It is known that aging causes marked changes in the morphology of the CBs (Hurst et al. 1985) and reduces catecholamine release and CSN output in response to hypoxia in rat isolated CBs (Conde et al. 2006). In contrast, it has been suggested that peripheral chemoreceptor activity in healthy humans is not impaired by aging (Smith et al. 2001; Pokorski et al. 2004). Functional experiments in in vivo models to deal with the gap between in vitro isolated preparations of the CBs and humans are required.

Occlusions of the common carotid artery during short periods (s) have been used as a functional model for accessing peripheral chemoreceptor responses (Alcayaga et al. 1986). The effects of aging on the cardiorespiratory reflexes induced by carotid ischemia are yet to be studied.

Dopamine has inhibitory effects on ventilation during rest (Zapata and Zuazo 1980) and in the presence of hypoxic exposure (Nishino and Lahiri 1981), which are mediated by D₂-receptors located in the CB (Gonzalez et al. 1994). No data are available concerning the effects of aging on D₂-receptors at the CB.

A₂-adenosine receptors in the CB are involved in the stimulation of breathing in basal conditions (McQueen and Ribeiro 1986; Monteiro and Ribeiro 1987) and in response to hypoxia (Conde and Monteiro 2004).

In the present work, CB function was assessed by measuring the ventilatory response induced by acute CB ischemia and exogenous administrations of adenosine, and dopamine, and their receptor antagonists. Changes in cAMP content, in response to oxygen concentrations in aged rats, were also investigated as a cellular indicator of CB function. It has been postulated that cAMP modulates the response of the CB to hypoxia, and is the common activation pathway of both dopamine and adenosine receptors in the CB (Batuca et al. 2003). The effects of aging on CB cAMP production have never been addressed. An additional reason for testing this hypothesis is the age-related alterations in the adenylate cyclase/cAMP system, which has been documented in other structures. For example, impairment of the activity of the catalytic subunit of adenylate cyclase was observed in rats (Kilts et al. 2002), and in the heart of >60-year-old humans (Brodde et al. 1995). However, on the other hand, it has been described that the maximal ability of forskolin to increase cAMP in the adrenal medulla and liver is enhanced in 24-month-old rats (Tumer et al. 1996).

The aging process is characterized by a decline in several physiological functions resulting in a reduction in the ability to maintain homeostasis (Troen 2003). However, there is increasing evidence that several physiological functions are well preserved in aging. Namely, several studies have examined the age-related changes in the ventilatory response to hypoxia and found a maintained ventilatory response throughout aging, suggesting that no alteration in peripheral chemoreception occurred in old rats (Pokorski and Antosiewicz 2010) or humans (Smith et al. 2001; Pokorski et al. 2004; Vovk et al. 2004).

The present work contributes indirectly to knowing the effect of aging in the oxygen-sensing mechanisms, and was focused in the peripheral control of ventilation mediated by the CB.

It was previously described that respiratory frequency decreases with the size increase of the animals and declines in a linear fashion with the age increase in rat (Soulage et al. 2004). Healthy elderly subjects at rest breathe with a V_E identical to that of younger subjects, but with smaller V_T and higher f_R (Janssens et al. 1999). In our model’s experimental conditions, a significant increase in rat weight was found after 12 months of age, but the basal values of f_R remained constant throughout aging. The animals were vagotomized and under the influence of sodium pentobarbitone anesthesia and consequently showed lower values of f_R (46.6±1.5 breaths min⁻¹). This could explain the absence of the bradypnea found by others in old rats where basal values of f_R were 104.7±3.4 breaths min⁻¹ (Soulage et al. 2004).

