In unstimulated tissues, numerous cellular mechanisms contribute to the influx and efflux of Ca²⁺ to and from the cytoplasm in order to maintain homeostasis of intracellular basal Ca²⁺ concentrations [_b(Ca²⁺)_i], a phenomenon that occurs in almost all cells (1-7). In smooth muscle at rest, _b[Ca²⁺]_i must be kept tightly within the range of 100 and 150 nM (8-15) to maintain an equilibrium between contraction and relaxation. In these cells, the processes of Ca²⁺ influx and efflux preserve the myogenic tone, resting membrane potential and sarcoplasmic reticulum (SR) Ca²⁺ refilling (1,10,16-18). It has been proposed that the influx process involves entry of extracellular Ca²⁺ through L-type voltage dependent Ca²⁺ channels (L-VDCCs) (10,19-22), receptor-operated Ca²⁺ channels (ROCCs) activated by agonists (23-28) and store-operated Ca²⁺ channels (SOCCs, capacitative Ca²⁺ entry) activated by SR-Ca²⁺ depletion (10,29-33). An additional cytosolic Ca²⁺ source is the SR, that is the main intracellular Ca²⁺ store, activated via inositol 1,4,5-trisphosphate (IP₃) receptor channels (30,34-36) and ryanodine-receptor (RyR) channels (35,37-40). Ca²⁺ extrusion from the cytoplasm is accomplished via the action of membrane and sarcoplasmic Ca²⁺ ATPases and Na⁺/Ca²⁺ exchanger (NCX) in its forward mode (41-49).

Pivotal work on basal Ca²⁺ influx performed in aortic vascular smooth muscle cells using a pharmacological approach, demonstrated two predominant mechanisms of basal Ca²⁺ entry: One associated with L-VDCCs, accounting for ~23-43% of the total Ca²⁺ entry, and another associated with SOCCs, which contributed ~30% of the total (50).

In a recent study on airway smooth muscle (ASM), the present authors observed that the basal Ca²⁺ entry was mediated by L-VDCCs and probably also a constitutively active transient receptor potential canonical 3 (TRPC3) channel (18), which is described below. However, the mechanisms that maintain their permeability to Ca²⁺ have yet to be elucidated.

In the present review, current knowledge regarding different structures that maintain the _b[Ca²⁺]_i in ASM, including those involving L- and T-VDCCs, TRPC3, membrane and sarcoplasmic Ca²⁺-ATPases, NCX in its forward mode, IP₃ and RyRs, is discussed, including the most recent findings associated with the phosphorylation of L- and T-VDCCs and the dependence of TRPC3 on diacylglycerol (DAG).

For a better understanding of the participation of each of these proteins in the _b[Ca²⁺]_i regulation of ASM, novel unpublished data from studies by our group have been included. Firstly, Fig. 1A shows the maximal reduction of intracellular Ca²⁺ concentration ([Ca²⁺]_i) produced under Ca²⁺ free medium. This maneuver allowed determination of the proportional effect of each protein in the handling of _b[Ca²⁺]_i.

L- and T-VDCCs have been described in different types of smooth muscle (19,51,52); in particular, L-VDCC expression has been abundantly reported in the ASM of different species, including human (20,21,53-56). Opening of both types of channel is dependent on membrane depolarization, allowing the entry of Ca²⁺_, which subsequently contributes to contraction and SR Ca²⁺ refilling (9,10,19,20,57).

Several subunits for L-VDCC have been described: Ca_V1.1, Ca_V1.2, Ca_V1.3 and Ca_V1.4 (58). In ASM, L-VDCC had generally been characterized by pharmacological and electrophysiological methods (19). However, the presence of all the subunits of this channel was recently reported in rat bronchial smooth muscle (59). Nevertheless, in bovine and guinea-pig tracheal myocytes, only Ca_V1.2 and Ca_V1.2-Ca_V1.3, respectively, were observed (21,60). As identified recently by the present authors and shown in Fig. 1B and E, in guinea-pig ASM, D-600 (methoxyverapamil hydrochloride), a blocker of L-VDCC, significantly decreased the _b[Ca²⁺]_i, corroborating that this channel is constitutively active and contributes towards maintaining the _b[Ca²⁺]_i (18). It is well known that this channel is greatly dependent on the membrane voltage, and in canine ASM our group observed that its membrane potential at rest is approximately-59 mV, and is held steady. Furthermore, when the tissue was stimulated with carbachol, a cholinergic agonist, its membrane was depolarized, and when the depolarization reached-45 mV, it started oscillating (20). These oscillations are nifedipine-sensitive, and therefore corresponded to the opening and closing of the L-VDCC (61). Since the membrane potential at rest is unchanging, it was highly improbable that the voltage was influencing its opening at this stage.

