- Email: [email protected]
Role of the pons in hypoxic respiratory depression in the neonatal rat
Respiration Physiology 111 (1998) 55 – 63
Role of the pons in hypoxic respiratory depression in the neonatal rat Y. Okada *, A. Kawai 1, K. Mu¨ckenhoff, P. Scheid Institut fu¨r Physiologie, Ruhr-Uni6ersita¨t Bochum, D-44780 Bochum, Germany Accepted 8 September 1997
Abstract The main purpose of this study was to evaluate the role of the pons in hypoxic respiratory depression (HRD) of the neonatal rat. Experiments were conducted using the isolated brainstem-spinal cord preparation of the neonatal rat (1–3 days old). The brainstem was transected at various levels. We found that ablation of the diencephalon decreased respiratory frequency (fR), and conversely, that ablation of the midbrain or pons increased fR. In the preparation with the pons intact (without the midbrain), hypoxia (superfusate PO2 = 56 mmHg) caused strong depression of respiratory activity, which was characterized by a steady decrease in fR and in integrated inspiratory burst amplitude ( Phr). In the preparation with the intact ventral pons (without midbrain and dorsal pons) we observed similar, though weaker, HRD. When the entire pons was ablated, Phr was little depressed by hypoxia and thus, HRD was further attenuated. We conclude that the pons contributes importantly to the induction of hypoxic respiratory depression in the neonatal rat. Both the ventral and dorsal portions of the pons are involved in the control of hypoxic respiratory depression. In addition, we show that the respiratory modulatory functions of the diencephalon (facilitating) and midbrain (inhibitory) are already expressed at the time of birth. © 1998 Elsevier Science B.V. Keywords: Control of breathing, hypoxia, development; Hypoxia, central respiratory depression; Brainstem, diencephalon, midbrain, pons, medulla oblongata; Mammals, neonatal rat; Development, control of breathing
1. Introduction * Corresponding author. Present address. Department of Medicine, Keio University, Tagata-gun, Tsukigase Rehabilitation Cente, Tsukigase 380-2, Amagiyugashima-cho, Shizuokaken, 410-3215 Japan. Tel.: + 81 558 851701; fax: +81 558 851810; e-mail:[email protected] 1 Present address: Department of Medicine, Keio University, 35 Shinanomachi, Shinjuku-ku, Tokyo, 160-8582 Japan.
The respiratory response to hypoxia in mammals is typically biphasic; an initial augmentation is followed by a depression of ventilation, the so-called hypoxic respiratory depression (HRD), which is thought to constitute a response of the central nervous system (Martin-Body, 1988
0034-5687/98/$19.00 © 1998 Elsevier Science B.V. All rights reserved. PII S 0 0 3 4 - 5 6 8 7 ( 9 7 ) 0 0 1 0 5 - 9
56
Y. Okada et al. / Respiration Physiology 111 (1998) 55–63
Fig. 1. Levels of brainstem transection: (A) Schematic diagram of the ventral brainstem surface; (B) Lateral view of the brainstem. 1, V, VI, IX, X, XI and XII: fifth, sixth, ninth, tenth, eleventh and twelfth cranial nerve roots, respectively. The sections are: 2 , midbrain-pontine junction; 3 – 3 %, transection level for the pontine ablation; 4 , transection diencephalo-midbrain junction; line for the ablation of the dorsal pons.
Martin-Body and Johnston, 1988; for a review see Neubauer et al., 1990). In rats, the respiratory response to hypoxia changes substantially with development and growth (Ballanyi et al., 1992; Eden and Hanson, 1987; Fukuda, 1992), whereby neuronal responses to hypoxia in adults and neonates are strikingly different (Haddad and Jiang, 1993). The mechanisms underlying HRD may therefore differ between adults and neonates. In adult animals the HRD-inducing mechanism has been reported to be located in the midbrain in the cat (Gallman et al., 1991), rabbit (Martin-Body and Johnston, 1988) and rat (Martin-Body, 1988). On the other hand, the upper lateral pons has been reported to play an important role in the depression of respiration-like movement in the fetal lamb in utero (Gluckman and Johnston, 1987). The localization of the HRD-inducing mechanism in neonatal animals remains, however, unknown. The reports placing the respiratory depressor mechanism in the pons of neonatal animals under normoxic conditions (Errchidi et al., 1990; Farber, 1990) led us to suspect that the pons may play an important role in the HRD of neonatal animals. The main purpose of the present study was thus, to investigate the functional role of the pons in the neonatal rat HRD. We used the in vitro brainstem preparation to avoid circulatory disturbances that
occur typically in in vivo brainstem sectioning experiments, and we analyzed the effects of ablating the pons, or parts of it, in the neonatal rat. In addition, we studied the functional expression of the diencephalon and mid-brain in central respiratory control of the neonatal rat. Preliminary results have been published in abstract form (Okada et al., 1993a).
2. Methods
2.1. Preparations A total of 38 neonatal rats were used in these experiments (1–3 days old, Sprague-Dawley). The procedure for obtaining the isolated brainstemspinal cord preparation has been described previously (Okada et al., 1993b). In brief, the animal was deeply anesthetized with diethyl ether, and the brainstem, together with the cervical spinal cord, was isolated in a dissecting chamber, filled with oxygenated mock cerebrospinal fluid (mock CSF). The rostral brainstem was transected at one of the following levels: the rostral end of the diencephalon, the diencephalo-midbrain junction (Fig. 1 ), or the midbrain-pontine junction (Fig. 1, level 2 ). The cerebellum was removed. 1, level
Y. Okada et al. / Respiration Physiology 111 (1998) 55–63
The preparation was then transferred to a recording chamber (vol. 2 ml) and mechanically fixed with miniature pins on a silicon rubber base with either ventral or lateral side up. The preparation was superfused at 26°C at a rate of 7 ml min − 1 with oxygenated control mock CSF, which was equilibrated with a gas mixture (90% O2, 2% CO2, balance N2; pH 7.8). The composition of this control mock CSF was (in mM): NaCl 125, KCl 4, NaHCO3 26, NaH2PO4 0.5, CaCl2 2, MgSO4 1, Glucose 30. Neural respiratory activity was recorded with a glass suction electrode from one C4 ventral root and integrated. The signal was recorded simultaneously on two chart recorders with different chart speeds (3 mm min − 1 and 20 mm min − 1). The respiratory frequency (fR) was obtained as the frequency of the C4 burst. The amplitude of the integrated C4 burst activity ( Phr) was used as an index of inspiratory activity. Using a pair of fine ophthalmologic scissors under a high (50× ) magnification stereomicroscope, the brainstem was further transected as described below.
