SCS | SHR | SME | SMR | UMSPC | UMST | SSR | Main Page

 

CALCIUM SIGNALING AND

CALCIUM-CONTROLLED CELLULAR FUNCTIONS

 

Stanko S. Stojilkovic, PhD, Head, Section on Cellular Signaling

Melanija Tomic, PhD, Staff Scientist
Silvana A. Andric, PhD,
Postdoctoral Fellow
Nenad Bojat, PhD,
Postdoctoral Fellow
Arturo E. Gonzales-Iglesias, PhD,
Postdoctoral Fellow
Mu-Lan He, PhD,
Postdoctoral Fellow
Zonghe Yan, MD, PhD,
Postdoctoral Fellow

 
 

We investigate the cellular signaling cascades in endocrine and neuroendocrine cells and the interactions between plasma membrane electrical events and receptor-controlled pathways. Our main objective is to elucidate the receptors and channels involved in calcium signaling and the role of calcium ions as messengers in control of signaling, secretion, and gene expression. Our approach has been to characterize calcium signaling from a biophysical, physiological, pharmacological, and molecular biology point of view. More recently, we have begun to address how the calcium signaling function in endocrine and neuroendocrine cells is related to the structures of individual receptors and channels. Our current investigations focus on the calcium signaling function of nucleotide ("purinergic") receptors, cyclic nucleotide-gated channels, and several G protein-coupled receptors.

Nucleotide-controlled calcium signaling and secretion

He, Tomic, Gonzales-Iglesias, Yan, Stojilkovic; in collaboration with Koshimizu, Zemkova

Anterior pituitary cells secrete ATP and express nucleotide-gated G protein-coupled P2 receptors (P2YRs) and cation-conducting channels (P2XRs). However, the identification of P2R subtypes and their native ligands and the distribution and function of these receptors within the secretory and nonsecretory pituitary cells remain incompletely characterized. Over the past year, we focused on the lactotroph subpopulation of cells. Our results showed that ATP and ADP, but not UTP and UDP, triggered calcium signaling in a majority of lactotrophs and prolactin (PRL) release in mixed pituitary cells. Consistent with the role of P2Rs in signaling and secretion, the actions of ATP and ADP were abolished in the presence of apyrase, an ectonucleotidase. We identified transcripts for Gq-coupled calcium-mobilizing P2Y1R, P2Y2R, P2Y4R, and P2Y6R, as well as for Gi-coupled P2Y12R, in mixed anterior pituitary cells. The ligand selectivity profile of calcium mobilization-dependent signaling and PRL secretion and the blockade of these responses by pyridoxal 5-phosphate 6-azophenyl-2',4'-disulphonic acid indicated that P2Y1R mediates the stimulatory action of ATP and ADP. Pituitary cells also express P2X2aR, P2X2bR, P2X3R, P2X4R, and P2X7R. Several lines of evidence suggested that the P2X4R subtype provides a major pathway for calcium influx-dependent signaling and PRL secretion. In ongoing experiments, we are focusing on P2R subtypes expressed in gonadotrophs and somatotrophs and on ectoATPase subtypes expressed in pituitary cells.

 

Recombinant rat P2XR channels differ with respect to their ligand preferences and channel kinetics during activation, desensitization, and recovery. However, the calcium signaling function of P2XRs and the contributions of distinct receptor subdomains to the subtype-specific behavior have been incompletely characterized. We studied the spatio-temporal characteristics of intracellular calcium signaling and its dependence on current signaling in excitable GT1 and nonexcitable HEK293 cells expressing wild-type and chimeric P2XRs. In both cell types, P2XR generated depolarizing currents during the sustained ATP stimulation, which desensitized in the following order (from rapidly desensitizing to nondesensitizing): P2X3R > P2X2b+X4R > P2X2bR > P2X2a+X4R > P2X4R > P2X2aR > P2X7R (see Figure 4.2).

HEK293 cells were not suitable for studies on P2XR-mediated calcium influx because of the coactivation of endogenously expressed calcium-mobilizing purinergic P2YRs. However, when expressed in GT1 cells, all wild-type and chimeric P2XRs responded to agonist binding with global calcium signals, which desensitized in the same order as current signals but in a significantly slower manner. The global distribution of calcium signals was present independent of the rate of current desensitization. The temporal characteristics of calcium signals were not affected by voltage-gated calcium influx and removal of extracellular sodium. Calcium signals reflected well the receptor-specific EC50s for ATP and the extracellular zinc and pH sensitivities of P2XRs. The results indicate that, when the host cells do not express other members of nucleotide receptors, intracellular calcium measurements are useful for characterizing the pharmacological properties and messenger functions of P2XRs as well as the kinetics of channel activity.

