The impact of topical Saint John's Wort (Hypericum perforatum) treatment on tissue tumor necrosis factor-alpha levels in plaque-type psoriasis: A pilot study.

CONTEXT: Psoriasis is an inflammatory disorder, formed by inappropriate interaction of T lymphocytes with keratinocytes, and consequent eruption of immune responses. High concentrations of tumor necrosis factor-alpha (TNFα) are found in the skin lesions and plasma of patients with psoriasis. Hypericum perforatum, a phytomedicine that has both anti-inflammatory and antiproliferative properties, has been recently reported to be clinically helpful for improvement of psoriatic lesions. AIMS: The aim of the present study was to investigate the effects of topical H. perforatum on TNFα levels in psoriatic lesions for possible identification of the mechanism by which Hypericum reduces inflammation and modulates the disease in patients with plaque-type psoriasis. SETTINGS AND
DESIGN: A double-blind, placebo-controlled, pilot study with intraindividual comparison was conducted on twenty patients with mild to moderate plaque-type psoriasis. SUBJECTS AND METHODS: TNFα levels in tissue samples were measured with immunohistochemistry method. Moreover, Psoriasis Area Severity Index (PASI) scores and histological and clinical changes were investigated after topical application of Hypericum extract. STATISTICAL ANALYSIS USED: The Wilcoxon signed-rank test was used to evaluate the possible differences between the drug and placebo group.
RESULTS: TNFα concentrations in dermis (p= 0.025), endothelial cells (p=0.033), and dendrite cells (p=0.014) were significantly reduced in lesions treated with drug and the reduction observed in epidermis was superior to placebo (p=0.046). Results of PASI scores showed that erythema, scaling, and thickness were significantly lower where the ointment had been applied compared to application of placebo (p=0.014, p=0.004, p=0.003, respectively). Moreover, significant improvement in clinical and histological features of treated lesions in comparison with untreated lesions was observed (P < 0.05).
CONCLUSIONS: H. perforatum ointment can help decrease PASI scores and TNFα levels in psoriatic tissue. Its efficacy is probably related to its effect on lowering cytokines including TNFα.

RCT Entities:   Population Interventions Outcomes

Introduction

Psoriasis is a chronic, multisystem inflammatory disorder which affects approximately 2%–4% of people in the United States.[1] Its prevalence has been reported to vary according to age, ethnicity, and geographic regions.[2] For instance, its prevalence has been estimated to be 0.3%–1.2% in Japan, 0.4% in China, and 4%–5.5% in Malaysia.[3] Psoriasis has a considerable effect on patients’ quality of life[4] and imposes high economic impact on country's health-care system.[1] Its pathogenesis has not been fully understood; however, recent studies have shown that defects in epidermal barrier-related genes are involved.[5] Psoriasis is also approached as a hyperplasia in epidermis[6] which is clinically categorized as a papulosquamous skin disorder.[7] However, in plaque-type psoriasis, erythematous plaques usually covered by silver, flaking scales are recognized.[8] Tumor necrosis factor-alpha (TNFα) is known to directly contribute to the characteristic plaque surmounted by silvery scales through induction of keratinocyte proliferation and prevention of keratinocyte apoptosis by increasing vasoactive intestinal peptide receptor and plasminogen activator inhibitor type 2, respectively.[9] Moreover, upregulation of vascular endothelial growth factor and intercellular adhesion molecule-1/vascular cell adhesion molecule-1 is done by TNFα which in return promotes angiogenesis and microvascular permeability and targeting of lymphocytes to the inflammatory lesions, respectively.[10] In addition, TNFα stimulates lymphocyte migration and increases expression of many proinflammatory cytokines including CCL27 (a skin-specific memory T-cell attractant) and nuclear factor kappa B (a ubiquitous transcription factor) which leads to the expression of many gene products that mediate inflammatory responses.[11]

The most appropriate treatment for patients with mild to moderate psoriasis is known to be topical therapy.[12] Mild to moderate psoriasis is defined as the Psoriasis Area Severity Index (PASI) score ≤10 in the present study. Topical immunomodulators are also known to be an emerging therapy for psoriasis.[13] Systemic and phototherapy are usually reserved for treatment of moderate to severe cases.[14] Hyperforin, one of the chief active ingredients of Hypericum's extract, is reported to have wound healing,[15161718] anti-inflammatory,[1920] and antibacterial properties.[2122]

It has been shown that Hypericum perforatum can be effective in patients with mild to moderate psoriasis by modulating the immune system through regulation of CD8-mediated cytotoxicity and inhibition of TNFα-induced apoptosis in vitro.[15] However, further larger studies are needed to arrive at more conclusive results.[23] Thus, the aim of this study was to investigate the effects of topical Hypericum on TNFα levels in psoriatic lesion in comparison with placebo in patients with plaque-type psoriasis.

