Introduction

Top Abstract Introduction Materials and Methods Results Discussion References

Proteinase-activated receptor-2 (PAR₂), the second member of a growing family (PAR_1-4) of G protein-coupled receptors activated by proteinases (Rasmussen et al., 1991; Vu et al., 1991a; Nystedt et al., 1994; Ishihara et al., 1997; Kahn et al., 1998; Xu et al., 1998; Macfarlane et al., 2001; Hollenberg and Compton, 2002), is triggered by the proteolytic unmasking of an amino terminal receptor sequence (S³⁷LIGRLDTP... ) that acts as a "tethered ligand" (Vu et al., 1991a; Chen et al., 1994; Nystedt et al., 1994). As was discovered for the thrombin-activated receptor PAR₁ (Vu et al., 1991a), for PAR₂, short synthetic peptides based on the proteolytically revealed tethered ligand sequence can activate PAR₂, so as to mimic the action of trypsin in many tissues and cultured cells (Nystedt et al., 1994; Al-Ani et al., 1995; Hollenberg et al., 1997; Kawabata et al., 1999).

Given the target specificity of trypsin (Bergmann et al., 1939; Walsh and Neurath, 1964), when PAR₂ was discovered (Nystedt et al., 1994), it was hypothesized that although thrombin could not activate PAR₂, trypsin might do so at a postulated R³⁴/S³⁵ murine PAR₂ cleavage/activation site that is homologous with the one (R⁴¹/S⁴²) hydrolyzed by thrombin in human PAR₁ to reveal the tethered receptor-activating sequence (Vu et al., 1991a). Yet, in rat PAR₂, apart from the cleavage/activation site, R³⁶/S³⁷, there are six other theoretical trypsin cleavage sites in the extracellular amino terminus of the receptor either upstream (R², R²⁴, K³⁴) or downstream (R⁴¹, K⁵¹, K⁷²) of the R³⁶/S³⁷ cleavage/activation site that might be targeted by trypsin.

In previous work (Compton et al., 2001; Al-Ani et al., 2002a), we established that either trypsin or tryptase at room temperature can rapidly hydrolyze the synthetic rat PAR₂-derived peptide [G³⁰PNSKGRSLIGRLDTP⁴⁵... ] into several fragments, indicating cleavage at all potential target sites within the peptide (K³⁴, R³⁶, R⁴¹) to yield the receptor-activating sequence S³⁷LIGRLDTP, plus other predicted peptides that would not cause receptor activation. Furthermore, Loew et al. (2000) demonstrated that a large segment of the extracellular N-terminal sequence of human PAR₂ (R³¹-P⁷⁹), expressed in Escherichia coli, was rapidly (within 5 min) cleaved by trypsin (2.5 nM) at all potential tryptic cleavage sites (K³⁴, R³⁶, K⁵¹, K⁷²), except for the K⁴¹, equivalent to R⁴¹ in rat PAR₂. We were particularly interested to investigate in rat PAR₂ expressed in intact cells, the accessibility to trypsin of residues, R⁴¹, K⁵¹, K⁷², located C-terminal to the trypsin cleavage/activation site R³⁶. Tryptic hydrolysis at these receptor sites would "disarm" PAR₂, as trypsin does for PAR₁ (Kawabata et al., 1999).

To test the ability of trypsin, acting on receptor expressed in intact cells, to cleave at PAR₂ residues either N-terminal (i.e., R³⁴) or C-terminal (i.e., R⁴¹, K⁵¹, or K⁷²) of the tethered ligand activation site (R³⁶), we prepared receptors with mutations (Fig. 1A) that abolished the trypsin cleavage site at either R³⁶ or R⁴¹ (or both) [PAR₂R³⁶A, PAR₂S³⁷P, PAR₂R⁴¹A, PAR₂R³⁶AR⁴¹A, and PAR₂R³⁶ASⁱ⁴² (wherein a serine, Sⁱ⁴², is inserted between R⁴¹ and L⁴² of PAR₂R³⁶A)]. Thus, three receptor mutants, PAR₂R³⁶A, PAR₂S³⁷P, and PAR₂R³⁶AR⁴¹A, were resistant to trypsin hydrolysis at R³⁶ and to trypsin-mediated exposure of the tethered ligand sequence; and the fourth mutant, PAR₂R³⁶ASⁱ⁴², also resistant to trypsin cleavage at R³⁶, had inserted into it a new potential site of trypsin cleavage, G⁴⁰R/Sⁱ⁴²L⁴³, that mimicked the sequence of the wild-type PAR₂ cleavage/activation site (G³⁵R/SL³⁸). It was our hypothesis that although trypsin acting in vitro may cleave synthetic or recombinantly produced polypeptides that represent the extracellular N-terminal sequence of PAR₂, the same polypeptide sequences, when expressed at the cell surface as the fully glycosylated receptor, would not necessarily be accessible to enzymatic hydrolysis (Compton et al., 2001). To test this hypothesis, the wild-type and mutated receptors were expressed in Kirsten virus-transformed rat kidney (KNRK) cells (Al-Ani et al., 1999a) and were tested first for activation (elevated intracellular calcium) by trypsin and the selective PAR₂-activating peptide SLIGRL-NH₂ (SL-NH₂); and second, for the release by trypsin of two antigenic determinants, one (antibody SLAW-A) entirely amino terminal to the R³⁶/S³⁷ cleavage/activation sequence and another (antibody B5) spanning the receptor cleavage/activation site (G^30--P⁴⁵), amino terminal to the potential cleavage sites K⁵¹/G⁵² and K⁷²/L⁷³ in rat PAR₂ (Compton et al., 2001). Disappearance of the antigenic determinants detected by SLAW-A would indicate receptor cleavage either at or downstream of the R³⁶/S³⁷ cleavage/activation site (Al-Ani et al., 2002a; Compton et al., 2001), whereas disappearance of the antigenic determinants detected by B5 would indicate cleavage C-terminal of the tethered ligand site, e.g., at PAR₂ residues K⁵¹ and K⁷².

