Inactivate the C5a Receptor Neutrophil Serine Proteases

Mechanism of Neutrophil Dysfunction: Neutrophil Serine Proteases Cleave and Inactivate the C5a Receptor

This information is current as Carmen W. van den Berg, Denise V. Tambourgi, Howard of September 27, 2021. W. Clark, S. Julie Hoong, O. Brad Spiller and Eamon P. McGreal J Immunol 2014; 192:1787-1795; Prepublished online 20 January 2014;

doi: 10.4049/jimmunol.1301920 Downloaded from http://www.jimmunol.org/content/192/4/1787

References This article cites 44 articles, 15 of which you can access for free at: http://www.jimmunol.org/content/192/4/1787.full#ref-list-1 http://www.jimmunol.org/

Why The JI? Submit online.

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

• No Triage! Every submission reviewed by practicing scientists by guest on September 27, 2021 • 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 © 2014 by The American Association of Immunologists, Inc. All rights reserved. Print ISSN: 0022-1767 Online ISSN: 1550-6606. The Journal of Immunology

Mechanism of Neutrophil Dysfunction: Neutrophil Serine Proteases Cleave and Inactivate the C5a Receptor

Carmen W. van den Berg,* Denise V. Tambourgi,† Howard W. Clark,‡ S. Julie Hoong,* O. Brad Spiller,* and Eamon P. McGreal*

Neutrophil dysfunction, resulting in inefficient bacterial clearance, is a feature of several serious medical conditions, including cystic fibrosis (CF) and sepsis. Poorly controlled neutrophil serine protease (NSP) activity and complement activation have been implicated in this phenomenon. The capacity for excess NSP secretion and complement activation to influence the expression and function of the important neutrophil-activating receptor C5aR was investigated. Purified NSPs cathepsin G (CG), neutrophil elastase (NE), and proteinase 3 cleaved C5aR to a 26- to 27-kDa membrane-bound fragment, thereby inactivating its C5a-induced signaling ability. In a supernatant transfer assay, NSPs released from neutrophils in response to C5a induced the cleavage of the C5aR on unstimulated cells. Stimulation of myeolomonocytic U937 cells and purified neutrophils with C5a resulted in downregulation of the C5aR on Downloaded from these cells, which, in the case of U937 cells, was largely caused by NSP-mediated cleavage of C5aR, but in the case of neutrophils, intracellular degradation was likely the main mediator in addition to a small role for NSPs. CG and NE in bronchoalveolar lavage fluid from CF patients both contributed to C5aR cleavage. We propose two converging models for C5a- and NSP-mediated neutrophil dysfunction whereby C5aR cleavage is induced by NSPs, secreted in response to: 1) excess C5a generation or other stimuli; or 2) necrosis. The consequent impairment of C5aR activity contributes to suboptimal local neutrophil priming and bacterial clearance. NSP inhibitors with specificity for both CG and NE may aid the treatment of pathologies associated with neutrophil dys- http://www.jimmunol.org/ function including sepsis and CF. The Journal of Immunology, 2014, 192: 1787–1795.

ell-regulated neutrophil activity is critical for the ef- IL-6, CXCR1, and soluble IL-6R (4, 5)] and innate immune rec- ficient clearance of infectious microbes; however, ognition molecules including the complement receptor (CR) for W uncontrolled neutrophil activity also contributes to C3b (CR1), its ligand C3bi (6, 7), and C-type lectins including neutrophil dysfunction, characterized by impaired clearance of surfactant proteins A and D (8, 9). NSP-mediated loss of CXCR1 pathogens, leading to chronic infection or death. Secreted actively (4) and CR1 (6) on neutrophils results in reduced phagocytosis (as a consequence of cell activation) or passively (as a consequence of Pseudomonas aeruginosa, which may partly explain the high by guest on September 27, 2021 of necrotic cell death), neutrophil serine proteases (NSPs; e.g., prevalence of chronic infection with P. aeruginosa and other neutrophil elastase [NE], cathepsin G [CG], and proteinase 3 [PR3]) bacteria in individuals with neutrophil-dominated airway disease, are major contributors to neutrophil dysfunction. NE is considered including CF. a major contributor to a range of pulmonary pathologies, includ- The importance of the complement system in clearing bacterial ing cystic fibrosis (CF) (1–3), and is an important therapeutic infection is well recognized. C3b opsonizes bacteria for phago- target (3). Under well-controlled conditions in otherwise healthy cytosis via CRs CR1 and CR3. C5a is a powerful chemoattractant individuals, NSPs contribute to the effective control of infection. for neutrophils, but also enhances C3b-induced phagocytosis However, excessive and poorly regulated NSP activity contributes and intracellular killing of bacteria (10, 11). C5a mediates its ef- to disease in a number of ways (3). In addition to their direct fects by interaction with the C5a receptor (C5aR/CD88), a seven- elastolytic activity toward lung tissue, NSPs also have the capacity transmembrane G-protein–coupled receptor (GPCR), which is highly to substantially modulate local inflammatory responses through expressed on cells of myeloid origin (12). Both C5-deficient proteolysis of cytokines and chemokines and their receptors [e.g., and C5aR-knockout mice exhibit impaired lung clearance of