FoxP3⁺ T regulatory cells in oral lichen planus and its correlation with the distinct clinical appearance of the lesions

Abstract

The aim of this study was to evaluate the presence of FoxP3⁺ cells in oral lichen planus (OLP) and to correlate the findings with clinical and histopathological features of these lesions. The sample consisted of 32 cases of OLP (17 reticular and 15 erosive cases) and 10 cases of inflammatory fibrous hyperplasia (IFH). Clinical examination, histopathological and histomorphometric analysis, and immunohistochemistry (anti-FoxP3 antibody) were performed. Cells were counted in juxtaepithelial and intraepithelial regions of the lesions, and the results are expressed as the mean and range. Most erosive lesions were keratinized and exhibited epithelial atrophy, whereas most reticular lesions were hyperkeratinized. Mean epithelial thickness and mean density of the inflammatory infiltrate were higher in reticular lesions than in erosive OLP. Juxtaepithelial FoxP3⁺ cells were slightly more frequent in erosive lesions (mean: 1.7 and range: 0–9.4) than in reticular lesions (mean: 1.5 and range: 0–8.3). There was a significant difference in the frequency of these cells between OLP (mean: 1.6 and range: 0–9.4) and IFH (mean: 0.5 and range: 0–1.4) (P < 0.05). The number of intraepithelial FoxP3⁺ cells was higher in reticular OLP and IFH when compared with erosive lesions. The larger number of juxtaepithelial FoxP3⁺ cells in OLP compared to IFH might be related to the distinct etiopathogenesis of these lesions. High disease activity or action of the oral microbiota may explain the slightly higher frequency of FoxP3⁺ cells in erosive lesions.

Introduction

Lichen planus is a relatively common chronic mucocutaneous disease that can affect the skin, oral and genital mucosa, scalp and nails. Although its aetiology remains unknown, an immune-mediated pathogenesis of this disease has been recognized (Ismail et al. 2007). The prevalence of oral lichen planus (OLP) reported in the literature ranges from 0.5 to 2.2%. Patients between the age of 30 and 60 years and females are more frequently affected (van der Waal 2009). In general, OLP clinically manifests in two distinct forms: the reticular form and the erosive form. The reticular form is more frequent and is characterized by the presence of white striations, known as Wickham striae, which are generally surrounded by discretely erythematous borders. This form is usually asymptomatic. The erosive form is symptomatic (pain and/or ardour) and manifests as erythematous areas frequently surrounded by finely radiating striae (Sousa & Rosa 2008). According to Sousa et al. (2009), the main histological characteristics of OLP include liquefaction degeneration of the basal layer of the lining epithelium, intense lymphocyte infiltration in the subepithelial region and a normal keratinization pattern of epithelial cells.

The immunopathogenesis of lichen planus has been studied by various investigators (Lodi et al. 2005; Anuradha et al. 2008; Farhi & Dupin 2010), who suggested that CD8 lesional T cells are activated by an MHC class I-associated antigen present on basal keratinocytes that undergo apoptosis. The nature of this antigen is unknown. This mechanism also involves Langerhans cells and CD4 T cells. Although CD8 cells predominate in the inflammatory infiltrate of OLP (Farhi & Dupin 2010), a subset of T cells, called T regulatory (Treg) cells, is present and might be involved in the pathogenesis of OLP (Tao et al. 2010). These cells play a role in the prevention of uncontrolled immune responses against pathogens or allergens, guarantee stability of the normal microflora and prevent the escape of tumours from immune surveillance. Treg cells are also involved in the development of cancer, autoimmunity, allergy and asthma (Boros & Bromberg 2009; Feuerer et al. 2009; Yan & Liu 2009).

The mechanisms of Treg cell-mediated suppression include inhibition of the function of B cells and self-reactive T cells through direct cell–cell contact and secretion of multiple cytokines (membrane binding and secretion of TGF-β, IL-10 and IL-35); induction of cytolytic molecules in effector cells such as Fas, granzymes and perforins; upregulation of cyclic AMP in target cells through the release of this molecule by Treg cells or increase in adenosine in the extracellular medium mediated by the expression of CD39 and CD73 by Treg cells (Skaggs et al. 2008; Sojka et al. 2008; Workman et al. 2009). The function of Treg cells has been studied in different diseases such as atopic dermatitis (Ito et al. 2009; Hayashida et al. 2011), graft versus host disease (Xie et al. 2009; Koreth et al. 2011), T cell lymphoma (Wada et al. 2010; Peng et al. 2011), myasthenia gravis (Mu et al. 2009; Masuda et al. 2010), B-cell lymphoma (Grille et al. 2010; Feng et al. 2011), atherosclerosis (Es et al. 2010; Zhao et al. 2011), multiple sclerosis (Vandenbark et al. 2009; Libera et al. 2011), systemic lupus erythematosus (Yang et al. 2009; Sobel et al. 2011), rheumatoid arthritis (Chen et al. 2008; Suzuki et al. 2011), cutaneous lichen planus (de Boer et al. 2007a) and OLP (Tao et al. 2010).

