|
Sign In to gain access to subscriptions and/or personal tools.
|
Gene Profiling in Human Periodontal Ligament Fibroblasts by Subtractive Hybridization
T. Yamamoto,
F. Myokai,
F. Nishimura,
T. Ohira,
N. Shiomi,
K. Yamashiro,
H. Arai,
Y. Murayama and
S. Takashiba*
Department of Periodontal Science, Okayama University Graduate School of Medicine and Dentistry, 2-5-1 Shikata-cho, Okayama 700-8525, Japan;
Correspondence: * corresponding author, stakashi{at}cc.okayama-u.ac.jp
 |
ABSTRACT
|
|---|
Genes expressed by human periodontal ligament fibroblasts (HPFs) are likely to be associated with specific functions of the ligament. The aim of this study is to profile genes expressed highly by HPFs. A library (6 x 103 pfu) was constructed, followed by subtraction of HPF cDNAs with human gingival fibroblast (HGF) cDNAs. Reverse-dot hybridization revealed that 33 clones expressed higher levels of specific mRNAs in HPFs than in HGFs. These were mRNAs for known genes, including several associated with maturation and differentiation of cells. None had been reported in PFs. One clone, PDL-29, identified as a COX assembly factor, showed much stronger mRNA expression in HPFs than in HGFs in culture. In rat periodontium, however, PDL-29 mRNA expression was similar in PFs and GFs. These results suggest that HPFs express many previously unreported genes associated with maturation and differentiation, but expression can differ in vitro and in vivo.
Key Words: gene profiling • periodontal ligament fibroblasts • subtractive hybridization • in situ hybridization
|
INTRODUCTION
|
|---|
Periodontal ligament fibroblasts (PFs) are thought to play important roles not only in the remodeling of the ligament itself but also in the regeneration and wound healing of periodontal tissues, because they produce many factors related to mineralization, a variety of inflammatory mediators, and several growth factors (Lekic and McCulloch, 1996). Moreover, the periodontal ligament may contain undifferentiated mesenchymal cells or progenitor cells important for periodontal regeneration (Lekic et al., 2001). Thus, PFs may express the genes highly to exert specific functions during regeneration.
Despite the identical genomic DNA in the periodontal ligament and gingival fibroblasts, the phenotypic differences between them are obvious. The gene expression profile of each cell type has not yet been elucidated, although each cell expresses a select set of characteristic genes. A typical eukaryotic cell line contains approximately 250,000 cytoplasmic mRNAs and as many as 10,000 different kinds of mRNAs. However, there is only a 2% difference in the mRNAs present in the two closely related lineages (Hedrick et al., 1984). We speculated that human periodontal ligament fibroblasts (HPFs) and human gingival fibroblasts (HGFs) represent a difference in the gene expression profile, because they share common embryonic development from cells of the cranial neural crest (Palmer and Lubbock, 1995).
In this study, we aimed to profile the genes highly expressed by HPFs. We constructed a library, followed by subtraction of HPF cDNA from HGF cDNA. The clones isolated from the library were examined for mRNA expression levels for both PFs and GFs in vitro and in vivo. Furthermore, they were analyzed for sequence homology to known genes.
|
MATERIALS & METHODS
|
|---|
Cell Culture and Synthesis of cDNA
Periodontal ligament was obtained from a periodontally healthy and non-carious human tooth extracted from a 21-year-old female for orthodontic reasons, and attached gingival tissue was obtained from the neighboring tooth socket. The subject gave informed consent to the protocol, which was approved by the Departmental Review Board. HPFs and HGFs were maintained and expanded as described previously (Ye et al., 1995). For all experiments, we used cells grown to a monolayer at the fifth or seventh passage.
The mRNA (5 µg) isolated from the fifth-passage cells was reverse-transcribed to first-strand cDNA by means of SuperScript II (GIBCO BRL, Grand Island, NY, USA) and oligo (dT)12–18 primer (GIBCO) at 45°C, and then blunt-ended second-strand cDNA was synthesized from the cDNA as described previously (Gubler and Hoffman, 1983).
To confirm the differences in cell types, we examined osteocalcin mRNA expression by polymerase chain-reaction (PCR) as a marker of osteoblastic characteristics (Ducy et al., 2000).
