Phagocytic neutrophils express formyl peptide receptors (FPRs; FPR1 and FPR2) that distinctly recognize peptides starting with an N-formylated methionine (fMet). This is a hallmark of bacterial metabolism; similar to prokaryotes, the starting amino acid in synthesis of mitochondrial DNA–encoded proteins is an fMet. Mitochondrial cryptic peptides (mitocryptides; MCTs) with an N-terminal fMet could be identified by our innate immune system; however, in contrast to our knowledge about bacterial metabolites, very little is known about the recognition profiles of MCTs. In this study, we determined the neutrophil-recognition profiles and functional output of putative MCTs originating from the N termini of the 13 human mitochondrial DNA–encoded proteins. Six of the thirteen MCTs potently activated neutrophils with distinct FPR-recognition profiles: MCTs from ND3 and ND6 have a receptor preference for FPR1; MCTs from the proteins ND4, ND5, and cytochrome b prefer FPR2; and MCT-COX1 is a dual FPR1/FPR2 agonist. MCTs derived from ND2 and ND4L are very weak neutrophil activators, whereas MCTs from ND1, ATP6, ATP8, COX2, and COX3, do not exert agonistic or antagonistic FPR effects. In addition, the activating MCTs heterologously desensitized IL-8R but primed the response to the platelet-activating factor receptor agonist. More importantly, our data suggest that MCTs have biased signaling properties in favor of activation of the superoxide-generating NADPH oxidase or recruitment of β-arrestin. In summary, we identify several novel FPR-activating peptides with sequences present in the N termini of mitochondrial DNA–encoded proteins, and our data elucidate the molecular basis of neutrophil activation by MCTs.

Phagocytic leukocytes (neutrophils and monocytes) constitute a security system in peripheral blood and tissues where they act as an important first line of cellular defense that can be rapidly mobilized to combat invading microbes. The innate immune system is well conserved, adopting recognition of molecular patterns to track down invasion by microbial pathogens, as well as imminent danger signals released during traumatic destruction of host tissues/cells (1, 2). The phagocytes are equipped with recognition molecules (receptors) that are designed to identify and respond to microbe-associated molecular patterns and danger-associated molecular patterns (DAMPs) (3). Among the different pattern recognition receptors expressed in phagocytes, formyl peptide receptors (FPRs) participate in the recruitment of inflammatory cells to sites of infection and/or tissue injury (4, 5). The molecular basis of neutrophil accumulation at the site of a bacterial infection has been attributed to the recognition of bacterial-derived N-formylated peptides as microbe-associated molecular patterns by FPRs (6, 7). Protein/peptide synthesis starting with an N-formylated methionine (fMet) residue is a hallmark of bacteria metabolism (8, 9), and the formyl group is subsequently cleaved by a deformylase to generate mature proteins, or it is left untouched on secreted peptides, such as the phenol-soluble modulins, peptides released from growing pathogenic Staphylococcus aureus (10). Accordingly, partial inhibition of the formylmethionine deformylase during bacterial growth results in a release of increasing amounts of formyl peptides (1113).

FPR1 was the first receptor identified that recognizes bacterial-derived formyl peptides with high affinity; however, fairly recently, FPR2 was also shown to recognize such peptides with high affinity (4, 14). FPR1 and FPR2 share 69% sequence identity and have distinct, but somewhat overlapping, ligand-recognition profiles (5). It has been suggested that the ligand-binding pocket in FPR1 has room for 4 or 5 aa, including the fMet residue, but the precise molecular structures that determine the respective recognition profiles have not been defined (15). Earlier studies have demonstrated that the N-formyl group is of primary importance for recognition by FPRs (4), but sequences in more distant parts of the ligand may also affect the receptor preference (16, 17). This is definitely the case for FPR2; it recognizes longer peptides in which residues located outside of the presumed binding pocket are of importance (16, 17).

Human mitochondrial DNA is organized as a double-stranded circular molecule, which, in addition to the sequences for specific ribosomal RNA and tRNA molecules, encodes for 13 proteins that are essential components of the respiratory chain generating the majority of cellular ATP via oxidative phosphorylation (18). Being of prokaryotic origin, mitochondrial DNA is transcribed and translated by a system independent of that used for nuclear host DNA, and the initiating amino acid in the mitochondrial-encoded proteins is an fMet, meaning that this microbial hallmark is also valid for mitochondrial-encoded proteins (19, 20). Tissue destruction and cell necrosis will ultimately lead to a release of otherwise hidden/cryptic endogenous danger molecules, including mitochondrial cryptic peptides (mitocryptides; MCTs) (21, 22). In addition to MCTs, mitochondria contain ATP and mitochondrial DNA, molecules that all potentially can act as DAMPs and induce a proinflammatory response (19, 23, 24). Therefore, it is reasonable to assume that receptors recognizing the fMet–peptide pattern have evolved to clear and resolve infectious and aseptic inflammatory processes in traumatized tissues. Indeed, mitochondrial-derived DAMPs have been shown to activate blood neutrophils and, hence, induce a sepsis-like state during severe injury, although the chemical structures of those molecules have not been determined (25). Further, mitochondrial DAMPs have also been demonstrated to be of importance for the recruitment of innate immune cells to aseptic inflammatory sites (26, 27). Hence, an understanding, at the molecular level, of innate immune recognition of mitochondrial-derived DAMPs, including the formyl peptides, is of direct clinical relevance in organ failure and severe injury/inflammation. To our knowledge, MCT-2, a 15-residue-long formyl peptide with a sequence identical to the N-terminus of mitochondrial DNA–encoded cytochrome b, is the only formyl peptide that has been isolated from host tissue (28, 29). The fact that 12 other proteins, in addition to cytochrome b, are encoded for by mitochondrial DNA suggests that other endogenous N-formylated MCTs may be formed during tissue destruction/necrosis and have ability to interact with the innate immune system (20, 30).

In an attempt to gain molecular insights into the innate immune–modulation properties of endogenous inflammatory mediators of mitochondrial origin, we synthesized 13 putative formylated MCTs with peptide sequences identical to the N termini of proteins encoded for by human mitochondrial DNA (Table I). Their effects on neutrophils and their receptor preferences were investigated. Our data reveal some differences in neutrophil-activation potency and receptor preference among these 13 MCTs. We identified five novel FPR-activating MCTs that induce cellular activity at nanomolar concentrations. Further, our data suggest that two of these peptides may also display a biased signaling property. In addition, our data reveal that FPR1- and FPR2-activating MCTs modulate the responses induced by the chemoattractants IL-8 and platelet-activating factor (PAF), a phenomenon that is achieved through receptor signaling cross-talk mechanisms in human neutrophils.

This study, conducted at Sahlgrenska Academy, included blood from buffy coats of healthy individuals obtained from the blood bank at Sahlgrenska University Hospital, Gothenburg, Sweden. According to Swedish legislation section code 4§ 3p SFS 2003:460 (Lag om etikprövning av forskning som avser människor), no ethical approval was needed, because the buffy coats were provided anonymously and could not be traced back to a specific donor.

The amino acid sequences of the 13 human mitochondrial DNA–encoded proteins were obtained from the Swiss-Prot database (Table I). The putative N-formylated peptides formed from these proteins, following proteolytic cleavage with trypsin or chymotrypsin, were predicted using the PeptideCutter tool (http://web.expasy.org/peptide_cutter/), with the exception of NADH dehydrogenase subunit 6 (ND6), because this is thought to be located in the transmembrane domain of the mitochondria. In this article, we temporarily named these 13 cleavage fragments independent of their activity as MCTs for easier understanding, although “mitocryptides” were originally defined as neutrophil-activating peptides that are derived from mitochondrial proteins (30). The cleavage sites in the parent proteins to generate the MCTs are as follows: ATP synthase subunit protein 6 (ATP6; at amino acid position 9), ATP8 (at amino acid position 12), cytochrome c oxidase subunit I (COX1; at amino acid position 13), COX2 (at amino acid position 23), COX3 (at amino acid position 12), ND1 (at amino acid position 10), ND2 (at amino acid position 10), ND3 (at amino acid position 5), ND4 (at amino acid position 20), ND4L (at amino acid position 5), ND5 (at amino acid position 28), and ND6 (at amino acid position 6) (see amino acid sequence in Table I). MCT-2 derived from human cytochrome b (at amino acid position 15) was predicted by homology to the earlier described porcine MCT-2 (28). All of these peptides were chemically synthesized by a solid-phase method using 9-fluorenylmethyloxycarbonyl and purified by reverse-phase high-performance liquid chromatography, and their homogeneity was confirmed by reverse-phase high-performance liquid chromatography and mass spectrometry.

All peptides were dissolved in DMSO to a concentration of 10 mM and stored at −80°C until use. Further dilutions were made in Krebs-Ringer phosphate buffer supplemented with glucose (10 mM), Ca2+ (1 mM), and Mg2+ (1.5 mM, pH 7.3) (KRG).

The hexapeptide WKYMVM (agonist for FPR2) was from Alta Bioscience (University of Birmingham, Birmingham, U.K.). The formyl peptide fMLF (the prototype agonist for human FPR1), IL-8, DMSO, BSA, and isoluminol were purchased from Sigma-Aldrich (St. Louis, MO). The receptor antagonist for FPR1, cyclosporine H (CysH), was kindly provided by Novartis Pharma (Basel, Switzerland), and the receptor antagonist for FPR2, PBP10 [see Ref. 31 for details about this inhibitory molecule], was obtained from Calbiochem (San Diego, CA). Allophycocyaninin- and PE-conjugated Abs against CD62L and CD11b were from Becton Dickinson (San Jose, CA), and FITC-conjugated Ab against CD66b was from AbD Serotec/Bio-Rad (Sundbyberg, Sweden). rTNF-α was from R&D Systems Europe (Abingdon, Oxon, U.K.), and HRP was from Boehringer Mannheim (Mannheim, Germany). Dextran and Ficoll-Paque were obtained from GE Life Sciences (Uppsala, Sweden).

Human peripheral blood neutrophils were isolated from buffy coats or from freshly drawn blood of healthy donors using dextran sedimentation and Ficoll-Paque gradient centrifugation, as described (32). The remaining erythrocytes were disrupted by hypotonic lysis and the neutrophils were washed, resuspended in KRG, and stored on melting ice until use. This isolation procedure permits cells to be purified with minimal granule mobilization.

NADPH oxidase activity and subsequent release of superoxide anions (O2) were determined using isoluminol-ECL (CL) (33, 34). The CL activity was measured in a six-channel Biolumat LB 9505 (Berthold, Wildbad, Germany), using disposable 4-ml polypropylene tubes with 900 μl of reaction mixture containing 105 cells, isoluminol (2 × 10−5 M), and HRP (4 U/ml). The tubes were equilibrated in the Biolumat for 5 min at 37°C, after which the stimulus (100 μl) was added, and light emission was recorded continuously (expressed as Mega cpm in the CL graphs). Receptor reactivation and cross-talk were achieved by stimulation of cells with a receptor-specific agonist; subsequently, when the response returned to baseline (desensitized state), the cells were reactivated by a second stimulation to release O2. When reactivation experiments were performed with antagonists, the antagonists were added to the CL reaction mixture 1 min before the second stimulation, whereas antagonists were added to the CL reaction mixture and incubated for 5 min at 37°C before stimulation with agonist for the other experiments involving inhibitors.

