Lanthanide labeling of a potent protease activated receptor-2 agonist for time-resolved fluorescence analysis

Abstract

Protease activated receptor-2 (PAR₂) is one of four G-protein coupled receptors (GPCRs) that can be activated by exogenous or endogenous proteases, which cleave the extracellular amino-terminus to expose a tethered ligand and subsequent G-protein signaling. Alternatively, PAR₂ can be activated by peptide or peptidomimetic ligands derived from the sequence of the natural tethered ligand. Screening of novel ligands that directly bind to PAR₂ to agonize or antagonize the receptor has been hindered by the lack of a sensitive, high-throughput, affinity binding assay. In this report we describe the synthesis and use of a modified PAR₂ peptidomimetic agonist, 2-furoyl-LIGRLO-(diethylenetriaminepentaacetic acid)-NH₂ (2-f-LIGRLO-dtpa), designed for lanthanide-based time resolved fluorescence screening. We first demonstrate that 2-f-LIGRLO-dtpa is a potent and specific PAR₂ agonist across a full spectrum of in vitro assays. We then show that 2-f-LIGRLO-dtpa can be utilized in an affinity binding assay to evaluate the ligand-receptor interactions between known high potency peptidomimetic agonists (2-furoyl-LIGRLO-NH₂, 2-f-LIGRLO; 2-aminothiazol-4-yl-LIGRL-NH₂, 2-at-LIGRL and; 6-aminonicotinyl-LIGRL-NH₂, 6-an-LIGRL) and PAR₂. A separate N-terminal peptidomimetic modification (3-indoleacetyl-LIGRL-NH₂, 3-ia-LIGRL) that does not activate PAR₂ signaling was used as a negative control. All three peptidomimetic agonists demonstrated sigmoidal competitive binding curves, with the more potent agonists (2-f-LIGRLO and 2-at-LIGRL) displaying increased competition. In contrast, the control peptide (3-ia-LIGRL) displayed limited competition for PAR₂ binding. In summary, we have developed a Europium-containing PAR₂ agonist that can be used in a highly sensitive affinity binding assay to screen novel PAR₂ ligands in a high-throughput format. This ligand can serve as a critical tool in the screening and development of PAR₂ ligands.

INTRODUCTION

PAR₂ is one of a four-member family of GPCRs that are activated in response to numerous different proteases^1–5. PAR₂ is expressed in a variety of cell types and associated with a number of diseases including arthritis, allergic asthma, cancer, and acute and chronic pain. Similar to the other PARs, proteolytic cleavage of the extracellular N-terminus exposes a tethered ligand sequence. This ligand intramolecularly binds to the receptor to initiate G-protein dependent and independent intracellular signaling pathways^1,5,6. Exposure of the primary PAR₂ protease cleavage site reveals a tethered ligand (peptide sequence ending in SLIGRL in rodent and SLIGKV in humans) that elicits two dominant intracellular signaling events: an increase in intracellular calcium ion concentration ([Ca²⁺]_i) and increased phosphorylation of mitogen activated protein kinases (MAPKs). Interestingly, alternative protease cleavage sites and subsequent ligands can induce biased MAPK signaling over Ca²⁺ signaling, and potentially, alternative physiological responses initiated at PAR^2,5–8. All of the possible signaling and physiological outcomes that follow PAR₂ activation share in common a ligand-receptor interaction. While this ligand-receptor interaction is important to initiate signaling, the interaction itself remains ill-defined.

Small activating peptides or peptidomimetics that mimic the binding properties of the natural tethered ligand can activate PAR₂ specifically and potently in the absence of proteolytic cleavage^1,3,5,9. Such ligands have the specific advantage of eliminating the potential off-target effects of proteases that limit the use of proteases and protease inhibitors in evaluating PAR₂ effects in vitro and in vivo. The continued screening and development of novel ligands that specifically bind PAR₂ to agonize or antagonize the receptor can serve as probes to better characterize downstream signaling pathways of PAR₂, its physiological and pathophysiological roles, and serve as templates for drug design. We recently developed a high-throughput in vitro screening assay that includes physiological and signaling pathway responses to PAR₂ ligands⁹. This method helped to establish full characterization of known and novel PAR₂ ligands; however, it lacked a biochemical approach to evaluate direct interactions between novel ligands and PAR₂.

