Modification of PI3K- and MAPK-Dependent Chemotaxis in Aortic Vascular Smooth Muscle Cells by Protein Kinase CII
http://www.100md.com
循环研究杂志 2005年第2期
the Department of Clinical Biochemistry and Metabolic Medicine, Queen’s University Belfast, Institute of Clinical Science, Royal Victoria Hospital, Belfast, UK
Department of Clinical Biochemistry, The Royal Group of Hospitals, Belfast, UK.
Abstract
Hyperglycemia increases expression of platelet-derived growth factor (PDGF)- receptor and potentiates chemotaxis to PDGF-BB in human aortic vascular smooth muscle cells (VSMCs) via PI3K and ERK/MAPK signaling pathways. The purpose of this study was to determine whether increased activation of protein kinase C (PKC) isoforms had a modulatory effect on the PI3K and ERK/MAPK pathways, control of cell adhesiveness, and movement. All known PKC isoforms were assessed but only PKC and PKCII levels were increased in 25 mmol/L glucose. However, only PKCII inhibition affected (decreased) PI3K pathway and MAPK pathway activities and inhibited PDGF- receptor upregulation in raised glucose, and specific MAPK inhibition was required to completely block the effect of glucose. In raised glucose conditions, activity of the ERK/MAPK pathway, PI3K pathway, and PKCII were all sensitive to aldose reductase inhibition. Chemotaxis to PDGF-BB (360 pmol/L), absent in 5 mmol/L glucose, was present in raised glucose and could be blocked by PKCII inhibition. Formation of lamellipodia was dependent on PI3K activation and filopodia on MAPK activation; both lamellipodia and filopodia were eliminated when PKCII was inhibited. FAK phosphorylation and cell adhesion were reduced by PI3K inhibition, and although MAPK inhibition prevented chemotaxis, it did not affect FAK phosphorylation or cell adhesiveness. In conclusion, chemotaxis to PDGF-BB in 25 mmol/L glucose is PKCII-dependent and requires activation of both the PI3K and MAPK pathways. Changes in cell adhesion and migration speed are mediated mainly through the PI3K pathway.
Key Words: atherosclerosis diabetes Akt PKC ERK
Introduction
Premature atherosclerosis occurs not only in diabetic subjects1 but also in those with impaired glucose tolerance who experience intermittent periods of elevated glucose.2 Atheroma formation involves migration of the vascular smooth muscle cells (VSMCs) into the subintimal space,3 and a change from a nonmotile, contractile phenotype to a proliferative, secretory phenotype. In vitro, several cell types including fibroblasts, monocytes, and endothelial cells express the platelet-derived growth factor (PDGF)- receptor and migrate to a physiological concentration of PDGF-BB if serum starved4; PDGF-BB is elevated in diabetes5 and in atherosclerosis.6 We have previously shown that aortic VSMC migration to PDGF-BB occurs in raised glucose conditions (>9 mmol/L glucose) without prior serum starvation.7 PDGF- receptor antibodies block this migration in vitro,7 and in an in vivo mouse model they reduce significantly the extent of atheroma formation.8
Cell movement can be directional (migration) or random. Moving cells polarize9; the leading edge becomes broad, forming a lamellipodium, and presents direction-sensing microspikules (filopodia), whereas the actin cytoskeleton organizes along the long axis of the cell to facilitate movement.10 Movement also requires the laying down at the front of the cell and removal at the back of small punctate structures, focal adhesions, that link the cell to its basement membrane and contain integrins and PDGF- receptors.11 Most analysis techniques for migration do not separate the speed and direction components or allow direct cell observation. Using the Dunn chemotaxis chamber, we have shown that VSMC exposed to raised glucose in a PDGF-BB gradient present morphological alterations associated with migration, with increased levels of f-actin and with filopodia at the leading edge.7,12
It is known that the serine/threonine kinase protein kinase C (PKC) can also affect VSMC chemotaxis.13 PKC is associated with PDGF- receptor upregulation and, like the Ras-Raf-MEK1/2-ERK1/2 MAPK and PI3K pathways, is sensitive to elevated glucose levels.7,14 PKC consists of approximately 12 isoforms in three subclasses: conventional (cPKC-, -I, -II, and -), novel (nPKC-, -, -, -, and -e?, and atypical (aPKC- and -/). Conventional isoforms are Ca2+-dependent and require diacylglycerol (DAG) and phosphatidylserine for their activation, novel isoforms are Ca2+-independent, and atypicals require only the presence of phosphatidylserine.15 Elevated glucose levels produce increased flux through the sorbitol pathway with resultant de novo synthesis of DAG, which activates PKC isoforms16 and Ras.17 DAG is present in human aortic VSMCs, and its level increases in raised glucose conditions with a peak being reached by 3 days.18 The rate-limiting enzyme of the sorbitol pathway, aldose reductase, is widely distributed in those mammalian tissues susceptible to diabetic complications and its inhibition prevents complications in galactosemic animal models.19 The response to aldose reductase inhibitors (ARI) in clinical trials has been disappointing,20 indicating that indirect effects of long-term exposure to glucose (advanced glycation end-product formation or o-glycosylation)21 bypass this pathway’s control.
We have recently shown that glucose-potentiated VSMC chemotaxis and the associated morphological changes are dependent on the PI3K and Ras/ERK MAPK pathways and also on cross-talk between these pathways12; both pathways also control PDGF- receptor upregulation and migration in raised glucose.7 In view of that, the object of this investigation was to determine at which point(s) PKC modifies glucose-potentiated PI3K and/or MAPK activity during human aortic VSMC migration (direction, speed and adhesion) induced by PDGF-BB.
Materials and Methods
Chemicals
The following chemicals were used: D-glucose and L-glucose (BDH, UK); recombinant human PDGF-BB (Insight Biotechnology Ltd, UK); LY294002, wortmannin, rapamycin, and rhodamine-phalloidin (Sigma, UK); PD98059, calphostin C and H7 (Calbiochem); farnesyltransferase inhibitor (FTI) H-D-Trp-D-Met-p-chloro-D-Phe-Gla-NH2 (Bachem, UK); alrestatin (Tocris, UK); active Akt1, active PKCII, and PKC lipid activator (Upstate); cognate peptides to PKC isoforms (Santa Cruz Biotechnology); and LY379196 and sorbinil (gifts from Eli Lilly and Pfizer, respectively).
For an expanded Materials and Methods, see the online data supplement available at http://circres.ahajournals.org.
