Soluble adenylyl cyclase in the locus coeruleus

Respiratory Physiology & Neurobiology 201 (2014) 34–37

Contents lists available at ScienceDirect

Respiratory Physiology & Neurobiology journal homepage: www.elsevier.com/locate/resphysiol

Soluble adenylyl cyclase in the locus coeruleus Ana Rita Nunes a,b , Emilia Monteiro b , Estelle Gauda a,∗ a b

Department of Pediatrics, Johns Hopkins Medical Institutions, Baltimore, MD, USA Centro de Estudos de Doenc¸as Crónicas (CEDOC), Nova Medical School, Universidade Nova de Lisboa, Portugal

a r t i c l e

i n f o

Article history: Accepted 28 May 2014 Available online 25 June 2014 Keywords: Cyclic adenosine monophosphate (cAMP) Locus coeruleus (LC) Soluble adenylyl cyclase (sAC) Transmembrane adenylyl cyclase (tmAC) Central chemoreceptors Bicarbonate/CO2 (HCO3 − /CO2 )

a b s t r a c t Although it has been demonstrated that the CO2 -sensitivity in the locus coeruleus (LC) is mediated by changes in pH, the involvement of HCO3 − in the CO2 -detection mechanism in these neurons cannot be excluded. In the present work, we characterized sAC for the first time in the LC and we asked whether this enzyme is important in the detection of changes in HCO3 − /CO2 levels in these neurons, using an approach that allowed us to isolate CO2 from pH stimulus. sAC mRNA expression and activity were upregulated from 0 mM HCO3 − /0% CO2 to 24 mM HCO3 − /5% CO2 in the LC but not in the cortex of the brain. Comparing the effects of sAC and tmAC inhibitors in the LC, we observed that both tmAC and sAC contribute to the generation of cAMP during normocapnic conditions but only sAC contributed to the generation of cAMP during isohydric hypercapnia. Furthermore, activation of tmAC induced an increase in sAC expression in LC, but not cortex. sAC may be involved in CO2 sensitivity in the LC, up to its threshold of saturation, with a particular contribution of this enzyme in situations when low HCO3 − concentrations occur. Its role should be further explored in pathological states to determine whether sAC activation with HCO3 − alters ventilation. © 2014 Elsevier B.V. All rights reserved.

1. Introduction The main players involved in the CO2 /pH-ventilatory response are dispersed in distinct regions throughout the brainstem, including the locus coeruleus (LC) (Nattie, 1999). The LC is located at the ponto medullary border, underneath the fourth ventricle in the pons. It is rich in catecholaminergic neurons and over 80% of its neuronal population is CO2 /pH-sensitive (Filosa et al., 2002). Yet, the CO2 -sensing mechanism in the LC is not fully understood. CO2 is in equilibrium with bicarbonate ions (HCO3 − ) and H+ through the action of the carbonic anhydrase, and therefore, the nature of the primary stimuli remains unclear. Although classically it had been demonstrated that the acidic stimulus (H+ ) induces an increase in the firing rate of these neurons (Johnson et al., 2008), evidences for an involvement of intracellular HCO3 − in the CO2 /pH sensing mechanism have been shown, mainly through the activation of L-type Ca2+ channels (Filosa and Putnam, 2003). Recent findings showed that L-type Ca2+ currents are activated by increases in CO2 through a HCO3 − dependent mechanism (Imber and Putnam, 2012). These finding suggest that intracellular HCO3 − levels could induce activation of soluble adenylyl cyclase (sAC) with consequent elevation

