- Email: [email protected]
The PPAR-γ activator, Rosiglitazone, inhibits actin polymerisation in monocytes: Involvement of Akt and intracellular calcium
BBRC Biochemical and Biophysical Research Communications 333 (2005) 455–462 www.elsevier.com/locate/ybbrc
The PPAR-c activator, Rosiglitazone, inhibits actin polymerisation in monocytes: Involvement of Akt and intracellular calcium Neenu Singh a, Richard Webb a, Rachel Adams a, Shelley-Ann Evans a, Ameena Al-Mosawi a, Marc Evans b, Aled W. Roberts b, Andrew W. Thomas a,* a
School of Applied Science, University of Wales Institute Cardiff, Western Avenue, Cardiff CF5 2YB, UK b University Hospital of Wales, Heath Park, Cardiff CF5 2XN, UK Received 17 May 2005 Available online 2 June 2005
Abstract Monocyte hyperactivation as seen in diabetes results in increased cytoskeletal rigidity and reduced cell deformability leading to microchannel occlusions and microvascular complications. The thiazolidinediones (TZDs) are PPAR-c agonists that have been reported to exert beneficial non-metabolic effects on the vasculature. This study demonstrates that the TZD, Rosiglitazone, significantly reduces f-MLP-induced actin polymerisation in human monocytic cells (p < 0.05). Two of the key signalling processes known to be involved in the regulation of cytoskeletal remodelling were investigated: PI3K-dependent Akt phosphorylation and intracellular calcium concentration [Ca2+]i. The PI3K inhibitor, Wortmannin, ameliorated f-MLP-induced actin polymerisation (p < 0.05), while the Ca2+ sequestration inhibitor, thapsigargin, induced actin depolymerisation (p < 0.05), confirming the involvement of both processes in cytoskeletal remodelling. Rosiglitazone significantly reduced f-MLP activation of Akt (p < 0.05), and significantly increased [Ca2+]i in both resting and f-MLP-stimulated cells (p < 0.05). Therefore, Rosiglitazone interacts with signalling events downstream of occupancy of the f-MLP receptor, to modulate cytoskeletal remodelling in a PPAR-c-independent manner. To our knowledge, these results are the first to present evidence that a PPAR-c agonist can modulate actin remodelling in monocytes, and may therefore be protective against microvascular damage in diabetes. 2005 Elsevier Inc. All rights reserved. Keywords: Rosiglitazone; Actin polymerisation; Calcium; Akt; Diabetes; Microvascular; PPAR-c-independent
Monocytic activation is a normal physiological process involving changes in cell morphology and motility and is involved in both inflammation and phagocytosis. However, it has been shown that in diabetes, monocytes can become hyperactivated, resulting in increased rigidity of cell membranes and reduced cell deformability. This can result in microchannel occlusions and contribute to microvascular complications such as diabetic retinopathy, nephropathy, and neuropathy [1,2]. Cell deformability is determined by the degree of cytoplasmic actin polymerisation. External signals (e.g. growth factors and chemoattractants) can induce poly*
Corresponding author. Fax: +442920416982. E-mail address: [email protected] (A.W. Thomas).
0006-291X/$ - see front matter 2005 Elsevier Inc. All rights reserved. doi:10.1016/j.bbrc.2005.05.127
merisation of globular G-actin into filamentous F-actin in order to bring about changes in cell morphology and motility. For example, actin polymerisation is involved in the pseudopod formation that has been shown to be an important physical parameter determining the activation of monocytes [3]. Pseudopods are stiffer than the main cell body and, consequently, circulation of hyperactivated monocytes may lead to a greater risk of entrapment of cells within capillaries. Thus, hyperactivation of monocytes may confer an increased risk for the progression of microvascular complications in diabetes. PPAR-c is a ligand-activated nuclear receptor that regulates the expression of diverse genes involved in glucose and lipid metabolism, cell differentiation,
456
N. Singh et al. / Biochemical and Biophysical Research Communications 333 (2005) 455–462
apoptosis, and inflammation [4]. The ligands of PPAR-c include oxidative metabolites of polyunsaturated fatty acids, prostaglandins such as 15d-PGJ2, and oral antihyperglycaemic agents of the thiazolidinedione (TZD) class such as Rosiglitazone. There is evidence that, apart from their insulin-sensitising activity, TZDs exert a plethora of effects on the process of atherothrombosis which may provide macrovascular benefits [5–7]. By contrast, there is a relative paucity of data evaluating the potential effects of TZDs on mechanisms involved in the development of microangiopathy. Some of these vascular effects seem to be mediated by a mechanism independent of PPAR-c-activated gene expression [8]. These so-called ‘‘non-genomic’’ effects of PPAR-c ligands have been shown to influence intracellular signals such as changes in cytosolic calcium ion concentrations ([Ca2+]i) or protein kinase (e.g. Akt) activity [9–11]. Both of these processes are known to be involved in the regulation of the actin cytoskeleton: chemoattractant-induced reorganisation of the actin cytoskeleton is dependent on phosphorylation of Akt [12], while Ca2+ entry into the cell and sustained increases in [Ca2+]i are involved in the Ca2+-dependent disassembly of the actin cytoskeleton mediated by gelsolin [13–15]. Therefore, abnormal Akt activity or [Ca2+]i dysregulation in the monocytes of individuals with diabetes could be responsible for inappropriate signalling, abnormal monocytic activation, capillary plugging, and consequently microangiopathy [16]. Thus, decreases in F-actin content may result in reduced activation and increased deformability of leukocytes, which may contribute to prevention of microvascular injury [17]. In the present study, we attempt to elucidate whether Rosiglitazone modulates monocytic actin polymerisation, and also how it affects the key intracellular signals involved in cytoskeletal remodelling (particularly phosphorylation of Akt and changes in [Ca2+]i). This study may therefore provide an insight into some of the mechanisms by which Rosiglitazone may potentially influence the pathogenesis of microangiopathy in diabetes.
Materials and methods Materials. All chemicals were purchased from Sigma–Aldrich (Poole, UK) unless stated otherwise. Cell permeabilisation reagents, protein assay reagents, FITC–phalloidin, and Fluo-3 AM were obtained from Harlan SERA-LAB (Loughborough, UK), Bio-Rad Laboratories (Herts, UK), Molecular Probes (Eugene, OR), and Calbiochem, EMD Biosciences (Darmstadt, Germany), respectively. Rosiglitazone was obtained from GlaxoSmithKline (Uxbridge, UK). Rabbit anti-phospho-Akt (Ser473), anti-total Akt, and anti-rabbit APconjugated secondary antibody were purchased from New England BioLabs (Hitchin, UK). Isolation of human peripheral mononuclear cells. For isolation of peripheral blood mononuclear cells, 40 ml of EDTA-anticoagulated blood was obtained from healthy patients. Ten millilitres of blood was
layered over 10 ml Ficoll–Hypaque and centrifuged at 450g for 35 min. The mononuclear cell suspension was carefully removed from the Ficoll–Hypaque interface, washed three times in PBS/EDTA, and resuspended in RPMI 1640 for use in all the experiments. Cell viability was >95% as measured by trypan blue staining. Flow cytometric analysis of the formation of F-actin. Actin polymerisation was determined by the method of Howard and Meyer [18]. Briefly, peripheral blood monocytic cells (1.0 · 105 cells/ml) were incubated at 37 C in PBS. After 60-min incubation with or without 20 lM Rosiglitazone and/or stimulation for 5 min with 100 nM fMLP, cells were fixed, permeabilised, and stained in a single step by addition of 0.1 ml of 37% phosphate-buffered formalin containing 1.65 · 10 6 M FITC–phalloidin and 100 lg lysophosphatidylcholine. In some cases, cells were also preincubated for 15 min with 100 nM Wortmannin or for 30 min with 1 lM thapsigargin. Cells were then incubated at 37 C for 60 min, washed, and harvested by centrifugation. The cell pellet was resuspended in 0.5 ml PBS before analysis on FACS. The samples were excited by an argon laser at 488 nm and emission was measured at 530 nm. Measurement of intracellular Ca2+ concentration. [Ca2+]i was measured by the method of Elsner et al. [19], with some minor modifications. Briefly, the fluorescence of the Ca2+ indicator Fluo-3 was used as an indicator of [Ca2+]i at defined time intervals after stimulation with 100 nM f-MLP/preincubation with 20 lM Rosiglitazone or 1 lM thapsigargin. Cell suspensions (7 · 106 ml in RPMI 1640 medium) were incubated with 3 lM Fluo-3 AM for 40 min at 37 C. Cells were washed twice and resuspended in RPMI 1640 at a final concentration of 1 · 106 ml. Ca2+ (1.0 mM) and Mg2+ (1.2 mM) were added to cell suspensions, and cells were incubated with and without 20 lM Rosiglitazone for up to 1 h at 37 C before stimulation with 100 nM f-MLP, and subsequent analysis on FACS as described above. Using the Grynkiewicz equation, and previously reported values for Fluo-3Õs association constant with Ca2+ [20,21], fluorescence readings were then estimated as absolute Ca2+ concentration data. Western blot analysis. To evaluate changes in the phosphorylation of Akt, 5 · 106 cells were incubated with 100 nM f-MLP, 20 lM Rosiglitazone or 100 nM Wortmannin at 37 C for the indicated times. Reactions were stopped by adding an excess of ice-cold PBS. The cells were collected by centrifugation at 4 C and resuspended in lysis buffer (10 mM Tris, pH 7.4, 100 mM NaCl, 1 mM EDTA, 1 mM EGTA, 1 mM NaF, 20 mM Na4P2O7, 2 mM Na3VO4, 1% Triton X100, 10% glycerol, 0.1% SDS, and 0.5% deoxycholate supplemented with 1 mM PMSF and protease inhibitor cocktail (1 in 20 dilution)) just prior to use and maintained on ice for 30 min. Cells were sonicated for 10 s and centrifuged at 12,000 rpm for 5 min at 4 C. Bio-Rad DC protein assays were used to determine the protein concentration of cell lysates, and Western blot analyses were performed using 50 lg protein in each case. Electrophoresis was performed (10% SDS–PAGE) and samples were transferred onto nitrocellulose membranes. The blots were blocked with 4% skimmed milk–Tris-buffered saline containing 0.1% Tween 20 (TBST) for 30 min at room temperature and further incubated overnight with anti-phospho-Akt antibody (1:1000 dilution), followed by the APlabelled anti-rabbit IgG antibody (1:2000 dilution). Phosphoprotein was detected by ECL-enhanced chemiluminescence. To detect changes in the expression of total Akt, the same membrane was incubated in a stripping solution (62.5 mM Tris–HCl, 2% SDS, and 100 mM mercaptoethanol) for 30 min at 50 C, and washed three times at room temperature in TBST. The membrane was blocked with 4% skimmed milk–TBST for 30 min. The membrane was reprobed with anti-total Akt antibody (1:1000 dilution), followed by the AP-labelled anti-rabbit IgG antibody (1:2000 dilution). Protein was detected by ECL-enhanced chemiluminescence. Activation of Akt was expressed as the ratio of phosphorylated to total Akt, as detected on Western blots using antibodies directed against Akt phosphorylated at serine residue 473 and total Akt, respectively.
