INTRODUCTION

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

MULTIPLE STUDIES FROM VARIOUS laboratories have shown that N-methyl-D-aspartate (NMDA) glutamate receptors within the dorsocaudal brain stem mediate important aspects of respiratory pattern generation and cardiovascular regulation and also underlie critical components of the hypoxic ventilatory response (HVR; Refs. 2, 19, 22, 28). NMDA glutamate receptors are structurally heterodimers, which are composed from NR1 and NR2 subunits (9). In hippocampal neurons, activation of the NMDA receptor involves tyrosine phosphorylation of the NR2 subunit as an early event (16). In addition, application of tyrosine kinase (TK) inhibitors will markedly attenuate NMDA receptor channel currents (7, 35, 36), whereas administration of tyrosine phosphatase (TP) inhibitors will result in enhanced neuronal discharge (35, 36). Conversely, anoxia and resultant neuronal depression in hippocampal slices elicit a decrease in TK activity, particularly of the src TK family, leading to tyrosine dephosphorylation of the NMDA NR2A/2B subunits (5). Therefore, TK-mediated pathways appear to mediate components of neuronal excitability within brain regions that do not subserve specific respiratory tasks.

In previous studies from our laboratory, we have shown that inhibition of protein kinase C (PKC) activity within the dorsocaudal brain stem resulted in significant attenuation of the HVR in conscious rats, whereas administration of membrane-permeable PKC activators induced marked ventilatory enhancements during normoxia (13, 14). However, the contribution of PKC activity changes to the HVR does not fully account for the overall increases in ventilatory output (12, 14). Thus it is conceivable that multiple kinase systems are involved in HVR. Based on the previously demonstrated modulation of NMDA receptor channel activity by TK in the hippocampus and the more recent observation that TK activity in vivo alters the baroreflex response in anesthetized rats (23), we hypothesized that administration of TK inhibitors to the dorsocaudal brain stem of adult rats would be associated with HVR attenuation. To examine this issue, normoxic and hypoxic cardioventilatory responses were assessed in chronically instrumented, freely behaving adult rats before and after administration of the selective TK inhibitors genistein (Gen) and tyrphostin A25 (A25).

	METHODS

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

Adult male Sprague-Dawley rats (280-320 g) were purchased from a commercial breeder (Charles River). The experimental protocols were approved by the Institutional Animal Use and Care Committee. Animals were provided with water and rat chow ad libitum, kept on a 12:12-h light-dark cycle, and at 22 ± 1°C ambient temperature for at least 1 wk of habituation before surgery and during the postoperative period. For habituation purposes, the animals spent at least 1-2 h each day in a whole body plethysmographic chamber.

Surgical animal preparation. Anesthesia was induced by pentobarbital sodium (50 mg/kg ip, Nembutal), and core temperature was maintained at 37.5°C via a servo-controlled Harvard rectal temperature probe connected to a warming blanket. A 1-cm incision of the left inguinal skin was performed, and an indwelling polyethylene catheter (PE-50) was inserted into the left femoral artery and advanced to the abdominal aorta for sampling of blood and measurement of cardiovascular parameters. The catheter was secured in the groin and exteriorized in the dorsal aspect of the neck. The arterial line was flushed with heparinized saline (1,000 U/ml), sealed with heat, and stored in a plastic cap sutured to the skin between the shoulder blades. The animals were placed in a stereotaxic apparatus (Kopf Instruments), and a small diameter hole was drilled into the occipital skull. A small cannula (22G; Plastics One, Roanoke, VA) was surgically implanted at or in close proximity to the nucleus of the solitary tract according to standard stereotaxic coordinates (13.85 mm bregma, 0.2 mm off midline, 8.0 mm depth; Ref. 27). After surgery, animals were allowed to recover for at least 48 h, as demonstrated by the return to normal feeding and sleep-waking patterns.

