| Sign In to gain access to subscriptions and/or personal tools. |
Identification of Markers of the Midface
1 Department of Orthodontics & Pediatric Dentistry, University of Michigan School of Dentistry 1011 N. University Avenue, Ann Arbor, MI 48109-1078, USA; Correspondence: * corresponding author, sggong{at}umich.edu.
Currently, much remains unknown of the genes that mediate the biological events during growth and fusion of the midfacial region, and the possible pathways through which these genes function. We took advantage of high-throughput microarray analysis to search for genes that may play a critical role in the growth and fusion of the midfacial region to become the primary palate. We identified several genes that were potentially expressed at different levels between tail somite (TS) 6-8 (pre-fusion) and TS 12-14 (fusion) in the 3 midfacial processes. Expression of 4 of these genes (Tbx14/15, Dickkopf-1, Fibroblast Growth Factor 8, and Keratin-18) was further verified by reverse-transcription/quantitative PCR and in situ hybridization at the 2 stages of midfacial development. With the identification of these genes, and possibly others, functional analyses can be conducted to improve our understanding of the mechanisms and pathways by which the midface forms.
Key Words: primary palate • microarray analysis • RT-qPCR
During early embryonic craniofacial development, the 2 olfactory placodes invaginate to form nasal grooves that form the lateral boundaries of the frontonasal mass. The placodes then begin to curl outward around the edges to give rise to the lateral and medial nasal processes (LNP and MNP), both of which grow and eventually fuse with the maxillary process (MxP) to form much of the upper lip and primary palate. Failure of the growth and fusion of the 3 facial processes gives rise to the most common craniofacial birth defect, clefts of the lip with or without clefts of the secondary palate. Various genes have been implicated in the biological processes involved in growth and fusion of the midfacial processes (reviewed in Francis-West et al., 1998; Young et al., 2000). The analyses and functions of most of the genes during craniofacial development have been obtained primarily from a combination of their expression patterns during growth and development of the midfacial region, analysis of the phenotypes generated through loss-of-function experiments, functional studies, and linkage analyses in human populations. As yet, many of the possible pathways that operate in the patterning and regulation of growth and fusion of this region remain unclear. We took advantage of the Clontech Atlas Arrays to search for genes that may play a critical role in the growth and fusion of the midfacial region to become the primary palate. We identified about 70 genes that were potentially expressed at different levels between TS 6-8 (pre-fusion) and TS 12-14 (fusion) in LNP, MNP, and MxP. We describe here the quantitative and spatial characterization of 4 of these genes, Tbx14/15, Fgf8, Keratin-18 (K18), and Dickkopf-1 (Dkk-1).
Tissue Dissection and RNA Isolation Normal C57BL/6 embryos were harvested at 10.5 days post-coitum (dpc) and staged by tail somite (TS) count for pre-fusion (tail somite, TS, 6-8) and fusion (TS 12-14) of the midfacial processes (Gong and Guo, 2003). All animals were handled according to procedures approved by the policies of the University Laboratory Animal Medicine Department, the University of Michigan. Total RNA from the midfacial processes of about 10 embryos (for TS 6-8) and 6 embryos (for TS 12-14), usually from one litter of embryos, was isolated. Four pools of tissues were collected from each stage of development: One paired pool was sent to Clontech for microarrray analysis, and the other 3 paired pools were used for real-time RT-qPCR.
Analysis of Gene Expression by Clontech Atlas Arrays
Reverse-transcriptase and Real-time Polymerase Chain-reaction (RT-qPCR)
In situ Hybridization of RNA
The 2 groups of pooled midfacial tissues (pre-fusion, TS 6-8, and fusion, TS 12-14) were analyzed for gene expression profiling by means of the Clontech Atlas Gene Arrays. About 70 genes (Appendix Table II) showed a differential pattern of expression between the 2 samples, with 70 showing two-fold or greater changes in expression level. Fifteen genes were chosen for verification of expression changes during fusion by RT-qPCR (Appendix Table I). Three specific criteria were utilized in the selection of genes for further analyses: first, high differential expression levels at the 2 developmental stages examined, e.g., Fgf8, LimK1, Keratin18 (K18), BFSP1, Rad50; second, possible roles during craniofacial development, e.g., T-box genes, Dlx2, Fzd7; and third, possible roles in important biological events, e.g., Dkk1, Bcl-2, BAX, TNFR in apoptosis. Of the 15 genes we selected for testing by RT-qPCR, differential expression was verified by RT-qPCR in 3 genes, i.e., Tbx14/15, Dkk1, and K18. Tissue-specific expression patterns of these 3 genes and a fourth gene, Fgf8, during midfacial development were examined further by in situ hybridization.