N. Singh et al. / Biochemical and Biophysical Research Communications 333 (2005) 455–462
Results Cell viability Cell viability, as determined by an MTB assay and trypan blue exclusion, was not significantly reduced by incubation of cells with maximal doses of either f-MLP (100 nM) or Rosiglitazone (20 lM) for >24 h [data not shown; p < 0.05].
457
reduction in f-MLP-stimulated actin polymerization was observed even after relatively short preincubation periods (>5 min) with the drug [data not shown]. Preincubation with the phosphatidylinositol-3-kinase (PI3K) inhibitor Wortmannin (100 nM) ameliorated the response to f-MLP (116.9 ± 3.5% of basal, Fig. 1B). In contrast, preincubation with the Ca2+ sequestration inhibitor thapsigargin (1 lM; 30 min) produced a net decrease in F-actin content (74.2 ± 7.4% of basal, p < 0.05, Fig. 1B).
Formation of F-actin f-MLP stimulation (100 nM) brought about a substantial increase in the F-actin content of cells within 5 min of stimulation (138.6 ± 15.7% of basal, p < 0.05; Fig. 1). However, preincubation with Rosiglitazone negatively modulated this increase in a dose–responsive manner (IC50 = 3.72 lM, Fig. 1A). A maximally reduced response (36.0 ± 7.7% lower when compared to f-MLP alone, p < 0.05) was observed with P10 lM Rosiglitazone. In subsequent experiments, a maximal inhibitory concentration of Rosiglitazone (20 lM) was routinely used. Importantly, preincubation with 20 lM Rosiglitazone alone did not significantly affect actin polymerization (100.7 ± 3.5% of basal F-actin levels, Fig. 1B). Routinely, cells were preincubated with Rosiglitazone for 60 min, but it should be noted that a significant
The effect of Rosiglitazone on the phosphorylation of Akt in f-MLP-stimulated peripheral blood monocytes Before observing the effect of Rosiglitazone on f-MLP-induced activation of Akt, a time course of the effect of f-MLP was obtained. Increased levels of phosphorylated Akt were detected as early as 2 min after stimulation with f-MLP and showed a maximum increase at 4 min, after which gradual dephosphorylation was observed such that at 8 min it showed no significant difference from basal levels (Fig. 2). Therefore, the 4-min time point was chosen to study the effect of Rosiglitazone on f-MLP-induced phosphorylation Akt. At this maximal time point, a 2-fold increase in the levels of phospho-Akt was observed in human monocytes stimulated with 100 nM f-MLP (phospho/
Fig. 1. The effect of Rosiglitazone on f-MLP-induced actin polymerization in human peripheral mononuclear cells. (A) Dose–response curve illustrating the inhibitory effect of increasing concentrations of Rosiglitazone on f-MLP (100 nM)-induced actin polymerization in isolated PMNCs (IC50 = 3.72 lM, p < 0.05, n = 6 [ANOVA]). (B) PMNCs were incubated in the presence or absence of 100 nM f-MLP, 20 lM Rosiglitazone, 100 nM Wortmannin, and 1 lM thapsigargin as indicated, and cellular F-actin content was determined. (*p < 0.05, n = 6).
458
N. Singh et al. / Biochemical and Biophysical Research Communications 333 (2005) 455–462
20 lM Rosiglitazone alone brought about an approximately 1.4-fold increase in phosphorylation of Akt above basal levels (phospho/total Akt ratio = 0.336 ± 0.013, p < 0.05; Fig. 3). Preincubation with the PI3K inhibitor, Wortmannin, inhibited f-MLP- or Rosiglitazone-induced phosphorylation of Akt (Fig. 4). The effect of Rosiglitazone on [Ca2+]i
Fig. 2. Time course illustrating activation of Akt by f-MLP. Data illustrating the phosphorylation of Akt in monocytic cells after addition of f-MLP, as determined by Western blot analysis (n = 6).
total Akt ratio = 0.247 ± 0.014 to 0.494 ± 0.013; p < 0.05; Fig. 3). Preincubation with 20 lM Rosiglitazone for 1 h significantly reduced f-MLP-induced phosphorylation of Akt to near basal levels (phospho/total Akt ratio = 0.256 ± 0.019; p > 0.05; Fig. 3). Therefore, preincubation with Rosiglitazone prevented f-MLP-induced phosphorylation of Akt. Interestingly, incubation with
We studied the effect of Rosiglitazone on [Ca2+]i in isolated peripheral blood monocytes (Fig. 5). As expected, [Ca2+]i was less than 100 nM in resting cells in either the presence or the absence of Rosiglitazone (86.3 ± 4.1 vs 84.1 ± 5.2 nM, respectively; p > 0.05), but incubation for over 5 min with Rosiglitazone significantly increased basal [Ca2+]i levels in these cells to approximately double (i.e., 170 nM). This increase was not apparent until after 5-min incubation with Rosiglitazone, and levels remained elevated for up to 1 h post-exposure (Fig. 5;
Fig. 4. The effect of the PI3K inhibitor, Wortmannin, on f-MLP and Rosiglitazone-dependent Akt phosphorylation. Isolated monocytic cells were incubated with 100 nM Wortmannin (15 min) prior to stimulation with either 100 nM f-MLP (4 min) or 20 lM Rosiglitazone (60 min). The phosphorylation status of Akt was determined using Western blotting (results are representative of three independent experiments).
Fig. 3. The modulatory action of Rosiglitazone on f-MLP-dependent phosphorylation of Akt. Monocytic cells were incubated in the absence or presence of 100 nM f-MLP (4 min) and 20 lM Rosiglitazone (60 min) (as indicated), and the phosphorylation status of Akt was determined using Western blotting (*p < 0.05, **p < 0.001; results are representative of three independent experiments).
Fig. 5. The effect of Rosiglitazone on intracellular Ca2+ concentrations. Ca2+-dependent fluorescence of Fluo-3 in isolated mononuclear cells after incubation for the indicated times with (j) and without (d) 20 lM Rosiglitazone (n = 3, p < 0.005).
N. Singh et al. / Biochemical and Biophysical Research Communications 333 (2005) 455–462
Fig. 6. The modulatory effect of Rosiglitazone on f-MLP-induced changes in intracellular Ca2+ concentrations. Ca2+-dependent fluorescence of Fluo-3 in isolated mononuclear cells stimulated with 100 nM f-MLP without (d), and with (j) a 60-min preincubation with 20 lM Rosiglitazone, is indicated (p < 0.005, n = 3).
p < 0.05). In contrast, incubation of cells with 1 lM thapsigargin for 30 min brought about a much larger increase in [Ca2+]i to well over 1 lM (1224 ± 127 nM), as measured by Fluo-3 fluorescence (data not shown). Exposure to 100 nM f-MLP brought about a rapid increase in [Ca2+]i levels, with or without preincubation for 60 min with 20 lM Rosiglitazone (Fig. 6). Maximum [Ca2+]i levels were observed within 5 s of addition of f-MLP and were sustained until 120 s. At 5 s poststimulation, [Ca2+]i was significantly larger in Rosiglitazone-pretreated cells as compared to untreated cells (1136 ± 32 nM vs 315 ± 16 nM, respectively, p = 0.0005). Subsequently, [Ca2+]i levels decreased, so that 120 s after the addition of f-MLP, [Ca2+]i had returned to approximately basal levels in the absence of Rosiglitazone, while Rosiglitazone-treated cells showed a lower, but still significantly elevated, [Ca2+]i of approximately 180 nM (p = 0.0003 when compared to baseline).
Discussion To our knowledge, this is the first study to demonstrate that preincubation of monocytes with non-toxic doses of the oral hypoglycaemic agent Rosiglitazone modulates actin reorganisation. This effect involves a net suppression of actin polymerisation that appears to be controlled by two mechanisms: stimulation of a Ca2+ signalling pathway involved in the suppression of actin polymerisation, and inhibition of signalling via Akt phosphorylation, which is involved in triggering the process.
