Evidence for spatial proximity of two distinct receptor regions in the substance P (SP)*neurokinin-1 receptor (NK-1R) complex obtained by photolabeling the NK-1R with p-benzoylphenylalanine3-SP.

Substance P (SP) belongs to the tachykinin family of bioactive peptides and exerts its many biological effects through functional interaction with its cell-surface, G protein-coupled neurokinin-1 receptor (NK-1R). Previous studies from our laboratory have shown that (125)I-Bolton-Hunter reagent-labeled p-benzoylphenylalanine(8)-SP (Bpa(8)SP) covalently attaches to Met(181), whereas (125)I-Bolton-Hunter reagent-labeled Bpa(4)SP covalently attaches to Met(174), both of which are located on the second extracellular loop (EC2) of the NK-1R. In this study, evidence has been obtained that at equilibrium, the photoreactive SP analogue (125)I-[D-Tyr(0)]Bpa(3)SP covalently labels residues in two distinct extracellular regions of the NK-1R. One site of (125)I-[D-Tyr(0)]Bpa(3)SP photoinsertion is located on EC2 within a segment of the receptor extending from residues 173 to 177; a second site of (125)I-[D-Tyr(0)]Bpa(3)SP photoinsertion is located on the extracellular N terminus within a segment of the receptor extending from residues 11 to 21, a sequence that contains both potential sites for N-linked glycosylation. Since competition binding data presented in this study do not suggest the existence of multiple peptide.NK-1R complexes, it is reasonable to assume that the receptor sequences within EC2 and N terminus identified by peptide mapping are in close proximity in the equilibrium complex.

The undecapeptide substance P (SP) 1 belongs to the tachykinin family of bioactive peptides, which are structurally characterized by the conserved carboxyl-terminal sequence of -Phe-X-Gly-Leu-Met-NH 2 , where X is either a ␤-branched aliphatic (Val, Ile) or an aromatic (Phe, Tyr) amino acid residue (1)(2)(3)(4)(5). Synthesized and secreted from both neural and non-neural tissues, SP (H-Arg-Pro-Lys-Pro-Gln-Gln-Phe-Phe-Gly-Leu-Met-NH 2 ) participates in many physiological processes in the cardiovascular, respiratory, gastrointestinal, immune, and nervous systems (3,6,7). Considerable evidence exists that SP plays a role in pain modulation and neurogenic inflammation (6,7), and recent studies have associated this peptide in the pathogenesis of certain affective disorders (8). The many diverse biological effects of SP are mediated by the functional interaction of the peptide with its cell-surface G proteincoupled receptor, the neurokinin-1 receptor (NK-1R).
In our laboratory, we have utilized the biochemical approach of photoaffinity labeling using SP analogues containing the photoreactive amino acid p-benzoylphenylalanine (Bpa) in different peptide positions to identify contact sites between specific residues of SP and side chains of receptor residues (9,10). We have shown that 125 I-Bolton-Hunter reagent (BH)-labeled Bpa 8 SP covalently labels Met 181 (9), whereas 125 I-BH-Bpa 4 SP covalently labels Met 174 (10) of the rat NK-1R (rNK-1R), information that has been integral to the three-dimensional modeling of the SP⅐NK-1R complex (see the accompanying article (27)).
Maggio and co-workers (11) previously reported that a photoreactive analogue of SP in which the Bpa residue is substituted at position 3 ( 125 I-[D-Tyr 0 ]Bpa 3 SP) covalently labels a residue within the initial 21 amino acid residues of the N terminus of the NK-1R present in the murine P388D 1 cell line. We found this result intriguing in view of our finding that when Bpa is substituted at position 4 of the peptide, Met 174 on the second extracellular loop (EC2) is the site of covalent attachment (10). We therefore decided that it would be of interest to study the site of photoincorporation of 125 I-[D-Tyr 0 ]Bpa 3 SP into rat NK-1Rs expressed in stably transfected Chinese hamster ovary (CHO) cells.

