High fat diet exacerbates early psoriatic skin inflammation independent of obesity: Saturated fatty acids as key players

Transcription

1 Accepted Manuscript High fat diet exacerbates early psoriatic skin inflammation independent of obesity: Saturated fatty acids as key players Diana Herbert, MSc, Sandra Franz, PhD, Yulia Popkova, MSc, Ulf Anderegg, PhD, Jürgen Schiller, PhD, Katharina Schwede, MD, PhD, Axel Lorz, MD, PhD, Jan C. Simon, MD, Prof, Anja Saalbach, PhD PII: DOI: Reference: JID 1385 S X(18)31858-X /j.jid To appear in: The Journal of Investigative Dermatology Received Date: 26 January 2018 Revised Date: 12 March 2018 Accepted Date: 18 March 2018 Please cite this article as: Herbert D, Franz S, Popkova Y, Anderegg U, Schiller J, Schwede K, Lorz A, Simon JC, Saalbach A, High fat diet exacerbates early psoriatic skin inflammation independent of obesity: Saturated fatty acids as key players, The Journal of Investigative Dermatology (2018), doi: /j.jid This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.

2 High fat diet exacerbates early psoriatic skin inflammation independent of obesity: Saturated fatty acids as key players Short title: Saturated fatty acids exacerbate skin inflammation Diana Herbert, MSc 1 ; Sandra Franz, PhD 1, Yulia Popkova, MSc 2 ; Ulf Anderegg, PhD 1 ; Jürgen Schiller, PhD 2 ; Katharina Schwede, MD, PhD 1 ; Axel Lorz, MD, PhD 1 ; Jan C. Simon, MD, Prof 1 ; Anja Saalbach, PhD* 1 1 Department of Dermatology, Venerology and Allergology, University Hospital Leipzig, Germany; 2 Institute of Medical Physics and Biophysics, University Leipzig, Germany *Corresponding author: Anja Saalbach, Ph.D. Dept. Dermatology, Venerology and Allergology, Medical Faculty of the University Leipzig Johannisallee Leipzig, Germany Phone: /25860 Fax: Abbreviations Anja.Saalbach@medizin.uni-leipzig.de AT: adipose tissue; DHA: docosahexaenoic acid; FFA: free fatty acid; H&E: hematoxylin and eosin; HFD: high fat diet; LA: lauric acid; MA: myristic acid; OA: oleic acid; PA: palmitic acid; PUFA: polyunsaturated fatty acid; SA: stearic acid; SFA: saturated fatty acid; TG: triglyceride 1

3 ABSTRACT In obesity, hypertrophic adipocytes secret high amounts of adipocytokines resulting in lowgrade inflammation- amplified by infiltrating pro-inflammatory macrophages, oxidative stress, hypoxia and lipolysis. It is known, that these chronic pro-inflammatory conditions support the development of type II diabetes and cardio vascular diseases, while mechanisms of obesity-related exacerbation of inflammatory skin disorders like psoriasis are unclear. In this study, we uncover nutritional saturated fatty acids (SFAs) as major risk factors for the amplification of skin inflammation, independent of obesity-related parameters, like fat mass extension, adipocytokines and glucose homeostasis. Correlation analyses in a cohort of psoriasis vulgaris patients revealed free fatty acid (FFA)- serum-levels as the only obesity-associated parameter affecting disease severity. Studies in high fat diet-induced obese mice with psoriasiform inflammation confirmed this critical role of FFAs. Importantly, an increase of FFAs in healthy, lean mice alone was sufficient to induce an exacerbation of psoriasiform inflammation. In particular, SFAs sensitize myeloid cells to an increased inflammatory response in answer to pro-inflammatory stimuli which in turn augments the activation of keratinocytes. Consequently, reduction of nutritional SFAs alone diminished the psoriatic phenotype in obese mice. Thus, our findings may open new perspectives for adjuvant dietary measures accompanying anti-inflammatory psoriasis therapies in lean and obese patients. 2

4 INTRODUCTION Obesity has risen tremendously during the past decades (Nguyen and El-Serag 2010). It results in hyperglycaemia, dyslipidaemia, hyperinsulinaemia and a massive increase of adipose tissue (AT) associated with enhanced secretion of adipokines, chemokines and cytokines (Ahima and Flier 2000; Bray 2004; Després and Lemieux 2006; Saltiel 2016). Furthermore, levels of free fatty acids (FFAs) are increased in obese patients contributing to inflammation and insulin resistance (Boden 2008; Shi et al. 2006). Thus, in obesity, tissue resident cells and immune cells are exposed to an altered composition of metabolites, like FFAs and adipocytokines, which can modify immune cell homeostasis and function (Chiricozzi et al. 2016; Milner and Beck 2012). Thus, obesity is linked to the risk and severity of various inflammatory disorders, including type II diabetes, cardiovascular diseases, hepatic steatosis, inflammatory bowel disease, arteriosclerosis, and psoriasis (Endo et al. 2017; Haslam and James 2005). Psoriasis is a chronic inflammatory skin disease affecting about 2% 3% of the world s population (Ayala-Fontánez et al. 2016). A misguided crosstalk between inflammatory cells, like macrophages, neutrophils, antigen presenting cells, T cells, and epidermal keratinocytes is pivotal in the development and maintenance of psoriatic inflammation (Alexander and Nestle 2017; Kim and Krueger 2017; Pasparakis et al. 2014). An association of psoriasis severity with obesity has been evidenced by numerous studies in humans and mice (Azfar, Rahat S. 2008; Bremmer and Voorhees 2010; Diallo 2012; Herron et al. 2005; Kanemaru et al. 2015; Marino et al. 2004; Raychaudhuri and Gross 2000; Shipman and Millington 2011; Stelzner et al. 2016; Sterry et al. 2007; Wolk et al. 2009; Yosipovitch et al. 2007). But, until now it is poorly understood how obesity-related alterations mechanistically augment psoriasis. In this study, we performed correlational analyses in a cohort of psoriasis vulgaris patients which revealed serum FFA levels as the only metabolic, obesity-associated parameter 3

