Targeting the KIT activating switch control pocket: a novel mechanism to inhibit neoplastic mast cell proliferation and mast cell activation

Activating mutations in the receptor tyrosine kinase KIT, most notably KIT D816V, are commonly observed in patients with systemic mastocytosis. Thus, inhibition of KIT has been a major focus for treatment of this disorder. Here we investigated a novel approach to such inhibition. Utilizing rational drug design, we targeted the switch pocket (SP) of KIT which regulates its catalytic conformation. Two SP inhibitors thus identified, DP-2976 and DP-4851, were examined for effects on neoplastic mast cell proliferation and mast cell activation. Autophosphorylation of both wild type (WT) and, where also examined, KIT D816V was blocked by these compounds in transfected 293T cells, HMC 1.1 and 1.2 human mast cell lines; and in CD34⁺-derived human mast cells activated by stem cell factor (SCF). Both inhibitors induced apoptosis in the neoplastic mast cell lines and reduced survival of primary bone marrow mast cells from patients with mastocytosis. Moreover, the SP inhibitors more selectively blocked SCF potentiation of FcεRI-mediated degranulation. Overall, SP inhibitors represent an innovative mechanism of KIT inhibition whose dual suppression of KIT D816V neoplastic mast cell proliferation and SCF enhanced mast cell activation may provide significant therapeutic benefits.

Secretion of interleukin-1 receptor antagonist from human mast cells after immunoglobulin E-mediated activation and after segmental antigen challenge

Mast cells produce substances with antiinflammatory properties in addition to their capacity to release proinflammatory mediators. To further probe the antiinflammatory aspect of mast-cell function we investigated the ability of human mast cells (huMCs) to produce interleukin (IL)-1 receptor antagonist (IL-1ra) in response to high-affinity Fc receptor for immunoglobulin E (Fcalpha RI) aggregation, and examined IL-1ra in bronchoalveolar lavage fluid (BALF) to determine whether it might be of mast-cell origin. Using a ribonuclease protection assay, flow cytometry, and enzyme-linked immunosorbent assay (ELISA), IL-1ra message and protein were found to be constitutively expressed in cultured huMCs. Upon stimulation through Fcalpha RI, IL-1ra message was upregulated in huMCs and IL-1ra protein secreted from cultured huMCs and isolated human lung mast cells. By immunoblot analysis, huMCs were found to produce the 17-kD form of IL-1ra and the presence of IL-1ra in human lung mast cells was confirmed by immunohistochemistry. In BALF obtained from allergic asthmatic subjects, IL-1ra production increased after specific antigen challenge, with the 17-kD isoform of IL-1ra predominating. These findings demonstrate that huMCs produce and release IL-1ra after Fcalpha RI aggregation, which may contribute to a local inhibition of IL-1-dependent effects on inflammation in the lung.

IgG-dependent activation of human mast cells following up-regulation of FcgammaRI by IFN-gamma

It has been reported that FcgammaRI is up-regulated on human mast cells (huMC) by IFN-gamma and aggregation of this receptor using mouse F(ab')(2) specific for receptor-bound, mouse anti-CD64 F(ab')(2) results in activation. To determine whether huMC can similarly be stimulated by aggregation of FcgammaRI-bound human IgG, IFN-gamma-treated, CD34(+)-derived, cultured huMC were sensitized with human immunoglobulins and activation was evaluated following addition of antibodies specific for each IgG isotype. Degranulation was also examined following simultaneous IgG- and IgE-dependent aggregation of FcgammaRI and Fc(epsilon)RI. Activation of IFN-gamma-treated huMC sensitized with 100 ng/ml IgG(1) resulted in 40% beta-hexosaminidase (beta-hex) release; minimal degranulation was observed using IgG(2), IgG(3) or IgG(4). IgG(1)-dependent activation led to PGD(2) and LTC(4) generation as well as elevated cytokine production, most notably TNF-alpha. Preincubation of cells with F(ab')(2) from CD64-specific clones 10.1 and 32.2 reduced IgG(1)-mediated beta-hex release by 46% and 74%, respectively. While IgG-dependent cell stimulation induced half-maximal degranulation by 11 min, IgE-dependent activation resulted in half maximal responses within 1 min. Simultaneous activation of huMC via FcgammaRI and Fc(epsilon)RI led to additive degranulation using suboptimal concentrations of IgG(1) and IgE. Activation of huMC thus may occur via monomeric IgG and FcgammaRI thereby providing a novel paradigm for huMC recruitment into inflammation.

Induction of Telomerase Activity During Development of Human Mast Cells from Peripheral Blood CD34+Cells: Comparisons with Tumor Mast-Cell Lines

To further characterize the development of mast cells from human hemopoietic pluripotent cells we have investigated the expression of telomerase activity in cultured human peripheral blood CD34+ cells, and CD34+ /CD117+ /CD13+ progenitor mast cells selected therefrom, with the idea that induction of telomerase is associated with clonal expansion of CD34+ /CD117+ /CD13+ cells. A rapid increase in telomerase activity preceded proliferation of both populations of cells in the presence of stem cell factor and either IL-3 or IL-6. The induction was transient, and telomerase activity declined to basal levels well before the appearance of mature mast cells. Studies with pharmacologic inhibitors suggested that this induction was initially dependent on the p38 mitogen-activated protein kinase and phosphatidylinositol 3'-kinase, but once cell replication was underway telomerase activity, but not cell replication, became resistant to the effects of inhibitors. Tumor mast cell lines, in contrast, expressed persistently high telomerase activity throughout the cell cycle, and this expression was unaffected by inhibitors of all known signaling pathways in mast cells even when cell proliferation was blocked for extended periods. These results suggest that the transient induction of telomerase activity in human progenitor mast cells was initially dependent on growth factor-mediated signals, whereas maintenance of high activity in tumor mast cell lines was not dependent on intracellular signals or cell replication.

Human mast cell apoptosis is regulated through Bcl-2 and Bcl-XL

It is well established that human mast cell proliferation and maturation are regulated by kit ligand (stem cell factor). Little is known, however, about how these two processes are negatively regulated and thus, how mast cell number is controlled in normal and pathologic conditions. We therefore first hypothesized that SCF-dependent human mast cells would undergo programmed cell death (apoptosis) on removal of SCF as has been shown for growth factor-dependent rodent mast cells. We then examined whether SCF acts as a survival factor through the regulation of the bcl-2 family of apoptosis-regulatory genes. As hypothesized, elimination of SCF from primary peripheral blood-derived human mast cell cultures resulted in a significant apoptotic process. During apoptosis, down-regulation of the two apoptosis-regulatory proteins Bcl-2 and Bcl-XL was observed. Moreover, a deregulated expression of these two proteins was found in two human mast cell lines which are SCF-independent. Thus, SCF functions as a survival factor by repressing apoptosis of human mast cells through Bcl-2 and Bcl-XL. Deregulated expression of these antiapoptotic proteins may contribute to proliferation and accumulation of mast cells in certain forms of systemic mast cell disorders.

