Substance P (Neurokinin-1) Receptor Is a Marker of Human Mucosal But Not Peripheral Mononulear Cells: Molecular Quantitation and Localization¹

There is increasing evidence for a bidirectional communication between the nervous system and the immune system. Intersystem cross-talk is mediated by common soluble factors such as neuropeptides and cytokines interacting with specific receptors present on both immune and neuroendocrine cells. Substance P (SP),³ one of the most extensively characterized neuropeptides, has been implicated in neuroimmune modulation (1, 2).

SP is an 11-amino acid peptide that belongs to the mammalian tachykinin family, other members of which include substance K (neurokinin A) and neuromedin K (neurokinin B) (3). SP is widely distributed in the central, peripheral, and enteric nervous systems. In accordance with its wide tissue distribution, SP participates in the regulation of diverse physiologic activities such as neurotransmission of nociception, contraction of smooth muscle, vasodilatation, and secretion from exocrine and endocrine glands (4).

The microanatomic juxtaposition of SP-producing nerve fibers with immune-effector cells in primary and secondary lymphoid organs and in gut-associated lymphoid tissue, lends anatomic support to the role of SP as an immune modulator (5, 6). As a proinflammatory agent, SP has been reported to stimulate T cell proliferation (7, 8); enhance Ig synthesis (9, 10, 11); act as a chemoattractant for monocytes, neutrophils, and fibroblasts (12, 13, 14); induce mast cell degranulation and histamine release (15); stimulate phagocytosis of macrophages (16); and induce the secretion of IL-1, IL-6, and TNF-α from monocytes (17). Evidence that macrophages and eosinophilic granulocytes produce and secrete SP has also been reported (18, 19).

To date, three types of tachykinin receptors have been identified, namely neurokinin-1 receptor (NK-1R), NK-2R, and NK-3R. SP has a high affinity for NK-1R (SPR) and binds to NK-2R with lower affinity (20). Sequence analysis of the cloned neurokinin receptors reveals that they belong to the superfamily of G protein-coupled receptors, and act primarily through the activation of phospholipase C, leading to phosphatidylinositide breakdown and intracellular accumulation of calcium (21).

A requirement for understanding the function of an immunomodulator is to identify the cell targets that express its receptor. Binding studies by Payan et al. have shown the presence of SP binding sites on a subset of human T lymphocytes and on the IM-9 B lymphoblastoid cell line (22, 23). Mantyh et al. have autoradiographically demonstrated SP binding sites in the germinal centers of lymph nodules in resected colonic tissue (24). In contrast to these findings, Roberts et al. dispute the presence of SP receptors on human lymphocytes in peripheral blood, spleen, or intestinal mucosa. Using radioligand-binding studies, they report the absence of SP binding sites on PBL, polymorphonuclear leukocytes, splenocytes, intraepithelial lymphocytes, or jejunal lamina propria lymphocytes (25). Recent evidence suggests that human monocytes express a non-neurokinin receptor for SP (26). It has also been reported that SP can activate T lymphocytes receptor independently (27). The evidence for SPR on human lymphocytes is, therefore, controversial.

The aim of the present study was to clarify whether SPR is expressed by human systemic and mucosal immunocytes. In particular, we wished to identify cellular sites of SPR expression within human colonic mucosa and its associated lymphoid tissue. SPR expression was analyzed directly and specifically in isolated lymphoid cells by RT-PCR and flow cytometry. Direct, specific localization of SPR expression within the colonic mucosa was achieved at the mRNA and protein level by both in situ hybridization and immunohistochemistry, respectively.

Peripheral blood specimens were obtained from healthy volunteers. Colonic tissue was obtained from patients undergoing surgical resection for colorectal adenocarcinoma. Only histologically normal tissue from an uninvolved area of the colon was used. All protocols were approved by University Teaching Hospital’s Ethics Committee (Cork, Ireland).

PBMC were isolated from heparinized blood by centrifugation over Ficoll-Hypaque gradients (Sigma, St. Louis, MO) (28). Cells were recovered, washed twice in DMEM, and resuspended in DMEM supplemented with 10% FCS.

