Phosphoinositide Metabolism in Airway Smooth Muscle 1 EDWIN R. CHILVERS2 and STEFAN R. NAHORSKI
Introduction The biochemical events underlying agonistinduced contraction of airway smooth muscle (ASM) can be divided into two distinct phases. The initial contractile response (phasic stage) is triggered by an increase in cytoplasmic free Ca 2 + ([Ca2+Ji), resulting in a calmodulin-dependent phosphorylation of the 20-kD myosin light chain, activation of actin-myosin ATPase, rapid cross-bridge cycling between the actin filaments and myosin headgroups, and hence tension formation (1). In contrast, the maintenance of contraction (tonic phase) appears independent of [Ca"]: since in the continued presence of agonist, both myosin light chain phosphorylation and [Ca"]: rapidly return to near resting values despite no loss in tension (2, 3). Early studies in ASM demonstrated that agonist-induced contraction occurred independently to changes in membrane potential, and hence, signal transduction appeared to rely on pharmacomechanical coupling as opposed to electrochemical coupling (4, 5). A further important series of experiments demonstrated that removing extracellular Ca2+ or the addition of calcium channel-blocking agents did not significantly impair the development of contraction or the [Ca"], transient consequent upon the addition of agonist (6, 7). These findings suggested that the activation of a surface receptor in the ASM cell resulted in Ca 2' release from intracellular stores. In common with Ca2+-mobilizing receptors present in other cells, the molecular basis of the signal-transduction mechanism underlying agonist-stimulated Ca 2' releasewas at that time unknown. Over the past few years it has been established that in a wide variety of cells, the receptor-mediated hydrolysis of a specific membrane inositol phospholipid, phosphatidylinositoI4,5-bisphosphate (PtdIns(4,5)P2), forms the second messengers inositol 1,4,5trisphosphate (Ins(1,4,5)P,) and diacylglycerol (DAG) via the activation of phospholipase C (8, and see previous review by Putney). Substantial evidence that Ins(1,4,5)P, is recognized by receptorsthat can open a Ca 2+ channel associated with a nonmitochondrial intracellular store has provided the fundamental link between the spatially separated events of membrane phosphoinositide metabolism and intracellular Ca 2+ mobilization (8). Although it is not appropriate here to discuss the complex metabolism of Ins(1,4,5)P" it should be emphasized that phosphorylation via a 3-kinase yields another inositol polyphosphate (inositol1,3,4,5-tetrakisphosphate) of potential physiologic significance. Recent evidence suggests that Ins(1,3,4,5)P. may control Ca2+ movements between intracellular stores and/or across the plasma membrane,
Contraction of airway smooth muscle (ASM) results from an increase in cytoplasmic free Ca" and appears to be independent of changes in membrane potential or the presence of extracellular Ca". Inositol 1,4,5-trisphosphate (Ins(1,4,5)P,),the product of receptor-mediated hydrolysis of a membrane phospholnosltlde. phosphatldylinositol 4,5-blsphosphate, is capable of releasIng Ca" from Intracellular stores in this tissue and has been proposed as the intracellular second messenger responsible for agonist-Induced changes in cytosollc Ca" and hence contraction. In this report, we discuss the evidence for this signaling system in ASM and present new data on mass changes in Ins(1,4,5)P, in carbachol-stimulated bovine tracheal smooth muscle. The rapid, transient accumulation of this second messenger In ASM Is thought to reflect regUlation of the enzymes Involved In the metabolism rather than the synthesis of this inositol polyphosphate. The potential mechanisms of this regulation are discussed in relation to the initiation and maintenance AM REV RE5PIR DI5 1990; 141:5137-5140 of contraction In ASM. SUMMARY
hence regulating the availability of Ca 2+ that is accessible to Ins(1,4,5)p, (9, 10).
A further major feature of phosphoinositide metabolism in ASM is that, in common with a number of other tissues, agonists linked to phospholipase C markedly stimulate the incorporation of either ["PO.) or ['H)inositol into the phosphoinositides when the tissue has not been labeled to isotopic equilibrium (11, 13, 15-17). The magnitude of this "incorporation effect" is virtually unique to ASM and may reflect either a discrete agonistlinked phosphoinositide pool or more probably verylowturnover of this pool under basal conditions. It is also of interest that agonists with very different abilities in terms of inositol phosphate generation (e.g.,bradykinin and histamine compared with carbachol) share very similar incorporation effects (17).
