- Email: [email protected]
Frequency responses to hypoxia and hypercapnia in carotid body-denervated conscious rats
Respiratory Physiology & Neurobiology 130 (2002) 113– 120 www.elsevier.com/locate/resphysiol
Frequency responses to hypoxia and hypercapnia in carotid body-denervated conscious rats Sharon K. Coles w, Rob Miller, Julie Huela, Patty Wolken, Evelyn Schlenker * Neuroscience Group, Di6ision of Basic Biomedical Sciences, Uni6ersity of South Dakota Medical School, 414 East Clark Street, Vermillion, SD 57069 -2390, USA Accepted 8 January 2002
Abstract The ventilatory response to brief, severe hypoxia is biphasic consisting of an initial facilitation followed by a slowing of breathing frequency ( f R). After the hypoxic stimulus is removed, f R drops below baseline levels. This phenomenon is called the post-hypoxic frequency decline (phfd). These f R changes are due to reciprocal changes in expiratory time (TE), mediated by the ventrolateral pontine A5 region (J. Physiol. (London) 497 (1996) 79; Am. J. Physiol. 274 (1998) R1546). The purpose of this study was to determine if carotid body input is required for full manifestation of phfd by quantifying ventilation in intact and carotid sinus denervated rats in response to hypoxic, and contrasted with hypercapnic stimuli. Following carotid denervation the initial facilitation of f R was eliminated in response to hypoxia, but the phfd remained. In contrast the pattern in response to increased CO2 remained constant before and after carotid denervation. These results suggest that phfd is not dependent upon carotid body stimulation, but is mediated centrally. © 2002 Elsevier Science B.V. All rights reserved. Keywords: Carotid body, post-hypoxic frequency decline; Control of breathing, hypoxic response; Hypoxia, post-hypoxic frequency decline; Mammals, rat; Pattern of breathing, hypoxic response
1. Introduction The ventilatory response to a brief exposure of severe hypoxia and its subsequent removal is triphasic in anesthetized rats (Dick and Coles, 2000). The biphasic response occurs during hypoxia and consists of an initial facilitation in breathing frequency ( f R) and amplitude followed * Corresponding author. Tel.: +1-605-677-5160; fax: + 1605-677-6381. E-mail address: [email protected] (E. Schlenker). w Deceased.
by an overall ‘depression’ in ventilation due primarily to a decrease in frequency. After hypoxia, f R falls below baseline. This post-stimulus effect has been termed post-hypoxic frequency decline (phfd) and constitutes the third phase of the ventilatory response to brief, severe hypoxia (Coles and Dick, 1996). phfd requires activation of pontine neurons. In anesthetized rats ventrolateral pontine neurons, specifically located in the noradrenergic A5 region, modulate frequency changes that persist after cessation of the hypoxic stimulus (Coles and Dick, 1996; Coles et al., 1998b). The phenomenon is
0034-5687/02/$ - see front matter © 2002 Elsevier Science B.V. All rights reserved. PII: S 0 0 3 4 - 5 6 8 7 ( 0 2 ) 0 0 0 0 5 - 1
114
S.K. Coles et al. / Respiratory Physiology & Neurobiology 130 (2002) 113–120
eliminated after disruption of ventrolateral pontine activity by lesioning or microinjection of muscimol. It is unclear whether phfd requires carotid chemoreceptor afferent input, whether ventrolateral pontine neurons themselves are chemosensitive, or they receive afferent input from chemosensitive brainstem neurons (Byrum and Guyenet, 1987; Erickson and Millhorn, 1994; Hayashi and Fukuda, 2000). Ventilatory ‘depression’ due to a decrease in frequency during hypoxia in carotid-denervated rats is eliminated after intercollicular transection (Martin-Body, 1988). In anesthetized rats under hyperoxic conditions, stimulation of the carotid sinus nerve (CSN) elicits frequency decreases during and after cessation of the stimulus (Hayashi et al., 1993). Collectively, these results suggest that pontine neurons may be chemosensitive and/or may receive input from other central chemoreceptors in addition to peripheral chemoreceptor input, or both. Furthermore, these findings imply that the same neuronal mechanisms are involved in frequency decreases that occur after hypoxia is removed. Against this background, we hypothesized that central nervous system neurons (possibly the ventrolateral pontine neurons) contain oxygen-sensing receptors that can elicit frequency declines independent of carotid body inputs. We addressed this possibility by quantifying ventilation before, during, and after brief exposure to severe hypoxia in awake, adult rats both before and after carotid body denervation. To determine if the responses were specific for hypoxia, another group of rats was also exposed to acute hyperoxic hypercapnia, which does not necessitate carotid body input (Dreshaj et al., 1999).
2. Methods
2.1. Animals Adult Sprague –Dawley, male rats, weighing 304–364 g were obtained from Harlan Sprague– Dawley (Chicago, IL). Three to four animals were housed in plastic cages with wire tops. Wa-
ter and food were made available ad libitum throughout the study. Lighting was maintained at 12 h on and off, respectively. The University of South Dakota Institutional Animal Use and Care Center approved all procedures in the study.
2.2. Plethysmography Whole-body plethysmography was used to measure inspiratory time (TI), expiratory time (TE), and f R. Conscious rats were placed in a cylindrical Plexiglas chamber (22.9 cm long and 15.2 cm in diameter). Air or test gases entered the chamber through one port. Pressure changes within the chamber related to ventilation in the rats were measured with a Validyne DP45-14 very low-pressure range differential pressure transducer. This was coupled to a Validyne MC1-3 3-channel module. The signal was then sent through a National Instruments BNC 2090 data acquisition board. Ventilation was recorded using a specially designed program (CWRU Core 7.vi) in National Instruments’ Labview 5.1. Air leaving the chamber was directed either to a Cole Parmer rotameter to monitor flow rate through the chamber, or to a Vacu-Med Fast Response O2 Analyzer.
2.3. Protocol to e6aluate 6entilatory responses to hypoxia or hypercapnia Each of 14 rats was placed in the chamber and acclimated for 20–25 min. Subsequently, ventilation was recorded in room air for a period of 20–45 sec. This measurement was used as a baseline control for each animal. The rat was then exposed to 8% O2 in 92% N2 for 45 sec and its ventilation recorded. Subsequently, the chamber was infused with air and ventilation was recorded during a 5-min recovery period. The ventilatory response to hypercapnia was evaluated in five animals to demonstrate the specificity of this phenomenon for hypoxia. The procedure used was identical to the hypoxic exposure except the animals were given 5% CO2 in 95% O2 rather than the 8% O2 in 92% N2 for 45 sec.
