Circulating Soluble IL-6R but Not ADAM17 Activation Drives Mononuclear Cell Migration in Tissue Inflammation

This information is current as Neele Schumacher, Stefanie Schmidt, Jeanette Schwarz, of October 2, 2021. Dana Dohr, Juliane Lokau, Jürgen Scheller, Christoph Garbers, Athena Chalaris, Stefan Rose-John and Björn Rabe J Immunol 2016; 197:3705-3715; Prepublished online 3 October 2016; doi: 10.4049/jimmunol.1600909 http://www.jimmunol.org/content/197/9/3705 Downloaded from

Supplementary http://www.jimmunol.org/content/suppl/2016/10/01/jimmunol.160090

Material 9.DCSupplemental http://www.jimmunol.org/ References This article cites 52 articles, 21 of which you can access for free at: http://www.jimmunol.org/content/197/9/3705.full#ref-list-1

Why The JI? Submit online.

• Rapid Reviews! 30 days* from submission to initial decision by guest on October 2, 2021

• No Triage! Every submission reviewed by practicing scientists

• Fast Publication! 4 weeks from acceptance to publication

*average

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. The Journal of Immunology

Circulating Soluble IL-6R but Not ADAM17 Activation Drives Mononuclear Cell Migration in Tissue Inﬂammation

Neele Schumacher,* Stefanie Schmidt,* Jeanette Schwarz,*,1 Dana Dohr,* Juliane Lokau,* Jurgen€ Scheller,† Christoph Garbers,* Athena Chalaris,* Stefan Rose-John,* and Bjo¨rn Rabe*

Neutrophil and mononuclear cell infiltration during inflammatory processes is highly regulated. The first cells at the site of infection or inflammation are neutrophils, followed by mononuclear cells. IL-6 plays an important role during inflammatory states. It has been shown in several models that the soluble form of IL-6R (sIL-6R) is involved in the recruitment of mononuclear cells by a mechanism called IL-6 trans-signaling. It had been speculated that sIL-6R was generated at the site of inflammation by shedding from neutrophils via activation of the metalloprotease ADAM17. Attempts to genetically delete the floxed ADAM17 gene selec- cre

tively in myeloid cells infiltrating an air pouch cavity upon injection of carrageenan failed because in transgenic mice, LysM did Downloaded from not lead to appreciable loss of the ADAM17 protein in these cells. We therefore used ADAM17 hypomorphic mice, which only express ∼5% of ADAM17 wild-type levels in all tissues and show virtually no shedding of all tested ADAM17 substrates, to clarify the role of ADAM17 during local inflammation in the murine air pouch model. In the present study, we demonstrate that although IL-6 and the trans-signaling mechanism is mandatory for cellular infiltration in this model, it is not ADAM17-mediated shedding of IL-6R within the pouch that orchestrates this inflammatory process. Instead, we demonstrate that sIL-6R is infiltrating from the

circulation in an ADAM17-independent process. Our data suggest that this inﬁltrating sIL-6R, which is needed for IL-6 trans- http://www.jimmunol.org/ signaling, is involved in the controlled resolution of an acute inﬂammatory episode. The Journal of Immunology, 2016, 197: 3705– 3715.