Inositol Phospholipid Metabolism in ASM Evidence for the involvementof the phosphoinositides in transmembrane signaling in ASM is based on a number of observations. First, Baron and colleagues (11) demonstrated that in canine tracheal smooth muscle, the muscarinic receptor-agonist carbachol caused a rapid decline in the mass of phosphatidylinositol (PtdIns), the parent lipid for polyphosphoinositide synthesis, with parallel increases in DAG and phosphatidic acid (PA), serial lipid products of phosphoinositide hydrolysis. Takawa and coworkers (12) later provided evidence that the concentrations of the polyphosInositol Phosphate Formation phoinositides PtdIns(4)p and PtdIns(4,5)p2 also fall rapidly in ASM after muscarinic The second line of evidence for the involvereceptor stimulation and that PtdIns(4,5)P., ment of the phosphoinositides in pharmacoas in other tissues, appears to be the major mechanical coupling in ASM is the demonand possibly exclusive phosphoinositide hy- stration that many contractile agonists stimudrolyzed by hormone-activated phospholi- late the accumulation of inositol phosphates. pase C. In bovine tracheal smooth muscle, Assays using ASM strips or slice preparations we have shown that PtdIns(4,5)p 2represents labeled with [lH)inositol and stimulated with 10.2 ± 1.2010 (mean ± SEM) of the total agonists in the presence of Li', an uncornpeti['H) inositol-labeled phosphoinositide pool tive inhibitor of inositol monophosphatase and have confirmed the substan-tial declines (18), which increases assay sensitivity, have in the amounts of radiolabeled PtdIns(4)P shown increased [lH)inositol phosphate (['H)and PtdIns(4,5)p2 seen following carbachol InsP) accumulation (predominantly Insl-.) stimulation (PtdIns(4)p, 51 ± 2% and secondary to a variety of muscarinic agonists, PtdIns(4,5)P 2,44 ± 3% of control values fol- including carbachol, methacholine, oxotremlowing 10-' M carbachol stimulation for 30 orine and McN-A-343 (19), histamine (16), min) (13). Significant declines in {'H]ptdIns 5-HT (20), and bradykinin (17);and also folwere only observed in the presence of 5 mM lowing electrical field stimulation (21). PreLi",whichpreventsthe recycling of free inositol liminary reports also indicate that the tachykiinto the phosphoinositides by inhibiting the nins, substance P, neurokinin A and B (22), breakdown of'the inositol monophosphates, and leukotrienes C. and D. (23) elicit phosThese findings suggested that, in contrast to phoinositide breakdown in the guinea pig aircertain other tissues where PtdIns(4,5)p2 mass is relatively well preserved at the expense of 1 From the Department of Pharmacology and PtdIns (14),in ASM the conversion of PtdIns Therapeutics, University of Leicester, Leicester, to PtdIns(4)P by a specific 4-kinase enzyme United Kingdom. , Medical Research Council (U.K.) Training may be rate-limiting under conditions of prolonged maximal carbachol stimulation (13). Fellow. 5137
5138
wayalthough a specificeffect of these agonists in a pure ASM preparation has yet to be demonstrated. In the presence of Li', carbachol results in a linear production of total [3H]lnsPs (> 90% InsP,) for at least 90 min with an EC,o of 2 11M (13), and contrary to previous suggestions (24) the InsP response to carbachol and histamine are nonadditive (17),indicating utilization of a single agonistsensitivephosphoinositide pool. In bovine tracheal smooth muscle, a direct relationship has been demonstrated between the dose-response curves for contraction and InsP accumulation for a range of muscarinic agonists (19) with a large receptor reserve evident for InsP formation for the full contractile agonists methacholine and oxotremorine. An important caveat with regard to agonist-induced InsP formation in ASM, which distinguishes it from ileal smooth muscle (25, 26), is the inability of K+, itself a spasmogen, to induce InsP formation (II, 27). This has two important implications: (1) the ASM cell is not fastidious about the source of activator Ca2+ and (2) increasing [Ca"], alone does not appear to activate phospholipase C as has been demonstrated in many other cell types (28). Although measuring totallnsP accumulation is a useful pharmacologic tool in assessing agonist-induced phosphoinositide hydrolysis, it still remains an indirect index of Ptdlns(4,5)p, hydrolysisand Ins(l,4,5)p3 formation. This may be a particularly important distinction since in some tissues hydrolysis of Ptdlns and Ptdlns(4)p has been proposed as a direct source of Ins(I)p, and Ins(l,4)P" a scheme that allowsongoing DAG formation (and hence protein kinase C activation) without the generation of physiologically active InsP species. It is therefore important to determine InsP 3, and in particular Ins(l,4,5)P 3, formation. In addition, in order for Ins(l,4,5)p3to qualify as a true secondmessenger of Ca2+ release in ASM, its production must precede any increase in [Ca2+li and the subsequent phosphorylation