HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS		QUICK SEARCH:   [advanced] Author: Keyword(s): Year:  Vol:  Page:

This Article Abstract Figures Only Full Text (PDF) Appendix Services Similar articles in this journal Similar articles in ISI Web of Science Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Citing Articles Citing Articles via Google Scholar Google Scholar Articles by Ohira, T. Articles by Takashiba, S. Search for Related Content PubMed PubMed Citation Articles by Ohira, T. Articles by Takashiba, S.

Identification of Genes Differentially Regulated in Rat Alveolar Bone Wound Healing by Subtractive Hybridization

Department of Pathophysiology-Periodontal Science, Okayama University Graduate School of Medicine and Dentistry, 2-5-1 Shikata-cho, Okayama 700-8525, Japan;

^* corresponding author, stakashi{at}cc.okayama-u.ac.jp

	ABSTRACT

TOP ABSTRACT INTRODUCTION MATERIALS & METHODS RESULTS DISCUSSION REFERENCES

 
Periodontal healing requires the participation of regulatory molecules, cells, and scaffold or matrix. Here, we hypothesized that a certain set of genes is expressed in alveolar bone wound healing. Reciprocal subtraction gave 400 clones from the injured alveolar bone of Wistar rats. Identification of 34 genes and analysis of their expression in injured tissue revealed several clusters of unique gene regulation patterns, including the up-regulation at 1 wk of cytochrome c oxidase regulating electron transfer and energy metabolism, presumably occurring at the site of inflammation; up-regulation at 2.5 wks of pro--2 type I collagen involving the formation of a connective tissue structure; and up-regulation at 1 and 2 wks and down-regulation at 2.5 and 4 wks of ubiquitin carboxyl-terminal hydrolase l3 involving cell cycle, DNA repair, and stress response. The differential expression of genes may be associated with the processes of inflammation, wound contraction, and formation of a connective tissue structure.

KEY WORDS: subtractive hybridization • gene expression • alveolar bone • wound healing

	INTRODUCTION

TOP ABSTRACT INTRODUCTION MATERIALS & METHODS RESULTS DISCUSSION REFERENCES

 
The healing of periodontal tissues damaged by any kind of injury requires the participation of several regulatory molecules and cell types, and involves a series of overlapping stages that include inflammation, granulation tissue formation, and tissue remodeling. The healing presumably involves several cell types: fibroblasts for soft connective tissues, cementoblasts for cementogenesis, osteoblasts for bone, and endothelial cells for angiogenesis. During the healing process, these cells must interact with a variety of mediators, and the course of healing may be directed by a combination of molecule-cell, cell-matrix, and cell-cell interactions. However, little is known about the signals that initiate and regulate these interactions in vivo.

The management of periodontal defects—including destruction of the periodontal ligament, cementum, and the formation of infrabony defects—has always been a challenge in clinical periodontics. Complete restoration of the alveolar bone is necessary for periodontal healing and regeneration. However, it does not usually occur on a clinically predictable basis once the destructive phase reaches the alveolar bone and other deep periodontal structures. Only a few growth factors—fibroblast growth factor-2 (Takayama et al., 2001; Murakami et al., 2003), bone morphogenetic protein 2 (Sigurdsson et al., 1995), and transforming growth factor (TGF) ß-1 (Wikesjö et al., 1998)—have been shown to enhance periodontal regeneration or wound healing in vivo, although several soluble factors and matrix have been suggested to regulate various cellular functions in periodontal tissue. Many factors, including genes unidentified to date, may be associated with the wound healing of alveolar bone. Therefore, it is important for our understanding of the basis of periodontal wound healing to identify the genes expressed in damaged alveolar bone.

Subtractive hybridization is aimed at identifying mRNA molecules that differ in abundance between target and driver pools. We have modified a subtractive hybridization technique and then amplified the target cDNA by polymerase chain-reaction (PCR). From a small amount of mRNA, we have recently succeeded in extracting the unique genes expressed in human periodontal ligament cells in vitro (Myokai et al., 2003).

In this study, we aimed to identify the genes whose expression is up-regulated or down-regulated in rat alveolar bone wound healing. The genes identified by the subtractive hybridization were examined for mRNA enrichment during the wound healing, and their sequence similarities with known genes were analyzed.

