Morphine is the dominating analgetic drug used in neonates, but opioid-induced respiratory depression limits its therapeutic use. In this study, we examined acute morphine effects on respiration during intermittent hypoxia in newborn Tac1 gene knockout mice (Tac1−/−) lacking substance P and neurokinin A. In vivo, plethysmography revealed a blunted hypoxic ventilatory response (HVR) in Tac1−/− mice. Morphine (10 mg/kg) depressed the HVR in wild-type animals through an effect on respiratory frequency, whereas it increased tidal volumes in Tac1−/− during hypoxia, resulting in increased minute ventilation. Apneas were reduced during the first hypoxic episode in both morphine-exposed groups, but were restored subsequently in Tac1−/− mice. Morphine did not affect ventilation or apnea prevalence during baseline conditions. In vitro, morphine (50 nM) had no impact on anoxic response of brain stem preparations of either strain. In contrast, it suppressed the inspiratory rhythm during normoxia and potentiated development of posthypoxic neuronal arrest, especially in Tac1−/−. Thus this phenotype has a higher sensitivity to the depressive effects of morphine on inspiratory rhythm generation, but morphine does not modify the reactivity to oxygen deprivation. In conclusion, although Tac1−/− mice are similar to wild-type animals during normoxia, they differed by displaying a reversed pattern with an improved HVR during intermittent hypoxia both in vivo and in vitro. These data suggest that opioids and the substance P-ergic system interact in the HVR, and that reducing the activity in the tachykinin system may alter the respiratory effects of opioid treatment in newborns.
- substance P
morphine is the most widely used analgetic and sedative drug in neonatal and pediatric intensive care units during painful procedures, postoperatively, and during mechanical ventilation. Its use, and thereby its effectiveness, is, however, often limited by its most common side effect, respiratory depression. The mechanism behind this depression is not fully understood, but is considered to be an effect on neural respiratory networks in the brain stem and to be mainly mediated via μ-opioid receptors (MORs) (12, 20, 53, 60). Neurons in the ventral respiratory group that express MORs also express neurokinin-1 (NK1) receptors (30), the preferred receptor of substance P (SP), suggesting a possible interaction between these systems. Since both endogenous opioids (enkephalins, endomorphins, and endorphins) and endogenous SP are released on hypoxic activation (40), these peptidergic systems may interact in the regulation of the hypoxic response.
The involvement of SP in respiratory control and the hypoxic ventilatory response (HVR) has been described previously both in vivo and in vitro (14, 26, 36, 56, 68, 69), and studies have indicated that SP may antagonize the respiratory depressive effect of morphine (59). In previous work from our laboratory, we used a strain of mice with a targeted deletion of tachykinin gene 1 (Tac1−/−) to study the effects of constitutionally lacking SP and neurokinin A (NKA) in the adult period (6). Although morphine depresses basal ventilation in these transgenic animals, intermittent hypoxia causes a relatively enhanced HVR in vivo in Tac1−/− compared with wild-type (WT) animals (7). This confirms previous findings of a complex regulatory system of respiration and that hypoxic pathways are partly different from those at work under baseline conditions.
The possibility of an antagonizing effect on the respiratory depressive effect of morphine is of great clinical interest, and since the respiratory system undergoes extensive changes during the neonatal period, and SP levels in the brain stem reach their maximum during the second postnatal week in rodents (48, 52, 67, 69), it may be of even greater importance at this age. Likewise, it is well known that neonates have an increased sensitivity to opioids. We, therefore, wanted to investigate the effects of intermittent hypoxia/anoxia in morphine-exposed newborn mice, lacking SP and NKA in an in vivo/in vitro model. Due to this increased importance of the tachykinin system in newborn mammals, we hypothesized that Tac1−/− mice would have an altered HVR following opioid exposure.
The generation of mice with a targeted mutation of the Tac1 gene has been described previously (72). Newborn mice (C57BL/6) of both sexes at postnatal day 2 (i.e., 48–72 h after birth) were divided in four groups: WT as controls (B&K Universal, Sollentuna, Sweden) and Tac1−/− mice (provided by Dr. Zimmer, University of Bonn, Bonn, Germany). Both groups were then subdivided according to their morphine or sodium chloride exposure. Tac1−/− animals have been crossed more than 10 generations to C57BL/6J mice and are, therefore, considered congenic for this genetic background. To avoid genetic drift, we restart the homozygous breeding colonies after each fourth generation using heterozygous breeding pairs received from back-crossing of animals from the knockout line with mice from the original background strain. Animals were born naturally and were kept on regular 12:12-h light-dark cycle under standardized conditions with food and water ad libitum, in accordance with the European Community guidelines approved by the local ethics committee, i.e., Stockholms Norra Djurförsöksetiska nämnd (Dnr N69/07).