The ability of FPR agonists to promote β-arrestin recruitment was evaluated using a PathHunter eXpress system with CHO-K1 cells overexpressing FPR1 or FPR2 (DiscoverX, Fremont, CA), as described (35). In brief, CHO cells overexpressing FPR1 or FPR2 were diluted in cell plating reagent before being seeded in tissue culture–treated 96-well plates (10,000 cells per well) and incubated for 20 h (37°C, 5% CO2). Cells were stimulated with agonists for 90 min at 37°C, followed by the addition of detection solution for 60 min and measurement of chemiluminescence using a Mithras LB940 plate reader (Berthold).

The level of surface expression of the receptors CD62L (L-selectin), CD11b (complement receptor 3), and CD66b (CEACAM8) was determined on isolated neutrophils incubated or not with MCT-ND4, MCT-ND6, and MCT-COX1 at 37°C for 10 min. The tripeptide fMLF was used as a positive control. After incubation, the cells were stained with Abs against the receptors for 30 min at 4°C in the dark. The neutrophils were washed, ≥10,000 (>95% purity) were gated based on their size (forward scatter) and granularity (side scatter) on an Accuri C6 flow cytometer (BD Biosciences), and mean fluorescence intensity was analyzed by FlowJo software (v. 7.6.5; TreeStar). The results in Fig. 6 are presented as the percentage of decreased/cleaved (CD62L) or increased surface expression (CD11b and CD66b) compared with cells incubated at 37°C in the absence of peptide.

Neutrophil chemotaxis was investigated using a classical Boyden/Transwell system. Briefly, neutrophils (2 × 106 per milliliter) in KRG with BSA (0.3%) were loaded on top of a 3-μm-pore filter and allowed to migrate toward stimuli loaded into the bottom wells of a chemotaxis plate (ChemoTx; Neuro Probe) for 90 min (37°C, 5% CO2), as described earlier (35). The migrated cells were quantified by measuring myeloperoxidase activity of the cell lysate. For each peptide, samples from each donor were run in triplicates. As a negative control (spontaneous migration), only KRG with BSA (0.3%) was added to the bottom wells, and the classical neutrophil chemoattractant fMLF was used as positive migration control. Data are shown as the percentage migration compared with that induced by fMLF (10 nM).

Data analysis was performed using GraphPad Prism 6.0 (GraphPad Software, San Diego, CA). Curve fitting was performed by nonlinear regression using the sigmoidal dose-response equation (variable slope).

It is assumed that N-formylated peptides can be proteolytically processed from mitochondrial proteins and that these are detected by phagocytes as signs of tissue injury (27). To determine the relevance of this assumption, we synthesized 13 putative MCTs in their formylated forms, as described in 2Materials and Methods, and screened for their ability to activate human neutrophils isolated from peripheral blood. NADPH oxidase–mediated superoxide release was used as a sensitive readout for neutrophil activation (33). Of the 13 MCTs (Table I) examined in the study, 6 MCTs potently activated the neutrophil NADPH oxidase to release superoxide (Fig. 1), and these were MCTs derived from proteins ND4, COX1, ND6, ND5, and ND3, as well as from the earlier described MCT-2.

Table I.
MCTs derived from mitochondrial proteins used in this study
MCT NameSwiss-Prot IDaaPeptide SequenceParent Protein
MCT-2 Q9G345 1–15 Formyl-MTPMRKINPLMKLIN Cytochrome b 
MCT-ATP6 P00846 1–9 Formyl-MNENLFASF ATP6 
MCT-ATP8 P03928 1–11 Formyl-MPQLNTTVWPT ATP8 
MCT-COX1 P00395 1–13 Formyl-MFADRWLFSTNHK COX1 
MCT-COX2 P00403 1–23 Formyl-MAHAAQVGLQDATSPIMEELITF COX2 
MCT-COX3 P00414 1–12 Formyl-MTHQSHAYHMVK COX3 
MCT-ND1 P03886 1–10 Formyl-MPMANLLLLI ND1 
MCT-ND2 P03891 1–10 Formyl-MNPLAQPVIY ND2 
MCT-ND3 P03897 1–5 Formyl-MNFAL ND3 
MCT-ND4 P03905 1–20 Formyl-MLKLIVPTIMLLPLTWLSKK ND4 
MCT-ND4L P03901 1–5 Formyl-MPLIY ND4L 
MCT-ND5 P03915 1–28 Formyl-MTMHTTMTTLTLTSLIPPILTTLVNPNK ND5 
MCT-ND6 P03923 1–6 Formyl-MMYALF ND6 
MCT NameSwiss-Prot IDaaPeptide SequenceParent Protein
MCT-2 Q9G345 1–15 Formyl-MTPMRKINPLMKLIN Cytochrome b 
MCT-ATP6 P00846 1–9 Formyl-MNENLFASF ATP6 
MCT-ATP8 P03928 1–11 Formyl-MPQLNTTVWPT ATP8 
MCT-COX1 P00395 1–13 Formyl-MFADRWLFSTNHK COX1 
MCT-COX2 P00403 1–23 Formyl-MAHAAQVGLQDATSPIMEELITF COX2 
MCT-COX3 P00414 1–12 Formyl-MTHQSHAYHMVK COX3 
MCT-ND1 P03886 1–10 Formyl-MPMANLLLLI ND1 
MCT-ND2 P03891 1–10 Formyl-MNPLAQPVIY ND2 
MCT-ND3 P03897 1–5 Formyl-MNFAL ND3 
MCT-ND4 P03905 1–20 Formyl-MLKLIVPTIMLLPLTWLSKK ND4 
MCT-ND4L P03901 1–5 Formyl-MPLIY ND4L 
MCT-ND5 P03915 1–28 Formyl-MTMHTTMTTLTLTSLIPPILTTLVNPNK ND5 
MCT-ND6 P03923 1–6 Formyl-MMYALF ND6 
FIGURE 1.

Representative kinetics of maximal oxidative burst activity induced by formylated MCTs originating from 13 mitochondrial DNA–encoded proteins. Superoxide release from human neutrophils upon peptide stimulation was measured using the isoluminol/HRP-amplified chemiluminescence system. The prototype E. coli–derived FPR1 agonist fMLF (100 nM) and FPR2 agonist WKYMVM (100 nM) were used for comparison. MCT-ND4 (10 nM), MCT-COX1 (100 nM), MCT-ND6 (100 nM), MCT-ND5 (100 nM), and MCT-2 (500 nM) display full agonistic activity; MCT-ND3 (1 μM) has partial agonistic activity; and MCT-ND2 (1 μM) and MCT-ND4L (1 μM) have only minimal activating effects on neutrophils. MCT-ND1, MCT-ATP6, MCT-ATP8, MCT-COX2, and MCT-COX3 (all 1 μM) do not activate neutrophils for superoxide production. One representative of five independent experiments is shown. Mcpm, Mega cpm (106 cpm).

FIGURE 1.

Representative kinetics of maximal oxidative burst activity induced by formylated MCTs originating from 13 mitochondrial DNA–encoded proteins. Superoxide release from human neutrophils upon peptide stimulation was measured using the isoluminol/HRP-amplified chemiluminescence system. The prototype E. coli–derived FPR1 agonist fMLF (100 nM) and FPR2 agonist WKYMVM (100 nM) were used for comparison. MCT-ND4 (10 nM), MCT-COX1 (100 nM), MCT-ND6 (100 nM), MCT-ND5 (100 nM), and MCT-2 (500 nM) display full agonistic activity; MCT-ND3 (1 μM) has partial agonistic activity; and MCT-ND2 (1 μM) and MCT-ND4L (1 μM) have only minimal activating effects on neutrophils. MCT-ND1, MCT-ATP6, MCT-ATP8, MCT-COX2, and MCT-COX3 (all 1 μM) do not activate neutrophils for superoxide production. One representative of five independent experiments is shown. Mcpm, Mega cpm (106 cpm).

Close modal

Stimulation of neutrophils with MCT-ND2 and MCT-ND4L at a high concentration (up to 1 μM) activated NADPH oxidase but gave rise to a very low level of superoxide release compared with that induced by the potent activating MCTs, described above (Fig. 1). MCTs derived from the N termini of ND1, ATP6, ATP8, COX2, and COX3 proteins did not trigger superoxide release when tested in concentrations up to 1 μM (Fig. 1).

Our initial screening of the 13 MCTs in activating the human neutrophil NADPH oxidase revealed six of them as strong superoxide inducers, and 2 out of 13, namely ND2 and ND4L, were shown to be very weak neutrophil activators (Fig. 1) when their oxidase responses were compared with those induced by the prototype FPR1 agonist fMLF and FPR2 agonist WKYMVM (4) (Fig. 1). The six potent neutrophil-activating MCTs were characterized further, and their activation potencies were compared. Determination of the maximal responses and dose-response curves for the six potent MCTs identified all as full agonists, and they triggered neutrophils to produce O2 in a dose-dependent manner (Fig. 2). The activation potency was in the order MCT-ND4 > MCT-COX1 > MCT-ND6 > MCT-ND5 > MCT-2 > MCT-ND3, with MCT-ND3 being the least potent activator (EC50 = 360 nM, Fig. 2). The most potent peptide was MCT-ND4, with a peptide sequence originating from ND4, which activated neutrophils at low nanomolar concentrations (EC50 = 1 nM, Fig. 2). In fact, MCT-ND4 was as potent as the S. aureus–generated FPR1 high-affinity peptide fMIFL, heretofore the most potent FPR agonist described (16).

FIGURE 2.

Dose-response curves of active MCTs for neutrophil superoxide production. The neutrophil-activating MCTs were chosen for further dose-response determination. Neutrophils were activated with various peptide concentrations, as indicated, and the release of superoxide was recorded. Data are presented in descending order of potency, determined in an isoluminol/HRP-amplified chemiluminescence system. EC50 values range from 1 to 370 nM. Data are normalized to the maximal response (100%) of a respective titration of MCT-ND4 (10 nM), MCT-COX1 (100 nM), MCT-ND6 (250 nM), MCT-ND5 (500 nM), MCT-2 (1 μM), and MCT-ND3 (2 μM) and presented with a fitted curve. EC50 values (dashed lines) and 95% confidence intervals were calculated from n = 3 independent experiments (mean ± SD).

FIGURE 2.

Dose-response curves of active MCTs for neutrophil superoxide production. The neutrophil-activating MCTs were chosen for further dose-response determination. Neutrophils were activated with various peptide concentrations, as indicated, and the release of superoxide was recorded. Data are presented in descending order of potency, determined in an isoluminol/HRP-amplified chemiluminescence system. EC50 values range from 1 to 370 nM. Data are normalized to the maximal response (100%) of a respective titration of MCT-ND4 (10 nM), MCT-COX1 (100 nM), MCT-ND6 (250 nM), MCT-ND5 (500 nM), MCT-2 (1 μM), and MCT-ND3 (2 μM) and presented with a fitted curve. EC50 values (dashed lines) and 95% confidence intervals were calculated from n = 3 independent experiments (mean ± SD).

