AJP - Regu Fuel your research with LabChart
HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
QUICK SEARCH:   [advanced]


     

Am J Physiol Regul Integr Comp Physiol 275: R1725-R1734, 1998;
0363-6119/98 $5.00
This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Right arrow Citation Map
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Gannon, G. A.
Right arrow Articles by Shephard, R. J.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Gannon, G. A.
Right arrow Articles by Shephard, R. J.
Vol. 275, Issue 6, R1725-R1734, December 1998

beta -Endorphin and natural killer cell cytolytic activity during prolonged exercise. Is there a connection?

G. A. Gannon1,2, S. G. Rhind1,2, M. Suzui3, J. Zamecnik2, B. H. Sabiston1,2, P. N. Shek1,2,4, and R. J. Shephard1,2,5,6

1 Graduate Programme in Exercise Sciences, 5 Faculty of Physical Education and Health, and 4 Department of Laboratory Medicine and Pathobiology, Faculty of Medicine, University of Toronto, Toronto M5S 1A1; 2 Defence and Civil Institute of Environmental Medicine, Toronto M3M 3B9; 6 Health Studies Programme, Brock University, St. Catharines, Ontario, Canada L2S 3A1; and 3 Meiji University, Tokyo, Japan

    ABSTRACT
Top
Abstract
Introduction
Methods
Results
Discussion
References

This study was designed to test whether a single 50-mg dose of the opioid antagonist naltrexone hydrochloride, ingested 60 min before 2 h of moderate-intensity exercise (i.e., 65% peak O2 consumption), influenced the exercise-induced augmentation of peripheral blood natural killer cell cytolytic activity (NKCA). Ten healthy male subjects were tested on four occasions separated by intervals of at least 14 days. A rested-state control trial was followed by three double-blind exercise trials [placebo (P), naltrexone (N), and indomethacin] arranged according to a random block design. The indomethacin exercise trial is discussed elsewhere (S. G. Rhind, G. A. Gannon, P. N. Shek, and R. J. Shepherd. Med. Sci. Sports Exerc. 30: S20, 1998). For both the P and N trials, plasma levels of beta -endorphin were increased (P < 0.05) at 90 and 120 min of exercise but returned to resting (preexercise) levels 2 h postexercise. CD3-CD16+CD56+ NK cell counts and NKCA were significantly (P < 0.05) elevated at each 30-min interval of exercise compared with correspondingly timed resting control values. However, there were no differences in NK cell counts or NKCA between P and N trials at any time point during the two trials. Changes in NKCA reflected mainly changes in NK cell count (r = 0.72; P < 0.001). The results do not support the hypothesis that the enhancement of NKCA during prolonged submaximal aerobic exercise is mediated by beta -endorphin.

naltrexone; natural immunity; cell adhesion; growth hormone; cortisol

    INTRODUCTION
Top
Abstract
Introduction
Methods
Results
Discussion
References

PHYSICAL ACTIVITY of sufficient intensity and duration stimulates the release of beta -endorphin from the anterior pituitary gland, increasing blood levels of this hormone (18). Although beta -endorphin is known to have immunomodulatory properties (38), the biological significance of the increased plasma concentrations that are seen during exercise remains unclear (19, 25). In vitro, beta -endorphin acutely enhances the cytolytic activity of peripheral blood natural killer (NK) cells through an opioid receptor-mediated, naloxone-reversible pathway (32, 34), suggesting that NK cells carry specific receptors for beta -endorphin (5, 41). The dose-response curve apparently has an inverted-U shape, although there is little agreement on the minimum concentration required to induce a positive effect; enhancement of NK cell-mediated cytolytic activity (NKCA) has been demonstrated at concentrations ranging widely from 10-14 to 10-6 mol/l (6, 12, 28).

Because physical activity exerts a well-demonstrated stimulatory influence on the NKCA of peripheral blood (14), the physiological plasma concentrations of beta -endorphin (i.e., 10-12 mol/l) induced by physical exercise could conceivably explain the acute effects of physical exercise on NKCA. In support of this, Fiatarone et al. (11) administered in a blind protocol either normal saline or the opioid antagonist naloxone (100 µg/kg) to eight healthy young women who then performed a maximal incremental cycle ergometer test. After naloxone administration, the rise in NKCA was no longer statistically significant, although the increase in peripheral blood NK cell count (identified with CD16 or CD56 surface markers) was similar to that seen in the placebo trial. In contrast, Kappel et al. (27) argued that the exercise-induced increase in NKCA could be explained entirely by a concomitant increase in peripheral blood NK cell number, secondary to sympathetic activation and the peripheral release of epinephrine, although others have argued that not all of the effect of exercise on NKCA can be attributed to changes in NK cell number (31, 52). In support of the hypothesis of Kappel et al. (27), a similar cytolytic activity per NK cell (per NKCA) was seen before, during, and subsequent to exercise (4). Further evidence against a role for beta -endorphin was provided by a study of seven healthy young men in which lumbar epidural anesthesia was used to block afferent nerve impulses from exercised skeletal muscles (30), thus preventing any increase in beta -endorphin during 20 min of recumbent cycle ergometry [60% peak O2 consumption (VO2 peak)] under hypoxic conditions. The anticipated exercise-induced increase in NKCA and NK cell concentration was unaffected by the sensory nerve blockade.

Because cell adhesion represents the first step of effector-target interaction, it is logical to reason that any stimulatory effect induced by beta -endorphin may be mediated by an increased expression of specific cell surface adhesion molecules. Lymphocyte function-associated antigen 1 (LFA-1; CD11a/CD18) and LFA-2 (CD2) are two important cell adhesion molecules; they are expressed by peripheral blood NK cells and function as "accessory" molecules during cell-cell communication and activation (44, 46). In particular, these molecules and their respective target cell ligands (intercellular adhesion molecule 1 and LFA-3) play a major role during the recognition, conjugation, and cytolysis of K562 target cells by peripheral blood NK cells (47, 53).

The present study used the nonselective opioid receptor antagonist naltrexone hydrochloride to investigate further the possible influence of beta -endorphin on NKCA during and after an acute bout of prolonged physical activity. Therapeutic doses were administered according to a randomized, double-blind, placebo-controlled protocol. We chose a 2-h bout of moderate aerobic activity (i.e., 60-70% VO2 peak), envisioning that the first hour of such exercise would increase the NK cell count but not plasma beta -endorphin and that the second hour would increase plasma beta -endorphin (48) without further increments in NK cell count. The effects of naltrexone hydrochloride on the mean surface density of CD11a (LFA-1alpha ) and CD2 cell surface adhesion molecule expression were examined, and we controlled for secondary effects of naltrexone hydrochloride on growth hormone (GH) and cortisol (10, 36), which have a demonstrated influence on lymphocyte trafficking and NKCA (39, 40).

    METHODS
Top
Abstract
Introduction
Methods
Results
Discussion
References

Subjects. Ten recreationally active (VO2 peak 44.0 ± 3.5 ml · kg-1 · min-1) male nonsmokers [age 26.3 ± 5.4 (SD) yr, mass 79.3 ± 10.3 kg, height 1.78 ± 0.07 m] volunteered to participate in this study under conditions approved by the University of Toronto and the Defence and Civil Institute of Environmental Medicine Human Experimentation Committees. Participation was contingent on a detailed medical examination and approval by a physician. A history of allergies or evidence of acute or chronic infection were criteria for exclusion from the study.