2.2. Experimental Protocol During superfusion with control mock CSF, the brainstem was transected at defined levels 1, 2, 3 – 3 %). In most cases, (Fig. 1, levels the transected tissue was left in its pre-transected position to minimize exposure of the cut surface to mock CSF. The effects of these transections on respiratory frequency were observed. Respiratory responses to hypoxia were examined in the following three types of preparation. In seven preparations, the midbrain was ablated but the entire pons was left intact (preparation 2 ). In seven with the pons intact; Fig. 1, level other preparations, the midbrain and the dorsal pons were both ablated, leaving the ventral pons connected with the medulla (preparation with the 4 ). In ten preparaventral pons; Fig. 1, level tions, the brainstem was transected between lev3 and 3 % (Fig. 1), and the entire pons was els separated (preparation without the pons). The transected pontine tissue was mechanically fixed in the pre-transected position with miniature pins.
57
A given preparation was used only once for a single hypoxic trial. Before any hypoxic test, the preparation was superfused for at least 20 min with control mock CSF to stabilize the respiratory output. Then the superfusate was switched to hypoxic mock CSF, which was equilibrated with a hypoxic gas mixture for 60 min (8% O2, 2% CO2, balance N2; pH 7.8). After 60 min of hypoxic exposure, superfusion was switched back to control mock CSF.
2.3. Data analysis In preliminary experiments the effects of the brainstem transection on fR were consistent. However, the effect on Phr was small and variable across preparations. Therefore, only fR was used for analysis in transection experiments. Because in each set of brainstem transection experiments, distributions of fR data before and after transection were normal, they were compared using a paired t-test. Because the absolute values of fR as well as of Phr varied considerably across preparations in our hypoxic response experiments, they were expressed as a percentage of normoxic control, and so was total respiratory output, calculated as fR · Phr. The indices fR, Phr and fR · Phr were compared during the entire hypoxic period across preparation groups. Because these values were not normally distributed, these differences were analyzed with a Mann-Whitney rank sum test at a significance level PB 0.05. Results are presented as mean9 SE.
3. Results
3.1. Effect of brainstem transection on respiratory output Removal of the diencephalon by transversely 1 ) immedisectioning the neuraxis (Fig. 1, level ately decreased fR (in breath min − 1) from 3.4 91.2 to 1.3 90.6 (PB 0.03, n= 5) (Fig. 2 A). On the other hand, in preparations without the dien2) cephalon, removal of the midbrain (Fig. 1, level increased fR from 1.5 9 0.6 to 3.0 9 1.0 (PB 0.05,
58
Y. Okada et al. / Respiration Physiology 111 (1998) 55–63
n= 7), occasionally accompanied by transient tonic C4 excitation (Fig. 2B). Removal of the entire pons in preparations which lacked the mid3 – 3 %) noticeably increased brain (Fig. 1, level fR from 3.991.0 to 7.490.8 (P B 0.005, n = 7) (Fig. 2C). When the midbrain and dorsal pons were ablated beforehand, removal of the remain3 – 3 %) markedly ing ventral pons (Fig. 1, level increased fR from 2.3 90.7 to 7.7 91.8 (P B0.03, n= 5) (Fig. 2D).
3.2. Effect of brainstem transection on hypoxic respiratory response Irrespective of the brainstem transection levels, some preparations in every group, when exposed to hypoxia, showed an initial increase in fR that lasted for several minutes (Fig. 3B,C). However, in no group was this initial increase statistically significant. In general, the decrease in fR became apparent within 20 min from the onset of the hypoxic exposure (Fig. 3). In preparations with
Fig. 2. Effect of brainstem transections on respiratory output (integrated C4 activity): (A) Transection at the diencephalo1 in Fig. 1). B. Transection at the midbrain junction level ( 2 in Fig. 1). Transient elevamidbrain-pontine junction level ( tion of the baseline after transection represents the tonic C4 excitation; (C) Transection at the ponto-medullary junction 3 – 3 % in Fig. level in the preparation with the pons intact ( 1); (D) Transection at the ponto-medullary junction level in 3 – 3 % in Fig. 1). the preparation with the ventral pons (
Fig. 3. Examples of respiratory responses to hypoxia: (A) Preparation without the pons; (B) Preparation with the ventral pons (without the dorsal pons); (C) Preparation with the pons intact (without the midbrain).
the pons intact, fR decreased progressively. The hypoxic decrease in fR, in preparations with the pons intact, was significantly greater than in preparations without the pons (PB 0.03; Fig. 4A). In preparations with the pons intact, Phr was depressed progressively by sustained hypoxia and this depression was more pronounced than in preparations with only the ventral pons intact (PB0.02; Fig. 4B). On the other hand, in preparations with only the ventral pons intact, depression of Phr was more prominent than in preparations without the pons (PB 0.001), in which Phr was only slightly suppressed by hypoxia, even after 60 min (Fig. 4B). A similar sequence in the amount of suppression was also observed for fR · Phr (PB 0.02; Fig. 4C).
4. Discussion The major finding in the present study is that the decline of both fR and Phr during sustained hypoxia was attenuated by removal of the entire pons. Ablation of the dorsal pons attenuated hypoxic decline of Phr. Further ablation of the ventral pons abolished hypoxic decline of Phr.
Y. Okada et al. / Respiration Physiology 111 (1998) 55–63
59
Fig. 4. Mean data for time course of respiratory responses to hypoxic exposure for 60 min: (A) Response of respiratory frequency (fR); (B). Response of the amplitude of integrated inspiratory burst activity ( Phr); (C) Response of the total respiratory output (fR · Phr). Data are presented as % of control values. Error bars indicate SE.