5 affecting disease severity. In a mouse model, both, obese and lean mice fed with high fat diet (HFD) developed a more severe early psoriasiform skin inflammation which was dependent on FFA levels, but independent of fat mass extension, blood glucose levels, and AT-derived mediators. SFAs, which were increased by HFD, were unraveled to sensitize myeloid cells resulting in an amplified pro-inflammatory immune response to toll-like receptor (TLR)-like stimuli which in turn augmented keratinocyte activation. Consequently, dietary reduction of SFAs dampened psoriasiform inflammation. In conclusion, we identified nutritional SFAs as major risk factors for the onset and severity of psoriatic inflammation even before an obese phenotype has developed. Thus, our data may open new options supporting conventional anti-psoriatic therapies by accompanying dietary measures.. 4

6 RESULTS Free fatty acids contribute to the exacerbation of psoriatic skin inflammation Correlation studies in a cohort of patients with psoriasis vulgaris substantiated the association of psoriasis disease severity (PASI) and obesity determined by waist to hip ratio (WHR; Table 1). Importantly, among all factors that were linked to obesity (Supplementary Table S1), including HbA1c, triglycerides (TGs), high density lipoprotein (HDL) and low density lipoprotein (LDL) cholesterol, insulin and C-peptide levels solely serum FFA levels correlated with PASI (Table 1). This finding suggests an important role of FFAs for the disease severity of psoriasis under obese conditions. To elucidate the molecular mechanisms of FFAs exacerbating psoriatic inflammation, we established a modified diet-induced obesity mouse model with psoriasiform inflammation (Stelzner et al. 2016). Mice were fed with a HFD (Supplementary Table S2) rich in SFAs (SFA high ) and low in polyunsaturated fatty acids (PUFA low ), typically seen in Western diets (Kanoski and Davidson 2011). As control, mice were fed with a low fat chow diet (SFA low - PUFA low ; Supplementary Table S2). After weeks (long-term) of HFD mice developed an obese phenotype with significantly increased subcutaneous and visceral fat mass (Figure 1a). Altered metabolism was mirrored by elevated serum insulin, fasting blood glucose, and FFA serum levels (Figure 1b, Supplementary Table S3). Psoriasiform skin inflammation was induced by one topical imiquimod (IMQ) application. Interestingly, already three days after IMQ treatment obese HFD-fed mice developed a more severe skin disease and displayed an increased epidermal thickening compared to lean controls (Figure 1c). We observed a significantly increased infiltration of myeloid cells (CD11b + ) in particular neutrophils and macrophages in skin lesions of obese mice (Figure 1d) accompanied by increased expression/secretion of pro-inflammatory mediators and chemokines (Figure 1e, f). However, mediators regulating adaptive immune response such as IL-23 and IL-17 were not altered at this early time point of inflammation (Figure 1e). 5

7 Performing a genome wide expression analysis we identified S100A7A, S100A8, S100A9, and ß-defensin as well as cytokines, and chemokines, all known as psoriasis-related genes in human, to be significantly up-regulated in lesional skin of obese compared to lean mice (Supplementary Table S4). Since myeloid cells and keratinocytes are critical players in the pathogenesis of psoriasis, the impact of these cells for the exacerbated psoriasiform skin inflammation in obese HFD-fed mice was evaluated. Myeloid cells isolated from affected skin areas from obese HFD-fed mice expressed elevated levels of mainly pro-inflammatory mediators, including interleukin- 1ß (IL-1ß), osteopontin (OP), and cyclooxygenase-2 (Cox2), whereas chemokine expression was not altered in these cells (Figure 1g). Lesional epidermal cells from obese mice showed, besides increased IL-1ß and tumor necrosis factor-α (TNFα) an upregulation of S100A8, S100A9, and chemokine expression (Figure 1h). Thus, upon IMQ treatment myeloid cells and epidermal cells contribute to increased levels of pro-inflammatory mediators and epidermal cells to elevated chemokine expression within obese, HFD-fed mice. In sum, obesity amplifies early events of skin inflammation in mice. Importantly, lean and obese mice that were not treated with IMQ showed no differences in epidermal thickness of the skin and the expression of pro-inflammatory mediators, S100 proteins and chemokines within the skin, myeloid skin cells and blood cells (Supplementary Table S3). Consequently, HFD-induced obesity per se did not alter the pro-inflammatory status of skin and immune cells, but rather renders them more susceptible to pro-inflammatory stimuli, like IMQ. In accordance to our study of the psoriatic patient cohort, correlation analyses with the IMQtreated mice revealed besides weight and body fat total FFA serum levels as the only metabolic factor correlating with inflammation severity (Table 2), but also with epidermal thickness and the expression of pro-inflammatory factors (Supplementary Table S5). In contrast, fasting blood glucose and serum insulin levels were not associated to disease severity (Table 2). Consistently, long-term HFD-fed C129/Sv mice which exhibited 6

8 increased body fat but maintained healthy fasting blood glucose under HFD (Jiang et al. 2015) also developed a more severe inflammation compared to chow controls (Supplementary Figure S1a, b). Reflecting these data we conclude that in our mouse model obesity-related amplification of skin inflammation occurs independently of blood glucose homeostasis, but is linked to body weight, fat mass and serum FFA levels. FFAs amplify psoriatic skin inflammation even before an obese phenotype has developed To distinguish between the impact of body weight, body fat and serum total FFAs mice were fed with HFD or chow control diet for just 5 weeks (short-term). Both diet groups did not exhibit any differences in the body fat mass (Figure 2a) and fasting glucose (Figure 2b), but total FFA serum levels were significantly elevated in short-term HFD-fed mice (Figure 2b). Again, application of IMQ resulted in a more severe psoriasiform inflammation in short-term HFD-fed mice (Figure 2c) which correlated with serum FFA levels (Table 2). Consistently to obese, long term-hfd fed mice, lesions of short-term HFD-fed mice showed increased epidermal thickening and CD11b + cell infiltration of neutrophils and macrophages (Figure 2c,d) as well as an elevated gene (Figure 2e) and protein expression (Figure 2f) of psoriasisrelated pro-inflammatory mediators compared to lesions of chow controls. Again, expression of mediators of adaptive immunity such as IL-23 and IL-17 was not affected by HFD at this early time point of inflammation (Figure 2e). As seen in obese HFD mice myeloid cells isolated from lesional skin of short-term HFD-fed mice displayed higher expression of proinflammatory mediators while chemokines were not altered (Figure 2g). Epidermal cells of HFD-fed mice displayed increased expression of both pro-inflammatory mediators and chemokines (Figure 2h) compared to chow controls. 7

