Inhibition of calcium-independent phospholipase A impairs agonist-induced calcium entry in keratinocytes

Background In many cells, depletion of intracellular calcium (Ca2+) reservoirs triggers Ca2+ entry through store-operated Ca2+ channels in the plasma membrane. However, the mechanisms of agonist-induced calcium entry (ACE) in keratinocytes are not fully understood. Objectives This study was designed to determine if pharmacological inhibition of calcium-independent phospholipase A (iPLA2) impairs ACE in normal human epidermal keratinocytes. Methods Confocal laser scanning microscopy was used to monitor the dynamics of Ca2+ signalling in keratinocytes loaded with the calcium-sensitive dye Fluo-4. Cells were stimulated with extracellular nucleotides [adenosine triphosphate (ATP) or uridine triphosphate (UTP)] or with lysophosphatidic acid (LPA), a bioactive lipid that regulates keratinocyte proliferation and differentiation. Results Both ATP and UTP induced Ca2+ release in primary human keratinocytes. This was not followed by robust Ca2+ influx when the experiments were performed in low Ca2+ (70 μmol L−1) medium. Upon elevation of extracellular Ca2+ to 1·2 mmol L−1, however, a biphasic response consisting of an initial Ca2+ peak followed by an elevated plateau was observed. The plateau phase was inhibited when cells were treated with bromoenol lactone, a specific pharmacological inhibitor of iPLA2. These findings indicate that iPLA2 activity is required for ACE in keratinocytes. LPA also evoked Ca2+ release in keratinocytes but failed to induce sustained Ca2+ entry even when extracellular Ca2+ was elevated to 1·2 mmol L−1. Conclusion Our results demonstrate for the first time an important role for iPLA2 in regulating ACE in primary human keratinocytes.

Calcium is a ubiquitous second messenger that regulates numerous cellular processes such as gene transcription, cell proliferation, exocytosis and contraction. 1 Free cytosolic calcium ([Ca 2+ ] i ) levels are tightly controlled by a complex network of receptors, channels and pumps located in the plasma membrane (PM) and on intracellular organelles such as the endoplasmic reticulum (ER), mitochondria and the Golgi apparatus. Stimulation of G protein-coupled receptors (GPCRs), tyrosine kinase receptors and nonreceptor tyrosine kinases activate phospholipase C (PLC) which in turn hydrolyses phosphatidylinositol 4,5-bisphosphate to diacylglycerol and inositol 1,4,5-trisphosphate (IP 3 ). 2 Binding of IP 3 to its receptors (IP 3 R) on the ER triggers the release of Ca 2+ from the ER lumen leading to store depletion. 1 In many cells, this initial release is followed by a sustained influx of Ca 2+ across the PM, a phenomenon known as store-operated calcium entry (SOCE), which is the dominant form of Ca 2+ entry in nonexcitable cells. 3 One model of SOCE involves a diffusible messenger or 'calcium influx factor' (CIF) that is released from the ER upon store depletion. 4 Although the identity of CIF is unknown, it appears to be a soluble factor of 600 Da that activates calcium-independent phospholipase A (iPLA 2 ) by displacement of inhibitory calmodulin (CaM) from iPLA 2 . 5 This leads to the production of lysophospholipids (lysoPLs) and free fatty acid. The lysoPLs, such as lysophosphatidylcholine, then activate SOCE at the PM by an uncharacterized process. Thus iPLA 2 activity appears to be required for SOCE.
A second model for SOCE has emerged recently, involving STIM1 and Orai1. STIM1, a Ca 2+ -sensing protein localized predominantly to the ER contains a low-affinity Ca 2+ -binding EF hand which resides in the ER lumen when the stores are full. Depletion of the stores by IP 3 -mediated Ca 2+ release, or by inhibition of the sarco-endoplasmic reticulum Ca 2+ -ATPase (SERCA) pump with thapsigargin (TG) causes Ca 2+ to dissociate from STIM1 inducing the re-organization of STIM1 into discrete puncta. 6,7 These complexes appear to associate with, and activate, the transmembrane protein Orai1, which appears to be the pore through which SOCE occurs. [8][9][10][11][12] Interestingly, STIM1 has also been reported to activate TRPC1, 13,14 a member of the transient receptor potential (TRP) family of proteins which have been implicated in cation entry. 15 Little is known about the mechanisms of Ca 2+ entry in keratinocytes. A requirement for PLCc for SOCE has been demonstrated, 16 along with the formation of a ternary complex composed of PLCc, TRPC1 and IP 3 R. However, the putative role of iPLA 2 in Ca 2+ entry in keratinocytes has not been examined. In the present study therefore, we have investigated the role of iPLA 2 in agonist-induced Ca 2+ entry (ACE) in normal human epidermal keratinocytes (NHEKs). We have performed our studies using physiological agonists such as adenosine triphosphate (ATP) and uridine triphosphate (UTP), which have been reported to promote NHEK proliferation in vitro 17,18 and lysophosphatidic acid (LPA, 1-acyl-sn-glycerol-3-phosphate), which can promote proliferation or differentiation depending on cell density. 19 These agonists are released by platelets recruited to the epidermis following injury, 17,20 indicating a role in epidermal homeostasis. We found that the extracellular nucleotides but not LPA evoked sustained ACE in NHEK and that this was mediated at least in part by iPLA 2 .