View larger version (31K): [in this window] [in a new window]   Fig. 1.   Amino acid sequences of rat PAR2 variants (A) and receptor epitopes (B). A, partial amino acid sequence of the amino terminus of PAR2wt is compared with the same region of five receptor variants (PAR2R36A, PAR2S37P, PAR2R41A, PAR2R36AR41A, and PAR2R36ASi42). Bold letters denote mutated residues and the underlined letter denotes an inserted residue. Potential trypsin cleavage sites are numbered and the R36/S37 tryptic receptor cleavage/activation site is denoted for PAR2wt in A by the symbol/. B, partial amino acid sequence of the amino terminus of rat PAR2. Bold letters denote the antigenic receptor polypeptide sequences.

	Materials and Methods

Top Abstract Introduction Materials and Methods Results Discussion References

Cloning and Expression of PAR₂. Rat PAR₂ was cloned from kidney cDNA as documented previously (Saifeddine et al., 1996; Al-Ani et al., 1999a,b) using the primer pairs forward primer, PAR₂-F (containing a Hind III site and Kozak sequence shown in bold), 5'-TCAAGCTTCCACCATGCGAAGTCTCAGCCTGGC-3' and reverse primer PAR₂-R (containing SmaI site shown in bold) 5'-CCCGGGCTCAGTAGGAGGTTTTAACAC-3'. Then the rat PAR₂ cDNA, for which sequence verification was done (Sanger et al., 1977; DNA services facility at the University of Calgary) was subcloned further into the pcDNA3 mammalian expression vector (Invitrogen, Carlsbad, CA), which was used to prepare all five receptor mutants shown in Fig. 1A. The receptor mutants described in Fig. 1A were prepared using the QuikChange site-directed mutagenesis kit (Stratagene, La Jolla, CA) according to the manufacturer's instructions. In PAR₂R³⁶A and PAR₂R⁴¹A, the trypsin cleavage/activation site (R³⁶), and R⁴¹ were changed to A, respectively; in PAR₂R³⁶AR⁴¹A, R³⁶ and R⁴¹ residues were both changed to A. In PAR₂S³⁷P, S³⁷ was changed to P, a trypsin-resistant residue. In PAR₂R³⁶ASⁱ⁴², an extra residue (S) was inserted at position 42, in the amino terminus of PAR₂R³⁶A construct. The wild-type PAR₂ and PAR₂ mutants in pcDNA3 were then transfected into KNRK cells (American Type Culture Collection, Manassas, VA), as described previously (Al-Ani et al., 1999a,b) to yield permanent cell lines for further study. Transfected cells (either vector alone or PAR₂-containing vectors) were subcloned in geneticin-containing medium (0.6 mg/ml), and PAR₂-expressing cells were isolated by fluorescence-activated cell sorting (FACS) with the use of the anti-receptor B5 antibody (Kong et al., 1997; Al-Ani et al., 1999b) generated against a peptide representing the cleavage/activation sequence of rat PAR₂: G³⁰PNSKGR/SLIGRLDTP⁴⁵-YGGC [/represents the tryptic cleavage/activation site of PAR₂ that yields the tethered ligand sequence (the nonreceptor sequence YGGC was added for potential Cys-linked conjugation and ¹²⁵I radiolabeling)]. In the cell lines so isolated, >95% of the populations (flow cytometry) were found to exhibit reactivity with the B5 antibody. In addition, in all mutants the B5 fluorescence intensity on a per cell basis was equivalent. In keeping with our previous work (Al-Ani et al., 1999b, 2002b), we only maintained and used permanent cell lines that expressed high levels of PAR₂ and exhibited equivalent average fluorescence yields on a per cell basis with the B5 antibody. Cells were routinely propagated in geneticin (0.6 mg/ml)-containing Dulbecco's modified Eagle's medium supplemented with 5% (v/v) fetal calf serum, using 80-cm² plastic T-flasks. Cells were subcultured by resuspension in calcium-free isotonic saline/EDTA solution, without the use of trypsin.