Mast cells migrate, but do not degranulate, in response to fractalkine, a membrane-bound chemokine expressed constitutively in diverse cells of the skin

Mast cells (MC) are anatomically located near nerves and blood vessels in skin and the gastrointestinal tract and tend to localize within certain cutaneous tumors such as neurofibromas. However, the molecular mechanisms by which MC home to these sites are not well characterized. Fractalkine (FK) is a membrane-bound CX3C chemokine that displays constitutive expression in dendritic cells as well as in non-hematopoietic tissues including mammalian brain. Here we show that FK is constitutively expressed by skin endothelial cells, dermal dendrocytes and cells within neurofibromas. By reverse transcription-PCR, FK receptor, CX3CR1, is expressed by cultured murine bone marrow-derived MC (BMMC) of both connective tissue and mucosal phenotypes. Non-activated human dermal MC isolated from neonatal foreskin similarly demonstrated CX3CR1 expression. In chemotaxis assays, FK attracted MC with maximal migration occurring between 25 - 125 ng / ml. BMMC were not stimulated to release proinflammatory mediators in the presence of FK as measured by granule-associated beta-hexosaminidase release. Thus, CX3CR1 is expressed by MC and effectively mediates chemotaxis without inducing degranulation. We propose that the constitutive expression of FK on certain cells in the skin may be a factor in the tissue-specific homing of MC.

Interaction of phosphorylated FcepsilonRIgamma immunoglobulin receptor tyrosine activation motif-based peptides with dual and single SH2 domains of p72syk. Assessment of binding parameters and real time binding kinetics

To examine the characteristics of the interaction of the FcepsilonRIgamma ITAM with the SH2 domains of p72(syk), the binding of an 125I-labeled dual phosphorylated FcepsilonRIgamma ITAM-based peptide to the p72(syk) SH2 domains was monitored utilizing a novel scintillation proximity based assay. The Kd for this interaction, determined from the saturation binding isotherm, was 1.4 nM. This high affinity binding was reflected in the rapid rate of association for the peptide binding to the SH2 domains. Competition studies utilizing a soluble C-terminal SH2 domain knockout and N-terminal SH2 domain knockouts revealed that both domains contribute cooperatively to the high affinity binding. Unlabeled dual phosphorylated peptide competed with the 125I-labeled peptide for binding to the dual p72(syk) SH2 domains with an IC50 value of 4.8 nM. Monophosphorylated 24-mer FcepsilonRIgamma ITAM peptides, and phosphotyrosine also competed for binding, but with substantially higher IC50 values. This, and other data discussed, suggest that high affinity binding requires both tyrosine residues to be phosphorylated and that the preferred binding orientation of the ITAM is such that the N-terminal phosphotyrosine occupies the C-terminal SH2 domain and the C-terminal phosphotyrosine occupies the N-terminal SH2 domain.

Rapid dephosphorylation of the GTPase dynamin after FcepsilonRI aggregation in a rat mast cell line

As part of our studies aimed at exploring the potential role(s) of protein phosphatases in mast cell signaling, we analyzed the phosphorylation status of tyrosine-containing proteins in a rat mast (RBL) cell line that expresses both native rat high affinity IgE receptors (FcepsilonRI) and functional human FcepsilonRIalpha. After FcepsilonRI aggregation, there was a rapid increase in the tyrosine phosphorylation of a number of proteins, including those of m.w. 72 and 110 kDa. Concurrent with these events, however, there was a rapid dephosphorylation of a 100-kDa protein that was constitutively phosphorylated in the unstimulated cells. Using a specific mAb, this 100-kDa protein was identified as the GTPase dynamin. Dynamin was shown to associate with the SH3 domain of the src-related tyrosine kinase p56lyn in RBL 2H3 cells both in vitro and in vivo. FcepsilonRI aggregation causes rapid internalization of the aggregated receptors via clathrin-coated pits and dynamin is known to play a role in clathrin-mediated endocytosis, so the dephosphorylation of dynamin may provide the signal for targeting the aggregated receptors to the endocytic pathway.

Rapid dephosphorylation of the GTPase dynamin after FcepsilonRI aggregation in a rat mast cell line

As part of our studies aimed at exploring the potential role(s) of protein phosphatases in mast cell signaling, we analyzed the phosphorylation status of tyrosine-containing proteins in a rat mast (RBL) cell line that expresses both native rat high affinity IgE receptors (FcepsilonRI) and functional human FcepsilonRIalpha. After FcepsilonRI aggregation, there was a rapid increase in the tyrosine phosphorylation of a number of proteins, including those of m.w. 72 and 110 kDa. Concurrent with these events, however, there was a rapid dephosphorylation of a 100-kDa protein that was constitutively phosphorylated in the unstimulated cells. Using a specific mAb, this 100-kDa protein was identified as the GTPase dynamin. Dynamin was shown to associate with the SH3 domain of the src-related tyrosine kinase p56lyn in RBL 2H3 cells both in vitro and in vivo. FcepsilonRI aggregation causes rapid internalization of the aggregated receptors via clathrin-coated pits and dynamin is known to play a role in clathrin-mediated endocytosis, so the dephosphorylation of dynamin may provide the signal for targeting the aggregated receptors to the endocytic pathway.