S.K. Coles et al. / Respiratory Physiology & Neurobiology 130 (2002) 113–120
2.4. Surgery After initial ventilatory testing, the CSN was transected. Sterile surgical methods were used. Rats were anaesthetized using an intraperitoneal injection 0.15 ml/100 g mixture of Ketamine (6.5 ml from 20 mg/ml) and Xylazine (8.7 ml from 100 mg/ml). Atropine (4 mg/ml) and dexamethazone (1 mg/ml) were also given to reduce tracheal secretions (0.2 ml each). The CSN was then sectioned bilaterally. Rats were allowed a 24 h recovery period before ventilation was again tested.
2.5. Data analysis Each animal served as its own control for the hypoxic and hypercapnic studies. The data was recorded and analyzed using the Case Western Reserve University’s customized computer program. Breathing frequency along with the duration of both inspiration (TI) and expiration (TE), and were measured from at a minimum of eight consecutive breaths. Breaths were taken from five different points in time: baseline (room air), peak hypoxia or hypercapnia, end hypoxia or hypercapnia, maximum phfd or post-hypercapnic decline, and at the end of the recovery (5 min) both before and after bilateral carotid body denervation. Mean values were calculated for each animal for each time point. Significant differences between these mean values were evaluated using a one-way analysis of variance (ANOVA) for repeated measures followed by a Dunnett’s test. Significance was accepted at P B 0.05.
3. Results
3.1. Responses to hypoxia before and after CSN transection The pre-surgical measurements served as a control for the CSN transection study. In air f R was 1129 6 bpm (Fig. 1(A)). The peak hypoxic response f R increased to 1659 9 bpm, whereas the f R at the end of the hypoxic episode had dropped to 1439 9 bpm. Frequency continued to decrease during the post-hypoxic period to a level of 7696
115
bpm (a 32% drop from baseline) and slowly returned towards baseline levels. Five minutes after the end of hypoxia f R was 99 9 7 bpm. The significant difference between the baseline f R prior to hypoxia and the f R at the lowest point during post-hypoxia demonstrates the phfd phenomenon (PB0.05). A similar significant difference between baseline f R and that recorded during the peak hypoxic response (PB 0.05). A reduction in TE followed by a further elongation at the end of hypoxia and continuing into the post-hypoxic period caused the changes noted in f R. Initially TE was 0.3590.02 sec, but decreased to 0.219 0.01 sec at the peak of the hypoxic response (Fig. 1(B) PB0.05). At the end of hypoxia TE had increased to 0.269 0.02 sec. During the post-hypoxic recovery TE increased 77% to 0.6290.06, relative to baseline (PB 0.05). At the end of the 5-min recovery period TE approached baseline levels at 0.449 0.04 sec. In contrast, TI remained essentially unchanged with no significant differences found between any of the five time points (Fig. 1(C)). After CSN transection an overall decrease in f R occurred due to an elongation of both TE and TI. Initial baseline measurements after CSN transection were 7192 bpm, a 39% decrease from intact baseline values (Fig. 1(A)). Similarly, the peak hypoxic f R and end hypoxic f R were markedly attenuated to 779 5 and 629 5 bpm, respectively. A very low post-hypoxic f of 4693 bpm was noted that showed a trend to return toward initial baseline levels. Although both baseline and post hypoxic f R were lower after carotid nerve denervation, the percent decrease in f R after hypoxia relative to baseline of 35% was similar to that noted prior to carotid nerve denervation. TE was increased at all five time-points (Fig. 1(B)). Initial recordings of TE showed an increase of 38.3% to 0.539 0.027 sec. Peak hypoxia and end hypoxia TE also demonstrated this increase of 0.519 0.037 and 0.749 0.74 sec, respectively. Post-hypoxic TE markedly increased to 1.019 0.075 sec. End recovery TE, 0.6759 0.054 sec, continued to exhibit a slow trend to return to initial levels. In contrast, there was a slight, but insignificant increase of TI (P\0.05, Fig. 1(C)).
116
S.K. Coles et al. / Respiratory Physiology & Neurobiology 130 (2002) 113–120
Fig. 1. Frequency (A), TE (B), and TI (C) responses to hypoxia before and after carotid body denervation. Mean values for each parameter ( 9S.D.) were calculated for three consecutive breaths before (baseline), during (peak) at the end of (end), following hypoxia (PHFD), and at 5 min post-hypoxia (5 min later) before (Pre -- -) CSN transection and after (Post — ) CSN transection. (*) PB 0.05 relative to baseline.
S.K. Coles et al. / Respiratory Physiology & Neurobiology 130 (2002) 113–120
The post-surgical hypoxic exposure demonstrated the presence of two separate phenomena in the hypoxic response. Recordings taken during the peak hypoxic period no longer showed a change in f R and TE relative to the initial pre-hypoxia measurements (P \ 0.05). Moreover, TI continued to show insignificant changes throughout the study. Finally, significant differences between the initial readings and those taken during post-hypoxia were still present as they had been during the pre-surgical studies for f R and TE (PB0.05).
3.2. Responses to hypercapnia before and after CSN Transection A pre-surgical exposure to hypercapnia was also evaluated (n= 5). The results found were very similar to those from the hypoxic study for f R, TE, and TI. Initially f R was 1149 9 bpm, showing a significant increase during the hypercapnic exposure to a peak of 1579 5 bpm (Fig. 2(A), PB0.05). At the end of the 45 sec of hypercapnia f R was 1369 9 bpm that decreased to 7496 bpm and returned to a f R of 99 9 9 bpm. Hypercapnia showed a phenomenon similar to the phfd experienced from the hypoxic episode with a significant difference between initial and the minimum post-hypercapnic f (P B 0.05). During the hypercapnic testing there was a similar effect on TE (Fig. 2(B)) as there had been during hypoxia. The initial TE was 0.349 0.02 sec. The peak response during hypercapnia resulted in a TE of 0.22 9 0.01 sec, a significant decrease from baseline (P B 0.05). A lengthening of TE occurred at the end of hypercapnia to 0.29 90.03 sec. The post-hypercapnic TE of 0.619 0.07 sec (PB 0.05, relative to baseline) demonstrated a similar phenomenon to the phfd noted during post-hypoxic recordings. Finally TE (0.42 9 0.04 sec) returned towards baseline levels at the end of the 5-min recovery. No effect of hypercapnia was noted on TI (Fig. 2(C)). After CSN transection the hypercapnic exposure showed a significant change between initial and peak hypercapnic f R (increase) and TE (decrease; PB0.05, Fig. 2 (A and B)). In addition, the significant difference between the initial and
117
post-hypercapnic f R (decrease) and TE (increase) remained (PB0.05) despite the decrease in parameters relative to pre-CSN transection values. Again no effect of hypercapnia or its removal was noted on TI.
