Central respiratory effects of substance P in neonatal mice: an in vitro study

Neuroscience Letters 266 (1999) 189±192

Central respiratory effects of substance P in neonatal mice: an in vitro study Krzysztof Ptak, GeÂrard Hilaire* UPR CNRS 9011, Neurobiology and Movements, Chemin J. Aiguier, 13402 Marseille Cedex 20, France Received 27 January 1999; received in revised form 22 March 1999; accepted 24 March 1999

Abstract Experiments were performed on neonatal mice to know whether substance P (SP) modi®ed the rhythm and the amplitude of the phrenic bursts generated in vitro in brainstem-cervical cord preparations. In OF1 and C3H neonatal preparations, SP or the tachykinin NK1 receptor agonist [Sar 9,Met(O2) 11] substance P both increased signi®cantly phrenic burst amplitude (10 27 M) but had no signi®cant effect on respiratory rhythm unless used at concentrations 10 times larger. In neonates from the monoamine oxidase-A de®cient transgenic Tg8 line, SP increased phrenic burst amplitude but had no effect on the respiratory rhythm at the tested concentrations. The role of SP in regulating neonatal respiratory activity is discussed on the basis of rat and mouse results. q 1999 Elsevier Science Ireland Ltd. All rights reserved. Keywords: Respiratory rhythm; Substance P; Neonates; Rodents

As reviewed recently [8], there is a large body of data suggesting that substance P (SP) is involved in the regulation of the mammalian respiratory network activity. Substance P is present in the sensory afferents from arterial baroreceptors and chemoreceptors, and at their ending sites within medullary `respiratory areas' [7,11,13,15]. Substance P which is likely to be involved in the respiratory response to hypoxia [16,19], may regulate respiratory activity since its local application excites medullary respiratory cells [18,20] increasing both respiratory frequency (RF) and tidal volume [4±6,22]. Although denervation experiments suggested that only a fraction of SP in the medulla originated from the periphery [7], whether the SP-induced respiratory changes re¯ected activation of central or peripheral mechanisms was dif®cult to know. Using Suzue's in vitro neonatal rat preparation [21] which circumvented peripheral information, SP respiratory effects were tested on an isolated respiratory network able to generate a phrenic respiratory-like activity for several hours. In neonatal rats, SP was shown both to exert a central facilitatory effect on RF via medullary tachykinin NK1 receptors and to enhance phrenic discharge amplitude via cervical tachykinin NK1 receptors [17,23]. The aim of the present work was to know whether the SP effects reported in neonatal rats * Corresponding author. Tel.: 133-491-164650; fax: 133-491775084. E-mail address: [email protected] (G. Hilaire)

existed in respiratory systems of other species of rodents such as mice. Therefore, we conducted in neonatal mice some of the in vitro experiments previously performed in neonatal rats, and our results con®rmed the presence of a SP-induced increase in phrenic burst amplitude but questioned the SP-induced facilitation of the respiratory rhythm generator. In vitro experiments were carried out on neonatal mice aged from postnatal day 0±5 which originated from three different strains, the OF1 and C3H wild lines and the transgenic Tg8 line, created from the C3H line by disrupting the gene encoding mono amine oxidase-A, the enzyme which degrades serotonin [3]. Brie¯y, the medulla and spinal cord were isolated and superfused with arti®cial cerebro-spinal ¯uid (aCSF, in mM: NaCl 129, KCl 3.35, CaCl2 1.26, MgCl2 1.15, NaHCO3 21.0, NaH2PO4 0.58, glucose 30.0; pH 7.4, gassed with 95% O2 ± 5% CO2, and maintained at 27 ^ 18C). As reported elsewhere [9,10], elimination of the pons is required in mice to obtain respiratory activity in vitro which was recorded on the C4 phrenic roots with suction electrodes, ®ltered (5±3000 Hz), ampli®ed (5 K) and fed to leaky integrator (time constant 50 ms). As in the neonatal rat study [17], the control RF and the control amplitude of the integrated phrenic bursts were measured under normal aCSF; thereafter, the preparations were superfused by aCSF containing drug (SP or the tachykinin NK1 agonist [Sar 9,Met(O2) 11] substance P from Sigma) for 5 min and the resulting values of RF and phrenic burst amplitude

0304-3940/99/$ - see front matter q 1999 Elsevier Science Ireland Ltd. All rights reserved. PII: S03 04 - 394 0( 9 9) 00 28 9- X

