Structure-Function Relationships in the Neuropeptide S Receptor

Neuropeptide S (NPS) and its receptor (NPSR) are thought to have a role in asthma pathogenesis; a number of single nucleotide polymorphisms within NPSR have been shown to be associated with an increased prevalance of asthma. One such single nucleotide polymorphism leads to the missense mutation N107I, which results in an increase in the potency of NPS for NPSR. To gain insight into structure-function relationships within NPS and NPSR, we first carried out a limited structural characterization of NPS and subjected the peptide to extensive mutagenesis studies. Our results show that the NH2-terminal third of NPS, in particular residues Phe-2, Arg-3, Asn-4, and Val-6, are necessary and sufficient for activation of NPSR. Furthermore, part of a nascent helix within the peptide, spanning residues 5 through 13, acts as a regulatory region that inhibits receptor activation. Notably, this inhibition is absent in the asthma-linked N107I variant of NPSR, suggesting that residue 107 interacts with the aforementioned regulatory region of NPS. Whereas this interaction may be at the root of the increase in potency associated with the N107I variant, we show here that the mutation also causes an increase in cell-surface expression of the mutant receptor, leading to a concomitant increase in the maximal efficacy (Emax) of NPS. Our results identify the key residues of NPS involved in NPSR activation and suggest a molecular basis for the functional effects of the N107I mutation and for its putative pathophysiological link with asthma.

the lung (12), although other studies have shown that expression of the receptor is restricted to the brain, in particular the hypothalamus, and the retina (7,9). Consistent with this latter pattern of expression, NPSR has been shown to be involved in sleep, anxiety (7), and feeding (13) in rodents.
The structural determinants of NPSR activation by NPS are for the most part unknown. A recent report localized the pharmacophore to the NH 2 -terminal half of the peptide and demonstrated that the N107I variant of NPSR is a gain-of-function mutant (14). We sought to better characterize structure-activity relationships within the receptor and its ligand as well as to define the molecular basis for the altered functional behavior of the N107I variant. We identify the key residues in NPS involved in activation of NPSR as well as demonstrate that the N107I mutation increases both intrinsic efficacy as well as intracellular trafficking of the receptor to the cell membrane.

EXPERIMENTAL PROCEDURES
Generation of NPSR Vector Constructs-The NPSR-A isoform was amplified from a hypothalamus Marathon-ready cDNA library (BD Biosciences) by PCR using the Accuprime Pfx DNA polymerase (Invitrogen) following manufacturer's instructions and using the following primers: "sense 1," 5Ј-GTCAGGTA-CCACCATGCCAGCCAACTTCACAGAGGG-3Ј; "antisense 1," 5Ј-TGCTCGATGCGGCCGCCTAGATGAATTCTGGC-TTGGAC-3Ј. These primers introduce a KpnI recognition sequence and a Kozak sequence at the 5Ј end, and a NotI cleavage site at the 3Ј end. The NPSR-B isoform was generated from the NPSR-A construct by Splicing Overlap Extension PCR. Briefly, in a first round of PCR, the 1051-bp sequence corresponding to the common region of the two NPSR isoforms was amplified by PCR using the NPSR-A isoform sequence as template, and a 108-bp NPSR-B-specific sequence was obtained by annealing and amplification of the following overlapping primers: "sense 2," 5Ј-GTCATCCGTCTCCGTCAGCTCCAGGA-GGCTGCGCTAATGCTCTGCCCTCAACGAGAGAACTG-GAAGGGTACTTG-3Ј; and "antisense 2," 5Ј-GCTAGTAAG-CGGCCGCTCACCTTGGAAGAGCCCAGGAAGGTACAC-CTGGCCAAGTACCCTTCCAGTTCTCTCGTTG-3Ј. These 2 PCR fragments were then spliced together in a second round of PCR using the primers "sense 1" and "antisense 2" (above) to generate the full-length NPSR-B isoform. The NPSR-A-N107I and NPSR-B-N107I constructs were generated from the above isoform-specific sequences by site-directed mutagenesis using the QuikChange Site-directed Mutagenesis kit (Stratagene, La Jolla, CA) following the manufacturer's instructions and using the following primers: "sense 3," 5Ј-GACAGATATTATTTG-GCGATTCACTGG-3Ј and "antisense 3," 5Ј-CCAGTGAATC-GCCAAATAATATCTGTC-3Ј. FLAG-tagged versions of the NPSR constructs were generated by PCR using the overlapping primers "Sp-flag sense," 5Ј-GTAGGTACCACCATGAAGAC-GATCATCGCCCTGAGCTACATCTTCTGCCTGGTATT-CGCCGACTAC-3Ј and "Sp-flag antisense," 5Ј-CTGCCTGG-TATTCGCCGACTACAAGGACGATGATGACGCCCCAG-CCAACTTCACAGAGGGCAGCTTCG-3Ј, which introduce a KpnI cleavage site, a signal peptide sequence, and a FLAG tag at the amino termini of the receptors (15). All constructs were subcloned into the vector pcDEF3 (16) at the KpnI/NotI restric-tion sites and their sequences confirmed by automated capillary DNA sequencing using the Big Dye Terminator 3.0 system on the ABI Prism 3100 Genetic Analyzer (Applied Biosystems, Foster City, CA).
