Oral Diseases (2015) 21, 762–769 doi:10.1111/odi.12344 © 2015 John Wiley & Sons A/S. Published by John Wiley & Sons Ltd All rights reserved www.wiley.com

ORIGINAL ARTICLE

Latent transforming growth factor-b binding proteins (LTBP-1 and LTBP-2) and gingiva keratinization M-S Chiang1,2, J-R Yang3, S-C Liao1, C-C Hsu1, C-W Hsu2,4, K Yuan1,2 1

Department of Oral Medicine, National Cheng Kung University Hospital, Tainan; 2Institute of Oral Medicine, College of Medicine, National Cheng Kung University, Tainan; 3Division of Physiology, Livestock Research Institute, Council of Agriculture, Executive Yuan, Tainan; 4Dental Department, Tainan Municipal Hospital, Tainan, Taiwan

OBJECTIVE: Transforming growth factor-beta (TGF-b) proteins are involved in epithelial keratinization. The major function of latent TGF-b binding proteins (LTBPs) is modulating TGF-b activity. However, whether LTBP-1 and LTBP-2 play roles in gingiva keratinization remains unclear. MATERIALS AND METHODS: Human keratinized gingiva and non-keratinized alveolar mucosa were processed for LTBP-1, LTBP-2, cytokeratin-1 (K1), cytokeratin-4 (K4), and TGF-b immunohistochemical (IHC) staining. Porcine heterotopically transplanted connective tissues and newly grown epithelia were harvested for IHC staining. The expression levels of LTBP-1 and LTBP-2 were compared between differentiated and undifferentiated human normal oral keratinocytes (hNOK). The expression of LTBP-1 and LTBP-2 was knocked down in a cell line (OEC-M1) to evaluate the effects on the expression of K1, K4, and involucrin (INV). RESULTS: In human and porcine specimens, LTBP-2 expression patterns distinguished keratinized and nonkeratinized oral epithelia. Western blotting results showed that K1, LTBP-1, and INV proteins were upregulated in differentiated hNOK. In OEC-M1 cells, LTBP-2 knockdown resulted in upregulated the expression of K1 and INV and downregulated the expression of K4. LTBP-1 knockdown resulted in opposite effects. CONCLUSION: The expression patterns of LTBP-2 differ in keratinized gingiva and non-keratinized mucosa. LTBP-1 and LTBP-2 are involved in the keratinization of oral epithelium; however, the underlying mechanism remains to be elucidated. Oral Diseases (2015) 21, 762–769

Correspondence: Kuo Yuan, Institute of Oral Medicine, College of Medicine, National Cheng Kung University, 1 University Road, Tainan 701, Taiwan. Tel: 886-6-235-3535 ext 5370, Fax: 886-6-276-2819, E-mail: [email protected] Received 12 December 2014; revised 31 March 2015; accepted 1 April 2015

Keywords: gingiva; keratinization; latent transforming growth factor-beta binding protein; transforming growth factor-beta; cytokeratin

Introduction Whether keratinized mucosa is essential to maintain the health of a natural tooth and dental implant is controversial (Wennstr€om and Derks, 2012; Gobbato et al, 2013; Lin et al, 2013; Brito et al, 2014). However, it is known that mucogingival grafting can benefit patients with a thin gingiva, high frenulum attachment, or shallow vestibular depth (Kim and Neiva, 2015). Various surgical procedures have been developed to increase the width and thickness of keratinized mucosa surrounding a natural tooth or dental implant (Cortellini and Pini Prato, 2012; Esposito et al, 2012; Kim and Neiva, 2015). However, surgical procedures to harvest autogenous mucosa grafts can cause intraor postoperative complications such as bleeding or paresthesia at the donor site (Griffin et al, 2006). Although ready-to-use xenogenic matrices can reduce surgical time and patient morbidity caused by autogenous graft harvesting (Esposito et al, 2012; Jepsen et al, 2013), they do not increase keratinized mucosa to the same extent as autogenous mucosa grafts do (Aroca et al, 2013). Increased understanding of the molecular mechanisms underlying oral mucosa keratinization might facilitate the development of low-invasive procedures for increasing keratinized mucosae. At a molecular level, keratins are useful markers of epithelial differentiation. The (cyto)keratin family contains ≥ 21 polypeptides in acidic and basic subgroups. Studies have confirmed that keratin 1 (K1) and keratin 10 (K10) are expressed by suprabasal cells in keratinized oral epithelium, whereas keratin 4 (K4) and keratin 13 (K13) are expressed by suprabasal cells in non-keratinized oral epithelium (Sawaf et al, 1991; Pritlove-Carson et al, 1997; Hsieh et al, 2010). It is generally considered that oral connective tissue can determine the differentiation of the overlying epithelia (Karring et al, 1975). In vitro studies