P. aeruginosa (13–15). Interestingly, knockout of C5aR reveals a critical role for this receptor, because clearance and control of *Institute of Molecular and Experimental Medicine, School of Medicine, Cardiff University, Cardiff CF14 4XN, United Kingdom; †Immunochemistry Laboratory, bacterial infection and mortality from P. aeruginosa infection is Butantan Institute, Saõ Paulo 05503-900, Brazil; and ‡Department of Child Health, significantly elevated in these mice, despite increased neutrophil Faculty of Medicine, University of Southampton, Southampton SO17 1BJ, United Kingdom influx. Blocking C5aR also inhibits phagocytosis and intracellular Escherichia coli Received for publication July 18, 2013. Accepted for publication December 2, 2013. killing of by human neutrophils (10). The importance of the C5a/C5aR system to antimicrobial defense is S.J.H. received a Wales Heart Research Institute, Cardiff University, professional training year scholarship. underlined by the elaborate strategies evolved by bacteria to avoid Address correspondence and reprint requests to Dr. Carmen W. van den Berg, Insti- the complement system, including blockade of C5a generation tute of Molecular and Experimental Medicine, School of Medicine, Cardiff Univer- and C5aR function (16). However, excess complement activation sity, Cardiff CF14 4XN, U.K. E-mail address: [email protected] and C5a generation is also associated with numerous patholo- Abbreviations used in this article: BALF, bronchoalveolar lavage fluid; CF, cystic gies, including ’09 and Unnewher ‘13 (11, 12). Paradoxically, fibrosis; CG, cathepsin G; CR, complement receptor; CytoD, cytochalasin D; GPCR, G-protein–coupled receptor; KHB, Krebs/HEPES/BSA; NE, neutrophil elastase; excess C5a has been observed to impair neutrophil function NSP, neutrophil serine protease; PMN, polymorphonuclear neutrophil; PR3, protein- and is associated with impaired complement-dependent phago- ase 3. cytosis and reduced C5aR expression both in vivo in animal Copyright Ó 2014 by The American Association of Immunologists, Inc. 0022-1767/14/$16.00 models and ex vivo in cells from human sepsis patients (17–21) www.jimmunol.org/cgi/doi/10.4049/jimmunol.1301920 1788 NEUTROPHIL DYSFUNCTION CAUSED BY NSP CLEAVAGE OF C5aR in whom relative expression of C5aR is suggested to be of prog- and blotted onto Hybond Nitrocellulose (GE Healthcare UK, Little Chalfont, nostic significance (11). U.K.). C5aR was detected as described (22) using mAb S5/1 or polyclonal The mechanisms of C5a-induced neutrophil dysfunction and anti-C5aR. Precision Plus All Blue standards (Bio-Rad, Hemel Hempstead, U.K.) were used to calculate the molecular weights of the bands. regulation of C5aR expression have not yet been elucidated. NSPs have been shown to proteolytically reduce the expression of several Flow cytometry cell-surface molecules involved in immunity, including CXCR1 Cell-surface C5aR expression was detected by flow cytometry as described (IL-8RA), CR1, CD16, CD43, and TNFRII, but their impact on (22) using mAb S5/1, and fluorescence was measured on an Accuri flow C5aR expression has not been investigated. Proteolytic cleavage cytometer (BD Biosciences). Results are expressed as average of mean of 6 and inactivation of C5aR has been demonstrated by a variety of arbitrary fluorescent intensity SD of experiments carried out in triplicate. endogenous and exogenous proteases, including an endogenous Calcium flux measurements metalloprotease, activated by the action of Loxosceles spider venom Cells (at 107c/ml) were loaded with 2 mM fura 2-AM for 30 min at room sphingomyelinase D, a serine protease from Porphyromonas gin- temperature. Cells were washed and resuspended in KHB buffer and incu- givalis, and a metalloprotease from venom of the spider Plec- bated with NSPs for 1 h at 37˚C. After washing, 200 ml cells (5 3 106/ml) treurys tristes (22–24). Direct proteolysis and inactivation of was stimulated with 2.5 nM C5a or 250 nM C5aR agonist at 37˚C. Changes C5aR by an endogenous protease has not been demonstrated so in intracellular calcium were measured as described (22, 26). far; however, a reduced chemotactic response to C5a after NE and Patient recruitment and bronchoalveolar lavage CG treatment has been reported (25). A recent study by Unnewehr et al. (21) suggested that the C5aR may be shed from the cell Four CF BALF samples were taken from a cohort recruited as previously described from children undergoing fiber optic bronchoscopy for diagnostic surface in response to C5a. reasons at Southampton General Hospital (9). The study was approved by Downloaded from Considering the importance of NSPs and C5a in neutrophil dys- the Southampton and Wessex Local Research Ethics Committees, and function and the roles of NSPs and C5a in pathology of neutrophil- written informed consent was obtained. BALF from cohort B was initially m associated pathologies, we aimed to investigate the effects of filtered through a 100- m nylon cell strainer (BD Biosciences) prior to centrifugation at 500 3 g for 10 min at 4˚C. Aspirated cell-free BALF was NSPs on the expression and the function of the C5aR. We inves- immediately stored at 280˚C, and elastase activity in these samples was tigated the effects of purified NSPs, CF bronchoalveolar lavage assayed as previously reported (9).

