Biased Signaling by Agonists of Protease Activated Receptor 2 - ACS

Feb 7, 2017 - Structure–activity relationships here for 26 compounds spanned a signaling bias over 3 log units, culminating in three small ligands a...
0 downloads 0 Views 2MB Size
Subscriber access provided by Fudan University

Article

Biased signaling by peptide agonists of protease activated receptor 2 Yuhong Jiang, Mei-Kwan Yau, W. Mei Kok, Junxian Lim, Kai-Chen Wu, Ligong Liu, Timothy A Hill, Jacky Y. Suen, and David P. Fairlie ACS Chem. Biol., Just Accepted Manuscript • DOI: 10.1021/acschembio.6b01088 • Publication Date (Web): 07 Feb 2017 Downloaded from http://pubs.acs.org on February 12, 2017

Just Accepted “Just Accepted” manuscripts have been peer-reviewed and accepted for publication. They are posted online prior to technical editing, formatting for publication and author proofing. The American Chemical Society provides “Just Accepted” as a free service to the research community to expedite the dissemination of scientific material as soon as possible after acceptance. “Just Accepted” manuscripts appear in full in PDF format accompanied by an HTML abstract. “Just Accepted” manuscripts have been fully peer reviewed, but should not be considered the official version of record. They are accessible to all readers and citable by the Digital Object Identifier (DOI®). “Just Accepted” is an optional service offered to authors. Therefore, the “Just Accepted” Web site may not include all articles that will be published in the journal. After a manuscript is technically edited and formatted, it will be removed from the “Just Accepted” Web site and published as an ASAP article. Note that technical editing may introduce minor changes to the manuscript text and/or graphics which could affect content, and all legal disclaimers and ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errors or consequences arising from the use of information contained in these “Just Accepted” manuscripts.

ACS Chemical Biology is published by the American Chemical Society. 1155 Sixteenth Street N.W., Washington, DC 20036 Published by American Chemical Society. Copyright © American Chemical Society. However, no copyright claim is made to original U.S. Government works, or works produced by employees of any Commonwealth realm Crown government in the course of their duties.

Page 1 of 28

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

ACS Chemical Biology

39x20mm (300 x 300 DPI)

ACS Paragon Plus Environment

ACS Chemical Biology

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Biased signaling by peptide agonists of protease activated receptor 2

Yuhong Jiang#, Mei-Kwan Yau#, W. Mei Kok, Junxian Lim, Kai-Chen Wu, Ligong Liu, Timothy A. Hill, Jacky Y. Suen*, David P. Fairlie*

Centre for Inflammation and Disease Research and Australian Research Council Centre of Excellence in Advanced Molecular Imaging, Institute for Molecular Bioscience, The University of Queensland, Brisbane, Qld 4072, Australia

#These authors contributed equally *Correspondence to [email protected], [email protected]

1

ACS Paragon Plus Environment

Page 2 of 28

Page 3 of 28

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

ACS Chemical Biology

Abstract

Protease activated receptor 2 (PAR2) is associated with metabolism, obesity, inflammatory, respiratory and gastrointestinal disorders, pain, cancer and other diseases. The extracellular N-terminus of PAR2 is a common target for multiple proteases, which cleave it at different sites to generate different N-termini that activate different PAR2-mediated intracellular signaling pathways. There are no synthetic PAR2 ligands that reproduce the same signaling profiles and potencies as proteases. Structure-activity relationships here for 26 compounds spanned a signaling bias over 3 log units, culminating in three small ligands as biased agonist tools for interrogating PAR2 functions. DF253 (2f-LAAAAI-NH2) triggered PAR2-mediated calcium release (EC50 2 μM) but not ERK1/2 phosphorylation (EC50 > 100 μM) in CHO cells transfected with hPAR2. AY77 (Isox-Cha-Chg-NH2) was a more potent calcium-biased agonist (EC50 40 nM, Ca2+; EC50 2 μM, ERK1/2), while its analogue AY254 (Isox-Cha-Chg-A-R-NH2) was an ERK-biased agonist (EC50 2 nM, ERK1/2; EC50 80 nM, Ca2+). Signaling bias led to different functional responses in human colorectal carcinoma cells (HT29). AY254, but not AY77 or DF253, attenuated cytokine-induced caspase 3/8 activation, promoted scratch-wound healing and induced IL-8 secretion, all via PAR2-ERK1/2 signaling. Different ligand components were responsible for different PAR2 signaling and functions, clues that can potentially lead to drugs that modulate different pathway-selective cellular and physiological responses.

2

ACS Paragon Plus Environment

ACS Chemical Biology

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Protease activated receptor 2 (PAR2) belongs to a unique class of G proteincoupled receptor (GPCR) that is activated on the cell surface by extracellular proteases, including trypsin, tryptase, factor VIIa, matriptase, neutrophil elastase, cathepsin S, granzyme A and others1-5. The mechanism of activation by many proteases involves cleavage of the extracellular N-terminus of PAR2 at a “canonical” site (Arg36↓Ser37), exposing a new N-terminal receptor sequence (SLIGKV-, human; SLIGRL-, rodent). This “tethered ligand” (TL) receptor sequence intramolecularly activates PAR2 and its coupled intracellular signaling pathways, including calcium mobilization, ERK1/2 phosphorylation, Rho activation, cAMP down-regulation and β-arrestin1/2 recruitment

6, 7

. Synthetic peptides that mimic the TL receptor sequence

(e.g. SLIGRL-NH2, SLIGKV-NH2, 2-furoyl-LIGRLO-NH2) bind to PAR2 and activate cell signaling in the absence of a protease8. PAR2 is a promising drug target due to its association with diseases, including inflammatory conditions, metabolic dysfunction, cancers, cardiovascular, respiratory, gastrointestinal and neurological disorders9-15. Agonists are no longer thought to activate all intracellular signaling pathways coupled to a given GPCR1, rather different agonists can favor activation of some signaling pathways over others, a phenomenon known as biased signaling16-21. This is also true for PAR2,4, 5, 22, 23 for which proteases cleave the extracellular N-terminus of PAR2 at different sites and use different TL sequences to initiate specific intracellular signaling. Thus, trypsin and tryptase cleave Arg36↓Ser37 leading to activation of all known PAR2-dependent signaling, but neutrophil elastase cleaves at Ser68↓Val69 and activates ERK1/2 phosphorylation but not Ca2+ release4, whereas cathepsin S cleaves at Glu56↓Thr57 and induces cAMP accumulation but has no effect on ERK1/2 or Ca2+ 5. These selective cleavages by extracellular proteases enable PAR2 to act as a sensor on 3