LTBPs and gingiva keratinization

M-S Chiang et al

have shown that several factors can affect the differentiation of keratinocytes (basal cells). For example, transforming growth factor-beta (TGF-b) and calcium ions (Ca2+) regulate the expression of keratinized markers in keratinocytes (Marchese et al, 1990). As TGF-b proteins are involved in epithelial keratinization (Reiss and Sartorelli, 1987) and the major function of latent TGF-b binding protein (LTBP) family is modulating TGF-b bioavailability and function in the extracellular matrix (ECM) (Todorovic and Rifkin, 2012), it is plausible that LTBPs play roles in the keratinization of oral mucosa. LTBP family has four isoforms (LTBP-1, LTBP2, LTBP-3, and LTBP-4). They competitively bind to fibrillins and contribute to the assembly of differing microfibrils, including elastic fibers, to perform various functions (Hirai et al, 2007). Our previous study results indicated that elastin modulates the keratinized status of the oral epithelium (Hsieh et al, 2010). Therefore, we hypothesized that protein complexes consisting of various members of elastin, fibrillins, LTBPs, and TGF-bs might influence the keratinized phenotypes of epithelial tissues. In this study, we focused on evaluating the expression of LTBP-1 and LTBP-2 in keratinized gingiva and nonkeratinized alveolar mucosa, and determining whether these proteins are involved in keratinization of the oral epithelium.

blocked for 20 min, they were incubated with primary antibodies for K1, K4, LTBP-2, LTBP-1, and TGF-b (2 lg/ml) overnight at 4 °C. The primary antibodies for K1 and K4 were purchased from Neomarkers (Labvision, Fremont, CA, USA). The antibodies for LTBP-2, LTBP-1, and TGF-b were purchased from Sigma-Aldrich (St. Louis, MO, USA). We used human and porcine skins as the positive control tissues in IHC and immunoblot assays and confirm the antibodies for LTBP-1 and LTBP-2 have cross-reactivity between human and porcine proteins. The antibody for TGF-b recognizes all the subtypes of TGF-b. On the subsequent day, the slides were incubated with biotin-conjugated secondary antibodies and streptavidin– horseradish peroxidase, according to the manufacturer’s instructions (Biocare Medical, Walnut Creek, CA, USA). Finally, peroxidase activity was detected using an AEC chromogen kit (Vector Laboratories, Burlingame, CA, USA) and the slides were counterstained with Mayer’s hematoxylin. For semiquantitation of the protein staining, two independent observers (MS Chiang & CC Hsu) scored the percent of positively staining cells per highpower field (2009) in the tissue sections. The number of positive cells per high-power field was assessed as 1 + = 0-25% staining positively of cells, 2+ = 25–75% of cells, and 3+ ≥ 75% of cells with positive staining. Three representative sections of each sample were analyzed.

Materials and methods

Porcine tissue recombination assay Our animal protocols were approved by the Institutional Animal Care and Use Committee of Livestock Research Institute (Aug.01/2010; #LRIIACUC99032). To investigate whether the LTBP expression patterns in the epithelial layer are determined by the underlying connective tissue, free grafts of connective tissue from the keratinized gingiva and non-keratinized alveolar mucosa near the upper deciduous canines were reciprocally transplanted onto counterpart recipient beds after deepithelialization in two miniature pigs, aged 4 months. Two surgical sites were created in each pig (one keratinized and one non-keratinized connective tissue grafts for each site). A small piece of tissue at the graft margin was dissected and processed for hematoxylin and eosin (H&E) staining to confirm the absence of residual epithelial cells. The grafts were stabilized using resorbable chromic gut sutures. The graft size was approximately 10 9 5 9 2 mm. After 2 months of healing, the pigs were euthanized. The transplanted connective tissues and their overlying newly grown epithelium were excised for K1, K4, LTBP-1, LTBP-2, and TGF-b IHC analyses. A total of eight specimens were collected.

Oral mucosal sampling and isolation of oral keratinocytes Ten human oral mucosa samples containing keratinized gingiva and non-keratinized alveolar mucosa were collected during periodontal surgery for crown lengthening. Each patient provided signed informed consent, and the study was approved by the institutional review board of the study university (Institutional Review Board, National Cheng Kung University Hospital, Feb. 27, 2013; #A-ER-101354). Five samples were fixed in 4% paraformaldehyde and embedded in paraffin for subsequent immunohistochemical (IHC) analyses. The remaining five samples were used in cell culture experiments. The methods used to isolate normal oral keratinocytes from gingiva have been described previously (Hsieh et al, 2010). All cultures were incubated in a humidified atmosphere of 5% carbon dioxide (CO2) in air at 37°C. The oral keratinocytes were maintained in a keratinocyte serum-free medium (KSFM; Gibco, Carlsbad, CA, USA) supplemented with penicillin (100 U/ml), streptomycin (0.1 mg/ml), and amphotericin B (0.25 lg/ml). Immunohistochemistry Paraffin-embedded samples were cut into 4-lm sections and placed on silane-coated slides, which were deparaffinized and rehydrated with serial xylene and ethyl alcohol. The slides were then incubated for 10 min in 3% hydrogen peroxide in methanol, to quench endogenous peroxidase activity, and rinsed three times in phosphate-buffered saline (PBS) for 5 min each rinse. The K1, K4, LTBP-2, LTBP-1, and TGF-b antigens were retrieved using heat treatment (10 mM citrate buffer, pH 6.0). After the slides were washed with PBS three times (5 min each rinse) and