The Forkhead box P3 (FoxP3) protein is a transcription factor of the forkhead family, which is required for the development and function of most thymus-derived, naturally occurring Treg cells (Banham et al. 2006). This molecule also plays a crucial role in the regulation of immune responses mediated by peripheral T cells, preventing autoimmunity and permitting the maintenance of Treg cells (Khattri et al. 2003). FoxP3 is the most specific marker of Treg cells currently available (Karube et al. 2004; Wada et al. 2010; Zhao et al. 2011).

Studies investigating the presence and function of FoxP3⁺ Treg cells in OLP and correlating the findings with clinical and histopathological features are scarce. In view of the uncertain aetiology and variable clinical presentation of the disease, studies investigating these differences are of fundamental importance for a more adequate diagnosis and treatment of this condition. The aim of this study was to determine the number of FoxP3⁺ Treg cells in the inflammatory infiltrate of OLP and to correlate the findings with the clinical and histopathological features of reticular and erosive OLP lesions.

Methods

The study was approved by the Ethics Committee of the Federal University of Rio Grande do Norte, Brazil (protocol 253/2009), and was conducted according to the Declaration of Helsinki. Informed written consent was obtained from all subjects.

Thirty-two cases of OLP (17 reticular and 15 erosive cases) and 10 cases of inflammatory fibrous hyperplasia (IFH) seen between 1970 and 2010 were obtained from the archives of the Discipline of Oral Pathology, Federal University of Rio Grande do Norte. The OLP lesions were classified according to the clinical data collected from the records. Asymptomatic lesions with white interlacing lines or appearing as papules were classified as the reticular form, and erythematous, ulcerated and symptomatic lesions were classified as the erosive form (Neville et al. 2009).

The mean patient age was 45.11 years (range: 20–67). Twenty (62.5%) patients were women and 12 (37.5%) were men. Most lesions affected the oral mucosa (n = 18, 58.1%), followed by the tongue and gingiva with 5 cases (16.1%) each.

Inflammatory fibrous hyperplasia was chosen for comparison with OLP lesions in this study because it occasionally exhibits an intense juxtaepithelial inflammatory infiltrate resembling that of OLP. However, these lesions have a distinct etiopathogenesis, with OLP being an immune-mediated disease, whereas IFH is usually caused by chronic and low-intensity mechanical trauma.

Morphological analysis

The specimens selected were cut into 5-μm histological sections, which were mounted on glass slides and stained with haematoxylin and eosin. The slides were examined by two observers together with the clinical data obtained from the patient records. The diagnostic criteria proposed by the WHO (1978) and modified by Meij and Waal (2003) were adopted. Next, the specimens were analysed under a light microscope at ×40, ×100 and ×400 magnifications to define their histopathological characteristics, including epithelial features such as keratinization and epithelial hyperplasia or atrophy.

Histomorphometric evaluation

Histomorphometric analysis was used to determine epithelial thickness and the density of the underlying inflammatory infiltrate. These parameters were measured on images of the histological sections acquired with a Nikon Eclipse E200 microscope (Nikon, Inc., Melville, NY, USA) connected to a Nikon Coolpix 5400 (E5400) camera at ×10 magnification. The measurements were taken with the ImageJ software, version 1.43u (National Institutes of Health, Bethesda, MD, USA). For the determination of epithelial thickness, five different measurements were taken by tracing lines that corresponded to the distance between the outermost part of the epithelial surface and the basal layer (adapted from López-Jornet et al. 2009). For evaluation of the density of the underlying inflammatory infiltrate, five measurements corresponding to the distance between the lower limit of the epithelium and the depth of the inflammatory infiltrate in the submucosa were obtained according to Seoane et al. (2004). Next, the mean and range of the measurements were calculated for each type of lesion.

Immunohistochemistry

For immunohistochemistry, 3-μm-thick histological sections were cut and stained by the streptavidin-biotin method. Antigen retrieval was performed with Tris-EDTA, pH 9.0, in a microwave for 10 min. The specimens were incubated overnight with anti-FoxP3 (clone H190, Santa Cruz Biotechnology, Inc., Santa Cruz, CA, USA), diluted 1:150, as primary antibody.

