Further definition of the substance P (SP)/neurokinin-1 receptor complex. MET-174 is the site of photoinsertion p-benzoylphenylalanine4 SP.

The covalent attachment site of a substance P (SP) analogue containing the photoreactive amino acid p-benzoyl-l-phenylalanine (Bpa) in position 8 of the C-terminal portion of the peptide was identified previously as Met-181 on the neurokinin-1 (NK-1) receptor. In this study, a second photoreactive SP analogue, Bpa(4)-SP, in which the Bpa residue is located in the N-terminal portion of the peptide, was used to define further the peptide-receptor interface. The NK-1 receptor expressed in Chinese hamster ovary cells was specifically and efficiently photolabeled with a radioiodinated derivative of Bpa(4)-SP. Fragmentation analysis of the photolabeled receptor restricted the site of photoincorporation of Bpa(4)-SP to an amino acid within the sequence Thr-173 to Arg-177 located on the N-terminal side of the E2 loop. To identify the specific amino acid in this sequence that serves as the covalent attachment site for Bpa(4)-SP, a small photolabeled receptor fragment was generated by chemical cleavage with cyanogen bromide. Matrix-assisted laser desorption/ionization time of flight mass spectrometric analysis of the purified fragment identified a single protonated molecular ion with a molecular mass of 1801.3 +/- 1.8, indicating that upon irradiation, the bound photoligand covalently attaches to the terminal methyl group of a methionine residue. This result, taken together with the results of the peptide mapping studies, establishes that the site of Bpa(4)-SP covalent attachment to the NK-1 receptor is Met-174.

The covalent attachment site of a substance P (SP) analogue containing the photoreactive amino acid pbenzoyl-L-phenylalanine (Bpa) in position 8 of the Cterminal portion of the peptide was identified previously as Met-181 on the neurokinin-1 (NK-1) receptor. In this study, a second photoreactive SP analogue, Bpa 4 -SP, in which the Bpa residue is located in the N-terminal portion of the peptide, was used to define further the peptide-receptor interface. The NK-1 receptor expressed in Chinese hamster ovary cells was specifically and efficiently photolabeled with a radioiodinated derivative of Bpa 4 -SP. Fragmentation analysis of the photolabeled receptor restricted the site of photoincorporation of Bpa 4 -SP to an amino acid within the sequence Thr-173 to Arg-177 located on the N-terminal side of the E2 loop. To identify the specific amino acid in this sequence that serves as the covalent attachment site for Bpa 4 -SP, a small photolabeled receptor fragment was generated by chemical cleavage with cyanogen bromide. Matrix-assisted laser desorption/ionization time of flight mass spectrometric analysis of the purified fragment identified a single protonated molecular ion with a molecular mass of 1801.3 ؎ 1.8, indicating that upon irradiation, the bound photoligand covalently attaches to the terminal methyl group of a methionine residue. This result, taken together with the results of the peptide mapping studies, establishes that the site of Bpa 4 -SP covalent attachment to the NK-1 receptor is Met-174.
Substance P (SP) 1 is a peptide neurotransmitter that has a high affinity (10 Ϫ10 M) interaction with the NK-1 receptor (also Substance P receptor) (1). Identification of side-chain interactions between SP and the NK-1 receptor is important to under-stand the molecular basis for high affinity peptide binding and receptor activation. In the absence of high resolution structural data, molecular biological and biochemical approaches, particularly when combined, have provided useful information that allows the localization of specific interactions between SP and the NK-1 receptor. We have developed a methodology using the direct approach of photoaffinity labeling for the determination of contact sites between a peptide ligand and its receptor (1). In previous work, an analogue of SP in which a photoreactive amino acid p-benzoyl-L-phenylalanine (Bpa) was incorporated into position 8 of the peptide Bpa 8 -SP was used to covalently label the NK-1 receptor (1,(3)(4)(5). Importantly, the introduction of the Bpa residue into this position does not alter the ability of the SP analogue to bind the NK-1 receptor with high affinity or to induce a functional response. Furthermore, Bpa 8 -SP was shown to be a specific and an efficient photoprobe of the NK-1 receptor expressed in Chinese hamster ovary (CHO) cells. The site of Bpa 8 -SP covalent attachment was identified by peptide mapping strategies and matrix-assisted laser desorption/ionization time of flight (MALDI-TOF) mass spectrometry to be Met-181 of the E2 loop of the receptor (2,3)).