763

Induction of keratinocyte differentiation Normal human keratinocytes can be selectively cultured and grown optimally in a medium with a calcium concentration < 0.1 mM. The addition of calcium (≤1.2 mM) to attached keratinocytes reduces cell proliferation, commits to complete differentiation, and upregulates the expression of K1, K10, and INV (Yuspa et al, 1989). In this study, human normal oral keratinocytes (hNOK) were maintained in KSFM (Gibco) prior to exposure to a high-calcium (1.2 mM) medium. Cell lysates and protein samples were Oral Diseases

LTBPs and gingiva keratinization

M-S Chiang et al

764

harvested 3 days after the addition of calcium for Western blot analyses. Cells without high-calcium treatment were analyzed as the control cells. Culturing of oral cancer cells and knockdown assays hNOK primary cells are more vulnerable to transfection procedures than cancer cells; therefore, a gingival cancer cell line was selected for knockdown assays. The oral cancer cell line (OEC-M1) was originally derived from an oral epidermoid carcinoma in gingiva. The OEC-M1 cells were maintained in an RPMI 1640 medium supplemented with 10% fetal bovine serum, 100 U/ml penicillin G, 100 lg/ml streptomycin sulfate, and 0.25 lg/ml amphotericin B at pH 7.4. All of the cultures were incubated in a humidified atmosphere of 5% CO2 in air at 37°C. LTBP-1 and LTBP-2 knockdown was performed in OEC-M1 cells using lentiviral transduction to stably express short hairpin RNAs (shRNAs) that targeted LTBP-1 and LTBP-2. A shRNA is a sequence of RNA containing a tight hairpin turn that can be used to silence gene expression through RNA interference. The shRNA-containing clones were purchased from the National RNAi Core Facility (Institute of Molecular Biology/Genomic Research Center, Academia Sinica, Taiwan). The LTBP-1 shRNA construct (TRCN0000053376, containing the shRNA target sequence 50 -CCGTTGAATACC GCCTTGAAT-30 for human LTBP-1), the LTBP-2 shRNA construct (TRCN0000053439, containing the shRNA target sequence 50 - CACATGGACATCTGCTGGAAA-30 for human LTBP-2), and the luciferase shRNA construct (TRC N0000072247, containing the shRNA target sequence 50 -G AATCGTCGTATGCAGTGAAA-30 for a negative control) were used to generate recombinant lentiviral particles. The virus particles were packaged by cotransfecting human embryonic kidney (HEK) 293T cells with PsPAX (Addgene plasmid 12260; Cambridge, MA, USA), pMD2.G (Addgene plasmid 12259), and the modified pLKO plasmids containing shRNA-coding oligonucleotides. The HEK 293T cells were transfected for 24 hours using calcium chloride (CaCl2), and then, a fresh medium was provided. The cell supernatants were harvested 36, 48, 60, and 72 h after transduction and filtered with a 0.45-lm low-protein binding filter. The virus pellets were further concentrated by centrifuging at 20 000 g at 4°C for 2.5 h and resuspended in a fresh medium. The lentivirus was introduced to the

(a)

(d)

Oral Diseases

(b)

(e)

OEC-M1 cells, at an appropriate multiplicity of infection, in a medium supplemented with 8 lg/ml polybrene to generate LTBP-1 and LTBP-2 knockdown (shLTBP-1 and shLTBP2) and negative control (shControl) cells. Western blotting confirmed LTBP-1 and LTBP-2 knockdown efficiency. Western blotting For Western blotting, cell lysates were collected from 10cm culture plates after adding 1 ml of lysis buffer to each plate. Sodium dodecyl sulfate–polyacrylamide gel electrophoresis was performed on 12% acrylamide gels. In each lane, 40 lg of protein from a cell lysate and 10 ll of molecular weight standards were applied to the gel. After electrophoresis, transferring, and blocking, a primary antibody against K1, K4, INV, LTBP-1, or LTBP-2 (0.1 lg/ mL) was incubated with the membrane for 1 h at 37°C. Polyclonal rabbit anti-mouse b-actin (Santa Cruz Biotechnology, Santa Cruz, CA, USA) was used as the internal control to ensure equal loading. After washing, a secondary antibody (0.05 lg/ml) and horseradish peroxidase were incubated with the membrane for 1 h at 37°C. After washing the membrane with PBS/Tween-20, an enhanced chemiluminescence kit (Amersham Bioscience, Piscataway, NJ, USA) was used for development before exposure to X-ray film. The blots were then analyzed densometrically (UVP, Upland, CA, USA) to quantify protein expression.

Results Immunohistochemical analyses of human and porcine oral mucosae H&E staining indicated the absence of inflammation in the human keratinized and non-keratinized mucosae specimens collected during crown lengthening procedures. The keratin expression patterns were consistent among the five patients. K1 was expressed in the suprabasal layer of the entire keratinized gingiva and in a small area of the adjacent alveolar mucosa, whereas K4 was expressed exclusively in the suprabasal layer of the non-keratinized alveolar mucosa (Figure 1). The expression of LTBP-1 and LTBP-2 was substantially lower in the connective tissue than in the oral epithelium. LTBP-1 expression in keratinized gingiva and non-keratinized alveolar mucosa

(c)

(f)

