Involvement of STIM1 in the proteinase‐activated receptor 1‐mediated Ca2+ influx in vascular endothelial cells

Thrombin increases the cytosolic Ca²⁺ concentrations and induces NO production by activating proteinase‐activated receptor 1 (PAR₁) in vascular endothelial cells. The store‐operated Ca²⁺ influx is a major Ca²⁺ influx pathway in non‐excitable cells including endothelial cells and it has been reported to play a role in the thrombin‐induced Ca²⁺ signaling in endothelial cells. Recent studies have identified stromal interaction molecule 1 (STIM1) to function as a sensor of the store site Ca²⁺ content, thereby regulating the store‐operated Ca²⁺ influx. However, the functional role of STIM1 in the thrombin‐induced Ca²⁺ influx and NO production in endothelial cells still remains to be elucidated. Fura‐2 and diaminorhodamine‐4M fluorometry was utilized to evaluate the thrombin‐induced changes in cytosolic Ca²⁺ concentrations and NO production, respectively, in porcine aortic endothelial cells transfected with small interfering RNA (siRNA) targeted to STIM1. STIM1‐targeted siRNA suppressed the STIM1 expression and the thapsigargin‐induced Ca²⁺ influx. The degree of suppression of the STIM1 expression correlated well to the degree of suppression of the Ca²⁺ influx. The knockdown of STIM1 was associated with a substantial inhibition of the Ca²⁺ influx and a partial reduction of the NO production induced by thrombin. The thrombin‐induced Ca²⁺ influx exhibited the similar sensitivity toward the Ca²⁺ influx inhibitors to that seen with the thapsigargin‐induced Ca²⁺ influx. The present study provides the first evidence that STIM1 plays a critical role in the PAR₁‐mediated Ca²⁺ influx and Ca²⁺‐dependent component of the NO production in endothelial cells. J. Cell. Biochem. 108: 499–507, 2009. © 2009 Wiley‐Liss, Inc.     

Spectroscopic determination of sarcoplasmic reticulum Ca2+ uptake and Ca2+ release

In this report we describe the application of spectroscopic methods to the study of Ca2+ release by isolated native sarcoplasmic reticulum (SR) membranes from rabbit skeletal muscle. To date, dual-wavelength spectroscopy of arsenazo III and antipyrylazo III difference absorbance have been the most common spectroscopic methods for the assay of SR Ca2+ transport. The utility of these methods is the ability to manipulate intraluminal Ca2+ loading of SR vesicles. These methods have also been useful for studying the effect of both agonists and antagonists upon SR Ca2+ release and Ca2+ uptake. In this study, we have developed the application of Calcium Green-2, a long-wavelength excitable fluorescent indicator, for the study of SR Ca2+ uptake and release. With this method we demonstrate how ryanodine receptor Ca2+ channel opening and closing is regulated in a complex manner by the relative distribution of Ca2+ between extraluminal and intraluminal Ca2+ compartments. Intraluminal Ca2+ is shown to be a key regulator of Ca2+ channel opening. However, these methods also reveal that the intraluminal Ca2+ threshold for Ca2+-induced Ca2+ release varies as a function of extraluminal Ca2+ concentration. The ability to study how the relative distribution of a finite pool of Ca2+ across the SR membrane influences Ca2+ uptake and Ca2+ release may be useful for understanding how the ryanodine receptor is regulated, in vivo.     

Orchestration of neuronal migration by activity of ion channels,neurotransmitter receptors,and intracellular Ca2+ fluctuations

