Resolvin D3 Is Dysregulated in Arthritis and Reduces Arthritic Hildur H. Arnardottir, Jesmond Dalli, Lucy V. Norling, Romain A. Colas, Mauro Perretti and Charles N. Serhan This information is current as of September 27, 2021. J Immunol published online 17 August 2016 http://www.jimmunol.org/content/early/2016/08/17/jimmun ol.1502268 Downloaded from

Supplementary http://www.jimmunol.org/content/suppl/2016/08/17/jimmunol.150226 Material 8.DCSupplemental

Why The JI? Submit online. http://www.jimmunol.org/

• Rapid Reviews! 30 days* from submission to initial decision

• No Triage! Every submission reviewed by practicing scientists

• Fast Publication! 4 weeks from acceptance to publication

*average by guest on September 27, 2021

Subscription Information about subscribing to The Journal of Immunology is online at: http://jimmunol.org/subscription Permissions Submit copyright permission requests at: http://www.aai.org/About/Publications/JI/copyright.html Email Alerts Receive free email-alerts when new articles cite this article. Sign up at: http://jimmunol.org/alerts

The Journal of Immunology is published twice each month by The American Association of Immunologists, Inc., 1451 Rockville Pike, Suite 650, Rockville, MD 20852 Copyright © 2016 by The American Association of Immunologists, Inc. All rights reserved. Print ISSN: 0022-1767 Online ISSN: 1550-6606. Published August 17, 2016, doi:10.4049/jimmunol.1502268 The Journal of Immunology

Resolvin D3 Is Dysregulated in Arthritis and Reduces Arthritic Inflammation

Hildur H. Arnardottir,* Jesmond Dalli,*,1 Lucy V. Norling,† Romain A. Colas,*,1 Mauro Perretti,† and Charles N. Serhan*

Uncontrolled inflammation is a unifying component of many chronic inflammatory , such as arthritis. Resolvins (Rvs) are a new family from the endogenous specialized proresolving mediators (SPMs) that actively stimulate resolution of inflam- mation. In this study, using lipid mediator metabololipidomics with murine joints we found a temporal regulation of endog- enous SPMs during self-resolving inflammatory arthritis. The SPMs present in self-resolving arthritic joints include the D-series Rvs, for example, RvD1, RvD2, RvD3, and RvD4. Of note, RvD3 levels were reduced in inflamed joints from mice with delayed-resolving arthritis when compared with self-resolving inflammatory arthritis. RvD3 was also reduced in serum

from rheumatoid arthritis patients compared with healthy controls. RvD3 administration reduced joint leukocytes as well as Downloaded from paw joint eicosanoids, clinical scores, and edema. Taken together, these findings provide evidence for dysregulated endogenous RvD3 levels in inflamed paw joints and its potent actions in reducing murine arthritis. The Journal of Immunology, 2016, 197: 000–000.

cute inflammation is a host-protective response against mediators (e.g., leukotrienes [LTs], PGs, and cytokines) (1), which

