Formyl Peptide Receptor
Formyl peptide receptor 1 (FPR1) was first discovered on human neutrophils through its ability to bind N-formylated peptides with high affinity (Schiffmann et al. 1975). Rabbit neutrophils exhibit similar binding properties. The 350-amino acid human FPR1 receptor was the first cloned leukocyte chemoattractant receptor (Boulay et al. 1990). Genes with homologous sequence (FPR2 and FPR3) were identified through low-stringency hybridization using the FPR1 cDNA. The FPR2 cDNA encodes a 351-residue protein and shares approximately 69% sequence identity with FPR1. FPR2 is a low-affinity receptor for the prototypic formyl peptide, N-formyl-Met-Leu-Phe (fMLF). It binds lipoxin A4 and therefore is termed FPR2/ALX (Ye et al. 2009). FPR3 encodes a 7TM receptor of 352 amino acids that shares 56% sequence identity with FPR1 but does not bind fMLF. Although these three human members of the formyl peptide receptor family are relatively similar in terms of sequence identity and receptor structure, they are quite divergent with respect to agonist selectivity and cellular and tissue distribution, indicating a variety of biological functions (Migeotte et al. 2006; Ye et al. 2009).
Being the first characterized leukocyte chemoattractant receptor, FPR1 is extensively studied for its signaling mechanisms. FPR1 proves to be extremely important because of its ability to facilitate a number of downstream events in neutrophils leading to microbial killing. It has been known that different levels of FPR1 occupancy can lead to signaling pathways that vary between chemotaxis and cytotoxic functions such as degranulation and superoxide production (Korchak et al. 1984). However, the initial phase of receptor activation is similar for every aspect of FPR1 activation after fMLF binds to the receptor. After ligand binding, FPR1 undergoes a conformational change that enables the receptor to interact with the pertussis toxin (PTX) sensitive heterotrimeric G protein of the Gi family. Upon activation, there is an exchange of GDP for GTP in the Gαi subunit which leads to dissociation of the Gα and βγ subunit complex (Bokoch 1995). Following dissociation from the α subunit, the βγ subunit activates phospholipase C β2 (PLCβ2) and phosphoinositide 3-kinase γ (PI3Kγ). Activation of PLCβ2 hydrolyzes phophoinositiol-4,5-bisphosphate (PIP2) into diacylglycerol (DAG) and inositiol trisphosphate (IP3). Both hydrolyzation products are important second messengers, DAG activates protein kinase C (PKC), whereas IP3 binds the IP3 receptor on intracellular stores and mobilizes Ca2+. The activation of PI3Kγ converts the membrane phosphoinositol-4,5-bisphosphate (PIP2) into phosphoinositiol-3,4,5-trisphosphate (PIP3). Activation of PKC, as well as numerous calcium-sensitive protein kinases, leads to further downstream signaling, which has been shown to account for a variety of neutrophil functions to aid in microbial killing (Bokoch 1995). For the remainder of this entry, the distinct signaling pathways for FPR1-activated processes such as chemotaxis, superoxide production, and degranulation will be identified separately to avoid confusion, although significant cross talk exists between these pathways.
FPR1-Mediated Phagocyte Functions
Chemotaxis. The directed migration, or chemotaxis, of neutrophils by N-formyl peptides was one of the initially identified functions mediated by FPR1 (Prossnitz and Ye 1997). Human neutrophils can detect a chemotactic gradient of N-formyl peptides which produces a bell-shaped curve in dose–response experiments indicating that past optimal chemoattractant concentrations chemotaxis is reduced (Ye et al. 2009). Two mechanisms have been proposed to contribute to this phenomenon. First, saturation of these receptors with a high concentration of agonists eventually eliminates the chemoattractant gradient, thereby reducing chemotaxis. Secondly, FPR1 is desensitized at high concentrations of agonists, specifically fMLF, therefore reducing responsiveness to additional agonist stimulation (Prossnitz and Ye 1997). A number of events occur to aid in the chemotaxis activity such as polarization of the cell, integrin-mediated adhesion, and physical rearrangements of the cytoskeleton, requiring a variety of different signaling pathways. As previously mentioned, the initiation of chemotaxis through FPR1 activation begins with signaling from Gβγ. The βγ subunit activates PI3Kγ, which increases membrane concentration of PIP3 and helps distinguish leading edge of the chemotactic neutrophil from the trailing edge, termed uropod. The exposure of neutrophils to a chemoattractant gradient creates an intracellular gradient of signaling molecules such as PIP3. The regulation of PIP3 at the leading edge of neutrophils during chemotaxis is also facilitated by the phosphatidylinositol-3 phosphatase, PTEN, converting PIP3 back to PIP2. PTEN is localized to the side and the back of a migrating cell and therefore the cell maintains a higher concentration of PIP3 at leading edge, allowing it to migrate. The localization of PTEN to these areas of the neutrophil appears to involve the p38 MAPK, although the direct mechanism of localization is unclear (Heit et al. 2008). The mechanisms that regulate PIP3 localization at the leading edge are important for polarization and forward migration of the cell have been shown to incorporate a number of downstream signaling molecules and an important feedback loop involving some of these molecules. One of these downstream signaling molecules is Akt (PKB). In a number of studies involving the Pleckstrin homology domain of Akt, there was evidence that PIP3 generation by PI3K was crucial for the translocation of Akt to the leading edge of the plasma membrane. Further studies involving knockouts of Akt isoforms, indicating that Akt2 translocation was vital for polarization and chemotaxis of neutrophils after chemoattractant stimulation. Generation of PIP3 has been indicated to be