Introduction

Top Abstract Introduction Materials & Methods Results Discussion References

The immune and nervous systems, once viewed as isolated, independent systems, are now accepted to be interdependent, continuously influencing the functioning and responsiveness of each other. Substantial evidence has demonstrated that the mode of communication between the two systems is governed by various mediators, such as neurotransmitters (Felten et al., 1987; Ignatowski and Spengler, 1995), neuropeptides (Shahabi and Sharp, 1995; Sibinga and Goldstein, 1988) and cytokines (Ignatowski and Spengler, 1994). These mediators participate in a neuroimmune network directing cellular regulatory mechanisms of immune cells and nerves, thereby potentially influencing neural, inflammatory and immunological processes.

In particular, opiates and opioid peptides, previously regarded as exclusively neural-derived mediators, have been demonstrated to function in an immunomodulatory fashion. In vitro studies demonstrated either enhancement or suppression of immune functions, such as lymphocyte proliferation (Hucklebridge et al., 1990; Puppo et al., 1985; Shahabi and Sharp, 1995), antibody production (Taub et al., 1991), monocyte chemotaxis (Perez-Castrillon et al., 1992; Van Epps and Saland, 1984) and macrophage phagocytosis (Casella et al., 1991; Prieto et al., 1989) depending on the concentration and/or class of opioid peptide used, as well as the type and/or activation/differentiated status of the effector cell monitored. Furthermore, all three classes of opioid peptides, mu, delta, kappa, although shown to affect various immune cell functions, have been localized constitutively within these immune cells (Lolait et al., 1986; Martin et al., 1987; Smith and Blalock, 1981). An endogenous autocrine or paracrine role for the presence of these peptides has been confirmed by molecular evidence for opioid receptor mRNA (Chuang et al., 1994; Gaveriaux et al., 1995; Sedqi et al., 1995), which strongly suggests that these neural receptors are present on cells of the immune system. However, radioligand binding studies directed at detecting these putative receptors have been inconclusive thus far (Sibinga and Goldstein, 1988). Although many reports exist for naloxone-sensitive opioid binding to lymphocytes (Madden et al., 1987; Mehrishi and Mills, 1983; Ovadia et al., 1989) and leukocyte cell lines (Carr et al., 1989, 1991; Dobrenis et al., 1995; Fiorica and Spector, 1988; Makman et al., 1995), these findings are not characteristically consistent with those reported for brain opioid receptors. Classic properties such as high-affinity binding, stereoselectivity and inhibition of binding by opioids and opioid peptides has not been attained, which suggests that immune cell binding sites for opioids may be different than the opioid sites in the central nervous system. However, previous studies from our laboratory have demonstrated and characterized brain-like, kappa opioid receptors on the R1.1, R1.G1, and R1EGO mouse thymoma cell lines by radioligand binding (Bidlack et al., 1992; Lawrence and Bidlack, 1992; Lawrence et al., 1995b) and kappa opioid-mediated inhibition of adenylyl cyclase activity (Lawrence and Bidlack, 1993; Lawrence et al., 1995b). In addition, the mRNA for the kappa opioid receptor in the R1.1 thymoma cells line has been reported (Belkowski et al., 1995). Although these findings established that cells from the immune system can express a brain-like opioid receptor, questions still remained regarding the presence of opioid receptors on normal immune cells. The disparity in the findings concerning opioid binding studies may have been caused by a lack of sensitivity of the methods used for the detection of opioid receptors, which may be few in number on cells of the immune system, or which may be differentially expressed only on selective subsets of heterogenous immune cell populations.