Results
PKCII Is Elevated in VSMC in 25 mmol/L Glucose After 24 Hours
All PKC isoforms except PKC and PKCe?were identified in human primary aortic VSMCs (Table 1). After 24 hour in 25 mmol/L glucose, the protein level of only two isoforms PKCII (+55%: P<0.01 versus 5 mmol/L glucose) and PKC (+20%: P<0.05 versus 5 mmol/L glucose) increased significantly. The level of serine/threonine phosphorylation of PKCII (+94%: P<0.01 versus 5 mmol/L glucose) and PKC (+20%: P<0.05 versus 5 mmol/L glucose) also increased significantly after 24 hours in 25 mmol/L glucose. Phosphorylation of PKCII showed a significant increase only in the membranous fraction (Figure 1A: +74%, P<0.001 versus 5 mmol/L glucose). The small increase in PKC phosphorylation occurred across both the cytoplasmic and membranous cellular fractions (see online Figure B available in the online data supplement). Activation of PKCII in an isoform specific kinase assay (Figure 1B) showed a significant increase in activity in 25 mmol/L glucose after 24 hours in the membranous fraction (+160%, P<0.01 versus 5 mmol/L glucose alone).
Chemotaxis to PDGF-BB in 25 mmol/L Glucose Is Sorbitol Pathway and PKCII-Dependent
To determine whether the PKC isoforms elevated in 25 mmol/L glucose had a functional effect on VSMCs, chemotaxis assays were undertaken. Chemotaxis without serum starvation does not occur to a physiological level of PDGF-BB (360 pmol/L) in 5 mmol/L glucose (P=NS) but does occur in 25 mmol/L glucose (P<0.01: Table 2). The aldose reductase inhibitors (ARI) sorbinil and alrestatin, which prevent de novo synthesis of DAG, prevented chemotaxis to PDGF-BB. The general PKC inhibitor calphostin C and the specific PKCII inhibitor LY379196 used at a concentration (200 nmol/L) where PKCI was unaffected blocked chemotaxis (see online Figure C and online Table A). Microinjection experiments showed that only affinity-purified antibodies to the PKCII isoform blocked chemotaxis in 25 mmol/L glucose (Table 2); this could be prevented by preincubating the antibody with a molar excess of its cognate peptide.
PKCII in PDGF-BB Treated VSMCs: Activation/Phosphorylation Is Sorbitol PathwayeCDependent, Protein Level Is mTOR PathwayeCDependent
Chemotaxis to PDGF-BB in 25 mmol/L glucose is PKCII- and sorbitol pathway-dependent. We have previously shown that glucose-potentiated chemotaxis depends on PI3K and MAPK pathways7; therefore, the relationship of these pathways to PKCII regulation was examined.
Global PKCII Phosphorylation
To examine the degree of phosphorylation, serine/threonine phosphorylation of the protein was measured (Figure 1C). Phosphorylation was unaffected by inhibitors in 5 mmol/L glucose, but increased after 24 hours in 25 mmol/L glucose (+94%, P<0.001 versus 5 mmol/L glucose alone). Reductions in phosphorylation were found in 25 mmol/L glucose after 24 hours with PI3K, mTOR, and sorbitol pathway inhibitors (P<0.01 versus 25 mmol/L glucose alone). Similar results to those above were found when the degree of phosphorylation on the Ser660 residue of PKCII (located at the cofactor binding site)22 was examined with a significant increase after 24 hours in 25 mmol/L glucose (+103%, P<0.001 versus 5 mmol/L glucose alone; Figure 1D).
PKCII Protein Level
In 5 mmol/L glucose, the total protein level depended on the PI3K-related mTOR pathway because mTOR inhibition (rapamycin) decreased it by 24% (P<0.01: Figure 1E). PKCII in 5 mmol/L glucose was unaltered by sorbinil (ARI), PD98059 (MEK1/2 inhibitor), or by the farnesyltransferase inhibitor (FTI), which prevents prenylation of Ras (MAPK pathway). The increased level of PKCII in 25 mmol/L glucose (+55%, versus 5 mmol/L glucose, P<0.01) was under similar control but sorbinil or alrestatin (not shown) were also able to decrease the total PKCII level and the effects of PI3K inhibition were significant.
PKC Inhibition Prevents 25 mmol/L Glucose-Potentiated PDGF- Receptor Upregulation in PDGF-BBeCTreated VSMCs
Investigation was made into possible PKC involvement in the pathways responsible for PDGF- receptor upregulation in conditions of 25 mmol/L glucose plus 360 pmol/L PDGF-BB (Figure 2A). In 5 mmol/L glucose, MAPK and PI3K dependence were demonstrated (P<0.05 versus untreated for both inhibitors). Receptor upregulation associated with 25 mmol/L glucose for 24 hours (+101%, P<0.01 versus 5 mmol/L glucose alone) was decreased significantly by PI3K, MAPK, general PKC, PKCII, or aldose reductase inhibition. However, only MAPK or aldose reductase inhibition returned the receptor back to the 5 mmol/L glucose level (P<0.01 versus 25 mmol/L and P=NS versus 5 mmol/L glucose alone). That PKC inhibition controlled the receptor at the transcriptional level was confirmed by RT-PCR (Figure 2B). The significant increase in receptor mRNA found in 25 mmol/L glucose (+83%, P<0.01) was inhibited partially by general PKC inhibition but totally by PKCII inhibition (P<0.01 versus 25 mmol/L glucose alone).
PKC Inhibition Prevents 25 mmol/L Glucose-Potentiated Akt and ERK Activation in PDGF-BBeCTreated VSMCs
Pathways upregulating the PDGF- receptor in 25 mmol/L glucose include PKC, PI3K, and MAPK but the relationship between these pathways has not yet been determined.
PI3K Pathway
Akt (a downstream target of PI3K) phosphorylation increased after 24 hours in 25 mmol/L glucose (+54%, P<0.05 versus 5 mmol/L glucose; Figure 3A). Calphostin C caused a modest (nonsignificant) decrease in the 25 mmol/L glucose-potentiated phosphorylation level but LY379196 and the ARI both reduced it significantly (P<0.05 versus 25 mmol/L glucose alone). However, only the ARI decreased the level to that found in 5 mmol/L glucose. To confirm that the weak effect of PKCII inhibition on Akt phosphorylation was significant with respect to activity, an Akt kinase assay was undertaken (Figure 3B). LY379196 did not affect Akt activation in 5 mmol/L glucose but reduced the ability of Akt to phosphorylate its target molecule, GSK-3, in 25 mmol/L glucose (+81% in 25 mmol/L glucose versus 5 mmol/L glucose alone, P<0.01; LY379196 treatment, +45%, P<0.05 versus 5 or 25 mmol/L glucose alone). PKC appears to act upstream of PI3K in the pathways leading to PDGF- receptor upregulation in 25 mmol/L glucose.