Generation of CHO-NPSR and HEK 293 T-NPSR Cells-NPSR-expressing cells were generated through transfection of Chinese hamster ovary (CHO) or HEK 293 T cells with pcDEF3 constructs containing the sequences for NPSR-A, NPSR-B, NPSR-A-N107I, or NPSR-B-N107I, or FLAGtagged versions thereof (see above) using the Lipofectamine Plus system (Invitrogen). For CHO-NPSR cells, transfected cells were selected with G418 (750 g/ml) to generate a bulk stable population and high activity clones were identified by plating single cells in flat-bottom 96-well Nunc plates (VWR, Mississauga, ON, Canada) using a FACSVantage cell sorter, allowing the clones to grow to confluence, and assessing their activity in the Ca 2ϩ -mobilization assay (described below). For HEK 293 T-NPSR cells, transfected cells were assessed in Ca 2ϩ -mobilization, binding, immunofluorescence, or enzyme-linked immunosorbent (ELISA) assays (all described below) as a bulk population, 48 h following transfection with the pcDEF3(NPSR) constructs.
Peptide Synthesis-C-terminally amidated peptides were synthesized at the 0.1 mM scale by solid-phase Fmoc (N-(9fluorenyl)methoxycarbonyl) chemistry on a Pioneer Peptide Synthesis System (Applied Biosystems, Foster City, CA), as described previously (17). The identity of all peptides was confirmed using matrix-assisted laser desorption ionization mass spectrometry. All peptides were determined to be Ͼ95% pure by reversed-phase high-performance liquid chromatography using a C 18 column as the solid phase and a H 2 O:acetonitrile gradient as the solution phase. Peptide concentrations were determined using the bicinchonic acid (BCA) protein assay (Pierce Chemical Co.).
Calcium Mobilization Assay-The calcium-mobilization assay was carried out as described previously (18), with the following modifications. CHO or HEK 293 T cells expressing the NPSR variants (see above) were plated (40,000 or 100,000 cells/ well for CHO and HEK 293 T cells, respectively) in 96-well black-walled, clear-bottom plates. After incubation overnight and washing with Hanks' balanced salt solution, 105 l/well of "no wash" calcium dye (Molecular Devices, Sunnyvale, CA) was loaded onto the cells and incubated for 45 min at 37°C in 5% CO 2 . Using the built-in sample addition feature of the FlexStation II instrument, 15 l of agonist diluted in Hank's balanced salt solution (containing 20 mM Hepes and 0.24% bovine serum albumin) was added to the cells and fluorescence emission ( exc ϭ 485 nm; emm ϭ 525 nm) was measured at 3-s intervals before, during, and after agonist addition for a total of 2 min. Data are reported as the difference between the maximal fluorescence and the baseline fluorescence. EC 50 values, representing the concentration of agonist at which 50% activity is obtained, are reported as mean Ϯ S.D. of multiple determinations.