Measurement of Calcium Signaling Using Fluorescence Emission. Measurements of trypsin and peptide-stimulated fluorescence emission (reflecting an increase in intracellular calcium) were done with cells grown to about 85% confluence and disaggregated with calcium-free isotonic phosphate-buffered saline containing 0.2 mM EDTA. PAR₂-transfected KNRK cells were loaded with the intracellular calcium indicator Fluo-3 (Molecular Probes, Eugene, OR) at a final concentration of 22 µM (25 µg ml¹) of Fluo-3 acetoxymethyl ester, as described previously (Kao et al., 1989; Minta et al., 1989; Al-Ani et al., 1999b; Kawabata et al., 1999). Fluorescence measurements, reflecting elevations of intracellular calcium, were conducted at 24°C using a fluorescence spectrometer (PerkinElmer Instruments, Norwalk, CT) with an excitation wavelength of 480 nm and an emission recorded at 530 nm. The fluorescence signals caused by the addition of test agonists (trypsin or SL-NH₂, added to 2 ml of a cell suspension of about 3 × 10⁵ cells ml¹) were compared with the fluorescence peak height yielded by replicate cell suspensions treated with 2 µM ionophore A23187 (Sigma-Aldrich, St. Louis, MO). This concentration of A23187 was at the plateau of its concentration-response curve for a fluorescence response. Under these conditions, the calculated values for intracellular calcium were approximately 30 nM under basal conditions and about 340 nM, upon exposure to A23187 (Kao et al., 1989; Minta et al., 1989). Previous work (Kawabata et al., 1999; Compton et al., 2000) has shown that the fluorescence response of a cell preparation, as a percentage relative to the signal generated by 2 µM A23187, is a valid reference standard for the comparative determination of calcium signals for all PAR agonists. In addition, in previous work we have observed, as expected, that the presence of the extracellular PAR₂APs in the cell suspensions do not affect the Fluo-3 signal generated by intracellular calcium indicator, in response to other agonists such as lysophosphatidic acid (Kawabata et al., 1999). Under the assay conditions, the addition of proteinase inhibitors (e.g., amastatin) did not potentate or diminish the fluorescence response caused by the soluble peptide SL-NH₂. Thus, routinely, proteinase inhibitors were not added to the assay cuvettes. Measurements were done using three or more replicate cell suspensions derived from two or more independently grown groups of cells.

Monitoring Trypsin Removal of the PAR₂ N-Terminal Antigenic Determinants in Intact Cells by Flow Cytometric Analyses and Immunocytochemistry. The wild-type and PAR₂ variant cell lines were grown to about 85% confluence. These clones possess an N-terminal sequence that is proximal to the receptor's cleavage/activation sequence and that is therefore potentially released from the cell upon trypsin cleavage of PAR₂ at site R³⁶. We generated a rabbit antiserum (SLAW-A) targeted to two discontinuous antigenic determinants (⁵SLAWLLG¹¹-G³⁰PNSKGR³⁶-GGYGGC) (receptor antigenic sequences represented by bold print; GGYGGC added for radiolabeling and cysteine coupling) in the proteinase-released sequence. The polyclonal antiserum (SLAW-A) was raised in rabbits as described previously (Kong et al., 1997; Al-Ani et al., 1999b) for the B5 anti-PAR₂ polyclonal antibody used by others and by us (Kong et al., 1997; Al-Ani et al., 2002b). Disappearance of the signal detected by the SLAW-A antiserum (potentially up to and including residue R³⁶) indicates a loss of the antigenic determinants N-terminal to the cleavage/activation site R³⁶/S³⁷ (Compton et al., 2001; Al-Ani et al., 2002a). The B5 antiserum recognizes antigenic determinants that span the PAR₂ receptor cleavage/activation sequence (G³⁰PNSKGR/SLIGRLDTP⁴⁵) and can recognize both the cleaved/activated receptor as well as the uncleaved receptor, but would not recognize receptor sequences beyond ⁴⁵P (Compton et al., 2001). Neither the B5 nor the SLAW-A antibodies react with KNRK cells transfected with vector alone and the reactivity of both antibodies with PAR₂-expressing KNRK cells is abolished by preabsorption with the immunizing peptide (Al-Ani et al., 1999b; Compton et al., 2001 for B5 antibody and Compton et al., 2001; Al-Ani et al., 2002a for SLAW-A antibody).

