INTRODUCTION

Top Abstract Introduction Methods Results Discussion References

CHEMICAL STIMULATION OF vagal C fiber endings in the lungs and heart evokes the well-known pulmonary depressor chemoreflex, coronary chemoreflex, and pulmonary respiratory chemoreflex (9, 10, 18). These reflexes decrease heart rate and blood pressure or induce apnea followed by rapid shallow breathing. Many pharmacological agents used to evoke these viscerovisceral reflexes also evoke a viscerosomatic reflex that inhibits somatomotor activity. For example, intravenous injection of the exogenous chemical phenyl biguanide inhibits the knee-jerk reflex in the -chloralose-anesthetized cat and locomotion in the conscious rat and decerebrate cat (12, 13, 21, 30). Similarly, intrapericardial nicotine injection inhibits the knee-jerk reflex and locomotion in the cat (28, 29). Like the sensory limb of the reflex cardiovascular and respiratory changes, the afferent limb of the reflex somatomotor inhibition travels in the vagus nerves (11, 14, 28, 30).

Several groups have simulated physiological conditions that evoke vagal C fiber discharge to evaluate the relevance of somatomotor inhibition evoked by stimulation of vagal endings in the cardiopulmonary region. Pulmonary congestion, induced by raising left atrial pressure using an inflatable balloon placed in the left atrium, inhibits locomotion via a vagal reflex in the decerebrate cat (30). Lung expansion to volumes similar to those occurring during exercise inhibits the knee-jerk reflex via a vagal reflex in the Nembutal-anesthetized dog (6). In addition, intravenous injection of the endogenous chemical serotonin inhibits locomotion in the decerebrate cat and the knee-jerk reflex in the -chloralose anesthetized cat (13, 30). These findings have supported the concept that the discharge of vagal C fibers from the lung (6, 13, 30), and possibly the heart (30), provides negative feedback to the somatomotor system, reflexly decreasing motor activity when demands made on the cardiopulmonary system exceed its capacity to meet them (16).

The purpose of the present study was to determine whether the inflammatory mediator thromboxane A₂ could be an endogenous chemical capable of reflexly inhibiting somatomotor activity. Thromboxane A₂ is produced in the lung during a variety of pathological conditions, including systemic infection, and during muscle and intestinal ischemia-reperfusion (3, 32, 34). In the lung, thromboxane A₂ elicits rapid shallow breathing through a reflex whose afferent arm consists of vagal C fibers (22, 31, 32). Because the reflex breathing pattern is similar to the breathing pattern evoked by the vagally mediated pulmonary respiratory chemoreflex (9, 10, 31), the possibility was investigated that thromboxane A₂ evokes a viscerosomatic reflex. The hypothesis was tested that the stable thromboxane A₂ mimetic U-46619 (7) inhibits the knee-jerk reflex via a vagal reflex from the lung in the -chloralose-anesthetized cat.

	METHODS

Top Abstract Introduction Methods Results Discussion References

General. Studies were performed on 11 cats weighing between 3.9 and 4.6 kg. All experiments were performed in accordance with the Guiding Principles for the Care and Use of Animals established by the American Physiological Society. Anesthesia was induced using halothane. The cat was placed in a sealed plastic chamber into which a gaseous mixture of oxygen (5 l/min) and halothane (5%) was introduced. After induction, anesthesia was maintained by delivering oxygen (2 l/min) and 3% halothane through a face mask placed over the cat's nose and mouth. Catheters were placed near the right atrium, near the aortic arch, and into the pericardial sac. The venous catheter was inserted into the right external jugular vein at the level of the cricoid cartilage and advanced 14-15 cm toward the right atrium. The intravenous catheter was used to infuse fluids and inject U-46619 into the right side of the heart. The intra-arterial catheter was inserted into the right common carotid artery at the level of the cricoid cartilage and advanced ~10-11 cm toward the aortic arch. The intra-arterial catheter was used to monitor blood pressure and inject U-46619 into the aorta. Placement of these catheters was confirmed postmortem. U-46619 was also delivered onto the surface of the heart. The pericardial sac was exposed as described previously (see Fig. 1 in Ref. 15). The cat was placed supine, and the ribs were retracted widely. A cradle surrounding the heart was formed from the pericardial sac. A small hole was placed through the ventral surface of the pericardial sac, through which U-46619 was delivered using a syringe. The trachea was intubated and the cat was ventilated using a Harvard Respirator (model 681). The respiratory effects of U-46619 injection (32) were not measured. The cervical vagi were isolated and loosely snared using umbilical tape. A complete vagotomy was confirmed during the protocol by the investigator's ability to remove the snares from the cut ends of each vagus nerve.