Recently, a study in rat cardiomyocytes demonstrated that extracellular signal-regulated kinases 1 and 2 (ERK1/2), the mitogen-activated protein kinases (MAPKs), are able to phosphorylate L-VDCC at two sites: On Ser⁴⁹⁶ of the β₂ subunit and Ser¹⁹²⁸ of the α₁ subunit. Phosphorylation on the β₂ subunit or the α₁ subunit decreased or increased the L-VDCC activity, respectively (62). Thus, it may be hypothesized that in ASM, MAPK kinase (MEK)-ERK1/2 signaling may be involved in the continual opening and closing of the channel under basal conditions. This pathway may be associated with receptor tyrosine kinases (RTKs), which are activated by basal cyto-kines or growth factors. Our group previously demonstrated that ERK1/2 are present in the phosphorylated state in unstim-ulated bovine ASM (9). Fig. 1D and E show that the addition of U-0126, an inhibitor of ERK1/2, to guineapig tracheal myocytes significantly diminished the [Ca²⁺ b ]i until reaching a plateau. The addition of D-600 did not further modify the [Ca²⁺]_i, confirming that phosphorylation of the L-VDCC through the MEK-ERK1/2 pathway is possibly involved in its constitutive active mode. Therefore, the ERK1/2 signaling pathway may be responsible for phosphorylating the β₂ Ser⁴⁹⁶ and α₁ Ser¹⁹²⁸ sites, serving to switch the L-VDCC between an open and closed state (Fig. 1F).

Treatment with mibefradil, a T-VDCC blocker, also signifi-cantly lowered [Ca²⁺ b ]i in the guinea-pig tracheal myocytes, implying the participation of this channel in sustaining [Ca²⁺ b ]i (Fig. 1C and E). The presence of T-VDCC has been reported in this tissue (19), and the expression of Ca_V3.1, Ca_V3.2 and Ca_V3.3 subunits has been detected in ASM by immunohistochemistry (63). In this context, unexpectedly our group found that the addition of mibefradil following U-0126 did not further diminish _b[Ca²⁺]_i (Fig. 1D). This finding suggested that T-VDCC could also be regulated by the ERK1/2 signaling pathway. Recent studies have shown that T-VDCC may be modified by several serine/threonine protein kinase pathways, suggesting that this channel is susceptible to undergo phosphorylation (64); however, further research is required in this regard to determine the functional impact that ERK1/2 signaling has on the T-VDCC. Notably, in sensitized guinea-pigs that developed an airway inflammatory state, the expression level of L-VDCC was not modified (60). This finding indicated that these channels appear not to participate in the modification of _b[Ca²⁺]_i that is observed in inflammatory ailments, such as asthma (65).

In smooth muscle, TRPC channel genes code for ROCC and SOCC, which have an important role in intracellular Ca²⁺ homeostasis, while recently transient receptor potential vanilloid 1 (TRPV1) was revealed to be involved in the modulation of ASM tone and Ca²⁺ handling during agonist-induced contraction (66). In general, due to their ionic permeability, all TRPC channels are considered to be non-selective cation channels (NSCCs) (67,68). Thus far, all known TRPC channel activity has been shown to be associated with a phospholipase C (PLC) signaling pathway (69,70). In this context, it has been proposed that certain TRPC channels, including TRPC1, -2 and -3, are dependent on SR-Ca²⁺ depletion due to IP₃ production [a process termed store-operated Ca²⁺ entry (SOCE)] (36,71-75). On the other hand, ROCCs also include TRPC channels (TRPC3, -4, -5, -6 and 7), although these are activated by DAG, the other metabolite of PLC activity, and are independent of SR-Ca²⁺ depletion (69,70,76). In this context, only TRPCs 3, 6 and 7 are directly activated by DAG not involving protein kinase C (69,76), whereas TRPCs 4 and 5 are inhibited by protein kinase C, since their activity may be observed when this kinase is blocked (70).