9 These data suggest that the amplification of early psoriasiform inflammation is independent of weight and body fat content. To corroborate this we analyzed AT-derived pro-inflammatory IL-1ß, IL-6, and TNFα known to mediate a low-grade inflammation status within obese subjects (Ahima and Flier 2000; Després and Lemieux 2006; Xue et al. 2013) in the visceral fat of short-term fed mice. Interestingly, we determined no differences in the levels of these adipocytokines in short-term chow- and HFD-fed mice (Supplementary Figure S1c) indicating that these AT-derived pro-inflammatory mediators do not contribute to the increase of inflammation in those mice. In sum, experiments with short-term HFD-fed mice substantiate the critical role of FFAs in the amplification of skin inflammation. Finally, these mice develop an exacerbated initial psoriasiform skin inflammation which is solely associated with increased FFA serum levels, but independent of AT-derived mediators and other obesity-related parameters like body weight, fat mass, blood glucose and insulin. Saturated fatty acids sensitize myeloid cells resulting in an amplified immune response To get insights into the mechanisms of FFA-mediated exacerbation of skin inflammation, we first characterized the types of FFAs in the serum of short- and long-term chow- and HFD-fed mice. Besides monounsaturated FAs, like oleic acid (OA), and n-6 PUFAs, like linoleic acid (LA), we determined markedly increased levels of the pro-inflammatory SFAs palmitic acid (PA) and stearic acid (SA) in HFD-fed mice (Figure 3a, b). Since Western diet (Kanoski and Davidson 2011) and our standard HFD (Supplementary Table S2) contain high levels of proinflammatory SFAs, the mechanisms by which SFAs affect myeloid cells and keratinocytes the key players in the pathogenesis of psoriasis was analyzed. Myeloid peritoneal cells were isolated from healthy, lean, chow-fed mice and incubated with 600 µm PA or SA which resembles increased SFA concentration found in obese human subjects (Stelzner et al. 2016). In addition, effects of OA as an example of mono-unsaturated 8

10 FA were investigated. Interestingly, we observed no pro-inflammatory activation of the peritoneal cells by PA, SA or OA in the absence of any additional stimulation. In contrast, LPS stimulation upon PA- or SA-pre-incubation amplified secretion of IL-1ß, TNFα, and PGE2, respectively (Figure 3c). In contrast, OA did not affect IL-1ß and TNFα secretion. This suggests that saturated FAs such as PA and SA sensitize myeloid cells to a subsequent inflammatory stimulus resulting in an amplified inflammatory response. Next, keratinocytes were isolated from healthy, lean, chow-fed mice and incubated with 500 µm PA prior subsequent stimulation with IL-17 an important signature cytokine during psoriasis. PA forced CXCL1 expression in unstimulated and stimulated keratinocytes while expression of S100A9 and CCL2 was not affected by PA (Figure 3d). Thus, other mechanisms than a direct effect of PA have to be responsible for the augmented expression of specific chemokines and S100 proteins in the lesional epidermis of HFD-fed mice. Since a misguided crosstalk between inflammatory myeloid immune cells and epidermal keratinocytes seems to be pivotal for the development and exacerbation of psoriatic inflammation (Alexander and Nestle 2017; Boyman et al. 2007; Kim and Krueger 2017; Lowes et al. 2014), we next investigated whether SFA-treated, overreacting myeloid cells influence keratinocyte activity. To address this issue, keratinocytes from healthy, lean, chowfed mice were cultured with supernatants of PA-sensitized and LPS-stimulated myeloid peritoneal cells. Indeed, expression and secretion of S100A9 and CCL2 from keratinocytes were amplified by soluble factors produced by these myeloid cells (Figure 3e, f). Blocking IL- 1ß signaling in the keratinocytes which is pivotal during inflammation onset (Dinarello 2015) by IL1-receptor antagonist (IL-1RA) abrogated the stimulation of S100A9 and CCL2 expression and secretion by PA-sensitized myeloid cells (Figure 3e, f). Taken together, we showed that SFAs, such as PA, do not have direct pro-inflammatory effects, but sensitize myeloid cells resulting in an amplified immune response upon exogenous stimuli. Increased secretion of pro-inflammatory mediators, such as IL-1ß, of these 9

11 overreacting immune cells facilitate the expression and secretion of chemokines and S100 proteins from keratinocytes building up to a vicious circle of mutual inflammatory activation. Reduction of dietary SFAs decreases severity of psoriasiform inflammation Since our data suggest an augmentation of early psoriasiform inflammation by HFD-increased SFA serum levels, we tested whether dietary SFA reduction might mitigate exacerbated skin inflammation. For this, mice received standard HFD (SFA high -PUFA low ) for 2.5 weeks which was then changed to a HFD with reduced SFA but enriched PUFA content (HFD-SFA low -PUFA high ; Supplementary Table S2) for additional 2.5 weeks. After 5 weeks weight was not significantly different between standard HFD-fed, chow-fed mice, and mice with diet change to HFD- SFA low -PUFA high (Figure 4a). But, serum FFAs were increased in standard HFD-fed and mice with diet change to HFD-SFA low -PUFA high compared to chow diet (Figure 4b). Analysis of specific FFAs revealed that HFD-SFA high -PUFA low resulted in high serum levels of SFAs (lauric acid, myristic acid, PA, SA) as well as OA while the switch from HFD-SFA high - PUFA low to HFD-SFA low -PUFA high strongly reduced serum levels of these pro-inflammatory FFAs in favor of increased levels of the anti-inflammatory n-3 PUFA docosahexaenoic acid (DHA) (Figure 4b). After IMQ-treatment mice with dietary change to HFD-SFA low -PUFA high showed significantly less psoriasiform skin inflammation compared to HFD-SFA high - PUFA low -fed mice (Figure 4c). Likewise, epidermal thickness and immune cell infiltration of myeloid cells (CD11b + ) was decreased in skin lesions of mice fed with HFD-SFA low - PUFA high (Figure 4c). Moreover, expression of pro-inflammatory mediators, S100 proteins, and chemokines within the lesional skin of these mice was diminished to levels found in chow-fed controls (Figure 4d). These findings prove that composition of FFAs within the diet affects the outcome of proinflammatory processes during the onset of skin inflammation. Next, we aimed to 10