During the reflex study the cat was maintained on anesthesia using -chloralose (50 mg/kg iv). Halothane was withdrawn slowly as -chloralose was administered. Surgical anesthesia was determined by the absence of a withdrawal reflex to noxious pinching of the cat's toe pad. Supplemental doses of -chloralose (25 mg/kg) were given as needed. Arterial pH, PCO₂, and PO₂ were monitored every 60 min using a Ciba-Corning 238 pH/blood gas analyzer. Arterial blood gas values were maintained within the normal range (pH: 7.32-7.43; PCO₂: 32-35 mmHg; and PO₂: >85 mmHg). Arterial pH and PCO₂ were corrected by infusing sodium bicarbonate intravenously and by adjusting the ventilator. PO₂ was maintained by bleeding 100% O₂ into the intake line of the ventilator. Atelectasis was prevented in this open-chest preparation by submerging the ventilator's exhalation tube 2-5 cm below water level.

Knee-jerk reflex. The knee-jerk reflex was elicited using methods previously described (32). A small hole was drilled through the distal end of the femoral shaft. A metal rod was inserted through the hole and the femur was immobilized by clamping the rod on both ends. With the cat placed supine, the knee-jerk reflex was evoked by striking the patellar tendon with solenoid-driven reflex hammer. The solenoid was activated by a Grass stimulator (S88) using square-wave pulses 10 ms in duration. Knee-jerk reflexes were elicited every 1.5 s. The reflex hammer was constructed of lightweight aluminum, and its striking tip was made of hard rubber. The hammer was rapidly retracted by a rigid spring attached to the hammer. The solenoid-reflex hammer assembly was positioned relative to the patellar tendon using a rack and pinion. With this preparation, the knee-jerk reflex can be elicited for at least 3 min with little loss in tension (15).

The magnitude of the knee-jerk reflex was determined using a force transducer (FT-10, Grass). A small, light-weight U-shaped bracket was rigidly fastened to the transducer's cantilever beam. The bracket cradled the tibia so that the cantilever beam was coupled directly to the tibia. The transducer measured the isometric force produced by extensor muscles of the thigh. The output of the transducer was amplified and played back onto a chart recorder (TA5000, Gould) and into a data acquisition system (RC Electronics).

U-46619 injection. The stable thromboxane A₂ mimetic U-46619 (Cayman Chemical, Ann Arbor, MI) was used in these experiments because thromboxane A₂ is unstable being rapidly metabolized to thromboxane B₂ (31). U-46619 binds to the same receptor as thromboxane A₂ (7) and has been used in studies of vagal reflexes from the lung (22, 31, 32). U-46619 injections (0.8 ± 0.08 µg/kg) were made near the right atrium (n = 11) and aortic arch (n = 7) and into the pericardial sac (n = 4) of the same cats. Injections into the right atrium and aorta were delivered rapidly (~2 s duration); injections into the pericardial sac were delivered slowly (~9 s duration). At the end of a protocol using an intrapericardial injection, U-46619 was washed from the pericardial sac using four successive injections and aspirations of 2-3 ml of saline to reduce absorption of the mimetic. All injections were separated by 20 min. Injections of U-46619 were repeated after cutting of the cervical vagus bilaterally.

Data analysis. The amplitude of each knee-jerk reflex was determined offline with the use of the data acquisition system. Amplitudes were transformed into three-point running averages. The minimum three-point running average during the 30 s before U-46619 injection and the minimum three-point running average during the 60 s after U-46619 injection were determined for each protocol in each cat. The three-point averages before and after U-46619 injection were compared before and after vagotomy using a two-way analysis with repeated measures for each of the three routes of U-46619 injection. Peak systolic blood pressures were compared similarly. When appropriate, a Student-Newman-Keuls post hoc test was performed. P < 0.05 was considered significant. All knee-jerk reflexes were normalized to the before injection value; thus the "before injection" amplitudes in Fig. 2 are presented as 100%.