In the present study, to define further the peptide binding domain of the NK-1 receptor as well as to orient the SP peptide within the binding pocket, a second photoprobe, Bpa 4 -SP, in which proline 4 is replaced by the photoreactive Bpa residue (Bpa 4 -SP), has been used to covalently label the NK-1 receptor. The introduction of the Bpa residue into this position does not alter the ability of this analogue to bind to the NK-1 receptor with high affinity or to function as an NK-1 receptor agonist. The attachment site of radioiodinated 125 I-Bpa 4 -SP was identified by peptide mapping and MALDI-TOF mass spectrometry to be Met-174.

EXPERIMENTAL PROCEDURES
Materials-p-Benzoyl-L-phenylalanine 4 -substance P (Bpa 4 -SP) was synthesized and radioiodinated with a specific activity of 2200 Ci/mmol using the Bolton-Hunter reagent as described previously (1,2).
Cell Transfection and Culture-CHO cells were stably transfected with the rat SP receptor cDNA as described previously (6). The transfected CHO cells expressing 1 ϫ 10 5 SP binding sites/cell were maintained in ␣-minimum Eagle's medium (Sigma) supplemented with 10% fetal bovine serum and 0.8 mg/ml G418 (Life Technologies, Inc.). For preparative experiments, cells were obtained in quantity by culture in suspension using spinner flasks to a cell density of 2-3 ϫ 10 6 cells/ml.
Receptor Binding Assay-For the saturation binding experiments, transfected CHO cells were incubated at 4°C for 2 h with increasing concentrations of 125 I-Bpa 4 -SP (0.1-5 nM) in HEPES buffer (20 mM HEPES, 120 mM NaCl, 5 mM KCl, 2.2 mM MgCl 2 , 1 mM CaCl 2 , pH 7.4), supplemented with 6 mg/ml glucose and 0.6 mg/ml bovine serum albumin. Binding experiments using photoaffinity ligands were performed in the dark. Nonspecific binding was determined by the addition of excess unlabeled SP (1 M). Binding was terminated by the addition of * This work is sponsored by National Institutes of Health Grants NS 3134 (to N. D. B.) and NCRR P41-RR10888 (to C. E. C.). The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked "advertisement" in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.
Intracellular Calcium Measurement-Cells were plated on 9 ϫ 22-mm glass coverslips in a tissue culture dish and maintained in ␣-minimum Eagle's medium containing 10% fetal bovine serum and 0.8 mg/ml G418. When the cells reached 70 -90% confluence, the coverslips were washed with a HEPES-buffered saline solution (10 mM HEPES, 140 mM NaCl, 5 mM KCl, 1 mM MgSO 4 , 1.8 mM CaCl 2 , 10 mM glucose, pH 7.4). The cells were loaded with the fluorescent calcium indicator, 4 M Fura-2/AM (Molecular Probes), with 0.01% pluronic acid (9:1, v/v) in 2 ml of HEPES-buffered saline and incubated for 25 min at room temperature. The coverslip was then washed with HEPES-buffered saline and placed into a fluorescence spectrophotometer in a cuvette containing HEPES-buffered saline. Upon the addition of peptide ligands, fluorescent emission was measured at 510 nm with excitation wavelengths of 340 and 380 nm (Fura-2 bound and unbound to calcium, respectively). [Ca 2ϩ ] i was determined by the ratio of the fluorescence at the two excitation wavelengths as described previously (7).
Photoaffinity Labeling of Transfected CHO Cells-Cells were resuspended at 10 7 cells/ml in HEPES buffer and incubated with 125 I-labeled Bpa 4 -SP (0.1 nM) for 2 h at 4°C in the dark. The cells were then irradiated at a distance of 6 cm from a 100-watt long-wave (365 nm) UV lamp (Blak-ray, San Gabriel, CA) for 15 min on ice. For large scale photolabeling, cells were obtained in large quantity (5-liter suspension culture) and incubated with 125 I-Bpa 4 -SP isotopically diluted with 127 Ilabeled Bpa 4 -SP (1:1000).
After photolysis, the cells were obtained by centrifugation and washed twice with HEPES buffer. The cell pellets were resuspended in 5 mM Tris-HCl containing 1 mM EDTA (pH 7.0), homogenized briefly, and centrifuged at 1000 ϫ g for 15 min to remove cellular debris. The photolabeled membrane pellet was obtained by centrifugation at 40,000 ϫ g for 30 min and stored at Ϫ20°C. To test the specificity of the photoaffinity ligand, labeling was also carried out in the presence of excess unlabeled SP (1 M).