injury or infection that when uncontrolled can lead to mount the initiation and propagation of the initial acute inflam- http://www.jimmunol.org/ A chronic inflammation, scarring, and eventual loss of matory response (3, 4). This novel genus of bioactive lipid me- tissue function (1, 2). The ideal outcome of acute inflammation is diators (LMs) biosynthesized from essential fatty acids (EFAs) in complete resolution, an active process now known to be orches- resolving inflammatory exudates, includes lipoxins (LXs), resol- trated by specialized proresolving mediators (SPMs) that limit fur- vins (Rvs), protectins (PDs), and maresins (MaRs) (1, 5). They ther neutrophil recruitment and counterregulate proinflammatory actively promote catabasis via potent anti-inflammatory and pro- resolving actions (e.g., stimulating leukocyte uptake of apoptotic cells [efferocytosis] and bacterial clearance [reviewed in Ref. 1]), *Center for Experimental Therapeutics and Reperfusion Injury, Department of Anesthesiology, Perioperative and Pain Medicine, Harvard Institutes of Medicine, providing a molecular basis for novel therapeutic approaches via Brigham and Women’s Hospital and Harvard Medical School, Boston, MA 02115; promoting resolution (1). by guest on September 27, 2021 † and William Harvey Research Institute, Barts and London School of Medicine, Arthritis is a significant clinical problem characterized by ex- Queen Mary University of London, London EC1M 6BQ, United Kingdom uberant inflammation leading to joint destruction and pain (6, 7). 1Current address: William Harvey Research Institute, Barts and London School of Medicine, Queen Mary University of London, London, U.K. Emerging evidence indicates that failure to engage proresolving ORCIDs: 0000-0002-5163-3946 (H.H.A.); 0000-0001-6328-3640 (J.D.); 0000-0001- pathways may contribute to persistent chronic inflammation, such 5316-9115 (L.V.N.); 0000-0003-2068-3331 (M.P.); 0000-0003-4627-8545 (C.N.S.). as in rheumatoid arthritis (RA) (1, 2). Current therapeutic ap- Received for publication October 21, 2015. Accepted for publication July 12, 2016. proaches aim to disrupt the inflammatory response, slow This work was supported in part by National Institutes of Health Grants P01GM095467 progression, and limit pain (2, 8). These include methotrexate, and R01GM38765 (to C.N.S.), a postdoctoral fellowship from the Arthritis Foundation anti-TNF biologics, and nonsteroidal anti-inflammatory drugs (8). (to H.H.A.), and by Arthritis Research UK Carrier Development Fellowship 19909 (to However, these treatments may have drawbacks in terms of in- L.V.N.). J.D. is funded by a Sir Henry Dale Fellowship jointly funded by the Wellcome Trust and Royal Society Grant 107613/Z/15/Z. fections and may prevent resolution and tissue repair (1, 8, 9). Address correspondence and reprint requests to Dr. Charles N. Serhan, Center for Hence, safe and effective treatments that can regulate arthritic Experimental Therapeutics and Reperfusion Injury, Brigham and Women’s Hospital inflammation are still in need. In that context, 17R-RvD1 (7S,8R,17R- and Harvard Medical School, 77 Avenue Louis Pasteur, HIM 829, Boston, MA trihydroxy-docosa-4Z,9E,11E,13Z,15E,19Z-hexaenoic acid) and the 02115. E-mail address: [email protected] D-series Rv pathway marker 17R-hydroxydocosahexaenoic acid The online version of this article contains supplemental material. (17R-HDHA) are log order more potent than opioids, cyclooxy- Abbreviations used in this article: CHO, Chinese hamster ovary; DHA, docosahexaenoic acid; EFA, essential fatty acid; KO, knockout; LC-MS/MS, liquid chromatography– genase inhibitors, or glucocorticoids in modulating inflammatory tandem mass spectrometry; LM, lipid mediator; LT, leukotriene; LTB4, leukotriene B4 pain in a preclinical model of arthritis in rats (10). (5S,12R-dihydroxy-eicosa-6Z,8E,10E,14Z-tetraenoic acid); LX, lipoxin; LXA4, lipoxin Resolvins of the D-series are enzymatically biosynthesized from A4 (5S,6R,15S-trihydroxy-eicosa-7E,9E,11Z,13E-tetraenoic acid); MaR, maresin; MaR1, maresin 1 (7R,14S-dihydroxy-docosa-4Z,8E,10E,12Z,16Z,19Z-hexaenoic acid); the omega-3 EFA, docosahexaenoic acid (DHA), including RvD1, a MS/MS, tandem mass spectrometry; PD, protectin; PGD2, 11-oxo-9 ,15S-dihydroxy- RvD2 (7S,16R,17S-trihydroxy-docosa-4Z,8E,10Z,12E,14E,19Z- a prosta-5Z,13E-dien-1-oic acid; PGE2, 9-oxo-11 ,15S-dihydroxy-prosta-5Z,13E-dien-1- hexaenoic acid), RvD3 (4S,11R,17S-trihydroxy-docosa-5Z,7E,9E, oic acid; PLS-DA, partial least-squares discriminant analysis; RA, rheumatoid arthritis; 17R-HDHA, 17R-hydroxydocosahexaenoic acid; Ri, resolution interval; Rv, resolvin; RvD1, 13Z,15E,19Z-hexaenoic acid), and RvD4 (4S,5R,17S-trihydroxy- resolvin D1 (7S,8R,17S-trihydroxy-docosa-4Z,9E,11E,13Z,15E,19Z-hexaenoic acid); RvD2, docosa-6E,8E,10Z,13Z,15E,19Z-hexaenoic acid) (1). Recently, we resolvin D2 (7S,16R,17S-trihydroxy-docosa-4Z,8E,10Z,12E,14E,19Z-hexaenoic acid); RvD3, resolvin D3 (4S,11R,17S-trihydroxy-docosa-5Z,7E,9E,13Z,15E,19Z-hexaenoic acid); completed the stereochemistry of RvD3, having established its po- RvD4, resolvin D4 (4S,5R,17S-trihydroxy-docosa-6E,8E,10Z,13Z,15E,19Z-hexaenoic tent proresolving actions and unique temporal profile late within a acid); SPM, specialized proresolving mediator; TxB2,thromboxaneB2 (9 ,11,15S- the resolution phase (11), suggesting it may have a specific role trihydroxy-thromba-5Z,13E-dien-1-oic acid); VIP, variable influence on projection. in regulating resolution. Given these properties, we investigated Copyright Ó 2016 by The American Association of Immunologists, Inc. 0022-1767/16/$30.00 the role and action of RvD3 in arthritis.