Published methods for evaluating PAR₂ ligand-receptor interactions include the use of peptidomimetic analogs (trans-cinnamoyl-LIGRLO-NH₂, 2-furoyl-LIGRL-NH₂ and 2-furoyl-LIGRLO-NH₂) modified by radiolabels or traditional fluorescence^10–13. These studies demonstrated binding curves and evaluated PAR₂ ligand binding (e.g., IC₅₀s^10,13, K_is^11–13, relative K_is (compared to SLIGKV-OH)^12,13 or relative IC₅₀s (compared to SLIGRL-NH₂)^10,11 following competitive binding assays in tissue culture plates and/or suspended cultures. While these assays showed the utility of modifying known peptidomimetic ligands to develop competitive binding assays, a distinct disadvantage of these ligands is their limited adaptability to high-throughput analyses, thus compromising their use for PAR₂ ligand discovery. The recent development of lanthanide series luminescence/fluorescence assays allows for fluorescence-detection based competitive binding assays in vitro using a multi-well plate format, as well as the potential to develop similar probes for in vivo PAR₂ studies^14,15

Lanthanides (e.g., Terbium, Tb; Dysprosium, Dy; Europium, Eu; Samarium, Sm) exhibit the unique characteristics of a low detection limit (10⁻¹²–10⁻¹⁵ M), narrow emission bands (10–20 nm), large Stokes shift (greater than 200 nm), and long-lived fluorescence (milliseconds for Eu)^16–18. Prolonged fluorescence enables the use of time-resolved fluorescence (TRF) measurements¹⁸, greatly improving the signal to noise ratio¹⁹ and thus, increases fluorescent measurement sensitivity. We wanted to take advantage of the physical properties of lanthanides and used a modified Dissociation Enhanced Lanthanide Fluoroimmunoassay (DELFIA; Wallac/PerkinElmer Life Sciences) to carry out competitive binding assays. This technique has proven to be successful in screening (e.g. ligand binding assays), biodistribution and biomedical applications and is the most successful lanthanide-based immunoassay reported to date^14,17,18,20.

In this study we sought to develop a potent, PAR₂-specific ligand, whose characteristics allow for a functional high-throughput screening assay to characterize ligand-receptor interactions. We synthesized a derivative of the highly potent 2-f-LIGRLO modified with diethylenetriaminepentaacetic acid (dtpa), which serves as the chelate for a Europium (III) to be used in a time-resolved fluorescence assay^14,17,18. The resulting compound, 2-f-LIGRLO-dtpa was first characterized for its ability to function as a full and specific PAR₂ agonist using both in vitro physiological and cellular signaling assays. We next showed that 2-f-LIGRLO-dtpa can be used in a competitive binding assay with potent, full PAR₂-specific agonists (2-f-LIGRLO, 2-at-LIGRL, and 6-an-LIGRL), and the inactive peptidomimetic 3-ia-LIGRL. We propose that 2-f-LIGRLO-dtpa can be used to evaluate PAR₂ ligand-receptor interactions to both screen novel compounds and to better understand the role of ligand-receptor interactions in the signaling pathway-specific activation of PAR₂.

EXPERIMENTAL PROCEDURES

Materials

Materials and chemicals were of molecular biology grade or higher. Sources are listed in the Supplemental Material.

PAR₂ ligand synthesis

Peptides and peptidomimetics (2-f-LIGRLO, 2-at-LIGRL, 6-an-LIGRL and 3-ia-LIGRL) were prepared as previously described²¹. The lanthanide-labeled peptidomimetic, 2-f-LIGRLO-dtpa was made by coupling dtpa to the ε-amino group of the ornithine in the parent PAR₂ ligand, 2-f-LIGRLO (Figure 1) followed by the chelation of a Europium (III) ion via a synthesis scheme previously described¹⁸. Specific details on syntheses and purity are provided in the Supplemental Material.