Subjects and Methods

Twenty patients were enrolled in the present double-blind, placebo-controlled, and intraindividual comparison pilot study. Inclusion criteria were patients with the age between 18 and 55 years old; diagnosis of mild to moderate plaque-type psoriasis on both sides of the body; and not being under medical treatment for the past 2 months. Exclusion criteria were pregnancy, lactation, suffering from a major illness (such as cardiovascular disease, acute renal or liver disease, cancer, or active malignancy), and not having plaque psoriasis on both sides of the body. The study was approved by the Ethics Committee of Islamic Azad University, Pharmaceutical Sciences Branch (Number= 7173) and all patients participated in the study were informed of the study procedure and signed written consent forms. Moreover, the study was registered in Iranian Clinical Trials Center (IRCT) (Registration Number: IRCT201503173106N24). Due to the design as a pilot study and obligation of Ethics Committee (since the study needed tissue sampling from the patients), the sample size was limited to 20. However, 8 patients refused to give tissue samples at the point of sampling. Thus, 12 patients remained in the study.

After scoring clinical features and taking one biopsy from a psoriatic lesion of each patient, they received both H. perforatum ointment and placebo (vehicle), one applied randomly to the right and the other to the left side of the body twice daily for duration of 4 weeks. After that, along with clinical investigations, two punch biopsies (6 mm) each from lesions of the right and left side (the preferred symmetrically) were taken to examine the difference between drug and placebo group before and after the treatment.

Patients’ outcomes were evaluated according to PASI scores which were obtained at baseline and the end of 4 weeks period, by a senior dermatologist. Moreover, immunohistochemical (IHC) staining of the tissues was done to compare TNFα levels in the psoriatic tissues. Any possible side effects were recorded. PASI score (according to erythema, thickness, and scaling) was determined before and after the treatment, which were based on a 0–3 grade system, where 0 indicated an absence of the symptom, and 3 showed maximum severity. For the IHC study, paraffinized sections (5 μm) of the skin tissue was stained using a mouse anti-human TNFα (28401.111) monoclonal antibody (Santa Cruz Biotechnology Company) as the primary antibody and a goat anti-mouse IgG detected with envision as the secondary antibody. The intensity of TNFα IHC staining was measured using NIS Elements Software (Nikon Instruments Inc., New York, USA).

The formulated ointment was prepared from an extract of H. perforatum L (5% wt/wt), vaseline (84% wt/wt), propylene glycol (10% wt/wt), and avicel (1% wt/wt). The Hypericum extract which was extracted from the Hypericum plant was provided by Poursina Co. (Tehran, Iran). The vehicle was the same in the placebo and the active ointments contained the same percentages of propylene glycol, vaseline, and avicel. The formulations were packed in airtight bottles and went through stability tests. They were stored at 0°C, 30°C, and 45°C at different predetermined intervals (i.e., 30, 60 and 90 days), and their appearance, spreadability, texture, and phase separation were studied. Physical changes to color, odor, smoothness, and phase separation were observed visually during the study.

Two types of bottles were prepared, one filled with the drug formulation and the other with the same amount of placebo. Thus, drug and placebo were exactly identical in terms of their appearance and could not be identified neither by the clinician nor the patient. Moreover, the bottles were coded by a third party who wrote down the codes in a table, and the third party himself decoded the bottles at the end of the study.

Obtained data were analyzed by SPSS software (version 18.0, SPSS Inc, Chicago, Illinois). The Wilcoxon signed-rank test was used to evaluate the possible differences between the drug and placebo group. P < 0.05 were assumed statistically significant.

Results

Three of the patients were male, and eight patients were females. The mean age of the patients was calculated to be 41.25 ± 14.24. Two patients had the history of only skin involvement; whereas six patients had history of both skin and joint involvement. Moreover, four patients had history of both skin and nail involvement, and two patients had the history of skin, nail, and joint involvement. Mean duration of therapy before enrolling in the present study was calculated to be 1.41 years. Moreover, it should be noted that none of the patients were on biologics before the study. In IHC-stained samples, TNFα concentration in epidermis, endothelial cells, and dendrite cells was significantly reduced in lesions treated with drug (p=0.025, p=0.033, p=0.014, respectively) and the reduction in TNFα concentration was reported to be superior to placebo in epidermis tissue [p=0.046, Table 1]. Images showing clinical photos before and after the treatment are shown in Figure 1.