ﬂuid (BALF), and supernatants of activated neutrophils on C5aR http://www.jimmunol.org/ NE activity expression and function using Western blotting and measurement of intracellular calcium responses. We further investigated the NE activity was measured as described (5). mechanism of C5a-induced downregulation of endogenous C5aR Calculations expression. These data lead us to propose two interlinked models of NSP- and C5a-induced neutrophil dysfunction, which are relevant Calculations of potential cleavage site fragment sizes were carried out using to neutrophil-mediated airway disease and sepsis. C5aR sequence P21730.1 and calculated using the Sequence Manipulation Site m.w. calculation tool available at http://www.bioinformatics.org/sms2/ protein_mw.html. Materials and Methods

Statistics by guest on September 27, 2021 Reagents Statistics were carried out using one-way ANOVA followed by Tukey Abs and other reagents were from the following sources: mouse anti-C5aR multiple comparisons test. Differences were considered significant at (S5/1) and rabbit anti-C5aR (SC-25774: H100) from Santa Cruz Bio- p , 0.05. Experiments were carried out at least three times and rep- technology (Wembley, U.K.); rabbit anti-mouse IgG HRP and goat anti- resentative results are shown. rabbit IgG HRP from Stratech (Soham, U.K.); CG inhibitor I, elastase inhibitor IV, and PR3 from Merck Chemicals (Nottingham, U.K.); human recombinant C5a from Hycult Biotech (Cambridge Bioscience, Cambridge, Results U.K.); C5aR agonist (H-Phe-Lys-Pro-D-Cha-Cha-D-Arg-OH) and fura 2- NSPs cleave and inactivate the C5aR AM from AnaSpec (Cambridge Bioscience, Cambridge, U.K.); and NE and CG from Athens Research & Technology (Athens, GA). All other agents The effects of NSPs on C5aR expression and function were in- including inhibitors were from Sigma-Aldrich (Poole, U.K.). Lysis buffer vestigated using purified neutrophils and differentiated U937 was composed of 25 mM 2-(N-morpholino)ethanesulfonic acid, 5 mM myelomonocytic cells. Cells were exposed to CG, NE, and PR3 at EDTA, 150 mM NaCl, and 1% Triton X-100 (pH 6.5) plus the following inhibitors: 5 mM 4-(2-aminoethyl)benzenesulfonyl fluoride hydrochloride, concentrations commonly found in BALF of patients with CF (5, 27, 5 mM PMSF, 0.8 mM aprotinin, 40 mM bestatin, 14 mM E-64, 20 mM 28). Using a mAb specific for the N terminus of C5aR, a large re- leupeptin, 15 mM pepstatin A, and 5 mM 1.10 phenanthroline. All cell duction in cell-surface C5aR expression, as detected by flow cytom- incubations were carried out in Krebs/HEPES/BSA (KHB) buffer: 25 mM etry, was observed after incubation with all three NSPs (Fig. 1A, 1B). HEPES, 120 mM NaCl, 4.8 mM KCl, 1.2 mM KH PO , 1.2 mM MgSO , 2 4 4 Calcium release from intracellular stores is a hallmark of C5aR 1.3 mM CaCl2 (pH 7.4), and 0.1% BSA. activation and one of the first events observed after C5a binding. Cell isolation, culture and treatments After incubation of both neutrophils and U937 cells with the NSPs, Human neutrophils, isolated using dextran sedimentation and Ficoll gra- concomitant with the reduction in C5aR cell-surface expression, dient centrifugation, were incubated at 107 c/ml in KHB with the indicated a reduction in C5a responsiveness, as measured by release of intra- stimuli for 1 h at 37˚C, spun for 3 min at 1800 3 g, and supernatants were cellular calcium, was observed (Fig. 1C, 1D). However, the calcium- collected and used for incubations with U937 cells or analyzed by Western release response to a small-molecule C5aR agonist was not affected blotting. Cells were analyzed by flow cytometry or extracted with lysis buffer at 107 c/ml (30 min on ice followed by 5 min, 5000 3 g at 4˚C). (Fig. 1E, 1F). Native U937 cells were from Peter Monk (Sheffield, U.K.) and were dif- Activation of C5aR requires C5a binding at two distinct sites: ferentiated for 3 d with 0.5 mM Bt2cAMP to induce C5aR expression. For the main ligand binding site being the extracellular N terminus (aa 7 C5aR cleavage experiments, cells at 10 c/ml in KHB were incubated with the 2–22), and the other signal-transducing site is formed by the ex- 3 indicated reagents for 1 h at 37˚C. Cells were spun for 3 minutes at 1800 g, a and cell pellets were analyzed by Western blotting or flow cytometry. tracellular loops formed by -helices III, VI, and VII (12, 29–31). Thus, the reduction in responsiveness to C5a but not the C5aR ag- SDS-PAGE and Western blotting onist suggested that the loss of binding of the anti–N-terminal Ab Cell lysates or cell supernatants were run on 12% SDS-polyacrylamide gels and the loss of C5a-induced calcium response was not a consequence under nonreducing conditions and in the presence of protease inhibitors of total loss of the C5aR but suggested that only the C5a-binding The Journal of Immunology 1789