It is well known that, in general, all anesthetic regimens cause respiratory and metabolic depression. Anesthesia predominantly depresses respiratory frequency, with little or no effect on V_T, and it has been reported that pentobarbitone depresses V_E by 30% (Schwenke and Cragg 2004). Pentobarbital anesthesia was elected because it produces a steady respiratory baseline (Young 1957) and it does not affect or even potentiate the respiratory reflex reactivity to chemoreceptor stimulation (Douglas et al. 1950). Also, Biscoe and Millar (1968) felt that pentobarbital did not affected carotid chemoreceptor discharge significantly (Biscoe and Millar 1968). Furthermore, some authors found that the magnitude of the V_E response to hypoxia and hypercapnia appears to be unaltered by pentobarbitone anesthesia (Schwenke and Cragg 2004). The initial anesthetic dose given to old animals was lower (40 mg kg⁻¹ ip) than that given to young adults (60 mg kg⁻¹ ip), and in general they require fewer anesthetic supplements during the experiments to suppress pain reflexes (see details in the online methods supplement). This difference has been attributed to pharmacokinetic issues in the elderly where the rapid intercompartimental clearance is 30% lower in the old when compared to young subjects, resulting in higher concentrations and more drugs available for distribution to the brain (Bowie and Slattum 2007). Anyway, no age differences in brain responsiveness were found: no change was detected in the action of GABA_A receptor modulators like pentobarbital (Griffith et al. 2000).

The absence of changes in f_R when the input of the major pulmonary stretch receptors is abolished by vagotomy and CNS activity depressed by the barbiturate, suggest that changes in f_R in the elderly could not be attributed to an impairment of the CB peripheral drive.

Wistar rats exhibited, in the present conditions, a reduction of tidal volume with aging. Although some authors did not find tidal volume to decrease with advancing age (Soulage et al. 2004), an age-related decrease was commonly observed in rats (Nagase et al. 1994) and humans (Janssens et al. 1999) with advancing age, and has been linked to alterations of the mechanical properties of lung and thorax (Chan and Welsh 1998).

The cardiovascular effects associated with aging are: a striking attenuation in the cardiac frequency in humans (Kronenberg and Drage 1973); no significant effect of age on resting HR of rats (Gordon 2008); and increase in blood pressure in both rats (Di Nardo et al. 2009) and humans (Fleg et al. 1995). In the present work, mean arterial blood pressure was lower in 24 months old rats than in young adults. This finding could be attributed to the predominant effect of anesthetic-mediated central cardiovascular depression (Schwenke and Cragg 2004) rather than the arterial dysfunction in aging. It is known that the risk of cardiovascular depression induced by barbiturates is higher in the elderly (Schwenke and Cragg 2004) and compatible with the lower values for BP found in the present work, despite the lower doses of anesthetic were needed to abolish sensitivity in old animals.

There is a consensual body of evidence about the changes in the morphology of the CB throughout aging: increase in extracellular matrix, reduction in number and volume of type I cells, and in the volume and density of mitochondria compared with young CB (Hurst et al. 1985; Pokorski et al. 2004; Conde et al. 2006). These morphological findings do not apparently concur with the absence of functional impairment. However, the increase in the number of type II stem cells could contribute towards the maintenance of CB function (Hurst et al. 1985; Porzionato et al. 2005). The present work supports previous findings in humans—absence of alterations in the peripheral control of ventilatory responses to hypoxia (Pokorski et al. 2004; Vovk et al. 2004), providing evidence that structural and neurochemical changes caused by aging in the carotid body are further compensated. The finding that changes in cAMP content at the CB in response to O₂ concentrations did not change with aging, suggests that the compensatory mechanism could occur within the CB as a whole. These results contrast with the reduction in carotid sinus nerve discharges in response to hypoxia observed in old rats (Conde et al. 2006). It is known that the glossopharyngeal nerve has, in addition to carotid sinus nerve afferent fibers, a parallel autonomic parasympathetic efferent pathway sensitive to hypoxia that is the source of CB inhibition (Campanucci et al. 2006). We can speculate that the reduction in CSN discharges found in old animals (Conde et al. 2006) could mainly be due to an impairment of the efferent fiber activity without changes in the chemosensory excitatory afferent pathway. Similarly, it was shown in humans that aging impairs the autonomic responses to pain but does not modify nociceptive perception (Hajduczok et al. 1991).