Figure 1 K1 and K4 immunoreactivity in human keratinized gingiva and non-keratinized alveolar mucosa. The photomicrographs show representative sections from five specimens. Scale bar = 100 lm. G denotes keratinized gingiva and AM denotes non-keratinized alveolar mucosa. (b) and (c) show photomicrographs from (a) K1 immunostaining at high magnification; (e) and (f) show photomicrographs from (d) K4 immunostaining at high magnification

LTBPs and gingiva keratinization

M-S Chiang et al

765

Figure 2 LTBP-1, LTBP-2, and TGF-b immunoreactivity in human keratinized gingiva and non-keratinized alveolar mucosa. The photomicrographs show representative sections from five specimens. Scale bar = 100 lm. G denotes keratinized gingiva and AM denotes non-keratinized alveolar mucosa. (b) and (c) show photomicrographs from (a) LTBP-1 immunostaining at high magnification; (e) and (f) show photomicrographs from (d) LTBP-2 immunostaining at high magnification; and (h) and (i) show photomicrographs from (g) TGF-b immunostaining at high magnification

(a)

(b)

(c)

(d)

(e)

(f)

(g)

(h)

(i)

exhibited nonsignificant differences. However, LTBP-1 immunoreactivity was substantially lower in the basal cell layer than in the suprabasal layer. The LTBP-2 expression patterns in keratinized gingiva and non-keratinized alveolar mucosa differed. In keratinized gingiva, LTBP-2 was expressed in all layers of the epithelium, and the basal cell layer exhibited the strongest reactivity. In non-keratinized alveolar mucosa, LTBP-2 expression was predominantly limited to the basal cell layer. We observed faint LTBP-2 immunoreactivity in the suprabasal layers. TGF-b immunoreactivity was detectable in the connective tissue and epithelium. TGF-b expression in the keratinized gingiva and non-keratinized alveolar mucosa exhibited nonsignificant differences (Figure 2). Immunohistochemical analyses of porcine transplanted connective tissue and newly developed epithelium The patterns of the expression of LTBP-1, LTBP-2, K1, K4, and TGF-b in porcine oral mucosae after recombination surgery were similar to those in native human mucosae (K1, K4, and TGF-b immunoreactivity data not shown). We observed LTBP-2 expression from the basal cell layer to the keratinized layer of the newly developed epithelium overlying the transplanted gingiva connective tissue. These patterns of immunoreactivity were similar to those in human native keratinized gingiva. In the newly developed epithelium overlying the transplanted alveolar mucosa, LTBP-2 was expressed predominantly in the basal cell layer. The patterns of LTBP-2 expression were similar to those in native human alveolar mucosa (Figure 3, Figures S1 and S2). These results indicated that the underlying connective tissues determined the phenotypes of the examined molecules in the epithelia. Western blot analyses of oral keratinocytes after differentiation A high Ca2+ concentration is known to inhibit the proliferation, and promote the differentiation, of mammal

keratinocytes (Yuspa et al, 1989). After culturing for 3 days in a high-calcium medium, the expression of INV, K1, and LTBP-1 was higher, and the expression of K4 and LTBP-2 was lower, than expression after culturing in a low-calcium medium. INV and K1 are major differentiation markers for cutaneous and oral keratinocytes (Yuspa et al, 1989; Hayashi et al, 2007). These results indicated the upregulation of LTBP-1, and downregulation of LTBP-2, during oral keratinocyte differentiation (Figure 4). Western blot analyses after knockdown assays in oral cancer cells Western blotting results indicated that the knockdown efficiency of LTBP-1 and LTBP-2 in OEC-M1 cells was 50% and 62%, respectively. The expression of K4 protein was markedly upregulated, whereas the expression of K1 and INV proteins was markedly downregulated, after LTBP-1 knockdown (Figure 5). By contrast, the expression of K1 and INV proteins was upregulated, whereas the expression of K4 protein was downregulated, after LTBP-2 knockdown (Figure 6). The expression of LTBP1 and LTBP-2 remained unchanged when that of their counterpart was knocked down (Figures 5 and 6). Therefore, LTBP-1 and LTBP-2 knockdown produced opposing effects on the expression of K1, K4, and INV in OEC-M1 cells.

Discussion In this study, we identified that LTBP-2 is a novel marker of the keratinized status of the mucosae surrounding teeth. Our results indicated higher expression of LTBP-1 and LTBP-2 in oral epithelia than in connective tissues, and differing LTBP-1 and LTBP-2 expression patterns in oral mucosae, particularly in non-keratinized alveolar mucosa. LTBP-2 expression was substantially higher in the keratinized gingiva than in the non-keratinized alveolar mucosa. Animal studies confirmed that the epithelial expression of Oral Diseases

LTBPs and gingiva keratinization

M-S Chiang et al

766

(a)

(b)

(c)

(d)

Figure 3 LTBP-1 and LTBP-2 immunoreactivity in heterotopically transplanted connective tissue grafts and the newly grown epithelia from porcine recombination studies. The photomicrographs show representative sections from two pigs. The specimens were collected after 2 months of wound healing. Scale bar = 100 lm. KCT denotes that the donor tissue derives from the connective tissue of keratinized gingiva after deepithelialization. NKCT denotes that the donor tissue derives from the connective tissue of non-keratinized alveolar mucosa after deepithelialization