Figure 1. Structure of 2-furoyl-LIGRLO-dtpa-NH₂ (2-f-LIGRLO-dtpa).

A) Chemical structure of the parent PAR₂ hexapeptidomimetic 2-furoyl-LIGRLO-NH₂ is shown in blue. The diethylenetriaminepentaacetic acid (dtpa) chelating a Europium (III) and conjugated with the parent compound via the ε ornithine is shown in red.

Epithelial cell culture and cellular signaling measurements

The 16HBE14o- human bronchial epithelial cell line was used for physiological and second messenger screening assays and in competitive binding experiments were grown as previously described⁹. PAR₂ and vector control-transfected Kirsten virus-transformed normal rat kidney (kNRK-PAR₂ and kNRK, respectively) cells were used to test specificity of the 2-f-LIGRLO-dtpa ligand. Cell growth and specifics on physiological and signaling analyses are described in the Supplemental Material.

Competitive binding assay

Black 96-well plates with white, opaque wells were initially incubated with a matrix coating solution (MCS consisting of 88% LHC basal medium, 10% bovine serum albumin (BSA), 1% bovine collagen type I and 1% human fibronectin for two h at 37°C. MCS was removed and plates allowed to dry for at least one h. 16HBE14o- cells were plated on at a density of 6.0 × 10⁴ cells per well and grown at 37°C, 5% CO₂ to confluence (4 – 5). On the day of the experiment, growth media was aspirated using a multi-well vacuum manifold (Corning Inc, Corning, NY) and cells were washed with 200 µl of pre-warmed (37°C) Hanks’ balanced saline solution additionally buffered with 25 mM HEPES, pH 7.4 (HBSS) using a multi-channel pipet. HBSS was replaced with 100 µl of a 2% BSA in HBSS solution and incubated for 75 min to block non-specific interactions. All peptide competition solutions were made at 2× final concentration in a deep well mixing plate. Competition solutions consisted of a fixed concentration of 2-f-LIGRLO-dtpa (final concentration: 300 nM) and concentration response concentrations of peptidomimetics (final concentrations: 1 nM to 600 µM in log or ½ log steps). 100 µl of competition solutions were added to each well in the 96-well plate (final volume of 200 µl) and incubated with the BSA-blocked live cells for 45 min at 25°C. Competition solutions were carefully removed, and cells were washed three times with a wash solution (Base medium: HBSS supplemented with 20 µM EDTA, 0.01% Tween, and 0.2% BSA). The wash solution was removed and replaced with 100 µl of DELFIA enhancement solution (Perkin Elmer; Waltham, MA) for 40 – 45 min at room temp and fluorescence determined as previously described^17,18. Briefly, 96-well plates were read on a Wallac VICTOR₃ instrument (Perkin Elmer; Waltham, MA) using standard Eu time-resolved fluorescence (TRF) measurement: excitation light (340 nm) followed by a 400 µsec delay before a 400 µsec emission collection (615 nm). The total measurement cycle was set to 1000 µsec per well.

Statistics

All statistical analyses were evaluated with GraphPad software (San Diego, CA). Multivariate comparisons were done with a one-way ANOVA with Tukey’s multiple comparison post-test. Pair-wise comparisons were done with a 2-tailed Student’s t test. Relative IC₅₀s were calculated using a one-site fit logIC50 non-linear regression curve-fit analysis. A value of p < 0.05 was used to establish a significant difference between samples. Data in figures are graphed as ± SEM unless otherwise noted.