Lamina propria mononuclear cells (LPMC) were isolated by a technique originally described by Bull and Bookman (29). The entire procedure was performed in 6 h without interruption. The colonic specimen was washed in saline, and adherent debris was removed. The mucosa was stripped free from underlying musculature and was cut into small pieces (1 × 3 cm). To remove epithelium, tissue was incubated in a shaking water bath at 37°C for 30 min in calcium-magnesium-free HBSS containing 1 mM EDTA and 50 μg/ml gentamicin. Penicillin (100 U/ml), streptomycin (100 μg/ml), and fungizone (0.25 μg/ml) were also added as a 1% fungi-bact solution. The incubations were repeated every 30 min with fresh medium until the supernatant was free of epithelial cells.

The remaining tissue was then minced into smaller pieces and digested in a shaking water bath at 37°C for 1 h with 0.5 mg/ml collagenase and 1 mg/ml hyaluronidase (Sigma) in RPMI 1640 containing 10% FCS, 2 mM l-glutamine, 1% fungi-bact solution, and 50 μg/ml gentamicin. The supernatant was collected, cells were pelleted by centrifugation (500 × g for 5 min), resuspended in the same medium without the enzyme solution, and then filtered through a 60-μm nylon mesh to remove particulate material. The filtrate was then centrifuged over a Ficoll-Hypaque density gradient. LPMC were recovered, washed twice in DMEM, and resuspended in DMEM supplemented with 10% FCS and antibiotics, as detailed above.

Two different activation procedures were used to activate PBMC. PBMC (1 × 10⁶/ml) were cultured with either PHA (3 μg/ml) or with PMA (10 ng/ml) plus ionomycin (500 ng/ml), in DMEM with 10% FCS at 37°C in a humidified 5% CO₂ atmosphere for 48 and 24 h, respectively.

Resting and PMA-activated PBMC (1 × 10⁶/ml) were incubated with SP (Bachem California, Torrance, CA) at graded concentrations ranging from 10⁻¹⁰ M to 10⁻⁶ M, in DMEM with 10% FCS at 37°C in a humidified 5% CO₂ atmosphere for 6, 24, and 48 h.

PBMC (1 × 10⁶/ml) were incubated with a mixture of proinflammatory cytokines containing IFN-γ, TNF-α, and IL-1β, each at a concentration of 10 ng/ml for 1, 4, 8, and 12 h. PBMC (1 × 10⁶/ml) were also stimulated with IL-2 (10 nM) and the chemokine IL-8 (1 nM) for 1, 4, 8, and 12 h, respectively. All cytokines were obtained from R&D Systems Europe (Abingdon, U.K.).

SP was dissolved in degassed sterile water containing 0.5% BSA, and adjusted to pH 3 with acetic acid. Stock solutions were aliquoted at 10⁻³ M in polypropylene microcentrifuge tubes and frozen at −70°C until use. Working concentrations were made up in polypropylene tubes that had been treated with 5% BSA in water for at least 1 h before use, to prevent peptide binding to the plastic.

Total RNA was isolated by phenol-chloroform extraction of guanidium isothiocyanate lysates (30). cDNA was synthesized using approximately 100 ng total RNA, 9 U AMV reverse transcriptase (Promega, Madison, WI), 40 U RNAsin (Promega), 500 μM dNTPs, and either 50 μM SPR-specific antisense primer GGATTTCATTTCCAGCCCCT or 125 pM random hexanucleotide primers (Boehringer Mannheim GmbH, Mannheim, Germany) per 30 μl reaction for 90 min at 42°C.

SPR PCR was performed on the specific-primed cDNA using the following sense and antisense primers, respectively: TGACCGCTACCACGAGCAAGTCTC and ATAGTCGCCGGCGCTGATGAAG corresponding to nucleotides 699–722 and 993–972 of human SPR cDNA, respectively. β-actin control PCR was performed on the random-primed cDNA using the following sense and antisense primers, respectively: CCTTCCTGGGCATGGAGTCCTG and GGAGCAATGATCTTGATCTTC corresponding to nucleotides 794–815 and 995–975 of human β-actin cDNA, respectively. PCR primers were designed using the DNASTAR Lasergene Primerselect program (DNASTAR, Madison, WI). Primers were selected that showed insignificant homology to any other genes in the EMBL DNA sequence database. Primer pairs were chosen to span introns in their genomic sequences, thus ensuring mRNA-specific amplification.