ACS Paragon Plus Environment

Page 4 of 28

Page 5 of 28

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

ACS Chemical Biology

the cell surface and to regulate intracellular signaling and cell function in response to changes in a particular extracellular protease. These differences in protease-mediated signaling highlight the importance of the N-terminal protease-generated TL sequence of PAR2 for activating cellular signaling. Synthetic peptides corresponding to different TL receptor sequences have not proved to be effective in distinguishing between different PAR2-mediated signaling pathways. Attempts to correlate effects of mutations in TL receptor sequences with changes in corresponding short synthetic peptides on PAR2 activation have not produced the same signaling outcomes

24, 25

. For example, MAPK signaling

occurs in response to trypsin following mutagenesis in the TL region of PAR2 (e.g. LSIGRL-, -AAIGRL-, -SLAAAA-), but only the –SLAAAA- tethered ligand sequence activated Ca2+ mobilization. Synthetic ligands with these sequences all failed to induce PAR2-mediated intracellular Ca2+, and only SLAAAA-NH2 was able to induce MAPK activation via PAR2 26. With the prospect of differentially controlling individual signaling pathways linked to different diseased states, one could envisage modifying synthetic peptide ligands to differentially alter signaling and function. The synthetic hexapeptide (SLIGKV-OH) has activity at µM concentrations in all PAR2-dependent signaling pathways, and potency can be increased 10-100 fold by changes to the sequence (SLIGRL), N-terminus (e.g. 2-furoyl-LIGRL, 5-isoxazolyl-LIGRL) or C-terminus (OH to –NH2)8, 27, 28. PAR2 binding affinity, Ca2+ and ERK1/2 responses were all enhanced (EC50 300 nM) for 2f-LIGRL-NH2 relative to SLIGRL leading to greater PAR2-mediated inflammatory responses in vivo

29, 30

, but no biased signaling was

observed. Here we report structure-activity relationships for 25 analogues of 2f-

4

ACS Paragon Plus Environment

ACS Chemical Biology

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

LIGRL-NH2 in assays for PAR2-binding affinity, iCa2+ and ERK1/2 signaling on CHO cells transfected with human PAR2 (CHO-hPAR2), with the finding of several examples of ligand-induced biased signaling through PAR2. Mechanistic studies were conducted to highlight the molecular determinants of signaling bias in the ligands, and to link functional responses to these signaling properties. The study indicates that PAR2 ligands can be potentially developed to modulate specific intracellular signaling pathways associated with different PAR2-mediated functions.

5

ACS Paragon Plus Environment

Page 6 of 28

Page 7 of 28

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

ACS Chemical Biology

Results and Discussion Ligand-induced biased signaling through GPCRs is a growing research area with potential advantages anticipated from fine-tuning compounds to modulate one intracellular signaling pathway associated with disease, without affecting other signaling through the same receptor and associated with normal physiology31. Changes to specific amino acids within a GPCR protein, known as PAR2, have previously been reported to differentially affect Ca2+, ERK1/2 or cAMP signaling linked to this receptor, although synthetic PAR2 agonist peptides have not yet been extensively investigated for differential effects on signaling pathways8, 32. In the present study, structure-activity relationships have been investigated in small synthetic agonists to identify possible ligand-induced PAR2-mediated biased signaling in CHO-hPAR2 cells, since untransfected CHO cells do not express PAR26. Importantly, the four key chemical biology tools in this study (Isox-Cha-Chg-AlaArg-NH2 (AY254), Isox-Cha-Chg-NH2 (AY77), 2-furoyl-Leu-Ala-Ala-Ala-Ala-IleNH2 (DF253), 2-furoyl-Leu-Ile-Gly-Arg-Leu-NH2) were PAR2-selective agonists that did not activate Ca2+ release or ERK1/2 phosphorylation in CHO cells (Supplementary Figure S1), unless they were transfected with human PAR2 (Figure 1 and 2). Alanine-scanning mutagenesis is a common strategy for modifying peptides because alanine (Ala) has the smallest side chain that has minimalist effects on peptide conformation

33

. Based on 2f-LIGRL-NH2, which is widely used by

researchers as a substitute for proteases in activating PAR2, alanine scans were conducted alone (Table 1), and in conjunction with other changes (Table 2) based on our previous studies

8, 34, 35

. Compounds were compared for PAR2 affinity, via

6

ACS Paragon Plus Environment

ACS Chemical Biology

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

competitive fluorescent ligand binding using Europium-labeled 2f-LIGRLO-NH2, for Ca2+ release, and for ERK1/2 phosphorylation. In addition, the agonist bias between Ca2+ and ERK signaling was distinguished by calculating a bias factor. According to the operational model of agonism, the symbol τ denotes the signaling efficacy of the agonist and is defined as the transducer constant. KA represents the equilibrium association constant of a compound. Log τ/KA is a transduction coefficient, which is a composite parameter used to indicate bias for a particular signaling pathway since either a selective affinity (KA) of an agonist to a given receptor and/or a differential efficacy (τ) toward specific pathways can lead to biased agonism 36-39. Further, to eliminate cell system-dependent factors between different pathways, the log (τ/KA) value was normalized by the activity of a reference peptide (2f-LIGRL-NH2) to give Δ log τ/KA (Δ log(τ/KA) = log(τ/KA)compound -log(τ/KA)2f-LIGRL-NH2). To compare Ca2+ and pERK1/2 signaling pathways, ΔΔ log τ/KA was also calculated for the difference in Δ log τ/KA between the two signaling pathways (ΔΔ log(τ/KA )= Δ log(τ/KA )Ca2+-Δ log(τ/KA)ERK1/2) , and ultimately 10 ΔΔ log τ/KA was identified as the bias factor 36-39. Ca2+ versus ERK1/2 signaling altered by alanine mutagenesis in 2f-LIGRL-NH2 Table 1 shows that the first two N-terminal residues of 2f-LIGRL-NH2 were vital for affinity and activity, with leucine (L) at position 2 being crucial for both calcium and ERK1/2 signaling. This is attributed to the bulky isobutyl sidechain of leucine fitting an important binding pocket in PAR2. Table 1 compares 2f-LIGRLNH2 with all six possible single alanine mutants for binding affinity, Ca2+ mobilization and ERK1/2 phosphorylation. The 2f and L at positions 1 and 2 were crucial for affinity, I and R residues at positions 3 and 4 were somewhat important,