Morphine hydrochloride (Meda AB, Solna, Sweden) was dissolved in sterile 0.9% saline and administered in a volume of 10 ml/kg intraperitoneally (ip) in in vivo experiments. In the sodium chloride group, only saline 0.9% was administered. The dose regimen was chosen after pilot experiments and represents a rather high dose to induce a respiratory depression. A dose of 2.5 mg/kg subcutaneously is considered a normal analgetic dose (66) and did not affect basal respiration or HVR in our experiment (data not shown) when given ip. When given 10 mg/kg ip, basal respiration was intact, but an effect on respiration during hypoxia was observed. Due to first-pass hepatic metabolism when given ip, plasma levels are likely to be one-third of that seen after a normal analgetic dose (18), and we consider it a clinically relevant dose. In vitro, the en bloc preparations were superfused with 10 nM, 50 nM, 200 nM, or 1 μM morphine hydrochloride dissolved in control artificial cerebrospinal fluid (ACSF). The effect of SP (Sigma-Aldrich AB, Stockholm, Sweden), 50 nM in ACSF, was studied in separate sessions.
In Vivo Protocols and Ventilatory Measurements
Ventilatory measurements were made using dual-chamber plethysmography (13). A mask covering the mouth and nostrils was affixed with dental impression material (Impregum F, 3M ESPE). The animals then breathed from one chamber (the headbox), while pressure changes are measured in a second, volume-calibrated, chamber (35 ml, MTA, Karolinska University Hospital, Sweden) using a pneumotach (Hans Rudolph). Humidified air or a hypoxic gas mixture (8% O2 in N2) was administered to the headbox by manually switching between gas reservoirs. Gas flow through the headbox was maintained at 50 ml/min, resulting in a complete change of gases within 15 s. The signal was digitally converted with a sampling rate of 200 Hz and analyzed using Power Lab software (Chart version 5.5.4, PowerLab Systems, AD Instruments, Colorado Springs, CO). The obtained values were then used to calculate respiratory frequency (f; breaths/min), inspiration time (Ti), expiration time (Te), tidal volume (Vt), and minute ventilation (V̇e). Vt and V̇e were divided by body weight and expressed as milliliters per gram and milliliters per gram per minute, respectively. All studies were performed between 0900 and 1500 to control for possible daily fluctuations in respiratory patterns. The ambient temperature within the chamber was measured continuously using a digital thermometer (model BAT-12, Physiotemp Instruments) and recorded in parallel with the flow signal. It was maintained at 33 ± 0.5°C in accordance with the thermoneutral range for mice of similar age by immersing the chamber in a thermostat-controlled water bath. The temperature in the mouth was controlled as the animal was placed into the box, before and after respiratory recordings were performed.
Animals received an ip injection of either morphine hydrochloride 10 mg/kg or saline at a volume of 10 ml/kg and were kept for 25 min in the chamber in room air for acclimatization before the recording started. During this time span, serum levels of morphine have peaked (∼10 min postinjection) (18). After injection and initial acclimatization, all animals were exposed to normoxic air for 3 min while baseline values were collected, followed by three episodes of intermittent hypoxia (8% O2 in N2). This ventilatory protocol was chosen to enable comparison with previous studies in this transgenic strain and as repeated hypoxic exposures has been described to reflect the development of long-term facilitation (47). Each hypoxic period lasted 5 min, followed by a normoxic recovery period of 2 min. Baseline values of the respiratory parameters were calculated as a mean of the last 3 min of normoxic conditions. During hypoxic and posthypoxic periods, values of the respiratory parameters were calculated as the mean of the first and the second half of the corresponding interval. The number of apneas (defined as ventilatory pauses longer than twice the duration of the preceding breath) was calculated manually during the entire intervals of basic, hypoxic, and posthypoxic conditions and is presented as mean value per minute. Animals were weighed after the final recording. The number of animals in each group was n = 6 (saline groups) and n = 5 (morphine groups), coming from at least two different litters.