of cAMP, PKA activation and phosphorylation of Ca2+ channels. Thus sAC could be a potential HCO3 − chemosensitive transducer in these neurons but its identification and characterization was never performed in the LC. Moreover a putative role of sAC/cAMP in cerebrovascular reactivity to CO2 can be also considered in the central nervous system. sAC is activated not only by HCO3 − but also by Ca2+ ions (Chen et al., 2000) and its activity seems to be important on tissues which function depends on changes on HCO3 - /CO2 levels (Tresguerres et al., 2011). Besides the sAC, there are nine different transmembrane AC isoforms (tmAC), which are sensitive to G-protein, forskolin (FSK), Ca2+ -signaling pathways and molecular CO2 (Halls and Cooper, 2011). The effect of CO2 on tmAC and sAC activity in the LC has not been reported. Thus, the aim of this work is to identify and to investigate the potential role of sAC and tmAC in the LC, through its activation by changes in HCO3 − /CO2 , in order to clarify whether HCO3 − /CO2 is an independent stimulus from pH in the CO2 sensitive mechanism in these neurons.

2. Materials and methods 2.1. Surgical procedures

∗ Corresponding author at: Department of Pediatrics, Division of Neonatology, Johns Hopkins Medical University, 600 N. Wolfe Street, CMSC 6-104, Baltimore, MD 21287-3200, USA. Tel.: +1 410 614 7232; fax: +1 410 614 8388. E-mail address: [email protected] (E. Gauda). http://dx.doi.org/10.1016/j.resp.2014.05.011 1569-9048/© 2014 Elsevier B.V. All rights reserved.

The Animal Care and Use Committee at the Johns Hopkins University School of Medicine approved all experimental protocols. Sprague-Dawley rats (Charles River Laboratories, Wilmington, MA)

A.R. Nunes et al. / Respiratory Physiology & Neurobiology 201 (2014) 34–37

35

of both sexes at 16 and 17 postnatal days were anesthetized briefly with isoflurane and immediately decapitated. The brainstem was removed en bloc and sliced manually in ∼500 ␮m increments with a micro-knife. LC samples were obtained from pontine slices at the level of the 4th ventricle and the tract of 7th cranial nerve. The LC and adjacent tissue was bilaterally excised from the slice. Punches of the cortex were obtained and used as control tissue because it is not believed to be chemosensitive to changes in HCO3 - /CO2 .

20 s, 62 ◦ C for 20 s, 72 ◦ C for 20 s, and a terminal extension period (72 ◦ C, 10 min). The specificity of the qRT-PCR product was analyzed by performing a melting curve with 0.5 ◦ C increments in temperature. Product formation during the exponential phase of the reaction was analyzed for relative quantification to reference gene based on the threshold cycle (CT ) for amplification as 2(CT ) , where CT = CT,reference − CT,target . Alternatively, we also analysed gene expression by the Pfaffl method (Pfaffl, 2001).

2.2. HCO3 − /CO2 -regulated sAC mRNA gene levels

2.5. HCO3 − /CO2 -regulated cAMP levels

The effect of different HCO3 − /CO2 concentrations (mM of HCO3 − /% of CO2 : 0/0, 24/5, 44/10) on sAC mRNA expression levels was tested in LC and cortex. The samples were pre-incubated for 30 min in 0 mM HCO3 − /0% CO2 /60% O2 Krebs modified solution (Nunes et al., 2013). Then, the samples were placed into one of the following fresh incubation media: (1) 0 mM HCO3 − /0% CO2 /60% O2 , (2) 24 mM HCO3 − /5% CO2 /60% O2 , and (3) 44 mM HCO3 − /10% CO2 /60% O2 , for 1 h at 37 ◦ C, pH 7.4. Changes in CO2 and HCO3 − concentrations were followed by adjustments in the osmolarity (NaCl) by adding NaCl, to allow the assessment of CO2 effect, without changes in pH. Tissues were processed for quantitative real time-PCR (qRT-PCR) as described below (Nunes et al., 2013).