Table 1

Tissue tumor necrosis factor-alpha levels before and after treatment (pg/ml)

	TNFα level in parakeratosis tissue	TNFα level in epidermis	TNFα level in basal layer	TNFα level in endothelial tissue	TNFα level in dendritic cells
Mean concentration before treatment	1.41	0.58	1.58	1.50	0.66
Mean concentration after treatment with drug	0.83	0.16	0.92	1.00	0.25
Mean concentration after treatment with placebo	1.08	0.42	1.41	1.41	0.42
Difference in mean concentration after treatment with drug	0.58	0.42	0.66	0.50	0.41
Difference in mean concentration after treatment with placebo	0.33	0.16	0.17	0.09	0.24
Before and after treatment with drug (p)	0.35	0.025	0.212	0.033	0.014
Before and after treatment with placebo (p)	0.157	0.564	0.004	0.564	0.180
Difference between drug and placebo group (p)	0.083	0.046	0.336	0.059	0.083

TNFα: Tumor necrosis factor-alpha

Figure 1

Clinical images before and after the treatment. (a) Left foot before treatment. (b) Left foot after treatment with drug. (c) Right foot before treatment. (d) Right foot after treatment with base ointment

According to the clinical investigations, disease severity was shown to be significantly reduced by 1-month application of the drug. However, reduction of thickness, scaling, and pruritus scores was more significant. Thickness and scaling scores were reduced by the placebo ointment as well, however, when compared with the drug ointment, reduction observed by application of the drug ointment was shown to be superior to the placebo ointment [Table 2]. Reduction of erythema, scaling, thickness, and pruritus scores after application of drug and placebo are compared and shown in Table 2. Moreover, images showing pathologic results with H and E and IHC staining are shown in Figure 2.

Table 2

P values obtained from Wilcoxon tests to compare reduction of Psoriasis Area Severity Index scores with drug and placebo before and after treatment

	After treatment with drug	After treatment with placebo	Comparison of placebo and formulated ointment after treatment
Erythema score	0.010	0.414	0.014
Scaling score	0.004	0.008	0.003
Thickness score	0.003	0.004	0.002
Pruritus score	0.005	0.096	0.008

Figure 2

Pathological images before and after the treatment. (a) Foot tissue before treatment. (b) Left foot tissue after treatment with drug. (c) Right foot tissue after treatment with base ointment (a-c: H and E, ×10). (d) Immunohistochemical staining before treatment. (e) Immunohistochemical staining after treatment with drug. (f) Immunohistochemical staining after treatment with base ointment (d-f: AEC and TMB [HRP], ×40)

All the histological features that show the active stage of the disease including spongiosis, acanthosis, parakeratosis, thinning of suprapapillary plates, Munro microabscesses, and spongiform pustules of Kogoj were significantly reduced in tissue samples of lesions treated with drug ointment (p < 0.05) while acanthosis, hypogranulosis, and suprapapillary thinning were more significantly reduced (p = 0.003, p = 0.008, p = 0.007), respectively. Except for parakeratosis and spongiosis which did not change with placebo treatment, other histological features were improved with placebo treatment as well (p < 0.05). Reduction reported in spongiosis, parakeratosis, and suprapapillary thinning in drug-treated tissues were significantly superior to placebo (p < 0.05). Diluted vessels and edema in papillary dermis and leukocyte infiltration did not change significantly during 1-month topical treatment but fibrosis which shows the entrance of lesions into remission stage was significantly increased in tissues treated with the drug ointment [p < 0.05, Table 3]. Different pathological features including parakeratosis, acanthosis, and fibrosis are compared after drug and placebo application and are shown in Table 3.

Table 3

Pathological features before and after treatment

	Before and after treatment with drug (p)	Before and after treatment with placebo (p)	Comparison of placebo and formulated ointment (p) after treatment
Parakeratosis	0.013	0.132	0.034
Acanthosis	0.003	0.011	0.739
Spongiosis	0.014	0.361	0.008
Hypogranulosis	0.008	0.005	0.058
Thinning of suprapapillary plates	0.007	0.003	0.034
Munro microabscesses	0.010	0.008	0.317
Spongiform pustules of Kogoj	0.023	0.035	0.059
Edema in papillary dermis	0.180	0.100	0.083
Dilated vessels	0.059	0.257	0.480
Lymphocyte	0.206	0.025	0.650
Neutrophil	0.011	0.059	0.083
Eosinophils	0.100	0.317	0.317
Plasma cells	0.083	0.564	0.157
Fibroblast	0.257	0.033	0.059
Fibrosis	0.046	0.317	0.180

Both drug and placebo ointment were well tolerated by the patients, and there was no report of allergic reactions and side effects. One patient had recurrence of the lesions.