FIGURE 1. NSPs cleave and inactivate the C5aR. Neutrophils and U937 cells were incubated with buffer (control), NE (5 mg/ml), CG C D (5 mg/ml), or PR3 (10 mg/ml) for 1 h. Cells were analyzed by flow cytometry (A, B)or C5aR-dependent calcium responses (C–F) and Western blotting (G–I). C5aR cell-surface expression detected using anti-C5aR mAb S5/1: neutrophils (A) and U937 (B). Results are presented as mean fluorescence intensity (MFI) 6 Downloaded from SD. Release of intracellular calcium of fura 2– loaded cells (C, E: neutrophils; D, F: U937) in response to (C) and (D) (C5a [2.5 nM]) and (E) and (F) (C5aR agonist [0.25 mM]) was mea- E F sured by fluorometry. Addition of C5a and C5aR agonist are indicated by vertical arrows. Western blotting of cell lysate of neutrophils (G, H)and http://www.jimmunol.org/ U937 (I) developed with mAb anti-C5aR S5/1 (G) or polyclonal C5aR (H, I). Solid arrows indicate intact C5aR; dashed arrows indicate C5aR fragment. *p , 0.001 compared with control cells. backgr, Background; C, control.

GHI by guest on September 27, 2021

N terminus was lost but the calcium signaling C terminus was membrane (Fig. 2A) and resulted in the downregulation of C5aR retained. Indeed, whereas after incubation with the NSPs, Western cell-surface expression (Fig. 2B). Cytochalasin D (CytoD), an actin blots of the cell lysates, developed using the mAb recognizing the inhibitor and enhancer of GPCR-induced granule release, increased N terminus, showed a large reduction in the intensity of a 42-kDa fMLF and C5a-induced C5aR cleavage activity in the neutrophil band, typical for native C5aR (Fig. 1G), detection of the C5aR supernatants (Fig. 2A, 2B). Supernatants of neutrophils stimulated using the polyclonal Ab raised against the C terminus of the C5aR with fMLF (6 CytoD) or the phorbol ester PMA, a known inducer showed a reduction in intensity of the native 42-kDa band con- of neutrophil degranulation, had similar effects on the cleavage of comitant with the appearance of a band with a Mr of 27 kDa (after the C5aR on U937 cells (data not shown). Analysis of neutrophil incubation with CG) or 26 kDa (after incubation with NE or PR), elastase activity in the polymorphonuclear neutrophil (PMN) super- demonstrating that cleavage of the C5aR had taken place (Fig. 1H, natants demonstrated the C5a-induced secretion of NE mirrored 1I). Thus, NSPs can reduce C5a-induced cell activation by re- the cleavage of the C5aR (Fig. 2C). ducing expression of the N terminus by enzymatic cleavage. To identify the nature of the proteases involved in the C5aR cleavage, the supernatants of C5a-activated neutrophils were Neutrophils secrete C5aR-cleaving enzymes upon activation preincubated with the broad-spectrum metalloprotease inhibitor with C5a 1,10-phenanthroline, the broad-spectrum serine protease inhibitor C5a is a well-known activator of neutrophils and known to cause PMSF, and the speciﬁc NE and CG inhibitors. The C5aR cleavage neutrophil degranulation. To investigate if C5a, as a natural acti- was inhibited by the NE inhibitor and the serine protease inhibitor vator of neutrophils, can induce the secretion of NSPs, leading PMSF, but not by the CG inhibitor or metalloprotease inhibitor, to C5aR cleavage on neighboring cells, neutrophils were stim- identifying the enzyme responsible for the cleavage as the serine ulated with C5a, and cell supernatants were incubated with C5aR- protease NE (Fig. 2D). These results demonstrate that NSPs, expressing U937 cells as an indicator neighboring cell. Supernatants secreted by neutrophils, can cleave C5aR on neighboring cells, of C5a-activated neutrophils caused a small amount of cleavage of which will result in downregulation and inactivation of this re- the C5aR, resulting in a 26-kDa fragment being retained in the ceptor on unstimulated bystander cells. 1790 NEUTROPHIL DYSFUNCTION CAUSED BY NSP CLEAVAGE OF C5aR

FIGURE 2. Neutrophil elastase, secreted upon stimulation of PMN with C5a, cleaves the C5aR. U937 cells were incubated with dilutions of the supernatants of neutrophils (PMN) and cells were analyzed by Western blotting (A) and flow cytometry (B). PMN had been incubated with buffer only (control), PMA (1027 M), C5a (100 nM) 6 CytoD (CD; 10 mM), or fMLF (1026 M) 6 CD (10 mM) for 1 h. As an additional control, U937 cells were incubated with buffer, instead of neutrophil supernatant, for 1 h. (A) Cell pellets were Western blotted and developed with rabbit anti-human C5aR. (B) Cells were stained for flow cytometry using mouse anti C5aR (S5/1). Results are presented as mean fluorescence intensity (MFI) 6 SD. (C) NE activity of the PMN supernatants measured using a chro- mogenic substrate. (D) U937 cells were incubated with supernatants of C5a + CD-stimulated PMN, in the presence or absence of protease inhibitors. Cell pellets were run on 12% SDS-PAGE gels, Western blotted, and developed with rabbit anti-human Downloaded from C5aR. Solid arrows indicate intact C5aR; dashed arrows indicate C5aR fragment. CG inhibitor (CGI): 20 mM; NE inhibitor (NEI): 20 mM; 1,10-phenantroline (Ph): 4 mM; PMSF (PF): 2 mM. *p , 0.001 compared with control cells. C, No inhibitor. http://www.jimmunol.org/