Functional contributions of the FcepsilonRIalpha and FepsilonRIgamma subunit domains in FcepsilonRI-mediated signaling in mast cells

The functional contributions of the alpha and gamma subunit domains of the high affinity receptor for IgE (Fcepsilon-RI) were determined following chimeric receptor aggregation. Chimeric receptors of the extracellular (EC) and cytoplasmic tail (CT) domains of FcepsilonRI and the IL-2R p55 subunit (I) were constructed and stably expressed in RBL-2H3 cells. Signaling (inositol phosphate production, tyrosine phosphorylation, Ca2+ mobilization, and secretion of histamine and arachidonic acid metabolites) via alpha/gamma/gamma or I/gamma/gamma was similar to the native rat receptor, and both were shown to associate with endogenous FcepsilonRIbeta and FcepsilonRIgamma subunits. Therefore, the contributions of the EC domains could not be evaluated. The chimeras alpha/I/gamma and I/I/gamma were found to be single polypeptide chains, as they did not associate with beta and gamma. Signaling via alpha/I/gamma resulted in the appearance of biochemical events common to the native receptor. Cross-linking I/I/gamma elicited histamine release, [14C]arachidonic acid metabolites, tyrosine phosphorylation, Ca2+ mobilization, and only inositol trisphosphate production, which were not of a similar magnitude to the native FcepsilonRI. No biochemical events were elicited by cross-linking alpha/I/I or I/I/I. These results demonstrate that both the FcepsilonRIalpha EC domain and the FcepsilonRIgamma CT domain are essential for the FcepsilonRI signaling process, and that while FcepsilonRIIgamma CT plays a critical role in FepsilonRI signaling, the EC domain of FcepsilonRIalpha has a major contribution in signaling, as well as a role in modulating the magnitude of the biochemical events.

Functional contributions of the FcepsilonRIalpha and FepsilonRIgamma subunit domains in FcepsilonRI-mediated signaling in mast cells

The functional contributions of the alpha and gamma subunit domains of the high affinity receptor for IgE (Fcepsilon-RI) were determined following chimeric receptor aggregation. Chimeric receptors of the extracellular (EC) and cytoplasmic tail (CT) domains of FcepsilonRI and the IL-2R p55 subunit (I) were constructed and stably expressed in RBL-2H3 cells. Signaling (inositol phosphate production, tyrosine phosphorylation, Ca2+ mobilization, and secretion of histamine and arachidonic acid metabolites) via alpha/gamma/gamma or I/gamma/gamma was similar to the native rat receptor, and both were shown to associate with endogenous FcepsilonRIbeta and FcepsilonRIgamma subunits. Therefore, the contributions of the EC domains could not be evaluated. The chimeras alpha/I/gamma and I/I/gamma were found to be single polypeptide chains, as they did not associate with beta and gamma. Signaling via alpha/I/gamma resulted in the appearance of biochemical events common to the native receptor. Cross-linking I/I/gamma elicited histamine release, [14C]arachidonic acid metabolites, tyrosine phosphorylation, Ca2+ mobilization, and only inositol trisphosphate production, which were not of a similar magnitude to the native FcepsilonRI. No biochemical events were elicited by cross-linking alpha/I/I or I/I/I. These results demonstrate that both the FcepsilonRIalpha EC domain and the FcepsilonRIgamma CT domain are essential for the FcepsilonRI signaling process, and that while FcepsilonRIIgamma CT plays a critical role in FepsilonRI signaling, the EC domain of FcepsilonRIalpha has a major contribution in signaling, as well as a role in modulating the magnitude of the biochemical events.

Functional characterization of the signal transduction events mediated by Fc epsilon RI alpha and gamma chimeric receptors

Chimeric receptors containing the Fc epsilon RI alpha and gamma subunit domains were constructed, stably transfected into RBL-2H3 cells, and characterized for the biochemical events which are elicited upon receptor aggregation. Chimeric receptors containing the extracellular (EC) domain of the human Fc epsilon RI alpha subunit, or the EC domain of the p55 subunit of the interleukin-2 receptor were fused to the human Fc epsilon RI gamma subunit transmembrane and cytoplasmic (CT) domains or only the CT domain. The chimeras generated included alpha/gamma/gamma, I/gamma/gamma, alpha/I/gamma or I/I/gamma. The results indicate that both the Fc epsilon RI alpha EC domain and the Fc epsilon RI alpha CT domain are essential for signalling.

Temporal regulation of the IgE-dependent 1,2-diacylglycerol production by tyrosine kinase activation in a rat (RBL 2H3) mast-cell line

We explored the possible role of tyrosine kinases in the IgE-dependent regulation of 1,2-diacylglycerol (DAG) production in RBL 2H3 cells. When triggered via their high-affinity IgE receptors (Fc epsilon RI), there was a rapid phosphorylation of tyrosine residues on a number of proteins. The phosphorylation of these proteins and ultimately histamine release were inhibited in a concentration-dependent manner by the tyrosine kinase inhibitor, tyrphostin. In cells labelled with [3H]myristic acid, we observed a characteristic biphasic increase in [3H]DAG production. In the presence of tyrosine kinase inhibitor, the initial increase in DAG was still observed, but the secondary increase, which was dependent on phosphatidylcholine-specific phospholipase D (PC-PLD) activation, was completely abolished. Tyrphostin significantly inhibited IgE-dependent activation of PC-PLD, suggesting that PC-PLD activation was regulated by tyrosine phosphorylation. Furthermore, when proteins from RBL 2H3 cells were immunoprecipitated with an anti-phosphotyrosine antibody, PC-PLD activity was recovered from the immunoprecipitated fraction. These results demonstrate that the secondary, but not the initial, phase of 1,2-DAG production in response to Fc epsilon RI aggregation is regulated by the initial activation of tyrosine kinases and that PC-PLD may be regulated directly by this mechanism.

Conservation of signal transduction mechanisms via the human Fc epsilon RI alpha after transfection into a rat mast cell line, RBL 2H3

The high affinity receptor for IgE (Fc epsilon RI) is present on mast cells and basophils, and the aggregation of IgE-occupied receptors by Ag is responsible for the release of allergic mediators. The Fc epsilon RI is composed of at least three different subunits, alpha, beta, and gamma, with the alpha subunit binding IgE. The series of biochemical events linking receptor aggregation to the release of mediators has not been fully delineated. As a step towards understanding these processes, and for the development of functional cell lines, we have transfected the human Fc epsilon RI alpha subunit into the rat mast cell line RBL 2H3. These human Fc epsilon RI alpha-transfected cell lines have been characterized with respect to the association of the human alpha subunit with endogenous rat beta and gamma subunits and the ability of aggregated Fc epsilon RI alpha subunits to mediate a variety of biochemical events. The signal transduction events monitored include phosphoinositide hydrolysis, Ca2+ mobilization, tyrosine phosphorylation, histamine release, and arachidonic acid metabolism. In all cases, the events mediated by aggregating human Fc epsilon RI alpha subunits were indistinguishable from those produced via the rat Fc epsilon RI alpha. These results demonstrate that the human Fc epsilon RI alpha subunit can functionally substitute for the rat Fc epsilon RI alpha subunit during signal transduction. The availability of this cell line will provide a means of evaluating potential Fc epsilon RI antagonists.