Sites of involvement were hands and feet, and there was no difference in the resolution of plaques depending on the site of involvement.

Discussion

It has been shown that topical preparations of H. perforatum can be effective in wound healing, atopic dermatitis, and mild to moderate psoriasis. However, further larger studies are needed to arrive at more conclusive results.[15] Thus, the present study was designed to investigate the effects of H. perforatum's extract on TNFα levels in psoriatic tissue for possible identification of the mechanism by which Hypericum, modulates the disease, and decreases inflammation.

Najafizadeh et al. investigated the clinical effects of topical H. perforatum in plaque-type psoriasis. According to the results, all factors of PASI scores including erythema, scaling, and thickness were decreased significantly where the formulated ointment was applied. However, conduction of further studies was suggested.[23]

The results of the present study were in consistence with the results of Najafizadeh et al.'s study. In other words, it was shown that Hypericum ointment could decrease erythema, scaling, and thickness in patients with mild to moderate plaque-type psoriasis. Superior reduction of thickness and scaling of the lesions with drug application can be the result of reduction of epidermal hyperproliferation and improvement of keratinocytes’ differentiation and natural formation of epidermal layers. Naftodiantrones in Hypericum extract such as hypericin and pseudohypericin can inhibit C-protein kinase; thus, be effective in reduction of proliferation rate of epidermal cells.[24] Hyperforin can also reduce keratinocytes’ proliferation by increasing apoptotic cell death.[25]

Moderate reduction in the area of involvement and erythema in lesions treated with drug can be the result of decrease in permeability and dilution of capillary vessels and reduction of edema in dermis of the lesions. Quercetin and amentoflavone Hypericum extract inhibit lipid inflammatory mediators production (which increase vascular permeability) and reduce TNFα and interleukin (IL)-6 which can lead to reduction of edema and inflammation.[2627] Reduction in neutrophils by the application of Hypericum could be due to the effects of quercetin, hyperforin, or chlorogenic acid on the inhibition of neutrophils’ chemoattraction to skin, through reduction of adhesion molecules or leukotrienes or both.[2627] Reduction of TNFα concentrations in tissues treated with drug could be the result of inhibitory effects of quercetin and amentoflavone on production of TNFα or the result of inhibitory effects of hyperforin on activation of T lymphocytes, and therefore, suppression of initiating process.[2628] Studies by Tobin and Kirby on effects of TNFα inhibitors on treatment of psoriasis support this result.[29]

Xiuying et al. studied the effects of H. perforatum on treatment of flu infection in mouse. According to the results, Hypericum's extract was reported to decrease TNFα and IL-6 concentrations and increase interferon γ and IL-10 levels in mice lungs and serum.[30] In another study investigating the effects of Hypericum's extract's on reduction of lung injury in mouse, the extract was reported to decrease TNFα and IL-1β.[31] Moreover, in another study, the effects of H. perforatum's extract on irritable bowel syndrome were evaluated in rats. According to the results, the extract decreased TNFα and inflammatory cells accumulation.[32] However, studies by Canning et al. showed that Hypericum had probably no significant effect on cytokines levels in premenstrual syndrome in women.[33]

It has been suggested that anti-inflammatory effects of H. perforatum components appear in higher concentrations than what is in the extract, except for pseudohypericin, and hyperforin whose effects appear at concentrations near to what's in the extract. However, there is a belief that all these components next to each other can synergistically exhibit a significant anti-inflammatory effect. Moreover, it has been suggested that four compounds, namely, pseudohypericin, quercetin, amentoflavone, and chlorogenic acid all together, while being purified out of extract, can reduce TNFα but while into extract cannot have such effect.[34]

Conclusion

H. perforatum ointment helps reduce PASI scores and TNFα levels in psoriatic tissue in patients with plaque-type psoriasis. Its efficacy is probably related to its effect on lowering cytokines including TNFα.

Financial support and sponsorship

Financial support was provided by the Islamic Azad University, Pharmaceutical Sciences Branch, a nongovernmental nonprofit organization.

Conflicts of interest

There are no conflicts of interest.