Up-regulation of Tbx15 during Midfacial Development
Down-regulation of K18 during Midfacial Development The initial microarray screening analysis showed a 75% decrease in K18 expression after fusion of the midfacial processes. Verification by RT-qPCR showed a 40-65% decrease in K18 expression from pre-fusion (TS 6-8) to fusion (TS 12-14) midfacial tissues. This result was consistent with the microarray and spatial expression analyses. At TS6, K18 was expressed very strongly in all epithelial tissues covering the craniofacial region, including the epithelial cells of the olfactory placodes that will eventually form olfactory cells (Figs. 2A  , 2B). By TS14, expression of the gene had decreased and was more restrictive, being absent from most epithelium, except that lining the stomadeum (arrowheads in Fig. 2C), nasal pit, and the eye primordium (Fig. 2D). By 11.5 dpc, K18 was expressed in much of the ectodermal layer covering the presumptive olfactory cells in the nasal pit (Fig. 2E) and the epithelial layer of the tongue and oral cavity (Fig. 2F).
Up-regulation of Dkk1 during Development Microarray analysis indicated an increase in Dkk1 expression from TS 6-8 to TS 12-14. We verified this increase by RT-qPCR, using gene-specific primers, and showed a two- to 2.6-fold increase in Dkk1 mRNA in 3 independent sample pairs of TS 6-8 and TS 12-14. The magnitude of change as detected by RT-qPCR was comparable with the 2.4-fold change determined by the microarray analysis. This temporal change in expression levels was confirmed by spatial analysis of gene expression. Prior to fusion of the facial processes (TS6), Dkk1 was strongly expressed in the mesenchymal layer underlying the thickened epithelium immediately posterior to the putative site of fusion in the MxP (white arrowhead, Fig. 3A ), appearing over a broad domain further anteriorly around the putative site of fusion in the LNP and MNP (arrow, Fig. 3A). The expression level was lower in tissues further from the site of fusion, and higher in the MNP compared with the LNP (TS6, arrow in Fig. 3B). A similar pattern, but higher level, of expression in the site of fusion was observed in the TS11 embryo, where fusion of the midfacial processed had already occurred, and the epithelial seam (bilayer epithelium) formation was evident (es in Figs. 3C, 3D). By TS16, expression of the gene was strong in the medial surface of the MxP (Fig. 3E) and along the site of fusion (fusion line denoted by smaller black arrowheads in F). By this stage, Dkk1 transcripts were also found immediately underlying the ectodermal layer of the MNP and LNP (Fig. 3F).
Spatial and Temporal Expression of Fgf8 Microarray analysis indicated a 3.7-fold decrease in Fgf8 expression from TS 6-8 to TS 12-14. However, by RT-qPCR, we did not detect any significant changes in Fgf8 expression between TS 6-8 and TS 12-14 in 3 sets of samples isolated independently. The lack of significant changes in Fgf8 expression levels at the 2 main stages analyzed was corroborated by spatial analysis of the expression pattern of the gene in the midface (Appendix, Fig. 4).