An alternative approach for the identification of receptors was established through the synthesis of fluorescently conjugated ligands for receptor labeling. Although several reports described labeling of histamine receptors on immune cells by this method (Muirhead et al., 1985; Osband et al., 1980), fluorophore-conjugated, high-affinity opioid ligands were ineffective in the specific labeling of cells/tissues (Kolb et al., 1983, 1985). In a related approach, successful labeling of the kappa opioid receptor on the R1EGO thymoma cell line, which expresses the kappa opioid receptor as measured by radioligand binding (Lawrence et al., 1995b), and on freshly isolated murine thymocytes has been reported by our laboratory through the use of an indirect immunofluorescence method and flow cytometric analysis (Lawrence et al., 1995a, 1997). Detection of PE fluorescence by an avidin-biotin amplification procedure allowed for the specific labeling of the kappa opioid receptor by a fluorescein-conjugated arylacetamide ligand, FITC-AA. Specific opioid labeling was undetectable for either R1EGO cells or thymocytes without this amplification procedure (Lawrence et al., 1997).

The present study was undertaken to determine the presence and/or absence of the kappa opioid receptor on specific subpopulations of mouse thymocytes. Multiparameter analysis capabilities of flow cytometry allowed for the simultaneous detection of cells possessing the kappa opioid receptor and various CD markers, which were used to determine immune cell phenotype. Thymocytes from C57BL/6ByJ mice had 55 ± 4% specific labeling for the kappa opioid receptor, as measured by the inclusion of the kappa selective antagonist, nor-BNI. Further analysis demonstrated that these thymocytes constitute a preponderance of double-positive, CD4⁺/CD8⁺, cells of which approximately 60% were CD3 positive. These findings, therefore, further elucidate that thymocytes of immature phenotype possess the kappa opioid receptor in an environment where opioids are known to be produced, possibly orchestrating many critical developmental and/or differentiation processes (Dardenne and Savino, 1994; Linner et al., 1995, 1996).

	Materials and Methods

Top Abstract Introduction Materials & Methods Results Discussion References

Cell culture. R1E/TL8x.1.G1.OUA^r.1 (R1EGO) cells, a derivative of the R1.1 thymoma cell line, were purchased from the American Type Culture Collection (Rockville, MD). Cells were cultured in RPMI 1640 (Gibco BRL, Grand Island, NY) buffered with 12.5 mM HEPES (pH 7.2) and containing 10% (vol/vol) iron-supplemented bovine calf serum (Hyclone, Logan, UT), 300 µg/ml L-glutamine, 100 µg/ml streptomycin, 100 U/ml penicillin, 50 µM 2-mercaptoethanol and 60 µM 2-aminoethanol.

Murine thymocyte preparation. Male C57BL/6ByJ mice, aged 6 to 8 weeks (Jackson Laboratories, Bar Harbor, ME), were used for all studies and were given access to food and water ad libitum. Mice were sacrificed by CO₂ inhalation, and thymi were removed aseptically as described by Mishell et al. (1980). Thymocytes were dissociated by pressing thymi gently between the frosted ends of sterile microscope slides in ice-cold HEPES-buffered balanced salt solution (HEPES-BSS), consisting of 15 mM HEPES, 3.4 mM K₂HPO₄, 0.6 mM KH₂PO₄, 150 mM NaCl, 5 mM KCl, 2.5 mM CaCl₂, 1.2 mM MgSO₄ and 1% (w/v) bovine serum albumin, pH 7.4. Cell suspensions were passed over sterile, glass-wool columns to remove dead cells and debris. After centrifugation at 200 × g for 10 min at 4°C, cell suspensions were treated as necessary with an isotonic ammonium chloride solution to lyse contaminating erythrocytes. Unfixed cells, washed twice by centrifugation at 200 × g for 10 min at 4°C and counted in a hemacytometer, were resuspended at a final concentration of 2 × 10⁶ cells in 200 µl HEPES-BSS per sample for optimal staining of the kappa opioid receptor as described below.