events that result in contraction and its ability to release Ca2+ from an appropriate internal site demonstrated. A number of groups have shown agoniststimulated increasesin total [3H] InsP3in ASM that precede tension formation (21, 29, 30). However, due to the presence of at least two agonist-stimulated InsP 3 isomers (figure I), the kinetics of these total InsP 3responses may be very different from that for Ins(I,4,5)P 3, which has been shown in other systems to be the only naturally occurring InsP 3 isomer involved in Ca2+ release (31, 32), reflecting the strict positional specificity of the InsP3binding site present on the intracellular Ca2+ store (33).This point appears particularly important in ASM because after carbachol stimulation, we have shown, by high performance liquid chromatography separation of Ins(l,3,4)P 3 and Ins(l,4,5)P 3and by an enzymatic method that specifically hydrolyzes the Ins(l,3,4)P 3 component of the InsP 3 fraction (34), that even as early as 60 s Ins(I,3,4)P3 is the major (> 80%) ['H]lnsP 3 isomer accumulating in
CHILVERS AND NAHORSKI
prelabeled bovine tracheal smooth muscle (13, 13a). This is a situation similar to that observed in a number of other tissues, including muscarinic-receptor stimulation in the parotid gland (35),angiotensin stimulation in the guinea pig hepatocyte (36), and thrombin stimulation in human platelets (37).This would suggest that the metabolism of Ins(l,4,5)P 3 to Ins(l,3,4)P3 via Ins(l,3,4,5)p4 (figure I) may be of considerable importance in ASM and the lack of a significant accumulation of ['H]InsP 4 at early time points « I min) in prelabeled bovine ASM (l3a) testifies to the initially rapid flux of the inositol headgroup through Ins(l,3,4,5)P4.Although an inositol tetrakisphosphate does accumulate at later times after carbachol (I3a), it is not yet established that this is the Ins(l,3,4,5)p4isomer because the accumulation of an Ins(I,3,4,6)p4 isomer after receptor stimulation has also been reported (38). It is very difficult to quantify the relative importance of the 5'-phosphatase and 3'-kinase pathways that regulate Ins(l,4,5)P 3concentration within the cell. Although the K m for Ins(I,4,5)P3 is much lower for the kinase, this enzyme has limited capacity and may be readily saturated by its substrate. Thus, the relative proportion of metabolism via phosphorylation or dephosphorylation pathways may relate to the basal and
stimulated accumulation of Ins(I,4,5)P3within discrete compartments in the cell. This is clearly an important area for further investigation even though a definitive physiological role for Ins(l,3,4,5)P4in smooth muscle is yet to be established. In view of the problems associated with quantifying changes in ['H]lns(l,4,5)P 3 in ['H]inositol prelabeled ASM (i.e., agoniststimulated increases in phosphoinositidespecific radioactivity, relatively low amounts of [3H]lnsP 3accumulating, and dependence on methods to separate Ins(l,3,4)P 3 and Ins(l,4,5)p3' usually high performance liquid chromatography), we have used an alternative radioreceptor assay method based on the specific binding of Ins(l,4,5)p3 to its putative intracellular receptor in an adrenal cortex preparation to follow receptor-mediated changes in Ins(l,4,5)P 3 (39, 40, 40a). The major advantages of this method are that it provides a measurement of Ins(l,4,5)p3 mass and that the Ins(l,3,4)P 3 isomer does not crossreact at this receptor. Using this assay, carbachol produces a rapid but transient increase in Ins(I,4,5)P3 concentration in bovine tracheal smooth muscle from a resting value of 12.9 ± 0.8 to 27 ± 1.5 pmol/mg protein at 5 s (figure 2). Since the muscarinic receptor linked to phospholipase C does not appear to
Agonist
---~--
~-B~
)LLDG
Ptdlns(4,5)P 2
I I
T
, I I
Fig 1. Receptor-mediated generation of inositol 1,4,5-trisphosphate from phosphatidylinositol 4,5-bisphosphate and its initial metabolism to inositol-1,3,4,5-tetrakisphosphate or inositol 1,4-bisphosphate.
5139
PHOSPHOINOSITIDE METABOLISM IN AIRWAY SMOOTH MUSCLE
'2
30 -.---
-
-
-
- --
-
-
'iii
Fig. 2. Time-course for carbachol-stimulated inositoI1,4,5-trisphosphate accumulationinbovinetracheal smoothmuscleslices. 100 I'M carbachol (e) or vehicle (0) was added to aliquotsof bovinetracheal smooth muscle slices and incubationscontinuedfor time periods indicated. Incubations wereterminated and tissue extracts prepared and assayed forIns(1 ,4,5)P3 as described inChalliss and colleagues (40). Results are presented as mean ± SEMforsixto nine independent observations made in three separate experiments.
o...
Q.
25
-
-
-
-
---,
•
0.1mM Carbacho l
o
Control
ential regulation together with the site of action of cAMP remains to be established. In conclusion, there is now firm evidence supporting a physiologic role of Ins(I,4,5)P 3 as a second messenger responsible for excitation-contraction coupling in ASM and this system affords several potential target sites for modifying ASM contraction.