NMR Spectroscopy-Nuclear magnetic resonance (NMR) spectra were acquired at a sample temperature of 25°C on a Varian Inova 600 MHz spectrometer equipped with a 5-mm triple resonance Z-axis pulsed field gradient probe. NPS sam-ples were analyzed at 3.1 mg/ml (1.4 mM) in 50 mM sodium phosphate buffer in 90% H 2 O, 10% D 2 O, pH 5.0. 1 H chemical shifts were referenced internally to sodium 3-(trimethylsilyl) propionate-2,2,3,3-d 4 at 0.00 ppm. The water resonance in all experiments was suppressed by low power saturation using an attenuated transmitter pulse during the relaxation delay. Sequence-specific chemical shift assignments were obtained with the combination of two-dimensional total correlated spectroscopy (TOCSY) (19,20) and nuclear Overhauser effect spectroscopy (NOESY) (21) experiments using standard methodologies (22). All two-dimensional experiments were acquired in the hypercomplex mode for phase-sensitive presentation (23). Clean TOCSY (19,24) spectra were recorded with 1 K complex points in t 2 and 512 points in t 1 consisting of 16 -32 transients per increment. Spin locking was achieved with an MLEV-17 (24) mixing sequence for a duration of 65 ms preceded by a G-90-G magnetization randomization sequence using a 2.0-ms gradient pulse prior to the relaxation delay. Data sets were multiplied by a shifted Gaussian apodization function and zerofilled to 2 K by 1 K complex points prior to Fourier transformation. NOESY spectra were acquired using a range of mixing times. Solvent suppression was achieved by selective saturation of the water resonance during the 1-s recycle delay, t 1 period, and mixing period. Data sets were acquired with 1 K complex points in t 2 and 512 points in t 1 with 96 transients per increment. The NOESY data were processed as described above.
CD Spectroscopy-Circular dichroism (CD) measurements were performed on a J-810 spectropolarimeter equipped with a Peltier temperature controller set at 25°C (JASCO, Easton, MD). Spectra were collected from 180 to 300 nm at a scan speed of 200 nm/min for 5 accumulations/sample with a data interval of 0.2 nm. Spectra were corrected for solvent background, and the peptide concentration was quantified by amino acid analysis. The raw CD signal was converted to molar ellipticity, [⍜], by [⍜] ϭ ⍜ observed (MRW/10⅐l⅐c), where MRW is the mean residue weight (molecular mass divided by number of peptide bonds), l is path length in cm, and c is concentration in mg/ml. Spectral data were input into the program SSE, version 1.00.00 (JASCO), which is based upon the least squares method of Chang-Wu-Yang (25,26) for estimation of secondary structure characteristics.
Binding Assay-HEK 293 T cells were plated (600,000 cells/ well) in poly-D-lysine-coated 6-well plates, incubated for 24 h, and transfected with 0.5 g of the various NPSR constructs as described above. 48 h following transfection, cells were washed with 2% glucose, 1% bovine serum albumin in phosphate-buffered saline and incubated with various concentrations of 125 I-NPS (Phoenix Pharmaceuticals, Belmont, CA) in binding buffer (1 mM tyrosine, 1 mM phenylalanine, 1 mg/ml glucose, 10 mg/ml bovine serum albumin in phosphate-buffered saline) and in the presence or absence of unlabeled NPS (at 100 times the concentration of 125 I-NPS) for 2 h at 4°C with mild agitation. Cells were then washed twice and lysed in the presence of 0.1 M NaOH for 30 min at 37°C. Bound radioligand was measured by scintillation counting on a 1450 Microbeta counter (PerkinElmer Life Sciences).
Immunofluorescence Microscopy and ELISA-HEK 293 T cells were plated in poly-D-lysine-coated 8-well slides or 24-well plates (at 50,000 or 150,000 cells per well for immunofluorescence microscopy and ELISAs, respectively), incubated for 24 h, and transfected with the various NPSR constructs as described above. Analyses were carried out 48 h following transfection, as follows. For analysis of non-permeabilized cells, cells were incubated in blocking buffer (0.5% bovine serum albumin in phosphate-buffered saline) for 15 min and incubated with anti-FLAG antibody (1:500 in blocking buffer) for 30 min. Cells were then washed, fixed in the presence of 3% paraformaldehyde (15 min, 4°C), and permeabilized with 0.2% Triton in blocking buffer. For immunofluorescence, cells were then incubated with Alexa Fluor 488-conjugated anti-mouse antibody (1:500, Invitrogen), coverslips were mounted on the slides, and images were acquired on an Axioplan2 imaging microscope (Zeiss, Thornwood, NY). For ELISAs, cells were incubated with horseradish peroxidase-conjugated anti-mouse antibody (1:500; Roche Applied Science, Laval, Canada) and cleavage of the horseradish peroxidase substrate o-phenylenediamine dihydrochloride (Sigma) was monitored as per the manufacturer's instructions by measuring spectral absorbance at 492 nm. For analysis of permeabilized cells, cells were fixed and permeabilized prior to incubation with anti-FLAG antibody.

