Recovery of respiratory function following C2 hemi and carotid body denervation in adult rats: influence of peripheral adenosine receptors

Experimental Neurology 191 (2005) 94 – 103 www.elsevier.com/locate/yexnr

Recovery of respiratory function following C2 hemi and carotid body denervation in adult rats: influence of peripheral adenosine receptors Han Baea, Kwaku D. Nantwib,*, Harry G. Goshgarianb b

a Department of Otolaryngology, Wayne State University School of Medicine, Detroit, MI 48202, USA Department of Anatomy and Cell Biology, Wayne State University School of Medicine, Detroit, MI 48202, USA

Received 5 April 2004; revised 2 September 2004; accepted 20 September 2004 Available online 21 November 2004

Abstract The efficacy of the methylxanthine, theophylline, as a respiratory stimulant has been demonstrated previously in an animal model of spinal cord injury. In this model, an upper cervical (C2) spinal cord hemi paralyzes the ipsilateral hemidiaphragm. Theophylline restores respiratory-related activity in the paralyzed hemidiaphragm via activation of a latent respiratory motor pathway. Antagonism of central adenosine A1 receptors mediates this action. Theophylline also enhances respiratory frequency, f, defined as breaths per minute. Thus, long-term use may result in respiratory muscle or motoneuron fatigue particularly after spinal cord injury. We assessed the effects of an adenosine A1 receptor agonist, N 6-psulfophenyladenosine ( p-SPA) on theophylline’s action in our model under standardized recording conditions. Four groups of rats, classified as hemisected/nonhemisected with the carotid bodies denervated (H-CBD or NH-CBD), and hemisected/ nonhemisected with the carotid bodies intact (H-CBI or NH-CBI ) were used in the study. Eight days after recovery from carotid denervation, a left C2 hemi was performed in H-CBD rats. C2 hemi was also performed in H-CBI animals, and 24 h later, electrophysiologic experiments on respiratory activity were conducted in both groups of animals. Two groups using nonhemisected controls were also employed as described above. In H-CBD rats, theophylline significantly ( P b 0.05) enhanced f and induced respiratory-related activity in the previously quiescent left phrenic nerve. In NH-CBD rats, theophylline significantly enhanced f. In both H-CBD and NH-CBD rats, p-SPA (0.25 mg/kg) did not significantly change theophylline-induced effects. In H-CBI rats, theophylline significantly ( P b 0.05) enhanced f and induced activity in the previously quiescent left phrenic nerve. In HCBI rats, p-SPA reduced the values to pre-theophylline discharge levels. Recovered activity was not obliterated with the agonist. In NH-CBI rats, p-SPA reduced theophylline-induced effects to pre-drug discharge levels. Adenosine A1 and A2A receptor immunoreactivity was detected in the carotid bodies. The significance of our findings is that theophylline-induced effects can be normalized to pre-drug levels by the selective activation of peripheral adenosine A1 receptors. The therapeutic benefits of theophylline, i.e., recovered respiratory function after paralysis, however, persists. The potential therapeutic impact is that respiratory muscle fatigue associated with long-term theophylline use may be minimized by a novel therapeutic approach. D 2004 Elsevier Inc. All rights reserved. Keywords: Adenosine A1 receptors; Theophylline; Respiratory function; Cervical spinal cord injury; Immunohistochemistry

Introduction * Corresponding author. Department of Anatomy and Cell Biology, Wayne State University, School of Medicine, 540 East Canfield Avenue, Detroit, MI 48202. Fax: +1 313 577 3125. E-mail address: [email protected] (K.D. Nantwi). 0014-4886/$ - see front matter D 2004 Elsevier Inc. All rights reserved. doi:10.1016/j.expneurol.2004.09.007

In an animal model of spinal cord injury, a latent respiratory motor pathway found in the nondamaged regions of the spinal cord can be activated to restore function to a hemidiaphragm paralyzed by an ipsilateral C2 hemi

H. Bae et al. / Experimental Neurology 191 (2005) 94–103

(Goshgarian, 1979; Goshgarian and Guth, 1977). Activation of the latent respiratory motor pathway can be induced by asphyxia or the systemic administration of the methylxanthine, theophylline (Goshgarian, 1979; Nantwi and Goshgarian, 1998, 2001; Nantwi et al., 1996, 2003). In asphyxia-induced functional restoration, enhanced central respiratory drive is the underlying mechanism (Goshgarian, 1979; Goshgarian and Guth, 1977). Theophyllineinduced functional restoration is centrally mediated, however, by blockade of adenosine receptors (Nantwi and Goshgarian, 1998), which in turn enhances central respiratory drive. Although respiration is primarily under the control of the central nervous system, peripherally located chemoreceptors are involved in the modulation of respiration (Brian and Chau, 1998; Fuller et al., 1987; McQueen and Ribeiro, 1981; Middlekauf et al., 1998). In particular, adenosine receptors located in the carotid bodies modulate respiratory activity in various species, including man (Brian and Chau, 1998; Lugliani et al., 1971; McQueen and Ribeiro, 1983; Monteiro and Ribeiro, 1986, 1987; Watt and Routledge, 1985). Adenosine receptors have been classified into A1, A2 (A, B) and A3 receptor subtypes (Ribeiro et al., 2003). The A1 and A3 receptors are coupled to an inhibitory G-protein while the A2A and A2B receptors are coupled to a stimulatory G-protein. Adenosine A1 and A2A receptors are involved in the physiologic actions of adenosine and the A2B are relevant primarily under pathologic conditions (Cunha, 2001; Ribeiro et al., 2003). The A3 receptor is a high-affinity receptor in humans, but has a low density in most tissues. While its physiologic function is unclear, it may be implicated in neuroprotection (Ribeiro, 2003). Physiologically, adenosine A1 and A2A receptors modulate respiratory activity in a specific manner; that is, the A1 receptors depress while the A2A receptors excite respiratory activity (Carley and Radulovacki, 1999; Monti et al., 1996; Brian and Chau, 1998; Monteiro and Ribeiro, 1987). In previous studies, we addressed the therapeutic potential of theophylline in spinal cord injured patients with respiratory deficits (Nantwi et al., 1996, Nantwi and Goshgarian, 1998, 2001; Nantwi et al., 2003). In addition to restoring paralyzed respiratory muscle function, it should be pointed out that theophylline also enhances respiratory frequency ( f ), defined as breaths per minute in rats, healthy adult humans as well as humans with respiratory disease and human infants (Mueller et al., 1981; Nantwi et al., 1996; Richmond, 1949). From our animal studies and the aforementioned clinical studies, we postulated that since theophylline enhances f, prolonged use of the drug could result in respiratory muscle or motoneuron fatigue and offset any therapeutic benefit derived from its use. However, since peripherally acting adenosine A1 receptors can reduce centrally mediated respiratory disturbances that occur during sleep (Monti et al., 1996; Carley and Radulovacki, 1999), we reasoned that it may be possible to normalize