Ventilatory and cardiovascular measurements. Cardiorespiratory measures were continuously measured in the freely behaving, unrestrained animal in a calibrated 3-liter barometric chamber (Buxco Electronics, Troy, NY) by using methods described by Barlett and Tenney (3) and Pappenheimer (29). To minimize the effect of signal drift due to external temperature and pressure changes, a reference chamber of equal size, in which the temperature was measured by a T-type thermocouple, was used. In addition, as previously recommended by Epstein and colleagues (10), a correction factor was incorporated into the software routine to account for inspiratory and expiratory barometric asymmetries. Chamber temperature was maintained slightly below the thermoneutral range (24-28°C). A calibration volume of 0.5 ml of air was repeatedly introduced into the chamber before and on completion of the recordings to monitor chamber calibration. At least 60 min before the start of each protocol, animals were allowed to acclimate to the chamber, in which humidified air (90% relative humidity) was passed through at a rate of 5 l/min by using a precision flow pump-reservoir system. Pressure changes in the chamber due to the inspiratory and expiratory changes were measured by using a high-gain differential pressure transducer (model MP45-1, Validyne; Ref. 8). Analog signals were continuously digitized and analyzed on-line by a microcomputer software program (Buxco Electronics). A rejection algorithm was included in the breath-by-breath analysis routine and allowed for accurate rejection of motion-induced artifacts, according to rejection parameters established for minimum acceptable tidal volume (VT), as well as minimum and maximum inspiratory times, and the acceptable ratios of inspiratory and expiratory VT values. VT, respiratory frequency (f), and minute ventilation (E) were computed breath by breath throughout all baseline and experimental periods and subsequently stored for off-line analysis.

Chemicals. The TK inhibitors Gen (4',5,7-trihydroxyisoflavone) and A25 [-cyano-(3,4,5-trihydroxy)cinnamonitrile] and the inactive TK inhibitor analog tyrphostin A1 [A1; (4-methoxybenzylidene)malononitrile; -cyano-(4-methoxy)cinnamonitrile] were purchased from Calbiochem (La Jolla, CA).

Experimental protocol. In preliminary experiments, the optimal dosages for Gen and A25 were determined by dissolving these drugs in 0.1 ml DMSO and 1.9 ml artificial CSF and microinjecting them (1 µl) at increasing concentrations ranging from 0.0-10 mM into the dorsocaudal brain stem of one to two rats per dose. The lowest dose that attained maximal inhibitory cardiorespiratory effects during hypoxia was then chosen as the optimal dosage for microinjection during the experiments.

Measurement of blood-gas values. Arterial blood samples were obtained from the implanted arterial catheter. After withdrawal of 75-100 µl of blood in the dead space of the catheter, another 150 µl were sampled for immediate analysis of PO₂, PCO₂, and pH with a blood-gas analyzer (model 178, Ciba Corning). Measurements were always performed in room air and during the last minute of each hypoxic challenge.

Tissue lysate preparation, immunoprecipitation, and immunoblotting procedures. Animals were killed by decapitation at time 0 and minutes 5, 30, 60, and 120 of hypoxic exposure to 10% O₂-balance 90% N₂. Heads were quickly frozen in 42°C isopentane for 5 min and stored at 70°C. For dissection, brains were warmed to 5°C. The obex was visually identified, and a coronal section 1 mm caudal to 1 mm rostral to the obex was performed. The dorsal regions of the caudal brain stem were identified, carefully removed by using a 17-gauge thin-walled hypodermic needle for punch-sampling, and stored at 70°C. Approximately 10 mg of tissue were obtained from each rat. Tissues corresponding to the dorsal regions of the caudal brain stem from six animals were pooled and homogenized at 4°C in lysis buffer (1 M Tris, pH 7.5, 250 mM EGTA, 250 mM MnCl₂, 10 mM sodium orthovanadate, 200 mM phenylmethylsulfonyl fluoride, 1 M -glycerophosphate). Crude synaptosomes were isolated by centrifuging the homogenates for 10 min at 1,000 g and further centrifugation of supernatants at 15,000 g for 20 min. The resultant pellet was resuspended, washed three times in HEPES (pH 7.2) containing 1 mM EDTA, and centrifuged at 15,000 g for 20 min. Protein content was assayed by using a commercially available kit (DC-Biorad protein assay). Protein (500 µg) was used for the immunoprecipitation of NMDA receptor protein and was incubated overnight at 4°C with 10 µg of NR1 subunit antibody (Chemicon, Temecula, CA) in a total volume of 100 µl. Protein A sepharose (20 µl; Pharmacia, Uppsala, Sweden) was added and incubated for 1 h at 4°C. The immunoprecipitates were washed three times with lysis buffer and resuspended in 20 µl SDS sample buffer (0.5 M Tris, pH 6.5, 20% glycerol, 4% SDS, 100 mM dithiothreitol). Samples were equally divided in two. Proteins were then separated by electrophoresis on two 4-12% gradient Tris-glycine gels (Novex, San Diego, CA) and transferred to 0.2-mm nitrocellulose membranes. Nonspecific binding was blocked by 1 h incubation with 5% BSA or nonfat dry milk in Tris-buffered saline-Tween 20. One of the membranes was incubated with an antiphosphotyrosine antibody (PY20, 1:1,000; Upstate Biotechnology, Lake Placid, NY) for 1 h. The second membrane was incubated overnight at 4°C with either NMDA NR2A subunit antibody (1:500; Chemicon) or NMDA NR2C antibody (1:500; Chemicon). Membranes were washed three times with Tris-buffered saline-Tween 20 and incubated with secondary antibodies for 1 h. After extensive washing, proteins were visualized by enhanced chemiluminescence (Amersham), and semiquantitative analysis of the bands was performed by scanning densitometry. It should be noted that, because the immunoprecipitation was performed by using an antibody against the NR1 subunit of the NMDA glutamate receptor, the phosphotyrosine band at mol wt 180 may correspond to both the NR2A and NR2B subunits, which migrate at identical molecular weights. In addition, although the antibody against NR2A used in these experiments is considered as specific (37), we cannot exclude with certainty that some proportion of the phosphotyrosine bands or of the bands migrating at mol wt 179-180 may in fact represent NR2B, which is tyrosine phosphorylated. Therefore, results were normalized across experiments by calculating the densitometry value ratios of antiphosphotyrosine to NR2A/B for each sample.