190

K. Ptak, G. Hilaire / Neuroscience Letters 266 (1999) 189±192

Fig. 1. Effect of aCSF containing SP at 10 26 M on the in vitro respiratory activity recorded in medulla-spinal cord preparations from neonatal mice. Traces represent the integrated phrenic nerve discharges recorded in in vitro medulla-spinal cord preparations of neonatal OF1 (A) and Tg8 (B) mice under normal aCSF (A1 and B1) and under aCSF containing SP at 10 26 M (A2 and B2). (A) in OF1 mice, SP at 10 26 M increased both RF and phrenic burst amplitude (note the tonic discharge in A2). (B) in Tg8 mice, SP at 10 26 M did not affect RF but increased phrenic burst amplitude (note the weak tonic discharge in B2).

were expressed as percent of the control values (100%). Experiments were repeated on several preparations with a standardized procedure to evaluate any changes in the RF and the phrenic burst amplitude (mean ^ SEM), which were taken to be signi®cant at P , 0:05 (paired t-test). The mean control RF was signi®cantly more elevated in C3H (9:0 ^ 0:7 c.min 21, n ˆ 19) and Tg8 (9:1 ^ 0:8, n ˆ 9) than in OF1 preparations (7:2 ^ 0:6 c/min 21, n ˆ 23). In C3H and OF1 preparations, aCSF containing SP at concentration of 10 27 M signi®cantly increased phrenic burst amplitude to 135 ^ 13% of the control in both line (no difference between C3H and OF1 lines), with a slight tonic activity elicited on cervical roots in 75% of the experiments, but did not change signi®cantly RF unless SP was applied at ten times larger concentrations (Fig. 1A). When the tachykinin NK1 receptor agonist was applied at 10 27 M, phrenic burst amplitude increased to 138 ^ 18% of control

in both line whereas no signi®cant effects were elicited on RF. However, RF was signi®cantly increased when the agonist was applied at 10 26 M (SP and NK1 agonist effects on RF were not signi®cantly different, see Table 1). In three further experiments, SP 10 26 was applied either to the medulla or to the cervical cord of C3H preparations (double bath experiments). As previously shown in rats [17], applying SP to the medulla (normal aCSF to the cervical cord) increased RF whereas applying SP to the cervical cord (normal aCSF to the medulla) increased phrenic burst amplitude. In Tg8 preparations, neither SP nor the tachykinin NK1 receptor agonist signi®cantly increased RF for all the tested concentrations whereas they both signi®cantly increased phrenic burst amplitude to 119 ^ 5% (SP at 10 26 M) and to 138 ^ 3% (NK1 receptor agonist at 10 27 M). Our results in neonatal mice from OF1 and C3H lines are

Table 1 Respiratory Frequency, (RF, mean ^ SEM, % of control) measured in medulla-spinal cord neonatal preparations of OF1, C3H and Tg8 mice superfused with aCSF containing either SP or [Sar 9,Met(O2) 11] substance P ([SarMet]SP) concentrations of 10 27 M and 10 26 M a RF (% of control)

SP 10 27

SP 10 26

[SarMet]SP 10 27

[SarMet]SP 10 26

OF1 mice

106 ^ 9 (n ˆ 10) ns 92 ^ 8 (n ˆ 3) ns

148 ^ 17 (n ˆ 7) * 121 ^ 9 (n ˆ 12) * 96 ^ 16 (n ˆ 4) ns

97 ^ 5 (n ˆ 5) ns 100 ^ 25 (n ˆ 4) ns 94 ^ 5 (n ˆ 4) ns

142 ^ 11 (n ˆ 3) * 128 ^ 16 (n ˆ 3) * 106 ^ 5 (n ˆ 4) ns

C3H mice Tg8 mice

a

Not tested

ns, non signi®cant; *signi®cant changes (P , 0:05) versus control, respectively.