CF BALF contains enzymes that degrade C5aR C5a induces the cleavage of endogenously expressed C5aR on As shown above, NSPs released upon neutrophil stimulation can U937 cells cleave the C5aR on neighboring cells. NSP-mediated cleavage C5a itself has been implicated in the downregulation of its own and inactivation of C5aR has important consequences for con- receptor as measured by flow cytometry (20, 21, 32). It is generally ditions in which excess NSP release and defective protease/ acknowledged that ligand binding to GPCRs such as the C5aR re- antiprotease balance are associated with pathology such as ob- sults in the internalization and recycling of the receptor or possibly served in CF. In CF, excess neutrophil influx results in apoptosis degradation. Thus, a reduced C5aR cell-surface expression could and necrosis of these neutrophils and release of excess NSPs. To suggest internalization followed by lysosomal degradation. by guest on September 27, 2021 investigate if lung fluids from CF patients have the ability to The effects of C5a on the C5aR expression on the stimulated cell enzymatically cleave the C5aR, two BALFs with and two BALFs was investigated, and when U937 cells were incubated with C5a, without detectable NE activity were tested for their ability to a reduction in the binding of the anti–N-terminal mAb was ob- cleave the C5aR. Only the two BALFs containing detectable NE served by flow cytometry (Fig. 4A); simultaneously, as detected activity induced C5aR cleavage (Fig. 3A, 3B). The broad- by Western blotting using the anti–C terminus Ab, a 27-kDa band spectrum serine protease inhibitor PMSF inhibited the C5aR appeared, demonstrating that C5a can reduce the expression of its cleavage by both BALFs, whereas the broad-spectrum metal- own receptor by inducing enzymatic cleavage (Fig. 4B). This loprotease inhibitor 1,10 phenanthroline had no effect, demon- cleavage was enhanced by the secretagogue and actin inhibitor strating that the cleavage was accomplished by a serine protease. CytoD, which enhances degranulation (Fig. 4A, 4B), thus demon- Further analysis using specific protease inhibitors showed that in strating that this process is mediated by C5a-induced local secretion one BALF (Fig. 3A), the enzyme responsible for the cleavage of enzymes, rather than internal lysosomal degradation. PMA, was CG, as only the CG inhibitor prevented the cleavage. a potent inducer of cell activation and degranulation, also induced However, in the other BALF (Fig. 3B), a different pattern the cleavage of endogenously expressed C5aR (Fig. 4A, 4B). emerged. Although the CG inhibitor did not prevent the proteolysis, it led to a slight reduction in the Mr of the observed fragment, resulting in a size normally observed when cells are incubated with NE (Fig. 3B). Furthermore, the NE inhibitor caused partial inhibition of cleavage. This suggests that both CG and NE contributed to the cleavage. Two BALFs that did not contain any detectable NE activity did not cause cleavage of the C5aR (data not shown). These data suggest a hierarchy exists between CG and NE FIGURE 3. CF BALF contains C5aR-degrading enzymes. U937 cells toward C5aR, and inhibition of CG enabled NE to cleave the were incubated with NE-containing BALFs from CF patients at a final NE m C5aR. Indeed, when purified CG and NE were coincubated with concentration of 2 g/ml in the presence or absence of protease inhibitors for 1 h, cells were spun and run on 12% SDS-PAGE gels, Western blotted, the U937 cells, a fragment size similar to that obtained with CG and developed with rabbit anti-human C5aR. (A) BALF original NE was observed, suggesting that CG is dominant over NE (Fig. 3C). concentration 4.2 mg/ml. (B) BALF original NE concentration 38 mg/ml. These results demonstrate that NE and CG in BALF from CF Solid arrows indicate intact C5aR; dashed arrow indicates C5aR fragment.2, patients have the ability to cleave the C5aR, which is likely to no inhibitor; CG inhibitor (CGI): 20 mM; NE inhibitor (NEI): 20 mM; 1,10- impair C5a-dependent functions such as phagocytosis and in- phenantroline (Ph): 4 mM; PMSF (PF): 2 mM. (C) Cells were incubated with tracellular killing of pathogens. buffer, CG (5 mg/ml), NE (5 mg/ml), or CG + NE (5 mg/ml each). The Journal of Immunology 1791

FIGURE 4. C5a and PMA induce CG-mediated cleavage of the C5aR on U937 cells. U937 cells were incubated with C5a (100 nM) in the presence or absence of CytoD (CD; 10 mM) or PMA (1027 M); cells were spun and analyzed by ﬂow cytometry (A) or run on 12% SDS-PAGE gels, Western blotted, and developed with rabbit anti-human C5aR (B). U937 were incubated with C5a (100 nM; C) and 10 mMCD(C)or1027 M PMA (D) in the absence or presence of protease inhibitors for 1 h, cells were spun and run on 12% SDS-PAGE gels, C D Western blotted, and developed with rabbit anti- human C5aR. Inhibitor concentrations: CG inhibitor (CGI): 20 mM; GM6001 (GM): 10 mg/ml; NE inhibitor (NEI): 20 mM; PMSF (PF): 2 mM. Solid arrows indicate intact C5aR; dashed arrows indicate C5aR fragment. 2, stimulant but no inhibitor.

*p , 0.001 compared with control cells. backgr, Downloaded from Background; C, control, no stimulant; MFI, mean ﬂuorescence intensity. http://www.jimmunol.org/ The nature of the enzymes involved in the C5a- and PMA- Probing of the Western blots with the anti–C terminus anti-C5aR induced cleavage of endogenously expressed C5aR on U937 cells Ab showed again a reduction in intensity of the 42-kDa native was assessed using various protease inhibitors. Addition of inhib- C5aR band; however, C5a on its own did not induce the cleavage itors speciﬁc for the serine proteases CG or NE before stimulation of the C5aR (Fig. 5B, bottom panel); only when CytoD was with C5a or PMA showed that only the CG inhibitor prevented the coincubated, in addition to the native C5aR 42-kDa band, a band induced cleavage (Fig. 4C, 4D), identifying the enzyme responsible of ∼27 kDa was observed, consistent with NSP-mediated cleav- for the C5a- and PMA-induced cleavage of endogenous C5aR in age. In contrast, fMLF on its own induced the cleavage of the U937 cells as CG. C5aR, and addition of CytoD resulted in complete elimination of