The real-time observation of cell movement in acute cerebellar slices reveals that granule cells alter their shape concomitantly with changes in the mode and rate of migration as they traverse different cortical layers. Although the origin of local environmental cues responsible for these position-specific changes in migratory behavior remains unclear, several signaling mechanisms involved in controlling granule cell movement have emerged. The onset of one such mechanism is marked by the expression of voltage-gated ion channels and neurotransmitter receptors in postmitotic cells prior to the initiation of their migration. Granule cells start their radial migration after the expression of N-type Ca²⁺ channels and the N-methyl-D -aspartate subtype of glutamate receptors on the plasmalemmal surface. Blockade of the channel or receptor activity significantly decreases the rate of cell movement, indicating that the activation of these membrane constituents provides an essential signal for the translocation of granule cells. Another signal that controls the rate of cell migration is embedded in the combined amplitude and frequency components of Ca²⁺ fluctuations in the somata of migrating granule cells. Interestingly, each phase of Ca²⁺ fluctuation controls a separate phase of saltatory movement in the granule cells: The cells move forward during the phase of transient Ca²⁺ elevation and remain stationary during the troughs. Consequently, the changes in the amplitude and frequency components of Ca²⁺ fluctuations directly affect granule cell movement: Reducing the amplitude or frequency of Ca²⁺ fluctuations slows down the speed of cell movement, while the enhancement of these components accelerates migration. These findings suggest that signaling molecules present in the local cellular milieu encountered on the migratory route control the shape and motility of granule cells by modifying Ca²⁺ fluctuations in the soma through the activation of specific ion channels and neurotransmitter receptors. © 1998 John Wiley & Sons, Inc. J Neurobiol 37: 110–130, 1998     

Calcium signaling in non-excitable cells: Ca2+ release and influx are independent events linked to two plasma membrane Ca2+ entry channels

The regulatory mechanism of Ca2+ influx into the cytosol from the extracellular space in non-excitable cells is not clear. The "capacitative calcium entry" (CCE) hypothesis suggested that Ca2+ influx is triggered by the IP(3)-mediated emptying of the intracellular Ca2+ stores. However, there is no clear evidence for CCE and its mechanism remains elusive. In the present work, we have provided the reported evidences to show that inhibition of IP(3)-dependent Ca2+ release does not affect Ca2+ influx, and the experimental protocols used to demonstrate CCE can stimulate Ca2+ influx by means other than emptying of the Ca2+ stores. In addition, we have presented the reports showing that IP(3)-mediated Ca2+ release is linked to a Ca2+ entry from the extracellular space, which does not increase cytosolic [Ca2+] prior to Ca2+ release. Based on these and other reports, we have provided a model of Ca2+ signaling in non-excitable cells, in which IP(3)-mediated emptying of the intracellular Ca2+ store triggers entry of Ca2+ directly into the store, through a plasma membrane TRPC channel. Thus, emptying and direct refilling of the Ca2+ stores are repeated in the presence of IP(3), giving rise to the transient phase of oscillatory Ca2+ release. Direct Ca2+ entry into the store is regulated by its filling status in a negative and positive manner through a Ca2+ -binding protein and Stim1/Orai complex, respectively. The sustained phase of Ca2+ influx is triggered by diacylglycerol (DAG) through the activation of another TRPC channel, independent of Ca2+ release. The plasma membrane IP(3) receptor (IP(3)R) plays an essential role in Ca2+ influx, by interacting with the DAG-activated TRPC, without the requirement of binding to IP(3).     

Endoplasmic reticulum Ca(2+) homeostasis and neuronal death

The endoplasmic reticulum (ER) is a universal signalling organelle, which regulates a wide range of neuronal functional responses. Calcium release from the ER underlies various forms of intracellular Ca²⁺ signalling by either amplifying Ca²⁺ entry through voltage-gated Ca²⁺ channels by Ca²⁺-induced Ca²⁺ release (CICR) or by producing local or global cytosolic calcium fluctuations following stimulation of metabotropic receptors through inositol-1,4,5-trisphosphate-induced Ca²⁺ release (IICR). The ER Ca²⁺ store emerges as a single interconnected pool, thus allowing for a long-range Ca²⁺ signalling via intra-ER tunnels. The fluctuations of intra-ER free Ca²⁺ concentration regulate the activity of numerous ER resident proteins responsible for post-translational protein folding and modification. Disruption of ER Ca²⁺ homeostasis results in the developing of ER stress response, which in turn controls neuronal survival. Altered ER Ca²⁺ handling may be involved in pathogenesis of various, neurodegenerative diseases including brain ischemia and Alzheimer dementia.     