Experimental design. This study was composed of five laboratory visits: 1) clinical, physical, and anthropometric assessment, 2) a nonexercise, resting control trial, and 3) three double-blind exercise tests ordered according to a random block design (placebo, naltrexone, and indomethacin). For the present purpose, only the resting control, placebo, and naltrexone trials will be considered. The trial involving administration of indomethacin has been described by Rhind et al. (42a).

Physical assessment. After clinical examination and medical clearance, but at least 1 wk before the control trial, VO2 peak and peak heart rate (HRpeak) were determined on a mechanically braked cycle ergometer (Monark Ergomedic 818E, Stockholm, Sweden). Subjects performed an incremental exercise test at a pedal cadence of 70 rpm (initial loading of 60 W, with 25-W/min increments). Volitional exhaustion was reached in 8-12 min. Expired gases, collected breath by breath, were analyzed for respiratory minute volume and oxygen consumption using a metabolic measurement cart (SensorMedics 2900C, Yorba Linda, CA). Heart rates were recorded at 5-s intervals using a Polar Vantage XL heart rate monitor (Polar USA). The work rate estimated to elicit 65% VO2 peak was determined for each subject from a plot of work rate versus oxygen consumption.

Control and experimental trials. Within 2 wk of the physical assessment, subjects performed a control trial followed by three randomized, counterbalanced exercise trials at intervals of at least 2 wk. On each test day, subjects reported to the laboratory at 0700 to 0730, having fasted overnight and abstained from strenuous physical activity for 36 h. Subjects were immediately instrumented with a heart rate monitor and a 21-gauge intravenous catheter (Insyte, BD Vascular Access, Sandy, UT). To standardize metabolic conditions, each subject consumed 1.1 MJ (250 cal) of a clinical dietary product (Ensure Plus; 9.4 g protein, 38 g carbohydrate, and 6.7 g fat) immediately after collection of the initial (T0) blood sample.

The nonexercise control trial was conducted in the same laboratory environment as the exercise trials, serving to familiarize subjects with personnel, protocol, and equipment. Beginning between 0730 and 0800, after the subjects had ~30 min of seated rest, we collected serial blood samples according to the following schedule: (in hours) time 0 (T0), T1, T2, T4, and T24.

The exercise trials involved administration of either placebo or naltrexone hydrochloride before a 2-h bout of cycle ergometry at a pedal cadence of 60 rpm and an intensity adjusted to demand 65% of the individual's VO2 peak. Oxygen consumption (SensorMedics 2900C, Yorba Linda, CA) and heart rate (Polar Vantage XL) were monitored at 15-min intervals during exercise, and the work load was adjusted as necessary to maintain the required intensity of effort. Participants were encouraged to consume 1.0-1.5 liters of water during each trial to minimize hemoconcentration. Venous blood samples of 45 ml were collected in sterile glass Vacutainers (Becton-Dickinson, Franklin Lakes, NJ) containing the necessary preservatives and anticoagulants at T0, T0.5, T1, T1.5, T2, T4, and T24 relative to the initiation of exercise.

Drug administration. Identical gelatin capsules containing lactose placebo (180 mg; Novopharm, Scarborough, Ontario, Canada) or naltrexone hydrochloride (50 mg Trexan, Du Pont Merck Pharmaceuticals, Wilmington, DE) were presented according to a double-blind, counterbalanced protocol immediately after collection of the initial blood sample and 60 min before exercise. Compliance was controlled by observation of drug and/or placebo intake on scheduled test days.

Hematological analyses. Determinations of total leukocyte counts, three-cell differential counts (granulocytes, monocytes, and lymphocytes), Hb, and hematocrit were performed on tripotassium ethylenediamine tetra-acetate-treated blood using an automated Coulter JT hematology analyzer (Coulter Electronics, Hialeah, FL). All blood cell counts were corrected to resting (T0) blood volumes using the observed Hb and hematocrit values and applying the formulas of Dill and Costill (9).

MAbs. The following mouse, anti-human monoclonal antibodies (MAbs) were used in this study: FITC anti-CD11a MAb B-B15 (IgG1) and FITC anti-CD2 MAb LT2 (IgG2b), purchased from Serotec Canada (Mississauga, Ontario, Canada); and FITC anti-CD3 MAb SK7 (IgG1), phycoerythrin (PE) anti-CD16 MAb B73.1 (IgG1), and PE anti-CD56 MAb MY31 (IgG1), purchased from Becton-Dickinson (Mississauga, Ontario, Canada).

NK cell immunophenotyping. NK cells (CD3-CD16+CD56+) were enumerated by dual-parameter immunophenotyping, using combinations of FITC- and PE-conjugated MAbs. All samples were analyzed on the day when they were collected. Briefly, 100 µl EDTA-whole blood was incubated with saturating amounts of FITC anti-CD3, PE anti-CD16, and PE anti-CD56 MAbs as previously described (13). Whole blood samples with leukocyte counts >9.8 × 109 cells/l were diluted with 1× isotonic PBS containing 0.1% sodium azide. Stained-cell suspensions were enumerated on a FACScan flow cytometer equipped with an air-cooled argon ion laser emitting 15 mW at 488 nm using standard operating methods [Becton-Dickinson Immunocytometry Systems (BDIS), San Jose, CA]. Daily calibration was performed using a mixture of monosized FITC- and PE-conjugated and unconjugated latex particles (CaliBRITE beads, BDIS) in conjunction with AutoCOMP software (BDIS). Forward-scatter and side-scatter gains, forward-scatter threshold, and fluorescence compensation levels were optimized using isotype-negative controls (anti-IgG1-FITC/anti-IgG1-PE) and anti-CD4-FITC/anti-CD8-PE double-stained whole blood samples. Electronic compensation was adjusted to eliminate spectral overlap between fluorescence channels. Digitized data were acquired and analyzed using CellQUEST software (BDIS).

Cell adhesion molecule surface density. Peripheral blood mononuclear cell (PBMC) suspensions were stained with saturating concentrations of PE anti-CD56 MAb and a second FITC-labeled MAb specific for either CD11a or CD2 (13). Stained-cell suspensions were analyzed for fluorescence on a multiparameter FACScan flow cytometer (BDIS). Acquired list mode data were analyzed with CELLQuest software (BDIS) for mean fluorescence intensity (MFI) of CD11a and CD2 on CD56+ lymphocytes (i.e., total cellular fluorescence of CD56-positive lymphocytes averaged over the number of positive events; MFI). For each analysis, 5,000 events positive for CD56 were acquired. The MFI served as an indicator of mean surface density of a given adhesion molecule.

Isolation of PBMCs. Fresh PBMCs were isolated from heparinized blood samples (143 USP units/10 ml of blood) by 30 min of density-gradient centrifugation (400 g, 20°C) over Ficoll-Hypaque (Pharmacia, Uppsala, Sweden). The mononuclear cell band was carefully harvested, washed, and reconstituted to a concentration of 2.0 × 107 cells/ml in 10% FCS-RPMI 1640.

K562 tumor cell culture. The NK-sensitive K562 tumor line (American Type Culture Collection, Rockville, MA) was used to provide target cells in cytolytic assays. The cell line was maintained as a continuous suspension in RPMI 1640 culture medium containing 10% FCS, 1% (wt/vol) penicillin-streptomycin, 20 mM HEPES (pH 7.3), and 2 mM L-glutamine (Gibco Life Technologies, Burlington, Ontario, Canada). A humidified, water-jacketed incubator (Revco Ultima; VWR Scientific, Toronto, Ontario, Canada) provided a 37°C atmosphere of 5% CO2 to maintain a pH of 7.2-7.3. To ensure that cells were in the logarithmic phase of growth, cultures were split into 3-5 × 109 cells/ml on a 24-h schedule 3 days before the experiment. Cell viability was assessed by trypan blue exclusion and was typically >90%.

