INTRODUCTION

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

Obesity has multiple causes, and results from complex interactions between genetic, psychological, socioeconomic, and cultural factors. Morbid obesity is frequently associated with respiratory deficits that often include dysfunction of respiratory control, chest wall limitations, and upper airway narrowing (1, 2). As a potential consequence of the pulmonary dysfunction, exercise capacity is also reduced in morbidly obese individuals (3, 4).

Endogenous opioids are peptides that are widely distributed throughout the nervous system (5). Endogenous opioids such as -endorphins exert important physiologic effects and are activated by a variety of stressors that include exercise and obesity (6, 7). Indeed, exercise stress has been shown to increase -endorphin levels by from 2- to 10-fold during intense exercise stress (> 70% maximum oxygen consumption [O₂max]) in individuals of normal weight (6, 8, 9). Once secreted, endogenous opioids influence a wide range of functions, including cardiovascular and respiratory control, satiety, renal function, thermoregulation, pain perception, and metabolic rate (10). The depressant effects of exogenous opioids such as morphine on ventilation and O₂ are well known (11). Moreover, naloxone, an opioid receptor antagonist, increases ventilation during intense (80% O_2max) exercise in humans (14). Thus, higher breathing levels may be achieved once the opioid system is turned off by naloxone. In normal humans, however, naloxone infusion and the increased ventilation during severe exercise do not lead to an increased O_2max (14).

The obese Zucker (Z) rat, a genetic model of morbid obesity, presents many of the same physiologic abnormalities as noted in obese humans, including increased resting levels of -endorphins (17). The elevated endorphin levels in obesity are linked to overeating and impaired ventilatory control. In resting young (6-wk-old) but not older (10-mo-old) obese Z rats, the increased endorphin levels are indeed responsible for a diminished minute ventilation [E] at rest and during CO₂ challenges (12). Whether endogenous opioids modulate O_2peak in morbid obesity, and whether the modulation is mediated by central or peripheral pathways, is unknown and formed the basis of our study. The purpose of this study was therefore to investigate whether endogenous opioids modulate O_2peak in morbidly obese Z rats. A second goal of the study was to distinguish effects attributed to opioid receptors located within the central nervous system (CNS) from those mediated by opioid receptors located within peripheral structures. We hypothesized that endogenous opioids acting specifically on central but not on peripheral opioid receptors modulate E and O_2peak in obese but not in lean Z rats. Moreover, since both the number and function of brain opioid receptors decline as a function of age (20, 21), we further hypothesized that the opioid modulation of E and O_2peak would be noted predominantly in young and not in older obese Z rats.

In the study, we assessed E and O_2peak in obese Z rats at 6 and 16 wk of age. Studies were conducted after the administration of saline (control), naloxone hydrochloride (N_HCl), an opioid antagonist that readily crosses the blood-brain barrier, and naloxone methiodide (N_M), an opioid antagonist that does not cross the blood-brain barrier. The agents were given in a blinded, randomized protocol with no less than 72-h recovery between successive E or O_2peak tests. A parallel study design was used, with lean, age-matched Z rats serving as controls.

	METHODS

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

Animals

The studies were performed on 16 male Z rats, of which eight were of the lean phenotype and eight were of the obese phenotype. Z lean (Fa/?) and obese (fa/fa) rats were purchased at 4 wk of age from Vassar College, Poughkeepsie, NY. One lean and one obese rat were housed per cage. Ambient temperature was maintained at 21° C, and the animals were kept on an artificial 12-h light-dark cycle. The light period began at 7:00 A.M. Rats were provided with standard laboratory chow (Ralston Purina, St. Louis, MO) and water ad libitum. All protocols were approved by the Institutional Animal Care and Use Committee of the State University of New York at Buffalo. Animals underwent testing at 6 wk of age and again at 16 wk of age. Sixteen weeks of age was selected as our second time point because lean animals at 16 wk weigh approximately the same as 6-wk-old obese Z rats (Figure 1).