Indirect immunofluorescence. The conjugation of FITC-AA has been described previously (Lawrence and Bidlack, 1993; Lawrence et al., 1995a). In a final volume of 200 µl HEPES-BSS, cells were incubated with 10 to 200 µM FITC-AA for 30 min at 25°C for determination of optimal kappa opioid receptor staining. The kappa selective antagonist nor-BNI (Portoghese et al., 1987) was also titrated (50-1000 µM) to determine the optimal concentration necessary for measurement of nonspecific fluorescence. A final concentration of 500 µM for both mu (DAMGO and morphine) and delta (ICI 174,864 and DPDPE) opioids (dissolved in sterile distilled water) was included to determine receptor specificity. Samples were chilled on ice, diluted with 1 ml HEPES-BSS and centrifuged at 400 × g for 3 min at 4°C. After aspirating the supernatants, cells were washed twice and resuspended in a final volume of 100 µl of HEPES-BSS, including 10 µl of biotinylated rabbit antifluorescein IgG (Molecular Probes, Eugene, OR), except in FITC-AA-only and PE-only controls. After incubation for 30 min at 4°C in the dark, samples were diluted with 1 ml HEPES-BSS and centrifuged at 400 × g for 3 min at 4°C. The supernatants were aspirated, and the cells were washed again. Cells were resuspended in 40 µl HEPES-BSS and 10 µl of extravidin-R-PE (Sigma Chemical Co., St. Louis, MO) for 15 min at 4°C in the dark. The cells were washed twice as described above and were resuspended in a final volume of 1 ml HEPES-BSS for flow cytometric analysis. Phenotypic characterization of thymocytes possessing the kappa opioid receptor was undertaken as above with minor modification. Fluorophore-conjugated, rat mAbs directed against the following mouse cell surface markers: FITC-CD4, QR-CD3, QR-CD4 and QR-CD8 obtained from Sigma Chemical Co. (St. Louis, MO) were assayed for optimal labeling before use. Optimal amounts of QR-CD3, -CD4 or -CD8 were added to appropriate tubes during the second 30-min incubation, followed by the subsequent steps listed above. For phenotypic analysis alone, thymocytes (1 × 10⁶ cells/100 µl HEPES-BSS) were incubated with 4 µl of QR-CD3, -CD8 or FITC-CD4 and QR-CD3/FITC-CD4 or QR-CD8/FITC-CD4 for 30 min at 4°C in the dark. Samples were diluted with 1 ml HEPES-BSS and were centrifuged at 400 × g for 3 min at 4°C. After aspiration of the supernatants, cells were washed two additional times, resuspended in 1 ml HEPES-BSS and analyzed for fluorescence labeling. Controls consisted of unstained thymocytes (autofluorescence controls), FITC-AA-only stained thymocytes, PE-only stained thymocytes and thymocytes stained with appropriate fluorophore-conjugated, isotype-matched control mAbs (nonspecific staining controls).