Reagents
Fluo-4-AM was purchased from Molecular Probes (Eugene, OR, U.S.A.), bromoenol lactone (BEL) from Sigma (Poole, Dorset, U.K.) and the iPLA 2 antibody from Santa Cruz (Santa Cruz, CA, U.S.A.). The iPLA 2 antibody recognizes iPLA2b but not iPLA 2 c according to the manufacturer. All other reagents were obtained from Sigma unless stated otherwise.
Cell culture NHEK were prepared from redundant foreskin with the approval of the Newcastle and North Tyneside local ethical committee. The cells were cultured in supplemented MCDB 153 culture medium as previously described. 21 For imaging, cells were seeded at passage 1 or 2 in 20 lL suspensions containing 10 000-20 000 cells, with the medium increased to 1 mL about 45 min to 1 h after seeding.

Calcium imaging
Subconfluent monolayers of cells, seeded in Willco glass-bottomed microwell dishes (Intracel, Royston, U.K.) 1 day before experiments, were loaded with 3 lmol L )1 of Fluo-4 acetoxymethyl (AM) ester (Molecular Probes) for 45 min at 37°C. Dye loading and all subsequent steps were performed with MCDB153 medium (Sigma) containing 70 lmol L )1 Ca 2+ unless indicated otherwise. To minimize uptake of the dye into organelles, 200 lmol L )1 of the anion transport inhibitor sulphinpyrazone was dissolved in dimethyl sulphoxide (DMSO) and included in the medium during loading and de-esterification. After loading, the cells were washed in Ca 2+ and Mg 2+ free phosphate-buffered saline (PBS) and incubated in MCDB 153 medium with 70 lmol L )1 Ca 2+ for 1 h at 37°C to allow de-esterification of the intracellular dye. Where indicated, Ca 2+ in the medium was raised to 1AE2 mmol L )1 at the start of the de-esterification phase. The iPLA 2 inhibitor BEL (10 or 20 lmol L )1 ) or vehicle (0AE1% DMSO) was added for the last 30 min of de-esterification.
The cells were maintained at 37°C during image acquisition with a heated stage. Changes in [Ca 2+ ] i were detected with a Leica TCS SP2 confocal laser scanning microscope equipped with an argon laser (Leica, Milton Keynes, U.K.). Fluorescence excitation was performed with the 488 nm line of the laser. Fluorescence emission was collected through a 500-550 nm window of the detector. Images were captured with a 63X Plan Apo objective (NA1.32) at 4-s intervals as 12-bit frames of 512 · 512 pixels. The perimeter of each cell was outlined to define the region of interest whose mean fluorescence intensity in each frame was determined by Leica confocal software. The changes in [Ca 2+ ] i were expressed as the ratio of the temporal fluorescence to the initial fluorescence (Ft ⁄F0).

Western blotting
Primary keratinocyte lysates were separated on a 10% Bis-Tris gel (Invitrogen, Paisley, U.K.), transferred to a nitrocellulose membrane and incubated overnight with 10 lg mL )1 of iPLA 2 antibody (goat polyclonal, Santa Cruz). After extensive washing the membrane was probed with a biotinylated secondary antibody for 2-3 h, processed for chemiluminescence using ABC reagents (Vector Laboratories, Burlingame, CA, U.S.A.) and ECL Advance TM (GE Healthcare, Little Chalfont, U.K.), then visualized on a phosphoimager.

