Observational study on Swedish plaque psoriasis patients receiving narrowband-UVB treatment show decreased S100A8/A9 protein and gene expression levels in lesional psoriasis skin but no effect on S100A8/A9 protein levels in serum

S100A8 and S100A9 proteins are highly upregulated in patients with psoriasis and have been proposed as potential biomarkers for psoriasis. The present study was designed to analyze the effect of narrowband ultraviolet B therapy on these proteins. S100A8, S100A9 gene expression and S100A8/A9 heterocomplex protein levels were analyzed in lesional and non-lesional skin before and after narrowband-UVB treatment in patients with chronic plaque type psoriasis. In addition, disease severity was measured by psoriasis area and severity index (PASI) and serum protein levels of S100A8/A9 were repeatedly analyzed. Narrowband-UVB treatment significantly reduced S100A8, S100A9 gene expression and S100A8/A9 protein levels in lesional skin while serum levels showed no significant change. No correlation between PASI and serum S100A8/A9 protein levels was found. These results implicate a role of S100A8/A9 in the anti-inflammatory effect of narrowband-UVB. Serum S100A8/A9 levels do not respond to treatment suggesting that serum S100A8/A9 does not originate from psoriasis skin keratinocytes. Serum S100A8/A9 levels do not correlate with PASI questioning serum S100A8/A9 as a biomarker for psoriasis skin activity. Trial Registration: DRKS 00014817.

The genes coding for S100A8 and S100A9 are located within the psoriasis susceptibility locus 4 (PSORS4) mapped to chromosome 1q21.S100A8 and S100A9 are multifunctional proteins, and it is hypothesized that S100A8 and S100A9 functions vary with concentration and oxidation of the S100A8/A9 heterocomplex.S100A8/A9 properties include oxidant scavenging, antimicrobial activity, leukocyte chemoattractant and chemokine-like activities suggesting a regulatory role in inflammation [3,4].S100A8 and S100A9 proteins are primarily expressed in neutrophils where they compose 40% of the cytosolic content and in monocytes/macrophages [5].Keratinocytes express S100A8/A9 in response to stress, including wound healing, tape stripping, and exposure to detergents [6].
S100A8/A9 has become a widespread biomarker used in the diagnosis and monitoring of inflammatory bowel disease (IBD) [7] and has been proposed as a potential biomarker for psoriasis and psoriasis arthritis (PsA) [8,9].S100A8/A9 is present at high levels in extracellular fluids (synovia, blood and sputum) in various inflammatory diseases such as rheumatoid arthritis and cystic fibrosis [3,5].Serum levels of S100A8/A9 in middle age healthy individuals correlate to blood neutrophil counts [10] and high levels of circulating S100A8/A9 are associated with increased risk for cardiovascular events and carotid arteriosclerosis [11].Plasma S100A8/A9 has a short half-life of approximately 5 hours [12].
In psoriasis mouse models, deletion of S100A9 improves psoriasis-like skin disease, and studies on adenovirus induced overexpression of S100A8/A9 in human keratinocytes show increased levels of TNF-alpha, IL-6 and IL-8 in cell medium suggesting a pro-inflammatory effect and important role for S100A8/A9 in psoriasis pathogenesis [13,14].
S100A8 and S100A9 are exceedingly upregulated in the epidermis in lesional skin of patients with psoriasis [13] and histopathological analysis of psoriatic lesions show increased levels of S100A8/A9 in keratinocytes compared to healthy skin [15][16][17].These findings led to the assumption that S100A8 and S100A9 might be potential therapeutic targets for the treatment of psoriasis [15].Previous genomic transcriptome studies on the effect of NB-UVB in psoriasis showed downregulation of S100A8 and S100A9 in response to NB-UVB [18,19].However, these studies included only 3 and 11 patients and did not analyse protein expression.Benoit et al. revealed increased S100A8/A9 protein serum levels in patients with psoriasis compared to healthy controls as well as a positive correlation with disease severity measured by PASI [8].The authors suggested keratinocytes from lesional skin to be the source of elevated serum S100A8/A9 levels.However, other studies have not been able to confirm a correlation between disease severity and S100A8/A9 serum levels [9,20,21] but proposed that S100A8/A9 might be a potential biomarker of PsA [9,20].