Enzymatic and Chemical Cleavage of Photolabeled NK-1 Receptors-Membrane preparations containing the photolabeled receptor were partially digested for 18 h with L-1-tosylamido-2-phenylethyl chloromethyl ketone-treated trypsin in 50 mM Tris-HCl and 1 mM CaCl 2 (pH 8.0) at room temperature. The amount of trypsin used per mg of membrane protein is specified in the figure legends. The digestion was stopped by the addition of SDS-PAGE sample buffer (10% glycerol, 1% SDS, 0.05% bromphenol blue, 125 mM Tris-HCl, pH 6.8). The solubilized peptide fragments were separated on an SDS-PAGE gel using the Tricine gel system of Schagger and von Jagow (8). Following electrophoresis, the gel was dried and exposed to x-ray film (Kodak XAR-5) with an intensifying screen (DuPont). The molecular masses of the radiolabeled fragments were determined using molecular mass markers from Amersham Pharmacia Biotech (2.35-46 kDa).
Following autoradiography, the radioactive bands were excised from dried gels and macerated into pieces. The partially digested receptor fragments were then eluted passively in either extraction buffer (0.1% SDS, 100 mM NH 4 HCO 3 , pH 7.8) for subcleavage by Staphylococcus aureus V8 protease or 0.1 N HCl for CNBr (cyanogen bromide) subcleavage. The eluted tryptic fragments were also incubated with 100 mM dithiothreitol (DTT) for 1 h at room temperature and then reanalyzed by SDS-PAGE.
For cleavage of the intact receptor with CNBr, photolabeled membranes were first absorbed to C18-derivatized silica beads (Alltech) at 1 mg of C18 beads/mg of membrane protein. The silica beads were washed twice with 0.1% trifluoroacetic acid in water, followed by 70% acetonitrile in 0.1% trifluoroacetic acid/water to remove the unbound ligand. The membrane-attached silica beads were then incubated overnight with 50 mg/ml CNBr in 0.1 N HCl in the dark at room temperature. CNBr-generated peptide fragments were eluted from the silica beads with SDS-PAGE sample buffer and were subject to electrophoresis and autoradiography.
To evaluate the effects of CNBr on the ligand, which also contains a methionine residue, 127 I-Bpa 4 -SP was treated with 50 mg/ml CNBr in 0.1 N HCl at room temperature overnight. The reaction mixture was further separated by C18 HPLC and analyzed by mass spectrometry.
Partial Purification of the Labeled Fragments by SDS-PAGE-For preparative experiments, a small receptor fragment (2 kDa) generated from CNBr cleavage was partially purified on preparative SDS-PAGE Tricine gel (3 mm). The small labeled receptor fragment was eluted from the gel in 0.1% trifluoroacetic acid for further purification by HPLC.
To determine the yield of photolabeled receptor fragments, the autoradiographs were aligned with the dried gels, the radiolabeled polypeptides were cut out, and the amount of radioactivity was measured.
Purification of the CNBr Cleavage Fragment by HPLC-The eluted receptor fragment (2 kDa) in 0.1% trifluoroacetic acid was further purified by reverse-phase HPLC using an Alltech C18 column (4.6 ϫ 250 mm) and a Vydac C4 column (4.6 ϫ 250 mm). The columns were eluted with increasing Solvent B (60% acetonitrile/0.08% trifluoroacetic acid/39% water) against Solvent A (0.08% trifluoroacetic acid in water) at a constant flow rate of 1.5 ml/min at 45°C. The concentration of Solvent B was raised from 20 to 60% at a rate of 1%/min. Fractions were collected every minute, and the 125 I radioactivity of eluted fractions was monitored by ␥ spectrometry. The UV absorbance of each fraction was also recorded at a wavelength of 210 nm.
Receptor Fragment Peptide Analysis by Mass Spectrometry-The molecular mass of the purified 2-kDa receptor fragment was determined at the Mass Spectrometry Resource at Boston University School of Medicine using matrix-assisted laser desorption/ionization time-of-flight mass spectrometry in less than picomole quantities. The Finnigan Vision 2000 instrument (Thermo BioAnalysis, Franklin, MA) is equipped with an LSI nitrogen laser (337 nm, 3-ns pulse duration). The matrix, 2,5-dihydroxybenzoic acid, was dissolved in 1:4 water/ acetonitrile (10 mg/ml).