involved in the positive feedback loop involving the Rho GTPases Rac and Cdc42. The activation of Rac has been linked to actin filament formation and subsequent membrane ruffling, therefore inducing the protrusion of the pseudopod and aiding in chemotaxis toward chemoattractants. Additionally, concentration of PIP2 to the uropod from the PTEN conversion of PIP3 to PIP2 mediates the activation of Rho, which regulates stress fiber and focal adhesions and therefore aiding in the uropod movement. In addition to the regulation of Rho by PIP3, there have been other signaling molecules implicated in the complex action of chemotaxis. Gα12 and Gα13 have been shown to be involved in a divergent pathway from the canonical Gαi pathway activated by FPR1, and have been directly linked to activating Rho, a Rho-dependent kinase (p160-ROCK), and myosin II and aiding in the “backness” of the neutrophil (Xu et al. 2003). In addition to the described mechanisms of chemotaxis in neutrophils after FPR1 activation, directed cell movement has also been linked to alternative pathways for its regulation. One of the pathways involves tyrosine kinases such as Src. It is thought that the first dissociation of the βγ subunit which activates PI3K can also increase the activity of Src-like kinases. Further characterization of this pathway was shown to involve the activation of the Shc-Grb2-Sos-Ras pathway, which leads to enhancement of the MAPK pathway involving Raf-MEK-ERK. Although the mechanism of this pathway is biochemically sound, the functional outcome of the activation of this pathway has not been specifically implicated in chemotaxis or other functions of neutrophils (Rabiet et al. 2007).
Degranulation. Degranulation is an important mechanism of host defense against invading microorganisms. However, proper regulation of granule release in neutrophils is vital to evade tissue damage and excess inflammation. Generally four types of granules are distinguished in neutrophils, including primary or azurophilic granules which contain neutrophil myeloperoxidase (MPO) and many hydrolytic enzymes such as β-glucuronidase (Borregaard 1997). There are three peroxidase-negative granules which are classified as secondary (specific) granules, tertiary (gelatinase) granules, and secretory vesicles. The release of these different granules follows a strict kinetic hierarchy in which secretory vesicles containing adhesion molecules and receptors such as the FPR1 receptor are released promptly after activation. Gelatinase granule release follows whereas specific and primary granule releases occur at a much slower rate. Neutrophils can discharge the contents of the granules both into a phagosome containing engulfed particles of bacteria or to the external environment when activated by soluble stimuli such as fMLF (Borregaard 1997). fMLF-induced activation of degranulation downstream of FPR requires 10–50 times higher concentration of the agonist but the cells were committed to full degranulation abilities after only 10 s of receptor-agonist interaction (Korchak et al. 1984). The FPR1-mediated signaling pathway leading to degranulation involves the secondary messengers PLCβ and PI3K. PLCβ leads to the activation of diacylglycerol and release of Ca2+ from intracellular stores, both of which have been shown to modulate the activation of PKC. The role of PKC in granule release has been shown to be both Ca2+ dependent in the case of primary granules, whereas secretory and specific granule contents are released in a Ca2+-independent manner. Pharmacological characterization of the signaling pathway leading to degranulation has implicated additional kinases, PI3K, p38 MAPK, and Src-family kinases such as Fgr and Hck (Ye et al. 2009). PI3K, which controls the phosphoinositide population of PIP2 and PIP3, is involved in granule release by regulating the actin cytoskeleton. Tyrosine kinases such as the above-mentioned Src-family kinases Fgr and Hck are required for adhesion-dependent degranulation and are speculated to initiate actin cytoskeletal rearrangements leading to granule mobilization to the plasma membrane. In addition, the small GTPases Rac and Rho and their respective GDP dissociation inhibitors (GDIs) and guanine nucleotide exchange factors (GEFs) are directly involved in the signaling pathway that leads to granule release, due to their effect on the actin cytoskeleton (Bokoch 1995). Proper docking of the vesicles in neutrophils is one of the final steps of granule release after fMLF activation and requires a set of exocytosis machinery proteins. Activation of the cytoskeletal contractile components is followed by phosphorylation of the soluble N-ethylmaleimide-sensitive factor-attachment protein (SNAP) and receptor proteins (SNAREs). Many of the classic, neural-type SNARE proteins such as syntaxin 1, VAMP-1, and SNAP-25 are not detected in granulocytes but homologs such as syntaxin 4 and VAMP-2 have been found in granulocytes. Other homologs have been found in leukocytes, one being SNAP-23 which is the homolog of SNAP-25. At resting state, the Munc-18 family of proteins that serve as binding partners for targeting membrane SNAREs (t-SNAREs) assumes a closed conformation that prevents the SNARE complex from forming. Phosphorylation of these proteins by kinases such as PKC can induce dissociation of Munc18 and syntaxins and therefore facilitates vesicular fusion with the plasma membrane. Specific evidence in a mast cell line, RBL, and human neutrophils indicates that Munc 18–3, SNAP-23, and syntaxins 2, 4, and 6 are phosporylated after fMLF stimulation. Further evidence identified the cGMP-dependent kinase PKG-1 and PI3K are both involved in phosphorylating Munc18–3, SNAP-23, and syntaxins 2 and 4 and the subsequent granule release. Further studies will be necessary to link direct phosphorylation by kinases downstream of fMLF signaling pathway and regulation of the SNARE proteins involved in vesicular fusion for granule release.