MAPK Pathway
ERK1/2 (a downstream target in the MAPK pathway) phosphorylation increased in 25 mmol/L glucose after 24 hours (+95%, P<0.01 versus 5 mmol/L glucose; Figure 3C). Calphostin C and LY379196 had a modest (nonsignificant) effect on the 25 mmol/L glucose-potentiated increase in phosphorylation, but sorbinil reduced the level (P<0.05 versus 25 mmol/L glucose alone). To identify if LY379196 modulated ERK phosphorylation, an ERK kinase assay was undertaken (Figure 3D). LY379196 did not affect ERK activation in 5 mmol/L glucose but reduced the ability of ERK to phosphorylate its target molecule ELK-1 in 25 mmol/L glucose (+189% in 25 mmol/L glucose versus 5 mmol/L glucose alone, P<0.01; LY379196 treatment, P<0.05 versus 5 or 25 mmol/L glucose alone). It would appear that PKCII acts upstream of ERK/MAPK in the pathways leading to PDGF- receptor upregulation in 25 mmol/L glucose.
To confirm that ERK phosphorylation was a downstream target of PKCII and Akt activation, microinjection experiments were undertaken (Figure 3E and Figure 3F). Injection of activated Akt caused a significant increase in ERK phosphorylation in 5 mmol/L glucose conditions. In 25 mmol/L glucose and the presence of alrestatin (20 e蘭ol/L), injection of activated Akt (Figure 3E) could partially restore the level of ERK phosphorylation (P<0.05 versus 5 mmol/L or 25 mmol/L glucose). In the presence of the PKCII inhibitor, LY379196 (200 nmol/L), injection of activated Akt could totally recover the level of ERK phosphorylation (P=NS versus 25 mmol/L glucose; Figure 3E). Injection of activated PKCII significantly increased ERK phosphorylation in both 5 and 25 mmol/L conditions (Figure 3F). The effect in 25 mmol/L glucose was partly lost in the presence of the ARI and totally lost in the presence of the MAPK inhibitor (Figure 3F). The microinjection of activated PKCII was sufficient to induce chemotaxis in 5 mmol/L glucose conditions (without prior serum starvation) and restored chemotaxis lost because of aldose reductase inhibition in 25 mmol/L glucose conditions (Table 2).
PKCII Inhibition Prevents 25 mmol/L Glucose-Potentiated Filopodia Formation
Migrating cells show characteristic changes in their ultrastructure. Membrane ruffles (lamellipodia) distribute around the rim of a nonmigrating cell: this is seen in a representative micrograph (Figure 4A) for a cell in 5 mmol/L glucose within a 360 pmol/L PDGF-BB gradient. In 25 mmol/L glucose, the cell polarizes with a lamellipodium localizing along the front and prominent actin bundles stretching along the long axis of the cell (Figure 4B). Characteristic filopodia (linked with direction sensing) protrude from the lamellipodium (see inset). PDGF- receptors are detected along filopodia in 25 mmol/L glucose (Figure 4C). With PKCII inhibition filopodia, lamellipodium and f-actin alignment are lost (Figure 4D).
PKN Signaling to Lamellipodia Is PKCII-Dependent in 25 mmol/L Glucose
Protein kinase N (PKN) is involved in formation of lamellipodia.23 We have recently shown that MAPK inhibitors affect filopodia but not lamellipodia formation in VSMCs.12 To clarify the role PKC has on lamellipodia formation, the effect of the pathway inhibitors on PKN was assessed (Figure 5A). The Ser/Thr phosphorylation of PKN was significantly increased in 25 mmol/L glucose (+69%, P<0.01 versus 5 mmol/L glucose). MAPK inhibition had no effect but PI3K, PKCII, or aldose reductase inhibition significantly decreased the activation level (P<0.01 versus 25 mmol/L glucose alone for all).
PKCII-Dependent Pathways Control Cell Speed in 25 mmol/L Glucose
Chemotaxis to PDGF-BB (360 pmol/L) was demonstrated in the presence of 25 mmol/L glucose. Migration also has a speed component. Speed in 25 mmol/L glucose is significantly reduced compared with matched cells (same source and passage) set up in 5 mmol/L glucose (random movement) in the chamber (Table 2). Speed was unaffected in 25 mmol/L glucose by MAPK inhibition (by chemical inhibitor or microinjected antibody; see online Table A), or by microinjection of conventional PKC isoform antibodies other than PKCII. PKCII isoform microinjection or LY379196 exposure prevented chemotaxis and restored speed to that found in 5 mmol/L glucose.
PKCII-Dependent Pathways Control Adhesion in 25 mmol/L Glucose
Changes in cellular adhesion may account for the decreased speed found in 25 mmol/L glucose; therefore, an adhesion assay was undertaken (Figure 5B). In 5 mmol/L glucose, MAPK (PD98059 or FTI), PKCII, or aldose reductase inhibition did not alter adhesion level, but PI3K inhibition caused a significant decrease in adhesion (LY294002, eC48%; wortmannin, eC42%; P<0.05 versus 5 mmol/L glucose alone). In 25 mmol/L glucose, adhesiveness increased significantly (+66%, P<0.05 versus 5 mmol/L glucose). MAPK inhibition had no effect, whereas PI3K inhibition showed a similar percentage reduction (LY294002, eC48%; wortmannin, eC54%), as found in 5 mmol/L glucose. PKCII inhibition and ARI, ineffective in 5 mmol/L glucose, significantly reduced adhesion in 25 mmol/L glucose. Similar results to those in Figure 5B were found when phosphospecific antibodies were microinjected into cells subsequently set up in an adhesion assay (Figure 5C). Again a pPKCII dependency was found in 25 mmol/L glucose conditions for control of adhesion. An antibody to phosphorylated FAK (a known component of focal adhesions) inhibited adhesion in both 5 mmol/L and 25 mmol/L glucose conditions. Microinjection of pAkt did not alter adhesion, indicating that this branch of the PI3K pathway is not involved in adhesion control (Figure 5C).
PKCII-Dependent Pathways Signal to Focal Adhesions in 25 mmol/L Glucose
Cellular adhesion is mainly controlled through focal adhesions, and focal adhesion kinase (FAK) can act as a marker of intracellular signaling to the substratum. The level of FAK phosphorylation increased in 25 mmol/L glucose (+69%, P<0.05 versus 5 mmol/L glucose; Figure 5D). MAPK inhibition had no effect on 25 mmol/L glucose-potentiated FAK phosphorylation, but the PI3K, PKCII, and aldose reductase inhibitors all restored it to 5 mmol/L glucose levels (P<0.05 versus 25 mmol/L glucose alone; Figure 5D). A representative micrograph is shown (Figure 4E) for a cell stained with phospho-FAK antibody (green) in 25 mmol/L glucose. The phospho-FAK(Tyr576/577) staining was punctate in nature and localized mainly to the cell membrane: this distribution is in keeping with that expected for focal adhesions. Microinjection of cells in 25 mmol/L glucose with affinity-purified phospho-FAK antibody did not alter PDGF- receptor level (online Figure D) nor prevent chemotaxis to PDGF-BB (Table 2); however, the cell speed (in 25 mmol/L glucose) was restored to that found in 5 mmol/L glucose.