Data analysis. Values are reported as means ± SE unless otherwise indicated. Changes in cardiovascular and ventilatory measurements were assessed between the average of stable 3-min period recordings for epochs preceding or following stimulus administration, such as microinjection or hypoxia. The hypoxic response was considered as the highest moving average of E breath-by-breath values measured over any given consecutive 3-min period during the 20-min hypoxic challenge. Differences among the various treatments groups (baseline, treatment, and hypoxia) were compared by analysis of variance (two-way ANOVA for repeated measures) followed by the Newman-Keuls test for each drug. A P value of <0.05 was set as statistically significant.

	RESULTS

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

Ventilatory responses. Microinjection of the TK inhibitors Gen (10 mM) and A25 (1 mM) was not associated with significant ventilatory changes [P = not significant (NS) vs. Veh]. Similarly, A1 administration did not elicit E, f, or VT changes (Figs. 1 and 2).

View larger version (14K): [in this window] [in a new window]   Fig. 1.   Representative hypoxic ventilatory response showing changes in minute ventilation (E) over time in same animal after vehicle (A) and genistein (10 mM; B) microinjections into dorsocaudal brain stem. A: on exposure to 10% O2 (solid arrows) for 20 min, a sharp increase in E was observed, which was sustained until return to normoxia (open arrows). B: a similar hypoxic protocol was administered after genistein treatment and showed an attenuated hypoxic ventilatory response.

View larger version (36K): [in this window] [in a new window]   Fig. 2.   Effects of tyrosine kinase inhibition on ventilatory measurements during hypoxia in conscious rat. Shown are respiratory frequency (f), tidal volume (VT), and E measurements (means ± SE) of stable 3-min epochs at baseline; after 1-µl dorsocaudal brain stem microinjection of vehicle (), genistein (, n = 13; A), tyrphostin A25 (, n = 9; B), and tyrphostin A1 (, n = 9; C); and subsequent responses to hypoxia (10% O2). br, Breaths. * P < 0.01.

Cardiovascular responses. Neither Gen (10 mM) nor A25 (1 mM) elicited significant alterations in MAP or HR during normoxia. Similarly, the chronotropic responses to hypoxia were unchanged after administration of TK inhibitors, such that HR increased by 8.6 ± 1.8, 7.5 ± 2.0, and 8.3 ± 2.1% in Veh, Gen, and A25, respectively (P = NS). No changes in MAP responses during hypoxia occurred in any of the treatment groups.

Tyrosine phosphorylation of NMDA receptor NR2 subunits. Significant increases in tyrosine phosphorylation of the NR2A/B subunits of the NMDA receptor (mol wt 179-180) occurred at 5, 30, and 60 min of hypoxia in the dorsocaudal brain stem and returned to baseline values at 120 min (Fig. 3). No changes in tyrosine phosphorylation of the NR2C subunit (mol wt 140) occurred during the hypoxic challenge.

View larger version (46K): [in this window] [in a new window]   Fig. 3.   Effect of acute hypoxia on tyrosine phosphorylation of N-methyl-D-aspartate receptor NR2A/2B subunits in rat dorsocaudal brain stem. A: dorsocaudal brain stem homogenates (500 µg) from rats in room air (0) and in hypoxia (5, 30, 60, and 120 min) were immunoprecipitated with antibodies specific for NR1, and precipitates were analyzed by immunoblotting with antibodies for NR2A/2B (top) antiphosphotyrosine (PY; bottom). B: bands corresponding to PY and NR2A/2B were scanned, and PY-to-NR ratio of optical densities for each condition was calculated to correct for inhomogeneities in immunoprecipitation across samples. Values (means ± SD) of 3 separate experiments are plotted for control (0) and hypoxia (5, 30, 60, and 120 min). Significant increases in PY occurred at 30 min hypoxia (* P < 0.01, 30 min vs. 0) and resolved at 120 min (* P < 0.01, 30 vs. 120 min).