View larger version (8K): [in this window] [in a new window]   Figure 1.   Body weights of 6-wk-old lean and obese (Y-L, Y-O) and more mature 16-wk-old lean and obese (M-L, M-O) Z rats. The symbols represent the average weight of individual animals measured over the 18-d experimental period.

Techniques and Measurements

Pulmonary ventilation. Breathing patterns at rest were recorded with the barometric technique, of which details have previously been provided (12, 22). A cylindrical Plexiglas chamber with a volume of 4.6 L was used for the measurement of metabolic rate and breathing pattern. The rat was placed in the chamber within a cylindrical tube that acted as a restraining device. A flow of gas through the chamber was provided either by a wall-mounted compressed air source (during the preliminary habituation period and for CO₂ washout, as described in the experimental protocol) or from pressurized gas tanks (BOC Gases). Flow rate was controlled by a flowmeter (Dwyer Instruments Inc., Michigan City, IN). Flow through the chamber was kept at a steady rate of 1.5 L/min during measurement of gas exchange, but was raised to 4 L/min for a few minutes to aid wash-in at the time of change of the gas mixture. To evaluate ventilation, the chamber was completely sealed after momentary interruption of the flow through it, and the oscillations in pressure caused by breathing were recorded by a sensitive pressure transducer (Model LCVR; Celesco Transducer Products). The signal was received and amplified by a DC driver (Model 7PCPA; Grass Instruments, Quincy, MA) and was displayed on an oscillographic strip-chart recorder (Model 7 Polygraph; Grass Instruments, Quincy, MA). Sampling was done at a chart paper speed of 10 mm/s. Injection and withdrawal of known volumes of gas was done several times during the recording, for calibration purposes. Barometric pressure values for the nearest hour were obtained from the internet posting of the U.S. National Weather Service at the Buffalo International Airport.

For each condition, the average VT and breathing frequency (f) were calculated over a period corresponding to at least 10 successive breaths. Pulmonary ventilation (E) was also calculated (E = VT × f), and was expressed at body temperature-atmospheric pressure-saturation (ml BTPS/kg/min). Colonic temperature was measured continuously from a rectal probe (Tele-thermometer; Yellow Springs Instruments, Yellow Springs, OH), and was taken as representative of body temperature (T_b). Chamber temperature and humidity were monitored by means of a flow-through probe (Fisher Scientific, Pittsburgh, PA) mounted within the chamber.

o_2peak. The exercise test done to elicit peak aerobic activity (O_2peak) was performed in a metabolic treadmill (Columbus Instruments, Columbus, OH). A constant flow rate of 5 L/min through the metabolic treadmill was provided from pressurized air tanks (BOC Gases). Flow was controlled by a flowmeter (Dwyer Instruments).

Because of differences in exercise capacity between lean and obese animals, the exercise protocols for the two phenotypes were slightly different. The exercise protocol for both phenotypes elicited O_2peak within 12 to 15 min. The treadmill slope was set at 20% for lean animals and 10% for obese animals, and remained constant throughout the exercise test. The protocol for lean animals consisted of an initial treadmill speed of 10 m/min followed by a 3 m/min increase in speed every 2 min until the animal could no longer continue to run. Obese rats began exercise at 10 m/min followed by a 3 m/min increase every 3 min. These exercise protocols were repeated with lean and obese rats at 16 wk of age.

Oxygen consumption and CO₂ production. Oxygen consumption (O₂) and CO₂ production (CO₂) were measured in the barometric chamber or during the exercise test. The concentrations of CO₂ and O₂ entering and exiting the chamber (barometric or treadmill) were monitored with a CO₂ gas analyzer (Model CD-3A; Amatek Applied Electrochemistry, Sunnyvale, CA) and an O₂ analyzer (Model S-3A/1; Amatek Applied Electrochemistry) arranged in series. The calibrations and linearities of the gas analyzers were checked twice daily, using certified calibration gases (BOC gases). O₂ and CO₂ were calculated from the differences between O₂ and CO₂ inflow and outflow multiplied by the gas flow. Data are presented at standard temperature and pressure under dry conditions (STPD), corrected for the effective body mass exponent according to Refinetti (23), and are expressed in kilograms to the power of 0.75 (ml O₂ STPD/kg^0.75/min). Effective body mass (EBM) for lean and obese rats was calculated as 1.00 M^0.75 and 0.86 M^0.75 for lean and obese animals, respectively (23). EBM was used to minimize differences between lean and obese rats.

