J Periodontal Implant Sci.  2017 Apr;47(2):116-131. 10.5051/jpis.2017.47.2.116.

Substrate roughness induces the development of defective E-cadherin junctions in human gingival keratinocytes

• Jin C
• Lee G
• Oh C
• Kim HJ
• Kim HM

• KMID: 2377746
• DOI: http://doi.org/10.5051/jpis.2017.47.2.116

Abstract

PURPOSE
The entry of bacteria or harmful substances through the epithelial seal of human gingival keratinocytes (HGKs) in the junctional epithelium (JE) is blocked by specialized intercellular junctions such as E-cadherin junctions (ECJs). However, the influence of roughened substrates, which may occur due to apical migration of the JE, root planing, or peri-implantitis, on the development of the ECJs of HGKs remains largely unknown.
METHODS
HGKs were cultured on substrates with varying levels of roughness, which were prepared by rubbing hydrophobic polystyrene dishes with silicon carbide papers. The activity of c-Jun N-terminal kinase (JNK) was inhibited with SP600125 or by transfection with JNK short hairpin RNA. The development of intercellular junctions was analyzed using scanning electron microscopy or confocal laser scanning microscopy after immunohistochemical staining of the cells for E-cadherin. The expression level of phospho-JNK was assessed by immunoblotting.
RESULTS
HGKs developed tight intercellular junctions devoid of wide intercellular gaps on smooth substrates and on rough substrates with low-nanometer dimensions (average roughness [Ra]=121.3±13.4 nm), although the ECJs of HGKs on rough substrates with low-nanometer dimensions developed later than those of HGKs on smooth substrates. In contrast, HGKs developed short intercellular junctions with wide intercellular gaps on rough substrates with mid- or high-nanometer dimensions (Ra=505.3±115.3 nm, 867.0±168.6 nm). Notably, the stability of the ECJs was low on the rough substrates, as demonstrated by the rapid destruction of the cell junction following calcium depletion. Inhibition of JNK activity promoted ECJ development in HGKs. JNK was closely associated with cortical actin in the regulation of ECJs in HGKs.
CONCLUSIONS
These results indicate that on rough substrates with nanometer dimensions, the ECJs of HGKs develop slowly or defectively, and that this effect can be reversed by inhibiting JNK.

Keyword

Actins; Cadherins; Dental implants; Keratinocytes; Re-epithelialization; Periodontal diseases

MeSH Terms

Actins
Bacteria
Cadherins*
Calcium
Dental Implants
Epithelial Attachment
Humans*
Immunoblotting
Intercellular Junctions
JNK Mitogen-Activated Protein Kinases
Keratinocytes*
Microscopy, Confocal
Microscopy, Electron, Scanning
Peri-Implantitis
Periodontal Diseases
Polystyrenes
Re-Epithelialization
RNA, Small Interfering
Root Planing
Silicon
Transfection
Actins
Cadherins
Calcium
Dental Implants
JNK Mitogen-Activated Protein Kinases
Polystyrenes
RNA, Small Interfering
Silicon

Figure

• 

  Figure 1 Model substrates. Substrates prepared in polystyrene dishes with varying levels of roughness were analyzed using atomic force microscopy. SV and LF of the substrates with varying levels of roughness.SV: surface views, LF: line profiles, S: smooth culture dish, R(4000): prepared with #4000 sandpaper, R(1200): prepared with #1200 sandpaper, R(200): prepared with #200 sandpaper.

• 

  Figure 2 Surface structures of HGKs attached to smooth substrates or substrates of low-nanometer dimensions (Ra=121.3±13.4 nm). Cells were processed for observation by FE-SEM 10 hours after cell seeding. (A) Membrane folding and furrows are rich on the surface of the underside of the cell bodies on the rough substrate (bar=5 µm). (B) Membrane folding and furrows gradually fade toward the end of the cell process, which is extended from the cell body on the rough substrate (bar=10 µm). (C) The cells on the smooth substrate display circumferential lamellipodia rich in filopodia (bar=10 µm). (D) Several cell processes extend from the cell body on the rough substrate. Except for the end of the cell processes, the margin of the cell membrane is linear without filopodia. The surface of the cell body is rich in membrane folding and furrows (bar=10 µm).HGK: human gingival keratinocyte, Ra: average roughness, FE-SEM: field emission scanning electron microscopy.