Discussion
This study has revealed significant differences in both the glucose- and PKC-dependency of PI3K and MAPK activation in human VSMCs (Figure 6). Most PKC isoforms were detected in primary explant-derived human aortic VSMCs including PKC-, -I, -II, -, -, and -. The time frame for the 25 mmol/L glucose experiments was short in terms of the chronicity of diabetes and was associated with upregulation of the conventional isoforms of PKC (PKC- and -II). We found that the total level of PKCII was regulated by mTOR-dependent pathways in VSMCs; mTOR acts on the serine/arginine-rich splicing factor (SRp40) that controls splicing of the PKC gene, altering production away from PKCI toward PKCII.24 The PI3K/Akt pathway regulates mTOR-dependent signaling.25 Aldose reductase inhibitors prevent increased flux through the sorbitol pathway, reduce PI3K/Akt activation, and mTOR-dependent signaling. PKCII is upregulated in many cell types in response to glucose26; exceptionally, one group has reported small decreases in PKCII activity in rat aorta,27,28 but cell cycle synchronization through serum starvation in those experiments meant that the S phaseeCdependent decrease in PKCII28 may have affected the results. In the present study in human aortic VSMCs, both PKC and PKCII were activated, but only PKCII demonstrated a major increase in the membrane fraction and was involved in glucose-potentiated chemotaxis.
We have previously shown that regulation of the PDGF- receptor depends on both the PI3K and MAPK pathways7 with mainly, but not exclusively, PI3K signaling to the MAPK pathway.12 Aldose reductase inhibition and blockade of PKCII prevent PDGF- receptor increases in rat VSMCs14; in this study, we confirm the effect in human aortic VSMCs and demonstrate that receptor protein expression is controlled at the transcriptional level. We demonstrated that PKCII inhibition suppresses, but does not abolish, activation of Akt and ERK in high glucose; PKCII inhibition has no effect on ERK activity in normal glucose (not shown). DAG can activate the MAPK pathway directly,17 and in our experiments, sorbitol pathway inhibition or MAPK inhibition (upstream and downstream of DAG, respectively) removed all of the high glucose effect on PDGF- receptor protein and chemotaxis. Under similar glucose conditions, migration of coronary VSMCs was also sorbitol pathway-dependent.13 Here, we show that high glucose induces both PKCII-independent and PKCII-dependent signaling to the MAPK pathway. PKCII-dependent signaling to MAPK occurs indirectly via PKCII-stimulated phosphorylation in the PI3K/Akt pathway with subsequent cross-talk to the ERK/MAPK pathway.12 The PKCII-independent signaling involves direct DAG activation of Ras upstream in the ERK/MAPK pathway bypassing both PKCII and the PI3K/Akt pathway.
Previously, we showed that MAPK pathway inhibition reduces filopodia but not lamellipodia formation12; in this study, PKCII inhibition affects both. The effect of PKCII on MAPK activation (via PI3K) may account for its effect on filopodia. To confirm that lamellipodia formation was PKCII/PI3K-dependent, we measured PKN activity, which is involved in lamellipodia formation,23 and demonstrated it was PKCII/PI3K-dependent and MAPK-independent. Speed of movement was higher in 5 mmol/L glucose (random movement) than in 25 mmol/L glucose (migration), provided cells were age- and source-matched. With PKCII inhibition (chemical inhibitor or microinjected antibody), migration was abolished and speed restored to that for 5 mmol/L glucose. Specific inhibitors of the MAPK pathway (PD98059 or microinjected anti-pERK antibody) prevented migration in high glucose without altering speed, indicating that speed is controlled at a higher point in the pathways. Cell adhesion in 25 mmol/L glucose was regulated in a similar manner to speed with PKCII inhibition reducing adhesion to that found with 5 mmol/L glucose levels, and MAPK inhibition having no effect. Focal adhesions, which control inside to outside cell signaling, contain PDGF- receptors and integrins linked via their cytoplasmic components to signaling molecules such as FAK.29 In this study, the phosphorylation site to which the FAK antibody was directed depends on PI3K signaling.30 We demonstrated that in 25 mmol/L glucose this signaling to FAK can be modified by input from PKCII. As demonstrated by phospho-FAK antibody microinjection, it is this PKCII/PI3K-dependent signaling to FAK that controls cell speed by altering adhesion (see Figure 6).
In summary, we have demonstrated that in glucose conditions that occur intermittently in diabetes, human aortic VSMCs migrate in a concentration of PDGF-BB such as is present in atherosclerotic plaques. This glucose-potentiated migration is sorbitol pathway-dependent for activation of the ERK/MAPK pathway and for PKCII signaling to the PI3K pathway. We have shown in this study that in addition to PKCII-independent signaling to the MAPK pathway from increased flux through the sorbitol pathway, there is PKCII-dependent signaling to MAPK. Increases in PDGF- receptor level depend on PKCII signaling to both the PI3K and ERK/MAPK pathways in these conditions. PKCII influence on PI3K leads to alterations in adhesion, FAK phosphorylation, cell speed, and lamellipodia formation, but the ERK/MAPK pathway is not involved in these processes.
Acknowledgments
This work was supported by a project grant from the Research and Development Office, Department of Health, Social Services, and Public Safety, Northern Ireland. LY379196 was a generous gift from Eli Lilly. We thank P. Anderson (for practical assistance and manuscript comments) and W.E. Allen (for manuscript comments).
References
Beckman JA, Creager MA, Libby P. Diabetes and atherosclerosis: epidemiology, pathophysiology, and management. JAMA. 2002; 287: 2570eC2581.
Ceriello A. Impaired glucose tolerance and cardiovascular disease: the possible role of post-prandial hyperglycemia. Am Heart J. 2004; 147: 803eC807.
Schwartz SM. Smooth muscle migration in atherosclerosis and restenosis. J Clin Invest. 1997; 100: S87eCS89.
Waltenberger J. Modulation of growth factor action: implications for the treatment of cardiovascular diseases. Circulation. 1997; 96: 4083eC4094.
Kawano M, Koshikawa T, Kanzaki T, Morisaki N, Saito Y, Yoshida S. Diabetes mellitus induces accelerated growth of aortic smooth muscle cells: association with overexpression of PDGF beta-receptors. Eur J Clin Invest. 1993; 23: 84eC90.
Ross R. The pathogenesis of atherosclerosis: a perspective for the 1990s. Nature. 1993; 362: 801eC809.