In ASM, previous studies have reported the presence of almost all TRPC channel subtypes (TRPC1, -2, -3, -4, -5 and -6), with the exception of TRPC7 (67,68). Several TRPC channels have been shown to be constitutively active in different types of tissue. For example, TRPC1 and -4 were proposed to be continuously active in C57 mice skeletal myocytes (77); likewise, TRPC7 in human embryonic kidney cells (76), while TRPC3 was also observed to be constitutively active in rabbit ear artery and mouse airway myocytes (78,79). In this regard, our recent study demonstrated that, in guinea-pig ASM, this channel was also involved in maintaining the _b[Ca²⁺]_i and preserving smooth muscle basal tone (18). The role of this channel in _b[Ca²⁺]_i is illustrated in Fig. 2, where the addition of 2-aminoethoxydiphenyl borate (2-APB), a blocker of the TRPC3 channel (80), markedly diminished the _b[Ca²⁺]_i (Fig. 2A and E). Furthermore, Pyr3, another specific TRPC3 channel blocker (81), also lowered _b[Ca²⁺]_i by a similar extent (Fig. 2B and E). These results suggested that TRPC3 is constitutively active in guineapig ASM, even though the mechanism underlying this phenomenon has yet to be fully elucidated.

Since almost all TRPC channel subtypes are expressed in ASM, in this review the DAG analog, 1-oleoyl-2-acetyl-sn-glic-erol (OAG), was used to investigate the possible functional role of the channels present in this tissue. Fig. 2C shows that the addition of OAG to tracheal myocytes induced a transient peak in the [Ca²⁺]_i followed by a plateau. This response could have been developed through TRPC3 and/or TRPC6 channels, since these are both directly activated by DAG (69). However, after having reached the Ca²⁺ plateau induced by OAG, the addition of Pyr3 led to a return of [Ca²⁺]_i to its basal level. This finding indicated that the predominant TRPC channel that is functionally active in guineapig ASM, is TRPC3. Our group has postulated that TRPC3 is one of the channels involved in the maintenance of _b[Ca²⁺]_i (18), probably in a DAG-dependent manner. This lipid molecule is produced via the PLC or phospholipase D (PLD) pathways. It has been reported in rabbit ear artery myocytes that the PLD pathway produces DAG to sustain the constitutive activity of TRPC3 that contributes to the resting membrane potential (78,82). In ASM, protein kinase A was reported to regulate PLD activity, and it has been postulated that this phospholipase may be involved in the molecular mechanism underlying cyclic adenosine 5′-phosphate (c-AMP)-mediated relaxation in this tissue (83). By contrast, PLC has been shown to be predominantly involved in the IP₃-Ca²⁺ signaling pathway and in contraction (35). Therefore, in this review, we investigated if PLC may participate in DAG production in ASM at rest by using tricyclodecan-9-yl xanthogenate (D-609, a relatively specific inhibitor of PLC) (84) to inhibit this enzyme activity. It was observed that the addition of Pyr3 following D-609 to tracheal myocytes did not result in any further notable perturbations of the b[Ca²⁺]_i (Fig. 2D). Thus, these results suggested that PLC generates DAG, which subsequently leads to the activation of TRPC3 under basal conditions in order to maintain _b[Ca²⁺]_i in ASM (Fig. 2F). Conceivably, the activity of PLC may be regulated by endogenous ligands of RTKs, or by G-protein-coupled receptors.

It has been demonstrated that the expression levels and activity of the TRPC3 channel are greatly augmented in ASM cells obtained from sensitized mice (79). This may lead to an increase in the _b[Ca²⁺]_i, which could contribute to airway hyperresponsiveness in asthma.

The TRPV receptors, which are other members of the TRP family, have been implicated in mechanical stretch-induced Ca²⁺ influx in human ASM (85). In this context, TRPV1 is expressed in these cells, and was shown to be involved in Ca²⁺ oscillations and the maintenance of contraction by cholinergic agonists (66). However, any role in terms of maintaining the _b[Ca²⁺]_i has not yet been elucidated, and this requires further research.