Cl
.§ '0
20
E
Q.
desensitizesignificantlyin this tissue, suggested by the continued breakdown of PtdIns(4,5)p 2 in the presence of agonist (12, 13, 17), the subsequent prompt decline in Ins(I,4,5)P3 concentration indicates a rapid activation of the pathways responsible for Ins(1,4,5)p 3metabolism within the cell. This response is somewhat different to that observed in rat cerebral cortex slices where Ins(1,4,5)p3concentration remains persistently elevated throughout the period of carbachol stimulation (40). This suggests that whereas in peripheral tissues, such as ASM, Ins(1,4,5)p3concentration and Ins(I,4,5)P.-induced Ca H release are closely related, in brain additional factors, such as Ca H inhibition of Ins(1,4,5)P 3 binding to its receptor (41), may allow subsequently independent regulation of Ins(1,4,5)P3 concentration and CaH release. In either case, this scheme would also allow ongoing DAG formation and protein kinase C activation, which are thought to be responsible for the maintenance of ASM contraction (42), without concomitant Ca H release. Finally, the ability of Ins(I,4,5)P 3 to release Ca H from nonmitochondrial, ATP-dependent cellular stores has been well established in a variety of cell types (8, 43), including ASM (29, 44). Although the absolute increase in Ins(1,4,5)P 3 induced by receptor-agonists in bovine ASM is relatively small (40a), this is almost certainly complicated by local changes in discrete compartments close to the plasma membrane and by basal accumulation of this second messenger not associated with the Ins(l,4,5)P 3 receptor. It is also worth considering the fact that whereas carbachol and histamine can both increase the accumulation of Ins(I,4,5)P3 in bovine ASM, the maximal increase is much larger with the muscarinic agonist (40a). Although both agonists act as full contractile agonists, it is tempting to ascribe the differences in receptor reserve for contraction to the ability to accumulate initially the second messenger Ins(l,4,5)P 3.
Regulation of Ins(1,4,S}P3 A fundamental prerequisite of any intracellular second-messenger molecule is that its formation and subsequent metabolism must be carefully regulated to obtain the desired phys-
15
--~--- - - - - - - - -- - - - - - - - - -----~
Acknowledgment The collaboration of Drs. I. H. Batty, R. A. J. Challiss, P. J. Barnes, and Ms. E. D. Kennedy is gratefully acknowledged together with the secretarial assistance of Ms. Jenny Batty.
10
o
60
12 0
2 40
1 80
300
Time (sec)
References
iologic response intended. A scheme is presented in figure 3 whereby Ins(l,4,5)P 3 concentration in the cell may be acutely regulated by Ca H through the activation of the 3'-kinase (45) and possibly 5'-phosphatase (46) enzymes. Protein kinase C activation has also been shown to activate the 5'-phosphatase (47) in addition to influencing Ins(l,4,5)P3formation (48). Certainly, activation of protein kinase C by 200 nM phorbol 12-myristate 13acetate completely blocks the release of internal Ca H by histamine in cultured canine ASM cells (49). Recently a number of groups have demonstrated that elevating cAMP in the cell is also able to antagonize agonist-induced phospholipase C activation (16, 50). In ASM, this effect is more pronounced with histamine than with carbachol stimulation, although the mechanism underlying this differ-
1. Rodger IW, Biochemistry of airway smooth muscle contraction. In: Barnes PJ, Rodger IW, Thomson N, eds. Asthma: basic mechanisms and clinical management. London: Academic Press Ltd., 1988; 57-79. 2. Aksoy MO, Murphy RA, Kamm KE. Role of Ca2• and myosin light chain phosphorylation in regulation of smooth muscle. Am J Physiol 1982; 242:CI09-16. 3. Park S, Rasmussen H. Carbachol-induced protein phosphorylation changes in bovine tracheal smooth muscle. J Biol Chem 1986; 261:15734-9. 4. Coburn RF. Electromechanical coupling in canine trachealis muscle: acetylcholine contractions. Am J Physiol 1979; 236:CI77-84. 5. Ahmed F, Foster RW, Small RC, Weston AH. Some features of the spasmogenic actions of acetylcholine and histamine in guinea-pig isolated trachea. Br J Pharmacol 1984; 83:227-33. 6. Kirkpatrick CT. Excitation and contraction in bovine tracheal smooth muscle. J Physiol 1975; 244:263-81. 7. Ito Y, Itoh T. The roles of stored calcium in contractions of cat tracheal smooth muscle pro-
Agonist
.J,
8~-G-------J
-/"",- -8~ _ -- --?