4. Discussion The first notable result of this study was the depression of ventilation observed after bilateral transection of the CSN. These findings are consistent with similar results following carotid body denervation demonstrated previously in the rat (Favier and Laccaisse, 1978; Cardenas and Zapata, 1983; Martin-Body et al., 1985; Olson et al., 1988), and also in such other species as the cat (Millhorn et al., 1984), rabbit (Wright, 1935; Korner et al., 1969), dog (Bouckaert et al., 1938; Nielsen et al., 1986), and ground squirrel (Webb and Milsom, 1994). Removal of this tonic input via primary peripheral chemoreceptors, the carotid bodies, significantly depressed ventilation in normoxic, hypoxic, and hypercapnic environments and lead to increases in arterial PCO2 levels due to relative hypoventilation (Gautier et al., 1997). In addition the peak hypoxic response to brief, severe hypoxia following chemoafferent denervation was eliminated. The initial facilitation in f R normally seen as part of the biphasic portion of the ventilatory response to hypoxia in both anesthetized (Hayashi et al., 1993; Dick and Coles, 2000) and awake rats (Martin-Body et al., 1985) was removed following CSN transection. The slowing of f R seen near the end of the hypoxic exposure known as the biphasic portion of the hypoxic ventilatory response was observed both before and post-denervation. These findings suggest that separate neural substrates may modulate the different characteristics of the hypoxic ventilatory response. The post-hypoxic response to brief, severe hypoxia shows a significant decrease in f R, due to a significant increase in TE (Coles and Dick, 1996; Coles et al., 1998b; Hayashi et al. 1993). A potential factor influencing this frequency decline is a decrease in arterial PCO2 (Ohatake et al., 1998),
118
S.K. Coles et al. / Respiratory Physiology & Neurobiology 130 (2002) 113–120
Fig. 2. Frequency (A), TE (B), and TI (C) responses to hypercapnia before and after carotid body denervation. Mean values for each parameter ( 9 S.D.) were calculated for three consecutive breaths before (baseline), during (peak) at the end of (end), following hypercapnia (post-hyper), and at 5 min post-hypercapnia (5min later) before (Pre -- -) CSN transection and after (Post —) CSN transection. (*) P B0.05 relative to baseline.
S.K. Coles et al. / Respiratory Physiology & Neurobiology 130 (2002) 113–120
since there was no attempt to control for this variable in that study. If, however, arterial PCO2 increased after CSN transection, but the post hypoxic frequency decline remained (as was also noted in the present study), this may suggest that changes in PCO2 alone could not fully explain our findings. Support for a central mechanism responsible for the phfd includes the elimination of the post hypoxic frequency decline with intercollicular transection (Martin-Body, 1988). Moreover, in anesthetized rats the disruption of the ventrolateral pontine, noradrenergic A5 region eliminates the phfd (Coles and Dick, 1996; Coles et al., 1998a). More recently, Dawid-Milner and colleagues reported that either electrical or glutametergic stimulation of the A5 region in anesthetized rats decreased f R by selectively increasing TE (DawidMilner et al., 2001). Moreover, this response was elicited if the rats’ ventilation were controlled to prevent changes in arterial PCO2 or the animals were allowed to breath spontaneously. Finally, preliminary results in our laboratory (Schlenker and Prestbo) indicate that lesioning the A5 region of the pons in rats causes a selective elimination of the post hypoxic frequency decline that had been present prior to lesioning. Moreover, the ventilatory response to hypoxia was similar before and after the A5 lesions in these conscious rats suggesting that they most likely hyperventilated in response to hypoxia and exhibited hypocapnia. It is also significant that TI remained unchanged during and after hypoxic and hypercapnic exposures before carotid body denervation. Post-transection recordings revealed an overall lengthening of TI, but throughout the hypoxic and hypercapnic exposures in denervated rats there were no significant changes from baseline levels. This stability of TI demonstrates not only an exclusive effect of chemoafferent denervation on the expiratory portion of ventilation, and hence f R, but is consistent with previous findings of when the ventrolateral pontine was perturbed (Jodkowski et al., 1997; Dawid-Milner et al., 2001). The exposure of animals to a hypercapnic stimulus relative to the hypoxia demonstrated specificity for the results presented in this study. A brief exposure to hypercapnia and its removal also
119
exhibits a reaction similar to that seen in response to the brief, severe hypoxia (Schlenker et al., 2000). An initial facilitation of f R, due again to a significant decrease in TE, was followed by an attenuation of f R near the end of the hypercapnia, due to a significant elongation of TE, was observed. Posthypercapnic recordings showed a phenomenon similar to the phfd. A potential common mechanism for both the decline in frequency after hypoxic and hypercapnic exposure may involve stimulation of a2− adenergic receptors (Bach and Mitchell, 1998; Bach et al., 1999). These investigators noted that the decline of frequency after repeated hypoxic or hypercapnic exposures could be markedly attenuated in anesthetized, paralyzed and vagotomized rats by prior administration of a2− receptor antagonists. In contrast, Coles and coworkers were unable to show that the post-hypoxic frequency decline after one brief hypoxic exposure was modulated by a2 adenergic receptors (Coles et al., 1998b). Whether the differences in the effects of the a2 receptor antagonists in the two studies are due to differences in the protocols or strain differences needs to be considered. Of importance in this study was that the frequency decline after hypercapnia was still present after CSN transection. The hypercapnic peak f R response was not dependent on CSN input, because the majority of hypercapnic chemoreception is centrally (Nattie, 1995). Furthermore, the slowing of f R seen at the end of the acute hypercapnic exposure and the depression following exposure remained after surgery. Therefore, this study demonstrated specific effects of carotid denervation on hypoxic stimulation. Another question that this study raises is whether the post hypoxic frequency decline originally defined by Hayashi and coworkers that was dependent on carotid body stimulation (either electrically or due to hypoxia) is the same as we noted after denervation of the carotid body (Hayashi et al., 1993). Although the ventilatory pattern of response is similar to that noted in their study, the present study was not designed to investigate specifically this question, but opens up the potential for future investigations. In conclusion, carotid body input is not necessary to elicit a decline of frequency after hypoxic
120
S.K. Coles et al. / Respiratory Physiology & Neurobiology 130 (2002) 113–120
exposure. These results suggest that central nervous system structures are necessary to elicit phfd. Future studies in which the A5 region of the pons are lesioned or inhibited chemically may allow us to determine if the phfd or elongation of TE observed following hypoxia is different from that noted following hypercapnia.