SR-Ca²⁺ depletion mediated by IP₃ induces the established mechanism of capacitative Ca²⁺ entry. The first studies on this were performed by Putney (31) in non-excitable cells. Capacitative Ca²⁺ entry also occurs in smooth muscle via Ca²⁺ influx through diverse membrane channels (32,86). One of these Ca²⁺ influx mechanisms involves two types of protein associated with the SOCE pathway: Stromal interaction molecules (STIMs) and Orai proteins (87,88), both of which have been characterized in vascular smooth muscle and ASM (89,90). Orai are plasma membrane proteins, and three isoforms from different genes have been characterized: Orai1, -2 and -3 (91). On the other hand, two homologs of STIM have been identified: STIM1 and STIM2, both of which are located in the SR membrane (88,92,93). Regarding the two protein groups, Orai1 and STIM1 are the proteins that are chiefly expressed in ASM, and are responsible for the capacitative Ca²⁺ entry (89,94). Briefly, STIM1 on the SR functions as a Ca²⁺ sensor, monitoring the organelle’s Ca²⁺ content (95). When the SR-Ca²⁺ store is depleted, STIM1 forms an aggregate with other STIM1 molecules, thereby forming structures designated as ‘puncta’, which interact with Orai1 plasma membrane proteins to promote capacitative Ca²⁺ entry (89). Additionally, in several cell types it has been postulated that STIM/Orai may interact with TRPC channels, thereby establishing an alternative mechanism for capacitative Ca²⁺ entry (89,96). It is noteworthy that, in ASM, IP₃ has been demonstrated to directly open membranal TRPC3 channels. This recent finding implies that IP₃ mediates SR-Ca²⁺ depletion (i.e., capacitative Ca²⁺ entry) and also a direct, independent Ca²⁺ influx by TRPC channels (36). In this context, in one of our previous studies, we demonstrated that, in unstimulated airway myocytes, capacitative Ca²⁺ entry was not activated unless the SR Ca²⁺ content fell below 50% (8). However, it is well known that capacitative Ca²⁺ entry is activated by contractile agonists that act through the PLC_β-IP₃ signaling cascade (32), therefore providing no certainty that it does contribute to the maintenance of _b[Ca²⁺]i.

The Na⁺/Ca²⁺ exchanger (NCX) is a membrane Ca²⁺-handling protein that introduces three Na⁺ ions to the cytoplasm, while extruding one Ca²⁺ when in its forward mode. By contrast, in its reverse mode, it introduces Ca²⁺ and extrudes Na⁺ (42). To activate the reverse mode (NCX_REV), the entry of Na⁺ through an NSCC, and probably L-VDCC in proximity to the NCX, is required (21,41,48,97). The NCX is encoded by three gene isoforms, which generate NCX1, -2 and -3 (98-100). NCX1, extensively distributed in mammalian cells, has 17 different splicing variants that are tissue-specific and define the exchanger’s ionic sensitivity and regulation (101). NCX2 has no splicing variants and is located predominantly in the brain, spinal cord, gastrointestinal and kidney tissues, whereas NCX3 has five splice variants expressed in brain and skeletal muscle (101). In ASM, the NCX1.3 splicing variant is the main isoform present (102).

In airway myocytes, it has been proposed that NCX participates in the physiology of [Ca²⁺]_i, including SR-Ca²⁺ refilling (10,57), although it has been given a minor role in Ca²⁺ homeostasis (43). In this context, we have observed that NCX blockade with amiloride, a blocker of both the forward and reverse NCX modes, or KB-R7943, a blocker of NCX_REV, had no noticeable effect on _b[Ca²⁺]_i, indicating a minor role of this protein in terms of _b[Ca²⁺]_i regulation (unpublished data). Nevertheless, its participation in Ca²⁺ regulation, accomplished mainly through NCX_REV, becomes evident when _b[Ca²⁺]_i is increased and acquires a new steady-state (Fig. 3A). In this context, in a murine chronic model of allergen-induced airway hyperresponsiveness, it was shown that the levels of NCX1 were significantly augmented, and that NCX_REV activity was increased (103). Furthermore, in human myocytes, the addition of pro-inflammatory cytokines, including tumor necrosis factor-α (TNFα) and interleukin (IL)-13, also increased the expression of NCX1 and favored NCX_REV activity (104). These findings suggested that, during inflammation, NCX_REV could significantly contribute to an increase in the _b[Ca²⁺]_i, which would predispose airway smooth muscle to hyperresponsiveness.