• 

  Figure 3 Development of the ECJs of HGKs depends on the substrate roughness in vitro, as shown by confocal laser-scanning microscopy. ECJ development was followed by immunocytochemical staining for the expression level of E-cadherin (red) at 5, 10, 20, and 40 hours after cell seeding. F-actin (green) was stained with FITC-phalloidin. Nuclei were stained with DAPI (blue). The lower right picture in each set of 4 pictures is a merged image of the E-cadherin, F-actin, and nuclei images. Refer to the results section for details regarding ECJ development (short bar=20 µm, long bar=160 µm).ECJ: E-cadherin junction, HGK: human gingival keratinocyte, FITC: fluorescein isothiocyanate-labeled, DAPI: 4′, 6-diamidino-2-phenylindole dihydrochloride, S: smooth culture dish, R(4000): prepared with #4000 sandpaper, R(1200): prepared with #1200 sandpaper, R(200): prepared with #200 sandpaper.

• 

  Figure 4 Development of the ECJs of HGKs depended on the substrate roughness in vitro, as shown by confocal laser-scanning microscopy. (A-D) F-actin (green) in HOK-16B cells cultured on the substrates with varying levels of roughness for 5 hours was stained with FITC-phalloidin to identify cortical actin (arrow heads) or lamellipodia (arrows). (E-H) E-cadherin (red) in HOK-16B cells cultured on substrates with varying levels of roughness for 40 hours was immunohistochemically stained to examine how ECJ development depended on the roughness of the substrates. (I-L) E-cadherin (red) of HOK-16B cells cultured on substrates with varying levels of roughness for 4 days were immunohistochemically stained to examine how ECJ development depended on the roughness of the substrates. Intercellular gaps (*) are clearly present between cells in (J) and (K). Refer to the results section for details regarding ECJ development (short bar=20 µm). Ra=121.3±13.4, 505.3±115.3, and 867.0±168.6 nm for R(4000), R(1200), and R(200), respectively.ECJ: E-cadherin junction, HGK: human gingival keratinocyte, FITC: fluorescein isothiocyanate-labeled, Ra: average roughness, R(4000): prepared with #4000 sandpaper, R(1200): prepared with #1200 sandpaper, R(200): prepared with #200 sandpaper, S: smooth substrate.

• 

  Figure 5 ECJ development in HGKs seeded at a high density on smooth or rough substrates was examined using confocal laser-scanning microscopy. ECJ development was followed by immunocytochemical staining for the expression level of E-cadherin (red) at 5, 10, and 20 hours after cell seeding. F-actin (green) was stained with FITC-phalloidin. Nuclei were stained with DAPI (blue). The lower right picture in each set of 4 pictures is a merged image of the E-cadherin, F-actin, and nuclei images. (A, D, G) Cells on the smooth substrates were cultured under normal conditions. Cells on the rough substrates with high-nanometer dimensions, R(200), were cultured under normal conditions (B, E, H) or JNK inhibition by 1 µM SP (C, F, I). (H) Arrows indicate the wide intercellular gaps remaining between the cells on the rough substrate with a high-nanometer dimension despite the high density of the culture (bar=20 µm). Ra=867.0±168.6 nm for R(200).ECJ: E-cadherin junction, HGK: human gingival keratinocyte, FITC: fluorescein isothiocyanate-labeled, DAPI: 4′, 6-diamidino-2-phenylindole dihydrochloride, JNK: c-Jun N-terminal kinase, S: smooth substrate, SP: SP600125, Ra: average roughness, R(200): prepared with #200 sandpaper.

• 

  Figure 6 Dissociation of ECJs after calcium depletion. HGKs were treated for the indicated times with 0.25% trypsin-EDTA after culturing for 4 days on the smooth substrate (A-C) or the rough substrate with low-nanometer dimensions (D-F). ECJ development was followed by immunocytochemical staining for the expression level of E-cadherin (red) at 0, 15, and 30 minutes after calcium depletion. F-actin (green) was stained with FITC-phalloidin. Nuclei were stained with DAPI (blue). The lower right picture in each set of 4 pictures is a merged image of the E-cadherin, F-actin, and nuclei images (bar=20 µm).ECJ: E-cadherin junction, HGK: human gingival keratinocyte, FITC: fluorescein isothiocyanate-labeled, DAPI: 4′, 6-diamidino-2-phenylindole dihydrochloride, S: smooth substrate, R(4000): prepared with #4000 sandpaper.