 

 

FIGURE 4.2

The receptor-specific pattern of ATP-induced current signals in HEK293 cells expressing extracellular nucleotide-activated purinergic receptors (P2XRs). The horizontal bar indicates the exposure time to 100 microM ATP, and the current traces shown are from cells clamped at -60 mV.

 

The structural bases of P2XR ligand preferences and channel kinetics have been incompletely characterized. We tested the hypothesis that the affinity of agonists for binding domain accounts for a ligand-specific desensitization pattern. We generated chimeras using receptors with sensitivity to ATP that varied in the following order: P2X4R > P2X2aR = P2X2bR >> P2X7R. Chimeras with the ectodomain Ile66-Tyr310 sequence of P2X2R and the Val61-Phe313 sequence of P2X7R in the backbone of P2X4R were expressed but formed nonfunctioning channels. P2X2a+X4R and P2X2b+X4R chimeras with the Val66-Tyr315 ectodomain sequence of P2X4R in the backbones of P2X2aR and P2X2bR were functional and exhibited increased sensitivity to ligands compared with both parental receptors. The chimeras also desensitized faster than parental receptors and in a non-ligand-specific manner. However, like parental P2X2bR and P2X2aR, chimeric P2X2b+X4R desensitized more rapidly than P2X2a+X4R, and the rate of desensitization of P2X2a+X4R increased by substituting its Arg371-Pro376 intracellular C-terminal sequence with the Glu376-Gly381 sequence of P2X4R. The results indicate the relevance of interaction between ectodomain and flanking regions around the transmembrane domains on ligand potency and receptor activation. Furthermore, the ligand potency positively correlated with the rates of receptor desensitization but did not affect the C-terminal-specific pattern of desensitization. We currently focus on crystallization and modeling of the ectodomain of P2X4R.
 

He M-L, Koshimizu T, Tomic M, Stojilkovic SS. Purinergic P2X2 receptor desensitization depends on

coupling between ectodomain and C-terminal domain. Mol Pharmacol 2002;62:1187-1197.

He M-L, Zemkova H, Koshimizu T, Tomic M, Stojilkovic SS. Intracellular calcium measurements as

a method in studies on activity of purinergic P2X receptor-channels. Am J Physiol Cell Physiol 2003;285:C467-C479.

He M-L, Zemkova H, Stojilkovic SS. Dependence of purinergic P2X receptor activity on ectodomain

structure. J Biol Chem 2003;278:10182-10188.

Koshimizu T, Ueno S, Tanoue A, Yanagihara N, Stojilkovic SS, Tsujimoto G. Heteromultimerization

modulates P2X receptor functions through participating extracellular and C-terminal subdomains. J Biol Chem 2002;277:46891-46899.

Cyclic nucleotides and calcium signaling and secretion

Andric, Gonzales-Iglesias, Kostic,a He, Tomic, Stojilkovic

The interrelationship between spontaneous and receptor-controlled electrical activity and cyclic nucleotide signaling has not been clarified in pituitary cells. Our results indicate that the cells re action potentials spontaneously and that the associated calcium influx is sufficient to trigger PRL release, whereas removal of extracellular calcium abolishes pacemaking activity and basal PRL release. Spontaneous calcium transients and PRL release were also abolished by removal of extracellular sodium, but were not affected by tetrodotoxin, a blocker of voltage-gated sodium channels. In addition, nickel, a blocker of T-type calcium channels and cyclic nucleotide-gated cation (CNG) channels, inhibited spontaneous calcium transients and PRL release. Consistent with the role of CNG channels in pacemaking, high calcium and magnesium inhibited calcium transients and PRL release. Furthermore, activation of adenylyl cyclase by forskolin and inhibition of phosphodiesterases by 3-isobutyl-1 methylxanthine (IBMX) initiated calcium transients in quiescent lactotrophs and increased the frequency of spiking and PRL release. Inhibition of protein kinase A did not affect the stimulatory actions of forskolin and IBMX on calcium signaling and PRL release. The mRNAs encoding the alpha subunits of RCNG1, OCNG1, and CCNG1 of channels were detected in mixed pituitary cells by RT-PCR, whereas GH3 immortalized pituitary cells expressed only RCNG1. The messages for the hyperpolarization-activated CNG channel (HCN) isoforms were also detected in pituitary and GH3 cells, with HCN3 and HCN4 subtypes expressed abundantly. Our current investigations focus on electrophysiological characterization of CNG and HCN channels in lactotrophs, gonadotrophs, and somatotrophs.