PCR was performed on 1% of the cDNA using a final concentration of 1.5 μM MgCl₂, 50 μM dNTPs, 0.1 μM each primer, and 1 U Taq DNA polymerase (Promega) per 50 μl reaction. Thermal cycling programs were as follows: SPR PCR, denaturation at 96°C for 15 s, annealing at 60°C for 30 s, and extension at 72°C for 1 min and 30 s for 40 cycles, followed by a final extension at 72°C for 10 min; β-actin PCR, denaturation at 96°C for 15 s, annealing at 55°C for 30 s, and extension at 72°C for 3 min for 35 cycles. Negative controls were performed by either omitting reverse transcriptase from cDNA synthesis or by omitting cDNA from the PCR amplifications. As a positive control, RNA from cells known to abundantly express SPR mRNA was used: IM-9 B lymphoblastoid cell line. Hot start PCR was employed to increase the specificity of the amplification. PCR products were analyzed by electrophoresis through 2% agarose gels and viewed under UV light after ethidium bromide staining. Product specificities were confirmed by DNA sequence analysis using an ABI Prism 310 Genetic Analyzer (Perkin-Elmer, Norwalk, CT). HaeIII-digested φX174 DNA size markers were used.

To facilitate quantitation of SPR mRNA, a competitive internal RNA standard was constructed (31). This standard is identical to target SPR sequence, except for an internal deletion of 71 bp. Construction of the competitive standard involved cloning a SPR PCR product (324 bp) into pBluescript (Stratagene, La Jolla, CA) at the EcoRV site. This SPR PCR product was obtained using the following sense and antisense primers, respectively: GACTCCTCTGACCGCTACCA and GGATTTCATTTCCAGCCCCT corresponding to nucleotides 691–710 and 1014–995 of the human SPR cDNA, respectively. Orientation of the PCR product insert within the plasmid was determined by restriction mapping with HinfI. Digestion at the BsrGI and BglII unique restriction sites within the cloned insert resulted in the deletion of a 71-bp fragment. Sticky ends of the plasmid were then filled in by Klenow DNA polymerase (Promega), and blunt-ended recircularization of the plasmid was performed using T4 DNA ligase (Promega). Following propagation in Escherichia coli, the deleted recombinant plasmid was subjected to in vitro transcription with T3 RNA polymerase (Promega) to synthesize deleted sense RNA transcripts (257 bp) for use as a competitive standard in qcRT-PCR.

For qcRT-PCR, varying amounts of RNA standard transcripts of known concentration were spiked into aliquoted target RNA sample, and the mixtures were then subjected to RT-PCR, as described above. As the internal standard is spiked in at the cDNA synthesis step, competition for both reverse transcription and PCR amplification occurs. Equivalence of PCR products occurs when target and standard templates are present in equal initial concentration, permitting quantitation of the target template. Equivalence was determined as the point at which target and competitive standard PCR products were of equal band intensity. Results for SPR mRNA quantitation are expressed as number of SPR mRNA molecules per microgram of total RNA isolated. Total RNA was quantified using a nucleic acid quantitation kit (Invitrogen, Leek, The Netherlands).

In situ hybridization was performed on paraffin-embedded surgically resected human colon sections (4 μm thick), mounted on aminopropylethoxysilane-treated slides. Following deparaffinization and rehydration, prehybridization treatments involved washing 2 × 5 min each in 1) PBS, 2) PBS, 0.1 M glycine, 3) PBS, 0.3% Triton X-100, and 4) PBS again. Sections were digested for 30 min at 37°C with proteinase K (10 μg/ml in 100 mM Tris-HCl, 50 mM EDTA, pH 8), fixed for 5 min at 4°C in 4% paraformaldehyde, PBS, and then acetylated for 2 × 5 min in fresh 0.25% acetic anhydride, 0.1 M triethanolamine (pH 8). Sections were incubated at 37°C for 10 min in a prehybridization buffer consisting of 50% formamide in 4× SSC. A digoxigenin-labeled RNA hybridization probe (324 bp) corresponding to codons 230–338 of the human SPR cDNA sequence was synthesized from a recombinant plasmid clone (also used to construct the competitive standard for qcRT-PCR) by in vitro transcription with digoxigenin-11-UTP and T7 RNA polymerase (Boehringer Mannheim). The nucleotide sequence of the SPR probe showed insignificant homology to any other sequences in the EMBL DNA sequence database. Hybridization was performed at 42°C overnight in hybridization buffer (50% formamide, 10% dextran sulfate, 1× Denhardt’s reagent, 4× SSC, 10 mM DTT, 500 μg/ml yeast tRNA, and 100 μg/ml heat-denatured herring sperm DNA) containing 1 ng/μl digoxigenin-labeled riboprobe. After hybridization, tissues were washed with increasing stringency to 0.1 × SSC at 37°C. Hybridized probe was detected immunologically using alkaline-phosphatase-conjugated sheep anti-digoxigenin Ab (Boehringer Mannheim) and visualized with NBT-BCIP (purple/black precipitating product). Control slides involved competitive inhibition of hybridization with a 10-fold excess of unlabeled riboprobe. This resulted in a marked reduction in signal intensity, thus confirming specificity of the hybridization.