the 42-kDa native C5aR band, with the appearance of an intense by guest on September 27, 2021 C5a, fMLF, and PMA induce reduction in neutrophil C5aR 27-kDa fragment (Fig. 5B, bottom panel). PMA had the same cell-surface expression, but reduction in response to C5a is effect as fMLF combined with CytoD (data not shown). only partially due to enzymatic degradation These results suggested that the mechanism of downregulation of In a recent study by Unnewehr et al. (21), it was suggested that the the C5aR in response to C5a on neutrophils was different from that C5a-induced downregulation of the C5aR on neutrophils was of the promonocytic cell line U937. It is well known that GPCRs caused by shedding of the receptor. In their study, an Ab recog- upon interaction with their ligand become internalized. This may nizing the N terminus of C5aR was used, and thus possible cleavage explain why the C5aR did not become cleaved upon C5a induced would not have been detected. Our results described above dem- activation as internalization would provide temporary protection onstrated that NSPs can cleave the C5aR and that, on the prom- against NSP-mediated cleavage. This C5aR internalization would onocytic cell line U937, C5a induces CG-mediated cleavage of not occur upon neutrophil activation by fMLF and PMA, and the endogenously expressed C5aR. To investigate if the reduction therefore, NSP secretion in response to these agents could thus in cell-surface expression of endogenously expressed C5aR on directly cleave and inactivate the C5aR. A time-course experiment neutrophils, in response to C5a, could be caused by NSPs-mediated showed that only after 3-h incubation with C5a, a small amount of enzymatic cleavage, neutrophils were assessed for C5aR expres- cleaved C5aR was observed, whereas the addition of CytoD greatly sion by ﬂow cytometry and Western blotting. Flow cytometry enhanced and accelerated this (Fig. 5C, bottom panel). The slight showed that C5a- and also fMLF-induced downregulation of increase in Mr of the native C5aR remained during this time period cell-surface expression (Fig. 5A). The C5a- and fMLF-induced (Fig. 5C, top panel), which was only clearly observed using the C5aR downregulation was enhanced by CytoD, resulting in a near anti–N terminus Ab. Unnewehr et al. (21) reported that the C5aR complete elimination of the N-terminal epitope from the cell was released in the supernatant of the neutrophils; however, we surface. were unable to detect any C5aR in the supernatants of C5a- or Western blotting using the anti–N terminus mAb as the de- fMLF-stimulated neutrophils using the anti–N terminus or anti–C tecting Ab also showed a reduction in intensity of the 42-kDa band terminus Ab after loading of supernatants equivalent to the cell upon incubation with C5a or fMLF (Fig. 5B, top panel). Although lysates (Fig. 5B), and our results suggest that the loss of C5aR addition of CytoD to fMLF caused the complete disappearance of may be caused by intracellular degradation after internalization. the C5aR, it only had a small effect on the C5a-induced C5aR Prolonged and overexposure of the blots in Fig. 5B also did not downregulation (Fig 5B, top panel). Interestingly, C5a induced reveal any sign of C5aR released in the supernatant. a slight increase in Mr of the C5aR; this was not observed with fMLF. The slight increase in Mr is likely caused by phosphory- Discussion lation of the C5aR upon C5aR stimulation, which has previously We have investigated the effects of NSPs on C5aR expression, been reported (33). integrity, and function and their role in downregulation of the C5aR 1792 NEUTROPHIL DYSFUNCTION CAUSED BY NSP CLEAVAGE OF C5aR

FIGURE 5. Effect of C5a, fMLF, and PMA on C5aR cleavage on neutrophils. Neutrophils were incubated with C5a (100 nM), fMLF (1026 M), or PMA (1027 M) in the absence or presence of CytoD (CD; 10 mM) for the indicated period of time. Cells were analyzed for the expression of the C5aR by flow cytometry (A) or Western blotting (B, C). (A) Cell-surface expression after 1-h incubation with the indicated agents was assessed using mAb S5/1. Western blotting of neutrophil lysates after stimulation with the indicated agents for 1 h (B) or the indicated period of time (C). (B and C) Top panels, Detection using anti-C5aR N terminus mAb S5/1; bottom panels, detection using polyclonal anti-C5aR C terminus Ab. Solid arrows indicate intact C5aR; dashed arrows indicate C5aR fragment. *p , 0.001 compared with control cells. MFI, Mean fluorescence intensity. Downloaded from http://www.jimmunol.org/ after C5a-induced cell activation. Furthermore, we have investigated cleaving C5aR and may also be an important therapeutic target. the action of BALF from CF patients, as a source of NSPs, on the This was exemplified by the observation that in one BALF, inhi- integrity of the C5aR. bition of CG effectively prevented cleavage, whereas in another We show in this study that C5aR expressed on neutrophils and BALF, when CG was inhibited, a fragment size typical of NE was the differentiated monocytic cell line U937 is sensitive to cleavage observed, and when CG and NE were combined, a fragment size by physiological concentrations of purified NSPs CG, NE, and typical for CG cleavage was detected. Thus, dual inhibition of both PR3, resulting in loss of the C5a-binding N terminus and non- CG and NE would be required to prevent C5aR cleavage and in- responsiveness of the cells to C5a stimulation (Fig. 1). These activation. The balance among NE, CG, and their respective natural by guest on September 27, 2021 C5aR-cleaving NSPs can be released from neutrophils and U937 inhibitors in BALF may determine the effective cleavage profile, cells upon stimulation with C5a itself. This can lead to cleavage of and further analysis of a larger patient cohort is the subject of on- the C5aR expressed on neighboring, unstimulated cells (Fig. 2), going investigation. In the U937 cell line, C5a-induced cleavage in addition to the stimulated cells themselves (Figs. 4, 5). The was predominantly mediated by CG, whereas NE appeared to be the enzymes responsible for this were identified as CG and NE. Fur- predominantly active NSP in supernatants of activated neutrophils. thermore, BALF from CF patients also contained this C5aR- This difference may reflect cell-specific differences in enzyme as cleaving activity, which could be prevented by CG and NE well as the relative abundance of their natural inhibitors. inhibitors (Fig. 3). Our data help to explain previously reported Our results from the supernatant transfer experiments (Fig. 2) defects of neutrophil function under pathological conditions in have important consequences for conditions in which large which excess unregulated NSP activity or excess C5a generation amounts of NSPs are secreted by neutrophils, as this has the are seen, such as CF lung disease and sepsis. Our data also show potential to render cells nonresponsive to C5a. The manner in that loss of C5aR on neutrophils under such conditions can be which NSPs are secreted could be in response to C5a but also caused by NSP-mediated degradation in response to both micro- other stimulants including the bacterial peptide fMLF [as we bial (e.g., formylated peptides) and endogenous stimuli (e.g., ex- demonstrate in this study (Figs. 2,5)]. Excess NSP secretion in cess C5a). We also show that excess C5a generation also leads to diseases such as CF can also be a result of dysregulated clearance intracellular degradation of the C5aR, following ligand induced of apoptotic neutrophils with consequent large-scale secondary internalization, but this requires further investigation. Recently necrosis. published findings from Unnewehr et al. (21) suggest that shed- We also observed that C5aR can be cleaved after stimulation of ding is one cause of reduced cells’ C5aR surface expression; cells by C5a; however, the extent of this cleavage was dependent however, we were unable to find evidence of shedding in this on the cell type used. Although stimulation of differentiated U937 study. A possible reason for this may be a possible difference in with C5a led to extensive CG-mediated cleavage of the endogenous methodology. C5aR (Fig. 4), on neutrophils, C5aR downregulation following Considering the importance of C5a and C5aR in clearance of C5a exposure was only partially explained by cleavage, although bacterial infections by potentiating phagocytosis and intracellular this was substantially enhanced by the secretagogue CytoD (Fig. 5). killing including of P. aeruginosa (10, 13, 14), our observation that Neutrophils are very dynamic cells, and it was recently suggested the C5aR is cleaved and inactivated by both purified NSPs and that C5aR is shed from the cell surface in response to C5a (21). NSPs found in CF BALF (Figs. 1, 3) may partly explain why However, despite the inclusion of a mixture of protease inhibitors chronic infection is such a recalcitrant aspect of CF airway disease. in the processing of the samples, we detected neither uncleaved NE is the most widely studied NSP in CF and is a major therapeutic nor cleaved C5aR in the supernatants of C5a-stimulated neutro- target. However, our data show that CG is also very effective at phils in our experiments (Fig. 5). The reason we observed only The Journal of Immunology 1793 limited generation of the 27 kDa C5aR cleavage fragment on (42) suggested that 50% of the C5aR was degraded after inter- neutrophils is likely to be explained by internalization of the C5aR nalization. Our results suggest that in U937 cells no or minimal in response to C5a, which would have protected it from degra- internal degradation may have taken place and that the degrada- dation by NSPs until recycling to the cell surface. However, the tion observed took place extracellularly, whereas in neutrophils, reduction in overall C5aR expression on neutrophils is likely due internal degradation may play a major role in reduction of C5aR to degradation of internalised C5aR. expression. Van Epps et al. (39) showed that C5aR has a very long C5a- and PMA-induced downregulation of C5aR has previ- internalization/recycling time, and this may explain why only after ously been observed (20, 21, 32, 34). Our study is the first, to our 3 h, some of the 26- to 27-kDa C5aR-degradation fragment was knowledge, to show that enzymatic degradation is a significant observed. Kinetics of internalization and shedding of the C5aR contributor to this phenomenon. Our demonstration that NSPs are expressed on U937 cells have not been investigated, which may responsible for this cleavage and functional inactivation has im- explain the differences in C5aR cleavage observed after C5a ex- portant consequences for when considering pathologies associated posure. Differences in cell types and experimental setup used in with chronic unregulated NSP activity as discussed above. Our the above-mentioned studies and our study may account for the data complement and add to observations from several previous differences observed. As GPCR receptors are quickly internalized studies. Preincubation with PMA has previously been reported following ligand binding, a process that happens within minutes, to reduce binding of C5a to neutrophils (34), which, on the basis they would be temporarily protected from extracellular degra- of data presented in this study, is likely to be due to enzymatic dation by NSPs, released from granules in response to GPCR cleavage of the C5aR in response to PMA-induced degranulation activation, which also happens within minutes. This could be an