Peptides and Other Reagents. The soluble selective PAR₂-activating peptide SLIGRL-NH₂ was synthesized by solid-phase methods at the peptide synthesis facility (University of Calgary). High-performance liquid chromatography analysis, mass spectral analysis, and quantitative amino acid analysis confirmed the composition and purity (>95%) of the peptide. Stock solutions, prepared in 25 mM HEPES buffer, pH 7.4, were standardized by quantitative amino acid analysis to verify peptide concentration. Porcine trypsin (14,900 units mg¹) was obtained from Sigma-Aldrich. A maximum specific activity of 20,000 units mg¹ was used to calculate the approximate molar concentration of trypsin in the incubation medium (1 unit ml¹  2 nM).

	Results

Top Abstract Introduction Materials and Methods Results Discussion References

Expression and Functional Analysis of PAR₂ Variants. Wild-type and mutated rat PAR₂ receptors were transfected into a KNRK cell line that lacks the expression of a functional PAR₂ (Böhm et al., 1996; Al-Ani et al., 1999b). FACS analysis using the B5 antibody revealed that all cell lines exhibited comparable average cell surface fluorescence and that in all cell lines, more than 95% of cells expressed receptor (data not shown). As illustrated in Fig. 1A, two potential trypsin cleavage residues, R³⁶ and R⁴¹, in the amino terminus were targeted by mutations (PAR₂R³⁶A, PAR₂R³⁶AR⁴¹A, and PAR₂S³⁷P) that would mitigate trypsin activation, because changing arginine³⁶ to alanine (Al-Ani et al., 2002b) or changing the P1' residue (serine³⁷) at the trypsin-reactive site (P1-P1') to proline (Nystedt et al., 1994) rendered the receptor cleavage/activation site resistant to trypsin. As illustrated in Fig. 2, B and C, the two receptor variants PAR₂R³⁶A and PAR₂S³⁷P failed to yield a calcium signal upon treatment with trypsin (40 nM) at twice the concentration (20 nM) required to activate PAR₂wt maximally. Yet, PAR₂R³⁶A and PAR₂S³⁷P were equally responsive to 50 µM SL-NH₂ (Fig. 2, B and C, middle panel). The R⁴¹ cleavage site, absent in the PAR₂R⁴¹A mutant, still allowed for receptor activation and cleavage at R³⁶, as indicated by the strong calcium signal shown in the left-hand tracing of Fig. 2D. In contrast, the construct PAR₂R³⁶AR⁴¹A, which has double mutations at R³⁶ and R⁴¹, was resistant to trypsin activation (Fig. 2E), but showed a calcium response to 50 µM SL-NH₂ equivalent to that of the other receptor mutants, including PAR₂wt (Fig. 2E, middle panel). This receptor mutant, although not activated by trypsin, allowed for an evaluation of trypsin accessibility to the K⁵¹ and K⁷² receptor sites without a possible interference from the other two potential trypsin cleavage sites (see below). The PAR₂R³⁶ASⁱ⁴² mutant that was resistant to trypsin activation at residue R³⁶ nonetheless provided for an internal sequence "G⁴⁰ R/Sⁱ⁴² L⁴² " the same as that targeted by trypsin (G³⁵R/SL³⁸) to reveal the wild-type tethered receptor-activating ligand. Thus, trypsin cleavage at the R⁴¹/Sⁱ⁴² bond could potentially reveal the N-terminal sequence [Sⁱ⁴² LDTPPP⁴⁷... ]. Notwithstanding, this receptor mutant failed to generate a calcium signal upon trypsin treatment (Fig. 2F, left tracing), whereas the calcium signal generated by 50 µM SL-NH₂ was equivalent to that of PAR₂wt and the other receptor mutants (Fig. 2F, middle tracing).

View larger version (20K): [in this window] [in a new window]   Fig. 2.   Tracings of calcium signals of PAR2 variants. Calcium signaling (E530, reflecting increases in intracellular calcium) by PAR2wt (A), PAR2R36A (B), PAR2S37P (C), PAR2R41A (D), PAR2R36AR41A (E), and PAR2R36ASi42 (F) was monitored in response to either trypsin () or the PAR2AP SLIGRL-NH2 (SL-NH2, O). Responses were compared relative to the E530 signal yielded in each cell sample by 2 µM ionophore A23187 (). The scale for time and calcium signal is shown to the right in tracing A. In cells treated with ionophore, the calculated intracellular calcium concentration rose from a basal level of 30 nM to a peak value of about 340 nM (Kao et al., 1989; Minta et al., 1989).