RESULTS

Evaluation of 2-f-LIGRLO-dtpa as a full PAR₂ agonist

We used a high PAR₂-expressing human bronchial epithelial cell line (16HBE14o-)²² to evaluate the ability of 2-f-LIGRLO-dtpa to activate PAR₂ with a three-tiered screening assay that included in vitro physiological, MAPK signaling and Ca²⁺ signaling assays⁹. To examine physiological activation of 16HBE14o- cells by 2-f-LIGRLO-dtpa, we used the xCELLigence Real Time Cell Analyzer (Roche; RTCA). Cells were exposed to a concentration response (0.05 µM–2.5 µM) of either the parent compound, 2-f-LIGRLO, or 2-f-LIGRLO-dtpa and examined for the ability of each compound to activate PAR₂ (Figure 2). We found 2-f-LIGRLO-dtpa stimulated a maximum Cell Index at a concentration of 2.5 µM. The calculated EC₅₀ of 2-f-LIGRLO-dtpa (388 nM; 95% CI, 336 – 448 nM) was slightly less potent than 2-f-LIGRLO (130 nM; 95% CI, 107 – 159 nM), and similar to another highly potent peptidomimetic PAR₂ agonist, 6-an-LIGRL⁹.

Figure 2. 2-f-LIGRLO-dtpa stimulates a physiological response in model epithelial cells.

(A) 16HBE14o- cells were plated onto E-plates, exposed to a concentration response (50 nM – 2.5 µM) of 2-f-LIGRLO-dtpa and Cell Index was measured every 15 s for 4 h. Cells showed a concentration-dependent physiological response over the 4 h experiment. (B) A concentration response curve was constructed for 2-f-LIGRLO-dtpa and the high potency agonist 2-f-LIGRLO from the area under the curve of individual experiments. EC₅₀ for 2-f-LIGRLO-dtpa (388 nM) is consistent with a potent full agonist, albeit slightly reduced from 2-f-LIGRLO (130 nM)⁹. Traces in (A) and curves in (B) are mean responses from 4 experiments, error bars are eliminated in (A) for clarity.

We next evaluated the ability of 2-f-LIGRLO-dtpa to induce MAPK signaling pathways. 16HBE14o- cells were exposed to a concentration response (1 – 10 µM) of 2-f-LIGRLO-dtpa for 5 min, and examined for activation of MAPK signaling using immunoblot of pERK/tERK (Figure 3). Exposure of 2.5 µM 2-f-LIGRLO-dtpa to 16HBE14o- cells induced a significant increase in pERK/tERK when compared with control. Increasing concentrations up to 10 µM 2-f-LIGRLO-dtpa induced higher pERK/tERK values. In the same assay, 2-f-LIGRLO induced a significant increase of pERK/tERK at 1 µM exposure, and reached a maximum by 5 µM exposure. Although 2-f-LIGRLO-dtpa was slightly less potent than 2-f-LIGRLO, it was comparable to alternative full peptidomimetic agonists 2-at-LIGRL and 6-an-LIGRL⁹.

Figure 3. 2-f-LIGRLO-dtpa stimulates MAPK signaling in model epithelial cells.

16HBE14o- cells were exposed to a concentration response of either 2-f-LIGRLO or 2-f-LIGRLO-dtpa for 5 min. Representative immunoblots show increased ERK1/2 phosphorylation (pERK) compared to total ERK1/2 (tERK) beginning at 1 µM agonist concentration and maintained at concentrations up to 10 µM for each ligand. Densitometric quantification (n=4) demonstrates significant increase of pERK at 1 µM and peak activation by 5 µM for 2-f-LIGRLO, while 2-f-LIGRLO-dtpa was slightly less potent, with a significant increase in pERK at 2.5 µM and a peak response by 10 µM. Significant difference from matching HBSS control are shown: “*”, p < 0.05; “**”, p < 0.01.