Figure 4 Representative Western blot analyses of differentiation markers and LTBPs in human oral keratinocytes with or without high-calcium treatment. The differentiation of oral keratinocytes was induced using 1.2 mM Ca2+. INV is a marker of keratinocyte differentiation, K1 is a maker of keratinized epithelium, and K4 is a marker of non-keratinized epithelium. b-actin was assessed as an internal control. The experiments were performed in triplicate and repeated twice

Figure 5 Expression of K1, K4, and INV in LTBP-1 knockdown and control OEC-M1 cells, evaluated using Western blotting. LTBP-1 expression was knocked down using a shRNA technique in a gingival carcinoma cell line (OEC-M1). b-actin was used as an internal control. The experiment was performed in duplicate and repeated twice. According to the density measurement in the first row, the knockdown efficiency of LTBP-1 is approximately 50%, while LTBP-2 expression remained unchanged

LTBP-1 and LTBP-2, similar to other differentiation markers, is regulated by the underlying connective tissues. In vitro cell culture assays indicated opposing trends in the expression of LTBP-1 and LTBP-2 during oral keratinocyte differentiation. In a gingiva cancer cell line, LTBP1 upregulated the expression of K1 and INV and downregulated the expression of K4, whereas LTBP-2 had opposite effects on the expression of K1, INV, and K4. According to our research, this study is the first to show an association between the expression of LTBP-1 and LTBP-2 and epithelial cell keratinization. Previous studies predominantly investigated the roles of LTBP-1 and

LTBP-2 in extracellular matrices and connective tissues (Hirai et al, 2007; Todorovic and Rifkin, 2012). Several factors are regulators of the terminal differentiation of keratinocytes, including TGF-b, retinoid acid, and calcium. Previous studies have indicated that a keratinocyte must escape the cell cycle to terminally and irreversibly differentiate (Reiss and Sartorelli, 1987; Yuspa et al, 1989). TGF-b is a pleiotropic cytokine involved in several physiologic and pathologic processes (Reiss and Sartorelli, 1987; Marchese et al, 1990; Chen et al, 2013). Exogenous TGF-b can inhibit keratinocyte cell division and DNA synthesis (Shipley et al, 1986; Kopan et al, 1987). In previous

Oral Diseases

LTBPs and gingiva keratinization

M-S Chiang et al

Figure 6 Western blot analyses of the expression of K1, K4, and INV in LTBP-2 knockdown and control OEC-M1 cells. A shRNA technique was used to knock down LTBP-2 expression in a gingival carcinoma cell line (OEC-M1). b-actin was used as an internal control. The experiment was performed in duplicate and repeated twice. According to the density measurement in the first row, the knockdown efficiency of LTBP-2 is approximately 60%, while LTBP-1 expression remained unchanged

analyses, the combination of a low TGF-b concentration and the presence of epidermal growth factor resulted in markedly increased normal keratinocyte differentiation (Reiss and Sartorelli, 1987). Choi and Fuchs (1990) suggested that the effects of TGF-b and retinoic acid on epidermal growth and differentiation are multifaceted and that the extent to which their actions are coupled in keratinocytes might differ in different conditions and/or species. Furthermore, TGF-b is secreted by cells in a latent form, and must be activated before interaction with TGF-b receptors on the cell surface (Lyons et al, 1988; Miyazono et al, 1988). Thus, in vitro studies might not be able to fully elucidate the functional relevance of TGF-b in various biological situations. These may partly explain why inconsistent results existed between our in vitro and in vivo analyses. LTBP-1, LTBP-2, LTBP-3, and LTBP-4 belong to a family of matrix proteins with crucial and complex functions in the extracellular space. Studies of knockout mice have revealed that LTBP-1 and LTBP-2 deficiencies result in lethal phenotypes (Shipley et al, 2000; Todorovic et al, 2007), whereas LTBP-3 and LTBP-4 deficiencies are nonlethal (Dabovic et al, 2002; Sterner-Kock et al, 2002). LTBPs are key regulators of TGF-b bioavailability and action. They also interact with other components of the ECM, such as fibronectin, fibrillins, and cytokines other than TGF-b, and affect the formation, structure, and activity of the ECM (Todorovic and Rifkin, 2012). The different LTBP members compete for binding to various fibrillin members to orchestrate microfibril composition and function. All LTBPs, with the exception of LTBP-3, bind to fibrillin-1 through their C-terminus. Competitive binding studies have suggested that LTBP-2 might be a negative regulator of LTBP-1 interaction with fibrillin-1 (Hirani et al, 2007). Overactivation of the TGF-b signaling factors, SMADs, was detected in mice and humans with mutations