95

theophylline-induced enhanced f without sacrificing respiratory muscle recovery by also employing a drug that acts on peripheral adenosine A1 receptors. In the present study, we tested the hypothesis that theophylline-enhanced f can be modulated back to normal levels after C2 hemi by the selective activation of adenosine A1 receptors in the carotid bodies with N 6-p-sulfophenyladenosine ( p-SPA), an adenosine A1 receptor agonist. We addressed the issue of whether excision of the carotid bodies alters the effectiveness of theophylline to induce recovery of respiratory-related activity following C2 hemi. Second, in order to confirm the presence of adenosine receptors in the carotid bodies, immunohistochemical analysis of carotid bodies was performed to identify adenosine A1 and A2A receptor immunoreactivity and expression. Lastly, we employed the systemic administration of (N 6-p-sulfophenyladenosine, p-SPA), a specific adenosine A1 receptor agonist that does not cross the blood–brain barrier (Gleeson and Zwillich, 1992), to determine if theophylline-enhanced respiratory frequency can be normalized in C2 spinal cord hemisected rats displaying recovery of the ipsilateral hemidiaphragm.

Materials and methods Four experimental groups of adult Sprague–Dawley female rats (260–340 g) were employed in the study. Animals were classified as hemisected or nonhemisected with the carotid body denervated (H-CBD, n = 9 or NHCBD, n = 7), and hemisected or nonhemisected with the carotid body intact (H-CBI, n = 7 or NH-CBI n = 6). Animal care was carried out in compliance with the guidelines set by the Division of Laboratory Animal Research at Wayne State University. Prior to surgery to excise the carotid bodies, animals were anesthetized with ketamine (70 mg/kg) and xylazine (7 mg/kg, i.m.) approximately 10 min after injection with atropine sulfate (0.1 mg/kg, i.m.) to minimize mucus secretions. To denervate the carotid bodies, a ventral midline incision was made in the neck to expose and visualize the carotid artery bifurcation and related structures. All tissue within the carotid bifurcation was then carefully excised with a pair of microscissors. Carotid body excision was conducted bilaterally as described previously (Olson et al., 1988). The excised tissue was placed in 4% paraformaldehyde for later processing for the presence of adenosine A1 and A2A receptors using immunohistochemical techniques. Functional assessment of a complete bilateral carotid body denervation was conducted at the conclusion of electrophysiologic experiments (as described below) by the i.v. administration of sodium cyanide (20 Ag/kg). Following carotid body denervation, animals were returned to clean litter-lined cages to recover. They were allowed to survive post-operatively for 7 days prior to any other procedures. After recovery from carotid body exci-

96

H. Bae et al. / Experimental Neurology 191 (2005) 94–103

sion, weight loss in most animals amounted to 20–25 g. Animals that survived the procedure were able to feed and water themselves remarkably well. On the 8th day after carotid body denervation, H-CBD animals were subjected to a left C2 spinal cord hemi (Liou and Goshgarian, 1994). Twenty-four hours later, they were prepared for electrophysiologic assessment of respiratory-related activity. NHCBD rats were subjected to carotid body denervation and allowed to recover for 7 days. On day 8, NH-CBD animals were prepared for assessment of respiratory-related activity in the phrenic nerves. Mortality following carotid body denervation In all animals (n = 22) that were subjected to carotid body excision, six (27%) did not survive the procedure. Usually, death occurred as a result of apnea, an irregular pattern of breathing, or from complications of excessive bleeding or laryngopharyngeal impairment. In the latter instance, the animals recovered from the surgery itself, but thereafter, they exhibited difficulty in swallowing. H-CBI animals were subjected to a left C2 hemi and allowed to recover in a clean litter-lined cage. Weight loss following hemi was minimal (5–10 g). Electrophysiologic assessment of respiratory activity in the phrenic nerves was conducted 24 h after hemi in H-CBI rats. NH-CBI rats were not subjected to any surgical procedures prior to assessment of respiratory-related activity in the phrenic nerves. Electrophysiology Prior to electrophysiologic experiments, animals were anesthetized with chloral hydrate (400 mg/kg, i.p.) approximately 10 min after administration (i.m.) of atropine sulfate to minimize mucus secretions. A tracheostomy was performed and an endotracheal tube (PE 160) inserted into the trachea, and the femoral vein and femoral artery cannulated (PE 50) to administer drugs and monitor blood pressure, respectively. Body temperature was monitored with a rectal probe and maintained with a thermostatic blanket (Bowdoinham, ME) at 37.0–37.28C. Blood pressure was maintained at 80–110 mm Hg by i.v. administration of a 5% dextrose and dextran solution. The left and right phrenic nerves were exposed in the neck, desheathed, and placed on platinum recording electrodes. The nerves were immersed in mineral oil to prevent desiccation. Electrophysiologic recordings were conducted under the following standardized conditions: (1) paralysis with pancuronium bromide (i.v. 0.5 mg/kg) to eliminate afferent activity associated with muscle contraction, (2) bilateral vagotomy to eliminate abdominal and aortic chemoreceptor inputs, (3) artificial ventilation on a small animal ventilator (Harvard Rodent Ventilator), and (4) endtidal CO2 was monitored continuously and maintained at a constant level (38–43 mm Hg) with a flow-through capnograph (Novametrix, Wallingford, CT).