60
Y. Okada et al. / Respiration Physiology 111 (1998) 55–63
4.1. Critique of methods In the isolated brainstem-spinal cord preparation, the deep brainstem region is hypoxic or anoxic (Brockhaus et al., 1993, Okada et al., 1993b). However, the superficial medullary layer (from the surface up to 400 mm), which is functionally important for respiration, is well oxygenated in this preparation (Okada et al., 1993b), and the oxygen level in this superficial layer can be manipulated by changing the superfusate oxygen fraction (Ballanyi et al., 1992). Indeed, respiratory activities in our study responded well to hypoxia. We are, thus, confident that this preparation is appropriate for the functional study of superficial brainstem regions. The nature of the respiratory output in this preparation has recently been criticized as reflecting gasping in a hypoxic animal (Fung et al., 1996). However, there is considerable evidence that is inconsistent with this criticism (see the discussion in Kawai et al., 1995, 1996). From the perspective of central respiratory control, the brainstem environments of the in vitro brainstemspinal cord preparation and in vivo hypoxic animals are fundamentally different. In the latter preparation, the whole brainstem is hypoxic. In contrast, in the former preparation, the superficial medullary tissue layers, where respiratory rhythm is supposedly generated, remains well oxygenated (Okada et al., 1993b; Duffin et al., 1995) and the deep layers, where gasping is generated (Fung et al., 1994), is totally anoxic (Okada et al., 1993b) and thus, probably not functioning at all. In fact, we have rarely found spontaneously firing neurons in deep tissue layers of this preparation (Okada et al., 1993c). These findings suggest that the respiratory output in the isolated brainstemspinal cord preparation may be fundamentally different in nature from that of in vivo hypoxic animals. Nevertheless, special care is needed when extrapolating the findings obtained in this preparation to in vivo situations. It would be desirable to confirm our in vitro findings in in vivo preparations. In brainstem transection experiments, the tissue micro-environment near the cut surface may change with exposure of the cut surface to mock
CSF. In our experiments, this possible influence was minimized by keeping the transected tissue in the pre-transected position. Furthermore, the effect of brainstem transection on respiratory output appeared immediately after the transection (Fig. 2). This time course of respiratory output change was too fast to be explained by any change in tissue microenvironment (Okada et al., 1993b). Therefore, it is unlikely that a change in the tissue microenvironment, which unavoidably accompanied the transection procedure, was primarily responsible for our results.
4.2. Effect of brainstem transection on respiratory output Our brainstem transection experiments showed that the diencephalon and midbrain have facilitatory and inhibitory influences, respectively, on the respiratory rhythm generator. These findings are compatible with previous reports on adult animal experiments in vivo (Hugelin, 1986). The rat is born in a relatively immature stage and the function of the higher central nervous system is not well developed at birth (Greer et al., 1996). To date, the functional expression of the neonatal rat higher brainstem in central respiratory control has rarely been investigated. We have demonstrated that the excitatory and inhibitory modulatory functions for central respiratory control are already expressed in the diencephalon and midbrain, respectively, despite the immaturity of the neonatal rat brain. This suggests that the in vitro neonatal rat brainstem preparation may be useful in developmental and functional studies of the higher brainstem. We also confirm the inhibitory influence of the pons on the medullary respiratory neuronal network in the neonatal animal. This finding is in agreement with previous in vivo experiments in the developing opossum (Farber, 1990) and the in vitro neonatal rat (Errchidi et al., 1990; Hamada et al., 1994).
4.3. Respiratory response to hypoxia In the present study, the common major respiratory response to hypoxia in most preparations
Y. Okada et al. / Respiration Physiology 111 (1998) 55–63
was steady depression of the total respiratory output, although some preparations showed a biphasic response: brief initial augmentation followed by a steady decline in respiratory activity (Figs. 3 and 4). This biphasic response suggests that the initial excitatory component is at least in part central in origin. This finding is consistent with previous experiments with unanesthetized peripherally chemodenervated animals with mild to moderate hypoxia (Neubauer et al., 1990). The mechanism for the brief initial augmentation is unknown. It may be caused either by hypoxic suppression of the mechanism that inhibits respiratory output (disinhibition), or by tissue acidification that stimulates central chemosensitivity. The preparation with the pons intact showed the strongest HRD among the three types of preparation, and this was due to decreases in both fR and Phr (Figs. 3 and 4). These patterns appear to be similar to those in the in vivo paralyzed, artificially ventilated and peripherally chemodenervated neonatal rat (Fung et al., 1996). The preparation with only the ventral pons intact showed similar but weaker response patterns. The preparation without the pons exhibited even weaker responses to hypoxia, and Phr in this type of preparation was very little depressed by sustained hypoxia. These results indicate that a considerable portion of HRD in the neonatal rat is induced by supramedullary mechanisms. Respiratory depression by hypoxia can be due, either to an active inhibition of the medullary respiratory network or to direct inhibition of respiratory neurons by O2 deprivation (Neubauer et al., 1990). The effects of hypoxia on central respiratory control are strikingly different in adult and neonatal animals (Ballanyi et al., 1992, Eden and Hanson, 1987; Haddad and Jiang, 1993). HRD in perinatal animals appears to be induced by an active mechanism (Gluckman and Johnston, 1987; Neubauer et al., 1990), and several neurochemicals may be involved in HRD (Neubauer et al., 1990). We and others have reported the possible involvement of adenosine in the HRD of perinatal animals (Dong and Feldman, 1995; Kawai et al., 1995; Koos et al., 1994), which
61
also has a direct neuron-protecting effect from hypoxic injury (Haddad and Jiang, 1993). We confirm that HRD in the neonate is induced mainly by an active mechanism because ablation of the pons abolished HRD. Before birth such a mechanism is thought to be important in protection of neurons through reduction of energy consumption.
4.4. Localization of the neural substrate for HRD In the adult cat (Gallman and Millhorn, 1988; Gallman et al., 1991), adult rat (Martin-Body, 1988) and neonatal rabbit (Martin-Body and Johnston, 1988), it has been reported that the midbrain is the major anatomical substrate for HRD. We have not studied the role of the midbrain in the neonatal rat HRD. However, as we have demonstrated, the midbrain executes a depressing action on respiration (Fig. 2B) and it is probable that the midbrain is involved in the neonatal rat HRD. The presence of a respiratory depressing mechanism in the pons has been reported in adult and neonatal animals under normoxic conditions (Errchidi et al., 1990; Farber, 1990; Hamada et al., 1994). Gluckman and Johnston (1987) also reported that the pons is involved in the depression of the respiration-like movement in the fetal lamb. However, a role of the pons in the HRD of neonatal animals has not yet been described in the literature. In the present study, Phr was little suppressed by hypoxia in preparations without the pons, but Phr was markedly suppressed in preparations with only the ventral pons intact (Fig. 4B). Although it has been reported in the rat, that not only the dorsal but also the ventral pons is involved in respiratory control (Errchidi et al., 1990; Hamada et al., 1994), our results indicate that it is mainly the ventral pons that is involved in the HRD of neonatal rats. Ablation of the dorsal pons also attenuated the hypoxic depression of Phr (Fig. 4B) and of fR · Phr (Fig. 4C), revealing an involvement of the dorsal pons in the HRD of the neonatal rat. This finding is consistent with a recent report which
62
Y. Okada et al. / Respiration Physiology 111 (1998) 55–63
indicates that the parabrachial nuclei participate in the control of the in vivo adult rat HRD (Mizusawa et al., 1995). Although we did not identify the pontine neurons that are involved in HRD, the A5 noradrenergic neurons in the ventral pons, for example, may be involved in the HRD-inducing mechanism. Noradrenaline, which inhibits respiration, is released from these neurons (Errchidi et al., 1990). A morphological connection also has been reported from the A5 noradrenergic neurons to the neurons in the ventrolateral reticular formation of the medulla (Byrum and Guyenet, 1987). Similarly, in the dorsal pons, A7 noradrenergic neurons in the parabrachial nuclei may be involved in the neonatal HRD. Further studies are needed to identify the neurons which induce HRD in neonatal animals.