	DISCUSSION

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

In this study, inhibition of TK activity, within the neuronal structures of the dorsocaudal brain stem targeted by our microinjections, did not elicit significant changes in normoxic ventilation. Thus within these neural regions TK appears to play little if any respiratory role during normoxia. However, TK inhibitors significantly attenuated HVR, and such attenuation was primarily due to reduced f responses. The absence of any alterations in HR or MAP responses further suggests that TK are specifically involved in the hypoxic chemotransduction reflex within these regions.

The NTS within the dorsocaudal brain stem provides the first synaptic relay for primary afferent fibers originating from the peripheral chemo- and baroreceptors (11, 17). A substantial body of evidence supports a critical role for NMDA glutamate receptors within this brain stem region in mediating HVR (2, 19, 23, 28). Indeed, glutamate microinjections within this region will increase ventilatory output in conscious rats, whereas systemic or localized administration of NMDA receptor antagonists such as MK-801 will result in marked attenuation or abolition of HVR (24, 28). Such attenuation is clearly not due to removal of excitatory components, allowing for full expression of hypoxia-induced central inhibition, because parenteral administration of MK-801 was associated with significant reductions in the ventilatory response to selective peripheral chemoreceptor activation with sodium cyanide (28).

The NMDA receptor is composed of two major types of subunits: NR1 and NR2 subunits. Whereas NR1 subunits will form fully functional homooligomeric channels, NR2 subunits can only form heterodimeric receptors with NR1, which serve to modulate the channel properties of such receptors (4, 26, 27). The intracellular domain of NMDA glutamate receptors contains multiple consensus sites for serine-threonine and tyrosine phosphorylation, suggesting that various protein kinases, such as PKC and TK, may modulate receptor channel activity (6, 32, 34). Previous work from our laboratory indicates that, within the dorsocaudal brain stem, increases in PKC activity occur during hypoxia and that PKC inhibition elicits significant HVR attenuation in conscious rats (12-14). However, differences in the magnitude of HVR reduction between NMDA receptor and PKC antagonists further suggested that other signal transduction pathways may be functionally implicated in HVR. Arguments to support TKs as likely candidates for a modulatory role in HVR stem from several lines of work. 1) In dorsal horn and hippocampal neurons, NMDA receptor-mediated currents were regulated by protein tyrosine phosphorylation, whereas marked attenuation of such currents was demonstrated when protein TPs were activated (35, 36). 2) In the cerebral cortex, the NR2A and NR2B subunits of the NMDA receptor were shown to undergo tyrosine phosphorylation in vivo (21), and induction of long-term potentiation in the rat dentate gyrus parallels increases in tyrosine phosphorylation of the NMDA receptor 2B subunit (31). 3) In postsynaptic densities, the NR2B subunit is the major protein undergoing tyrosine phosphorylation (25). 4) Protein TKs of the src and fyn families control current fluxes through NMDA receptor channels (20). 5) Transient ischemia is associated with increased tyrosine phosphorylation of NMDA glutamate receptors (18, 33). 6) Finally, topical application of Gen to the dorsal surface of the medulla attenuates the phenylephrine-induced baroreflex in the rat (23). In this study, we now demonstrate that in vivo tyrosine phosphorylation of the NR2A/2B subunits, but not of the NR2C subunit, occurs during mild hypoxia within the dorsocaudal brain stem (Fig. 3).

The HVR attenuation in rats receiving dorsocaudal brain stem microinjections of either Gen or A25 further indicates that TK activation in the dorsocaudal brain stem during hypoxia may underlie important functional respiratory components of the HVR. In addition, increases in TK activity during hypoxia appear to modulate ventilatory frequency components and had no effect on VT. It should be stressed at this point that, although TK inhibitors affected the HVR, they did not modify any cardiovascular or ventilatory parameter during normoxia, suggesting that either the regions encompassed by the inhibitor or the TK basal activity in these neural sites do not appear to play a role in maintenance of eucapnic normoxic ventilation and cardiovascular regulatory mechanisms in the conscious rat. This is somewhat in contradistinction with Man et al. (23), who found that the chronotropic and baropressor responses to intravenous phenylephrine were mildly attenuated by topical application of Gen to the surface of the dorsocaudal brain stem of anesthetized rats. These discrepant results can be accommodated under the assumption that state (waking vs. anesthesia) may modify the dependency of the response to a stimulus, such that, when redundant drives are removed by the anesthesia, the functional role of TK in the modulation of the response to the challenge more fully emerges. Alternatively, and in clear agreement with our present findings, Gen application to the surface of the dorsocaudal brain stem did not modify the basal cardiovascular measurements of the anesthetized rats (23), further suggesting that TKs have little if any role in the tonic maintenance of HR or MAP.