eutrophils are the most abundant cells in human blood and this complex associates with the ubiquitously expressed re- from where they rapidly infiltrate tissues upon wounding, ceptor gp130 (5). IL-6 only shows measurable affinity to IL-6R N infection, and inflammation (1). Resolution of the neu- but not to gp130. Also, IL-6R alone does not bind to gp130. Only trophil infiltrate is associated with subsequent recruitment of the complex of IL-6 and IL-6R binds to and activates the gp130 mononuclear cells (2). We have shown earlier that in acute bac- protein. Because IL-6R is only expressed on few cell types of the by guest on October 2, 2021 terial peritonitis, IL-6 signaling orchestrates the switch from the human body, only these cell types can be stimulated by IL-6 (6). neutrophilic to the mononuclear phase of the inflammatory pro- However, a soluble form of IL-6R (sIL-6R) can bind to IL-6 and cess and thereby contributes to the resolution of the inflammatory associate with gp130 on cells not expressing the membrane-bound process (2). IL-6R. Such cells in the absence of sIL-6R would not be able to IL-6 is an inflammatory cytokine, the level of which is elevated respond to IL-6. This paradigm has been called IL-6 trans- in most if not all inflammatory states (3, 4). IL-6 binds to IL-6R, signaling (7). Furthermore, a soluble form of gp130 (sgp130) generated via *Institute of Biochemistry, Medical Faculty, Christian Albrechts University, 24098 alternative splicing (8) is present in the circulation (9) and represents Kiel, Germany; and †Institute of Biochemistry and Molecular Biology II, Medical the natural inhibitor of IL-6 trans-signaling (10). A recombinant Faculty, Heinrich Heine University, 40225 Dusseldorf,€ Germany form of sgp130 dimerized by a human IgG1 Fc (sgp130Fc) was 1Current address: Department of Biomedical Sciences and Biotech Research and shown to be an even 10- to 50-fold more effective inhibitor of IL-6 Innovation Center, University of Copenhagen, Copenhagen, Denmark. trans-signaling than was monomeric sgp130 (10, 11). ORCID: 0000-0003-4939-6950 (C.G.); 0000-0002-7519-3279 (S.R.-J.). Using sgp130Fc as a molecular tool, it has been demonstrated Received for publication May 25, 2016. Accepted for publication August 31, 2016. that IL-6 responses via the membrane-bound IL-6R are rather This work was supported by Deutsche Forschungsgemeinschaft (Bonn, Germany) protective and regenerative whereas IL-6 trans-signaling appears to be Grant CRC877 (Proteolysis as a Regulatory Event in Pathophysiology, projects A1 and A10, and the Cluster of Excellence “Inflammation at Interfaces”). involved in the proinflammatory activities of this cytokine (6, 12). We have previously shown that IL-6 trans-signaling plays a N.S., A.C., S.R.-J., and B.R. designed the study, collected data, performed the statistical analysis, and interpreted the data; S.R.-J. wrote the manuscript; and S.S., J.S., decisive role in the switch from the neutrophilic to the mononu- D.D., J.L., J.S., and C.G. performed measurements and analyzed data. All coauthors clear phase (13, 14) in the mouse air pouch model (15, 16). Upon critically reviewed all versions of the manuscript. injection of the irritant carrageenan, local sIL-6R levels increased Address correspondence and reprint requests to Prof. Stefan Rose-John, Institute of during the course of inflammation, whereas neutralizing sIL-6R Biochemistry, Christian Albrechts University, Olshausenstrasse 40, 24098 Kiel, Ger- many. E-mail address: [email protected] activity with a specific Ab significantly reduced the recruitment of The online version of this article contains supplemental material. mononuclear cells (13). Furthermore, we generated mice in which Abbreviations used in this article: DppI, dipeptidyl peptidase I; PR3, proteinase 3; transgenic overexpression of sgp130Fc resulted in blockade of IL-6 sgp130, soluble form of gp130; sgp130Fc, recombinant form of sgp130 dimerized by trans-signaling. Consequently, invasion of immune cells into the 2/2 a human IgG1 Fc; sIL-6R, soluble form of IL-6R; sTNFRI, soluble form of TNFRI; inflamed air pouch was impaired as seen in IL-6 animals, in- wt, wild-type. dicating that IL-6 trans-signaling is essential for correct cellular Copyright Ó 2016 by The American Association of Immunologists, Inc. 0022-1767/16/$30.00 infiltration (14). As depletion of neutrophils led to decreased sIL-6R www.jimmunol.org/cgi/doi/10.4049/jimmunol.1600909 3706 IL-6 TRANS-SIGNALING AND INFLAMMATION levels within the air pouch, it has been speculated that sIL-6R CD45+CD11b+Ly6G2Ly6Clow cells, and inflammatory monocytes as + + 2 high generation takes place locally via ADAM17-mediated shedding CD45 CD11b Ly6G Ly6C cells (23, 30). from infiltrating neutrophils (17). ELISAs In humans, sIL-6R levels of 40–75 ng/ml are found, which together with sgp130 levels of ∼400 ng/ml constitute a buffer for Cytokines and cytokine receptors (murine sIL-6R, IL-6, KC, MCP-1, and a soluble form of TNFRI [sTNFRI]) were measured by ELISA (R&D Sys- IL-6 in the blood (3). Interestingly, a single nucleotide polymor- tems, Wiesbaden-Nordenstadt, Germany). Human sIL-6R was measured phism has been identified, which leads to 50% higher sIL-6R by ELISA as described (13). levels (18). This IL-6R single nucleotide polymorphism was as- Western blot analysis sociated with a significant decreased risk of coronary heart disease events (19, 20), which was explained with the higher capacity of Cells were lysed in 50 mM Tris, 150 mM NaCl2,2mMEDTA,and1% 3 the IL-6 buffer in the blood (21). IGEPAL (v/v) with 1 cOmplete protease inhibitors (Roche Diagnos- ex/ex tics, Mannheim, Germany) and 10 mM 1,10-phenanthroline (Sigma- We have recently developed ADAM17 hypomorphic (ADAM17 ) Aldrich, St. Louis, MO). Proteins were separated by SDS-PAGE, mice, which in all investigated tissues expressed only ∼5% of transferred to polyvinylidene difluoride membranes (Millipore, Darmstadt, ADAM17 levels (22). Using these mice, we could show that Germany), and detected with Abs against ADAM17 (18.2 or 10.1) during bacterial infection, IL-6R cleavage by ADAM17 is induced (23, 31), ADAM10 (Pin1) (23), and b-actin (Sigma-Aldrich, Deisenhofen, on the surface of different leukocyte populations. ADAM17 was Germany). also required for IL-6R cleavage and release during endotoxemia Microvesicle preparation (23). Furthermore, it was demonstrated that lung macrophages To clear supernatants from microvesicles, supernatants were subjected to shed IL-6R from the cell surface during bleomycin-induced pul- ultracentrifugation at 100,000 3 g for 2 h at 4˚C (Beckman Optima TLX). monary fibrosis in an ADAM17-dependent manner (24). In this Downloaded from study, we analyzed whether ADAM17-dependent IL-6R shedding drives IL-6 trans-signaling and is needed for the transition from Results the neutrophilic to the mononuclear phase during acute inflam- Genetic deletion of ADAM10 and ADAM17 in myeloid cells mation. However, unexpectedly, we demonstrate that ADAM17 Injection of sterile air in the back induces formation of an air pouch activity is not required for the coordination of this cellular infil- lined by cells equivalent to endothelial cells. Subcutaneous ad- tration. The sIL-6R needed for IL-6 trans-signaling is not gener- ministration of an irritant such as carrageenan leads to the infil- http://www.jimmunol.org/ ated within the air pouch but rather infiltrates from the circulation tration of neutrophils followed by mononuclear cells. The model in an ADAM17-independent manner. allows for the analysis of invading cells and of soluble mediators, which accumulate within the air pouch (15, 16). We have previ- Materials and Methods ously demonstrated that IL-6 and in particular IL-6 trans-signaling Mice plays an important role in the precise orchestration of the neutrophilic and mononuclear phase of the local inflammatory Hypomorphic ADAM17 mice (ADAM17ex/ex) and IL-6R–deficient mice 2 2 process (13, 14). Additionally, it was speculated that shedding (IL-6R / ) have been described (22, 25, 26). LysMcre 3 ADAM10flox/flox Dmyeloid cre flox/flox Dmyeloid of the IL-6R from neutrophils at the site of inflammation by the (ADAM10 ) and LysM 3 ADAM17 (ADAM17 ) by guest on October 2, 2021 mice were homozygous for the floxed Adam10 or Adam17 allele, respec- metalloprotease ADAM17 was responsible for local sIL-6R tively, and heterozygous for the Cre recombinase under the control of the generation, thereby driving the resolution of the inflammatory lysozyme M promoter (27). Dipeptidyl peptidase I (DppI) knockout insult (17). (DppI2/2) mice and proteinase 3 (PR3) knockout (PR32/2) mice were described previously (28, 29). Animal experiments were approved by the As ADAM17 is the major IL-6R sheddase in vivo (23) and the local Institutional Review Committee and performed with 8- to 14-wk-old air pouch infiltrate at the investigated time points mostly consists mice on a C57BL/6J background. of neutrophils and mononuclear cells, we crossed ADAM17flox/flox mice with LysMcre mice (27) to accomplish ADAM17 deletion in Air pouch model myeloid cells (ADAM17Dmyeloid) (32). However, as shown in The air pouch model was performed as described (14). Blood was collected Fig. 1A, we found no significant reduction in ADAM17 protein from the submandibular vein. Pouches were washed with 3 ml of 2 mM levels in cells recovered from the air pouch of ADAM17Dmyeloid EDTA/PBS, which was immediately cooled on ice. For serum preparation, blood was incubated overnight at 4˚C and centrifuged at 1000 3 g for mice 72 h after carrageenan injection. Importantly, this was not cre 10 min at 4˚C. Recombinant human sIL-6R (10 mg) was administered i.p. due to impaired LysM activity toward the ADAM17 gene in 6 h prior to analysis. Recombinant sgp130Fc (100 mg) was injected i.p. 6 h general, because an efficient knockout of ADAM17 at protein prior to carrageenan injection. level was observed in bone marrow–derived macrophages of the Dmyeloid Generation of bone marrow–derived macrophages same ADAM17 animals (Fig. 1B). In keeping with this, Horiuchi et al. (32) reported efficient ADAM17 deletion in bone Bone marrow was collected from femur and tibia of 8-wk-old mice and cells marrow–derived macrophages of ADAM17Dmyeloid mice. More- were differentiated in DMEM supplemented with 10% FCS, 1 mM sodium Dmyeloid pyruvate, 2 mM L-glutamine, 10 mM HEPES, 100 U/ml penicillin, 100 mg/ml over, using ADAM10 mice (33), we likewise found no streptomycin, and 40 ng/ml murine M-CSF (ImmunoTools, Friesoythe, visible reduction in ADAM10 protein expression in cells recov- 6 Germany). After 7 d, 1 3 10 bone marrow–derived macrophages were ered from inflamed air pouches (Supplemental Fig. 1), although seeded on six-well plates and lysed after 24 h. deletion of the ADAM10 gene in bone marrow–derived macro- Flow cytometry phages was effective (23). We conclude from these experiments that LysMcre mice are not suited to drive deletion of ADAM10 or Cells (0.5 3 106) were used for FACS analysis. Blood and pouch cells were incubated with 1:100 mouse serum and Fc Block (anti-mouse CD16/ ADAM17 genes in myeloid cells infiltrating the murine air pouch. ex/ex 32; BioLegend, San Diego, CA) in FACS buffer for 15 min to minimize We therefore turned to hypomorphic ADAM17 mice (ADAM17 unspecific Ab binding. Staining was performed on ice with fluorochrome- mice), which express only 5% of ADAM17 mRNA levels due to conjugated mAbs at concentrations suggested by the providers. Fluorochrome- insertion of an artificial exon harboring an in-frame translational labeled anti-CD45 mAb (30-F11), CD11b mAb (M1/70), Ly6G (1A8), anti-Ly6C mAb (HK1.4), CD62L mAb (MEL-14), and IL-6Ra/CD126 stop codon into the ADAM17 gene (22). We have previously mAb (D7715A7) were purchased from BioLegend. Neutrophilic shown that in these mice shedding of various ADAM17 substrates, granulocytes were defined as CD45+CD11b+Ly6Ghigh cells, monocytes as including IL-6R, is severely compromised (22, 23, 34–36). As The Journal of Immunology 3707