5. Summary This study demonstrates that the pons contributes importantly to the induction of hypoxic respiratory depression. Both the ventral and dorsal portions of the pons are involved in the control of hypoxic respiratory depression of the neonatal rat. Although the neonatal rat is born in an immature state, the function of the supramedullary brainstem in respiratory modulation is already expressed at birth. The diencephalon shows facilitatory and the midbrain and pons inhibitory, influences on the medullary respiratory neuronal network.
Acknowledgements We thank F.L. Eldridge and D. Ballantyne for critical comments and suggestions.
References Ballanyi, K., Kuwana, S., Vo¨lker, A., Morawietz, G., Richter, D.W., 1992. Developmental changes in the hypoxia tolerance of the in vitro respiratory network of rats. Neurosci. Lett. 148, 141 – 144.
Brockhaus, J., Ballanyi, K., Smith, J.C., Richter, D.W., 1993. Microenvironment of respiratory neurons in the in vitro brainstem-spinal cord of neonatal rats. J. Physiol. (London) 462, 421 – 445. 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. Dong, X., Feldman, J.L., 1995. Modulation of inspiratory drive to phrenic motoneurons by presynaptic adenosine A1 receptors. J. Neurosci. 15, 3458 – 3467. Duffin, J., Ezure, K., Lipski, J., 1995. Breathing rhythm generation: focus on the rostral ventrolateral medulla. News Physiol. Sci. 10, 133 – 140. Eden, G.J., Hanson, M.A., 1987. Maturation of the respiratory response to acute hypoxia in the newborn rat. J. Physiol. (London) 392, 1 – 9. Errchidi, S., Hilaire, G., Monteau, R., 1990. Permanent release of noradrenaline modulates respiratory frequency in the newborn rat: an in vitro study. J. Physiol. (London) 429, 497 – 510. Farber, J.P., 1990. Effects on breathing of rostral pons glutamate injection during opossum development. J. Appl. Physiol. 69, 189 – 195. Fukuda, Y., 1992. Changes in ventilatory response to hypoxia in the rat during growth and aging. Pflu¨g. Arch. 421, 200 – 203. Fung, M., Wang, W., St. John, W.M., 1994. Medullary loci critical for expression of gasping in adult rats. J. Physiol. (London) 480, 597 – 611. Fung, M., Wang, W., Darnall, R.A., St. John, W.M., 1996. Characterization of ventilatory responses to hypoxia in neonatal rats. Respir. Physiol. 103, 57 – 66. Gallman, E.A., Millhorn, D.E., 1988. Two long-lasting central respiratory responses following acute hypoxia in glomectomized cats. J. Physiol. (London) 395, 333 – 347. Gallman, E.A., Lawing, W.L., Millhorn, D.E., 1991. Mesencephalic stimulation elicits inhibition of phrenic nerve activity in cat. J. Physiol. (London) 436, 405 – 420. Gluckman, P.D., Johnston, B.M., 1987. Lesions in the upper lateral pons abolish the depression of breathing in unanesthetized fetal lambs in utero. J. Physiol. (London) 382, 373 – 383. Greer, J.J., Carter, J.E., Allan, D.W., 1996. Respiratory rhythm generation in a precocial rodent in vitro preparation. Respir. Physiol. 103, 105 – 112. Haddad, G.G., Jiang, C., 1993. O2 deprivation in the central nervous system: on mechanisms of neuronal response, differential sensitivity and injury. Prog. Neurobiol. 40, 277 – 318. Hamada, O., Garcia-Rill, E., Skinner, R.D., 1994. Electrical activation and inhibition of respiration in vitro. Neurosci. Res. 19, 131 – 142. Hugelin, A., 1986. Forebrain and midbrain influence on respiration. In: Cherniack, N.S., Widdicombe, J.G. (Eds.), Handbook of Physiology, section 3: The Respiratory System, vol. II: Control of Breathing, part 1. American Physiological Society, Washington DC, pp. 69 – 91.
Y. Okada et al. / Respiration Physiology 111 (1998) 55–63 Kawai, A., Okada, Y., Mu¨ckenhoff, K., Scheid, P., 1995. Theophylline and hypoxic response in the rat isolated brainstem-spinal cord. Respir. Physiol. 100, 25–32. Kawai, A., Ballantyne, D., Mu¨ckenhoff, K., Scheid, P., 1996. Chemosensitive medullary neurons in the brainstem-spinal cord preparation of the neonatal rat. J. Physiol. (London) 492, 277 – 292. Koos, B.J., Mason, B.A., Punla, O., Adinolfi, A.M., 1994. Hypoxic inhibition of breathing in fetal sheep: relationship to brain adenosine concentrations. J. Appl. Physiol. 77, 2734 – 2739. 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., Johnston, B.M., 1988. Central origin of the hypoxic depression of breathing in the newborn. Respir. Physiol. 71, 25–32.
.
63
Mizusawa, A., Ogawa, H., Kikuchi, Y., Hida, W., Shirato, K., 1995. Role of the parabrachial nucleus in ventilatory responses of awake rats. J. Physiol. (London) 489, 877 – 884. Neubauer, J.D., Melton, J.E., Edelman, N.H., 1990. Modulation of respiration during brain hypoxia. J. Appl. Physiol. 68, 441 – 451. Okada, Y., Kawai, A., Mu¨ckenhoff, K., Scheid, P., 1993a. Localization of the mechanism of the hypoxic ventilatory depression in the brainstem. XXXII IUPS Abst. 185. Okada, Y., Mu¨ckenhoff, K., Holtermann, G., Acker, H., Scheid, P., 1993b. Depth profiles of pH and PO2 in the isolated brain stem-spinal cord of the neonatal rat. Respir. Physiol. 93, 315 – 326. Okada, Y., Mu¨ckenhoff, K., Scheid, P., 1993c. Hypercapnia and medullary neurons in the isolated brain stem-spinal cord of the rat. Respir. Physiol. 93, 327 – 336.