Campbell M, Allen WE, Silversides JA, Trimble ER. Glucose-induced phosphatidylinositol 3-kinase and mitogen-activated protein kinase-dependent upregulation of the platelet-derived growth factor-beta receptor potentiates vascular smooth muscle cell chemotaxis. Diabetes. 2003; 52: 519eC526.
Sano H, Sudo T, Yokode M, Murayama T, Kataoka H, Takakura N, Nishikawa S, Nishikawa SI, Kita T. Functional blockade of platelet-derived growth factor receptor-beta but not of receptor-alpha prevents vascular smooth muscle cell accumulation in fibrous cap lesions in apolipoprotein EeCdeficient mice. Circulation. 2001; 103: 2955eC2960.
Servant G, Weiner OD, Herzmark P, Balla T, Sedat JW, Bourne HR. Polarization of chemoattractant receptor signaling during neutrophil chemotaxis. Science. 2000; 287: 1037eC1040.
Yamboliev IA, Gerthoffer WT. Modulatory role of ERK MAPK-caldesmon pathway in PDGF-stimulated migration of cultured pulmonary artery SMCs. Am J Physiol Cell Physiol. 2001; 280: C1680eCC1688.
Schneller M, Vuori K, Ruoslahti E. Alphavbeta3 integrin associates with activated insulin and PDGFbeta receptors and potentiates the biological activity of PDGF. EMBO J. 1997; 16: 5600eC5607.
Campbell M, Allen WE, Sawyer C, Vanhaesebroeck B, Trimble ER. Glucose-potentiated chemotaxis in human vascular smooth muscle is dependent on cross-talk between the PI3K and MAPK signaling pathways. Circ Res. 2004; 95: 380eC388.
Yasunari K, Kohno M, Kano H, Minami M, Yoshikawa J. Aldose reductase inhibitor improves insulin-mediated glucose uptake and prevents migration of human coronary artery smooth muscle cells induced by high glucose. Hypertension. 2000; 35: 1092eC1098.
Nakamura J, Kasuya Y, Hamada Y, Nakashima E, Naruse K, Yasuda Y, Kato K, Hotta N. Glucose-induced hyperproliferation of cultured rat aortic smooth muscle cells through polyol pathway hyperactivity. Diabetologia. 2001; 44: 480eC487.
Idris I, Gray S, Donnelly R. Protein kinase C activation: isozyme-specific effects on metabolism and cardiovascular complications in diabetes. Diabetologia. 2001; 44: 659eC673.
Williamson JR, Chang K, Frangos M, Hasan KS, Ido Y, Kawamura T, Nyengaard JR, van den EM, Kilo C, Tilton RG. Hyperglycemic pseudohypoxia and diabetic complications. Diabetes. 1993; 42: 801eC813.
Ebinu JO, Bottorff DA, Chan EY, Stang SL, Dunn RJ, Stone JC. RasGRP, a Ras guanyl nucleotide-releasing protein with calcium- and diacylglycerol-binding motifs. Science. 1998; 280: 1082eC1086.
Lee IK, Koya D, Ishi H, Kanoh H, King GL. d-Alpha-tocopherol prevents the hyperglycemia induced activation of diacylglycerol (DAG)-protein kinase C (PKC) pathway in vascular smooth muscle cell by an increase of DAG kinase activity. Diabetes Res Clin Pract. 1999; 45: 183eC190.
Collier A, Small M. The role of the polyol pathway in diabetes mellitus. Br J Hosp Med. 1991; 45: 38eC40.
Oka M, Kato N. Aldose reductase inhibitors. J Enzyme Inhib. 2001; 16: 465eC473.
Aronson D, Rayfield EJ. How hyperglycemia promotes atherosclerosis: molecular mechanisms. Cardiovasc Diabetol. 2002; 1: 1eC10.
Edwards JL, Apicella MA. The Molecular Mechanisms Used by Neisseria gonorrhoeae To Initiate Infection Differ between Men and Women. Clin Microbiol Rev. 2004; 17: 965eC981.
Dong LQ, Landa LR, Wick MJ, Zhu L, Mukai H, Ono Y, Liu F. Phosphorylation of protein kinase N by phosphoinositide-dependent protein kinase-1 mediates insulin signals to the actin cytoskeleton. Proc Natl Acad Sci U S A. 2000; 97: 5089eC5094.
Patel NA, Chalfant CE, Watson JE, Wyatt JR, Dean NM, Eichler DC, Cooper DR. Insulin regulates alternative splicing of protein kinase C beta II through a phosphatidylinositol 3-kinase-dependent pathway involving the nuclear serine/arginine-rich splicing factor, SRp40, in skeletal muscle cells. J Biol Chem. 2001; 276: 22648eC22654.
Oudit GY, Sun H, Kerfant BG, Crackower MA, Penninger JM, Backx PH. The role of phosphoinositide-3 kinase and PTEN in cardiovascular physiology and disease. J Mol Cell Cardiol. 2004; 37: 449eC471.
Way KJ, Katai N, King GL. Protein kinase C and the development of diabetic vascular complications. Diabet Med. 2001; 18: 945eC959.
Patel NA, Chalfant CE, Yamamoto M, Watson JE, Eichler DC, Cooper DR. Acute hyperglycemia regulates transcription and posttranscriptional stability of PKCbetaII mRNA in vascular smooth muscle cells. FASEB J. 1999; 13: 103eC113.
Yamamoto M, Acevedo-Duncan M, Chalfant CE, Patel NA, Watson JE, Cooper DR. Acute glucose-induced downregulation of PKC-betaII accelerates cultured VSMC proliferation. Am J Physiol Cell Physiol. 2000; 279: C587eCC595.
Schaller MD. Biochemical signals and biological responses elicited by the focal adhesion kinase. Biochim Biophys Acta. 2001; 1540: 1eC21.
Sundberg LJ, Galante LM, Bill HM, Mack CP, Taylor JM. An endogenous inhibitor of focal adhesion kinase blocks Rac1/JNK but not Ras/ERK-dependent signaling in vascular smooth muscle cells. J Biol Chem. 2003; 278: 29783eC29791.(Malcolm Campbell, Elisabe)
Department of Clinical Biochemistry, The Royal Group of Hospitals, Belfast, UK.