Construction of a Subtractive cDNA Library
Subtractive hybridization was carried out as described previously (Klickstein, 1992), with slight modifications. In brief, the blunt-ended cDNA from HPFs was ligated with an EcoRI adapter (Takara, Kyoto, Japan) to both ends by means of T4 DNA ligase (NEB, Beverly, MA, USA). Complementary DNA less than 400 by was removed by means of a column (Size Sep 400 Spun Columns; Pharmacia Biotech, Uppsala, Sweden). The HGF cDNA was digested with both Alu I and Rsa I (NEB) to make the blunt-ended short fragment. The HPF cDNA (40 ng) with the EcoRI site was subtracted from the blunt-ended HGF cDNA (2 µg) under the following conditions: The cDNA mixture was heated to 99°C for 5 min, cooled on ice immediately, and then kept at 37°C for 24 hrs. The remaining double-stranded cDNA with the EcoRI site was ligated into a ZAP Express Predigested Vector (Stratagene, La Jolla, CA, USA) and packaged with the use of Gigapack III Gold Packaging Extracts (Stratagene) according to the manufacturer’s instructions. The phage library was screened in the presence of 1 M isopropylthio-β-D-galactoside and 250 mg/mL of 5-bromo-4-chloro-3-indoyl-β-D-galactoside (Sigma Chemical, St. Louis, MO, USA). Positive phages were stocked in 1 mL of SM buffer, and some were converted to pBK-CMV phagemid for sequence analysis.
Reverse Dot-blot Hybridization
The cloned DNA was amplified by PCR with 1 µL of phage stock and M13-20 and BK Reverse primers (Stratagene). The PCR product (2 µL) was dot-blotted on a Hybond N+ membrane (Amersham Biosciences, Tokyo, Japan) and incubated with the probe at a concentration of 1 x 106 cpm/mL in the buffer as previously described (Davis et al., 1994) at 42°C for 36 hrs. After the RNAs were recovered from fifth-passage HPFs and HGFs, cDNA pools were synthesized as described above, and then labeled with [ -32P] dCTP (Amersham Biosciences) by means of the Bca Best Labeling Kit (Takara). After a final wash with 4 x SSC containing 0.1% SDS at 42°C, the hybridization signals were visualized in a Bio Imaging Analyzer (BAS 2000; FUJI, Tokyo, Japan). The signal intensity of each DNA sample was quantified with NIH Image Ver. 1.62, and normalized against that of β-actin. After stripping the HPF probe, we performed hybridization with the HGF probe, and then quantified and normalized the signal intensity of each DNA sample as above. Clones were selected for sequence analysis and Northern analysis when signal ratios differed by more than two-fold between the two probes.
DNA Sequencing and Homology Search
The phagemid DNA was purified by means of a Plasmid Mini Kit (Qiagen, Hilden, Germany), and the sequences from both ends were partially determined by the dideoxy sequencing procedure with an Automatic 377 sequencer (Perkin-Elmer, Foster City, CA, USA). The nucleotide sequence homology was analyzed by use of the BLASTN and BLASTX homology programs through GenBank DNA databases (final searches in April, 2002). We performed the functional classification of the genes identified as previously described (Adams et al., 1993), by using the Locus Link program from the National Center for Biotechnology Information.
Northern Blot Analysis
Messenger RNA (1 µg) from the seventh-passage cells was separated on 1.0% agarose gel with formaldehyde and transferred to a Hybond N+ membrane (Amersham Biosciences). The membrane was incubated with the probe at a concentration of 5 x 105 cpm/mL in the hybridization buffer as previously described (Church and Gilbert, 1984) at 65°C for 16 hrs. After a final wash with 2 x SSC containing 0.1% SDS at 65°C, the hybridization signals were visualized and quantified as described above. DNA insert from the phagemid clone was amplified by PCR, and labeled with [-32P] dCTP by the method described above, and then used for hybridization. In addition, we collected mRNA from seventh-passage cells from two different donors and used it for Northern hybridization.