Flow cytometric analysis. Samples were analyzed on a Becton Dickinson FACScan (San Jose, CA) equipped with a 15-mW argon-ion laser for excitation (488 nm) of FITC and PE with band pass filters of 530 ± 15 nm and 585 ± 21 nm, respectively. The red fluorescence channel (FL3) for QR (which emits at 670 nm) uses a long-pass emission filter which transmits long-wavelength light beams above 650 nm. In each sample, 10,000 R1EGO cells or 25,000 thymocytes were analyzed. Data [forward angle and right angle (side) light scatter, as well as green (FITC), orange (PE) and red (QR) fluorescence] were measured, collected and stored into list mode data files by a Macintosh Quadra 650 computer system with CELLQuest software (Becton Dickinson Immunocytometry Systems, San Jose, CA). Median peak values of relative fluorescence intensity distributions were used to compare the fluorescence among samples, which assumed that only a single population was labeled. Unlabeled cells (autofluorescence) and cells labeled with only FITC-AA were used as negative controls to ensure that fluorescein did not contribute to the PE signal as measured in the PE fluorescence channel (FL2). This spectral overlap was corrected by adjusting the electronic compensation such that the intensity of cells labeled with only FITC-AA, as measured in the FL2 channel, was identical with that of the autofluorescence controls. Background controls, consisting of cells incubated with only biotin-conjugated, antifluorescein IgG and extravidin-R-PE, were used to reestablish base-line PE emission by taking into account the nonspecific staining of both the antifluorescein IgG and the extravidin-R-PE. PI (Sigma Chemical Co., St. Louis, MO) at a concentration of 5 µg/ml was added to each sample (not containing PE or QR) before analysis for live/dead cell discrimination (Loken and Stall, 1982). Viabilities were >95% as assessed by PI exclusion. In samples containing PE or QR, PI as a viability stain was impeded by spectral overlap. In these cases, light scatter gating provided reasonable live/dead cell discrimination because <5% PI positive cells were detected within the applied gates in samples stained with PI only. To report a relative quantification for the thymocyte subsets identified by mAb labeling, list mode data were analyzed with gate 1 (R1) set to enclose thymocytes; lysed erythrocytes and dead cells were excluded from analysis, because mAbs and ligands bind to/enter dead cells contributing to background fluorescence (Loken and Stall, 1982). With gate R1 set, the QR-fluorescence histogram was displayed for the thymocytes, along with the corresponding dot plot demonstrating QR-fluorescence versus FSC. The cells labeled positively for each particular QR-mAb were gated on (designated R2 in the QR vs. FSC dot plot and R3 in the QR histogram) and viewed in the PE histogram for dual kappa opioid receptor labeling. Appropriate PE histograms were then overlaid to demonstrate both background and specific labeling for the kappa opioid receptor. The relative number of kappa opioid receptor positive/QR-mAb positive stained cells from R1 gated thymocytes was then enumerated by histogram subtraction with CELLQuest software. The percent specific fluorescence for labeling of the kappa opioid receptor was calculated by the relative median phycoerythrin fluorescence intensity values as: [(Total fluorescence  background)  (Fluorescence with nor-BNI  background)/(Total fluorescence  background)] × 100.

	Results

Top Abstract Introduction Materials & Methods Results Discussion References

Kappa opioid receptor expression on freshly isolated thymocytes. FITC-AA bound with high affinity to the kappa opioid receptor in both guinea pig brain and R1.1 thymoma cell membranes (Lawrence et al., 1995a). The R1EGO cell line, a murine thymoma cell line which expresses kappa but not mu or delta opioid receptors (Lawrence et al., 1995b), served as a positive control for fluorescence labeling. R1EGO cells, incubated with 30 µM FITC-AA followed by biotinylated-antifluorescein IgG and extravidin-R-PE, demonstrated specific kappa opioid receptor labeling (83 ± 5%), defined as PE fluorescence inhibited by nor-BNI (table 1). Thymocytes from 6- to 8-week-old C57BL/6ByJ male mice also displayed 55 ± 4% specific labeling (table 1), which correlates well with a previous report (Lawrence et al., 1995a). Figure 1 illustrates: 1) the gating applied in the forward versus side scatter dot plots, designated gate R1, for all thymocyte analyses (fig. 1A) and 2) the PE fluorescence histograms of the FITC-AA/PE-stained cells, showing both the comparison of background to total and nonspecific fluorescence (fig. 1B), as well as specific fluorescence (defined as the difference between the total and nonspecific histograms) in the absence of background staining (fig. 1C) for labeling of the kappa opioid receptor. The relative percentage of thymocytes stained for the kappa opioid receptor was determined to be 58 ± 2% of the R1 gated population (100%) as per histogram subtraction. The lack of inhibition for FITC-AA/PE labeling (as compared with nor-BNI) by both delta (ICI 174,864 and DPDPE) and mu (DAMGO and morphine) selective opioids further demonstrates specificity of labeling for the kappa opioid receptor (fig. 2).

