Inhibitory effect of pentoxifylline on the synthesis and deposition of glycosaminoglycans in hypothyroid skin

Abstract The skin manifestations in hypothyroidism are common but their advanced forms do not have adequate therapy. Along with achievement of euthyroidism, thyroid-associated dermatopathy (TAD) needs an inhibition of the proliferation and activation of fibroblasts, and of the glycosaminoglycans synthesis. Pentoxifylline (PTX) exerts multifaceted effects and has shown properties to inhibit certain constitutive activities of dermal fibroblasts. The aim of this study was to determine the histopathological changes and to establish the amount and distribution of mucin in hypothyroid skin, with or without treatment with PTX. Hypothyroid skin showed abundant diffuse mucin deposition in all dermal compartments in comparison with euthyroid skin. After administration of PTX in dose 100 mg/kg daily, there was a significant decreased amount of mucin in the dermis. The improvement of the dermal histological manifestations, in general demonstrated by narrowing between the dermal collagen fibers and by a decrease in mucin deposition, might be explained with dose-dependent PTX inhibition of the synthesis and accumulation of glycosaminoglycans. The histological results of this study confirmed several clinical observations about the beneficial effect of PTX on clinical skin manifestations in hypothyroid state. Therefore, we may propose PTX as a new potential therapeutic agent for TAD treatment. Further studies are needed to fully establish the effect of PTX on dermal fibroblasts after continuous administration.


Introduction
Skin is commonly affected from thyroid disorders and the skin changes are presented as thyroid-associated dermatopathy (TAD) [1,2]. Myxedema is an infiltrative TAD; it is either localized (as a pretibial myxedema) in thyrotoxicosis, or disseminated in overt hypothyroidism [3,4]. Pretibial skin lesions are typical and are related to mechanical factors. Other localization of TAD indicates a systemic involvement. Almost all cases with TAD are presented with relatively severe ophthalmopathy which appears first. In Graves' disease and Hashimoto's thyroiditis both humoral and cellular mechanisms are involved in the stimulation of fibroblasts and in the production of large amounts of acid glycosaminoglycans in the dermis [5]. Six major groups of glycosaminoglycans compose dermal mucin (MPS), whose increased amount is a characteristic histopathologic feature of TAD. Glycosaminoglycans absorb water and bind to extracellular protein. Additionally, they could activate macrophages, dendritic cells and neutrophils, and may inhibit the effects of tumor necrosis factor-alpha (TNF-α) and interleukin 6 (IL-6) [6,7].
About 50% of the mild cases of TAD achieve complete remission after several years without treatment. Advanced forms with nodules or elephantiasis are treated along with achievement of euthyroidism with intralesional and topical glucocorticoids or other off-label therapies but their outcome is similar to untreated mild cases. Unfortunately, current treatment modalities for severe TAD and acropachy are mostly palliative.
Pentoxifylline (PTX) is a methyl-xanthine derivative with a variety of anti-inflammatory properties. The Pentoxifylline (PTX); hypothyroidism; thyroid-associated dermatopathy (TAD); colloidal iron; glycosaminoglycans; mucin drug is effective for many dermatological conditions, including fibroblast mediated diseases [8,9]. It has been found that PTX inhibit in vitro the proliferation and activation of fibroblasts, and glycosaminoglycan synthesis [10,11]. Furthermore, PTX suppresses certain constitutive and TNF-α-induced activities of normal dermal fibroblasts [12] and has beneficial effect on thyroid-associated ophthalmopathy [13].
The goal of this study was to investigate the effect of PTX on skin changes in a state of hypothyroidism, and to establish the amount and distribution of mucin in the skin of untreated rats and rats treated with different doses of PTX. We also aimed to determine the level of TNF-α and IL-6 in the skin of the same hypothyroid rats treated or not treated with PTX. These cytokines are thought to be mainly involved in inflammation and might stimulate dermal fibroblasts to synthesize collagen and glycosaminoglycans.

Ethics statement
All animals were treated in agreement with the general regulations for treatment of experimental animals, established by the Ethics Committee of the Medical University of Sofia (study was approved by protocol № 68/12.09.2016 from KENIMUS), in agreement with EU Directive 2010/63/EU of the protection of animals used for scientific purposes.