EXPERIMENTAL PROCEDURES
Materials-Purified SP was purchased from Sigma, and [D-Tyr 0 ]Bpa 3 SP was purchased from Quality Controlled Biochemicals, Inc., following high performance liquid chromatography (HPLC) purification and mass spectrometric analysis. 125 I-Labeled Bolton-Hunter reagent and 125 I (each with a specific activity of 2200 Ci/mmol) were obtained from PerkinElmer Life Sciences, and 127 I-labeled Bolton-Hunter reagent was synthesized by B. Tomczuk (Eastman Kodak Co.).
Preparation of Iodinated Ligands-Purified SP was iodinated by coupling the Lys 3 residue ⑀-NH 2 group to the monoiodinated 125 Ilabeled Bolton-Hunter reagent (N-succinimidyl-3-(4-hydroxyphenyl) propionate) as described previously (12,13). Purified [D-Tyr 0 ]Bpa 3 SP was iodinated by coupling the N-terminal D-Tyr 0 residue to either 125 I or 127 I using the solid-phase oxidant 1,3,4,6-tetrachloro-3␣,6␣-diphenylglycouril (IODO-GEN ® iodination reagent, Pierce). Briefly, [D-Tyr 0 ]Bpa 3 SP was dissolved at room temperature in 0.1 M sodium borate buffer (pH 8.5) and incubated with 125 I or 127 I for 15 min at room temperature in IODO-GEN ® reagent-treated 10 ϫ 75-mm borosilicate culture tubes. Reversed-phase HPLC on a 0.1% (v/v) trifluoroacetic acid-equilibrated and derivatized silica gel C 18 column was then per-formed to separate the iodination reaction products using an acetonitrile/water/trifluoroacetic acid solvent system. The acetonitrile concentration in the eluant was raised by a gradient controller/pump system by 0.7%/min, and fractions were collected every minute at a flow rate of 1.5 ml/min. 127 I-[D-Tyr 0 ]Bpa 3 SP was identified by measuring the UV absorbance at 262 nm, and 125 I-[D-Tyr 0 ]Bpa 3 SP was identified by ␥-emission spectrometry. 20% (v/v) ␤-mercaptoethanol was added to both the 127 I-[D-Tyr 0 ]Bpa 3 SP and 125 I-[D-Tyr 0 ]Bpa 3 SP peptide fractions, and the samples were heated at 90°C for 2 h (to reduce the methionine sulfoxide on Met 11 of the peptides to its thioether form). Mass spectrometry was then used to confirm that the 127 I had coupled to the D-Tyr 0 residue of 127 I-[D-Tyr 0 ]Bpa 3 SP. Both 127 I-[D-Tyr 0 ]Bpa 3 SP and 125 I-[D-Tyr 0 ]Bpa 3 SP were shown to coelute in the reversed-phase HPLC system described above.
Equilibrium Displacement Competition Assay-rNK-1R-transfected CHO cells were harvested and resuspended in ice-cold KRH buffer (20 mM HEPES, 1 mM CaCl 2 , 2.2 mM MgCl 2 , 5 mM KCl, and 120 mM NaCl (pH 7.4)) supplemented with 6 mg/ml glucose and 0.6 mg/ml bovine serum albumin. Cells were incubated for 2 h at 4°C with the radiolabeled ligand 125 I-BH-SP, and binding was measured either alone or in the presence of increasing concentrations of unlabeled SP or [D-Tyr 0 ]Bpa 3 SP. In all experiments, nonspecific 125 I-BH-SP binding was defined as the binding in the presence of 1 M unlabeled SP. To separate bound ligand from free ligand, cells were filtered after incubation through Whatman GF/C filter paper (soaked Ͼ2 h in 0.1% polyethyleneimine) and washed three times in ice-cold KRH buffer (pH 7.4) with a Brandel Harvester apparatus. Bound radioactivity on the filters was then quantified by ␥-emission spectrometry. Competition assays were performed in triplicate and were repeated at least three times.
