点击显示 收起
Johnson & Johnson Pharmaceutical Research and Development, LLC, San Diego, California
Abstract
Prokineticins 1 and 2 (PK1 and PK2) have been recently identified from humans and other mammals and play multiple functional roles. PK proteins are ligands for two G protein-coupled receptors, PK receptor 1 (PKR1) and PK receptor 2 (PKR2). Here, we report the molecular cloning and pharmacological characterization of an alternatively spliced product of the PK2 gene encoding 21 additional amino acids compared with PK2, designated PK2L (for PK2 long form). PK2L mRNA is broadly expressed, as is PK2. However, PK2L mRNA expression is lower in brain, undetectable in kidney, and much higher in lung and spleen than that of PK2. We expressed PK2L in mammalian cells and characterized the resulting peptide in comparison with PK1 and PK2. Biochemical characterization indicates that secreted PK2L protein is processed into a smaller peptide by proteolytic cleavage. We designate this smaller form of peptide as PK2. Coexpression of furin with PK2L significantly increased the PK2 processing efficiency. Functional studies showed that PK1, PK2, and PK2 stimulate intracellular Ca2+ responses in PKR1-expressing cells with similar potencies. However, the PK2 stimulus of Ca2+ responses in PKR2-expressing cells is at least 10-fold less potent than that of PK1 or PK2. Differences in receptor selectivity combined with differential tissue expression patterns suggest PK2 and PK2 might have different functions in vivo. PKRs have been reported to couple to Gq and Gi proteins. In this report, we show that PKs not only stimulate Ca2+ mobilization but also induce cAMP accumulation in PKR-expressing cells.
Recently, two cysteine-rich peptides, prokineticin 1 (PK1) and prokineticin 2 (PK2), have been identified and shown to stimulate gastrointestinal smooth muscle contractions (Li et al., 2001). PK1, also known as endocrine gland vascular endothelial growth factor, stimulates endocrine glands cell proliferation/migration and promotes angiogenesis in the mouse ovary (LeCouter et al., 2001). PK2, or mammalian Bv8, is believed to affect behavioral circadian rhythms in the suprachiasmatic nucleus and to promote angiogenesis in the testis (Cheng et al., 2002; LeCouter et al., 2003). PK1 and PK2 are highly homologous to each other as well as to a mamba intestinal protein (Schweitz et al., 1990, 1999) and a frog skin secreted protein, Bv8. Bv8 is a potent stimulator of gastrointestinal smooth muscle contractions (Mollay et al., 1999) and stimulates the sensitization of peripheral nociceptors (Negri et al., 2002). PKs bind and activate two closely related G protein-coupled receptors, prokineticin receptor 1 (PKR1) and 2 (PKR2), which are 87% identical by sequence (Lin et al., 2002a; Masuda et al., 2002; Soga et al., 2002). PKs stimulate Ca2+ mobilization in PKR-expressing cells, presumably through a receptor/Gq protein interaction (Lin et al., 2002a; Masuda et al., 2002; Soga et al., 2002). Pertussis toxin (PTX) inhibits PK1-induced mitogen-activated protein kinase signaling (Lin et al., 2002b), suggesting that PKRs may also couple to Gi proteins.
The nonselectivity of PK1 and PK2 versus their receptors (PKR1 and PKR2) in vitro raises the question of which ligand(s) activate(s) which receptor(s) in vivo. To investigate the ligand/receptor relationship of PK1, PK2, and their receptors, we systematically analyzed PK1, PK2, PKR1, and PKR2 mRNA expression in different human tissues and found two splice forms of PK2 mRNA. We isolated the cDNA for the alternatively spliced PK2 mRNA, designated PK2L, which encodes 21 additional amino acids compared with PK2. In this article, we report that the expression of PK2L results in the production of a short form of the peptide, which we refer to as PK2. Functional characterization of PK2 in comparison with PK1 and PK2 indicates that PK2 displays strong receptor selectivity for PKR1 over PKR2. In addition, signal transduction studies showed that PKs induce cAMP accumulation in PKR-expressing cells, indicating that PKRs are also coupled to Gs proteins.
Materials and Methods
cDNA Cloning of PKR1 and PKR2. The cDNA coding regions for both PKR1 and PKR2 were amplified by polymerase chain reaction (PCR) from human fetal brain cDNA (BD Biosciences Clontech, Palo Alto, CA). The primers used for PKR1 were P1, 5'-ACG TGA ATT CGC CAC CAT GGA GAC CAC CAT GGG GTT CAT G-3' and P2, 5'-ACG TAG CGG CCG CTT ATT TTA GTC TGA TGC AGT CCA CCT C-3'. The primers used for PKR2 were P3, 5'-ACG CGA ATT CGC CAC CAT GGC AGC CCA GAA TGG AAA CAC-3' and P4, 5'-ACG CAT GCG GCC GCG TCA CTT CAG CCT GAT ACA GTC CAC-3'. The PCR conditions were 94°C for 40 s, 65°C for 40 s, and 72°C for 3 min (40 cycles). Platinum TaqDNA polymerase (Invitrogen, Carlsbad, CA) was used for all PCR reactions. The PCR products were cloned into pCIneo (Promega, Madison, WI) vector and the insert regions were sequenced using an automated DNA sequencer (Applied Biosystems, Foster City, CA).
