and phenytoin- induced gingival overgrowth

Archives of Oral Biology 66 (2016) 38–43

Contents lists available at ScienceDirect

Archives of Oral Biology journal homepage: www.elsevier.com/locate/aob

Biological roles of KGF, CTGF and TGF-b in cyclosporine-A- and phenytoin- induced gingival overgrowth: A comparative experimental animal study Nadia Saeed Al-hamillya , Lobna R.S. Radwanb , Mohamed Abdul-rahmanb , Mohamed I. Mouradc, Mohammed E. Grawishb,d,* a

Ministry of Health, Faculty of Dentistry, Tripoli University, Libya Oral Biology, Faculty of Dentistry, Mansoura University, Egypt c Oral Pathology, Faculty of Dentistry, Mansoura University, Egypt d Oral Biology, Faculty of Oral and Dental Medicine, Delta University for Science and Technology, Gamasa, Mansoura, Egypt b

A R T I C L E I N F O

A B S T R A C T

Article history: Received 6 June 2015 Received in revised form 3 February 2016 Accepted 4 February 2016

Objective: To identify the possible biological roles of keratinocyte growth factor (KGF), connective tissue growth factor (CTGF) and transforming growth factor-b (TGF-b) in cyclosporine-A (CsA) and phenytoin (PNT)-induced gingival overgrowth (GO) and to correlate them with each other. Methods: Sixty adult male albino rats were selected and divided into 3 equal groups. Group I rats received no treatment. Group II rats were administrated CsA for 7 weeks. Group III were administrated PNT for the same period. Rats were euthanized at the end of the experiment and routine tissue processing was carried out. The obtained specimens were stained with H&E, KGF, CTGF and TGF-b antibodies. Results: One-way MANOVA test for KGF, CTGF and TGF-b revealed an overall significant difference between the different groups (P < 0.001). LSD post hoc test for multiple comparisons revealed a significant difference between each two groups. Two-tailed Pearson correlation for group II revealed non-significant weak positive correlations between KGF & CTGF and between CTGF & TGF-b. Nonsignificant weak negative correlation was found between KGF & TGF-b. Meanwhile, group III revealed non-significant weak positive correlation between KGF & TGF-b and between CTGF & TGF-b. Significant moderate positive correlation was found between KGF & CTGF. Conclusion: The findings of the present study indicated that KGF, CTGF and TGF-b have biological roles in progression of CsA- and PNT- induced GO. KGF plays a greater role in CsA- induced GO than in PNTinduced GO. Meanwhile, CTGF and TGF-b play a role in PNT- induced GO greater than in CsA- induced GO. ã 2016 Elsevier Ltd. All rights reserved.

Keywords: Drug induced gingival overgrowth Cyclosporine-A Phenytoin Keratinocyte growth factor Connective tissue growth factor Transforming growth factor-beta

1. Introduction Drug induced gingival overgrowth (DIGO) is known as an adverse side effect commonly related to three major classes of drugs (1) anticonvulsants phenytoin (PNT), (2) an immunosuppressive cyclosporine A (CsA), and (3) antihypertensive calcium channel blockers (Subramani, Rathnavelu, Yeap, & Alitheen, 2013; Trackman & Kantarci, 2015). The pathogenesis of CsA- and PNTinduced gingival overgrowth (GO) remains unclear and contradictory (Clementini et al., 2008). The imbalance between the metabolism of the extracellular matrix (ECM) synthesis and

* Corresponding author at: Oral Biology Department, Faculty of Dentistry, Mansoura University, PO Box 35516, Egypt. E-mail address: [email protected] (M.E. Grawish). http://dx.doi.org/10.1016/j.archoralbio.2016.02.006 0003-9969/ ã 2016 Elsevier Ltd. All rights reserved.

degradation has been suggested as a mechanism for accumulation of excessive connective tissue components within the gingival tissue. Kato et al. (2005) suggested that such imbalance leads to decreased collagen degradation, not to an increase on its synthesis. In contrary, Chung et al. (2015) found that these drugs enhanced type 1 collagen formation, and alpha-smooth muscle actin expression in human gingival fibroblasts. The epithelial–mesenchymal interactions play an important role in tissue biology. The mutual interactions between them control tissue homeostasis through gradients of morphogens and differential sensitivity of target cells for these diffusible mediators (Nieto & Cano, 2012) To test this interaction, in vivo testing using animal models is often needed because it is better suited for observing the overall effects of an experiment and offer conclusive insights about the nature of medicine and disease (Dahiya & Ogden, 2010).