459
Previous investigations have shown that a correlation exists between the rigidity of leukocytes and the viscosity of blood, and that monocytes from individuals with diabetes are more rigid than those from normal controls [22]. Taken together with the observation that elevated blood viscosity is associated with both microvascular and macrovascular complications, this has led to the proposal that decreased leukocytic deformability plays a critical role in the initiation of diabetic microangiopathy [2]. Deformability of leukocytes has been shown to be governed by their F-actin content, and therefore our observation that Rosiglitazone inhibits actin polymerisation may be of important clinical relevance to the aetiology of diabetic microangiopathy. We and others have previously noted that, in addition to its metabolic effects, Rosiglitazone has been shown to have several anti-atherogenic properties, including reduced synthesis of pro-inflammatory cytokines and vascular adhesion molecules [23–25]. This study demonstrates that Rosiglitazone may also affect a key monocytic process involved in the development of microangiopathy, namely actin remodelling in response to chemotactic stimulation. We have shown that actin polymerisation in response to f-MLP stimulation is ameliorated in Rosiglitazonepreincubated cells (Fig. 1), and that polymerisation is significantly inhibited in cells even after a relatively short incubation period (P5 min). Binding of f-MLP to its receptor on the cell surface initiates signalling via several different pathways, including via PI3K and the hydrophobic second messenger molecule, phosphatidylinositol 3,4,5-trisphosphate [PIP3], which stimulate phosphorylation of the downstream kinase Akt [26–28]. Akt phosphorylation has been reported to promote actin polymerisation [29]. Our data are in agreement with these findings, with maximum increases in both Akt phosphorylation and actin polymerisation being seen 4–5 min after stimulation with f-MLP (p < 0.05; Figs. 1 and 2), and inhibition of both processes being evident in cells pretreated with the PI3K inhibitor, Wortmannin (Figs. 1 and 4). Preincubation of cells with Rosiglitazone inhibited fMLP-stimulated Akt activation, reducing its phosphorylation to levels not significantly different to basal (Fig. 3). The mechanism by which this occurs is not yet clear, but our observation that Rosigliazone alone can induce Akt phosphorylation in a Wortmannin-inhibitable manner (Fig. 4) suggests that f-MLP- and Rosiglitazone-initiated signalling pathways may converge antagonistically at a point upstream of Akt (perhaps at PI3K). In support of this, Meyer et al. [38] have reported increases in Akt phosphorylation after treatment with TZD drugs. Another signalling pathway involved in actin remodelling involves changes in intracellular levels of calcium ions. [Ca2+]i are routinely kept very low
460
N. Singh et al. / Biochemical and Biophysical Research Communications 333 (2005) 455–462
([Ca2+]i < 0.1 lM), with controlled releases of the ion into the cytoplasm, either from internal stores or from the external milieu, acting as triggers for the activation of a variety of intracellular processes. It has been reported that stimulation of f-MLP receptors results in production of the intracellular second messenger inositol triphosphate, which mobilizes calcium from intracellular stores [31]. Our experiments conformed to this general model: stimulation with f-MLP caused a marked but transient increase in [Ca2+]i. However, preincubation with Rosiglitazone caused elevated basal [Ca2+]i levels (Fig. 5), and a significantly larger increase in Ca2+-dependent fluorescence than in control cells after stimulation with f-MLP (Fig. 6). We suggest that the contrasting mode of the response to Rosiglitazone alone compared to that induced by f-MLP may reflect a distinct mechanism. It is noteworthy that the timescale of the former response resembles that of cells subjected to treatment with inhibitors of SERCA2b, the ÔhousekeepingÕ Ca2+ pump that is responsible for a sequestration of Ca2+ into the ER compensatory to non-specific leakage of Ca2+ into the cytoplasm [32]. Our demonstration of gradual increases in [Ca2+]i in cells treated with Rosiglitazone suggests that, rather than triggering a Ca2+ signalling cascade, the drug may non-specifically disrupt Ca2+ homeostasis in such a way as to cause Ca2+ ions to leak into the cytoplasm. Although some have suggested that cytoskeletal remodelling occurs independently of [Ca2+]i in leukocytes [26,33], others have reported that Ca2+ triggers actin depolymerisation in neutrophils and monocytes [34]. The mechanism by which Ca2+ controls depolymerisation may be via the activation of gelsolin, a 91-kDa glycoprotein that severs actin filaments and prevents filament elongation by capping fast-growing filamentous ends in a Ca2+-dependent manner [35]. Thus, the local increases in [Ca2+]i seen immediately after monocyte activation may fulfil a negative regulatory role in preventing excessive actin polymerisation in some locations. However, if Ca2+ homeostasis is disrupted by the use of ionophores, energy depletion, or Ca2+ pump inhibitors, the result is disintegration of the actin cytoskeleton due to large-scale Ca2+-dependent activation of gelsolin [15,36,37]. Hence, we observe that if cells are pretreated with thapsigargin, then their F-actin content is reduced below basal levels (74.2 ± 7.4%, p < 0.05, Fig. 1B). In contrast, the retention of a classical ‘‘Ca2+ spike’’ in f-MLP-stimulated cells after pretreatment with Rosiglitazone (Fig. 6), and indeed the drugÕs lack of cytotoxicity, suggests that Rosiglitazone does not produce such drastic changes. Instead, we suggest that treatment with Rosiglitazone causes a small non-specific increase in [Ca2+]i, and so may sensitise Ca2+ signalling mechanisms to chemotactic stimulation. As Ca2+ signalling in response to f-MLP is suppressed in diabetic
neutrophils [16], these data provide a mechanism by which treatment with Rosiglitazone may restore normal Ca2+ signalling to leukocytes from patients with diabetes. However, while incubation of monocytes with Rosiglitazone alone brought about significant increases in both Akt phosphorylation and resting [Ca2+]i, it must be emphasised that the drug alone did not stimulate actin polymerisation (Fig. 1B). Nevertheless, we hypothesise that, while not directly evoking cytoskeletal responses, by modulation of Akt and [Ca2+]i, Rosiglitazone suppresses the signalling events required to promote actin polymerisation. In support of this hypothesis, there is now a growing body of evidence that PPAR-c agonists exert anti-inflammatory effects through so-called ‘‘non-genomic’’ or ‘‘PPAR-c-independent’’ events, in which ligand-activated PPAR-c receptors (or possibly the ligands acting independently) may directly crosstalk with, and therefore modulate, multiple signalling pathways. For example, PPAR-c ligands have been shown to activate ERK1/2 and p38 MAPK in VSMCs and HT-29 colon cancer cells [9,40]. Moreover, the ability of PPAR-c ligands to influence the release of bioactive mediators from platelets, which lack nuclei, is further evidence of the non-genomic nature of at least some of the effects of PPAR-c ligands [41]. It remains to be seen whether the cytoskeletal remodelling effects we have observed in monocytes also occur in other cell types, but it may be relevant that the PPAR-c ligands, troglitazone and ciglitizone, can inhibit migration of endothelium cells via inhibition of leptin-dependent Akt signalling [30], while Knock et al. [39] have demonstrated that troglitazone (and to a lesser extent Rosiglitazone) can disrupt [Ca2+]i homeostasis in aortic myocytes. As the actin cytoskeleton of such cells is involved in the maintenance of a healthy vasculature, this supports the hypothesis that TZD drugs can act as modulators of the cytoskeletal remodelling involved in the aetiology of vascular complications of diabetes. However, it should be stressed that, while the modulation of actin remodelling could have clinically beneficial effects such as reducing cell rigidity in the microvasculature of patients with diabetes, such modulation could be deleterious in other contexts. For example, impaired neutrophil actin polymerisation has been shown to lead to reduced primary granule exocytosis and chemotaxis [16]. In conclusion, we hypothesise that Rosiglitazone interacts with signalling events downstream of occupancy of the f-MLP receptor (i.e., modulation of Akt phosphorylation and [Ca2+]i) to inhibit monocyte actin polymerisation in a non-genomic manner. This study demonstrates that TZD therapy not only influences the expression of genes involved in glucose and lipid metabolism, and the release of inflammatory mediators, but
N. Singh et al. / Biochemical and Biophysical Research Communications 333 (2005) 455–462
also modulates basic monocyte physiology, and that the true benefits of TZD therapy may be due to a combination of PPAR-c-independent and PPAR-c-dependent events. Our observations support the concept that TZD therapy may influence both macrovascular complications and microangiopathy through mechanisms independent of metabolic effects in diabetes.