12 discriminate whether the reduction of dietary SFAs alone or the enrichment of PUFAs was responsible for the improvement of psoriasiform inflammation. In this approach we used obese mice to mimic the clinical situation of obese psoriatic patients. To this end, obese mice fed with standard HFD (SFA high -PUFA low ) for 25 weeks were switched to low fat chow diet without PUFA supplementation (SFA low -PUFA low ; Supplementary Table S2) for one week. Mice receiving this dietary change were still obese reflected by increased weight (Figure 4e) but displayed significantly reduced total serum FFA levels, in particular of proinflammatory SFAs like PA and SA (Figure 4f). This reduction of dietary SFAs for one week was sufficient to reduce psoriasiform inflammation compared to permanently HFD-fed mice (Figure 4g) as shown on the low expression level of pro-inflammatory mediators, S100 proteins and chemokines within the lesions of mice after diet change (Figure 4h). These data show the importance of dietary SFA reduction, either with or without PUFA supplementation, for attenuation of early psoriasiform skin inflammation in lean as well as in obese subjects. 11

13 Discussion Obesity exacerbates a wide variety of inflammatory diseases (Pietro A. Tataranni and Emilio Ortega 2005; Armstrong et al. 2012; Henseler and Christophers 1995; Milner and Beck 2012; Nakamizo et al. 2016; Naldi et al. 2014; Panagiotakos et al. 2005; Wolk et al. 2009). Due to today s human lifestyle resulting in persistently rising prevalence of obesity the problem of obesity-mediated exacerbation of inflammatory diseases is strikingly increasing. Consistently, in the current study HFD-induced obese mice developed an augmented psoriasiform inflammation which is in line with other studies (Kanemaru et al. 2015; Nakamizo et al. 2016; Stelzner et al. 2016). However, the causal relationship by which obesity strengthens this inflammatory response are still poorly understood. In this study, we identified mechanisms of obesity mediated exacerbation of psoriasiform skin inflammation. Nutritional SFAs were figured out as major risk factors for the amplification of skin inflammation, independent of obesity-related parameters, like fat mass extension, adipocytokines and glucose homeostasis. Consistently to previous human studies (Henseler and Christophers 1995; Naldi et al. 2014), in the present study WHR positively correlated with disease severity in a cohort of plaque type psoriasis patients. Interestingly, from all other crucial obesity-related factors only total serum FFA levels were also significantly associated to disease severity, suggesting an important role of these metabolites in the inflammatory processes. In accordance, in our mouse model of psoriasiform inflammation in obese mice inflammation severity also correlated only with FFA serum levels among all metabolic parameters. Moreover, five weeks (short-term) high fat nutrition did not increase weight, body fat and fasting glucose but elevated total FFA levels. Worsened early psoriasiform inflammation in these short-term HFD-fed mice underlines the pivotal regulatory role of FFAs in skin inflammation. Importantly, our data indicate that high fat nutrition exacerbates psoriasiform 12

14 inflammation even before an obese phenotype, with increased weight, fat content, disturbed glucose homeostasis and increased secretion of pro-inflammatory adipocytokines, has developed. Our data are underlined by the observation that Western diet feeding resulted in the priming of the immune system that triggers enhanced innate immune responses toward an inflammatory trigger in an experimental model of atherosclerosis (Christ et al. 2018). Several mechanisms have been reported to contribute to obesity-mediated exacerbation of psoriasiform inflammation. Nakamizo et al. (2017) described an increase of γδ T cellrecruiting chemokines and accumulation of IL-17A-producing γδ T cells in the skin of obese untreated mice which leads to an exacerbation of psoriatic inflammation. In accordance to Zhang et al. (2015) we did not find elevated levels of T-cells (TCs) in skin of untreated obese mice and thus TC-derived mediators such as IL-17were not differently expressed. Another study showed that long high fat feeding over nine month promotes the accumulation of specific CD11c + macrophages in skin in an E-FABP dependent manner. E-FABP primes skin to induce IL-1β and IL-18 signaling in response to environmental stimuli (Zhang et al. 2015). However, after 5 month of HFD we did not find elevated amounts of myeloid cells. Attenuation of psoriasiform inflammation in ob/ob obese mice revealed a central role of leptin in linking obesity and inflammation (Johnston et al. 2008; Nakamizo et al. 2017). Thus, leptin or other adipokines might contribute to amplification of psoriasiform inflammation under obese conditions. Since expression of these adipokine increases with body fat gain it seems unlikely that these mediators are involved in the exacerbated psoriasiform inflammation in short, lean HFD-fed mice (Wasim et al. 2015). Taken together, our study identified high fat nutrition mediated elevation of FFA levels as one important factor amplifying psoriasiform inflammation. Thus, several mechanisms contribute to both HFD- and obesity-mediated exacerbation of psoriasiform inflammation. Elevated levels of SFAs in HFD-fed mice and the reported pro-inflammatory effect of SFAs (Clark and Kupper 2006; Teng et al. 2014) supposed a pivotal role of SFAs in HFD-induced 13

15 exacerbation of psoriasiform inflammation. Indeed, excessive concentrations of PA caused sensitization of myeloid peritoneal cells and resulted in an increased expression and secretion of pro-inflammatory mediators upon LPS stimulation. Consistently, we observed an increased pro-inflammatory activation of myeloid cells within the lesional skin of both short-term HFDfed lean mice and obese mice. FFA can affect inflammation through a variety of mechanisms (Calder PC. 2011). They can bind to cell surface and intracellular receptors/sensors that control inflammatory cell signaling and gene expression. Several reports demonstrate a pro-inflammatory response in macrophages by SFA depending on TLR activation (Schwartz et al. 2010; Shi et al. 2006; Chait et al. 2010, Snodgrass et al. 2013). The PA-mediated pro-inflammatory stimulation is associated with the activation of p38 and JNK kinases, inflammasome activation and the NFκB pathway (Suganami et al. 2007, Snodgrass et al. 2013, Haversen et al. 2009). However, we did not observe any direct FFA-mediated activation of myeloid cells, neither by PA or SA challenge in vitro nor in the skin of untreated, long-term HFD-fed mice in the absence of additional danger signals. In accordance with our in vitro data, Wen et al (2014) and Chang et al. (2012) also described that treatment of RAW264.7 cells or bone marrow-derived macrophages with SFA in the absence of LPS stimulation has no effect. FFA can change the membrane composition and thus modify membrane fluidity, lipid raft formation, and cell signaling (Levental et al. 2017). SFA are able to amplify the pro-inflammatory response of myeloid cells. Schwartz et al (2010) describe in addition to a direct pro-inflammatory effect of SFA an amplification of the LPS-induced proinflammatory response of monocytes by SFA. Uptake of SFA leads to enhanced ceramide generation, which in turn activates PKC-ζ and MAPK resulting in an increased IL-6 and IL-8 secretion upon LPS stimulation (Schwartz et al. 2010). Consistent to several studies we observed an amplification of the pro-inflammatory response of myeloid cells by SFA (Schwartz et al. 2010; Chang et al. 2012; Wen et al. 2014). 14