Abstract
Hyperglycemia increases expression of platelet-derived growth factor (PDGF)- receptor and potentiates chemotaxis to PDGF-BB in human aortic vascular smooth muscle cells (VSMCs) via PI3K and ERK/MAPK signaling pathways. The purpose of this study was to determine whether increased activation of protein kinase C (PKC) isoforms had a modulatory effect on the PI3K and ERK/MAPK pathways, control of cell adhesiveness, and movement. All known PKC isoforms were assessed but only PKC and PKCII levels were increased in 25 mmol/L glucose. However, only PKCII inhibition affected (decreased) PI3K pathway and MAPK pathway activities and inhibited PDGF- receptor upregulation in raised glucose, and specific MAPK inhibition was required to completely block the effect of glucose. In raised glucose conditions, activity of the ERK/MAPK pathway, PI3K pathway, and PKCII were all sensitive to aldose reductase inhibition. Chemotaxis to PDGF-BB (360 pmol/L), absent in 5 mmol/L glucose, was present in raised glucose and could be blocked by PKCII inhibition. Formation of lamellipodia was dependent on PI3K activation and filopodia on MAPK activation; both lamellipodia and filopodia were eliminated when PKCII was inhibited. FAK phosphorylation and cell adhesion were reduced by PI3K inhibition, and although MAPK inhibition prevented chemotaxis, it did not affect FAK phosphorylation or cell adhesiveness. In conclusion, chemotaxis to PDGF-BB in 25 mmol/L glucose is PKCII-dependent and requires activation of both the PI3K and MAPK pathways. Changes in cell adhesion and migration speed are mediated mainly through the PI3K pathway.
Key Words: atherosclerosis diabetes Akt PKC ERK
Introduction
Premature atherosclerosis occurs not only in diabetic subjects1 but also in those with impaired glucose tolerance who experience intermittent periods of elevated glucose.2 Atheroma formation involves migration of the vascular smooth muscle cells (VSMCs) into the subintimal space,3 and a change from a nonmotile, contractile phenotype to a proliferative, secretory phenotype. In vitro, several cell types including fibroblasts, monocytes, and endothelial cells express the platelet-derived growth factor (PDGF)- receptor and migrate to a physiological concentration of PDGF-BB if serum starved4; PDGF-BB is elevated in diabetes5 and in atherosclerosis.6 We have previously shown that aortic VSMC migration to PDGF-BB occurs in raised glucose conditions (>9 mmol/L glucose) without prior serum starvation.7 PDGF- receptor antibodies block this migration in vitro,7 and in an in vivo mouse model they reduce significantly the extent of atheroma formation.8
Cell movement can be directional (migration) or random. Moving cells polarize9; the leading edge becomes broad, forming a lamellipodium, and presents direction-sensing microspikules (filopodia), whereas the actin cytoskeleton organizes along the long axis of the cell to facilitate movement.10 Movement also requires the laying down at the front of the cell and removal at the back of small punctate structures, focal adhesions, that link the cell to its basement membrane and contain integrins and PDGF- receptors.11 Most analysis techniques for migration do not separate the speed and direction components or allow direct cell observation. Using the Dunn chemotaxis chamber, we have shown that VSMC exposed to raised glucose in a PDGF-BB gradient present morphological alterations associated with migration, with increased levels of f-actin and with filopodia at the leading edge.7,12
It is known that the serine/threonine kinase protein kinase C (PKC) can also affect VSMC chemotaxis.13 PKC is associated with PDGF- receptor upregulation and, like the Ras-Raf-MEK1/2-ERK1/2 MAPK and PI3K pathways, is sensitive to elevated glucose levels.7,14 PKC consists of approximately 12 isoforms in three subclasses: conventional (cPKC-, -I, -II, and -), novel (nPKC-, -, -, -, and -e?, and atypical (aPKC- and -/). Conventional isoforms are Ca2+-dependent and require diacylglycerol (DAG) and phosphatidylserine for their activation, novel isoforms are Ca2+-independent, and atypicals require only the presence of phosphatidylserine.15 Elevated glucose levels produce increased flux through the sorbitol pathway with resultant de novo synthesis of DAG, which activates PKC isoforms16 and Ras.17 DAG is present in human aortic VSMCs, and its level increases in raised glucose conditions with a peak being reached by 3 days.18 The rate-limiting enzyme of the sorbitol pathway, aldose reductase, is widely distributed in those mammalian tissues susceptible to diabetic complications and its inhibition prevents complications in galactosemic animal models.19 The response to aldose reductase inhibitors (ARI) in clinical trials has been disappointing,20 indicating that indirect effects of long-term exposure to glucose (advanced glycation end-product formation or o-glycosylation)21 bypass this pathway’s control.
We have recently shown that glucose-potentiated VSMC chemotaxis and the associated morphological changes are dependent on the PI3K and Ras/ERK MAPK pathways and also on cross-talk between these pathways12; both pathways also control PDGF- receptor upregulation and migration in raised glucose.7 In view of that, the object of this investigation was to determine at which point(s) PKC modifies glucose-potentiated PI3K and/or MAPK activity during human aortic VSMC migration (direction, speed and adhesion) induced by PDGF-BB.
Materials and Methods
Chemicals
The following chemicals were used: D-glucose and L-glucose (BDH, UK); recombinant human PDGF-BB (Insight Biotechnology Ltd, UK); LY294002, wortmannin, rapamycin, and rhodamine-phalloidin (Sigma, UK); PD98059, calphostin C and H7 (Calbiochem); farnesyltransferase inhibitor (FTI) H-D-Trp-D-Met-p-chloro-D-Phe-Gla-NH2 (Bachem, UK); alrestatin (Tocris, UK); active Akt1, active PKCII, and PKC lipid activator (Upstate); cognate peptides to PKC isoforms (Santa Cruz Biotechnology); and LY379196 and sorbinil (gifts from Eli Lilly and Pfizer, respectively).
For an expanded Materials and Methods, see the online data supplement available at http://circres.ahajournals.org.
Results
PKCII Is Elevated in VSMC in 25 mmol/L Glucose After 24 Hours
All PKC isoforms except PKC and PKCe?were identified in human primary aortic VSMCs (Table 1). After 24 hour in 25 mmol/L glucose, the protein level of only two isoforms PKCII (+55%: P<0.01 versus 5 mmol/L glucose) and PKC (+20%: P<0.05 versus 5 mmol/L glucose) increased significantly. The level of serine/threonine phosphorylation of PKCII (+94%: P<0.01 versus 5 mmol/L glucose) and PKC (+20%: P<0.05 versus 5 mmol/L glucose) also increased significantly after 24 hours in 25 mmol/L glucose. Phosphorylation of PKCII showed a significant increase only in the membranous fraction (Figure 1A: +74%, P<0.001 versus 5 mmol/L glucose). The small increase in PKC phosphorylation occurred across both the cytoplasmic and membranous cellular fractions (see online Figure B available in the online data supplement). Activation of PKCII in an isoform specific kinase assay (Figure 1B) showed a significant increase in activity in 25 mmol/L glucose after 24 hours in the membranous fraction (+160%, P<0.01 versus 5 mmol/L glucose alone).