Tumor cell labeling. K562 tumor cells were first washed two times in 10 ml RPMI 1640 medium without FCS and centrifuged for 5 min (400 g, 20°C); they were then resuspended at a concentration of 1 × 107 cells/ml. K562 tumor cells were labeled with PKH26 (Sigma, St. Louis, MO), a stable lipophilic membrane dye with an emission peak at 567 nm. To label cells, we rapidly added 0.5 ml of the target cell suspension to 0.5 ml of PKH26 (4 µM) in a 12 × 75 mm polystyrene culture tube. After incubation at 25°C for 2-5 min, the reaction was stopped by the addition of 1 ml of 100% FCS for 1 min. After centrifugation (400 g, 25°C) for 5 min, cells were washed (3 times) in 10 ml of supplemented RPMI 1640 medium and resuspended to a final concentration of 2 × 105 cells/ml for the cytolytic assay.

NKCA. The total cytolytic activity of peripheral blood NK cells was assessed by an in vitro flow cytometric assay (22). To expedite data acquisition, we modified this method by using a higher target cell concentration (43). With this method, the plasma membrane integrity of PKH26-labeled K562 tumor cells, after 4 h of incubation with PBMC (i.e., NK cells), is determined with flow cytometry using the DNA-intercalating dye propidium iodide (PI).

Briefly, 100 µl of freshly isolated PBMC (effectors, at 1 × 107 cells/ml) were gently mixed with 100 µl of PKH26-labeled K562 tumor cells (targets, at 2 × 105 cells/ml) and 25 µl (1 µg/ml) of PI solution (Sigma) at an effector:target ratio of 50:1. Cell mixtures were centrifuged for 5 min (20°C, 50 g) to promote optimal effector-to-target cell conjugation and incubated for 4 h (37°C, 5% CO2). The assay was stopped by addition of cold cell wash to the cultures. Samples were placed on ice until same-day analysis.

Samples were analyzed in triplicate, using a FACScan flow cytometer and CellQUEST software (BDIS). PKH26+ target cells were defined and live gated via a histogram of FL2 fluorescence. A minimum of 5,000 PKH26+ target cell events (corresponding to ~200,000 list mode events) were acquired per sample. Dead K562 cells were differentiated from live K562 cells based on the FL3 fluorescence of PI. Spontaneous target cell death was determined by incubating 100 µl of PKH26-labeled K562 tumor cells with 25 µl of PI in the absence of effector cells. Percent specific lysis was calculated by subtracting the mean percentage of spontaneously dead target cells from the percentage of target cells killed in the test sample. The corresponding absolute number of dead target cells was calculated by multiplying the percent lysis by the total number of target cells used in a given assay. The intra-assay coefficient of variation was consistently 4% among triplicate samples, and the between-trial coefficients of variation for the same subject were typically 5%.

Neuroendocrine analyses. Total plasma beta -endorphin concentrations were determined in duplicate, using an affinity gel extraction and 125I RIA (Incstar, Stillwater, MN). Total plasma concentrations of cortisol (ICN Biomedicals, Costa Mesa, CA) and GH (Allegro Nichols Institute, San Juan Capistrano, CA) were determined in duplicate using a competitive solid-phase 125I RIA technique. Plasma hormone concentrations were adjusted for estimated changes in plasma volume using the formulas of Dill and Costill (9).

Plasma naltrexone and 6-beta -naltrexol analyses. Plasma samples (0.5 ml) were analyzed by gas chromatography (GC) and mass spectrometry (MS), using negative ion chemical ionization and a selective ion-monitoring mode. The method was adapted from Monti et al. (35), with methane-5% ammonia as the reagent gas and deuterium-labeled naltrexone and naltrexol as internal standards. Unknown plasma samples and plasma spiked with standards were extracted with toluene, dried at 40°C under a stream of nitrogen, and treated with pentafluoropropionic anhydride for 45 min at 80°C. Excess pentafluoropropionic anhydride was removed by drying with N2, and 1-ml aliquots were injected into a Finnigan TSQ-700 GC-MS/MS capillary column equipped with 15m DB-5 (JandW Scientific, low load, 0.25 mm ID). The GC injector was held at 250°C; the column was ramped from 140 to 250°C at 25°C/min and then held at this temperature for 10 min. The mass spectrometer conditions were an ion source at 110°C, and reagent gas was optimized using negative chemical ionization fragment m/z 633 of pentafluoro tributal amine, 70 eV, ion current 200 mA, single ion monitoring m/z 779 and m/z 782 for beta -naltrexol pentafluoro propionic anhydride (PFP) and D3-naltrexol PFP, and m/z 761 and m/z 764 for naltrexone and D3-naltrexone, with dwell times of 30 ms. The calibration curves were constructed by spiking blank plasma with 0.5, 2.5, 10, and 20 ng/ml of naltrexone and 5 ng/ml of D3 naltrexone internal standard (I.S.) and 5, 50, 100, 200 ng/ml of beta -naltrexol and 50 ng/ml of D3 beta -naltrexol I.S. Linear regressions and sample concentration calculations were done using the spreadsheet Microsoft Excel.

Statistical analyses. Analyses were performed using the Statistical Package for the Social Sciences (SPSS/PC+, Windows version 7.0). Data are presented as means ± SE unless otherwise noted. To determine circadian effects, we analyzed resting control data using a one-way (time) repeated-measures (RM) ANOVA across five time points (T0, T1, T2, T4, and T24). A two-way (condition × time) RM ANOVA was used to determine the main effects of condition (control vs. placebo vs. naltrexone), time (T0, T0.5, T1, T1.5, T2, T4, and T24), and condition × time interactions. Because values for a given measure are highly correlated across time, the Greenhouse-Geisser correction was applied to reduce the risk of a type I error. When a significant F ratio was demonstrated, differences among treatment means were determined by a Newman-Keuls post hoc analysis.

    RESULTS
Top
Abstract
Introduction
Methods
Results
Discussion
References

Physiological response to acute exercise. Mean oxygen costs of the exercise conditions (i.e., placebo and naltrexone) were maintained at ~65% VO2 peak as demonstrated by condition (F1,9 = 0.10, P = 0.76), time (F3,27 = 0.17, P = 0.91), and condition × time interaction (F3,27 = 0.67, P = 0.58) statistics (2 × 4 RM ANOVA). Absolute power outputs were 127 ± 14 and 125 ± 12 W for placebo and naltrexone, respectively, eliciting 82.0 ± 4.5 and 78.6 ± 4.2% of HRpeak.

Plasma naltrexone and 6-beta -naltrexol. Plasma naltrexone concentrations reached 9.7 ± 1.7 ng/ml within the first 30 min of exercise (i.e., 90 min postingestion), and values were still within 2.0 ± 0.2 ng/ml 2 h postexercise (Fig. 1A). The plasma concentration of the metabolite 6-beta -naltrexol peaked at 60 min of exercise (81.5 ± 7.6 ng/ml), falling to 11.1 ± 1.3 ng/ml by T24 (Fig. 1B). A plasma naltrexone concentration of 2 ng/ml provides effective opioid blockade (54).


