| Sign In to gain access to subscriptions and/or personal tools. |
DMP1-targeted Cre Expression in Odontoblasts and Osteocytes
1 Department of Oral Biology, School of Dentistry, University of Missouri-Kansas City, Kansas City, MO 64108, USA; and Correspondence: * corresponding author, jfeng{at}bcd.tamhsc.edu
Odontoblasts in dentin and osteocytes in bone contain dendritic processes. To test if their dendrites share a common feature, we compared their cellular morphology as visualized using scanning electron microscopy. Analysis of our data showed that both cells share an identical dendritic canalicular system and express extensive processes forming a complex network within the mineralized matrix. Because dentin matrix protein 1 (DMP1), an extracellular matrix protein, is highly expressed in both types of cells, we next tested, using a transgenic approach, whether a 9.6-kb Dmp1 promoter-4-kb 1st intron would be able to target Cre cDNA in these cells for expression/deletion of other genes in odontoblasts and osteocytes. We determined the specificity and efficiency of Cre activity by crossing Dmp1-Cre mice with ROSA26 reporter mice. Results showed that odontoblasts and osteocytes were specifically targeted, suggesting that this animal model will be useful for the preferential study of gene functions in both types of cells.
Key Words: DMP1 • transgenic mouse • Cre recombinase • osteocytes • odontoblasts
Cells that become embedded and remain embedded in a mineralized matrix for decades have unique characteristics compared with other cells in the body. The functions of these cells have been difficult to discern, due to this unique property—their localization in mineral. Targeted gene/protein expression and deletion approaches are common for many cell types and many tissues, but not for cells partially embedded in a mineralized matrix, such as odontoblasts, or for cells completely embedded in a mineralized matrix, such as osteocytes derived from osteoblasts (Bonewald, 2006a). Note that the odontoblast shares characteristics with both the osteoblast (a cell outside the mineralized matrix) and the osteocyte (numerous processes that penetrate the mineralized matrix). Whereas the major function of an odontoblast is clear, that of making and maintaining predentin and dentin, the function of the osteocyte, is less clear. The odontoblast most likely has functions other than making and mineralizing matrix, and may have mechanosensory functions similar to those proposed for the osteocyte. The osteocyte cells make up over 90–95% of all bone cells in the adult skeleton, are viable for decades, and are thought to sense mechanical strain and coordinate adaptive bone-remodeling responses (Bonewald, 2006b). Recently, it has been proposed that these cells may regulate phosphorus homeostasis, and that the osteocyte lacuno-canalicular network in bone may function as an endocrine organ (Feng et al., 2006). However, due to the difficulties in accessing these cells in vivo, or isolating and growing these cells in vitro, very little is known about this bone cell compared with other cells, such as osteoclasts and osteoblasts. Numerous key genes/proteins must be responsible for both odontoblast and osteocyte function. The gene-targeting approach is very powerful to ’dissect out’ the function of an individual gene from the entire genome. A prime example is the deletion of the Dmp1 gene that leads to defects in odontoblast morphology and function (Ye et al., 2004), as well as osteocyte abnormalities (Feng et al., 2006). However, unlike Dmp1 conventional deletion, deletion of many other genes results in embryonic or early postnatal lethality, or severe defects (Davey et al., 2004). The Cre/loxP approach is now the method of choice for the specific targeting of various cell types (Nagy, 2000). Cre is a recombinase that mediates intra-molecular and inter-molecular site-specific recombination between loxP sites (two 13-bp inverted repeats separated by a 9-bp asymmetric spacer region). The precise removal of DNA between two loxP sites can then be used to eliminate an endogenous gene of interest (Gu et al., 1994). This system can also be used to activate a transgene by cutting out an intervening lacZ gene sequence between the promoter and the transgene (Fukuda et al., 2006). For studies of genes important to odontoblast cells, investigators have established a Cre mouse line using a promoter for dentin sialophosphoprotein (Sreenath et al., 2003b); however, there is no report on the use of this line for targeting genes in odontoblasts. Currently, there is no Cre line targeted to osteocytes. DMP1 is a matrix protein, initially isolated from dentin (George et al., 1993), that is highly expressed in osteocytes (Toyosawa et al., 2001; Feng et al., 2002) and odontoblasts (D’Souza et al., 1997). The murine Dmp1 promoter has been successfully used to drive GFP in osteocytes (Kalajzic et al., 2004) or lacZ in odontoblasts (Lu et al., 2005). In this study, we hypothesized that both odontoblast processes and osteocyte dendrites share similarity in their canalicular systems, and that a 9.6-kb Dmp1 promoter-4 kb 1st intron, highly active in odontoblasts and osteocytes, would be able to target Cre cDNA in these two cells for expression/deletion of other genes in odontoblasts and osteocytes.