Flow Cytometric Analyses and Immunocytochemistry. Because comparable EC₅₀ values for the PAR₂ agonist SLIGRL-NH₂ as well as equivalent maximal calcium responses and receptor densities were observed in all KNRK-PAR₂ variants (Fig. 2; data not shown), it proved possible to use flow cytometric analyses and an immunocytochemistry approach to evaluate the comparative loss, upon trypsin treatment, of the antigenic determinants detected by the anti-receptor antibodies B5 and SLAW-A (Al-Ani et al., 1999b, 2002a; Compton et al., 2001). These antibodies scan receptor sequences within (B5) or N-terminal (SLAW-A) to the PAR₂ trypsin cleavage/activation site (Fig. 1B). As already described, the cell surface antigenic determinants detected by B5 span the putative cleavage/activation site (G³⁰PNSKGR³⁶/SLIGR⁴¹LDTP⁴⁵) in rat PAR₂. As also mentioned above, in the absence of trypsin treatment, the reactivity of all expressed mutant receptors with the B5 antibody was comparable with that of the wild-type receptor PAR₂wt (data not shown). Exposure of the PAR₂R³⁶ASⁱ⁴²-expressing cell line that failed to yield a calcium signal upon trypsin treatment (Fig. 2F, left tracing) by morphometric and flow cytometric analyses showed about a 90% loss of detectable B5 reactivity from the cell surface when treated with 40 nM trypsin for 5 min at room temperature (Fig. 3, G and H). In contrast, as determined in three independent replicate assays, the other receptor mutants retained 98 to 100% of their B5 reactivity (flow cytometric analysis), when treated with 40 nM trypsin for 5 min at room temperature or at 37°C (Fig. 3, A-D). The reactivity of the receptor mutants with the SLAW-A antibody that was generated using the receptor sequences S⁵LAWLLG¹¹ and G³⁰PNSKGR³⁶ of PAR₂ as antigenic determinants (Fig. 1B) is also shown in Fig. 3. Unfortunately, even before trypsin treatment, the SLAW-A antibody failed to detect its antigenic determinants on the cell surface in KNRK cells expressing receptors with an R³⁶A mutation. Thus, it was not possible to use this reagent to assess trypsin cleavage for PAR₂R³⁶A, PAR₂R³⁶AR⁴¹A, and PAR₂R³⁶ASⁱ⁴². However, a full signal was observed using SLAW-A for cells expressing PAR₂wt, PAR₂R⁴¹A, and PAR₂S³⁷P (Fig. 3, top, left-hand and middle open histograms). For the wild-type and PAR₂R⁴¹A mutant, in contrast with B5 reactivity, the majority of the SLAW-A reactivity was removed by incubation with trypsin (40 nM) for 5 min at room temperature, whereas in the PAR₂S³⁷P mutant, which was fully reactive with SLAW-A before trypsin treatment, exposure to 20 nM enzyme failed to remove the SLAW-A reactivity (Fig. 3, top, compare the first and fourth hatched histograms with the third from the left). At 40 nM (compared with 0.3 nM trypsin used by Nystedt et al., 1994), trypsin overcame the hydrolysis resistance caused by the proline mutation and removed about 40% of the SLAW-A reactivity in PAR₂S³⁷P (data not shown); yet, under comparable conditions, as indicated above, trypsin did not generate a calcium signal in this receptor mutant (Fig. 2C, left tracing). The results obtained with the immunocytochemical staining method that detects cell surface receptor using the B5 and SLAW-A anti-receptor antibodies (Fig. 3, bottom) were comparable with those obtained with flow cytometry (Fig. 3, top). Before trypsin treatment, staining of PAR₂ at the cell surface of transfected KNRK cells was visualized using the B5 antibody in PAR₂wt (Fig. 3A), PAR₂R³⁶AR⁴¹A (Fig. 3C), PAR₂R³⁶ASⁱ⁴² (Fig. 3G), as well as for PAR₂R³⁶A, PAR₂R⁴¹A, and PAR₂S³⁷P (data not shown). Using a morphometric analysis approach documenting the percentage of cells with visible cell membrane staining (see Materials and Methods), cell surface receptor B5 reactivity was observed in about 80 to 90% of the cell population in these cell lines prior to trypsin treatment. Using the SLAW-A antibody, it was possible to detect the receptor in >80% of cells expressing PAR₂wt (Fig. 3E), PAR₂S³⁷P (Fig. 3I) and PAR₂R⁴¹A (photomicrograph not shown). However, in keeping with the flow cytometry data (Fig. 3, top), immunocytochemistry using the SLAW-A antibody failed to visualize the receptor in the PAR₂R³⁶A, PAR₂R³⁶AR⁴¹A, and PAR₂R³⁶ASⁱ⁴² constructs that had their R³⁶ residues mutated to A (data not shown). After trypsin treatment (40 nM for 5 min at room temperature), the B5 antibody was still able to detect receptor in >90% of cells expressing PAR₂wt (Fig. 3B), PAR₂R³⁶AR⁴¹A (Fig. 3D), as well as in PAR₂R³⁶A, PAR₂R⁴¹A, and PAR₂S³⁷P (photomicrographs not shown), presumably because of retention near the cell surface of the tethered ligand sequence [SLIGRLDTP... ] against which B5 had been targeted. In contrast, trypsin treatment (20 nM) removed more than 90% of the cell surface determinants detected by the B5 antibody in cells expressing PAR₂R³⁶ASⁱ⁴² (Fig. 3H). Trypsin treatment (20 nM) also removed the epitopes visualized by the SLAW-A antibody in cells expressing PAR₂wt and PAR₂R⁴¹A (Fig. 3F; data not shown); less than 10% of the cells were positive by morphometric analysis after trypsin treatment. However, trypsin at 20 nM did not remove the epitopes visualized by SLAW-A in cells expressing PAR₂S³⁷P (Fig. 3J). Morphometric analysis revealed that >80% of the PAR₂S³⁷P cells were still positive for cell surface immunoreactivity with SLAW-A after trypsin treatment. As recorded under Materials and Methods, no immunoreactivity was detected by the two antibodies either in the "empty" vector-transfected KNRK cell line or in experiments where the B5 and SLAW-A antibodies were preabsorbed with the receptor-derived peptide immunogens (Al-Ani et al., 1999b, 2002a; data not shown). Thus, the data using the immunohistochemical approach for receptor detection were entirely in accord with the data obtained using the flow cytometry method.