Close modal

The neutrophil NADPH oxidase activity induced by the neutrophil-activating MCTs was, when comparing the time-course/kinetics of superoxide release, very similar to that induced by the bacterial-derived prototype FPR1 agonist fMLF and the FPR2 agonist WKYMVM (Fig. 1). Accordingly, involvement of FPRs expressed by human neutrophils (FPR1 and FPR2) was determined, and this was achieved through two approaches. In the first, we determined the inhibitory effects mediated by the well-characterized FPR antagonists CysH (specific for FPR1) and PBP10 (specific for FPR2) when added alone or in combination, before stimulation with the MCTs. In accordance with the well-known selectivity of CysH for FPR1 and PBP10 for FPR2 (31, 36), CysH completely abolished the fMLF response, but not the WKYMVM response, whereas PBP10 inhibited the response induced by WKYMVM but not fMLF (Fig. 3A). The inhibitory profile of fMLF also applied to the activity induced by MCT-ND3 and MCT-ND6 (i.e., the responses induced by these MCTs were completely inhibited by CysH but not by PBP10) (Fig. 3A); thus, these two MCTs preferentially activate neutrophils through FPR1. In agreement with earlier published data showing that MCT-2 is an FPR2-specific agonist, addition of the FPR2-specific antagonist PBP10, but not CysH, completely abolished the release of superoxide induced by MCT-2 (Fig. 3A). Similarly, the responses induced by MCT-ND4 and MCT-ND5 were also completely inhibited in the presence of PBP10, but not CysH (Fig. 3A), strongly suggesting that they display a receptor preference for FPR2 in human neutrophils. However, the response induced by MCT-COX1 was only partially inhibited by CysH or PBP10; to obtain a complete inhibition, the two antagonists had to be combined (Fig. 3A). Thus, MCT-COX1 is a dual agonist that activates neutrophils through the engagement of FPR1 and FPR2. Our second approach to confirm the receptor specificity for the two most potent FPR-selective MCTs (FPR1-selective MCT-ND6 and FPR2-selective MCT-ND4) was to perform receptor-desensitization experiments. The well-characterized FPR-selective agonist fMLF (selective for FPR1) and WKYMVM (selective for FPR2) (37) were used. The results obtained from the second approach confirmed the receptor preference shown in Fig. 3A. Briefly, neutrophils activated with WKYMVM were desensitized to MCT-ND4 and vice versa, further supporting an FPR2 preference of MCT-ND4 (Fig. 3B). In line with this, fMLF-desensitized cells were still fully responsive to MCT-ND4 stimulation (Fig. 3B). When MCT-ND4 was replaced by MCT-ND6, a different desensitization pattern was observed. Cells desensitized with WKYMVM were still responsive to MCT-ND6, but MCT-ND6 and fMLF could desensitize each other, further supporting an FPR1 preference of MCT-ND6 (Fig. 3C).

FIGURE 3.

MCTs activate neutrophil superoxide production through distinct interaction with FPR1 and FPR2. (A) FPR1 or FPR2 specificity was determined with optimal concentrations of peptides in combination with FPR1 inhibitor CysH (1 μM), FPR2 inhibitor PBP10 (1 μM), or both. E. coli–derived FPR1 agonist fMLF (100 nM) and FPR2 agonist WKYMVM (100 nM) were used for comparison. MCT-ND6 (50 nM) and MCT-ND3 (500 nM) activate the neutrophil NADPH oxidase through FPR1 (black bars showing CysH sensitivity), whereas MCT-ND4 (2.5 nM), MCT-ND5 (50 nM), and MCT-2 (500 nM) are specific for FPR2 (gray bars showing PBP10 sensitivity). MCT-COX1 (50 nM) is a dual agonist with affinity for FPR1 and FPR2. The percentage of inhibition of MTC-mediated superoxide production in the presence of the respective FPR antagonists was calculated from individual agonist peak responses in the absence of antagonist (n = 3, mean + SD); 100% inhibition is represented by the dashed line. (B) Homologous desensitization between the FPR2 agonist WKYMVM (100 nM) and MCT-ND4 (2.5 nM, solid line). In reverse order, MCT-ND4 also desensitized cells for WKYMVM (dotted line), and FPR1 agonist fMLF (100 nM) had no effect on the MCT-ND4 (2.5 nM) response (dashed line). One representative of n = 3 independent experiments is shown. (C) Homologous desensitization between the FPR1 agonist fMLF (100 nM) and MCT-ND6 (100 nM; dashed line). In a similar manner, MCT-ND6 desensitized cells for fMLF (dotted line), whereas the FPR2 agonist WKYMVM (100 nM) had no effect on the MCT-ND6 response (100 nM; solid line). Agonists were added sequentially (arrows) (i.e., FPRs were desensitized by a first stimulation, followed by a second stimulation when the response had diminished). Data from n = 3 independent experiments are shown. Mcpm, Mega cpm (106 cpm).

FIGURE 3.

MCTs activate neutrophil superoxide production through distinct interaction with FPR1 and FPR2. (A) FPR1 or FPR2 specificity was determined with optimal concentrations of peptides in combination with FPR1 inhibitor CysH (1 μM), FPR2 inhibitor PBP10 (1 μM), or both. E. coli–derived FPR1 agonist fMLF (100 nM) and FPR2 agonist WKYMVM (100 nM) were used for comparison. MCT-ND6 (50 nM) and MCT-ND3 (500 nM) activate the neutrophil NADPH oxidase through FPR1 (black bars showing CysH sensitivity), whereas MCT-ND4 (2.5 nM), MCT-ND5 (50 nM), and MCT-2 (500 nM) are specific for FPR2 (gray bars showing PBP10 sensitivity). MCT-COX1 (50 nM) is a dual agonist with affinity for FPR1 and FPR2. The percentage of inhibition of MTC-mediated superoxide production in the presence of the respective FPR antagonists was calculated from individual agonist peak responses in the absence of antagonist (n = 3, mean + SD); 100% inhibition is represented by the dashed line. (B) Homologous desensitization between the FPR2 agonist WKYMVM (100 nM) and MCT-ND4 (2.5 nM, solid line). In reverse order, MCT-ND4 also desensitized cells for WKYMVM (dotted line), and FPR1 agonist fMLF (100 nM) had no effect on the MCT-ND4 (2.5 nM) response (dashed line). One representative of n = 3 independent experiments is shown. (C) Homologous desensitization between the FPR1 agonist fMLF (100 nM) and MCT-ND6 (100 nM; dashed line). In a similar manner, MCT-ND6 desensitized cells for fMLF (dotted line), whereas the FPR2 agonist WKYMVM (100 nM) had no effect on the MCT-ND6 response (100 nM; solid line). Agonists were added sequentially (arrows) (i.e., FPRs were desensitized by a first stimulation, followed by a second stimulation when the response had diminished). Data from n = 3 independent experiments are shown. Mcpm, Mega cpm (106 cpm).

Close modal

Taken together, our data reveal five novel FPR1- and/or FPR2-specific MCTs of mitochondrial origin that display activity at nanomolar concentrations. Among these, the most potent was shown to be FPR2-selective MCT-ND4, followed by the dual agonist MCT-COX1, and then FPR1-selective MCT-ND6.

The MCTs with peptide sequences present in ND1, COX2, COX3, ATP6, and ATP8 were all nonactivating in the neutrophil NADPH oxidase system, whereas MCT-ND2 and MCT-ND4L showed a low-activating profile (Fig. 1). It is clear from the data presented in Fig. 3 that neutrophils activated by an agonist become nonresponding when triggered with the same agonist once again or with another agonist recognized by the same receptor. This is achieved through a mechanism termed “homologous receptor desensitization” (37); however, a nonactivating receptor-specific ligand can hypothetically mediate inhibitory (antagonistic) effects by blocking binding of a receptor-activating agonist (4). In accordance with this, we next determined whether any of these nonactivating peptides were FPR antagonists/inhibitors. To study neutrophil activation in an antagonism mode, we treated neutrophils with the respective MCTs, followed by stimulation with fMLF or WKYMVM, and the inhibitory effects of the MCTs were monitored. As shown for MCT-ATP6, cells preincubated with this peptide did not affect the fMLF response (Fig. 4A) or the WKYMVM response (Fig. 4B), revealing that MCT-ATP6 lacks FPR-antagonistic activity. This was true for all nonactivating MCTs, and the two low-activating MCTs only inhibited the response induced by fMLF/WKYMVM to a very limited degree when used in concentrations up to 2 μM (Fig. 4C). The small inhibitory effect of MCT-ND2 and MCT-ND4L is most probably due to homologous receptor desensitization. Taken together, we show that the MCTs designed from the N termini of the mitochondrial proteins ND1, COX2, COX3, ATP6, and ATP8 lack the ability to activate or antagonize FPRs.

FIGURE 4.

Nonactivating MCTs do not inhibit FPR-mediated neutrophil superoxide production. All MCTs with no or low agonistic activity were added to cells for 5 min at 37°C before stimulation with fMLF or WKYMVM, and O2 was measured by isoluminol/HRP-amplified chemiluminescence. (A) Representative kinetics of the effect of MCT-ATP6 (1 μM) on FPR1-mediated superoxide release induced by the FPR1-specific agonist fMLF (50 nM). (B) Representative kinetics of the effect of MCT-ATP6 (1 μM) on the FPR2-mediated response induced by the FPR2-specific hexapeptide agonist WKYMVM (50 nM). (C) Summary of the inhibitory profiles of all mitochondrial peptides (1 μM) on the neutrophil response induced by fMLF (50 nM) and WKYMVM (50 nM). Inert MCT-ATP6, MCT-ATP8, MCT-COX2, and MCT-COX3 (all 1 μM) only exert minimal inhibitory effects on fMLF or WKYMVM activity. Low-activity MCT-ND2 (1 μM) and MCT-ND4L (1 μM) have weak inhibitory effects on FPR1 and FPR2 activity (n = 3, mean + SD). The 50% inhibition cutoff is represented by the dashed line.

FIGURE 4.

Nonactivating MCTs do not inhibit FPR-mediated neutrophil superoxide production. All MCTs with no or low agonistic activity were added to cells for 5 min at 37°C before stimulation with fMLF or WKYMVM, and O2 was measured by isoluminol/HRP-amplified chemiluminescence. (A) Representative kinetics of the effect of MCT-ATP6 (1 μM) on FPR1-mediated superoxide release induced by the FPR1-specific agonist fMLF (50 nM). (B) Representative kinetics of the effect of MCT-ATP6 (1 μM) on the FPR2-mediated response induced by the FPR2-specific hexapeptide agonist WKYMVM (50 nM). (C) Summary of the inhibitory profiles of all mitochondrial peptides (1 μM) on the neutrophil response induced by fMLF (50 nM) and WKYMVM (50 nM). Inert MCT-ATP6, MCT-ATP8, MCT-COX2, and MCT-COX3 (all 1 μM) only exert minimal inhibitory effects on fMLF or WKYMVM activity. Low-activity MCT-ND2 (1 μM) and MCT-ND4L (1 μM) have weak inhibitory effects on FPR1 and FPR2 activity (n = 3, mean + SD). The 50% inhibition cutoff is represented by the dashed line.