• 

  Figure 7 Influence of JNK on ECJ development. (A) HGKs were cultured on substrates with varying levels of roughness. Then, the expression levels of p-JNK were compared by western blotting. Values under the blots for p-JNK indicate the relative expression level of p-JNK to the total JNK for each group, which was analyzed using Image J. (B) HGKs were treated with SP (1 µM) to inhibit JNK activity for 24 hours in culture starting 12 hours after cell seeding on the smooth and rough substrates. CNT indicates the control culture without treatments with SP. (C) HGKs transfected using lentiviruses expressing shJNK1/2 to selectively inhibit JNK1/2 activity were re-plated for 24 hours on the rough substrate with low-nanometer dimensions (Ra=121.3±13.4 nm). ECJ development was followed by immunocytochemical staining for the expression level of E-cadherin (red). F-actin (green) was stained with FITC-phalloidin. Nuclei were stained with DAPI (blue). The lower right picture in each set of 4 pictures is a merged image of the E-cadherin, F-actin, and nuclei images. shCNT, scrambled shRNA. (D) HGKs were treated with anisomycin (20 ng/mL) to activate JNK for 20 hours in culture starting 3 days after cell seeding on the rough substrates. CNT indicates the control culture without treatments with anisomycin (bar=20 µm). Ra=121.3±13.4, 505.3±115.3, and 867.0±168.6 nm for R(4000), R(1200), and R(200), respectively.S: smooth substrate, SP: SP600125, JNK: c-Jun N-terminal kinase, ECJ: E-cadherin junction, HGK: human gingival keratinocyte, CNT: carbon nanotube, Ra: average roughness, p-JNK: phospho-c-Jun N-terminal kinase, shRNA: small hairpin RNA, FITC: fluorescein isothiocyanate-labeled, DAPI: 4′, 6-diamidino-2-phenylindole dihydrochloride, R(4000): prepared with #4000 sandpaper, R(1200): prepared with #1200 sandpaper, R(200): prepared with #200 sandpaper.

• 

  Figure 8 Influence of JNK on the organization of F-actin during ECJ development. (A-C) The development of cortical actin was assessed using CLSM after treating the cells with SP (1 µM) (B), or anisomycin (20 ng/mL) (C) for 20 hours from the start of cell culture on the smooth substrate. CNT (A) indicates the control culture without treatments regulating JNK activity. F-actin was stained with FITC-phalloidin (bar=20 µm). (D-F) Cortical actin development was assessed using CLSM after HGKs infected with lentiviruses expressing shJNK1 (E) or shJNK2 (F) to selectively inhibit JNK1 or JNK2 were cultured for 8 hours on the rough substrate with low-nanometer dimensions (Ra=121.3±13.4 nm). Control cells (shCNT) (D) were cultured after transfection with scrambled shRNA. F-actin was stained with FITC-phalloidin (bar=20 µm). (G-I) Cortical actin development and ECJ development (arrows) were assessed using CLSM after treating the cells with anisomycin (20 ng/mL) (G) or co-treating the cells with anisomycin (20 ng/mL) and Y-27632 (20 µM) (H, I) for 20 hours from the start of cell culture on the smooth substrate. F-actin (green) was stained with FITC-phalloidin. ECJ development was followed by immunocytochemical staining for the expression level of E-cadherin (red) (bar=20 µm).JNK: c-Jun N-terminal kinase, ECJ: E-cadherin junction, CNT: carbon nanotube, Ra: average roughness, SP: SP600125, CLSM: confocal laser-scanning microscopy, FITC: fluorescein isothiocyanate-labeled, sh: small hairpin.

Reference

1. Bosshardt DD, Lang NP. The junctional epithelium: from health to disease. J Dent Res. 2005; 84:9–20. PMID: 15615869.

2. Hara AT, Livengood SV, Lippert F, Eckert GJ, Ungar PS. Dental surface texture characterization based on erosive tooth wear processes. J Dent Res. 2016; 95:537–542. PMID: 26848070.

3. Merijohn GK. Management and prevention of gingival recession. Periodontol 2000. 2016; 71:228–242. PMID: 27045439.

4. Solís Moreno C, Santos A, Nart J, Levi P, Velásquez A, Sanz Moliner J. Evaluation of root surface microtopography following the use of four instrumentation systems by confocal microscopy and scanning electron microscopy: an in vitro study. J Periodontal Res. 2012; 47:608–615. PMID: 22494068.

5. Amid R, Kadkhodazadeh M, Fekrazad R, Hajizadeh F, Ghafoori A. Comparison of the effect of hand instruments, an ultrasonic scaler, and an erbium-doped yttrium aluminium garnet laser on root surface roughness of teeth with periodontitis: a profilometer study. J Periodontal Implant Sci. 2013; 43:101–105. PMID: 23678394.

6. Santos FA, Pochapski MT, Leal PC, Gimenes-Sakima PP, Marcantonio E Jr. Comparative study on the effect of ultrasonic instruments on the root surface in vivo . Clin Oral Investig. 2008; 12:143–150.