(Fig. 5A), resulting in the generation of a membrane-bound 26- to autoprotective mechanism, as when the GPCR recycles to the cell Downloaded from 27-kDa C-terminal fragment (not shown). Other studies showed surface, the NSPs may have diffused away or could have been that preincubation of neutrophils with C5a reduced binding of an inactivated. However, this would only be relevant if the GPCR is anti-C5aR Ab as measured by ﬂow cytometry (20, 21, 32). This not degraded intracellularly, as observed for the C5aR in neu- preincubation with C5a resulted in neutrophil dysfunction, char- trophils. acterized by reduced phagocytosis of zymosan and P. aeruginosa The NE- and CG-induced cleavage described in this study re-

(20, 32). None of these studies investigated cleavage of the C5aR sulted in the loss of the extracellular C5a-binding N terminus and http://www.jimmunol.org/ as a potential mechanism. C5a is known to induce the release of retention of a 26- to 27-kDa C-terminal fragment in the membrane, granule associated enzymes from neutrophils (35), and TNF- and reflected in a reduction in Mr of 15 to 16 kDa. CG and NE have GM-CSF–induced elastase release correlates with reduced C5a binding and inhibition of chemotaxis (36). Our novel observations of NSP-mediated cleavage of the C5aR and that C5a and fMLF- induced release of NSPs that subsequently cleave and inactivate the C5aR combine the above studies and explain findings of reduced C5a and anti-C5aR binding to the cells and the reduction in complement-dependent phagocytosis and intracellular killing (20, by guest on September 27, 2021 32, 34). A study by Tralau et al. (25) is particularly interesting, as upon exposure to NSPs, neutrophils exhibited impaired responses to C5a but not to fMLF, IL-8, leukotriene B4, or platelet-activating factor. This is despite one other report of NE-mediated inactivation of the IL-8RA/CXCR1, in which the authors suggested that inactivation of this receptor was the main cause of neutrophil dysfunction in CF (4). The results of Tralau et al. (25) may point to a difference in sensitivities of the C5aR, IL-8RA/CXCR1, and the receptor for fMLF and leukotriene B4 to NSP-mediated inactivation. Desensitization of neutrophils to C5a is thought to be caused by internalization of the C5aR following initial ligand binding. This becomes temporarily unavailable, before being recycled to the cell surface, at which point the cells can be restimulated. Our data using the monocytic cell line U937 support a model in which C5a stimulation leads to NSP secretion, which subsequently inactivates the C5aR. The cells only become responsive again after resynthe- sis of the C5aR. Desensitization to C5a after stimulation with fMLF (termed cross-desensitization) has been reported (37), and as we show, this is caused by fMLF-stimulated release of NSPs, resulting in C5aR cleavage. Desensitization and downregulation after C5a stimulation of neutrophils is likely mostly caused by inter- FIGURE 6. Proposed models for NSP-induced neutrophil dysfunction. nalization followed by intracellular degradation. However, C5a- Model 1, High levels of systemically activated C5a or formylated bacte- induced NSP secretion can inactivate the C5aR on neighboring rial peptides as found in, for example, sepsis induces the release of NSPs, which subsequently cleave the C5aR on own cells and on neighboring cells that have not been in contact with C5a. Internalization of the bystander cells. Model 2, Excess neutrophil infiltration, associated with GPCRs followed by endosomal degradation and/or recycling to excess unregulated secretion of NSPs by necrotic neutrophils, resulting in the cell surface is well documented and a very quick process that the cleavage of the C5aR on neighboring newly infiltrated bystander cells. happens within minutes (38). Several studies investigated the fate Both mechanisms render the affected cells nonresponsive to further C5a of C5aR after ligand binding. Although some studies suggested stimulation, thereby impairing their ability to phagocytose and kill bac- that no internal degradation took place (39, 40, 41), another study teria. 1794 NEUTROPHIL DYSFUNCTION CAUSED BY NSP CLEAVAGE OF C5aR different amino acid preferences for cleavage of substrates. CG CXCR1 on neutrophils disables bacterial killing in cystic fibrosis lung disease. Nat. Med. 13: 1423–1430. cleaves preferentially after L, H, and F, whereas NE cleaves after 5. McGreal, E. P., P. L. Davies, W. Powell, S. Rose-John, O. B. Spiller, I. Doull, V, I, and A. Cleavage after H(100) by CG and cleavage after A S. A. Jones, and S. Kotecha. 2010. Inactivation of IL-6 and soluble IL-6 receptor (108) by NE would result in a predicted retained 27- to 28-kDa by neutrophil derived serine proteases in cystic fibrosis. Biochim. Biophys. Acta 1802: 649–658. C-terminal fragment, similar to our experimental observations. 6. Berger, M., R. U. Sorensen, M. F. Tosi, D. G. Dearborn, and G. Do¨ring. 1989. These cleavage events are predicted to result in a loss of 11 and Complement receptor expression on neutrophils at an inflammatory site, the 12 kDa, respectively, whereas we observed a loss of 15 and 16 Pseudomonas-infected lung in cystic fibrosis. J. Clin. Invest. 84: 1302–1313. 7. Tosi, M. F., H. Zakem, and M. Berger. 1990. Neutrophil elastase cleaves C3bi on kDa, respectively. This is probably explained by differences in the opsonized pseudomonas as well as CR1 on neutrophils to create a functionally calculated Mr (39 kDa) and our observed Mr (42 kDa) of native important opsonin receptor mismatch. J. Clin. Invest. 86: 300–308. 8. Liau, D. F., N. X. Yin, J. Huang, and S. F. Ryan. 1996. Effects of human C5aR, likely due to N-linked glycosylation at N at position 5 (43). polymorphonuclear leukocyte elastase upon surfactant proteins in vitro. Biochim. A single cleavage event would still retain the N-terminal epitope Biophys. Acta 1302: 117–128. in the membrane as a 15- to 16-kDa fragment. However, such 9. Duvoix, A., R. M. Mackay, N. Henderson, E. McGreal, A. Postle, K. Reid, and H. Clark. 2011. Physiological concentration of calcium inhibits elastase-induced a fragment was not observed using the anti–N terminal mAb, cleavage of a functional recombinant fragment of surfactant protein D. Immu- suggesting that a second cleavage event occurred most likely after nobiology 216: 72–79. L33 (CG) and V35 (NE), near the membrane, thereby releasing 10. Mollnes, T. E., O. L. Brekke, M. Fung, H. Fure, D. Christiansen, G. Bergseth, V. Videm, K. T. Lappega˚rd, J. Ko¨hl, and J. D. Lambris. 2002. Essential role of the N-terminal epitope. Although potential cleavage sites are pres- the C5a receptor in E coli-induced oxidative burst and phagocytosis revealed by ent in the other extracellular loops, these sites could be protected a novel lepirudin-based human whole blood model of inflammation. Blood 100: 1869–1877. due to steric hindrance. A C5aR agonist, which only requires the 11. Guo, R. F., and P. A. Ward. 2005. Role of C5a in inflammatory responses. Annu. C-terminal C5a binding site, was still active, demonstrating that Rev. Immunol. 23: 821–852. Downloaded from the C terminus had remained intact. 12. Monk, P. N., A. M. Scola, P. Madala, and D. P. Fairlie. 2007. Function, structure and therapeutic potential of complement C5a receptors. Br. J. Pharmacol. 152: On the basis of the data presented in this study, we propose the 429–448. following interlinked models for NSP-induced neutrophil dysfunc- 13. Larsen, G. L., B. C. Mitchell, T. B. Harper, and P. M. Henson. 1982. The pul- tion (Fig. 6). In the first model, C5a or other neutrophil stimulants monary response of C5 sufficient and deficient mice to Pseudomonas aeruginosa. Am. Rev. Respir. Dis. 126: 306–311. such as fMLF are the initiating factor, and under conditions of 14. Cerquetti, M. C., D. O. Sordelli, J. A. Bellanti, and A. M. Hooke. 1986. Lung excess systemic complement activation and C5a generation and defenses against Pseudomonas aeruginosa in C5-deficient mice with different http://www.jimmunol.org/ genetic backgrounds. Infect. Immun. 52: 853–857. bacteremia, as found in sepsis, these stimulants activate neu- 15. Ho¨pken, U. E., B. Lu, N. P. Gerard, and C. Gerard. 1996. The C5a chemo- trophils, resulting in NSP secretion and cleavage and inactivation attractant receptor mediates mucosal defence to infection. Nature 383: 86–89. of the C5aR on the activated or unstimulated bystander cells, 16. Lambris, J. D., D. Ricklin, and B. V. Geisbrecht. 2008. Complement evasion by human pathogens. Nat. Rev. Microbiol. 6: 132–142. resulting in dysfunctional neutrophils and reduced bacterial clear- 17. Seely, A. J., J. F. Naud, G. Campisi, B. Giannias, S. Liu, A. DiCarlo, L. E. Ferri, ance. In the second model, excess secretion of unregulated NSPs, J. L. Pascual, J. Tchervenkov, and N. V. Christou. 2002. Alteration of chemo- as in the case of CF, is the initiating factor. In this case, circulating attractant receptor expression regulates human neutrophil chemotaxis in vivo. Ann. Surg. 235: 550–559. neutrophils are attracted to the lungs by locally produced che- 18. Huber-Lang, M. S., N. C. Riedeman, J. V. Sarma, E. M. Younkin, S. R. McGuire, mokines such as IL-8 and also by C5a itself. Upon arrival in the I. J. Laudes, K. T. Lu, R. F. Guo, T. A. Neff, V. A. Padgaonkar, et al. 2002.