Close modal

Signals generated by activated FPRs affect the functions of other GPCRs, and this receptor cross-talk is hierarchical. The hierarchy is illustrated by the fact that FPR agonists desensitize neutrophils homologously and that CXCR1/2 (IL-8Rs) and the PAF receptor (PAFR) are also affected; CXCR1/2 are heterologously desensitized, whereas the response induced by the agonist-occupied PAFR is augmented (3841). It is important to point out that the augmented PAFR response is achieved through a novel receptor cross-talk mechanism that relies on signals downstream of PAFR to reactivate the desensitized FPRs (40, 41). To gain additional insights into the immunomodulatory effects of the FPR-activating peptides of mitochondrial origin, we used the above-identified MCT-ND6 (selective for FPR1) and MCT-4 (selective for FPR2), together with the dual FPR1/FPR2-agonistic MCT-COX1 (Fig. 3), and examined their effects on the responses induced by IL-8 (CXCR1/2 agonist) and PAF (agonist for PAFR). In agreement with the established model of cross-talk hierarchy, MCT-ND6–activated/desensitized neutrophils displayed an increased response upon a second stimulation with PAF (Fig. 5A, upper panel), but a decreased IL-8 response, when PAF and IL-8 responses were compared with their respective naive response (Fig. 5C). The primed PAF response in FPR-desensitized cells is primarily achieved through a reactivation of FPRs; this conclusion is drawn from the fact that the primed PAF response in MCT-ND4–, MCT-ND6–, and MCT-COX1–desensitized neutrophils was inhibited by the antagonist specific for the corresponding FPR (Fig. 5A, lower panel, Fig. 5B). In agreement with the above findings (Fig. 3), these data further strengthen the receptor selectivity for MCT-ND4 (FPR2), MCT-ND6 (FPR1), and MCT-COX (both FPRs). Similar to MCT-ND6, neutrophils desensitized with MCT-ND4 or MCT-COX1 also did not respond when activated by a second stimulation with IL-8 (Fig. 5C, 5D), further confirming the hierarchical cross-talk between FPR1/2 and CXCR1/2 in human neutrophils. It is worth mentioning that, in an experiment designed to study the effects of receptor antagonists, it is important to adjust the concentration to optimal (high submaximal) levels when comparing different agonists with different activation potencies.

FIGURE 5.

Cross-talk between FPRs stimulated with MCTs and the receptors for IL-8 and PAF. (A) Cells were stimulated with FPR1-specific MCT-ND6 (100 nM, first arrow); after the response had diminished, cells were reactivated with PAF (100 nM, second arrow) (upper panel). The PAF response derived from naive cells is shown for comparison (dotted line). Cells were stimulated with FPR1-specific MCT-ND6 (100 nM, first arrow); after the response had diminished, the FPR1 inhibitor CysH (1 μM, second arrow, dashed line) was added 1 min before PAF stimulation (third arrow) (lower panel). The PAF response derived from naive cells is shown for comparison (dotted line). (B) Summary of inhibitory profiles of FPR1 and FPR2 antagonists on the secondary PAF responses when added just prior to PAF stimulation. Cells were activated/desensitized with MCT-ND4 (5 nM), MCT-ND6 (100 nM), or MCT-COX1 (50 nM) before reactivation by PAF (100 nM). Data are mean + SD from n = 3 independent experiments. (C) Heterologous desensitization of IL-8–induced superoxide production (second arrow) through initial stimulation with MCT-ND6 (100 nM, first arrow) compared with IL-8 stimulation (500 ng/ml) of naive cells (dashed line) (n = 3). (D) Summary of MCT-ND4–mediated (5 nM), MCT-ND6–mediated (100 nM), and MCT-COX1–mediated (50 nM) receptor cross-talk with PAFR and of their inhibitory effect on IL-8R, with regard to neutrophil superoxide production, as a consequence of heterologous desensitization (n = 3, mean + SD). Naive responses are represented by the dashed line. Mcpm, Mega cpm (106 cpm).

FIGURE 5.

Cross-talk between FPRs stimulated with MCTs and the receptors for IL-8 and PAF. (A) Cells were stimulated with FPR1-specific MCT-ND6 (100 nM, first arrow); after the response had diminished, cells were reactivated with PAF (100 nM, second arrow) (upper panel). The PAF response derived from naive cells is shown for comparison (dotted line). Cells were stimulated with FPR1-specific MCT-ND6 (100 nM, first arrow); after the response had diminished, the FPR1 inhibitor CysH (1 μM, second arrow, dashed line) was added 1 min before PAF stimulation (third arrow) (lower panel). The PAF response derived from naive cells is shown for comparison (dotted line). (B) Summary of inhibitory profiles of FPR1 and FPR2 antagonists on the secondary PAF responses when added just prior to PAF stimulation. Cells were activated/desensitized with MCT-ND4 (5 nM), MCT-ND6 (100 nM), or MCT-COX1 (50 nM) before reactivation by PAF (100 nM). Data are mean + SD from n = 3 independent experiments. (C) Heterologous desensitization of IL-8–induced superoxide production (second arrow) through initial stimulation with MCT-ND6 (100 nM, first arrow) compared with IL-8 stimulation (500 ng/ml) of naive cells (dashed line) (n = 3). (D) Summary of MCT-ND4–mediated (5 nM), MCT-ND6–mediated (100 nM), and MCT-COX1–mediated (50 nM) receptor cross-talk with PAFR and of their inhibitory effect on IL-8R, with regard to neutrophil superoxide production, as a consequence of heterologous desensitization (n = 3, mean + SD). Naive responses are represented by the dashed line. Mcpm, Mega cpm (106 cpm).

Close modal

In summary, our results show that neutrophil-activating MCTs also modulate the responses induced by the CXCR1/2–IL-8 and PAFR–PAF receptor–agonist pairs, and this receptor communication is hierarchical.

In addition to their ability to activate the neutrophil NADPH oxidase, most FPR agonists trigger mobilization/secretion of neutrophil granule constituents. The ability of the three most potent MCTs with different FPR preferences (MCT-ND4 selective for FPR2, MCT-ND6 selective for FPR1, and the FPR1/2 dual agonist MCT-COX1) to trigger secretion has been demonstrated earlier using neutrophil-like HL-60 cells (21). We could confirm their ability to also trigger secretion in human neutrophils, as determined through measurement of the cleavage of CD62L (L-selectin) from the neutrophil cell surface and mobilization of neutrophil granule markers. The latter was measured by surface upregulation of CD11b (CR3) and CD66b, membrane markers stored primarily in the gelatinase and specific granules, respectively (42). All three MCTs (MCT-ND4, MCT-ND6, and MCT-COX1) promoted shedding of CD62L (Fig. 6A) and increased the expression of CD11b (Fig. 6B) and CD66b (Fig. 6C), demonstrating that they are also neutrophil secretagogues with similar potency as fMLF (Fig. 6).

FIGURE 6.

MCTs trigger L-selectin shedding and granule mobilization in human neutrophils. Neutrophil exposure to MCT-ND4 (10 and 100 nM, FPR2), MCT-ND6 (100 nM, FPR1), or MCT-COX1 (100 nM, dual agonist) for 10 min at 37°C induces CD62L shedding (A), upregulation of cell surface CD11b (B), and upregulation of cell surface CD66b (C) in a manner similar to the FPR1 agonist fMLF (10 and 100 nM). Data are presented as percentage of cleavage/upregulation compared with neutrophils incubated with buffer for 10 min at 37°C (mean + SD, n = 3). Dashed lines represent the unchanged control levels, depicted as 100%.

FIGURE 6.

MCTs trigger L-selectin shedding and granule mobilization in human neutrophils. Neutrophil exposure to MCT-ND4 (10 and 100 nM, FPR2), MCT-ND6 (100 nM, FPR1), or MCT-COX1 (100 nM, dual agonist) for 10 min at 37°C induces CD62L shedding (A), upregulation of cell surface CD11b (B), and upregulation of cell surface CD66b (C) in a manner similar to the FPR1 agonist fMLF (10 and 100 nM). Data are presented as percentage of cleavage/upregulation compared with neutrophils incubated with buffer for 10 min at 37°C (mean + SD, n = 3). Dashed lines represent the unchanged control levels, depicted as 100%.

Close modal

Many GPCRs promote, upon receptor-specific agonist binding, β-arrestin translocation, which is suggested to be of importance for the subsequent desensitization, internalization, and/or initiation of β-arrestin–regulated signaling pathways (43). The precise role of β-arrestin in FPR activation and neutrophil function has not been clarified, but our earlier studies clearly illustrate that some FPR agonists could trigger neutrophil NADPH oxidase activation without β-arrestin recruitment (35). We next determined the abilities of all 13 MCTs to recruit β-arrestin using PathHunter enzyme fragment complementation technology (DiscoverX). CHO cells expressing β-arrestin, together with FPR1 or FPR2 (35), were activated by MCTs. In line with the receptor specificity for the prototype FPR1/FPR2 agonists, fMLF induced β-arrestin recruitment solely in FPR1-expressing cells, whereas WKYMVM induced recruitment only in cells expressing FPR2 (Fig. 7A). In agreement with the activation profile obtained in the neutrophil NADPH oxidase assay (Fig. 1), the nonactivating MCTs derived from ATP6, ATP8, COX2, COX3, and ND1 also were unable to trigger β-arrestin recruitment in FPR-expressing cells (determined in concentrations up to 2 μM, Fig. 7B). Of note, MCT-ND4L, which was a very weak NADPH oxidase activator (Fig. 1), also triggered weak translocation of β-arrestin in cells overexpressing FPR1 (Fig. 7B).

FIGURE 7.

MCTs induce recruitment of β-arrestin in FPR-overexpressing CHO cells. PathHunter CHO cells that coexpress ProLink-tagged FPR1 or FPR2 and enzyme acceptor tagged β-arrestin 2 were challenged with FPR agonists, and arrestin binding was measured as β-galactosidase activity through enzyme fragment complementation. (A) Representative results for FPR1-overexpressing cells stimulated with the control agonist fMLF (100 nM) and for FPR2-overexpressing cells stimulated with the control agonist WKYMVM (100 nM). Data are mean + SD from n = 3 independent experiments. (B) FPR1 and FPR2 cells were stimulated with low/nonactivating MCT-ATP6, MCT-ATP8, MCT-COX2, MCT-COX3, MCT-ND1, or MCT-ND4L (all 1 μM), and β-arrestin recruitment was determined (n = 3, mean + SD). The positive control responses for fMLF from FPR1 cells and WKYMVM from FPR2 cells are highlighted as 100% (dashed line). (C) COX1 (250 nM) induces β-arrestin recruitment in FPR1 cells and FPR2 cells to the same extent as the respective controls (represented by the dashed cutoff line). ND6 (250 nM) mediates β-arrestin recruitment in FPR1 cells, and ND2 (1 μM) and ND3 (1 μM) partially activate FPR1 cells. MCT-2 (250 nM), ND4 (25 nM), and ND5 (100 nM) fully activate FPR2 cells for β-arrestin recruitment (n = 3, mean + SD).

FIGURE 7.