.t-- ----.,1 C-AMP
11...----'
P LC
__ 1~
0
P HORBOL ESTERS
)
.~
P
>-=- 8 -
Introduction The biochemical events underlying agonistinduced contraction of airway smooth muscle (ASM) can be divided into two distinct phases. The initial contractile response (phasic stage) is triggered by an increase in cytoplasmic free Ca 2 + ([Ca2+Ji), resulting in a calmodulin-dependent phosphorylation of the 20-kD myosin light chain, activation of actin-myosin ATPase, rapid cross-bridge cycling between the actin filaments and myosin headgroups, and hence tension formation (1). In contrast, the maintenance of contraction (tonic phase) appears independent of [Ca"]: since in the continued presence of agonist, both myosin light chain phosphorylation and [Ca"]: rapidly return to near resting values despite no loss in tension (2, 3). Early studies in ASM demonstrated that agonist-induced contraction occurred independently to changes in membrane potential, and hence, signal transduction appeared to rely on pharmacomechanical coupling as opposed to electrochemical coupling (4, 5). A further important series of experiments demonstrated that removing extracellular Ca2+ or the addition of calcium channel-blocking agents did not significantly impair the development of contraction or the [Ca"], transient consequent upon the addition of agonist (6, 7). These findings suggested that the activation of a surface receptor in the ASM cell resulted in Ca 2' release from intracellular stores. In common with Ca2+-mobilizing receptors present in other cells, the molecular basis of the signal-transduction mechanism underlying agonist-stimulated Ca 2' releasewas at that time unknown. Over the past few years it has been established that in a wide variety of cells, the receptor-mediated hydrolysis of a specific membrane inositol phospholipid, phosphatidylinositoI4,5-bisphosphate (PtdIns(4,5)P2), forms the second messengers inositol 1,4,5trisphosphate (Ins(1,4,5)P,) and diacylglycerol (DAG) via the activation of phospholipase C (8, and see previous review by Putney). Substantial evidence that Ins(1,4,5)P, is recognized by receptorsthat can open a Ca 2+ channel associated with a nonmitochondrial intracellular store has provided the fundamental link between the spatially separated events of membrane phosphoinositide metabolism and intracellular Ca 2+ mobilization (8). Although it is not appropriate here to discuss the complex metabolism of Ins(1,4,5)P" it should be emphasized that phosphorylation via a 3-kinase yields another inositol polyphosphate (inositol1,3,4,5-tetrakisphosphate) of potential physiologic significance. Recent evidence suggests that Ins(1,3,4,5)P. may control Ca2+ movements between intracellular stores and/or across the plasma membrane,
Contraction of airway smooth muscle (ASM) results from an increase in cytoplasmic free Ca" and appears to be independent of changes in membrane potential or the presence of extracellular Ca". Inositol 1,4,5-trisphosphate (Ins(1,4,5)P,),the product of receptor-mediated hydrolysis of a membrane phospholnosltlde. phosphatldylinositol 4,5-blsphosphate, is capable of releasIng Ca" from Intracellular stores in this tissue and has been proposed as the intracellular second messenger responsible for agonist-Induced changes in cytosollc Ca" and hence contraction. In this report, we discuss the evidence for this signaling system in ASM and present new data on mass changes in Ins(1,4,5)P, in carbachol-stimulated bovine tracheal smooth muscle. The rapid, transient accumulation of this second messenger In ASM Is thought to reflect regUlation of the enzymes Involved In the metabolism rather than the synthesis of this inositol polyphosphate. The potential mechanisms of this regulation are discussed in relation to the initiation and maintenance AM REV RE5PIR DI5 1990; 141:5137-5140 of contraction In ASM. SUMMARY
hence regulating the availability of Ca 2+ that is accessible to Ins(1,4,5)p, (9, 10).
A further major feature of phosphoinositide metabolism in ASM is that, in common with a number of other tissues, agonists linked to phospholipase C markedly stimulate the incorporation of either ["PO.) or ['H)inositol into the phosphoinositides when the tissue has not been labeled to isotopic equilibrium (11, 13, 15-17). The magnitude of this "incorporation effect" is virtually unique to ASM and may reflect either a discrete agonistlinked phosphoinositide pool or more probably verylowturnover of this pool under basal conditions. It is also of interest that agonists with very different abilities in terms of inositol phosphate generation (e.g.,bradykinin and histamine compared with carbachol) share very similar incorporation effects (17).