	RESULTS

Top Abstract Introduction Methods Results Discussion References

Knee-jerk reflex in response to intravenous U-46619. The mean maximum tension produced by the knee-jerk reflex was 306 ± 21 g (range 154-471 g, n = 22). Figure 1 is a recording demonstrating the effect of intravenous U-46619 injection on the knee-jerk reflex and arterial blood pressure before and after vagotomy in one cat. The knee-jerk reflex was inhibited by 28% and peak systolic pressure increased 60 mmHg (Fig. 1A). Cutting the cervical vagus bilaterally abolished the inhibition, but the pressor response remained intact (Fig. 1B). Figure 2A contains group data showing the effect of intravenous U-46619 injection on the knee-jerk reflex before and after vagotomy in 11 cats. Before vagotomy, intravenous U-46619 injection significantly decreased the magnitude of the knee-jerk reflex by 25 ± 6%. Visual inspection of individual records indicated that the onset of inhibition occurred within 17.7 ± 1.8 s (range 3-31.5 s) of intravenous U-46619 injection. Vagotomy abolished the U-46619-induced inhibition of the knee-jerk reflex, reducing the inhibition to 1 ± 2% (n = 10). In 1 of the 11 cats, vagotomy clearly did not abolish the U-46619-induced inhibition (31% before and 58% after vagotomy).

View larger version (38K): [in this window] [in a new window]   View larger version (42K): [in this window] [in a new window]   Fig. 1.   Change in knee-jerk reflex in response to intravenous U-46619 before (A) and after vagotomy (B) in the same cat. Horizontal bar indicates timing and duration of U-46619 injection. Insets are individual knee-jerk reflexes using an expanded time based on the corresponding single-headed arrow. Graph was drawn from digitized data collected for 0.5 s (shown in insets) every 1.5 s using the "burst mode" of the data acquisition system. This mode of data acquisition introduced an artifact into the blood pressure tracing by omitting 1 s of the arterial pressure trace every 1.5 s. Double-headed arrows pointing to the relatively horizontal lines indicate where a data point at the end of 0.5 s of data collection is connected to the next data point at the beginning of the 0.5 s of data collection. x-Axis represents real time.

View larger version (19K): [in this window] [in a new window]   Fig. 2.   Magnitude of knee-jerk reflex in response to intravenous (A), intra-arterial (B), and intrapericardial (C) U-46619 injection before and after vagotomy. *P < 0.05 compared with before injection.

Knee-jerk reflex in response to intra-arterial U-46619. Figure 2B shows the effect of intra-arterial U-46619 injection on the knee-jerk reflex before and after vagotomy in seven cats. Intra-arterial U-46619 did not significantly change the knee-jerk reflex before (7 ± 5%) or after (2 ± 3%) vagotomy. However, in two of these seven cats intra-arterial U-46619 decreased the amplitude of the knee-jerk reflex by 21 and 22%. Similarly, intravenous U-46619 in these same two cats decreased the amplitude of the knee-jerk reflex by 24 and 31%, respectively. The latency to the onset of inhibition was nearly twice as long after intra-arterial U-46619 injection compared with intravenous U-46619 injection (51.0 vs. 21.0 s and 34.5 vs. 15.0 s, respectively, for each of the 2 cats). Cutting the cervical vagus bilaterally abolished the inhibition induced by intra-arterial U-46619 in one cat (from 21% before to 0% after vagotomy) and reduced the inhibition in the other cat (from 22% before to 14% after vagotomy).

Knee-jerk reflex in response to intrapericardial U-46619. Figure 2C shows the effect of intrapericardial U-46619 injection on the knee-jerk reflex before and after vagotomy in four cats. Intrapericardial U-46619 did not significantly change the knee-jerk reflex before (0 ± 6%) or after (0 ± 3%) vagotomy. By contrast, intravenous U-46619 decreased the knee-jerk reflex 23 ± 9% in these same four cats, and intra-arterial U-46619 decreased the knee-jerk reflex 22% in one of these four cats.

Blood pressure. Figure 3 contains group data showing peak systolic blood pressure before and after U-46619 injection and before and after vagotomy. Intravenous U-46619 (Fig. 3A) injection significantly increased peak systolic pressure 53 ± 7 mmHg (from 207 ± 13 before to 259 ± 16 mmHg after injection, n = 11). The hypertensive response to U-46619 persisted after vagotomy, with the peak systolic pressure increasing 65 ± 9 mmHg (from 225 ± 20 before to 290 ± 23 mmHg after injection).

View larger version (18K): [in this window] [in a new window]   Fig. 3.   Peak systolic blood pressures in response to intravenous (A), intra-arterial (B), and intrapericardial (C) U-46619 injection before and after vagotomy. *P < 0.05 compared with before injection.

Similarly, intra-arterial U-46619 (Fig. 3B) significantly increased peak systolic pressure before and after vagotomy. Before vagotomy, peak systolic pressure increased 41 ± 11 mmHg (from 211 ± 29 before to 251 ± 36 mmHg after injection, n = 7); after vagotomy, peak systolic pressure increased 58 ± 12 mmHg (from 204 ± 32 before to 262 ± 36 mmHg after injection).