Mice All animal studies were reviewed and approved by the Institutional Animal Care and Use Committee of the University of Missouri-Kansas City. The Dmp1-null mice (Feng et al., 2003) and the 9.6-kb –lacZ mice (Lu et al., 2005) were generated as described previously. A CD-1 background was used in this study.
Generation of the Dmp1-Cre Transgenic Mice
Analysis of Cre Expression and Recombination in Osteocytes and Odontoblasts
Sample Preparation and β-galactosidase Assay
Imaging Resin-cast Osteocyte Dendrites and Odontoblast Processes by Scanning Electronic Microscopy (SEM)
Odontoblasts and Osteocytes Had Similar Dendritic Morphology, as Determined by Three-dimensional (3-D) Images. To compare the three-dimensional structures of canalicular systems in odontoblast and osteocyte cells at EM levels, we used the acid-etched resin-cast technique (Feng et al., 2006). With this technique, a polished surface from resin-embedded bone or dentin was etched with acid to remove mineral, leaving a relief cast of the non-mineralized structures that had been penetrated by resin. Dentinal tubules from odontoblasts branched into numerous anastomosing canalicular systems, penetrating the entire dentin (Fig. 1a  ). Similarly, the alveolar bone was filled with an extensive osteocyte lacuno-canalicular system, which penetrated the entire bone matrix (Fig. 1b). Analysis of the data showed a striking similarity in the morphological distribution of both canalicular systems in dentin and alveolar bone.
The 9.6-kb Murine Dmp1 Promoter was Highly Active in Osteocytes, as Determined with the LacZ Reporter. Previously, we reported the development of a mouse model in which the LacZ reporter was driven by a 9.6-kb murine Dmp1 promoter, with high activity in odontoblasts (Lu et al., 2005). In the present study, we further analyzed and compared the lacZ expression pattern of this promoter in osteocytes with that of Dmp1 lacZ knock-in mice. Six-day-old femoral sections from the 9.6-kb Dmp1-lacZ mouse line (Fig. 2 , middle panel) displayed an expression pattern very similar to that in the Dmp1 lacZ knock-in mice (Fig. 2, left panel). As expected, the wild-type control showed no signal.
The 9.6-kb Murine Dmp1 Cre was highly active in odontoblasts. Because the 9.6-kb murine Dmp1 promoter was active in both odontoblasts (Lu et al., 2005) and osteocytes (Fig. 3 ), we next generated Dmp1-Cre transgenic mice using the same promoter and the first intron linked to Cre recombinase cDNA. We obtained 11 founder mice carrying the Dmp1-Cre transgene, as determined by PCR analysis of tail DNAs. Three of the founder mice were partially characterized (data not shown), and a detailed analysis on one founder was performed and is presented in this report.
To test Cre recombination in odontoblasts, we crossed Dmp1-Cre transgenic mice with the ROSA26R mice. In this transgenic mouse line, the lacZ gene was inserted into the ubiquitously expressed ROSA locus, where a transcriptional stop cassette was flanked by two loxPs (Soriano, 1999). In the presence of Cre, the blocking DNA sequence flanked by loxP sequences was excised, and the lacZ gene then activated. The presence of both ROSA26R and Dmp1-Cre transgenes was confirmed by PCR for Cre and β-Gal, and activation of lacZ was assayed by x-Gal staining. Functional analysis of Cre recombinase, reflected by x-Gal staining, was carried out from embryonic day 18.5 (E18.5) to the age of 4 mos.
Previously, we have reported that the 9.6-kb Dmp1-lacZ mice showed a weak lacZ expression during early odontogenesis, but strong promoter activity during postnatal odontogenesis (Lu et al., 2005). In the double-transgenic mice, there was little or no lacZ expression detected in odontoblast cells at E18.5 (data not shown). In contrast, lacZ signal was high in odontoblasts by either whole-mount x-gal staining (Fig. 3a
The 9.6-kb Murine Dmp1-Cre Transgene was Highly Active in Osteocytes.