FoxP3⁺ lymphocytes in OLP and IFH were counted by one observer under a light microscope. Only cells morphologically compatible with lymphocytes were counted in juxtaepithelial and intraepithelial areas in 10 consecutive fields per specimen (×400 magnification). The number of labelled cells was determined as the mean for each case (adapted from Loddenkemper et al. 2009) and is expressed as the mean ± standard deviation for each lesion (adapted from Di Stefano et al. 2009).

Statistical analysis

The results were analysed by descriptive statistics, Fisher's exact test and Student's t-test using the Statistical Package for the Social Sciences, version 17.0 (SPSS, Chicago, IL, USA). A P value < 0.05 was considered to indicate statistical significance.

Results

Histopathological analysis of the OLP cases in haematoxylin/eosin-stained sections normally revealed the presence of a stratified pavement epithelium. The papillary and reticular lamina propria consisted of fibrous connective tissue of variable density. With respect to the presence or absence of epithelial hyperkeratinization, most erosive OLP lesions were parakeratinized/orthokeratinized and most reticular lesions were hyperparakeratinized/hyperorthokeratinized, with this difference being statistically significant (P < 0.05). Most erosive OLP lesions (n = 12; 80.0%) exhibited epithelial atrophy. In contrast, in reticular lesions, hyperplasia was identified in 9 (52.9%) cases and atrophy in 8 (47.0%) (Figure 1a,b and Table 1).

Figure 1.

(a) Erosive oral lichen planus (OLP) showing a parakeratinized surface, epithelial atrophy, lower epithelial thickness and lower density of the inflammatory infiltrate (HE, ×200). (b) Reticular OLP showing hyperkeratinization, hyperplasia, greater epithelial thickness and higher density of the inflammatory infiltrate (HE, ×200). (c) Intraepithelial FoxP3⁺ Treg cells (arrows) in reticular OLP (SABC, ×400). (d) Intraepithelial and juxtaepithelial FoxP3⁺ Treg cells (arrows) in erosive OLP (SABC, ×400). (e) Juxtaepithelial FoxP3⁺ Treg cells (arrows) in reticular OLP (SABC, ×400). (f) Intraepithelial and juxtaepithelial FoxP3⁺ Treg cells (arrows) in inflammatory fibrous hyperplasia (SABC, ×400).

Table 1.

Distribution of cases of erosive and reticular oral lichen planus (OLP) according to the type of keratinization

	Parakeratinized/Orthokeratinized n (%)	Hyperparakeratinized/Hyperorthokeratinized n (%)	P value*	Hyperplasia n (%)	Atrophy n (%)	P value*
Erosive OLP	11 (68.8)	4 (25.0)	0.032	3 (25.0)	12 (60.0)	0.076
Reticular OLP	5 (31.3)	12 (75.0)		9 (75.0)	8 (40.0)
Total	16 (100.0)	16 (100.0)		12 (100.0)	20 (100.0)

^*Fisher exact test.

Epithelial thickness and the density of the underlying inflammatory infiltrate were then determined in the specimens analysed. Mean epithelial thickness was greater in reticular OLP lesions (mean: 192.4 and range: 46.5–480.9 μm) than in erosive lesions (mean: 81.7 and range: 49.7–142.5 μm). In addition, the mean density of the inflammatory infiltrate was also higher in reticular OLP lesions (mean: 216.4 and range: 79.0–366.2 μm) when compared to erosive lesions (mean: 154.6 and range: 63.2–372.7 μm). These differences were significant (P < 0.05) (Figure 1a,b and Table 2).

Table 2.

Epithelial thickness and density of the inflammatory infiltrate in connective tissue of erosive and reticular oral lichen planus (OLP) lesions. Values are reported as the mean (range) and 95% confidence interval

		Confidence interval
		Minimum	Maximum	P value*
Epithelial thickness (μm)
Erosive OLP	81.7 (49.7–142.5)	67.5	96.0	0.004
Reticular OLP	192.4 (46.5–480.9)	123.4	261.4
Density of the inflammatory infiltrate (μm)
Erosive OLP	154.6 (63.2–372.7)	109.3	199.9	0.048
Reticular OLP	216.4 (79.0–366.2)	171.2	261.5

^*Student t-test.