Only scarce data is available on the response of S100A8/A9 protein expression to psoriasis treatment.One previous study showed that etanercept treatment reduces S100A8/A9 protein levels in serum [21].
The present study was designed to analyse skin and serum expression of S100A8/A9 before and after NB-UVB treatment in patients with chronic plaque type psoriasis.The correlation between psoriasis area and severity index (PASI) and serum levels was studied to investigate a potential role of serum S100A8/A9 as a biomarker for skin disease severity.

Study subjects
Patients with chronic plaque psoriasis eligible for NB-UVB treatment were recruited for the study.According to the national treatment recommendations of the Swedish Society of Dermatology and Venereology NB-UVB is primarily offered to plaque psoriasis patients with a PASI value of 3-9 where topical treatment is insufficient to obtain disease control.All participants were included at the dermatology outpatient clinic at Ryhov hospital, Jo ¨nko ¨ping, Sweden (December 2013 to March 2016).Exclusion criteria were ongoing systemic antiinflammatory treatment (including any systemic treatment for psoriasis), PsA or other rheumatological diseases, IBD, other inflammatory dermatoses, pregnancy, intense UV exposure two weeks prior to study start (i.e.vacation, tanning salon), chronic infectious diseases, exposure to photosensitizing drugs and age < 18 years.
The study was conducted in compliance with good clinical practice and according to the Declaration of Helsinki Principles.All subjects received written study information prior to recruitment and written informed consent was obtained.Study protocols were approved by the ethical committee at Linko ¨ping University, Linko ¨ping, Sweden (Dnr.2012/428-31) (S10.1 and S10.2).Age, sex, body mass index (BMI) and current medication were recorded for all patients.A minimum sample size of n = 25 was determined based on the assumption that NB-UVB would have an effect size corresponding to the magnitude of the difference of serum S100A8/A9 levels from patients with psoriasis selected at random in the dermatology department (n = 47) and healthy blood donors (n = 8) at Ryhov hospital, Jo ¨nko ¨ping, Sweden.Power (1-β) being set to 0.8 and α to 0.05 (S1 Sample size).Out of 30 recruited patients, 27 completed a full NB-UVB treatment (21.4±4.2NB-UVB sessions).Dropouts were due to severe illness, change of residence and unspecified personal reason.Dropout average baseline disease severity was similar to subjects who fulfilled the study (PASI 7.6 vs PASI 7.9).

Study design
NB-UVB (311nm) therapy was administered using a Waldmann 7002 cabin (Waldmann Medizintechnik, Villingen-Schwenningen, Germany).Patients were treated 2.3 times (±0.7) per week and the mean treatment period was 10.4 weeks (±3.6).The mean maximum dose reached was 2.64 J/cm2 (±1.2) at the end of the treatment period.Energy output was measured with a standard intrinsic UV meter.Initial dose was dependent on skin phototype and increased 20% at each visit if well tolerated.When a previous treatment resulted in erythema, the dose was maintained or the dose was decreased, depending on whether the erythema was asymptomatic or severe and painful.