View larger version (17K):
[in this window]
[in a new window]
  		Fig. 1.   Plasma concentrations of naltrexone (A) and 6-beta -naltrexol (B) during naltrexone exercise condition. ND, not determined. Data are means ± SE.

Monocyte and NK cell counts. Initial (T0) circulating monocyte (data not shown) and CD3-CD16+CD56+ NK cell (Fig. 2A) counts fell within normal resting ranges and did not differ significantly between conditions. In the control condition, no significant differences were found between T0 and T24 for monocyte counts. However, NK cell counts demonstrated a significant main effect of time (F2.3,21.1 = 3.29, P < 0.05); counts at T24 (0.23 ± 0.03 × 109 cells/l) were significantly greater (P < 0.05) than counts at T0 (0.17 ± 0.02 × 109 cells/l).


View larger version (50K):
[in this window]
[in a new window]
  		Fig. 2.   Absolute counts of CD3-CD16+CD56+ natural killer (NK) cells (A), total NK cell-mediated cytolytic activity (NKCA), measured as %lysis of K562 target cells per fixed number of peripheral blood mononuclear cells at a 50:1 effector:target cell ratio (B), change (Delta ) in %lysis relative to baseline values within each condition (cond) (C), and change in number of K562 target cells killed by each CD3-CD16+CD56+ NK cell (per NKCA) relative to baseline values within each condition, as shown in rectangular enclosure (D). P, placebo; N, naltrexone; C, control. In A, numbers within bars represent relative percentage of lymphocytes. Two-way repeated-measures ANOVA statistics are shown, as well as significant pairwise differences. c × t, Condition × time. * P < 0.05 vs. control; ** P < 0.01 vs. control; a P < 0.05 vs. baseline [time 0 (T0)]; and b P < 0.05 vs. previous time point.

Over the three conditions, monocyte counts (data not shown) showed significant main effects of condition (F1.8,16.4 = 4.32, P < 0.05) and time (F2.4,21.5 = 21.84, P < 0.001), with a condition × time interaction (F4.0,36.2 = 5.73, P < 0.001). Relative to the corresponding control data, the moderately intensive placebo cycling protocol elevated the total circulating monocyte count by 104% (P < 0.05) at 30 min of exercise (T0.5), with further significant increments at 2 h of exercise (T2). By 2 h postexercise, the number of circulating monocytes had returned to control levels. The exercise-induced mobilization of monocytes was similar under both placebo and naltrexone conditions.

Significant main effects of condition (F1.5,13.6 = 42.92, P < 0.001) and time (F1.8,15.8 = 40.44, P < 0.001) and a condition × time interaction (F3.2,28.9 = 28.85, P < 0.001) were found for changes in circulating CD3-CD16+CD56+ NK cell counts over the three conditions (Fig. 2A). During the placebo exercise condition, CD3-CD16+CD56+ NK cell counts had increased by 270% by 30 min of exercise (T0.5), accounting for ~20% of the total circulating lymphocyte pool. Despite the steady-state nature of the exercise protocol, counts continued to increase to 330% of resting control values at 2 h of exercise (T2). The NK lymphocytosis had abated by 2 h postexercise (T4), with counts dropping to 75% of the T4 control value. Normal resting values were achieved by 24 h postbaseline (T24). There were no statistically significant differences between the placebo and naltrexone conditions.

In vitro NKCA. The total cytolytic activity of NK cells against K562 target cells, measured as percent lysis (Fig. 2B), showed a significant main effect of time (F2.6,23.2 = 81.86, P < 0.001) and a significant interaction effect (F3.0,27.2 = 16.33, P < 0.001). Under placebo conditions, NKCA increased from 26.3 ± 3.4% lysis (T0) to 49.4 ± 6.0% lysis (T2) during exercise (85% increase, P < 0.01) but fell 28% below resting control values at T4 2 h postexercise (P < 0.01). A very similar trend was found in the naltrexone condition; post hoc analysis suggested that percent lysis values were not significantly different between conditions at any time point. However, 24-h postbaseline placebo data showed a significant reduction (P < 0.05) relative to corresponding control values, whereas the naltrexone data did not.

Measures of NKCA at baseline (T0) showed a nonsignificant trend to a difference (P > 0.05) between placebo and naltrexone conditions (Fig. 2B). We thus decided to express the percent lysis data for each condition as a change relative to their corresponding baseline (T0) values (Fig. 2C). This made the trend to differences between placebo and naltrexone conditions more apparent. Relative to corresponding baseline (T0) values, there was a significant (P < 0.01) 33% drop in percent lysis at 2 h postexercise (T4) during the naltrexone trial, whereas the 20% drop at 2 h postexercise during the placebo condition was not significant. Taken together, these results suggest that naltrexone administration may have reduced the natural cytotoxic capacity of the whole blood compartment early in recovery from prolonged aerobic exercise.

Per NKCA. The CD3-CD16+CD56+ NK cells accounted for an increasing proportion of the total circulating lymphocyte pool during exercise but returned approximately to baseline levels at 2 and 24 h postexercise (Fig. 2A). Exercise also changed the relative concentrations of circulating monocytes (data not shown), leading to significant changes in the composition of the 1.0 × 106 mononuclear cells (i.e., PBMC) incubated at a 50:1 ratio with K562 cells. Thus, to determine the exact effector-to-target ratio (i.e., CD3-CD16+CD56+ NK cell:K562 cell) at each time point, adjustments were made to account for the exercise-induced changes in the proportion of NK cells and monocytes. The percentage of monocytes in 1.0 × 106 PBMCs showed only minor shifts (interaction effect; F3.7,33.7 = 1.54, P = 0.21); however, at 2 h postexercise, both placebo and naltrexone values exceeded the corresponding control values (16.2 and 16.8%, respectively, vs. 12.5%). The calculated NK:K562 ratio changed from ~4:1 at baseline (T0), over 9:1 during exercise (P < 0.01), and between 4 and 5:1 postexercise (T4 and T24) for both placebo and naltrexone conditions.

When NKCA was expressed on a per-CD3-CD16+ CD56+ NK cell basis (i.e., per NKCA), there was no statistically significant interaction effect of time versus condition (Fig. 2D). For the placebo condition, values of per NKCA showed a decreasing trend from a baseline (T0) value of 7.3 × 10-2 dead K562 cells to below 6.0 × 10-2 dead K562 cells during exercise, while increasing to ~6.3 × 10-2 dead K562 cells at T4 and T24. No differences were found between the placebo and naltrexone conditions.

Because per NKCA did not change significantly during either placebo or naltrexone conditions, it appears that much of the change in the natural cytotoxic capacity of whole blood (i.e., difference between T0 and T2) can be accounted for by changes in the concentration of circulating NK cells (Pearson product moment correlation; r = 0.72, P < 0.001). To evaluate the relationship between NKCA and NK cell count, we fitted a linear regression relating the number of CD3-CD16+CD56+ NK cells to the number of dead K562 cells for each subject at each time point (Fig. 3); this demonstrated significant correlations for both placebo and naltrexone (P < 0.01), although the goodness of fit was slightly tighter for placebo (r = 0.818) than for naltrexone (r = 0.695).


View larger version (21K):
[in this window]
[in a new window]
  		Fig. 3.   Scatter plot of calculated NK cell count per assay tube vs. dead K562 cell per assay tube.

