| Sign In to gain access to subscriptions and/or personal tools. |
HUGO (FNDC3A): a New Gene Overexpressed in Human Odontoblasts
1 Université de Lyon, Villeurbanne, F-69000, France; Correspondence: * corresponding author, IGFL/INSERM ERI 16, Faculté d’Odontologie, Rue Guillaume Paradin, 69372 LYON Cedex 08, France, bleicher{at}univ-lyon1.fr
Previously, we established a subtractive cDNA library enriched in odontoblast-specific genes and hypothesized that new, previously unidentified, markers would be present, associated with the odontoblast phenotype. In this paper, we report the first characterization of a new gene we have named HUGO, and its associated deduced protein sequence. This gene expression is under the control of two alternative promoters, resulting in the synthesis of two proteins, one of which, HUGO2, is included in the other, HUGO1. HUGO proteins are mainly composed of a proline-rich region at the N-terminus, 8 type III-fibronectin modules, and a transmembranous helix at the C-terminus. In odontoblasts, the proteins are located in Golgi vesicles. However, they display a broader expression pattern, since they are also expressed by nerve fibers in the dental pulp and other tissues (e.g., trachea, brain, kidney), as demonstrated by immunohistochemistry and qPCR, respectively. Their location in odontoblasts suggests a role in collagen and glycosaminoglycan synthesis.
Key Words: tooth • human odontoblast • new gene • type III-Fibronectin module • HUGO
Odontoblasts are post-mitotic, neural-crest-derived cells, which differentiate according to tooth-specific temporospatial patterns. They are organized as a layer of palisaded cells along the interface between the dental pulp (soft connective tissue) and dentin. Their terminal differentiation is characterized by several steps, implying withdrawal from the cell cycle, elongation, and cytological polarization (Ruch et al., 1995). These cells play a central role during the formation of dentin, in that they synthesize and secrete the organic matrix macromolecules (pre-dentin) and actively participate in the transportation and accumulation of calcium at the mineralization front (Linde and Lundgren, 1995; Ruch et al., 1995). They are also known to participate in dentin regeneration following caries (Goldberg and Smith, 2004). However, if dentin production is the main function of odontoblasts, it is not the only one. These cells are also thought to play a role in the protection of the dental pulp tissue. Indeed, odontoblasts are the first line of defense against cariogenic bacteria by initiating the innate immune response (Durand et al., 2006), and are suspected to play a role in tooth pain transmission by operating as sensors cells (Allard et al., 2006). However, while the phenotype of dental pulp mesenchymal cells, during tooth initiation and morphogenesis stages, has been extensively documented (Jernvall and Thesleff, 2000), little is known about odontoblast cytodifferentiation and the resultant phenotype. This lack of information is due to the difficulty in micro-dissection and isolation of a pure odontoblast population, but the development of cultures of pulp cells differentiating into odontoblasts has reduced this barrier (Kuo et al., 1992; Stanislawski et al., 1997; Mesgouez et al., 2006). Previously, we established a human pulp explant culture system supportive of odontoblast differentiation at both morphological and functional levels (Couble et al., 2000). A cDNA library was then established by the application of suppression-subtractive hybridization to characterize odontoblast-specific genes (Buchaille et al., 2000a). Several differentially expressed cDNA clones were isolated, including extracellular matrix proteins, mitochondrial proteins, transcription factors, and unknown genes. A reverse Northern dot blot procedure applied to these unknown genes revealed the overexpression of a clone (clone D6 in Buchaille et al., 2000a) in odontoblast-like cells compared with pulpal precursor cells. In this study, we report the identification and initial characterization of this gene, which we have named HUGO (human gene expressed in odontoblasts). The aim of this study was to identify a new marker associated with the odontoblast phenotype.
Cell Culture Dental pulps were collected with informed consent from individuals in accordance with French legal requirements (article 672-1, Public Health Code). Human precursor pulp cells and odontoblast-like cells were obtained as previously described (Couble et al., 2000). COS-1 cells were cultured in DMEM medium (Invitrogen, Cergy-Pontoise, France) supplemented with 10% fetal calf serum (Eurobio, Les Ulis, France), 100 UI/mL penicillin-50 mg/mL streptomycin, and fungizone (Invitrogen). HUGO2 cDNA sequence was cloned in frame with the V5 tag in pcDNA 3.1, and transfected into COS-1 cells with the use of LipofectAmine 2000 (Invitrogen), according to the manufacturer’s instructions. Geneticin 1.2 3g/mL was added for the selection of these transfected cells.