	MATERIALS & METHODS

TOP ABSTRACT INTRODUCTION MATERIALS & METHODS RESULTS DISCUSSION REFERENCES

 
Mechanical Injury, Tissue Preparation, and cDNA Synthesis
Twenty Wistar rats (male, 10–12 wks old), each weighing from 300 to 350 g, were used. The experimental protocol was carried out according to the guidelines for animal care of Okayama University Dental School. Rats were deeply anesthetized with an intraperitoneal injection of 5% sodium pentobarbital (Nembutal, Dianippon Pharmaceutical Co., Suita, Japan) at a dose of 30 mg/kg. A cavity approximately 3 mm deep was prepared in the alveolar bone of the maxillary first molar after a full-thickness flap had been made. One, 2, 2.5, and 4 wks after the flap had been repositioned, the full-thickness flap was removed, and the tissues proliferating in the cavity were then harvested from the rats by means of dental curettes (Fig. 1e). For a healthy control, alveolar bone was taken from the maxilla at the first molar on the opposite side in the same rat. Total RNA (300 ng) was isolated from the two tissues by the acid guanidinium thiocyanate-phenol-chloroform extraction method (Chomczynski and Sacchi, 1987). Target and driver single-stranded (ss) cDNAs bound to the oligo(dT)-coupled magnetic beads (Dynal, Lake Success, NY, USA) were synthesized from the total RNA by reverse transcriptase (Superscript II; Invitrogen, Carlsbad, CA, USA) at 42°C for 1 hr, and they were used for subtraction.

View larger version (48K): [in this window] [in a new window]   Figure 1. Histology of wound and detection of genes. (A) Histological findings of alveolar bone wound. Periodontium 1 wk after injury: Granulation tissues were observed in the defect (a,b). Two wks after injury: Granulation tissues and blood vessels were observed in the defect (c). Similar changes were observed at 2.5 wks (data not shown). Four wks after injury: Granulation tissues were contracted and connective tissue was partially remodeled (d). Bar equals 300 µm. After removal of the full-thickness flap, tissue proliferating in the bone cavity was recovered by dental curette (e). (B) General procedure of subtractive hybridization. The target c-sscDNA was synthesized from the target sscDNA-beads, and an auto-subtraction was performed. The target c-sscDNA was subtracted twice from the driver sscDNA-beads. The target c-sscDNA (up-regulated and down-regulated genes) was amplified by PCR. The PCR products were subjected to electrophoresis and used for cloning. (C) Representation of GAPDH cDNA after sequential subtraction. The amount of GAPDH cDNA in the sample was analyzed by PCR with primers designed for the 3' non-coding region of the rat GAPDH cDNA: sense, 5'-TGAAGGTCGGTGTCAACGGATTTGGC-3'; antisense, 5'-CATGTAGGCCATGAGGTCCACCAC-3'. The following templates were used: cDNA from the one-week injured tissue (cDNA-injury), cDNA-injury subtracted once (cDNA-sub1), and cDNA-injury subtracted twice (cDNA-sub2). The amplification was performed for 20, 25, 30, and 35 cycles. (D) Display of amplified cDNAs followed by two-round subtraction. The PCR products underwent gel electrophoresis. Lane 1, one-week up-regulated genes; lane 2, two-week up-regulated genes; lane 3, 2.5-week up-regulated genes; lane 4, four-week up-regulated genes; lane 5, one-week down-regulated genes; lane 6, two-week down-regulated genes; lane 7, 2.5-week down-regulated genes; lane 8, four-week down-regulated genes; and lane M, 100-bp DNA ladder.