View this table: [in this window] [in a new window]   TABLE 1 Kappa opioid receptor labeling on R1EGO cells and mouse thymocytes Kappa opioid receptor labeling was carried out as described under "Materials and Methods." Autofluorescence control intensities were 4.3 ± 0.2 arbitrary units (a.u.) for R1EGO cells and 1.6 ± 0.1 a.u. for gated (R1) thymocytes. Background staining (biotinylated antifluorescein IgG/PE in the absence of FITC-AA) was 124 ± 26 a.u. for R1EGO cells and 2.9 ± 0.1 a.u. for thymocytes. Data represent the mean ± S.E.M. of relative median peak fluorescence intensity values from three to seven experiments performed in triplicate, with the background fluorescence subtracted. Percent specific fluorescence was calculated by relative median fluorescence values as:

View larger version (19K): [in this window] [in a new window]   Fig. 1.   Labeling of kappa opioid receptors on mouse thymocytes. (A) Gating in the forward (FSC) vs. side (SSC) scatter dot plot along with PI exclusion was used to discriminate thymocytes from erythrocytes and dead cells. Gate 1 (R1) includes the cells of interest for subsequent analysis. (B) Total fluorescence for kappa opioid receptor labeling is depicted by overlaid PE fluorescence histograms; the relationship of background staining to that of PE fluorescence in the absence and in the presence of the kappa opioid receptor antagonist, nor-BNI, is shown. (C) Specific labeling of the kappa opioid receptor as demonstrated by overlaid histograms with the background subtracted. Data represent seven separate experiments.

View larger version (15K): [in this window] [in a new window]   Fig. 2.   Specificity of FITC-AA/PE labeling for the kappa opioid receptor. The kappa selective opioid antagonist nor-BNI was compared with the delta selective peptides ICI 174,864 and DPDPE and with the mu selective opioids DAMGO and morphine for inhibition of FITC-AA/PE labeling of thymocytes. All compounds were used at a final concentration of 500 µM. Refer to the legend in table 1 for the percent specific labeling formula. Data are the mean ± S.E.M. from three experiments, performed in triplicate.

View larger version (16K): [in this window] [in a new window]   Fig. 3.   Titrations of FITC-AA and nor-BNI for optimal kappa opioid receptor labeling. (A) Thymocytes were incubated with 10 to 200 µM FITC-AA followed by the amplification procedure for labeling. Data are presented as median peak PE fluorescence intensity values (in arbitrary units). Data are the mean ± S.E.M. from four experiments. (B) Thymocytes were incubated with 30 µM FITC-AA and 50 to 1000 µM nor-BNI followed by the amplification procedure. Data are the mean ± S.E.M. from four experiments demonstrating a concentration-dependent increase in specific labeling for the kappa opioid receptor with increasing concentrations of the kappa selective antagonist nor-BNI.

Relationship of CD3, CD4 and CD8 expression on thymocytes. To determine which subpopulation of thymocytes expressed the kappa opioid receptor, it was first necessary to characterize the phenotypic subpopulations of thymocytes from 6- to 8-week-old C57BL/6ByJ mice. Flow cytometry was further used to correlate several antigenic properties of cells within the phenotypically heterogenous thymocyte population. Thymocytes were stained with fluorescently conjugated mAbs directed against mouse cell surface markers, such as FITC-CD4 (used to identify T-helper cells) and/or QR-CD3 (a pan-T-cell marker) or QR-CD8 (used to identify T-cytotoxic cells) to determine the phenotypic maturity of the cells, as well as the percentage of cells within each of the various phenotypic subpopulations. As shown in table 2, 91% of thymocytes analyzed were CD4⁺. Likewise, 86% of the cells were determined to be CD8⁺. Double labeling thymocytes for both CD4 and CD8 demonstrated that 82% of the cells expressed both antigenic markers, constituting a majority of CD4⁺/CD8⁺ (double positive) cells (table 2 and fig. 4). However, only approximately 60% of the total thymocytes analyzed expressed the CD3 antigen. Of the CD3⁺ thymocytes, three subpopulations consisting of CD3⁺ low/dim or immature (45%), CD3⁺ high/bright or mature (11%), as well as CD3 or very immature (~40%) thymocytes were evident, as shown in table 2, which agrees with that reported for CD3 expression on cells from human thymi (Lanier et al., 1986). Further analysis of these cells demonstrated that of the 45% CD3⁺ low/dim cells, 94% are also CD4⁺ (and probably CD8⁺, because 82% of all thymocytes analyzed are double positive for CD4/CD8), whereas of the 11% CD3⁺ high/bright cells, 81% are double positive for CD4 (and probably CD8, because increased expression of CD3 is indicative of mature T cells), which indicates that a major portion of the CD3⁺ cells are also CD4⁺.