Inflammatory infiltration depends on IL-6 signaling but not on ADAM17 We first determined IL-6R expression on the cell surface of pe- ripheral neutrophils as well as Ly6Clow and Ly6Chigh monocytes isolated from wild-type (wt) mice. As we have previously shown (23, 30), flow cytometric measurement revealed IL-6R expression on all cell types analyzed with high-level expression on Ly6Chigh inflammatory monocytes (Fig. 2A; quantified in Fig. 2B). Inter- estingly, analysis of IL-6R expression on the cell surface of the same cell types during inflammatory activation by the irritant carrageenan revealed a strong trend to decreased IL-6R levels upon transmigration into the air pouch cavity (Fig. 2), although it only reached significance in the case of monocytes. A similar result was found when we analyzed the cell surface expression of L-selectin (CD62L), an additional well-studied substrate of ADAM17 (37, 38). On circulating neutrophils and inflammatory monocytes from ADAM17ex/ex mice, cell surface expression of L-selectin was significantly higher as compared with wt mice. Ad-

ditionally, on all cell types isolated from the air pouch, L-selectin Downloaded from expression was signiﬁcantly higher in the absence of ADAM17 due to the lack of shedding activity (Supplemental Fig. 2). We speculate that the observed downmodulation of membrane- bound IL-6R and L-selectin from the surface of neutrophils and monocytes during transmigration is associated with the generation

of the soluble form by ADAM17-mediated proteolytic cleavage of http://www.jimmunol.org/ the membrane-bound IL-6R. To determine whether innate immune cells shed the membrane- bound IL-6R during transendothelial migration into inflamed areas, we studied cell migration in the air pouch model in hypomorphic ADAM17ex/ex mice. We first analyzed the total number of cells found in the air pouch 4 and 72 h after injection of carrageenan. The cellular infiltrate 4 h after carrageenan injection mainly contains neutrophils, whereas macrophages represent the dominant cell population at 72 h after by guest on October 2, 2021 induction of inflammation (13, 14). As shown in Fig. 3A, no difference in total cell number was detected in ADAM17ex/ex mice compared with wt mice. Because evidence for the importance of sIL-6R in the regulation of cell migration in the air pouch model has been provided only by inhibitor studies (13, 14), we additionally performed this model using IL-6R2/2 mice (26). As shown in Fig. 3B, the total cell number recovered from the air pouch was significantly reduced in IL-6R2/2 animals, which is in line with earlier studies using sIL-6R inhibitors (13, 14). Taken together, although ADAM17 seems not to be essential for cellular infiltration, the regulatory role of the IL-6R could be confirmed. We next investigated the impact of ADAM17 and IL-6R on the cellularity of neutrophils, mononuclear cells, and inflammatory monocytes in the blood and in the air pouch 72 h upon carrageenan 2/2 FIGURE 1. Deletion of ADAM17 in isolated cells from the air pouch. injection. In IL-6R mice, we found significantly fewer neu- Cells from air pouches were isolated 72 h after carrageenan injection and trophils in the blood whereas the numbers of monocytes and ADAM17 expression was monitored by Western blot. (A) Representative Ly6Chigh inflammatory monocytes remained unchanged (Fig. 3C). ADAM17 Western blot analysis of cells isolated from air pouch exudates Consistent with reduced total cell numbers in IL-6R2/2 mice, we of ADAM17fl/fl or ADAM17Dmyeloid mice. (B) Bone marrow–derived Dmyeloid also found fewer neutrophils, mononuclear cells, and inflamma- macrophages isolated from ADAM17 and control mice were lysed tory monocytes in the inflamed air pouch, although only mono- and ADAM17 expression was monitored by Western blot. (C) Represen- cytes and Ly6Chigh inflammatory monocytes reached statistical tative ADAM17 Western blot analysis of cells isolated from air pouch exudates of ADAM17ex/ex mice and wt control mice. significance (Fig. 3D). In the absence of ADAM17, fewer neutrophils, mononuclear cells, and inflammatory monocytes were detected in the blood, but only differences in monocyte and shown in Fig. 1C, ADAM17 protein levels are drastically reduced Ly6Chigh inflammatory monocyte numbers were statistically sig- in cells from air pouches of ADAM17ex/ex mice, indicating that nificant (Fig. 3E). As expected, as total cell numbers in the the role of ADAM17 in the regulation of immune cell migration in inflamed pouch were identical in the absence of ADAM17 (Fig. 3A), this inflammatory model can be studied in these mice. no changes in the respective immune cell subpopulations were 3708 IL-6 TRANS-SIGNALING AND INFLAMMATION

FIGURE 2. IL-6R cell surface expression on leukocytes from blood and air pouch of hypomorphic ADAM17ex/ex and control mice. (A) IL-6R cell surface expression on neutrophils, monocytes, and Ly6Chigh inflammatory monocytes from the blood and air pouch cavity of ADAM17wt/wt and ADAM17ex/ex mice was monitored by flow cytom- Downloaded from etry. (B) IL-6R mean fluorescence intensity (MFI) of leukocytes from blood and air pouch of ADAM17wt/wt (n = 4) and ADAM17ex/ex (n =5) mice. Significance was calculated using two-way ANOVA with Bonferroni , , posttests. *p 0.05, ***p 0.001. http://www.jimmunol.org/ by guest on October 2, 2021

seen in ADAM17ex/ex mice either (Fig. 3F). Note that the cell Proteolytic processing of IL-6R levels of wt mice shown in Fig. 3D and 3F are different. The Although we had seen earlier with transfected cell lines over- 2/2 reason is that the genetic background of the IL-6R mice is expressing human or murine IL-6R that murine ADAM17 did only ex/ex C57BL/6 whereas the ADAM17 mice are on a mixed 129/ inefficiently shed the murine IL-6R (39), we have in the meantime C57BL/6 background, and we have used wt mice on a matched established in vitro (40) and in vivo (23) that ADAM17 is largely genetic background. responsible for stimulated IL-6R shedding. Therefore, we ana- Inflammatory infiltration depends on IL-6 trans-signaling lyzed sIL-6R levels in the inflamed air pouch. Concentrations of We further asked whether the leukocyte influx remained dependent sIL-6R increased during the course of inflammation, with sIL-6R on IL-6 trans-signaling or whether compensatory mechanisms protein levels in the air pouch being higher at 72 h than after 4 h have arisen due to the lack of ADAM17. As shown in Fig. 4A, we following carrageenan injection (Fig. 5A). Although no differ- injected ADAM17ex/ex mice 6 h prior to carrageenan injection ences between wt and hypomorphic mice were observed 4 h after with 100 mg of recombinant sgp130Fc protein, which selectively induction of inflammation, sIL-6R levels were moderately, but ex/ex blocks IL-6 trans-signaling (10). The sgp130Fc protein was significantly, reduced in ADAM17 mice 72 h after carra- readily detectable in the serum of the mice, and in accordance geenan injection (Fig. 5A). Furthermore, we did not detect a with our earlier report (14) ∼5% of the blood levels were detected compensatory upregulation of the ADAM10 protein in the absence in the air pouch (Fig. 4B). The total number of infiltrated cells of ADAM17 (Supplemental Fig. 3), indicating that during the after 72 h was significantly decreased in sgp130Fc-injected mice course of the air pouch experiment ADAM10 seems not to com- (Fig. 4C). The decrease was comparable to the one seen in IL- pensate for the loss of the ADAM17 protease. Given the crucial 6R2/2 mice, suggesting that IL-6 acted via IL-6 trans-signaling. role of ADAM17 as the major inducible IL-6R sheddase, we were Accordingly, the ratios of neutrophils, monocytes, and inflam- surprised by the relatively modest effect of ADAM17 deficiency matory monocytes were similar to the experiment with IL-6R2/2 on sIL-6R release and hypothesized that an alternative protease mice (Figs. 3B, 3D, 4D). Surprisingly, the levels of the chemokine may also contribute to IL-6R processing in this inflammatory CCL2 remained unchanged in this experimental setting (Fig. 4E). context. We conclude from this experiment that the leukocyte influx Because IL-6R can be readily shed from dying neutrophils and remained dependent on IL-6 trans-signaling and that no com- neutrophils represent the predominant cell population especially in pensatory mechanisms had arisen in the absence of ADAM17. the early phase of this inflammatory model, we assessed whether The Journal of Immunology 3709 Downloaded from http://www.jimmunol.org/ by guest on October 2, 2021