Reciprocal Subtractive Hybridization and Cloning
Reciprocal subtractive hybridization between the two cDNAs from injured and control tissues was performed, and the general procedure is outlined in Fig. 1B. The procedure has been described previously (Myokai et al., 2003). Briefly, the target complementary sscDNA (c-sscDNA) was synthesized from the target sscDNA-beads by the KlenTaq polymerase reaction (Clontech, Palo Alto, CA, USA) with an EcoRI-dT primer (5'-GGCGAATTCTGCAGTTTTTTTTTTTTTT-3'), and an auto-subtraction was performed at 75°C for 24 hrs. The target c-sscDNA was subtracted twice from the driver sscDNA-beads in 1 x KlenTaq PCR Buffer (Clontech) at 75°C for 24 hrs. The target c-sscDNA solution was recovered, and 1 µL of this solution was used for PCR with the EcoRI-dT primer. The PCR products were separated on a 3% agarose gel and visualized by ethidium bromide staining. The amplified cDNA fragments longer than 400 bp were recovered from the gel and cloned into the EcoRI site of a pUC118 plasmid vector (Takara, Otsu, Japan). All plasmids were prepared for further analysis with the use of Qiagen Plasmid Miniprep Kits (Qiagen, Hilden, Germany). We monitored the efficiency of each round of subtraction by analyzing the cDNA encoding glyceraldehyde-3-phosphate dehydrogenase (GAPDH) by PCR.

Reverse Northern Hybridization
Reverse Northern hybridization was performed by the method described previously (Myokai et al., 2003). Plasmids containing cDNA fragments longer than 400 bp were used as target genes for hybridization. In addition, we selected 7 known cDNAs as targets for hybridization: osteocalcin (BGP), core-binding factor a1 (Cbfa1), TGFß-1, activin receptor-like kinase (ALK) 5, type II receptor for TGFß (TGFßRII), type III receptor for TGFß (TGFßRIII), and ß-actin. The plasmid (500 ng), digested with EcoRI, was subjected to 3% agarose gel electrophoresis and transferred to a Hybond N⁺ membrane (Amersham Bioscience, Tokyo, Japan). Total RNA (100 ng) isolated from the injured and control tissues was reverse-transcribed with the use of Superscript II (Invitrogen), and then labeled with [-³²P] dCTP with a Bca BEST labeling kit (Takara) according to the manufacturer’s instructions. The membranes were incubated at 68°C for 1 hr in ExpressHyb hybridization solution (Clontech) containing the probe at a concentration of 5 x 10⁵ cpm/mL, and then washed finally with 1 x SSC containing 0.1% SDS at 68°C for 30 min. The hybridization signals were visualized in a Bio Imaging Analyzer (BAS 2000; FUJI, Tokyo, Japan). The signal intensity of each cDNA was quantified with NIH Image (Ver. 1.62) and normalized against that of GAPDH. Data analysis was performed by the k-means clustering technique, with the use of GeneSpring software version 6 (Silicon Genetics, Redwood, CA, USA). To confirm the reverse Northern hybridization data, we made the quintuple blots using the 7 kinds of cDNAs for targets, and then hybridized the blots with the mixture of probes at different concentrations (Fig. 2A). The hybridization signals were visualized and quantified as mentioned above.

View larger version (36K): [in this window] [in a new window]   Figure 2. Confirmation of quantitation by reverse Northern hybridization. (A) Detection of known cDNAs. Seven known cDNAs (BGP, Cbfa1, TGFß-1, ALK5, TGFß-RII, TGFß-RIII, and ß-actin) were amplified by PCR and cloned. Each clone (500 ng) was digested with EcoRI, subjected to gel electrophoresis (a), and then transferred to the membrane. Lanes: 1, 100-bp ladder; 2, BGP; 3, Cbfa1; 4, TGFß-a; 5, ALK5; 6, TGFß-RII; 7, TGFß-RIII; and 8, ß-actin. The membranes were hybridized with a mixture of probes at different concentrations as described in the lower table (b-f). To standardize the total amount of labeled probe, 5 x 105 cpm/mL of GAPDH probe was added to the mixture. The PCR primers used were: BGP sense, 5'-CTGAGTCTGACAAAGCCTTC-3', and BGP antisense, 5'-CCATAGAT GCGCTTGTAGGC-3'; Cbfa1 sense, 5'-ACCTCTGACTTCTGCCTCTG-3', Cbfa1 antisense, 5'-CGCCAAACAGACTCATCCAT-3'; TGFß-1 sense, 5'-CATGACATGAACCGGCCCTT-3', TGFß-1 antisense, 5'-AAATATA GGGGCAGGGTCCC-3'; ALK5 sense, 5'-GGACGCAGCTGTGGTTGGTG-3', ALK5 antisense, 5'-TTCCACCAATAGAACAGCGT-3'; TGFßRII sense 5'-CTTGACCTGTTGCCTGTGTG-3', TGFßRII antisense 5'-CATGCTCTCC ACACAGGGGT-3'; and TGFßRIII sense 5'-TACACCATCATCG AGAACAT-3', TGFßRIII antisense 5'-GAGTAG ATGTACCACAAGGC-3'. The ß-actin primers were purchased from Clontech (Rat Control Amplimer Set). Complementary DNA (1 ng) from injured rat tissue or mouse embryo was amplified by PCR according to the primers described above. After cDNAs were cloned, the nucleotide sequences were confirmed. (B) Quantitation of hybridization signals. The signal intensity of each cDNA was quantified with NIH Image and normalized against that of ß-actin. The mean value of 7 kinds of targets to the same probe concentration is plotted, and error bars indicate standard deviation.