Serum S100A8/A9 protein and PASI levels were measured before the first, 6th, 11th, 16th, 21st, 26th and final treatment.The exact number of treatment sessions was not predetermined by the study design but based on clinical response and motivation of the study subjects to continue treatment.The NB-UVB regimen was designed to be as similar to the ordinary clinical setting as possible.Before the first and the final NB-UVB treatment session two 2 mm skin biopsies from the central parts of a single psoriasis plaque (target lesion) and two 2 mm skin biopsies from non-lesional skin 10 cm from target lesion site were taken for S100A8/A9 gene and protein analysis (Fig 1).Target lesions were selected on the basis of being representative of the patients general disease (assessed by dermatologists AD, MA and OS), being distinct as to be easily identified in follow up, not located on face, chest, distal parts of arms and hands to avoid scars in these locations.Target lesion psoriasis severity and total body PASI were recorded before biopsy sampling.PASI was first introduced by Fredriksson and Pettersson in 1978 and is currently a widespread method for disease evaluation [22,23].Target lesion severity was calculated by adding PASI scores (0-4) for erythema, induration and desquamation of the target lesion and was denoted "target PASI".Participants were instructed not to treat the target lesion with topical treatment other than moisturizing cream.For other lesions, topical treatment was restricted to moisturizing cream and mometason 0.1%.The latter was to be applied to a maximum of 30% of the total body surface area.S100A8/A9 protein analysis.Venous blood samples for protein analysis were collected in BD Vacutainer gel tubes (Becton, Dickinson and Company).After 30 minutes the tubes were centrifuged at 2000xg for 10 minutes at room temperature.300 μl serum was frozen at -80˚C for subsequent protein analysis for each sample.
Skin biopsies sampled for protein analysis were immediately stored at -80˚C.Before protein analysis, samples were homogenized in RIPA Lysis Buffer System (Santa Cruz Biotechnology) using Tissue Lyser II (Qiagen) set at 3 + 3 minutes, 30Hz using 7 mm stainless beads.Serum and skin levels of S100A8/A9 heterocomplex were analysed using Phadia EliA Calprotectin assay (Thermo Fisher Scientific).
S100A8/A9 RNA purification and gene expression analysis.Skin biopsies sampled for gene expression analysis were immediately immersed in RNAlater RNA Stabilization Reagent (Qiagen) and stored at 4˚C.After 24 hours RNAlater was removed and samples were frozen at -196 ˚C.
Total RNA was purified according to manufacturer's instructions.Skin biopsies were homogenized using a TissueRuptor and disposable probes (Qiagen).RNA was purified using the RNeasy Fibrous Tissues mini kit (Qiagen).Concentration and purity was measured using Nanodrop ND-1000 (Thermo Fisher Scientific Inc.), and RNA integrity was assessed with the 2100 Bioanalyzer (Agilent technologies).
RNA reverse transcription was performed using the High capacity cDNA reverse transcription kit with RNase inhibitor (Applied Biosystems), according to the manufacturer's instructions, and resulting cDNA was stored at -80 ˚C.Gene expression was analyzed on the 7500 Fast real-time PCR system (Applied Biosystems) and the standard run mode using Taq-Man Universal Master Mix no UNG (Applied Biosystems) and Taqman Gene Expression Assays (Applied Biosystems) for S100A8 (Hs00374264_g1), S100A9 (Hs00610058_m1), TBP (Hs00427620_m1), ACTB (Hs99999903_m1) and GAPDH (Hs03929097_g1).For each assay and sample, cDNA based on 10 ng total RNA were analyzed in a total volume of 20 μl.
Threshold cycle (Ct) values were established using the 7500 software (version 2.0.6,Applied Biosystems).Reference genes (TBP, ACTB and GAPDH) were evaluated for low sample-tosample variation using the NormFinder [24] algorithm implemented in GenEx Professional software (version 5.4.2.128, MultiD Analyses AB).Ct values were normalized to the TBP reference gene showing the best stability value and GenEx Professional software was used to compute relative gene expression based on the comparative Ct (2 -ΔΔct ) method [25].