MCTs induce recruitment of β-arrestin in FPR-overexpressing CHO cells. PathHunter CHO cells that coexpress ProLink-tagged FPR1 or FPR2 and enzyme acceptor tagged β-arrestin 2 were challenged with FPR agonists, and arrestin binding was measured as β-galactosidase activity through enzyme fragment complementation. (A) Representative results for FPR1-overexpressing cells stimulated with the control agonist fMLF (100 nM) and for FPR2-overexpressing cells stimulated with the control agonist WKYMVM (100 nM). Data are mean + SD from n = 3 independent experiments. (B) FPR1 and FPR2 cells were stimulated with low/nonactivating MCT-ATP6, MCT-ATP8, MCT-COX2, MCT-COX3, MCT-ND1, or MCT-ND4L (all 1 μM), and β-arrestin recruitment was determined (n = 3, mean + SD). The positive control responses for fMLF from FPR1 cells and WKYMVM from FPR2 cells are highlighted as 100% (dashed line). (C) COX1 (250 nM) induces β-arrestin recruitment in FPR1 cells and FPR2 cells to the same extent as the respective controls (represented by the dashed cutoff line). ND6 (250 nM) mediates β-arrestin recruitment in FPR1 cells, and ND2 (1 μM) and ND3 (1 μM) partially activate FPR1 cells. MCT-2 (250 nM), ND4 (25 nM), and ND5 (100 nM) fully activate FPR2 cells for β-arrestin recruitment (n = 3, mean + SD).

Close modal

Also in agreement with the activation profile in the neutrophil NADPH oxidase system, MCTs from COX1, ND2, ND3, ND6, cytochrome b, ND4, and ND5 proteins that potently triggered superoxide release (Figs. 1, 2) also recruited β-arrestin, and the respective receptor preference was the same in the two assay systems: MCTs shown to be FPR1 selective in the neutrophil/NADPH oxidase assay system (i.e., MCT-ND3 and MCT-ND6) selectively induced β-arrestin recruitment in FPR1-overexpressing cells, and the FPR2-selective MCTs (i.e., MCT-2, MCT-ND4, and MCT-ND5) preferred this receptor in both assay systems (Fig. 7C). MCT-COX1, which was shown to be a dual FPR1/FPR2 agonist (Fig. 3A), potently triggered β-arrestin translocation in FPR1- and FPR2-overexpressing cells (Fig. 7C). The efficacy (Emax) of β-arrestin recruitment of FPR2-specific MCT-2, MCT-ND4, and MCT-ND5 was in a similar range as that for the receptor-specific prototype agonist WKYMVM (Fig. 7C, dashed line), and this was also true for the FPR1-specific MCT-ND6 compared with the FPR1 receptor–specific prototype agonist fMLF (Fig. 7C, dashed line). When the maximal responses in β-arrestin recruitment induced by MCT-ND2 and MCT-ND3 were compared, MCT-ND2 was a stronger stimulus than MCT-ND3 (Fig. 7C) (i.e., the order of potency with regard to these two FPR1-selective MCTs was reversed with regard to β-arrestin recruitment and NADPH oxidase activation) (Fig. 1). The fact that the intrinsic efficacy of FPR1-specific MCT-ND2 and MCT-ND3 was reduced (compared with fMLF) suggests that these should be regarded as partial β-arrestin–recruiting agonists. An overview with detailed MCT dose-response curves is shown in Fig. 8. In summary, we show that all NADPH oxidase–activating MCTs also have the ability to promote FPRs to recruit β-arrestin. In addition, their activation profile in cells overexpressing FPR1 or FPR2 confirmed their receptor preference in primary neutrophils obtained above with receptor-specific agonists and antagonists.

FIGURE 8.

Dose-response curves of active MCTs for β-arrestin recruitment. Recruitment of β-arrestin was measured in CHO cells overexpressing ProLink-tagged FPR1 or FPR2 as β-galactosidase activity. MTCs are listed in descending order of potency, EC50 values range from 10 to 387 nM, as represented by the dashed lines. Data are normalized to the maximal response (100%) of a respective titration and presented with a fitted curve. EC50 values and 95% confidence intervals were calculated from n = 3 independent experiments (mean + SD).

FIGURE 8.

Dose-response curves of active MCTs for β-arrestin recruitment. Recruitment of β-arrestin was measured in CHO cells overexpressing ProLink-tagged FPR1 or FPR2 as β-galactosidase activity. MTCs are listed in descending order of potency, EC50 values range from 10 to 387 nM, as represented by the dashed lines. Data are normalized to the maximal response (100%) of a respective titration and presented with a fitted curve. EC50 values and 95% confidence intervals were calculated from n = 3 independent experiments (mean + SD).

Close modal

Our data presented above demonstrate that MCT-ND2, when compared with MCT-ND3 (EC50 ∼ 400 nM), is a very modest agonist that poorly triggers the release of superoxide in the NADPH oxidase system; however, in the β-arrestin–recruitment system, MCT-ND2 was stronger than MCT-ND3 (Figs. 1, 7). This suggests that MCTs may display biased signaling properties downstream of FPRs. To gain more insights into potential MCT signaling bias, we determined their potencies for β-arrestin recruitment and compared it with their respective potencies to activate NADPH oxidase. The prototype FPR agonists fMLF and WKYMVM were used to generate the reference ratios (0.3 for fMLF and 0.5 for WKYMVM, Table II). The recruitment of β-arrestin by MCT-ND2 was very similar (or even somewhat higher in potency) to that of MCT-ND3; however, because MCT-ND2 was very weak at triggering superoxide release (Fig. 1), it was not possible to calculate its potency ratio (Table II). A possible signaling bias was also noted for MCT-ND4, because the relative potencies in the activation for NADPH oxidase and the ability to recruit β-arrestin indicated a preference for triggering NADPH oxidase activation over the recruitment of β-arrestin (potency ratio of β-arrestin/superoxide release was 9.4, Table II). The potency ratio for MCT-ND6 (∼4) and MCT-COX1 (∼4) was greater than that for fMLF and WKYMVM, indicating some degree of biased agonism for MCT-ND6 and MCT-COX1 as well (Table II). MCT-ND5 and MCT-2 generated potency ratios similar to that of fMLF or WKYMVM (Table II).

Table II.
Properties of activating peptides and MCTs
Peptide or MCTReceptor AffinityEC50 Values (nM)aFold Changeb O2 → β-Arrestin
O2 Productionβ-Arrestin Recruitment
fMLF FPR1 20 0.35 
WKYMVM FPR2 40 20 0.5 
MCT-2 FPR2 160 80 0.5 
MCT-COX1 FPR1/2 22 60/100 2.73/4.55 
MCT-ND2 FPR1 ND 353 ND 
MCT-ND3 FPR1 370 416 1.12 
MCT-ND4 FPR2 9.4 9.4 
MCT-ND5 FPR2 31 24 0.77 
MCT-ND6 FPR1 27 107 3.96 
Peptide or MCTReceptor AffinityEC50 Values (nM)aFold Changeb O2 → β-Arrestin
O2 Productionβ-Arrestin Recruitment
fMLF FPR1 20 0.35 
WKYMVM FPR2 40 20 0.5 
MCT-2 FPR2 160 80 0.5 
MCT-COX1 FPR1/2 22 60/100 2.73/4.55 
MCT-ND2 FPR1 ND 353 ND 
MCT-ND3 FPR1 370 416 1.12 
MCT-ND4 FPR2 9.4 9.4 
MCT-ND5 FPR2 31 24 0.77 
MCT-ND6 FPR1 27 107 3.96 
a

EC50 values were obtained from Figs. 2 and 8.

b

Fold change = EC50 (β-arrestin)/EC50 (O2).

Taken together, our data reveal some signaling differences among MCTs, and when comparing NADPH oxidase activity in neutrophils with their ability to recruit β-arrestin in CHO cells, it is evident that MCT-ND4 is most potent and is also a seemingly biased FPR2 agonist.

To measure the ability of MCTs to induce neutrophil chemotaxis, we used a classical Boyden/Transwell chamber system. Initially, the chemotactic properties of the three most potent MCTs (FPR2-activating MCT-ND4, FPR1-activating MCT-ND6, and the dual agonist MCT-COX1) were examined. The neutrophils were placed on top of the filter, allowing migration toward the presumed attractant placed below the filter. A dose-dependent neutrophil migration was induced by all three MCTs, with potencies in the same range as the prototype FPR1 agonist fMLF (Fig. 9, fMLF was used as the 100% control). Interestingly, we observed that the potency required for MCT-ND4 (Fig. 9B) in cell migration was very similar to that in β-arrestin recruitment but not in NADPH oxidase activation. Similarly, MCT-ND6 and MCT-COX1 displayed comparable potencies for β-arrestin recruitment and chemotaxis (Fig. 9C, 9D, Table II). This indicates an association between β-arrestin recruitment and migration. We studied the migration potency of MCT-ND2, a very weak agonist in the NADPH oxidase system (Fig. 1) but a fairly strong trigger with regard to β-arrestin recruitment (Fig. 8, Table II). MCT-ND2 also induced a dose-dependent migration (Fig. 9A), and the potency was comparable to that required for β-arrestin recruitment (Fig. 8). The notion that β-arrestin recruitment associates with chemotaxis is in agreement with our recently published data showing that an FPR2-specific pepducin activates NADPH oxidase but lacks the ability to recruit β-arrestin and does not induce neutrophil chemotaxis (35). Taken together, these data demonstrate that the novel nanomolar-potent MCTs also induce chemotaxis. In addition, their potencies in chemotaxis are very similar to those in recruiting β-arrestin, supporting a critical role for β-arrestin in neutrophil migration.

FIGURE 9.

Neutrophil migration induced by MCTs. The chemotactic potential of different concentrations of MCT-ND2 (25–3000 nM) (A), MCT-ND4 (0.5–2000 nM) (B), MCT-ND6 (0.5–2000 nM) (C), and MCT-COX1 (1–2000 nM) (D) was examined via cell migration for 90 min through a 3-μm-pore filter membrane toward various concentrations of peptides, which were added to the bottom wells of the respective Transwell chambers. Quantitative analysis was performed by absorbance-based myeloperoxidase activity measurements, and the fraction of cells (percentage of those added) in the lower compartments was determined. Data are presented as percentage of control neutrophil chemoattractant fMLF (10 nM). Concentrations of 0 nM indicate spontaneous migration (mean + SD, n = 3).

FIGURE 9.

Neutrophil migration induced by MCTs. The chemotactic potential of different concentrations of MCT-ND2 (25–3000 nM) (A), MCT-ND4 (0.5–2000 nM) (B), MCT-ND6 (0.5–2000 nM) (C), and MCT-COX1 (1–2000 nM) (D) was examined via cell migration for 90 min through a 3-μm-pore filter membrane toward various concentrations of peptides, which were added to the bottom wells of the respective Transwell chambers. Quantitative analysis was performed by absorbance-based myeloperoxidase activity measurements, and the fraction of cells (percentage of those added) in the lower compartments was determined. Data are presented as percentage of control neutrophil chemoattractant fMLF (10 nM). Concentrations of 0 nM indicate spontaneous migration (mean + SD, n = 3).