Expression and Purification of Prokineticins. Human PK1 mature peptide coding region was PCR amplified from human fetal brain cDNA (BD Biosciences Clontech) using two primers: P5, 5'-TCA TCA CGA ATT CGA TGA CGA CGA TAA GGC TGT GAT CAC AGG GGC CTG TGA GCG GGA TG-3' and P6, 5'-ACG ATA GGA TCC CTA AAA ATT GAT GTT CTT CAA GTC CAT G-3'. Human PK2 mature peptide and PK2L propeptide coding regions were PCR amplified from human fetal brain cDNA (BD Biosciences Clontech) using two primers: P7, 5'-CAT CAC GAA TTC GAT GAC GAC GAT AAG GCC GTG ATC ACC GGG GCT TGT GAC AAG-3' and P8, 5'-ACG ATA GGA TCC TTA CTT TTG GGC TAA ACA AAT AAA TCG-3'. The PCR conditions were 94°C for 40 s, 65°C for 40 s, and 72°C for 1 min (40 cycles). The PCR products for PK1, PK2, and PK2L were cloned into a modified pCMV-sport1 (Invitrogen, Carlsbad, CA) expression vector, which encodes an peptide signal sequence followed by a FLAG tag (Liu et al., 2003). The PK cDNAs were cloned in-frame after the FLAG coding sequence, and the insert regions were sequenced to confirm the identities. The resulting expression vectors encode fusion proteins with a mammalian secreted protein signal peptide followed by a FLAG peptide, an enterokinase cleavage site, and the relevant PKs without their natural signal peptide sequences. The PK1-, PK2-, and PK2L-expressing plasmids were transfected into COS-7 cells using LipofectAMINE (Invitrogen). Three days after transfection, the cell culture supernatants were collected and run through anti-FLAG M2 agarose (Sigma-Aldrich, St. Louis, MO) affinity columns, respectively. The columns were washed with phosphate-buffered saline (PBS) and eluted with 0.1 mM glycine-HCl, pH 3.0. The eluted protein fractions were immediately neutralized with 1 M Tris-HCl, pH 8.0, and cleaved with enterokinase (Novagen, Madison, WI). The cleaved proteins, which are free of the FLAG tag, were then further purified by reverse phase HPLC using a C4 column (Vydac, Hisperia, CA).
Western Blot. The recombinant PK protein expression was monitored by Western blot. Briefly, in a 1.5-ml tube, 20 e of anti-FLAG M2 agarose beads slurry were added to 1 ml of cell culture media from COS-7 cells expressing either PK1, PK2, or PK2L; coexpressing PK2L and furin; or to medium from control COS-7 cells. At the same time, corresponding cell samples were lysed with lysis buffer (100 mM Tris-HCl, pH 8.0, 150 mM NaCl, 1% Triton X-100, and 1% protease inhibitor cocktail; Sigma-Aldrich) and mixed with the anti-FLAG beads. The tubes were incubated at 4°C on a rocking platform overnight. The beads were centrifuged and washed twice with ice-cold Tris-buffered saline/Tween 20 (50 mM Tris-HCl, pH 7.5, 150 mM NaCl, and 0.05% Tween 20). The immunoprecipitated proteins were run onto a 4 to 20% SDS-polyacrylamide gel under reducing conditions and transferred onto a polyvinylidene difluoride membrane (Invitrogen). The membrane was blotted first with anti-FLAG M2 antibody (Sigma-Aldrich) and then with goat anti-mouse IgG (horseradish peroxidase-conjugated; Sigma-Aldrich). The Western blot membrane was then developed with an Amersham enhanced chemiluminescence kit and imaged by an X-ray film.
Expression, Purification, and Iodination of C-Terminal FLAG-Tagged PK2. C-terminal FLAG-tagged PK2 (PK2-f) was constructed as described previously (Soga et al., 2002). The primers P9, 5'-ATC GAG AAT TCG CCA CCA TGA GGA GCC TGT GCT GCG CCC-3' and P10, 5'-ACC TGA GGA TCC CTA CTT ATC GTC GTC ATC CTT ATA ATC CTT TTG GGC TAA ACA-3' were used to amplified human whole brain cDNA (BD Biosciences Clontech). The PCR-amplified PK2-f was cloned into a mammalian expression vector pCMV-sport1 (Invitrogen). The resulting clones were sequenced to confirm the identities and transfected into COS-7 cells using LipofectAMINE (Invitrogen). Three days after transfection, the cell culture supernatant was collected and run through an anti-FLAG M2 agarose (Sigma-Aldrich) affinity column. The column was washed with PBS and eluted with 0.1 mM glycine HCl, pH 3.0. The eluted protein fraction was immediately neutralized with 1 M Tris-HCl, pH 8.0, and then further purified by reverse phase HPLC using a C4 column (Vydac). The purified recombinant PK2-f protein was iodinated using Iodogen reagent (Pierce Chemical, Rockford, IL) and 125I-NaI (PerkinElmer Life and Analytical Sciences, Boston, MA) as described by Pierce Chemical. The iodinated PK2-f was purified by a G-50 (Amersham Biosciences Inc., Piscataway, NJ) gel filtration column.