View larger version (43K): [in this window] [in a new window]   Fig. 4.   Double-labeling of thymocytes with anti-CD4-FITC and anti-CD8-QR mAbs. Boundary markers were set by isotype-matched (IgG2a-QR and -FITC) mAbs for nonspecific staining. The numbers in each quadrant are the percentages of cells located within the boundaries based on the total number of cells that satisfy the gating criteria (Gate 1 = R1), which indicate the various phenotypic percentages (also shown in table 3). Data represent three separate experiments.

Phenotypic determination of kappa opioid receptor-labeled thymocytes. To determine the extent to which the kappa opioid receptor was expressed on these phenotypic subpopulations of thymocytes, QR-conjugated mAbs directed against CD3, CD4 and CD8 were used along with the amplification procedure for kappa opioid receptor labeling. Analysis of the double-labeled thymocytes for CD4 and the kappa opioid receptor is shown in figure 5. The dot plot (fig. 5A) and histogram (fig. 5B) for QR-CD4 fluorescently labeled thymocytes (gate R1) are displayed. Subsequently, gates R2/R3 were set to enclose CD4⁺ thymocytes, which were then selected and analyzed for the expression of the kappa opioid receptor in the PE fluorescence histogram. Examination of figure 5, C and D, and table 3 show that most thymocytes label for both CD4 and the kappa opioid receptor. Likewise, applying similar analysis to thymocytes double-labeled for both the kappa opioid receptor and CD8 (fig. 6A) or CD3 (fig. 6B) demonstrates that CD3⁺, CD4⁺ and CD8⁺ thymocytes, respectively, all possess specific labeling for the kappa opioid receptor (table 3). In each case, the fluorescence distribution of thymocytes labeled with 30 µM FITC-AA/PE is increased over background (fig. 5C). Subtraction of background PE median peak fluorescence intensity demonstrated a reduction in the median fluorescence value with FITC-AA/PE when nor-BNI was included as an inhibitor (table 3). Histogram (FL2 fluorescence channel for PE) subtraction of cells labeled with each mAb and the kappa opioid receptor in the presence of nor-BNI (nonspecific labeling) from cells labeled with each mAb and the kappa opioid receptor in the absence of nor-BNI (total labeling) provided the resultant relative percentage of cells specifically labeled for both the particular mAb and the receptor. These double-positive percentages ranged from 38% for CD3⁺ high/bright cells to 64% for CD4⁺ cells, as summarized in table 3.

View larger version (44K): [in this window] [in a new window]   Fig. 5.   Double-labeling of thymocytes for the kappa opioid receptor and the CD4 antigenic marker. Thymocytes were labeled via the amplification procedure for the kappa opioid receptor and/or anti-CD4-QR. Control samples were labeled with an isotype-matched (IgG2a) control conjugated to QR. (A) CD4+ thymocytes were gated (R2) in the forward scatter vs. QR fluorescence dot plot for subsequent kappa opioid receptor analysis. (B) The relationship of CD4+ thymocytes to control mAb staining (background) is shown in the FL3 (QR) fluorescence histogram. CD4+ cells were gated on (R3) for further analysis of the cells in the FL2 (PE) fluorescence channel. (C) Kappa opioid receptor labeling of CD4+ thymocytes is demonstrated by overlaid histograms. (D) Specific labeling of the kappa opioid receptor on CD4+ thymocytes as demonstrated after histogram subtraction of the background PE fluorescence. Note the downward and leftward shift of the PE histogram in the presence of nor-BNI. Data represent three separate experiments.