Cell Membrane Preparation-Membranes were prepared by collecting photolabeled cells through centrifugation and resuspension in Tris/ EDTA buffer (5 mM Tris and 1 mM EDTA (pH 7.4)) containing 0.1 mM phenylmethylsulfonyl fluoride. Cell mixtures were sonicated for 10 s twice with a Sonicator Cell Disrupter to ensure complete homogenization and centrifuged at 500 ϫ g for 10 min to remove nuclear and cellular debris. The remaining supernatants were then sedimented at 38,000 ϫ g for 1 h to collect the membrane pellets. Following the removal of noncovalently attached radioligand from the membranes by a hypertonic acid wash (0.2 M acetic acid and 0.5 M NaCl (pH 2.4)), the membranes were washed twice by resuspension and centrifugation in Tris/EDTA buffer (pH 7.4) and stored at Ϫ20°C.
Identification of the Photolabeled Receptor-125 I-[D-Tyr 0 ]Bpa 3 SP-labeled membranes were solubilized in sample buffer (0.125 M Tris, 2% SDS, 10% glycerol, and 0.01% bromphenol blue (pH 6.8)), heated at 55°C for 10 min, and then subjected to SDS-polyacrylamide gel electrophoresis (PAGE) as described by Laemmli (17). Following electrophoresis, the gels were dried on filter paper and exposed to x-ray film (Kodak XAR-5). Prestained molecular mass standards (14.3-220 kDa; Amersham Pharmacia Biotech) were used to determine the molecular mass of the radiolabeled complex.
Tryptic Digestion of Photolabeled Membranes-125 I-[D-Tyr 0 ]Bpa 3 SP-labeled membranes were resuspended in 0.1% SDS, 50 mM Tris, and 1 mM CaCl 2 (pH 8.0) and digested at room temperature for 2 h with the amount of L-1-tosylamido-2-phenylethyl chloromethyl ketone-treated bovine trypsin specified in the legend to Fig. 3. N ␣ -p-tosyl-L-lysine chloromethyl ketone was added to the reaction mixtures in a 1:1000 dilution following incubation to terminate enzymatic activity, and the samples were then added to sample buffer Ϯ 30 mM DL-dithiothreitol (DTT). The 125 I-[D-Tyr 0 ]Bpa 3 SP photoprobe itself is protected from tryptic cleavage under the conditions used. Tryptic cleavage fragments were then separated and analyzed using the Tricine gel system of SDS-PAGE (18). Prestained molecular mass standards (2.35-46 kDa; Amersham Pharmacia Biotech) were used to determine the molecular masses of the radiolabeled tryptic fragments. Endoglycosidase F Digestion of Photolabeled Tryptic Fragments-The radioactive bands of the limit tryptic fragments were passively eluted from macerated dried Tricine gel slices in extraction buffer (5 mM Tris and 1 mM EDTA (pH 8.0)) for 1-4 days at room temperature. The eluted radiolabeled limit tryptic fragments were then dried by speed vacuum; resuspended in 5 mM Tris, 1 mM EDTA, and 0.5% n-octyl ␣-D-glucopyranoside (pH 8.0); and digested with Flavobacterium meningosepticum endoglycosidase F (Roche Molecular Biochemicals) overnight at 37°C. Digestion products were analyzed with either the Tricine gel system of SDS-PAGE (18) or the NuPAGE 4 -12% BisTris gradient gel system (Novex) using MES running buffer. For preparative scale experiments, the radioactive N-terminal limit tryptic fragment was electroeluted from Tricine/SDS-polyacrylamide gels into extraction buffer as described above and then adsorbed to wheat germ agglutininagarose beads overnight at 4°C. The wheat germ agglutinin-agarose beads were then washed twice in 1% SDS, 5 mM Tris, and 1 mM EDTA (pH 8.0) and pelleted by centrifugation. The supernatants were transferred to Centricon-10 microconcentrators (Amicon, Inc.) and centrifuged at 5000 ϫ g for 1 h. The concentrates were washed in extraction buffer and reconcentrated twice before being dried by speed vacuum. Dried radiolabeled pellets were then resuspended in 5 mM Tris, 1 mM EDTA, and 0.5% n-octyl ␣-D-glucopyranoside (pH 8.0) and digested with endoglycosidase F overnight at 37°C as described above.