In a complementary signaling assay to evaluate PAR₂ activation by 2-f-LIGRLO-dtpa, we examined the activation of Ca²⁺ signaling in 16HBE14o- cells using digital imaging microscopy^9,21. Upon addition of 5 µM 2-f-LIGRLO-dtpa, an initial increase in [Ca²⁺]_i was observed within 30 sec, with a peak response observed between 45 and 60 sec (Figure 4; supplemental video). The majority of cells in the field of view responded within the 3 min experiment. An advantage of this imaging method is the detection of [Ca²⁺]_i changes in individual cells and thus, an increased sensitivity in the PAR₂ response. A concentration response curve (0.5 µM – 25 µM 2-f-LIGRLO-dtpa) is shown in Figure 4B. The EC₅₀ for 2-f-LIGRLO-dtpa (1.34 µM; 95% CI, 0.85–2.11 µM), was slightly less efficacious than that reported for 2-f-LIGRLO but more efficacious than another high potency PAR₂ specific agonist, 2-at-LIGRL^9,21. In summary, 2-f-LIGRLO-dtpa displays signaling responses across a variety of assays that are similar to other high potency, full PAR₂ agonists.

Figure 4. 2-f-LIGRLO-dtpa stimulates Ca²⁺ signaling in model epithelial cells.

16HBE14o-cells were exposed to 2-f-LIGRLO-dtpa and [Ca²⁺]_i was monitored over time. (A) Individual panels depict [Ca²⁺]_i in all cells within a field of view at listed times (sec) following application of 5 µM 2-f-LIGRLO-dtpa at time 0. White lines approximate individual cell borders in each image and estimated [Ca²⁺]_i is depicted in the color bar shown in upper left panel. 2-f-LIGRLO-dtpa stimulates a rapid increase in [Ca²⁺]_i that reaches peak changes by 45 – 60 sec in most cells and begins to return towards base line by 90 sec. (B) Percentage of cells in the field of view that show [Ca²⁺]_i responses within the 3 min experiment are graphed as a concentration response to 2-f-LIGRLO-dtpa. The calculated EC₅₀ of 2-f-LIGRLO-dtpa for a Ca²⁺ response is 1.34 µM, comparable to known full PAR₂ agonists⁹. A minimum of 3 and maximum of 9 experiments are plotted for each concentration shown in (B).

Specificity of 2-f-LIGRLO-dtpa to PAR₂

To test the specificity of 2-f-LIGRLO-dtpa, we used Kirsten virus-transformed normal rat kidney cells, a low-PAR₂ expressing epithelial cell line. We compared the activation of Ca²⁺ signaling in PAR₂-transfected kNRK cells (kNRK-PAR₂) to those transfected with a control vector (kNRK) (Figure 5). Exposure of kNRK-PAR₂ cells to 5 µM or 2.5 µM 2-f-LIGRLO-dtpa resulted in Ca²⁺ signaling in a high percentage of cells (90.1% ± 4.8 and 55.1% ± 11.5, respectively). Similar experiments using kNRK cells resulted in limited Ca²⁺ signaling (2.98% ± 0.09 and 0.42% ± 0.42, respectively). These results indicate our novel 2-f-LIGRLO-dtpa ligand requires PAR₂ for initiating cellular signaling and thus, is a specific agonist to PAR₂.

Figure 5. 2-f-LIGRLO-dtpa requires PAR₂ expression to induce Ca²⁺ signaling.

PAR₂-transfected (kNRK-PAR₂) or control vector-transfected (kNRK) rat kidney cells were exposed to 2-f-LIGRLO-dtpa and evaluated for Ca²⁺ response as described in Fig. 4. Addition of 5 or 2.5 µM 2-f-LIGRLO-dtpa resulted in a robust Ca²⁺ response in PAR₂ expressing cells that was largely absent in control cells. “**” denotes p < 0.01 and “***” denotes p < 0.001 when compared to matched control; n ≥ 4 for each experiment.