7

ACS Paragon Plus Environment

Page 8 of 28

Page 9 of 28

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

ACS Chemical Biology

but G and L at positions 5 and 6 were no more important than alanine. For inducing Ca2+ release, the order of importance was position 2 > 1 > 3 = 4 = 5, while the terminal leucine could be replaced by alanine with a 10-fold gain in function. On the other hand, for inducing ERK1/2 phosphorylation, the order was 2 > 5 > 1 = 3 > 4 = 6 which had no effect. For double alanine substitutions, the order was 1, 2 > 5, 6 >> 3, 4. Binding affinity correlated better with ERK1/2 than Ca2+ signaling (Supplementary Figure S2). Importantly, alanine substitution of arginine at position 5 significantly reduced ERK1/2 phosphorylation (pEC50 5.0 ± 0.3) and binding affinity (pIC50 4.8 ± 0.2) compared to 2f-LIGRL-NH2 (pEC50 6.4 ± 0.1, pIC50 6.5 ± 0.1), but Ca2+ release was not much affected. The peptide 2f-LIGAL-NH2 had ΔΔ log τ/KA = 1.0 ± 0.3, the positive number indicating a bias toward Ca2+ signaling pathway (Figure 1f). Because affinity for PAR2 was reduced, isoleucine (I) at position 3 also impacted the ERK1/2 pathway, but 2f-LAGRL-NH2 had a smaller bias (ΔΔ log τ/KA = 0.6 ± 0.2) than 2fLIGAL-NH2. In contrast, alanine substitution of glycine (G) at position 4 caused no change in PAR2 affinity, but 2f-LIARL-NH2 containing arginine reversed this bias slightly toward ERK1/2 (ΔΔ log τ/KA = -0.2 ± 0.1), while leucine at position 6 was not important for Ca2+ or ERK1/2 activity. Since both isoleucine and arginine affected ERK1/2 activation, it was not surprising that the double mutant (2f-LAGAL-NH2) could not stimulate ERK1/2 signaling. Binding affinity decreased more in the absence of R5 and L6 (2f-LIGAANH2, pIC50 4.7 ± 0.2) than I3 and G4 (2f-LAARL-NH2, pIC50 5.7 ± 0.2). 2f-LAARLNH2 without isoleucine stimulated weak ERK1/2 activation and was biased toward Ca2+ (ΔΔ log τ/KA = 0.7 ± 0.1), whereas 2f-LIGAA-NH2 without arginine completely

8

ACS Paragon Plus Environment

ACS Chemical Biology

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

lost ERK1/2 potency (Δ log (τ/KA)

ERK1/2

ND and ΔΔ log τ/KA ND), indicating

extreme bias toward the Ca2+ pathway. Due to the importance of 2f-L, AAIGRL-NH2 did not induce any signaling at all. Thus, 2f-L was retained and I-G-R-L was replaced with alanine to give 2fLAAAA-NH2 (Table 1). It has been reported that SLAAAA-NH2, with four alanine substitutions relative to SLIGRL-NH2, partially induced MAPK activation but not Ca2+ release in KNRK cells 24. In contrast, the weakly binding 2f-LAAAA-NH2 was extremely biased toward the Ca2+ pathway in our study using CHO-hPAR2 cells (Δ log (τ/KA)

ERK1/2

ND and ΔΔ log τ/KA ND), with no ERK1/2 phosphorylation and this

was similar for 2f-LIGAA-NH2. These results are consistent with arginine at position 5 being extremely important in 2f-LIGRL-NH2 for PAR2-dependent ERK1/2 signaling.

Calcium signaling bias for DF253 and AY77 Although alanine scans had identified three PAR2 ligands with Ca2+ bias (2fLIGAL-NH2, 2f-LIGAA-NH2, 2f-LAAAA-NH2), all had very weak binding affinity. Our previous studies showed that extending the peptide length can improve PAR2 affinity28. Therefore, isoleucine was added to the C-terminus to produce 2f-LIGAAINH2 and 2f-LAAAAI-NH2 (DF253), which indeed enhanced PAR2 binding by ~10 fold (pIC50 5.4 ± 0.2 and 5.3 ± 0.3, respectively). Although the former began to induce ERK1/2 activation, it was still biased toward Ca2+ (ΔΔ log τ/KA = 0.5 ± 0.1). On the other hand, DF253 had better affinity and Ca2+ release (pEC50 5.7 ± 0.1) without ERK1/2 activation, so it was a more potent biased ligand (Δ log (τ/KA) ERK1/2 ND and ΔΔ log τ/KA ND , indicating extreme bias towards Ca2+) than 2f-LIGAL-NH2 and 2f-

9

ACS Paragon Plus Environment

Page 10 of 28

Page 11 of 28

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

ACS Chemical Biology

LIGAA-NH2 (Figure 1d). Therefore, DF253 reveals that extending the peptide length without adding arginine can promote bias for calcium signaling over ERK1/2, and could conceivably guide discovery of antagonists of PAR2-dependent ERK1/2 signaling. An alternative to S-L or 2f-L at positions 1 and 2 was Isox-Cha which was found to aid PAR2 affinity and selectivity in our previous study6. The two-residue peptide 2f-L-NH2 was not potent enough to stimulate any PAR2 activation, but the analogue Isox-Cha-NH2 bound weakly to PAR2 (pIC50 4.3 ± 0.3) and induced some Ca2+ (pEC50 4.8 ± 0.1) although no ERK phosphorylation in CHO-hPAR2 cells (Tables 1 & 2). Recently, we found that adding cyclohexylglycine to the C-terminus to give IsoxCha-Chg-NH2 (AY77) substantially increased Ca2+ stimulation in CHO-hPAR2 cells while maintaining PAR2 selectivity, including selectivity for PAR2 over PAR1

40

.

AY77 has been found here (Figure 2, Table 1) to be a biased agonist for Ca2+ (pEC50 7.5 ± 0.3) over ERK1/2 (pEC50 5.6 ± 0.2) signaling. AY77 is thus a valuable PAR2 agonist with a bias toward Ca2+ over ERK1/2 signaling (ΔΔ log τ/KA = 1.5 ± 0.2). It did not compete as well with europium-labeled 2f-LIGRLO-NH2 for binding (pIC50 5.7 ± 0.2) to PAR2, consistent with no competition with arginine in that peptide.