T cell receptor-mediated Ca2+ signaling: Release and influx are independent events linked to different Ca2+ entry pathways in the plasma membrane

In this study, we showed that cross-linking CD3 molecules on the T cell surface resulted in Ca²⁺ release from the intracellular stores followed by a sustained Ca²⁺ influx. Inhibition of release with TMB-8 did not block the influx. However, inhibition of phospholipase C activity suppressed both Ca²⁺ release and influx. Once activated, the influx pathway remained open in the absence of further hydrolysis of PIP2. Thapsigargin, a microsomal Ca²⁺ -ATPase inhibitor, stimulated Ca²⁺ entry into the cells by a mechanism other than emptying Ca²⁺ stores. In addition, Ca²⁺ entry into the Ca²⁺ -depleted cells was stimulated by low basal level of cytosolic Ca²⁺, not by the emptying of intracellular Ca²⁺ stores. Both the Ca²⁺ release and influx were dependent on high and low concentrations of extracellular Ca²⁺. At low concentrations, Mn²⁺ entered the cell through the Ca²⁺ influx pathway and quenched the sustained phase of fluorescence; whereas, at higher Mn²⁺ concentration both the transient and the sustained phases of fluorescence were quenched. Moreover, Ca²⁺ release was inhibited by low concentrations of Ni²⁺, La³⁺, and EGTA, while Ca²⁺ influx was inhibited by high concentrations. Thus, in T cells Ca²⁺ influx occurs independently of IP3-dependent Ca²⁺ release. However, some other PIP2 hydrolysis-dependent event was involved in prolonged activation of Ca²⁺ influx. Extracellular Ca²⁺ influenced Ca²⁺ release and influx through the action of two plasma membrane Ca²⁺ entry pathways with different pharmacological and biochemical properties.     

Spatial and dye correlation analysis of intracellular Ca2+ distribution

Intracellular Ca²⁺ is an important regulator of many cellular processes. Besides ion channels and transporters in the plasmalemma, changes in [Ca]_i can be mediated by uptake and release mechanisms of internal organelles. Theoretical and experimental procedures are developed aiming to reveal the distribution of internal Ca²⁺ pools and their role in generating complicated spatial patterns of [Ca]_i gradients. Cultured pyramidal neurons from rat hippocampus were loaded with Ca²⁺-sensitive fluorescent dyes, fura-2 and fluo-3. Cell images were partitioned according to pixel amplitude and highlighted pictures were characterized by their intensity, relative area and connectivity. This approach facilitates the localization of the sites of Ca²⁺ release from internal stores induced by application of different agents. After each trial, neurons were stained with dyes, acridine orange or DiOC₆, which bind preferentially to nucleus and endoplasmic reticulum. A correlation between images confirmed the spatial localization of Ca²⁺ release sites. Application of the partition procedure also gave a clear evidence for the importance of Ca²⁺ influx in the mechanism of [Ca]_i oscillations.     

Characterization of a 1,25(OH)2-vitamin D3-responsive capacitative Ca2+ entry pathway in rat osteoblast-like cells

We investigated the existence of a capacitative Ca2+ entry (CCE) pathway in ROS 17/2.8 osteoblast-like cells and its responsiveness to 1,25-dihydroxy-vitamin D3 [1,25(OH)2D3]. Depletion of inner Ca2+ stores with thapsigargin or 1,25(OH)2D3 in the absence of extracellular Ca2+ transiently elevated cytosolic Ca2+ ([Ca2+]i); after recovery of basal values, Ca2+ re-addition to the medium markedly increased Ca2+ entry, reflecting pre-activation of a CCE pathway. Recovery of the Ca2+ overshoot that followed the induced CCE was mainly mediated by the plasma membrane Ca2+-ATPase. Addition of 1,25(OH)2D3 to the declining phase of the thapsigargin-induced CCE did not modify further [Ca2+]i, indicating that steroid activation of CCE was dependent on store depletion. Pre-treatment with 1 microM Gd3+ inhibited 30% both thapsigargin- and 1,25(OH)2D3-stimulated CCE, whereas 2.5 microM Gd3+ was required for maximal inhibition ( approximately 85%). The activated CCE was permeable to both Mn2+ and Sr2+. Mn2+ entry sensitivity to Gd3+ was the same as that of the CCE. However, 1-microM Gd3+ completely prevented capacitative Sr2+ influx, whereas subsequent Ca2+ re-addition was reduced only 30%. These results suggest that in ROS 17/2.8 cells CCE induced by thapsigargin or 1,25(OH)2D3 is contributed by at least two cation entry pathways: a Ca2+/Mn2+ permeable route insensitive to very low micromolar (1 microM) Gd3+ accounting for most of the CCE and a minor Ca2+/Sr2+/Mn2+ permeable route highly sensitive to 1 microM Gd3+. The Ca2+-mobilizing agonist ATP also stimulated CCE resembling the Ca2+/Sr2+/Mn2+ permeable entry activated by 1,25(OH)2D3. The data demonstrates for the first time, the presence of a hormone-responsive CCE pathway in an osteoblast cell model, raising the possibility that it could be an alternative Ca2+ influx route through which osteotropic agents influence osteoblast Ca2+ homeostasis. Copyright Wiley-Liss, Inc.     