Brain Stem-Spinal Cord Preparation
Whereas in vivo measurements also reflect the influence of peripheral factors, experiments on isolated central structures offer insight into central regulation. Experiments were, therefore, performed on the brain stem-spinal cord preparation of postnatal day 2 pups. The isolated brain stem and cervical spinal cord were obtained under isoflurane anesthesia, as described previously (44). The isolated preparation was continuously perfused at the rate of 3.5–4.5 ml/min in a 2-ml chamber with a solution containing (in mM) 130 NaCl, 5.4 KCl, 0.8 KH2PO4, 0.8 CaCl2, 1 MgSO4, 26 NaHCO3, and 30 glucose and equilibrated with 95% O2 and 5% CO2 at 28°C; pH was set at 7.4 (control ACSF).
In Vitro Protocols, Data Recording, and Analysis
Inspiratory discharges of respiratory motoneurons were monitored by extracellular recording with glass suction electrodes applied to the proximal cut end of C4 and C3 ventral roots of spinal nerves. Burst frequency was analyzed and calculated as the number of C4 bursts per minute. After the preparation was superfused with control ACSF for 40 min and C4 activity reached a steady state, the control perfusate was replaced by testing solutions. There were four experimental protocols. In protocol 1 and to study the effect of morphine concentration, control ACSF was replaced for 30 min by solutions equilibrated with 95% O2 and 5% CO2 containing either 10 nM, 50 nM, 200 nM, or 1 μM of morphine hydrochloride. In protocol 2, the intermittent anoxia consisting of three 3-min intervals of anoxia separated by 5 min of normoxia was applied to the preparation. The anoxic solution was ACSF equilibrated with 95% N2 and 5% CO2. The last anoxic episode was followed by a 40-min interval of washing out with control ACSF. In protocol 3, anoxic responses to intermittent anoxia 15 min after morphine application were studied. In protocol 4, the effect of 50 nM SP was investigated in separate sessions. After the preparations were superfused with control ACSF for 40 min, SP was bath applied for 3 min.
The control values of the inspiratory burst frequency were calculated during application of normoxic ACSF as the mean of the last 3 min before testing a modified ACSF application. During morphine application, the burst frequency was calculated at 12 and 27 min as the mean of the following 3 min of modified ACSF application. During anoxia and SP application, the burst frequency was calculated as the mean over 1 min when the maximal effect was observed.
Axoscope software and Digidata 1200B interface (Axon Instruments, Foster, CA) were used to collect electrophysiological data. Offline analysis was performed employing Clampfit 8.02 (Axon Instruments, Foster, CA), DATAPAC 2K2 (RUN Technologies, Laguna Hills, CA), and Origin 6.0 (Microcal Software, Northampton, MA) software for PC.
Statistical analysis of paired comparisons was performed by Student's t-test. For multiple statistical comparisons, two- and three-way ANOVA (Statistica software, Statsoft Scandinavia, Uppsala, Sweden) designs for repeated measurements were used where appropriate. Post hoc analysis was conducted by Tukey's test. Values are shown as means ± SE. Unless stated otherwise, P values refer to three-way ANOVA. Significance level was set at P < 0.05.
WT and Tac−/− mice were born naturally and did not differ in body weight [1.52 ± 0.04 g in WT (n = 39) mice and 1.52 ± 0.05 g in Tac −/− (n = 29)]. During normoxia, there were no differences in respiratory parameters (f, Vt, V̇e) between groups (Table 1).
During the first hypoxic period, WT animals displayed a normal HVR with an increase from baseline in f (61.2 ± 12.7%) and in V̇e (133 ± 39.5%, P < 0.05 for both parameters, t-test), but without significant Vt changes (40.2 ± 18.9%). During subsequent hypoxic challenges, the normal biphasic pattern with a secondary decline in f was replaced by a further increase, which contributed to a significant increase in V̇e (P < 0.05). In contrast, Tac1−/− mice displayed an attenuated HVR with no significant increase from baseline in either f (18.6 ± 7.2%), Vt (19.5 ± 11.8%), or V̇e (32 ± 22.5%, Figs. 1 and 2). There were no differences in Ti and Te between strains during normoxia, hypoxia, and posthypoxia (Table 2).