ERK1/2 signaling bias for AY254 Structure-activity relationships on peptides containing Isox-Cha were also investigated to identify bias toward the ERK1/2 pathway. Adding amino acids to Isox-Cha significantly increased PAR2 affinity and signaling responses. Isox-ChaAAA-NH2 was a more potent Ca2+-activating agonist than Isox-Cha-AGA-NH2, suggesting that alanine was more tolerated than glycine at position 4. However, unlike

10

ACS Paragon Plus Environment

ACS Chemical Biology

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 12 of 28

2f-L-AAAA-NH2, Isox-Cha-AAAA-NH2 was not a biased agonist, almost equally activating Ca2+ (pEC50 6.0 ± 0.1) and ERK1/2 (pEC50 6.4 ± 0.1), like 2f-LIGRL-NH2. When glycine was added to Isox-Cha-Chg-NH2 (AY77), shown above to be biased for Ca2+ over ERK1/2, the compound Isox-Cha-Chg-G-NH2 improved receptor affinity and ERK1/2 activation, but there was still a bias to Ca2+ signaling (ΔΔ log (τ/KA)Ca2+ = 1.2 ± 0.1, Figure 2f). Addition of arginine to the fifth position (Isox-ChaChg-G-R-NH2) switched the bias towards ERK1/2 (ΔΔ log (τ/KA)

ERK1/2

= 1.7 ± 0.4).

Substituting glycine with alanine in 2f-LIGRL-NH2 biased signaling towards ERK1/2 (Table 1), so the same replacement was made here (Isox-Cha-Chg-A-R-NH2, named AY254). AY254 showed improved PAR2 affinity (pIC50 7.9 ± 0.2), Ca2+ release (pEC50 7.4 ± 0.1) and ERK1/2 phosphorylation (pEC50 8.7 ± 0.2), leading to slightly enhanced bias to ERK1/2 (ΔΔ log (τ/KA)ERK1/2 = 1.8 ± 0.1). Binding affinity correlated better with ERK1/2 than with Ca2+ for Isox-Cha derived peptides (Supplementary Figure S2). Thus, AY254 is the first potent ERK1/2 activating PAR2 agonist that is biased away from Ca2+ and shows a distinct difference from the widely used PAR2 agonist 2f-LIGRLO-NH2 41.

PAR2 ligands attenuated cytokine-induced cleavage of caspases 3 and 8 in colorectal carcinoma cells via ERK1/2 signaling PAR2 is implicated in colon cancer

42,43

and is highly expressed in HT-29

colorectal carcinoma cells. PAR2 activation has been reported to inhibit TNF-α and IFN-γ-induced cleavage of caspases in a MEK1/2 and PI3K-dependent manner, implying that PAR2 may have anti-apoptotic effects in colon cancer 44. ERK1/2 is involved in cell proliferation, cell survival and cytokine release in colon cancer

11

ACS Paragon Plus Environment

44-46

.

Page 13 of 28

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

ACS Chemical Biology

Our discovery of Ca2+-biased PAR2 agonists (DF253, AY77) and an ERK1/2-biased agonist (AY254) provides valuable new chemical probes relative to unbiased agonists (SLIGRL-NH2, 2f-LIGRL-NH2) for dissecting mechanistic roles of PAR2 signaling in cellular functions. When IFN-γ and TNF-α were added to HT-29 cells, apoptosis was induced and cleaved caspases 3 and 8 were measured by western blot (Supplementary Figure S3). The unbiased agonist 2f-LIGRL-NH2 (1 µM) attenuated caspase cleavage (Figure 3a, 3b). Agonists without arginine at position 5 (2f-LIGAR-NH2, 2f-LIGAA-NH2, DF253), that did not strongly activate ERK1/2 at 10 µM (Figure 1), failed to block caspase cleavages. Consistent with this finding, other alanine mutated compounds that were similarly unable to potently activate ERK1/2 at 10 µM were also unable to prevent cytokine-induced cleavage of caspases (Supplementary Figure S4). The different Ca2+-biased PAR2 agonist AY77 (1 µM) that induced weak ERK1/2 signaling, also failed to reduce caspases 3 and 8 cleavages in cells pre-treated with IFN-γ and TNF-α. On the other hand, the potent ERK-activating PAR2 agonist AY254 did inhibit the cleavages. Although a biased ERK1/2 activator, AY254 at low concentrations was still effective in preventing apoptosis, as observed by reduction of cleaved caspase 3/8, indicating it is more potent than 2f-LIGRL-NH2 (Supplementary Figure S5). These results support an association between PAR2-activation of ERK1/2 and cytokine-induced caspase cleavage, which can be selectively induced by AY254. Further experiments using known signaling pathway inhibitors supported this mechanistic association between PAR2, ERK and cytokine-induced cleavage of caspases 3 and 8. The MEK inhibitor (U0126), more specific ERK1/2 inhibitor (FR180204), and calcium inhibitor (BAPTA), were each used to probe this

12

ACS Paragon Plus Environment

ACS Chemical Biology

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 14 of 28

mechanism. Although the inhibitors could slightly affect western blot of cleaved caspases (Supplementary Figure S3 and S7), adding inhibitors of MEK, ERK1/2 or both kinases abolished 2f-LIGRL-NH2 induced inhibition of caspase 3/8 cleavages, while the calcium inhibitor had no effect (Figures 3e, 3f). Thus, PAR2 biased ligands that stimulate ERK1/2 signaling would be expected to inhibit cytokine-induced caspase cleavage, and associated apoptosis in HT-29 cells.

PAR2 agonists promote scratch-wound healing in colorectal carcinoma cells through ERK1/2 activation. The in vitro scratch assay creates a gap, into which cells migrate and gap size is measured. PAR2 ligands that promote cell migration and proliferation might have benefits in wound healing. Tissue factor (TF) promotes scratch-wound healing through PAR2 activation in myoblasts and human glioma

47, 48

. Here, PAR2 biased

peptides were compared for their capacity to stimulate HT-29 cell migration into the scratch gap, using FBS as a positive control. PAR2 activating ligands promote migration of HT-29 cells over 48 h (Figures 4a and 4b). Compared with 82% reduction in the scratch gap by FBS, 2f-LIGRL-NH2 narrowed the scratch gap to ~56%, the more potent ERK1/2 activator AY254 was more effective (~67%), while AY77 only slightly triggered gap closure (~30%) which was not significant compared to control. Similarly, DF253 failed to induce much cell movement (~10%), indicating ERK1/2 may be important in this assay. ERK1/2 signaling has been suggested to modulate myoblasts in scratch-wound healing 47, but its role was not investigated in colonocytes.

13

ACS Paragon Plus Environment

Page 15 of 28

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

ACS Chemical Biology

To support the importance of ERK1/2 in this assay, known pathway inhibitors were used to probe 2f-LIGRL-NH2 induced cell migration. Inhibitors of MEK, ERK1/2 or both almost abolished the effect of 2f-LIGRL-NH2 in HT-29 cells with the scratch gap size significantly less (80%) than for 2f-LIGRL-NH2 alone (*** p < 0.001). The calcium inhibitor had no effect (Fig. 4C). These results indicated that PAR2 agonist-promoted migration of HT-29 cells was ERK1/2-dependent, such activation may be important in wound healing during colitis and colorectal carcinoma.