N.S. Al-hamilly et al. / Archives of Oral Biology 66 (2016) 38–43

Growth factors and cytokines play an important role in regulation of the gingival ECM turnover (Ma et al., 2014). Knowing the degree of correlation between these diffusible mediators is required to understand the overall pathogenesis of GO and could change our ideas about treatment modalities. Keratinocyte growth factor (KGF), a member of the fibroblast growth factor family, is a paracrine-acting epithelial mitogen produced by cells of mesenchymal origin (Andreadis, Hamoen, Yarmush, & Morgan, 2001). Gingival epithelial keratinocytes and mesenchymal fibroblasts interact with each other to dynamically regulate KGF gene expression, which may have an effect on the gingival condition following treatment with nifedipine (Di et al., 2013). Connective-tissue growth factor (CTGF) is a member of the CCN family secreted proteins implicated in multiple cellular events. The name of this family is derived from the first three family members identified: cysteine-rich 61, CTGF, and nephroblastoma overexpressed proteins. CTGF/CCN2 is expressed both in the connective tissue stroma and in gingival epithelial cells in vivo in fibrotic tissues (Kantarci et al., 2006). CTGF stimulates fibroblasts to produce ECM constituents, so its expression correlates positively with the degree of gingival fibrosis (Heng, Huang, & Trackman, 2006). CTGF have been reported to function as downstream mediator of the some of the effects of TGF-b (Sakuta et al., 2001). TGF-b is another protein involved in multiple cellular responses including cell proliferation, differentiation, senescence, and apoptosis. TGF-b expressed by most cells, including gingival inflammatory cells, endothelial cells and fibroblasts. It is chemotactic for fibroblasts and selectively stimulates the production of some ECM components and being implicated in several fibrotic diseases (Prime, Pring, Davies, & Paterson, 2004). In this context, this study was carried out to identify and outline the possible biological roles of KGF, CTGF and TGF-b in CsA- and PNT- induced GO and to correlate them with each other in an animal model. The research null hypothesis was that no relationship between the growth factors “TGF-b, CTGF, KGF” and GO-induced by these medications “CsA, PNT”. 2. Materials and methods 2.1. Animals’ preparation, medicine supplementation and euthanization Sixty adult male albino rats, weighing 150–200 g were selected, housed and cared according to the guidelines of the Medical Experimental Research Center, Faculty of Medicine, Mansoura University. They received commercial soft diet and water. The rats were divided randomly into 3 equal groups as follows. Group I rats received no treatment and considered as negative control. Group II rats were administrated CsA (30 mg\kg, Sandimmune, Novartis

39

Pharmaceuticals Co., Hanover, Germany) in 16.5 cm3 mineral oil via gastric feeding daily for 7 weeks (Paik et al., 2004). Group III rats were administered PNT (100 mg, El-Nasr Pharmaceuticals Chemical Co, A.R.E) after trituration with 0.5% tween 80 solution (0.5 ml tween 80 dissolved in 99.5 ml distilled water, El-Nasr Pharmaceuticals Chemical Co., A.R.E) (Tamamori, Tamura, Yamazaki, & Ohya, 2005). They received PNT via subcutaneous injection twice daily for 7 weeks. All experimental procedures were performed under the approved protocol of the ethical committee of the Faculty of Dentistry, Mansoura University, Egypt. At the end of the experiment, all the rats were euthanized with an overdose of halothane and the lower jaw was dislocated by pulling it down. A spoon excavator was used to reflect the marginal gingiva and then it was continually resected from the right and left sides of the mandible opposite molar region using a scalpel. 2.2. Histological examination The soft tissue specimens were fixed immediately in 10% formaldehyde in phosphate-buffered saline (PBS) for 24 h. The obtained specimens were sliced into 4 mm sections and stained with H&E stain, KGF (1:250 dilution, nuclear; AbD SeroTec Ltd., Kidlington, Oxford, UK), CTGF (1:250, cytoplasmic, Bioss Antibody, wo-burn, MA, USA) and TGF-b (1:50 cytoplasmic, AbDSeroTec, kidlington, Oxford UK) immunohistochemical stains. 2.3. Evaluation of immunostain Slides were photographed using Olympus1 digital camera installed on Olympus1 microscope with 1/2 photo adaptor, using X objective. The resulted images were analyzed on Intel1 Core I31 based computer using Video Test Morphology1 software (SaintPetersburg, Russia) with a specific built-in automated object counting and stain quantification routines. Immunoreactivity of the cells was scored as the follow: (0) = 100% of cells were negative, mild (1) = (>0–10– 50%) of the cells were positive. 2.4. Statistical analysis The data were tabulated, coded, and then analyzed using Statistical Package for the Social Sciences (SPSS version 17.0). For the analytical statistics, significant differences were tested using multivariate analyses of variance to compare between more than two groups for the numerical parametric data, followed by posthoc LSD for multiple comparisons. Moreover, bivariate Pearson correlation was used to test the association between two contentious numerical variables. The statistical tests were based