www.jimmunol.org/cgi/doi/10.4049/jimmunol.1502268 2 RvD3 AND ARTHRITIS

Materials and Methods K/BxN serum transfer inflammatory arthritis model Animals For self-resolving and delayed-resolving arthritis, C57BL/6 male mice were m Male C57BL/6 mice (8 wk old) were obtained from Charles River Labo- administered arthritogenic K/BxN serum (100 l, i.p.) at days 0 and 2 for ratories (Newton, MA). All animal experiments were in accordance with the self-resolving arthritis, and with additional challenge on day 8 for delayed- Harvard Medical Area Standing Committee on Animals (protocol no. 02570). resolving arthritis. Clinical scores were assessed (score criteria per paw: 0, For some experiments, male C57BL/6 mice (12 wk old) were obtained from no signs of inflammation; 1, inflammation in one of the following aspects— Charles River UK (Kent, U.K.) or ALX/fpr2/3 knockout (KO) mice were individual phalange joints, localized wrist/ankle, or swelling on surface of generated on a C57BL/6 background (and backbackcrossed 10 times) in- paw; 2, inflammation on two aspects of paw; 3, major swelling on all aspects house at the William Harvey Research Institute, Barts and the London of paw; maximum score of 12 per mouse) to monitor disease development. School of Medicine, Queen Mary University of London (London, U.K.), as Joints were harvested at indicated time intervals, and LM metabololipidomics described (12). All animal experiments were performed in accordance with was performed as described. In some experiments, mice were taken to institutional guidelines. moorFLPI-2 laser Doppler imaging (Moor Instruments, Wilmington, DE) at day 15. Liquid chromatography–tandem mass spectrometry-based LM For RvD3 treatments, mice (C57BL/6 or ALX/fpr2/3 KO mice) were m metabololipidomics administered vehicle (0.1% ethanol) or RvD3 (100 ng) in 100 l of saline i.p. daily, starting from day 0 until completion of experiments at day 6. Deidentified serum from healthy volunteers and RA patients were obtained from Dx Clinical scores were assessed as above, and paw edema assessed by water Biosamples (San Diego, CA). Their demographics are shown in Supplemental displacement plethysmometry (Ugo Basile, Comerio, Italy). On termina- Table I. RA patients were diagnosed with stage III RA by a board-certified phy- tion of the experiments, arthritic paws and joints were collected for his- sician and were treated with standard therapies (Supplemental Table I). For LM tology staining and LM metabololipidomics. profiling, ice-cold methanol containing deuterated internal standards d8-5S-HETE, d4-LTB4 (5S,12R-dihydroxy-eicosa-6Z,8E,10E,14Z-tetraenoic acid), d5-LXA4 Partial least-squares discriminant analysis Downloaded from (5S,6R,15S-trihydroxy-eicosa-7E,9E,11Z,13E-tetraenoic acid), d4-PGE2 (9-oxo-11a,15S-dihydroxy-prosta-5Z,13E-dien-1-oic acid), and d5-RvD2, rep- Partial least-squares discriminant analysis (PLS-DA) was performed using resenting each chromatographic region of identified LM (500 pg each), was SIMCA 13.0.3 software (Umetrics, San Jose, CA) following mean centering added to samples prior to processing to facilitate quantification and assess- and unit variance scaling of LM amounts. The score plot shows the sys- ment of sample recovery. Liquid chromatography–tandem mass spectrometry tematic clusters among the observations (closer plots presenting higher (LC-MS/MS)-based metabololipidomics were performed as described in similarity in the data matrix). Loading plots describe the magnitude and the Colas et al. (13). RvD3 was prepared and qualified by LC-MS/MS and manner (positive or negative correlation) in which the measured LMs-SPMs spectroscopic analysis as in Dalli et al.(11),andsecondsyntheticRvD3was contribute to the cluster separation in the score plot (15). http://www.jimmunol.org/ obtained from Cayman Chemical (Ann Arbor, MI). Also, the complete ste- reochemistry of RvD4 was recently determined (14). Multiple reaction Histological analysis monitoring was used with signature ion pairs Q1 (parent ion)–Q3 (charac- teristic daughter ion) for each molecule. Data acquisition was performed in Joints were fixed and decalcified for 2 wk prior to embedding in paraffin. negative ionization mode, and identification was conducted in accordance Sections (8 mm) were stained with H&E and standard light microscopy with published criteria (13), with a minimum of six diagnostic ions. was used to determine the degree of neutrophil infiltration. by guest on September 27, 2021

FIGURE 1. SPM temporal regulation during self-limited inflammatory arthritis. Mice were challenged with K/BxN serum (100 ml, i.p.) on days 0 and 2. Disease progression was monitored and paws were collected on the indicated intervals. (A) Clinical scores. (B) H&E staining; blue arrowheads denote leukocytic infiltration; black arrowheads denote bone erosion. Original magnification 310. (C and D) LM levels were determined in paws at the indicated intervals using LC-MS/MS–based LM profiling. (C) MS/MS spectrum employed for the identification of RvD3. (D) Principal component analysis of the LM profiles. Left panel, Two-dimensional score plot; right panel, two-dimensional loading plot. Results in (A)aremean6 SEM; n = 4 mice per interval. Results in (B)–(D) are representative of n =16mice. The Journal of Immunology 3

Ligand selectivity using electric cell-substrate identified in accordance with published criteria as illustrated for impedance sensing RvD3 (see Fig. 1D, Materials and Methods, Supplemental Fig. 1). hALX-transfected Chinese hamster ovary (CHO) cells were plated at 1 3 Distinct LM profiles were identified in joints for each time interval 105 cells/well in Ham’s F-12 buffer containing 10% FCS and G418 (37˚C) prior to (i.e., day 0; naive) and after induction of self-resolving for 36 h. Then the media were changed to serum-free Ham’s media, and arthritis using principal component analysis (Fig. 1D). There were vehicle or RvD3 (100 nM) was added and impedance across the cell higher levels of RvD4, MaR1 (7R,14S-dihydroxy-docosa-4Z,8E, monolayer was measured using electric cell-substrate impedance sensing 10E,12Z,16Z,19Z-hexaenoic acid), RvE1 (5S,12R,18R-trihydroxy- (Applied Biophysics, Troy, NY) to monitor ligand–receptor interactions as in Krishnamoorthy et al. (16). eicosa-6Z,8E,10E,14Z,16E-pentaenoic acid) and 15R-LXA4 (5S,6R, 15R-trihydroxy-eicosa-7E,9E,11Z,13E-tetraenoic acid) in naive com- Statistical analysis pared with arthritic paws. Following serum challenge there was Data are presented as means 6 SEM. The criterion for statistical signi- an increase in proinflammatory eicosanoids, including PGE2,PGD2 , ficance was p 0.05 using a Student t test (two groups), one-way ANOVA (11-oxo-9a,15S-dihydroxy-prosta-5Z,13E-dien-1-oic acid) TXB2, (multiple groups), or two-way repeated measures ANOVA (two groups and LTB , with a concomitant decrease in most SPMs identified over time) followed by a post hoc Bonferroni test using GraphPad Prism 6. 4 (Fig. 1D, Supplemental Fig. 1) demonstrated by a downward shift in the 4- and 8-d clusters (arthritis) on the score plot (Fig. 1D, left). Results This was followed by a counterclockwise shift in the 16-d cluster SPMs are temporally regulated in joints during self-limited (resolution) back toward the 0-d cluster (Fig. 1D). Of note, levels of inflammatory arthritis specific SPMs, including the D-series resolvins RvD1, RvD2 and