PAR2 induced ERK1/2 signaling in IL-8 secretion in HT-29 cells. Interleukin-8 (IL-8) is a major chemokine that activates neutrophils in colorectal carcinoma and modulates metastasis, proliferation and angiogenesis in colonic cancer

49, 50

. A MEK inhibitor has been reported to block PAR2-induced IL-8

secretion in HT-29 cells, suggesting a mechanism involving ERK1/2 signaling 46. The potent ERK-activating PAR2 agonist, AY254 (1 μM), was found to trigger strong IL8 production in HT29 cells, while 2f-LIGRL-NH2 at a higher concentration (10 μM) also stimulated IL-8 secretion (Figure 5a). In contrast, the Ca2+-biased PAR2 agonists AY77 and DF253 (at 10 μM) did not significantly promote IL-8 secretion in HT-29. These findings are consistent with PAR2 activation stimulating IL-8 release in HT-29 cells via an ERK1/2-dependent pathway 49, 50. MEK, ERK1/2 and calcium inhibitors were used to dissect involvement of PAR2-dependent ERK1/2 signaling in IL-8 production in these cells. Figure 5b shows that Ca2+ induction was not involved, whereas the ERK1/2 inhibitor alone or in combination with the MEK inhibitor inhibited IL-8 induced by PAR2 agonist 2fLIGRL-NH2.

14

ACS Paragon Plus Environment

ACS Chemical Biology

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Thus, two different approaches, one using biased PAR2 compounds and the other using pathway inhibitors, have indicated that cytokine-induced caspase cleavage, cell migration scratch-wound healing, and IL-8 cytokine release, were all dependent in HT-29 cells on ERK1/2 signaling. The results suggest that biased agonists that activate (or inhibit) PAR2-ERK1/2 signaling could be valuable tools to study PAR2 in colon function and colorectal disease. In conclusion, residues at the first (2f), second (L) and fifth (R) positions of 2fLIGRL-NH2 were key molecular determinants of PAR2-mediated signaling bias. For example, 2f-LAAAAI-NH2 (DF253) with no arginine preferentially activated the Ca2+ pathway. Similarly, Isox-Cha-Chg-NH2 (AY77), corresponding to the first three residues 2f-L-I and reported by us to be a PAR2-selective agonist for inducing Ca2+ release,40 is shown here to be far more effective in activating Ca2+ than ERK1/2. This is consistent with just three residues directing PAR2-mediated Gq coupling to induce intracellular Ca2+. Importantly, adding alanine and arginine to AY77 (Isox-Cha-ChgA-R-NH2, AY254) induced potent ERK1/2 signaling at much lower concentrations than for inducing Ca2+. This ligand-induced PAR2-mediated biased signaling in CHOhPAR2 cells was also observed for HT-29 cells (which do not express PAR1), where it translated into functional bias with activation of three distinct PAR2-mediated responses (cleavage of caspases 3 and 8 and attenuation of apoptosis, induction of cell migration on a PAR2 agonist gradient, and promotion of IL-8 secretion). For example, the ERK1/2-activating agonist AY254, but not the Ca2+-activating agonists DF253 or AY77, attenuated cytokine-induced apoptosis, promoted in vitro cell migration in scratch-wound healing, and induced IL-8 release in human HT29 cells in a PAR2ERK1/2 dependent manner (Figure 6). These newly developed PAR2 biased ligands

15

ACS Paragon Plus Environment

Page 16 of 28

Page 17 of 28

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

ACS Chemical Biology

(DF253, AY77, AY254) provide valuable tools for interrogating PAR2-dependent signaling and cell functions in physiology and disease.

16

ACS Paragon Plus Environment

ACS Chemical Biology

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Methods Details of experimental methods are provided in Supporting Information. Associated Content Supporting Figures S1-S7 and compound characterization. Acknowledgments We thank National Health and Medical Research Council for SPR Fellowships (1027369, 1117017) and grants (1084083, 1047759) and the Australian Research Council for grants (DP130100629, CE140100011) to DP Fairlie; the Chinese Scholarship Council for an award to Y Jiang; W Xu (University of Queensland) for help in analyzing data; and the Australian Cancer Research Foundation (ACRF) for funding a Cancer Biology Imaging Facility (Brisbane, Qld, Australia) which provided access to microscopes.