RNA Extraction and cDNA Synthesis
5'-Rapid Amplification of cDNA Ends (5'-RACE)
Northern Blot Analysis
Real-time PCR
In situ Hybridization
Antiserum Preparation
Western Blotting
Flow Cytometry
Immunochemistry
Electron Microscopy Peroxidase Immunolabeling
Characterization of HUGO at the Gene and Protein Levels We applied a 5'-RACE PCR strategy to isolate the complete HUGO cDNA from odontoblast-like cells, which resulted in the isolation of a cDNA fragment of about 4600 bp. Sequencing of this band revealed the presence of two cDNAs showing a difference in their 5' UTR. Their sequences were deposited in the GenBank database under references AJ749706 (HUGO1) and AJ749707 (HUGO2). A Northern blot analysis on mRNA extracted from odontoblast-like cells confirmed the expression of the gene in these cells as two transcripts of 8500 and 6500 nucleotides, respectively (Fig. 1A  ). According to genome resources (www.ensembl.org), these genes are located on chromosome 13q14.2 and have a genomic organization consisting of 26 exons and 25 introns for HUGO1 and 24 exons and 23 introns for HUGO2 (Fig. 1B). Moreover, cDNA sequence alignment on chromosome 13 indicates that HUGO1 and HUGO2 have the last 23 exons in common, but differ at their 5' end, using different sets of exons (Fig. 1B). The open reading frame of HUGO1 cDNA is 3594 nucleotides long, coding for a 1198-amino-acid protein with a predicted pI of 6.68 and a molecular mass of 131.8 kDa, whereas the open reading frame of HUGO2 is 3402 nucleotides long, encoding a 1134-amino-acid protein with a pI of 7.07 and a molecular mass of 125 kDa. Alignment between the deduced protein sequences (Corpet, 1988) showed that the HUGO2 protein sequence is contained in the HUGO1 sequence (Fig. 1C). Protein sequence analysis demonstrated that the two HUGO isoforms are mainly composed of 8 type III-fibronectin modules. A proline-rich region is present at the N-terminus and a transmembranous helix at the C-terminus of the proteins (Fig. 1C). Evolutionary analysis of the protein showed that HUGO is highly conserved, since human and rat sequences shared 84% homology of identity and human and Tetraodon sequences shared 56% homology of identity (data not shown). The evolutionary tree also demonstrates the existence of a related protein family, named FNDC3B (cf. APPENDIX).
Characterization of HUGO Gene Expression Pattern Real-time PCR, realized with primers in the common sequence of both transcripts, showed that HUGO is 10-fold overexpressed in odontoblast-like cells compared with the pulpal precursor cells (Fig. 2A ), and is also highly expressed in the trachea and, to a lesser extent, in the brain, kidney, lung, and liver (Fig. 2B). The same pattern of expression was obtained with specific primers for HUGO1 or HUGO2 transcripts (Figs. 2A, 2B). However, COS-1 cells express only HUGO2 transcripts (Fig. 2A). In situ hybridization on human dental pulp sections confirmed the expression of HUGO in the odontoblast layer (Fig. 2D), as well as in nerve fibers in the pulp (Fig. 2E).
Characterization of HUGO Protein Expression Pattern A purified polyclonal antibody, raised against two human peptides of HUGO (Fig. 1C  ), was characterized by Western blot with proteins extracted from odontoblast-like cells, precursor pulp cells, and from COS-1 cells overexpressing V5-6xHis-tagged HUGO2 protein (Fig. 3A). The V5-6xHis-tag added 47 amino acids (5114 Da) to the HUGO2 C-terminal end. No band was detected by the V5-antibody on protein extracts from odontoblast-like cells, precursor pulp cells, and non-transfected COS-1 cells. One band of about 130 kDa appeared with this antibody on HUGO2-transfected COS-1 cells, corresponding to the tagged protein. The anti-HUGO antibody reacted with one band of 120 kDa in non-transfected COS-1 cells, corresponding to HUGO2. Two bands of about 130 kDa and 120 kDa (V5-6xHis-tagged HUGO2 and HUGO2, respectively) were detected in HUGO2-transfected COS-1 cells, whereas two bands of the same size were seen in odontoblast-like cells and precursor pulp cells (but corresponding to HUGO1 and HUGO2, respectively). Controls (see Fig. 3, legend) did not reveal any non-specific staining. Analysis of these data, together, indicates that the polyclonal antibody is specific to HUGO. Flow cytometry analysis indicates the cytoplasmic localization of the protein, since labeling of both precursor pulp cells and odontoblast-like cells was seen only when a permeabilization step was applied to the cells (Fig. 3B). Immunohistochemistry on human dental pulp sections showed a specific staining of the odontoblast layer (Fig. 3C) and nerve fibers (Fig. 3E). Double immunostaining for HUGO and β2 tubulin on nerve fibers demonstrated that both proteins did not co-localize (Fig. 4K). In odontoblast-like cells, a punctiform staining was seen in the Golgi zone around the nucleus (Figs. 3F, 3G). A double staining with an anti-Golgi zone antibody showed a co-localization between this marker and HUGO, both in vivo and in vitro (Fig. 4). Electron microscopy on these cells confirmed that the protein was localized in Golgi saccule-presenting aggregates (Fig. 3H).