To investigate whether SPMs are temporally regulated in joints during RvD3, increased in arthritic joints during the resolution phase (day Downloaded from inflammation–resolution, we used an established murine model 16; Supplemental Fig. 1). These findings demonstrate that endog- relevant to human arthritis that is driven by the innate immune re- enous SPM pathways are active in murine joints that are temporally sponse (17, 18). Mice injected with K/BxN arthritogenic serum and differentially regulated during self-resolving inflammatory (100 ml/mouse, i.p.) at day 0 and day 2 developed a self-resolving arthritis. polyarthritic inflammatory response that reached a maximum clinical Shorter resolution intervals and specific SPMs in score at day 11, and subsequently declined (Fig. 1A). The reduction http://www.jimmunol.org/ in clinical scores corresponded with loss of leukocytes from the joints self-resolving arthritis (Fig. 1B), a key histological feature of resolution (1). Because selected SPMs were temporally regulated in self-resolving Using LC-MS/MS–based metabololipidomics, we next investi- joint inflammation, we investigated their levels during delayed-resolving gated the temporal LM-SPM profiles in self-resolving joints. LMs inflammatory arthritis to determine which of these tissue-protective identified include E- and D-series Rvs, PDs, MaRs from the omega-3 molecules may be directly involved in resolving joint inflammation. (i.e., eicosapentaenoic acid and DHA), bioactive metabolomes, and First, to provide quantitative analyses of resolution (i.e., remission) of LXs as well as prostanoids and LTs from the arachidonic acid bio- joint inflammation, we applied resolutionindices,asarewidelyusedto active metabolome (Fig. 1C, Supplemental Fig. 1). All LMs were calculate the loss of neutrophils from site of inflammation (19), using by guest on September 27, 2021

FIGURE 2. Delayed-resolving versus self-resolving arthritis: prolonged reso- lution interval and altered LM-SPM profile. Mice were challenged with K/BxN serum (100 ml, i.p.) on days 0 and 2 (self-resolving) or days 0, 2, and 8 (delayed-resolving). (A) Clinical scores were determined and used to calculate the Ri (see Materials and Methods). Results are means 6 SEM from n =5mice.(B) Doppler imaging from paws with arthritis (left), self-resolving arthritis (middle), and without K/BxN serum (right). Results are expressed as perfusion units (which are arbitrary) and arerepresentativefromn = 5 mice per group. LM from self-resolving (left) and delayed-resolving (right) paw joints was determined using LC-MS/MS–based LM metabololipidomics. (C) PLS-DA. Endog- enous LM: two-dimensional score (left) and two-dimensional loading (right) plots of LMs-SPMs obtained from self-resolving (blue) and delayed-resolving (red) paw joints. LMs with VIP scores of .1.0 are color coded in accordance with associated cluster. Results are from four mice per group. 4 RvD3 AND ARTHRITIS