Endoproteinase Glu-C (Protease V8) Subcleavage of Photolabeled Tryptic Fragments-The radioactive bands of the limit tryptic fragments were passively eluted from macerated dried Tricine gel slices as described above and then subjected to subcleavage at a final concentration of 1 mg/ml endoproteinase Glu-C (protease V8, Worthington) overnight at 37°C. Note that in the TE buffer (5 mM Tris and 1 mM EDTA (pH 8.0)) used for these subcleavage experiments, protease V8 hydrolyzes peptide and ester bonds specifically at the carboxylic side of both Glu and Asp. Subcleavage products were analyzed with either the Tricine gel system of SDS-PAGE (18) or the NuPAGE 4 -12% BisTris gradient gel system using MES running buffer.
Disulfide Bond Reduction in Photolabeled Tryptic Fragments-The radioactive bands of the limit tryptic fragments were passively eluted from macerated dried Tricine gel slices as described above and then incubated with 30 mM DTT for 1 h at room temperature. The DTTtreated labeled fragments were analyzed using the Tricine gel system of SDS-PAGE (18).
Tryptic This confirms the presence of a disulfide bond in these extended fragments of the rNK-1R, which has previously been shown to exist between Cys 105 and Cys 180 , linking the first and second extracellular domains (15).
Peptide mapping analysis using the theoretical tryptic fragmentation restriction map of the rNK-1R suggests that the ϳ40-kDa tryptic fragment observed under both nonreducing and reducing conditions corresponds to the 125 I-[D-Tyr 0 ]Bpa 3 SP photoligand (calculated molecular mass of 1.8 kDa) covalently attached to an amino acid within the segment of the rNK-1R extending from residues 1 to 61 (calculated molecular mass of 6.8 kDa ϩ ϳ30-kDa carbohydrate residues). This region of the rNK-1R (designated the N-terminal tryptic fragment; see Fig.  5) includes the extracellular N terminus (containing both consensus sequences for N-linked receptor glycosylation: Asn 14 and Asn 18 ), the first transmembrane region, and the proximal portion of the first intracellular loop. This ϳ40-kDa tryptic fragment was not reduced further in size by digestion with higher concentrations of trypsin, indicating that it represents a limit tryptic fragment of the photolabeled rNK-1R.
We can also use the theoretical tryptic fragmentation restriction map of the rNK-1R to conclude that the 5.1-kDa tryptic fragment, when analyzed under both nonreducing and reducing conditions, corresponds to the 125 I-[D-Tyr 0 ]Bpa 3 SP photoligand (calculated molecular mass of 1.8 kDa) covalently attached to an amino acid within the segment of the rNK-1R extending from residues 149 to 177 (calculated molecular mass of 3.3 kDa). This region of the rNK-1R (designated the core tryptic fragment; see Fig. 5) includes the fourth transmembrane region and the proximal portion of EC2. This 5.1-kDa fragment was also not further reduced in size by digestion with higher concentrations of trypsin, indicating that it represents a second limit tryptic fragment of the photolabeled rNK-1R.
The intermediate tryptic fragments observed in Fig. 3a were reduced in size to 6.7 kDa upon addition of DTT, confirming the presence of a disulfide bond. Remarkably, this tryptic digestion pattern closely resembles the tryptic digestion pattern observed with the 125 I-BH-Bpa 4 SP-labeled rNK-1R (10). Work in our laboratory has shown that when the rNK-1R is photolabeled with 125 I-BH-Bpa 4 SP, the identified 6.7-kDa tryptic fragment corresponds to the 125 I-BH-Bpa 4 SP photoligand (calculated molecular mass of 1.8 kDa) covalently attached to an amino acid within the segment of the rNK-1R extending from residues 149 to 190 (calculated molecular mass of 4.9 kDa) (10). This segment of the rNK-1R contains EC2 residue Cys 180 , which forms a disulfide bond with EC1 residue Cys 105 , linking the first and second extracellular domains of the receptor (15). Our laboratory subsequently reported that the direct site of covalent attachment of 125 I-BH-Bpa 4 SP to the rNK-1R is to EC2 residue Met 174 (10). The inability of trypsin to fully cleave the 125 I-BH-Bpa 4 SP-labeled (and in this case, the 125 I-[D-Tyr 0 ]Bpa 3 SP-labeled) rNK-1R at the rNK-1R Arg 177 -Val 178 cleavage site might suggest that (i) the photoligand in some manner limits the enzyme from accessing this particular site, or (ii) the spatial constraint of the disulfide bond linking the first and second extracellular domains of the receptor partially protects this receptor region from the active site of the enzyme by positioning it near the plasma membrane.