PAR₂ competitive binding assay using 2-f-LIGRLO-dtpa

The primary advantage of 2-f-LIGRLO-dtpa compared to other PAR₂ ligands is its use to evaluate novel PAR₂ ligand-receptor interactions in live cells using time-resolved fluorescence. We used 2-f-LIGRLO-dtpa in a modified Dissociation Enhanced Lanthanide Fluoroimmunoassay (DELFIA) based Europium-DTPA competitive binding assay, similar to the methods previously used^17,18, to evaluate the ligand-receptor interaction between novel ligands and PAR₂. We competed 2-f-LIGRLO-dtpa (300 nM) with its parent compound, 2-f-LIGRLO, and a recently described highly potent and specific PAR₂ peptidomimetic agonist, 2-at-LIGRL^9,21. Incubation with concentration responses of 2-f-LIGRLO or 2-at-LIGRL resulted in sigmoidal competitive binding curves (Figure 6), with apparent IC₅₀s of 3.29 µM (95% CI: 2.51 – 4.30 µM) and 2.06 µM (95% CI: 1.61 – 2.65 µM) respectively. Similar experiments conducted in the absence of cells did not result in competitive binding, although the highly fluorescent 2-f-LIGRLO-dtpa did display significant signal (15.1 % ± 4.4 of the total signal measured), indicative of an interaction with the reflective plate surface. Because of this background, we are only reporting apparent IC₅₀s for competitive binding. To fully demonstrate the utility of this assay, we tested a less potent ligand, 6-an-LIGRL, and a similar peptidomimetic, 3-ia-LIGRL, which does not activate PAR₂²¹. 6-an-LIGRL displayed a typical sigmoidal binding curve that was significantly shifted to the right of the 2-f-LIGRLO and 2-at-LIGRL curves with an apparent IC₅₀ of 18.54 µM (95% CI, 14.11 – 24.36 µM; Figure 6), indicative of reduced competition for binding at PAR₂. In contrast, 3-ia-LIGRL did not compete for PAR₂ binding (Figure 6). We conclude that our novel PAR₂ agonist, 2-f-LIGRLO-dtpa, can be utilized in a unique, sensitive, high-throughput screening assay to examine PAR₂ ligand binding.

Figure 6. Competitive binding of 2-f-LIGRLO-dtpa with peptidomimetic PAR₂ agonists and controls.

16HBE14o- cells plated onto 96-well plates were simultaneously exposed to 300 nM 2-f-LIGRLO-dtpa and various concentrations of known agonists 2-f-LIGRLO, 2-at-LIGRL, 6-an-LIGRL or a control peptidomimetic, 3-ia-LIGRL and assessed for competitive binding. Binding is expressed as a percentage of the maximal net Eu fluorescence measured from 2-f-LIGRLO-dtpa (300 nM) binding (% Max). The two most potent agonists, 2-f-LIGRLO and 2-at-LIGRL displayed sigmoidal curves consistent with high competition for PAR₂ binding. The slightly less potent full agonist, 6-an-LIGRL, displayed a right-shifted sigmoidal binding curve, indicative of reduced ability to compete against 2-f-LIGRLO-dtpa for receptor binding. The control peptidomimetic, which does not initiate PAR₂ signaling, displayed limited binding, and was best fit with a linear function. Apparent IC₅₀ values for agonists showing competitive binding for PAR₂ are as follows: 2-f-LIGRLO: 3.29 µM; 2-at-LIGRL, 2.06 µM and; 6-an-LIGRL, 18.54 µM. Data of all agonists are from n ≥ 3 samples at each concentration of all agonists in each multi-well experiment, and from at least two independent multi-well experiments. Control peptidomimetic (3-ia-LIGRL) data is from one multi-well experiment.

DISCUSSION

In this report we introduce the Europium (III) containing compound, 2-f-LIGRLO-dtpa, as a potent and specific PAR₂ agonist and demonstrate its use as a probe in a novel, radioligand-free, competitive binding assay to evaluate ligand-receptor interactions at PAR₂. In the live-cell affinity binding experiments using three known full agonists (2-f-LIGRLO, 2-at-LIGRL and 6-an-LIGRL) and a control peptidomimetic (3-ia-LIGRL) a rank order of apparent IC₅₀s was obtained (2-f-LIGRL ≈ 2-at-LIGRL > 6-an-LIGRL >>> 3-ia-LIGRL). This rank order potency closely matches the rank order potency of these compounds tested across physiological assays^9,21. These data strongly suggest that differences in the described ligand potency is directly related to their avidity for PAR₂ and highlight the importance of evaluating PAR₂ binding.

