Selective inhibition of thrombin receptor-mediated Ca2+ entry by protein kinase C beta.

Thrombin initiates many physiological processes in platelets and other megakaryocyte-lineage cells by interacting with surface receptors and generating rises in cytoplasmic Ca2+; these rises result from both Ca2+ release from intracellular stores and receptor-mediated Ca2+ entry. Regulators that limit Ca2+ entry after its initiation by thrombin have not been identified. In this study, prevention of expression of a single protein kinase C isoenzyme (PKCβ) by antisense cDNA overexpressed in HEL cells, a human megakaryoblastic cell line that expresses thrombin receptors, promotes thrombin receptor-mediated Ca2+ entry without altering thrombin-induced intracellular release of Ca2+. The cytoplasmic Ca2+ rise initiated by endoperoxide analogs was not affected by inhibiting PKCβ. Overexpression of a cDNA encoding wild-type PKCβ mutated to prevent recognition by the antisense cDNA abolished the enhancement of Ca2+ influx following thrombin. Thus, PKCβ appears to be a specific negative regulator of thrombin receptor-mediated Ca2+ entry.


Thrombin initiates many physiological processes in platelets and other megakaryocyte-lineage cells by interacting with surface receptors and generating rises in cytoplasmic Ca
; these rises result from both Ca 2؉ release from intracellular stores and receptor-mediated Ca 2؉ entry. Regulators that limit Ca 2؉ entry after its initiation by thrombin have not been identified. In this study, prevention of expression of a single protein kinase C isoenzyme (PKC␤) by antisense cDNA overexpressed in HEL cells, a human megakaryoblastic cell line that expresses thrombin receptors, promotes thrombin receptor-mediated Ca 2؉ entry without altering thrombin-induced intracellular release of Ca 2؉ . The cytoplasmic Ca 2؉ rise initiated by endoperoxide analogs was not affected by inhibiting PKC␤. Overexpression of a cDNA encoding wild-type PKC␤ mutated to prevent recognition by the antisense cDNA abolished the enhancement of Ca 2؉ influx following thrombin. Thus, PKC␤ appears to be a specific negative regulator of thrombin receptor-mediated Ca 2؉ entry.
The protease thrombin is generated at sites of vascular injury and is a central mediator of hemostasis, thrombosis, inflammation, and vascular proliferation (1). Thrombin stimulates several cell types, including platelets and other cells of megakaryocytic lineage, monocytes, endothelium, and vascular smooth muscle, by triggering surface receptors to generate intracellular messengers (2,3). The addition of thrombin to platelets initiates a rise in cytoplasmic Ca 2ϩ ([Ca 2ϩ ] in ) 1 that results from release of Ca 2ϩ from intracellular stores (4) followed by Ca 2ϩ entry via a receptor-regulated cation channel (5,6). Several messengers, including an unidentified Ca 2ϩ influx factor (8) and inositol 1,3,4,5-phosphate (9), have been postulated to initiate receptor-mediated Ca 2ϩ influx (10,11); how-ever, few regulators that limit receptor-mediated Ca 2ϩ entry, which is necessary to prevent excessive [Ca 2ϩ ] in , have been characterized. Activation of the intracellular phospholipid-dependent protein kinase C (PKC) with phorbol esters prevents both entry and intracellular mobilization of Ca 2ϩ induced by thrombin (12,13). PKC can also be activated by generation of lipid regulators (14) that are mobilized following thrombin stimulation (2); some isoenzymes of PKC (e.g. PKC␤) are regulated by a rise in [Ca 2ϩ ] in (15). Since limiting or negative mediators of Ca 2ϩ influx might be reasonably considered to be regulated by a process requiring Ca 2ϩ (16), we asked whether selective inhibition of a Ca 2ϩ -dependent PKC isoenzyme would modify the thrombin receptor-mediated Ca 2ϩ influx. For these experiments, we utilized human erythroleukemic (HEL) cells (17), a megakaryoblastic cell line that has functional thrombin receptors and shares with platelets many components of the thrombin signaling mechanism (12,18). HEL cells offer an additional advantage for the present studies, in that they express only one (PKC␤) of the Ca 2ϩ -responsive PKC isoenzymes (12,19) and therefore required only a single intervention to eliminate Ca 2ϩ -dependent PKC activity from the cells. As there are few chemical PKC inhibitors that are specific for individual isoenzymes, a strategy based on antisense DNA was utilized to reduce selectively PKC␤ in HEL cells.