Radioligand Binding Assays. PKR1 and PKR2 in the expression vector pCIneo (Promega) were transfected into COS-7 cells using LipofectAMINE (Invitrogen). Two days after transfection, cells were detached from the culture dishes with 10 mM EDTA in PBS, washed with Dulbecco's modified Eagle's medium (DMEM) and seeded in 96-well opaque polylysine-coated plates (BD Biosciences, San Jose, CA) at a density of 50,000 cells per well. Two hours after the seeding, competition binding assays were carried out in the 96-well plates in presence of 100 pM 125I-labeled PK2-f and various concentrations of unlabeled PK1, PK2, or PK2 as competitors. The binding assays were performed in DMEM plus 50 mM HEPES, pH 7.2, and 1% bovine serum albumin in a final volume of 100 e. The binding assays were incubated at room temperature for 1 h. The binding buffer was aspirated, and the cells in 96-well plates were washed three times with ice-cold PBS. Microscint-40 (PerkinElmer Life and Analytical Sciences) was added to each well (50 e/well), and the plates were counted on a microscintillation counter (Topcount NTX; PerkinElmer Life and Analytical Sciences).
Intracellular Ca2+ Mobilization Assays. PKR1 or PKR2 expression construct, either alone or in cotransfection with a chimeric G protein, Gqi5 (Conklin et al., 1993), were transfected into HEK293 cells using LipofectAMINE (Invitrogen). Two days after transfection, cells were detached using PBS containing 10 mM EDTA and seeded in poly-D-lysine-coated 96-well black wall tissue culture plates (BD Biosciences). Ligand stimulated Ca2+ mobilization was assayed using Fluo-3 Ca2+ dye (TEF Labs, Austin, TX) in a fluorometric imaging plate reader (Molecular Devices, Sunnyvale, CA) as described previously (Liu et al., 2001b).
PK Stimulation of cAMP Accumulation in PKR-Expressing Cells. PK stimulated cAMP accumulation assays were performed using SK-N-MC/-Gal cells (Liu et al., 2001a) stably expressing PKR1 or PKR2. SK-N-MC/-Gal cells harbor a -galactosidase reporter gene under the control of the cAMP response element (CRE). The stable cell lines were created under selection with 400 mg/l G418 (Geneticin; Sigma-Aldrich) after the transfection of PKR1 or PKR2 expression vectors. Increase of the intracellular cAMP concentration is associated with the activation of the CRE promoter, which leads to higher -galactosidase expression whose activity is measured using chlorophenol red--D-galactopyranoside as the substrate. Cells were seeded in 96-well tissue culture plates, stimulated with different concentrations of PK1, PK2, or PK2. Intracelluar cAMP concentrations were indirectly measured by assaying the -galactosidase activities in the cells as described previously (Liu et al., 2001a).
In a different experiment, PKR1 or PKR2 in the expression vectors were cotransfected with a Gs protein expression plasmid into HEK293 cells (American Type Culture Collection, Manassas, VA) using LipofectAMINE (Invitrogen). Human Gs protein long form complete coding region (Bray et al., 1986) was cloned into a mammalian expression vector pcDNA 3.1 (Invitrogen). Two days after transfection, the cells were detached with 10 mM EDTA in PBS, resuspended in DMEM/Ham's F-12 media, and then plated on 96-well plates at a density of 50,000 cells per well. Two hours after the seeding, cells culture medium was replaced with DMEM/Ham's F-12 containing 2 mM 3-isobutyl-1-methylxanthine (Sigma-Aldrich) and incubated for 30 min. Different concentrations of PK1, PK2, or PK2 were added to the cells, which were incubated for an additional 30 min in a final volume of 200 e/well. The reactions were stopped and cAMP was extracted by adding 20 e of 0.5 N HCl to each well. Cell extracts were assayed for cAMP concentrations using the 125I-cAMP FlashPlate assay kit (PerkinElmer Life and Analytical Sciences) as described by the manufacturer.
