| Sign In to gain access to subscriptions and/or personal tools. |
Mouse Amelogenin Exons 8 and 9: Sequence Analysis and Protein Distribution
Department of Pediatric Dentistry, Dental School, University of Texas Health Science Center San Antonio, MSC 7888, 7703 Floyd Curl Drive, San Antonio, TX 78229-3900, USA; and Correspondence: * corresponding author, macdougall{at}uthscsa.edu
Amelogenin is the major protein of the developing enamel. Two additional exons, termed 8 and 9, have been characterized in the rat. Our aim was: to identify the mouse amelogenin exons 8/9 sequences; to investigate the potential presence of the alternative spliced isoforms of amelogenin exons 8/9; and to immunolocalize proteins containing sequences encoded by exons 8/9 during odontogenesis. RT-PCR analysis with exon 9 anti-sense primer generated 2 major amplicons with the use of a mouse tooth cDNA library and dental cell lines. DNA sequence analysis showed 93% identify with the rat exons 8/9 sequence. Alternative splicing of exon 3 was also found, but only in cDNAs lacking exons 8 and 9. Immunohistochemistry localized exons 8/9-encoded proteins in ameloblasts, young odontoblasts, and stratum intermedium cells. Analysis of our data supports the hypothesis that: (1) AMELX contains 2 additional exons; (2) ameloblasts and odontoblasts synthesize amelogenin 8/9; and (3) amelogenin splice variants may have unique functions during tooth formation.
Key Words: amelogenin • ameloblasts • odontoblasts • alternative splicing • exons 8 and 9
Amelogenin is the most predominant and well-characterized protein of the developing enamel matrix (Sasaki and Shimokawa, 1995). Amelogenins are expressed as a heterogeneous group of peptides derived through mRNA alternative splicing as well as post-translational protein cleavage (Simmer, 1994; Brookes et al., 1995; Zeichner-David et al., 1995). Amelogenins are encoded mainly by the amelogenin gene located on the X chromosome (AMELX), which contains 7 exons in both humans and the mouse. To date, 11 different mRNA isoforms of mouse AMEL, derived by extensive alternative splicing of exons 2 through 7, have been isolated and characterized (Simmer, 1994; Li et al., 1995). In addition, Northern-blot analysis has showed several amelogenin RNA transcripts larger than the predicted full-length mRNA size, suggesting the existence of additional coding regions (Wurtz et al., 1996; Papagerakis et al., 1999). Studies in rats have identified 2 additional coding exons, termed 8 and 9, which are expressed at the protein level during amelogenesis (Li et al., 1998; Baba et al., 2002). However, no amelogenin cDNA containing exons 8 and 9 sequences has been isolated to date. Amelogenins are critical for normal enamel formation, as has been demonstrated by human genetics and experimental laboratory studies (Hart et al., 2000; Gibson et al., 2001; Greene et al., 2002; Paine et al., 2002; Li et al., 2003). In addition, recent in vivo and in vitro studies have suggested that amelogenin proteins might have signaling potential for instructing epithelial and mesenchymal cells during tooth development (Zeichner-David, 2001; Veis, 2003). In particular, it has been suggested that the inclusion or exclusion of amelogenin exon 4 can specifically direct changes in the phenotype of the rat muscle fibroblast toward acquiring osteogenic or chondrogenic potential (Veis et al., 2000). Therefore, full characterization of amelogenin-spliced isoforms would provide the foundation for understanding the biological roles of the various amelogenin peptides. The goal of this study was to: (1) determine potential mouse amelogenin exon 8/9 sequences; (2) characterize the alternative spliced isoforms of amelogenin exons 8–9; and (3) immunolocalize amelogenin exons 8/9 encoded proteins during tooth development. These findings will help to delineate the mechanisms of enamel formation and mineralization as directed by the major group of extracellular matrix proteins.