FIGURE 3. Cellularity in air pouch and blood during acute inflammation. Cells from inflamed air pouches were isolated at indicated time points and counted using a Cellometer automated cell counter. (A) Cell counts of infiltrated pouch cells isolated from ADAM17wt/wt (4 h, n = 12; 72 h, n = 18) and ADAM17ex/ex (4 h, n =9;72h,n = 17) mice after carrageenan injection at indicated time points. (B) Total cell numbers of infiltrated pouch cells isolated from IL-6R+/+ (n = 9) and IL-6R2/2 (n = 7) mice 72 h after injection with carrageenan or PBS (control, n = 3). Neutrophils, monocytes, and Ly6Chigh inflammatory monocytes from blood and air pouches 72 h after carrageenan injection were analyzed by flow cytometry. Leukocyte populations in the blood (C) and the air pouch (D) of IL-6R2/2 mice are shown, as are leukocyte populations in the blood (E) and the air pouch (F) of hypomorphic ADAM17ex/ex mice. Significance was calculated using a Student t test. *p , 0.05. neutrophil-derived serine proteases contribute to sIL-6R accu- PR3 activity in DPPI2/2 mice is not complete, as opposed to mulation within the inflamed air pouch. Neutrophil proteases have cathepsin G and neutrophil elastase (43). However, in PR32/2 previously been implicated in the shedding of the IL-6R (41). We mice, sIL-6R serum levels (Fig. 5E) and levels in the inflamed air performed the air pouch model with mice deficient in DppI. This pouch (Fig. 5F) were not changed as compared with those in wt protease is critical for activation of the neutrophil proteases neu- mice. In conclusion, neither neutrophil-derived serine proteases trophil elastase, PR3, and cathepsin G, and consequently their neutrophil elastase, cathepsin G, and PR3 nor ADAM17 appears proteolytic activity is severely reduced in DPPI2/2 mice (42). As to be responsible for inflammation-induced sIL-6R release in shown in Fig. 5B, in DppI2/2 mice the total number of infiltrating this model. These results indicate that although the absence cells 72 h after carrageenan injection was comparable to that in wt of ADAM17 resulted in lower sIL-6R levels in the inflamed mice, indicating that neutrophil proteases are not involved in the air pouch, this absence was not sufficient to impair cellular regulation of cellular infiltration in this local inflammation model. infiltration. Likewise, sIL-6R serum levels (Fig. 5C) and levels in the inflamed We have recently shown that IL-6R is also present on circulating air pouch (Fig. 5D) remained unchanged in the absence of DppI. microvesicles, establishing microvesicle release as a distinct route We included PR32/2 mice in our study, because abrogation of of sIL-6R generation (44, 45). Microvesicle-associated IL-6R can 3710 IL-6 TRANS-SIGNALING AND INFLAMMATION

FIGURE 4. Diminished cell infiltration in ADAM17ex/ex mice after inhibition of IL-6 trans-signaling. (A) Air pouches were generated by injection of sterile air at day 0 and day 3. Six hours prior to induction of inflammation with carrageenan at day 6, human recombinant sgp130Fc protein was injected i.p. and cell infiltration into the air pouch cavity was analyzed after 72 h. (B) gp130Fc levels in blood and inflamed air pouches of gp130Fc- or PBS (control)-treated mice. (C) Total cell numbers of infiltrated pouch cells isolated from hypomorphic ADAM17ex/ex mice 72 h after in- Downloaded from jection with carrageenan treated with either gp130Fc or PBS (control). (D) Neutrophils, monocytes, and Ly6Chigh inflammatory monocytes from air pouches of ADAM17ex/ex mice treated with either gp130Fc or

PBS (control) were analyzed by ﬂow http://www.jimmunol.org/ cytometry. (E) CCL2 levels in the air pouches of ADAM17ex/ex treated with either gp130Fc or PBS (control) mice were measured by ELISA. Data depict the mean of n = 4 mice per group. Signiﬁcance was calculated using a Student t test. *p , 0.05. by guest on October 2, 2021

be distinguished from proteolytically cleaved sIL-6R by ultra- thereby influence local sIL-6R concentrations. sTNFRI levels were centrifugation at 100,000 3 g, a treatment that leads to pelleting significantly lower in the serum of ADAM17ex/ex mice as com- of microvesicle-associated IL-6R (44, 45). As shown in pared with wt mice (Fig. 6B), which confirmed that steady-state Supplemental Fig. 4, quantitative analysis by ELISA revealed that levels of sTNFRI are in fact controlled by ADAM17. However, air pouch levels of sIL-6R (Supplemental Fig. 4A) as well as 72 h after carrageenan injection, no inflammation-induced rise in ex/ex sTNFRI (Supplemental Fig. 4B) were not significantly different serum sTNFRI was observed in either wt or ADAM17 ani- before and after ultracentrifugation. This excludes the possibility mals, verifying that, similar to sIL6R, elevated sTNFRI levels in that the observed rise in sIL-6R levels in the inflamed air pouch the inflamed air pouch are not due to globally increased sTNFRI was due to increased microvesicle release. concentrations. Moreover, increased local sIL-6R concentrations in the inflamed air pouch cannot be explained by an increase in sIL-6R and sTNFR in the blood I global soluble receptor levels throughout the experimental period. As reported before, the steady-state levels of circulating sIL-6R in ADAM17ex/ex mice are not significantly different from those of wt IL-6, CXCL1, and CCL2 levels in the air pouch mice. In contrast, short-term cellular release of sIL-6R, which can We next asked whether the levels of relevant cytokines and che- be measured in the serum 1 h after LPS injection, is strictly de- mokines were affected in the gene-deficient animals used in this pendent on ADAM17 and is completely abrogated in ADAM17ex/ex study. As shown in Fig. 7A, in the presence or absence of mice (23). Because in mice no alternative splicing mechanism of ADAM17, IL-6 levels were high 4 h after carrageenan injection the IL-6R mRNA exists, which might lead to the generation of and decreased after 72 h. There was a tendency toward lower sIL-6R protein (45), these findings suggest that proteases other CXCL1 levels in the absence of ADAM17, although these dif- than ADAM17 might contribute to steady-state levels of circu- ferences were not significant (Fig. 7B, left panel). We detected no lating sIL-6R. To exclude the possibility that the local inflam- difference in CCL2 levels in the absence or presence of ADAM17 mation within the air pouch could have an impact on circulating (Fig. 7B, right panel). Interestingly, also in IL-6R2/2 mice, no sIL-6R levels, we gauged sIL-6R serum levels during the time significant differences in the levels of IL-6, CXCL1, and CCL2 course of the air pouch experiment. As shown in Fig. 6A, levels of were detected in the air pouch as compared with wt mice sIL-6R were similar in wt and ADAM17ex/ex mice throughout the (Fig. 7C). Moreover, we observed that two additional monocyte- experimental period, essentially ruling out the possibility that attracting chemokines, CCL5 and CCL12, were differentially elevated sIL-6R levels in the blood leak into the air pouch and regulated during the course of the air pouch experiment. Whereas The Journal of Immunology 3711