	RESULTS

TOP ABSTRACT INTRODUCTION MATERIALS & METHODS RESULTS DISCUSSION REFERENCES

 
Genes Isolated from Tissue
Changes in mRNA expression for 6 known genes (BGP, Cbfa1, TGFß-1, ALK5, TGFßRII, and TGFßRIII) were detected in the injured tissue (Fig. 3A and Appendix). In addition, histological changes were observed in the injured periodontium (Fig. 1A). These results suggest that cDNA obtained from the tissues in this rodent model was suitable for the identification of genes whose expression is regulated in wound healing.

View larger version (22K): [in this window] [in a new window]   Figure 3. Expression patterns of genes and phases of wound repair. (A-E) Clustering of genes regulated in alveolar bone wound healing. On the basis of the changes in level of expression, 34 clones from injured tissues (Table) and 6 known genes were clustered into 5 groups: A, Cluster I (8 clones and ALK5); B, Cluster II (11 clones and BGP); C, Cluster III (3 clones, Cbfa1, and TGFß1); D, Cluster IV (9 clones and TGFßRIII); and E, Cluster V (3 clones and TGFßRII). (F) Phases of wound repair. The wound-healing process has been divided into three phases: (1) inflammation, (2) re-epithelialization, and granulation tissue formation, and (3) matrix formation and remodeling. This is modified from the figure described by Clark (1996).

After the second round of subtraction and amplification by PCR, highly expressed genes that were responsive to the injury were detected by gel electrophoresis (Fig. 1D). The 250 clones for up-regulated genes and 150 clones for down-regulated genes were isolated, and then 68 containing fragments longer than 400 bp were examined for mRNA enrichment.

Verification of Reverse Northern Hybridization Results
To confirm the reverse Northern hybridization data, we selected the 7 known cDNAs as targets, and hybridized the quintuple blots with the mixture of probes at different concentrations (Fig. 2A). The intensity of the band depended on the concentration of the probe, suggesting that this hybridization method succeeded in the quantitation of the cDNA synthesized from the tissues (Fig. 2B).

Clustering of Genes Regulated in Alveolar Bone Wound Healing
To visualize typical gene expression patterns, we clustered into five groups the 34 clones whose mRNA expression was detected in the injured tissue (Fig. 3, Table, and Appendix). Clusters I (21% of the clones) and V (9% of the clones) included mainly genes whose expression was up-regulated at 1 wk and recovered their basal levels thereafter. Cluster II (32% of the clones) displayed up-regulated expression at 1 and 2.5 wks and down-regulated expression at 2 and 4 wks, and Cluster IV (29% of the clones) showed up-regulated expression at 1 and 2.5 wks. However, Cluster III (9% of the clones) included genes that showed no significant change in mRNA level during wound healing. In general, wound contraction occurred at 1 wk, and collagen accumulated thereafter (Fig. 3F). The wound contraction phase corresponded to up-regulation of mRNA expression in clusters I and V; however, we did not see a relationship between the other cluster and a particular stage of wound healing.