Materials A mouse tooth cDNA library (MacDougall et al., 1997) and enamel organ epithelium (MEOE-3M) and odontoblast (MO6-G3) cell lines (MacDougall et al., 1995) were used to characterize mouse AMEL exon 8/9 sequences and alternative spliced profiles. The NIH-3T3 fibroblast cell line was used as a negative control. Hemi-mandibles of newborn and four-day-old Swiss mice were micro-dissected, washed in PBS for 10 min, fixed overnight in 10% buffered formalin, and demineralized in formic acid for 3–5 days. Tissues were paraffin-embedded, and 5-micron sections were prepared. All animal experiments were approved by the IACUC and were carried out under the NIH guidelines for proper animal handling.
RT-PCR
Immunohistochemistry
Gene Expression PCR was performed with AMEL exon 2 or exon 5 forward primers and exon 5 or exon 9 reverse primers, based on the published mouse and rat sequences. Multiple transcripts were generated for mouse amelogenin compared with the negative control (no added cDNA), with the use of cDNA target from the mouse tooth library and dental cell lines. Two amplification products were produced as previously described when the exon 2 forward and exon 5 reverse primers were used (Papagerakis et al., 2003). In addition, 2 major amplicons of approximately 350-bp and 750-bp fragments were obtained when exon 9 reverse primer was used with the exon 2 forward primer (Fig. 1  ). These fragments were subcloned, and the sequence was determined.
DNA Sequence Analysis Sequence analysis of AMEL exon 2–5 products showed that mouse amelogenin exon 3 was alternatively spliced, as has previously been shown for other species. However, exon 3 was always present when exons 8 and 9 were present. Sequencing of the amelogenin products containing exons 8/9 (GenBank accession # AY842498) showed 93% identity with the published rat amelogenin exons 8 and 9 sequence (Fig. 1  ). The large fragment (750 bp) contained amelogenin exons 2, 3, 5, 6, 8, and 9, while the 350-bp product contained exons 2, 3, 5, partial 6 (6D), 8, and 9. Both fragments lacked exons 4 and 7 and differed only in the partial omission of exon 6. Exons 8 and 9 consisted of a total of 119 bp (Fig. 1) and were always found together. Due to the presence of an internal stop codon within exon 9, the putative translation of mouse exons 8 and 9 would result in a peptide segment of 24 amino acids. This putative protein sequence (GenBank accession # AAW28903) shares 87% identity with the previously deduced rat sequence (Li et al., 1998), varying by only 3 amino acids (Fig. 1). Moreover, using the BLAST program against the mouse genome, we have characterized the AMELX intron sequence and length among exons 6, 8, and 9 (Table).
Immunohistochemistry Immunodetection showed positive staining in dental cells when the MEOE-3M ameloblast (Figs. 2A , 2B) and the MO6-G3 odontoblast (Fig. 2C) mouse cell lines were used. NIH-3T3 cells were negative (not shown). In developing mouse tissues, amelogenin 8/9 expression ranged from little or no expression (within the undifferentiated inner enamel epithelium) to high expression [associated with pre-ameloblasts (Fig. 2E), differentiated ameloblasts (Fig. 2F), and post-secretory ameloblasts (Fig. 2G)], to a complete loss of expression [associated with maturation stage cells (Figs. 2I–2M)]. In addition, staining was observed in the stratum intermedium cells associated with early stages of amelogenesis (Fig. 2E). Young odontoblasts clearly exhibited expression of amelogenin exons 8/9 peptides (Figs. 2E, 2F). In contrast, pre-odontoblasts (Fig. 2E) and mature odontoblasts (Figs. 2I–2M) were devoid of detectable staining. Within the enamel matrix, a strong biphasic pattern of staining was observed (Figs. 2J, 2K). Immunostaining was first concentrated at the dentino-enamel junction (Fig. 2J). Later, a second phase of intense staining was observed close to the ameloblast cell layer (Figs. 2J, 2K). No staining was seen after omission of the primary antibody (Figs. 2D, 2H).