Quantification of respiratory-related activity The mean areas under the integrated waveform (AUC F SEM) of 10 consecutive respiratory bursts in the left (AUCl) and right (AUCr) phrenic nerves were calculated before and after theophylline (15 mg/kg) administration in all animals. Once a stable level of theophylline-induced respiratory changes had been attained, the adenosine A1 agonist (N 6-psulfophenyladenosine, p-SPA, at a dose of 0.25 mg/kg) was injected i.v. The theophylline dose chosen was based on previous studies in which we demonstrated that at 15 mg/kg, the drug elicits optimal effects (Nantwi and Goshgarian, 1998, 2001). The dose of p-SPA was based on pilot dose response experiments. Changes in respiratory burst pattern and f after p-SPA administration were quantitatively analyzed. It has been demonstrated that the integrated phrenic waveform most closely linked with respiratory function is amplitude (Eldridge, 1971). However, the area under the curve analysis was used as a measure of respiratory activity because this approach represents a more global assessment of phrenic motor output. Previous studies in our lab have demonstrated that using both parameters to assess phrenic nerve activity showed that changes in amplitude correlated with significant changes in area (Hadley et al., 1999). Other respiratory parameters measured were inspiratory burst duration (Ti), inter-burst interval (Te), and respiratory rates ( f ). Statistical analysis Data for drug-induced actions are expressed as means (FSEM) and compared using a one-way ANOVA test (Newman–Keuls). Significance was set at P b 0.05. Immunohistochemistry The excised carotid bodies were fixed in 4% paraformaldehyde overnight and then cryoprotected in 30% sucrose. Following cryoprotection, the tissues were cut into 50-Am s on a cryostat. The cut s were washed (3) with Tris-buffered saline (TBS) (0.1 M, pH = 7.4) and endogenous peroxidase was quenched (for 30 min) with 0.5% hydrogen peroxide (H2O2) in TBS and then washed again (1) with TBS. The s were incubated (1 h) in 0.02% Triton X-100 in TBS and then blocked with 2% normal goat serum (NGS) and 2% bovine serum albumin (BSA) in TBS at room temperature for 2 h. Sections were then incubated overnight either in a rabbit anti-adenosine A1 receptor antibody or a rabbit anti-adenosine A2A receptor antibody in a 1% NGS/1% BSA TBS solution (diluted 1:250). Reagents for immunohistochemistry, Diaminobenzidine (DBA, Sigma) specific adenosine A1 and A2A antibodies, bovine serum albumin (BSA), normal goat serum (NGS), goat anti-rabbit IgG biotinylated secondary antibody were purchased from Vector Laboratory. Following incubation with the primary antibody and washing 3

H. Bae et al. / Experimental Neurology 191 (2005) 94–103

(10 min per wash), the s were processed according to the Vector Laboratory ABC protocol utilizing the goat antirabbit IgG biotinylated secondary antibody and 3,3Vdiaminobenzidine (DAB + Ni) to develop and visualize the immunoreactive areas. Drugs and reagents Theophylline (Sigma) and N 6-p-sulfophenyladenosine, p-SPA), (RBI) were prepared separately in physiologic saline. Sodium cyanide (Sigma) was prepared as a stock solution in physiologic saline and refrigerated. Adequate volumes were dispensed and used as necessary during experiments.

Results H-CBD animals Fig. 1 shows a recording taken from a H-CBD animal before and after administration of theophylline (15 mg/kg) and p-SPA (0.25 mg/kg). In hemisected animals (H-CBD and H-CBI), a total absence of respiratory-related activity in the left phrenic nerve prior to any drug administration was

97

indicative of a functionally complete hemi (Fig. 1A). Only animals that showed a total absence of respiratory activity were included in data analysis. In this group (n = 9) of animals, functional recovery was induced by systemic theophylline (15 mg/kg) administration in seven of the animals tested (Fig. 1B). Onset of changes in respiratory-related activity induced after theophylline was within 10 min. By 15, min the changes attained a stable level. The adenosine A1 receptor agonist, p-SPA, was administered after theophyllineinduced effects were stable (15 min). A standard dose (0.25 mg/kg) of p-SPA was chosen based on preliminary experiments that showed that at higher doses (0.5–2 mg/kg), p-SPA elicited a characteristic (apneustic) pattern of respiratory activity (data not shown). We reasoned that at higher p-SPA doses, the apneustic pattern would mask any other effect and therefore the low dose of 0.25 mg/kg that did not alter the respiratory pattern was employed. p-SPA had no effect in carotid body denervated animals (Fig. 1C). At the conclusion of the experiment, sodium cyanide (20 Ag/kg) was administered i.v. to assess functional completeness of the carotid body denervation. An absence of enhanced f (in marked contrast to H-CBI animals) confirmed that in this animal, excision of the carotid bodies was functionally complete (Fig. 1D).