Conservation of signal transduction mechanisms via the human Fc epsilon RI alpha after transfection into a rat mast cell line, RBL 2H3

The high affinity receptor for IgE (Fc epsilon RI) is present on mast cells and basophils, and the aggregation of IgE-occupied receptors by Ag is responsible for the release of allergic mediators. The Fc epsilon RI is composed of at least three different subunits, alpha, beta, and gamma, with the alpha subunit binding IgE. The series of biochemical events linking receptor aggregation to the release of mediators has not been fully delineated. As a step towards understanding these processes, and for the development of functional cell lines, we have transfected the human Fc epsilon RI alpha subunit into the rat mast cell line RBL 2H3. These human Fc epsilon RI alpha-transfected cell lines have been characterized with respect to the association of the human alpha subunit with endogenous rat beta and gamma subunits and the ability of aggregated Fc epsilon RI alpha subunits to mediate a variety of biochemical events. The signal transduction events monitored include phosphoinositide hydrolysis, Ca2+ mobilization, tyrosine phosphorylation, histamine release, and arachidonic acid metabolism. In all cases, the events mediated by aggregating human Fc epsilon RI alpha subunits were indistinguishable from those produced via the rat Fc epsilon RI alpha. These results demonstrate that the human Fc epsilon RI alpha subunit can functionally substitute for the rat Fc epsilon RI alpha subunit during signal transduction. The availability of this cell line will provide a means of evaluating potential Fc epsilon RI antagonists.

Phosphatidylcholine-specific phospholipase D-derived 1,2-diacylglycerol does not initiate protein kinase C activation in the RBL 2H3 mast-cell line

We examined the role of phosphatidylcholine-specific phospholipase D (PC-PLD) in the IgE-dependent activation of protein kinase C (PKC) in RBL 2H3 cells (a model for mast-cell function). Cells were sensitized with mouse monoclonal anti-trinitrophenol (TNP) IgE (0.5 micrograms/ml) and were then triggered with an optimal concentration (10 ng/ml) of TNP-ovalbumin conjugate (TNP-OVA). This resulted in an immediate biphasic increase in the production of 1,2-diacylglycerol (DAG) and activation of PKC. The initial increase in DAG production reached a peak within 30 s, and the second phase reached a plateau within 5 min after stimulation. TNP-OVA-induced PC-PLD activation followed the initial increase in DAG formation in response to IgE-receptor cross-bridging, but coincided with the second peak. Phosphatidic acid (PA), derived from the PC-PLD pathway, is metabolized to DAG by the action of PA phosphohydrolase (PAPase). Propranolol (0.3 mM), which inhibits PAPase, blocked the IgE-dependent increase in DAG, activation of PKC, and subsequently degranulation. The PKC inhibitor staurosporine (0.1 microM) inhibited the second, but not first, peak of DAG accumulation, reversed PKC translocation after 10 min and inhibited subsequent mediator release. Taken together, these results demonstrate that PC-PLD does not initiate, but may play a latent role in, IgE-dependent DAG production, PKC activation and mediator release from RBL 2H3 cells.

The role of calcium and protein kinase C in the IgE-dependent activation of phosphatidylcholine-specific phospholipase D in a rat mast (RBL 2H3) cell line

Our previous studies have suggested that phosphatidylcholine-specific phospholipase D (PtdCho-PLD) plays a role in IgE-dependent diacylglycerol production, protein kinase C activation and mediator release in the RBL 2H3 mast cell line. We have extended these studies to examine the mechanisms by which PtdCho-PLD may be regulated in these cells. RBL 2H3 cellular lipids were labeled with [14C]arachidonic acid or [3H]myristic acid, then PtdCho-PLD activity was monitored by the formation of radiolabeled phosphatidylethanol when ethanol was included in the incubation medium. Trinitrophenol-ovalbumin conjugate (10 ng/ml), when added to cells previously sensitized with anti-(trinitrophenelated mouse IgE) (0.5 microgram/ml), ionomycin (1 microM) and thapsigargin (0.1 microM), stimulated PtdCho-PLD activation and mediator release in cells incubated in buffer containing 1.8 mM calcium, but not in cells incubated in calcium-free, buffer. Phorbol 12-myristate 13-acetate (0.1 microM) activated PtdCho-PLD in both buffers, but on its own did not trigger mediator release. When intracellular calcium was chelated with 5,5'-dimethyl-1,2-bis(2- aminophenoxy)ethane-N,N,N',N'-tetraacetic acid, trinitrophenol-ovalbumin conjugate failed to activate PtdCho-PLD and histamine release. Similarly, down-regulation of protein kinase C activity by long-term exposure to the phorbol ester (0.1 microM) and preincubation of the cells with protein kinase inhibitors resulted in the loss of the trinitrophenol-ovalbumin response on PtdCho-PLD activity and histamine release. Taken together, the above results suggest that IgE-dependent PtdCho-PLD activation is dependent on both activation of protein kinase C and a rise in the intracellular free calcium concentration.

Differential effects of propranolol on the IgE-dependent, or calcium ionophore-stimulated, phosphoinositide hydrolysis and calcium mobilization in a mast (RBL 2H3) cell line

Our previous studies demonstrated that propranolol, an inhibitor of phosphatidic acid phosphohydrolase (PAPase) (EC 3.1.3.4) blocks the IgE-dependent mediator release from a rat mast (RBL 2H3) cell line. To continue these studies, we examined the ability of propranolol to inhibit the IgE-dependent or ionomycin-mediated phosphoinositide hydrolysis and calcium mobilization in RBL 2H3 cells. RBL 2H3 cells, sensitized with mouse monoclonal anti-trinitrophenol IgE (anti-TNP IgE), were stimulated to release both histamine and peptidoleukotrienes (LT) in response to a suboptimal concentration of trinitrophenol-ovalbumin conjugate (TNP-OVA) or ionomycin. Preincubation of the cells with d,l-propranolol (300 microM) significantly (P less than 0.05) inhibited the effects of both TNP-OVA and ionomycin on histamine and LT release. There was no difference in potency for the different isomers of propranolol, indicating that these effects were not a consequence of an effect on beta 2-adrenergic receptors. TNP-OVA produced a rapid hydrolysis of phosphoinositides resulting in a time-dependent increase in mono- (IP1), di- (IP2), tri- (IP3), and total inositol phosphate production. Ionomycin also produced a rapid increase in total inositol phosphate production; however, this largely reflected an accumulation of IP1. Both secretagogues produced a rapid elevation in cytosolic free calcium ([Ca2+]i); however, the effect of ionomycin maximized within a much shorter time frame than the effect of TNP-OVA. The effects of TNP-OVA on phosphoinositide hydrolysis and increase in [Ca2+]i were inhibited by propranolol over exactly the same concentration range as the effects of this compound on TNP-OVA-stimulated mediator release. In contrast, propranolol had no effect on the increase in [Ca2+]i and phosphoinositide hydrolysis in response to ionomycin. Taken together, these results suggest that PAPase/phospholipase D (PLD) (EC 3.1.4.4) activation may be a prerequisite for both IgE-dependent and ionomycin-stimulated mediator release from RBL 2H3 cells. Although other explanations are possible, the data further suggest that receptor-mediated, but not ionophore-stimulated, phosphoinositide hydrolysis and [Ca2+]i in RBL 2H3 cells may be regulated by a propranolol-sensitive pathway involving possible activation of PAPase.