Experimental protocol. Animals were tested 30 min after a subcutaneous injection of equal volumes of saline (vehicle: 1 ml/kg), N_HCl (5 mg/kg), or N_M (5 mg/kg) (12). The solutions were prepared daily and placed in vials labeled as solutions I, II, or III. The agents were given in a blinded, randomized design with 72 h elapsing between successive tests. The investigators involved in the testing were blinded to the contents of the vials and remained so until both the ventilatory and exercise tests were completed and the data analyzed. Ventilation or exercise tests were performed on six separate occasions with no less than a 72 h recovery period between successive tests. Half the animals (four lean and four obese rats) were designated as Group A, and the other half were designated as Group B. Group A animals underwent ventilatory measurements first, followed by exercise testing, whereas Group B animals underwent initial exercise testing followed by the ventilatory function test. Ventilatory and exercise tests on any given animal were thus completed within an 18-d period. In an attempt to reduce the stress level during the study, we habituated all animals to the restraining device within the barometric chamber for 30 min, and to treadmill walking for 10 min at a speed of 10 m/min 2 d before the first ventilation or exercise testing period, respectively. To minimize any potential differences related to circadian rhythms, each rat was injected and tested at the same approximate time of day. Studies were performed between 8:00 A.M. and 4:00 P.M.

Statistical Analysis

A main effect was evaluated by repeated measures analysis of variance followed by post hoc pairwise comparisons using the least-significant-difference test. Specific group comparisons of interest were the effect of saline compared with N_HCl and of saline compared with N_M on physiologic parameters evaluated in lean and obese rats at a given age. In all cases, a difference was considered statistically significant at p < 0.05. All data presented in the text, tables, and figures represent mean ± SD.

	RESULTS

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

Six-week-old obese rats weighed about 30% more than age-matched lean animals (401 ± 44 g [mean ± SD] versus 286 ± 35 g, p < 0.01). At 16 wk of age, the obese animals remained approximately 30% heavier than the lean age-matched group (610 ± 31 g versus 417 ± 23 g, p < 0.01). However, 16-wk- old lean Z rats weighed the same as 6-wk-old obese animals (417 ± 23 g versus 401 ± 44 g, respectively; p = NS). Body weights of individual animals are shown graphically in Figure 1.

Ventilatory Parameters

In lean Z rats, ventilation was unaffected by the administration of either N_HCl or N_M. Thus, as compared with control values (saline), ventilatory parameters (E, f, VT) during room air breathing and during hypercapnic challenges were found to be unaltered in lean animals by the administration of N_HCl or N_M (Table 1). These findings were consistent for all ventilatory parameters measured, and were made in phenotypically lean Z rats at both 6 wk and 16 wk of age (Table 1, Figure 2). Values are shown for individual animals in Figures 3 and 4, and the ventilatory parameters shown in Figures 2-4 are corrected for body weight.

View larger version (42K): [in this window] [in a new window]   Figure 2.   Effects of saline, NM, and NHCl infusion on E in 6-wk-old lean (Y-L), 6 wk-old obese (Y-O), 16-wk-old lean (M-L), and 16-wk-old obese (M-O) Z rats breathing room air (top panel ), breathing 4% CO2 (middle panel ), and breathing 8% CO2 (bottom panel ). Bars indicate mean values ± SD. *p < 0.05 naloxone versus saline, **p < 0.01 naloxone versus saline.

View larger version (27K): [in this window] [in a new window]   Figure 3.   Ventilation, corrected for body weight, of individual 6-wk- old lean (open triangles) and obese (closed circles) Z rats breathing room air (top panel ), 4% CO2 (middle panel ), or 8% CO2 (bottom panel ) after administration of NM (left panels) and NHCl (right panels). The solid line represents the line of identity. Symbols falling above the line of identity represent animals whose ventilation increased after naloxone infusion and symbols falling below the line of identity represent animals whose ventilation decreased after naloxone infusion.