In contrast, intrapericardial U-46619 injection (Fig. 3C) did not change peak systolic pressure (from 218 ± 37 before to 223 ± 40 mmHg after injection, n = 4). After vagotomy, peak systolic blood pressure also did not change in response to intrapericardial U-46619 injection (193 ± 55 before and 195 ± 56 mmHg after injection).

	DISCUSSION

Top Abstract Introduction Methods Results Discussion References

This study demonstrated that the thromboxane A₂ mimetic U-46619 inhibited the knee-jerk reflex. The somatomotor inhibition was a vagal reflex because it was abolished by cutting the cervical vagi. The afferent arm of this vagally mediated viscerosomatic reflex likely arose from sensory nerve endings in the pulmonary region because U-46619 injected near the right atrium significantly inhibited the knee-jerk reflex whereas U-46619 injected onto the surface of the heart did not inhibit the knee-jerk reflex and U-46619 injected into the aortic arch infrequently inhibited the knee-jerk reflex. In the two instances where aortic injection clearly inhibited the knee-jerk reflex, the latency to the onset of attenuation was longer compared with right atrial injection. The longer latency after intra-arterial injection is consistent with recirculation of U-46619 to the lung. Alternatively, the inhibition evoked by intra-arterial U-46619 injection might have been caused by an extravagal mechanism.

Activation of vagal C fibers likely caused the somatomotor inhibition. U-46619 undoubtedly stimulates vagal afferent C fibers (22). However, U-46619 also activates slowly (SAR) and rapidly (RAR) adapting pulmonary receptors, but their discharge frequency is four times weaker, on average, compared with the C fiber discharge (22). Moreover, SARs and RARs do not contribute to the vagally mediated respiratory reflexes initiated by U-46619, at least in the cat (5, 22). Therefore it seems unlikely, although not proven directly, that vagal afferent fibers other than C fibers caused the inhibition of the knee-jerk reflex in the present study.

The rise in peak systolic blood pressure was not responsible for the somatomotor inhibition because vagotomy abolished the U-46619-induced inhibition of the knee-jerk reflex but did not abolish the pressor response. In fact, peak systolic blood pressures were slightly higher after vagotomy. In the one cat where vagotomy did not abolish the somatomotor inhibition, an extravagal mechanism may underlie the inhibition. Similarly, vagal and extravagal mechanisms underlie the inhibition of the knee-jerk reflex produced by intravenous serotonin injection, although the extravagal pathway remains unknown (13). It is worthwhile noting in this one cat that U-46619 evoked a pressor effect greater than the pressor effect in any of the other 10 cats (84 vs. 71 mmHg before vagotomy and 109 vs. 86 mmHg after vagotomy, respectively). The hypertension or circulation of U-46619 to the carotid sinus may have caused the inhibition reflexly via the ninth cranial nerve because mechanical and chemical stimulation of the carotid sinus nerve can inhibit motor output (8, 17). Alternatively, the inhibition may have been unrelated to the large change in blood pressure but caused by stimulation of group III and IV muscle afferents (23), the discharge of which may inhibit motor output (20).

The pressor response to intra-arterial and intravenous U-46619 suggests that the U-46619-induced vasoconstriction likely predominated in the systemic circulation. Thromboxane A₂ contracts vascular smooth muscle in the systemic and pulmonary circulations (31, 32). U-46619 injected into the abdominal aorta (23) increases arterial blood pressure to a level similar to that measured in the present study after intra-arterial U-46619 injection. Intravenous U-46619 injection consistently increases pressure in the pulmonary circulation, but it can increase, decrease, or have no effect on pressure in the systemic circulation. For example, in the chloralose-urethan-anesthetized cat, right ventricular pressure increases by nearly 100%, but systemic arterial pressure decreases by nearly 20% during very slow (6-8 min duration, ~1.5 µg) intravenous U-46619 infusion (32). Slightly faster (3-4 min duration, ~1.8 µg) U-46619 infusion increases right ventricular pressure only 75% and increases systemic arterial pressure 3-17% in the cat (22). However, in the chloralose-urethan-anesthetized rabbit right ventricular pressure increases by 60% but systemic arterial pressure decreases by 9% during rapid (10 s duration, 0.5 µg/kg) intravenous U-46619 injection (5). In the unanesthetized goat, pulmonary and systemic arterial pressures increase 350 and 30%, respectively, during moderately slow (5 min duration, 2 µg/kg) intravenous U-46619 infusion (4). Carrithers et al. (4, 5) suggest that the relative strength of systemic versus pulmonary vasoconstriction determines the change in systemic arterial blood pressure. Pulmonary vasoconstriction would decrease left ventricular end-diastolic filling, thereby decreasing cardiac output. The relative strength of the systemic vasoconstriction would determine the hypotensive effect. In the present experiments, right ventricular pressures were not measured, but rapid (2 s duration, 0.8 µg/kg) intravenous U-46619 injection increased peak systolic pressure by nearly 25%. It seems reasonable to suggest that the rapid injection of U-46619 provided a short duration over which this vasoconstrictor was present in the pulmonary circulation and thus increased the relative strength of systemic versus pulmonary vasoconstriction. Clearly, the pressor response was not a vagal reflex because the hypertension persisted after vagotomy.