Tissue Preparation and in situ Hybridization
We took periodontal tissues from the maxillary first molars of male Wistar rats (10–12 wks old) under deep anesthesia and fixed them for 3 hrs in a solution containing 4% paraformaldehyde. The experimental protocol was carried out according to the guidelines for animal care of Okayama University Dental School. The tissues were decalcified in a 10% EDTA (pH 7.4) solution at 4°C for 4 wks and then embedded in paraffin according to standard procedures.
In situ hybridization was performed according to the method described by Noji et al.(1989), with minor modifications. Briefly, the tissue sections were deparaffinized in xylene, rehydrated, acetylated, and pre-hybridized in a solution of 2 x SSC, 50% formamide at 50°C. The PstI-HindIII gene fragment (433 bp) of a PDL-29 cDNA was subcloned into pBluescript SK(–) (Stratagene), and then sense and anti-sense riboprobes were synthesized by T3 and T7 RNA polymerases from the construct, according to the instructions supplied with the DIG RNA Labeling kit (Roche Diagnostics, Mannheim, Germany). We applied a 20-µL quantity of hybridization mixture (50% de-ionized formamide, 4 mg/mL BSA, 4 mg/mL tRNA, 10% dextran sulfate, and 2 x SSC) to each slide. The specimens were incubated for 15–17 hrs in a humid chamber at 50°C with the probe, at a concentration of 0.4 ng/µL for hybridization. The sections were washed with a solution of 50% formamide and 2 x SSC at 50°C for 1 hr, treated with 20 mg/L RNase A (Roche Diagnostics) at 37°C for 30 min in NTE buffer (0.5 M NaCl, 10 mM Tris [pH 8], 1 mM EDTA), washed further with 0.1 x SSC at 50°C for 1 hr, and then developed by means of a DIG Nucleic Acid Detection Kit (Roche Diagnostics).
|
RESULTS
|
|---|
Clones Isolated from a Subtractive Library
By reverse-transcription PCR analysis, the difference in osteocalcin mRNA expression between the two cells suggests the isolation of PFs from the periodontal tissues (Fig. 1A ).
View larger version (26K):
[in this window]
[in a new window]
|
Figure 1. Unique gene expression in HPFs and HGFs. (A) Osteocalcin mRNA expression in HPFs and HGFs as revealed by PCR analysis. The arrowhead at the right indicates the position of the amplicon. Complementary DNA (10 ng) from each cell was amplified by PCR with human osteocalcin primers (5'-ATGAGAGCCCTCACACTCCTC-3' and 5'-CAGGGGATCCGGGTAGGGGAC-3'). The amplification was performed for 40 cycles, and the annealing temperature was 55°C. M: 100-bp ladder. (B) Messenger RNA expression of PDL-29 in HPFs and HGFs. Messenger RNA from HPFs and HGFs was hybridized with PDL-29 cDNA probe. For an internal control, the mRNAs were also hybridized with human β-actin probe. The donor is #1 in Fig. 1C . (C) Densitometric analysis of signal intensity. Relative signal intensities for PDL-29 mRNA from three different donors are shown after normalization against the signal for β-actin mRNA.
|
|
The β-actin mRNA was present in 1 of 400 clones (0.25%) from the subtractive library, whereas it was present in 1.4 x 102 of 1.2 x 104 clones (1.1%) from the initial HPF library. These results suggested that our subtraction procedure enriched unique gene products and reduced the abundance of common cDNA in the subtractive library.
Among 6 x 103 plaques from the subtractive library, 400 independent clones containing DNA inserts between 400 bp and 2400 bp were obtained and used for reverse dot-blot hybridization. Thirty-three clones were highly expressed by HPFs, and all of them had similarities with known human genes (Table).
Messenger RNA Expression of the Clones in vitro
By Northern hybridization, we detected mRNA expression of all the clones isolated in both HPFs and HGFs. After normalization of the signal for each clone against that of β-actin, PDL-29 showed a five-fold increased expression in HPFs, and this difference was greatest among the 33 clones (Figs. 1B, 1C). A single 2.2-kb transcript was detected in both HPFs and HGFs, and the size of the transcript was consistent with that of the DNA insert in the clone. In the case of the other two donors. the PDL-29 mRNA level in HPFs was much higher than that in HGFs (Fig. 1C).