Statistics
Results are expressed in tables as mean ± standard deviation (SD).The Shapiro-Wilk test was used to test for normality (S1 Table ) and Wilcoxon signed rank test to assess the effect of NB-UVB on S100A8/A9 levels.Kendall´s tau rank correlation (correlation coefficient τ) was used to measure non-parametric ordinal association.Statistical analyses and graphical presentation were performed using IBM SPSS statistics (Version 22) and GraphPad Prism (Version 7.0b).G � Power (Version 3.1.9.3) was used to calculate sample size.P<0.05 was considered statistically significant.Dropout subjects were not included in statistical analysis.

Treatment response
Out of 30 recruited study participants, 27 fulfilled NB-UVB treatment as described in the method section (Fig 1).82% of patients reached PASI50 (at least 50% improvement of PASI) and 52% PASI75.Demographics of study population (age, gender, BMI) and details of treatment regimen, duration, and outcome are summarized in Table 1.
Increased S100A8/A9 expression in lesional skin compared to non-lesional skin.S100A8 and S100A9 gene expression in lesional skin was significantly elevated compared to non-lesional skin before start of NB-UVB treatment (Figs 2 and 3).Results from protein analysis were consistent with gene expression results showing significantly increased S100A8/A9 protein levels in lesional skin compared to non-lesional skin before NB-UVB therapy (Fig 4).S100A8/A9 expression is significantly decreased in lesional skin after NB-UVB treatment.After NB-UVB treatment, skin biopsies from lesional skin showed significantly reduced S100A8 and S100A9 gene expression compared to levels before NB-UVB treatment (Figs 2 and 3).Gene expression results were consistent with skin protein expression of S100A8/A9, which also showed a significant decrease after NB-UVB (Fig 4).No significant effect of NB-UVB treatment on serum S100A8/A9 levels.Serum protein levels of S100A8/A9 heterocomplex before (357ng/ml ± 214) and after (369ng/ml ± 200)

Discussion
Several S100 proteins and particularly S100A8 and S100A9 are highly upregulated in psoriasis plaques [6,26].Hence, potential roles for S100A8 and S100A9 both as treatment targets and biomarkers for psoriasis disease severity have been proposed.
The present study shows for the first time that NB-UVB treatment significantly suppress elevated S100A8/A9 protein heterocomplex levels in lesional skin to levels comparable to those found in non-lesional skin.These findings are supported by corresponding changes in skin gene expression.Interestingly, our results show that treatment with NB-UVB has no significant effect on serum levels of S100A8/A9.These results suggest that elevated serum S100A8/A9 levels in individuals with psoriasis do not originate from lesional skin keratinocytes as previously suggested by Benoit et al. [8].As proposed in earlier studies on S100A8/A9 in healthy middle-aged individuals and patients with systemic lupus erythematosus the origin of serum S100A8/A9 could be circulating neutrophils or thrombocytes.Elevated S100A8/A9 levels in these studies were associated with increased risk for cardiovascular disease [10,27] implicating that increased S100A8/A9 serum concentrations could be a marker of systemic inflammation rather than skin inflammation in individuals with psoriasis.Since etanercept treatment induces significant serum S100A8/A9 reduction in patients with psoriasis [21] S100A8/A9 might be involved in the systemic anti-inflammatory effect of biologics.The expression of S100A15 (Koebnerisin) and S100A7 (Psoriasin) are increased in circulating peripheral blood mononuclear cells (PBMCs) in patients with psoriasis [28].If S100A8 and S100A9 share similar expression patterns then it is possible that PBMCs could be a source of elevated S100A8/A9 in NB-UVB is thought to exert immunosuppression in psoriasis partly by inhibiting Th17 and interferon signalling pathways in the skin [18].It is not fully elucidated if NB-UVB therapy has effect on systemic inflammation in psoriasis.Some studies have shown decreased levels of IL-17A, TNF-α, S100A15, S100A7, S100A9 and IL-6 gene expression in PBMCs after NB-UVB treatment in patients with psoriasis [28,29] whereas other pro-inflammatory mediators such as sVCAM-1, sICAM-1, sE-selectin, MMP-9, and MPO showed no significant reduction [30].The latter proposing an anti-inflammatory effect of NB-UVB solely restricted to the skin which is consistent with our data showing no effect of NB-UVB on serum S100A8/A9 levels.