Unlike previous results observed with the peptide mapping of both 125 I-BH-Bpa 4 SP-labeled (10) and 125 I-BH-Bpa 8 SP-labeled (9) rNK-1Rs, peptide mapping of the 125 I-[D-Tyr 0 ]Bpa 3 SP-labeled rNK-1R revealed two limit tryptic fragments, suggesting that 125 I-[D-Tyr 0 ]Bpa 3 SP covalently attaches to two distinct residues far apart from one another within the rNK-1R primary sequence: one site of covalent attachment being to a receptor amino acid between residues 1 and 61 on the extracellular N terminus and the other site of covalent attachment being to a receptor amino acid between residues 149 and 177 on EC2.
Subcleavage Mapping of the 125 I-[D-Tyr 0 ]Bpa 3 SP/rNK-1R Contact Points-As expected, treatment of the 5.1-kDa core tryptic fragment with additional reducing agents did not alter the size of the fragment due to the lack of the cysteine-cysteine disulfide bond (Fig. 4a, lane 3). However, digestion of the 5.1-kDa core tryptic fragment with protease V8 resulted in the generation of a 2.4-kDa fragment (Fig. 4a, lane 5). This protease V8-generated fragment corresponds to the 125 I-[D-Tyr 0 ]Bpa 3 SP photoligand (calculated molecular mass of 1.8 kDa) covalently attached to an amino acid within the segment of the rNK-1R extending from residues 173 to 177 (calculated molecular mass of 0.6 kDa) (see Fig. 5). Therefore, these results show that one site of covalent attachment of 125 I-[D-Tyr 0 ]Bpa 3 SP to the rNK-1R is to an amino acid located on the proximal portion of EC2 between residues 173 and 177. Treatment of the ϳ40-kDa N-terminal tryptic fragment with endoglycosidase F resulted in the generation of an 8.6-kDa fragment (Fig. 4b, lane 8). This result shows that the N-terminal tryptic fragment contains ϳ30 kDa of carbohydrate residues and must therefore also contain the two asparagine residues on the extracellular N terminus capable of being glycosylated (Asn 14 and Asn 18 ). The deglycosylated 8.6-kDa fragment itself corresponds to the 125 I-[D-Tyr 0 ]Bpa 3 SP photoligand (calculated molecular mass of 1.8 kDa) covalently attached to an amino acid within the segment of the rNK-1R extending from residues 1 to 61 (calculated molecular mass of 6.8 kDa). Subsequent digestion of the 8.6-kDa fragment with protease V8 resulted in the generation of a 3.0-kDa fragment (Fig. 4b, lane 9). This protease V8-generated fragment corresponds to the 125 I-[D-Tyr 0 ]Bpa 3 SP photoligand (calculated molecular mass of 1.8 kDa) covalently attached to an amino acid within the segment of the rNK-1R extending from residues 11 to 21 (calculated molecular mass of 1.2 kDa) (Fig. 5). Therefore, these results show that a secondary site of covalent attachment of 125 I-[D-Tyr 0 ]Bpa 3 SP to the rNK-1R is to an amino acid located on the extracellular N terminus between residues 11 and 21; interestingly, this segment of the rNK-1R contains both consensus sequences for N-linked glycosylation, the importance of which is still undetermined. Importantly, the same 3.0-kDa protease V8-generated 125 I-[D-Tyr 0 ]Bpa 3 SP-labeled rNK-1R fragment was generated when the N-terminal tryptic fragment was analyzed in the converse manner, i.e. when the glycosylated N-terminal tryptic fragment was first subjected to protease V8 digestion and then to treatment with endoglycosidase F (Fig. 4b, lanes 11 and 12). DISCUSSION [D-Tyr 0 ]Bpa 3 SP is, in our studies to date, the only photoreactive analogue of SP that covalently labels residues on two distinct domains of the rNK-1R. The sites of [D-Tyr 0 ]Bpa 3 SP covalent attachment to the rNK-1R are located at a considerable distance from