In addition to forskolin, cholera toxin and several Gs protein-coupled receptors stimulated calcium signaling and increased cAMP/cGMP levels, whereas Gi/o protein-coupled receptors inhibited spontaneous calcium transients and decreased cAMP/cGMP levels. The stimulatory action of adenylyl cyclase activators on calcium signaling and the parallelism in the changes of cAMP and cGMP levels is consistent with the literature's reports that calcium stimulates nitric oxide (NO) synthase activity. Furthermore, we showed the expression of endothelial and neuronal NO synthases in pituitary cells as well as the expression of NO-sensitive soluble guanlylyl cyclase (sGC). Surprisingly, we found that the abolition of calcium signaling did not block receptor-controlled up- and down-regulation of cyclic nucleotide accumulation, suggesting that cAMP production affects cGMP accumulation. Inhibition of NO synthase activity abolished basal and stimulated cGMP production, but the rise in NO levels above basal was not essential for agonist- and forskolin-induced increase in cGMP production. Adenylyl cyclase-dependent cGMP production was mimicked by expression of a constitutive active catalytic subunit of protein kinase A and accompanied by phosphorylation of the native and recombinant alpha-1-sGC subunit. Inhibition of the phosphorylation by a protein kinase A inhibitor and a dominant negative mutant of regulatory protein kinase A subunit abolished adenylyl cyclase-dependent cGMP production without affecting basal and NO-donor stimulated cGMP production. The results indicate that phosphorylation of the alpha-1 subunit of sGC by protein kinase A enhances the NO-dependent sGC activity, most likely by stabilizing the NO/sGC complex. Our current investigations focus on identification of residues that are phosphorylated by protein kinase A.

The coupling between the NO-cGMP signaling pathway and PRL release in pituitary lactotrophs has been previously established. However, the messenger that mediates the action of this signaling pathway on hormone secretion and the affected calcium-dependent or -independent secretory mechanism have not yet been identified. Inhibition of constitutively expressed neuronal NO synthase decreased NO and cGMP levels and increased basal PRL release. The addition of a slowly releasable NO donor increased cGMP levels and inhibited basal PRL release in a time-dependent manner. Expression of inducible NO synthase also increased NO and cGMP levels and inhibited basal, depolarization-induced, and TRH-induced PRL release, whereas inhibition of the enzyme decreased NO and cGMP production and recovered PRL release. None of these treatments affected spontaneous and stimulated voltage-gated calcium influx. At basal NO levels, the addition of permeable cGMP analogs did not inhibit PRL secretion. At elevated NO levels, inhibition of cGMP production and facilitation of its degradation did not reverse inhibited PRL secretion. These experiments indicate that NO inhibits calcium-dependent PRL secretion in a cGMP-independent manner and downstream of voltage-gated calcium influx.

Andric SA, Gonzalez-Iglesias AE, Tomic M, Stojilkovic SS. Nitric oxide inhibits prolactin secretion in pi

tuitary cells downstream of voltage-gated calcium influx. Endocrinology 2003;144:2912-2921. Kostic TS, Andric SA, Stojilkovic SS. Receptor-controlled phosphorylation of alpha-1 soluble guany-lyl

cyclase enhances nitric oxide-dependent cGMP production in pituitary cells. Mol Endocrinol 2003;Nov 20 [Epub ahead of print].

Kostic TS, Tomic M, Andric SA, Stojilkovic SS. Calcium-independent and cAMP-dependent modula

tion of soluble guanylyl cyclase activity by G protein-coupled receptors in pituitary cells. J Biol Chem 2002;277:16412-16418.

Stojilkovic SS. Calcium signaling. In: The Encyclopedia of Hormones 2003; in press.

Receptor-controlled calcium mobilization and secretion

Tomic, Andric, Zivadinovic,b Van Goor,c Stojilkovic; in collaboration with Basta

Several G protein-coupled receptors expressed in lactotrophs stimulate or inhibit spontaneous voltage-gated calcium influx and PRL secretion. Here we focus on two calcium-mobilizing receptors: thyrotropin-releasing hormone (TRH) and endothelin-A (ETA). The TRH receptor is expressed in lactotrophs and GH immortalized cells and is coupled to the phospholipase C signaling pathway through Gq/11 protein. Activation of the receptors leads to the rapid increase in calcium mobilization from intracellular stores and PRL release accompanied by sustained facilitation of calcium influx and secretion. Whereas the calcium-mobilizing action of TRH receptors is well defined, the mechanism of facilitation of sustained calcium influx and PRL secretion has not been clarified. Consistent with the hypothesis that spontaneously active Kir and/or erg channels participate in the control of pacemaking in lactotrophs, we observed the stimulatory effects of cesium on cytosolic calcium in a fraction of cells. Others have shown the expression of mRNA for cesium-sensitive Kir1.0 and 2.0 subfamilies and cesium-sensitive erg channels in rat pituitary and immortalized GH3 cells. Furthermore, it has been suggested that the stimulatory action of calcium-mobilizing TRH agonist on calcium influx in lactotrophs occurs through partial inhibition of Kir and/or erg channels. In our experiments, blockers of these channels did not affect calcium signaling and PRL release. A blocker of M channels was also ineffective in inhibiting TRH-induced biphasic calcium response. The lack of effects of these blockers on TRH-induced calcium signaling and secretion does not argue against the agonist action on currents observed by others but rather indicates that the currents are not critical for generating the agonist-specific bidirectional and biphasic patterns of signaling and secretion.