RESULTS
Functional Properties of NPSR Variants-It has previously been demonstrated that a single nucleotide polymorphism leading to mutation of residue Asn-107 to Ile in NPSR is associated with an increased risk for asthma (4). More recently, it was shown that this mutation causes an increase in the potency of NPS for NPSR (14). To confirm the latter finding, we assessed NPS-dependent intracellular Ca 2ϩ -mobilization of the wild type (WT) and N107I variants of both NPSR-A and NPSR-B splice isoforms in multiple clones of CHO-NPSR cells. As shown in Table 1, NPS activates NPSR-A-N107I with a ϳ10fold higher potency than its WT counterpart NPSR-A. This pattern is also observed for NPSR-B-N107I and NPSR-B and was confirmed in transiently transfected HEK 293 T cells (as described below). A comparison of the two splice variants themselves, however, reveals no significant difference in their potency for NPS, regardless of the nature of residue 107 (Table 1).
Secondary Structure of NPS-To study the molecular basis for the functional difference between the WT and N107I variants, we first sought to investigate structure-function relation-

CHO-NPSR cells HEK-NPSR cells
ships in the WT receptor and its ligand. As an initial approach, we assessed the secondary structure of the peptide by NMR spectroscopy. Sequence-specific 1 H resonance assignments were obtained for the peptide with the combined use of TOCSY and NOESY spectra (22). Qualitative secondary structure characterization was accomplished primarily by two-dimensional NOE studies. As shown in Fig. 1A, the amide proton region of the NOESY spectrum displays a contiguous stretch of moderately intense amide NH to NH NOEs from Gly-5 to Thr-13 (with the exception of NOEs between Lys-11 and Lys-12, which cannot be resolved because of the close proximity of the chemical shifts of these two residues), consistent with helical secondary structure in the region spanning these residues. However, the long range proton-proton interactions, d ␣N (i, iϩ3) and d ␣ ␤ (i, iϩ3), which are characteristic of a defined ␣-helix were not observed. In addition, CD spectroscopy analysis (not shown) was consistent with approximately one-third of the peptide adopting a ␤-turn-like, rather than ␣-helical, conformation. Taken together, the NMR and CD data indicate the presence of a nascent helix from residue Gly-5 to Thr-13, which may adopt a true ␣-helical conformation upon receptor binding (see "Discussion" and Ref. 27). The putative three-dimensional conformation of NPS containing such a helix is shown in Fig. 1B.

Functional Roles of Residues in NPS-
To investigate the role of specific amino acid residues within NPS, we carried out mutagenesis of the peptide (see Fig. 2) and assessed the effect of these mutations on NPSR activation in the CHO-NPSR cells using the Ca 2ϩ -mobilization assay. Examples of concentrationresponse curves for NPS and selected truncated mutant peptides are shown in Fig. 3, A and B, whereas a summary of the EC 50 values obtained for all truncated mutants is shown in Fig.  3C and Table 2. The results demonstrate that residues in the NH 2 -terminal third of the peptide are necessary and sufficient for receptor activation. Indeed, removal of the C-terminal 14 residues (peptide 1-6) has limited effect on the potency of the peptide, whereas removing Ser-1 (peptide 2-20) is detrimental to function and removing the first two (peptide [3][4][5][6][7][8][9][10][11][12][13][14][15][16][17][18][19][20] or three (peptide 4-20) NH 2 -terminal residues results in largely inactive peptides. Interestingly, peptides 1-7, 1-8, and 1-10 are significantly less potent compared with both full-length NPS and peptide 1-6, indicating that residues in this region, although not critical for function, can affect receptor activation.