Role of the pons in hypoxic respiratory depression in the neonatal rat Y. Okada *, A. Kawai 1, K. Mu¨ckenhoff, P. Scheid Institut fu¨r Physiologie, Ruhr-Uni6ersita¨t Bochum, D-44780 Bochum, Germany Accepted 8 September 1997
Abstract The main purpose of this study was to evaluate the role of the pons in hypoxic respiratory depression (HRD) of the neonatal rat. Experiments were conducted using the isolated brainstem-spinal cord preparation of the neonatal rat (1–3 days old). The brainstem was transected at various levels. We found that ablation of the diencephalon decreased respiratory frequency (fR), and conversely, that ablation of the midbrain or pons increased fR. In the preparation with the pons intact (without the midbrain), hypoxia (superfusate PO2 = 56 mmHg) caused strong depression of respiratory activity, which was characterized by a steady decrease in fR and in integrated inspiratory burst amplitude ( Phr). In the preparation with the intact ventral pons (without midbrain and dorsal pons) we observed similar, though weaker, HRD. When the entire pons was ablated, Phr was little depressed by hypoxia and thus, HRD was further attenuated. We conclude that the pons contributes importantly to the induction of hypoxic respiratory depression in the neonatal rat. Both the ventral and dorsal portions of the pons are involved in the control of hypoxic respiratory depression. In addition, we show that the respiratory modulatory functions of the diencephalon (facilitating) and midbrain (inhibitory) are already expressed at the time of birth. © 1998 Elsevier Science B.V. Keywords: Control of breathing, hypoxia, development; Hypoxia, central respiratory depression; Brainstem, diencephalon, midbrain, pons, medulla oblongata; Mammals, neonatal rat; Development, control of breathing
1. Introduction * Corresponding author. Present address. Department of Medicine, Keio University, Tagata-gun, Tsukigase Rehabilitation Cente, Tsukigase 380-2, Amagiyugashima-cho, Shizuokaken, 410-3215 Japan. Tel.: + 81 558 851701; fax: +81 558 851810; e-mail:[email protected] 1 Present address: Department of Medicine, Keio University, 35 Shinanomachi, Shinjuku-ku, Tokyo, 160-8582 Japan.
The respiratory response to hypoxia in mammals is typically biphasic; an initial augmentation is followed by a depression of ventilation, the so-called hypoxic respiratory depression (HRD), which is thought to constitute a response of the central nervous system (Martin-Body, 1988
0034-5687/98/$19.00 © 1998 Elsevier Science B.V. All rights reserved. PII S 0 0 3 4 - 5 6 8 7 ( 9 7 ) 0 0 1 0 5 - 9
56
Y. Okada et al. / Respiration Physiology 111 (1998) 55–63
Fig. 1. Levels of brainstem transection: (A) Schematic diagram of the ventral brainstem surface; (B) Lateral view of the brainstem. 1, V, VI, IX, X, XI and XII: fifth, sixth, ninth, tenth, eleventh and twelfth cranial nerve roots, respectively. The sections are: 2 , midbrain-pontine junction; 3 – 3 %, transection level for the pontine ablation; 4 , transection diencephalo-midbrain junction; line for the ablation of the dorsal pons.
Martin-Body and Johnston, 1988; for a review see Neubauer et al., 1990). In rats, the respiratory response to hypoxia changes substantially with development and growth (Ballanyi et al., 1992; Eden and Hanson, 1987; Fukuda, 1992), whereby neuronal responses to hypoxia in adults and neonates are strikingly different (Haddad and Jiang, 1993). The mechanisms underlying HRD may therefore differ between adults and neonates. In adult animals the HRD-inducing mechanism has been reported to be located in the midbrain in the cat (Gallman et al., 1991), rabbit (Martin-Body and Johnston, 1988) and rat (Martin-Body, 1988). On the other hand, the upper lateral pons has been reported to play an important role in the depression of respiration-like movement in the fetal lamb in utero (Gluckman and Johnston, 1987). The localization of the HRD-inducing mechanism in neonatal animals remains, however, unknown. The reports placing the respiratory depressor mechanism in the pons of neonatal animals under normoxic conditions (Errchidi et al., 1990; Farber, 1990) led us to suspect that the pons may play an important role in the HRD of neonatal animals. The main purpose of the present study was thus, to investigate the functional role of the pons in the neonatal rat HRD. We used the in vitro brainstem preparation to avoid circulatory disturbances that
occur typically in in vivo brainstem sectioning experiments, and we analyzed the effects of ablating the pons, or parts of it, in the neonatal rat. In addition, we studied the functional expression of the diencephalon and mid-brain in central respiratory control of the neonatal rat. Preliminary results have been published in abstract form (Okada et al., 1993a).
2. Methods
2.1. Preparations A total of 38 neonatal rats were used in these experiments (1–3 days old, Sprague-Dawley). The procedure for obtaining the isolated brainstemspinal cord preparation has been described previously (Okada et al., 1993b). In brief, the animal was deeply anesthetized with diethyl ether, and the brainstem, together with the cervical spinal cord, was isolated in a dissecting chamber, filled with oxygenated mock cerebrospinal fluid (mock CSF). The rostral brainstem was transected at one of the following levels: the rostral end of the diencephalon, the diencephalo-midbrain junction (Fig. 1 ), or the midbrain-pontine junction (Fig. 1, level 2 ). The cerebellum was removed. 1, level
Y. Okada et al. / Respiration Physiology 111 (1998) 55–63
The preparation was then transferred to a recording chamber (vol. 2 ml) and mechanically fixed with miniature pins on a silicon rubber base with either ventral or lateral side up. The preparation was superfused at 26°C at a rate of 7 ml min − 1 with oxygenated control mock CSF, which was equilibrated with a gas mixture (90% O2, 2% CO2, balance N2; pH 7.8). The composition of this control mock CSF was (in mM): NaCl 125, KCl 4, NaHCO3 26, NaH2PO4 0.5, CaCl2 2, MgSO4 1, Glucose 30. Neural respiratory activity was recorded with a glass suction electrode from one C4 ventral root and integrated. The signal was recorded simultaneously on two chart recorders with different chart speeds (3 mm min − 1 and 20 mm min − 1). The respiratory frequency (fR) was obtained as the frequency of the C4 burst. The amplitude of the integrated C4 burst activity ( Phr) was used as an index of inspiratory activity. Using a pair of fine ophthalmologic scissors under a high (50× ) magnification stereomicroscope, the brainstem was further transected as described below.