The development and use of PAR₂-specific activating peptides is an increasingly prevalent method to study downstream signaling pathways of PAR₂, the physiological and pathophysiological roles of PAR₂, and provides an avenue for the discovery of therapeutics targeting PAR₂-associated pathologies. Although many methods have been employed for the development and screening of ligands, the current screening of such ligands that directly bind PAR₂ has been limited by the availability of a high-throughput competitive binding assay. Because we had previously used lanthanide-labeled ligand targeting to evaluate a variety of GPCRs^15,17,20, we hypothesized that lanthanide-labeled PAR₂ peptidomimetic agonists would be viable candidates for the protease activated receptor family of GPCRs. The lanthanide containing 2-f-LIGRLO-dpta compound reported herein was synthesized by coupling a diethylenetriaminepentacetic acid (dtpa) molecule, with its inherent ability to effectively chelate a Europium (III), to the ornithine residue of the potent PAR₂ agonist 2-f-LIGRLO. The ornithine had previously been used to attach fluorescent markers without greatly affecting biological activity or receptor binding ability of the PAR₂ parent peptidomimetic (e.g., Alexa Fluor 594 or rhodamine labeled 2-f-LIGRLO^10,23). Because of the highly charged and relatively large chelate, we found it necessary to verify 2-f-LIGRLO-dtpa maintained the ability to interact with PAR₂ through a series of in vitro assays designed to screen and assess the structure-activity relationship of novel PAR₂ ligands⁹. In the in vitro physiological response assay (RTCA), 2-f-LIGRLO-dtpa displayed full agonist properties at both acute (e.g., 10 min) and long-term (e.g., 4 h) time points (Figure 2). Similarly, 2-f-LIGRLO-dtpa acted as a full agonist in cell signaling assays (Figures 3 and 4). A final test for use of 2-f-LIGRLO-dtpa was to determine specificity for PAR₂, which was accomplished using a kNRK cell model (Figure 5). From these assays, it was determined that 2-f-LIGRLO-dtpa acts as a full and specific agonist to PAR₂ with slightly reduced potency when compared with its parent 2-f-LIGRLO compound. As such, it can be considered a suitable probe for competitive binding experiments.

Competitive binding assays for PAR₂ have been reported with both tritiated and fluorescent compounds directed at PAR₂^10–13. Radioactive tritiated compounds (e.g.,[³H]propionyl-2-furoyl-LIGRLO-NH₂) serve as sensitive binding probes, however use of radiolabeled compounds limits ability for adaptation to high-throughput analysis. In contrast, our modified DELFIA competitive binding assay allows for rapid assessment of multiple ligands in one 96-well test plate while providing sensitivity consistent with radioligand binding experiments^17,20,24. Studies using modified ligands with Alexa series dyes have also been used in competitive binding experiments with PAR₂¹⁰. The relatively small Stokes shift (27 nm) which increases sampling error due to scattering of the excitation light by instrument optics, and significant background fluorescence from sample vessels, matrix or cells during measurement of Alexa Fluor 594 reduces measurement sensitivity and presents a disadvantage in using this dye in high-throughput screening. Further disadvantages include the inherent aqueous quenching and short-lived fluorescence that also reduce signal to noise ratio¹⁴. We overcome these limitations by utilizing a modified DELFIA technique combined with the luminescence properties of lanthanide series ions (e.g., Eu). The Eu present in our compound has relatively narrow emission bands and a Stokes shift of ~275 nm, much larger than any available fluorescent dye. These properties significantly reduce noise contributions due to spectral overlap between excitation and emission spectra. Furthermore, our use of the dtpa chelate saturates the lanthanide coordinating sphere and the micellar environment present during fluorescence measurement protect the Eu ion from aqueous quenching^14,20. The long-lived luminescence of the Eu that allows for a delay between excitation and emission measurement readings prevents inherent short-lived background fluorescence from contributing to measurements, and thus, increases signal to noise fidelity.