References [1] S. Schroder, W. Palinski, G.W. Schmid-Schonbein, Activated monocytes and granulocytes, capillary nonperfusion, and neovascularization in diabetic retinopathy, Am. J. Pathol. 139 (1991) 81– 100. [2] K. Miyamoto, Y. Ogura, S. Kenmochi, Y. Honda, Role of leukocytes in diabetic microcirculatory disturbances, Microvasc. Res. 54 (1997) 43–48. [3] M.C. Mazzoni, G.W. Schmid-Schonbein, Mechanisms and consequences of cell activation in the microcirculation, Cardiovasc. Res. 32 (1996) 709–719. [4] E. Hu, J.B. Kim, P. Sarraf, B.M. Spiegelman, Inhibition of adipogenesis through MAP kinase-mediated phosphorylation of PPARgamma, Science 274 (1996) 2100–2103. [5] J. Plutzky, The vascular biology of atherosclerosis, Am. J. Med. 115 (Suppl. 8A) (2003) 55S–61S. [6] M. Ricote, A.C. Li, T.M. Willson, C.J. Kelly, C.K. Glass, The peroxisome proliferator-activated receptor-gamma is a negative regulator of macrophage activation, Nature 391 (1998) 79–82. [7] N. Marx, D. Walcher, N. Ivanova, K. Rautzenberg, A. Jung, R. Friedl, V. Hombach, C.R. de, G. Basta, M.P. Wautier, J.L. Wautiers, Thiazolidinediones reduce endothelial expression of receptors for advanced glycation end products, Diabetes 53 (2004) 2662–2668. [8] G. Hong, B. Davis, N. Khatoon, S.F. Baker, J. Brown, PPAR gamma-dependent anti-inflammatory action of rosiglitazone in human monocytes: suppression of TNF alpha secretion is not mediated by PTEN regulation, Biochem. Biophys. Res. Commun. 303 (2003) 782–787. [9] K. Takeda, T. Ichiki, T. Tokunou, N. Iino, A. Takeshita, 15Deoxy-delta 12,14-prostaglandin J2 and thiazolidinediones activate the MEK/ERK pathway through phosphatidylinositol 3kinase in vascular smooth muscle cells, J. Biol. Chem. 276 (2001) 48950–48955. [10] N. Marx, U. Schonbeck, M.A. Lazar, P. Libby, J. Plutzky, Peroxisome proliferator-activated receptor gamma activators inhibit gene expression and migration in human vascular smooth muscle cells, Circ. Res. 83 (1998) 1097–1103. [11] S. Goetze, F. Eilers, A. Bungenstock, U. Kintscher, P. Stawowy, F. Blaschke, K. Graf, R.E. Law, E. Fleck, M. Grafe, PPAR activators inhibit endothelial cell migration by targeting Akt, Biochem. Biophys. Res. Commun. 293 (2002) 1431–1437. [12] D. Chodniewicz, D.V. Zhelev, Chemoattractant receptor-stimulated F-actin polymerization in the human neutrophil is signaled by 2 distinct pathways, Blood 101 (2003) 1181–1184. [13] V. Vasioukhin, C. Bauer, M. Yin, E. Fuchs, Directed actin polymerization is the driving force for epithelial cell–cell adhesion, Cell 100 (2000) 209–219. [14] W. Siess, Molecular mechanisms of platelet activation, Physiol. Rev. 69 (1989) 58–178. [15] G.P. Downey, C.K. Chan, S. Trudel, S. Grinstein, Actin assembly in electropermeabilized neutrophils: role of intracellular calcium, J. Cell Biol. 110 (1990) 1975–1982.
461
[16] A. Advani, S.M. Marshall, T.H. Thomas, Impaired neutrophil store-mediated calcium entry in Type 2 diabetes, Eur. J. Clin. Invest. 34 (2004) 43–49. [17] P.L. Sonkin, S.F. Freedman, D. Needham, K.M. Rao, D.L. Hatchell, Pentoxifylline modulates deformability, F-actin content, and superoxide anion production of polymorphonuclear leukocytes from diabetic cats, Exp. Eye Res. 55 (1992) 831–838. [18] T.H. Howard, W.H. Meyer, Chemotactic peptide modulation of actin assembly and locomotion in neutrophils, J. Cell Biol. 98 (1984) 1265–1271. [19] J. Elsner, J. Norgauer, G.J. Dobos, A. Emmendorffer, E. Schopf, A. Kapp, J. Roesler, Flow cytometry reveals different lag times in rapid cytoplasmic calcium elevations in human neutrophils in response to N-formyl peptide, J. Cell. Physiol. 157 (1993) 637–643. [20] G. Grynkiewicz, M. Poenie, R.Y. Tsien, A new generation of Ca2+ indicators with greatly improved fluorescence properties, J. Biol. Chem. 260 (1985) 3440–3450. [21] J.E. Merritt, S.A. McCarthy, M.P. Davies, K.E. Moores, Use of fluo-3 to measure cytosolic Ca2+ in platelets and neutrophils. Loading cells with the dye, calibration of traces, measurements in the presence of plasma, and buffering of cytosolic Ca2+, Biochem. J. 269 (1990) 513–519. [22] G. Athanassiou, P. Matsouka, V. Kaleridis, Y. Missirlis, Deformability and filterability of white blood cell subpopulations. Evaluation of these parameters in the cell line HL-60 and in type II diabetes mellitus, Clin. Hemorheol. Microcirc. 22 (2000) 35–43. [23] C. Jiang, A.T. Ting, B. Seed, PPAR-gamma agonists inhibit production of monocyte inflammatory cytokines, Nature 391 (1998) 82–86. [24] A.W. Roberts, A. Thomas, A. Rees, M. Evans, Peroxisome proliferator-activated receptor-gamma agonists in atherosclerosis: current evidence and future directions, Curr. Opin. Lipidol. 14 (2003) 567–573. [25] J. Plutzky, The potential role of peroxisome proliferator-activated receptors on inflammation in type 2 diabetes mellitus and atherosclerosis, Am. J. Cardiol. 92 (2003) 34J–41J. [26] Y.D. Cui, O. Inanami, T. Yamamori, K. Niwa, H. Nagahata, M. Kuwabara, FMLP-induced formation of F-actin in HL60 cells is dependent on PI3-K but not on intracellular Ca2+, PKC, ERK or p38 MAPK, Inflamm. Res. 49 (2000) 684–691. [27] Y.I. Miller, D.S. Worrall, C.D. Funk, J.R. Feramisco, J.L. Witztum, Actin polymerization in macrophages in response to oxidized LDL and apoptotic cells: role of 12/15-lipoxygenase and phosphoinositide 3-kinase, Mol. Biol. Cell 14 (2003) 4196–4206. [28] A. Ptasznik, E.R. Prossnitz, D. Yoshikawa, A. Smrcka, A.E. Traynor-Kaplan, G.M. Bokoch, A tyrosine kinase signaling pathway accounts for the majority of phosphatidylinositol 3,4,5trisphosphate formation in chemoattractant-stimulated human neutrophils, J. Biol. Chem. 271 (1996) 25204–25207. [29] V. Cenni, A. Sirri, M. Riccio, G. Lattanzi, S. Santi, A. de Pol, N.M. Maraldi, S. Marmiroli, Targeting of the Akt/PKB kinase to the actin skeleton, Cell. Mol. Life Sci. 60 (2003) 2710–2720. [30] S. Goetze, A. Bungenstock, C. Czupalla, F. Eilers, P. Stawowy, U. Kintscher, C. Spencer-Hansch, K. Graf, B. Nurnberg, R.E. Law, E. Fleck, M. Grafe, Leptin induces endothelial cell migration through Akt, which is inhibited by PPARgamma-ligands, Hypertension 40 (2002) 748–754. [31] E.R. Prossnitz, O. Quehenberger, C.G. Cochrane, R.D. Ye, Signal transducing properties of the N-formyl peptide receptor expressed in undifferentiated HL60 cells, J. Immunol. 151 (1993) 5704–5715. [32] L.M. John, J.D. Lechleiter, P. Camacho, Differential modulation of SERCA2 isoforms by calreticulin, J. Cell Biol. 142 (1998) 963– 973. [33] F.A. al Mohanna, M.B. Hallett, Actin polymerization in neutrophils is triggered without a requirement for a rise in cytoplasmic Ca2+, Biochem. J. 266 (1990) 669–674.
462
N. Singh et al. / Biochemical and Biophysical Research Communications 333 (2005) 455–462
[34] R.C. May, L.M. Machesky, Phagocytosis and the actin cytoskeleton, J. Cell Sci. 114 (2001) 1061–1077. [35] H.L. Yin, J.H. Albrecht, A. Fattoum, Identification of gelsolin, a Ca2+-dependent regulatory protein of actin gel–sol transformation, and its intracellular distribution in a variety of cells and tissues, J. Cell Biol. 91 (1981) 901–906. [36] K. Kanno, Y. Sasaki, Smooth muscle gelsolin and a Ca2+sensitive contractile cell model, J. Cell. Physiol. 139 (1989) 58–67. [37] W. Kuhne, M. Besselmann, T. Noll, A. Muhs, H. Watanabe, H.M. Piper, Disintegration of cytoskeletal structure of actin filaments in energy-depleted endothelial cells, Am. J. Physiol. 264 (1993) H1599–H1608. [38] M.M. Meyer, K. Levin, T. Grimmsmann, N. Perwitz, A. Eirich, H. Beck-Nielsen, H.H. Klein, Troglitazone treatment increases protein kinase B phosphorylation in skeletal muscle of normo-
glycemic subjects at risk for the development of type 2 diabetes, Diabetes 51 (2002) 2691–2697. [39] G.A. Knock, S.K. Mishra, P.I. Aaronson, Differential effects of insulin-sensitizers troglitazone and rosiglitazone on ion currents in rat vascular myocytes, Eur. J. Pharmacol. 368 (1999) 103–109. [40] F. Chen, M. Wang, J.P. OÕConnor, M. He, T. Tripathi, L.E. Harrison, Phosphorylation of PPARgamma via active ERK1/2 leads to its physical association with p65 and inhibition of NFkappabeta, J. Cell. Biochem. 90 (2003) 732–744. [41] F. Akbiyik, D.M. Ray, K.F. Gettings, N. Blumberg, C.W. Francis, R.P. Phipps, Human bone marrow megakaryocytes and platelets express PPARgamma, and PPARgamma agonists blunt platelet release of CD40 ligand and thromboxanes, Blood 104 (2004) 1361–1368.