Statistical analysis
Results of the Ca 2+ imaging experiments are presented as means (± SEM) which were determined in GraphPad Prism (GraphPad Software, San Diego, CA, U.S.A.) or Microsoft Excel. Statistical significance was calculated using the unpaired two-tailed Student's t-test. A P-value < 0AE05 was considered significant.

Agonist-induced Ca 2+ entry in primary keratinocytes
Several studies have shown that ATP and UTP evoke [Ca 2+ ] i elevation in primary keratinocytes, 17,22 but the extent to which this promotes a sustained increase in [Ca 2+ ] i (indicative of Ca 2+ entry) was not clear. When preconfluent monolayers of NHEKs were stimulated with UTP, a single [Ca 2+ ] i transient was observed in the majority of responsive cells (Fig. 1a,b). The rise in [Ca 2+ ] i began immediately following the addition of the agonist, and reached a peak about 16 s thereafter. The entire [Ca 2+ ] i signal lasted just over 2 min before returning to baseline levels. In 6% (6 of 98 responsive cells from four experiments), a second transient of smaller amplitude was observed at later time points (Fig. 1d). Stimulation of the cells with ATP also induced a single 2 min [Ca 2+ ] i transient in most of the responsive cells (Fig. 1c), with a second [Ca 2+ ] i peak in 23% of the cell population (16 of 70 responsive cells pooled from five experiments; Fig. 1e).
In these experiments, which were performed in medium containing 70 lmol L )1 Ca 2+ we consistently failed to observe sustained [Ca 2+ ] i elevation. We therefore repeated the assays after switching to medium containing 1AE2 mmol L )1 [Ca 2+ ] o for~1 h. Over several hours 1 mmol L )1 extracellular calcium itself stimulates a significant rise in [Ca 2+ ] i and promotes differentiation. 23 Both UTP and ATP induced a biphasic response under these conditions, with a defined initial peak followed by an elevated plateau indicative of Ca 2+ entry (Fig. 2a,b). Taken together, these results indicate that the extracellular calcium ([Ca 2+ ] o ) level needs to be in the millimolar range before the electrochemical gradient is sufficiently high to promote robust ACE in keratinocytes following stimulation with exogenous nucleotides.