Fig. 1. H&E stained sections of group I showing keratinized stratified squamous epithelium (Epi) covering a core of connective tissue (Ct) that exhibits scattered blood vessels (Bv) and few randomly distributed chronic inflammatory cells (Ic) (A). Group II showing hyperplasia of the spinous cell layer (Sp) and hyperorthokeratosis (Ke) with collagen bundles (C) run in a wavy pattern (B). Group III showing mild acanthotic epithelium (Epi) with narrow and slightly elongated ret processes (Rp) with highly cellular connective tissue (Ct) that contains numerously distributed vascular channels (Bv) and inflammatory cells (Ic) (C) (H&EX100).

40


Fig. 2. KGF immunostained sections of group I showing mild level of immunoreactivity with the reaction mostly limited to the basal layer (asterisks) (A), high moderate immunoreactivity throughout the epithelium (star) with inconspicuous reaction among the basal cells (asterisks) in group II (B), low moderate immunoreactivity with the reaction mostly appears at the basal cell layer (star) in group III (C) (ABCX400).

on a type 1 error value of 5% (a = 0.05) and on a power of 0.85 sample size. 3. Results 3.1. Hematoxylin and eosin The specimens of group I exhibited a keratinized stratified squamous epithelium covering a core of connective tissue. The epithelium revealed normal arrangement of its layers. Each layer could be distinguished from the others; basal cells were cuboidal with dark stained basophilic nuclei. They were arranged in a single layer perpendicular to the basement membrane. The spinous cell layer was about 3–4 rows. Its cells were spherical in shape with rounded faintly stained basophilic nuclei. The granular cells appeared flat and the keratin was well adhered to it. The basement membrane was flat with no extended rete processes. The underlying connective tissue showed delicate collagen fibers and bundles, fibroblasts, scattered blood vessels and few randomly distributed chronic inflammatory cells (Fig. 1A). The epithelium of group II rats revealed variable microscopical features. Majority of rats exhibited extensively hyperkeratotic acanthotic epithelium. The epithelial layers exhibited several cytological changes. The cells of the basal layer were crowded. The size of spinous cell layer appeared to be larger than normal. Also, the number of rows of the spinous cells was more than those of the control group. Moreover, some spinous cells exhibited perinuclear cytoplasmic vacuoles while others revealed pyknotic nuclei. The suprabasal layers showed mitotic activity. The epithelium was hyperplastic with bulbous, broad rete processes and limited surface hyperkeratosis. Few rats of this group (5%) showed thin and slender rete processes with abundant hyperkeratosis. The hyperkeratosis was hyperorthokeratosis at all specimens of this group. The underlying connective tissue was hypovascular with dense and coarse collagen bundles running in tangentially wavy pattern.