The SPR Ab was raised in rabbits against a synthetic peptide (MDNVLPVDSDLSP) corresponding to the extracellular N-terminal amino acids 1–13 of human SPR. The IgG fraction was affinity purified on an AH-Sepharose column to which the peptide had been coupled. Ab specificity was confirmed by extensive RIA and Western blotting.

Immunohistochemistry was performed on paraffin-embedded surgically resected human colon sections (4 μm thick), mounted on aminopropylethoxysilane-treated slides. After deparaffinization and rehydration, sections were microwaved in 10 mM citrate buffer, pH 6, at 350 W for 5 min to assist Ag retrieval. Sections were immediately cooled in TBS (50 mM Tris-HCl, pH 7.6, 150 mM NaCl, and 5 mM KCl) and then washed twice for 5 min in wash buffer (TBS, 0.001% saponin). Endogenous peroxidase activity was quenched with 0.9% hydrogen peroxide in distilled water for 30 min. Sections were then washed in wash buffer containing 1% normal goat serum (used for all subsequent wash steps). Nonspecific binding sites were blocked by incubation in wash buffer containing 5% normal goat serum for 1 h. Sections were washed twice and then incubated overnight at 4°C with the rabbit polyclonal anti-human SPR-specific IgG at a dilution of 1/50 in 50 mM Tris-HCl, pH 7.6, and 1% normal goat serum. Ab binding was localized using a biotinylated goat anti-rabbit IgG, followed by avidin-biotin-conjugated horseradish peroxidase (Vectastain Elite ABC detection kit; Vector Laboratories, Burlingame, CA). Ab binding was visualized using 3′,3′ diaminobenzidine (Vector Laboratories) yielding a brown reaction product. Staining with isotype-matched rabbit IgG was performed as a negative control. A specificity control was performed involving preadsorption of SPR Ab with the immunizing peptide (100 μM) for 16 h at 4°C before staining of sections.

Sections were first processed for SPR immunohistochemistry, as detailed above. Following incubation with diaminobenzidine, sections were washed twice in TBS and then incubated for 1 h at room temperature with mouse anti-human CD45-specific IgG (Dako, Glostrup, Denmark) at a dilution of 1/70 in RPMI, 10% FCS, and 0.15 mM sodium azide. Ab binding was localized using rabbit anti-mouse Igs (Dako), followed by alkaline phosphatase/anti-alkaline phosphatase (APAAP) complexes (Dako). Ab binding was visualized using Fast Blue RR chromagen (Sigma) yielding a purple reaction product.

PBMC and LPMC isolations were fixed with 2% paraformaldehyde for 10 min on ice, permeabilized with 70% methanol for 2 min on ice, and then washed twice in PBS containing 2% FCS (used for all subsequent wash steps). A total of 5 × 10⁵ cells was incubated with rabbit polyclonal anti-human SPR-specific IgG (as used for immunohistochemistry) at a dilution of 1/50 for 30 min on ice. Cells were washed twice, and FITC-conjugated anti-rabbit IgG (Dako, Carpenteria, CA) was added for 30 min on ice. Cells were washed twice again and incubated with a mAb to either CD4, CD8, CD19 (R-phycoerythrin-Cy5 conjugated), CD45RO, CD45RA, or CD14 (R-phycoerythrin-conjugated) (Dako) for 30 min on ice, followed by two washes. Analysis was performed using an Epics Elite flow cytometer (Coulter, Hialeah, FL). Isotype-matched control Abs were used in negative control staining. A total of 10,000 cells was analyzed for each determination. Events contained within the forward/light scatter window corresponding to lymphocytes were selected and analyzed.