Close modal

It has been well accepted that mitochondrial release of danger molecules that are able to modulate innate immune reactivity and inflammation are of pathophysiological importance (23, 24, 27). Such danger molecules include reactive oxygen species generated during mitochondrial respiration, as well as ATP, mitochondrial DNA, and proteins/peptides with an fMet in their N terminus encoded for by this DNA, and they are supposed to be released during tissue destruction and cell necrosis. Thus, formylated peptides that originate from mitochondria have been regarded as endogenous danger molecules that are sensed by immune cells as DAMPs, but knowledge about the molecular link and the importance of such mitochondrial peptides and inflammation remains limited. In this study, we measured neutrophil recognition and associated activation, mediated by peptides predicted to be proteolytic cleavage products formed from the proteins encoded by the mitochondrial DNA which are translated with an fMet in their respective N termini. Such N-terminal formylated MCTs were synthesized for all 13 mitochondrial-encoded proteins; 8 of these are neutrophil activators, 6 of which were found to be very potent triggers of the neutrophil superoxide–generating NADPH oxidase that is active in nanomolar concentrations. These data confirm earlier results obtained with endogenous N-formylated mitochondrial FPR-activating peptides (21), but add several new mitochondrial agonists, two of which (MCT-ND4 and MCT-ND6) are potent but have a preference for FPR2 and FPR1, respectively. We also show that MCT-ND4 and MCT-ND2 may act as FPR-biased agonists that preferentially activate neutrophil NADPH oxidase or recruitment of β-arrestin, respectively. Further, our data demonstrate that the mitochondrial peptides also modulate neutrophil functions induced by IL-8 and PAF through receptor cross-talk mechanisms.

Earlier studies, in which the activity of synthetic peptides derived from the 13 mitochondrial proteins was investigated, demonstrated that MCTs derived from cytochrome b (MCT-2), COX1, ND4, ND5, and ND6 activate neutrophil-like HL-60 cells (21). In this study, we examined the effects of these peptides on primary human neutrophils and focused on the activation of superoxide-generating NADPH oxidase and the basal characterization of their targeting receptors. Using a sensitive neutrophil NADPH oxidase–activation assay system, we could confirm the activity of MCTs MCT-2, MCT-COX1, MCT-ND4, MCT-ND5, and MCT-ND6. These peptides activated human neutrophils to release superoxide in nanomolar concentrations, with the most potent MCT-ND4 having an EC50 value of 1 nM. In addition, we found that MCT-ND2 and MCT-ND3 activate neutrophils but with much lower potencies than MCT-ND4 (EC50 of 370 nM for MCT-ND3; an EC50 value for MCT-ND2 could not be determined because it is such a poor NADPH oxidase activator). We also confirm that MCT-ND1, MCT-COX2, MCT-COX3, MCT-ATP6, MCT-ATP8, and MCT-ND4L are unable to activate neutrophils. With respect to the nonactivating MCT-ND1, it is interesting to note that the amino acid sequence of human ND1 differs significantly from that in rat, from which a peptide (known to bind the MHC class I H2M3 complex) has been demonstrated to be a very potent activator of human and mouse neutrophils (44). Moreover, the peptides lacking the ability to activate neutrophil NADPH oxidase also lacked the ability to recruit β-arrestin and to inhibit (antagonize) FPR1/FPR2-induced activity.

The neutrophil-activating MCT-2 derived from mitochondrial cytochrome b has been isolated from porcine heart (28), suggesting that this peptide may be formed in vivo; however, whether the other cryptic formylated peptides really are present in tissues or are generated/released when tissues are damaged remains to be determined, but it is very conceivable that peptides with identical or very similar sequences could be generated in vivo. This possibility relies on the following assumptions: that the mitochondrial proteins are cleaved by endogenous proteases at the predicted cleavage sites and that these peptides retain their N-terminal formylated group. It is clear from animal models and in vivo/in situ data on the effects of receptor-specific inhibitors/antagonists that agonists for the FPRs are of importance in the initiation, as well as maintenance and resolution, of inflammatory reactions. However, to precisely define the ligands involved, advanced biochemical separation techniques and peptide sequence analysis of normal and destroyed tissues have to be performed, and identification of this type of host-derived cryptic danger peptide is of importance for understanding their role in immunomodulation in health and disease. It is worth noting that, in addition to the formyl peptides, nonformylated cryptic peptides may be formed by proteolytic cleavage of proteins released during organ/tissue destruction. It has been shown that peptides derived from cytochrome c oxidase subunit VIII can also activate neutrophils; however, unlike the formylated ones, they may bypass the FPRs and directly activate the signaling Gi protein (45). Moreover, nonformylated peptides originating from the N terminus of the Ca2+-regulated phospholipid-binding protein annexin I interact with FPRs, despite the lack of an N-terminal fMet (15). However, it is important to note that most of the reported nonformylated peptides require micromolar or even higher concentrations to activate neutrophils, whereas the formyl peptides used in this study display activity in the low nanomolar range. To facilitate our understanding of the biological roles of mitochondrial-derived, as well as nonmitochondrial-derived, neutrophil-activating peptides and the corresponding receptors, an additional strategy would be to generate specific Abs against the active peptides (46, 47) or the receptors. These approaches could be powerful tools that enable us to localize such peptides/receptors and determine their biological roles, but they lie beyond the scope of this study.

At the receptor level, all mitochondrial DNA–encoded neutrophil-activating peptides studied mediate their functions through interaction with FPRs, but they are recognized differently by FPR1 and FPR2. Similar to MCT-2, MCT-ND4 and MCT-ND5 mediate their responses preferentially through FPR2, whereas MCT-ND3 and MCT-ND6 display receptor preference for FPR1. MCT-COX1 was recognized by both FPRs and should be regarded as a dual agonist with fairly equal affinity for the two receptors. The precise structure present in these host-derived formyl peptides that determines receptor preference is not known; however, previous findings on the structure–activity relationship of different peptides produced and released by S. aureus bacteria, including phenol-soluble modulins, show that the amino acids close to fMet constitute the basis for the receptor preference. Yet, also the size of the peptide and the properties of the peptide structures present outside of the presumed binding pocket in the receptor are of importance in determining receptor preference (16, 17). It seems that addition of new amino acids to the C terminus of a short FPR1-selective peptide shifts the preference to FPR2. In contrast, C-terminal truncations of peptides that prefer FPR2 shift the preference to FPR1 (16). This trend also seems to fit for the mitochondrial peptides: the longer MCT-ND5 (28 aa residues) and MCT-ND4 (20 aa residues) prefer FPR2, the shorter ones (MCT-ND3, 5 aa residues and MCT-ND6, 6 aa residues) prefer FPR1, and the one in the middle, MCT-COX1 (13 aa residues), is a dual FPR agonist. The receptor preference shift from FPR2 to FPR1, achieved through a C-terminal truncation, is obviously relevant for MCT-ND4 and MCT-ND6, as illustrated by the fact that shorter variants of these peptides have been shown to be dual FPR2 agonists in receptor-overexpressing HL-60 cells (48). There is no direct correlation between peptide size and activation potency, as clearly illustrated by the fact that the most potent, MCT-ND4, has 20 aa residues compared with the much less potent MCT-ND5 and MCT-2, which have 28 and 15 aa residues, respectively. MCT-ND6 and MCT-ND3 have almost the same number of amino acids, but they differ with regard to oxidase-activating potency.

It is clear from our data that there is a cross-talk between FPRs and the receptor for IL-8 (CXCR1/2), but this inhibitory receptor cross-talk is by no means unique to mitochondrial peptides (38, 39). The novel cross-talk between PAFR and the agonist-occupied FPRs that leads to reactivation of desensitized receptors occupied with mitochondrial peptides was described for other FPR agonists (40, 41). Our data on receptor cross-talk add another layer of complexity to immune regulation and the role of neutrophils in organ injury that probably is not unique to host-derived formyl peptides but, rather, is a general phenomenon of many inflammatory mediators.

Similar to other FPR agonists, the formylated peptides originating from mitochondrial peptides triggered directional neutrophil migration. We have shown earlier that a biased FPR2 agonist (35) that lacks the ability to recruit β-arrestin also lacks the ability to recruit neutrophils chemotactically (35), suggesting that the signaling pathway downstream of β-arrestin may have a critical role in the sensing of chemoattractants. Interestingly, in this study, we have identified two mitochondrial peptides that presumably mediate a functional selective activation profile: MCT-ND2 presumably being a biased FPR1-selective agonist that preferentially recruits β-arrestin over the signaling pathway, leading to superoxide release and MCT-ND4 being a biased FPR2-selective agonist that preferentially activates NADPH oxidase over β-arrestin recruitment. Further characterization of MCT-ND2 and MCT-ND4 in chemotaxis experiments revealed that MCT-ND2 is a potent chemoattractant, supporting the critical role of β-arrestin translocation in mediating neutrophil migration (35). In comparison with MCT-ND2, the FPR2 agonist MCT-ND4 is a potent oxidase inducer with an EC50 value of 1 nM; however, a much higher concentration is required to trigger β-arrestin translocation, indicating that MCT-ND4 may have a signaling bias toward NADPH oxidase activation. Biased agonism is a newly emerged concept that is highly appreciated in GPCR research; in simple terms, it describes agonists that can selectivity trigger one signaling pathway over another (49). The biological relevance of biased agonism has just started to be explored. For FPRs, a recent study using an FPR-biased small compound has revealed that this agonist has superior cardioprotection compared with the well-characterized unbiased/balanced FPR agonist compound 43 (50). The identification of endogenous FPR-biased agonists should provide novel molecular tools to dissect different signaling pathways and a particular cellular response (i.e., chemotaxis, secretion, and NADPH oxidase activation) downstream of FPRs in vitro and in vivo. However, it is important to mention that, although the topic of biased agonism gained much attention recently (49), it is still very challenging to study biased signaling properties in primary cells, including neutrophils (51). For FPRs, the ideal way to conduct such studies would be to use the same cell system, preferably neutrophils, because they would be the relevant cell type. However, the strategy commonly used by us and other investigators has been to determine β-arrestin recruitment in transfected cells overexpressing the receptor of interest, because no method allows quantitative measurement of GPCR-mediated β-arrestin recruitment in primary neutrophils. An alternative approach would be to determine a signaling event downstream of β-arrestin; however, no such signaling event has been described in neutrophils. It is clear that ERK phosphorylation is an event downstream of β-arrestin for many GPCRs expressed by other cells, but FPR-mediated ERK1/2 phosphorylation is G protein dependent and also occurs in cells that lack β-arrestin (52, 53). These results are in agreement with our unpublished observations obtained when comparing ERK phosphorylation in neutrophils activated with a conventional FPR ligand triggering β-arrestin recruitment and an FPR-specific pepducin that does not trigger arrestin recruitment (35). Despite the differences in their ability to recruit β-arrestin, both of these FPR agonists induced ERK1/2 phosphorylation (H. Forsman, M. Gabl, and M. Sundqvist, unpublished observations).