We tested the effect of different concentrations of HCO3 − /CO2 on cAMP production in the LC and cortex. The tissues were preincubated in 0 mM HCO3 − /0%CO2 Krebs modified solution with MDL-12,330A (500 ␮M), a tmAC inhibitor. After 30 min, the tissues were placed in one of the following media: 0 mM HCO3 − /0% CO2 ; 12 mM HCO3 − /2.5% CO2 ; 24 mM HCO3 − /5% CO2 ; or 44 mM HCO3 − /10% CO2 , for 30 min, pH 7.4, 37 ◦ C, in the presence of MDL12,330A (500 ␮M). All the experiments were conducted in the presence of IBMX (500 ␮M, a non-specific inhibitor for phosphodiesterase). Tissues were then processed for cAMP extraction and quantification by enzyme immunoassay (EIA, RPN 2255, GE Healthcare Bio-Sciences AB, Piscataway, NJ) as described before (Nunes et al., 2013).

2.3. Forskolin-regulated sAC mRNA gene expression We investigated the effect of forskolin (FSK), a tmAC activator, on the regulation of sAC gene expression in LC and cortex. The tissues were pre-incubated for 30 min in 24 mM HCO3 − /5% CO2 /60% O2 media then placed in fresh 24 mM HCO3 − /5% CO2 /60% O2 media in the presence or absence of FSK (100 ␮M), for 1 h at 37 ◦ C, pH 7.4. Tissues were processed for qRT-PCR as described below (Nunes et al., 2013). 2.4. Quantitative real-time (qRT) PCR LC and cortex homogenate tissues used to determine the level of sAC gene expression were processed to obtain total RNA (Micro-to-Midi Total RNA Purification, Invitrogen, Carlsbad, CA) according to the manufacturer’s instructions. DNase treatment (PureLink DNase, Invitrogen) was performed to avoid genomic DNA contamination. RNA yield and quality was measured at 260 and 280 nm using a UV spectrophotometer (Beckman Du 530). Total RNA (about 1 ␮g) was used for first-strand cDNA synthesis using an iSCRIPt cDNA synthesis kit (Bio-Rad Laboratories, Hercules, CA). The primer sequences used to assess sAC expression were: sense 5 -catgagtaaggaatggtggtactca-3 and anti-sense 5 -gtgagggttcacccacttgt-3 (Nunes et al., 2013). Relative expression levels for sAC gene between tissues and conditions were standardized using the reference gene glucose-6-phosphate dehydrogenase (G6PDH): sense 5 -gaagcctggcgtatcttcac-3 and anti-sense 5 -gtgagggttcacccacttgt-3 (Nunes et al., 2013). Results were further validated using two additional reference genes: glyceraldehyde 3-phosphate dehydrogenase (GAPDH) and betaactin (␤-actin) (Nunes et al., 2013). The expression of the three reference genes in LC and cortex was evaluated by combining four different algorithms used to determine the stability of the reference genes: geNorm, Normfinder, Bestkeeper, and the comparative CT method (http://www.leonxie.com/ referencegene.php?type=reference).qRT-PCR was performed using a MyiQ iCycler RT-PCR system (Bio-Rad) with the SYBR Green detection system. Each PCR reaction consisted of 1 ␮l of cDNA and 3.5 ␮l of 300 nM primers diluted in DEPC-H2 O and 10 ␮l of SYBR Green Supermix (Bio-Rad) for a final volume of 20 ␮l. Triplicates were performed for each sample. PCR conditions for sAC expression were: denaturing 5 min at 95 ◦ C, followed by 40 cycles at 95 ◦ C for