Activation of phospholipase D in a rat mast (RBL 2H3) cell line. A possible unifying mechanism for IgE-dependent degranulation and arachidonic acid metabolite release

RBL 2H3 cells (a model of mast cell function) were sensitized with anti-TNP IgE (0.5 micrograms/ml) and triggered to secrete both histamine and arachidonic acid (AA) metabolites by the addition of TNP-OVA (0 to 100 ng/ml). After a 3-min delay, the release of both groups of mediators proceeded in a parallel manner. In cells labeled with [14C]-AA, TNP-OVA produced a rapid increase in phosphatidic acid (PA), and subsequently, 1,2-diacylglycerol (DAG) and intracellular AA levels. Concurrently, there was a decrease in [14C]-AA labeled phosphatidylcholine. The release of labeled AA from phosphatidylcholine in response to TNP-OVA was paralleled by a liberation of free choline but no evidence of liberation of phosphorylcholine. When ethanol (0.05 to 2% v/v) was included in the culture medium, phosphatidylethanol was synthesized at the expense of PA and DAG, with a resulting inhibition of secretion. D,1 propranolol, an inhibitor of PA phosphohydrolase, inhibited the IgE-dependent production of [14C]-DAG, and [14C]-free fatty acid but not [14C]-PA. The IgE-dependent release of both histamine and AA metabolites was completely inhibited by pretreatment with propranolol. Taken together, the above results suggest that phospholipase D is activated upon cross-bridging of IgE receptors on the surface of RBL 2H3 cells and that this may be a pivotal step in the signal transduction cascade leading to the release of both presynthesized and de novo synthesized mediators.

Activation of phospholipase D in a rat mast (RBL 2H3) cell line. A possible unifying mechanism for IgE-dependent degranulation and arachidonic acid metabolite release

RBL 2H3 cells (a model of mast cell function) were sensitized with anti-TNP IgE (0.5 micrograms/ml) and triggered to secrete both histamine and arachidonic acid (AA) metabolites by the addition of TNP-OVA (0 to 100 ng/ml). After a 3-min delay, the release of both groups of mediators proceeded in a parallel manner. In cells labeled with [14C]-AA, TNP-OVA produced a rapid increase in phosphatidic acid (PA), and subsequently, 1,2-diacylglycerol (DAG) and intracellular AA levels. Concurrently, there was a decrease in [14C]-AA labeled phosphatidylcholine. The release of labeled AA from phosphatidylcholine in response to TNP-OVA was paralleled by a liberation of free choline but no evidence of liberation of phosphorylcholine. When ethanol (0.05 to 2% v/v) was included in the culture medium, phosphatidylethanol was synthesized at the expense of PA and DAG, with a resulting inhibition of secretion. D,1 propranolol, an inhibitor of PA phosphohydrolase, inhibited the IgE-dependent production of [14C]-DAG, and [14C]-free fatty acid but not [14C]-PA. The IgE-dependent release of both histamine and AA metabolites was completely inhibited by pretreatment with propranolol. Taken together, the above results suggest that phospholipase D is activated upon cross-bridging of IgE receptors on the surface of RBL 2H3 cells and that this may be a pivotal step in the signal transduction cascade leading to the release of both presynthesized and de novo synthesized mediators.

The effects of the protein kinase C inhibitors staurosporine and H7 on the IgE dependent mediator release from RBL 2H3 cells

RBL 2H3 cells, a model for mast cell function, sensitized with rat IgE, released histamine and peptidoleukotrienes (LT) in response to rabbit anti-rat IgE in a concentration-dependent manner. The calcium ionophore, A23187 also stimulated the release of both mediators but to a greater extent. The protein kinase C activator, 12-O-tetradecanoyl phorbol-13-acetate (TPA) failed to influence mediator release when added alone, but when added with either A23187 or anti-IgE, TPA significantly enhanced the release of both histamine and LT. The effects of anti-IgE, TPA and A23187 were completely inhibited by prior addition of the protein kinase C inhibitors staurosporine and 1-(5-isoquinolinesulfonyl)-2-methylpiperazine dihydrochloride (H7) but not by N-(2-guanidinoethyl)-5-isoquinoline-sulfonamide dihydrochloride (HA1004), a compound which has similar potency to H7 as an inhibitor of some protein kinases but is less potent as a protein kinase C inhibitor. Although other explanations are possible, these results support the hypothesis that the release of histamine and leukotrienes from RBL 2H3 cells resulting from the cross bridging of the IgE receptors, is dependent on activation of protein kinase C.

Pertussis toxin pretreatment reveals differential effects of adenosine analogs on IgE-dependent histamine and peptidoleukotriene release from RBL-2H3 cells

The effects of adenosine (A) and the nonmetabolizable adenosine analogs, N-ethylcarboxamidoadenosine (NECA), L-phenylisopropyladenosine (L-PIA), D-PIA and 2-chloroadenosine (2CHA) were examined on the IgE-dependent mediator release from RBL-2H3 cells, a model for mast-cell function. Adenosine and the adenosine analogs failed to influence mediator release from cells, previously sensitized with monoclonal anti-TNP mouse immunoglobulin E (anti-TNP IgE), when added alone. When added prior to conjugated trinitrophenol-ovalbumin (TNP-OVA), adenosine and the adenosine analogs (10(-8)-10(-4) M) significantly potentiated the release of both histamine (marker for degranulation) and peptidoleukotrienes (LT) (marker for de novo synthesized mediators). The effects were concentration-dependent with the potency order being L-PIA greater than NECA greater than A greater than D-PIA, 2CHA. The stimulatory effect on both histamine and LT release were reversed by prior treatment of the cells with pertussis toxin but not by the purinoceptor antagonists, theophylline and 8-phenyltheophylline, nor adenosine uptake blockers. At higher concentrations (above 10(-5) M), adenosine and adenosine analogs were also inhibitory on LT but not on histamine release. This inhibition was more evident on pertussis-toxin-treated cells in which there was no effect of adenosine or adenosine analogs on histamine release, but a concentration-dependent inhibition of IgE-dependent LT release. These findings demonstrate that adenosine analogs have two distinct mechanisms on mediator release from RBL-2H3 cells; a stimulatory effect on both histamine and LT release, mediated via a pertussis-toxin-sensitive G protein and an inhibitory effect on LT release via a pertussis-toxin-insensitive pathway. An abstract of this work has been published.