16 Although epidermal layers of lesional skin of mice with increased serum FFA levels displayed higher chemokine and S100 protein expression, PA did not directly alter the expression of S100 proteins or chemokines except CXCL1 in keratinocytes. But, we showed that amplified immune responses of myeloid cells in the presence of PA subsequently resulted in an IL-1ß-depending upregulation of S100A9 and CCL2 production in keratinocytes. We hypothesize that this amplified keratinocyte activation might contribute to the increased epidermal thickening and the strong recruitment of myeloid cells and build up the vicious cycle of inflammation within the IMQ-induced lesions of HFD-fed mice. Since previous reports described pro-inflammatory effects of SFAs and omega-6 PUFAs, and anti-inflammatory properties of omega-3 PUFAs (Georgiadi and Kersten 2012) we asked whether modulation of dietary FFAs is beneficial for improvement of psoriasiform inflammation. In our study the concurrent reduction of SFA levels and enrichment of omega-3 PUFAs in our mouse model significantly reduced the severity of initial psoriasiform inflammation. There is a controversial discussion on the impact and efficacy of PUFA supplementation as a therapeutic measure in inflammatory disorders such as psoriasis (Mari et al. 2017; Mayser et al. 2002; Park et al. 2016; Qin et al. 2014; Teng et al. 2014). Interestingly, our data indicate that a dietary reduction of SFA alone without PUFA supplementation was sufficient to improve psoriasiform inflammation in obese mice independent of weight loss which was not shown before. In summary, this study identifies elevated dietary SFAs, which are typical for Western diets, as one important risk factor for the onset and severity of early psoriatic inflammation in obese and lean subjects independently of obesity-associated parameters. Importantly, reduction of dietary SFAs with or without PUFA supplementation improves psoriasiform inflammation independently of weight and fat content. These findings may open new translational perspectives for adjuvant dietary measures supporting conventional anti-inflammatory therapies in lean and obese patients with severe psoriatic inflammation. 15

17 METHODS Patients Serum was collected from 72 (32 males, 40 females) overnight-fasted patients with psoriasis vulgaris after written informed patient consent. Metabolic parameters (Supplementary Table S1) and Psoriasis Area and Severity Index (PASI) were analyzed. The study was approved by the local ethics committee (# ethic vote: 332/13-ff). Mouse studies Four to five weeks old male C57BL/6J or C129/VS mice were fed with high fat diet (HFD; EF R/M D12331 diet modified by Surwit), chow diet or PUFA enriched HFD (Surwit, ssniff, Soest, Germany; Supplementary Table S2). Body fat percentage was determined using nuclear magnetic resonance technology with EchoMRI700 instrument (Echo Medical Systems, Houston, TX, USA) and fasting glucose was measured after 16 h starvation. A psoriasiform skin inflammation was induced by topical application of 100 mg imiquimod (IMQ, Aldara, Meda GmbH, Wiesbaden, Germany) on the shaved back of C57/BL6 or C129/Sv mice. A cumulative severity score including erythema, scaling and affected area (0: none; 1: mild; 2: moderate; 3: severe; 4: very severe) was determined by two independent researchers All animal experiments were performed according to institutional and state guidelines. The Committee on Animal Welfare of Saxony (Germany) approved animal protocols used in this study (TVV65/13, TVV03/16). 16

18 For information on PCR, genome-wide expression analysis, ELISA, cell isolation, histology, Western Blot, detection of FFA and flow cytometry see Supplementary Materials and Methods. Statistics For correlational and statistical analyses GraphPad Prism7 (GraphPad Software, Inc., La Jolla, USA) was used. Data analysis was performed using Mann-Whitney rank-sum test. Correlational studies were performed with Spearman correlations. Values of p less than 0.05 were considered to be significant. The different degrees of significance were indicated as followed: * p < 0.05; ** p < 0.01; *** p <

19 ACKNOWLEDGEMENT We thank Tobias Janik, Heidi Gedicke, Eva Böge, and Danny Gutknecht for their excellent technical assistance. This work was supported by Deutscher Psoriasis Bund e.v. and PsoNet Leipzig/Westsachsen and by the Deutsche Forschungsgemeinschaft (to AS: SA863/2-3, to UA: SFB Transregio 67, project B4, and AN276/6-1, to AS and JCS: SFB1052 project B5). CONFLICT OF INTEREST The authors have declared that no conflict of interest exists. 18

20 REFERENCES Ahima RS, Flier JS. Adipose Tissue as an Endocrine Organ. 2000;11(8): Ahima RS, Mitchell A. Lazar. The Health Risk of Obesity Better Metrics Imperative. Science (80-. ). 2013;341:856 8 Alexander H, Nestle FO. Pathogenesis and immunotherapy in cutaneous psoriasis. Curr. Opin. Rheumatol. 2017;29(1):71 8 Armstrong A, Harskamp CT, Armstrong EJ. The association between psoriasis and obesity: a systematic review and meta-analysis of observational studies. Nutr. Diabetes. Nature Publishing Group; 2012;2(12):e54 Ayala-Fontánez N, Soler DC, McCormick TS. Current knowledge on psoriasis and autoimmune diseases. Psoriasis Targets Ther. Dove Press; 2016;6:7 32 Azfar, Rahat S. GJ. Psoriasis and Metabolic Disease: Epidemiology and Pathophysiology. Curr Opin Rheumatol. 2008;20(4): Boden G. Obesity and FFAs. Endocrinol Metab Clin North Am. 2008;37(3):635 ix Boß M, Newbatt Y, Gupta S, Collins I, Brüne B, Namgaladze D. AMPK-independent inhibition of human macrophage ER stress response by AICAR. Nat. Publ. Gr. Nature Publishing Group; 2016;(April):1 10 Boyman O, Conrad C, Tonel G, Gilliet M, Nestle FO. The pathogenic role of tissue-resident immune cells in psoriasis. Trends Immunol. 2007;28(2):51 7 Bray GA. Medical Consequences of Obesity. J. Clin. Endocrinol. Metab. 2004;89(6): Bremmer S, Voorhees A Van. Obesity and psoriasis: from the Medical Board of the National Psoriasis Foundation. J Am Acad Dermatal. American Academy of Dermatology, Inc.; 2010;63(6): Calder PC. Omega-3 polyunsaturated fatty acids and inflammatory processes: nutrition or pharmacology? Br J Clin Pharmacol. 2013;75(3): Chang CF, Chau YP, Kung HN, Lu KS. The lipopolysaccharide-induced pro-inflammatory 19