RT-PCR Detection of PK2L mRNA Expression in Different Human Tissues. Eleven cDNA pools (BD Biosciences) from various human tissues were analyzed for expression of PK1, PK2, PK2L, PKR1, and PKR2 mRNA using PCR amplification method. The PCR primers used in the reactions were P7 and P8 as described above for PK2 and PK2L; P11 (5'-ACG TAA GAA TTC GCC ACC ATG AGA GGT GCC ACG CGA GTC TCA-3') and P12 (5'-ACG TAA GAA TTC CTA AAA ATT GAT GTT CTT CAA GTC CAT GGA-3') for PK1; P13 (5'-CAA CTT CAG CTA CAG CGA CTA TGA TAT GCC TTT GG-3') and P14 (5'-GAC GAG GAC CGT CTC GGT GGT GAA GTA GGC GGA AG-3') for PKR1; and P15 (5'-TCT CCT TTA ACT TCA GTT ATG GTG ATT ATG ACC TC-3') and P16 (5'-CGA TGG GAT GGC AAT GAG AAT GGA CAC CAT CCA GA-3') for PKR2. All of the PCR reactions were performed using Platinum TaqDNA polymerase (Invitrogen) at the conditions of 94°C 40 s, 65°C 30 s, and 72°C 1 min for 40 cycles. The PCR products were run on 2% agarose gels, transferred onto nitrocellulose membranes, and hybridized with 32P-labeled oligo probes specific for PK1 (5'-ACC TGT CCT TGC TTG CCC AAC CTG CTG TGC TCC AGG TTC-3'), PK2 and PK2 (5'-TGG GCA AAC TGG GAG ACA GCT GCC ATC CAC TGA CTC GTA-3'), PKR1 (5'-CTG ATT GCC TTG GTG TGG ACG GTG TCC ATC CTG ATC GCC ATC C-3'), and PKR2 (5'-CGG ATG AAT TAT CAA ACG GCC TCC TTC CTG ATC GCC TTG G-3'), respectively. PCR detection of human -actin gene expression was used for all tissue as a control for the quality of the cDNAs. The primers for PCR detection of human -actin mRNA expression were 5'-GAG AAG AGC TAC GAG CTG CCT GAC GGC CAG GTC-3' and 5'-AAG GGT GTA ACG CAA CTA AGT CAT AGT CCG CCT A-3'.
Results and Discussion
Identification of the PK2L cDNA and Molecular Characterization of PK2L mRNA Tissue Expression Pattern. In the course of characterizing the PK2 mRNA tissue expression profile using RT-PCR, we identified a PCR fragment with slightly larger size than the predicted PK2 PCR product. Molecular cloning and DNA sequencing of that PCR product indicated that it has a 63-base pair (bp) insertion in the coding region of PK2, which resulted in a protein 21 residues longer (Fig. 1A) and is designated as PK2L. A GenBank search indicated that our sequence encodes a protein that is identical to a protein sequence in National Center for Biotechnology Information protein database (GenBank accession no. Q9HC23) published by Wechselberger et al. (1999). Using a similar method, we have also isolated a rat PK2L cDNA from a rat lung cDNA pool. The complete cDNA sequences for human and rat PK2L have been submitted to GenBank (accession nos. AY349131 and AY348322 ). Comparison of the deduced protein sequences indicate the rat and mouse (GenBank accession no. NP_056583 ) PK2L proteins are essentially identical and are highly related to human PK2L with about 90% amino acid identity (Fig. 1B). A DNA sequence comparison of human PK2, PK2L cDNA, and PK2 genomic DNA shows that the PK2 gene contains a putative exon region (63 bp) that is used by PK2L mRNA but not by PK2 mRNA (Fig. 1C), indicating that PK2L mRNA is an alternatively spliced isoform from the PK2 gene. The 63-bp insertion in PK2L mRNA leads to a 21-amino acid insertion between Lys47 and Val48 of the mature PK2 protein. A very similar splice variant was also found in the rat and has been reported in the mouse (Wechselberger et al., 1999), suggesting that the function of PK2L is conserved among species.
We have analyzed the mRNA expression profile of PK2L in parallel with that of PK1, PK2, PKR1, and PKR2 in 11 different human tissues using an RT-PCR method. As shown in Fig. 2, our results indicate that each of them has its own unique expression pattern. PK1 mRNA is found predominantly expressed in the placenta, whereas PK2 mRNA is found in all tissues. PK2L mRNA was detected in most tissues tested and expression was found to be highest in the lung and spleen, barely detected in the brain, and not detectable in the kidney where PK2 mRNA was detected. PKR1 mRNA was detected in the brain, lung, liver, spleen, and mammary gland. PKR2 mRNA had a very dominant expression in the brain with lower levels of expression in the spleen and mammary gland.
The mRNA expression analysis of PK2L indicates that the PK2L mRNA expression pattern is different from that of PK2, suggesting that the protein encoded by PK2L may function differently. The relatively abundant PK2L mRNA expression in the lung and spleen, where PKR1 mRNA is also expressed, suggests that PK2L may participate some immune functions through activation of PKR1. A chemoattractive effect of PKs has been shown for adrenal cortical capillary endothelial cells expressing PKR (LeCouter et al., 2003). It will be interesting to see whether immune cells express PKR and chemoattract in response to PKs.