17

ACS Paragon Plus Environment

Page 18 of 28

Page 19 of 28

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

ACS Chemical Biology

References 1. Cenac, N., Coelho, A. M., Nguyen, C., Compton, S., Andrade-Gordon, P., MacNaughton, W. K., Wallace, J. L., Hollenberg, M. D., Bunnett, N. W., GarciaVillar, R., Bueno, L., and Vergnolle, N. (2002) Induction of intestinal inflammation in mouse by activation of proteinase-activated receptor-2, Am. J. Pathol. 161, 19031915. 2. Corvera, C. U., Déry, O., McConalogue, K., Böhm, S. K., Khitin, L. M., Caughey, G. H., Payan, D. G., and Bunnett, N. W. (1997) Mast cell tryptase regulates rat colonic myocytes through proteinase-activated receptor 2, J. Clin. Investi. 100, 13831393. 3. Takeuchi, T., Harris, J. L., Huang, W., Yan, K. W., Coughlin, S. R., and Craik, C. S. (2000) Cellular localization of membrane-type serine protease 1 and identification of protease-activated receptor-2 and single-chain urokinase-type plasminogen activator as substrates, J. Biol. Chem. 275, 26333-26342. 4. Zhao, P., Lieu, T., Barlow, N., Sostegni, S., Haerteis, S., Korbmacher, C., Liedtke, W., Jimenez-Vargas, N. N., Vanner, S. J., and Bunnett, N. W. (2015) Neutrophil Elastase Activates Protease-activated Receptor-2 (PAR2) and Transient Receptor Potential Vanilloid 4 (TRPV4) to Cause Inflammation and Pain, J. Biol. Chem. 290, 13875-13887. 5. Zhao, P., Lieu, T., Barlow, N., Metcalf, M., Veldhuis, N. A., Jensen, D. D., Kocan, M., Sostegni, S., Haerteis, S., Baraznenok, V., Henderson, I., Lindstrom, E., Guerrero-Alba, R., Valdez-Morales, E. E., Liedtke, W., McIntyre, P., Vanner, S. J., Korbmacher, C., and Bunnett, N. W. (2014) Cathepsin S causes inflammatory pain via biased agonism of PAR2 and TRPV4, J. Biol. Chem. 289, 27215-27234. 6. Suen, J. Y., Cotterell, A., Lohman, R. J., Lim, J., Han, A., Yau, M. K., Liu, L., Cooper, M. A., Vesey, D. A., and Fairlie, D. P. (2014) Pathway-selective antagonism of proteinase activated receptor 2, Br. J. Pharmacol. 171, 4112-4124. 7. Adams, M. N., Ramachandran, R., Yau, M. K., Suen, J. Y., Fairlie, D. P., Hollenberg, M. D., and Hooper, J. D. (2011) Structure, function and pathophysiology of protease activated receptors, Pharmacol. Ther. 130, 248-282. 8. Maryanoff, B. E., Santulli, R. J., McComsey, D. F., Hoekstra, W. J., Hoey, K., Smith, C. E., Addo, M., Darrow, A. L., and Andrade-Gordon, P. (2001) Proteaseactivated receptor-2 (PAR-2): structure-function study of receptor activation by diverse peptides related to tethered-ligand epitopes, Arch. Biochem. Biophys. 386, 195-204. 9. Afkhami-Goli, A., Noorbakhsh, F., Keller, A. J., Vergnolle, N., Westaway, D., Jhamandas, J. H., Andrade-Gordon, P., Hollenberg, M. D., Arab, H., Dyck, R. H., and Power, C. (2007) Proteinase-activated receptor-2 exerts protective and pathogenic cell type-specific effects in Alzheimer's disease, J. Immunol. 179, 5493-5503. 10. Rothmeier, A. S., and Ruf, W. (2012) Protease-activated receptor 2 signaling in inflammation, Sem. immunopathol. 34, 133-149. 11. Hollenberg, M. D., Mihara, K., Polley, D., Suen, J. Y., Han, A., Fairlie, D. P., and Ramachandran, R. (2014) Biased signalling and proteinase-activated receptors (PARs): targeting inflammatory disease, Br. J. Pharmacol. 171, 1180-1194. 12. Lim, J., Iyer, A., Liu, L., Suen, J. Y., Lohman, R. J., Seow, V., Yau, M. K., Brown, L., and Fairlie, D. P. (2013) Diet-induced obesity, adipose inflammation, and metabolic dysfunction correlating with PAR2 expression are attenuated by PAR2 antagonism, FASEB. J. 27, 4757-4767.

18

ACS Paragon Plus Environment

ACS Chemical Biology

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

13. Jahan, I., Fujimoto, J., Alam, S. M., Sato, E., Sakaguchi, H., and Tamaya, T. (2007) Role of protease activated receptor-2 in tumor advancement of ovarian cancers, Ann. Oncol. 18, 1506-1512. 14. Zhou, B., Zhou, H., Ling, S., Guo, D., Yan, Y., Zhou, F., and Wu, Y. (2011) Activation of PAR2 or/and TLR4 promotes SW620 cell proliferation and migration via phosphorylation of ERK1/2, Oncol. Rep. 25, 503-511. 15. Antoniak, S., Rojas, M., Spring, D., Bullard, T. A., Verrier, E. D., Blaxall, B. C., Mackman, N., and Pawlinski, R. (2010) Protease-activated receptor 2 deficiency reduces cardiac ischemia/reperfusion injury, Arterioscler. Thromb. Vasc. Biol. 30, 2136-2142. 16. Galandrin, S., Oligny-Longpré, G., and Bouvier, M. (2007) The evasive nature of drug efficacy: implications for drug discovery, Trends in pharmacological sciences 28, 423-430. 17. Violin, J. D., and Lefkowitz, R. J. (2007) β-Arrestin-biased ligands at seventransmembrane receptors, Trends in pharmacological sciences 28, 416-422. 18. Rajagopal, S., Rajagopal, K., and Lefkowitz, R. J. (2010) Teaching old receptors new tricks: biasing seven-transmembrane receptors, Nature reviews. Drug discovery 9, 373-386. 19. Evans, B. A., Sato, M., Sarwar, M., Hutchinson, D. S., and Summers, R. J. (2010) Ligand‐directed signalling at β‐adrenoceptors, British journal of pharmacology 159, 1022-1038. 20. Zhao, P., Metcalf, M., and Bunnett, N. W. (2013) Biased signaling of proteaseactivated receptors, Front. Endocrinol. 5, 67-67. 21. Rankovic, Z., Brust, T. F., and Bohn, L. M. (2016) Biased agonism: An emerging paradigm in GPCR drug discovery, Bioorg. Med. Chem. Lett. 26, 241-250. 22. Ramachandran, R., Mihara, K., Chung, H., Renaux, B., Lau, C. S., Muruve, D. A., DeFea, K. A., Bouvier, M., and Hollenberg, M. D. (2011) Neutrophil elastase acts as a biased agonist for proteinase-activated receptor-2 (PAR2), J. Biol. Chem. 286, 24638-24648. 23. Elmariah, S. B., Reddy, V. B., and Lerner, E. A. (2014) Cathepsin S Signals via PAR2 and Generates a Novel Tethered Ligand Receptor Agonist, PloS one 9, e99702. 24. Ramachandran, R., Mihara, K., Mathur, M., Rochdi, M. D., Bouvier, M., DeFea, K., and Hollenberg, M. D. (2009) Agonist-biased signaling via proteinase activated receptor-2: Differential activation of calcium and mitogen-activated protein kinase pathways, Mol. Pharmacol. 76, 791-801. 25. Al-Ani, B., and Hollenberg, M. D. (2003) Selective tryptic cleavage at the tethered ligand site of the amino terminal domain of proteinase-activated receptor-2 in intact cells, J. Pharmacol. Exp. Ther. 304, 1120-1128. 26. Lau, C., Lytle, C., Straus, D. S., and DeFea, K. A. (2011) Apical and basolateral pools of proteinase-activated receptor-2 direct distinct signaling events in the intestinal epithelium, Am. J. Physiol. - Cell Physiol. 300, C113-C123. 27. Ferrell, W. R., Lockhart, J. C., Kelso, E. B., Dunning, L., Plevin, R., Meek, S. E., Smith, A. J., Hunter, G. D., McLean, J. S., McGarry, F., Ramage, R., Jiang, L., Kanke, T., and Kawagoe, J. (2003) Essential role for proteinase-activated receptor-2 in arthritis, J. Clin. Invest. 111, 35-41. 28. Barry, G. D., Suen, J. Y., Low, H. B., Pfeiffer, B., Flanagan, B., Halili, M., Le, G. T., and Fairlie, D. P. (2007) A refined agonist pharmacophore for protease activated receptor 2, Bioorg. Med. Chem. Lett. 17, 5552-5557. 29. Kawabata, A., Kanke, T., Yonezawa, D., Ishiki, T., Saka, M., Kabeya, M., Sekiguchi, F., Kubo, S., Kuroda, R., Iwaki, M., Katsura, K., and Plevin, R. (2004) 19