View larger version (26K): [in this window] [in a new window]   Figure 4.   Ventilation, corrected for body weight, of individual 16-wk- old lean (open triangles) obese (closed circles) Z rats breathing room air (top panel ), 4% CO2 (middle panel ), or 8% CO2 (bottom panel ) after administration of NM (left panels) and NHCl (right panels). The solid line represents the line of identity. Symbols falling above the line of identity represent animals whose ventilation increased after naloxone infusion and symbols falling below the line of identity represent animals whose ventilation decreased after naloxone infusion.

In sharp contrast, ventilatory parameters in obese Z rats revealed an age-specific modulation by naloxone administration. Administration of N_M had no significant effect on any pulmonary parameter in either 6- or 16-wk-old obese Z rats (Figure 2). N_HCl infusion, however, altered ventilation in 6-wk- old but not in 16-wk-old obese Z rats (Figure 2). N_HCl infusion increased resting E in 6-wk-old obese rats by 34% over the values obtained with saline (721 ± 154 ml/kg/min versus 965 ± 216 ml/kg/min, p < 0.05), which was specifically the result of an increased VT (Table 1). In 16-wk-old obese Z rats, N_HCl infusion had no effect on either E or on the pattern of breathing (Table 1).

Ventilation in response to hypercapnic gas challenges was also increased after the administration of N_HCl as compared with vehicle (Figure 2) in 6-wk-old, but not 16-wk-old, obese Z rats. In 6-wk-old obese Z rats, N_HCl infusion significantly increased E in response to 4% CO₂, by 24% (925 ± 110 versus 1,212 ± 201 ml/kg/min, p < 0.01), and in response to 8% CO₂, by 21% (1,233 ± 172 versus 1,565 ± 327 ml/kg/min, p < 0.01). E values are shown for individual animals in Figure 3. The higher E values achieved after administration of N_HCl were the result of a larger VT and not of an increase in f (Table 1). Once the obese rats reached 16 wk of age, N_HCl no longer increased E in response to either 4% or 8% CO₂ challenges (Figure 4).

Exercise Test

In accord with the ventilatory data, O_2peak in lean animals was unchanged relative to control values after infusion of N_M or N_HCl (Table 2). The inability of naloxone to alter O_2peak was consistent, whether lean animals were studied at 6 wk or 16 wk of age (Figure 5).

View larger version (27K): [in this window] [in a new window]   Figure 5.   Peak oxygen consumption corrected for effective body mass ( O2peak/EBM) of individual lean (open triangles) and obese (closed circles) Z rats treated with NM (left panels) and NHCl (right panels). The solid line represents the line of identity. Symbols falling above the line of identity represent animals whose O2peak increased after naloxone infusion and symbols falling below the line of identity represent animals whose O2peak decreased after naloxone infusion.

In obese Z rats, N_M failed to alter O_2peak as compared with values obtained with saline treatment in both 6-wk and 16-wk- old animals. In contrast, N_HCl significantly increased values of O_2peak in all eight 6-wk-old obese rats (average increase: 8.9 ± 6.9%). N_HCl had no effect on O_2peak in obese animals retested at 16 wk of age (Table 2). O_2peak values following N_HCl and N_M administration as compared with control (saline) values are shown for individual lean and obese animals in Figure 5.

	DISCUSSION

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

Our major findings in this study can be summarized as follows: (1) endogenous opioids do not modulate ventilation in phenotypically lean Z rats; (2) ventilation appears to be modulated by endogenous opioids in young obese Z rats; (3) opioid modulation of ventilation in obese Z rats is not observed in more mature animals; (4) blocking of the effect of endogenous opioids in young obese rats allows them to achieve an increased O_2peak; and (5) the modulation of ventilation and peak metabolism is attributed to endogenous opioids acting on receptors located within the CNS.