Fibroblasts were plump and chronic inflammatory cells were more frequent than control group. Blood vessels were small slit like channels (Fig. 1B). The epithelium of group III rats revealed slight hyperplasia as compared to that of the control group, but it appeared lesser than that of group II. Rete processes were slightly elongated as compared to the control group. Also, they were narrower and less extensive than those of group II. The hyperkeratosis was detected, but it was lesser than that of group II. All of the specimens of this group revealed hyperorthokeratosis. The epithelial layers were less organized than control group. Basal cell layer was crowded while prickle cells showed mild acanthosis and vacuoles. The inflammatory cells were distributed throughout the connective tissue. The cases with highly collagenized stroma revealed hypovascularity and absence of inflammatory cells while those with cellular stroma showed numerous vascular channels (Fig. 1C). 3.2. KGF, CTGF and TGF-b immunohistochemical findings Regarding KGF, it was expressed with mild levels (7.20 0.25) in group I and the reaction was limited to the basal layer (Fig. 2A). However, group II had a high moderate epithelial reaction (40.08 0.52) throughout its whole thickness but immunoreactivity of basal cell layer was inconspicuous (Fig. 2B). Group III exhibited low moderate reaction (31.93 0.73) throughout the epithelial layers. Most of the reaction was detected at the basal layer (Fig. 2C). However, CTGF was mildly expressed with very low level within connective tissue (10.28 0.34) while epithelium was negative in group I (Fig. 3A). The fibrous tissue cores of groups II exhibited moderate reaction (37.19 0.49) while epithelium showed mild expression. The reaction within epithelium of group II had no definite pattern (Fig. 3B). Group III revealed fibrous tissue intense reaction (73.33 0.39), while epithelium had mild expression.

Fig. 3. CTGF immunostained section of group I showing very mild connective tissue immunoreactivity (star) and negative epithelium immunoreactivity (asterisks) (A), moderate reaction mostly in the connective tissue (star) and mild epithelium immunoreactivity with no definite pattern (asterisks) in group II (B), intense reaction mostly in the connective tissue (star) and mild epithelium immunoreactivity mostly appears at the basal cells with no definite pattern (asterisks) in group III (C) (ABCX400).


41

Fig. 4. TGF-b immunostained sections of group I showing mild expression at epithelium (asterisks) and connective tissue (star) (A), moderate immune expression at basal and suprabasal layers of the epithelium (asterisks) and connective tissue (star) in group II (B), intense immune reaction in both epithelium (asterisks) and connective tissue (star) in group III (C) (ABCX400).

Unlike group II, most epithelial reaction was at basal cell layer in group III (Fig. 3C). In addition, TGF-b immunostain of group I showed mild expression (3.35 0.26) (Fig. 4A). Moderate reaction for this protein was detected at group II (30.21 0.46). This reaction was seen within the epithelium and underlying connective tissue (Fig. 4B). Within epithelium, most of the reaction was at the basal and suprabasal layers. The highest immune reaction was seen in group III (44.42 0.33) (Fig. 4C). One-way MANOVA test for KGF, CTGF and TGF-b revealed an overall significant difference between the different groups (P < 0.001). In addition, the LSD post hoc test for multiple comparisons revealed a significant difference between each two groups (Fig. 5 & Table 1). Two-tailed Pearson correlation revealed non-significant weak positive correlation between TGF-b & CTGF at the three studied groups (P = 0.43, P = 0.94, P = 0.99; respectively). Also, there was non-significant weak positive correlation between KGF and CTGF in group II (P = 0.38). In addition, group III had significant moderate positive correlation between KGF & CTGF (P = 0.03) and non-significant weak positive correlation between KGF & TGF-b (r = 0.001, P = 0.99). Meanwhile, non-significant negative weak correlation was found between KGF & CTGF (P = 0.19) in group I and between KGF & TGF-b in groups I and II (P = 0.18, P = 0.17; respectively) (Table 2).

4. Discussion It is well established that gingival overgrowth can be due to three causes: chronic inflammatory hyperplasia; noninflammatory, hyperplastic reaction to the medication; or a combined enlargement due to drug-induced hyperplasia and chronic inflammation. Drug-induced gingival enlargement can be minimized, but not prevented, by elimination of local irritants, meticulous oral hygiene, and regular periodontal recall (Hall, 1997). In addition, Quinchia-Rios et al. (2008) found that human gingival fibroblasts exhibit proliferative responses following epidermal growth factor exposure, which are thought to enhance periodontal regeneration in the absence of bacterial products such as lipopolysacharide. In the current work, inflammation wasn’t considered because all the studied groups had nearly the same inflammatory responses due to the similar experimental conditions of all rats. Also, CsA and PNT always induce GO in all experimental rats but in human patients, there is an incidence for the induction of GO due the different and variable factors such as dose, duration, age . . . etc. This observation is supported by Afonso, Bello Vde, Shibli, and Sposto (2003) who indicated that GO induced by CsA may vary according to the individual sensitivity of each patient and may or may not be correlated with other local factors. Moreover, Afonso et al. (2014) suggested that the expression of vascular endothelial

Fig. 5. Bar chart for KGF, CTGF and TGF-b (%) immunoreactivity. The symbols y, z, and # represent significant difference between different groups for KGF, CTG F and TGF-b, respectively. P < 0.05 represents level of significance.