21 response in RAW264.7 cells is attenuated by an unsaturated fatty acid-bovine serum albumin complex and enhanced by a saturated fatty acid-bovine serum albumin complex. Inflamm. Res. 2012;61(2): Chait A, Kim F. Saturated fatty acids and inflammation: who pays the toll? Arterioscler Thromb Vasc Biol. 2010;30(4): Chiricozzi A, Raimondo A, Lembo S, Fausti F, Dini V, Costanzo A, et al. Crosstalk between skin inflammation and adipose tissue-derived products: pathogenic evidence linking psoriasis to increased adiposity. Expert Rev Clin Immunol. 2016;12(12): Christ A, Günther P, Lauterbach MAR, Duewell P, Biswas D, Pelka K, et al. Western Diet Triggers NLRP3-Dependent Innate Immune Reprogramming. Cell. 2018;172(1 2): e14 Clark RA, Kupper TS. Misbehaving macrophages in the pathogenesis of psoriasis. J. Clin. Invest. 2006;116(8): Després J, Lemieux I. Abdominal obesity and metabolic syndrome. Nature. 2006; 444 (December):881 7 Diallo M. Psoriasis Epidemiology. J. Clin. Case Reports. 2012;2(8) Dinarello C. Interleukin-1 in the pathogenesis and treatment of inflammatory diseases. Blood. 2015;117(14): Endo Y, Yokote K, Nakayama T. The obesity-related pathology and Th17 cells. Cell. Mol. Life Sci. Springer International Publishing; 2017;74(7): van der Fits L, Mourits S, Voerman JSA, Kant M, Boon L, Laman JD, et al. Imiquimod- Induced Psoriasis-Like Skin Inflammation in Mice Is Mediated via the IL-23/IL-17 Axis. J. Immunol. 2009;182(9): Furukawa S, Fujita T, Shumabukuro M, Iwaki M, Yamada Y, Makajima Y, et al. Increased oxidative stress in obesity and its impact on metabolic syndrome. J. Clin. Invest. 2004; 114 (12):

22 Georgiadi A, Kersten S. Mechanisms of Gene Regulation by Fatty Acids. Adv. Nutr. 2012; 3(2): Haslam DW, James WPT. Obesity. Lancet. 2005;366(9492): Haversen L, Danielsson KN, Fogelstrand L, Wiklund O: Induction of proinflammatory cytokines by long-chain saturated fatty acids in human macrophages. Atherosclerosis 2009;202(2): Henseler T, Christophers E. Disease concomitance in psoriasis. J. Am. Acad. Dermatol. 1995;32(6):982 6 Herron MD, Hinckley M, Hoffman MS, Papenfuss J, Hansen CB, Callis KP, et al. Impact of Obesity and Smoking on Psoriasis Presentation and Management. Arch. Dermatol. 2005;141(12):1 80 Jiang X, Yang L, Luo Y. Animal Models of Diabetic Retinopathy. Curr. Eye Res. 2015;40(8): Johnston A, Arnadottir S, Gudjonsson JE, Aphale A, Sigmarsdottir AA, Gunnarsson SI, et al. Obesity in psoriasis: Leptin and resistin as mediators of cutaneous inflammation. Br. J. Dermatol. 2008;159(2): Kanemaru K, Matsuyuki A, Nakamura Y, Fukami K. Obesity exacerbates imiquimod-induced psoriasis-like epidermal hyperplasia and interleukin-17 and interleukin-22 production in mice. Exp. Dermatol. 2015;24(6): Kanoski SE, Davidson TL. Western Diet Consumption and Cognitive Impairment: Links to Hippocampal Dysfunction and Obesity. Physiol Behav. 2011;103(1):59 68 Kim J, Krueger JG. Highly Effective New Treatments for Psoriasis Target the IL-23/Type 17 T Cell Autoimmune Axis. Annu. Rev. Med. 2017;68(1): Levental KR, Surma MA, Skinkle AD, Lorent JH, Zhou Y, Klose C, Chang JT, Hancock JF, Levental I. omega-3 polyunsaturated fatty acids direct differentiation of the membrane 21

23 phenotype in mesenchymal stem cells to potentiate osteogenesis. Sci Adv 2017;3(11):eaao1193. Lowes MA, Suárez-Farinas M, Krueger JG. Immunology of Psoriasis. Annu. Rev. Immunol. 2014;32: Mari NL, Simão ANC, Dichi I. N-3 Polyunsaturated Fatty Acids Supplementation in Psoriasis: a Review. Nutrire. Nutrire; 2017;42(1):5 Marino MG, Carboni I, De Felice C, Maurici M, Maccari F, Franco E. Risk factors for psoriasis: a retrospective study on 501 outpatients clinical records. Ann Ig. 2004;16(6):753 8 Mayser P, Grimm H, Grimminger F. n-3 Fatty Acids in Psoriasis. Br. J. Nutr. 2002;87:S77-82 Milner JJ, Beck M a. The impact of obesity on the immune response to infection. Proc. Nutr. Soc. 2012;71(2): Minihane AM, Vinoy S, Russell WR, Baka A, Roche HM, Tuohy KM, et al. Low-grade inflammation, diet composition and health: current research evidence and its translation. Br. J. Nutr. 2015;114(7): Mohamed-Ali V, Pinkney JH, Coppack SW, Hoffmeister D, Pallua N, Mohamed-Ali V, et al. Adipose tissue as an endocrine and paracrine organ. Int. J. Obes. Relat. Metab. Disord. 1998;22(12): Nakamizo S, Honda T, Adachi A, Nagatake T, Kunisawa J, Kitoh A, Otsuka A, Dainichi T, Nomura T, Ginhoux F, Ikuta K, Egawa G, KabashimaK. High fat diet exacerbates murine psoriatic dermatitis by increasing the number of IL-17-producing γδ T cells. Scientific Reports, 2017; 7(1): Naldi L, Conti A, Cazzaniga S, Patrizi A, Pazzaglia M, Lanzoni A, et al. Diet and physical exercise in psoriasis: A randomized controlled trial. Br. J. Dermatol. 2014;170(3): Nguyen DM, El-Serag HB. The Epidemiology of Obesity. Gastroenterol Clin North Am. 2010;39(1):1 7 Nickoloff BJ. Cracking the cytokine code in psoriasis. Nat. Med. 2007;13(3):