PBMC and LPMC isolations were analyzed for SPR mRNA expression by RT-PCR (Fig. 1). Results consistently show that isolated LPMC express SPR mRNA, whereas resting PBMC are negative for SPR mRNA expression (n = 10). When PBMC were activated with either PHA or PMA plus ionomycin, SPR mRNA expression was not induced. Incubation of resting or activated PBMC with SP did not induce SPR mRNA expression (n = 5). In addition, stimulation with a proinflammatory cytokine mixture containing IFN-γ, TNF-α, and IL-1β, with IL-2, or with the chemokine IL-8, did not induce PBMC expression of SPR mRNA (n = 5). The RT-PCR assay was capable of detecting 50 SPR mRNA molecules. The assay was controlled by equalization of input RNA for each cell isolation. Comparable amplification efficiencies were achieved in all RNA samples, as evidenced by the uniformity of control β-actin RT-PCR product yields. Negative control RT-PCRs were consistently negative.

A qcRT-PCR assay was developed to accurately quantify SPR mRNA expression. The accuracy, sensitivity, and reproducibility of the qcRT-PCR for SPR mRNA have all been validated in control experiments (31). The assay has been validated by reproducibly demonstrating that when equal amounts of competitor and target RNA are spiked into the cDNA synthesis step, an equivalence point is seen, whereby both target and competitor PCR products are of equal band intensity. The assay is sufficiently sensitive to quantitate 100 SPR mRNA copies (corresponding to 10³ SPR mRNA transcripts/μg RNA isolated).

Figure 2 illustrates the quantitation of SPR mRNA expressed in a particular LPMC isolation. Equivalence is seen at 5 × 10⁴ SPR mRNA molecules, in which target (295 bp) and competitive standard (228 bp) PCR products are of equal band intensity. When adjusted for the amount of total RNA in the assay, this represents a level of expression of 6.7 × 10⁵ SPR mRNA transcripts/μg RNA. The higher m.w. band of ∼360 bp seen on the gel is due to heteroduplex formation between target and competitor PCR products (confirmed by S1 nuclease digestion, data not shown). As the heteroduplex consists of a hybrid of one strand from target and competitor products, it does not bias the normal occurrence of equivalence (32). We confirmed this in control experiments: when equal known amounts of target and competitive standard RNA transcripts were mixed before qcRT-PCR, product equivalence was always obtained, independently of heteroduplex formation.

The mean level of SPR mRNA expression found in LPMC isolations was 7.7 × 10⁵ ± 1.2 × 10⁵ SPR mRNA transcripts/μg RNA (n = 5). Since 10⁵ cells yielded ∼1 μg cellular RNA, this level of expression is equivalent, on average, to ∼7.7 ± 1.2 SPR transcripts/cell.

While LPMC isolations were obtained routinely with a high lymphoid content (>95% CD45 positive), the possibility of contamination by epithelial and neuroendocrine cells could not be excluded. In situ hybridization was therefore used to examine SPR mRNA expression in LPMC in vivo. As shown in Figure 3, in situ hybridization localized SPR mRNA to mucosal lymphoid cells. Expression was evident in LPMC and in lymphoid follicles. SPR mRNA was also detected in both surface and crypt epithelium. It was observed that lymphoid follicles within the same colonic section varied in their expression of SPR mRNA. SPR mRNA expression by individual LPMC was also variable, but LPMC were predominantly positive. Hybridization specificity was confirmed, as a 10-fold excess of unlabeled riboprobe resulted in a dramatic reduction in hybridization signal intensity.

SPR protein was detected immunohistochemically in LPMC and in lymphoid follicles of resected colonic sections (Fig. 4). SPR protein expression was also evident in surface and crypt epithelium. SPR specificity of Ab detection was confirmed as staining was competitively inhibited by preincubation of the primary Ab with the SPR-immunizing peptide. The pattern of SPR protein expression detected by immunohistochemistry closely matched that of SPR-mRNA expression detected in consecutive sections by in situ hybridization. Colocalization of SPR mRNA and protein to LPMC and epithelial cells confirms that these cells indeed express SPR.

To conclusively verify that LPMC express SPR, dual immunohistochemical staining was performed for the simultaneous detection of SPR and the leukocyte common Ag (CD45) in resected colonic sections. SPR-immunoperoxidase (brown) staining colocalized with CD45-immunophosphatase (purple) staining, resulting in a brown/purple double stain in LPMC and lymphoid follicles (Fig. 5). This colocalization of SPR with CD45 confirms that CD45-positive LPMC coexpress SPR. Although some individual cells were singly stained for CD45 (purple), the majority of LPMC were doubly stained, indicating predominant SPR expression by LPMC.