Currently, very little is known about the biological roles of mitochondrial-derived inflammatory modulating molecules, including their formyl peptides and mitochondrial DNA. Our data demonstrating that these host-derived formylated MCTs are highly potent in activating neutrophils support the notion that, when released from disintegrating mitochondria, these molecules may play important roles in the inflammatory response resulting from tissue injury (21, 23, 24). Indeed, elevated levels of mitochondrial-derived formyl peptides have been detected in bronchoalveolar lavage fluid and serum in patients with acute respiratory distress syndrome (54). Activation of circulating neutrophils by mitochondrial-derived formyl peptides during cell damage has also been described in trauma patients and has been the suggested cause of systemic inflammatory response syndrome with sepsis-like clinical manifestations (25). Moreover, when given to animals, mitochondrial crude extracts trigger a marked acute inflammation–associated tissue injury (25, 55). In many diseases, including rheumatoid arthritis and lung injury, neutrophils are the major cell type that causes inflammation-associated tissue damage through a release of granule-stored proteases (5658). Identification and determination of chemical structures of potent neutrophil-activating factors released from host tissue during sterile inflammation would be the final proof of the presence of these MCTs in vivo, and this knowledge is essential for a better understanding of the underlying disease pathogenesis and for more effective treatment of these neutrophil-dominant inflammatory disorders. Many investigators have attempted to address this issue by demonstrating that human mitochondrial extracts, as well as necrotic cells, release activating compounds suggested to target FPR, based on the inhibitory effects of FPR antagonist (22, 25, 59), rather than a direct identification from an in vivo source. In addition, the endogenous neutrophil agonists originating from mitochondrial/cell extracts have not been shown to be formyl peptides. To our knowledge, among the 13 mitochondrial DNA–encoded proteins, MCT-2 purified from porcine heart extracts is the only endogenous formyl peptide that has been identified (28). We are still awaiting the identification of mitochondrial peptides in human tissues, but we have very recently started a novel approach by generating specific mAbs against the formyl peptides used in this study, which will be used to probe the peptides and inhibit their activities. We have successfully generated Abs against the N-terminal sequence of human MCT-2 (46) and MCT-COX1 used in this study, and immunoreactivity against MCT-2 and MCT-COX1 is detected in mitochondrial DAMPs prepared from human cultured cells (T. Marutani and H. Mukai, unpublished observations). In addition, generation of Abs against other MCT peptides used in this study are in progress. We hope that specific mAbs against these “theoretic mitochondrial-derived formyl peptides” will facilitate future investigations to understand the role of formyl peptides in aseptic inflammation reactivity and tissue injury.

In summary, the current study reveals several N-terminal formylated MCTs originating from proteins encoded by mitochondrial DNA as potent FPR agonists with activity in the low nanomolar range. These MCTs display different receptor preferences for FPR1 and/or FPR2 expressed in human neutrophils, and some of them also seem to exhibit a biased signaling profile for G protein activation versus β-arrestin recruitment. Our data highlight a key role for FPRs in bacterial infections, as well as in aseptic inflammation and tissue injury, through the recognition of mitochondrial-derived formyl peptides. Future attempts to identify and determine the structure of these endogenous formylated MCTs may facilitate our understanding of their pathophysiological roles and lead to the development of FPR-based therapeutic strategies for organ failure and tissue injury.

We thank members of the Phagocyte Research Group for valuable suggestions.

This work was supported by the Swedish Research Council, the King Gustaf V 80-Year Foundation, the Swedish government under the Avtal om Läkarutbildning och Forskning agreement, the Clas Groschinsky Memorial Foundation, the Ingabritt and Arne Lundberg Foundation, the Wilhelm and Martina Lundgrens Scientific Foundation, the Rådman and Mrs. Ernst Collianders Foundation, and by a research grant from the Ministry of Education, Culture, Sports, Science and Technology of Japan (25350971).

Abbreviations used in this article:

ATP6

ATP synthase subunit protein 6

CL

isoluminol-ECL

COX1

cytochrome c oxidase subunit I

CysH

cyclosporine H

DAMP

danger-associated molecular pattern

fMet

formylated methionine

FPR

formyl peptide receptor

KRG

Krebs–Ringer phosphate buffer supplemented with glucose (10 mM), Ca2+ (1 mM), and Mg2+ (1.5 mM, pH 7.3)

MCT

mitochondrial cryptic peptide (mitocryptide)

ND6

NADH dehydrogenase subunit 6

PAF

platelet-activating factor

PAFR

PAF receptor.