24 Panagiotakos DB, Pitsavos C, Yannakoulia M, Chrysohoou C, Stefanadis C. The implication of obesity and central fat on markers of chronic inflammation : The ATTICA study. 2005;183: Park MK, Li W, Paek SY, Li X, Wu S, Li T, et al. Consumption of polyunsaturated fatty acids and risk of incident psoriasis and psoriatic arthritis from the Nurses Health Study II. Br. J. Dermatol. 2016; Pasparakis M, Haase I, Nestle FO. Mechanisms regulating skin immunity and inflammation. Nat. Rev. Immunol. 2014;14(5): Qin S, Wen J, Bai XC, Chen TY, Zheng RC, Zhou G Bin, et al. Endogenous n-3 polyunsaturated fatty acids protect against imiquimod-induced psoriasis-like inflammation via the IL-17/IL-23 axis. Mol. Med. Rep. 2014;9(6): Raychaudhuri SP, Gross J. A comparative study of pediatric onset psoriasis with adult onset psoriasis. Pediatr. Dermatol. 2000;17(3):174 8 Saltiel AR. New therapeutic approaches for the treatment of obesity. Sci. Transl. Med. 2016;8(323):323rv2 Schwartz EA, Zhang WY, Karnik SK, Borwege S, Anand VR, Laine PS, et al. Nutrient modification of the innate immune response: A novel mechanism by which saturated fatty acids greatly amplify monocyte inflammation. Arterioscler. Thromb. Vasc. Biol. 2010;30(4):802 8 Shi H, Kokoeva M V, Inouye K, Tzameli I, Yin H, Flier JS. TLR4 links innate immunity and fatty acid induced insulin resistance. J. Clin. Invest. 2006;116(11): Shipman AR, Millington GWM. Obesity and the skin. Br. J. Dermatol. 2011;165(4): Stelzner K, Herbert D, Popkova Y, Lorz A, Schiller J, Gericke M, et al. Free fatty acids sensitize dendritic cells to amplify TH1/TH17-immune responses. Eur. J. Immunol. 2016;46(8):

25 Snodgrass RG, Huang S, Choi IW, Rutledge JC, Hwang DH: Inflammasome-mediated secretion of IL-1beta in human monocytes through TLR2 activation; modulation by dietary fatty acids. J Immunol 2013;191(8): Sterry W, Strober BE, Menter A. Obesity in psoriasis: The metabolic, clinical and therapeutic implications. Report of an interdisciplinary conference and review. Br. J. Dermatol. 2007;157(4): Tataranni PA, Ortega E. Does an Adipokine-induced activation of immune system mediate the effect of overnutrition on Type 2 Diabetes? Diabetes. 2005;54(1): Teng K-T, Chang C-Y, Chang LF, Nesaretnam K. Modulation of obesity-induced inflammation by dietary fats: mechanisms and clinical evidence. Nutr. J. Nutrition Journal; 2014;13(1):12 Wasim M. Role of leptin in obesity. J. Obes. Weight Loss Ther. 2015;5(2) Wen H, Gris D, Lei Y, Jha S, Zhang L, Huang MT, Brickey WJ, Ting JP. Fatty acid-induced NLRP3-ASC inflammasome activation interferes with insulin signaling. Nat Immunol 2011;12(5): Wolk K, Mallbris L, Larsson P, Rosenblad A, Vingård E, Ståhle M. Excessive body weight and smoking associates with a high risk of onset of plaque psoriasis. Acta Derm. Venereol. 2009;89(5):492 7 Xu X, Grijalva A, Skowronski A, Van Eijk M, Serlie MJ, Ferrante AW. Obesity activates a program of lysosomal-dependent lipid metabolism in adipose tissue macrophages independently of classic activation. Cell Metab. Elsevier Inc.; 2013;18(6): Xue K, Liu H, Jian Q, Liu B, Zhu D, Zhang M, et al. Leptin induces secretion of proinflammatory cytokines by human keratinocytes in vitro - a possible reason for increased severity of psoriasis in patients with a high body mass index. Exp. Dermatol. 2013;22(6): Yosipovitch G, DeVore A, Dawn A. Obesity and the skin: Skin physiology and skin 24

26 manifestations of obesity. J. Am. Acad. Dermatol. 2007;56(6): Zhang Y, Li Q, Rao E, Sun Y, Grossmann ME, Morris RJ, et al. Epidermal Fatty Acid Binding Protein Promotes Skin Inflammation Induced by High-Fat Diet. Immunity. Elsevier Inc.; 2015;42(5):

27 FIGURE LEGENDS Figure 1. HFD-induced obesity amplifies psoriasiform skin inflammation in mice. C57BL/6 mice were fed with high fat diet (HFD) or chow diet for weeks (long-term). (a) Weight of subcutaneous (sc) and visceral (vis) fat, (b) fasting glucose, and serum levels of free fatty acids (FFA) were determined. (c-h) Psoriasiform skin inflammation was induced by topical application of imiquimod on the shaved back. Inflammation was analyzed after three days. (c) Severity score, epidermal thickness quantified in hematoxylin and eosin stainings. (d) Quantification of myeloid cell (CD11b + ), neutrophile (Ly6G+) and macrophage (F4/80+) infiltration by flowcytometry. (e) Relative gene expression (RS36 normalized) within lesional skin was analyzed by qpcr and (f) protein levels by multiplexing. (g) Myeloid cells and (h) epidermal cells were isolated from lesional skin and qpcr was performed to analyze relative gene expression (RS36 normalized). IL: interleukin; TNF: tumor necrosis factor; OP: osteopontin; S: S100 protein; CCL: CC-chemokine ligand; Cox; cyclooxygenase; CXCL: chemokine (C-X-C motif) ligand. Each dot represents one mouse. ns: not significant; * p < 0.05; ** p < 0.01; *** p < Figure 2. HFD amplifies psoriasiform skin inflammation independent of an obese phenotype. C57BL/6 mice were fed with high fat diet (HFD) or chow diet for 5 weeks (short-term). (a) Weight of subcutaneous (sc) and visceral (vis) fat, (b) fasting glucose, and serum levels of free fatty acids (FFA) were determined. (c-h) Psoriasiform skin inflammation was induced by topical application of imiquimod on the shaved back. Inflammation was analyzed after three days. (c) Severity score, epidermal thickness quantified in hematoxylin and eosin stainings. (d) Quantification of myeloid cell (CD11b + ), neutrophile (Ly6G + ) and macrophage (F4/80 + ) infiltration by flowcytometry (e) Relative gene expression (RS36 normalized) within lesional skin was analyzed by qpcr and (f) protein levels by multiplexing. (g) Myeloid cells and (h) 26