The PPAR-c activator, Rosiglitazone, inhibits actin polymerisation in monocytes: Involvement of Akt and intracellular calcium Neenu Singh a, Richard Webb a, Rachel Adams a, Shelley-Ann Evans a, Ameena Al-Mosawi a, Marc Evans b, Aled W. Roberts b, Andrew W. Thomas a,* a
School of Applied Science, University of Wales Institute Cardiff, Western Avenue, Cardiff CF5 2YB, UK b University Hospital of Wales, Heath Park, Cardiff CF5 2XN, UK Received 17 May 2005 Available online 2 June 2005
Abstract Monocyte hyperactivation as seen in diabetes results in increased cytoskeletal rigidity and reduced cell deformability leading to microchannel occlusions and microvascular complications. The thiazolidinediones (TZDs) are PPAR-c agonists that have been reported to exert beneficial non-metabolic effects on the vasculature. This study demonstrates that the TZD, Rosiglitazone, significantly reduces f-MLP-induced actin polymerisation in human monocytic cells (p < 0.05). Two of the key signalling processes known to be involved in the regulation of cytoskeletal remodelling were investigated: PI3K-dependent Akt phosphorylation and intracellular calcium concentration [Ca2+]i. The PI3K inhibitor, Wortmannin, ameliorated f-MLP-induced actin polymerisation (p < 0.05), while the Ca2+ sequestration inhibitor, thapsigargin, induced actin depolymerisation (p < 0.05), confirming the involvement of both processes in cytoskeletal remodelling. Rosiglitazone significantly reduced f-MLP activation of Akt (p < 0.05), and significantly increased [Ca2+]i in both resting and f-MLP-stimulated cells (p < 0.05). Therefore, Rosiglitazone interacts with signalling events downstream of occupancy of the f-MLP receptor, to modulate cytoskeletal remodelling in a PPAR-c-independent manner. To our knowledge, these results are the first to present evidence that a PPAR-c agonist can modulate actin remodelling in monocytes, and may therefore be protective against microvascular damage in diabetes. 2005 Elsevier Inc. All rights reserved. Keywords: Rosiglitazone; Actin polymerisation; Calcium; Akt; Diabetes; Microvascular; PPAR-c-independent
Monocytic activation is a normal physiological process involving changes in cell morphology and motility and is involved in both inflammation and phagocytosis. However, it has been shown that in diabetes, monocytes can become hyperactivated, resulting in increased rigidity of cell membranes and reduced cell deformability. This can result in microchannel occlusions and contribute to microvascular complications such as diabetic retinopathy, nephropathy, and neuropathy [1,2]. Cell deformability is determined by the degree of cytoplasmic actin polymerisation. External signals (e.g. growth factors and chemoattractants) can induce poly*
Corresponding author. Fax: +442920416982. E-mail address: [email protected] (A.W. Thomas).
0006-291X/$ - see front matter 2005 Elsevier Inc. All rights reserved. doi:10.1016/j.bbrc.2005.05.127
merisation of globular G-actin into filamentous F-actin in order to bring about changes in cell morphology and motility. For example, actin polymerisation is involved in the pseudopod formation that has been shown to be an important physical parameter determining the activation of monocytes [3]. Pseudopods are stiffer than the main cell body and, consequently, circulation of hyperactivated monocytes may lead to a greater risk of entrapment of cells within capillaries. Thus, hyperactivation of monocytes may confer an increased risk for the progression of microvascular complications in diabetes. PPAR-c is a ligand-activated nuclear receptor that regulates the expression of diverse genes involved in glucose and lipid metabolism, cell differentiation,
456
N. Singh et al. / Biochemical and Biophysical Research Communications 333 (2005) 455–462
apoptosis, and inflammation [4]. The ligands of PPAR-c include oxidative metabolites of polyunsaturated fatty acids, prostaglandins such as 15d-PGJ2, and oral antihyperglycaemic agents of the thiazolidinedione (TZD) class such as Rosiglitazone. There is evidence that, apart from their insulin-sensitising activity, TZDs exert a plethora of effects on the process of atherothrombosis which may provide macrovascular benefits [5–7]. By contrast, there is a relative paucity of data evaluating the potential effects of TZDs on mechanisms involved in the development of microangiopathy. Some of these vascular effects seem to be mediated by a mechanism independent of PPAR-c-activated gene expression [8]. These so-called ‘‘non-genomic’’ effects of PPAR-c ligands have been shown to influence intracellular signals such as changes in cytosolic calcium ion concentrations ([Ca2+]i) or protein kinase (e.g. Akt) activity [9–11]. Both of these processes are known to be involved in the regulation of the actin cytoskeleton: chemoattractant-induced reorganisation of the actin cytoskeleton is dependent on phosphorylation of Akt [12], while Ca2+ entry into the cell and sustained increases in [Ca2+]i are involved in the Ca2+-dependent disassembly of the actin cytoskeleton mediated by gelsolin [13–15]. Therefore, abnormal Akt activity or [Ca2+]i dysregulation in the monocytes of individuals with diabetes could be responsible for inappropriate signalling, abnormal monocytic activation, capillary plugging, and consequently microangiopathy [16]. Thus, decreases in F-actin content may result in reduced activation and increased deformability of leukocytes, which may contribute to prevention of microvascular injury [17]. In the present study, we attempt to elucidate whether Rosiglitazone modulates monocytic actin polymerisation, and also how it affects the key intracellular signals involved in cytoskeletal remodelling (particularly phosphorylation of Akt and changes in [Ca2+]i). This study may therefore provide an insight into some of the mechanisms by which Rosiglitazone may potentially influence the pathogenesis of microangiopathy in diabetes.
Materials and methods Materials. All chemicals were purchased from Sigma–Aldrich (Poole, UK) unless stated otherwise. Cell permeabilisation reagents, protein assay reagents, FITC–phalloidin, and Fluo-3 AM were obtained from Harlan SERA-LAB (Loughborough, UK), Bio-Rad Laboratories (Herts, UK), Molecular Probes (Eugene, OR), and Calbiochem, EMD Biosciences (Darmstadt, Germany), respectively. Rosiglitazone was obtained from GlaxoSmithKline (Uxbridge, UK). Rabbit anti-phospho-Akt (Ser473), anti-total Akt, and anti-rabbit APconjugated secondary antibody were purchased from New England BioLabs (Hitchin, UK). Isolation of human peripheral mononuclear cells. For isolation of peripheral blood mononuclear cells, 40 ml of EDTA-anticoagulated blood was obtained from healthy patients. Ten millilitres of blood was
layered over 10 ml Ficoll–Hypaque and centrifuged at 450g for 35 min. The mononuclear cell suspension was carefully removed from the Ficoll–Hypaque interface, washed three times in PBS/EDTA, and resuspended in RPMI 1640 for use in all the experiments. Cell viability was >95% as measured by trypan blue staining. Flow cytometric analysis of the formation of F-actin. Actin polymerisation was determined by the method of Howard and Meyer [18]. Briefly, peripheral blood monocytic cells (1.0 · 105 cells/ml) were incubated at 37 C in PBS. After 60-min incubation with or without 20 lM Rosiglitazone and/or stimulation for 5 min with 100 nM fMLP, cells were fixed, permeabilised, and stained in a single step by addition of 0.1 ml of 37% phosphate-buffered formalin containing 1.65 · 10 6 M FITC–phalloidin and 100 lg lysophosphatidylcholine. In some cases, cells were also preincubated for 15 min with 100 nM Wortmannin or for 30 min with 1 lM thapsigargin. Cells were then incubated at 37 C for 60 min, washed, and harvested by centrifugation. The cell pellet was resuspended in 0.5 ml PBS before analysis on FACS. The samples were excited by an argon laser at 488 nm and emission was measured at 530 nm. Measurement of intracellular Ca2+ concentration. [Ca2+]i was measured by the method of Elsner et al. [19], with some minor modifications. Briefly, the fluorescence of the Ca2+ indicator Fluo-3 was used as an indicator of [Ca2+]i at defined time intervals after stimulation with 100 nM f-MLP/preincubation with 20 lM Rosiglitazone or 1 lM thapsigargin. Cell suspensions (7 · 106 ml in RPMI 1640 medium) were incubated with 3 lM Fluo-3 AM for 40 min at 37 C. Cells were washed twice and resuspended in RPMI 1640 at a final concentration of 1 · 106 ml. Ca2+ (1.0 mM) and Mg2+ (1.2 mM) were added to cell suspensions, and cells were incubated with and without 20 lM Rosiglitazone for up to 1 h at 37 C before stimulation with 100 nM f-MLP, and subsequent analysis on FACS as described above. Using the Grynkiewicz equation, and previously reported values for Fluo-3Õs association constant with Ca2+ [20,21], fluorescence readings were then estimated as absolute Ca2+ concentration data. Western blot analysis. To evaluate changes in the phosphorylation of Akt, 5 · 106 cells were incubated with 100 nM f-MLP, 20 lM Rosiglitazone or 100 nM Wortmannin at 37 C for the indicated times. Reactions were stopped by adding an excess of ice-cold PBS. The cells were collected by centrifugation at 4 C and resuspended in lysis buffer (10 mM Tris, pH 7.4, 100 mM NaCl, 1 mM EDTA, 1 mM EGTA, 1 mM NaF, 20 mM Na4P2O7, 2 mM Na3VO4, 1% Triton X100, 10% glycerol, 0.1% SDS, and 0.5% deoxycholate supplemented with 1 mM PMSF and protease inhibitor cocktail (1 in 20 dilution)) just prior to use and maintained on ice for 30 min. Cells were sonicated for 10 s and centrifuged at 12,000 rpm for 5 min at 4 C. Bio-Rad DC protein assays were used to determine the protein concentration of cell lysates, and Western blot analyses were performed using 50 lg protein in each case. Electrophoresis was performed (10% SDS–PAGE) and samples were transferred onto nitrocellulose membranes. The blots were blocked with 4% skimmed milk–Tris-buffered saline containing 0.1% Tween 20 (TBST) for 30 min at room temperature and further incubated overnight with anti-phospho-Akt antibody (1:1000 dilution), followed by the APlabelled anti-rabbit IgG antibody (1:2000 dilution). Phosphoprotein was detected by ECL-enhanced chemiluminescence. To detect changes in the expression of total Akt, the same membrane was incubated in a stripping solution (62.5 mM Tris–HCl, 2% SDS, and 100 mM mercaptoethanol) for 30 min at 50 C, and washed three times at room temperature in TBST. The membrane was blocked with 4% skimmed milk–TBST for 30 min. The membrane was reprobed with anti-total Akt antibody (1:1000 dilution), followed by the AP-labelled anti-rabbit IgG antibody (1:2000 dilution). Protein was detected by ECL-enhanced chemiluminescence. Activation of Akt was expressed as the ratio of phosphorylated to total Akt, as detected on Western blots using antibodies directed against Akt phosphorylated at serine residue 473 and total Akt, respectively.