Protection of innate immunity by C5aR antagonist in septic mice. FASEB J. 16: by guest on September 27, 2021 lung, neutrophils are exposed to unopposed NSP activity due to 1567–1574. the presence of large numbers of dead and dying neutrophils in the 19. Guo, R. F., N. C. Riedemann, K. D. Bernacki, V. J. Sarma, I. J. Laudes, absence of appropriate phagocytic efferocytosis (44) and the ab- J. S. Reuben, E. M. Younkin, T. A. Neff, J. D. Paulauskis, F. S. Zetoune, and P. A. Ward. 2003. Neutrophil C5a receptor and the outcome in a rat model of sence of appropriate antiprotease defense. These NSPs cleave the sepsis. FASEB J. 17: 1889–1891. C5aR, resulting in these cells becoming impaired as in model 1. 20. Conway Morris, A., K. Kefala, T. S. Wilkinson, K. Dhaliwal, L. Farrell, This results in a vicious cycle of newly arrived neutrophils be- T. Walsh, S. J. Mackenzie, H. Reid, D. J. Davidson, C. Haslett, et al. 2009. C5a mediates peripheral blood neutrophil dysfunction in critically ill patients. Am. J. coming dysfunctional and releasing their contents, too, resulting in Respir. Crit. Care Med. 180: 19–28. further neutrophil dysfunction and tissue damage. 21. Unnewehr, H., D. Rittirsch, J. V. Sarma, F. Zetoune, M. A. Flierl, M. Perl, S. Denk, M. Weiss, M. E. Schneider, P. N. Monk, et al. 2013. Changes and In conclusion, C5a-induced NSP release or NSPs released regulation of the C5a receptor on neutrophils during septic shock in humans. J. through other mechanisms cleave and inactivate C5aR. Subse- Immunol. 190: 4215–4225. quent hyporesponsiveness of cells to C5a may lead to defective 22. van den Berg, C. W., R. M. Gonçalves-de-Andrade, C. K. Okamoto, and D. V. Tambourgi. 2012. C5a receptor is cleaved by metalloproteases induced by phagocytosis and intracellular killing of opsonized bacteria. The sphingomyelinase D from Loxosceles spider venom. Immunobiology 217: 935– use of NSP inhibitors in pathologies associated with excess NSP 941. secretion but also with excess C5a and formyl peptide generation 23. Jagels, M. A., J. A. Ember, J. Travis, J. Potempa, R. Pike, and T. E. Hugli. 1996. Cleavage of the human C5A receptor by proteinases derived from Porphyr- may be beneficial for patients with CF or sepsis and may prevent omonas gingivalis: cleavage of leukocyte C5a receptor. Adv. Exp. Med. Biol. neutrophil dysfunction and aid the clearance of bacterial infections. 389: 155–164. 24. Siciliano, S. J., T. E. Rollins, J. DeMartino, Z. Konteatis, L. Malkowitz, G. Van Both CG and NE were found to cleave the C5aR in biological Riper, S. Bondy, H. Rosen, and M. S. Springer. 1994. Two-site binding of C5a by samples, and thus, therapeutic inhibition of both enzymes should be its receptor: an alternative binding paradigm for G protein-coupled receptors. considered. Proc. Natl. Acad. Sci. USA 91: 1214–1218. 25. Tralau, T., U. Meyer-Hoffert, J. M. Schro¨der, and O. Wiedow. 2004. Hu- man leukocyte elastase and cathepsin G are specific inhibitors of C5a- Disclosures dependent neutrophil enzyme release and chemotaxis. Exp. Dermatol. 13: The authors have no financial conflicts of interest. 316–325. 26. Al-Mohanna, F. A., and M. B. Hallett. 1988. The use of fura-2 to determine the relationship between cytoplasmic free Ca2+ and oxidase activation in rat neutrophils. Cell Calcium 9: 17–26. References 27. Witko-Sarsat, V., L. Halbwachs-Mecarelli, A. Schuster, P. Nusbaum, I. Ueki, 1. Do¨ring, G. 1994. The role of neutrophil elastase in chronic inflammation. Am. J. S. Canteloup, G. Lenoir, B. Descamps-Latscha, and J. A. Nadel. 1999. Pro- Respir. Crit. Care Med. 150: S114–S117. teinase 3, a potent secretagogue in airways, is present in cystic fibrosis sputum. 2. Voynow, J. A., B. M. Fischer, and S. Zheng. 2008. Proteases and cystic fibrosis. Am. J. Respir. Cell Mol. Biol. 20: 729–736. Int. J. Biochem. Cell Biol. 40: 1238–1245. 28. Rubio, F., J. Cooley, F. J. Accurso, and E. Remold-O’Donnell. 2004. Linkage of 3. Griese, M., M. Kappler, A. Gaggar, and D. Hartl. 2008. Inhibition of airway neutrophil serine proteases and decreased surfactant protein-A (SP-A) levels in proteases in cystic fibrosis lung disease. Eur. Respir. J. 32: 783–795. inflammatory lung disease. Thorax 59: 318–323. 4. Hartl, D., P. Latzin, P. Hordijk, V. Marcos, C. Rudolph, M. Woischnik, S. Krauss- 29. DeMartino, J. A., G. Van Riper, S. J. Siciliano, C. J. Molineaux, Z. D. Konteatis, Etschmann, B. Koller, D. Reinhardt, A. A. Roscher, et al. 2007. Cleavage of H. Rosen, and M. S. Springer. 1994. The amino terminus of the human C5a The Journal of Immunology 1795

receptor is required for high affinity C5a binding and for receptor activation by factor alpha and granulocyte macrophage colony-stimulating factor. Exp. Clin. C5a but not C5a analogs. J. Biol. Chem. 269: 14446–14450. Immunogenet. 16: 212–225. 30. Higginbottom, A., S. A. Cain, T. M. Woodruff, L. M. Proctor, P. K. Madala, 37. Tomhave, E. D., R. M. Richardson, J. R. Didsbury, L. Menard, R. Snyderman, J. D. Tyndall, S. M. Taylor, D. P. Fairlie, and P. N. Monk. 2005. Comparative and H. Ali. 1994. Cross-desensitization of receptors for peptide chemo- agonist/antagonist responses in mutant human C5a receptors define the ligand attractants. Characterization of a new form of leukocyte regulation. J. Immunol. binding site. J. Biol. Chem. 280: 17831–17840. 153: 3267–3275. 31. Buck, E., H. Bourne, and J. A. Wells. 2005. Site-specific disulfide capture of 38. Cottrell, G. S. 2013. Roles of proteolysis in regulation of GPCR function. Br. J. agonist and antagonist peptides on the C5a receptor. J. Biol. Chem. 280: 4009– Pharmacol. 168: 576–590. 4012. 39. Van Epps, D. E., S. Simpson, J. G. Bender, and D. E. Chenoweth. 1990. Reg- 32. Furebring, M., L. Ha˚kansson, P. Venge, and J. Sjo¨lin. 2006. C5a, interleukin-8 ulation of C5a and formyl peptide receptor expression on human polymorpho- and tumour necrosis factor-alpha-induced changes in granulocyte and monocyte nuclear leukocytes. J. Immunol. 144: 1062–1068. expression of complement receptors in whole blood and on isolated leukocytes. 40. Van Epps, D. E., S. J. Simpson, and D. E. Chenoweth. 1992. C5a and formyl Scand. J. Immunol. 63: 208–216. peptide receptor regulation on human monocytes. J. Leukoc. Biol. 51: 393–399. 33. Ali, H., R. M. Richardson, E. D. Tomhave, J. R. Didsbury, and R. Snyderman. 41. Naik, N., E. Giannini, L. Brouchon, and F. Boulay. 1997. Internalization and 1993. Differences in phosphorylation of formylpeptide and C5a chemoattractant recycling of the C5a anaphylatoxin receptor: evidence that the agonist-mediated receptors correlate with differences in desensitization. J. Biol. Chem. 268: internalization is modulated by phosphorylation of the C-terminal domain. J. 24247–24254. Cell Sci. 110: 2381–2390. 34. Bender, J. G., D. E. Van Epps, and D. E. Chenoweth. 1987. Independent regu- 42. Suvorova, E. S., J. M. Gripentrog, and H. M. Miettinen. 2005. Different endo- lation of human neutrophil chemotactic receptors after activation. J. Immunol. cytosis pathways of the C5a receptor and the N-formyl peptide receptor. Traffic 139: 3028–3033. 6: 100–115. 35. Goldstein, I. M., and G. Weissmann. 1974. Generation of C5-derived lysosomal 43. Pease, J. E., and M. D. Barker. 1993. N-linked glycosylation of the C5a receptor. enzyme-releasing activity (C5a) by lysates of leukocyte lysosomes. J. Immunol. Biochem. Mol. Biol. Int. 31: 719–726. 113: 1583–1588. 44. Vandivier, R. W., P. M. Henson, and I. S. Douglas. 2006. Burying the dead: the 36. Binder, R., A. Kress, and M. Kirschfink. 1999. Modulation of C5a-mediated impact of failed apoptotic cell removal (efferocytosis) on chronic inflammatory effector functions of human polymorphonuclear leukocytes by tumor necrosis lung disease. Chest 129: 1673–1682. Downloaded from http://www.jimmunol.org/ by guest on September 27, 2021