To further probe the role of residues in the NH 2 -terminal half of NPS, we individually substituted residues 1 to 10 with Ala and assessed the ability of the resulting peptides to activate NPSR-A. As shown in Fig. 4 and Table 2, substitution of residues 2, 3, and 4 with Ala results in substantial to complete inactivation of the peptide, as does the G7A mutation. Taken together with the findings obtained with the truncated NPS mutants (above) these results indicate that the first 6 residues, in particular residues Phe-2, Arg-3, Asn-4, and Val-6, are necessary for receptor activation, whereas residue Gly-7 is critically located and can modulate the inherent activity of the peptide.
Interaction between NPSR Residue Asn-107 and Residues in NPS-Having identified regions of NPS that are critical for NPSR activity, we wished to assess the possible role of residue  The one-letter code for amino acids is used. Positions at which residues were mutated to Ala are underlined in the Ala scan mutants.
Asn-107 (in the receptor) on ligand binding and activation. We hypothesized that we might see an altered behavior of the mutant peptides on NPSR-A-N107I activation if this residue is involved in receptor activation and/or ligand binding. We therefore assessed the NH 2 -terminal truncated and Ala scan mutants on the N107I variant of NPSR-A. As shown in Fig. 5 (with EC 50 values shown in Table 2), the pattern of activation with the NH 2 -terminal truncated mutants is significantly altered with the N107I variant compared with its WT counterpart. Whereas peptides 1-10, 1-8, and 1-7 have EC 50 values for NPSR-A that are 21-, 17-, and 116-fold higher than WT NPS, respectively, these same peptides are equipotent to the WT peptide (with perhaps a modest 2-fold increase in EC 50 for peptide 1-10) on the NPSR-A-N107I variant. In addition, although substitutions of residues 2-4 with Ala results in partial to complete inactivation on both receptor variants, confirming their importance in receptor activation, the G7A mutation is considerably more detrimental to NPSR-A activity compared with NPSR-A-N107I. Taken together, these results suggest that Asn-107 plays a key role in modulating receptor activation and that its interaction with Gly-7 of NPS plays a regulatory role in NPSR-A, but not in NPSR-A-N107I.
Functional Effect of the N107I Mutation in a Transient Expression System-Whereas the results described above strongly suggest that the N107I mutation directly affects the intrinsic activation/binding of NPSR, we, nevertheless, wished to assess whether it might have indirect effects on function by altering receptor density at the cell surface. In the absence of suitable NPSR antibodies for such studies, we opted to tag the receptor with a FLAG epitope on the NH 2 -terminal end of the receptor. In addition, to circumvent the relatively low cell surface expression of NPSR in CHO cells, which precludes functional determinations in transiently transfected cells (data not shown and see Ref. 14), we tested a number of alternate cell  Table 2. Peptides for which EC 50 values could not be obtained due to low activity and absence of a maximal plateau (1-5, 3-20, and 4-20) are shown as having an EC 50 of 2000 nM. Values above bars represent % of maximal activity (based on WT peptide) obtained for these mutant peptides at a concentration of 2000 nM.

TABLE 2 Activity of mutant NPS peptides on NPSR-A and NPSR-A-N107I variants
EC 50 values (mean Ϯ S.D.) were obtained from multiple Ca 2ϩ -mobilization concentration-response curves generated in single clones of CHO-NPSR-A or CHO-NPSR-A-N107I cells. Major findings were confirmed in two other clones of each variant. In cases where EC 50 values could not be determined due to low potency, they are shown as being Ͼ2000 nM. lines for these studies. Based on comparative immunofluorescent microscopy analyses with various cell types (not shown), we identified HEK 293 T cells as being the most conducive to high cell surface expression of NPSR. Consistent with this, HEK 293 T cells show a concentration-dependent increase in Ca 2ϩmobilization in the presence of NPS following transfection with FLAG-NPSR-A (Fig. 6A). To investigate the effect of the N107I mutation on cell surface expression, we first confirmed the gain-of-function phenotype of this variant in both the NPSR-A and NPSR-B backgrounds in this system ( Fig. 6A and Table 1).