N. Singh et al. / Biochemical and Biophysical Research Communications 333 (2005) 455–462
Results Cell viability Cell viability, as determined by an MTB assay and trypan blue exclusion, was not significantly reduced by incubation of cells with maximal doses of either f-MLP (100 nM) or Rosiglitazone (20 lM) for >24 h [data not shown; p < 0.05].
457
reduction in f-MLP-stimulated actin polymerization was observed even after relatively short preincubation periods (>5 min) with the drug [data not shown]. Preincubation with the phosphatidylinositol-3-kinase (PI3K) inhibitor Wortmannin (100 nM) ameliorated the response to f-MLP (116.9 ± 3.5% of basal, Fig. 1B). In contrast, preincubation with the Ca2+ sequestration inhibitor thapsigargin (1 lM; 30 min) produced a net decrease in F-actin content (74.2 ± 7.4% of basal, p < 0.05, Fig. 1B).
Formation of F-actin f-MLP stimulation (100 nM) brought about a substantial increase in the F-actin content of cells within 5 min of stimulation (138.6 ± 15.7% of basal, p < 0.05; Fig. 1). However, preincubation with Rosiglitazone negatively modulated this increase in a dose–responsive manner (IC50 = 3.72 lM, Fig. 1A). A maximally reduced response (36.0 ± 7.7% lower when compared to f-MLP alone, p < 0.05) was observed with P10 lM Rosiglitazone. In subsequent experiments, a maximal inhibitory concentration of Rosiglitazone (20 lM) was routinely used. Importantly, preincubation with 20 lM Rosiglitazone alone did not significantly affect actin polymerization (100.7 ± 3.5% of basal F-actin levels, Fig. 1B). Routinely, cells were preincubated with Rosiglitazone for 60 min, but it should be noted that a significant
The effect of Rosiglitazone on the phosphorylation of Akt in f-MLP-stimulated peripheral blood monocytes Before observing the effect of Rosiglitazone on f-MLP-induced activation of Akt, a time course of the effect of f-MLP was obtained. Increased levels of phosphorylated Akt were detected as early as 2 min after stimulation with f-MLP and showed a maximum increase at 4 min, after which gradual dephosphorylation was observed such that at 8 min it showed no significant difference from basal levels (Fig. 2). Therefore, the 4-min time point was chosen to study the effect of Rosiglitazone on f-MLP-induced phosphorylation Akt. At this maximal time point, a 2-fold increase in the levels of phospho-Akt was observed in human monocytes stimulated with 100 nM f-MLP (phospho/
Fig. 1. The effect of Rosiglitazone on f-MLP-induced actin polymerization in human peripheral mononuclear cells. (A) Dose–response curve illustrating the inhibitory effect of increasing concentrations of Rosiglitazone on f-MLP (100 nM)-induced actin polymerization in isolated PMNCs (IC50 = 3.72 lM, p < 0.05, n = 6 [ANOVA]). (B) PMNCs were incubated in the presence or absence of 100 nM f-MLP, 20 lM Rosiglitazone, 100 nM Wortmannin, and 1 lM thapsigargin as indicated, and cellular F-actin content was determined. (*p < 0.05, n = 6).
458
N. Singh et al. / Biochemical and Biophysical Research Communications 333 (2005) 455–462
20 lM Rosiglitazone alone brought about an approximately 1.4-fold increase in phosphorylation of Akt above basal levels (phospho/total Akt ratio = 0.336 ± 0.013, p < 0.05; Fig. 3). Preincubation with the PI3K inhibitor, Wortmannin, inhibited f-MLP- or Rosiglitazone-induced phosphorylation of Akt (Fig. 4). The effect of Rosiglitazone on [Ca2+]i
Fig. 2. Time course illustrating activation of Akt by f-MLP. Data illustrating the phosphorylation of Akt in monocytic cells after addition of f-MLP, as determined by Western blot analysis (n = 6).
total Akt ratio = 0.247 ± 0.014 to 0.494 ± 0.013; p < 0.05; Fig. 3). Preincubation with 20 lM Rosiglitazone for 1 h significantly reduced f-MLP-induced phosphorylation of Akt to near basal levels (phospho/total Akt ratio = 0.256 ± 0.019; p > 0.05; Fig. 3). Therefore, preincubation with Rosiglitazone prevented f-MLP-induced phosphorylation of Akt. Interestingly, incubation with
We studied the effect of Rosiglitazone on [Ca2+]i in isolated peripheral blood monocytes (Fig. 5). As expected, [Ca2+]i was less than 100 nM in resting cells in either the presence or the absence of Rosiglitazone (86.3 ± 4.1 vs 84.1 ± 5.2 nM, respectively; p > 0.05), but incubation for over 5 min with Rosiglitazone significantly increased basal [Ca2+]i levels in these cells to approximately double (i.e., 170 nM). This increase was not apparent until after 5-min incubation with Rosiglitazone, and levels remained elevated for up to 1 h post-exposure (Fig. 5;
Fig. 4. The effect of the PI3K inhibitor, Wortmannin, on f-MLP and Rosiglitazone-dependent Akt phosphorylation. Isolated monocytic cells were incubated with 100 nM Wortmannin (15 min) prior to stimulation with either 100 nM f-MLP (4 min) or 20 lM Rosiglitazone (60 min). The phosphorylation status of Akt was determined using Western blotting (results are representative of three independent experiments).
Fig. 3. The modulatory action of Rosiglitazone on f-MLP-dependent phosphorylation of Akt. Monocytic cells were incubated in the absence or presence of 100 nM f-MLP (4 min) and 20 lM Rosiglitazone (60 min) (as indicated), and the phosphorylation status of Akt was determined using Western blotting (*p < 0.05, **p < 0.001; results are representative of three independent experiments).
Fig. 5. The effect of Rosiglitazone on intracellular Ca2+ concentrations. Ca2+-dependent fluorescence of Fluo-3 in isolated mononuclear cells after incubation for the indicated times with (j) and without (d) 20 lM Rosiglitazone (n = 3, p < 0.005).
N. Singh et al. / Biochemical and Biophysical Research Communications 333 (2005) 455–462
Fig. 6. The modulatory effect of Rosiglitazone on f-MLP-induced changes in intracellular Ca2+ concentrations. Ca2+-dependent fluorescence of Fluo-3 in isolated mononuclear cells stimulated with 100 nM f-MLP without (d), and with (j) a 60-min preincubation with 20 lM Rosiglitazone, is indicated (p < 0.005, n = 3).
p < 0.05). In contrast, incubation of cells with 1 lM thapsigargin for 30 min brought about a much larger increase in [Ca2+]i to well over 1 lM (1224 ± 127 nM), as measured by Fluo-3 fluorescence (data not shown). Exposure to 100 nM f-MLP brought about a rapid increase in [Ca2+]i levels, with or without preincubation for 60 min with 20 lM Rosiglitazone (Fig. 6). Maximum [Ca2+]i levels were observed within 5 s of addition of f-MLP and were sustained until 120 s. At 5 s poststimulation, [Ca2+]i was significantly larger in Rosiglitazone-pretreated cells as compared to untreated cells (1136 ± 32 nM vs 315 ± 16 nM, respectively, p = 0.0005). Subsequently, [Ca2+]i levels decreased, so that 120 s after the addition of f-MLP, [Ca2+]i had returned to approximately basal levels in the absence of Rosiglitazone, while Rosiglitazone-treated cells showed a lower, but still significantly elevated, [Ca2+]i of approximately 180 nM (p = 0.0003 when compared to baseline).
Discussion To our knowledge, this is the first study to demonstrate that preincubation of monocytes with non-toxic doses of the oral hypoglycaemic agent Rosiglitazone modulates actin reorganisation. This effect involves a net suppression of actin polymerisation that appears to be controlled by two mechanisms: stimulation of a Ca2+ signalling pathway involved in the suppression of actin polymerisation, and inhibition of signalling via Akt phosphorylation, which is involved in triggering the process.