clinical scores to quantify the resolution interval (Ri). In this study, the (TxB2,9a,11,15S-trihydroxy-thromba-5Z,13E-dien-1-oic acid), and Ri is the interval between when clinical scores reach maximum (day 8) PGE2 (Fig. 2C, right). Taken together, these findings indicate that andwhentheyreach50%remission(t50; day 12), giving an Ri of 4 d delayed resolution dysregulates endogenous resolution programs (Fig. 2A). Additional challenge with K/BxN serum at day 8 prolonged in arthritic joints, increasing local proinflammatory mediators arthritis with delayed resolution, with a t50 of 17 d, or an Ri of 9 d and diminishing SPMs. (Fig. 2A). At day 15, mice were taken for laser Doppler imaging to further assess the degree of joint inflammation–resolution in compari- Dysregulated LM-SPM profiles in human RA patients son with unchallenged mice (without K/BxN serum, Fig. 2B). Increases Toward human translation, we also compared the LM-SPM profiles in tissue perfusion monitored via laser Doppler were consistent with in serum from healthy individuals and patients with RA. All LMs increases in joint inflammation. These results indicate that resolution of were identified in accordance with published criteria as illustrated inflammatory arthritis is delayed following an additional K/BxN serum for RvD1 and RvD3 (Fig. 3A). Quantification using multiple re- challenge, which may be associated with an altered LM-SPM profile. action monitoring gave significantly lower levels of RvD3 (4.3 6 We next assessed the local endogenous LM-SPM profile in self- 1.4 versus 17.7 6 2.3 pg/ml, p = 0.008) and PGD2 (46.8 6 9.8 resolving joints and compared it to the delayed-resolving joints. versus 95.7 6 13.9 pg/ml, p = 0.046), with a similar trend for Differences in local paw joint LM-SPM profiles were assessed RvD4 (1.1 6 0.3 versus 3.3 6 0.9 pg/ml, p = 0.083), found in using PLS-DA (Fig. 2C). To identify which variables were respon- serum from RA patients compared with healthy individuals sible for the separation between the clusters (Fig. 2C, left), a variable (Supplemental Table I), results that will need to be corroborated in influence on projection (VIP) score of .1.0 was used to select the a prospective study. Following identification using the LC-MS/MS most significant variables contributing to the cluster separation. The fractions with matched mass spectra diagnostic ions and retention Downloaded from self-resolving cluster was characterized by higher levels of RvD3, time of authentic standards, the data were subjected to multivariate MaR1, and PD1 (10R,17S-dihydroxy-docosa-4Z,7Z,11E,13E,15Z,19Z- analysis. The PLS-DA score plot demonstrated a clear separation hexaenoic acid, also known as neuroprotectin D1) (VIP score of .1.0; between the healthy and RA serum clusters (Fig. 3B, left). The Fig. 2C, right), whereas the delayed-resolving cluster was associated loading plot, indicating the magnitude and manner (positive or with higher levels of proinflammatory LM, such as thromboxane B2 negative correlation) that each of the LM-SPM profiles contributes http://www.jimmunol.org/ by guest on September 27, 2021

FIGURE 3. Human arthritis patients: altered LM-SPM profiles. LMs obtained from deidentified serum from healthy individuals or RA patients were identified by LC-MS/MS–based LM metabololipidomics (see Materials and Methods). (A) Representative MS/MS spectra employed for identification of, for example, RvD1 (left) and RvD3 (right). Inset, diagnostic ions; M, molecular mass. (B) PLS-DA. Endogenous LMs: two-dimensional score (left) and two-dimensional loading (right) plots of LMs-SPMs obtained from healthy (blue) and RA (red) serum. LMs with VIP scores .1.0 are color coded in accordance with associated cluster. Results are from three individuals in each group. The Journal of Immunology 5 to the cluster separation, showed that serum from healthy individ- uals was characterized (VIP score $1.0) by higher RvD3, RvD4, RvE3 (17R,18R-dihydroxy-eicosa-5Z,8Z,11Z,13E,15E-pentaenoic acid), 15R-LXA4, and PGD2 levels, whereas the RA serum was associated with higher TxB2 levels (Fig. 3B, right). Hence, these results in human RA patients together with the in vivo findings in murine arthritis suggest that selected SPMs, such as RvD3, are deregulated in arthritis. RvD3 limits arthritis severity and joint inflammation in mice Having found that RvD3 was dysregulated both in delayed- resolving arthritic joints (Fig. 2C) and in human RA serum (Fig. 3), we next sought to determine its pharmacological actions in inflam- matory arthritis. Mice were challenged with K/BxN serum (100 ml, i.p.) on days 0 and 2. Daily administration of RvD3 (100 ng/mouse, i.p.) starting at day 0 significantly reduced clinical scores by 35–60% (Fig. 4A) and hind paw edema by 50–85% (Fig. 4B) during the course of the experiments, with a significant reduction observed as early as day 2 (Fig. 4A, 4B). Histological analysis showed fewer leukocytes at Downloaded from day 6 in joints from arthritic mice administered RvD3 (Fig. 4C). Hence, these results demonstrate that RvD3 alleviates arthritis pro- gression, reducing edema and limiting leukocyte numbers in murine K/BxN serum–induced arthritis. RvD3 reduces local eicosanoid levels in murine inflammatory arthritis http://www.jimmunol.org/ Because RvD3 reduces eicosanoid production in acute peritonitis (11), we next assessed its ability to regulate local joint eicosanoid levels in murine inflammatory arthritis. We preformed LC-MS/ MS–based LM metabololipidomics with murine arthritic joints (day 6) to identify potential mechanisms underlying the antiar- thritic actions of RvD3. Treatment with RvD3 resulted in signif- icant (p , 0.01) reduction in local joint levels of LTB4 (34.7 6