28 epidermal cells were isolated from lesional skin and qpcr was performed to analyze relative gene expression (RS36 normalized). IL: interleukin; TNF: tumor necrosis factor; OP: osteopontin; S: S100 protein; CCL: CC-chemokine ligand; Cox; cyclooxygenase; CXCL: chemokine (C-X-C motif) ligand. Each dot represents one mouse. ns: not significant; * p < 0.05; ** p < 0.01; *** p < Figure 3. PA sensitization of myeloid cells results in an amplified immune response upon stimulation resulting in augmented keratinocyte activation. (a) Mice were fed with HFD or chow diet for 5 weeks (short-term) or (b) weeks (longterm) and serum free fatty acid (FFA) profile was determined. PA: palmitic acid; SA: stearic acid; OA: oleic acid; LA: linolic acid. (c) Myeloid peritoneal cells from lean chow-fed mice were incubated with 600 µm palmitic acid (PA), stearic acid (SA) or oleic acid (OA) -bovine serum albumin (BSA) complexes or BSA alone followed by LPS (lipopolysaccharides) stimulation. Cells without LPS stimulation served as controls (ctr). Protein levels within supernatants were detected by ELISA. mean ± SD; n=10 mice. (d) Keratinocytes from lean chow-fed mice were incubated with 500 µm PA (PA500)-BSA complexes or BSA alone followed by interleukin-17 (IL-17) stimulation. Relative RNA expression levels (RS36 normalized) was detected by PCR. n=4 mice. mean ± SD. (e+f) Keratinocytes from lean chow-fed mice were co-cultured with supernatants of PA/BSA/LPS stimulated myeloid cells or controls. IL-1ß was neutralized by pre-incubation with IL-receptor agonist (IL1RA). (e) Relative RNA expression and (f) protein expression was detected by RT-PCR and ELISA. mean ± SD; n=7 mice. AU: arbitrary units; Cox: cyclooxygenase; PG: prostaglandin; S: S100 protein; CCL: CC-chemokine ligand; CXCL: chemokine (C-X-C motif) ligand. * p < 0.05; ** p < 0.01; *** p <

29 Figure 4. Severity of psoriatic inflammation decreases by reduction of dietary SFAs. Mice were fed with standard high fat diet high in saturated and low in polyunsaturated fatty acids (red, HFD-SFA high -PUFA low ) or chow control diet (blue). After 2.5 weeks HFD-SFA high - PUFA low diet was switched to HFD-SFA low -PUFA high (beige) for further 2.5 weeks. Mice were then treated with imiquimod and analyzed on d3. (a) Weight and (b) serum free fatty acids (FFA) were determined. (c) Severity score, epidermal thickness within lesions quantified in hematoxylin and eosin stainings, and immune cell infiltrate within lesions quantified by flowcytometry. (d) Relative RNA expression (RS36 normalized) within lesional skin quantified by qpcr. (e-h) Mice were fed with high fat diet (HFD) (obese; red) or chow diet (lean; blue) for weeks. HFD was switched to chow control diet, low in saturated fatty acids (SFA low ), for 7 days (grey). Mice were treated with imiquimod and analyzed on d3. (e) Weight and (f) FFAs were determined. (g) Severity score on d3 and (h) relative RNA expression (RS36 normalized) within lesional skin quantified by qpcr. LAA: lauric acid; MA: myristic acid; PA: palmitic acid; SA: stearic acid; OA: oleic acid; DHA: docosahexaenoic acid. AU: arbitrary units; IL: interleukin; TNF: tumor necrosis factor; OP: osteopontin; Cox: cyclooxygenase; S: 100 protein; CCL: CC-chemokine ligand; CXCL: chemokine (C-X-C motif) ligand. Each dot represents one mouse. ns: not significant. * p < 0.05; ** p < 0.01; *** p <

30 Table 1. Psoriatic inflammation severity correlates with FFA levels in patients with plaque type psoriasis. PASI 1 r p value WHR FFA [mm] HbA1c [mmol/mol] Triglycerides [mmol/l] Cholesterol [mmol/l] HDL-C [mmol/l] LDL-C [mmol/l] Insulin [pmol/l] C-peptide [pmol/l] n= 72 patients of plaque type psoriasis without psoriasis pustulosa and system therapy; PASI range: ; r= Spearman correlation coefficient; p < 0.05 means significant 2 WHR: waist to hip ratio; WHR range: PASI: Psoriasis Area and Severity Index; FFA: free fatty acids; HDL-C: high density lipoprotein cholesterol; LDL-C: low density lipoprotein cholesterol. 29

31 Table 2: In mice severity of psoriatic inflammation correlates with weight, body fat and serum FFA levels. d3 severity [score] 2 long-term HFD short-term HFD r p value r p value body composition 1 weight [g] , body fat [%] subcutaneous fat [g] visceral fat [g] metabolism 1 serum FFAs [mm] < fasting glucose [mmol/l] ns -0, serum insulin [ng/ml] ns 1 weight, body fat percentage, fasting glucose and insulin levels were measured before imiquimod treatment. subcutaneous and visceral fat depot mass and FFA levels were determined on d3 upon imiquimod treatment. 2 n= 15 per group (long-term chow and high fat diet-fed mice); r: Spearman correlation coefficient; ns: not significant; p < 0.05 means significant. FFA: free fatty acid 30

32 AC C EP TE D M AN U SC RI PT

33 AC C EP TE D M AN U SC RI PT

34

35 AC C EP TE D M AN U SC RI PT

36 本文献由 学霸图书馆 - 文献云下载 收集自网络, 仅供学习交流使用 学霸图书馆 ( 是一个 整合众多图书馆数据库资源, 提供一站式文献检索和下载服务 的 24 小时在线不限 IP 图书馆 图书馆致力于便利 促进学习与科研, 提供最强文献下载服务 图书馆导航 : 图书馆首页文献云下载图书馆入口外文数据库大全疑难文献辅助工具