42


Table 1 One way-MANOVA and LSD post hoc test for KGF, CTGF and TGF-b (%) for all groups. One way-MANOVA

LSD post hoc

Mean SD

Groups

Groups

KGF

CTGF

TGF-b

Control CsA PNT F ratio P1 Value

7.20 0.25 40.08 0.52 31.93 0.73 19773.23 0.001*

10.28 0.34 37.19 0.49 73.33 0.39 114126.30 0.001*

3.35 0.26 30.21 0.46 44.42 0.33 66563.96 0.001*

Control CsA Control PNT CsA PNT

KGF

CTGF

TGF-b

P2 value

P2 value

P2 value

0.001** 0.001** 0.001**

0.001** 0.001** 0.001**

0.001** 0.001** 0.001**

P < 0.05. * One way-MANOVA. ** LSD.

Table 2 Bivariate Pearson correlations (r) for KGF, CTGF and TGF-b for all groups. Groups

CTGF (I)

TGF-b (I)

KGF (I)

0.30 0.19

0.30 0.18

KGF (II) KGF (III) TGF-b (I)

CTGF (II)

TGF-b (II)

0.20 0.38

0.31 0.17

r P value

CTGF (III)

TGF-b (III)

0.48 0.03

0.001 0.99

0.18 0.43

TGF-b (II) TGF-b (III)

growth factor, nitric oxide synthase 1 and 3, and Ki-67 have a role in the inflammation processes associated with immunosuppressive therapy. In the present study, the epithelium of group II had moderate reaction to KGF throughout its whole thickness. This finding is consistent with Hyland, McKeown, Mackenzie, and Irwin (2004), Das, Newman, and Olsen (2002) and Das, Parkar, and Olsen (2001). They stated that KGF has a potent mitogenic effect on epithelial cells, and, therefore, could be involved in the pathogenesis of GO. Moreover, they suggested that KGF-receptor pathway plays an important role in the molecular pathogenesis of gingival hyperplasia. In current study, the fibrous tissue cores of group III revealed an intense reaction to CTGF. This is consistent with the findings of Heng et al. (2006) who attributed fibrosis in PNT- induced GO to the CTGF. They reported that PNT-induced lesions are the most fibrotic, and CsA- induced lesions are not fibrotic. In addition, Uzel et al. (2001) found that CTGF is highly expressed in PNT- induced GO, whereas it is not expressed in CsA- induced overgrowth. Moderate reaction for TGF-b was detected within the epithelium and underlying connective tissue in group II of the present study. This is in agreement with the findings of Chung et al. (2015) who reported that CsA enhanced collagen type 1 (COL1), a-smooth muscle actin (a-SMA), sonic hedgehog (Shh) and TGF-b expressions in human gingival fibroblasts. The highest immune reaction to TGF-b among the studied groups was seen in group III. This coincides with Saito, Mori, Iwakura, and Sakamoto (1996) who found significant immunostaining against TGF-b, basic fibroblast growth factor (bFGF) and heparan sulphate glycosaminoglycan (HSGAG) in DIGO. The findings of their study suggested that the increased synthesis of TGF-b, bFGF and HSGAG may be related to the pathogenesis of drug-induced gingival hyperplasia. The Immunohistochemical findings of the present study suggested that intensification of KGF, CTGF and TGF-b signaling