Inositol Phospholipid Metabolism in ASM Evidence for the involvementof the phosphoinositides in transmembrane signaling in ASM is based on a number of observations. First, Baron and colleagues (11) demonstrated that in canine tracheal smooth muscle, the muscarinic receptor-agonist carbachol caused a rapid decline in the mass of phosphatidylinositol (PtdIns), the parent lipid for polyphosphoinositide synthesis, with parallel increases in DAG and phosphatidic acid (PA), serial lipid products of phosphoinositide hydrolysis. Takawa and coworkers (12) later provided evidence that the concentrations of the polyphosInositol Phosphate Formation phoinositides PtdIns(4)p and PtdIns(4,5)p2 also fall rapidly in ASM after muscarinic The second line of evidence for the involvereceptor stimulation and that PtdIns(4,5)P., ment of the phosphoinositides in pharmacoas in other tissues, appears to be the major mechanical coupling in ASM is the demonand possibly exclusive phosphoinositide hy- stration that many contractile agonists stimudrolyzed by hormone-activated phospholi- late the accumulation of inositol phosphates. pase C. In bovine tracheal smooth muscle, Assays using ASM strips or slice preparations we have shown that PtdIns(4,5)p 2represents labeled with [lH)inositol and stimulated with 10.2 ± 1.2010 (mean ± SEM) of the total agonists in the presence of Li', an uncornpeti['H) inositol-labeled phosphoinositide pool tive inhibitor of inositol monophosphatase and have confirmed the substan-tial declines (18), which increases assay sensitivity, have in the amounts of radiolabeled PtdIns(4)P shown increased [lH)inositol phosphate (['H)and PtdIns(4,5)p2 seen following carbachol InsP) accumulation (predominantly Insl-.) stimulation (PtdIns(4)p, 51 ± 2% and secondary to a variety of muscarinic agonists, PtdIns(4,5)P 2,44 ± 3% of control values fol- including carbachol, methacholine, oxotremlowing 10-' M carbachol stimulation for 30 orine and McN-A-343 (19), histamine (16), min) (13). Significant declines in {'H]ptdIns 5-HT (20), and bradykinin (17);and also folwere only observed in the presence of 5 mM lowing electrical field stimulation (21). PreLi",whichpreventsthe recycling of free inositol liminary reports also indicate that the tachykiinto the phosphoinositides by inhibiting the nins, substance P, neurokinin A and B (22), breakdown of'the inositol monophosphates, and leukotrienes C. and D. (23) elicit phosThese findings suggested that, in contrast to phoinositide breakdown in the guinea pig aircertain other tissues where PtdIns(4,5)p2 mass is relatively well preserved at the expense of 1 From the Department of Pharmacology and PtdIns (14),in ASM the conversion of PtdIns Therapeutics, University of Leicester, Leicester, to PtdIns(4)P by a specific 4-kinase enzyme United Kingdom. , Medical Research Council (U.K.) Training may be rate-limiting under conditions of prolonged maximal carbachol stimulation (13). Fellow. 5137
5138
wayalthough a specificeffect of these agonists in a pure ASM preparation has yet to be demonstrated. In the presence of Li', carbachol results in a linear production of total [3H]lnsPs (> 90% InsP,) for at least 90 min with an EC,o of 2 11M (13), and contrary to previous suggestions (24) the InsP response to carbachol and histamine are nonadditive (17),indicating utilization of a single agonistsensitivephosphoinositide pool. In bovine tracheal smooth muscle, a direct relationship has been demonstrated between the dose-response curves for contraction and InsP accumulation for a range of muscarinic agonists (19) with a large receptor reserve evident for InsP formation for the full contractile agonists methacholine and oxotremorine. An important caveat with regard to agonist-induced InsP formation in ASM, which distinguishes it from ileal smooth muscle (25, 26), is the inability of K+, itself a spasmogen, to induce InsP formation (II, 27). This has two important implications: (1) the ASM cell is not fastidious about the source of activator Ca2+ and (2) increasing [Ca"], alone does not appear to activate phospholipase C as has been demonstrated in many other cell types (28). Although measuring totallnsP accumulation is a useful pharmacologic tool in assessing agonist-induced phosphoinositide hydrolysis, it still remains an indirect index of Ptdlns(4,5)p, hydrolysisand Ins(l,4,5)p3 formation. This may be a particularly important distinction since in some tissues hydrolysis of Ptdlns and Ptdlns(4)p has been proposed as a direct source of Ins(I)p, and Ins(l,4)P" a scheme that allowsongoing DAG formation (and hence protein kinase C activation) without the generation of physiologically active InsP species. It is therefore important to determine InsP 3, and in particular Ins(l,4,5)P 3, formation. In addition, in order for Ins(l,4,5)p3to qualify as a true secondmessenger of Ca2+ release in ASM, its production must precede any increase in [Ca2+li and the subsequent phosphorylation