The absence of a depressor response after intrapericardial U-46619 injection suggests that thromboxane A₂ is not an endogenous chemical capable of evoking the coronary chemoreflex. Previous studies demonstrate that intravenous U-46619 evokes a breathing pattern (22, 31, 32) similar to the vagally mediated pulmonary respiratory chemoreflex (9, 10), and many chemicals, including serotonin, phenyl biguanide, and nicotine, that evoke the pulmonary respiratory chemoreflex also evoke the pulmonary depressor chemoreflex and coronary chemoreflex (10, 25, 33, 35). Thus it seemed reasonable to expect that intravenous and intrapericardial U-46619 would evoke a reflex hypotension. However, evidence showing that U-46619 induces the pulmonary depressor chemoreflex and the coronary chemoreflex was not obtained. The absence of a depressor response after intravenous U-46619 in the present study and in previous studies (4, 22) might be due to U-46619's direct vasoconstrictor effect (31, 32) overriding the pulmonary depressor chemoreflex. Although a distinction has been made between the coronary chemoreflex [evoked by chemical injection into the coronary arteries (10)] and the epicardial chemoreflex [evoked by chemical injection into the pericardial sac (33)], the same chemicals often evoke both of these reflexes (10, 25, 33, 35). However, in the rabbit for example, intravascular but not intrapericardial phenyl biguanide injection elicits a depressor reflex from the heart (1). U-46619's lack of effect on cardiac reflexes in the cat could be confirmed by injection into the coronary vessels.

Although intravenous U-46619 inhibited the knee-jerk reflex via a vagal reflex, the latency to the onset of inhibition after injection was surprisingly long. Somatomotor inhibition attributable to sensory receptors in lung typically occurs within 2-7 s and almost always within 14 s after right atrial injections of serotonin, phenyl biguanide, and nicotine (11, 17). In addition, vagal C fibers often discharge within 2.5 s of right atrial phenyldiguanide injection (27). These short onset latencies have been used to argue for the heart and the lungs as the site of the reflex's origination because the circulatory times are sufficiently short that the injectate would not have had time to reach other tissues (14). In the present experiments, the average response latency after intravenous U-46619 injection was 18 s. Interestingly, this latency is similar to the latency of other vagal responses after intravenous U-46619 injection. For example, vagal C fiber discharge increases 20 s after U-46619 injection (see Figs. 2 and 3 in Ref. 22). In addition, vagally mediated respiratory reflexes occur at least 25 s after U-46619 injection in the cat (see Fig. 3 in Ref. 32). Furthermore, the response of group III and IV muscle afferents to abdominal aortic injection of U-46619 (23) is substantially longer (27 s) than the response to capsaicin (~6 s) or bradykinin (~21 s). These long latencies raise the possibility that U-46619 binding to its receptors (7) may not be the proximate event that depolarizes the receptive endings of vagal C fibers.

Perspectives

The present study provides additional evidence for the role of the vagus nerve during illness. The stable thromboxane A₂ analog U-46619 reflexly inhibited the knee-jerk reflex via a vagal reflex from the lung. Systemic infection, produced experimentally by LPS injection, also increases thromboxane A₂ (3, 26), and the increased thromboxane A₂ activates respiratory reflexes whose afferent arm consists of vagal afferents from the lung (3). Because U-46619 reflexly inhibited central circuits associated with muscle activity in the present study, endogenous thromboxane A₂ released during infection or inflammation may contribute reflexly to the reduced motor activity accompanying sickness. It is worthwhile noting within this context that vagal reflexes from the thoracic viscera may contribute to exercise intolerance when demands made on the cardiopulmonary system exceed its capacity to meet them, such as in patients with congestive heart failure (16, 29, 30). Future research is needed to clarify how information contained within sensory feedback from visceral organs contributes to the organism's motor state.