10.9%), PGE2 (67.7 6 6.9%), PGD2 (54.3 6 15.9%), PGF2a by guest on September 27, 2021 (9a,11a,15S-trihydroxy-prosta-5Z,13E-dienoic acid) (61.0 6 10.1%), and TxB2 (50.0 6 3.6%; Fig. 4D). We also identified the FIGURE 4. RvD3 limits joint inflammation in arthritic mice. Mice were nonenzymatic 8-iso-PGF a, indicative of , which 2 challenged with K/BxN serum (100 ml, i.p.) on days 0 and 2 for induction was also reduced in RvD3-treated mice (Fig. 4E). Taken together, of arthritis and were administered either vehicle (veh; saline plus 0.1% these results indicate that RvD3’s potent actions limiting paw joint EtOH) or RvD3 (100 ng/mouse, i.p.) daily. (A) Clinical scores and (B) hind inflammation include regulating local eicosanoids, characteristic paw edema were assessed daily (see Materials and Methods). (C) On day of SPMs (1). 6, joints were collected and taken for histological microscopic analysis with H&E. Original magnification 310 (top); inset box shows leukocyte The arthritis-protective actions of RvD3 are lost in infiltrates with original magnification 340 (bottom). Results are (A and B) ALX/FPR2-deficient mice mean 6 SEM or (C) representative from n = 10 mice per group. LMs were We next tested whether RvD3 activates the proresolving receptor assessed in arthritic paws (day 6) using LC-MS/MS–based LM metab- ololipidomics following solid phase extraction. (D and E) Reduction in ALX/FPR2, which mediates the actions of LXA4 and RvD1 (1). Expression of this receptor in transfected CHO cells demonstrated joint eicosanoids with RvD3 treatment expressed as percentage reduction 6 that RvD3 at 100 nM directly activated this receptor (Fig. 5A). from arthritis plus vehicle mice. Results are means SEM from n =3 mice per group. *p , 0.05, **p , 0.01, ***p , 0.001, ****p , 0.0001 Because mice lacking ALX/FPR2 exhibit a more pronounced and versus arthritis plus vehicle mice. prolonged joint response compared with wild-type counterparts (20), we next investigated whether the arthritis-protective actions of RvD3 were altered in these mice. We found that in ALX/FPR2- RvD3 alleviated arthritis severity and reduced paw edema, joint deficient mice, RvD3 did not reduce clinical scores, hind paw leukocytes, and eicosanoid levels, actions that were lost in ALX/ edema, or LTB4 levels (Fig. 5B–D). Thus, taken together, these FPR2-deficient mice. Taken together, these results implicate a results suggest that RvD3 activates ALX/FPR2 in vivo in mice protective role for RvD3 in arthritis. that, at least in part, mediates RvD3 protective actions in this mu- Disruption of endogenous resolution programs may contribute to rine model of experimental arthritis. persistent chronic inflammation, a pathological feature of many widely occurring diseases, including RA (1, 2). To our knowledge, Discussion the present study is the first to investigate the temporal regulation In the present study, using LM metabololipidomics, we investigated of SPMs in self-resolving arthritis to potentially identify specific the temporal regulation of LMs-SPMs in joints during self-limited SPMs that may be upregulated in resolution of joint inflammation. inflammatory arthritis in mice. We identified RvD3 as an SPM that In this context, evidence from experimental models of inflam- was deregulated both delayed-resolving murine arthritis and in matory arthritis suggests that disruption of SPM biosynthetic or serum obtained from RA patients. When administered to mice, signaling pathways leads to exacerbation of disease (20, 21). Arthritic 6 RvD3 AND ARTHRITIS

FIGURE 5. RvD3’s protective actions in arthritis are lost in receptor-deficient mice. (A) Ligand-receptor–dependent changes in impedance with CHO cells overexpressing human ALX continuously recorded with real-time monitoring across cell monolayers using an electric cell–substrate impedance sensing system. Results are representative from four independent traces obtained from incubations of RvD3 (100 nM) or vehicle with CHO-ALX/fpr– expressing cells. (B–D) ALX-deficient (ALX/fpr2/3 KO) and wild-type mice were challenged with K/BxN serum (100 ml, i.p.) on days 0 and 2 for in- duction of arthritis and were administered either vehicle (veh; saline plus 0.1% EtOH) or RvD3 (100 ng/mouse) i.p. daily. Reduction in (B) arthritis score,

(C) joint edema, and (D) local LTB4 levels at day 6. Results are expressed as percentage reduction by RvD3 from vehicle treated mice in wild-type and ALX/fpr2/3 KO mice. Results in (B)–(D) are mean 6 SEM from n = 8 mice for treatment with vehicle (saline plus 0.1% EtOH), n = 7 mice for treatment with RvD3 in ALX/fpr2/3 KO mice, and n = 10 for RvD3 or vehicle treatment in wild-type mice. *p , 0.05, **p , 0.01 versus arthritis plus vehicle mice. Downloaded from WT, wild-type. mice deficient in 12/15-lipoxygenase, a key in SPM bio- of interest. In this regard, recent randomized controlled human synthesis, have reduced LXA4 levels and exacerbated disease (21). trials indicate that omega-3 intake elevated 17R-HDHA and RvD1