2.6. Contribution of sAC and tmAC on cAMP levels in response to normocapnia and isohydric hypercapnia To study the contribution of tmAC and sAC on cAMP accumulation under normocapnia and isohydric hypercapnia, we pre-incubated LC tissues in 24 mM HCO3 − /5% CO2 Krebs modified media, with or without MDL-12,330A (500 ␮M) or KH7 (100 ␮M, a selective sAC inhibitor) or both. After 30 min the tissues were placed either in 24 mM NaHCO3 /5%CO2 or 44 mM NaHCO3 /10%CO2 , with IBMX (500 ␮M), in the presence or absence of MDL-12,330A (500 ␮M), KH7 (100 ␮M), or both, for 30 min at 37 ◦ C. Tissues were then processed for cAMP extraction and quantification by enzyme immunoassay (EIA, RPN 2255, GE Healthcare Bio-Sciences AB, Piscataway, NJ) (Nunes et al., 2013). 2.7. Data analysis and statistical procedures The data are represented as mean ± SEM and differences between the experimental groups were determined using statistical software from GraphPad Prism (GraphPad Software Inc., version 4, San Diego, CA) or SPSS (SPSS Inc., version 12, Chicago, IL). Statistical significance was set at p < 0.05. 3. Results 3.1. HCO3 − /CO2 -regulated sAC mRNA gene levels Augments of HCO3 − /CO2 levels from 0 mM HCO3 − /0%CO2 to 24 mM HCO3 − /5%CO2 up-regulated sAC gene expression in the LC but higher concentrations did not induce further changes (Fig. 1A). This result was not observed in the cortex, a central non-chemoreceptor tissue, where increases in HCO3 − /CO2 did not induce significant changes on the level of sAC mRNA gene expression (Fig. 1A). These findings suggest that sAC could be functional in the LC. 3.2. HCO3 − /CO2 -regulated cAMP levels We further investigated the effect of different concentrations of HCO3 − /CO2 (mM/%: 0/0,12/2.5, 24/5, 44/10) on cAMP levels, as an indirect measurement of sAC activity, in the presence of a

36

A.R. Nunes et al. / Respiratory Physiology & Neurobiology 201 (2014) 34–37

Fig. 1. Effect of increasing HCO3 − /CO2 concentrations on (a) sAC gene expression and (b) cAMP levels in the locus coeruleus (LC) and cortex (Cx). O, 24 and 44 correspond to 0 mM HCO3 − /0%CO2 , 24 mM HCO3 − /5%CO2 , 44 mM HCO3 − /10%CO2 , respectively.

3.3. Contribution of sAC and tmAC on cAMP levels in response to normocapnia and isohydric hypercapnia To investigate the contribution of sAC and tmAC to HCO3 − /CO2 sensing mechanism in the LC, we compared the effect of sAC and tmAC inhibitors on cAMP levels. MDL-12,330A (500 ␮M) reduced cAMP levels by 49.2 ± 8.8% (n = 16) under normocapnia and by −0.7 ± 14.8% (n = 16) under isohydric hypercapnia. KH7 (100 ␮M) reduced cAMP levels by 23.3 ± 11.8% (n = 15) under normocapnia and by 51.6 ± 11.3% (n = 8) under isohydric hypercapnia. The effect on cAMP levels in the presence of both inhibitors in both normocapnia and isohydric hypercapnia was 66.7 ± 5.9% (n = 7) and 36.5 ± 23.2% (n = 9), respectively (Fig. 2). These results suggest that during normocapnic conditions both tmAC and sAC contribute to the generation of cAMP (thereby potentially modifying the activity of chemoreceptive neurons in the LC). During elevated HCO3 − /CO2 only sAC contributes to cAMP generation.

3.4. Contribution of tmAC on the regulation of sAC expression We have shown above that changes from 0 mM HCO3 − /0%CO2 to normocapnic conditions up-regulated sAC gene expression in the LC but not in cortex (Fig. 1A). Here, we investigated whether cAMP produced by tmAC contributes to the regulation of sAC gene expression, in normocapnic conditions. We observed that FSK (100 ␮M) increased sAC gene expression in LC by 57% (n = 9, p = 0.00, Mann–Whitney test) but not in the cortex (n = 5, p = 0.98, Mann–Whitney test). These findings suggested to us that a cross talk between sAC and tmAC activity may occur in the LC.