Neuroendocrine response. Resting (T0) plasma beta -endorphin levels (3.7 ± 0.7 pmol/l, averaged across conditions) were in the normal range and did not differ significantly between placebo and naltrexone conditions (Fig. 4A). In the placebo condition, beta -endorphin increased by 235% (P < 0.05) and 300% (P < 0.05) at 90 (T1.5) and 120 min (T2) of exercise, respectively, relative to the corresponding baseline (T0) value (1-way RM ANOVA; F1.6,14.8 = 7.51, P < 0.01). A significant main effect of condition (F1,9 = 9.18, P = 0.01) indicated that, compared with placebo, beta -endorphin concentrations were higher throughout the naltrexone condition. In the naltrexone condition, values were 250 and 400% of baseline (T0) at T1.5 and T2, respectively.


View larger version (30K):
[in this window]
[in a new window]
  		Fig. 4.   Plasma concentrations of beta -endorphin (A), growth hormone (B), and cortisol (C). * P < 0.05 vs. control; ** P < 0.01 vs. control; and a P < 0.05 vs. baseline (T0).

Resting values for total plasma GH (Fig. 4B) were within the expected range of 1-5 µg/dl. Exercise induced a significant increase (P < 0.01) in total GH concentrations, but values returned to the nonexercise control value 2 h postexercise (T4). There were no significant differences between placebo and naltrexone conditions.

Total plasma cortisol concentrations (Fig. 4C) showed significant main effects of condition (F1.8,12.8 = 4.34, P < 0.05) and time (F3.2,22.0 = 3.15, P < 0.05), with a significant interaction effect (F6.8,47.6 = 2.29, P < 0.05). The mean resting (T0) values fell within the expected range of 5-20 µg/dl, and subjects displayed the expected circadian changes (F2.2,15.1 = 4.69; P < 0.05); values decreased significantly (P < 0.05) from 11.5 ± 1.8 µg/dl at 0800 (T0) to 7.0 ± 1.3 µg/dl at 1200 (T4). Exercise values at T0 through T1.5 were not significantly different from the corresponding control values. However, cortisol concentrations at T2 were ~100% greater (P < 0.01) than the corresponding control values for both placebo and naltrexone conditions. There were no significant differences between placebo and naltrexone conditions.

Adhesion activation molecules. The mean surface density of CD11a (i.e., LFA-1alpha ) and CD2 (LFA-2) on CD56+ lymphocytes was analyzed at each time point for all three experimental conditions (Fig. 5). Significant interaction effects were found for relative changes in CD11a surface densities (F2.8,25.0 = 3.79, P < 0.05) and CD2 surface densities (F2.5,22.1 = 3.28, P < 0.05).


View larger version (28K):
[in this window]
[in a new window]
  		Fig. 5.   Changes in mean fluorescence intensity (MFI) of CD11a (A) and CD2 (B) surface adhesion molecules on CD56+ lymphocytes. Data are expressed as change relative to respective baseline values at T0, as shown in insets. * P < 0.05 vs. C; ** P < 0.01 vs. C; and dagger  P < 0.01 between P and N conditions.

In the placebo condition, the mean CD11a surface density of circulating CD56+ lymphocytes had increased (P < 0.05) by ~25 MFI units at 30 min of exercise (Fig. 5A). This level of expression was maintained until 2 h of exercise (T2). Similar changes were found during the naltrexone condition. Two hours postexercise (T4), the mean surface density had dropped to ~9 and 14 MFI units below the resting control for placebo and naltrexone conditions, respectively (P > 0.05). The only significant difference between the two conditions was found at 24 h postbaseline (P < 0.05), when the mean surface density was 26 MFI units above baseline (T0) for the placebo condition, whereas it was only 5 MFI units above baseline for naltrexone; neither placebo nor naltrexone values differed significantly from the corresponding resting control value, which was 9 MFI units above its initial baseline.

The mean CD2 surface density of circulating CD56+ lymphocytes was reduced at each time point between T0.5 and T4 (Fig. 5B). Statistically, this drop of 8-14 MFI units did not reach significance until T4 in both placebo (P < 0.01) and naltrexone (P < 0.05) conditions. Differences between placebo and naltrexone conditions were not statistically significant.

    DISCUSSION
Top
Abstract
Introduction
Methods
Results
Discussion
References

The possible contribution of beta -endorphin secretion to the exercise-mediated increase in peripheral blood NKCA remains controversial. Although a role for this endogenous opioid was demonstrated during short-term, exhaustive exercise (11), no effect was seen during short-term, moderate-intensity exercise under hypoxic conditions (30). Several components of the present study support the view that the endogenous release of beta -endorphin does not mediate the enhancement of peripheral blood NKCA during prolonged moderate-intensity exercise.

First, NKCA increased by 85% relative to baseline levels during the first 60 min of exercise, despite the absence of any significant change in plasma beta -endorphin levels over the same period. Furthermore, if beta -endorphin were acting to enhance the cytolytic activity of NK cells, then measured levels of NKCA should have shown an increase during the last 60 min of exercise (when blood levels of beta -endorphin were elevated) and measurements of per NKCA should have reflected this change. However, when NKCA was expressed on a per-NK cell basis, values during exercise were slightly lower than preexercise levels (P > 0.05).

Second, any direct stimulatory effect mediated by beta -endorphin should be blocked by the potent, nonselective, opioid receptor antagonist naltrexone hydrochloride. However, this was not observed.

Third, the in vitro beta -endorphin-mediated augmentation of NKCA has been attributed to accelerated kinetics of lysis (i.e., cell activation) and an enhanced effector-tumor cell conjugate formation (34). Both cellular activation and NK cell-target cell conjugate formation are dependent on initial binding events, in which cellular adhesion molecules such as LFA-1 and CD2 play a major role (44). Conceptually, a beta -endorphin-mediated upregulation of LFA-1 or CD2 expression on NK cells could thus provide a mechanism by which exercise, secondary to the release of beta -endorphin, increases NKCA. However, the present results speak against such an explanation of the in vivo data. Although the expression of LFA-1alpha and CD2 on peripheral blood NK cells changed during exercise, the mechanism of this change could not be attributed to a beta -endorphin-mediated mechanism because the in vivo administration of naltrexone had no effect on this response. Nevertheless, the exercise-induced changes in LFA-1alpha and CD2 surface expression seem likely to influence the cytolytic response of NK cells during periods of physical exercise, and they thus deserve further examination.

Fourth, previous reports have suggested that the normal exercise-induced elevation in GH may be enhanced (10), inhibited (51), or unaffected (3) by opioid antagonism. Cortisol levels are typically enhanced, although more so in early recovery than during exercise (10). Such neuroendocrine changes could have secondary influences on NKCA. The in vivo effects of GH are unclear, but potentiating effects on human natural immunity have been described (7). Cortisol mediates both in vitro and in vivo inhibition of NKCA in humans (15, 16, 24). However, in the present study, naltrexone had no effect on circulating levels of either GH or cortisol, supporting the view that beta -endorphin bears neither direct nor indirect responsibility for the exercise-induced increase in NKCA.

Given our essentially negative conclusions regarding the role of beta -endorphin in modulating NKCA, it is necessary to consider possible limitations of our experiments. The patterns of exercise differed from that of Fiatarone et al. (11) but nevertheless were sufficient to yield a substantial production of beta -endorphin during the second hour of exercise. Nonexercise control values of plasma beta -endorphin were not determined for reasons of cost. However, circadian variations in resting concentrations of plasma beta -endorphin were not anticipated, because beta -endorphin is not secreted tonically, but rather requires some stimulation of the nervous system to be formed and released (2). The bout of activity was also sufficient to induce the anticipated changes in the NKCA of peripheral blood (14), and cytolytic activity was measured by a well-accepted, standard technique (i.e., in vitro incubation of PBMC with K562 tumor cells at an effector:target ratio of 50:1) (37). We may thus conclude that our results cannot be explained by an inadequate exercise stress or by problems in measuring NKCA.