K. Ptak, G. Hilaire / Neuroscience Letters 266 (1999) 189±192

only partly in agreement with results gained in neonatal rats [17,23] where SP exerts a potent facilitatory effect on respiratory activity by acting on both the medullary respiratory rhythm generator (RF changes) and the cervical phrenic motoneurons (phrenic amplitude changes). Both SP and NK1 receptor agonist at 10 27 M increase phrenic burst amplitude in mice as well as in rats, but their effects on RF seem different in the two species. Whereas weak SP concentrations signi®cantly increase RF in rats (20, 70 and 90% of increase with SP at 10 29, 10 28, and 10 27 M, respectively, from [17]), SP has to be applied at 10 26 M to signi®cantly increase RF in mice. The speci®city of the RF effect obtained with SP at 10 26 M in mice may be even questioned since we did not test the effect of SP antagonist. In addition, the resulting RF increase is weaker than that observed at 10 28 M in rats. Therefore, the functional significance of the SP effects on RF in mice must be thoroughly discussed in view of understanding why RF changes are weak and why large doses of SP have to be used in mice. SP has been shown to affect RF by acting directly on some respiratory neurons of the rostral ventrolateral medulla [23], present both in rat and mouse neonates, and which are thought to constitute the primary respiratory rhythm generator (for a review see [8]). It has been ®rst reported in rats [23] that SP effects on RF are strongly facilitatory in preparations having a low resting RF (2±4 c/min 21), but are only weakly facilitatory, or even inhibitory, in preparations having an elevated RF (around 10 c/ min 21). The existence of SP inhibitory effect on RF, checked again in rats by others, was denied [17]. In mice, it is noteworthy that the SP-induced RF facilitation is strong in mouse strain having low RF, i.e. the OF1strain. Comparing rat and mouse results reveals that the in vitro control RF is lower in rats than in mice (around 4 against 7±9 c/min 21). The RF difference may arise from technical reasons since the pontine area is retained only in rat preparations. In both species, the respiratory rhythm generator is inhibited by A5 pontine structures [8,9], but the inhibition is so strong in mice that elimination of the pons is required to obtain respiratory activity in vitro [9], leading therefore to a higher RF in mouse preparations (pons transected) than in rat preparations (pons retained). Thus, different control RFs between mice and rats can, at least partly, explain the quantitative difference in SP-induced RF changes. Moreover, when the pons is transected in rat preparations, RF increases to around 10 c/min 21, and application of SP at 10 27 M weakly but signi®cantly increased RF (see Fig. 2 of [17]). Since SP at 10 27 M still increases RF in rat preparations after elimination of the pons, why a 10 times greater SP concentration is needed in mice to affect RF remains unanswered. The hypothesis that endogenous protease activity is higher in mice than in rats may be put forward; in this view, a potent degradation of exogenous SP by endogenous protease in mice could explain that large concentrations of SP have to be used. It is occasionally believed that larger doses of drugs and longer duration of application are needed

191

in smaller animals but this assumption is not fully satisfactory since; (i) it seems easier for a drug to reach its target cells in a small-sized preparation by passive diffusion, and (ii) SP at 10 27 M increases mouse phrenic burst amplitude (cervical effect) but not frequency (medullary effect). Another possible explanation is that RF regulation by SP is not yet fully mature at birth in mice; comparing SPinduced RF changes in mice at postnatal days 0±2 and 3± 5 did not reveal any age-related differences, however. Thus, it sounds possible that interspecies differences exist between respiratory regulation by SP at birth in mice and rats: RF may be regulated by SP in neonates of both species, but this process is less sensitive in mice and requires large amounts of SP to be turned on. Finally, we have shown in neonates of the transgenic Tg8 line that SP increases phrenic burst amplitude but has no effect on RF, even at 10 26 M. The Tg8 line was selected because disruption of the gene encoding monoamine oxidase-A resulted in very elevated 5-HT endogenous levels [3,14] which have disturbed the maturation of the central respiratory network and its activity at birth [1,2]. The lack of effect of SP on RF in Tg8 neonates might be due to their elevated serotonin levels [3,14] since serotonin and SP mutually reverse their excitatory effects on the nucleus tractus solitarius neurons [12]. This is one of the most simple explanations and it cannot be excluded that serotonin excess during the Tg8 prenatal period [3,14] has disturbed the maturation of the mechanisms through which SP regulates respiratory activity. The point of interest is that SP does not facilitate RF in Tg8 neonates without drastic consequences for their survival at birth; this may suggest either that the role of SP in respiratory activity has been over-estimated or that some compensatory mechanisms have been developed in Tg8 mice. To conclude, the role of SP in respiratory control appears different in neonatal mice and rats: SP activates the phrenic discharge in both species but abnormally large concentrations of SP have to be used to slightly affect RF in mice, suggesting a possible interspecies difference in respiratory regulation mechanisms at birth. The authors gratefully acknowledge Edward De Maeyer and Isabelle Seif (UMR CNRS 146, Orsay, France) for their gift of the C3H and Tg8 mice and Michelle Bevengut (UPR CNRS 9011, Marseille, France) for her comments on the manuscript. They thank Marie Gardette (UPR CNRS 9011, Marseille, France) for the iconography. [1] Bou, C., Lajard, A., Seif, I., De Maeyer, E., Monteau, R. and Hilaire, G., Serotonin and maturation of the medullary respiratory network in the neonatal mouse. Eur. J. Neurosci., 10 (S10) (1998) 152.09. [2] Bou, C., Lajard, A., Seif, I., De Maeyer, E., Monteau, R. and Hilaire, G., Serotonin and maturation of the neonatal mouse phrenic motoneurons. Eur. J. Neurosci., 10 (S10) (1998) 152.10. [3] Cases, O., Seif, I., Grimsby, J., Gaspar, P., Chen, K., Pournin,

192

[4]

[5] [6] [7]

[8] [9] [10] [11]

[12]

[13]