in fibrillin-1 (Lannoy et al, 2014). Deficiency of LTBP-2 might favor LTBP-1 interactions with fibrillin-1 and therefore reduce the availability of TGF-b. LTBP-1 and LTBP-4 can also bind to fibrillin-2, whereas LTBP-2 cannot. LTBP1 and LTBP-2 interact with fibulin-4 and fibulin-5, respectively, to have different roles in elastogenesis. LTBPs exhibit different affinity for three isoforms of TGF-b latency-associated peptide (LAP). LTBP-1 and LTBP-3 bind all three TGF-b LAP isoforms with high affinity, whereas LTBP-4 shows a weak binding capacity only for TGF-b1 LAP. LTBP-2 does not bind to any form of latent TGF-b (Todorovic and Rifkin, 2012). Few studies have investigated the roles of LTBPs in the epithelium. Limited data indicate that LTBP-1 is expressed by the normal ovary epithelium (Henriksen et al, 1995), embryonic epithelial cells (Gualandris et al, 2000), and normal oral epithelium (Karatsaidis et al, 2003); LTBP-2 is expressed by immortalized nasopharyngeal epithelial cells (Chen et al, 2012) and cutaneous keratinocytes (Langton et al, 2012); and LTBP-4 is expressed by immortalized keratinocytes (Koli et al, 2001). LTBP expression is generally higher in normal epithelial cells than in diseased epithelial cells. It is plausible that LTBPs tightly control the availability and function of TGF-b and maintain the epithelia in a state of homeostasis. Our in vitro assays showed LTBP-1 and LTBP-2 had opposite effects on oral epithelial cell differentiation. IHC results exhibited the same pattern of TGF-b expression in keratinized gingiva and non-keratinized alveolar mucosa. There might be no significant difference in the synthesis and secretion of latent TGF-b between two types of epithelial cells. However, relative amounts of LTBP-1 and LTBP-2 might control the availability of active TGF-b and modulate epithelial keratinizing phenotypes. Our IHC results correspond well with those of Langton et al (2012), who observed that LTBP-2 is expressed specifically in the epidermal basal layer in young skin. In aged epidermis, LTBP-2 is expressed in the basal and suprabasal layers of the epidermis. Our IHC results consistently showed that LTBP-2 is predominantly expressed in the basal layer of non-keratinized alveolar mucosa, and in the basal and suprabasal layers of keratinized gingiva. It is generally agreed that a higher level of differentiation occurs in the epithelium of keratinized gingiva than in non-keratinized alveolar mucosa. Langton et al (2012) suggest that elastic fiber in the connective tissue might function to maintain a low differentiated or young state. Our previous study about oral mucosa (Hsieh et al, 2010) suggested that elastin is essential for maintaining the non-keratinized phenotype of the oral epithelium. Langton et al (2012) observed the significant upregulation of LTBP-2 in the epidermis with increasing age. Young skin has abundant elastic fibers, which increase skin elasticity. However, aged skin loses its elasticity and becomes more lax and wrinkle when the network of elastic fibers disintegrates. It is plausible that more elastic fibers in the non-keratinized oral mucosa or young skin downregulate LTBP-2 expression in the overlying epithelium and maintain a low-keratinized state. However, the results from our in vitro assays did not show agreement with our IHC findings. The data suggested that LTBP-2 upregulated the expression of K4 and downregulated the expression of K1 and INV in a gingiva cancer cell line and that

767

Oral Diseases

LTBPs and gingiva keratinization

M-S Chiang et al

768

LTBP-1 had opposing effects to LTBP-2. Inconsistency between the in vivo and in vitro experimental results might predominantly derive from the absence of other interacting molecules from adjacent ECM and cells in the in vitro system. As previously described, TGF-b activation requires a complicated process (Lyons et al, 1988; Miyazono et al, 1988). Another major reason might be the discrepancy between normal and cancer cells. Nevertheless, the knockdown assays clearly demonstrated that LTBP-1 and LTBP-2 are involved in keratinization of oral epithelial cells. Another limitation of our study is that we did not perform similar analyses on LTBP-3 and LTBP-4 because of the absence of the appropriate LTBP-3 and LTBP-4 antibodies at study initiation. In future studies, we aim to evaluate the roles of LTBP-3, LTBP-4, and other components of microfibrils, such as fibrillins, various subtypes of TGF-b, and fibronectin in keratinization of the oral mucosa. Current clinical dental implant surgery focuses on increasing the width of the keratinized mucosa. Inadequate keratinized mucosa adjacent to the implant increases inflammation and recession of the marginal peri-implant soft tissues (Lin et al, 2013). Thus, it is considered that reconstructing the keratinized mucosa surrounding an implant might help a patient to maintain adequate oral hygiene, facilitate restorative procedures, and improve esthetics. Currently, autogenous soft tissue grafting is the gold standard for increasing keratinized mucosa. The procedures are surgical, require donor and recipient surgical sites, and typically cause considerable patient discomfort (Oates et al, 2003). Increased understanding of the molecular mechanisms underlying the keratinization of oral mucosa can facilitate the development of low-invasive procedures (e.g., delivering key recombinant proteins or siRNAs) to increase keratinized mucosa. In conclusion, LTBP-2 is a novel marker for distinguishing keratinized gingiva from non-keratinized alveolar mucosa in humans and pigs. Its patterns of expression in oral epithelia are determined by the underlying connective tissues. Knockdown experiments confirm that LTBP-1 and LTBP-2 play opposite roles in keratinization of oral epithelial cells. The mechanisms underlying the effects of LTBPs on oral epithelium keratinization might be modulated by other components of the microfibrils and involve the regulation of TGF-b activity. Additional investigations to develop novel strategies for increasing keratinized mucosa are warranted. Acknowledgments This study was supported by grants from the National Science Council (NSC) 99-2314-B-006-035-MY3) and Cheng Kung University Hospital, Taiwan (NCKUH-10104005). The RNAi clones were obtained from the National RNAi Core Facility Platform, located at the Institute of Molecular Biology/Genomic Research Center, Academia Sinica, and supported by the National Core Facility Program for Biotechnology Grants of NSC (NSC 100-2319-B-001-002).