Sarco-endoplasmic reticulum Ca2+ ATPase (SERCA) inhibitors identify a novel calcium pool in the central nervous system

Ca2+ uptake into the endoplasmic reticulum (ER) is mediated by Ca2+ ATPase isoforms, which are all selectively inhibited by nanomolar concentrations of thapsigargin. Using ATP/Mg2+-dependent 45Ca2+ transport in rat brain microsomes, tissue sections, and permeabilized cells, as well as Ca2+ imaging in living cells we distinguish two ER Ca2+ pools in the rat CNS. Nanomolar levels of thapsigargin blocked one component of brain microsomal 45Ca2+ transport, which we designate as the thapsigargin-sensitive pool (TG-S). The remaining component was only inhibited by micromolar thapsigargin, and thus designated as thapsigargin resistant (TG-R). Ca2+ ATPase and [32P]phosphoenzyme assays also distinguished activities with differential sensitivities to thapsigargin. The TG-R Ca2+ uptake displayed unique anion permeabilities, was inhibited by vanadate, but was unaffected by sulfhydryl reduction. Ca2+ sequestered into the TG-R pool could not be released by inositol-1,4,5-trisphosphate, caffeine, or cyclic ADP-ribose. The TG-R Ca2+ pool had a unique anatomical distribution in the brain, with selective enrichment in brainstem and spinal cord structures. Cell lines that expressed high levels of the TG-R pool required micromolar concentrations of thapsigargin to effectively raise cytoplasmic Ca2+ levels. TG-R Ca2+ accumulation represents a distinct Ca2+ buffering pool in specific CNS regions with unique pharmacological sensitivities and anatomical distributions.     

The possible involvement of protein phosphatase 1 in thrombin-induced Ca2+ influx of human platelets

Protein phosphatase 1 is considered to be involved in thrombin-induced platelet activation (Murata et al., Biochem Int 26:327–334, 1992). To clarify the mechanism, we examined the effects of protein phosphatase 1 and 2A inhibitors (calyculin A, tautomycin, okadaic acid) on Ca²⁺ influx. In the presence of 1 mM Ca²⁺, thrombin- (0.1 U/ml) induced platelet aggregation and ATP release were inhibited by calyculin A, while this inhibitory effect was abolished in the absence of Ca²⁺ (EGTA 1 mM). Furthermore, thrombin-induced Mn²⁺ influx but not intracellular Ca²⁺ mobilization was inhibited by calyculin A in a dose-related manner. Calyculin A also blocked the ongoing Ca²⁺ influx when added 3 min after thrombin stimulation. Similar inhibitory effects were observed with okadaic acid and tautomycin in the same potency sequence as the reported one for protein phosphatase 1 (calyculin A > tautomycin > okadaic acid). These results suggest that the anti-platelet effects of phosphatase inhibitors are due to the inhibition of Ca²⁺ influx and that protein phosphatase 1 plays a key role in the regulation of receptor operated Ca²⁺ channel of human platelets.     

The activation state of the inositol 1,4,5-trisphosphate receptor regulates the velocity of intracellular Ca2+ waves in bovine aortic endothelial cells