MATERIALS AND METHODS
Construction, Expression, and Detection of anti-PKC␤-To achieve selective inhibition of PKC␤, a segment of cDNA (55 base pairs) specific to PKC␤ (5ЈTCCGGCTCCCCGCGCGCAAGATGGCTGACCCGGCT-GCGGGGCCGCCGCCGAGCGA3Ј) from Ϫ20 to 35 (initiating ATG (underlined) numbered 1-3) was cloned into pcDNA1/Neo (Invitrogen) in an antisense orientation. The nucleotide sequence of the first 20 and the last 35 base pairs of the cDNA corresponded to that of the 5Ј-end of the untranslated and the beginning of the translated regions (respectively) of human PKC␤ cDNA.
Low passage HEL cells (a generous gift of Thalia Papayannopoulou, University of Washington, Seattle) were either transfected with the antisense PKC␤ construct or with the vector only (control cells) by electroporation. Cells that stably expressed anti-PKC␤ cDNA (anti-␤ cells) were selected by limiting dilution and on the basis of cell survival in the presence of Geneticin (1.2 mg/ml). For Northern blot analysis, total RNA was prepared from the cells using the method described previously (12). The cDNA probes of PKC␤, PKC␦, and PKC are the same as described previously (19). The cDNA probes for ␣ IIb , ␤-thromboglobulin, and the thrombin receptor were generous gifts from Drs. Peter Newman (Blood Research Institute of Southeastern Wisconsin), Mortimer Poncz, and Lawrence Brass (University of Pennsylvania), respectively; the anti-thrombin receptor antibody was kindly provided by Dr. Brass.
Construction and Expression of mut-PKC␤-To create a mutated PKC␤ that was not inhibited by the antisense PKC␤ construct, the 5Ј-end of the untranslated and the beginning of the translated region of the cDNA encoding rat PKC␤ (a generous gift of Dr. I. B. Weinstein, Columbia University) was mutated to reduce the complementarity to anti-␤ cDNA. The mutant cDNA was cloned into pREP4 vector (Invitrogen) carrying the hygromycin resistance gene. Anti-␤ cells were transfected with the mut-PKC␤ construct by electroporation. Stable transfectants were selected by limiting dilution and on the basis of cell survival in the presence of 100 g/ml hygromycin.
[Ca 2ϩ ] in Measurements-Cytoplasmic ionized Ca 2ϩ ([Ca 2ϩ ] in ) was measured in Fura-2 loaded HEL cells as described previously (12). Briefly, HEL cells were washed and resuspended in HEPES-Tyrode's buffer (12). Fura-2/AM was added in a final concentration of 2 M to the cells, which were incubated for 30 min at 37°C. Fluorescence measurements were obtained using a dual excitation wavelength spectrofluorometer (SPEX Fluorolog-2, Edison, NJ). Fura-2 signals were calibrated as described previously (20). All measurements were performed * This work was supported by Grants HL38820 and HL47032 from the National Institutes of Health (to J. A. W.) and a fellowship (to Y. X.) from the Massachusetts Affiliate of the American Heart Association. The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked "advertisement" in accordance with 18  on cells suspended in HEPES-Tyrode's buffer containing 1 mM Ca 2ϩ ; in some experiments, NiCl 2 , EGTA, or MnCl 2 was added just before stimulation with thrombin.

RESULTS AND DISCUSSION
Expression of both RNA and protein of PKC␤ was significantly reduced in anti-␤ cells compared with that in either wild-type HEL cells or the control cells (Fig. 1, a and b). In addition to PKC␤, the predominant isoenzymes expressed by HEL cells are PKC␦ and PKC, which are also expressed in platelets (12,19); neither PKC␦ nor PKC was inhibited by anti-PKC␤ cDNA (Fig. 1a). Inhibition of PKC␤ provoked no obvious alteration of differentiation, as shown by the similar expression of integrin ␣ IIb , ␤-thromboglobulin, and thrombin receptor (Fig. 1c).
Because of the effect that overall PKC activation exerts on [Ca 2ϩ ] homeostasis, we asked whether this sharp reduction in PKC␤ would alter [Ca 2ϩ ] in following thrombin. The thrombininduced [Ca 2ϩ ] in in clonal populations of anti-␤ HEL cells loaded with the Ca 2ϩ -sensitive fluorophore Fura-2 was significantly enhanced when compared with that of the control cells (Fig. 2, a and b). This result suggests that thrombin-induced elevation of [Ca 2ϩ ] in in HEL cells is normally inhibited by PKC␤; this effect appeared to be specific for thrombin, because [Ca 2ϩ ] in induced by the endoperoxide analog U46619, which activates the thromboxane A 2 receptor, is not affected in anti-␤ cells (Fig. 2, c and d).