Immunohistochemistry showed a mean number of FoxP3⁺ cells in the juxtaepithelial area of 1.5 (range: 0–8.3) in reticular OLP, of 1.7 (range: 0–9.4) in erosive OLP, and of 0.5 (range: 0–1.4) in IFH (Figure 1c–f). Comparison of the number of FoxP3⁺ cells between reticular OLP and IFH, erosive OLP and IFH, all OLP lesions and IFH, and reticular and erosive OLP lesions only showed a statistically significant difference between all OLP lesions (mean: 1.6 and range: 0–9.4) and IFH (P < 0.05) (Table 3). In the intraepithelial area, the mean number of FoxP3⁺ cells was 0.9 (range: 0–7.5) in reticular OLP, 0.4 (range: 0–1.4) in erosive OLP and 0.6 (range: 0–2.2) in IFH (Figure 1c–f). Comparison between the different types of lesions revealed no significant difference (P > 0.05) (Table 4).

Table 3.

Number of juxtaepithelial FoxP3⁺ cells in erosive and reticular oral lichen planus (OLP) and inflammatory fibrous hyperplasia (IFH) and correlation between the different types of lesions. Values are reported as the mean (range) and 95% confidence interval

			Confidence interval (95%)
	Mean (range)	P value*	Minimum	Maximum
Juxta-epithelial cells
Reticular OLP	1.5 (0–8.3)		0.421	2.667
Erosive OLP	1.7 (0–9.4)		0.304	3.182
IFH	0.5 (0–1.4)		0.214	0.906
Correlations
Reticular OLP × IFH		0.162	−0.4243	2.3918
Erosive OLP × IFH		0.105	−0.2798	2.6455
OLP × IFH		0.019	0.1848	1.9686
Reticular × Erosive OLP		0.814	−1.9192	1.5209

^*Student's t-test.

Table 4.

Number of intraepithelial FoxP3⁺ cells in erosive and reticular oral lichen planus (OLP) and inflammatory fibrous hyperplasia (IFH) and correlation between the different types of lesions. Values are reported as the mean (range) and 95% confidence interval

			Confidence interval (95%)
	Mean (range)	P value*	Minimum	Maximum
Intraepithelial cells
Reticular OLP	0.9 (0–7.5)		0.034	1.941
Erosive OLP	0.4 (0–1.4)		0.133	0.667
IFH	0.6 (0–2.2)		0.163	1.137
Correlations
Reticular OLP × IFH		0.576	−0.8900	1.5650
Erosive OLP × IFH		0.294	−0.7324	0.2324
OLP × IFH		0.889	−0.8468	0.9734
Reticular × Erosive OLP		0.244	−0.4228	1.5978

^*Student's t-test.

Discussion

In this study, reticular OLP lesions were found to be more hyperkeratinized than erosive lesions and most of the cases exhibited epithelial atrophy. Sousa and Rosa (2008) and Neville et al. (2009) also demonstrated the presence of variable degrees of keratinization of the epithelium in OLP depending on the clinical form analysed. Brant et al. (2008) observed a higher degree of epithelial atrophy in erosive OLP than in reticular lesions and both types of lesions were atrophic when compared with healthy mucosa. According to Xia et al. (2006), the red colour of erosive OLP lesions is caused by local inflammation and/or epithelial atrophy.

Mean epithelial thickness was significantly greater in reticular lesions than in erosive OLP. Similar results have been reported by Bagan Sebastian et al. (1991), Karatsaidis et al. (2003) and Brant et al. (2008). In contrast, López-Jornet et al. (2009) found no significant differences in epithelial thickness between the two clinical forms of OLP. These differences between studies might be related to variations in the method used for the measurement of epithelial thickness.

The mean density of the inflammatory infiltrate was higher in reticular OLP than in erosive lesions. In contrast, Bagan Sebastian et al. (1991) and Seoane et al. (2004) found no significant correlation between the clinical form of OLP and whole inflammatory infiltrate density. The authors suggested an association between clinical presentation of the lesion and qualitative features of the inflammatory infiltrate. Thus, a higher lesion activity and occurrence of the atrophic-erosive form do not depend on the quantity of the subepithelial inflammatory infiltrate but rather on its quality.

Immunohistochemical analysis showed that FoxP3⁺ cells were more frequent in OLP lesions than in IFH. This result may be explained by the different aetiology of these lesions: IFH is caused by traumatic injury or infection, whereas OLP is probably due to one or more permanent antigens of uncertain origin that initiate and maintain the inflammatory response (Kubo et al. 2004; Lodi et al. 2005; Scully & Carrozzo 2008; Wing & Sakaguchi 2010).