Hyperventilation induced by bilateral brief (<15 s) occlusions of the common carotid arteries are abolished by carotid sinus nerve section (Monteiro and Ribeiro 1989), constituting an alternative model for hypoxia in order to study CB chemosensory activation in vivo. In the present work, the cardiorespiratory reflexes induced by carotid ischemia were similar in young and old rats. Previous work has addressed the impairment of carotid ischemia on cerebral oxidative balance caused by permanent ligation of the right carotid artery (Macri et al. 2010) in old animals. These are long-term ischemia and the authors did not address the effects of the ventilatory reflexes induced by carotid ischemia in aged animals, shown for the first time in the present work. Carotid sinus baroreflex function throughout aging has never been studied using this model but the persistency of carotid baroreflex function in advanced age rats was previously described in anesthetized Fisher rats (Wei et al. 1986). In humans, there is no difference in carotid body baroreflexes throughout aging (Fiocchi et al. 1985; Shi et al. 1996).

Another aspect to consider in our discussion pertains to the age-dependent variations in cAMP levels and responses to hypoxia. Aging did not cause appreciable changes in cAMP levels in the CB. The normoxic cAMP levels do not differ in the four ages studied, and the pattern of cAMP production in response to different O₂ concentrations remained almost unchanged in old animals. The cAMP levels found in the rat CB in the present study, expressed in unit weight, are nearly identical to the values reported in the rabbit CB (Perez-Garcia et al. 1990; Chen et al. 1997). At all ages, cAMP levels were maximal in normoxic conditions. Compared to normoxia, hyperoxia caused a decrease in cAMP in the CB in all ages. Adenosine effects via A₂ receptors, which are known to be expressed in the CB (Gauda 2002), can represent one of the mechanisms maintaining high cAMP levels in hypoxia. It is worth noting that the release of adenosine in the CB is maximal at mild levels of hypoxia (10%O₂), decreasing with higher intensities of hypoxic stimulation and that the most intense hypoxia used in this study does not increase the release of adenosine in the SCG or CA (Conde and Monteiro 2004). Of course adenosine is not the only modulator of cAMP levels in the CB, as it is well known that hypoxia also increases the release of many other neurotransmitters, particularly dopamine (Vicario et al. 2000). It is also known that the CB expresses high levels of D₂ dopamine receptors (Gauda 2002) which are negatively coupled to adenylate cyclase (Kebabian and Calne 1979). In fact, it would appear that the much higher rate of dopamine release in the rat vs. the rabbit CB in hypoxia (Vicario et al. 2000) could explain the inability of hypoxia to increase the cAMP levels in the rat CB above those found in normoxia, as it has been previously observed in the rabbit (Cachero et al. 1996). It should be mentioned that Mir et al. (Mir et al. 1983), in their pioneer study, also found that 5% O₂ administered in vivo did not significantly modify cAMP levels in the rat CB.

The absence of significant changes in cAMP production in old animals is not incompatible with individual changes in adenylate cyclase activity and/or metabotropic receptors but implies that if individual changes occurred the overall response is maintained.

The effects of aging in cAMP accumulation in other preparations show impairment of the activity of the catalytic subunit of adenylate cyclase in rats (Kilts et al. 2002) and in the heart of >60-year-old humans (Brodde et al. 1995). However, it has been described that the maximal ability of forskolin to increase cAMP in the adrenal medulla and liver is enhanced in 24-month-old rats (Tumer et al. 1996).

If a reduction in both A₂-adenosine (positively coupled to adenylate cyclase) and D₂-dopamine (negatively coupled to adenylate cyclase) receptor-mediated responses coexist in the CB, then compensatory effects in cAMP production can take place. In fact, the results obtained in the present work with exogenous adenosine and dopamine as well as manipulating their endogenous effects through antagonists showed that neither adenosine or dopamine receptors are significantly impaired in old animals. These dopamine and adenosine effects were mediated through carotid body chemoreceptors, seeing as these actions disappeared after bilateral section of the carotid sinus nerves.