The given morphine dose did not cause any significant changes in respiratory parameters during normoxic conditions in either experimental group. However, morphine administration did affect respiratory parameters during hypoxic conditions in both strains, albeit with a different pattern in WT and Tac1−/− mice. In WT animals during hypoxic challenge, morphine administration decreased net ventilation, i.e., V̇e (20.2 ± 6.0%) compared with nonexposed WT animals (P < 0.01). This effect was mainly due to a reduction of f (38.3 ± 2.8%, P < 0.01), whereas no significant changes was observed in Vt (36.1 ± 16.4%).
In Tac1−/− animals, the already attenuated f response to hypoxia was not affected by morphine during hypoxia or reoxygenation. Ti was increased during hypoxia, but Te was not significantly affected, which contributed to more stable frequency response in Tac1−/− mice (Table 2). An increase in Vt remained in Tac1−/− mice after morphine exposure during hypoxia compared with saline group (70.3 ± 10.1% during the second hypoxic period, P < 0.01). These changes resulted in a gradual reinforcement of V̇e in Tac1−/− mice, which became significant during the third hypoxic period (32.5 ± 9.22%, P < 0.05), and which was opposite to the depressed ventilation observed in WT animals.
Following hypoxic episodes, a decrease in f was seen in both strains (35.7 ± 8.1% in WT and 39.9 ± 7.4% in Tac−/− mice, P < 0.01, respectively). Such a posthypoxic frequency decline was seen in WT animals also following morphine exposure (50.1 ± 7.2%, P < 0.01), whereas, in Tac1−/− mice, it was considerably attenuated (22.6 ± 9.3%, nonsignificant).
Thus, although V̇e during hypoxia was considerably greater in WT intact mice, after morphine administration, both strains displayed a similar net ventilation patterns, reaching the same V̇e levels (P > 0.05, Fig. 1). WT and Tac1−/− mice demonstrated similar frequencies of apneas during normoxic conditions, but there were significant differences between strains during hypoxia, displayed as a reduction of the number of apneas in WT mice (P < 0.0001), whereas no change was seen in Tac1−/− mice (Fig. 3). In normoxic conditions, morphine administration did not affect apnea prevalence in either strain. However, during intermittent hypoxia, morphine decreased apnea prevalence, both in WT and Tac1−/− mice compared with controls (P < 0.01, respectively). Apneas remained low in WT mice during subsequent hypoxic challenges, whereas Tac1−/− mice instead displayed a significant increase compared with WT during each hypoxic challenge, reaching baseline levels (P < 0.001, hypoxic periods 2 and 3, Fig. 3). During posthypoxic or reoxygenation periods, all four experimental groups of mice demonstrated severalfold increase in the apnea prevalence, with a tendency to increase more in WT compared with Tac1−/− mice (498 ± 79.7 vs. 307 ± 71.9%, P = 0.08).
Effects of morphine.
In control (normoxic) conditions, the basic burst frequency was similar in both experimental groups: 6.76 ± 0.17 bursts/min (n = 30) in WT vs. 6.31 ± 0.29 bursts/min (n = 26) in Tac1−/−. The effect of acute morphine administration was different in WT and Tac1−/− preparations, with the exception of the effect of 10 μM, which, within 1 min, completely suppressed the inspiratory rhythm in both strains. Administration of ACSF with 1 μM of morphine quickly suppressed the bursting in all Tac1−/− preparations (n = 5), while, in WT preparations (n = 4), the inspiratory rhythm was suppressed in only two preparations, whereas, in the other two, bursting gradually declined during the 30 min of morphine application. Effects of lower morphine concentrations are shown in Table 3. The suppression of burst frequency in Tac1−/− preparation was greater compared with WT (P < 0.001, two-way ANOVA). Since 10 nM of morphine did not affect the respiratory rhythm in WT preparations, 50 nM, which induced mild frequency reduction in both strains, was chosen to study the effects of intermittent anoxia on inspiratory burst frequency (Fig. 4).
Effects of morphine on changes in inspiratory rhythm during intermittent anoxia.
Fifteen minutes after morphine application, the frequency was significantly more depressed in Tac1−/− mouse preparations than in WT and was similar to the changes observed in 30 min shown in Table 3. In both Tac1−/− and WT mouse brain stem preparations, the anoxic response was characterized by an initial transient increase in burst frequency and a reduction in burst amplitude, with less attenuation in Tac1−/− preparations (Fig. 4C), as shown earlier (6).