Acknowledgements These studies were supported in part by grants P20 RR15567 of the Center of Biomedical Research Excellence and EPSCoR NSF HHS-2-T35HL-007760-08.
References Bach, K.B., Mitchell, G.S., 1998. Hypercapnia-induced longterm depression of respiratory activity requires a2− adreergic receptors. J. Appl. Physiol. 84, 2099 –2105. Bach, K.B., Kinkead, R., Mitchell, G.S., 1999. Post-hypoxia frequency decline in rats: sensitivity to repeated hypoxia and a2− adrenoreceptor antagonism. Brain. Res. 817, 25 –33. Bouckaert, J., Hemans, C., Samaan, C., 1938. The role of the carotid sinus and vagal chemoreceptors in the respiratory and vasomotor effects of hypoxaemia in anaesthetized and normal dogs. J. Physiol. (London) 94, 4 –5. Byrum, C.E., Guyenet, P.G., 1987. Afferent and efferent connections of the A5 noradrenergic cell group in the rat. J. Comp. Neurol. 261, 529 – 542. Cardenas, H., Zapata, P., 1983. Ventilatory reflexes originated from carotid and extracarotid chemoreceptors in rats. Am. J. Physiol. 244, R119 –125. Coles, S.K., Dick, T.E., 1996. Neurones in the ventrolateral pons are required for post-hypoxic frequency decline in rats. J. Physiol. (London) 497, 79 –94. Coles, S.K., Ernsberger, P., Dick, T.E., 1998a. A role for NMDA receptors in post-hypoxic frequency decline in the rat. Am. J. Physiol. 274, R1546 –1555. Coles, S.K., Ernsberger, P., Dick, T.E., 1998b. Post-hypoxic frequency decline does not depend on a2− adrenergic receptors in the adult rat. Brain. Res. 794, 267 –273. Dawid-Milner, M.S., Lara, J.P., Gonzale`z-Baro´ n, S., Spyer, K.M., 2001. Respiratory effects of stimulation of cell bodies of the A5 region on the anaesthetized rat. Pflu¨ gers. Arch. Eur. J. Physiol. 441, 434 –443. Dick, T.E., Coles, S.K., 2000. Ventrolateral pons mediates short-term depression of respiratory frequency after brief hypoxia. Respir. Physiol. 121, 87 –100. Dreshaj, I.A., Haxhiu, M.A., Abu-Shaweesh, J., Carey, R.E., Martin, R.J., 1999. CO2-induced prolongation of expiratory
time during early development. Respir. Physiol. 116, 125 – 132. Erickson, J.T., Millhorn, D.E., 1994. Hypoxia and electrical stimulation of the carotid sinus nerve induce Fos-like immunoreactivity within the catecholaminergic and serotoninergic neurons of the rat brainstem. J. Comp. Neurol. 348, 161 – 182. Favier, R., Laccaisse, A., 1978. O2 chemoreflex drive of ventilation in the awake rat. J. Physiol. (London) 74, 411 – 417. Gautier, H., Murariu, C., Bonora, M., 1997. Ventilatory and metabolic responses to ambient hypoxia or hypercapnia in rats exposed to CO hypoxia. J. Appl. Physiol. 83, 253 – 261. Hayashi, F., Fukuda, Y., 2000. Neuronal mechanisms mediating the integration of respiratory responses to hypoxia. Jpn. J. Physiol. 50, 15 – 24. Hayashi, F., Coles, S.K., Bach, K.B., Mitchell, G.S., McCrimmon, D.R., 1993. Time- dependent phrenic nerve responses to carotid afferent activation: intact vs. decerebrate rats. Am. J. Physiol. 265, R811 – R819. Jodkowski, J.S., Coles, S.K., Dick, T.E., 1997. Prolongation in expiration evoked from ventrolateral pons of adult rats. J. Appl. Physiol. 82, 377 – 381. Korner, P.I., Uther, J.B., White, S.W., 1969. Central nervous integration of the circulatory and respiratory responses to arterial hypoxaemia in the rabbit. Circ. Res. 24, 757 –776. Martin-Body, R.L., 1988. Brain transections demonstrate the central origin of hypoxic ventilatory depression in carotid body-denervated rats. J. Physiol. (London) 407, 41 – 52. Martin-Body, R.L., Robson, G.J., Sinclair, J.D., 1985. Respiratory effects of sectioning the carotid sinus glossopharyngeal and abdominal vagal nerves in the awake rat. J. Physiol. (London) 361, 35 – 45. Millhorn, D.E., Eldridge, F.L., Kiley, J.P., Waldrop, T.G., 1984. Prolonged inhibition of respiration following acute hypoxia in glomectomized cats. Respir. Physiol. 57, 331 – 340. Nattie, E., 1995. Central Chemoreception. In: Dempsey, A.J., Pack, A.I. (Eds.), Regulation of Breathing, second ed. Marcel Dekker, New York, pp. 473 – 510. Nielsen, A.M., Bisgard, G.E., Mitchell, G.S., 1986. Phrenic nerve responses to hypoxia and CO2 in decerebrate dogs. Respir. Physiol. 65, 267 – 283. Ohatake, P.J., Torres, J.E., Gozal, Y.M., Graff, G.R., Gozal, D., 1998. NMDA receptors mediate peripheral chemoreceptor afferent input in the conscious rat. J. Appl. Physiol. 84, 853 – 861. Olson, E.B., Vidruck, E.H., Dempsey, J.A., 1988. Carotid body excision significantly changes ventilatory control in awake rats. J. Appl. Physiol. 64, 666 – 671. Schlenker, E.A., Burbach, J.A., Singh, Y.N., 2000. Effects of thyroxine and naloxone administration on metabolism and ventilation in hamsters. Respir. Physiol. 123, 63 – 70. Webb, C.L., Milsom, W.K., 1994. Ventilatory responses to acute and chronic hypoxic hypercapnia in the ground squirrel. Respir. Physiol. 98, 127 – 152. Wright, S., 1935. Mode of action of oxygen lack and carbon dioxide excess on the respiration in the rabbit. Q. J. Exp. Physiol. 24, 169 – 175.