459
Previous investigations have shown that a correlation exists between the rigidity of leukocytes and the viscosity of blood, and that monocytes from individuals with diabetes are more rigid than those from normal controls [22]. Taken together with the observation that elevated blood viscosity is associated with both microvascular and macrovascular complications, this has led to the proposal that decreased leukocytic deformability plays a critical role in the initiation of diabetic microangiopathy [2]. Deformability of leukocytes has been shown to be governed by their F-actin content, and therefore our observation that Rosiglitazone inhibits actin polymerisation may be of important clinical relevance to the aetiology of diabetic microangiopathy. We and others have previously noted that, in addition to its metabolic effects, Rosiglitazone has been shown to have several anti-atherogenic properties, including reduced synthesis of pro-inflammatory cytokines and vascular adhesion molecules [23–25]. This study demonstrates that Rosiglitazone may also affect a key monocytic process involved in the development of microangiopathy, namely actin remodelling in response to chemotactic stimulation. We have shown that actin polymerisation in response to f-MLP stimulation is ameliorated in Rosiglitazonepreincubated cells (Fig. 1), and that polymerisation is significantly inhibited in cells even after a relatively short incubation period (P5 min). Binding of f-MLP to its receptor on the cell surface initiates signalling via several different pathways, including via PI3K and the hydrophobic second messenger molecule, phosphatidylinositol 3,4,5-trisphosphate [PIP3], which stimulate phosphorylation of the downstream kinase Akt [26–28]. Akt phosphorylation has been reported to promote actin polymerisation [29]. Our data are in agreement with these findings, with maximum increases in both Akt phosphorylation and actin polymerisation being seen 4–5 min after stimulation with f-MLP (p < 0.05; Figs. 1 and 2), and inhibition of both processes being evident in cells pretreated with the PI3K inhibitor, Wortmannin (Figs. 1 and 4). Preincubation of cells with Rosiglitazone inhibited fMLP-stimulated Akt activation, reducing its phosphorylation to levels not significantly different to basal (Fig. 3). The mechanism by which this occurs is not yet clear, but our observation that Rosigliazone alone can induce Akt phosphorylation in a Wortmannin-inhibitable manner (Fig. 4) suggests that f-MLP- and Rosiglitazone-initiated signalling pathways may converge antagonistically at a point upstream of Akt (perhaps at PI3K). In support of this, Meyer et al. [38] have reported increases in Akt phosphorylation after treatment with TZD drugs. Another signalling pathway involved in actin remodelling involves changes in intracellular levels of calcium ions. [Ca2+]i are routinely kept very low
460
N. Singh et al. / Biochemical and Biophysical Research Communications 333 (2005) 455–462
([Ca2+]i < 0.1 lM), with controlled releases of the ion into the cytoplasm, either from internal stores or from the external milieu, acting as triggers for the activation of a variety of intracellular processes. It has been reported that stimulation of f-MLP receptors results in production of the intracellular second messenger inositol triphosphate, which mobilizes calcium from intracellular stores [31]. Our experiments conformed to this general model: stimulation with f-MLP caused a marked but transient increase in [Ca2+]i. However, preincubation with Rosiglitazone caused elevated basal [Ca2+]i levels (Fig. 5), and a significantly larger increase in Ca2+-dependent fluorescence than in control cells after stimulation with f-MLP (Fig. 6). We suggest that the contrasting mode of the response to Rosiglitazone alone compared to that induced by f-MLP may reflect a distinct mechanism. It is noteworthy that the timescale of the former response resembles that of cells subjected to treatment with inhibitors of SERCA2b, the ÔhousekeepingÕ Ca2+ pump that is responsible for a sequestration of Ca2+ into the ER compensatory to non-specific leakage of Ca2+ into the cytoplasm [32]. Our demonstration of gradual increases in [Ca2+]i in cells treated with Rosiglitazone suggests that, rather than triggering a Ca2+ signalling cascade, the drug may non-specifically disrupt Ca2+ homeostasis in such a way as to cause Ca2+ ions to leak into the cytoplasm. Although some have suggested that cytoskeletal remodelling occurs independently of [Ca2+]i in leukocytes [26,33], others have reported that Ca2+ triggers actin depolymerisation in neutrophils and monocytes [34]. The mechanism by which Ca2+ controls depolymerisation may be via the activation of gelsolin, a 91-kDa glycoprotein that severs actin filaments and prevents filament elongation by capping fast-growing filamentous ends in a Ca2+-dependent manner [35]. Thus, the local increases in [Ca2+]i seen immediately after monocyte activation may fulfil a negative regulatory role in preventing excessive actin polymerisation in some locations. However, if Ca2+ homeostasis is disrupted by the use of ionophores, energy depletion, or Ca2+ pump inhibitors, the result is disintegration of the actin cytoskeleton due to large-scale Ca2+-dependent activation of gelsolin [15,36,37]. Hence, we observe that if cells are pretreated with thapsigargin, then their F-actin content is reduced below basal levels (74.2 ± 7.4%, p < 0.05, Fig. 1B). In contrast, the retention of a classical ‘‘Ca2+ spike’’ in f-MLP-stimulated cells after pretreatment with Rosiglitazone (Fig. 6), and indeed the drugÕs lack of cytotoxicity, suggests that Rosiglitazone does not produce such drastic changes. Instead, we suggest that treatment with Rosiglitazone causes a small non-specific increase in [Ca2+]i, and so may sensitise Ca2+ signalling mechanisms to chemotactic stimulation. As Ca2+ signalling in response to f-MLP is suppressed in diabetic
neutrophils [16], these data provide a mechanism by which treatment with Rosiglitazone may restore normal Ca2+ signalling to leukocytes from patients with diabetes. However, while incubation of monocytes with Rosiglitazone alone brought about significant increases in both Akt phosphorylation and resting [Ca2+]i, it must be emphasised that the drug alone did not stimulate actin polymerisation (Fig. 1B). Nevertheless, we hypothesise that, while not directly evoking cytoskeletal responses, by modulation of Akt and [Ca2+]i, Rosiglitazone suppresses the signalling events required to promote actin polymerisation. In support of this hypothesis, there is now a growing body of evidence that PPAR-c agonists exert anti-inflammatory effects through so-called ‘‘non-genomic’’ or ‘‘PPAR-c-independent’’ events, in which ligand-activated PPAR-c receptors (or possibly the ligands acting independently) may directly crosstalk with, and therefore modulate, multiple signalling pathways. For example, PPAR-c ligands have been shown to activate ERK1/2 and p38 MAPK in VSMCs and HT-29 colon cancer cells [9,40]. Moreover, the ability of PPAR-c ligands to influence the release of bioactive mediators from platelets, which lack nuclei, is further evidence of the non-genomic nature of at least some of the effects of PPAR-c ligands [41]. It remains to be seen whether the cytoskeletal remodelling effects we have observed in monocytes also occur in other cell types, but it may be relevant that the PPAR-c ligands, troglitazone and ciglitizone, can inhibit migration of endothelium cells via inhibition of leptin-dependent Akt signalling [30], while Knock et al. [39] have demonstrated that troglitazone (and to a lesser extent Rosiglitazone) can disrupt [Ca2+]i homeostasis in aortic myocytes. As the actin cytoskeleton of such cells is involved in the maintenance of a healthy vasculature, this supports the hypothesis that TZD drugs can act as modulators of the cytoskeletal remodelling involved in the aetiology of vascular complications of diabetes. However, it should be stressed that, while the modulation of actin remodelling could have clinically beneficial effects such as reducing cell rigidity in the microvasculature of patients with diabetes, such modulation could be deleterious in other contexts. For example, impaired neutrophil actin polymerisation has been shown to lead to reduced primary granule exocytosis and chemotaxis [16]. In conclusion, we hypothesise that Rosiglitazone interacts with signalling events downstream of occupancy of the f-MLP receptor (i.e., modulation of Akt phosphorylation and [Ca2+]i) to inhibit monocyte actin polymerisation in a non-genomic manner. This study demonstrates that TZD therapy not only influences the expression of genes involved in glucose and lipid metabolism, and the release of inflammatory mediators, but
N. Singh et al. / Biochemical and Biophysical Research Communications 333 (2005) 455–462
also modulates basic monocyte physiology, and that the true benefits of TZD therapy may be due to a combination of PPAR-c-independent and PPAR-c-dependent events. Our observations support the concept that TZD therapy may influence both macrovascular complications and microangiopathy through mechanisms independent of metabolic effects in diabetes.