Dose-dependent treatment with PTX
After reaching the hypothyroid state (proved by the levels of the peripheral hormones), the rats were divided into the following groups: untreatedhypothyroid and hypothyroid treated with PTX (25, 50, 100 mg/kg, p.o, respectively) for 14 days, five animals in each group [15]. A control (euthyroid) group was used for a comparison. The animals were kept at room temperature (25 ± 0.5 °C), standard humidity (60 ± 1%) and a light/dark cycle.
After 2 weeks, PTX-treated-and untreated hypothyroid rats were anaesthetized with thiopental (40 mg/kg i.p.) and 6 mm punch biopsies were taken from the femur region after shaving of the skin. The skin tissues used for measurement of the cytokines were placed in cold 0.9% NaCl, homogenized and centrifuged at 4 °C and 6000 g for 10 min. The supernatant was collected and used for TNF-α and IL-6 measurement by enzyme-linked immunosorbent assay (rat tissue lysates ELISA kit; Sigma-Aldrich ® , St. Louis, USA). ELISA was performed according to the manufacturer's protocols for TNF-α and IL-6 with determination of a standard curve and calibration solutions in pg/mL. The ELISA results were multiplied by the primary dilution, recalculated in ng/mL.
The other biopsies were fixed with 10% formalin. Histology was performed with H&E, Van-Gieson (for collagen) and colloidal iron (marker used to distinguish dermal mucin) stains according to the standard protocols.

Statistical analysis
Data obtained in the experiments were expressed in terms of mean values with standard deviation (± SD). Statistical significance of data was assessed by analysis of variance (one-way ANOVA) followed by a comparison between different groups using Bonferroni's post-hoc test. Differences were considered statistically significant at the p < 0.05 level.
The distribution of colloidal iron stained mucin was studied by implementing threshold analysis by using ImageJ software (http://rsb.info.nih.gov/ij, written by Wayne Rasband at the National Institute of Health (Bethesda, MD, USA)) with a Threshold Colour Plugin created by Gabriel Landini, which is a modification of Bob Dougherty's BandPass2 Filter (http://www.dentistry.bham.ac.uk/landinig/software/software.html). We selected only the pixels corresponding to the acid mucin (visualized as bright blue color) and we calculated the percentage of total area it occupies in five slides from each group with the same magnification and similar structure.

Results
The experimental hypothyroid model was proven by a significant decrease in fT4 for hypothyroid rats (0.74 ± 0.31 ng/L) in comparison with euthyroid (22.41 ± 0.28 ng/l) rats. The hypothyroid state demonstrated a negative impact on the growth and weight gain of the rats, and slowed the metabolism of the animals, similar to the previously published data [16].
In comparison with euthyroid skin (Figure 1a and  d), the histopathological examination of the skin of hypothyroid animals revealed a typical thinning hyperkeratotic epidermis, dermal mucin deposition, mainly in the lower dermis, and mild perivascular and periadnexal lymphocytic infiltrate (Figure 1b). Mucin deposits revealed a separation of collagen bundles in the dermis clearly visible with van Gieson staining in hypothyroid skin (Figure 1e). The amount and distribution of dermal mucin was highlighted by specific colloidal iron stain, where mucin was visualized as electric blue colored stringy material in between and attached to collagen fibers [17]. As illustrated in Figure 2a, the hypothyroid skin showed an abundant diffuse mucin deposition in all dermal compartments in comparison with euthyroid skin (Figure 2c). In euthyroid skin, scarce thin mucin fibers were seen mainly subepidermally and  . in the propylthiouracil (ptu)-treated rats, the mucin was significantly more than in the controls and in the hypothyroid rats treated with pentoxiphylline (pentoxi). there was no significant difference between the control and the pentoxi groups.
around hair follicles, glands and vessels. In contrast with healthy skin, there was significantly increased mucin material, histologically seen as thick fibers and lamellas almost filling and expanding the spaces between collagen fibers in hypothyroid skin. The mucin distribution was most pronounced in the deep reticular dermis (Figure 2a) in hypothyroid rats.
The treatment with 25-and 50 mg/kg PTX for 14 days did not show any recognizable amelioration in dermatohistology (pictures are not shown here). Only after treatment with 100 mg/kg PTX for 2 weeks, histological analysis of van Gieson and colloidal iron stains revealed a visible improvement of the skin histology, mainly in collagen fiber separation (Figure 1c and f) and mucin deposition (Figure 2b), without changes in the epidermis. After treatment with PTX, hypothyroid skin showed a decreased amount of MPS in papillary and superficial reticular dermis, and moderate amount in deep reticular dermis, arranged as a thin fiber net (Figure 2b).
Additionally, we evaluated the cutaneous level of TNF-α and IL-6 in our hypothyroid model (Figure 3). The ELISA test revealed a decreased TNF-α in the hypothyroid compared to euthyroid skin (p < 0.01). In the untreated hypothyroid group the average level of TNF-α (5.82 ± 1.4 ng/mL) was detected significantly lower than in the control (euthyroid) group (13 ± 2 ng/ mL). After treatment with 25 mg/kg PTX no difference was seen (5.75 ± 1.0 ng/mL), compared with untreated hypothyroid rats. Application of 50 or 100 mg/kg PTX led to slight, but statistically significant (p < 0.05) diminishing of TNF-α level up to 3.8 ± 1.8 and 3.2 ± 0.2 ng/mL, respectively, compared to the untreated hypothyroid skin. No difference in IL-6 in the skin was found between euthyroid (1.29 ± 0.4 ng/mL) and hypothyroid (1.26 ± 0.3 ng/mL) animals. PTX application slightly decreased IL-6.