Complete characterization of the AMELX gene structure is critical to our understanding not only of the complexity of enamel formation and biomineralization, but also of the role of amelogenin as a signaling molecule. Previous studies have showed that several amelogenin RNA transcripts are expressed by ameloblasts and odontoblasts (Simmer, 1994; Li et al., 1998; Papagerakis et al., 1999, 2003). In this study, we isolated and sequenced 3 novel mouse AMELX alternative spliced transcripts. Two of these transcripts contained sequences derived from mouse AMELX exons 8 and 9, which were characterized for the first time. Furthermore, a new mouse amelogenin alternative spliced variant, lacking exon 3, was also identified. Finally, amelogenin proteins encoded from alternatively spliced transcripts containing exons 8/9 have been shown to be expressed during odontogenesis in a tissue- and development-stage-specific manner. AMELX contains 7 exons encoding for a 196-amino-acid protein in mice and rats. Recently, Li et al.(1998) described an alternative 3' end rat transcript due to the presence of 2 additional exons downstream of exon 7. However, no sequence information has been provided for the human or mouse gene. Our study, in which we used RT-PCR and a tooth cDNA library or dental cell cDNAs, identified 2 major AMELX transcripts containing mouse amelogenin exons 8 and 9. Sequence analysis was performed and showed that the largest amplicon contained exons 2, 3, 5, 6, 8, and 9, with exons 4 and 7 being alternatively spliced. In contrast, the small amplicon contained exons 2, 3, 5, 6D, 8, and 9, with exons 4, 7, and a large part of exon 6 (6A-C) being spliced. In addition, a third newly characterized alternative spliced isoform has been isolated, with both exons 3 and 4 being deleted. In summary, 3 new alternative spliced forms have been characterized, adding to the complexity of amelogenin isoforms expressed during odontogenesis. The specific role of these alternative spliced isoforms remains to be determined. Many studies of splicing mechanisms have focused upon the basis for splice boundary selection for AMELX (for review, see Yuan et al., 2001). Amelogenin relative splicing potential depends on intron/exon sequence composition and length (Yuan et al., 2001). Here, exons 8 and 9 were always found transcribed together. Furthermore, intron regions, at the borders of exons 6, 8, and 9, important for binding of the splicing machinery, were highly conserved between species (personal communication), suggesting that optimal exon inclusion conditions exist for both exons 8 and 9. Exon 4 is the most commonly spliced (99%) exon (Simmer, 1994). It has been suggested that the inclusion or exclusion of amelogenin exon 4 may have a role in progenitor fibroblast cell differentiation (Veis et al., 2000). Our study failed to isolate cDNAs containing exon 4. It should be noted that some RT-PCR experiments resulted in a faint product with the expected size of a mouse amelogenin exon 8/9 full-length transcript. Further investigations are necessary to clarify the inclusion or exclusion of amelogenin exon 4 from AMEL exons 8/9 containing transcripts, as well as their potential function(s) during tooth development. In addition to the exclusion of exon 4, exon 7 was also absent from all transcripts containing exons 8 and 9. Amelogenin exon 7 contains a stop codon and poly (A) signal and is detected in all mouse AMEL cDNAs previously characterized (Simmer, 1994). The newly characterized exon 9 also contains a stop codon and poly (A) signal similar to those of exon 7 (this study), strongly suggesting that 2 alternative COOH-terminals exist for mouse and rat AMELX transcripts. However, the biological consequences of the alternative use between the 2 COOH-terminals need to be clarified. Immunolocalization was performed and showed that amelogenin 8/9 transcripts are translated into proteins in mouse tooth organs. Very intense staining was detected in early ameloblast cytodifferentiation; however, as secretory ameloblasts transitioned to mature ameloblasts, expression levels dramatically decreased to undetectable levels. In contrast to results reported in a previous study (Baba et al., 2002), our study also revealed staining in the mouse stratum