Expression, Purification, and Biochemical Characterization of PKs. We have expressed PK1, PK2, and PK2L as secreted fusion proteins with an N-terminal FLAG tag in COS-7 cells. The secreted fusion proteins in cell culture supernatants were purified using anti-FLAG M2 affinity columns. The affinity-purified proteins were cleaved with enterokinase and further purified by reverse phase HPLC. The HPLC-purified proteins were greater than 98% pure. The sizes of PK1 (10 kDa) and PK2 (9 kDa) agreed with our prediction. However, the size (6eC7 kDa) of the purified protein from COS-7 cells expressing PK2L is much smaller than what was predicted (11.5 kDa) according to the PK2L cDNA. Since the PK2L coding region encodes 21 additional amino acids compared with PK2, we expected that PK2L should have a higher molecular weight than PK2. However, the purified protein from PK2L transfected cells culture medium has a molecular weight smaller than PK2, strongly suggesting that there is a proprotein cleavage process for PK2L protein. Therefore, we designate the full-length protein as PK2L and the cleaved protein as PK2. Western Blot analysis of the PK2L-expressing cell lysate and cell culture medium indicated that FLAG-PK2L was made in the cells as predicted (13.5 kDa), which is bigger than FLAG-PK1 (11.7 kDa) and FLAG-PK2 (10.8 kDa) (Fig. 3A). Although trace amounts of FLAG-PK2L are detected in the culture medium, the majority of FLAG-PK2L present in the conditioned medium is processed into a smaller form (8eC9 kDa), namely, FLAG-PK2 (Fig. 3A). Based on the size of the cleaved PK2, the protease cleavage site is predicted to be in the stretch of 21 additional amino acids present in PK2L. Protein sequence analysis of PK2L indicates that in the 21-amino acid insert region, there exist two putative furin cleavage sites (Arg-Arg-Lys-Arg60 and Arg-Ser-Lys-Arg65), which fit the Arg-X-Lys-Arg or Arg-X-Arg-Arg motif for furin cleavage sites (Steiner et al., 1992; Nakayama, 1997). Similar furin cleavage sites are also present in mouse and rat PK2L but are absent in PK1 and PK2 peptides. Since furin is expressed by many different cells, including COS-7 cells (Yanagita et al., 1993), PK2L is probably cleaved by endogenous furin before secretion from COS-7 cells. Coexpression of furin facilitates the cleavage process (Fig. 3A), which supports our theory. The doublet bands of the FLAG-PK2 that occurred in the Western blot suggested the differential processing of the PK2 occurred at the two different furin cleavage sites. The lower band of the mature PK2 was further purified by reverse phase HPLC and used for pharmacological characterization. Since the basic 21-amino acid insert in PK2L could be a substrate for many different serine proteases, including furin and other prohormone convertases, what proteases are involved in the maturation of PK2 under the natural conditions remains to be further studied.
PK2 Selectively Activates PKR1. PK1 and PK2 have been reported to stimulate Ca2+ mobilization in PKR-Expressing Cells (Lin et al., 2002a; Soga et al., 2002). We compared PK1, PK2, and PK2 stimulation of Ca2+ mobilization in PKR-expressing cells. Our results showed that PK1, PK2, and PK2 stimulate Ca2+ mobilization in PKR1-expressing HEK293 cells at nanomolar concentrations. Unlike PK1 and PK2, which have high potency for both receptors, PK2 only shows high potency for PKR1. The EC50 values for all Ca2+ assays are summarized in Table 1. The recombinant PK2L protein has been expressed in bacteria and was shown to be active for both PK receptors with receptor selectivity similar to that of PK2, although with significantly lower potency (with an EC50 value 500 nM for PKR1 and an EC50 value >1000 nM for PKR2) (Bullock et al., 2004). It seems that the cleavage and removal of the 21 amino acids (mostly basic) are important for full activity of PK2. We also tested N-terminally FLAG-tagged PKs as agonists for PKRs in the Ca2+ assay and found that they are inactive, suggesting that the exposure of the natural N termini of PKs is critical for the agonist function of those peptides. Modifications of the N terminus of PK1 described in a recent report (Bullock et al., 2004) abolishes the agonist activity of PK1, indicating the N terminus of PK plays an important role in the ligand receptor interaction. In contrast, tagging the C terminus of PK does not affect the ligand activity significantly. PK2-FLAG, reported in this study and by Soga et al. (2002), binds and activates both PKR1 and PKR2 at high affinity.
Values are representative of triplicate experiments.