Role of Endogenous Opioids in Lean Z Rats

In lean Z rats, naloxone failed to alter any ventilatory parameter or O_2peak. Although exogenous opiates are known to exert a depressant action on ventilation, endogenously released opioids are reported to have no major influence on the control of breathing, O₂ uptake, or CO₂ output in healthy humans of normal weight (24) or in lean Z rats (12). Exercise stress, however, is a well-known stimulus to the release of endogenous endorphins. In fact, maximal exercise elicits a 2- to 10-fold increase in circulating endorphin levels (15). Despite significant endorphin release during strenuous exercise in humans of normal weight, naloxone infusion in humans failed to modify O_2max during exercise in several studies (15, 16, 24), despite some reports of greater ventilation at peak exercise after naloxone infusion (14). Since in normal humans O_2max is not limited by constraints imposed by the ventilatory system, increases in E may not translate into increases in O_2max.

Role of Endogenous Opioids in Obese Z Rats

In accord with our earlier observation (12), N_HCl in young obese Z rats increased both resting ventilation and ventilation in response to CO₂. In our previous study (12), the increased E following N_HCl infusion might have been due to antagonism of opioid receptors located in either the central or peripheral nervous system or both, and/or naloxone could have exerted a direct potentiating effect on respiratory muscle contractility (25). In the present study, however, use of N_M, a quaternary opioid receptor antagonist that does not penetrate the blood-brain barrier after systemic administration, was used in conjunction with N_HCl to distinguish peripheral from central actions of opioids (26). Some studies have suggested that peripheral hypoxic chemoreception and cardiovascular responses in other species may be mediated by peripheral opioid receptors (27). Nevertheless, in the present study, peripheral opioid antagonism failed to alter body temperature, ventilatory parameters during quiet breathing or in response to CO₂ challenges, or O_2peak in lean or obese Z rats at either of the two ages studied. The findings in the study allow us to conclude that the increased resting endogenous -endorphin levels noted in obese Z rats (17) depress ventilatory drive by acting exclusively on receptors located within the CNS. Opioid receptors are widely distributed in the medulla and pons, especially in the area of the nucleus tractus solitarius and the nucleus ambiguus, both of which structures are linked to respiratory control (5, 28). Central respiratory chemosensitivity has been ascribed to CO₂-sensitive neurons located on the ventral brainstem surface. Naloxone applied topically to the caudal ventral medullary surface augments the activity of these CO₂-sensitive neurons by more than 300% (28). Our finding that N_HCl increased E both at rest and during CO₂ challenges suggests that endogenous opioids may be involved in the central regulation of respiration by modulating CO₂ sensitivity. At the molecular level, endorphins function as inhibitors of neural activity, generating inhibitory pre- or postsynaptic hyperpolarizing potentials via K⁺ channels (29). Thus, our results imply that the hyperendorphinemia associated with obesity exerts an important role in central control of breathing in young obese Z rats.

The increased ventilation and CO₂ chemosensitivity noted in the young obese Z rats in our study after N_HCl infusion, and the associated increased exercise performance noted as a higher O_2peak (Figure 5), suggest that these responses may be interrelated. In humans, exercise ventilation correlates positively with ventilatory chemoresponsiveness (30). Thus, low responders to hypoxic and hypercapnic challenges breathe less during exercise than do high responders. Assuming that the same relationship exists in rats, one could speculate that obese Z rats achieved a higher level of ventilation during exercise after N_HCl infusion. At present, however, we have no means of measuring E in rats during exercise, and simply achieving a higher E during exercise would not necessarily translate into a higher O_2peak unless maximal aerobic power were constrained by the respiratory system. Although the respiratory system is not considered a limiting factor to oxygen consumption during exercise under normal conditions, owing to a large ventilatory reserve at peak exercise, this may not be so in certain pathologic situations that affect the respiratory system, such as aging, lung disease, and obesity (3, 4). The mass loading due to fat deposition in and around the chest wall may indeed restrict ventilation in morbid obesity. Moreover, as shown in the present study, increased -endorphin levels may blunt the ventilatory drive and result in a lower E being attained at O_2peak. Thus, we can speculate that after the infusion of N_HCl, the increased chemosensitivity and enhanced respiratory drive seen in young obese Z rats in our study resulted in higher E and a concomitant increase in O_2peak.