intermedium cells (SI), not generally associated with amelogenin expression. Technical differences between the 2 studies may explain these results. Our data, showing amelogenin expression by SI cells, are supported by in vivo bovine AMELX promoter studies with the β-galactosidase reporter gene (Chen et al., 1994) and RNA studies that used in situ hybridization (Papagerakis et al., 2003). The function of amelogenin in the SI is not known and perhaps is related to cell signaling. Our results also demonstrated amelogenin exons 8/9-protein expression in young odontoblasts. Until recently, the localization of amelogenin was strictly confined to the ameloblasts, but novel evidence has showed amelogenin expression in porcine and rat odontoblasts (Veis et al., 2000; Papagerakis et al., 2003). Similarly, recombinant amelogenin M179 antibody (corresponding to exons 2–6) reacted positively within secretory-stage ameloblasts and weakly within odontoblasts during normal (Diekwisch et al., 2002) and pathological (Goldberg et al., 2002) tooth development. Our study demonstrated that amelogenin 8/9 mRNA and proteins are also transiently expressed by odontoblasts. Interestingly, similar alternative spliced profiles were found in odontoblasts and ameloblasts cells, suggesting that amelogenin alternative splicing is not cell-type-specific, as has also previously been suggested (Veis et al., 2000; Papagerakis et al., 2003; this study). Results from our study differ from the previously reported data describing that exon 7 COOH-terminal specific immunoreactivity was detected only at the outer layer of the immature layer of enamel (Uchida et al., 1991). This pattern has been attributed to rapid COOH-terminal proteolytic cleavage following amelogenin secretion into the enamel matrix (Brookes et al., 1995; Moradian-Oldak et al., 2001; Simmer and Hu, 2002). Analysis of our data clearly shows intense staining at the enamel-dentin junction and in the newly formed enamel proximal to the ameloblasts. The new COOH-terminal conferred by alternative splicing with exons 8/9, reported here, is apparently not being cleaved in the same manner as proteins containing the exon 7 COOH-terminal. Our study suggests that exons 8/9 COOH-terminal might have unique functions related to different matrix compartments of the developing enamel. Amelogenins self-assemble within the matrix to form nanospheres and interact with the crystal surfaces in this form (Paine et al., 2003; Shaw et al., 2004). It has also been suggested that the COOH- and NH2-terminals have different roles to play in this respect (Paine et al., 2003; Shaw et al., 2004), with the COOH-terminal at the exterior of the nanosphere. A quick comparison between exons 7 and 8/9 COOH-terminal sequences reveals that the exons 8/9 splice variant would have a less charged terminal compared with that of the exon 7 product. Analysis of these data suggests that the exons 8/9 COOH-terminal may have a different localization within the nanospheres, as well as a different affinity to hydroxyapatite and different ability to inhibit crystal growth. Additional studies are needed to evaluate the properties and potential roles of amelogenin exons 8/9 proteins in enamel formation and maturation. The newly characterized exons 8 and 9 may play important role(s) in enamel biomineralization and/or epithelial mesenchymal interactions during tooth development. However, it is not yet known if combinations of exclusion/inclusion of exons 4 and 6A-C may influence the potential role(s) of exons 8 and 9. The sequence information and alternative spliced profiles obtained in this investigation will be important for further evaluation of the functional significance of peptides produced from these novel exons.
This research was supported by grants from the ADEA William Gies Award (D-STAR), NIDCR T-32 DE 14318 (CO-STAR), and NIDCR RO1 DE 09875 (MM). Received for publication August 23, 2004. Revision received March 4, 2005. Accepted for publication April 8, 2005.
Journal of Dental Research, Vol. 84, No. 7,
613-617 (2005) This article has been cited by other articles:
|
|||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
 |
|
|
 |
|
Sign In to gain access to subscriptions and/or personal tools.
TOP
ABSTRACT
INTRODUCTION