Nor does it seem possible that the plasma concentrations of naltrexone were insufficient to block the effects of beta -endorphin. A plasma naltrexone concentration of 2 ng/ml offers effective opioid blockade (54). Given effective blood levels of 2-9 ng/ml (Fig. 1A), the dose and timing of naltrexone administration must be judged as effective. However, the results could have been confounded by agonistic effects of naltrexone. Naloxone, an opioid antagonist with structural similarity to naltrexone, has either no direct effect (26, 29) or has donor- and dose-dependent stimulatory and inhibitory effects on the cytolytic function of NK cells (33). Until further study resolves this issue, any potential direct in vivo effects of naltrexone on NKCA remain speculative.

A further factor potentially confounding the present results is the possible involvement of naltrexone-insensitive opioid receptors in the regulation of peripheral blood NKCA. Classically, opioid receptor binding involves the NH2-terminal sequence Tyr-Gly-Gly-Phe of beta -endorphin (1). However, lymphocytes also carry naltrexone-insensitive, nonopioid receptors that bind the COOH-terminal residues of the molecule (23). Whether opioid or nonopioid receptors are activated depends on the duration of exposure to and concentration of beta -endorphin (42), which likely explains the inverted-U dose-response effect of beta -endorphin. The removal of competitive binding between these receptors by a naltrexone-mediated blockade of classical opioid receptors might allow activation of nonopioid receptors. Binding of the COOH-terminal segments of beta -endorphin to nonopioid receptors on lymphocytes would inhibit cytolytic function (55), providing a further mechanism for the slight trend to a postexercise depression of NKCA seen during naltrexone administration.

Alternative hypothesis. In a rested individual, the cytolytic activity of PBMCs is mediated almost exclusively by the NK cell subset; this accounts for 5-15% of the total PBMC population (45). The measured NKCA thus reflects both the concentration of NK cells and the cytolytic activity of each NK cell; thus increases in either factor could enhance NKCA. The most obvious alternative to the beta -endorphin hypothesis, supported by our results, is that the exercise-induced increase in NKCA is secondary to an increased count of circulating NK cells (27). There is no simple relationship between changes in NK cell number and NKCA, but several researchers have suggested that the enhanced NKCA observed during a bout of physical exercise can be explained largely by an increased NK cell count rather than by an enhanced cytolytic capability per NK cell (4, 27, 49). In the present study, both the absolute and relative numbers of peripheral blood NK cells increased more than twofold during exercise, resulting in a 100% increase in the NK cell-to-K562 cell ratio (4:1 at rest compared with 9:1 during exercise). Furthermore, a tight correlation was found between the number of NK cells and the number of dead K562 in each assay tube (r = 0.82, P < 0.01), suggesting that increases in NK cell numbers could account for the demonstrated increase in NKCA.

Although not a primary objective of this study, postexercise changes in the study variables were examined. It is well accepted that exercise of sufficient intensity or duration can depress the peripheral blood NKCA postexercise (14). The trend for naltrexone to affect the postexercise NKCA may thus be an important observation. When naltrexone and placebo data were compared with the 2-h postexercise control value, there was a 32% drop in percent lysis in the naltrexone condition (P < 0.01) but only a 20% drop in percent lysis during the placebo condition (P > 0.05). This may reflect a release of cortisol in parallel with beta -endorphin. ACTH is released from the pituitary gland concomitantly with beta -endorphin (21), and it in turn stimulates the release of cortisol. In agreement with this, beta -endorphin and cortisol levels showed a similar kinetic response in the present study. Cortisol is potently inhibitory to the cytolytic activity of NK cells both in vitro and in vivo (15, 16, 24); however, addition of physiological concentrations of beta -endorphin can ameliorate the in vitro cortisol-induced inhibition of human NKCA (17). Conceptually, beta -endorphin release during the later stages of prolonged moderate physical exercise may thus prevent the overshoot of glucocorticoid-dependent immunosuppression, specifically counteracting the negative effects of elevated cortisol levels on NKCA. Therefore, the removal of the beta -endorphin-mediated, counter-regulatory mechanism may have tipped the balance between these two opposing modulatory hormones in favor of cortisol-induced immunosuppression. It is already known that peripheral blood lymphocytes downregulate their sensitivity and binding capacity for glucocorticoids after strenuous exercise (8, 50). The exercise-induced release of beta -endorphin may provide a further mechanism to prevent any untoward effects of elevated blood levels of cortisol. Although naltrexone did not affect the release of cortisol at the time points examined in this study, the previously described action of opioid antagonists in augmenting cortisol release during early recovery (i.e., 0-60 min) from moderate- to high-intensity exercise (10, 20) may have further promoted cortisol-induced immunosuppression.

It must be stressed that our study has not ruled out the possibility that the potentiating effect of beta -endorphin may require its concomitant presence during the cytolytic process. It has been well established that the in vitro exposure of NK cells to beta -endorphin can indeed promote NKCA in a dose-dependent fashion, perhaps stimulating what actually happens in vivo. In this study, the culture of NK cells ex vivo, in the absence of beta -endorphin, may diminish or negate its putative potentiating effect. This postulate appears to reconcile the apparent difference between the cytolytic stimulatory effect of beta -endorphin in direct contact with the effector cells during the culture period and the lack of apparent potentiation observed in this study, in which the effector cells were previously exposed to an exercise-induced physiological concentration of beta -endorphin but the neuropeptide was absent during the NK cell-mediated cytolytic process.

With the assumption that confounding mechanisms can be excluded, the results of this study do not support the hypothesis that beta -endorphin is an important mediator of the acute augmentation of peripheral blood NKCA during prolonged, moderate-intensity exercise because 1) NKCA is enhanced during the early stage of prolonged moderate-intensity exercise despite there being no concomitant increase in plasma beta -endorphin, 2) a later increase in plasma beta -endorphin does not further increase NKCA, and 3) the nonselective opioid antagonist naltrexone had no statistically measurable effect on the response of NKCA or on the regulation of other neuroendocrine or cell adhesion factors that could influence NKCA. We conclude that the primary explanation for the exercise-induced changes in NKCA during prolonged exercise is a concomitant change in NK cell count with essentially no change in per NKCA. On the other hand, the present results do not exclude a possible regulatory effect of beta -endorphin on NKCA early in the postexercise period.

    ACKNOWLEDGEMENTS

We gratefully acknowledge the expert technical assistance of Steven Combden, Sheila Petrongolo, Garry Seabrook, Capt. Yvonne Severs, and Ingrid Smith.

    FOOTNOTES

A preliminary report of the results of this paper was presented during the 44th Meeting of the American College of Sports Medicine held in Denver, CO (May 1997).

This research was supported in part by research grants from the Defence and Civil Institute of Environmental Medicine and Canadian Tire Acceptance Limited.

The costs of publication of this article were defrayed in part by the payment of page charges. The article must therefore be hereby marked "advertisement" in accordance with 18 U.S.C. §1734 solely to indicate this fact.

Address for reprint requests: G. A. Gannon, Human Protection and Performance Section, Defence and Civil Institute of Environmental Medicine, 1133 Sheppard Ave. West, Toronto, Ontario, Canada M3M 3B9.