0.01 0.94 0.00

is one of the underlying causes of CsA- and PNT- induced GO lesions. Therefore, this unwanted side effect might be prevented through inhibition of the conjugation between these proteins and their receptors. This deduction is supported by Ko et al. (2003) who found that the down-regulation of KGF-1 expression in oral fibroblast cell lines potentially impairs the proliferation of overlying keratinocytes, which could partially explain the frequent epithelial atrophy observed in chronic areca chewers in vivo. Moreover, Hattori, Matsunaga, Nakazono, and Wang (2006) demonstrated that Saiko contained in Saireito, a Chinese herbal medicine, inhibited the nifedipine-induced proliferation of gingival fibroblasts by reducing the release of FGF and induced inhibition of nifedipine-stimulated proliferation and collagen synthesis in gingival fibroblasts. In addition, Condé, Bastos, Vieira, and Aarestrup (2009) stated that down-regulation of TGFb2 expression may be an important mechanism by which roxithromycin inhibits GO. In addition, Guo, Carter, and Leask (2011) found that controlling TGF-b activity may be useful in controlling responses to mechanical strain in the gingiva and may be of value in controlling fibroproliferative conditions such as gingival hyperplasia. 5. Conclusion The Immunohistochemical findings of the present study suggested that intensification of KGF, CTGF and TGF-b signaling is one of the underlying causes of CsA- and PNT- induced GO lesions. Therefore, this unwanted side effect might be prevented through inhibition of the conjugation between these proteins and their receptors. KGF plays a greater role in CsA- induced GO than in PNT- induced GO. Meanwhile, CTGF and TGF-b play a role in PNTinduced GO greater than in CsA- induced GO. Significant positive correlation was found only between KGF & CTGF in group administered PNT.


Conflict of interest All authors deny any conflict of interest. Funding Not financially supported. References Afonso, M., Bello Vde, O., Shibli, J. A., & Sposto, M. R. (2003). Cyclosporin A-induced gingival overgrowth in renal transplant patients. Journal of Periodontology, 74 (1), 51–56. Afonso, M., Perrotti, V., Rapani, M., Iaculli, F., Piccirilli, M., Onuma, T., et al. (2014). Immunoexpression of angiogenesis and proliferation markers in patients treated with cyclosporin A. Minerva Stomatologica, 63(3), 59–67. Andreadis, S. T., Hamoen, K. E., Yarmush, M. L., & Morgan, J. R. (2001). Keratinocyte growth factor induces hyperproliferation and delays differentiation in a skin equivalent model system. FASEB Journal, 15(6), 898–906. Chung, Y., Fu, E., Chin, Y. T., Tu, H. P., Chiu, H. C., Shen, E. C., et al. (2015). Role of Shh and TGF in cyclosporine-enhanced expression of collagen and a-SMA by gingival fibroblast. Journal of Clinical Periodontology, 42(1), 29–36. Clementini, M., Vittorini, G., Crea, A., Gualano, M. R., Macrì, L. A., Deli, G., et al. (2008). Efficacy of AZM therapy in patients with gingival overgrowth induced by Cyclosporine A: a systematic review. BioMed Cental Oral Health, 8, 34. Condé, S. A., Bastos, M. G., Vieira, B. J., & Aarestrup, F. M. (2009). Down-regulation of transforming growth factor beta-2 expression is associated with the reduction of cyclosporin induced gingival overgrowth in rats treated with roxithromycin: an experimental study. BioMed Cental Oral Health, 9, 33. Dahiya, P., & Ogden, B. E. (2010). Animal ethics in SIRS research. Frontiers in Bioscience (Scholar Edition), 2, 5–10. Das, S. J., Newman, H. N., & Olsen, I. (2002). Keratinocyte growth factor receptor is up-regulated in cyclosporin A-induced gingival hyperplasia. Journal of Dental Research, 81(10), 683–687. Das, S. J., Parkar, M. H., & Olsen, I. (2001). Upregulation of keratinocyte growth factor in cyclosporin A-induced gingival overgrowth. Journal of Periodontology, 72(9), 745–752. Di, C. P., Sun, Y., Zhao, L., Li, L., Ding, C., Xu, Y., et al. (2013). Effect of nifedipine on the expression of keratinocyte growth factor and its receptor in cocultured/ monocultured fibroblasts and keratinocytes. Journal of Periodontal Research, 48 (6), 740–747. Guo, F., Carter, D. E., & Leask, A. (2011). Mechanical tension increases CCN2/CTGF expression and proliferation in gingival fibroblasts via a TGFb-dependent mechanism. Public Library of Science One, 6(5), e19756. Hall, E. E. (1997). Prevention and treatment considerations in patients with druginduced gingival enlargement. Current Opinion in Periodontology, 4, 59–63. Hattori, T., Matsunaga, S., Nakazono, Y., & Wang, P. L. (2006). Inhibition of nifedipine-induced proliferation of cultured human gingival fibroblasts by Saiko, a Chinese herbal medicine. Phytotherapy Research, 20(8), 704–707.