	DISCUSSION

TOP ABSTRACT INTRODUCTION MATERIALS & METHODS RESULTS DISCUSSION REFERENCES

 
The regulation of alveolar bone wound healing is a complex process involving hormones and local factors acting in an autocrine and/or paracrine manner on the generation and activity of differentiated bone cells. Connective tissue wound healing has been arbitrarily divided into three phases: (1) inflammation, (2) re-epithelialization and granulation tissue formation, and (3) matrix formation such as deposition of proteoglycan and collagen fibrils, and tissue remodeling, including both continued collagen synthesis and collagen catabolism (Fig. 3F). Bone remodeling also occurred during the third phase. In our study, the granulation tissue for the second phase, which overlaps with the wound contraction, was observed at 1 and 2 wks (Figs. 1A-a,b,c), and the wound contraction and remodeled tissue for the third phase were seen at 4 wks (Figs. 1A-d). We focused on genes expressed in the injured tissues at the second and third phases, because bone remodeling, whereby osteoclasts resorb and osteoblasts reform bone, is an essential function in the repair of alveolar bone wounds.

Thirty-four genes were assigned to clusters based on their changes in level of expression, and the known genes were analyzed in the light of their functional annotation and wound-healing phase. Cluster I included COX subunit II and VIa, which are terminal enzymes of the mitochondrial respiratory chain and regulate both electron transfer and energy transduction. In the first phase of wound healing, injury causes the infiltration of white blood cells into the tissues and induces the continuous synthesis and secretion of growth factors and cytokines. The up-regulation of the COXs suggests that high-energy metabolism occurs at the site of inflammation. Cluster II contained the pro--2 type I collagen (Table), which belongs to the collagen superfamily comprised mainly of extracellular structural proteins involved in the formation of a connective tissue structure. Up-regulation of the gene at 2.5 wks seems to be consistent with collagen accumulation shortly after the onset of granulation tissue formation (Fig. 3). Cluster IV contained the uch l3, whose expression was up-regulated at 1 and 2 wks, and down-regulated at 2.5 and 4 wks. The UCHs are implicated in the proteolytic processing of polymeric ubiquitin, and the carboxyl terminal processing of ubiquitin precursors and ubiquitin-like proteins is essential for their subsequent conjugation to target protein. Since ubiquitin-mediated protein degradation plays a critical role in cellular functions such as cell cycle, DNA repair, and stress response (Finley and Chau, 1991), the UCH l3 may act in the cellular functions in the injured tissue. This cluster also contained DSSP, which displayed no significant change in expression during wound healing (Fig. 3, Appendix). It is interesting that DSSP gene mRNA was detected in alveolar bone wound healing. Because DSSP was recently shown to be expressed not only in dentin and odontoblasts but also in bone, it may have a role in osteogenesis (Qin et al., 2002). In addition, clusters I and V included mainly genes whose expressions were up-regulated during the wound contraction phase (Fig. 3). Myofibroblasts are specialized fibroblasts considered to be responsible for granulation tissue contraction (Martin, 1997), and a marker of fibroblast-myofibroblast modulation is the neo-expression of -smooth-muscle (-SM) actin (Skalli et al., 1986; Darby et al., 1990). However, we did not detect -SM actin in both clusters. This may be due to the lack of difference in -SM actin mRNA levels between injured and control tissues. Among 4 kinds of unknown genes, we may find molecules responsible for wound contraction.

This study showed that the genes expressed differentially in alveolar bone wound healing could be assigned to clusters based on their changes in level of expression. The windows of time that were used here were broad, so that dynamic changes in gene expression were not detected. However, we could propose clusters displaying different gene expression patterns that might be associated with alveolar bone wound healing. In addition, from these clusters, including newly identified genes, we may find new molecules that could contribute to periodontal healing.

In summary, we identified and clustered the genes whose expressions are differentially regulated and analyzed their relationships to alveolar bone wound healing. The clusters appear to display different gene expression patterns that may be associated with the various phases of alveolar bone wound healing. The differential expression of genes, including newly identified genes, may be associated with the processes of inflammation, wound contraction, and formation of a connective tissue structure.