View larger version (53K): [in this window] [in a new window]   Fig. 3.   Detection of cell surface reactivity of PAR2 variants using antibodies B5 and SLAW-A. Top, histograms from flow cytometric analysis. As outlined under Materials and Methods, variants of PAR2-expressing KNRK cells, PAR2wt, PAR2R36A, PAR2S37P, PAR2R41A, PAR2R36AR41A, and PAR2R36ASi42, were treated (hatched histograms) or not (open histograms) with trypsin (40 nM, except for PAR2S37P, 20 nM) at room temperature for 5 min at which point soya trypsin inhibitor (1 µg/ml) was added to terminate proteolysis. Cell surface antibody reactivity was then detected by flow cytometry using the SLAW-A anti-receptor antibody. Values as a percentage of the fluorescence yield of untreated cells (% SLAW-A) represent the averages (± S.E.M., bars) of three different experiments done using separately grown groups of receptor-expressing cells. Note that even before trypsin treatment, SLAW-A failed to react with receptors having an R36A mutation. Bottom, immunohistochemistry with B5 and SLAW-A antibodies. As for the flow cytometric analysis, cells expressing wild-type (wt) or mutant receptors were treated (B, D, F, H, and J) or not (A, C, E, G, and I) with trypsin and processed for immunohistochemistry using either the B5 (A-D, G, and H) or SLAW-A (E, F, I, and J) antisera. Receptor detection was observed as a dark ring-like staining at the cell periphery (filled arrows), which was quantified further by morphometric analysis as described under Materials and Methods.