Ro 19-3704 directly inhibits immunoglobulin E-dependent mediator release by a mechanism independent of its platelet-activating factor antagonist properties

Rat basophilic leukemia (RBL 2H3) cells were passively sensitized by exposure to monoclonal anti-trinitrophenol mouse immunoglobulin E (anti-trinitrophenol IgE) (0.5 microgram/ml) and triggered by exposure to a sub-optimal concentration of trinitrophenol ovalbumin conjugate (5 ng/ml). At this concentration, trinitrophenol-ovalbumin increased histamine release from a basal rate of 4.8 +/- 0.5 to 28.5 +/- 4.6% and peptidoleukotrienes from less than 0.1 to 4.2 +/- 1.3 ng/10(6) cells in the activated cells. Ro 19-3704 and Ro 19-1400, platelet activating factor (PAF) antagonists which are structural analogs of PAF, potently inhibited both the IgE-dependent release of histamine (IC50 values of 3.0 and 3.6 microM, respectively) and LT release (IC50 values of 5.0 microM for both compounds) from the cells. These effects appeared to be independent to the ability of the compounds to act as PAF antagonists since PAF on its own had no effect on mediator release, and WEB 2086 and BN 52021, structurally distinct PAF antagonists, were relatively ineffective as inhibitors of mediator release. Ro 19-3704 and Ro 19-1400 were observed to be potent inhibitors of the soluble phospholipase A2 activity in synovial fluid from rheumatoid arthritic patients (IC50 values of 6.5 and 8.4 microM, respectively). In contrast, WEB 2086 and BN 52021 had no effect on this phospholipase A2. Ro 19-3704 significantly inhibited the IgE-dependent formation of inositol phosphates in RBL 2H3 cells (IC50 value of 7.0 microM). These data suggest that the mediator release inhibitory action of these compounds may be related to the ability of these compounds to inhibit phospholipase A2 and/or phospholipase C.

Functional evidence for involvement of P2 purinoceptors in the ATP stimulation of phosphatidylcholine secretion in type II alveolar epithelial cells

There is evidence that phosphatidylcholine secretion in type II pneumocytes is stimulated by adenosine and adenine nucleotides and that the effect of adenosine is mediated by the A2 subtype of the P1 purinoceptor. To determine if the effect of ATP is also mediated by the same receptor following its catabolism to adenosine or by the P2 purinoceptor we compared the effects of adenosine and ATP. Adenosine and terbutaline stimulated phosphatidylcholine secretion approx. 2-fold, while ATP stimulated it by more than 3-fold, essentially to the same extent as the protein kinase C activator, 12-O-tetradecanoylphorbol 13-acetate. The stimulatory effect of adenosine but not of ATP was abolished by adenosine deaminase. The effect of ATP was markedly diminished by the P2 desensitizing agent alpha,beta-methylene ATP, but only slightly by the P1 antagonist 8-phenyltheophylline. Adenosine increased the cAMP content of type II cells while ATP had little effect. The effects of ATP and terbutaline were additive while those of adenosine and terbutaline were not. These data show that ATP and adenosine stimulate phosphatidylcholine secretion via different mechanisms. Therefore, the effect of ATP is not mediated via catabolism to adenosine. Metabolically resistant analogs of ATP also stimulated secretion in a concentration-dependent manner although none were as potent as ATP. The order of potency was ATP greater than beta,gamma-methylene ATP = 2-methylthio ATP = 2-deoxy ATP greater than or equal to 8-bromo ATP greater than alpha,beta-methylene ATP. The facts that ATP analogs also stimulate secretion and that the effect of ATP was antagonized by alpha,beta-methylene ATP suggest that the stimulatory effect of ATP is mediated by the P2 purinoceptor.

Effect of 1-oleoyl-2-acetylglycerol and other lipids on phosphatidylcholine synthesis and cholinephosphate cytidylyltransferase activity in cultured type II pneumocytes

We previously reported that addition of phosphatidylglycerol to the culture medium stimulates phosphatidylcholine synthesis and cholinephosphate cytidylyltransferase activity in type II pneumocytes. In view of the known biological effects of diacylglycerols and since phosphatidylglycerol could be metabolized to diacylglycerol, we now examined the effects of diacylglycerols on the same parameters. The rate of choline incorporation into phosphatidylcholine was increased 30-60% by 10 microM phosphatidylglycerol, diolein, mixed diacylglycerols and 1-oleoyl-2-acetylglycerol (OAG). The effects of phosphatidylglycerol and OAG were not additive, suggesting a similar mechanism of action. The diacylglycerols and phosphatidylglycerol increased the activity of cholinephosphate cytidylyltransferase in type II cell sonicates by 35-50%, but had no effect on the activities of choline kinase, cholinephosphotransferase or 1-acylglycerophosphocholine acyltransferase. Again, the effects of OAG and phosphatidylglycerol on cytidylyltransferase were not additive. It is known that addition of lipids to the assay mixture increases the activity of cholinephosphate cytidylyltransferase in vitro and inclusion of the above lipids (1.1 mM) in the in vitro assay mixture increased cytidylyltransferase activity in type II cell sonicates. In addition, the stimulatory effects of OAG and of diolein, as well as of phosphatidylglycerol as reported previously, in the culture medium on cytidylyltransferase activity in type II cells were diminished or abolished when the assay was carried out in the presence of sufficient amounts of the same lipids to stimulate maximally the activity in vitro. These data show that lipids in the culture medium stimulate phosphatidylcholine biosynthesis in type II cells by direct activation of cholinephosphate cytidylyltransferase.