The availability of a rapid competitive binding assay described herein can now be used to discern relative ability of a ligand to interact at a site that results in full activation of PAR₂ signaling. This will be a critical assessment tool since the evaluation of ligand-receptor interactions is a fundamental screening assay in drug development. Combining this assay with structure-activity relationship data obtained from our in vitro physiological and cell signaling screening assay (e.g.,⁹) provides a potent selection process to identify novel drug candidates to the most common activation site of PAR₂. It should be noted that PAR₂, like other G-protein coupled receptors (GPCRs), can activate alternative signaling pathways^4,6,25–27; however, mechanisms for such diverse activation of GPCRs and PARs remain ill-defined. The development of highly sensitive and high-throughput competitive binding assays as described herein or to potential alternative ligand binding sites on PAR₂, will help us to better understand these novel signaling mechanisms.

ABBREVIATIONS USED

16HBE14o-

Human bronchial epithelial cell line

2-f-LIGRLO-dtpa

2-furoyl-LIGRLO-(diethylenetriaminepentaacetic acid)-NH₂

2-f-LIGRLO

2-furoyl-LIGRLO-NH₂

2-at-LIGRL

2-aminothiazol-4-yl-LIGRL-NH₂

6-an-LIGRL

6-aminonicotinyl-LIGRL-NH₂

3-ia-LIGRL

3-indoleacetyl-LIGRL-NH₂

[Ca²⁺]_i

calcium ion concentration

Aloc

allyloxycarbonyl

BSA

Bovine serum albumin

tBu

tert-butyl

DCM

dichloromethane

DIC

N,N’-diisopropylcarbodiimide

DIEA

diisopropylethylamine

DELFIA

Dissociation Enhanced Lanthanide Fluoroimmunoassay

DMEM

Dulbecco’s Modified Eagle Medium

DMF

N,N’-dimethylformamide

DMSO

dimethylsulfoxide

dtpa

diethylenetriamine-N¹,N²,N³,N⁴-pentaacetic acid

Eu

Europium (III)

E-plate

96-well tissue culture plate lined with gold electrodes used in RTCA

TRF

Time-resolved fluorescenceESI-MS Electrospray ionization - mass spectrometry

Fmoc

(9H-fluoren-9-ylmethoxy)carbonyl

FT-ICR

Fourier Transform - Ion Cyclotron Resonance

GPCR

G-protein coupled receptor

HBSS

Hanks’ balanced salt solution additionally buffered with 25 mM HEPES, pH 7.4

HBTU

2-(1H-benzotriazol-1-yl)-1,1,3,3-tetramethyluronium hexafluoro-phosphate

HOBt

N-hydroxybenzotriazole

HOCt

6-chloro-1H-hydroxybenzotriazole

kNRK

Kirsten virus-transformed normal rat kidney cells transfected with a vector control

kNRK-PAR₂

Kirsten virus-transformed normal rat kidney cells transfected with PAR₂

LHC

Lechner and LaVeck basal medium

MALDI

Matrix Assisted Laser Desorption Ionization - Time of Flight

MAPK

mitogen activated protein kinase

MCS

matrix coating solution

MEM

Minimal Essential Medium with Earle’s salts

PAR₂

Protease activated receptor-2

Pbf

2,2,4,6,7-pentamethyl-dihydrobenzofuran-5-sulfonyl

Pd(TPP)₄

Palladium (0) tetrakis triphenylphoshine

RP-HPLC

reverse-phase high performance liquid chromatography

RTCA

xCELLigence real time cell analyzer (Roche)

SPE

solid-phase extraction

SPPS

solid-phase peptide synthesis

TEAA

triethylammonium acetate

THF

tetrahydrofuran

TFA

trifluoroacetic acid

TRF

time-resolved fluorescence