transcribed using a murine EcoRI-HindIII cDNA fragment (nucleotides 1601-2424) cloned in a pBS vector and digested with the appropriate restriction enzymes. Mouse embryo tissue sections were prepared and RNA in situ hybridization performed as described (54). In all the experiments performed, the sense probes did not show any specific hybridization. Autoradiographs were exposed for 3 days. Slides were dipped in Kodak NTB2 emulsion and exposed for 3 weeks.
Genetic mapping in the mouse
SSCP was performed on the Mid2 PCR product from nucleotides 3162-3560. Primers TO8343 (5[prime]-ATCTGGGGACAAGATTGCAC-3[prime]) and TO8344 (5[prime]-AGGGGATGCAAATGAATGAA-3[prime]) were used (58°C annealing) to amplify 25 ng of genomic DNA templates derived from the Jackson Laboratory, (Bar Harbor, ME) interspecific backcross panel (C57BL/6JEi×SPRET/Ei)F1×C57BL/6JEi (called Jackson BSB) (30).
Cell culture, transfection and immunofluorescence
Cos7 cells were grown in Dulbecco's modified Eagle's medium with 10% fetal calf serum. Transfections were performed by the calcium phosphate method (55).
Indirect immunofluorescence was performed on paraformaldehyde-fixed transfected Cos7 cells. Cells were then permeabilized with 0.2% Triton X-100, blocked with porcine serum and incubated with anti-[beta]-tubulin monoclonal antibody (clone KMX-1; Boehringer Mannheim, Mannheim, Germany). Staining was obtained after incubation with tetramethylrhodamine isothiocyanate (TRITC)-conjugated anti-mouse antibody.
ACKNOWLEDGEMENTS
We thank Sandro Banfi for helpful discussions, M. Smith for preparation of the manuscript, and the PAC Screening Centre at the San Raffaele Biomedical Science Park. We also wish to thank all the clinicians who provided us with the patient material. This work was supported by the Italian Telethon Foundation, and by the EC under grant nos BMH4-CT96-1134 and BMH4-CT96-0889.
REFERENCES
*To whom correspondence should be addressed. Tel: +39 02 21560202; Fax: +39 02 21560220; Email: franco{at}tigem.it
This page is run by Oxford University Press, Great Clarendon Street, Oxford OX2 6DP, as part of the OUP Journals
Comments and feedback: jnl.info{at}oup.co.uk
Last modification:
Copyright© Oxford University Press, 1999.
CiteULike 
Connotea 
Del.icio.us What's this?
This article has been cited by other articles:
|
|
 |
|
  B. B. Cheung, J. Bell, A. Raif, A. Bohlken, J. Yan, B. Roediger, A. Poljak, S. Smith, M. Lee, W. D. Thomas, et al. The Estrogen-responsive B Box Protein Is a Novel Regulator of the Retinoid Signal J. Biol. Chem., June 30, 2006; 281(26): 18246 - 18256. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 M. W. Yap, S. Nisole, C. Lynch, and J. P. Stoye Trim5{alpha} protein restricts both HIV-1 and murine leukemia virus PNAS, July 20, 2004; 101(29): 10786 - 10791. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 T. V. O. Hansen, N. A. Hammer, J. Nielsen, M. Madsen, C. Dalbaeck, U. M. Wewer, J. Christiansen, and F. C. Nielsen Dwarfism and Impaired Gut Development in Insulin-Like Growth Factor II mRNA-Binding Protein 1-Deficient Mice Mol. Cell. Biol., May 15, 2004; 24(10): 4448 - 4464. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
M. A. Benson, C. L. Tinsley, and D. J. Blake Myospryn Is a Novel Binding Partner for Dysbindin in Muscle J. Biol. Chem., March 12, 2004; 279(11): 10450 - 10458. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 P. A. Stein, C. P. Toret, A. N. Salic, M. M. Rolls, and T. A. Rapoport A novel centrosome-associated protein with affinity for microtubules J. Cell Sci., January 9, 2002; 115(17): 3389 - 3402. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 T. C. Cox, L. R. Allen, L. L. Cox, B. Hopwood, B. Goodwin, E. Haan, and G. K. Suthers New mutations in MID1 provide support for loss of function as the cause of X-linked Opitz syndrome Hum. Mol. Genet., October 1, 2000; 9(17): 2553 - 2562. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 J. A. Spencer, S. Eliazer, R. L. Ilaria , Jr., J. A. Richardson, and E. N. Olson Regulation of Microtubule Dynamics and Myogenic Differentiation by MURF, a Striated Muscle RING-Finger Protein J. Cell Biol., August 21, 2000; 150(4): 771 - 784. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
A. Gassama-Diagne, F. Hullin-Matsuda, R. Y. Li, M. Nauze, A. Ragab, V. Pons, C. Delagebeaudeuf, M.-F. Simon, J. Fauvel, and H. Chap Enterophilins, a New Family of Leucine Zipper Proteins Bearing a B30.2 Domain and Associated with Enterocyte Differentiation J. Biol. Chem., May 18, 2001; 276(21): 18352 - 18360. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||