Expression of PDL-29 mRNA in vivo
By in situ hybridization on the healthy rat periodontal tissue sections with the anti-sense probe, dense signals indicating PDL-29 mRNAs were detected in both GFs and PFs (Fig. 3). However, little difference in signal intensity was seen between them in the tissue. No signal was detected in the tissue in the case of the sense probe.
View larger version (101K):
[in this window]
[in a new window]
|
Figure 3. Expression of PDL-29 mRNA in rat periodontium. In situ hybridization on periodontal tissue sections from Wistar rat was performed with an anti-sense probe from human PDL-29 cDNA (A,B,C). Dense signals are observed in GFs and PFs. When a sense riboprobe is used, no signal is observed in the tissue (D,E). Neighboring sections were stained with hematoxylin and eosin (F,G). Gingival tissue (B,D,F) and periodontal ligament (C,E,G) correspond to the areas in panel A. Bar equals 300 µm. to, tooth; gt, gingival connective tissue; ab, alveolar bone; pl, periodontal ligament.
|
|
|
DISCUSSION
|
|---|
PFs are thought to be important not only for the remodeling of the ligament itself but also for the regeneration and wound healing of periodontal tissues. However, the gene expression profile of PFs is largely unknown. In this study, 33 clones were isolated from a library constructed by subtraction of HPF from HGF cDNAs, and analyzed for similarities to known genes and mRNA expression in both PFs and GFs.
It is interesting that none of the clones had been reported as a unique gene of PFs. Recently, 15 human periodontal ligament fibroblast-specific cDNAs were isolated by subtractive hybridization between cultured HPFs and HGFs (Park et al., 2001). However, the profile of the clones isolated here is completely different from that of the cDNAs. This may be due to a difference in stringency in the subtractive hybridization. Among the clones isolated, only 3 (PDL-28, -29, -33) have been shown to be expressed in tissues or cell types (http://bodymap.ims.u-tokyo.ac.jp/). Therefore, we classified them according to known functions. The "Energy Metabolism" category contains COX assembly factor (PET112), which is involved in energy metabolism at amino acid residues (Petruzzella et al., 1998). The "Signal Transduction" and "Hormones and Cytokines" categories include D1S155E (Jeffers et al., 1990), NUDC (Miller et al., 1999), RSP-1 (Cutler et al., 1992), TMSB4 (Gondo et al., 1987), and CIN85 (Take et al., 2000). These are related to either an altered proliferative response or the developmental potential of cells. The "Transcription and Translation machinery" category includes ribosomal components and elongation factors, which may be involved in the high turnover and remodeling of matrix proteins. The "Other Metabolism" category includes the gene for annexin A2, which has been identified as a factor enhancing osteoclast formation and bone resorption (Takahashi et al., 1994). The annexin A2 and osteocalcin produced by HPFs may play roles in bone remodeling during periodontal regeneration. These functional annotations suggest that the gene expression profile is associated with a characteristic of HPFs.
From healthy human periodontal tissue, PDL-29, identified as a COX assembly factor, showed much stronger mRNA expression by PFs than that by GFs in vitro. The similarities between PFs and GFs were revealed in the other two donors. However, little difference in PDL-29 mRNA expression was seen between the two cell types in healthy rat periodontal tissue. These results suggest that HPFs express PDL-29 mRNA highly, at least in vitro. The PDL-29 expression by PFs may vary from that by GFs in the magnitude of mRNA in response to some exogenous stimuli such as growth factors or inflammatory mediators in vivo.
The PDL-29 cDNA contained a sequence in the 3'-untranslated region (3'-UTR) just before the poly (A)-tail of PET112 (Fig. 2A). An additional homology search with human genomic BLAST in the NCBI revealed that the insertion sequence was located in exon 15 of the PET112 genome (NT_022851). These findings suggest that PDL-29 is an alternative-splicing variant of the COX assembly factor. Several known cis-acting elements (Ogbourne and Antalis, 1998) are found in the insertion sequence of the cDNA (Fig. 2B). It is well-known that the promoter element in the 3'-UTR of the eukaryote gene—namely, silencers—also controls the localization, stability, and translation of its transcripts (Decker and Parker, 1995). Since the silencers identified here usually destabilize RNA, the protein production of PDL-29 may not parallel its transcription.