Fig. 1. The effects of systemically administered theophylline and the adenosine A1 receptor agonist, N 6-p-sulfophenyladenosine, p-SPA on respiratory-related activity was assessed in an animal subjected to carotid body denervation and left C2 hemisection (H-CBD). Twenty-four hours after hemisection, electrophysiologic assessment was conducted. The upper tracing in each pair is an integrated waveform and the lower tracing shows the raw compound action potentials from each nerve. (A) Activity in the contralateral (right) phrenic nerve (RPN) is evident while the left phrenic nerve (LPN) is totally devoid of respiratory-related activity indicating a functionally complete hemisection. (B) Ten to fifteen minutes after administration of theophylline, activity in the previously quiescent LPN is evident while the RPN shows enhanced activity. Note also that respiratory frequency is enhanced (compare with A). (C) Administration of p-SPA in this carotid body denervated animal does not result in any significant change in respiratory activity or frequency (compare B and C). (D) Administration of sodium cyanide (20Ag/kg) (to activate peripheral chemoreceptors in the carotid body) did not enhance respiratory activity, in particular f, and thus confirmed that the carotid bodies had been denervated.

98

H. Bae et al. / Experimental Neurology 191 (2005) 94–103

A summary of the quantitative analysis of theophylline and p-SPA actions in H-CBD rats is shown in Table 1. Interestingly, theophylline significantly decreased Te from 0.87 F 0.07 to 0.62.0 F 0.03 s. However, the drug did not significantly change Ti. The effect on Te was apparently more important in the drug’s effect on f than Ti. In two H-CBD animals, respiratory activity in the left phrenic nerve was not induced by theophylline. However, theophylline enhanced AUCr from pre-drug values of 35.62 and 42.25 AV s and to 52.67 and 55.78 AV s, respectively. A test of asphyxia (ventilator was turned off) showed that only the right phrenic nerve responded with enhanced amplitude, the left phrenic nerve remained quiescent in both cases suggesting that activation of the latent respiratory pathway could not be induced either chemically or pharmacologically in the two cases. Alternatively, it is possible that the left phrenic nerve may have been damaged in both instances. H-CBI Fig. 2 is a recording from an H-CBI animal that was administered theophylline and p-PSA 24 h after surgery. As in all cases after hemi, functional completeness of the lesion was confirmed by the total absence of any respiratory-related activity in the left phrenic nerve prior to drug administration. Theophylline (15 mg/kg) induced recovery of respiratory activity in the previously quiescent left phrenic nerve, enhanced the amplitude in the right phrenic nerve and significantly increased f. Administration of p-SPA (0.25 mg/kg) significantly attenuated theophylline-induced effects; however, recovered activity in the left phrenic nerve did not disappear; that is, the effect of theophylline in inducing recovery was not blocked. The apparent net effect of p-SPA was that f was normalized back to near predrug levels. A summary of the quantitative analysis of changes in all respiratory parameters measured in H-CBI rats after theophylline and p-SPA is shown in Table 1. Administration of sodium cyanide (20 Ag/kg) at the conclusion of electrophysiologic recording demonstrated a marked enhancement in respiratory activity in contrast to the absence of enhanced activity in H-CBD animals (compare Figs. 1D and 3D).

NH-CBD Pre-drug discharge levels were significantly ( P b 0.05) slower in animals in this group compared to the other groups (Table 2). Pre-drug f in all animals in this group amounted to 34.2 F 1.39 breaths/min. Theophylline administration significantly enhanced frequency to 58.73 F 1.46 breaths/ min. Theophylline-induced changes were observed within 10 min after drug injection. Administration of p-SPA after a stable level of theophylline-induced effect did not significantly alter Ti, Te, or f for up to1 h post administration. A summary of the effects of theophylline and p-SPA in all NH-CBD animals is shown in Table 2. NH-CBI Theophylline enhanced respiratory-related activity in the phrenic nerves in animals in this group. Onset of theophylline-induced affects was apparent within 10–15 min after drug injection. The effects of theophylline in NH-CBI animals were parallel to its effects in the H-CBI group. Administration of the agonist, p-SPA, at the standard dose modulated theophylline-induced changes back to normal discharge levels. A summary of the quantitative analysis of NH-CBI animals before and after drug administration is shown in Table 2. Adenosine A1 and A2A expression in carotid bodies Fig. 3 demonstrates adenosine positive A1 and A2A receptor immunoreactivity using a light-field photomicrograph of DAB positive carotid body s. The pooled s were incubated with rabbit adenosine A1 receptor antibody and A2A receptor antibody. Inserts show the immunoreactivity in the whole carotid bodies.

Discussion The results of the present study demonstrate that theophylline stimulates respiratory-related activity in animals subjected to a left C2 hemi with the carotid body

Table 1 Respiratory parameters in H-CBD and H-CBI animals after systemic administration of theophylline and P-SPA Treatment

AUCr (AV s)

AUCl (AV s)

Ti (s) F SEM

Te (s) F SEM

F (breaths/min) F SEM

H-CBD pre-drug H-CBD post-theophylline H-CBD post-p-SPA H-CBI pre-drug H-CBI post-theophylline H-CBI post-p-SPA

22.2 F 59.78 F 56.48 F 41.3 F 98.9 F 46.1F

ND 22.78 24.45 ND 40.2 32.5

0.69 0.64 0.62 0.45 0.41 0.41

0.87 0.62 0.72 0.91 0.69 0.91

42.6 48.8 49.18 44.5 56.0 47.0

1.2 3.5* 3.17* 3.1 7.0* 3.7

F1.7 F 2.8 F 1.2** F 1.2**

F F F F F F

0.03 0.02 0.02 0.01 0.01** 0.01**

F F F F F F

0.07 0.03* 0.03* 0.03 0.02* 0.03

F F F F F F

1.2 1.5** 1.5** 3.30 1.63** 2.5

AUCr = area under the curve, right phrenic nerve. AUCl = area under the curve, left phrenic nerve. ND = respiratory activity in the left phrenic nerve of hemisected animals prior to drug administration was not detected. Ti = inspiratory burst duration (s). Te = inter-burst interval (s). F = respiratory frequency (breaths/min). * Statistically significant P b 0.001, compared to pre-drug. ** Statistically significant P b 0.05, compared to pre-drug.