FIG. 1. Inhibition of PKC␤ expression in megakaryocyte-lineage cells.
a, Northern transfer analysis of RNAs extracted from wild-type (HEL), control, and anti-␤ HEL cells. The blot was hybridized with cDNA probes complementary to PKC␤, PKC␦, PKC, or glyceraldehyde-phosphate dehydrogenase (GAP-DH) as a control for RNA loading, as indicated. Approximate mRNA sizes are: 2.6 kb for PKC␤; 2.2 kb for PKC␦; and 2.7 kb for PKC. b, immunoblot of total protein extracted from wild-type, control, and anti-␤ HEL cells with a PKC␤-specific monoclonal antibody (Seikagaku America, Inc.). c, Northern transfer analysis of RNA extracted from wild-type (HEL), control, and anti-␤ cells. The blot was hybridized with cDNA probes encoding portions of ␣ IIb , ␤-thromboglobulin, and the thrombin receptor, as indicated. Approximate mRNA sizes are: 3.5 kb for the thrombin receptor; 2.5 kb for ␣ IIb ; and 1.6 kb for ␤-thromboglobulin (␤TG).
anti-thrombin receptor antibody (data not shown) or by expression of mRNA encoding the thrombin receptor (Fig. 1c).
To determine whether enhanced [Ca 2ϩ ] in resulted specifically from the reduction of PKC␤ expression by the antisense construct, we restored the expression of PKC␤ by stably transfecting cDNA encoding the full-length PKC␤ into clonal populations of anti-␤ cells. To prevent inhibition of expression of the transfected PKC␤ by the constitutively expressed anti-␤ cDNA, the degeneracy of the genetic code was exploited to generate a mutant PKC␤ cDNA (mut-PKC␤) that had minimal complementarity with the anti-␤ cDNA but still encoded the same amino acid sequence as native PKC␤ (Fig. 3a). Expression of the mut-PKC␤ was not inhibited by the anti-␤ construct, as verified by cotransfection of the two constructs in COS7 cells, which do not normally express PKC␤, followed by Northern transfer analysis (data not shown). Transfection of mut-PKC␤ into the anti-␤ cells restored the expression of PKC␤, as assessed by immunoblotting (Fig. 3b), and abolished the enhancement of thrombin-induced [Ca 2ϩ ] in in the anti-␤ cells (Fig. 3, c  and d). Thus, the reversal of the changes induced by antisense restoration of PKC␤ in the same clone of cells strengthens the argument that PKC␤ specifically inhibits the thrombin-induced increase in [Ca 2ϩ ] in .
This inhibition of the thrombin-induced increase in [Ca 2ϩ ] in by PKC␤ might result from an effect on Ca 2ϩ entry, on Ca 2ϩ release from intracellular stores, or on both. To distinguish among these possibilities, the contribution of Ca 2ϩ entry from extracellular medium to [Ca 2ϩ ] in was eliminated by either briefly chelating extracellular Ca 2ϩ with EGTA or adding Ni 2ϩ , which blocks the Ca 2ϩ entry via receptor-operated cation channels (5,7). Following these interventions, [Ca 2ϩ ] in was not significantly different between anti-␤ and control cells (Fig. 4,  a and b). Furthermore, the divalent cation entry, assessed by measuring the Mn 2ϩ quench of intracellular Fura-2 fluorescence (6) after addition of thrombin, was less in the control cells than in the anti-␤ cells (Fig. 4c). Thus, PKC␤ reduces the magnitude of thrombin-induced Ca 2ϩ entry but has little effect on release of Ca 2ϩ from intracellular stores. These results also demonstrate that some aspects of thrombin receptor function were not affected by inhibition of PKC␤; additionally, PKCmediated inhibition of thrombin-induced mobilization of Ca 2ϩ from intracellular storage sites (12,13) does not require the presence of known Ca 2ϩ -regulated PKC isoenzymes in HEL cells.
After addition of thrombin (but not ADP) to platelets (6, 7), a measurable delay precedes Ca 2ϩ entry, suggesting that the Ca 2ϩ channel is not directly linked to the receptor but instead is activated by intracellular mediators, generated perhaps by depletion of intracellular Ca 2ϩ stores (10,11) or by phospholipid hydrolysis (21). Thus, PKC␤ might interfere with a mediator that initiates or potentiates Ca 2ϩ influx; such a mediator would presumably be generated by thrombin but not endoperoxide analogs such as U46619. An alternative possibility is that  3). b, bar graphs of the mean Ϯ S.E. of the differences between peak thrombin-induced and basal levels of [Ca 2ϩ ] in in control and anti-␤ cells suspended in buffer to which 2 mM NiCl 2 has been added (n ϭ 3). c, representative tracing of Fura-2 fluorescence changes (360 nM excitation wavelength) of control and anti-␤ cells suspended in medium containing Mn 2ϩ (0.1 mM) shortly after stimulation with thrombin (0.5 unit/ml). PKC␤, once activated, might phosphorylate a receptor-mediated Ca 2ϩ channel specifically associated with the thrombin receptor. Either model has implications for the specificity of agonist effect. Thus, although other mechanisms for limiting Ca 2ϩ entry may exist for other agonists, this reduction in thrombin receptor-mediated Ca 2ϩ influx by PKC␤, a Ca 2ϩregulated PKC, represents a novel and selective cross-regulatory mechanism that could prevent excessive accumulation of cytoplasmic Ca 2ϩ following thrombin stimulation.