Author contributions M-S Chiang and K Yuan designed the study, analyzed the data, and drafted the manuscript. J-R Yang, S-C Liao, and C-W Hsu Oral Diseases

performed the animal studies. M-S Chiang, S-C Liao, and C-W Hsu helped collect specimens. M-S Chiang and C-C Hsu carried on the in vitro assays. All authors read the manuscript and approved the submitted version.

Conflict of interest The authors declare that they have no conflict of interests.

References Aroca S, Molnar B, Wiindisch P et al (2013). Treatment of multiple adjacent Miller class I and II gingival recessions with a Modified Coronally Advanced Tunnel (MCAT) technique and a collagen matrix or palatal connective tissue graft: a randomized, controlled clinical trial. J Clin Periodontol 40: 713–720. Brito C, Tenenbaum HC, Wong BK, Schmitt C, Nogueira-Filho G (2014). Is keratinized mucosa indispensable to maintain peri-implant health? A systematic review of the literature. J Biomed Mater Res B Appl Biomater 102: 643–650. Chen H, Ko JMY, Wong VCL et al (2012). LTBP-2 confers pleiotropic suppression and promotes dormancy in a growth factor permissive microenvironment in nasopharyngeal carcinoma. Cancer Lett 325: 89–98. Chen YK, Huang AHC, Cheng PH, Yang SH, Lin LM (2013). Overexpression of Smad proteins, especially Smad7, in oral epithelial dysplasias. Clin Oral Invest 17: 921–932. Choi Y, Fuchs E (1990). TGF-beta and retinoic acid: regulators of growth and modifiers of differentiation in human epidermal cells. Cell Regul 1: 791–809. Cortellini P, Pini Prato G (2012). Coronally advanced flap and combination therapy for root coverage. Clinical strategies based on scientific evidence and clinical experience. Periodontol 2000 59: 158–184. Dabovic B, Chen Y, Colarossi C, Zambuto L, Obata H, Rifkin DB (2002). Bone defects in latent TGF-b binding protein (Ltbp)-3 null mice; a role for Ltbp in TGF-b presentation. J Endocrinol 175: 129–141. Esposito M, Maghaireh H, Grusovin MG, Ziounas I, Worthington HV (2012). Soft tissue management for dental implants: what are the most effective techniques? A Cochrane systematic review. Eur J Oral Implantol 5: 221–238. Gobbato L, Avila-Ortiz G, Sohrabi K, Wang CW, Karimbux N (2013). The effect of keratinized mucosa width on peri-implant health: a systematic review. Int J Oral Maxillofac Implants 28: 1536–1545. Griffin TJ, Cheung WS, Zavras AI, Damoulis PD (2006). Postoperative complications following gingival augmentation procedures. J Periodontol 77: 2070–2079. Gualandris A, Annes JP, Arese M, Noguera I, Jurukovski V, Rifkin DB (2000). The latent transforming growth factor-b-binding protein-1 promotes in vitro differentiation of embryonic stem cells into endothelium. Mol Biol Cell 11: 4295–4308. Hayashi N, Kido J, Kido R et al (2007). Regulation of calprotectin expression by interleukin-1a and transforming growth factor-b in human gingival keratinocytes. J Periodont Res 42: 1–7. Henriksen R, Gobl A, Wilander E, Oberg K, Miyazono K, Funa K (1995). Expression and prognostic significance of TGF-beta isotypes, latent TGF-beta 1 binding protein, TGF-beta type I and type II receptors, and endoglin in normal ovary and ovarian neoplasms. Lab Invest 73: 213–220. Hirai M, Horiguchi M, Ohbayashi T, Kita T, Chien KR, Nakamura T (2007). Latent TGF-beta-binding protein 2 binds to DANCE/ fibulin-5 and regulates elastic fiber assembly. EMBO J 26: 3283–3295.