2.2. Experimental Protocol During superfusion with control mock CSF, the brainstem was transected at defined levels 1, 2, 3 – 3 %). In most cases, (Fig. 1, levels the transected tissue was left in its pre-transected position to minimize exposure of the cut surface to mock CSF. The effects of these transections on respiratory frequency were observed. Respiratory responses to hypoxia were examined in the following three types of preparation. In seven preparations, the midbrain was ablated but the entire pons was left intact (preparation 2 ). In seven with the pons intact; Fig. 1, level other preparations, the midbrain and the dorsal pons were both ablated, leaving the ventral pons connected with the medulla (preparation with the 4 ). In ten preparaventral pons; Fig. 1, level tions, the brainstem was transected between lev3 and 3 % (Fig. 1), and the entire pons was els separated (preparation without the pons). The transected pontine tissue was mechanically fixed in the pre-transected position with miniature pins.
57
A given preparation was used only once for a single hypoxic trial. Before any hypoxic test, the preparation was superfused for at least 20 min with control mock CSF to stabilize the respiratory output. Then the superfusate was switched to hypoxic mock CSF, which was equilibrated with a hypoxic gas mixture for 60 min (8% O2, 2% CO2, balance N2; pH 7.8). After 60 min of hypoxic exposure, superfusion was switched back to control mock CSF.
2.3. Data analysis In preliminary experiments the effects of the brainstem transection on fR were consistent. However, the effect on Phr was small and variable across preparations. Therefore, only fR was used for analysis in transection experiments. Because in each set of brainstem transection experiments, distributions of fR data before and after transection were normal, they were compared using a paired t-test. Because the absolute values of fR as well as of Phr varied considerably across preparations in our hypoxic response experiments, they were expressed as a percentage of normoxic control, and so was total respiratory output, calculated as fR · Phr. The indices fR, Phr and fR · Phr were compared during the entire hypoxic period across preparation groups. Because these values were not normally distributed, these differences were analyzed with a Mann-Whitney rank sum test at a significance level PB 0.05. Results are presented as mean9 SE.
3. Results
3.1. Effect of brainstem transection on respiratory output Removal of the diencephalon by transversely 1 ) immedisectioning the neuraxis (Fig. 1, level ately decreased fR (in breath min − 1) from 3.4 91.2 to 1.3 90.6 (PB 0.03, n= 5) (Fig. 2 A). On the other hand, in preparations without the dien2) cephalon, removal of the midbrain (Fig. 1, level increased fR from 1.5 9 0.6 to 3.0 9 1.0 (PB 0.05,
58
Y. Okada et al. / Respiration Physiology 111 (1998) 55–63
n= 7), occasionally accompanied by transient tonic C4 excitation (Fig. 2B). Removal of the entire pons in preparations which lacked the mid3 – 3 %) noticeably increased brain (Fig. 1, level fR from 3.991.0 to 7.490.8 (P B 0.005, n = 7) (Fig. 2C). When the midbrain and dorsal pons were ablated beforehand, removal of the remain3 – 3 %) markedly ing ventral pons (Fig. 1, level increased fR from 2.3 90.7 to 7.7 91.8 (P B0.03, n= 5) (Fig. 2D).
3.2. Effect of brainstem transection on hypoxic respiratory response Irrespective of the brainstem transection levels, some preparations in every group, when exposed to hypoxia, showed an initial increase in fR that lasted for several minutes (Fig. 3B,C). However, in no group was this initial increase statistically significant. In general, the decrease in fR became apparent within 20 min from the onset of the hypoxic exposure (Fig. 3). In preparations with
Fig. 2. Effect of brainstem transections on respiratory output (integrated C4 activity): (A) Transection at the diencephalo1 in Fig. 1). B. Transection at the midbrain junction level ( 2 in Fig. 1). Transient elevamidbrain-pontine junction level ( tion of the baseline after transection represents the tonic C4 excitation; (C) Transection at the ponto-medullary junction 3 – 3 % in Fig. level in the preparation with the pons intact ( 1); (D) Transection at the ponto-medullary junction level in 3 – 3 % in Fig. 1). the preparation with the ventral pons (
Fig. 3. Examples of respiratory responses to hypoxia: (A) Preparation without the pons; (B) Preparation with the ventral pons (without the dorsal pons); (C) Preparation with the pons intact (without the midbrain).
the pons intact, fR decreased progressively. The hypoxic decrease in fR, in preparations with the pons intact, was significantly greater than in preparations without the pons (PB 0.03; Fig. 4A). In preparations with the pons intact, Phr was depressed progressively by sustained hypoxia and this depression was more pronounced than in preparations with only the ventral pons intact (PB0.02; Fig. 4B). On the other hand, in preparations with only the ventral pons intact, depression of Phr was more prominent than in preparations without the pons (PB 0.001), in which Phr was only slightly suppressed by hypoxia, even after 60 min (Fig. 4B). A similar sequence in the amount of suppression was also observed for fR · Phr (PB 0.02; Fig. 4C).
4. Discussion The major finding in the present study is that the decline of both fR and Phr during sustained hypoxia was attenuated by removal of the entire pons. Ablation of the dorsal pons attenuated hypoxic decline of Phr. Further ablation of the ventral pons abolished hypoxic decline of Phr.
Y. Okada et al. / Respiration Physiology 111 (1998) 55–63
59
Fig. 4. Mean data for time course of respiratory responses to hypoxic exposure for 60 min: (A) Response of respiratory frequency (fR); (B). Response of the amplitude of integrated inspiratory burst activity ( Phr); (C) Response of the total respiratory output (fR · Phr). Data are presented as % of control values. Error bars indicate SE.