ACS Paragon Plus Environment

Page 20 of 28

Page 21 of 28

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

ACS Chemical Biology

Potent and metabolically stable agonists for protease-activated receptor-2: evaluation of activity in multiple assay systems in vitro and in vivo, J. Pharmacol. Exp. Ther. 309, 1098-1107. 30. Kanke, T., Ishiwata, H., Kabeya, M., Saka, M., Doi, T., Hattori, Y., Kawabata, A., and Plevin, R. (2005) Binding of a highly potent protease-activated receptor-2 (PAR2) activating peptide, [3H]2-furoyl-LIGRL-NH2, to human PAR2, Br. J. Pharmacol. 145, 255-263. 31. Violin, J. D., Crombie, A. L., Soergel, D. G., and Lark, M. W. (2014) Biased ligands at G-protein-coupled receptors: promise and progress, Trends. Pharmacol. Sci. 35, 308-316. 32. Al-Ani, B., Hansen, K. K., and Hollenberg, M. D. (2004) Proteinase-activated receptor-2: key role of amino-terminal dipeptide residues of the tethered ligand for receptor activation, Mol. Pharmacol. 65, 149-156. 33. Lefèvre, F., Rémy, M.-H., and Masson, J.-M. (1997) Alanine-stretch scanning mutagenesis: a simple and efficient method to probe protein structure and function, Nucleic. Acid. Res. 25, 447-448. 34. Barry, G. D., Suen, J. Y., Le, G. T., Cotterell, A., Reid, R. C., and Fairlie, D. P. (2010) Novel agonists and antagonists for human protease activated receptor 2, J. Med. Chem. 53, 7428-7440. 35. Suen, J. Y., Barry, G. D., Lohman, R. J., Halili, M. A., Cotterell, A. J., Le, G. T., and Fairlie, D. P. (2012) Modulating human proteinase activated receptor 2 with a novel antagonist (GB88) and agonist (GB110), Br. J. Pharmacol. 165, 1413-1423. 36. Black, J., and Leff, P. (1983) Operational models of pharmacological agonism, Proc. Royal. Soc. Lond. B: Biol. Sci. 220, 141-162. 37. Weichert, D., Banerjee, A., Hiller, C., Kling, R. C., Hübner, H., and Gmeiner, P. (2015) Molecular Determinants of Biased Agonism at the Dopamine D2 Receptor, J. Med. Chem. 58, 2703-2717. 38. Shonberg, J., Herenbrink, C. K., López, L., Christopoulos, A., Scammells, P. J., Capuano, B., and Lane, J. R. (2013) A structure–activity analysis of biased agonism at the dopamine D2 receptor, J. Med. Chem. 56, 9199-9221. 39. Weiss, D. R., Ahn, S., Sassano, M. F., Kleist, A., Zhu, X., Strachan, R., Roth, B. L., Lefkowitz, R. J., and Shoichet, B. K. (2013) Conformation guides molecular efficacy in docking screens of activated β-2 adrenergic G protein coupled receptor, ACS Chem. Biol. 8, 1018-1026. 40. Yau, M.-K., Suen, J. Y., Xu, W., Lim, J., Liu, L., Adams, M. N., He, Y., Hooper, J. D., Reid, R. C., and Fairlie, D. P. (2015) Potent Small Agonists of Protease Activated Receptor 2, ACS Med. Chem. Lett. 7, 105-110. 41. Barry, G. D., Suen, J. Y., Le, G. T., Cotterell, A., Reid, R. C., and Fairlie, D. P. (2010) Novel agonists and antagonists for human protease activated receptor 2, J. Med. Chem. 53, 7428-7440. 42. Elste, A. P., and Petersen, I. (2010) Expression of proteinase-activated receptor 14 (PAR 1-4) in human cancer, J. Mol. Histol. 41, 89-99. 43. Darmoul, D., Marie, J., Devaud, H., Gratio, V., and Laburthe, M. (2001) Initiation of human colon cancer cell proliferation by trypsin acting at protease-activated receptor-2, Br. J. Cancer. 85, 772-779. 44. Iablokov, V., Hirota, C. L., Peplowski, M. A., Ramachandran, R., Mihara, K., Hollenberg, M. D., and MacNaughton, W. K. (2014) Proteinase-activated receptor 2 (PAR2) decreases apoptosis in colonic epithelial cells, J. Biol. Chem. 289, 3436634377.

20

ACS Paragon Plus Environment

ACS Chemical Biology

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

45. Darmoul, D., Gratio, V., Devaud, H., and Laburthe, M. (2004) Protease-activated receptor 2 in colon cancer - Trypsin-induced MAPK phosphorylation and cell proliferation are mediated by epidermal growth factor receptor transactivation, J. Biol. Chem. 279, 20927-20934. 46. Wang, H., Moreau, F., Hirota, C. L., and MacNaughton, W. K. (2010) Proteinaseactivated receptors induce interleukin-8 expression by intestinal epithelial cells through ERK/RSK90 activation and histone acetylation, FASEB. J. 24, 1971-1980. 47. Borensztajn, K., Stiekema, J., Nijmeijer, S., Reitsma, P. H., Peppelenbosch, M. P., and Spek, C. A. (2008) Factor Xa stimulates proinflammatory and profibrotic responses in fibroblasts via protease-activated receptor-2 activation, Am. J. Pathol. 172, 309-320. 48. Gessler, F., Voss, V., Dutzmann, S., Seifert, V., Gerlach, R., and Kogel, D. (2010) Inhibition of tissue factor/protease-activated receptor-2 signaling limits proliferation, migration and invasion of malignant glioma cells, Neurosci. 165, 1312-1322. 49. Ning, Y., Manegold, P. C., Hong, Y. K., Zhang, W., Pohl, A., Lurje, G., Winder, T., Yang, D., LaBonte, M. J., and Wilson, P. M. (2011) Interleukin‐8 is associated with proliferation, migration, angiogenesis and chemosensitivity in vitro and in vivo in colon cancer cell line models, Int. J. Cancer. 128, 2038-2049. 50. Gross, V., Andus, T., Daig, R., Aschenbrenner, E., Schölmerich, J., and Falk, W. (1995) Regulation of interleukin-8 production in a human colon epithelial cell line (HT-29), Gastroenterol. 108, 653-661.