View larger version (35K):
[in this window]
[in a new window]
|
Figure 2. Structure of the PDL-29 gene. (A) PDL-29, PET112 cDNA sequences, and their chromosomal localization. Nucleotides 1-1984 of PDL-29 and 28-2011 of PET112 completely match, whereas only PDL-29 contains the insertion sequence just before the poly (A)-tail. Nucleotides 1-28 of PET112 and 1-161 of PDL-29 are localized to exon 1, and nucleotides 1530-2180 of PDL-29 are localized to exon 15. Filled area in PDL-29: insertion sequence (197 bp) just before the poly (A)-tail. Filled area in PET112: 5' region of PET112 different from that of PDL-29. Full-length PDL-29 (2180 bases) was submitted to GenBank (Nov. 1, 1998) and has been assigned accession no. AB019410.
(B) Insertion sequence in the 3'-UTR of PDL-29. Nucleotides 1984-2180 of PDL-29 and the poly (A)-tail are displayed. Horizontal bars indicate putative cis-acting elements of known molecules, and similarities are indicated in percentages. GH, bovine growth hormone; TCR Vβ2.2, human T-cell receptor Vβ2.2; OC, rat osteocalcin; ML-MuMTV, mouse mammarytumor-virus long terminal repeat; GPT, mouse N-acetylglucosamine-l-phosphate transferase; LYZ, chicken lysozyme; and PAI-2, human plasminogen activator inhibitor type-2.
|
|
In conclusion, the 33 known isolated genes have not been reported in PFs, and they include several cDNAs, the annotated functions of which are associated with the maturation and differentiation of cells. PDL-29, COX assembly factor, displayed much stronger mRNA expression in HPFs than that in HGFs in vitro, whereas there was little difference in its mRNA expression between PFs and GFs in rat periodontium. These results suggest that HPFs represent many genes which have not previously been reported, and whose functions are associated with maturation nd differentiation. HPFs may express higher PDL-29 mRNA than HGFs, at least in vitro.
|
ACKNOWLEDGMENTS
|
|---|
This study was supported in part by Grants-in-Aid for Scientific Research from the Ministry of Education, Science, Sports and Culture of Japan (No. 14207082 to F.M.; No. 14370710 to S.T.; No. 13470463 to F.N.), the Ryobi Teien Foundation, the Inamori Foundation, and Tokubetsu Haibun from Okayama University (F.M.). We thank Dr. Charles F. Shuler (Director, Center for Craniofacial Molecular Biology, School of Dentistry, University of Southern California) for helpful discussion.
Received for publication May 31, 2002.
Revision received May 5, 2003.
Accepted for publication May 8, 2003.
|
REFERENCES
|
|---|
- Adams MD, Kerlavage AR, Fields C, Venter JC (1993). 3,400 new expressed sequence tags identify diversity of transcripts in human brain. Nat Genet 4:256–267.[CrossRef][Medline]
[Order article via Infotrieve]
- Church GM, Gilbert W (1984). Genomic sequencing. Proc Natl Acad Sci USA 81:1991–1995.[Abstract/Free Full Text]
- Cutler ML, Bassin RH, Zanoni L, Talbot N (1992). Isolation of rsp-1, a novel cDNA capable of suppressing v-Ras transformation. Mol Cell Biol 12:3750–3756.[Abstract/Free Full Text]
- Davis LG, Kuehl WM, Battey JF, editors (1994). Basic methods in molecular biology, 2nd ed. Norwalk: Appleton & Lange.