Received 29 May 1998; accepted in final form 31 July 1998.

    REFERENCES
Top
Abstract
Introduction
Methods
Results
Discussion
References

1.   Akil, H., E. Young, S. J. Watson, and D. H. Coy. Opioid binding properties of naturally occurring N- and C-terminus beta -endorphins. Peptides 2: 289-292, 1981[Medline].

2.   Aronin, N., M. Wiesen, A. S. Liotta, G. C. Schussler, and D. T. Krieger. Comparative metabolic clearance rates of beta-endorphin and beta-lipotropin in humans. Life Sci. 29: 1265-1269, 1981[Medline].

3.   Brammert, M., and B. Hokfelt. The influence of naloxone on exercise-induced increase in plasma pituitary hormones and the subjectively experienced level of exhaustion in healthy males. Acta Endocrinol. 115: 125-130, 1987.

4.   Brenner, I., Y. D. Severs, P. N. Shek, and R. J. Shephard. Impact of heat exposure and moderate, intermittent exercise on cytolytic cells. Eur. J. Appl. Physiol. 74: 162-171, 1996.

5.   Carr, D. J., B. R. DeCosta, C.-H. Kim, A. E. Jacobson, K. L. Bost, K. C. Rice, and J. E. Blalock. Anti-opioid receptor antibody recognition of a binding site on brain and leukocyte opioid receptors. Neuroendocrinology 51: 552-560, 1990[Medline].

6.   Chiappelli, F., N. Yamashita, M. Faisal, M. Kemeny, R. Bullington, L. Nguyen, L. T. Clement, and J. L. Fahey. Differential effect of beta-endorphin on three human cytotoxic cell populations. Int. J. Immunopharmacol. 13: 291-297, 1991[Medline].

7.   Crist, D. M., and J. C. Kraner. Supplemental growth hormone increases the tumor cytotoxic activity of natural killer cells in healthy adults with normal growth hormone secretion. Metabolism 39: 1320-1324, 1990[Medline].

8.   DeRijk, R. H., J. Petrides, P. Deuster, P. W. Gold, and E. M. Sternberg. Changes in corticosteroid sensitivity of peripheral blood lymphocytes after strenuous exercise in humans. J. Clin. Endocrinol. Metab. 81: 228-235, 1996[Abstract].

9.   Dill, D., and D. Costill. Calculation of percentage changes in volumes of blood, plasma, and red cells in dehydration. J. Appl. Physiol. 37: 247-248, 1974[Free Full Text].

10.   Farrell, P., A. Gustafson, T. Garthwaite, R. Kalkhoff, A. J. Cowley, and W. Morgan. Influence of endogenous opioids on the response of selected hormones to exercise in humans. J. Appl. Physiol. 61: 1051-1057, 1986[Abstract/Free Full Text].

11.   Fiatarone, M., J. Morley, E. Bloom, D. Benton, T. Makinodan, and G. Solomon. Endogenous opioids and the exercise-induced augmentation of natural killer cell activity. J. Lab. Clin. Med. 112: 544-552, 1988[Medline].

12.   Froelich, C., and A. Bankhurst. The effect of beta -endorphin on natural cytotoxicity and antibody dependent cellular cytotoxicity. Life Sci. 35: 261-265, 1984[Medline].

13.   Gannon, G. A., S. G. Rhind, M. Suzui, P. N. Shek, and R. J. Shephard. Circulating levels of peripheral blood leucocytes and cytokines following competitive cycling. Can. J. Appl. Physiol. 22: 133-147, 1997[Medline].

14.   Gannon, G. A., P. N. Shek, and R. J. Shephard. Natural killer cells: modulation by intensity and duration of exercise. Exerc. Immunol. Rev. 1: 26-48, 1995.

15.   Gatti, G., R. Cavallo, M. L. Sartori, D. Del Ponte, R. Masera, A. Salvadori, R. Carignola, and A. Angeli. Inhibition by cortisol of human natural killer (NK) cell activity. J. Steroid Biochem. 26: 49-58, 1987[Medline].

16.   Gatti, G., R. Cavallo, M. L. Sartori, C. Marinone, and A. Angeli. Cortisol at physiological concentrations and prostaglandin E2 are additive inhibitors of human natural killer cell activity. Immunopharmacology 11: 119-128, 1986[Medline].

17.   Gatti, G., R. G. Masera, L. Pallavicini, M. L. Sartori, A. Staurenghi, F. Orlandi, and A. Angeli. Interplay in vitro between ACTH, beta -endorphin, and glucocorticoids in the modulation of spontaneous and lymphokine-inducible human natural killer (NK) cell activity. Brain Behav. Immun. 7: 16-28, 1993[Medline].

18.   Goldfarb, A., B. Hatfield, D. Armstrong, and J. Potts. Plasma beta-endorphin concentration: response to intensity and duration of exercise. Med. Sci. Sports Exerc. 22: 241-244, 1990[Medline].

19.   Goldfarb, A. H., and A. Z. Jamurtas. beta -Endorphin response to exercise: an update. Sports Med. 24: 8-16, 1997[Medline].

20.   Grossman, A., P. Bouloux, P. Price, P. L. Drury, K. S. L. Lam, T. Turner, J. Thomas, G. M. Besser, and J. Sutton. The role of opioid peptides in the hormonal responses to acute exercise in man. Clin. Sci. (Colch.) 67: 483-491, 1984[Medline].

21.   Guillemin, R., T. Vargo, J. Rossier, S. Minick, N. Ling, C. Rivier, W. Vale, and F. Bloom. beta -Endorphin and adrenocorticotropin are secreted concomitantly by the pituitary gland. Science 197: 1367-1369, 1977[Abstract/Free Full Text].

22.   Hatam, L., S. Schuval, and V. Bonagura. Flow cytometric analysis of natural killer cell function as a clinical assay. Cytometry 16: 59-68, 1994[Medline].

23.   Hazum, E., K.-J. Chang, and P. Cuatrecasas. Specific nonopiate receptors for beta -endorphin. Science 205: 1033-1035, 1979[Abstract/Free Full Text].

24.   Holbrook, N. J., W. I. Cox, and H. C. Horner. Direct suppression of natural killer activity in human peripheral blood leukocyte cultures by glucocorticoids and its modulation by interferon. Cancer Res. 43: 4019-4025, 1983[Abstract/Free Full Text].

25.   Jonsdottir, I. H., P. Hoffman, and P. Thoren. Physical exercise, endogenous opioids and immune function. Acta Physiol. Scand. 161: 47-50, 1997[Medline].

26.   Jonsdottir, I. H., C. Johansson, A. Asea, K. Hellstrand, P. Thoren, and P. Hoffman. Chronic intracerebroventricular administration of beta -endorphin augments natural killer cell cytotoxicity in rats. Regul. Pept. 62: 113-118, 1996[Medline].

27.   Kappel, M., N. Tvede, H. Galbo, P. Haahr, M. Kjaer, M. Linstow, K. Klarlund, and B. K. Pedersen. Evidence that the effect of physical exercise on NK cell activity is mediated by epinephrine. J. Appl. Physiol. 70: 2530-2534, 1991[Abstract/Free Full Text].

28.   Kay, N., J. Allen, and J. E. Morley. Endorphins stimulate normal human peripheral blood lymphocyte natural killer activity. Life Sci. 35: 53-59, 1984[Medline].