There were no differences between WT (n = 8) and Tac1−/− (n = 7) preparations in anoxic frequency response, and frequency depression induced by morphine did not affect the anoxic frequency response in either group (Fig. 4, B and D). Resumption of oxygenation induced posthypoxic neural arrest (PHNA) in WT and Tac1−/− preparations without morphine treatment. The PNHA appeared significantly more often in Tac1−/− (100%, 7/7) than in WT preparations (37.5%, 3/8, z-test, P < 0.05). After morphine treatment, the PNHA was seen in 100% of the preparations, in both WT (n = 12) and Tac1−/− (n = 9) mice (Fig. 4A). Moreover, in 7/9 Tac1−/− preparations (78%), the inspiratory burst rhythm did not recover between anoxic periods, in contrast to WT preparations, where the bursting always recovered during postanoxic period (z-test, p < 0.01).
Effects of SP administration on inspiratory rhythm.
SP administration for 3 min had excitatory effects on bursting activity in both strains, but with a significantly greater frequency increase in Tac1−/− animals. Control frequency values were similar in Tac1−/− (5.46 ± 0.21 bursts/min, n = 4) and WT group (5.62 ± 0.45 bursts/min, n = 4), but, following SP application, the frequency increased in Tac1−/− to 17.25 ± 0.25 bursts/min (215 ± 0.24%) vs. 11.5 ± 1.89 bursts/min (115 ± 0.33%) in WT (P < 0.05, two-way ANOVA).
In the present study, we demonstrate that morphine modifies the HVR differently in newborn mice with a hereditary deficiency in the tachykinin SP/NKA system. When exposed to morphine, Tac1−/− mice display an enhanced HVR, whereas, in WT mice, the response is severely suppressed by morphine. Tac1−/− mice also display a higher sensitivity to the depressive effects of morphine in respiratory rhythm generation mechanisms, but no impact on the central reactivity to oxygen deprivation during severe anoxia.
As seen in previous studies, Tac1−/− neonatal mice displayed a normal respiratory pattern during normoxic conditions in vivo, but failed to increase their respiration adequately in response to hypoxia (6). The purpose of this study was to further examine the control of respiration and investigate whether the tachykinin system is of importance when respiration is suppressed by opioids. The chosen dose of morphine (10 mg/kg), corresponding to a high postoperative analgetic treatment for rodents, had no effect on the respiration at rest. Morphine-exposed WT animals displayed an attenuated HVR, which was mainly due to failure to increase their f. This pattern strikingly resembles that seen in Tac1−/− mice without morphine exposure. In contrast, morphine-exposed Tac1−/− mice demonstrated a significant and gradual increase in Vt throughout the whole hypoxic session, without any further suppression of f. These changes resulted in reinforced V̇e during intermittent hypoxia and thus restored a normal HVR, contrasting from the depressed ventilation observed in WT animals.
Our laboratory has previously demonstrated in adult Tac1−/− mice a relatively greater resistance to the inhibitory effects on ventilation exerted by morphine, compared with WT animals. In adults, that effect was mainly due to a lesser suppression of the f (7), thereby indicating a developmental change in the morphine reactivity in Tac1−/− mice. Contrasting from the biphasic HVR reported in the newborn (38), we registered a sustained increase in ventilation in WT animals, a pattern that was blunted in Tac1−/− animals. As suggested also for Swiss CD-1 and C57BL/6 mice (8, 54), a sustained increase in ventilation during hypoxia in neonatal mice could reflect the early development of chemosensitive reflexes and pathways, but also a diminished reduction in metabolism in response to acute hypoxia (8, 38, 54). The decrease in f following hypoxia seen in both strains is similar to the posthypoxic frequency decline previously described in adult rats (10) and mice (17, 24). This may be of particular interest, since the response requires functional integrity of the ventrolateral pons, an area rich in receptors for both tachykinins and opioids. The frequency decline appears to be the result of a hypoxia-induced central inhibition of respiration, as described previously in neonatal Swiss CD-1 mice (48).