Similarly, increased disease severity is observed in mice lacking the plasma levels in patients with chronic kidney diseases (28) as well http://www.jimmunol.org/ receptor ALX/FPR2 (20). In mice, RvD1 signals via ALX/FPR2 as in chronic knee effusions and plasma in arthritis patients (29). (16), an observation that also holds for RvD3, as demonstrated by Thus, these and our present findings may have implications for loss of activity in ALX/FPR2-deficient mice (Fig. 5B, 5C). These human chronic inflammatory diseases. Moreover, SPMs are present findings are consistent with the notion that ALX/FPR2 is a key re- at bioactive concentrations in synovial fluid from RA patient (30), ceptor in the resolution mechanisms in inflammation. These findings, including RvD1 and RvD3 (24). Therefore, in clinical settings taken together with the findings that SPMs are temporally regulated RvD3 may represent one of the mediators involved in the etiopa- during resolution and that its delay is associated with deregulation in thology of RA. When taken together with our present results dem- specificSPMsinarthriticmurinejoints, implicate a role for these onstrating that RvD3 is dysregulated in both human and murine SPMs and their pathways in modulating inflammatory arthritis. arthritis, enhancing the endogenous D-series Rvs may provide new by guest on September 27, 2021 RA is a chronic progressive disease with predominant inflam- therapeutic approaches for controlling arthritis and inflammation- mation driven by proinflammatory signals, such as PGs, LTB4, and mediated diseases. Moreover, these results with D-series Rvs pro- TNF-a (6). Most current drugs used to treat RA aim to reduce, vide a new molecular basis that may be responsible for some of the antagonize, or inhibit the production of these proinflammatory beneficial actions of omega-3 EFA in human RA. molecules (8, 22). Although these medications can help relieve symptoms, they can also delay or prevent complete resolution and Acknowledgments are thus considered resolution toxic, because they block the pro- We thank M.H. Small for expert assistance in manuscript preparation and duction of local PGs that are important for resolution (1). One such Dr. B. Schmidt at the Department of , Children’s Hospital Boston, example is the cyclooxygenase-2 inhibitor NS-398, which delays for expert histopathology and microscopic evaluation of arthritic joints. resolution of murine autoimmune arthritis via disrupting the PGE2- mediated LXA4 induction and production (9). Of note, when added Disclosures back in animal disease models, SPMs, such as Rvs, PDs, and LXs, The authors have no financial conflicts of interest. actively counterregulate inflammation and promote its termination (reviewed in Ref. 1). In that context, 17R-RvD1 as well as a D-series References resolution precursor, 17R-HDHA, potently alleviate joint stiffness, 1. Serhan, C. N. 2014. Pro-resolving lipid mediators are leads for resolution inflammation, and inflammatory pain in a preclinical model of ar- . Nature 510: 92–101. thritisinrats(10).BothLXA4 and 17R-RvD1 counterregulate pro- 2. Tabas, I., and C. K. Glass. 2013. Anti-inflammatory therapy in chronic disease: duction of TNF-a and TNF-a actions, and they are protective in challenges and opportunities. Science 339: 166–172. 3. Samuelsson, B. 2012. Role of basic science in the development of new medi- experimental models of arthritis (10, 23, 24). Of interest, SPMs, cines: examples from the eicosanoid field. J. Biol. Chem. 287: 10070–10080. including RvD3, enhance host-directed microbial responses and thus 4. Dinarello, C. A., A. Simon, and J. W. van der Meer. 2012. Treating inflammation by blocking interleukin-1 in a broad spectrum of diseases. Nat. Rev. Drug Dis- do not provoke immune suppression (1, 19, 25), a potential major cov. 11: 633–652. side effect associated with many current RA therapeutics (2, 8). 5. Serhan, C. N., S. Hong, K. Gronert, S. P. Colgan, P. R. Devchand, G. Mirick, and Thus, taken together with our present results on RvD3’s potent an- R. L. Moussignac. 2002. Resolvins: a family of bioactive products of omega-3 fatty acid transformation circuits initiated by aspirin treatment that counter tiarthritic actions, SPMs may represent a new class of molecules that proinflammation signals. J. Exp. Med. 196: 1025–1037. are well suited to treat inflammation and inflammatory pain associ- 6. Koopman, J., and L. Moreland, eds. 2005. Arthritis and Allied Conditions: A ated with arthritis without compromising host-directed responses. Textbook of Rheumatology. Lippincott Williams & Wilkins, Philadelphia. 7. Yacoubian, S., and C. N. Serhan. 2007. New endogenous anti-inflammatory and Omega-3 polyunsaturated fatty acids, such as eicosapentaenoic proresolving lipid mediators: implications for rheumatic diseases. Nat. Clin. acid and DHA, are now appreciated to exert beneficial actions in Pract. Rheumatol. 3: 570–579, quiz 1, 589. 8. Siebert, S., A. Tsoukas, J. Robertson, and I. McInnes. 2015. Cytokines as several human diseases with underlying inflammatory conditions, therapeutic targets in rheumatoid arthritis and other inflammatory diseases. such as RA (26, 27), but their mechanisms have remained a subject Pharmacol. Rev. 67: 280–309. The Journal of Immunology 7