FIGURE 5. sIL-6R levels in the air pouch of protease-deficient mice. Levels of sIL-6R in inflamed air pouches are shown. (A) Levels of sIL-6R protein in inflamed air pouches of ADAM17wt/wt (4 h, n = 12; 72 h, n = 18) and ADAM17ex/ex (4 h, n = 11; 72 h, n = 17) mice were assessed by ELISA 4 or 72 h after carrageenan injection. (B) Total cell numbers in inflamed air pouches of DppI-deficient mice. Levels of sIL- Downloaded from 6R protein in the blood (C) and in inflamed air pouches (D) of DppI+/+ (n = 5) and DppI2/2 (n = 5) mice were assessed by ELISA 72 h after carrageenan injection. Levels of sIL- 6R protein in the blood (E) and in inflamed air pouches (F) of PR3+/+ http://www.jimmunol.org/ (n = 4) and PR32/2 (n = 4) mice were assessed by ELISA 72 h after carrageenan injection. Significance was calculated using a Student t test. ***p , 0.0005. by guest on October 2, 2021

CCL5 was downregulated in ADAM17ex/ex mice as compared Mice transgenic for sgp130Fc show levels of ∼30 mg/ml human with wt mice, CCL5 levels remained unchanged in IL-6R2/2 sgp130Fc protein in the blood (14). Interestingly, increasing levels of mice. Conversely, CCL12 was downregulated in IL-6R2/2 mice sgp130Fc were found within the air pouch 4 and 72 h after irritant as compared with wt mice, whereas CCL12 levels were un- application, indicating progressive leakage of plasma protein into the changed in ADAM17ex/ex mice (data not shown). As has been air pouch after carrageenan injection (Fig. 8A). Because the dimeric observed in many studies, chemokines seem to functionally sgp130Fc protein has a molecular mass of .200 kDa, it is not well compensate for each other (46) and there seem to be differences comparable to the 55-kDa sIL-6R. We therefore chose to directly between ADAM17ex/ex mice and IL-6R2/2 mice. Our results in- measure the infiltration of sIL-6R from the blood into the air pouch. dicate that in our particular model the milieu of relevant cytokines To avoid an influence of the injected recombinant sIL-6R on the and chemokines within the air pouch is not decisive for the impaired inflammatory process, we chose to inject the recombinant human cellular infiltration response in the absence of IL-6 signaling in sIL-6R, which does not interact with murine IL-6 and therefore IL-6R2/2 mice (Fig. 3B) or in the absence of IL-6 trans-signaling shows no biologic activity in the mouse (47). after injection of sgp130Fc protein (Fig. 4). To estimate the endothelial permeability in hypomorhpic ADAM17ex/ex mice, we i.p. injected the recombinant human sIL- ADAM17-independent leakage of plasma sIL-6R 6R 6 h prior to analysis of 72-h-old air pouches (Fig. 8B). As Because IL-6R and in particular IL-6 trans-signaling via sIL-6R shown in Fig. 8C, i.p. injection of 10 mg of human sIL-6R resulted (13, 14) is important for cellular infiltration, but shedding within in similar serum levels of human sIL-6R in wt and ADAM17ex/ex the air pouch by the proteases analyzed was not relevant in this mice. In the air pouches of wt and ADAM17ex/ex mice, levels of model, we reasoned that time-dependent infiltration of the sIL-6R 2.0 and 2.5 ng/ml human sIL-6R were detected after 72 h protein from the blood could be the source of sIL-6R found in the (Fig. 8D). These differences were not significant and reflected the air pouch. We therefore decided to directly analyze the influence slightly higher sIL-6R levels in the blood of ADAM17ex/ex mice. of ADAM17 on the leakage of injected sIL-6R from the blood into Interestingly, the sIL-6R levels detected in the air pouch were the air pouch. about a factor of 20 lower than the levels detected in the blood. 3712 IL-6 TRANS-SIGNALING AND INFLAMMATION

FIGURE 6. Unaltered sIL-6R serum levels during acute inﬂamma- tion. (A) sIL-6R and (B) sTNFRI serum concentrations from wt and hypomorphic ADAM17ex/ex mice with air pouches injected with carrageenan for 4 or 72 h were measured by ELISA. Signiﬁcance was calculated using a Student t test. **p , 0.01, ***p , 0.001.

We next asked whether the endothelial permeability changed Discussion during the course of the inflammatory period. We therefore injected There are four major findings in this study. First, ADAM17 is not the recombinant human sIL-6R protein 6 h prior to analysis at 24 important for the orchestration of the cellular infiltration during the and72h.AsshowninFig.8E,atbothtimepoints,thehuman local inflammatory process in the murine air pouch model. Second, sIL-6R levels within the air pouch were identical. We conclude we confirmed earlier indications of the importance of IL-6 trans- from these experiments that the endothelial permeability did not signaling via sIL-6R on the quantity and quality of the cellular Downloaded from change with progression of inflammation. Overall, our data indi- infiltration detected within the air pouch. Third, we could exclude cate that substantial leakage of proteins from the blood to the site a role for the neutrophil-derived serine proteases neutrophil elas- of inflammation occurred after injection of the irritant and that this tase, PR3, and cathepsin G both in sIL-6R release and orches- leakage was not affected by the presence or absence of the tration of the cellular influx in this model. Finally, we show that ADAM17 protease. most of the sIL-6R needed for trans-signaling is not generated by http://www.jimmunol.org/ by guest on October 2, 2021

FIGURE 7. IL-6 and chemokine levels in hypomorphic ADAM17ex/ex and IL-6R2/2 mice. Levels of IL-6 (A), CXCL1 (B, left panel), and CCL2 (B, right panel) in the air pouch of hypomorphic ADAM17ex/ex and wt mice were measured by ELISA 4 and 72 h after injection with carrageenan. Levels of IL-6 (C, left panel), CXCL1 (C, middle panel) and CCL2 (C, right panel) in the air pouch of IL-6R2/2 and wt mice were measured by ELISA 72 h after injection with carrageenan or PBS (control). Signiﬁcance was calculated using a Student t test. ***p , 0.0001. The Journal of Immunology 3713

FIGURE 8. Increased serum protein levels at site of infection with progression of acute inﬂammation. (A) Levels of transgenic sgp130Fc in the inﬂamed air pouch of sgp130Fc transgenic mice 4 and 72 h after carrageenan injection were assessed by ELISA. (B) Air pouch model course after stimulation with carrageenan and i.p. injection of recombinant human sIL-6R. (C and D) ex/ex