Peptide
Interestingly, in addition to the ϳ12-fold difference in EC 50 values between the WT and N107I variants, a 70 Ϯ 23% increase in maximal activation (E max ) is also observed for the mutant variants compared with the corresponding WT variants. Despite the important shift in EC 50 values, there was no significant difference in the binding affinity (K d ϭ 5.7 Ϯ 3.6 and 3.3 Ϯ 1.1 nM for NPSR-A and NPSR-A-N107I, respectively) of NPS for the two receptor variants (Fig. 6B), consistent with what has been previously demonstrated (14). On the other hand, we did observe a significant difference in maximal 125 I-NPS binding (not shown) between the two variants, suggesting that the N107I mutation affects cell surface expression of the receptor.
Cell Surface Expression of WT and N107I Variants of NPSR and Impact on Function-We explored the question of cell surface expression of the two NPSR variants using this transient expression system and taking advantage of the availability of anti-FLAG antibodies. As shown in Fig. 7A, immunofluorescence analysis of cells transfected with either variant of NPSR-A shows a higher level of N107I versus WT receptor expressed at the cell surface (non-permeabilized cells) despite similar levels of total receptor expression (permeabilized cells). We confirmed this finding quantitatively by measuring cell surface and total NPSR expression by ELISA (Fig. 7B) and flow cytometry (not shown). Our results are consistent with a higher degree of cell surface expression, resulting in higher receptor density, for the N107I receptor compared with the WT receptor. To assess whether receptor density might affect receptor function (28) in our system, we transfected HEK 293 T cells

. Activation of NPSR-A by alanine point mutants of human NPS.
Concentration-response curves in the Ca 2ϩ -mobilization assay were generated as described under "Experimental Procedures" and the EC 50 values thus obtained are shown in Table 2 Table 2. Peptides for which EC 50 values could not be obtained due to low activity and absence of a maximal plateau (F2A, R3A, N4A, and G7A) are shown as having an EC 50 of 2000 nM. Values above bars represent % of maximal activity (based on WT peptide) obtained for these mutant peptides at a concentration of 2000 nM.

FIGURE 5. Comparison of EC 50 values of mutant NPS peptides on NPSR-A and NPSR-A-N107I variants.
Concentration-response curves in the Ca 2ϩmobilization assay were generated as described under "Experimental Procedures" and the EC 50 values thus obtained are shown in Table 2. EC 50 values of the various mutants were normalized to those obtained for WT peptide on each variant and are expressed as -fold increase over WT peptide. Each value is the mean Ϯ S.E. of at least three separate determinations for each peptide. Peptides for which the -fold increase in EC 50 is higher than 200 are shown as having a 200-fold increase. NPSR-A, filled bars; NPSR-A-N107I, empty bars. A, summary of -fold increases in EC 50 values for C-terminal-truncated peptides. B, summary of -fold increases in EC 50 values for alanine point mutant peptides.
with varying amounts of pcDEF3(FLAG-NPSR-A-N107I) DNA to alter receptor expression, and measured the impact of this change on the potency (EC 50 ) of NPS and its maximal efficacy (E max ) on NPSR in the Ca 2ϩ -mobilization assay. To ensure that only receptor expression per cell was affected in this experiment (and not total number of cells transfected), we kept the total amount of DNA constant for each transfection by supplementing with "empty" pcDEF3 vector (see "Experimental Procedures"). As shown in Fig. 8, transfecting HEK 293 T cells with increasing amounts of NPSR-A-N107I DNA (0.003 to 0.1 g) causes a dose-dependent increase in E max without significantly affecting EC 50 .

DISCUSSION
The data presented herein provide information on structurefunction relationships within NPSR and its ligand, and propose a molecular basis for the functional differences observed between the WT receptor and the asthma-linked mutant N107I. The gain-of-function phenotype of this variant of NPSR is associated with an increase in both the intrinsic efficacy of its ligand NPS (associated with a decrease in EC 50 ) and in its cell surface expression (leading to an increase in E max ). Taken together with the finding that the N107I mutation is associated with an increase in asthma susceptibility (4,10,11), our data suggest that NPSR antagonism may be a suitable therapeutic approach for the treatment of this disease.