21

ACS Paragon Plus Environment

Page 22 of 28

Page 23 of 28

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

ACS Chemical Biology

Figure 1. Calcium versus ERK1/2 signaling by PAR2 activating peptides related to 2f-LIGRL-NH2 in CHO-hPAR2 cells. (A) Competitive binding by 2f-LIGRL-NH2 and alanine-substituted peptides versus 300 nM Eu-tagged 2f-LIGRLO-NH2. (B) Concentration-dependent Ca2+ signaling by PAR2 agonists. (C) Concentrationdependent ERK1/2 phosphorylation by PAR2 agonists. Each data point represents mean ± SEM (n ≥ 3). (D, E, F) Bias factor quantified by operational agonist model. (D, E) Δlog (τ/KA) is difference in transduction coefficient between compounds and 2f-LIGRL-NH2 in Ca2+ (D) and ERK (E) pathways. (F) ΔΔlog (τ/KA) is difference in transduction coefficient of compounds between two signaling pathways. ΔΔ log(τ/KA)=Δ log(τ/KA)Ca2+-Δ log(τ/KA)ERK1/2. Bias factor=10ΔΔ log(τ/KA). ND= no signal detected. Δ log(τ/KA)ERK1/2 ND and ΔΔlog (τ/KA) ND, indicating extreme bias toward Ca2+.

22

ACS Paragon Plus Environment

ACS Chemical Biology

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Figure 2. PAR2 agonists differentially activate Ca2+ vs ERK1/2 signaling in CHOhPAR2 cells. (A) Concentration-dependent competitive binding with Eu-tagged 2f-LIGRLO-NH2 (300 nM) of four PAR2 agonists. (B) Concentration-dependent Ca2+ release induced by PAR2 agonists. (C) Concentration-dependent ERK1/2 phosphorylation induced by PAR2 agonists. (n ≥ 3). (D, E, F) Bias factor quantified by operational model of agonism. (D, E) Δlog (τ/KA) is difference in transduction coefficient between compounds and 2f-LIGRL-NH2 in one pathway. (F) ΔΔlog (τ/KA) is difference in transduction coefficient between two pathways. Statistical analysis by one-way ANOVA followed by Dunnett’s multiple comparisons test, *** p < 0.001 (n ≥ 3).

23

ACS Paragon Plus Environment

Page 24 of 28

Page 25 of 28

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

ACS Chemical Biology

Figure 3. PAR2 agonist attenuation of cytokine-induced caspase cleavage in HT-29 cells depends on ERK1/2 signaling. (A, B) 2f-LIGRL-NH2 (1 μM) attenuated IFN-γ/TNF-α induced apoptosis over 6h in HT-29 cells by inhibiting cleavage of (A) caspase 3 and (B) caspase 8, whereas other peptides (10 μM) had no effect. *** p < 0.001 (n ≥ 3). (C, D) AY254 (1 μM) but not AY77 (1 μM) reduced cytokine-induced cleavage of caspases 3 and 8. *** p < 0.001 (n ≥ 3). (E, F) Effect of inhibitors of Ca2+ (BAPTA), MEK (U0126), ERK1/2 (FR180204), or MEK and ERK prior to adding 2f-LIGRL-NH2, on cytokine-induced cleavage of (E) caspase 3 and (F) caspase 8. *** p < 0.001 (n ≥ 3). 2f-LIGRL-NH2 is blue, 2f-LIGAL-NH2 is red, 2f-LIGAA-NH2 is purple, DF253 is green, AY77 is orange and AY254 is pink. “+” means

24

ACS Paragon Plus Environment

ACS Chemical Biology

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

addition of cytokines or PAR2 agonists whereas “-” indicates no addition. Panel E represents the effect of signaling pathway inhibitors that are not PAR2 ligands, so we use black. Only PAR2 agonists are shown colored. All panels are individual experiments. Statistical analysis by one-way ANOVA followed by Dunnett’s multiple comparisons test.

Figure 4. PAR2 agonists promote HT-29 cell migration in a scratch assay via ERK1/2 signaling. (A) PAR2 agonists stimulate ERK1/2 activation in HT-29 cells to induce cell migration. (B). Quantification of the scratch gap for HT-29 cells treated with 2f-LIGRL-NH2 (1 μM, blue), DF253 (10 μM, green), AY77 (10 μM, orange), AY254 (1 μM, pink) versus 10% FBS (positive control, red) and serum-free medium (negative control). % scratch gap size = (gap area at 48 h / initial gap area at 0 h) × 100% to compare effects of PAR2 agonists. (C) Inhibitors of MEK and ERK1/2 block cell migration by PAR2-activated HT-29 cells after 48h. Inhibitors of Ca2+ (10 μM BAPTA), MEK (10 μM U0126), ERK1/2 (10 μM FR180204), or MEK and ERK1/2, were added 30 min before scratch test. “+” indicates addition of 10% FBS or 2f-LIGRL-NH2 or inhibitors, whereas “-” indicates no addition. Statistical analysis by one-way ANOVA followed by Dunnett’s multiple comparisons test, *** p < 0.001(n ≥ 3).

25

ACS Paragon Plus Environment

Page 26 of 28

Page 27 of 28

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

ACS Chemical Biology

Figure 5. IL-8 secretion is stimulated by PAR2 agonists in HT-29 via ERK1/2 signaling. (A) Release of cytokine IL-8 after incubation of PAR2 agonists in HT-29 cells for 24 h. IL-8 production induced by 2f-LIGRL-NH2 (10 μM), AY254 (1 μM), AY77 (10 μM) or DF253 (10 μM) was detected by ELISA. (B) Inhibitors of MEK and ERK1/2 prevent PAR2-induced IL-8 secretion in HT-29 cells. Inhibitors (10 μM) of Ca2+ (BAPTA), MEK (U0126), ERK1/2 (FR180204), or both MEK and ERK1/2, were added 30 min prior to 2f-LIGRL-NH2. Statistical analysis by one-way ANOVA followed by Dunnett’s multiple comparisons test, *** p < 0.001(n ≥ 3).

26

ACS Paragon Plus Environment

ACS Chemical Biology

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Figure 6. Summary of PAR2 agonist structure-signaling-function relationships. The arrows indicate comparative relative potencies for inducing Ca2+ versus ERK signaling, and subsequent functional responses.

27

ACS Paragon Plus Environment

Page 28 of 28