The expression and functional characterization of PK2 strongly suggest that PKs have two domains. The N-terminal domain alone, amino acids 1eC47 of PK1 or PK2 mature peptides, already posses both receptor binding and activation abilities. The function of the C-terminal domain (amino acids 48eC86 of PK1 or 48eC81 of PK2) remains to be studied, but we speculate it may have functions such as maintaining structural stability or interacting with the extracellular matrix to form ligand gradients for chemotaxis, which has been reported for adrenal cortical capillary endothelial cells (LeCouter et al., 2003).
Natural disulfide bond assignments are important for organic peptide synthesis and future mutagenesis of peptides for detailed study of the structure-activity relationships of peptides. PKs posses 10 Cys residues, which presumably form five pairs of disulfide bonds. It is very difficult to assign the correct disulfide bridges between the 10 Cys residues. The results that we demonstrated by expression of PK2, which dissected the PK peptides into two physical domains, indicate that the six Cys residues in the N-terminal domains (amino acids 1eC47) form three pair of disulfide bonds, whereas the four Cys residues in the C-terminal domains (amino acids 48eC86 for PK1 and 48eC81 for PK2) form two pairs of disulfide bonds. The results presented in this report will help detailed structural study of PK in the future.
Prokineticins Bind PKR1 and PKR2 with Different Affinities. To further characterize PK2 pharmacologically, we investigated the binding properties of PK2 to PKR1 or PKR2 in comparison with that of PK1 and PK2. We attempted to use human PK1 labeled with 125I at Tyr75 as the radioligand in binding assays. However, we obtained very little specific binding using either PKR1- or PKR2-expressing cells. Since PK2 does not have a Tyr, we expressed the C-terminal FLAG-tagged PK2, which has been reported to bind prokineticin receptors with high affinity (Soga et al., 2002). PK2-FLAG has a Tyr in the FLAG-tag and can be labeled with 125I. Our results showed that 125I-PK2-FLAG binds PKR1 and PKR2 with high affinities, producing an average signal-to-noise ratio of 8:1 in the binding assays and was therefore used as the tracer in the binding assay. COS-7 cells transiently expressing PKR1 and PKR2 were used in competition binding assays. The results indicate that PK2 showed the highest affinity for both PKR1 and PKR2. PK1 demonstrated moderately high affinities for both PKRs. PK2 demonstrated a moderately high affinity for PKR1 but showed very low affinity for PKR2. The ligand rank order of potency for PKR1 is PK2 > PK2b PK1. The ligand rank order of potency for PKR2 is PK2 > PK1 >> PK2. The IC50 values of PK1, PK2, and PK2 for PKR1 and PKR2 are listed in Table 2.
IC50 values are representative of triplicate experiments in radioligand competition binding assays.
In the Ca2+ assay, PK1 demonstrated similar potency to that of PK2 for both PKRs. The binding assay revealed that compared with PK2, PK1 has significantly lower affinity for both receptors. The lower affinity of PK1 may explain the reduced specific binding observed when using 125I-PK1 as the radioligand. Comparing the results from the Ca2+ assay and binding assay revealed significant differences between the EC50 values and IC50 values. The difference between the IC50 (from the binding assay) and EC50 (from the Ca2+ assay) values could be a result of the differences in the assay mechanisms. Whereas IC50 values from the binding assay reflect the affinity (or Kd) of ligand-receptor interactions that reach equilibrium, the EC50 values from the Ca2+ assay may only represent the association rate of the ligand-receptor interactions. In the Ca2+ assay, the assay readout is measured seconds after the addition of ligand, which is, in most cases, long before the equilibrium is reached. Ligands with faster association rates (not necessarily higher Kd or affinity value) therefore tend to show higher potency in the Ca2+ assays.
Prokineticin Receptors Are Coupled to Multiple Signal Transduction Pathways. It has been reported that PTX inhibits PK-stimulated MAP kinase signaling (Lin et al., 2002b), suggesting that PKR activates MAP kinase through activation of Gi-related proteins. A phenomenon we observed in the Ca2+ mobilization assays is that the maximum ligand stimulated Ca2+ mobilization in PKR2-expressing cells was consistently significantly lower than that in PKR1-expressing cells. Our observation is consistent with what has been reported previously (Lin et al., 2002a). However, when PKR2 was coexpressed with a chimeric G protein (Gqi5), which shifts receptor/Gi coupling to Ca2+ mobilization signaling (Conklin et al., 1993), the maximum ligand stimulated Ca2+ mobilization in PKR2-expressing cells is dramatically increased to approximately the same level of that from PKR1-expressing cells. This suggests that PKR2 may also be coupled with Gi-related G proteins, in agreement with the previous report by Lin et al. (2002b). The Ca2+ mobilization in PKR1-expressing cells is not significantly affected by co-expression of Gqi5.