4. Discussion We show for the first time that soluble adenylyl cyclase is expressed and functional in the LC, a putative CO2 central chemoreceptor. Our findings suggest a relationship between sAC activity/cAMP generation and intracellular bicarbonate concentrations. sAC and tmAC contribute to the accumulation of cAMP in isohydric hypo- and normocapnic conditions, saturating at normocapnia at which point only sAC contributes to cAMP accumulation. Our results open door to study the effect of HCO3 − /sAC/cAMP in the CO2 central chemosensitivity. These findings do not invalidate the classical assumption of acidic-mediated CO2 detection in central chemoreceptors (Johnson et al., 2008) and supports the involvement of more than one CO2 transducer. Although intracellular pH is a key transducer for CO2 chemoreception, there is evidence that CO2 and HCO3 − (independent of pH) can also contribute in different contexts (Filosa and Putnam, LC_24mM HCO3-/5%CO2 LC_44mM HCO3- /10%CO2

0

MDL-12,330A 500 μM

MDL 12,330A 500μM + KH7 100 μM

KH7 100 μM

-10

% effect of AC inhibitors in LC

tmAC inhibitor, MDL-12330A (500 ␮M), and a PDE inhibitor, IBMX (500 ␮M). Increases from 0 mM/0%CO2 to 24 mM/5%CO2 (normocapnic conditions) augmented cAMP levels in the LC, but not in the cortex (Fig. 1B). Higher concentrations of HCO3 − /CO2 (44 mM HCO3 − /10%CO2 , isohydric hypercapnia) had no additional effect on cAMP levels in the central chemosensitive tissues (Fig. 1B). These results suggested to us that sAC activity perhaps reached a level of saturation under physiological conditions, 24 mM HCO3 − /5%CO2 (normocapnia conditions), in the LC and that control of sAC will be particularly important in hypocapnic situations.

(16)

-20 -30

(15)

-40 -50 -60

(16)

-80

(9)

(8)

-70

**

(7)

-90 -100

Fig. 2. Effect of MDL-12,330A (tmAC inhibitor, 500 ␮M) and KH7 (sAC inhibitor, 100 ␮M) on cAMP levels in normocapnia (24 mM HCO3 − /5%CO2 ) and isohydric hypercapnia (44 mM HCO3 − /10%CO2 ) in locus coeruleus (LC). Paired t-test, ** p < 0.01.

A.R. Nunes et al. / Respiratory Physiology & Neurobiology 201 (2014) 34–37

2003; Imber and Putnam, 2012). Further it has been suggested that AC generation and cAMP accumulation participate in the CO2 chemoreception in the carotid body glomus cells (Summers et al., 2002) as well as centrally (Imber and Putnam, 2012), via altering L-type calcium channel activity. Here we show that CO2 /HCO3 − activate tmAC and sAC and generate cAMP in the LC; thus the LC may also show intrinsic CO2 chemosensitivity via similar mechanisms. That shows we cannot exclude the possibility that changing sAC/tmAC activity is induced by changing intracellular pH (rather than CO2 /HCO3 − per se, as has been shown by others (Filosa and Putnam, 2003)). Although we have not tested the tmAC and sAC inhibitors in hypocapnia conditions (2.5%CO2 /12 mM HCO3 − ), the high degree of cAMP inhibition in normocapnia could suggest a more important role for cAMP in the hypocapnic range. What is clear from Fig. 2 is that during normocapnic conditions both tmAC and sAC contribute to the generation of cAMP (thereby potentially modifying the activity of chemoreceptive neurons in the LC). During elevated CO2 /HCO3 − only sAC contributes to cAMP generation. These results do not rule out the possibility that PKA activity increases in the hypercapnic range, due to AC/cAMP-independent pathways. Thus this short communication emphasizes the interest of a future work looking into the effect of cAMP and sAC inhibitors on PKA activity and Ca2+ currents. Our results suggest that cAMP activity operates mainly in the hypocapnic range, and reduced AC activity may contribute to apnea. Cheyne–Stokes breathing, ventilatory overshoots during obstructive apnea and excessive periodic breathing in premature infants (Di Fiore et al., 2013) are all associated with hypocapnia prolonging the apneic event and precipitating hypoxemia. Moreover, in chronic ventilated patients, isohydric hypocapnia can occur when there is a decrease in PaCO2 compensated by a decrease in the renal reabsorption of HCO3 − . Thus, the study of the role of sAC in pathological states to determine whether its activation by HCO3 − modifies ventilation, would be of interest. The experiments were performed in homogenates of the LC so we cannot exclude the effect of sAC activity on vessels and glia. However, the absence of sAC activity in cortex homogenates (non-chemosensitive neurons plus glia and vessels) emphasizes the link between chemosensitive neurons and sAC activity. Future