Frequency responses to hypoxia and hypercapnia in carotid body-denervated conscious rats Sharon K. Coles w, Rob Miller, Julie Huela, Patty Wolken, Evelyn Schlenker * Neuroscience Group, Di6ision of Basic Biomedical Sciences, Uni6ersity of South Dakota Medical School, 414 East Clark Street, Vermillion, SD 57069 -2390, USA Accepted 8 January 2002
Abstract The ventilatory response to brief, severe hypoxia is biphasic consisting of an initial facilitation followed by a slowing of breathing frequency ( f R). After the hypoxic stimulus is removed, f R drops below baseline levels. This phenomenon is called the post-hypoxic frequency decline (phfd). These f R changes are due to reciprocal changes in expiratory time (TE), mediated by the ventrolateral pontine A5 region (J. Physiol. (London) 497 (1996) 79; Am. J. Physiol. 274 (1998) R1546). The purpose of this study was to determine if carotid body input is required for full manifestation of phfd by quantifying ventilation in intact and carotid sinus denervated rats in response to hypoxic, and contrasted with hypercapnic stimuli. Following carotid denervation the initial facilitation of f R was eliminated in response to hypoxia, but the phfd remained. In contrast the pattern in response to increased CO2 remained constant before and after carotid denervation. These results suggest that phfd is not dependent upon carotid body stimulation, but is mediated centrally. © 2002 Elsevier Science B.V. All rights reserved. Keywords: Carotid body, post-hypoxic frequency decline; Control of breathing, hypoxic response; Hypoxia, post-hypoxic frequency decline; Mammals, rat; Pattern of breathing, hypoxic response
1. Introduction The ventilatory response to a brief exposure of severe hypoxia and its subsequent removal is triphasic in anesthetized rats (Dick and Coles, 2000). The biphasic response occurs during hypoxia and consists of an initial facilitation in breathing frequency ( f R) and amplitude followed * Corresponding author. Tel.: +1-605-677-5160; fax: + 1605-677-6381. E-mail address: [email protected] (E. Schlenker). w Deceased.
by an overall ‘depression’ in ventilation due primarily to a decrease in frequency. After hypoxia, f R falls below baseline. This post-stimulus effect has been termed post-hypoxic frequency decline (phfd) and constitutes the third phase of the ventilatory response to brief, severe hypoxia (Coles and Dick, 1996). phfd requires activation of pontine neurons. In anesthetized rats ventrolateral pontine neurons, specifically located in the noradrenergic A5 region, modulate frequency changes that persist after cessation of the hypoxic stimulus (Coles and Dick, 1996; Coles et al., 1998b). The phenomenon is
0034-5687/02/$ - see front matter © 2002 Elsevier Science B.V. All rights reserved. PII: S 0 0 3 4 - 5 6 8 7 ( 0 2 ) 0 0 0 0 5 - 1
114
S.K. Coles et al. / Respiratory Physiology & Neurobiology 130 (2002) 113–120
eliminated after disruption of ventrolateral pontine activity by lesioning or microinjection of muscimol. It is unclear whether phfd requires carotid chemoreceptor afferent input, whether ventrolateral pontine neurons themselves are chemosensitive, or they receive afferent input from chemosensitive brainstem neurons (Byrum and Guyenet, 1987; Erickson and Millhorn, 1994; Hayashi and Fukuda, 2000). Ventilatory ‘depression’ due to a decrease in frequency during hypoxia in carotid-denervated rats is eliminated after intercollicular transection (Martin-Body, 1988). In anesthetized rats under hyperoxic conditions, stimulation of the carotid sinus nerve (CSN) elicits frequency decreases during and after cessation of the stimulus (Hayashi et al., 1993). Collectively, these results suggest that pontine neurons may be chemosensitive and/or may receive input from other central chemoreceptors in addition to peripheral chemoreceptor input, or both. Furthermore, these findings imply that the same neuronal mechanisms are involved in frequency decreases that occur after hypoxia is removed. Against this background, we hypothesized that central nervous system neurons (possibly the ventrolateral pontine neurons) contain oxygen-sensing receptors that can elicit frequency declines independent of carotid body inputs. We addressed this possibility by quantifying ventilation before, during, and after brief exposure to severe hypoxia in awake, adult rats both before and after carotid body denervation. To determine if the responses were specific for hypoxia, another group of rats was also exposed to acute hyperoxic hypercapnia, which does not necessitate carotid body input (Dreshaj et al., 1999).
2. Methods
2.1. Animals Adult Sprague –Dawley, male rats, weighing 304–364 g were obtained from Harlan Sprague– Dawley (Chicago, IL). Three to four animals were housed in plastic cages with wire tops. Wa-
ter and food were made available ad libitum throughout the study. Lighting was maintained at 12 h on and off, respectively. The University of South Dakota Institutional Animal Use and Care Center approved all procedures in the study.
2.2. Plethysmography Whole-body plethysmography was used to measure inspiratory time (TI), expiratory time (TE), and f R. Conscious rats were placed in a cylindrical Plexiglas chamber (22.9 cm long and 15.2 cm in diameter). Air or test gases entered the chamber through one port. Pressure changes within the chamber related to ventilation in the rats were measured with a Validyne DP45-14 very low-pressure range differential pressure transducer. This was coupled to a Validyne MC1-3 3-channel module. The signal was then sent through a National Instruments BNC 2090 data acquisition board. Ventilation was recorded using a specially designed program (CWRU Core 7.vi) in National Instruments’ Labview 5.1. Air leaving the chamber was directed either to a Cole Parmer rotameter to monitor flow rate through the chamber, or to a Vacu-Med Fast Response O2 Analyzer.
2.3. Protocol to e6aluate 6entilatory responses to hypoxia or hypercapnia Each of 14 rats was placed in the chamber and acclimated for 20–25 min. Subsequently, ventilation was recorded in room air for a period of 20–45 sec. This measurement was used as a baseline control for each animal. The rat was then exposed to 8% O2 in 92% N2 for 45 sec and its ventilation recorded. Subsequently, the chamber was infused with air and ventilation was recorded during a 5-min recovery period. The ventilatory response to hypercapnia was evaluated in five animals to demonstrate the specificity of this phenomenon for hypoxia. The procedure used was identical to the hypoxic exposure except the animals were given 5% CO2 in 95% O2 rather than the 8% O2 in 92% N2 for 45 sec.