Superoxide Generation. Superoxide production is initiated in neutrophils when the oxidase accepts an electron from reduced nicotinamide adenine dinucleotide phosphate (NADPH) and donates it to molecular oxygen, generating O2− in the phagosome or to the extracellular environment in cases of soluble stimuli such as fMLF. The O2− that is generated can be converted to cytotoxic products such as hydrogen peroxide by superoxide dismutase. The hydrogen peroxide can then be converted to the most potent bactericidal product, hypochlorous acid (HOCl), by the granular enzyme MPO. fMLF activation of superoxide generation in neutrophils requires a number of different proteins, and regulation of these proteins, as well as the coupling of responses to granulation, produces the appropriate amount of microbicidal response without resulting in injury to surrounding tissue. fMLF activation of FPR1 has been shown to require higher concentrations of the agonist than necessary for functions such as chemotaxis and degranulation. The mechanism underlying different concentration requirement of fMLF in activating these functions have not been fully appreciated; however, it is plausible that it helps to prevent tissue injury from migrating neutrophils. It could implicate that additional signaling pathways are required due to different receptor occupancy. Unlike degranulation that showed commitment to full granule release after finite receptor occupancy, O2− generation was shown to require continuous occupation of the receptor to initiate and maintain full superoxide response and generation (Korchak et al. 1984; Boxer et al. 1979).
Activation of the signaling pathway to stimulate O2− production downstream of FPR1 involves a number of additional proteins that are not necessary for chemotaxis and degranulation. The production of O2− is dependent on the translocation and activation of membrane and cytosolic components forming a fully active complex. In the resting state, the membrane components gp91phox and p22phox create the stable membrane complex cytochrome b558. The cytosolic components of the NADPH oxidase, p47phox, p67phox, p40phox, and the small GTPase Rac are physically separate from the membrane components and require phosphorylation and translocation to the membrane to assemble the functioning complex. P47phox contains an N-terminal phox homology (PX) domain that targets p47phox to the membrane to interact with the membrane component p22phox and PIP2. Phosophorylation of p47phox is required before translocation to change the conformation and expose the PX domain for the membrane. Evidence has been shown for a number of kinases to phosphorylate p47phox. Several PKC isoforms, including α, β, δ, and ζ have been implicated in p47phox phosphorylation (Ye et al. 2009). With regard to fMLF-specific activation of p47phox and subsequent superoxide production, the PKCδ isoform has been implicated and shown to directly phosphorylate the autoinhibitory region of p47phox (Babior et al. 2002). Additional kinases, including AKT and the MAPKs, ERK and p38 have been shown to phosphorylate p47phox such as the SH3 domain and the C-terminal proline-rich domain. The p67phox protein is a larger protein that interacts with almost all proteins involved in the NADPH complex. Its N-terminal tetratricopeptide repeat domains (TPR) interact with activated Rac. Phosphorylation of p67phox in its activation domain enhances binding of the protein to the gp91phox membrane component of the oxidase. Additional interaction with both p47phox and p40phox proteins allows for full activation and regulation of the oxidase system. P40phox has been indicated as a regulatory protein in the NADPH oxidase complex, specifically through its association with 67phox in the resting state. Both basal phosphorylation and subsequent phosphorylation of p40phox by a PKC isoform after stimulation with fMLF have been suggested for its involvement in both regulation and activation. Rac is a cytosolic component of the NADPH oxidase complex essential for activation (Abo et al. 1991). Rac2 binds directly to p67phox at the membrane but does not bind p40phox or p47phox. The activation of Rac involves guanine nucleotide exchange factors that exchange the inactive GDP for GTP. A Rac GEF, P-Rex1, is activated by PIP3 and Gβγ downstream of FPR1 stimulation, and therefore generates active GTP bound Rac. Additional GEFs, such as Vav1, have also been shown to play a role in fMLF-induced activation of NADPH oxidase (Welch et al. 2002).