Hypomorphic ADAM17 (n =5) Downloaded from or control (n = 7) mice were injected i.p. with 10 mg of recombinant human sIL-6R 6 h prior to analysis of 72-h-old air pouches. Human sIL-6R protein levels in the blood (C)and in inﬂamed air pouches (D)72h after carrageenan stimulation were http://www.jimmunol.org/ assessed by ELISA. (E) C57BL/6J mice were injected i.p. with 10 mg of recombinant human sIL-6R 6 h prior to analysis of 24-h-old (n = 10) and 72-h-old (n = 12) air pouches. Signiﬁcance was calculated using a Student t test. **p , 0.005. hsIL-6R, recombinant human sIL-6R. by guest on October 2, 2021

local activation of the metalloprotease ADAM17 and subsequent air pouch is generated by leakage of plasma protein from the proteolytic release of the IL-6R ectodomain, but rather it enters blood. Indeed, we found that the serum protein sgp130Fc in the site of inflammation via leakage from the blood, a process that sgp130Fc transgenic mice increased with time at the site of in- is apparently independent of ADAM17 activity. flammation. Furthermore, the i.p. injection of the biologically We postulated earlier that upon local inflammation, ADAM17 on inert human recombinant sIL-6R led to substantial amounts of this apoptotic neutrophils was activated, resulting in IL-6R cleavage protein at the site of inflammation, although the extent of infil- and subsequent release of sIL-6R into the air pouch (13, 17). Other tration was not dependent on the time point of injection. This inflammatory states such as bacterial peritonitis were additionally finding indicates that once initiated by carrageenan injection, the hypothesized to depend on the sIL-6R generated by infiltrating permeability of the endothelium does not change during the ac- neutrophils (2, 48). Such a mechanism would serve as a gauge to cumulation of blood proteins at the site of inflammation. measure the extent of tissue damage or inflammation and recruit That LysMcre transgenic mice could not be used for cellular the appropriate number of mononuclear cells. Our present study ADAM17 and ADAM10 gene ablation in the context of the air pouch revealed that ADAM17 activity is not required for the regulated model is of interest to the scientific community. It indicates that influx of immune cells, whereas the IL-6 response via trans- cells found in the air pouch differ in terms of genetic age from the signaling was confirmed to be critically involved in the orches- bone marrow–derived macrophages, in which efficient ablation of tration of cellular infiltration. Although we measured some ADAM17 and ADAM10 genes was achieved. The cells found in the ADAM17-dependent cleavage of IL-6R within the air pouch, this air pouch might have been derived from their cognate circulating amount was not sufficient to affect total cellularity of the in- progenitor cells so recently that the long-lived protease proteins flammatory reaction, because no difference in cellular composi- were still present even though the recombination had already tion of the air pouch exudate was seen in the absence of ADAM17. occurred. This information could be relevant for other studies using Therefore, we hypothesize that most functional sIL-6R within the LysMcre transgenic mice for the ablation of floxed genes. 3714 IL-6 TRANS-SIGNALING AND INFLAMMATION

Interestingly, in a recent study it was shown that ADAM17 is soluble interleukin-6 receptor transsignaling responses. Eur. J. Biochem. 268: 160–167. involved in downregulation of the chemokine receptor CXCR2. 11. Waetzig, G. H., and S. Rose-John. 2012. Hitting a complex target: an update on Downregulation of CXCR2 is associated with impaired neutrophil interleukin-6 trans-signalling. Expert Opin. Ther. Targets 16: 225–236. recruitment (49). Moreover, ADAM17 deficiency in all immune 12. Jones, S. A., J. Scheller, and S. Rose-John. 2011. Therapeutic strategies for the clinical blockade of IL-6/gp130 signaling. J. Clin. Invest. 121: 3375–3383. cells led to improved survival during experimental polymicrobial 13. Chalaris, A., B. Rabe, K. Paliga, H. Lange, T. Laskay, C. A. Fielding, sepsis (50). However, during acute pulmonary inflammation, S. A. Jones, S. Rose-John, and J. Scheller. 2007. Apoptosis is a natural stimulus leukocytes require ADAM10 but not ADAM17 for migration into of IL6R shedding and contributes to the proinflammatory trans-signaling function of neutrophils. Blood 110: 1748–1755. the alveolar space (51). In contrast, we had found that upon 14. Rabe, B., A. Chalaris, U. May, G. H. Waetzig, D. Seegert, A. S. Williams, bacterial infection, the absence of ADAM17 in all cells led to S. A. Jones, S. Rose-John, and J. Scheller. 2008. Transgenic blockade of inter- higher bacteria levels in spleen and liver (23). Mishra et al. (50) leukin 6 transsignaling abrogates inflammation. Blood 111: 1021–1028. 15. Edwards, J. C., A. D. Sedgwick, and D. A. Willoughby. 1981. The formation of a showed that ADAM17 deficiency led to an increase of peritoneal structure with the features of synovial lining by subcutaneous injection of air: an neutrophils. In contrast, Yan et al. (23) found that in infected mice, in vivo tissue culture system. J. Pathol. 134: 147–156. 16. Garcıá-Ramallo, E., T. Marques, N. Prats, J. Beleta, S. L. Kunkel, and granulocytes and inflammatory monocytes were increased. Al- N. Godessart. 2002. Resident cell chemokine expression serves as the major though these are very different experimental settings, it can be mechanism for leukocyte recruitment during local inflammation. J. Immunol. concluded that, depending on the model used, ADAM17 clearly 169: 6467–6473. 17. DeLeo, F. R. 2007. Attractive shedding. Blood 110: 1711–1712. has a substantial influence on the defense of the body against 18. Garbers, C., N. Monhasery, S. Aparicio-Siegmund, J. Lokau, P. Baran, bacterial infection and on the infiltration of immune cells. M. A. Nowell, S. A. Jones, S. Rose-John, and J. Scheller. 2014. The interleukin-6 In conclusion, we show that ADAM17 plays no decisive role in receptor Asp358Ala single nucleotide polymorphism rs2228145 confers increased proteolytic conversion rates by ADAM proteases. Biochim. Biophys. local sIL-6R generation, which rather leaks to the site of inflam- Acta 1842: 1485–1494. Downloaded from mation from the circulation. We further confirmed the importance 19. Interleukin-6 Receptor Mendelian Randomisation Analysis (IL6R MR) Con- of IL-6R for the local inflammatory response, where it directs the sortium. 2012. The interleukin-6 receptor as a target for prevention of coronary heart disease: a Mendelian randomisation analysis. Lancet 379: 1214–1224. infiltration of neutrophils and mononuclear cells to the site of 20. Sarwar, N., A. S. Butterworth, D. F. Freitag, J. Gregson, P. Willeit, D. N. Gorman, inflammation via IL-6 trans-signaling, a process that could be P. Gao, D. Saleheen, A. Rendon, C. P. Nelson, et al; IL6R Genetics Consortium inhibited with anti–IL-6R agents such as tocilizumab (52), but Emerging Risk Factors Collaboration. 2012. Interleukin-6 receptor pathways in coronary heart disease: a collaborative meta-analysis of 82 studies. Lancet also with the sgp130Fc protein, a specific IL-6 trans-signaling 379: 1205–1213. inhibitor, which is currently undergoing clinical trials for the 21. Scheller, J., and S. Rose-John. 2012. The interleukin 6 pathway and athero- http://www.jimmunol.org/ sclerosis. Lancet 380: 338. treatment of inflammatory diseases (11, 53). Apparently, thera- 22. Chalaris, A., N. Adam, C. Sina, P. Rosenstiel, J. Lehmann-Koch, P. Schirmacher, peutic proteins such as the anti–TNF-a agents adalimumab and D. Hartmann, J. Cichy, O. Gavrilova, S. Schreiber, et al. 2010. Critical role of the etanercept or anti–IL-6R agents such as tocilizumab and sgp130Fc disintegrin metalloprotease ADAM17 for intestinal inflammation and regenera- tion in mice. J. Exp. Med. 207: 1617–1624. will readily leak from the blood to the inflamed tissue to modulate 23. Yan, I., J. Schwarz, K. Lucke,€ N. Schumacher, V. Schumacher, S. Schmidt, local inflammatory responses. B. Rabe, P. Saftig, M. Donners, S. Rose-John, et al. 2016. ADAM17 controls IL- 6 signaling by cleavage of the murine IL-6Ra from the cell surface of leukocytes during inflammatory responses. J. Leukoc. Biol. 99: 749–760. Acknowledgments 24. Le, T. T., H. Karmouty-Quintana, E. Melicoff, T. T. Le, T. Weng, N. Y. Chen, M. Pedroza, Y. Zhou, J. Davies, K. Philip, et al. 2014. Blockade of IL-6 trans