9. Chan, M. M., and A. R. Moore. 2010. Resolution of inflammation in murine effects on leukocyte responses and experimental inflammation. J. Immunol. autoimmune arthritis is disrupted by cyclooxygenase-2 inhibition and restored 184: 2611–2619. by prostaglandin E2-mediated lipoxin A4 production. J. Immunol. 184: 6418– 21. Kro¨nke, G., J. Katzenbeisser, S. Uderhardt, M. M. Zaiss, C. Scholtysek, 6426. G. Schabbauer, A. Zarbock, M. I. Koenders, R. Axmann, J. Zwerina, et al. 2009. 10. Lima-Garcia, J. F., R. C. Dutra, K. da Silva, E. M. Motta, M. M. Campos, and 12/15-Lipoxygenase counteracts inflammation and tissue damage in arthritis. J. B. Calixto. 2011. The precursor of resolvin D series and aspirin-triggered J. Immunol. 183: 3383–3389. resolvin D1 display anti-hyperalgesic properties in adjuvant-induced arthritis 22. Anderson, P. J. 2005. Tumor factor inhibitors: clinical implications of in rats. Br. J. Pharmacol. 164: 278–293. their different immunogenicity profiles. Semin. Arthritis Rheum. 34(Suppl. 1): 11. Dalli, J., J. W. Winkler, R. A. Colas, H. Arnardottir, C. Y. Cheng, N. Chiang, 19–22. N. A. Petasis, and C. N. Serhan. 2013. Resolvin D3 and aspirin-triggered 23. Conte, F. P., O. Menezes-de-Lima, Jr., W. A. Verri, Jr., F. Q. Cunha, C. Penido, resolvin D3 are potent immunoresolvents. Chem. Biol. 20: 188–201. and M. G. Henriques. 2010. Lipoxin A4 attenuates zymosan-induced arthritis by 12. Dufton, N., and M. Perretti. 2010. Therapeutic anti-inflammatory potential of modulating endothelin-1 and its effects. Br. J. Pharmacol. 161: 911–924. formyl-peptide receptor agonists. Pharmacol. Ther. 127: 175–188. 24. Norling, L. V., S. E. Headland, J. Dalli, H. H. Arnardottir, O. Haworth, 13. Colas, R. A., M. Shinohara, J. Dalli, N. Chiang, and C. N. Serhan. 2014. H. R. Jones, D. Irimia, C. N. Serhan, and M. Perretti. 2016. Proresolving and Identification and signature profiles for pro-resolving and inflammatory lipid cartilage-protective actions of resolvin D1 in inflammatory arthritis. JCI Insight mediators in human tissue. Am. J. Physiol. Cell Physiol. 307: C39–C54. 1: e85922. 14. Winkler, J. W., S. K. Orr, J. Dalli, C. Y. Cheng, J. M. Sanger, N. Chiang, 25. Norris, P. C., H. Arnardottir, J. M. Sanger, D. Fichtner, G. S. Keyes, and N. A. Petasis, and C. N. Serhan. 2016. Resolvin D4 stereoassignment and its C. N. Serhan. 2016. Resolvin D3 multi-level proresolving actions are host pro- novel actions in host protection and bacterial clearance. Sci. Rep. 6: 18972. tective during infection. Prostaglandins Leukot. Essent. Fatty Acids DOI: 10.1016/ 15. Janes, K. A., and M. B. Yaffe. 2006. Data-driven modelling of signal- j.plefa.2016.01.001. transduction networks. Nat. Rev. Mol. Cell Biol. 7: 820–828. 26. Reed, G. W., K. Leung, R. G. Rossetti, S. Vanbuskirk, J. T. Sharp, and 16. Krishnamoorthy, S., A. Recchiuti, N. Chiang, G. Fredman, and C. N. Serhan. R. B. Zurier. 2014. Treatment of rheumatoid arthritis with marine and botanical 2012. Resolvin D1 receptor stereoselectivity and regulation of inflammation and oils: an 18-month, randomized, and double-blind trial. Evid. Based Complement. proresolving microRNAs. Am. J. Pathol. 180: 2018–2027. Alternat. Med. 2014: 857456. 17. Misharin, A. V., C. M. Cuda, R. Saber, J. D. Turner, A. K. Gierut, G. K. Haines, 27. Miles, E. A., and P. C. Calder. 2012. Influence of marine n-3 polyunsaturated III, S. Berdnikovs, A. Filer, A. R. Clark, C. D. Buckley, et al. 2014. Nonclassical fatty acids on immune function and a systematic review of their effects on Downloaded from Ly6C2 monocytes drive the development of inflammatory arthritis in mice. Cell clinical outcomes in rheumatoid arthritis. Br.J.Nutr.107(Suppl. 2): S171–S184. Rep. 9: 591–604. 28. Mas, E., A. Barden, V. Burke, L. J. Beilin, G. F. Watts, R. C. Huang, I. B. Puddey, 18. Monach, P., K. Hattori, H. Huang, E. Hyatt, J. Morse, L. Nguyen, A. Ortiz- A. B. Irish, and T. A. Mori. 2016. A randomized controlled trial of the effects of Lopez, H. J. Wu, D. Mathis, and C. Benoist. 2007. The K/BxN mouse model of n-3 fatty acids on resolvins in chronic kidney disease. Clin. Nutr. 35: 331–336. inflammatory arthritis: theory and practice. Methods Mol. Med. 136: 269–282. 29. Barden, A. E., M. Moghaddami, E. Mas, M. Phillips, L. G. Cleland, and 19. Chiang, N., G. Fredman, F. Ba¨ckhed, S. F. Oh, T. Vickery, B. A. Schmidt, and T. A. Mori. 2016. Specialised pro-resolving mediators of inflammation in in- C. N. Serhan. 2012. Infection regulates pro-resolving mediators that lower an- flammatory arthritis. Prostaglandins Leukot. Essent. Fatty Acids 107: 24–29.

tibiotic requirements. Nature 484: 524–528. 30. Giera, M., A. Ioan-Facsinay, R. Toes, F. Gao, J. Dalli, A. M. Deelder, http://www.jimmunol.org/ 20. Dufton, N., R. Hannon, V. Brancaleone, J. Dalli, H. B. Patel, M. Gray, C. N. Serhan, and O. A. Mayboroda. 2012. Lipid and lipid mediator profiling of F. D’Acquisto, J. C. Buckingham, M. Perretti, and R. J. Flower. 2010. Anti- human synovial fluid in rheumatoid arthritis patients by means of LC-MS/MS. inflammatory role of the murine formyl-peptide receptor 2: ligand-specific Biochim. Biophys. Acta 1821: 1415–1424. by guest on September 27, 2021