An additional interest to test the effects of dopamine and adenosine in the present model is their specific therapeutic indications in clinical practice: shock and supraventricular tachycardia, respectively. These clinical situations are associated with heart failure, more prevalent in the elderly, and the cardiorespiratory effects of exogenous dopamine and adenosine in aged humans or animals have never been investigated.

Dopamine has been shown to impair the ventilatory drive in response to hypoxemia and depress minute ventilation and oxygen saturation in heart failure patients, even when they are breathing room air (van de Borne et al. 1998). The present work provides evidence that these undesirable effects should be of the same magnitude in aged subjects.

Domperidone is a D₂-selective antagonist that does not cross the blood–brain barrier (Baudry et al. 1979) and, administered alone, it increases basal V_E as described in previous work (Lahiri et al. 1984; Gamboa et al. 2003). We have found statistically significant differences between 3 and 24 months old rats. This means that the inhibitory basal tonus of dopamine at the CB activity is less marked in old rats, probably resulting from a reduction in the number of receptors throughout aging. This is not an unexpected finding considering that it has previously been found that the density of striatal D₂-receptors is significantly reduced in aged rats between 30% and 80% depending on the study (Marshall and Joyce 1988; Petkov et al. 1988; Han et al. 1989; Popoli et al. 1998). However, in the CB this reduction is not enough to reduce the effect of exogenous dopamine which has the same magnitude in old rats and controls.

The BP response triggered by dopamine, a short-lived rise followed to a return to baseline, occurred independent of the carotid sinus nerve afferentation and blockade of dopamine D₂-receptors. This result warrants that the prompt hypertensive response evoked by dopamine challenge apparently occurs beyond the baroreceptor afferentation and may depend on D₁-dopamine receptors at the heart and/or adrenoceptors.

Despite its dyspnoeic effect mediated by chemoreceptor activation (Burki et al. 2005), adenosine is a clinically useful tool to treat supraventricular tachycardia (Biaggioni et al. 1987; Riccardi et al. 2008), a common cardiac rhythm disturbance more frequently observed in the elderly (Medi et al. 2009). However, studies about this effect on aged subjects are yet to be made. Although the present work was not focused in the crono/dromo and batmotropic effects of adenosine, an interesting finding is the lesser bradycardic effect observed during its exogenous application in aged animals, suggesting that the efficacy of adenosine to revert supraventricular tachycardia could be attenuated in the elderly. This could be attributed to a reduction in the density of adenosine receptors in the aged heart and may also involve an age-related reduction in the intrinsic ability of nodal tissue to respond to adenosine receptor activation (Hinschen et al. 2001). Some authors find an absence of changes in A₁-receptors density and G alpha protein levels but an adenosine A₁-receptor function in rat ventricles to decrease with age and this was related to a reduction in the coupling between adenosine A₁-receptor and their G proteins (Cai et al. 1997).

The excitatory effects of adenosine found in old rats suggest that the A₂-adenosine receptors in the carotid body are well preserved and that the chest discomfort caused by its exogenous administration will remain in the elderly. The excitatory effects obtained with low doses of SCH 58261 by itself, an adenosine A_2A-antagonist that crosses the blood–brain barrier (El Yacoubi et al. 2000), are in agreement with the presence of A_2A receptors with inhibitory effects on the control of breathing (Mayer et al. 2006) and that central inhibitory control is apparently more relevant in basal conditions than the excitatory effects peripherally mediated. Therefore, low doses of A_2A antagonists that cross BBB (like SCH 58261) could be useful to treat central mediated ventilatory depressions without major actions in the cardiovascular system because the dose that antagonize the ventilatory effects did not modify BP and HR.