60
Y. Okada et al. / Respiration Physiology 111 (1998) 55–63
4.1. Critique of methods In the isolated brainstem-spinal cord preparation, the deep brainstem region is hypoxic or anoxic (Brockhaus et al., 1993, Okada et al., 1993b). However, the superficial medullary layer (from the surface up to 400 mm), which is functionally important for respiration, is well oxygenated in this preparation (Okada et al., 1993b), and the oxygen level in this superficial layer can be manipulated by changing the superfusate oxygen fraction (Ballanyi et al., 1992). Indeed, respiratory activities in our study responded well to hypoxia. We are, thus, confident that this preparation is appropriate for the functional study of superficial brainstem regions. The nature of the respiratory output in this preparation has recently been criticized as reflecting gasping in a hypoxic animal (Fung et al., 1996). However, there is considerable evidence that is inconsistent with this criticism (see the discussion in Kawai et al., 1995, 1996). From the perspective of central respiratory control, the brainstem environments of the in vitro brainstemspinal cord preparation and in vivo hypoxic animals are fundamentally different. In the latter preparation, the whole brainstem is hypoxic. In contrast, in the former preparation, the superficial medullary tissue layers, where respiratory rhythm is supposedly generated, remains well oxygenated (Okada et al., 1993b; Duffin et al., 1995) and the deep layers, where gasping is generated (Fung et al., 1994), is totally anoxic (Okada et al., 1993b) and thus, probably not functioning at all. In fact, we have rarely found spontaneously firing neurons in deep tissue layers of this preparation (Okada et al., 1993c). These findings suggest that the respiratory output in the isolated brainstemspinal cord preparation may be fundamentally different in nature from that of in vivo hypoxic animals. Nevertheless, special care is needed when extrapolating the findings obtained in this preparation to in vivo situations. It would be desirable to confirm our in vitro findings in in vivo preparations. In brainstem transection experiments, the tissue micro-environment near the cut surface may change with exposure of the cut surface to mock
CSF. In our experiments, this possible influence was minimized by keeping the transected tissue in the pre-transected position. Furthermore, the effect of brainstem transection on respiratory output appeared immediately after the transection (Fig. 2). This time course of respiratory output change was too fast to be explained by any change in tissue microenvironment (Okada et al., 1993b). Therefore, it is unlikely that a change in the tissue microenvironment, which unavoidably accompanied the transection procedure, was primarily responsible for our results.
4.2. Effect of brainstem transection on respiratory output Our brainstem transection experiments showed that the diencephalon and midbrain have facilitatory and inhibitory influences, respectively, on the respiratory rhythm generator. These findings are compatible with previous reports on adult animal experiments in vivo (Hugelin, 1986). The rat is born in a relatively immature stage and the function of the higher central nervous system is not well developed at birth (Greer et al., 1996). To date, the functional expression of the neonatal rat higher brainstem in central respiratory control has rarely been investigated. We have demonstrated that the excitatory and inhibitory modulatory functions for central respiratory control are already expressed in the diencephalon and midbrain, respectively, despite the immaturity of the neonatal rat brain. This suggests that the in vitro neonatal rat brainstem preparation may be useful in developmental and functional studies of the higher brainstem. We also confirm the inhibitory influence of the pons on the medullary respiratory neuronal network in the neonatal animal. This finding is in agreement with previous in vivo experiments in the developing opossum (Farber, 1990) and the in vitro neonatal rat (Errchidi et al., 1990; Hamada et al., 1994).
4.3. Respiratory response to hypoxia In the present study, the common major respiratory response to hypoxia in most preparations
Y. Okada et al. / Respiration Physiology 111 (1998) 55–63
was steady depression of the total respiratory output, although some preparations showed a biphasic response: brief initial augmentation followed by a steady decline in respiratory activity (Figs. 3 and 4). This biphasic response suggests that the initial excitatory component is at least in part central in origin. This finding is consistent with previous experiments with unanesthetized peripherally chemodenervated animals with mild to moderate hypoxia (Neubauer et al., 1990). The mechanism for the brief initial augmentation is unknown. It may be caused either by hypoxic suppression of the mechanism that inhibits respiratory output (disinhibition), or by tissue acidification that stimulates central chemosensitivity. The preparation with the pons intact showed the strongest HRD among the three types of preparation, and this was due to decreases in both fR and Phr (Figs. 3 and 4). These patterns appear to be similar to those in the in vivo paralyzed, artificially ventilated and peripherally chemodenervated neonatal rat (Fung et al., 1996). The preparation with only the ventral pons intact showed similar but weaker response patterns. The preparation without the pons exhibited even weaker responses to hypoxia, and Phr in this type of preparation was very little depressed by sustained hypoxia. These results indicate that a considerable portion of HRD in the neonatal rat is induced by supramedullary mechanisms. Respiratory depression by hypoxia can be due, either to an active inhibition of the medullary respiratory network or to direct inhibition of respiratory neurons by O2 deprivation (Neubauer et al., 1990). The effects of hypoxia on central respiratory control are strikingly different in adult and neonatal animals (Ballanyi et al., 1992, Eden and Hanson, 1987; Haddad and Jiang, 1993). HRD in perinatal animals appears to be induced by an active mechanism (Gluckman and Johnston, 1987; Neubauer et al., 1990), and several neurochemicals may be involved in HRD (Neubauer et al., 1990). We and others have reported the possible involvement of adenosine in the HRD of perinatal animals (Dong and Feldman, 1995; Kawai et al., 1995; Koos et al., 1994), which
61
also has a direct neuron-protecting effect from hypoxic injury (Haddad and Jiang, 1993). We confirm that HRD in the neonate is induced mainly by an active mechanism because ablation of the pons abolished HRD. Before birth such a mechanism is thought to be important in protection of neurons through reduction of energy consumption.
4.4. Localization of the neural substrate for HRD In the adult cat (Gallman and Millhorn, 1988; Gallman et al., 1991), adult rat (Martin-Body, 1988) and neonatal rabbit (Martin-Body and Johnston, 1988), it has been reported that the midbrain is the major anatomical substrate for HRD. We have not studied the role of the midbrain in the neonatal rat HRD. However, as we have demonstrated, the midbrain executes a depressing action on respiration (Fig. 2B) and it is probable that the midbrain is involved in the neonatal rat HRD. The presence of a respiratory depressing mechanism in the pons has been reported in adult and neonatal animals under normoxic conditions (Errchidi et al., 1990; Farber, 1990; Hamada et al., 1994). Gluckman and Johnston (1987) also reported that the pons is involved in the depression of the respiration-like movement in the fetal lamb. However, a role of the pons in the HRD of neonatal animals has not yet been described in the literature. In the present study, Phr was little suppressed by hypoxia in preparations without the pons, but Phr was markedly suppressed in preparations with only the ventral pons intact (Fig. 4B). Although it has been reported in the rat, that not only the dorsal but also the ventral pons is involved in respiratory control (Errchidi et al., 1990; Hamada et al., 1994), our results indicate that it is mainly the ventral pons that is involved in the HRD of neonatal rats. Ablation of the dorsal pons also attenuated the hypoxic depression of Phr (Fig. 4B) and of fR · Phr (Fig. 4C), revealing an involvement of the dorsal pons in the HRD of the neonatal rat. This finding is consistent with a recent report which
62
Y. Okada et al. / Respiration Physiology 111 (1998) 55–63
indicates that the parabrachial nuclei participate in the control of the in vivo adult rat HRD (Mizusawa et al., 1995). Although we did not identify the pontine neurons that are involved in HRD, the A5 noradrenergic neurons in the ventral pons, for example, may be involved in the HRD-inducing mechanism. Noradrenaline, which inhibits respiration, is released from these neurons (Errchidi et al., 1990). A morphological connection also has been reported from the A5 noradrenergic neurons to the neurons in the ventrolateral reticular formation of the medulla (Byrum and Guyenet, 1987). Similarly, in the dorsal pons, A7 noradrenergic neurons in the parabrachial nuclei may be involved in the neonatal HRD. Further studies are needed to identify the neurons which induce HRD in neonatal animals.