To further investigate the signal transduction pathways used by PKR1 and PKR2, we examined the effects of the PKs on the stimulation of cAMP accumulation in PKR1- and PKR2-expressing cells. We established PKR1 and PKR2 cell lines in SK-N-MC cells harboring a -galactosidase gene under the control of a CRE promoter. In the host cells, increased cAMP concentration activated the CRE promoter, which led to increased -galactosidase expression whose enzyme activity was measured using chlorophenol red--D-galactopyranoside as the substrate. Our results indicate that PK1, PK2, and PK2 stimulated -galactosidase activity in PKR-expressing cells in a dose-dependent manner. Without PKR expression, SK-N-MC cells showed no response to PKs. The EC50 values for PKs to stimulate -galactosidase activity in PKR-expressing cells are shown in Table 3.
EC50 values are representative of triplicate experiments.
To confirm our observation, we measured the cAMP accumulation in HEK293 cells transiently expressing PKR1 or PKR2. Our results indicate that PK stimulates cAMP accumulation in PKR-expressing cells specifically. HEK293 cells without PKR expression did not respond to PK stimulation. The ligand-stimulated cAMP accumulation is significantly increased if the Gs protein is coexpressed with PKRs (Fig. 4). In the cAMP accumulation assays using HEK293 cells expressing Gs and PKR1 or PKR2, dose-response curves have been generated and the ligand rank order of potency for PK1, PK2, and PK2 to either PKR1- or PKR2-expressing HEK293 cells are similar to those from SK-N-MC cells expressing PKRs. The EC50 values for PKs in stimulation of cAMP accumulation in PKR1- or PKR2-expressing HEK293 cells are also shown in Table 3.
Some G protein-coupled receptors have been shown to interact with different G proteins, including Gq, Gi, and Gs proteins (Chabre et al., 1994; Liu et al., 2002). PKs stimulate Ca2+ mobilization in PKR-expressing cells (Lin et al., 2002a; Soga et al., 2002), suggesting that PKRs are coupled with Gq proteins. The fact that PK-induced activation of mitogen-activated protein kinase is PTX-sensitive (Lin et al., 2002b) suggests that PKR may be also coupled with Gi proteins. This hypothesis is supported by our results, which showed that the coexpression of Gqi5 with PKR2 increases PK stimulated Ca2+ response in PKR2-expressing cells. Our results showing PK stimulation of cAMP accumulation in PKR-expressing cells indicate that PKRs are capable of coupling to Gs proteins. These different lines of evidence strongly suggest that PKRs can couple to different G proteins.
Conclusions
We have identified and characterized PK2 as a selective ligand for PKR1. Our results, in addition to adding new knowledge to the PK/PKR ligand/receptor system, provided a potential tool for in vivo functional study of PKR1. Our results also indicate that PK peptides possess two domains with the receptor binding and activation domain located at the N terminus. The results from the recombinant expression of PK2 in mammalian cells also physically dissected PK peptides in two segments, which provide useful information for the future study of the structure-activity relationships of PK peptides. Functional studies indicate that, in addition to Ca2+ mobilization, PKs also stimulate cAMP accumulation in PKR-expressing cells. Gqi5 enhances Ca2+ response in PKR2-expressing cells, suggesting possible PKR/Gi coupling. Our results in combination with previous reports (Lin et al., 2002a,b; Masuda et al., 2002; Soga et al., 2002) strongly suggest that PKRs are capable of coupling to multiple G proteins. PKs are multifunctional peptides. Different natural cells expressing PKRs may have different G protein expression patterns and hence respond to PK differently, thus allowing those cells to perform different physiological functions in response to the same ligand stimulation.
doi:10.1124/mol.105.011619.
References
Bray P, Carter A, Simons C, Guo V, Puckett C, Kamholz J, Spiegel A, and Nirenberg M (1986) Human cDNA clones for four species of G alpha s signal transduction protein. Proc Natl Acad Sci USA 83: 8893eC8897.
Bullock CM, Li JD, and Zhou QY (2004) Structural determinants required for the bioactivities of prokineticins and identification of prokineticin receptor antagonists. Mol Pharmacol 65: 582eC588.
Chabre O, Conklin BR, Brandon S, Bourne HR, and Limbird LE (1994) Coupling of the 2A-adrenergic receptor to multiple G-proteins. A simple approach for estimating receptor-G-protein coupling efficiency in a transient expression system. J Biol Chem 269: 5730eC5734.
Cheng MY, Bullock CM, Li C, Bermak JC, Belluzzi J, Weaver DR, Leslie FM, and Zhou QY (2002) Prokineticin 2 transmits the behavioural circadian rhythm of the suprachiasmatic nucleus. Nature (Lond) 417: 405eC410.
Conklin BR, Farfel Z, Lustig KD, Julius D, and Bourne HR (1993) Substitution of three amino acids switches receptor specificity of Gq to that of Gi. Nature (Lond) 363: 274eC280.