Trypsin Cleavage in PAR₂ Variants Preactivated with SLIGRL-NH₂. The results described in the preceding sections indicated that potential trypsin cleavage sites C- terminal to the activation/cleavage sequence (R³⁶/S³⁷) seem to be sequestered at the cell surface, so as to be resistant to trypsin cleavage. We hypothesized that upon receptor activation, the conformation of the receptor might change so as to make residues K⁵¹ and K⁷² accessible to trypsin. We therefore tested this hypothesis by activating the receptor with SLIGRL-NH₂ concurrent with trypsin treatment, using mutants that did not possess trypsin cleavage sites at R³⁶. According to the hypothesis, activation of these receptors with SLIGRL-NH₂ (Fig. 2) could lead to a conformational change that might enable trypsin to cleave sites K⁵¹ or K⁷², thereby removing the antigenic determinants (from rat PAR₂ residues G³⁰ to P⁴⁵) detected by the B5 antiserum. Compared with untreated cells, the PAR₂wt, PAR₂R³⁶A and PAR₂R³⁶AR⁴¹A cells treated with either 100 µM SLIGRL-NH₂ or 40 nM trypsin at room temperature for 5 min showed no loss of cell surface reactivity with the B5 antiserum monitored either by flow cytometry or by immunocytochemical detection (data not shown). Similarly, incubation of these cells at room temperature first with 100 µM SLIGRL-NH₂ (sufficient to generate a robust calcium signal in all mutants; Fig. 2), followed by treatment with trypsin at concentrations up to 100 nM (more than sufficient to remove all PAR₂wt reactivity with SLAW-A; Fig. 3F) failed to cause a loss of B5 reactivity with any of these receptors expressed at the cell surface, as determined by FACS analysis and immunocytochemistry (data not shown). In contrast, even without receptor preactivation, the PAR₂ mutant with an insertion of serine at receptor residue 42 (PAR₂R³⁶ASⁱ⁴²) was susceptible to trypsin cleavage (presumably at R⁴¹), resulting in a complete loss of the antigenic determinants detected by B5 (Fig. 3H). Thus, if tryptic cleavage of preactivated PAR₂ mutants PAR₂R³⁶A or PAR₂R³⁶AR⁴¹A had occurred at residues K⁵¹ or K⁷² to release the N-terminal sequence, reactivity toward B5 would have been lost.

	Discussion

Top Abstract Introduction Materials and Methods Results Discussion References

The main finding of this study using receptors expressed in intact cells was that in mutants of rat PAR₂, in which trypsin cleavage was prevented at the receptor R³⁶/S³⁷ activation site, further cleavage by trypsin, under conditions that yielded a full calcium signal in PAR₂wt (5 min at 24°C), did not occur at three potential trypsin cleavage sites R⁴¹, K⁵¹, and K⁷², located downstream of the putative tethered ligand sequence beginning with S³⁷. Thus, upon activation of PAR₂ by trypsin to generate a calcium signal, the receptor-activating sequence [SLIGRLDTP... ] would remain persistently attached to the receptor. This persistent attachment of the tethered ligand, which may not occur for other PARs or for the activation of PAR₂ by serine proteinases other than trypsin, may potentially affect the further down-regulation and trafficking of PAR₂. This finding is in contrast with our own previous data demonstrating that trypsin and tryptase can cleave the PAR₂-derived synthetic peptide [GPNSKGRSLIGRLDTP... ] at all of its putative serine proteinase cleavage sites, including the site R³⁶/S³⁷ that reveals the receptor-activating peptide [SLIGRL... ] (Compton et al., 2001; Al-Ani et al., 2002a). More importantly, the data differ substantially from the results of Loew et al. (2000), who showed that a relatively long polypeptide (R³¹-P⁷⁹) representing a considerable portion of the N-terminal domain of human PAR₂ (expressed in E. coli), which in principle should be able to adopt considerable secondary structure in solution, was susceptible to hydrolysis by trypsin (2.5 nM, much lower than the 20 nM we used) at a number of potential cleavage sites (especially, K⁵¹ and K⁷²) that would remove (i.e., disarm) the receptor's tethered ligand sequence, thereby silencing the PAR₂ system. This cleavage would remove the antigenic determinants detected by either the B5 or SLAW-A antibodies. In marked contrast, when reacting with an intact cell expression system, trypsin, working at higher concentrations and under conditions comparable with those used to monitor R³¹-P⁷⁹ polypeptide hydrolysis in vitro, was not able to remove the N-terminal antigenic determinants of PAR₂ variants that were not susceptible to cleavage at the tethered ligand site (R³⁶) but were nonetheless potentially cleaved at residues K⁵¹ and K⁷². In keeping with the results obtained with the recombinant R³¹-P⁷⁹ human PAR₂ polypeptide exposed to trypsin in vitro (Loew et al., 2000), we were not able to detect removal of the N-terminal B5 antigenic determinants of PAR₂ via potential cleavage of R⁴¹/L⁴² in the PAR₂ mutant PAR₂R³⁶A. Notwithstanding, the receptor mutant PAR₂R³⁶ASⁱ⁴², in which an extra amino acid was added to mimic at residues G⁴⁰R⁴¹Sⁱ⁴²L⁴², the sequence at the cleavage/activation site (G³⁵R³⁶/S³⁷L³⁸), but which was resistant to hydrolysis at R³⁶A, was readily cleaved by trypsin (presumably at the R⁴¹/Sⁱ⁴² bond), so as to lose its N-terminal antigenic determinants detected by the B5 antiserum. Evidently, the epitopes detected by B5 must reside N-terminal to L⁴² (removed by trypsin for PAR₂R³⁶ASⁱ⁴², but not for PAR₂R³⁶A). Importantly, the PAR₂R³⁶ASⁱ⁴² mutant did not yield a calcium signal in response to trypsin, and the trypsin-revealed sequence [Sⁱ⁴²LDTP... ] seems incapable of receptor activation. Furthermore, taken together, the data indicate, as found by Loew et al., (2000), that the receptor sequence GR⁴¹LDT is somehow intrinsically resistant to trypsin cleavage. However, in contrast with the results of Loew et al. (2000), our data strongly suggest that when expressed at the cell surface, either because of its glycosylation status (Compton et al., 2001), its secondary structure, or for other reasons yet to be determined, PAR₂ trypsin target sites that might potentially "silence" the receptor by removing the tethered ligand sequence, remain inaccessible to the enzyme, such that there is preferential tryptic cleavage/activation of the receptor at R³⁶/S³⁷, leaving the tethered ligand sequence physically attached to the body of the receptor.