29.   Kay, N. E., J. E. Morley, and J. I. Allen. Interaction between endogenous opioids and IL-2 on PHA-stimulated human lymphocytes. Immunology 70: 485-491, 1990[Medline].

30.   Klokker, M., M. Kjaer, N. M. Secher, B. Hanel, L. Worm, M. Kappel, and B. K. Pedersen. Natural killer cell response to exercise in humans: effect of hypoxia and epidural anesthesia. J. Appl. Physiol. 78: 709-716, 1995[Abstract/Free Full Text].

31.   Mackinnon, L. T., T. W. Chick, A. van As, and T. B. Tomasi. Effects of prolonged intense exercise on natural killer cell number and function. In: Exercise Physiology: Current Selected Research., edited by C. O. Dotson, and J. H. Humphrey. New York: AMS, 1998, vol. 3, p. 77-89.

32.   Mandler, R. N., W. E. Biddison, R. Mandler, and S. A. Serrate. beta -Endorphin augments the cytolytic activity and interferon production of natural killer cells. J. Immunol. 136: 934-939, 1986[Abstract].

33.   Martin-Kleiner, I., and J. Gabrilovac. Naloxone modulates NK cell activity of human peripheral blood lymphocytes like an opioid agonist. Immunopharmacol. Immunotoxicol. 15: 179-197, 1993[Medline].

34.   Mathews, P. M., C. J. Froelich, W. L. Sibbitt, and A. D. Bankhurst. Enhancement of natural cytotoxicity by beta -endorphin. J. Immunol. 130: 1658-1662, 1983[Abstract].

35.   Monti, K. M., R. L. Foltz, and D. M. Chinn. Analysis of naltrexone and beta -naltrexol in plasma and urine by gas chromatography/negative ion chemical ionization mass spectrometry. J. Anal. Toxicol. 15: 136-140, 1991[Medline].

36.   Moretti, C., A. Fabbri, L. Gnessi, M. Cappa, A. Calzolari, F. Fraioli, A. Grossman, and G. M. Besser. Naloxone inhibits exercise-induced release of PRL and GH in athletes. Clin. Endocrinol. (Oxf.) 18: 135-138, 1983[Medline].

37.   Ortaldo, J. R. Cytotoxicity by natural killer cells: analysis of large granular lymphocytes. Methods Enzymol. 132: 445-457, 1986[Medline].

38.   Paneri, A. E., and P. Sacerdote. beta -Endorphin in the immune system: a role at last? Immunol. Today 19: 317-319, 1997.

39.   Pedersen, B., and H. Ullum. NK cell response to physical activity: possible mechanisms of action. Med. Sci. Sports Exerc. 26: 140-146, 1994[Medline].

40.   Pedersen, B. K., and J. M. Beyer. Characterization of the in vitro effects of glucocorticosteroids on NK cell activity. Allergy 41: 220-224, 1986[Medline].

41.   Plaut, M. Lymphocyte hormone receptors. Annu. Rev. Immunol. 5: 621-669, 1987[Medline].

42.   Prete, P., E. R. Levin, and A. Pedram. The in vitro effects of endogenous opiates on natural killer cells, antigen-specific cytolytic T cells, and T-cell subsets. Exp. Neurol. 92: 349-359, 1986[Medline].

42a.   Rhind, S. G., G. A. Gannon, P. N. Shek, and R. J. Shephard. Effects of 2 h of exercise and in vivo indomethacin on circulating PGE2 levels and NK cell activity (Abstract). Med. Sci. Sports Exerc. 30: S20, 1998.

43.   Rhind, S. G., G. A. Gannon, M. Suzui, P. N. Shek, and R. J. Shephard. Determination of natural killer cell cytolytic activity by flow cytometry: applications in studies of exercise immunology (Abstract). Int. J. Sports Med. 18: S113, 1997.

44.   Robertson, M., M. Caligiuri, T. Manley, H. Levine, and J. Ritz. Human natural killer cell adhesion molecules: differential expression after activation and participation in cytolysis. J. Immunol. 145: 3194-3201, 1990[Abstract].

45.   Robertson, M. J., and J. Ritz. Biology and clinical relevance of human natural killer cells. Blood 76: 2421-2438, 1990[Free Full Text].

46.   Schmidt, R. E., G. Bartley, H. Levine, S. F. Schlossman, and J. Ritz. Functional characterization of LFA-1 antigens in the interaction of human NK clones and target cells. J. Immunol. 135: 1020-1025, 1985[Abstract].

47.   Schmidt, R. E., J. P. Caulfield, J. Michon, A. Hein, M. M. Kamada, R. P. MacDermott, R. L. Stevens, and J. Ritz. T11/CD2 activation of cloned human natural killer cells results in increased conjugate formation and exocytosis of cytolytic granules. J. Immunol. 140: 991-1002, 1988[Abstract].

48.   Schwarz, L., and W. Kindermann. Changes in beta-endorphin levels in response to aerobic and anaerobic exercise. Sports Med. 13: 25-36, 1992[Medline].

49.   Shinkai, S., S. Shore, P. N. Shek, and R. J. Shephard. Acute exercise and immune function: relationship between lymphocyte activity and changes in subset counts. Int. J. Sports Med. 13: 452-461, 1992[Medline].

50.   Song, L.-N. Stress-induced changes in glucocorticoid receptors: molecular mechanisms and clinical implications. Mol. Cell. Endocrinol. 80: C171-C174, 1991[Medline].

51.   Spiler, I. J., and M. E. Molitch. Lack of modulation of pituitary hormone stress response by neural pathways involving opiate receptors. J. Clin. Endocrinol. Metab. 50: 516-520, 1980[Abstract].

52.   Targan, S., L. Britvan, and F. Dorey. Activation of human NKCC by moderate exercise: increased frequency of NK cells with enhanced capability of effector-target lytic interactions. Clin. Exp. Immunol. 45: 352-360, 1981[Medline].

53.   Timonen, T., C. G. Gahmberg, and M. Patarroyo. Participation of CD11a-c/CD18, CD2 and RGD-binding receptors in endogenous and interleukin-2-stimulated NK activity of CD3-negative large granular lymphocytes. Int. J. Cancer 46: 1035-1040, 1990[Medline].

54.   Verebey, K., J. Volavka, S. J. Mule, and R. B. Resnick. Naltrexone: disposition, metabolism, and effects after acute and chronic dosing. Clin. Pharmacol. Ther. 20: 315-328, 1976[Medline].

55.   Williamson, S., R. Knight, S. Lightman, and J. Hobbs. Differential effects of beta -endorphin fragments on human natural killing. Brain Behav. Immun. 1: 329-335, 1987[Medline].

Am J Physiol Regul Integr Compar Physiol 275(6):R1725-R1734


This article has been cited by other articles:


Home page
 Br. J. Sports. Med.Home page
R J Shephard
Sepsis and mechanisms of inflammatory response: is exercise a good model?
Br. J. Sports Med., August 1, 2001; 35(4): 223 - 230.
[Abstract] [Full Text] [PDF]

Home page
 Am. J. Physiol. Regul. Integr. Comp. Physiol.Home page
S. G. Rhind, G. A. Gannon, M. Suzui, R. J. Shephard, and P. N. Shek
Indomethacin inhibits circulating PGE2 and reverses postexercise suppression of natural killer cell activity
Am J Physiol Regulatory Integrative Comp Physiol, May 1, 1999; 276(5): R1496 - R1505.
[Abstract] [Full Text] [PDF]