LTBPs and gingiva keratinization

M-S Chiang et al

Hirani R, Hanssen E, Gibson MA (2007). LTBP-2 specifically interacts with the amino-terminal region of fibrillin-1 and competes with LTBP-1 for binding to this microfibrillar protein. Matrix Biol 26: 213–223. Hsieh PC, Jin YT, Chang CW, Huang CC, Liao SC, Yuan K (2010). Elastin in oral connective tissue modulates the keratinization of overlying epithelium. J Clin Periodontol 37: 705–711. Jepsen K, Jepsen S, Zucchelli G et al (2013). Treatment of gingival recession defects with a coronally advanced flap and a xenogeneic collagen matrix: a multicenter randomized clinical trial. J Clin Periodontol 40: 82–89. Karatsaidis A, Schreurs O, Axell T, Helgeland K, Schenck K (2003). Inhibition of the transforming growth factor-b/Smad signaling pathway in the epithelium of oral lichen. J Invest Dermatol 121: 1–8. Karring T, Lang NP, L€ oe H (1975). The role of gingival connective tissue in determining epithelial differentiation. J Periodontal Res 10: 1–11. Kim DM, Neiva R (2015). Periodontal soft tissue non-root coverage procedure: a systematic review from the AAP regeneration workshop. J Periodontol 86(2 Suppl): S56–S72. Koli K, Saharinen J, K€arkk€ainen M, Keski-Oja J (2001). Novel non-TGF-b-binding splice variant of LTBP-4 in human cells and tissues provides means to decrease TGF-b deposition. J Cell Sci 114: 2869–2878. Kopan R, Traska G, Fuchs E (1987). Retinoids as important regulators of terminal differentiation: examining keratin repression in individual epidermal cells at various stages of keratinization. J Cell Biol 105: 427–440. Langton AK, Sherratt MJ, Griffiths CE, Watson RE (2012). Differential expression of elastic fibre components in intrinsically aged skin. Biogerontology 13: 37–48. Lannoy M, Slove S, Jacob MP (2014). The function of elastic fibers in the arteries: beyond elasticity. Pathol Biol (Paris) 62: 79–83. Lin GH, Chan HL, Wang HL (2013). The significance of keratinized mucosa on implant health: a systematic review. J Periodontol 84: 1755–1767. Lyons RM, Keski-Oja J, Moses HL (1988). Proteolytic activation of latent transforming growth factor-beta from fibroblast-conditioned medium. J Cell Biol 106: 1659–1665. Marchese C, Rubin J, Ron D et al (1990). Human keratinocyte growth factor activity on proliferation and differentiation of human keratinocytes: differentiation response distinguishes KGF from EGF family. J Cell Physiol 144: 326–332. Miyazono K, Hellman U, Wernstedt C, Heldin CH (1988). Latent high molecular weight complex of transforming growth factor beta 1. Purification from human platelets and structural characterization. J Biol Chem 263: 6407–6415. Oates TW, Robinson M, Gunsolley JC (2003). Surgical therapies for the treatment of gingival recession. A systematic review. Ann Periodontol 8: 303–320. Pritlove-Carson S, Charlesworth S, Morgan PR, Palmer RM (1997). Cytokeratin phenotypes at the dento-gingival junction in relative health and inflammation, in smokers and non-smokers. Oral Dis 3: 19–24. Reiss M, Sartorelli AC (1987). Regulation of growth and differentiation of human keratinocytes by type beta transforming growth factor and epidermal growth factor. Cancer Res 47: 6705–6709.

Sawaf MH, Ouhayoun JP, Forest N (1991). Cytokeratin profiles in oral epithelia: A review and a new classification. J Biol Buccale 19: 187–198. Shipley GD, Pittelkow MR, Wille JJ, Scott RE, Moses HL (1986). Reversible inhibition of normal prokeratinocyte proliferation by type beta transforming growth factorgrowth inhibitor in serum-free medium. Cancer Res 46: 2068– 2071. Shipley JM, Mecham RP, Maus E et al (2000). Developmental expression of latent transforming growth factor beta binding protein 2 and its requirement early in mouse development. Mol Cell Biol 20: 4879–4887. Sterner-Kock A, Thorey IS, Koli K et al (2002). Disruption of the gene encoding the latent transforming growth factor-b binding protein 4 (LTBP-4) causes abnormal lung development, cardiomyopathy, and colorectal cancer. Genes Dev 16: 2264–2273. Todorovic V, Rifkin DB (2012). LTBPs, more than just an escort service. J Cell Biochem 113: 410–418. Todorovic V, Frendewey D, Gutstein DE et al (2007). Long form of latent TGF-b binding protein 1 (Ltbp1L) is essential for cardiac outflow tract septation and remodeling. Development 134: 3723–3732. Wennstr€ om JL, Derks J (2012). Is there a need for keratinized mucosa around implants to maintain health and tissue stability. Clin Oral Implants Res 23 Suppl 6: 136–146. Yuspa SH, Kilkenny AE, Steinert PM, Roop DR (1989). Expression of murine epidermal differentiation markers is tightly regulated by restricted extracellular calcium concentrations in vitro. J Cell Biol 109: 1207–1217.

769

Supporting Information Additional Supporting Information may be found in the online version of this article: Figure S1 LTBP-1 immunoreactivity in heterotopically transplanted connective tissue grafts and the newly grown epithelia from porcine recombination studies. The photomicrographs show representative sections from four surgical sites (one keratinized and one nonkeratinized connective tissue grafts for each site) of two pigs. Scale bar = 100 lm. KCT denotes that the donor tissue derives from the connective tissue of keratinized gingiva after deepithelialization. NKCT denotes that the donor tissue derives from the connective tissue of nonkeratinized alveolar mucosa after deepithelialization. Figure S2 LTBP-2 immunoreactivity in heterotopically transplanted connective tissue grafts and the newly grown epithelia from porcine recombination studies. The photomicrographs show representative sections from four surgical sites (one keratinized and one nonkeratinized connective tissue grafts for each site) of two pigs. Scale bar = 100 lm. KCT denotes that the donor tissue derives from the connective tissue of keratinized gingiva after deepithelialization. NKCT denotes that the donor tissue derives from the connective tissue of nonkeratinized alveolar mucosa after deepithelialization.

Oral Diseases