In previous work with the thrombin receptor (PAR₁), Vu et al. (1991b) sequentially changed three potential thrombin cleavage residues, R⁴¹, R⁴⁶, and R⁷⁰, to alanines, finding that R⁴¹ represented the key cleavage site that revealed the putative tethered receptor-activating ligand. This work established unequivocally the ability of the tethered ligand sequence, S⁴²FLLR⁴⁶NPNDKYEPF⁵⁵ to activate PAR₁. Nonetheless, the sequence [LLR⁴⁶/N⁴⁷PN... ] represents another potential thrombin cleavage site. Unfortunately, the original studies did not determine whether thrombin cleavage had or had not occurred at sites (e.g., R⁴⁶ and R⁷⁰) other than the cleavage/activation site (R⁴¹/S⁴²). These other PAR₁ sites in principle could be targeted by a serine proteinase. Their conclusion was that R⁴¹ is critical for thrombin receptor activation and that residues R⁴⁶ and R⁷⁰ play no role in PAR₁ activation; but these two latter residues in principle could play a role in receptor disarming/silencing not only by thrombin but also by other serine proteinases. Subsequent work by Ishii et al. (1993), using two antibodies, one (AP) that detects PAR₁ residues located upstream of the putative thrombin cleavage/activation site (R⁴¹), and a second (HIR) that detects PAR₁ epitopes C-terminal of R⁴¹, starting at Y⁵², indicated that thrombin probably does not cleave at residue R⁷⁰. However, our own work (Kawabata et al., 1999) demonstrated that, as opposed to thrombin, trypsin can both activate and disarm PAR₁, presumably by cleaving both at the activation site to reveal the tethered ligand and at targets downstream of R⁴¹. The site(s) at which trypsin can silence PAR₁, in terms of its subsequent activation by thrombin (Kawabata et al., 1999) remains to be determined.

The results obtained with the PAR₂S³⁷P mutant merit comment. As indicated by our data, this receptor mutant was not able to generate a calcium signal even at trypsin concentrations (40 nM) sufficient to remove 40% of the upstream antigenic epitope detected by SLAW-A (Fig. 2C; data not shown). Because the results with the R³⁶A receptor mutants showed that trypsin was not capable of hydrolyzing the receptor at other downstream target lysine residues (retention of B5 reactivity), one must conclude from our work that 40 nM trypsin was sufficient to cleave (albeit inefficiently) the R³⁶P³⁷ bond, to reveal a tethered sequence beginning: [PLIGRLDTP... ]. It would therefore seem that this sequence as a tethered ligand, like the sequence, [Sⁱ⁴²LDTP... ], has little or no ability to activate PAR₂. It remains an open question as to whether a comparable sequence in PAR₁, i.e., PFLLRNPN... , might be similarly inactive as a tethered ligand moiety. Although Ishii et al. (1993) demonstrated that 10 nM thrombin was not able to cleave an R⁴¹/P⁴² site in a mutated PAR₁, they did not test higher concentrations of thrombin to see whether it would overcome the proline resistance to hydrolysis, in keeping with our results with trypsin cleavage of the PAR₂S³⁷P mutant. In summary, in accord with the work by Ishii et al. (1993) using site-targeted antibodies to assess receptor cleavage of PAR₁ by thrombin, we have established that for PAR₂ expressed in intact cells, as opposed to data obtained for trypsin acting on PAR₂ polypeptide sequences in vitro, trypsin preferentially cleaves principally if not exclusively at the PAR₂ R³⁶/S³⁷ activation site, so as to release the N-terminal portion of the receptor and to liberate the activating tethered ligand, while leaving the remaining N-terminal domain persistently attached to the receptor. Our work highlights the importance of using an intact cell expression system, rather than using synthetic or recombinant polypeptides for future studies of the potential impact of various enzymes on the activation/disarming of PAR₂.