Functional evidence for adenosine A2 receptor regulation of phosphatidylcholine secretion in cultured type II pneumocytes

We examined the effects of P1 purinoceptor agonists on secretion of phosphatidylcholine in primary cultures of rat type II pneumocytes. Adenosine and its nonmetabolizable analogs, 5'-N-ethylcarboxyamidoadenosine (NECA), N6-phenylisopropyladenosine (PIA) and 2-chloroadenosine, increased secretion, and this effect was dependent on concentration. At the highest concentration, all agonists stimulated phosphatidylcholine secretion approximately 2-fold. NECA, with an EC50 of 8.9 X 10(-8) M, was the most potent agonist. The order of potency was NECA greater than 2-chloroadenosine = L-PIA greater than or equal to adenosine greater than D-PIA. The stimulatory effect of 10(-6) M NECA was diminished by the P1 antagonists theophylline and 8-phenyltheophylline. The degree of stimulation by the adenosine analogs as well as its time course was the same as that induced by terbutaline. The effects of the adenosine analogs and of terbutaline were not additive, suggesting a similar mode of action for the beta adrenergic and purinoceptor agonists. Terbutaline and the adenosine analogs increased intracellular cyclic AMP levels. Again, NECA was the most potent adenosine analog. For NECA, L-PIA and adenosine, there was a significant correlation between effects on secretion and on cyclic AMP content. These data suggest that the effect of adenosine on phosphatidylcholine secretion in type II pneumocytes is mediated by the A2 subtype of the P1 purinoceptor.

Effects of thiazinamium chloride and other antihistamines on phosphatidylcholine secretion in rat type II pneumocyte cultures

Thiazinamium chloride (TCl) stimulated phosphatidylcholine secretion in cultures of adult rat type II pneumocytes in a concentration-dependent manner in the range 10(-9)-10(-6) M. At the optimal concentration, secretion was stimulated by 46% which is approximately half the stimulatory effect of the beta-agonists terbutaline and isoproterenol. TCl did not increase the rate of choline incorporation into cellular phosphatidylcholine or of lactate dehydrogenase release so its effect on secretion was not secondary to phosphatidylcholine synthesis or cell injury. Since TCl has antihistaminic properties, we examined the effects of other antihistamines. The H-1 antagonists promethazine, which is structurally similar to thiazinamium, and pyrilamine, which has a different structure, also stimulated secretion but the H-2 antagonist, cimetidine, did not. The effects of TCl and pyrilamine were additive to those of terbutaline, suggesting that the mechanisms of action of the antihistamines and the beta-agonist were different. Although we were unable to demonstrate an inhibitory effect of histamine itself on either basal or terbutaline-stimulated phosphatidylcholine secretion, it is possible that histamine plays a regulatory role in lung surfactant secretion.

Purinoceptor agonists stimulate phosphatidylcholine secretion in primary cultures of adult rat type II pneumocytes

We examined the effect of purinoceptor agonists on phosphatidylcholine secretion in primary cultures of type II pneumocytes from adult rats. Surfactant is a major product of the type II cell and phosphatidylcholine is its principal component. Adenosine, AMP, ADP and ATP stimulated phosphatidylcholine secretion in a concentration-dependent manner. At the optimum concentration (1 mM), adenosine and AMP stimulated phosphatidylcholine secretion more than 2-fold, while ATP stimulated 5-fold and ADP almost 7-fold. Because of the magnitude of the response it is tempting to speculate that secretion of surfactant may be under purinoceptor regulation. None of these agents influenced cellular phosphatidylcholine synthesis or lactate dehydrogenase release into the medium, so the effects were primarily on secretion and were not secondary to effects on synthesis or cell damage. Non-metabolizable analogs of adenosine, 5'-N-ethyl-carboxyamidoadensoine (NECA) and L-N6-phenylisopropyladenosine (L-PIA), stimulated secretion to the same extent as adenosine and the effect of NECA was antagonized by 8-phenyltheophylline, suggesting a P1 purinoceptor-mediated mechanism. The stimulatory effect of ATP was diminished by alpha, beta-methylene ATP but only slightly by 8-phenyltheophylline, suggesting that, although part of the ATP effect could be explained by catabolism to adenosine, the P2 purinoceptor may also be involved in regulation of surfactant secretion.

Comparison of the enzyme activities of phosphatidylcholine, phosphatidylglycerol and phosphatidylinositol synthesis in freshly isolated type II pneumocytes and whole lung from the adult rat

We compared the activities of enzymes of phosphatidylcholine, phosphatidylglycerol and phosphatidylinositol synthesis in whole lung tissue and freshly isolated type II pneumocytes from adult rats. The activities of 1-acylglycerophosphocholine acyltransferase and CDPdiacylglycerol-glycerol-3-phosphate 3-phosphatidyltransferase were 2.9- and 4.4-fold higher, respectively, in type II cell sonicates than in whole lung homogenates. There was little difference between the type II cells and whole lung in the activities of choline kinase, choline-phosphate cytidyltransferase, cholinephosphotransferase, phosphatidate phosphatase, phosphatidate cytidylytransferase or CDPdiacylglycerol-inositol 3-phosphatidyltransferase. Since the type II cell is the source of pulmonary surfactant, and disaturated phosphatidylcholine and phosphatidylglycerol are major components of surfactant, it is of interest that this cell is enriched in the activities of enzymes exclusively involved in the synthesis of these lipids. In view of possible proteolytic damage during isolation we compared freshly isolated type II cells with those cultured for 1 day. The rates of incorporation of [methyl-3H]choline and [2-3H]glycerol into phospholipids, L-[U-14C]phenylalanine into protein and [methyl-3H]thymidine into DNA were the same in the freshly isolated and cultured cells. The composition of the phospholipids synthesized from [2-3H]glycerol and sodium [1-14C]acetate were also the same. The freshly isolated cells were at least 90% pure and did not release significant amounts of lactate dehydrogenase. Since use of freshly isolated cells avoids cell loss during culture they provide an attractive alternative, particularly in studies requiring large amounts of material.