events that result in contraction and its ability to release Ca2+ from an appropriate internal site demonstrated. A number of groups have shown agoniststimulated increasesin total [3H] InsP3in ASM that precede tension formation (21, 29, 30). However, due to the presence of at least two agonist-stimulated InsP 3 isomers (figure I), the kinetics of these total InsP 3responses may be very different from that for Ins(I,4,5)P 3, which has been shown in other systems to be the only naturally occurring InsP 3 isomer involved in Ca2+ release (31, 32), reflecting the strict positional specificity of the InsP3binding site present on the intracellular Ca2+ store (33).This point appears particularly important in ASM because after carbachol stimulation, we have shown, by high performance liquid chromatography separation of Ins(l,3,4)P 3 and Ins(l,4,5)P 3and by an enzymatic method that specifically hydrolyzes the Ins(l,3,4)P 3 component of the InsP 3 fraction (34), that even as early as 60 s Ins(I,3,4)P3 is the major (> 80%) ['H]lnsP 3 isomer accumulating in
CHILVERS AND NAHORSKI
prelabeled bovine tracheal smooth muscle (13, 13a). This is a situation similar to that observed in a number of other tissues, including muscarinic-receptor stimulation in the parotid gland (35),angiotensin stimulation in the guinea pig hepatocyte (36), and thrombin stimulation in human platelets (37).This would suggest that the metabolism of Ins(l,4,5)P 3 to Ins(l,3,4)P3 via Ins(l,3,4,5)p4 (figure I) may be of considerable importance in ASM and the lack of a significant accumulation of ['H]InsP 4 at early time points « I min) in prelabeled bovine ASM (l3a) testifies to the initially rapid flux of the inositol headgroup through Ins(l,3,4,5)P4.Although an inositol tetrakisphosphate does accumulate at later times after carbachol (I3a), it is not yet established that this is the Ins(l,3,4,5)p4isomer because the accumulation of an Ins(I,3,4,6)p4 isomer after receptor stimulation has also been reported (38). It is very difficult to quantify the relative importance of the 5'-phosphatase and 3'-kinase pathways that regulate Ins(l,4,5)P 3concentration within the cell. Although the K m for Ins(I,4,5)P3 is much lower for the kinase, this enzyme has limited capacity and may be readily saturated by its substrate. Thus, the relative proportion of metabolism via phosphorylation or dephosphorylation pathways may relate to the basal and
stimulated accumulation of Ins(I,4,5)P3within discrete compartments in the cell. This is clearly an important area for further investigation even though a definitive physiological role for Ins(l,3,4,5)P4in smooth muscle is yet to be established. In view of the problems associated with quantifying changes in ['H]lns(l,4,5)P 3 in ['H]inositol prelabeled ASM (i.e., agoniststimulated increases in phosphoinositidespecific radioactivity, relatively low amounts of [3H]lnsP 3accumulating, and dependence on methods to separate Ins(l,3,4)P 3 and Ins(l,4,5)p3' usually high performance liquid chromatography), we have used an alternative radioreceptor assay method based on the specific binding of Ins(l,4,5)p3 to its putative intracellular receptor in an adrenal cortex preparation to follow receptor-mediated changes in Ins(l,4,5)P 3 (39, 40, 40a). The major advantages of this method are that it provides a measurement of Ins(l,4,5)p3 mass and that the Ins(l,3,4)P 3 isomer does not crossreact at this receptor. Using this assay, carbachol produces a rapid but transient increase in Ins(I,4,5)P3 concentration in bovine tracheal smooth muscle from a resting value of 12.9 ± 0.8 to 27 ± 1.5 pmol/mg protein at 5 s (figure 2). Since the muscarinic receptor linked to phospholipase C does not appear to
Agonist
---~--
~-B~
)LLDG
Ptdlns(4,5)P 2
I I
T
, I I
Fig 1. Receptor-mediated generation of inositol 1,4,5-trisphosphate from phosphatidylinositol 4,5-bisphosphate and its initial metabolism to inositol-1,3,4,5-tetrakisphosphate or inositol 1,4-bisphosphate.
5139
PHOSPHOINOSITIDE METABOLISM IN AIRWAY SMOOTH MUSCLE
'2
30 -.---
-
-
-
- --
-
-
'iii
Fig. 2. Time-course for carbachol-stimulated inositoI1,4,5-trisphosphate accumulationinbovinetracheal smoothmuscleslices. 100 I'M carbachol (e) or vehicle (0) was added to aliquotsof bovinetracheal smooth muscle slices and incubationscontinuedfor time periods indicated. Incubations wereterminated and tissue extracts prepared and assayed forIns(1 ,4,5)P3 as described inChalliss and colleagues (40). Results are presented as mean ± SEMforsixto nine independent observations made in three separate experiments.
o...
Q.
25
-
-
-
-
---,
•
0.1mM Carbacho l
o
Control
ential regulation together with the site of action of cAMP remains to be established. In conclusion, there is now firm evidence supporting a physiologic role of Ins(I,4,5)P 3 as a second messenger responsible for excitation-contraction coupling in ASM and this system affords several potential target sites for modifying ASM contraction.