Ca(2+) is a highly versatile second messenger that plays a key role in the regulation of many cell processes. This versatility resides in the fact that different signals can be encoded spatio-temporally by varying the frequency and amplitude of the Ca(2+) response. A typical example of an organized Ca(2+) signal is a Ca(2+) wave initiated in a given area of a cell that propagates throughout the entire cell or within a specific subcellular region. In non-excitable cells, the inositol 1,4,5-trisphosphate receptor (IP(3) R) is responsible for the release of Ca(2+) from the endoplasmic reticulum. IP(3) R activity can be directly modulated in many ways, including by interacting molecules, proteins, and kinases such as PKA, PKC, and mTOR. In the present study, we used a videomicroscopic approach to measure the velocity of Ca(2+) waves in bovine aortic endothelial cells under various conditions that affect IP(3) R function. The velocity of the Ca(2+) waves increased with the intensity of the stimulus while extracellular Ca(2+) had no significant impact on wave velocity. Forskolin increased the velocity of IP(3) R-dependent Ca(2+) waves whereas PMA and rapamycin decreased the velocity. We used scatter plots and Pearson's correlation test to visualize and quantify the relationship between the Ca(2+) peak amplitude and the velocity of Ca(2+) waves. The velocity of IP(3) R-dependent Ca(2+) waves poorly correlated with the amplitude of the Ca(2+) response elicited by agonists in all the conditions evaluated, indicating that the velocity depended on the activation state of IP(3) R, which can be modulated in many ways.     

Relative contribution of the Na(+)/Ca(2+) exchanger, mitochondria and endoplasmic reticulum in the regulation of cytosolic Ca(2+) and catecholamine secretion of bovine adrenal chromaffin cells

The relative importance of mitochondria, the Na(+)/Ca(2+) exchanger (NCX) and the endoplasmic reticulum (ER) in the regulation of the cytosolic Ca(2+) concentration ([Ca(2+)](i)) were examined in bovine chromaffin cells using fura-2 for average [Ca(2+)](i) and amperometry for secretory activity, which reflects the local Ca(2+) concentration near the exocytotic sites. Chromaffin cells were stimulated by a high concentration of K(+) when the three Ca(2+) removal mechanisms were individually or simultaneously inhibited. When the mitochondrial Ca(2+) uptake was inhibited, the [Ca(2+)](i) decayed at a significantly slower rate and the secretory activity was higher than the control cells. The NCX appears to function only in the initial phase of [Ca(2+)](i) decay and when the ER Ca(2+) pump is blocked. Similarly, the ER had a significant effect on the [Ca(2+)](i) decay and on the secretion only when the NCX was blocked. Inhibition of all three mechanisms leads to a substantial delay in [Ca(2+)](i) recovery and an increase in the secretion. The results suggest that the three mechanisms work together in the regulation of the Ca(2+) near the Ca(2+) channels and exocytotic sites and therefore modulate the secretory activity. When Ca(2+) diffuses away from the exocytotic sites, the mitochondrial Ca(2+) uptake becomes the dominant mechanism.     

Arachidonic acid and docosahexaenoic acid suppress thrombin-evoked Ca2+ response in rat astrocytes by endogenous arachidonic acid liberation

Arachidonic (AA) and docosahexaenoic acid (DHA) are the major polyunsaturated fatty acids (PUFAs) in the brain. However, their influence on intracellular Ca2+ signalling is still widely unknown. In astrocytes, the amplitude of thrombin- induced Ca2+ response was time-dependently diminished by AA and DHA, or by the AA tetraynoic analogue ETYA, but not by eicosapentaenoic acid (EPA). Thrombin-elicited Ca2+ response was reduced (20-30%) by 1-min exposure to AA or DHA. Additionally, 1-min application of AA or DHA together with thrombin in Ca2+-free medium blocked Ca2+ influx, which followed after readdition of extracellular Ca2+. EPA and ETYA, however, were ineffective. Long-term treatment of astrocytes with AA and DHA, but not EPA reduced the amplitude of the thrombin-induced Ca2+ response by up to 80%. AA and DHA caused a comparable decrease in intracellular Ca2+ store content. Only DHA and AA, but not EPA or ETYA, caused liberation of endogenous AA by cytosolic phospholipase A2 (cPLA2). Therefore, we reasoned that the suppression of Ca2+ response to thrombin by AA and DHA could be due to release of endogenous AA. Possible participation of AA metabolites, however, was excluded by the finding that specific inhibitors of the different oxidative metabolic pathways of AA were not able to abrogate the inhibitory AA effect. In addition, thrombin evoked AA release via activation of cPLA2. From our data we propose a novel model of positive/negative-feed-back in which agonist-induced release of AA from membrane phospholipids promotes further AA release and then suppresses agonist-induced Ca2+ responses.     