S.K. Coles et al. / Respiratory Physiology & Neurobiology 130 (2002) 113–120
2.4. Surgery After initial ventilatory testing, the CSN was transected. Sterile surgical methods were used. Rats were anaesthetized using an intraperitoneal injection 0.15 ml/100 g mixture of Ketamine (6.5 ml from 20 mg/ml) and Xylazine (8.7 ml from 100 mg/ml). Atropine (4 mg/ml) and dexamethazone (1 mg/ml) were also given to reduce tracheal secretions (0.2 ml each). The CSN was then sectioned bilaterally. Rats were allowed a 24 h recovery period before ventilation was again tested.
2.5. Data analysis Each animal served as its own control for the hypoxic and hypercapnic studies. The data was recorded and analyzed using the Case Western Reserve University’s customized computer program. Breathing frequency along with the duration of both inspiration (TI) and expiration (TE), and were measured from at a minimum of eight consecutive breaths. Breaths were taken from five different points in time: baseline (room air), peak hypoxia or hypercapnia, end hypoxia or hypercapnia, maximum phfd or post-hypercapnic decline, and at the end of the recovery (5 min) both before and after bilateral carotid body denervation. Mean values were calculated for each animal for each time point. Significant differences between these mean values were evaluated using a one-way analysis of variance (ANOVA) for repeated measures followed by a Dunnett’s test. Significance was accepted at P B 0.05.
3. Results
3.1. Responses to hypoxia before and after CSN transection The pre-surgical measurements served as a control for the CSN transection study. In air f R was 1129 6 bpm (Fig. 1(A)). The peak hypoxic response f R increased to 1659 9 bpm, whereas the f R at the end of the hypoxic episode had dropped to 1439 9 bpm. Frequency continued to decrease during the post-hypoxic period to a level of 7696
115
bpm (a 32% drop from baseline) and slowly returned towards baseline levels. Five minutes after the end of hypoxia f R was 99 9 7 bpm. The significant difference between the baseline f R prior to hypoxia and the f R at the lowest point during post-hypoxia demonstrates the phfd phenomenon (PB0.05). A similar significant difference between baseline f R and that recorded during the peak hypoxic response (PB 0.05). A reduction in TE followed by a further elongation at the end of hypoxia and continuing into the post-hypoxic period caused the changes noted in f R. Initially TE was 0.3590.02 sec, but decreased to 0.219 0.01 sec at the peak of the hypoxic response (Fig. 1(B) PB0.05). At the end of hypoxia TE had increased to 0.269 0.02 sec. During the post-hypoxic recovery TE increased 77% to 0.6290.06, relative to baseline (PB 0.05). At the end of the 5-min recovery period TE approached baseline levels at 0.449 0.04 sec. In contrast, TI remained essentially unchanged with no significant differences found between any of the five time points (Fig. 1(C)). After CSN transection an overall decrease in f R occurred due to an elongation of both TE and TI. Initial baseline measurements after CSN transection were 7192 bpm, a 39% decrease from intact baseline values (Fig. 1(A)). Similarly, the peak hypoxic f R and end hypoxic f R were markedly attenuated to 779 5 and 629 5 bpm, respectively. A very low post-hypoxic f of 4693 bpm was noted that showed a trend to return toward initial baseline levels. Although both baseline and post hypoxic f R were lower after carotid nerve denervation, the percent decrease in f R after hypoxia relative to baseline of 35% was similar to that noted prior to carotid nerve denervation. TE was increased at all five time-points (Fig. 1(B)). Initial recordings of TE showed an increase of 38.3% to 0.539 0.027 sec. Peak hypoxia and end hypoxia TE also demonstrated this increase of 0.519 0.037 and 0.749 0.74 sec, respectively. Post-hypoxic TE markedly increased to 1.019 0.075 sec. End recovery TE, 0.6759 0.054 sec, continued to exhibit a slow trend to return to initial levels. In contrast, there was a slight, but insignificant increase of TI (P\0.05, Fig. 1(C)).
116
S.K. Coles et al. / Respiratory Physiology & Neurobiology 130 (2002) 113–120
Fig. 1. Frequency (A), TE (B), and TI (C) responses to hypoxia before and after carotid body denervation. Mean values for each parameter ( 9S.D.) were calculated for three consecutive breaths before (baseline), during (peak) at the end of (end), following hypoxia (PHFD), and at 5 min post-hypoxia (5 min later) before (Pre -- -) CSN transection and after (Post — ) CSN transection. (*) PB 0.05 relative to baseline.
S.K. Coles et al. / Respiratory Physiology & Neurobiology 130 (2002) 113–120
The post-surgical hypoxic exposure demonstrated the presence of two separate phenomena in the hypoxic response. Recordings taken during the peak hypoxic period no longer showed a change in f R and TE relative to the initial pre-hypoxia measurements (P \ 0.05). Moreover, TI continued to show insignificant changes throughout the study. Finally, significant differences between the initial readings and those taken during post-hypoxia were still present as they had been during the pre-surgical studies for f R and TE (PB0.05).
3.2. Responses to hypercapnia before and after CSN Transection A pre-surgical exposure to hypercapnia was also evaluated (n= 5). The results found were very similar to those from the hypoxic study for f R, TE, and TI. Initially f R was 1149 9 bpm, showing a significant increase during the hypercapnic exposure to a peak of 1579 5 bpm (Fig. 2(A), PB0.05). At the end of the 45 sec of hypercapnia f R was 1369 9 bpm that decreased to 7496 bpm and returned to a f R of 99 9 9 bpm. Hypercapnia showed a phenomenon similar to the phfd experienced from the hypoxic episode with a significant difference between initial and the minimum post-hypercapnic f (P B 0.05). During the hypercapnic testing there was a similar effect on TE (Fig. 2(B)) as there had been during hypoxia. The initial TE was 0.349 0.02 sec. The peak response during hypercapnia resulted in a TE of 0.22 9 0.01 sec, a significant decrease from baseline (P B 0.05). A lengthening of TE occurred at the end of hypercapnia to 0.29 90.03 sec. The post-hypercapnic TE of 0.619 0.07 sec (PB 0.05, relative to baseline) demonstrated a similar phenomenon to the phfd noted during post-hypoxic recordings. Finally TE (0.42 9 0.04 sec) returned towards baseline levels at the end of the 5-min recovery. No effect of hypercapnia was noted on TI (Fig. 2(C)). After CSN transection the hypercapnic exposure showed a significant change between initial and peak hypercapnic f R (increase) and TE (decrease; PB0.05, Fig. 2 (A and B)). In addition, the significant difference between the initial and
117
post-hypercapnic f R (decrease) and TE (increase) remained (PB0.05) despite the decrease in parameters relative to pre-CSN transection values. Again no effect of hypercapnia or its removal was noted on TI.