Characterization of Iodinated Bpa 4 -SP-
The addition of a 127 I-labeled Bolton-Hunter conjugate of Bpa 4 -SP to transfected CHO cells expressing the NK-1 receptor induced a Ca 2ϩ response comparable with that induced by the same concentration of the parent peptide SP (Fig. 1). The response evoked by both peptides was completely inhibited in the presence of 1 M RP-67,580, a specific nonpeptide antagonist of the rat NK-1 receptor.
The binding affinity of radioiodinated 125 I-Bpa 4 -SP to intact CHO cells expressing the NK-1 receptor was determined by saturation binding experiments. The observed equilibrium binding was characterized by K d ϭ 1.4 Ϯ 0.2 nM, a value about 5-fold higher than that previously reported for the corresponding radioiodinated conjugate of the parent peptide (3). In contrast, the number of NK-1 receptors/cell (B max ) was the same for both derivatives.
Photoaffinity Labeling of the Rat NK-1 Receptor and Fragmentation Analysis-NK-1 receptors expressed on intact CHO cells were photolabeled with 125 I-Bpa 4 -SP, and the resulting ligand-receptor complex was analyzed by SDS-PAGE and autoradiography. Photolabeled receptors were observed as a broad band centered at 80 kDa (Fig. 2, lane 1). The diffuseness of the band is attributable to heterogeneous glycosylation of NK-1 receptors expressed in CHO cells (5). Quantitative analysis of the radioactivity incorporated into the NK-1 receptor indicated that photolabeling was highly efficient; ϳ40% of the specifically bound photoligand became covalently linked to the receptor upon UV exposure.
Membrane preparations of photolabeled receptors were treated under nonreducing conditions with trypsin, and the resulting fragments were analyzed by SDS-PAGE. Three major photolabeled fragments of 22, 14, and 5 kDa were observed (Fig. 2, lane 2). These bands were eluted from the gel. Further digestion by trypsin of the eluted fragments converted the 22and 14-kDa fragments to the 5-kDa fragment (data not shown), thus establishing that the 5-kDa fragment is the limit peptide produced by trypsin digestion.
Previously we have shown that the NK-1 receptor contains a disulfide bond linking Cys-105 on the E1 loop to Cys-180 on the E2 loop and that this bond plays an important role in the high affinity binding of SP (3). The analysis of receptor fragments generated in the absence or presence of DTT is valuable for the identification of receptor fragments held together by a disulfide bond between Cys-105 and Cys-180. Therefore, each of the photolabeled bands obtained from tryptic digestion under nonreducing conditions was further treated with DTT and reanalyzed by SDS-PAGE. Upon reduction by DTT, the mobility of the 22-kDa band increased to 14 kDa, whereas the 14-kDa band increased to 7 kDa (Fig. 3, lanes 1-4). In contrast, the mobility of the 5-kDa fragment was not changed by treatment with DTT, indicating that this photolabeled receptor fragment, unlike the other two larger fragments, lacks the disulfide bond (Fig. 3, lanes 5 and 6). Based on the experimentally determined molecular mass of the limit fragment and extended forms, together with their susceptibility to reductive cleavage by DTT, we can conclude that digestion of the photolabeled receptor with trypsin generates an ϳ5-kDa fragment corresponding to residues Leu-142 to Arg-177, which has a calculated molecular mass of 5.8 kDa when the mass of the covalently attached probe, 1.8 kDa, is added.
This 5-kDa fragment contains a single cleavage site for S. aureus V8 protease at residue Glu-172. To define more closely the residue that serves as the site of photoincorporation, the limit tryptic fragment of M r 5000 was treated with S. aureus V8 protease. After cleavage of the 5-kDa fragment with S. aureus V8 protease, a small fragment of 2.4 kDa in agreement with the calculated mass (2.3 kDa) of the photoligand covalently attached to a receptor fragment extending from residues 173 to 177 (TMPSR) was obtained (Fig. 4, lane 2). The M r 5000 limit tryptic fragment was also treated with CNBr, a reagent specific for cleavage after Met, generating a fragment of M r ϳ2000 (Fig.  4, lane 3), a value that is close to that of the photoligand itself.