Dual-fluorochrome flow cytometry was used to analyze SPR expression levels in PBMC and in distinct LPMC subpopulations. A polyclonal Ab specific for SPR was used in conjunction with specific mAbs that define distinct lymphocyte subsets, thereby enabling characterization of both SPR distribution and expression levels on the different lymphoid cell types. CD4 mAb was used to identify helper/inducer T cells, CD45RO for memory T cells, CD45RA for naive T cells, CD8 for cytotoxic/suppressor T cells, CD19 for B cells, and CD14 for macrophages. The fluorescence profile of SPR expression in each LPMC subset is shown in Figure 6.

SPR immunofluorescence was not detected in PBMC, but was detected in all LPMC subpopulations examined, with varying degrees of intensity. The majority of SPR-expressing cells (60–70%) corresponded to CD4 cells. The fluorescence profile of CD4 cells, relative to that of the other subsets examined, showed a broad range of SPR fluorescence intensities. This is indicative of both high and low levels of SPR expression within the CD4 population.

SP is known to exert a spectrum of immunomodulatory effects. Despite the substantive evidence for SP as an immunomodulator, evidence for expression of SPR on lymphocytes has not been conclusive. Binding studies by Payan et al. have shown the presence of SP binding sites on a subset of human T lymphocytes (22). In contrast, binding studies by Roberts et al. dispute the presence of SP binding sites on human lymphocytes in peripheral blood (25). Recent evidence suggests that SP can activate T lymphocytes receptor independently (27). It has also been reported that human B lymphocytes and monocytes express a non-neurokinin receptor for SP (26, 33).

The conflicting results regarding the presence of SP receptors on lymphocytes could be due to technical difficulties inherent to binding studies, as suggested by Roberts et al. (25). Other factors, such as the internalization of receptor, yield false-negative results in binding studies. Rapid agonist-induced endocytosis of SPR has been observed both in vivo and in vitro (34, 35). In addition, SP can bind with a lower affinity to many non-neurokinin receptors. This poses the additional problem of false-positive results in binding studies. Binding of SP to the serpin-enzyme complex receptor on hepatocytes and monocytes has been reported (36). Certain NK-1 antagonists are known to bind to the bombesin receptor (37), while the chemotactic activity of SP on rabbit neutrophils has been attributed to binding of SP to the FMLP receptor (38).

Clearly, determination of SPR expression in immunocytes demands the use of more direct, specific methods of detection. In this study, we present a molecular approach to investigate the expression and cellular localization of SP receptors in human peripheral and mucosal lymphocytes. Using the highly sensitive technique of RT-PCR, we demonstrate that human PBMC do not express SPR mRNA. This finding is in agreement with that of Roberts et al. (25). It would also seem to suggest that the receptor-independent mechanisms of action of SP on T lymphocytes and the binding of SP to a non-neurokinin receptor on B lymphocytes and monocytes reported by Kavelaars et al. (26, 27, 33), although operable at high nonphysiologic concentrations, may indeed be the only mechanistic theory of SP relevant to PBMC, to date. Recently, NK-1 mRNA expression was detected in murine T lymphocytes (39). It would therefore seem that the expression of SP receptors by PBL is species specific.

The evidence for expression of SP receptors by mucosal lymphocytes in vitro (25, 40) and in vivo (24, 41, 42) has also been conflicting. Using the technique of RT-PCR, we found that LPMC isolated from resected human colonic tissue express SPR mRNA. Quantitation of SPR mRNA expression in LPMC isolations by qcRT-PCR revealed a mean level of expression of 7.7 ± 1.2 SPR transcripts per cell. This level of expression is consistent with previous reports concerning G protein-coupled receptors, which are generally associated with low abundance mRNA. For example, the β₂-adrenergic receptor has been shown to express 15 transcripts per cell (43). Our finding of SPR mRNA expression by LPMC in vitro is at variance with the observation of Roberts et al. (25). It should, however, be noted that the isolation procedure for LPMC involves lengthy enzymatic treatments that may influence surface expression of receptors, and thus interfere with binding studies. This further illustrates the advantage of a molecular approach. The fact that PBMC put through the LPMC isolation process do not express SPR mRNA (unpublished observation) rules out the possibility that the LPMC isolation procedure itself is inductive of receptor expression.