each other in the primary amino acid sequence of the receptor molecule: one is located within the proximal portion of EC2, and the other is located within the extracellular N terminus. An interpretation of the results presented in this report is that two distinct ligand⅐receptor complexes exist at equilibrium: one complex in which an amino acid of the receptor between residues 173 and 177 on EC2 is in close spatial proximity to the Bpa 3 residue of 125 I-[D-Tyr 0 ]Bpa 3 SP and a second complex in which an amino acid of the receptor between residues 11 and 21 on the extracellular N terminus is in close spatial proximity to the Bpa 3 residue of 125 I-[D-Tyr 0 ]Bpa 3 SP. However, since the results of our competition binding studies comparing the binding properties of Bpa 3 SP with those of SP for the rNK-1R do not suggest the existence of multiple peptide⅐rNK-1R complexes, a more likely interpretation of the photolabeling data showing the ability of 125 I-[D-Tyr 0 ]Bpa 3 SP to photolabel the aforementioned receptor segments is that the glycosylated segment of the extracellular N-terminal domain of the rNK-1R is in close spatial proximity to the experimentally determined proximal region of EC2 in the high affinity SP⅐NK-1R equilibrium complex.
Previously, the same photoligand ([D-Tyr 0 ]Bpa 3 SP) has been used to map the peptide-binding domains of the NK-1R present in the macrophage/monocyte cell line P388D 1 (11). The conclusion of this study was that the site of 125 I-[D-Tyr 0 ]Bpa 3 SP covalent attachment to the NK-1R was to an amino acid of the NK-1R within the extracellular N-terminal segment extending from residues 1 to 21. In this report, we have confirmed these previous results documenting 125 I-[D-Tyr 0 ]Bpa 3 SP photolabeling of the N terminus of the NK-1R and have restricted the site of covalent attachment to an amino acid between residues 11 and 21, a segment of the N terminus that contains the two consensus sequences for N-linked glycosylation.
In the previous study using the 125 I-[D-Tyr 0 ]Bpa 3 SP photoligand (11), however, no photolabeling of EC2 was reported. The difference in the results obtained is probably due to the differences in the methods used for analysis of the photolabeled receptor fragments. We analyzed the 125 I-[D-Tyr 0 ]Bpa 3 SP-photolabeled NK-1R tryptic fragments and subcleavage products by an SDS-PAGE system modified for small peptide analysis (17,18) and autoradiography. In contrast, Li et al. (11) used HPLC analysis following receptor trypsinization. In our experience, hydrophobic tryptic fragments, such as those derived from EC2, are resistant to elution by standard HPLC procedures.
Thus, our data add new information that contributes experimentally determined spatial constraints that are important for the three-dimensional modeling of the SP⅐NK-1R complex (see accompanying article (27)). Whereas large glycoprotein hormones (e.g. thyrotropin and follicle-stimulating hormone) bind to the extended extracellular N-terminal region of their G protein-coupled receptors (24,25), small neurotransmitters (e.g. acetylcholine and norepinephrine) bind to multiple transmembrane-spanning sequences of their G protein-coupled receptors (25,26). The mechanisms by which small flexible peptides such as SP interact with their membrane-bound G protein-coupled receptors is still unknown. The insights we are obtaining on the multiple contacts established between SP and the NK-1R should prove useful for the study of the structurefunction relationships of other peptides of similar size with their G protein-coupled receptors.