In this report, we describe the identification and initial characterization of a novel gene whose expression is highly up-regulated in odontoblast-like cells. We named it HUGO (HUman Gene expressed in Odontoblasts), but it is now referred to as FNDC3A in databases. Northern blot analysis and sequencing of the 5'RACE-PCR fragment demonstrated that this gene is expressed as two isoforms under the control of alternative promoters. The alternative usage of these promoters leads to the synthesis of two proteins, one of which (HUGO2) is included in the other (HUGO1). Many eukaryotic genes contain multiple promoters, each subjected to different regulatory factors controlling temporal or spatial expression. Alternative promoter usage often has physiological implications, as is the case for the acetylcholinesterase gene (Meshorer et al., 2004). Five alternative first exons have been described for this gene, one of which encodes an N-terminal extension of the protein, resulting in a transmembrane protein variant with distinct properties. The N-terminus specific sequence of HUGO1 does not seem to contain any characterized signature domain, and neither protein presents a signal peptide. However, only the HUGO2 promoter is active in COS-1 cells, as demonstrated by qPCR and Western blot, showing a cell-type-specific regulated expression. Both isoforms are mainly composed of 8 type III-fibronectin domains, which are widely spread among contemporary animal proteins, having thus far been identified in more than 50 different proteins (Pankov and Yamada, 2002). These include many extracellular proteins (for example, fibronectin itself), some intracellular proteins (titin, twitchin), and the extracellular domains of many kinds of membrane-receptor proteins. The main function of this module seems to be to mediate protein-protein or protein-glycosaminoglycan (GAG) interactions. Immunolocalization of HUGO in odontoblasts indicates that the protein is confined to Golgi vesicles, as demonstrated by confocal co-immunolocalization with a Golgi zone antibody and, more precisely, in Golgi vesicles corresponding to those showing parallel bands, as shown by electron microscopy. The labeling seems to be associated with the membranes of these vesicles, which can be explained by the presence of a transmembranous helix at the C-terminus of HUGO. These saccules have been described in native odontoblasts (Weinstock and Leblond, 1974) and are actively involved in the synthesis of pro-collagen molecules from the pro a-chains and their post-translational modifications, i.e., glycosylation and oxidative deamination of lysine residues. Moreover, these same saccules have been demonstrated to be involved in the secretion of GAGs (Takagi et al., 1981). Analysis of these data led us to suggest that HUGO should be involved in the synthesis pathway of typeI collagen or GAG, which are the main molecules of the dentin matrix. The overexpression of HUGO in odontoblast-like cells, compared with the precursor pulp cells, supports this hypothesis. Indeed, odontoblast-like cells synthesize a larger amount of extracellular matrix than do precursor pulp cells in culture (Couble et al., 2000). However, this gene is not specific for odontoblasts, since it is also expressed in nerves in the dental pulp and in different organs such as the trachea, and, to a lesser extent, in the kidney and brain. In dental nerve fibers, HUGO is most probably expressed in Schwann cells, since it does not co-localize with β2 tubulin, a marker of nerve fibers. Other preliminary immunochemical studies have demonstrated that only particular cell types are labeled in these tissues (data not shown). Further identification of these cells will allow us to identify the precise role of the HUGO protein. But whatever it may be, it seems to be essential for the cells, since its sequence is well-conserved throughout the phylogenic tree. The next steps will be to characterize the HUGO partner (GAG or protein) and make a conditional deletion of the gene in transgenic mice to elucidate the function of this protein.
This work was supported by IFRO, GIS "Maladies Rares", and the Rhône-Alpes region. We thank the CeCIL (Lyon) and the CTµ (Lyon) for providing technical facilities.
A supplemental appendix to this article is published electronically only at http://jdr.iadrjournals.org/cgi/content/full/87/2/131/DC1. Received for publication July 6, 2007. Revision received November 21, 2007. Accepted for publication November 21, 2007.
Journal of Dental Research, Vol. 87, No. 2,
131-136 (2008)
|
|
|||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
 |
|
|
 |
|
Sign In to gain access to subscriptions and/or personal tools.
TOP
ABSTRACT
INTRODUCTION