4. Discussion The first notable result of this study was the depression of ventilation observed after bilateral transection of the CSN. These findings are consistent with similar results following carotid body denervation demonstrated previously in the rat (Favier and Laccaisse, 1978; Cardenas and Zapata, 1983; Martin-Body et al., 1985; Olson et al., 1988), and also in such other species as the cat (Millhorn et al., 1984), rabbit (Wright, 1935; Korner et al., 1969), dog (Bouckaert et al., 1938; Nielsen et al., 1986), and ground squirrel (Webb and Milsom, 1994). Removal of this tonic input via primary peripheral chemoreceptors, the carotid bodies, significantly depressed ventilation in normoxic, hypoxic, and hypercapnic environments and lead to increases in arterial PCO2 levels due to relative hypoventilation (Gautier et al., 1997). In addition the peak hypoxic response to brief, severe hypoxia following chemoafferent denervation was eliminated. The initial facilitation in f R normally seen as part of the biphasic portion of the ventilatory response to hypoxia in both anesthetized (Hayashi et al., 1993; Dick and Coles, 2000) and awake rats (Martin-Body et al., 1985) was removed following CSN transection. The slowing of f R seen near the end of the hypoxic exposure known as the biphasic portion of the hypoxic ventilatory response was observed both before and post-denervation. These findings suggest that separate neural substrates may modulate the different characteristics of the hypoxic ventilatory response. The post-hypoxic response to brief, severe hypoxia shows a significant decrease in f R, due to a significant increase in TE (Coles and Dick, 1996; Coles et al., 1998b; Hayashi et al. 1993). A potential factor influencing this frequency decline is a decrease in arterial PCO2 (Ohatake et al., 1998),
118
S.K. Coles et al. / Respiratory Physiology & Neurobiology 130 (2002) 113–120
Fig. 2. Frequency (A), TE (B), and TI (C) responses to hypercapnia before and after carotid body denervation. Mean values for each parameter ( 9 S.D.) were calculated for three consecutive breaths before (baseline), during (peak) at the end of (end), following hypercapnia (post-hyper), and at 5 min post-hypercapnia (5min later) before (Pre -- -) CSN transection and after (Post —) CSN transection. (*) P B0.05 relative to baseline.
S.K. Coles et al. / Respiratory Physiology & Neurobiology 130 (2002) 113–120
since there was no attempt to control for this variable in that study. If, however, arterial PCO2 increased after CSN transection, but the post hypoxic frequency decline remained (as was also noted in the present study), this may suggest that changes in PCO2 alone could not fully explain our findings. Support for a central mechanism responsible for the phfd includes the elimination of the post hypoxic frequency decline with intercollicular transection (Martin-Body, 1988). Moreover, in anesthetized rats the disruption of the ventrolateral pontine, noradrenergic A5 region eliminates the phfd (Coles and Dick, 1996; Coles et al., 1998a). More recently, Dawid-Milner and colleagues reported that either electrical or glutametergic stimulation of the A5 region in anesthetized rats decreased f R by selectively increasing TE (DawidMilner et al., 2001). Moreover, this response was elicited if the rats’ ventilation were controlled to prevent changes in arterial PCO2 or the animals were allowed to breath spontaneously. Finally, preliminary results in our laboratory (Schlenker and Prestbo) indicate that lesioning the A5 region of the pons in rats causes a selective elimination of the post hypoxic frequency decline that had been present prior to lesioning. Moreover, the ventilatory response to hypoxia was similar before and after the A5 lesions in these conscious rats suggesting that they most likely hyperventilated in response to hypoxia and exhibited hypocapnia. It is also significant that TI remained unchanged during and after hypoxic and hypercapnic exposures before carotid body denervation. Post-transection recordings revealed an overall lengthening of TI, but throughout the hypoxic and hypercapnic exposures in denervated rats there were no significant changes from baseline levels. This stability of TI demonstrates not only an exclusive effect of chemoafferent denervation on the expiratory portion of ventilation, and hence f R, but is consistent with previous findings of when the ventrolateral pontine was perturbed (Jodkowski et al., 1997; Dawid-Milner et al., 2001). The exposure of animals to a hypercapnic stimulus relative to the hypoxia demonstrated specificity for the results presented in this study. A brief exposure to hypercapnia and its removal also
119
exhibits a reaction similar to that seen in response to the brief, severe hypoxia (Schlenker et al., 2000). An initial facilitation of f R, due again to a significant decrease in TE, was followed by an attenuation of f R near the end of the hypercapnia, due to a significant elongation of TE, was observed. Posthypercapnic recordings showed a phenomenon similar to the phfd. A potential common mechanism for both the decline in frequency after hypoxic and hypercapnic exposure may involve stimulation of a2− adenergic receptors (Bach and Mitchell, 1998; Bach et al., 1999). These investigators noted that the decline of frequency after repeated hypoxic or hypercapnic exposures could be markedly attenuated in anesthetized, paralyzed and vagotomized rats by prior administration of a2− receptor antagonists. In contrast, Coles and coworkers were unable to show that the post-hypoxic frequency decline after one brief hypoxic exposure was modulated by a2 adenergic receptors (Coles et al., 1998b). Whether the differences in the effects of the a2 receptor antagonists in the two studies are due to differences in the protocols or strain differences needs to be considered. Of importance in this study was that the frequency decline after hypercapnia was still present after CSN transection. The hypercapnic peak f R response was not dependent on CSN input, because the majority of hypercapnic chemoreception is centrally (Nattie, 1995). Furthermore, the slowing of f R seen at the end of the acute hypercapnic exposure and the depression following exposure remained after surgery. Therefore, this study demonstrated specific effects of carotid denervation on hypoxic stimulation. Another question that this study raises is whether the post hypoxic frequency decline originally defined by Hayashi and coworkers that was dependent on carotid body stimulation (either electrically or due to hypoxia) is the same as we noted after denervation of the carotid body (Hayashi et al., 1993). Although the ventilatory pattern of response is similar to that noted in their study, the present study was not designed to investigate specifically this question, but opens up the potential for future investigations. In conclusion, carotid body input is not necessary to elicit a decline of frequency after hypoxic
120
S.K. Coles et al. / Respiratory Physiology & Neurobiology 130 (2002) 113–120
exposure. These results suggest that central nervous system structures are necessary to elicit phfd. Future studies in which the A5 region of the pons are lesioned or inhibited chemically may allow us to determine if the phfd or elongation of TE observed following hypoxia is different from that noted following hypercapnia.