Regulation of FPR1
Activation of FPR1 is tightly regulated through mechanisms including phosphorylation, desensitization, and internalization, which are necessary for the regulation of the levels of G protein activation and subsequent expansion of the signal through secondary messengers. The first step in FPR1 desensitization is the rapid phosphorylation of the C-terminal tail which is both time- and agonist-concentration dependent. Phosphorylation of FPR1 is PKC independent unlike other GPCR chemoattractant receptors such as the C5a receptor. The C-terminal tail of FPR1 contains a total of 11 serine and threonine residues, which are arranged in two domains characteristic of G protein receptor kinase (GRK) phosphorylation sites. GRK2 has been shown as a primary kinase in phosphorylating FPR1, whereas other GRKSs have minimal (GRK3) or no effect (GRK5,6) on the receptor (Prossnitz 1997). The phosphorylation of the clusters of serine and threonine residues modulates the affinity of the receptor to both β-arrestins and agonists. The affinity of FPR1 for β-arrestins is controlled by the phosphorylation level on the C-terminal tail of the receptor, allowing β-arrestin to form a high affinity complex and sterically inhibit Gai protein coupling. Desensitization of FPR1 also uncouples it from G proteins independently of β-arrestin, indicating that phosphorylation of the receptor alone is sufficient in uncoupling the receptor from G protein–mediated signaling cascade. In addition to uncoupling receptors from their respective G proteins, β-arrestins have also been shown to co-localize with FPR1 during internalization of the receptor through an endocytic pathway. There are a number of reviews that address desensitization and internalization of GPCRs with details of the involvement of GRKs and β-arrestins (Ye et al. 2009; Prossnitz and Ye 1997; Rabiet et al. 2007; Migeotte et al. 2006).
Diversity of Formyl Peptide Receptor Family Ligand Binding and Downstream Effects
Although this review focuses on the downstream signaling and functional response of FPR1 activation after fMLF stimulation, it is important to note that different members of the family, such as FPR2, can mediate response through an entirely separate system or pathway. One notable difference is in the primary response, calcium mobilization, after receptor activation. The cyclic adenosine 5′-diphosphate ribose (cADPR) is a known regulator of calcium signaling. It is synthesized by CD38 ADP ribosyl cyclase and regulates intracellular calcium release from the ryanodine receptor–gated stores (Migeotte et al. 2006). Human FPR2 has been shown to be dependent on the presence of cyclic ADP-ribose for intracellular calcium release, whereas FPR1 calcium release is independent of this specific calcium metabolite. There is some evidence that FPR2 can couple to the Gq family of G proteins, which could lead to differential downstream effects for the calcium signaling. In addition to calcium mobilization, there is evidence for different endocytic machinery involved in FPR1 and FPR2 signaling. The classical GPCR internalization pathway involving β-arrestin, dynamin, and clathrin, is absent in FPR1 internalization. However, FPR2 undergoes clathrin-mediated and dynamin-dependent endocytosis and has also been shown to require at least one β-arrestin for internalization. Another interesting phenomenon is the diversity in ligand binding between FPR1 and FPR2. These two receptors share 68% sequence identity, yet they bind the same agonists, fMLF, with drastically different affinity (Rabiet et al. 2005). It is surprising that FPR2 can bind a wider range and more structural diverse group of agonists than FPR1, including the eicosanoid lipoxin A4. These agonists can induce similar signaling pathways as described above; however, many can also trigger divergent signaling pathways with different functional outcomes (Ye et al. 2009).
FPR1 is a “classic” chemoattractant receptor characterized with its ability to mediate strong neutrophil activation of a full set of bactericidal functions, and functionally couple to the Gαi family of G proteins. FPR1 is one of the first cloned and well-characterized leukocyte chemoattractant receptors, and is often used as a model for studies of G protein–mediated signaling pathways in leukocytes. Although a tremendous amount of work has been completed on the formyl peptide receptor family, many questions still remain. These include the diverse ligand-binding properties of the FPR family receptors, the ability to mediate both pro-inflammatory and anti-inflammatory activities, and the physiological functions of these receptors when bound to bacterial formyl peptides compared to endogenous ligands. Studies have begun in recent years for the characterization of mouse FPR family receptors, which may result in genetic models for a better understanding of the signaling properties of this family of G protein–coupled chemoattractant receptors.