Although we cannot definitively rule out the possibility that the TK inhibitors used in this study may have diffused to neighboring brain stem structures to elicit the alterations in HVR, it is noteworthy that, when the injection cannula were implanted within nondorsocaudal brain stem regions (data not shown), no significant effects on HVR occurred after administration of the TK inhibitors. Obviously the possibility exists that the diffusion of methylene blue and that each of the TK inhibitors employed herein may differ within the dorsocaudal brain stem. In such instances, the extent of such diffusion would not be precisely appreciated from extrapolations based on the methylene blue assessments. Thus it is possible that microinjections of TK inhibitors may have diffused more efficiently and, therefore, affected sites more remotely located than those suggested from the postmortem examination. Notwithstanding such unavoidable constraints in a freely behaving preparation, the gradient of drug concentrations created by the diffusion process would point to a higher concentration in the immediate vicinity of the cannula tip and, therefore, would suggest that most of the physiological alterations elicited by administration of TK inhibitors may be preferentially ascribed to neural regions encompassed within close proximity to the NTS. Of note, all animals studied demonstrated the predicted ventilatory and cardiovascular response to L-glutamate when it was microinjected in the NTS, further consolidating the implications of our findings.

The HVR effects are probably not related to the inherent properties of the particular TK inhibitor employed in these studies, because members of both common classes of TK inhibitors were used and elicited similar results. The first inhibitor used in this study, Gen, acts by binding to the ATP binding site (1). Although this inhibitor is widely used as a specific TK inhibitor, the possibility exists that Gen may have induced nonspecific inhibition of serine-threonine kinases (1). However, such a possibility is remote because A25, which inhibits TK activity by binding to the substrate-binding site of the enzyme (22), was associated with HVR attenuations of similar magnitude. In addition, the inactive analog to A25, A1, had no effect on the ventilatory response to either normoxia or hypoxia. Therefore, the results obtained in this study indicate that the HVR attenuation induced by the two TK inhibitors is indeed due to TK inhibition and most probably does not represent modulation of other kinases such as PKC or protein kinase A.

The physiological role(s) of the various TKs within neurons underlying cardiorespiratory functions has yet to be established. The present study merely indicates that hypoxia induces temporally constrained changes in tyrosine phosphorylation of particular NMDA glutamate receptor subunits and that the HVR is modified by application of TK inhibitors to the dorsocaudal brain stem. From presently available evidence, TKs and TPs are thought to act together and thus create a delicate balance controlling neuronal activity, such that in isolated neurons for example, NMDA currents are potentiated by TK and attenuated by TP (35). Thus in the context of HVR constituting a predominantly glutamatergic response, the early activation of functionally relevant TK within respiratory neurons could potentiate the early HVR. In contrast, subsequent activation of TP could allow for more precise modulation of neuronal discharge and thereby permit onset of hypoxic ventilatory roll-off. Furthermore, in analogy to the hippocampal slice preparation, in which anoxic stimulation leads to a selective inhibition of protein TK activity leading to tyrosine-dephosphorylation of the NMDA receptor (5), accentuation of the severity of hypoxemia would be expected to induce rapid deactivation of the functionally relevant TK and lead to apnea. Thus improved understanding of the various signal transduction elements mediating the intracellular response magnitude and temporal domains may allow for development of better pharmacological interventions during disease states associated with hypoxia.

In summary, administration of TK inhibitors within the dorsocaudal brain stem attenuates frequency components and the magnitude of the ventilatory response to hypoxia in conscious rats, whereas it has no apparent effect on cardiovascular responses. Furthermore, we have shown that tyrosine phosphorylation of the NR2A/NR2B subunits of NMDA receptors, a critical voltage-gated receptor in HVR within the dorsocaudal brain stem, occurs during acute hypoxic challenges. We postulate that protein tyrosine phosphorylation events within neuronal populations of the dorsocaudal brain stem mediate functionally important signal transduction pathways of HVR.