References [1] S. Schroder, W. Palinski, G.W. Schmid-Schonbein, Activated monocytes and granulocytes, capillary nonperfusion, and neovascularization in diabetic retinopathy, Am. J. Pathol. 139 (1991) 81– 100. [2] K. Miyamoto, Y. Ogura, S. Kenmochi, Y. Honda, Role of leukocytes in diabetic microcirculatory disturbances, Microvasc. Res. 54 (1997) 43–48. [3] M.C. Mazzoni, G.W. Schmid-Schonbein, Mechanisms and consequences of cell activation in the microcirculation, Cardiovasc. Res. 32 (1996) 709–719. [4] E. Hu, J.B. Kim, P. Sarraf, B.M. Spiegelman, Inhibition of adipogenesis through MAP kinase-mediated phosphorylation of PPARgamma, Science 274 (1996) 2100–2103. [5] J. Plutzky, The vascular biology of atherosclerosis, Am. J. Med. 115 (Suppl. 8A) (2003) 55S–61S. [6] M. Ricote, A.C. Li, T.M. Willson, C.J. Kelly, C.K. Glass, The peroxisome proliferator-activated receptor-gamma is a negative regulator of macrophage activation, Nature 391 (1998) 79–82. [7] N. Marx, D. Walcher, N. Ivanova, K. Rautzenberg, A. Jung, R. Friedl, V. Hombach, C.R. de, G. Basta, M.P. Wautier, J.L. Wautiers, Thiazolidinediones reduce endothelial expression of receptors for advanced glycation end products, Diabetes 53 (2004) 2662–2668. [8] G. Hong, B. Davis, N. Khatoon, S.F. Baker, J. Brown, PPAR gamma-dependent anti-inflammatory action of rosiglitazone in human monocytes: suppression of TNF alpha secretion is not mediated by PTEN regulation, Biochem. Biophys. Res. Commun. 303 (2003) 782–787. [9] K. Takeda, T. Ichiki, T. Tokunou, N. Iino, A. Takeshita, 15Deoxy-delta 12,14-prostaglandin J2 and thiazolidinediones activate the MEK/ERK pathway through phosphatidylinositol 3kinase in vascular smooth muscle cells, J. Biol. Chem. 276 (2001) 48950–48955. [10] N. Marx, U. Schonbeck, M.A. Lazar, P. Libby, J. Plutzky, Peroxisome proliferator-activated receptor gamma activators inhibit gene expression and migration in human vascular smooth muscle cells, Circ. Res. 83 (1998) 1097–1103. [11] S. Goetze, F. Eilers, A. Bungenstock, U. Kintscher, P. Stawowy, F. Blaschke, K. Graf, R.E. Law, E. Fleck, M. Grafe, PPAR activators inhibit endothelial cell migration by targeting Akt, Biochem. Biophys. Res. Commun. 293 (2002) 1431–1437. [12] D. Chodniewicz, D.V. Zhelev, Chemoattractant receptor-stimulated F-actin polymerization in the human neutrophil is signaled by 2 distinct pathways, Blood 101 (2003) 1181–1184. [13] V. Vasioukhin, C. Bauer, M. Yin, E. Fuchs, Directed actin polymerization is the driving force for epithelial cell–cell adhesion, Cell 100 (2000) 209–219. [14] W. Siess, Molecular mechanisms of platelet activation, Physiol. Rev. 69 (1989) 58–178. [15] G.P. Downey, C.K. Chan, S. Trudel, S. Grinstein, Actin assembly in electropermeabilized neutrophils: role of intracellular calcium, J. Cell Biol. 110 (1990) 1975–1982.
461
[16] A. Advani, S.M. Marshall, T.H. Thomas, Impaired neutrophil store-mediated calcium entry in Type 2 diabetes, Eur. J. Clin. Invest. 34 (2004) 43–49. [17] P.L. Sonkin, S.F. Freedman, D. Needham, K.M. Rao, D.L. Hatchell, Pentoxifylline modulates deformability, F-actin content, and superoxide anion production of polymorphonuclear leukocytes from diabetic cats, Exp. Eye Res. 55 (1992) 831–838. [18] T.H. Howard, W.H. Meyer, Chemotactic peptide modulation of actin assembly and locomotion in neutrophils, J. Cell Biol. 98 (1984) 1265–1271. [19] J. Elsner, J. Norgauer, G.J. Dobos, A. Emmendorffer, E. Schopf, A. Kapp, J. Roesler, Flow cytometry reveals different lag times in rapid cytoplasmic calcium elevations in human neutrophils in response to N-formyl peptide, J. Cell. Physiol. 157 (1993) 637–643. [20] G. Grynkiewicz, M. Poenie, R.Y. Tsien, A new generation of Ca2+ indicators with greatly improved fluorescence properties, J. Biol. Chem. 260 (1985) 3440–3450. [21] J.E. Merritt, S.A. McCarthy, M.P. Davies, K.E. Moores, Use of fluo-3 to measure cytosolic Ca2+ in platelets and neutrophils. Loading cells with the dye, calibration of traces, measurements in the presence of plasma, and buffering of cytosolic Ca2+, Biochem. J. 269 (1990) 513–519. [22] G. Athanassiou, P. Matsouka, V. Kaleridis, Y. Missirlis, Deformability and filterability of white blood cell subpopulations. Evaluation of these parameters in the cell line HL-60 and in type II diabetes mellitus, Clin. Hemorheol. Microcirc. 22 (2000) 35–43. [23] C. Jiang, A.T. Ting, B. Seed, PPAR-gamma agonists inhibit production of monocyte inflammatory cytokines, Nature 391 (1998) 82–86. [24] A.W. Roberts, A. Thomas, A. Rees, M. Evans, Peroxisome proliferator-activated receptor-gamma agonists in atherosclerosis: current evidence and future directions, Curr. Opin. Lipidol. 14 (2003) 567–573. [25] J. Plutzky, The potential role of peroxisome proliferator-activated receptors on inflammation in type 2 diabetes mellitus and atherosclerosis, Am. J. Cardiol. 92 (2003) 34J–41J. [26] Y.D. Cui, O. Inanami, T. Yamamori, K. Niwa, H. Nagahata, M. Kuwabara, FMLP-induced formation of F-actin in HL60 cells is dependent on PI3-K but not on intracellular Ca2+, PKC, ERK or p38 MAPK, Inflamm. Res. 49 (2000) 684–691. [27] Y.I. Miller, D.S. Worrall, C.D. Funk, J.R. Feramisco, J.L. Witztum, Actin polymerization in macrophages in response to oxidized LDL and apoptotic cells: role of 12/15-lipoxygenase and phosphoinositide 3-kinase, Mol. Biol. Cell 14 (2003) 4196–4206. [28] A. Ptasznik, E.R. Prossnitz, D. Yoshikawa, A. Smrcka, A.E. Traynor-Kaplan, G.M. Bokoch, A tyrosine kinase signaling pathway accounts for the majority of phosphatidylinositol 3,4,5trisphosphate formation in chemoattractant-stimulated human neutrophils, J. Biol. Chem. 271 (1996) 25204–25207. [29] V. Cenni, A. Sirri, M. Riccio, G. Lattanzi, S. Santi, A. de Pol, N.M. Maraldi, S. Marmiroli, Targeting of the Akt/PKB kinase to the actin skeleton, Cell. Mol. Life Sci. 60 (2003) 2710–2720. [30] S. Goetze, A. Bungenstock, C. Czupalla, F. Eilers, P. Stawowy, U. Kintscher, C. Spencer-Hansch, K. Graf, B. Nurnberg, R.E. Law, E. Fleck, M. Grafe, Leptin induces endothelial cell migration through Akt, which is inhibited by PPARgamma-ligands, Hypertension 40 (2002) 748–754. [31] E.R. Prossnitz, O. Quehenberger, C.G. Cochrane, R.D. Ye, Signal transducing properties of the N-formyl peptide receptor expressed in undifferentiated HL60 cells, J. Immunol. 151 (1993) 5704–5715. [32] L.M. John, J.D. Lechleiter, P. Camacho, Differential modulation of SERCA2 isoforms by calreticulin, J. Cell Biol. 142 (1998) 963– 973. [33] F.A. al Mohanna, M.B. Hallett, Actin polymerization in neutrophils is triggered without a requirement for a rise in cytoplasmic Ca2+, Biochem. J. 266 (1990) 669–674.
462
N. Singh et al. / Biochemical and Biophysical Research Communications 333 (2005) 455–462
[34] R.C. May, L.M. Machesky, Phagocytosis and the actin cytoskeleton, J. Cell Sci. 114 (2001) 1061–1077. [35] H.L. Yin, J.H. Albrecht, A. Fattoum, Identification of gelsolin, a Ca2+-dependent regulatory protein of actin gel–sol transformation, and its intracellular distribution in a variety of cells and tissues, J. Cell Biol. 91 (1981) 901–906. [36] K. Kanno, Y. Sasaki, Smooth muscle gelsolin and a Ca2+sensitive contractile cell model, J. Cell. Physiol. 139 (1989) 58–67. [37] W. Kuhne, M. Besselmann, T. Noll, A. Muhs, H. Watanabe, H.M. Piper, Disintegration of cytoskeletal structure of actin filaments in energy-depleted endothelial cells, Am. J. Physiol. 264 (1993) H1599–H1608. [38] M.M. Meyer, K. Levin, T. Grimmsmann, N. Perwitz, A. Eirich, H. Beck-Nielsen, H.H. Klein, Troglitazone treatment increases protein kinase B phosphorylation in skeletal muscle of normo-
glycemic subjects at risk for the development of type 2 diabetes, Diabetes 51 (2002) 2691–2697. [39] G.A. Knock, S.K. Mishra, P.I. Aaronson, Differential effects of insulin-sensitizers troglitazone and rosiglitazone on ion currents in rat vascular myocytes, Eur. J. Pharmacol. 368 (1999) 103–109. [40] F. Chen, M. Wang, J.P. OÕConnor, M. He, T. Tripathi, L.E. Harrison, Phosphorylation of PPARgamma via active ERK1/2 leads to its physical association with p65 and inhibition of NFkappabeta, J. Cell. Biochem. 90 (2003) 732–744. [41] F. Akbiyik, D.M. Ray, K.F. Gettings, N. Blumberg, C.W. Francis, R.P. Phipps, Human bone marrow megakaryocytes and platelets express PPARgamma, and PPARgamma agonists blunt platelet release of CD40 ligand and thromboxanes, Blood 104 (2004) 1361–1368.