View larger version (19K): [in this window] [in a new window]   Fig. 6.   Anti-CD8-QR and anti-CD3-QR labeling of thymocytes. (A) The QR fluorescence histogram demonstrates the relative fluorescence intensity of CD8+ stained cells. Gate R3 was used to select the CD8+ cells for analysis of kappa opioid receptor expression. The dotted line represents staining of thymocytes with an isotype-matched-QR control (IgG2a) for background staining. (B) Thymocytes were stained with anti-CD3-QR or IgG2b-QR control mAb. Histograms from control samples (dotted, left peak) were overlaid with histograms from anti-CD3 stained samples. The percentage of positive cells expressing CD3 was determined by histogram subtraction, resulting in the quantitation of cells within regions R3 and R4 (see also table 3). Note the two peaks of QR fluorescence for CD3+ cells which are characteristic of thymocytes. Data represent three separate experiments, each performed in triplicate.

	Discussion

Top Abstract Introduction Materials & Methods Results Discussion References

Opioid receptors of the central nervous system have been extensively characterized by use of a variety of opioid compounds in conjunction with both radioligand binding and competition binding assays. In the immune system, however, implementation of the same methods for the detection of opioid receptors has been met with many inconsistencies when compared characteristically to brain opioid receptors (Sibinga and Goldstein, 1988). Although several studies provide functional evidence for opioid receptors on cells of the immune system (Linner et al., 1995, 1996; Shahabi and Sharp, 1995; Taub et al., 1991), identification of these classical, neural receptors has been difficult. The present reportdocuments the expression of kappa opioid receptors on murine thymocytes through the use of indirect immunofluorescence and flow cytometry. By use of this approach, it was possible to assess the co-expression of various cell surface antigens along with receptor expression, in a phenotypically heterogenous cell population, such as the thymus (Mathieson and Fowlkes, 1984).

Our investigations demonstrated that the high-affinity kappa opioid ligand, FITC-AA, which is structurally similar to the kappa selective agonist U50,488, specifically labeled thymocyte kappa opioid receptors (table 1 and fig. 1). These results closely agree with previous findings from our laboratory and others, which have shown FITC-AA-labeled kappa opioid receptors on murine thymocytes (Lawrence et al., 1995a) as well as on human microglia (Chao et al., 1996). Whereas inhibition of kappa opioid receptor labeling was ineffective by the addition of the delta selective peptides ICI 174,864 or DPDPE, inclusion of the mu opioids, DAMGO or morphine, resulted in a slight inhibition of the labeling, probably because of the mu opioids binding to the kappa site at the concentrations used (Bidlack et al., 1992). Sixty percent of all gated thymocytes demonstrated specific labeling for the kappa opioid receptor. The 40% unlabeled cells within gate R1 may constitute a subpopulation of immature T cells that does not express the kappa opioid receptor or does not express the kappa opioid receptor at a concentration high enough to be detected by this method. The kappa opioid receptor was then determined to be present on thymocytes expressing CD3, CD4 and CD8 antigenic markers. Although simultaneous, three-color labeling for both the kappa opioid receptor and two mAbs was not possible because of the use of fluorescein and PE in the amplification labeling procedure, these findings suggest that most thymocytes are in one of two phenotypic categories, CD3⁺/CD4⁺/CD8⁺ or CD3/CD4⁺/CD8⁺, which agrees with other reports (Lanier et al., 1986; Scollay et al., 1984). Kappa opioid receptor expression on thymocytes of relatively immature phenotype is further substantiated by reports of mRNA for this receptor being expressed in the more immature, CD3/CD4/CD8, R1.1 thymoma cell line and in double-negative, immature thymocytes (Belkowski et al., 1995), as well as by radioligand binding for this receptor on the R1.1 cell line (Lawrence et al., 1995b). As thymocytes differentiate into mature T cells, expression of the CD3 antigenic marker increases (Lanier et al., 1986; Mathieson and Fowlkes, 1984). The present studies further demonstrate a concurrent decrease in kappa opioid receptor expression on the mature subpopulation of thymocytes (table 3), which suggests that the kappa opioid receptor may be a characteristic marker of the developing, immature T cell.