Identification of the Residue on the NK-1 Receptor That Is Covalently Labeled by Bpa 4 SP-To obtain a sufficient amount of the 2-kDa fragment for accurate determination of molecular mass MALDI-TOF mass spectrometric analysis, transfected CHO cells (10 9 ) were photolabeled with 125 I-Bpa 4 -SP isotopically diluted with 127 I-Bpa 4 -SP to a specific activity of 2 Ci/ mmol and then subjected to solid-phase CNBr cleavage (2). Approximately 100 pmol of the photolabeled receptor was absorbed onto C18-derivatized silica beads. Following treatment with CNBr, the generated fragments were resolved using Tricine-based SDS-PAGE. The 2-kDa radiolabeled fragment was eluted and further purified by HPLC using both C18 and C4 columns (Fig. 5, A and B). The final yield of the purified 2-kDa fragment was 8 pmol, representing 8% of the starting photolabeled receptor, an amount more than sufficient to permit MALDI-TOF mass spectrometric analysis.
MALDI-TOF mass spectrometric analysis of the CNBr-generated fragment yielded a (MϩH) ϩ with an m/z value of 1801.3 Ϯ 1.8 (Fig. 6). This mass spectrum not only confirms the purity of the isolated fragment but also defines the chemical structure of the fragment. The observed molecular mass (1801.3 Ϯ 1.8) could only be generated from a photolabeled receptor by the covalent attachment of the probe 127 I-Bpa 4 -SP (molecular weight ϭ 1776.9) to the methyl group of a methionine residue, followed by CNBr cleavage of the bond between the ␥-carbon and sulfur on the methionine side chain to form a thiocyanate (ϪCH 2 SCN) derivative. In addition, the C-terminal methionine amide of the covalently attached probe is converted by CNBr to a homoserine lactone. The observed (MϩH) ϩ of m/z 1801.3 Ϯ 1.8 of the final cleavage product is in good agreement with the calculated (MϩH) ϩ of m/z 1802.9.
Because there is only one methionine residue (Met-174) in the photolabeled receptor fragment (residues 173-177), we can conclude that the site of photoincorporation of Bpa 4 -SP is Met-174 (Fig. 7). DISCUSSION Our previous results have shown that Bpa 8 -SP, which contains the photoreactive Bpa residue in the eighth position of the SP peptide within the conserved C-terminal region that defines the tachykinin peptide family, covalently attaches to a single residue, Met-181, on the NK-1 receptor (2). The present study characterizes the site of a covalent attachment of a second photoreactive analogue of SP with the photoreactive residue The substitution of the 4th position of SP (Pro-4) by Bpa is well tolerated by the receptor, as evidenced by its high affinity binding to the NK-1 receptor and its ability to stimulate an increase in intracellular calcium in the in vitro assay. The results presented in this report show that Bpa 4 -SP covalently attaches to a single residue Met-174, implying that position 4 of SP is in close spatial proximity to Met-174 when bound to the NK-1 receptor.
Interestingly, in a separate study, Girault et al. (9) identified Met-174 as the attachment site on the human NK-1 receptor for the SP analogue Bpa 8 -Pro 9 -SP, the same residue that is labeled in the present study by Bpa 4 -SP. Although the basis for the discrepancy in these studies is unknown, the introduction of a proline residue at the 9th position of SP could potentially alter the confirmation of the peptide in such a manner that the Bpa residue of Bpa 8 -Pro 9 -SP is positioned in close proximity to Met-174 rather than Met-181. Additional explanations could include (i) species differences (human versus rat) or (ii) experimental differences (e.g. labeling of membrane preparations versus labeling of intact cells).
The results obtained here with radioiodinated Bpa 4 -SP, when combined with our previous findings (2, 3) with radioiodinated Bpa 8 -SP, establish the importance of the initial sequence of the E2 loop extending beyond transmembrane 4 of the NK-1 receptor (in peptide binding). Based on these findings, it is likely that the bound SP peptide is oriented parallel to this region of the receptor with its 4th position adjacent to Met-174 and its 8th position adjacent to Met-181. Because of its location in the E2 loop just after it emerges from the lipid bilayer, Met-174 is positioned close to the membrane. Furthermore, although Met-181 is located in the middle of the E2 loop,  5. Purification of the CNBr-generated 2-kDa receptor fragment by HPLC. The partially purified 2-kDa receptor fragment from CNBr cleavage was isolated from the SDS-PAGE gel and was further purified by reverse-phase HPLC using a C18 column (A), followed by a C4 column (B). Radioactivity was monitored by ␥-emission spectrometry. it is adjacent to Cys-180, which participates in a disulfide bond with Cys-105 (3), a residue that is close to the membrane interface. Thus, Met-181 is also positioned near the membrane interface of the peptide. Interestingly, we have recently obtained evidence (10) that these two methionines on the E2 loop are also spatially close to each other.