Chemotaxis to PDGF-BB in 25 mmol/L Glucose Is Sorbitol Pathway and PKCII-Dependent
To determine whether the PKC isoforms elevated in 25 mmol/L glucose had a functional effect on VSMCs, chemotaxis assays were undertaken. Chemotaxis without serum starvation does not occur to a physiological level of PDGF-BB (360 pmol/L) in 5 mmol/L glucose (P=NS) but does occur in 25 mmol/L glucose (P<0.01: Table 2). The aldose reductase inhibitors (ARI) sorbinil and alrestatin, which prevent de novo synthesis of DAG, prevented chemotaxis to PDGF-BB. The general PKC inhibitor calphostin C and the specific PKCII inhibitor LY379196 used at a concentration (200 nmol/L) where PKCI was unaffected blocked chemotaxis (see online Figure C and online Table A). Microinjection experiments showed that only affinity-purified antibodies to the PKCII isoform blocked chemotaxis in 25 mmol/L glucose (Table 2); this could be prevented by preincubating the antibody with a molar excess of its cognate peptide.
PKCII in PDGF-BB Treated VSMCs: Activation/Phosphorylation Is Sorbitol PathwayeCDependent, Protein Level Is mTOR PathwayeCDependent
Chemotaxis to PDGF-BB in 25 mmol/L glucose is PKCII- and sorbitol pathway-dependent. We have previously shown that glucose-potentiated chemotaxis depends on PI3K and MAPK pathways7; therefore, the relationship of these pathways to PKCII regulation was examined.
Global PKCII Phosphorylation
To examine the degree of phosphorylation, serine/threonine phosphorylation of the protein was measured (Figure 1C). Phosphorylation was unaffected by inhibitors in 5 mmol/L glucose, but increased after 24 hours in 25 mmol/L glucose (+94%, P<0.001 versus 5 mmol/L glucose alone). Reductions in phosphorylation were found in 25 mmol/L glucose after 24 hours with PI3K, mTOR, and sorbitol pathway inhibitors (P<0.01 versus 25 mmol/L glucose alone). Similar results to those above were found when the degree of phosphorylation on the Ser660 residue of PKCII (located at the cofactor binding site)22 was examined with a significant increase after 24 hours in 25 mmol/L glucose (+103%, P<0.001 versus 5 mmol/L glucose alone; Figure 1D).
PKCII Protein Level
In 5 mmol/L glucose, the total protein level depended on the PI3K-related mTOR pathway because mTOR inhibition (rapamycin) decreased it by 24% (P<0.01: Figure 1E). PKCII in 5 mmol/L glucose was unaltered by sorbinil (ARI), PD98059 (MEK1/2 inhibitor), or by the farnesyltransferase inhibitor (FTI), which prevents prenylation of Ras (MAPK pathway). The increased level of PKCII in 25 mmol/L glucose (+55%, versus 5 mmol/L glucose, P<0.01) was under similar control but sorbinil or alrestatin (not shown) were also able to decrease the total PKCII level and the effects of PI3K inhibition were significant.
PKC Inhibition Prevents 25 mmol/L Glucose-Potentiated PDGF- Receptor Upregulation in PDGF-BBeCTreated VSMCs
Investigation was made into possible PKC involvement in the pathways responsible for PDGF- receptor upregulation in conditions of 25 mmol/L glucose plus 360 pmol/L PDGF-BB (Figure 2A). In 5 mmol/L glucose, MAPK and PI3K dependence were demonstrated (P<0.05 versus untreated for both inhibitors). Receptor upregulation associated with 25 mmol/L glucose for 24 hours (+101%, P<0.01 versus 5 mmol/L glucose alone) was decreased significantly by PI3K, MAPK, general PKC, PKCII, or aldose reductase inhibition. However, only MAPK or aldose reductase inhibition returned the receptor back to the 5 mmol/L glucose level (P<0.01 versus 25 mmol/L and P=NS versus 5 mmol/L glucose alone). That PKC inhibition controlled the receptor at the transcriptional level was confirmed by RT-PCR (Figure 2B). The significant increase in receptor mRNA found in 25 mmol/L glucose (+83%, P<0.01) was inhibited partially by general PKC inhibition but totally by PKCII inhibition (P<0.01 versus 25 mmol/L glucose alone).
PKC Inhibition Prevents 25 mmol/L Glucose-Potentiated Akt and ERK Activation in PDGF-BBeCTreated VSMCs
Pathways upregulating the PDGF- receptor in 25 mmol/L glucose include PKC, PI3K, and MAPK but the relationship between these pathways has not yet been determined.
PI3K Pathway
Akt (a downstream target of PI3K) phosphorylation increased after 24 hours in 25 mmol/L glucose (+54%, P<0.05 versus 5 mmol/L glucose; Figure 3A). Calphostin C caused a modest (nonsignificant) decrease in the 25 mmol/L glucose-potentiated phosphorylation level but LY379196 and the ARI both reduced it significantly (P<0.05 versus 25 mmol/L glucose alone). However, only the ARI decreased the level to that found in 5 mmol/L glucose. To confirm that the weak effect of PKCII inhibition on Akt phosphorylation was significant with respect to activity, an Akt kinase assay was undertaken (Figure 3B). LY379196 did not affect Akt activation in 5 mmol/L glucose but reduced the ability of Akt to phosphorylate its target molecule, GSK-3, in 25 mmol/L glucose (+81% in 25 mmol/L glucose versus 5 mmol/L glucose alone, P<0.01; LY379196 treatment, +45%, P<0.05 versus 5 or 25 mmol/L glucose alone). PKC appears to act upstream of PI3K in the pathways leading to PDGF- receptor upregulation in 25 mmol/L glucose.
MAPK Pathway
ERK1/2 (a downstream target in the MAPK pathway) phosphorylation increased in 25 mmol/L glucose after 24 hours (+95%, P<0.01 versus 5 mmol/L glucose; Figure 3C). Calphostin C and LY379196 had a modest (nonsignificant) effect on the 25 mmol/L glucose-potentiated increase in phosphorylation, but sorbinil reduced the level (P<0.05 versus 25 mmol/L glucose alone). To identify if LY379196 modulated ERK phosphorylation, an ERK kinase assay was undertaken (Figure 3D). LY379196 did not affect ERK activation in 5 mmol/L glucose but reduced the ability of ERK to phosphorylate its target molecule ELK-1 in 25 mmol/L glucose (+189% in 25 mmol/L glucose versus 5 mmol/L glucose alone, P<0.01; LY379196 treatment, P<0.05 versus 5 or 25 mmol/L glucose alone). It would appear that PKCII acts upstream of ERK/MAPK in the pathways leading to PDGF- receptor upregulation in 25 mmol/L glucose.