Adult C57BL/6J mice are known to exhibit spontaneous central apneas, which may be related to perturbations within the pontomedullary central pattern generator (58) and which involves tachykinin and opioid regulation. We could also demonstrate spontaneous apneas in neonatal mice and assessed respiratory stability, analyzing the frequency of apneas during hypoxia. Again, no effects by morphine were seen during normoxia, illustrating how important physiological effects on respiration may be impossible to discern under basal conditions. During hypoxia, a reduced number of apneas were seen in WT animals, which is the normal reaction to hypoxia in neonates, shown also in Swiss CD-1 mice and C57BL/6 (6, 54). In contrast, the prevalence of apneas was unaffected by hypoxia in transgenic mice, but morphine administration considerably reduced it to the range of values shown in WT animals. Thereafter, levels gradually increased during the following hypoxic challenges, returning to levels seen in Tac1−/− animals not treated with morphine. The particular ventilatory mechanism underlying the changes in respiration in Tac1−/− mice and modulatory effects of morphine is not clear. Since morphine did not affect the respiratory parameters at rest in either strain, we assume that O2 saturation level was similar (39) and CO2 apnea threshold was not affected either (31, 65). However, during intermittent hypoxia in two strains, gradual decreasing/increasing of partial O2 pressure could affect peripheral CO2 chemosensitivity and cerebrovascular CO2 reactivity, which, in turn, could affect the dynamics of the hypoxic response (1). This issue remains to be further investigated.
During posthypoxic/recovery periods Tac1−/− mice displayed rather strong inclination to less apneas, simultaneously with gradual ventilation enhancement with repeated hypoxic exposure. Our data suggest a restoration of respiration during hypoxia in mice lacking SP/NKA when the opioid system is activated. This is in accordance with previous observations that adult Tac1−/− mice experience less severe posthypoxic depression (6).
In vitro data showed that, during normoxic conditions, morphine induced a considerable reduction in burst frequency with skipped bursts during some inspiratory cycles. This appeared in a concentration-dependent manner, and findings are in concordance with results showing that MOR activation affects the respiratory rhythm-generating mechanisms and slows down the inspiratory rhythm in slice and brain stem preparations in mice (15, 21), rats (16, 41, 56, 59, 60, 66), and lamprey (42). Under normoxic conditions, Tac1−/− preparations were also more sensitive to morphine administration. This higher sensitivity to morphine suggests some tonic effect of an intact tachykinin system on respiratory rhythm generation, although there is some evidence that SP is not involved in rhythm generation at rest (35). Our data are also in concordance with observations collected in vitro and in situ that endogenously released SP is involved in the maintenance of normal respiratory motor output (51).
Interestingly, anoxic stimulation partly overcame the depressive effects of morphine in both strains. Morphine application did not affect frequency and the pattern of anoxic responses in either strain, but strongly potentiated the development of PHNA and especially so in Tac1−/− mice. Our PHNA data are in line with observations that opioids affect the duration of PHNA in neonatal rats (23). Moreover, anoxic response and PHNA seem to demonstrate a dissociated behavior, suggesting that morphine and anoxia recruit different synaptic/cellular mechanisms.
It is well established that morphine, via μ-receptor Gi/o- protein suppression of the cAMP-PKA signal transduction pathway, activates a set of K+ channels. It also inhibits the activity of voltage-gated Ca+2 channels and thus pre- and postsynaptically suppresses the efficacy of neurotransmitter release and decreases the neuronal excitability (64). Acute hypoxia/anoxia seems to directly affect the chemosensitive cells within the medullary respiratory networks and initiate release of multiple neurotransmitters (5, 25, 43). Some of these are known to antagonize the suppressive effects of opioids and to stimulate the rhythmogenesis, as has been demonstrated for subset of serotonin and dopamine receptors, thyrotropin-release hormone, acetylcholine, and SP (4, 21, 30, 34, 55, 59, 66).
Our data extend previous observations (6, 50, 62) and suggest that compensatory mechanisms developing during early ontogeny can maintain a rather normal respiratory activity in ordinary conditions, but are not able to support the transient stressor states during, e.g., hypoxic challenge. The mechanism of the hypoxic response is not fully elucidated, nor is the nature of the developmental compensation, which interacts with morphine mediating the HVR in the Tac1−/− phenotype.