The foregoing argument linking the increase in O_2peak to an increased ventilatory capacity with N_HCl becomes more convincing when one considers that N_HCl failed to alter either E or O_2peak in obese Z rats restudied at 16 wk of age. We have previously reported (12) that N_HCl infusion did not alter resting E or ventilatory chemoresponsiveness to hypoxic or hypercapnic challenges in 10-mo-old Z rats. The current findings show that the modulating role of the endogenous endorphins is lost at a much earlier age than previously demonstrated. In accord with our findings, Piva and colleagues (21) reported an age-related decline of µ-opioid receptors in the brain and hypothalamus of aged male rats.

Can the results in the present study have been artifactual? Naloxone was administered subcutaneously, and it can be argued that uptake of the exogenous compound will depend on peripheral circulation. With aging, it is possible that peripheral circulation is compromised, such that naloxone was not adequately absorbed in the older morbidly obese animals. Although naloxone infusion had no demonstratable effect on E or O_2peak in the obese rats at 16 wk, we did note that obese rats studied at 6 wk and 16 week of age had identical decreases in body temperature after N_HCl administration (Table 1), providing evidence that the drug was indeed being absorbed in 16-wk-old obese animals, and at a minimum was antagonizing receptors located in the CNS that are responsible for thermoregulation. It is also possible that the central depressant effects of the increased -endorphin levels in the young obese Z rats were reversed after naloxone infusion, but that the increased central output from the respiratory centers did not increase E because of peripheral limitations of the ventilatory system. Indeed, obese Z rats show an increased chest-wall stiffness (31) and weak respiratory muscles (32, 33), both of which features are exacerbated by age. Additional studies, however, would need to be conducted in order to address this issue.

Other nonventilatory factors may have contributed to the increase in O_2peak observed in obese Z rats after naloxone infusion. The endogenous opioids are neuromodulators affecting a wide range of systems. Fatigue in a O_2peak test, defined as the inability to maintain the required aerobic power, is a complex phenomenon. Since motivation to run was provided by mild foot shock in our study, naloxone infusion may have increased pain sensitivity to the electric shock, which would have provided additional motivation to continue to run. In trained humans, naloxone infusion increased the perception of pain but had no effect on performance (34). It is difficult to conceive that this mechanism, acting via altered pain perception, would be present at 6 wk of age and not at 16 wk.

Significance of Study Findings

Since the data in our study were generated with obese Z rats, the question should be raised of whether increased -endorphin levels have any effect in morbidly obese humans (18). Recent appetite-suppressant therapy in morbidly obese humans has used long-term opioid antagonism with success (35), providing evidence that increased levels of endorphins do indeed exert a physiological effect. Do they also modulate ventilation and thereby influence exercise performance? Indirect evidence does suggest that ventilation in human obesity is under endogenous endorphin modulation. Orlowski and colleagues (36) presented a case study of a 20-mo-old female child admitted to a clinic with respiratory failure caused by obesity hypoventilation syndrome. The child weighed 24 kg (200% of ideal body weight) at the time of admission. A single dose of naloxone (10 µg/kg) given early in the course of the child's respiratory failure resulted in a dramatic improvement in ventilation that lasted approximately 3 to 4 h. Continuous naloxone infusion led to further improvements in the child's respiratory status. Despite these responses, naloxone infusion was discontinued after 5 d, and progressive respiratory deterioration recurred (36). The next question is whether the respiratory modulation caused by increased endorphin levels is lost as obese humans age. Again, no controlled study linking the blunted ventilatory drive, often observed in morbid obesity, to the elevated endogenous endorphins has to our knowledge been reported. Atkinson and colleagues (37), however, administered naloxone to 10 obese humans with sleep apnea. Nine of the patients had a lower desaturation index after naloxone treatment. It was concluded that opioid antagonists hold promise in the treatment of sleep apnea, and that the endogenous opioid system may contribute to the development of sleep apnea. We can therefore speculate that the endogenous opioids exert a significant physiologic effect on breathing in obese adults. Additional studies will be required to investigate whether exercise performance in obese humans is modulated by endogenous endorphins, and whether this modulation is modified with age.