1
Hoffmann
,
J. A.
,
F. C.
Kafatos
,
C. A.
Janeway
,
R. A.
Ezekowitz
.
1999
.
Phylogenetic perspectives in innate immunity.
Science
284
:
1313
1318
.
2
Medzhitov
,
R.
,
C. A.
Janeway
Jr.
1997
.
Innate immunity: the virtues of a nonclonal system of recognition.
Cell
91
:
295
298
.
3
Nauseef
,
W. M.
,
N.
Borregaard
.
2014
.
Neutrophils at work.
Nat. Immunol.
15
:
602
611
.
4
Dahlgren
,
C.
,
M.
Gabl
,
A.
Holdfeldt
,
M.
Winther
,
H.
Forsman
.
2016
.
Basic characteristics of the neutrophil receptors that recognize formylated peptides, a danger-associated molecular pattern generated by bacteria and mitochondria.
Biochem. Pharmacol.
114
:
22
39
.
5
Ye
,
R. D.
,
F.
Boulay
,
J. M.
Wang
,
C.
Dahlgren
,
C.
Gerard
,
M.
Parmentier
,
C. N.
Serhan
,
P. M.
Murphy
.
2009
.
International union of basic and clinical pharmacology. LXXIII. Nomenclature for the formyl peptide receptor (FPR) family.
Pharmacol. Rev.
61
:
119
161
.
6
Fu
,
H.
,
J.
Karlsson
,
J.
Bylund
,
C.
Movitz
,
A.
Karlsson
,
C.
Dahlgren
.
2006
.
Ligand recognition and activation of formyl peptide receptors in neutrophils.
J. Leukoc. Biol.
79
:
247
256
.
7
Schiffmann
,
E.
,
B. A.
Corcoran
,
S. M.
Wahl
.
1975
.
N-formylmethionyl peptides as chemoattractants for leucocytes.
Proc. Natl. Acad. Sci. USA
72
:
1059
1062
.
8
Adams
,
J. M.
,
M. R.
Capecchi
.
1966
.
N-formylmethionyl-sRNA as the initiator of protein synthesis.
Proc. Natl. Acad. Sci. USA
55
:
147
155
.
9
Adams
,
J. M.
1968
.
On the release of the formyl group from nascent protein.
J. Mol. Biol.
33
:
571
589
.
10
Wang
,
R.
,
K. R.
Braughton
,
D.
Kretschmer
,
T. H.
Bach
,
S. Y.
Queck
,
M.
Li
,
A. D.
Kennedy
,
D. W.
Dorward
,
S. J.
Klebanoff
,
A.
Peschel
, et al
.
2007
.
Identification of novel cytolytic peptides as key virulence determinants for community-associated MRSA.
Nat. Med.
13
:
1510
1514
.
11
Marasco
,
W. A.
,
S. H.
Phan
,
H.
Krutzsch
,
H. J.
Showell
,
D. E.
Feltner
,
R.
Nairn
,
E. L.
Becker
,
P. A.
Ward
.
1984
.
Purification and identification of formyl-methionyl-leucyl-phenylalanine as the major peptide neutrophil chemotactic factor produced by Escherichia coli.
J. Biol. Chem.
259
:
5430
5439
.
12
Fu
,
H.
,
C.
Dahlgren
,
J.
Bylund
.
2003
.
Subinhibitory concentrations of the deformylase inhibitor actinonin increase bacterial release of neutrophil-activating peptides: a new approach to antimicrobial chemotherapy.
Antimicrob. Agents Chemother.
47
:
2545
2550
.
13
Southgate
,
E. L.
,
R. L.
He
,
J. L.
Gao
,
P. M.
Murphy
,
M.
Nanamori
,
R. D.
Ye
.
2008
.
Identification of formyl peptides from Listeria monocytogenes and Staphylococcus aureus as potent chemoattractants for mouse neutrophils.
J. Immunol.
181
:
1429
1437
.
14
He
,
H. Q.
,
R. D.
Ye
.
2017
.
The formyl peptide receptors: diversity of ligands and mechanism for recognition.
Molecules
22
:
E455
.
15
Movitz
,
C.
,
L.
Brive
,
K.
Hellstrand
,
M. J.
Rabiet
,
C.
Dahlgren
.
2010
.
The annexin I sequence gln(9)-ala(10)-trp(11)-phe(12) is a core structure for interaction with the formyl peptide receptor 1.
J. Biol. Chem.
285
:
14338
14345
.
16
Forsman
,
H.
,
M.
Winther
,
M.
Gabl
,
S. L.
Skovbakke
,
F.
Boulay
,
M. J.
Rabiet
,
C.
Dahlgren
.
2015
.
Structural changes of the ligand and of the receptor alters the receptor preference for neutrophil activating peptides starting with a formylmethionyl group.
Biochim. Biophys. Acta
1853
:
192
200
.
17
Kretschmer
,
D.
,
M.
Rautenberg
,
D.
Linke
,
A.
Peschel
.
2015
.
Peptide length and folding state govern the capacity of staphylococcal β-type phenol-soluble modulins to activate human formyl-peptide receptors 1 or 2.
J. Leukoc. Biol.
97
:
689
697
.
18
Anderson
,
S.
,
A. T.
Bankier
,
B. G.
Barrell
,
M. H.
de Bruijn
,
A. R.
Coulson
,
J.
Drouin
,
I. C.
Eperon
,
D. P.
Nierlich
,
B. A.
Roe
,
F.
Sanger
, et al
.
1981
.
Sequence and organization of the human mitochondrial genome.
Nature
290
:
457
465
.
19
Martin
,
W. F.
,
S.
Garg
,
V.
Zimorski
.
2015
.
Endosymbiotic theories for eukaryote origin.
Philos. Trans. R. Soc. Lond. B Biol. Sci.
370
:
20140330
.
20
Taanman
,
J. W.
1999
.
The mitochondrial genome: structure, transcription, translation and replication.
Biochim. Biophys. Acta
1410
:
103
123
.
21
Marutani
,
T.
,
T.
Hattori
,
K.
Tsutsumi
,
Y.
Koike
,
A.
Harada
,
K.
Noguchi
,
Y.
Kiso
,
H.
Mukai
.
2016
.
Mitochondrial protein-derived cryptides: are endogenous N-formylated peptides including mitocryptide-2 components of mitochondrial damage-associated molecular patterns?
Biopolymers
106
:
580
587
.
22
McDonald
,
B.
,
K.
Pittman
,
G. B.
Menezes
,
S. A.
Hirota
,
I.
Slaba
,
C. C.
Waterhouse
,
P. L.
Beck
,
D. A.
Muruve
,
P.
Kubes
.
2010
.
Intravascular danger signals guide neutrophils to sites of sterile inflammation.
Science
330
:
362
366
.
23
Monlun
,
M.
,
C.
Hyernard
,
P.
Blanco
,
L.
Lartigue
,
B.
Faustin
.
2017
.
Mitochondria as molecular platforms integrating multiple innate immune signalings.
J. Mol. Biol.
429
:
1
13
.
24
Jin
,
H. S.
,
H. W.
Suh
,
S. J.
Kim
,
E. K.
Jo
.
2017
.
Mitochondrial control of innate immunity and inflammation.
Immune Netw.
17
:
77
88
.
25
Zhang
,
Q.
,
M.
Raoof
,
Y.
Chen
,
Y.
Sumi
,
T.
Sursal
,
W.
Junger
,
K.
Brohi
,
K.
Itagaki
,
C. J.
Hauser
.
2010
.
Circulating mitochondrial DAMPs cause inflammatory responses to injury.
Nature
464
:
104
107
.
26
Kolaczkowska
,
E.
,
P.
Kubes
.
2013
.
Neutrophil recruitment and function in health and inflammation.
Nat. Rev. Immunol.
13
:
159
175
.
27
Pittman
,
K.
,
P.
Kubes
.
2013
.
Damage-associated molecular patterns control neutrophil recruitment.
J. Innate Immun.
5
:
315
323
.
28
Mukai
,
H.
,
T.
Seki
,
H.
Nakano
,
Y.
Hokari
,
T.
Takao
,
M.
Kawanami
,
H.
Tsukagoshi
,
H.
Kimura
,
Y.
Kiso
,
Y.
Shimonishi
, et al
.
2009
.
Mitocryptide-2: purification, identification, and characterization of a novel cryptide that activates neutrophils.
J. Immunol.
182
:
5072
5080
.
29
Seki
,
T.
,
A.
Fukamizu
,
Y.
Kiso
,
H.
Mukai
.
2011
.
Mitocryptide-2, a neutrophil-activating cryptide, is a specific endogenous agonist for formyl-peptide receptor-like 1.
Biochem. Biophys. Res. Commun.
404
:
482
487
.
30
Ueki
,
N.
,
K.
Someya
,
Y.
Matsuo
,
K.
Wakamatsu
,
H.
Mukai
.
2007
.
Cryptides: functional cryptic peptides hidden in protein structures.
Biopolymers
88
:
190
198
.
31
Forsman
,
H.
,
E.
Andréasson
,
J.
Karlsson
,
F.
Boulay
,
M. J.
Rabiet
,
C.
Dahlgren
.
2012
.
Structural characterization and inhibitory profile of formyl peptide receptor 2 selective peptides descending from a PIP2-binding domain of gelsolin.
J. Immunol.
189
:
629
637
.
32
Bøyum
,
A.
,
D.
Løvhaug
,
L.
Tresland
,
E. M.
Nordlie
.
1991
.
Separation of leucocytes: improved cell purity by fine adjustments of gradient medium density and osmolality.
Scand. J. Immunol.
34
:
697
712
.
33
Bylund
,
J.
,
H.
Björnsdottir
,
M.
Sundqvist
,
A.
Karlsson
,
C.
Dahlgren
.
2014
.
Measurement of respiratory burst products, released or retained, during activation of professional phagocytes.
Methods Mol. Biol.
1124
:
321
338
.
34
Dahlgren
,
C.
,
A.
Karlsson
,
J.
Bylund
.
2007
.
Measurement of respiratory burst products generated by professional phagocytes.
Methods Mol. Biol.
412
:
349
363
.
35
Gabl
,
M.
,
A.
Holdfeldt
,
M.
Sundqvist
,
J.
Lomei
,
C.
Dahlgren
,
H.
Forsman
.
2017
.
FPR2 signaling without β-arrestin recruitment alters the functional repertoire of neutrophils.
Biochem. Pharmacol.
145
:
114
122
.
36
Stenfeldt
,
A. L.
,
J.
Karlsson
,
C.
Wennerås
,
J.
Bylund
,
H.
Fu
,
C.
Dahlgren
.
2007
.
Cyclosporin H, Boc-MLF and Boc-FLFLF are antagonists that preferentially inhibit activity triggered through the formyl peptide receptor.
Inflammation
30
:
224
229
.
37
Christophe
,
T.
,
A.
Karlsson
,
C.
Dugave
,
M. J.
Rabiet
,
F.
Boulay
,
C.
Dahlgren
.
2001
.
The synthetic peptide Trp-Lys-Tyr-Met-Val-Met-NH2 specifically activates neutrophils through FPRL1/lipoxin A4 receptors and is an agonist for the orphan monocyte-expressed chemoattractant receptor FPRL2.
J. Biol. Chem.
276
:
21585
21593
.
38
Heit
,
B.
,
S.
Tavener
,
E.
Raharjo
,
P.
Kubes
.
2002
.
An intracellular signaling hierarchy determines direction of migration in opposing chemotactic gradients.
J. Cell Biol.
159
:
91
102
.
39
Fu
,
H.
,
J.
Bylund
,
A.
Karlsson
,
S.
Pellmé
,
C.
Dahlgren
.
2004
.
The mechanism for activation of the neutrophil NADPH-oxidase by the peptides formyl-Met-Leu-Phe and Trp-Lys-Tyr-Met-Val-Met differs from that for interleukin-8.
Immunology
112
:
201
210
.
40
Önnheim
,
K.
,
K.
Christenson
,
M.
Gabl
,
J. C.
Burbiel
,
C. E.
Müller
,
T. I.
Oprea
,
J.
Bylund
,
C.
Dahlgren
,
H.
Forsman
.
2014
.
A novel receptor cross-talk between the ATP receptor P2Y2 and formyl peptide receptors reactivates desensitized neutrophils to produce superoxide.
Exp. Cell Res.
323
:
209
217
.
41
Forsman
,
H.
,
K.
Önnheim
,
E.
Andréasson
,
K.
Christenson
,
A.
Karlsson
,
J.
Bylund
,
C.
Dahlgren
.
2013
.
Reactivation of desensitized formyl peptide receptors by platelet activating factor: a novel receptor cross talk mechanism regulating neutrophil superoxide anion production.
PLoS One
8
:
e60169
.
42
Clemmensen
,
S. N.
,
L.
Udby
,
N.
Borregaard
.
2014
.
Subcellular fractionation of human neutrophils and analysis of subcellular markers.
Methods Mol. Biol.
1124
:
53
76
.
43
Lefkowitz
,
R. J.
,
E. J.
Whalen
.
2004
.
Beta-arrestins: traffic cops of cell signaling.
Curr. Opin. Cell Biol.
16
:
162
168
.
44
Winther
,
M.
,
A.
Holdfeldt
,
M.
Gabl
,
J. M.
Wang
,
H.
Forsman
,
C.
Dahlgren
.
2016
.
Formylated MHC class Ib binding peptides activate both human and mouse neutrophils primarily through formyl peptide receptor 1.
PLoS One
11
:
e0167529
.
45
Mukai
,
H.
,
Y.
Hokari
,
T.
Seki
,
T.
Takao
,
M.
Kubota
,
Y.
Matsuo
,
H.
Tsukagoshi
,
M.
Kato
,
H.
Kimura
,
Y.
Shimonishi
, et al
.
2008
.
Discovery of mitocryptide-1, a neutrophil-activating cryptide from healthy porcine heart.
J. Biol. Chem.
283
:
30596
30605
.
46
Hattori
,
T.
,
T.
Yamada
,
H.
Morikawa
,
T.
Marutani
,
K.
Tsutsumi
,
K.
Nishino
,
T.
Shimizu
,
Y.
Nishi
,
Y.
Kiso
,
H.
Mukai
.
2017
.
Generation of monoclonal antibodies against mitocryptide-2: toward a new strategy to investigate the biological roles of cryptides.
J. Pept. Sci.
23
:
610
617
.
47
Hattori
,
T.
,
K.
Nakashima
,
T.
Marutani
,
Y.
Kiso
,
Y.
Nishi
,
H.
Mukai
.
2015
.
Successful acquisition of a neutralizing monoclonal antibody against a novel neutrophil-activating peptide, mitocryptide-1.
Biochem. Biophys. Res. Commun.
463
:
54
59
.
48
Rabiet
,
M. J.
,
E.
Huet
,
F.
Boulay
.
2005
.
Human mitochondria-derived N-formylated peptides are novel agonists equally active on FPR and FPRL1, while Listeria monocytogenes-derived peptides preferentially activate FPR.
Eur. J. Immunol.
35
:
2486
2495
.
49
Pupo
,
A. S.
,
D. A.
Duarte
,
V.
Lima
,
L. B.
Teixeira
,
L. T.
Parreiras-E-Silva
,
C. M.
Costa-Neto
.
2016
.
Recent updates on GPCR biased agonism.
Pharmacol. Res.
112
:
49
57
.
50
Qin
,
C. X.
,
L. T.
May
,
R.
Li
,
N.
Cao
,
S.
Rosli
,
M.
Deo
,
A. E.
Alexander
,
D.
Horlock
,
J. E.
Bourke
,
Y. H.
Yang
, et al
.
2017
.
Small-molecule-biased formyl peptide receptor agonist compound 17b protects against myocardial ischaemia-reperfusion injury in mice.
Nat. Commun.
8
:
14232
.
51
Seifert
,
R.
2013
.
Functional selectivity of G-protein-coupled receptors: from recombinant systems to native human cells.
Biochem. Pharmacol.
86
:
853
861
.
52
Gripentrog
,
J. M.
,
H. M.
Miettinen
.
2005
.
Activation and nuclear translocation of ERK1/2 by the formyl peptide receptor is regulated by G protein and is not dependent on beta-arrestin translocation or receptor endocytosis.
Cell. Signal.
17
:
1300
1311
.
53
Gripentrog
,
J. M.
,
H. M.
Miettinen
.
2008
.
Formyl peptide receptor-mediated ERK1/2 activation occurs through G(i) and is not dependent on beta-arrestin1/2.
Cell. Signal.
20
:
424
431
.
54
Dorward
,
D. A.
,
C. D.
Lucas
,
M. K.
Doherty
,
G. B.
Chapman
,
E. J.
Scholefield
,
A.
Conway Morris
,
J. M.
Felton
,
T.
Kipari
,
D. C.
Humphries
,
C. T.
Robb
, et al
.
2017
.
Novel role for endogenous mitochondrial formylated peptide-driven formyl peptide receptor 1 signalling in acute respiratory distress syndrome.
Thorax
72
:
928
936
.
55
Raoof
,
M.
,
Q.
Zhang
,
K.
Itagaki
,
C. J.
Hauser
.
2010
.
Mitochondrial peptides are potent immune activators that activate human neutrophils via FPR-1.
J. Trauma
68
:
1328
1332
, discussion 1332–1334.
56
Hunninghake
,
G. W.
,
J. E.
Gadek
,
T. J.
Lawley
,
R. G.
Crystal
.
1981
.
Mechanisms of neutrophil accumulation in the lungs of patients with idiopathic pulmonary fibrosis.
J. Clin. Invest.
68
:
259
269
.
57
Kolahian
,
S.
,
I. E.
Fernandez
,
O.
Eickelberg
,
D.
Hartl
.
2016
.
Immune mechanisms in pulmonary fibrosis.
Am. J. Respir. Cell Mol. Biol.
55
:
309
322
.
58
Gadek
,
J. E.
1992
.
Adverse effects of neutrophils on the lung.
Am. J. Med.
92
(
6A
):
27S
31S
.
59
Stenfeldt
,
A. L.
,
C.
Wennerås
.
2004
.
Danger signals derived from stressed and necrotic epithelial cells activate human eosinophils.
Immunology
112
:
605
614
.

The authors have no financial conflicts of interest.