We thank Melanie Boss for help with the experiments. We further thank by guest on October 2, 2021 signaling attenuates pulmonary fibrosis. J. Immunol. 193: 3755–3768. Oliver So¨hnlein (Institute for Cardiovascular Prevention, Ludwig Maximi- 25. McFarland-Mancini, M. M., H. M. Funk, A. M. Paluch, M. Zhou, P. V. Giridhar, 2/2 2/2 lians University, Munich, Germany) for the gift of DPPI and PR3 C. A. Mercer, S. C. Kozma, and A. F. Drew. 2010. Differences in wound healing mice and for many fruitful discussions and experimental advice. in mice with deficiency of IL-6 versus IL-6 receptor. J. Immunol. 184: 7219– 7228. 26. Sommer, J., E. Engelowski, P. Baran, C. Garbers, D. M. Floss, and J. Scheller. Disclosures 2014. Interleukin-6, but not the interleukin-6 receptor plays a role in recovery The authors have no financial conflicts of interest. from dextran sodium sulfate-induced colitis. Int. J. Mol. Med. 34: 651–660. 27. Clausen, B. E., C. Burkhardt, W. Reith, R. Renkawitz, and I. Fo¨rster. 1999. Conditional gene targeting in macrophages and granulocytes using LysMcre mice. Transgenic Res. 8: 265–277. References 28. Pham, C. T., and T. J. Ley. 1999. Dipeptidyl peptidase I is required for the 1. Mayadas, T. N., X. Cullere, and C. A. Lowell. 2014. The multifaceted functions processing and activation of granzymes A and B in vivo. Proc. Natl. Acad. Sci. of neutrophils. Annu. Rev. Pathol. 9: 181–218. USA 96: 8627–8632. 2. Hurst, S. M., T. S. Wilkinson, R. M. McLoughlin, S. Jones, S. Horiuchi, 29. Kessenbrock, K., L. Fro¨hlich, M. Sixt, T. La¨mmermann, H. Pfister, A. Bateman, N. Yamamoto, S. Rose-John, G. M. Fuller, N. Topley, and S. A. Jones. 2001. IL- A. Belaaouaj, J. Ring, M. Ollert, R. Fa¨ssler, and D. E. Jenne. 2008. Proteinase 3 6 and its soluble receptor orchestrate a temporal switch in the pattern of leu- and neutrophil elastase enhance inflammation in mice by inactivating antiin- kocyte recruitment seen during acute inflammation. Immunity 14: 705–714. flammatory progranulin. J. Clin. Invest. 118: 2438–2447. 3. Schaper, F., and S. Rose-John. 2015. Interleukin-6: biology, signaling and 30. Hoge, J., I. Yan, N. Ja¨nner, V. Schumacher, A. Chalaris, O. M. Steinmetz, strategies of blockade. Cytokine Growth Factor Rev. 26: 475–487. D. R. Engel, J. Scheller, S. Rose-John, and H. W. Mittrucker.€ 2013. IL-6 controls 4. Wolf, J., S. Rose-John, and C. Garbers. 2014. Interleukin-6 and its receptors: a the innate immune response against Listeria monocytogenes via classical IL-6 highly regulated and dynamic system. Cytokine 70: 11–20. signaling. J. Immunol. 190: 703–711. 5. Rose-John, S., J. Scheller, and F. Schaper. 2015. “Family reunion”—a structured 31. Schwarz, J., C. Broder, A. Helmstetter, S. Schmidt, I. Yan, M. Muller,€ view on the composition of the receptor complexes of interleukin-6-type and D. Schmidt-Arras, C. Becker-Pauly, F. Koch-Nolte, H. W. Mittrucker,€ et al. 2013. interleukin-12-type cytokines. Cytokine Growth Factor Rev. 26: 471–474. Short-term TNFa shedding is independent of cytoplasmic phosphorylation or 6. Scheller, J., A. Chalaris, D. Schmidt-Arras, and S. Rose-John. 2011. The pro- furin cleavage of ADAM17. Biochim. Biophys. Acta 1833: 3355–3367. and anti-inflammatory properties of the cytokine interleukin-6. Biochim. Bio- 32. Horiuchi, K., T. Kimura, T. Miyamoto, H. Takaishi, Y. Okada, Y. Toyama, and phys. Acta 1813: 878–888. C. P. Blobel. 2007. Cutting edge: TNF-a-converting enzyme (TACE/ADAM17) 7. Rose-John, S., and P. C. Heinrich. 1994. Soluble receptors for cytokines and inactivation in mouse myeloid cells prevents lethality from endotoxin shock. J. growth factors: generation and biological function. Biochem. J. 300: 281–290. Immunol. 179: 2686–2689. 8. Diamant, M., K. Rieneck, N. Mechti, X. G. Zhang, M. Svenson, K. Bendtzen, 33. Jorissen, E., J. Prox, C. Bernreuther, S. Weber, R. Schwanbeck, L. Serneels, and B. Klein. 1997. Cloning and expression of an alternatively spliced mRNA A. Snellinx, K. Craessaerts, A. Thathiah, I. Tesseur, et al. 2010. The disintegrin/ encoding a soluble form of the human interleukin-6 signal transducer gp130. metalloproteinase ADAM10 is essential for the establishment of the brain cortex. FEBS Lett. 412: 379–384. J. Neurosci. 30: 4833–4844. 9. Narazaki, M., K. Yasukawa, T. Saito, Y. Ohsugi, H. Fukui, Y. Koishihara, 34. Caolo, V., G. Swennen, A. Chalaris, A. Wagenaar, S. Verbruggen, S. Rose-John, G. D. Yancopoulos, T. Taga, and T. Kishimoto. 1993. Soluble forms of the D. G. Molin, M. Vooijs, and M. J. Post. 2015. ADAM10 and ADAM17 have interleukin-6 signal-transducing receptor component gp130 in human serum opposite roles during sprouting angiogenesis. Angiogenesis 18: 13–22. possessing a potential to inhibit signals through membrane-anchored gp130. 35. Marczynska, J., A. Ozga, A. Wlodarczyk, M. Majchrzak-Gorecka, P. Kulig, Blood 82: 1120–1126. M. Banas, D. Michalczyk-Wetula, P. Majewski, A. Hutloff, J. Schwarz, et al. 10. Jostock, T., J. Mullberg,€ S. Ozbek, R. Atreya, G. Blinn, N. Voltz, M. Fischer, 2014. The role of metalloproteinase ADAM17 in regulating ICOS ligand- M. F. Neurath, and S. Rose-John. 2001. Soluble gp130 is the natural inhibitor of mediated humoral immune responses. J. Immunol. 193: 2753–2763. The Journal of Immunology 3715