H. Bae et al. / Experimental Neurology 191 (2005) 94–103

99

Fig. 2. A representative set of tracings from an animal subjected to a left C2 hemisection without carotid body denervation (H-CBI) 24 h prior to electrophysiologic assessment of respiratory-related activity. (A) Activity in the RPN is evident while the LPN is devoid of respiratory-related activity indicating that the hemisection was functionally complete. (B) Following administration of theophylline (15 mg/kg), RPN activity is enhanced while the previously quiescent LPN demonstrates robust respiratory-related activity synchronous with RPN activity. Note that f is enhanced. (C) Administration of pSPA does not block recovered activity or activity in RPN. However, f is decreased to pre-drug levels. Sodium cyanide (20 Ag/kg) administration (1 h after pSPA) elicited enhanced respiratory activity in marked contrast to its lack of effect in carotid body denervated animals (compare D with Fig 1D.).

denervated or intact, as well as in animals not subjected to a hemi with denervated or intact carotid body. Furthermore, the results also show that in animals with the carotid bodies intact, the respiratory effects induced after theophylline administration can be modulated in a specific manner by the adenosine A1 receptor agonist, p-SPA. Specifically, p-SPA normalizes theophylline-enhanced respiratory frequency without sacrificing recovery in the phrenic nerve ipsilateral to hemi. Conversely, in animals with the carotid bodies excised, p-SPA does not normalize frequency. The results also show that adenosine A1 receptors that modulate theophylline-induced respiratory-related activities are immunoreactive in the carotid bodies along with adenosine A2A receptors. The findings in the present study are

discussed in terms of adenosinergic mechanisms involved in respiration, peripheral modulation of respiration, and the potential therapeutic implications of modulating the central control of respiration via the specific activation of peripherally located adenosine A1 receptors. It is generally known that theophylline (an antagonist at adenosine A1 and A2 receptors) stimulates respiration in various species including man (Eldridge et al., 1985; Richmond, 1949; Rudolphi et al., 1992; Rall, 1990). We have demonstrated previously in our model of spinal cord injury that the activation of a latent respiratory motor pathway with theophylline restores respiratory function in C2 hemisected rats and this action is mediated centrally via adenosine A1 receptor blockade (Nantwi and Goshgarian,

Table 2 Respiratory parameters in NH-CBD and NH-CBI animals after systemic administration of theophylline and p-SPA Treatment

AUCr (AV s) F S.E.M.

AUCl (AV s) F S.E.M.

Ti (s) F S.E.M.

Te (s) F S.E.M.

F (breaths/min) F S.E.M.

NH-CBD pre-drug NH-CBD post theophylline NH-CBD post p-SPA NH-CBI pre-drug NH-CBI post theophylline NH-CBI post p-SPA

46.66 F 1.5 65.2 F 2.10* 64.4 F 1.9* 34.73 F 1.1 51.96 F 1.73+ 36.11 F 1.4

44.2 58.8 59.33 29.53 67.77 31.5

0.52 0.52 0.50 0.45 0.37 0.45

0.89 0.46 0.51 0.98 0.72 0.98

34.12 58.73 57.56 40.86 54.38 41.73

F F F F F F

1.8 2.18** 1.80** 1.1 1.84++ 1.2

F F F F F F

0.02 0.03 0.03 0.01 0.01+ 0.01

F F F F F F

0.06 0.03* 0.03* 0.02 0.02+ 0.02

F F F F F F

2.15 1.46* 1.36* 2.72 1.34++ 1.84

AUCr = area under the curve, right phrenic nerve. AUCl = area under the curve, left phrenic nerve. Ti = inspiratory burst duration (s). Te = inter-burst interval (s). F = respiratory frequency (breaths/min). * Statistically significant, compared to pre-drug levels, P b 0.001. ** Statistically significant, compared to pre-drug levels, P b 0.01. + Statistically significant, P b 0.001, compared to pre-drug post, p-SPA. ++ Statistically significant, P b 0.05, compared to pre-drug post, p-SPA.

100

H. Bae et al. / Experimental Neurology 191 (2005) 94–103

Fig. 3. Light field photomicrographs of DAB + Ni positive carotid body sections incubated with rabbit adenosine A1 antibody (A) and A2A (B). The sections were incubated in 0.02% Triton X-100 blocked with normal goat serum (NGS) and incubated with the adenosine receptor antibody in 1% NGS/1%BSA diluted at 1:250. The sections are 50 Am apart and the demarcated areas represent cells reactive/positive for A1 (A) and A2A (B), respectively. Inserts represent the entire cross section of the carotid body. A control section of the pooled carotid bodies (incubated without adenosine receptor antibody) is shown in C for comparison.