Inhibition of iPLA 2 impairs agonist-induced calcium entry in primary keratinocytes
A recent study by Bolotina and co-workers showed that pharmacological inhibition or RNAi knockdown of iPLA 2 impaired TG-induced entry in mouse smooth muscle cells. 4 Although the TG-induced SOCE has historically been considered as a mechanistic parallel of ACE, differences are beginning to emerge. 24,25 Thus, we asked if iPLA 2 activity was required for ACE in NHEKs. For these studies, we used the suicide substrate, BEL, to inhibit iPLA 2 activity. BEL is a specific inhibitor of iPLA 2 , with a 1000-fold selectivity for iPLA 2 over cytosolic PLA 2 . 26 Paired assays were performed on keratinocytes from the same donors. As shown in Figure 3, stimulation with UTP in the presence of 1AE2 mmol L )1 [Ca 2+ ] o led to a sustained [Ca 2+ ] i plateau in control cells exposed to vehicle (DMSO), consistent with the results shown in Figure 2a. In contrast, the elevated [Ca 2+ ] i phase was impaired in cells treated with BEL (Fig. 3a), returning to baseline by 300 s in contrast to control cells. It is important to note that the control cells in Figure 3a were treated with dimethylsulphoxide which might explain why their responses differed somewhat to those in Figure 2a. Nevertheless, the data suggest that BEL treatment impairs Ca 2+ entry in NHEK. Furthermore, similar results were obtained on the HaCaT keratinocyte cell line. 39 The expression of iPLA 2 in NHEK was assessed by Western blotting (Fig. 3c). An 85-kDa band corresponding to the expected size of iPLA 2 was specifically detected in NHEK lysates probed with an antibody against iPLA 2 confirming that iPLA 2 is expressed in these cells.  (Fig. 4b). This result also controls against the possibility that raised (1AE2 mmol L )1 ) [Ca 2+ ] o was exerting a nonspecific effect enhancing effect on ACE in keratinocytes. In addition, the duration of the transient was not significantly altered by the inclusion of 10 mmol L )1 EGTA in the medium, indicating that sustained Ca 2+ entry did not occur (data not shown). Thus LPA does not appear to induce significant Ca 2+ influx in keratinocytes. This observation is consistent with the findings of others on T cells and fibroblasts. 27,28 The inabil- ity of LPA to evoke ACE in our experiments was not due to submaximal stimulation because dose-response curves revealed that LPA-mediated Ca 2+ release peaked at about 1 lmol L )1 (Fig. 4c). Furthermore, LPA was a more potent inducer of Ca 2+ release than UTP, with an EC 50 of 40 nmol L )1 compared with 265 nmol L )1 for UTP.   The findings presented in this study indicate for the first time that iPLA 2 activity is required for ACE in human keratinocytes. Evidence from the Gross laboratory indicates that iPLA 2 exists as a ternary complex with Ca 2+ ⁄CaM. 29 Displacement of CaM from this complex leads to activation of iPLA 2 which in turn cleaves phospholipids to generate fatty acids such as arachidonic acid and lysoPLs. 5 Bolotina and colleagues recently demonstrated a role for lysoPLs iPLA 2 in TG-induced SOCE in rodent cells. 4 Our results extend their observations by showing that ACE in human keratinocytes also appears to be mediated at least in part by iPLA 2 . How does iPLA 2 mediate ACE? In the CIF-iPLA 2 model of SOCE, the lysoPLs generated by iPLA 2 are postulated to activate store-operated channels directly. 5 However, a ternary complex composed of PLCc, TRPC1 and IP 3 R has been detected in NHEKs. 16 Although the formation of the complex did not appear to be dependent on store depletion, knockdown of PLCc or TRPC1 (and TRPC4) suggested a role in Ca 2+ entry. Could iPLA 2 participate in the formation or localization of this complex? Recent evidence indicates that PLCc interacts with the TRPC3 to form a functional intermolecular pleckstrin homology (PH) domain that binds lipids 30 and that this enhances surface expression of TRPC3 in HEK293 cells. One possibility then is that lysoPLs generated by iPLA 2 enhance the localization of TRP proteins to the cell surface of keratinocytes, and that inhibition of iPLA 2 impairs this process. However, we cannot exclude a role for arachidonic acid generated by iPLA 2 activity, given that it has also been implicated in ACE 31 although this seems to occur only at submaximal agonist concentrations. Further investigations will be required to delineate the respective functions of lysoPLs and arachidonic acid in Ca 2+ entry in keratinocytes.

LPA-induced [Ca
Although When extracellular Ca 2+ was raised to 1AE2 mmol L )1 , the [Ca 2+ ] i peak generated by the application of exogenous nucleotides was followed by an elevated plateau. Importantly, addition of Mn 2+ to the medium after UTP stimulation led to quenching of the Fluo-4 signal (data not shown), indicating that under these conditions UTP activated Ca 2+ entry. In contrast to extracellular nucleotides, LPA stimulation did not produce an elevated Ca 2+ plateau in keratinocytes even in the presence of millimolar levels of [Ca 2+ ] o (Fig. 4b). The reason LPA fails to activate ACE in these cells is unclear but studies on Jurkat T cells and a lung fibroblast cell line also found that LPA did not induce Ca 2+ entry in these cells. 27,28 Thus it appears that despite its ability to mobilize Ca 2+ from internal stores, LPA does not promote robust Ca 2+ entry in several distinct cell types. The contrast between UTP and LPA induced [Ca 2+ ] i mobilization may be due to differential coupling of LPA receptor activation to STIM1 or Orai1 ⁄CRACM1, two newly discovered mediators of SOCE. [35][36][37][38] Indeed we have observed that although both UTP and LPA evoke translocation of STIM1 to the PM, the duration of STIM1 localization to the PM is significantly shorter in LPA treated cells. 39 In conclusion, the work presented here demonstrates that extracellular nucleotides trigger Ca 2+ release in cultured human keratinocytes, and when external Ca 2+ is in the millimolar range, Ca 2+ release follows sustained ACE. Stimulation with LPA also evoked Ca 2+ release, but without inducing robust Ca 2+ influx. Pharmacological inhibition of iPLA 2 impaired ACE, highlighting the importance of iPLA 2 in ACE in keratinocytes.