Leukotrienes stimulate phosphatidylcholine secretion in cultured type II pneumocytes

We previously reported that arachidonic acid stimulates secretion of phosphatidylcholine in cultures of type II pneumocytes and, based on studies with cyclooxygenase and lipoxygenase inhibitors, suggested that this effect was mediated by lipoxygenase products of arachidonic acid metabolism (Gilfillan, A.M. and Rooney, S.A. (1985) Biochim. Biophys. Acta 833, 336-341). We have now examined the effect of leukotrienes on phosphatidylcholine secretion in type II cells as well as the effect of a leukotriene antagonist, FPL55712, on the stimulatory effect of arachidonic acid. Leukotrienes C4, D4 and E4 stimulated phosphatidylcholine secretion and this effect was dependent on concentration in the range 10(-12)-10(-6) M. Leukotriene E4 was the most stimulatory, followed by D4 and C4. Leukotriene B4 had no effect. Incubation of the cells with 10(-7) M leukotriene E4 for 90 min resulted in a 107% increase in the rate of phosphatidylcholine secretion. Incubation with 10(-6) M leukotrienes D4 and C4 for the same period resulted in 81% and 63% stimulation, respectively. The leukotrienes had no effect on cellular phosphatidylcholine synthesis or on lactate dehydrogenase release. The stimulatory effects of leukotrienes E4 and D4 were abolished by FPL55712. Similarly, the stimulatory effect of 6 X 10(-6) M arachidonic acid on phosphatidylcholine secretion was reduced from 74% to 25% by 10(-5) M FPL55712. Thus, the stimulatory effect of arachidonic acid on surfactant phospholipid secretion in type II cells is mediated at least in part by leukotrienes.

Phosphatidylglycerol stimulates cholinephosphate cytidylyltransferase activity and phosphatidylcholine synthesis in type II pneumocytes

Phosphatidylcholine synthesis in type II pneumocytes is stimulated by inclusion of phosphatidylglycerol and other phospholipids in the culture medium (Gilfillan, A.M., Chu, A.J. and Rooney, S.A. (1984) Biochim. Biophys. Acta 794, 269-273). We have now examined the effect of phosphatidylglycerol in the medium on enzymes of de novo phosphatidylcholine synthesis in adult rat type II cells. Activities of choline kinase, cholinephosphate cytidylyltransferase and cholinephosphotransferase in homogenates of whole lung and type II cells were generally similar. Phosphatidate phosphatase activity in type II cells, however, was only 16% that in whole lung. Addition of phosphatidylglycerol (10 microM) to the culture medium had no effect on choline kinase, cholinephosphotransferase or phosphatidate phosphatase activities in type II cells but it increased the activity of cholinephosphate cytidylyltransferase by 56%. Since it is known that cholinephosphate cytidylyltransferase is stimulated in vitro by addition of phospholipids to the assay mixture, we also measured its activity in the presence of sufficient phosphatidylglycerol (1.1 mM) to maximally stimulate in vitro. Even under these conditions cholinephosphate cytidylyltransferase activity in type II cells cultured in the presence of phosphatidylglycerol was 32% greater than in control cells. These data show that the stimulatory effect of phospholipid in the culture medium on phosphatidylcholine synthesis in type II cells is mediated by increased cholinephosphate cytidylyltransferase activity. The mechanism of increased cytidylyltransferase activity remains to be elucidated but it is not due to direct in vitro activation by the phospholipid.

Arachidonic acid metabolites stimulate phosphatidylcholine secretion in primary cultures of type II pneumocytes

There is evidence from whole animal and intact lung studies that prostaglandins are involved in the regulation of surfactant secretion. To explore this further we examined the effect of arachidonic acid on secretion of phosphatidylcholine in primary cultures of adult rat type II pneumocytes. Arachidonic acid stimulated phosphatidylcholine secretion and this effect was dependent on concentration in the range 1-8 microM. Arachidonic acid (8 microM) stimulated secretion by 79% from a basal rate of 1.17% total cellular phosphatidylcholine secreted in 90 min to 2.09%. We examined the effects of inhibitors of arachidonic acid metabolism on the stimulatory effect. Nordihydroguairaretic acid (0.1 microM), a lipoxygenase inhibitor, reduced the stimulatory effect by 64%. The same concentration of cyclooxygenase inhibitors had no effect. We conclude that arachidonic acid metabolites stimulate surfactant secretion in type II cells. Whether this effect is mediated by leukotrienes or other products remains to be established.

Stimulation of phosphatidylcholine synthesis by exogenous phosphatidylglycerol in primary cultures of type II pneumocytes

We examined the effect of exogenous phospholipids on phosphatidylcholine synthesis in primary cultures of adult rat type II pneumocytes. Incubation of the cells with 10 microM phosphatidylglycerol for 2 h stimulated the rate of [3H]choline and [3H]glycerol incorporation into phosphatidylcholine by 72% and 50%, respectively. The effect appeared to be specific for phosphatidylcholine synthesis and was largely on the unsaturated species. Synthesis of disaturated phosphatidylcholine was little stimulated. The stimulatory effect of the lipid is unlikely to be a consequence of increased substrate, since it was not mimicked by glycerol, glycerol 3-phosphate or palmitic acid. Neither does it appear to be due to increased cell growth, since rates of protein and DNA synthesis were not increased. The relevance of these findings to surfactant turnover and reutilization warrants investigation.

Single plate separation of lung phospholipids including disaturated phosphatidylcholine

We have developed an improved thin-layer chromatographic method for separation of lung phospholipids. Individual phospholipids are completely separated in the first dimension. All phospholipids, except phosphatidylcholine, are then removed. The phosphatidylcholine-containing area is reacted with osmium tetroxide and saturated phosphatidylcholine species are separated from the unsaturated oxidation products by subsequent chromatography in the second dimension. This method should prove useful in studies on lung surfactant metabolism and secretion, in prediction of fetal lung maturity by analysis of amniotic fluid phospholipids, and in studies on surfactant obtained by bronchoscopy in adult lung diseases.

The pharmacological modulation of [3H]-disaturated phosphatidylcholine overflow from perifused lung slices of adult rats: a new method for the study of lung surfactant secretion

Lung slices from adult rats incubated in [methyl-3H]-choline chloride formed [3H]-disaturated phosphatidylcholine ( [3H]-DSPC) which was used as an index of lung surfactant. The slices were perifused after 3 h incubation in [methyl-3H]-choline chloride and the overflow of [3H]-DSPC, as a rate coefficient, was used as a measure of surfactant secretion. The basal overflow of [3H]-DSPC rapidly declined over the first 30 min of perifusion and then declined slowly. Salbutamol induced a prolonged, and sometimes delayed, increase in [3H]-DSPC overflow, which was reduced by (+/-)-propranolol. Potassium chloride produced an immediate, and usually transient, increase in [3H]-DSPC overflow which was not modified by atropine or (+/-)-propranolol. Adenosine 5'-triphosphate, but not phenylephrine, also increased [3H]-DSPC overflow. This method can measure the magnitude and time-course of lung surfactant secretion induced by drugs.