K. Ptak, G. Hilaire / Neuroscience Letters 266 (1999) 189±192 S., Muller, U., Aguet, M., Babinet, C., Chen Shih, J. and De Maeyer, E., Aggressive behavior and altered amounts of brain serotonin and norepinephrine in mice lacking MAOA. Science, 208 (1995) 1763±1766. Chen, Z., Hedner, J. and Hedner, T., Hypoventilation and apnea induced by the susbtance P antagonist D-Pro2, DTrp7±9 in the ventrolateral medulla. Acta Physiol. Scand., 134 (1988) 153±163. Chen, Z., Hedner, J. and Hedner, T., Substance P in the ventrolateral medulla oblongata regulates ventilatory responses. J. Appl. Physiol., 68 (1990) 2631±2639. Chen, Z., Hedner, J. and Hedner, T., Local effects of Substance P on respiratory regulation in the rat medulla oblongata. J. Appl. Physiol., 68 (1990) 693±703. Gillis, R., Helke, C., Hamilton, B., Norman, N. and Jacobowitz, D., Evidence that substance P is a neurotransmitter of baro- and chemoreceptor afferents in nucleus tractus solitarius. Brain Res., 181 (1980) 476±481. Hilaire, G. and Duron, B., Maturation of the mammalian respiratory system. Physiol. Rev., (1999) in press. Hilaire, G., Bou, C. and Monteau, R., Rostral ventrolateral medulla and respiratory rhytmogenesis in mice. Neurosci. Lett., 224 (1997) 13±16. Hilaire, G., Bou, C. and Monteau, R., Serotonergic modulation of central respiratory activity in the neonatal mouse: an in vitro study. Eur. J. Pharmacol., 329 (1997) 115±120. Holtman, J., Immunohistochemical localisation of serotonin and substance P containing ®bers around respiratory muscle motoneurons in the nucleus ambiguus of the cat. Neuroscience, 26 (1988) 169±178. Jacquin, T., Denavit-saubieÂ, M. and Champagnat, J., Substance P and serotonin mutually reverse their excitatory effects in the rat nucleus tractus solitarius. Brain Res., 502 (1989) 214±222. Kalia, M., Fuxe, K., Hok¯et, T., Johansson, O., Lang, R., Ganten, D., Cuello, C. and Terenius, L., Distribution of neuropeptide immunoreactive nerve terminals within the subnuclei of the tractus solitarius of the rat. J. Comp. Neurol., 222 (1984) 409±444.

[14] Lajard, A., Bou, C., Monteau, R. and Hilaire, G., Serotonin levels are abnormally elevated in the fetus of the MAOAde®cient transgenic mouse. Neurosci. Lett., 261 (1999) 41± 44. [15] Liebstein, A., Dermietzel, R., Willenberg, I. and Pauschert, R., Mapping of different neuropeptides in the lower brainstem of the rat with special reference to the ventral surface. J. Autonom. Nerv. Syst., 14 (1985) 299±313. [16] Lindefors, N., Yamamoto, Y., Pantaleo, T., Lagercrantz, H., Brodin, E. and Ungerstedt, U., In vivo release of substance P in the nucleus tractus solitarii increases during hypoxia. Neurosci. Lett., 69 (1986) 94±97. [17] Monteau, R., Ptak, K., Broquaire, N. and Hilaire, G., Tachykinins and central respiratory activity: an in vitro study on the newborn rat. Eur. J. Pharmacol., 314 (1996) 41±50. [18] Morin-Surun, M-P., Jordan, D., Champagnat, J., Spyer, K. and Denavit-SaubieÂ, M., Excitatory effects of iontophoretically applied substance P on neurons in the nucleus solitarius of the cat: lack of interaction with opiates and opioids. Brain Res., 307 (1984) 388±398. [19] Prabhakar, N., Runold, M., Yamamoto, Y., Lagercrantz, H. and Von Euler, C., Effects of substance P antagonist on the hypoxia-induced carotid chemoreceptor activity. Acta Physiol. Scand., 121 (1984) 301±303. [20] Rampin, O., Pierre®che, O. and Denavit-SaubieÂ, M., Effects of serotonin and substance P on bulbar respiratory neurons in vivo. Brain Res., 622 (1993) 185±193. [21] Suzue, T., Respiratory rhythm generation in the in vitro brain stem spinal cord preparation of the neonatal rat. J. Physiol., 354 (1984) 173±183. [22] Yamamoto, Y. and Lagercrantz, H., Some effects of substance P on central respiratory control in rabbit pups. Acta Physiol. Scand., 124 (1985) 449±455. [23] Yamamoto, Y., Onimaru, H. and Homma, I., Effects of substance P on respiratory rhythm and pre-inspiratory neurons in the ventrolateral structure of rostral medulla oblongata: an in vitro study. Brain Res., 599 (1992) 272± 274.