RNA--induced silencing of the plasma membrane Ca2+-ATPase 2 in neuronal cells: effects on Ca2+ homeostasis and cell viability

Intraneuronal calcium ([Ca(2+)](i)) regulation is altered in aging brain, possibly because of the changes in critical Ca(2+) transporters. We previously reported that the levels of the plasma membrane Ca(2+)-ATPase (PMCA) and the V(max) for enzyme activity are significantly reduced in synaptic membranes in aging rat brain. The goal of these studies was to use RNA(i) techniques to suppress expression of a major neuronal isoform, PMCA2, in neurons in culture to determine the potential functional consequences of a decrease in PMCA activity. Embryonic rat brain neurons and SH-SY5Y neuroblastoma cells were transfected with in vitro--transcribed short interfering RNA or a short hairpin RNA expressing vector, respectively, leading to 80% suppression of PMCA2 expression within 48 h. Fluorescence ratio imaging of free [Ca(2+)](i) revealed that primary neurons with reduced PMCA2 expression had higher basal [Ca(2+)](i), slower recovery from KCl-induced Ca(2+) transients, and incomplete return to pre-stimulation Ca(2+) levels. Primary neurons and SH-SY5Y cells with PMCA2 suppression both exhibited significantly greater vulnerability to the toxicity of various stresses. Our results indicate that a loss of PMCA such as occurs in aging brain likely leads to subtle disruptions in normal Ca(2+) signaling and enhanced susceptibility to stresses that can alter the regulation of Ca(2+) homeostasis.     

Biophysical re-equilibration of Ca2+ fluxes as a simple biologically plausible explanation for complex intracellular Ca2+ release patterns

Physiological regulation of Ca(2+) release from the endoplasmic reticulum (ER) is critical for cell function. Recent direct measurements of free [Ca(2+)] inside the ER ([Ca(2+)](ER)) revealed that [Ca(2+)](ER) itself is a key regulator of ER Ca(2+) handling. However, the role of this new regulatory process in generating various patterns of Ca(2+) release remains to be elucidated in detail. Here, we incorporate the recently quantified experimental correlations between [Ca(2+)](ER) and Ca(2+) movements across the ER membrane into a mathematical model ER Ca(2+) handling. The model reproduces basic experimental dynamics of [Ca(2+)](ER). Although this was not goal in model design, the model also exhibits mechanistically unclear experimental phenomena such as "quantal" Ca(2+) release, and "store charging" by increasing resting cytosolic [Ca(2+)]. While more complex explanations cannot be ruled out, on the basis of our data we propose that "quantal release" and "store charging" could be simple re-equilibration phenomena, predicted by the recently quantified biophysical dynamics of Ca(2+) movements across the ER membrane.     

L-type voltage-operated Ca(+2) channels modulate transient Ca(+2) influx triggered by activation of Sertoli cell surface L-selectin

Near the base of mammalian seminiferous epithelium, Sertoli cells are joined by tight junctions, which constitute the blood-testis barrier. Differentiating germ cells are completely enveloped by Sertoli cells and must traverse the tight junctions during spermatogenic cycle. Following the specific ligand activation of L-selectin, the up-regulated Rho family small G-proteins have been implicated as important modulators of tight junctional dynamics. Although the activation of L-selectin transmits subsequent intracellular signals in a Ca(+2)-dependent fashion in various cell types, little is understood regarding the signaling pathways utilized by L-selectin in Sertoli cells. Therefore, we have examined the possible resultant calcium influx triggered by specific ligand-activation of cell surface L-selectin receptors or by cross-linking of L-selectin with anti-L-selectin. Spectrofluorimetric studies demonstrate increase of intracellular Ca(+2) levels immediately after the treatment of the L-selectin ligands, fucoidan and sialyl Lewis-a, or after treatment with anti-L-selectin antibody. We then determined the mechanism of Ca(+2) influx by investigating L- and T-type voltage-operated Ca(+2) channels, which have been suggested to present in the membranes of Sertoli cells. Data demonstrate that Sertoli cells treated with L-type voltage-operated Ca(+2) channel antagonists, nifedipine, diltiazem, or verapamil, lead to dose-dependent blockage of L-selectin-induced Ca(+2) influx. Cells treated with mibedradil, a T-type voltage-operated Ca(+2) channel antagonist, results in little or no blocking effect. Therefore, we conclude that activation of Sertoli cell L-selectin induces Ca(+2) influx, which is at least partially regulated by L-type voltage-operated Ca(+2) channels.