37

experiments with purified neurons to look for activated PKA or other cAMP marker in neurons and glia should be performed. In conclusion, there is a relationship between HCO3 − concentrations and sAC/cAMP-mediated signaling in the LC, and although it seems not to be relevant in hypercapnia, it can be important to revert respiratory depression associated with chronic hypocapnia. Acknowledgments AR Nunes was supported by Science and Technology Foundation Fellowship (FCT, SFRH/BD/39473/2007). We would like to thank Dr. Shereé M. Johnson for her help with locus coerelus dissection. References Chen, Y., Cann, M.J., Litvin, T.N., Iourgenko, V., Sinclair, M.L., Levin, L.R., Buck, J., 2000. Soluble adenylyl cyclase as an evolutionarily conserved bicarbonate sensor. Science 289 (5479), 625–628. Di Fiore, J.M., Martin, R.J., Gauda, E.B., 2013. Apnea of prematurity—perfect storm. Respir. Physiol. Neurobiol. 189 (2), 213–222. Filosa, J.A., Dean, J.B., Putnam, R.W., 2002. Role of intracellular and extracellular pH in the chemosensitive response of rat locus coeruleus neurones. J. Physiol. 541, 493–509. Filosa, J.A., Putnam, R.W., 2003. Multiple targets of chemosensitive signaling in locus coeruleus neurons: role of K+ and Ca2+ channels. Am. J. Physiol. Cell Physiol. 284, C145–C155. Imber, A.N., Putnam, R.W., 2012. Postnatal development and activation of L-type Ca2+ currents in locus coeruleus neurons: implications for a role for Ca2+ in central chemosensitivity. J. Appl. Physiol. 112, 1715–1726. Johnson, S.M., Haxhiu, M.A., Richerson, G.B., 2008. GFP-expressing locus ceruleus neurons from Prp57 transgenic mice exhibit CO2 /H+ responses in primary cell culture. J. Appl. Physiol. 105, 1301–1311. Halls, M.L., Cooper, D.M., 2011. Regulation by Ca2+− signaling pathways of adenylyl cyclades. Cold Spring Harb. Perspect Biol. 3, a004143. Nattie, E., 1999. CO2 , brainstem chemoreceptors and breathing. Prog. Neurobiol. 59, 299–331. Nunes, A.R., Holmes, A.P., Sample, V., Kumar, P., Cann, M.J., Monteiro, E.C., Zhang, J., Gauda, E.B., 2013. Bicarbonate-sensitive soluble and transmembrane adenylyl cyclases in peripheral chemoreceptors. Respir. Physiol. Neurobiol. 188 (2), 83–93. Pfaffl, M.W., 2001. A new mathematical model for relative quantification in real-time RT-PCR. Nucleic Acids Res. 29, e45. Summers, B.A., Overholt, J.L., Prabhakar, N.R., 2002. CO (2) and pH independently modulate L-type Ca (2+) current in rabbit carotid body glomus cells. J. Neurophysiol 88, 604–612. Tresguerres, M., Levin, L.R., Buck, J., 2011. Intracellular cAMP signaling by soluble adenylyl cyclase. Kidney Int. 79 (12), 1277–1288.