LeCouter J, Kowalski J, Foster J, Hass P, Zhang Z, Dillard-Telm L, Frantz G, Rangell L, DeGuzman L, Keller GA, et al. (2001) Identification of an angiogenic mitogen selective for endocrine gland endothelium. Nature (Lond) 412: 877eC884.
LeCouter J, Lin R, Tejada M, Frantz G, Peale F, Hillan KJ, and Ferrara N (2003) The endocrine-gland-derived VEGF homologue Bv8 promotes angiogenesis in the testis: localization of Bv8 receptors to endothelial cells. Proc Natl Acad Sci USA 100: 2685eC2690.
Li M, Bullock M, Knauer J, Ehlert FJ, and Zhou QY (2001) Identification of two prokineticin cDNAs: recombinant proteins potently contract gastrointestinal smooth muscle. Mol Pharmacol 59: 692eC698.
Lin DC, Bullock CM, Ehlert FJ, Chen JL, Tian H, and Zhou QY (2002a) Identification and molecular characterization of two closely related G protein-coupled receptors activated by prokineticin/endocrine gland vascular endothelial growth factor. J Biol Chem 277: 19276eC19280.
Lin R, LeCouter J, Kowalski J, and Ferrara N (2002b) Characterization of endocrine gland-derived vascular endothelial growth factor signaling in adrenal cortex capillary endothelial cells. J Biol Chem 277: 8724eC8729.
Liu C, Eriste E, Sutton S, Chen J, Roland B, Kuei C, Farmer N, Jornvall H, Sillard R, and Lovenberg TW (2003) Identification of relaxin-3/INSL7 as an endogenous ligand for the orphan G-protein-coupled receptor GPCR135. J Biol Chem 278: 50754eC50764.
Liu C, Ma XJ, Wilson SJ, Hofstra CL, Blevitt J, Pyati J, Li X, Chai W, Carruthers N, and Lovenberg TW (2001a) Cloning and pharmacological characterization of a fourth histamine receptor (H4) expressed in bone marrow. Mol Pharmacol 59: 420eC426.
Liu C, Wilson SJ, Kuei C, and Lovengerg TW (2001b) Comparison of human, mouse, rat and guinea pig histamine H4 receptors reveals substantial pharmacological species variation. J Pharmacol Exp Ther 299: 121eC130.
Liu F, Usui I, Evans LG, Austin DA, Mellon PL, Olefsky JM, and Webster NJ (2002) Involvement of both Gq/11 and Gs proteins in gonadotropin-releasing hormone receptor-mediated signaling in LT2 cells. J Biol Chem 277: 32099eC32108.
Masuda Y, Takatsu Y, Terao Y, Kumano S, Ishibashi Y, Suenaga M, Abe M, Fukusumi S, Watanabe T, Shintani Y, et al. (2002) Isolation and identification of EG-VEGF/prokineticins as cognate ligands for two orphan G-protein-coupled receptors. Biochem Biophys Res Commun 293: 396eC402.
Mollay C, Wechselberger C, Mignogna G, Negri L, Melchiorri P, Barra D, and Kreil G (1999) Bv8, a small protein from frog skin and its homologue from snake venom induce hyperalgesia in rats. Eur J Pharmacol 374: 189eC196.
Nakayama K (1997) Furin: a mammalian subtilisin/Kex2p-like endoprotease involved in processing of a wide variety of precursor proteins. Biochem J 327: 625eC635.
Negri L, Lattanzi R, Giannini E, Metere A, Colucci M, Barra D, Kreil G, and Melchiorri P (2002) Nociceptive sensitization by the secretory protein Bv8. Br J Pharmacol 137: 1147eC1154.
Schweitz H, Bidard JN, and Lazdunski M (1990) Purification and pharmacological characterization of peptide toxin from the black mamba (Dendroaspis polylepis) venom. Toxicon 28: 847eC856.
Schweitz H, Pacaud P, Moinier D, Frelin C, and Lazdunski M (1999) MIT(1), a black mamba toxin with a new and highly potent activity on intestinal contraction. FEBS Lett 461: 183eC188.
Soga T, Matsumoto S, Oda T, Saito T, Hiyama H, Takasaki J, Kamohara M, Ohishi T, Matsushime H, and Furuichi K (2002) Molecular cloning and characterization of prokineticin receptors. Biochim Biophys Acta 1579: 173eC179.
Steiner DF, Smeekens SP, Ohagi S, and Chan SJ (1992) The new enzymology of precursor processing endoproteases. J Biol Chem 267: 23435eC23438.
Wechselberger C, Puglisi R, Engel E, Lepperdinger G, Boitani C, and Kreil G (1999) The mammalian homologues of frog Bv8 are mainly expressed in spermatocytes. FEBS Lett 462: 177eC181.
Yanagita M, Hoshino H, Nakayama K, and Takeuchi T (1993) Processing of mutated proinsulin with tetrabasic cleavage sites to mature insulin reflects the expression of furin in nonendocrine cell lines. Endocrinology 133: 639eC644.