Odontoblast processes in dentin and osteocyte dendrites in bone contact each other through thin ’tunnels’ (100–300 nm) called canaliculi. A high mineral content where canaliculi are captured greatly challenges studies of the morphologies of these structures. In this study, we used acid-etched resin samples from 1st molars and mandibles and compared similarities of canaliculi at EM levels. The striking similarity in their three-dimensional structures suggests that both canalicular systems may work similarly in mechanical signal transduction. A recent report estimated that approximately 10% of mouse genes have been conventionally deleted, and that these animal models were extremely useful in studies of gene function and identification of human mutations (Austin et al., 2004). For example, studies of dentin sialophosphoprotein (DSPP) null mice (Sreenath et al., 2003a) led to the conclusion that this phosphoprotein is essential for normal tooth development. Studies of Msx1 null mice (Satokata and Maas, 1994) resulted in identification of the homeodomain missense mutation, which causes selective tooth agenesis in humans (Vastardis et al., 1996). With the discovery of the Cre-loxP system, gene-targeting technology has become more specific with regard to tissue and age selectivity. However, progress toward understanding of the functions of odontoblast and osteocyte cells is minimal compared with what is known about other cells and cells which play a role in mineralized tissue, such as osteoblasts or osteoclasts. To target genes in both odontoblasts and osteocytes, we utilized the specific expression of a promoter fragment of DMP1, a matrix protein highly expressed in odontoblast and osteocyte cells. First to determine optimal expression, a lacZ reporter transgenic mouse line under the control of the 9.6-kb murine Dmp1 promoter was generated and showed high lacZ expression in mature odontoblast cells (Lu et al., 2005). In parallel, Kalajzic and his colleagues generated Dmp1-GFP mice using a 7.9-kb murine Dmp1 promoter targeted to osteocytes (Kalajzic et al., 2004). Since we had already identified an enhancer domain between the –7.9-kb and the –9.6-kb region (Lu et al., 2005), we generated Dmp1-Cre mice using this 9.6-kb murine promoter with the whole 4-kb intron 1. Surprisingly, the Dmp1-Cre mouse line showed little or no Cre activity (reflected by lacZ expression) in odontoblast or osteocyte cells during embryonic development. In contrast, the Cre activity was high in both odontoblast and osteocyte cells by the age of 4 mos and showed little or no ectopic expression in the muscle, kidney, liver, or brain (data not shown). We speculate that this phenomenon could be due to position effects on the transgene, or due to reduced efficiency of Cre-mediated deletion of the floxed sequences in early developmental stages. Note that Cre efficiency is in the 60–90% range and rarely reaches 100%, even in the most efficient cases (Xu et al., 1999; Li et al., 2003; Coumoul et al., 2005). Therefore, the Cre mouse line described here should prove particularly useful for targeting genes in odontoblast and osteocyte cells during postnatal or adult stages. Although here we demonstrated Cre activity under control of the 9.6-kb Dmp1 promoter by breeding onto ROSA26R mice, the final definitive determination of usefulness of this Cre mouse line will be to delete genes in odontoblast or osteocyte cells. Our future plans are to cross this Dmp1-Cre mouse line with β-catenin-loxP mice (Brault et al., 2001), or bone morphogenic protein receptor 1A (Mishina et al., 2002) or E11/gp38, thought to play a role in the formation of dendritic processes (Zhang et al., 2006). This Cre mouse line will also provide a valuable tool for other investigators to study the critical function of globally expressed genes by specific, targeted deletion in osteocytes and odontoblasts postnatally. Finally, we expect that the Dmp1-Cre transgenic mouse line will be used to determine the functions of numerous genes/proteins in odontoblasts and osteocytes. Investigators, including ourselves, will be able to cross these mice with a growing collection of loxP mice that have been generated or will be available in the future. We also propose that many of these genes play a role in dendrite formation and function.
The authors appreciate the use of the University of Missouri-Kansas City SEM Facility (J.D. Eick, Director). This study was supported by US National Institutes of Health grants to J.Q.F. (AR051587, AR046798) and L.F.B. (AR046798), and by a Chancellor Fellowship from the University of Missouri-Kansas City (to Y.L.). Received for publication November 7, 2006. Revision received January 30, 2007. Accepted for publication February 5, 2007.
Journal of Dental Research, Vol. 86, No. 4,
320-325 (2007) This article has been cited by other articles:
|
|
||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
 |
|
|
 |
|
Sign In to gain access to subscriptions and/or personal tools.
TOP
ABSTRACT
INTRODUCTION