- Decker CJ, Parker R (1995). Diversity of cytoplasmic functions for the 3' untranslated region of eukaryotic transcripts. Curr Opin Cell Biol 7:386–392.[CrossRef][Medline]
[Order article via Infotrieve]
- Ducy P, Schinke T, Karsenty G (2000). The osteoblast: a sophisticated fibroblast under central surveillance. Science 289:1501–1504.[Abstract/Free Full Text]
- Gondo H, Kudo J, White JW, Barr C, Selvanayagam P, Saunders GF (1987). Differential expression of the human thymosin-beta 4 gene in lymphocytes, macrophages, and granulocytes. J Immunol 139:3840–3848.[Abstract]
- Gubler U, Hoffma BJ (1983). A simple and very efficient method for generating cDNA libraries. Gene 25:263–269.[Medline]
[Order article via Infotrieve]
- Hedrick SM, Nielsen EA, Kavaler J, Cohen DI, Davis MM (1984). Sequence relationships between putative T-cell receptor polypeptides and immunoglobulins. Nature 308:153–158.[CrossRef][Medline]
[Order article via Infotrieve]
- Jeffers M, Paciucci R, Pellicer A (1990). Characterization of unr; a gene closely linked to N-ras. Nucleic Acids Res 18:4891–4899.[Medline]
[Order article via Infotrieve]
- Klickstein LB (1992). Production of a subtracted cDNA library. In: Current protocols in molecular biology (Suppl 18). Ausubel FM, Brent R, Kingston RE, Moore DD, Seidman JG, Smith JA, et al., editors. New York: John Wiley & Sons Inc., pp. 9–15.
- Lekic P, McCulloch CA (1996). Periodontal ligament cell population: the central role of fibroblasts in creating a unique tissue. Anat Rec 245:327–341.[CrossRef][Medline]
[Order article via Infotrieve]
- Lekic PC, Rajshankar D, Chen H, Tenenbaum H, McCulloch CA (2001). Transplantation of labeled periodontal ligament cells promotes regeneration of alveolar bone. Anat Rec 262:193–202.[Medline]
[Order article via Infotrieve]
- Miller BA, Zhang MY, Gocke CD, De Souza C, Osmani AH, Lynch C, et al. (1999). A homolog of the fungal nuclear migration gene nudC is involved in normal and malignant human hematopoiesis. Exp Hematol 27:742–750.[CrossRef][Medline]
[Order article via Infotrieve]
- Noji S, Yamaai T, Koyama E, Taniguch S (1989). Localization of epidermal growth factor and nerve growth factor mRNAs in mouse submandibular gland by simplified in situ hybridization with riboprobes. Radioisotopes 38:366–372.[Medline]
[Order article via Infotrieve]
- Ogbourne S, Antalis TM (1998). Transcriptional control and the role of silencers in transcriptional regulation in eukaryotes. Biochem J 331:1–14.[Medline]
[Order article via Infotrieve]
- Palmer RM, Lubbock MJ (1995). The soft connective tissues of the gingiva and periodontal ligament: are they unique? Oral Dis 1:230–237.[Medline]
[Order article via Infotrieve]
- Park JC, Kim YB, Kim HJ, Jang HS, Kim HS, Kim BO, et al. (2001). Isolation and characterization of cultured human periodontal ligament fibroblast-specific cDNAs. Biochem Biophys Res Commun 282:1145–1153.[CrossRef][Medline]
[Order article via Infotrieve]
- Petruzzella V, Tiranti V, Fernandez P, Ianna P, Carrozzo R, Zeviani M (1998). Identification and characterization of human cDNAs specific to BCS 1, PET 112, SCO 1, COX 15 and COX 11, five genes involved in the formation and function of the mitochondrial respiratory chain. Genomics 54:494–504.[CrossRef][Medline]
[Order article via Infotrieve]
- Takahashi S, Reddy SV, Chirgwin JM, Devlin R, Haipek C, Anderson J, et al. (1994). Cloning and identification of annexin II as an autocrine/paracrine factor that increases osteoclast formation and bone resorption. J Biol Chem 269:28696–28701.[Abstract/Free Full Text]
- Take H, Watanabe S, Takeda K, Yu ZX, Iwata N, Kajigaya S (2000). Cloning and characterization of a novel adaptor protein, CIN85, that interacts with c-Cbl. Biochem Biophys Res Commun 268:321–328.[CrossRef][Medline]
[Order article via Infotrieve]
- Ye J, Nishimura F, Orman R, Terranova VP (1995). Isolation, purification, and partial characterization of an autocrine periodontal ligament cell chemotactic factor. J Dent Res 74:1303–1309.
Journal of Dental Research, Vol. 82, No. 8,
641-645 (2003)
DOI: 10.1177/154405910308200814
 CiteULike  Complore  Connotea  Del.icio.us  Digg  Reddit  Technorati  Twitter What's this?
|
|
TOP
ABSTRACT
INTRODUCTION