A first objective of this work was to better characterize structure-function relationships within NPSR and its ligand. Limited structural characterization of the NPS peptide by NMR (see Fig.  1) and CD spectroscopy identified a putative nascent helix between residues Gly-5 and Thr-13. The speculative ␣-helical conformation for this region of NPS shown in Fig. 1B is based on the observation that nascent helices are expected to stabilize into ␣-helices in constrained environments, such as might occur in the context of receptor binding (27). Whatever its true nature, this region of defined structure seems to have a role in modulating receptor activity (as discussed below), but is not required for receptor activation per se. Thus, a peptide in which most of the structured region of NPS is removed (peptide 1-6) is largely equipotent with the full-length WT peptide on the NPSR-A and NPSR-A-N107I receptors (see Figs. 3 and 5 and Table 2), indicating that the residues necessary and sufficient for receptor activation are located in the NH 2 -terminal third of the peptide. Combined with the Ala scan mutagenesis studies   Table 2) these results identify residues Phe-2, Arg-3, Asn-4, and Val-6 as critical for receptor activation, whereas Ser-1 and Gly-5 appear to play limited roles.
As stated above, the structured domain between residues 5 and 13 of NPS is not necessary for receptor activity but rather appears to have a modulatory role. Indeed, in the context of a C-terminal truncated peptide, residues in this region cause a significant decrease in receptor activity (compare peptides 1-7, 1-8, and 1-10 to WT and 1-6 in Figs. 3 and 5 and Table 2), suggesting that this domain has an inherent inhibitory function. Residue 7 is particularly central to this role, because peptide 1-7 is particularly less potent than either the WT peptide or peptide 1-6 (see Fig. 3 and Table 2), and peptide G7A is almost completely lacking in activity on the NPSR-A receptor (see Fig.  4 and Table 2). Because these effects are not observed (or much reduced in the case of the G7A mutant) with the N107I variant of NPSR-A (see Fig. 5 and Table 2), our results are consistent with an interaction, direct or indirect, between residue 107 of NPSR and residue 7 of NPS that limits receptor activation in NPSR-A but not in NPSR-A-N107I. Interestingly, presence of the C-terminal residues of NPS reverses, at least partially, the unfavorable interactions of shorter peptides with NPSR-A, perhaps by allowing formation of the structured helical region described above. One possible structural explanation for the striking difference in activation of the two NPSR variants by peptide 1-7 is that there is electrostatic repulsion between the two residues in the WT variant of NPSR, which becomes more prevalent when Gly-7 is mutated to the more hydrophobic Ala. The N107I variant, on the other hand, might be predicted to interact favorably with Gly-7 due to the hydrophobic nature of Ile, with only a slight steric disruption of the interaction in the presence of the moderately larger side chain of an Ala residue in the G7A mutant peptide.
As first demonstrated by Reinscheid and co-workers (14) and confirmed herein (Fig. 6, Table 1), mutation of residue Asn-107 of NPSR to Ile leads to an increase in the potency of NPS for the receptor. The use of a transient expression system has allowed us to demonstrate that in addition to this increase in potency, the maximal efficacy (E max ) of NPS is also increased on the N107I variant compared with the WT variant. We also show that this increase in E max is likely secondary to a significantly higher level of cell surface receptor density of the mutant compared with the WT receptor. Thus, as demonstrated by immunohistochemistry and ELISA (Fig. 7), the cell surface expression of NPSR-A-N107I is significantly higher than that of NPSR-A. In addition, artificially altering receptor expression by transfecting cells with different amounts of NPSR vector DNA leads to changes in E max (Fig. 8). On the other hand, the decrease in EC 50 observed with the N107I variant is likely related not to cell surface expression of NPSR, but rather to differences in the intrinsic potency of NPS on its receptor, because increasing the expression of NPSR as described above does not significantly affect EC 50 values (Fig. 8). In addition, the increase in potency is not associated with a change in the affinity of NPS for NPSR ( Fig. 6B; see also Ref. 14), but appears to be due to a difference in the intrinsic efficacy of NPS, which is its ability to activate the receptor once it is bound. Mechanistically, it is tempting to speculate that this is related to the interaction between residue 107 of the receptor and the helical regulatory region of NPS, in particular residue Gly-7, as described above.
In conclusion, we have identified key residues in the NPSR receptor and NPS ligand involved in receptor activation. These results provide a foundation for future structural and functional studies on this novel receptor and define the molecular basis for the increased asthma susceptibility associated with the N107I variant of NPSR.