Ca²⁺-ATPases form part of a large family of membrane proteins defined as P-type ATPases, including the plasmalemmal Ca²⁺-ATPase (PMCA) and the SR Ca²⁺-ATPase (SERCA, or sarco/endoplasmic reticulum Ca²⁺-ATPase) (105).

The PMCA extrudes Ca²⁺ against a high concentration gradient to contribute to _b[Ca²⁺]_i. It exists in a 1:1 relation-ship with ATP, is electroneutral via H⁺/Ca²⁺ exchange, and its affinity for Ca²⁺ and transport efficiency is increased by calmodulin. PMCA1-4 are the products of four different genes with several splice variants (105). PMCA1 and -4 are ubiquitous, and have lower affinity for calmodulin, whereas PMCA2 and PMCA3 have high calmodulin affinity (105,106).

In ASM, the primordial function of PMCA in Ca²⁺ homeostasis was demonstrated late in the 20th century (43). Shortly afterwards, the expression of this pump in canine ASM was reported (107). More recently, in rat bronchial myocytes, the presence of PMCA1 and PMCA4 was confirmed, and the participation of these two isoforms in Ca²⁺ homeostasis was demonstrated (108).

On the other hand, SERCA is, in part, electrogenic, since it introduces two Ca²⁺ ions to the SR, at the same time releasing at least four H⁺ ions to the cytoplasm (105). Additionally, it has been demonstrated that SERCA transports two Ca²⁺ ions for each hydrolyzed ATP molecule, and it appears to be the main system for controlling [Ca²⁺]_i in muscular cells (105).

SERCA pumps are produced by three genes: SERCA1, -2 and -3. They are subjected to alternative splicing, resulting in the isoforms, SERCA1a-b, SERCA2a-c and SERCA3a-f (105,109). In smooth muscle cells, the SERCA isoforms predominantly present are 2a and 2b (109), whereas in ASM, SERCA2b is the predominant isoform (110).

By measuring [Ca²⁺]_i in the absence of extracellular Ca²⁺, the addition of thapsigargin, a SERCA blocker, to rat bronchial nmyocytes produced a transient Ca²⁺ peak that returned to its basal value. At this point, lanthanum, a PMCA blocker, induced a sustained [Ca²⁺]_i increment that promoted apoptosis (108), demonstrating the central functional role of the two pumps in Ca²⁺ handling in ASM. In this regard, it has been proposed that there is a functional coupling between PMCA and SERCA to maintain Ca²⁺ homeostasis (49). Under physiological conditions (i.e., in the presence of extracellular Ca²⁺), we found in guineapig tracheal myocytes that thapsigargin increased [Ca²⁺]_i until a plateau was reached (Fig. 3A). It is well known that, in ASM, this Ca²⁺ increment is due to capacitative Ca²⁺ entry (i.e., SOCE) predominantly via the TRPC3 channel, a process that also produces membrane depolarization due to the entry of Na⁺ (79,111), consequently leading to L-VDCC opening and further Ca²⁺ and Na⁺ entry (10,18,21,36,79,112). At this stage, the NCX may change to its reverse mode (i.e., NCX_REV) due to the Na⁺ entry, thereby becoming the main contributor towards sustaining the Ca²⁺ plateau due to SERCA blockade. This proposition was corroborated using an NCX_REV-mode blocker, KB-R7943, which brought [Ca²⁺]_i to a new basal Ca²⁺ steady state (Fig. 3A) that was maintained by the PMCA activity. At this point, the addition of lanthanum, a non-specific PMCA blocker, led to a marked increase in [Ca²⁺]_i, probably inducing cellular apoptosis, as was suggested by a previous study (108). Taken together, these results corroborated that, under physiological conditions, SERCA and PMCA exert a primordial role in regulating [Ca²⁺]_i homeostasis, whereas NCX_REV only participates when b[Ca²⁺]_i is modified and acquires a new steady state (Fig. 3A and B).