	ACKNOWLEDGMENTS

 
This study was supported by Grants-in-Aid for Scientific Research from the Japan Society for the Promotion of Science (No. 16659579 to FM, No.14370710 to ST), by the Ryobi Teien Foundation, the Inamori Foundation, and the Kobayashi Magobe Memorial Medical Foundation (FM).

	FOOTNOTES

 
A supplemental appendix to this article is published electronically only at http://www.dentalresearch.org.

Received June 2, 2003; Last revision April 30, 2004; Accepted May 6, 2004

	REFERENCES

TOP ABSTRACT INTRODUCTION MATERIALS & METHODS RESULTS DISCUSSION REFERENCES

 
Adams MD, Kerlavage AR, Fields C, Venter JC (1993). 3,400 new expressed sequence tags identify diversity of transcripts in human brain. Nature Genet 4:256–267.[ISI][Medline]

Chomczynski P, Sacchi N (1987). Single-step method of RNA isolation by acid guanidinium thiocyanate-phenol-chloroform extraction. Anal Biochem 162:156–159.[ISI][Medline]

Clark RAF (1996). Wound repair: overview and general considerations. In: The molecular and cellular biology of wound repair. Clark RAF, editor. New York: Plenum Press, pp. 3–50.

Darby I, Skalli O, Gabbiani G (1990). -smooth muscle actin is transiently expressed by myofibroblasts during experimental wound healing. Lab Invest 63:21–29.[ISI][Medline]

Finley D, Chau V (1991). Ubiquitination. Annu Rev Cell Biol 7:25–69.[ISI][Medline]

Martin P (1997). Wound healing—aiming for perfect skin regeneration. Science 276:75–81.[Abstract/Free Full Text]

Murakami S, Takayama S, Kitamura M, Shimabukuro Y, Yanagi K, Ikezawa K, et al. (2003). Recombinant human basic fibroblast growth factor (bFGF) stimulates periodontal regeneration in class II furcation defects created in beagle dogs. J Periodontal Res 38:97–103.[ISI][Medline]

Myokai F, Oyama M, Nishimura F, Ohira T, Yamamoto T, Arai H, et al. (2003). Unique genes induced by mechanical stress in periodontal ligament cells. J Periodontal Res 38:255–261.[ISI][Medline]

Qin C, Brunn JC, Cadena E, Ridall A, Tsujigiwa H, Nagatsuka H, et al. (2002). The expression of dentin sialophosphoprotein gene in bone. J Dent Res 81:392–394.[Abstract/Free Full Text]

Sanger F, Nicklen S, Coulson AR (1977). DNA sequencing with chain-terminating inhibitors. Proc Natl Acad Sci USA 74:5463–5467.[Abstract/Free Full Text]

Sigurdsson TJ, Lee MB, Kubota K, Turek TJ, Wozney JM, Wikesjö UM (1995). Periodontal repair in dogs: recombinant human bone morphogenetic protein-2 significantly enhances periodontal regeneration. J Periodontol 66:131–138.[ISI][Medline]

Skalli O, Ropraz P, Trzeciak A, Benzonana G, Gillessen D, Gabbiani G (1986). A monoclonal antibody against -smooth muscle actin: a new probe for smooth muscle differentiation. J Cell Biol 103:2787–2796.[Abstract/Free Full Text]

Takayama S, Murakami S, Shimabukuro Y, Kitamura M, Okada H (2001). Periodontal regeneration by FGF-2 (bFGF) in primate models. J Dent Res 80:2075–2079.[Abstract/Free Full Text]

Wikesjö UM, Razi SS, Sigurdsson TJ, Tatakis DN, Lee MB, Ongpipattanakul B, et al. (1998). Periodontal repair in dogs: effect of recombinant human transforming growth factor-ß1 on guided tissue regeneration. J Clin Periodontol 25:475–481.[ISI][Medline]

This Article Abstract Figures Only Full Text (PDF) Appendix Services Similar articles in this journal Similar articles in ISI Web of Science Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Citing Articles Citing Articles via Google Scholar Google Scholar Articles by Ohira, T. Articles by Takashiba, S. Search for Related Content PubMed PubMed Citation Articles by Ohira, T. Articles by Takashiba, S.