7. Lee HJ, Lee J, Lee JT, Hong JS, Lim BS, Park HJ, et al. Microgrooves on titanium surface affect peri-implant cell adhesion and soft tissue sealing; an in vitro and in vivo study. J Periodontal Implant Sci. 2015; 45:120–126. PMID: 26131372.

8. Lee HJ, Yang IH, Kim SK, Yeo IS, Kwon TK. In vivo comparison between the effects of chemically modified hydrophilic and anodically oxidized titanium surfaces on initial bone healing. J Periodontal Implant Sci. 2015; 45:94–100. PMID: 26131369.

9. Lee G, Kim HJ, Kim HM. RhoA-JNK regulates the E-cadherin junctions of human gingival epithelial cells. J Dent Res. 2016; 95:284–291. PMID: 26635280.

10. Zheng L, Kim HM. Low-Rac1 activity downregulates MC3T3-E1 osteoblastic cell motility on a nanoscale topography prepared on polystyrene substrates in vitro . J Biomed Mater Res A. 2013; 101:1629–1636. PMID: 23184573.

11. Torres-Gallegos I, Zavala-Alonso V, Patiño-Marín N, Martinez-Castañon GA, Anusavice K, Loyola-Rodríguez JP. Enamel roughness and depth profile after phosphoric acid etching of healthy and fluorotic enamel. Aust Dent J. 2012; 57:151–156. PMID: 22624754.

12. Barkmeier WW, Erickson RL, Kimmes NS, Latta MA, Wilwerding TM. Effect of enamel etching time on roughness and bond strength. Oper Dent. 2009; 34:217–222. PMID: 19363978.

13. Singh S, Uppoor A, Nayak D. A comparative evaluation of the efficacy of manual, magnetostrictive and piezoelectric ultrasonic instruments--an in vitro profilometric and SEM study. J Appl Oral Sci. 2012; 20:21–26. PMID: 22437673.

14. Vastardis S, Yukna RA, Rice DA, Mercante D. Root surface removal and resultant surface texture with diamond-coated ultrasonic inserts: an in vitro and SEM study. J Clin Periodontol. 2005; 32:467–473. PMID: 15842261.

15. Park NH, Min BM, Li SL, Huang MZ, Cherick HM, Doniger J. Immortalization of normal human oral keratinocytes with type 16 human papillomavirus. Carcinogenesis. 1991; 12:1627–1631. PMID: 1654226.

16. Samarin SN, Ivanov AI, Flatau G, Parkos CA, Nusrat A. Rho/Rho-associated kinase-II signaling mediates disassembly of epithelial apical junctions. Mol Biol Cell. 2007; 18:3429–3439. PMID: 17596509.

17. Lecuit T. “Developmental mechanics”: cellular patterns controlled by adhesion, cortical tension and cell division. HFSP J. 2008; 2:72–78. PMID: 19404474.

18. Maghzal N, Kayali HA, Rohani N, Kajava AV, Fagotto F. EpCAM controls actomyosin contractility and cell adhesion by direct inhibition of PKC. Dev Cell. 2013; 27:263–277. PMID: 24183651.

19. Sahai E, Marshall CJ. ROCK and Dia have opposing effects on adherens junctions downstream of Rho. Nat Cell Biol. 2002; 4:408–415. PMID: 11992112.

20. Huang C, Rajfur Z, Borchers C, Schaller MD, Jacobson K. JNK phosphorylates paxillin and regulates cell migration. Nature. 2003; 424:219–223. PMID: 12853963.

21. Otto IM, Raabe T, Rennefahrt UE, Bork P, Rapp UR, Kerkhoff E. The p150-Spir protein provides a link between c-Jun N-terminal kinase function and actin reorganization. Curr Biol. 2000; 10:345–348. PMID: 10744979.

22. Samak G, Gangwar R, Crosby LM, Desai LP, Wilhelm K, Waters CM, et al. Cyclic stretch disrupts apical junctional complexes in Caco-2 cell monolayers by a JNK-2-, c-Src-, and MLCK-dependent mechanism. Am J Physiol Gastrointest Liver Physiol. 2014; 306:G947–G958. PMID: 24722904.

23. Hamel M, Kanyi D, Cipolle MD, Lowe-Krentz L. Active stress kinases in proliferating endothelial cells associated with cytoskeletal structures. Endothelium. 2006; 13:157–170. PMID: 16840172.

24. Mengistu M, Brotzman H, Ghadiali S, Lowe-Krentz L. Fluid shear stress-induced JNK activity leads to actin remodeling for cell alignment. J Cell Physiol. 2011; 226:110–121. PMID: 20626006.