43

Heng, E. C., Huang, Y., Black, S. A. Jr., & Trackman, P. C. (2006). CCN2, connective tissue growth factor, stimulates collagen deposition by gingival fibroblasts via module 3 and alpha6- and beta1 integrins. Journal of Cellular Biochemistry, 98(2), 409–420. Hyland, P. L., McKeown, S. T., Mackenzie, I. C., & Irwin, C. R. (2004). Regulation of keratinocyte growth factor and scatter factor in cyclosporin-induced gingival overgrowth. Journal of Oral Pathology and Medicine, 33(7), 391–397. Kantarci, A., Black, S. A., Xydas, C. E., Murawel, P., Uchida, Y., Yucekal-Tuncer, B., et al. (2006). Epithelial and connective tissue cell CTGF/CCN2 expression in gingival fibrosis. The Journal of Pathology, 210(1), 59–66. Kato, T., Okahashi, N., Kawai, S., Kato, T., Inaba, H., Morisaki, I., et al. (2005). Impaired degradation of matrix collagen in human gingival fibroblasts by the antiepileptic drug phenytoin. Journal of Periodontology, 76(6), 941–950. Ko, S. Y., Lin, S. C., Chang, K. W., Liu, C. J., Chang, S. S., Lu, S. Y., et al. (2003). Modulation of KGF-1 gene expression in oral fibroblasts by ripe areca nut extract. Journal of Oral Pathology and Medicine, 32(7), 399–407. Ma, S., Liu, P., Li, Y., Hou, L., Chen, L., & Qin, C. (2014). Cyclosporine A inhibits apoptosis of rat gingival epithelium. Journal of Periodontology, 85(8), 1126–1134. Nieto, M. A., & Cano, A. (2012). The epithelial-mesenchymal transition under control: global programs to regulate epithelial plasticity. Seminars in Cancer Biology, 22(5–6), 361–368. Paik, J. W., Kim, C. S., Cho, K. S., Chai, J. K., Kim, C. K., & Choi, S. H. (2004). Inhibition of cyclosporin A-induced gingival overgrowth by azithromycin through phagocytosis: an in vivo and in vitro study. Journal of Periodontology, 75(3), 380– 387. Prime, S. S., Pring, M., Davies, M., & Paterson, I. C. (2004). TGF-beta signal transduction in oro-facial health and non-malignant disease (part I). Crticial Reviews in Oral Biology & Medicine, 15(6), 324–336. Quinchia-Rios, B. H., Guerrero, M., Abozeid, S., Bainbridge, B., Darveau, R., Compton, T., et al. (2008). Down-regulation of epidermal growth factor receptordependent signaling by Porphyromonas gingivalis lipopolysaccharide in lifeexpanded human gingival fibroblasts. Journal of Periodontal Research, 43(3), 290–304. Saito, K., Mori, S., Iwakura, M., & Sakamoto, S. (1996). Immunohistochemical localization of transforming growth factor beta, basic fibroblast growth factor and heparan sulphate glycosaminoglycan in gingival hyperplasia induced by nifedipine and phenytoin. Journal of Periodontal Research, 31(8), 545–555. Sakuta, H., Suzuki, R., Takahashi, H., Kato, A., Shintani, T., Iemura, Si., et al. (2001). Ventroptin: a BMP-4 antagonist expressed in a double-gradient pattern in the retina. Science, 293(5527), 111–115. Subramani, T., Rathnavelu, V., Yeap, S. K., & Alitheen, N. B. (2013). Influence of mast cells in drug-induced gingival overgrowth. Mediators of Inflammation, 2013, 275172. Tamamori, Y., Tamura, Y., Yamazaki, T., & Ohya, K. (2005). Establishment of rat model of drug-induced gingival overgrowth induced by continuous administration of phenytoin. Journal of Pharmacological Sciences, 98(3), 290–297. Trackman, P. C., & Kantarci, A. (2015). Molecular and clinical aspects of drug-induced gingival overgrowth. Journal of Dental Research, 94(4), 540–546. Uzel, M. I., Kantarci, A., Hong, H. H., Uygur, C., Sheff, M. C., Firatli, E., et al. (2001). Connective tissue growth factor in phenytoin-induced gingival overgrowth. Journal of Periodontology, 72(7), 921–931.