5. Summary This study demonstrates that the pons contributes importantly to the induction of hypoxic respiratory depression. Both the ventral and dorsal portions of the pons are involved in the control of hypoxic respiratory depression of the neonatal rat. Although the neonatal rat is born in an immature state, the function of the supramedullary brainstem in respiratory modulation is already expressed at birth. The diencephalon shows facilitatory and the midbrain and pons inhibitory, influences on the medullary respiratory neuronal network.
Acknowledgements We thank F.L. Eldridge and D. Ballantyne for critical comments and suggestions.
References Ballanyi, K., Kuwana, S., Vo¨lker, A., Morawietz, G., Richter, D.W., 1992. Developmental changes in the hypoxia tolerance of the in vitro respiratory network of rats. Neurosci. Lett. 148, 141 – 144.
Brockhaus, J., Ballanyi, K., Smith, J.C., Richter, D.W., 1993. Microenvironment of respiratory neurons in the in vitro brainstem-spinal cord of neonatal rats. J. Physiol. (London) 462, 421 – 445. 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. Dong, X., Feldman, J.L., 1995. Modulation of inspiratory drive to phrenic motoneurons by presynaptic adenosine A1 receptors. J. Neurosci. 15, 3458 – 3467. Duffin, J., Ezure, K., Lipski, J., 1995. Breathing rhythm generation: focus on the rostral ventrolateral medulla. News Physiol. Sci. 10, 133 – 140. Eden, G.J., Hanson, M.A., 1987. Maturation of the respiratory response to acute hypoxia in the newborn rat. J. Physiol. (London) 392, 1 – 9. Errchidi, S., Hilaire, G., Monteau, R., 1990. Permanent release of noradrenaline modulates respiratory frequency in the newborn rat: an in vitro study. J. Physiol. (London) 429, 497 – 510. Farber, J.P., 1990. Effects on breathing of rostral pons glutamate injection during opossum development. J. Appl. Physiol. 69, 189 – 195. Fukuda, Y., 1992. Changes in ventilatory response to hypoxia in the rat during growth and aging. Pflu¨g. Arch. 421, 200 – 203. Fung, M., Wang, W., St. John, W.M., 1994. Medullary loci critical for expression of gasping in adult rats. J. Physiol. (London) 480, 597 – 611. Fung, M., Wang, W., Darnall, R.A., St. John, W.M., 1996. Characterization of ventilatory responses to hypoxia in neonatal rats. Respir. Physiol. 103, 57 – 66. Gallman, E.A., Millhorn, D.E., 1988. Two long-lasting central respiratory responses following acute hypoxia in glomectomized cats. J. Physiol. (London) 395, 333 – 347. Gallman, E.A., Lawing, W.L., Millhorn, D.E., 1991. Mesencephalic stimulation elicits inhibition of phrenic nerve activity in cat. J. Physiol. (London) 436, 405 – 420. Gluckman, P.D., Johnston, B.M., 1987. Lesions in the upper lateral pons abolish the depression of breathing in unanesthetized fetal lambs in utero. J. Physiol. (London) 382, 373 – 383. Greer, J.J., Carter, J.E., Allan, D.W., 1996. Respiratory rhythm generation in a precocial rodent in vitro preparation. Respir. Physiol. 103, 105 – 112. Haddad, G.G., Jiang, C., 1993. O2 deprivation in the central nervous system: on mechanisms of neuronal response, differential sensitivity and injury. Prog. Neurobiol. 40, 277 – 318. Hamada, O., Garcia-Rill, E., Skinner, R.D., 1994. Electrical activation and inhibition of respiration in vitro. Neurosci. Res. 19, 131 – 142. Hugelin, A., 1986. Forebrain and midbrain influence on respiration. In: Cherniack, N.S., Widdicombe, J.G. (Eds.), Handbook of Physiology, section 3: The Respiratory System, vol. II: Control of Breathing, part 1. American Physiological Society, Washington DC, pp. 69 – 91.
Y. Okada et al. / Respiration Physiology 111 (1998) 55–63 Kawai, A., Okada, Y., Mu¨ckenhoff, K., Scheid, P., 1995. Theophylline and hypoxic response in the rat isolated brainstem-spinal cord. Respir. Physiol. 100, 25–32. Kawai, A., Ballantyne, D., Mu¨ckenhoff, K., Scheid, P., 1996. Chemosensitive medullary neurons in the brainstem-spinal cord preparation of the neonatal rat. J. Physiol. (London) 492, 277 – 292. Koos, B.J., Mason, B.A., Punla, O., Adinolfi, A.M., 1994. Hypoxic inhibition of breathing in fetal sheep: relationship to brain adenosine concentrations. J. Appl. Physiol. 77, 2734 – 2739. 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., Johnston, B.M., 1988. Central origin of the hypoxic depression of breathing in the newborn. Respir. Physiol. 71, 25–32.
.
63
Mizusawa, A., Ogawa, H., Kikuchi, Y., Hida, W., Shirato, K., 1995. Role of the parabrachial nucleus in ventilatory responses of awake rats. J. Physiol. (London) 489, 877 – 884. Neubauer, J.D., Melton, J.E., Edelman, N.H., 1990. Modulation of respiration during brain hypoxia. J. Appl. Physiol. 68, 441 – 451. Okada, Y., Kawai, A., Mu¨ckenhoff, K., Scheid, P., 1993a. Localization of the mechanism of the hypoxic ventilatory depression in the brainstem. XXXII IUPS Abst. 185. Okada, Y., Mu¨ckenhoff, K., Holtermann, G., Acker, H., Scheid, P., 1993b. Depth profiles of pH and PO2 in the isolated brain stem-spinal cord of the neonatal rat. Respir. Physiol. 93, 315 – 326. Okada, Y., Mu¨ckenhoff, K., Scheid, P., 1993c. Hypercapnia and medullary neurons in the isolated brain stem-spinal cord of the rat. Respir. Physiol. 93, 327 – 336.