Only scarce data is available describing the effect of NB-UVB treatment on the expression of S100A8/A9 in skin from patients with psoriasis.One previous study showed that NB-UVB downregulates S100A9 gene expression in psoriasis skin and another revealed downregulation of S100A8.Both studies were performed on small samples, including 11 and 3 patients [18,19].Acute UV radiation induces in vivo upregulation of S100A8 in the skin [31] whereas UVA radiation induces S100A8, but not S100A9 in murine epidermis and keratinocyte cell lines possibly by oxidative stress [32].It is completely unknown how UVB interferes with S100A8/ A9 expression but it is interesting to speculate that the anti-inflammatory effect of NB-UVB treatment might partly be mediated by the reduction of S100A8/A9 expression in lesional skin.
Our data show no correlation between PASI and serum S100A8/A9 concentration before, during or after UVB treatment suggesting serum S100A8/A9 to be unsuitable to quantify skin psoriasis disease activity.Corresponding results were obtained in a recent study showing that S100A7 and S100A12 serum levels correlate with disease severity whereas S100A8/A9 does not [21].However, previous studies on serum S100A8/A9 concentration and PASI have shown contradictory results.Hansson et al found no correlation in a study including 65 patients with psoriasis and PsA [9] while Benoit et al showed a significant correlation between PASI and S100A8/A9 serum concentration [8].Naik et al found a correlation between S100A8/A9 levels, vascular inflammation and PASI [33].However, this study included patients with systemic treatment which may have affected serum S100A8/A8 levels.Due to inconsistent results in these previous studies the final role for S100A8/A9 as a psoriasis biomarker cannot be defined.Larger studies are needed to further investigate the correlation between skin inflammation, PASI and serum S100A8/A9 concentration.Notably, our results show that S100A8/A9 plaque protein levels correlate with target plaque PASI after NB-UVB but not before treatment.Target PASI in itself is a blunt scale and might not be sufficient to accurately describe the severity of a single plaque.
Limitations in the present study lie in the inclusion of patients with mainly low PASI scores (22 out of 27 had PASI �10).Patients with more extensive and severe skin disease might show higher levels of S100A8/A9 and needs to be investigated.The number of patients included in our study may have been too small assuming that the effect of NB-UVB on serum S100A8/A9 was overestimated in our sample size calculation.However, our data revealed no trend towards serum reduction despite the substantial changes seen in lesional skin.The present sample size limited further subgroup analyses and did not allow comparison neither of non-responders versus responders nor severe disease versus moderate disease.NB-UVB treatment intervention in the current study followed the standard clinical setting and in this regard, other confounding factors biasing the results cannot be completely excluded.The patients included did not use systemic drugs for psoriasis and were not allowed to have systemic inflammatory diseases such as arthritis, IBD or chronic infections.However, it cannot be completely ruled out that other diseases (e.g.hypertension, hyperlipidemia) and treatments could have affected the present results.We decided not to include a control group since the present study was conducted to analyze differences between lesional and non-lesional skin and since NB-UVB treatment theoretically could increases the risk of skin cancer in healthy individuals.
In conclusion, the current study shows for the first time a significant reduction in S100A8/ A9 skin protein levels after NB-UVB treatment while serum S100A8/A9 protein concentration remained unaffected.These results suggest that skin keratinocytes may not be the major source of elevated circulating S100A8/A9 in patients with psoriasis as previously suggested by Benoit [8].Further studies are required to identify the source of elevated serum S100A8/A9 levels, which could lead to a better understanding of systemic inflammation and comorbidities in psoriasis.The present study shows no correlation between PASI and serum S100A8/A9 concentration questioning serum S100A8/A9 as biomarker for psoriasis skin disease activity.