36. Vidal, P. M., E. Lemmens, A. Avila, T. Vangansewinkel, A. Chalaris, S. Rose- 45. Schumacher, N., D. Meyer, A. Mauermann, J. von der Heyde, J. Wolf, John, and S. Hendrix. 2013. ADAM17 is a survival factor for microglial cells J. Schwarz, K. Knittler, G. Murphy, M. Michalek, C. Garbers, et al. 2015. in vitro and in vivo after spinal cord injury in mice. Cell Death Dis. 4: e954. Shedding of endogenous interleukin-6 receptor (IL-6R) is governed by a dis- 37. Wang, Y., A. C. Zhang, Z. Ni, A. Herrera, and B. Walcheck. 2010. ADAM17 integrin and metalloproteinase (ADAM) proteases while a full-length IL-6R activity and other mechanisms of soluble L-selectin production during death isoform localizes to circulating microvesicles. J. Biol. Chem. 290: 26059–26071. receptor-induced leukocyte apoptosis. J. Immunol. 184: 4447–4454. 46. Kolaczkowska, E., and P. Kubes. 2013. Neutrophil recruitment and function in 38. Walcheck, B., A. H. Herrera, C. St Hill, P. E. Mattila, A. R. Whitney, and health and inflammation. Nat. Rev. Immunol. 13: 159–175. F. R. Deleo. 2006. ADAM17 activity during human neutrophil activation and 47. Peters, M., S. Jacobs, M. Ehlers, P. Vollmer, J. Mullberg,€ E. Wolf, G. Brem, apoptosis. Eur. J. Immunol. 36: 968–976. K. H. Meyer zum Buschenfelde,€ and S. Rose-John. 1996. The function of the 39. Garbers, C., N. Ja¨nner, A. Chalaris, M. L. Moss, D. M. Floss, D. Meyer, F. Koch- Nolte, S. Rose-John, and J. Scheller. 2011. Species specificity of ADAM10 and soluble interleukin 6 (IL-6) receptor in vivo: sensitization of human soluble IL-6 ADAM17 proteins in interleukin-6 (IL-6) trans-signaling and novel role of receptor transgenic mice towards IL-6 and prolongation of the plasma half-life of ADAM10 in inducible IL-6 receptor shedding. J. Biol. Chem. 286: 14804– IL-6. J. Exp. Med. 183: 1399–1406. 14811. 48. McLoughlin, R. M., J. Witowski, R. L. Robson, T. S. Wilkinson, S. M. Hurst, 40. Matthews, V., B. Schuster, S. Schutze,€ I. Bussmeyer, A. Ludwig, C. Hundhausen, A. S. Williams, J. D. Williams, S. Rose-John, S. A. Jones, and N. Topley. 2003. T. Sadowski, P. Saftig, D. Hartmann, K. J. Kallen, and S. Rose-John. 2003. Interplay between IFN-g and IL-6 signaling governs neutrophil trafficking and Cellular cholesterol depletion triggers shedding of the human interleukin-6 re- apoptosis during acute inflammation. J. Clin. Invest. 112: 598–607. ceptor by ADAM10 and ADAM17 (TACE). J. Biol. Chem. 278: 38829–38839. 49. Mishra, H. K., C. Long, N. S. Bahaie, and B. Walcheck. 2015. Regulation of 41. Lokau, J., R. Nitz, M. Agthe, N. Monhasery, S. Aparicio-Siegmund, CXCR2 expression and function by a disintegrin and metalloprotease-17 N. Schumacher, J. Wolf, K. Mo¨ller-Hackbarth, G. H. Waetzig, J. Gro¨tzinger, (ADAM17). J. Leukoc. Biol. 97: 447–454. et al. 2016. Proteolytic cleavage governs interleukin-11 trans-signaling. Cell 50. Mishra, H. K., T. J. Johnson, D. M. Seelig, and B. Walcheck. 2016. Targeting Reports 14: 1761–1773. ADAM17 in leukocytes increases neutrophil recruitment and reduces bacterial 42. Pagano, M. B., M. A. Bartoli, T. L. Ennis, D. Mao, P. M. Simmons, spread during polymicrobial sepsis. J. Leukoc. BiolDOI: 10.1189/jlb.3VMAB1115- R. W. Thompson, and C. T. Pham. 2007. Critical role of dipeptidyl peptidase I in 496RR. neutrophil recruitment during the development of experimental abdominal aortic 51. Pruessmeyer, J., F. M. Hess, H. Alert, E. Groth, T. Pasqualon, N. Schwarz, aneurysms. Proc. Natl. Acad. Sci. USA 104: 2855–2860. 43. Adkison, A. M., S. Z. Raptis, D. G. Kelley, and C. T. Pham. 2002. Dipeptidyl S. Nyamoya, J. Kollert, E. van der Vorst, M. Donners, et al. 2014. Leukocytes Downloaded from peptidase I activates neutrophil-derived serine proteases and regulates the de- require ADAM10 but not ADAM17 for their migration and inflammatory re- velopment of acute experimental arthritis. J. Clin. Invest. 109: 363–371. cruitment into the alveolar space. Blood 123: 4077–4088. 44. Effenberger, T., J. von der Heyde, K. Bartsch, C. Garbers, K. Schulze-Osthoff, 52. Tanaka, T., M. Narazaki, and T. Kishimoto. 2012. Therapeutic targeting of the A. Chalaris, G. Murphy, S. Rose-John, and B. Rabe. 2014. Senescence- interleukin-6 receptor. Annu. Rev. Pharmacol. Toxicol. 52: 199–219. associated release of transmembrane proteins involves proteolytic processing 53. Calabrese, L. H., and S. Rose-John. 2014. IL-6 biology: implications for clinical by ADAM17 and microvesicle shedding. FASEB J. 28: 4847–4856. targeting in rheumatic disease. Nat. Rev. Rheumatol. 10: 720–727. http://www.jimmunol.org/ by guest on October 2, 2021 ADAM10fl/fl ADAM10Δmyeloid

250

130

100 ADAM10

β-actin

Supplemental figure 1: Deletion of ADAM10 in isolated cells from the air pouch. Cells from air pouches were isolated 72 h after carrageenan injection and ADAM10 expression was monitored by Western blot. Representative ADAM10 Western blot analysis of cells isolated from air pouches of ADAM10fl/fl or ADAM10Δmyeloid mice. A Neutrophils Monocytes Infl. Monocytes

Isotype ADAM17wt/wt

blood ADAM17ex/ex

count pouch

CD62L B 15000 *** *** *** ADAM17wt/wt 10000 *** ADAM17ex/ex

5000 * *** 1000 ***

CD62L (MFI) ** 500

0 blood air pouch blood air pouch blood air pouch Neutrophils Monocytes Infl. Monocytes

Supplemental figure 2: L-selectin (CD62L) cell surface expression on leukocytes from blood and air pouches of hypomorphic ADAM17ex/ex and control mice. (A) CD62L cell surface expression on neutrophils, monocytes and Ly6Chigh inflammatory monocytes from the blood and air pouch cavity of ADAM17wt/wt and ADAM17ex/ex mice was monitored by flow cytometry. (B) CD62L median fluorescence intensity (MFI) of leukocytes from blood and air pouch of ADAM17wt/wt (n=4) and ADAM17ex/ex (n=5) mice. Significance was calculated using 2-way ANOVA with Bonferroni posttests and is indicated by * p<0.05, ** p<0.01 and *** p<0.001. ADAM17wt/wt ADAM17ex/ex 130

100 ADAM10 (low exposure) 70 130

100 ADAM10 70 (high exposure)

β-actin

130 ADAM17 100

β-actin

Supplemental figure 3: ADAM10 protein expression in isolated cells from the air pouch of hypomorphic ADAM17ex/ex mice. Cells from air pouches of hypomorphic ADAM17ex/ex mice were isolated 72 h after carrageenan injection and ADAM10 and ADAM17 expression were monitored by Western blot. A B 4 ** 20 ** ADAM17wt/wt ADAM17wt/wt ADAM17ex/ex ADAM17ex/ex 3 15

2 10 [ng/pouch] I *

1 5 sIL-6R [ng/pouch] sTNFR

0 0 Ø UC Ø UC Ø UC Ø UC Ø UC Ø UC Ø UC Ø UC 4 h 72 h 4 h 72 h

Supplemental figure 4: sIL-6R and sTNFRI within the air pouch is not associated with microvesicles. Levels of soluble IL-6R (sIL-6R) and soluble TNFRI (sTNFRI) in inflamed air pouches before (Ø) and after ultracentrifugation (UC). Air pouch lavages from inflamed air pouches of ADAM17wt/wt and ADAM17ex/ex mice were isolated 4 h or 72 h after carrageenan injection and centrifuged at 100,000 x g for 2 h. Levels of sIL-6R (A) and of sTNFRI (B) were measured by ELISA. Data are represented as mean values of 3 animals. Significance was calculated using Student’s t test and is indicated by * p<0.05 and ** p<0.01.