H. Bae et al. / Experimental Neurology 191 (2005) 94–103

1998, 2001). The demonstration that theophylline restores respiratory function in C2 hemisected rats with bilaterally excised carotid bodies extends our previous findings (Nantwi and Goshgarian, 1998, 2001; Nantwi et al., 1996). Other pharmacological studies have revealed latent pathways in spinally injured animals and have also demonstrated recovery of respiratory activity in animals subjected to a left C2 hemi (Ling et al., 1994; Zhou and Goshgarian, 1999, 2000). In these studies, however, the carotid bodies were intact. As far as we are aware, the present study is the first to demonstrate that after carotid body denervation, recovery of respiratory activity in animals subjected to a left C2 hemi can be induced with theophylline. The effects of theophylline clearly demonstrate that the carotid bodies per se may not be essential for the pharmacological activation of the latent respiratory pathway to restore respiratory function. This finding, while intriguing, was not entirely unexpected inasmuch as the actions of theophylline are centrally mediated and central mechanisms are predominant in the control of respiration (Horn and Waldrop, 1998; Nantwi and Goshgarian, 1998, 2001, 2002; Nantwi et al., 1996). Perhaps of paramount importance is that theophyllineinduced stimulatory effects on respiratory frequency can be modulated back to near normal discharge levels by activation of adenosine A1 receptors located in the carotid bodies. Thus, in the present study, we have demonstrated that it is feasible to modulate a centrally mediated drug action by targeting peripherally located receptors. Our study has demonstrated that modulation of centrally mediated adenosinergic events by specific activation of peripheral A1 receptors is a novel approach that may be employed in spinal cord injury research. It derives primarily from the recognition that respiration is modulated by peripheral structures and secondarily from the influence of adenosine in carotid bodies (Fuller et al., 1987; Lugliani et al., 1971; McQueen and Ribeiro, 1981; Monteiro and Ribeiro, 1986, 1987; McQueen, 1993; Brian and Chau, 1998; Koos et al., 2001). Furthermore, it underscores a principal role of central mechanisms in the control of respiration (Bianchi et al., 1995; Eldridge et al., 1985; Horn and Waldrop, 1998). Adenosine is a neuromodulator and, in particular, the A1 receptors are inhibitory while A2A receptors are excitatory (Gauda et al., 2000). In the respiratory system, A1 receptors depress and A2A receptors excite respiratory activity (Monti et al., 1996; Brian and Chau, 1998; Carley and Radulovacki, 1999; Monteiro and Ribeiro, 1987; Reid et al., 1991). Our findings confirm that adenosine A1 receptors in the periphery can be specifically activated to modulate a centrally mediated event, a finding similar to previous studies (Carley and Radulovacki, 1999; Monti et al., 1995). In their investigations, the authors in the aforementioned studies showed that the incidence of centrally mediated sleep apneas can be modulated by activation of A1 receptors in the periphery and concluded that it is feasible to modulate a centrally mediated event

101

via peripheral mechanisms. We concur with the conclusion. Furthermore, the findings in the present study are the first to demonstrate that in an animal model of spinal cord injury, selective activation of peripheral adenosine A1 receptors can modulate recovered respiratory activity back to normal discharge levels without negating the principal mechanism that underlies functional recovery. The latter issue is particularly important since theophylline has been shown to improve respiratory muscle drive in human tetraplegia (Ferguson et al., 1999). Our results also show that after carotid body denervation in nonhemisected animals, basal discharge f is slower than in nonhemisected animals with the carotid bodies intact (see Table 2). This finding is consistent with previous studies that have shown that bilateral carotid body denervation results in hypoventilation in rats (Coles et al., 2002; MartinBody et al., 1985; Olson et al., 1988) in cats (Millhorn et al., 1984) and dogs (Nakayama et al., 2003). In our study, respiratory parameters were monitored under standardized recording conditions while the other studies were conducted in conscious animals following denervation. However, the observation of respiratory depression, i.e., respiratory frequency slower than in animals with intact carotid bodies, was consistent with hypoventilation reported in awake as well as anesthetized animals. Nakayama et al. (2003) reported hypoventilation after bilateral carotid body denervation and concluded that carotid body denervation decreases the drive to breathe. While we concur with this conclusion, we observed that when a left C2 hemi was performed in addition to bilateral carotid body excision, the basal f prior to any drug administration was higher than in animals subjected to carotid body excision alone (see Tables 1 and 2). Our findings that basal f in animals subjected to a left C2 hemi in addition to carotid body denervation was higher was not unexpected since our laboratory previously showed that after hemi f increases significantly (Goshgarian et al., 1986). This suggests that in the control of respiration, central mechanisms play a dominant role although peripheral mechanisms are modulatory. Alternatively, it may be inferred that enhanced respiratory drive following hemi can override the depressed drive to breathe following carotid body denervation (Moreno et al., 1992; Nakayama et al., 2003). The net effect is that in animals subjected to both denervation and hemi, basal f is similar to normal f prior to drug administration. Immunoreactivity of adenosine A1 and A2A receptors in the carotid bodies Our findings that A1 and A2A receptors are immunoreactive in the carotid bodies of rats supports similar findings from previous studies in rats and rabbits (Gauda, 2000, Gauda et al., 2000; Kobayashi et al., 2000; Rocher et al., 1999). Gauda et al. (2000) reported that while the A1 receptor is not expressed in the carotid body per se, it is abundantly expressed in the petrosal ganglion. It must be

102

H. Bae et al. / Experimental Neurology 191 (2005) 94–103

pointed out that in the present study, no attempt was made to identify the types of cells in the carotid body. Our approach was to excise the entire carotid body; undoubtedly, this would include the glomus cell types as well as fibers from the carotid sinus nerve that have cell bodies located in the petrosal ganglion. Our finding of A1 receptor immunoreactivity in the carotid body supports the suggestion that these receptors could be located on carotid sinus nerves (Gauda et al., 2000). Significance The significance of the findings in the present study is that administration of the specific adenosine A1 receptor agonist, p-SPA, can modulate theophylline-induced recovery of respiratory function in animals subjected to C2 hemi without negating recovered activity. This finding is important because although theophylline can be therapeutically beneficial to spinal cord injured patients with respiratory deficits, long-term use of the drug may be compromised because of the fact that it enhances f. Thus, it is possible that over prolonged periods of theophylline therapy, respiratory muscle fatigue may limit its therapeutic benefits to patients. However, as shown from our study, a novel approach of specifically targeting peripheral structures retains the capability of theophylline to restore respiratory function after C2 hemi, and more importantly, fine-tunes theophyllineenhanced f to approximate normal discharge rates and or patterns. With the approach, the likelihood of respiratory muscle fatigue from long-term theophylline use may be offset and the therapeutic benefit retained.