Discussion
Although cytokines and growth factors have been postulated to play a role in TAD, the pathogenesis of pretibial myxedema remains unclear. Recent evidence proved the important role of the cutaneous microenvironment (such as antibodies, fibroblasts, T cells, B cells, plasma cells) in the development of pretibial myxedema [18].
The multifaceted properties of PTX were proved in the last 15 years and it became a subject of a number of scientific studies. PTX is known to have a diminishing effect on the production of pro-inflammatory cytokines such as TNF-α, IL-6 and iNOS, which allows PTX to reduce the inflammatory response [15,19] and to decrease the oxidative stress [20]. Moreover, PTX exerts a wide spectrum of possibilities for application in dermatology. A high incidence of hypothyroidism has been reported within granuloma annulare patients [21]. The beneficial effect of PTX on generalized granuloma annulare was established [22][23][24]. Several clinical data also confirmed the ameliorating and in some cases the reversing fibrosis effect of PTX [25].
Here, we found an improvement of the histological changes in the skin of the hypothyroid model after 14 days oral administration of 100 mg/kg/24h PTX. The cutaneous histological manifestations demonstrated by narrowing between the dermal collagen fibers and by a statistically significant decrease in mucin deposition might be explained with a dose-dependent PTX inhibition of the synthesis and accumulation of glycosaminoglycans. Our results support the data reported by Chang et al. [10], who observed in vitro inhibition of the glycosaminoglycan synthesis for different kind of fibroblasts after exposure to PTX. Additionally, it has been reported that PTX inhibits TNF-α induction of fibroblasts to produce collagen and glycosaminoglycans [12]. In our study, the lower TNF-α level in hypothyroid compared to its level in euthyroid skin could be considered as a result of a decrease in the overall metabolic rates in the PTU-induced hypothyroid model [14]. Evidently, the effects of PTX on fibroblast proliferation and glycosaminoglycans production could be attributed not only to its anti-TNF-α effect but also to separate mechanisms.
The present study complements the clinical observations for a good response of pretibial myxedema to oral PTX alone or in combination with topical glucocorticoids [26,27]. . comparative assessment of tnF-α and il-6 concentrations in the skin of euthyroid rats and in hypothyroid rats untreated (ptu) and treated with 25, 50 and 100 mg/kg pentoxiphylline (ptX). We observed a statistically significant decrease in the concentrations of tnF-α not only in the hypothyroid skin, but also in the hypothyroid animals treated with 50 mg/kg and 100 mg/kg ptX. no difference in the cutaneous il-6 was found between euthyroid and hypothyroid animals.

Conclusions
To our knowledge, this is the first study in the literature to evaluate both, the mucin deposition in hypothyroid skin, and the effect of PTX on skin changes. PTU-induced hypothyroidism showed a statistically higher mean level of mucin in the skin. It significantly diminished after 14 day of 100 mg/kg PTX administration. It is important to consider the potential clinical utility of PTX on the histological and clinical skin manifestations associated with hypothyroidism. Furthermore, it may be helpful for treatment of dermatoses with an increased dermal mucin, like lichen myxedematosus, cutaneous lupus erythematoses, etc. Further studies are needed to fully establish its effect on dermal fibroblasts after continuous administration.