 

Like TRH, the ETA receptor also induces rapid calcium release from intracellular stores and PRL secretion but, in contrast to TRH, sustained activation of ETA receptor leads to inhibition of voltage-gated calcium influx and PRL secretion. Both the activation of adenylyl cyclase by forskolin and the addition of cell-permeable 8Br-cAMP stimulated PRL secretion but did not affect the ET-1-induced sustained inhibition of voltage-gated calcium influx, suggesting that the cAMP-protein kinase A signaling pathway does not mediate the inhibitory action of ET-1 on calcium influx. However, the addition of ET-1, an ETA receptor agonist, induced depolarization of cells and enhancement of calcium influx upon calcium mobilization in lactotrophs treated overnight with pertussis toxin. Consistent with the role of pertussis toxin-sensitive potassium channels in ET-1-induced hyperpolarization of controls but not in pertussis toxin-treated cells, ET-1 decreased the cell input resistance and activated cesium-sensitive potassium current. On the other hand, apamin and paxilline, specific blockers of calcium-activated SK- and BK-type potassium channels, respectively, E-4031, a blocker of the ether a-go-go potassium channel, and linopirdine, a blocker of the M-type potassium channel, did not affect the agonist-specific patterns of calcium signaling and PRL secretion. The results suggest that ET-1 inhibits voltage-gated calcium influx through activation of cesium-sensitive channels, presumably the Gi/o-controlled inward rectifier potassium channels, and that the agonist also inhibits PRL release, but downstream of calcium influx. Currently, we are analyzing the effects of ET-1 on PRL release in pertussis toxin-treated cells.

In collaboration with Milan Basta, we also examined the relevance of high-dose intravenous immunoglobulin (IVIG) on calcium signaling by scavenging complement fragments C3b and C4b in order to test the hypothesis that exogenous immunoglobulin molecules also bind to the anaphylatoxins C3a and C5a and thereby neutralize their proinammatory effects. Single-cell calcium measurements in human HMC-1 mast cells showed that a rise in intracellular calcium caused by C3a and C5a is inhibited in a concentration-dependent manner with IVIG, F(ab)'2-IVIG, and nonrelated human monoclonal antibody. Exogenous immunoglobins also suppressed C3a/C5a-induced thromboxane generation and histamine release from HMC-1 cells and whole-blood basophils. In a mouse model of asthma, immunoglobulin treatment reduced cellular migration to the lung. Lethal C5a-mediated circulatory collapse in pigs was prevented by pretreatment with F(ab)'2-IVIG. Molecular modeling, surface plasmon resonance, and Western blot analyses suggest physical association between anaphylatoxins and the constant region of F(ab)'2. Such binding could interfere with the role of C3a and C5a in inflammation.

Basta M, Van Goor F, Luccioli S, Billings EM, Vortmeyer AO, Baranyi L, Szebeni J, Alving CR, Carroll MC, Berkower I,

Stojilkovic SS, Metcalfe DD. F(ab)'2-mediated neutralization of C3a and C5a anaphylatoxins: a novel effector function of immunoglobulins. Nature Med 2003;9:431-438.

Stojilkovic SS. Ion channels and electrical signaling. In: Conn PM, ed. Neuroscience in Medicine, 2nded., Totowa, NJ: Humana

Press, 2003;37-68.

Tomic M, Andric SA, Stojilkovic SS. Dependence of prolactin release on coupling between Ca2+ mobilization and voltage-gated

Ca2+ influx pathways in rat lactotrophs. Endocrine 2003;2045-2052.

Tomic M, Van Goor F, He M-L, Zivadinovic D, Stojilkovic SS. Ca2+mobilizing endothelin-A receptors inhibit voltage-gated Ca2+

influx through Gi/o signaling pathway in pituitary lactotrophs. Mol Pharmacol 2002;61:1329-1339.

Zivadinovic D, Tomic M, Yuan D, Stojilkovic SS. Cell-type specific messenger functions of extracellular calcium in the anterior

pituitary. Endocrinology 2002;143:445-455.

COLLABORATORS

Milan Basta, MD, PhD, Neuronal Excitability Section, NINDS, Bethesda MD

Taka-aki Koshimizu, MD, PhD, National Research Institute for Child Health and Development, Tokyo, Japan

Hana Zemkova, PhD, Czech Academy of Sciences, Prague, Czech Republic

aTatjana S. Kostic, PhD, former Postdoctoral Fellow

cFredrick Van Goor, PhD, former Postdoctoral Fellow

bDragoslava Zivadinovic, PhD, former Postdoctoral Fellow

 

For further information, contact stankos@helix.nih.gov