Several challenges exist in analyses of midfacial growth. Midfacial growth is complex and involves morphogenetic changes that are intricate and often difficult to isolate experimentally. Furthermore, many of these morphogenetic events occur over a relatively short period of time. We chose to focus on a critical period of midfacial growth that we could stage efficiently and quickly, i.e., fusion of the midface. The results of our experiments showed that high-throughput technology was an effective way to identify the genes involved at 2 specific stages of midfacial morphogenesis. Our results also showed that our staging and dissection of the midfacial region worked well, as evidenced by the fact that all 3 genes have a restrictive pattern of expression, for the most part, to the midfacial region. It is clear from this experiment that replications with multiple samples are critical in minimizing false-positives. In addition, it is important that other techniques—such as RT-qPCR and in situ hybridization, as described here—be used to verify any potential changes. Since we were aware of the need for replications and statistical analysis, we were careful to verify only a limited number of genes by RT-qPCR and performed in situ hybridization only on genes that have been identified to be quantitatively changed with fusion of the midfacial processes. Results and experience obtained from this study will be the framework within which further experiments with the identification of new genes during midfacial development can be performed. Advances in high-throughput technology (e.g., Affymetrix® GeneChip technology) have created new possibilities for novel gene identification in the midface. Of the 4 genes that we analyzed, the expression pattern of Fgf8 in the midfacial region has been the most well-documented (Bachler and Neubuser, 2001; Firnberg and Neubuser, 2002). The absence of change in expression levels at the 2 stages was not surprising, in view of the fact that the rapid growth of the midfacial processes that occurs from TS6 to 12 would require that the expression levels of a molecule involved in directing outgrowth, such as Fgf8, be maintained during the period. Unlike Fgf8, the expression of Dkk1 in the midface has not been described previously. The Dkk1 gene is a family member of secreted proteins and is a potent inhibitor of wingless (Wnt)/beta-catenin signaling (Glinka et al., 1998), and appears to play a key role in embryonic patterning and specification of anterior structures (Glinka et al., 1998; Mukhopadhyay et al., 2001). Dkk1 is also believed to mediate epithelial-mesenchymal interactions and Bmp4-induced apoptosis during limb development (Mukhopadhyay et al., 2001; Grotewold and Ruther, 2002a,b). Bmp4 appeared to be expressed in the ectoderm overlying mesenchymal expression of Dkk1 in the fusion site (Gong and Guo, 2003). The increase in Dkk1 expression in the midface from TS6 to TS12, a period where extensive changes in both the growth and fusion of the midfacial processes occur, suggests a role for this gene in midfacial development. Functional studies must be performed to determine whether Dkk1 plays a role in mediating apoptosis during midfacial development. We also saw a four-fold decrease in K18 gene expression. K18 is also a substrate for caspase digestion during the course of epithelial cell apoptosis (Oshima, 2002), a cellular event that is believed to occur when the midfacial processes fuse (Gong and Guo, 2003). Analysis of our Tbx14/15 data suggests that it likely plays a role in the patterning process of the maxillary primordium and is probably not involved in mediating the fusion event of the midface. In this regard, it is interesting to note that several members of the T-box gene family, of which Tbx14/15 is a member, are involved in craniofacial development. Several human disorders have been linked to mutations in T-box genes, e.g., cleft palate with ankyloglossia/TBX22 (Packham and Brook 2003). Also, the mutation in the Dancer spontaneous mouse mutant, in which homozygotes present with clefts of the lip and palate, has been mapped to the Tbx10 gene (Bush et al., 2004). The 4 genes can be used as markers for further studies of midfacial growth and development. It is highly likely that these genes can act in concert to modulate the growth and fusion of the midfacial region. Analyses of the possible functions of these and other genes will lead to a better understanding of and appreciation for the biological events and molecular players and pathways involved in normal and abnormal midfacial morphogenesis.
The excellent technical expertise of Ms. Chiao Guo is greatly appreciated. This research was supported by the American Association of Orthodontists Foundation and by grant #1-FY02-180 from the March of Dimes Foundation, White Plains, NY 10605, USA.
A supplemental appendix to this article is published electronically only at http://www.dentalresearch.org. Received for publication February 5, 2004. Revision received September 30, 2004. Accepted for publication October 4, 2004.
Journal of Dental Research, Vol. 84, No. 1,
69-72 (2005)
|
|
||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
 |
|
|
 |
|
Sign In to gain access to subscriptions and/or personal tools.
TOP
ABSTRACT
INTRODUCTION
days old) were harvested from C57BL/6 timed-pregnant females mated and subjected to in situ hybridization as described by 