Acknowledgments This work was supported by U.S. Public Health Service Grant (HD 35766). The authors would like to acknowledge the technical expertise of Warren Alilain for immunohistochemical analysis of adenosine A1 and A2A receptors and Elysia James for technical assistance in the surgical procedures involved in carotid body denervation.

References Bianchi, A.L., Denavit-Saubie, M., Champagnat, J., 1995. Central control of breathing in mammals: neuronal circuitry, membrane properties and neurotransmitters. Physiol. Rev. 75, 1 – 75. Brian, J.K., Chau, A., 1998. Fetal cardiovascular and breathing responses to an adenosine A2A receptor agonist in sheep. Am. J. Physiol. 274, R152 – R159. Carley, D.W., Radulovacki, M., 1999. Role of peripheral adenosine A1 receptors in regulation of sleep apneas in rats. Exp. Neurol. 159, 545 – 550. Coles, S.K., Miller, R., Huela, W. Schelenker, J.P.E., 2002. Frequency responses to hypoxia and hypercapnia in carotid body-denervated conscious rats. Respir. Physiol., Neurobiol. 130, 113 – 120.

Cunha, R.A., 2001. Adenosine as a neuromodulator and as a homeostatic regulator in the nervous system: different roles, different sources and different receptors. Neurochem. Int. 38 (2), 107 – 125. Eldridge, F.L., 1971. Relationship between phrenic nerve activity and ventilation. Am. J. Physiol. 221, 535 – 543. Eldridge, F.L., Milhorn, D.E., Killey, J.P., 1985. Antagonism by theophylline of respiratory inhibition induced by adenosine. J. Appl. Physiol. 59, 1428 – 1433. Ferguson, G.T., Khanchandani, N., Lattin, C.D., Goshgarian, H.G., 1999. Clinical effects of theophylline therapy on inspiratory muscle drive in tetraplegia. NeuroRehabilitation Neural Repair 13 (3), 191 – 197. Fuller, R.W., Maxwell, D.L., Conradson, T.-B.G., Dixon, C.M.S., Barnes, P.J., 1987. Circulatory and respiratory effects of infused adenosine in conscious man. Br. J. Clin. Pharmacol. 24, 309 – 317. Gauda, E.B., 2000. Expression and localization of A2A and A1-adenosine receptor genes in the carotid body and petrosal ganglia. A2A and A1adenosine receptor mRNA in the rat carotid body. Adv. Exp. Med. Biol. 475, 549 – 558. Gauda, E.B., Northington, F.J., Linden, J., Rosin, D.L., 2000. Differential expression of A2A, A1-adenosine and D2-dopamine receptor genes in rat peripheral chemoreceptors during postnatal development. Brain Res. 872 (1–2), 1 – 10. Gleeson, K., Zwillich, C.W., 1992. Adenosine stimulation, ventilation and arousal from sleep. Am. Rev. Respir. Dis. 145, 453 – 457. Goshgarian, H.G., 1979. Developmental plasticity in the respiratory pathway of the adult rat. Exp. Neurol. 66, 547 – 555. Goshgarian, H.G., Guth, L., 1977. Demonstration of functionally ineffective synapses in the guinea pig spinal cord. Exp. Neurol. 57, 211 – 225. Goshgarian, H.G., Moran, M.F., Prcevski, P., 1986. Effect of cervical spinal cord hemisection and hemidiaphragm paralysis on arterial blood gases, pH, and respiratory rate in the adult rat. Exp. Neurol. 93 (2), 440 – 445. Hadley, S.D., Walker, P.D., Goshgarian, H.G., 1999. Effects of serotonin synthesis inhibitor p-CPA on the expression of the crossed phrenic phenomenon 4h following C2 spinal cord hemisection. Exp. Neurol. 160, 479 – 488. Horn, E.M., Waldrop, T.G., 1998. Suprapontine control of respiration. Respir. Physiol. 114 (3), 201 – 211. Kobayashi, S., Conforti, L., Millhorn, D.E., 2000. Gene expression and function of adenosine A2A receptor in the rat carotid body. Am. J. Physiol. 279 (2), L273 – L282. Koos, B.J., Maeda, T., Jan, C., 2001. Adenosine A(1) and A(2) receptors modulate sleep state and breathing in fetal sheep. J. Appl. Physiol. 91 (1), 343 – 350. Lugliani, R., Whipp, Seard, B.J.C., Wasserman, C., 1971. Effect of bilateral carotid-body resection on ventilatory control at rest and during exercise. N. Engl. J. Med. 285, 1105 – 1111. Ling, L., Bach, L.K., Mitchell, G.S., 1994. Serotonin reveals ineffective spinal pathways to contralateral phrenic motoneurons in spinally hemisected rats. Exp. Brain Res. 101, 35 – 43. Liou, W.W., Goshgarian, H.G., 1994. Quantitative assessment of the effect of chronic phrenicotomy on the induction of the crossed phrenic phenomenon. Exp. Neurol. 127 (1), 145 – 153. Martin-Body, R.L., Robson, G.J., Sinclair, J.D., 1985. Respiratory effects of sectioning the carotid sinus glossopharyngeal and abdominal vagal nerves in the awake rat. J. Physiol. 361, 35 – 46. McQueen, D.S., 1993. Does adenosine stimulate rat carotid body chemoreceptors? Adv. Exp. Med. Biol. 337, 289 – 293. McQueen, D.S., Ribeiro, J.A., 1981. Effect of adenosine on carotid chemoreceptor activity in the cat. Br. J. Pharmacol. 74, 129 – 136. McQueen, D.S., Ribeiro, J.A., 1983. On the specificity and type of receptor involved in carotid body activation by adenosine in the cat. Br. J. Pharmacol. 80, 347 – 354. Middlekauf, H.R., Rivkees, S.A., Raubould, H.E., Bitticaca, M., Goldhaber, J.I., Weiss, J.N., 1998. Localization and functional effects of adenosine A1 receptors in cardiac vagal afferents in adult rats. Am. J. Physiol. 274, H441 – H447.