Due to the fact that LPMC preparations could contain contaminating cells of epithelial and neuroendocrine origin, the technique of in situ hybridization was used to investigate SPR expression in LPMC in vivo. Examination of resected human colonic tissue revealed localization of SPR mRNA to LPMC throughout the lamina propria, to lymphoid follicles, and to epithelial cells. Regarding SPR mRNA expression in lymphoid follicles, it was observed that follicles within the same colonic section varied in their expression of SPR mRNA, suggesting that SP exerts its effects locally. SPR mRNA expression by individual LPMC was also variable, but LPMC were predominantly positive. This receptor distribution was reproduced in parallel sections processed for SPR immunohistochemistry, in which SPR immunoreactivity was detected in LPMC, lymphoid follicles, and epithelial cells. Colocalization of SPR mRNA and protein to LPMC and epithelial cells confirms that these cells indeed express SPR.

When colonic sections were treated simultaneously for both SPR and CD45 immunohistochemistry, both immunoreactivities colocalized. Colocalization of SPR with the leukocyte common Ag (CD45) confirms LPMC expression of SPR. The distribution pattern of SPR expression in lamina propria (which contains predominantly T cells) and lymphoid follicles (which contain predominantly B cells) suggests that both T and B mucosal lymphocyte subsets express SPR. Dual-fluorochrome flow-cytometric analysis of LPMC preparations revealed that SPR expression was not subset restricted. SPR expression was detected in helper and cytotoxic T cells of both the memory and naive phenotypes, in B cells, and in macrophages. The CD4 subset exhibited a broad range of SPR fluorescence intensities, indicating both high and low SP receptor density on CD4 cells. This observation may be indicative of CD4 cells that have differential sensitivity to SP, and may therefore have functional significance.

It is noteworthy that two isoforms of the NK-1R have been identified in the human brain (44). The two isoforms differ only in the length of their carboxyl-terminal tail. The primers we have chosen for RT-PCR and the riboprobe we use for in situ hybridization are specific for the long isoform of the SPR. The SPR Ab is specific for the extracellular N terminus, and accordingly detects both the long and the short isoforms. Due to the fact that the SPR mRNA distribution pattern detected by in situ hybridization closely matched SPR protein distribution detected by immunohistochemistry, it is unlikely that there is disparate distribution of the short isoform of SPR in human colonic mucosa.

It is interesting that we find that in contrast to peripheral lymphocytes, mucosal lymphocytes express SPR. Lymphocytes from lymphoid organs are considered to be in a state of activation relative to the resting population of peripheral blood (45). Activation of lymphocytes is usually accompanied by an increase in intracellular calcium. One could postulate that expression by LPMC of SPR is triggered via the calcium-inducible promoter sequence in the SPR gene. However, when PBMC were cultured with either PHA (mitogen) or PMA (activator of protein kinase C) plus ionomycin (Ca²⁺ ionophore), SPR expression was not induced.

Cytokines play a pivotal role in mucosal immune function. Both mucosal lymphoid and epithelial cells have the capacity for cytokine production (46). Expression of neuropeptide receptors by lymphocytes can be cytokine inducible (47). Indeed, SPR mRNA expression by bone marrow stroma has been shown to be induced by IL-1 and stem cell factor (48). However, when we stimulated PBMC with physiologic concentrations of cytokines normally present in the mucosal environment; namely IFN-γ, TNF-α, and IL-1β, IL-2 and IL-8SPR mRNA expression were not induced.

Another possible reason as to why LPMC express SPR could be the fact that their local microenvironment of gastrointestinal mucosa is rich in SP-producing neurons. However, incubation of PBMC with SP did not induce SPR expression. These findings lead us to the conclusion that SPR expression is distinctive of human mucosal mononuclear cells, and not of peripheral mononuclear cells. Indeed, the mucosal microenvironment rich in immunocytes and SP would seem to be the appropriate location for interactions between the nervous system and the immune system. Our finding that human LPMC express SPR provides the evidence for such intersystem communication and supports a direct role for SP in mucosal immunomodulation.

Evidence for the NK-1 receptor in monocytes using nested RT-PCR has recently been reported (49).

We thank Dr. Gary Lee (Department of Pathology, Mercy Hospital, Cork, Ireland) for providing colonic tissue sections; Fiona O’Brien, Bernie Crowley, and Maurice O’Donoghue for assistance with flow cytometry; and Jim O’Callaghan for technical assistance. We also thank the CURE peptide synthesis, Ab, and microscopic imaging Cores and Dr. Joe Reeve.