Studies associated with the effects of pro-inflammatory cytokines on the ASM SERCA have demonstrated that over-night exposure of human airway myocytes to TNFα or IL-13 decreases the expression of SERCA that, in turn, diminishes the reuptake of SR-Ca²⁺ (113). Notably, these authors also revealed nthat, unlike other species, e.g., in porcine airways (114), human ASM SERCA does not express phospholamban, but is directly phosphorylated by Ca²⁺/calmodulin-dependent protein kinase II (113). Thus, it is possible that in an inflammatory process such as asthma, SR-ATPase activity is decreased, which may lead to an increase in the _b[Ca²⁺]_i to a new steady state, favoring an augmented response to bronchoconstrictor agonists. The same phenomenon may also be occurring as far as the PMCA is concerned; however, further research is required in this field.

RyR is a non-selective cation channel that releases Ca²⁺ from the SR and, in mammals, its three isoforms, RyR1, -2 and -3, are the products of different genes (115). All three isoforms are expressed in smooth muscle, including ASM (115,116). Cyclic ADP-ribose (cADPR) is considered to be their endogenous ligand in airway myocytes, which is regulated by the membrane-bound protein, CD38 (117). This protein has ADP-ribosyl cyclase and hydrolase activity, and is involved in the synthesis or degradation of cADPR, respectively (118,119).

The IP₃ receptor (ITPR) is another non-selective cation channel that releases Ca²⁺ from the SR via IP₃ generated by the G_qα signaling pathway (35). It has three isoforms (ITPR1, -2 and -3) derived from different genes, which share ~60-80% amino acid homology (120,121). These receptors have also been identified in different smooth muscles types, including ASM (36,122-124).

In 1993, Ca²⁺ ‘sparks’ were described in heart muscle (125), and these were associated with the Ca²⁺-induced Ca²⁺ release from RyRs (126). In guineapig tracheal myocytes, the presence of spontaneous Ca²⁺ sparks was observed for the first time in 1998 (127). Subsequently, in urinary bladder smooth muscle, these Ca²⁺ sparks were characterized as the elementary release of Ca²⁺ from RyRs (128), and this finding was later corroborated in mouse ASM, occurring predominantly through RyR2 (116,129). In this context, studies on the pulmonary artery revealed that Ca²⁺ sparks are activated by Ca²⁺ released via ITPR (130), as well as in ASM (129). The physiological role of these Ca²⁺ sparks in guineapig tracheal myocytes was well established. Essentially, they produce spontaneous transient outward currents caused by large-conductance Ca²⁺-activated K⁺ channels; they also induce spontaneous transient inward currents accomplished through Ca²⁺-activated Cl-channels (127). Therefore, all these components may serve an important role in the basal state regulation of the ASM by stabilizing the membrane potential, the _b[Ca²⁺]_i and the basal contractile tone.

Interestingly, further lines of research have demonstrated that pro-inflammatory cytokines (predominantly TNFα), promote the augmentation of CD38-cADPR signaling and increase Ca²⁺ responses to agonists (117,131), a phenomenon that is probably mediated by an augmentation of _b[Ca²⁺]_i. Furthermore, TNFα also enhances G_qα protein expression, thereby increasing the ASM response to carbachol (132). However, upregulation of the IP₃-Ca²⁺ signaling pathway and any consequent modification of the _b[Ca²⁺]_i in an inflammatory context, such as in asthma, has not readily been identified, and this requires further research.

The current review has discussed how several Ca²⁺ handling mechanisms are finely tuned to regulate the _b[Ca²⁺]_i, summarized in Fig. 4. It is conceivable that alterations in any of these processes could render ASM susceptible to developing the type of hyperresponsiveness that is commonly observed in ailments such as asthma, and this warrants further study.

The present study was partly supported by grants from Consejo Nacional de Ciencia y Tecnología, Ciudad de México, México (grant no. 219859) and Dirección General de Asuntos del Personal Académico (DGAPA), Universidad Nacional Autónoma de México (grant no. IN201216) to LMM.

The datasets presented in the current review are available from the corresponding author on reasonable request.

With particular regard to the previously unpublished work presented herein, the contribution of each author was as follows. JRG and ACG performed the assays of intracellular Ca²⁺ levels. EFS performed enzymatic isolation of tracheal myocytes, participated in the assays of intracellular Ca²⁺ levels and data analysis, and provided critical ideas during the writing of the manuscript. BS contributed to the data analysis and writing of the manuscript. LMM contributed to the design and global supervision of the study, data analysis and writing of the manuscript, and was responsible for submitting the paper for publication. All authors read and approved the final manuscript.

Not applicable.

Not applicable.

The authors declare no competing interests.