H. Bae et al. / Experimental Neurology 191 (2005) 94–103 Millhorn, D.E., Eldridge, F.L., Kiley, J.P., Waldrop, T.G., 1984. Prolonged inhibition of respiration following acute hypoxia in glomectomized cats. Respir. Physiol. 57, 331 – 340. Monteiro, E.C., Ribeiro, J.A., 1986. Adenosine and carotid body chemoreceptor regulation of respiration in the rat. Pfluegers Arch., Suppl. 1, 116P. Monteiro, E.C., Ribeiro, J.A., 1987. Ventilatory effects of adenosine mediated by carotid body chemoreceptors in the rat. NaunynSchmiedeberg’s Arch. Pharmacol. 335, 143 – 148. Monti, D., Carley, D.W., Radulovacki, M., 1996. p-SPA, a peripheral adenosine A1 analog, reduces sleep apnea in rats. Pharmacol. Biochem. Behav. 53 (2), 341 – 345. Moreno, D.E., Yu, X.-J., Goshgarian, H.G., 1992. Identification of axon pathways which mediate functional recovery of a paralyzed hemidiaphragm following spinal cord hemisection in the adult rat. Exp. Neurol. 116, 216 – 228. Mueller, R.A., Lundberg, D.B., Breese, G.R., 1981. Alteration of aminophylline-induced respiratory stimulation by perturbation of biogenic systems. J. Pharmacol. Exp. Ther. 218, 593 – 599. Nakayama, H., Smith, C.A., Rodman, J.R., Skatrud, J.B., Dempsey, J.A., 2003. Carotid body denervation eliminates apnea in response to transient hypocapnia. J. Appl. Physiol. 94 (1), 155 – 164. Nantwi, K.D., Basura, G.J., Goshgarian, H.G., 2003. Effects of long-term theophylline exposure on recovery of respiratory function and expression of adenosine A1 mRNA in cervical spinal cord hemisected adult rats. Exp. Neurol. 182, 232 – 239. Nantwi, K.D., Goshgarian, H.G., 1998. Theophylline-induced recovery in a hemidiaphragm paralyzed by hemisection in rats contribution of adenosine receptors. Neuropharmacology 37, 113 – 121. Nantwi, K.D., Goshgarian, H.G., 2001. Alkylxanthine-induced recovery of respiratory function following cervical spinal cord injury in adult rats. Exp. Neurol. 168, 123 – 134. Nantwi, K.D., Goshgarian, H.G., 2002. Effects of specific adenosine receptor A1 and A2 agonists and antagonists in recovery of phrenic

103

motor output following upper cervical spinal cord injury in adult rats. Clin. Exp. Pharmacol. Physiol. 29, 915 – 923. Nantwi, K.D., El-Bohy, A., Goshgarian, H.G., 1996. Actions of systemic theophylline on hemdiaphragmatic recovery in rats following cervical spinal cord hemi. Exp. Neurol. 140, 53 – 59. Olson Jr., E.B., Vidruk, E.H., Dempsey, J.A., 1988. Carotid body excision significantly changes ventilatory control in awake rats. Am. J. Physiol. 64 (2), 666 – 671. Rall, T.W., 1990. Drugs used in treatment of asthma. The methylxanthines, cromolyn blue, and other agents. In: Gilman, A.G., Rall, T.W., Nies, A.S., Taylor, P. (Eds.), The Pharmacologic Basis of Therapeutics. Pergamon Press. Reid, P.G., Watt, A.H., Penny, W.J., Newby, A.C., Smith, A.P., Routledge, P.A., 1991. Plasma adenosine concentrations during adenosine-induced respiratory stimulation in man. Eur. J. Clin. Pharmacol. 40, 175 – 180. Ribeiro, J.A., Sebastiao, A.M., de Mendonca, A., 2003. Adenosine receptors in the nervous sytem: pathophysiological implications. Prog. Neurobiol. 68, 377 – 392. Richmond, G.H., 1949. Actions of caffeine and aminophylline as respiratory stimulants in man. J. Appl. Physiol. 2, 16 – 23. Rocher, A., Gonzalez, C., Almaraz, L., 1999. Adenosine inhibits L-type Ca+ current and catecholamine release in the rabbit carotid body chemoreceptor cells. Eur. J. Neurosci. 11, 673 – 681. Rudolphi, K.A., Schubert, P., Parkinson, F.E., Fredholm, B.B., 1992. Adenosine and brain ischemia. Cerebrovasc. Brain Metab. Rev. 4, 346 – 349. Watt, A.H., Routledge, P.A., 1985. Adenosine stimulates respiration in man. Br. J. Pharmacol. 20, 503 – 506. Zhou, S.Y., Goshgarian, H.G., 1999. Effects of serotonin on crossed phrenic nerve activity in cervical spinal cord hemisected rats. Exp. Neurol. 160 (2), 446 – 453. Zhou, S.Y., Goshgarian, H.G., 2000. 5-Hydrotryptophan-induced respiratory recovery after cervical spinal cord hemisection in rats. J. Appl. Physiol. 89, 1528 – 1536.