Acknowledgements These studies were supported in part by grants P20 RR15567 of the Center of Biomedical Research Excellence and EPSCoR NSF HHS-2-T35HL-007760-08.
References Bach, K.B., Mitchell, G.S., 1998. Hypercapnia-induced longterm depression of respiratory activity requires a2− adreergic receptors. J. Appl. Physiol. 84, 2099 –2105. Bach, K.B., Kinkead, R., Mitchell, G.S., 1999. Post-hypoxia frequency decline in rats: sensitivity to repeated hypoxia and a2− adrenoreceptor antagonism. Brain. Res. 817, 25 –33. Bouckaert, J., Hemans, C., Samaan, C., 1938. The role of the carotid sinus and vagal chemoreceptors in the respiratory and vasomotor effects of hypoxaemia in anaesthetized and normal dogs. J. Physiol. (London) 94, 4 –5. Byrum, C.E., Guyenet, P.G., 1987. Afferent and efferent connections of the A5 noradrenergic cell group in the rat. J. Comp. Neurol. 261, 529 – 542. Cardenas, H., Zapata, P., 1983. Ventilatory reflexes originated from carotid and extracarotid chemoreceptors in rats. Am. J. Physiol. 244, R119 –125. Coles, S.K., Dick, T.E., 1996. Neurones in the ventrolateral pons are required for post-hypoxic frequency decline in rats. J. Physiol. (London) 497, 79 –94. Coles, S.K., Ernsberger, P., Dick, T.E., 1998a. A role for NMDA receptors in post-hypoxic frequency decline in the rat. Am. J. Physiol. 274, R1546 –1555. Coles, S.K., Ernsberger, P., Dick, T.E., 1998b. Post-hypoxic frequency decline does not depend on a2− adrenergic receptors in the adult rat. Brain. Res. 794, 267 –273. Dawid-Milner, M.S., Lara, J.P., Gonzale`z-Baro´ n, S., Spyer, K.M., 2001. Respiratory effects of stimulation of cell bodies of the A5 region on the anaesthetized rat. Pflu¨ gers. Arch. Eur. J. Physiol. 441, 434 –443. Dick, T.E., Coles, S.K., 2000. Ventrolateral pons mediates short-term depression of respiratory frequency after brief hypoxia. Respir. Physiol. 121, 87 –100. Dreshaj, I.A., Haxhiu, M.A., Abu-Shaweesh, J., Carey, R.E., Martin, R.J., 1999. CO2-induced prolongation of expiratory
time during early development. Respir. Physiol. 116, 125 – 132. Erickson, J.T., Millhorn, D.E., 1994. Hypoxia and electrical stimulation of the carotid sinus nerve induce Fos-like immunoreactivity within the catecholaminergic and serotoninergic neurons of the rat brainstem. J. Comp. Neurol. 348, 161 – 182. Favier, R., Laccaisse, A., 1978. O2 chemoreflex drive of ventilation in the awake rat. J. Physiol. (London) 74, 411 – 417. Gautier, H., Murariu, C., Bonora, M., 1997. Ventilatory and metabolic responses to ambient hypoxia or hypercapnia in rats exposed to CO hypoxia. J. Appl. Physiol. 83, 253 – 261. Hayashi, F., Fukuda, Y., 2000. Neuronal mechanisms mediating the integration of respiratory responses to hypoxia. Jpn. J. Physiol. 50, 15 – 24. Hayashi, F., Coles, S.K., Bach, K.B., Mitchell, G.S., McCrimmon, D.R., 1993. Time- dependent phrenic nerve responses to carotid afferent activation: intact vs. decerebrate rats. Am. J. Physiol. 265, R811 – R819. Jodkowski, J.S., Coles, S.K., Dick, T.E., 1997. Prolongation in expiration evoked from ventrolateral pons of adult rats. J. Appl. Physiol. 82, 377 – 381. Korner, P.I., Uther, J.B., White, S.W., 1969. Central nervous integration of the circulatory and respiratory responses to arterial hypoxaemia in the rabbit. Circ. Res. 24, 757 –776. Martin-Body, R.L., 1988. Brain transections demonstrate the central origin of hypoxic ventilatory depression in carotid body-denervated rats. J. Physiol. (London) 407, 41 – 52. Martin-Body, R.L., Robson, G.J., Sinclair, J.D., 1985. Respiratory effects of sectioning the carotid sinus glossopharyngeal and abdominal vagal nerves in the awake rat. J. Physiol. (London) 361, 35 – 45. Millhorn, D.E., Eldridge, F.L., Kiley, J.P., Waldrop, T.G., 1984. Prolonged inhibition of respiration following acute hypoxia in glomectomized cats. Respir. Physiol. 57, 331 – 340. Nattie, E., 1995. Central Chemoreception. In: Dempsey, A.J., Pack, A.I. (Eds.), Regulation of Breathing, second ed. Marcel Dekker, New York, pp. 473 – 510. Nielsen, A.M., Bisgard, G.E., Mitchell, G.S., 1986. Phrenic nerve responses to hypoxia and CO2 in decerebrate dogs. Respir. Physiol. 65, 267 – 283. Ohatake, P.J., Torres, J.E., Gozal, Y.M., Graff, G.R., Gozal, D., 1998. NMDA receptors mediate peripheral chemoreceptor afferent input in the conscious rat. J. Appl. Physiol. 84, 853 – 861. Olson, E.B., Vidruck, E.H., Dempsey, J.A., 1988. Carotid body excision significantly changes ventilatory control in awake rats. J. Appl. Physiol. 64, 666 – 671. Schlenker, E.A., Burbach, J.A., Singh, Y.N., 2000. Effects of thyroxine and naloxone administration on metabolism and ventilation in hamsters. Respir. Physiol. 123, 63 – 70. Webb, C.L., Milsom, W.K., 1994. Ventilatory responses to acute and chronic hypoxic hypercapnia in the ground squirrel. Respir. Physiol. 98, 127 – 152. Wright, S., 1935. Mode of action of oxygen lack and carbon dioxide excess on the respiration in the rabbit. Q. J. Exp. Physiol. 24, 169 – 175.