Cl
.§ '0
20
E
Q.
desensitizesignificantlyin this tissue, suggested by the continued breakdown of PtdIns(4,5)p 2 in the presence of agonist (12, 13, 17), the subsequent prompt decline in Ins(I,4,5)P3 concentration indicates a rapid activation of the pathways responsible for Ins(1,4,5)p 3metabolism within the cell. This response is somewhat different to that observed in rat cerebral cortex slices where Ins(1,4,5)p3concentration remains persistently elevated throughout the period of carbachol stimulation (40). This suggests that whereas in peripheral tissues, such as ASM, Ins(1,4,5)p3concentration and Ins(I,4,5)P.-induced Ca H release are closely related, in brain additional factors, such as Ca H inhibition of Ins(1,4,5)P 3 binding to its receptor (41), may allow subsequently independent regulation of Ins(1,4,5)P3 concentration and CaH release. In either case, this scheme would also allow ongoing DAG formation and protein kinase C activation, which are thought to be responsible for the maintenance of ASM contraction (42), without concomitant Ca H release. Finally, the ability of Ins(I,4,5)P 3 to release Ca H from nonmitochondrial, ATP-dependent cellular stores has been well established in a variety of cell types (8, 43), including ASM (29, 44). Although the absolute increase in Ins(1,4,5)P 3 induced by receptor-agonists in bovine ASM is relatively small (40a), this is almost certainly complicated by local changes in discrete compartments close to the plasma membrane and by basal accumulation of this second messenger not associated with the Ins(l,4,5)P 3 receptor. It is also worth considering the fact that whereas carbachol and histamine can both increase the accumulation of Ins(I,4,5)P3 in bovine ASM, the maximal increase is much larger with the muscarinic agonist (40a). Although both agonists act as full contractile agonists, it is tempting to ascribe the differences in receptor reserve for contraction to the ability to accumulate initially the second messenger Ins(l,4,5)P 3.
Regulation of Ins(1,4,S}P3 A fundamental prerequisite of any intracellular second-messenger molecule is that its formation and subsequent metabolism must be carefully regulated to obtain the desired phys-
15
--~--- - - - - - - - -- - - - - - - - - -----~
Acknowledgment The collaboration of Drs. I. H. Batty, R. A. J. Challiss, P. J. Barnes, and Ms. E. D. Kennedy is gratefully acknowledged together with the secretarial assistance of Ms. Jenny Batty.
10
o
60
12 0
2 40
1 80
300
Time (sec)
References
iologic response intended. A scheme is presented in figure 3 whereby Ins(l,4,5)P 3 concentration in the cell may be acutely regulated by Ca H through the activation of the 3'-kinase (45) and possibly 5'-phosphatase (46) enzymes. Protein kinase C activation has also been shown to activate the 5'-phosphatase (47) in addition to influencing Ins(l,4,5)P3formation (48). Certainly, activation of protein kinase C by 200 nM phorbol 12-myristate 13acetate completely blocks the release of internal Ca H by histamine in cultured canine ASM cells (49). Recently a number of groups have demonstrated that elevating cAMP in the cell is also able to antagonize agonist-induced phospholipase C activation (16, 50). In ASM, this effect is more pronounced with histamine than with carbachol stimulation, although the mechanism underlying this differ-
1. Rodger IW, Biochemistry of airway smooth muscle contraction. In: Barnes PJ, Rodger IW, Thomson N, eds. Asthma: basic mechanisms and clinical management. London: Academic Press Ltd., 1988; 57-79. 2. Aksoy MO, Murphy RA, Kamm KE. Role of Ca2• and myosin light chain phosphorylation in regulation of smooth muscle. Am J Physiol 1982; 242:CI09-16. 3. Park S, Rasmussen H. Carbachol-induced protein phosphorylation changes in bovine tracheal smooth muscle. J Biol Chem 1986; 261:15734-9. 4. Coburn RF. Electromechanical coupling in canine trachealis muscle: acetylcholine contractions. Am J Physiol 1979; 236:CI77-84. 5. Ahmed F, Foster RW, Small RC, Weston AH. Some features of the spasmogenic actions of acetylcholine and histamine in guinea-pig isolated trachea. Br J Pharmacol 1984; 83:227-33. 6. Kirkpatrick CT. Excitation and contraction in bovine tracheal smooth muscle. J Physiol 1975; 244:263-81. 7. Ito Y, Itoh T. The roles of stored calcium in contractions of cat tracheal smooth muscle pro-
Agonist
.J,
8~-G-------J
-/"",- -8~ _ -- --?
.t-- ----.,1 C-AMP
11...----'
P LC
__ 1~
0
P HORBOL ESTERS
)
.~
P
>-=- 8 -