The higher SP sensitivity of the respiratory rhythm generator (RRG) seen in our in vitro experiments is supported by earlier observations that levels of NK1 receptor mRNA were greater in Tac1−/− mice, suggesting a compensatory increase in this receptor type. Upregulation of NK1 receptors is in concordance with the concept that G protein-coupled receptors are regulated by ligand levels, as shown for different opioid receptors in transgenic mice (9, 72). It is possible that, in this strain, tonic activation of NK1 receptors is maintained by the tachykinin peptide NKB, which is encoded by a different gene and thus still expressed (54, 63, 72). NKB can originate from the cortical level, from hypothalamus nuclei, and from medullar reticular formation (33). It is known that endogenous tachykinins can stimulate all three types of tachykinin receptors at physiological concentrations (28), and stimulation of NK1 and NK3 receptors have facilitatory effects on the RRG (35). Thus it is possible that increased NK1 receptor activity may contribute to maintain the normal respiratory function in Tac1−/− mice.
The biphasic inspiratory response seen in in vitro preparations with a similar and strong excitatory phase in both strains suggests that central oxygen sensing is not affected in Tac1−/− mice, regardless of whether they are treated with morphine or not. These data suggest that, during weak morphine-induced depression, the RRG seems to be protected against increasingly severe hypoxia. This also suggests that the activation phase of the inspiratory rhythm generator in response to acute anoxia could participate in development of hyperpnea in vivo. However, the hypoxic response in vivo was blunted in Tac1−/− mice and in neither strain displayed the depressive phase, which developed in the brain stem preparation within 1.5 min. This discrepancy between in vitro and in vivo results implies that regulation of the hypoxic response in vivo in our experimental conditions is controlled outside of the RRG and could include the dorsal respiratory group, pontine, or more rostral brain areas and the peripheral chemoreceptor and pulmonary sensory receptor regulation (32, 37, 45, 57, 68, 70).
It is well established that hyperpnea is regulated also by a sustained activation of the carotid body, and carotid body denervation considerably suppresses V̇e (11, 19, 46). This suggests that the blunted HVR in Tac1−/− mice may originate, at least partly, from the carotid body chemoreflex pathways. Augmented activity of the SP-NK1 receptor system in response to anoxia has been demonstrated for different parts of these pathways, including the carotid body, nucleus of the solitary tract (NTS), raphe nuclei, and pons, while blocking the SP-NK1 receptor activity markedly attenuated hypoxic response in vivo and in vitro (7, 26).
MORs were shown in all structures involved in rhythm and pattern generation (15, 22, 27, 29, 49, 70). In an observation in humans with intrathecal morphine administration, the authors suggest that the prolonged suppressed HVR may result from mainly a direct action of morphine on central MORs rather (3). In addition, it was recently demonstrated in rats that NTS, via MOR activation, might considerably suppress the HVR through the afferent pathways of the carotid body (71).
Both in vivo and in vitro studies have demonstrated that local MOR activation within the respiratory rhythm and pattern-generating structures, including the pre-Bötzinger complex, induce rhythm facilitation, as well as Vt increase (27, 61). Furthermore, tachykinins in the NTS were shown to play a diverse role modifying synaptic activity in different neuronal populations in vitro via the release of the glutamate or GABA (2). These experiments highlight the complexity of neuronal networks and that disruption of the normal SP/opioid interacting systems may give rise to unexpected effects. Thus the more robust and exaggerated hypoxic response seen in Tac1−/− mice could be caused by a modified balance between the opioid and tachykinin system in Tac1−/− mice and by a compensatory substitution of neuromodulators in the brain networks responsible for respiratory rhythm generation.
Although further studies are needed to elucidate the level of interactions, as well as the possible effects of changes in metabolism, the present results suggest that reducing the activity in the tachykinin system may be a possible method to minimize the life-threatening respiratory depressive effect of opioid treatment in newborns.
This study was supported by grants from the Swedish Medical Research Council (521-2009-4884), Sällskapet Barnavård, Stiftelsen Frimurare Barnhuset i Stockholm, and The Swedish Heart and Lung Foundation.
No conflicts of interest, financial or otherwise, are declared by the author(s).
Author contributions: J.B., Y.S., A.Z., and R.W. conception and design of research; J.B. and Y.S. performed experiments; J.B. and Y.S. analyzed data; J.B., Y.S., and R.W. interpreted results of experiments; J.B. and Y.S. prepared figures; J.B., Y.S., and R.W. drafted manuscript; J.B., Y.S., and R.W. edited and revised manuscript; J.B., Y.S., A.Z., and R.W. approved final version of manuscript.
- Copyright © 2012 the American Physiological Society