Our findings also showed a larger number of FoxP3⁺ cells in the juxtaepithelial region of erosive OLP when compared with reticular lesions, but the difference was not significant. This lack of statistical significance might be due to the small size of the sample. These results are different from those reported by Tao et al. (2010) who detected a larger number of FoxP3⁺ Treg cells in reticular lesions. In that study, the number of FoxP3⁺ Treg cells was negatively correlated with disease activity. On the other hand, studies investigating FoxP3⁺ Treg cells in different diseases, including some autoimmune disorders, demonstrated that the higher the disease activity, the larger the number of Treg cells. De Boer et al. (2007b) investigated the frequency of FoxP3⁺ Treg cells in different stages of atherosclerosis lesions using immunohistochemistry. The authors observed that T cells were virtually absent in normal vessel fragments, whereas the number of FoxP3⁺ Treg cells was higher in high-risk lesions compared to stable lesions. Lin et al. (2007) demonstrated a larger number of CD4⁺ FoxP3⁺ Treg cells in peripheral blood of patients with systemic lupus erythematosus when compared with the control group. In addition, the number of these cells was associated with increased disease activity. These studies agree with the present investigation in which higher disease activity was observed for erosive lesions when compared to reticular OLP. This higher disease activity may lead to an increase in the number of Treg cells in an attempt to control the exacerbated immune response. Indeed, it seems that alterations in the number or function of these cells influence the clinical course of OLP.

An increase in the number of Treg cells in clinically active immune-mediated diseases has been reported (Ou et al. 2004; Lee et al. 2007; Bonelli et al. 2008; Yan et al. 2008; Zhang et al. 2008; Ito et al. 2009; Jury et al. 2010). According to Farhi and Dupin (2010), in OLP, keratinocytes or Langerhans cells present an MHC class II-associated antigen to CD4⁺ helper T cells, thus increasing the number of these cells. CD4⁺ helper T cells secrete large amounts of IL-2 and interferon-γ. Increased concentrations of IL-2, in turn, induce the proliferation of Treg cells (Almeida et al. 2006; Miyara & Sakaguchi 2007). Kalogerakou et al. (2008) observed a larger number of IL-2-secreting cells in erosive OLP when compared to reticular lesions. Taken together, these findings suggest that the larger number of Treg cells in erosive lesions seen in the present study may be associated with a higher frequency of activated IL-2-producing cells.

The action of the oral microbiota is another possibility to explain the alterations in the number of Treg cells in erosive lesions. Bornstein et al. (2008) observed higher amounts of bacteria on the surface of reticular OLP lesions when compared to normal oral mucosa specimens. In addition, ulcerated lesions exhibit breakdown of the epithelial barrier, an event facilitating bacterial colonization and increasing inflammation (Greenberg & Pinto 2003). Some studies have demonstrated elevated numbers of Treg cells in bacterial infections (Lundgren et al. 2003; Kandulski et al. 2008; Ertelt et al. 2009). This increase might be due to expansion of natural Treg cells and/or conversion of naive T cells to adaptive Treg cells (Curotto de Lafaille & Lafaille 2009).

In the present study, a smaller number of intraepithelial FoxP3⁺ Treg cells were detected in erosive OLP compared to reticular and IFH lesions. This finding might be related to the epithelial atrophy seen in erosive OLP lesions, which is greater than in reticular lesions (Brant et al. 2008). The difference in epithelial atrophy between OLP lesions may modify the degree of intraepithelial inflammatory infiltration, thus preventing a reliable measurement of FoxP3⁺ Treg cells in this area. On the other hand, epithelial hyperplasia is usually observed in IFH, which is associated with greater epithelial thickness and greater susceptibility to FoxP3⁺ Treg cell infiltration.

In conclusion, the present results showed an association between reticular OLP and a higher frequency of hyperkeratosis, greater epithelial thickness and a higher density of the underlying inflammatory infiltrate. In contrast, the erosive form was associated with a higher degree of atrophy, lower epithelial thickness, and lower density of the underlying inflammatory infiltrate. These results support the hypothesis that the distinct clinical appearance of the two variants of OLP is associated with specific morphological characteristics of each form that directly influence the clinical behaviour of these lesions. A larger number of FoxP3⁺ Treg cells are found in OLP compared to IFH. This finding is probably related to the different etiopathogenesis of these lesions, because persistent antigen induction occurs in OLP which promotes perpetuation of the lesion. The number of FoxP3⁺ cells tended to be higher in erosive OLP when compared to reticular lesions. High disease activity or the action of the oral microbiota may explain the larger number of FoxP3⁺ cells in erosive lesions.

Funding

There is no source of funding to declare.

Declaration of interest

The authors declare there is no conflict of interest that could be perceived as prejudicing the impartiality of the research reported.