A very small proportion of the thymocytes analyzed were of the mature T-cell CD4⁺/CD8 (9%) or CD4/CD8⁺ (3%) phenotype (table 2 and fig. 4). Because 6- to 8-week-old mice were used for these experiments, it followed that older mice might be a viable source for the acquisition of greater numbers of mature thymocytes. However, several studies have determined a lack of age-dependent alterations in the proportion of CD4⁺, CD8⁺ thymus cell subsets in a related mouse strain (C57BL/6) (Dubiski et al., 1989; Oughton et al., 1995). Therefore, although thymocytes represent a more homogenous cell population than splenocytes, splenocytes will be used in future studies for the detection of opioid receptors on mature T-lymphocytes (CD3⁺/CD4⁺/CD8 and CD3⁺/CD4/CD8⁺). Preliminary data from our laboratory demonstrate that nylon-wool purified splenocytes (to enrich for T cells) express the kappa opioid receptor (Bidlack et al., 1996). On-going investigations will further elucidate the immune cell types which possess this receptor.

The presence of the kappa opioid receptor on thymocytes raises a fundamental question regarding its purpose. Because the thymus is innervated by opioid peptides (Dardenne and Savino, 1994), the expression of kappa opioid receptors suggests either a temporal or a tonic regulatory role in the development or differentiation of thymocytes. However, temporal regulation by opioids would preclude the presence of the receptor on all thymocytes. As suggested by the present results, whereby this receptor appears to be present on most thymocytes regardless of CD antigen marker expression, a tonic regulation of these immune cells would appear to be more appropriate. Furthermore, because the cellular mechanisms involved in the selection of self-tolerant thymocytes for completion of maturation and subsequent release into the periphery is not yet fully understood, the presence of this opioid receptor may play an integral role in this selection process, because preliminary studies demonstrate labeling for this receptor on splenocytes (Bidlack et al., 1996). Demonstration of immune cell kappa opioid receptor expression also supports an endogenous role, as well as paracrine and possibly autocrine roles, for opioid peptides that are produced by these same cells (Belkowski et al., 1995; Linner et al., 1995, 1996; Martin et al., 1987).

Opioid receptor expression on immune cells has many other wide-reaching implications. Opiate drug abuse has been linked to a decrease in immunocompetence (reviewed in Peterson et al., 1993), potentially exposing drug abusers to other infectious diseases, such as HIV infection, ultimately leading to the development of AIDS. Studies by Chao and colleagues have investigated the enhancement of HIV-1 expression in brain/promonocyte cocultures by kappa (Chao et al., 1995) and mu (Peterson et al., 1994) opioids, demonstrating opioid-induced immunomodulatory effects. More recently, however, these investigators have demonstrated kappa opioid receptor labeling on microglia cells by use of FITC-AA and flow cytometry (Chao et al., 1996). They also showed kappa opioid receptor mRNA in fetal human microglia. In addition, kappa opioids actually decrease HIV replication in these particular cells, providing provocative evidence as to the immunoinhibitory effects of kappa opioids. Research progressing in this area may ultimately lead to novel findings regarding opioid involvement in AIDS and other immune-related diseases, as well as inflammation- and stress-mediated immune responses, where opioid involvement has been implicated (Lewis et al., 1985).

Studies in progress with various fluorescein-conjugated opioids will further characterize the types of immune cells possessing mu, delta and kappa opioid receptors. Findings from these investigations may provide a much needed "missing link," joining results of functional studies, which suggest the presence of opioid receptors to findings from molecular studies, which provide evidence that mRNA exists within various immune cells for opioid peptides and their receptors. Therefore, the "mapping" of opioid receptors in the immune system will provide strong and convincing evidence for direct opioid involvement in immunocompetence, and thereby add a significant contribution to the ever-expanding neuroimmune network.