To confirm that ERK phosphorylation was a downstream target of PKCII and Akt activation, microinjection experiments were undertaken (Figure 3E and Figure 3F). Injection of activated Akt caused a significant increase in ERK phosphorylation in 5 mmol/L glucose conditions. In 25 mmol/L glucose and the presence of alrestatin (20 e蘭ol/L), injection of activated Akt (Figure 3E) could partially restore the level of ERK phosphorylation (P<0.05 versus 5 mmol/L or 25 mmol/L glucose). In the presence of the PKCII inhibitor, LY379196 (200 nmol/L), injection of activated Akt could totally recover the level of ERK phosphorylation (P=NS versus 25 mmol/L glucose; Figure 3E). Injection of activated PKCII significantly increased ERK phosphorylation in both 5 and 25 mmol/L conditions (Figure 3F). The effect in 25 mmol/L glucose was partly lost in the presence of the ARI and totally lost in the presence of the MAPK inhibitor (Figure 3F). The microinjection of activated PKCII was sufficient to induce chemotaxis in 5 mmol/L glucose conditions (without prior serum starvation) and restored chemotaxis lost because of aldose reductase inhibition in 25 mmol/L glucose conditions (Table 2).
PKCII Inhibition Prevents 25 mmol/L Glucose-Potentiated Filopodia Formation
Migrating cells show characteristic changes in their ultrastructure. Membrane ruffles (lamellipodia) distribute around the rim of a nonmigrating cell: this is seen in a representative micrograph (Figure 4A) for a cell in 5 mmol/L glucose within a 360 pmol/L PDGF-BB gradient. In 25 mmol/L glucose, the cell polarizes with a lamellipodium localizing along the front and prominent actin bundles stretching along the long axis of the cell (Figure 4B). Characteristic filopodia (linked with direction sensing) protrude from the lamellipodium (see inset). PDGF- receptors are detected along filopodia in 25 mmol/L glucose (Figure 4C). With PKCII inhibition filopodia, lamellipodium and f-actin alignment are lost (Figure 4D).
PKN Signaling to Lamellipodia Is PKCII-Dependent in 25 mmol/L Glucose
Protein kinase N (PKN) is involved in formation of lamellipodia.23 We have recently shown that MAPK inhibitors affect filopodia but not lamellipodia formation in VSMCs.12 To clarify the role PKC has on lamellipodia formation, the effect of the pathway inhibitors on PKN was assessed (Figure 5A). The Ser/Thr phosphorylation of PKN was significantly increased in 25 mmol/L glucose (+69%, P<0.01 versus 5 mmol/L glucose). MAPK inhibition had no effect but PI3K, PKCII, or aldose reductase inhibition significantly decreased the activation level (P<0.01 versus 25 mmol/L glucose alone for all).
PKCII-Dependent Pathways Control Cell Speed in 25 mmol/L Glucose
Chemotaxis to PDGF-BB (360 pmol/L) was demonstrated in the presence of 25 mmol/L glucose. Migration also has a speed component. Speed in 25 mmol/L glucose is significantly reduced compared with matched cells (same source and passage) set up in 5 mmol/L glucose (random movement) in the chamber (Table 2). Speed was unaffected in 25 mmol/L glucose by MAPK inhibition (by chemical inhibitor or microinjected antibody; see online Table A), or by microinjection of conventional PKC isoform antibodies other than PKCII. PKCII isoform microinjection or LY379196 exposure prevented chemotaxis and restored speed to that found in 5 mmol/L glucose.
PKCII-Dependent Pathways Control Adhesion in 25 mmol/L Glucose
Changes in cellular adhesion may account for the decreased speed found in 25 mmol/L glucose; therefore, an adhesion assay was undertaken (Figure 5B). In 5 mmol/L glucose, MAPK (PD98059 or FTI), PKCII, or aldose reductase inhibition did not alter adhesion level, but PI3K inhibition caused a significant decrease in adhesion (LY294002, eC48%; wortmannin, eC42%; P<0.05 versus 5 mmol/L glucose alone). In 25 mmol/L glucose, adhesiveness increased significantly (+66%, P<0.05 versus 5 mmol/L glucose). MAPK inhibition had no effect, whereas PI3K inhibition showed a similar percentage reduction (LY294002, eC48%; wortmannin, eC54%), as found in 5 mmol/L glucose. PKCII inhibition and ARI, ineffective in 5 mmol/L glucose, significantly reduced adhesion in 25 mmol/L glucose. Similar results to those in Figure 5B were found when phosphospecific antibodies were microinjected into cells subsequently set up in an adhesion assay (Figure 5C). Again a pPKCII dependency was found in 25 mmol/L glucose conditions for control of adhesion. An antibody to phosphorylated FAK (a known component of focal adhesions) inhibited adhesion in both 5 mmol/L and 25 mmol/L glucose conditions. Microinjection of pAkt did not alter adhesion, indicating that this branch of the PI3K pathway is not involved in adhesion control (Figure 5C).
PKCII-Dependent Pathways Signal to Focal Adhesions in 25 mmol/L Glucose
Cellular adhesion is mainly controlled through focal adhesions, and focal adhesion kinase (FAK) can act as a marker of intracellular signaling to the substratum. The level of FAK phosphorylation increased in 25 mmol/L glucose (+69%, P<0.05 versus 5 mmol/L glucose; Figure 5D). MAPK inhibition had no effect on 25 mmol/L glucose-potentiated FAK phosphorylation, but the PI3K, PKCII, and aldose reductase inhibitors all restored it to 5 mmol/L glucose levels (P<0.05 versus 25 mmol/L glucose alone; Figure 5D). A representative micrograph is shown (Figure 4E) for a cell stained with phospho-FAK antibody (green) in 25 mmol/L glucose. The phospho-FAK(Tyr576/577) staining was punctate in nature and localized mainly to the cell membrane: this distribution is in keeping with that expected for focal adhesions. Microinjection of cells in 25 mmol/L glucose with affinity-purified phospho-FAK antibody did not alter PDGF- receptor level (online Figure D) nor prevent chemotaxis to PDGF-BB (Table 2); however, the cell speed (in 25 mmol/L glucose) was restored to that found in 5 mmol/L glucose.
Discussion
This study has revealed significant differences in both the glucose- and PKC-dependency of PI3K and MAPK activation in human VSMCs (Figure 6). Most PKC isoforms were detected in primary explant-derived human aortic VSMCs including PKC-, -I, -II, -, -, and -. The time frame for the 25 mmol/L glucose experiments was short in terms of the chronicity of diabetes and was associated with upregulation of the conventional isoforms of PKC (PKC- and -II). We found that the total level of PKCII was regulated by mTOR-dependent pathways in VSMCs; mTOR acts on the serine/arginine-rich splicing factor (SRp40) that controls splicing of the PKC gene, altering production away from PKCI toward PKCII.24 The PI3K/Akt pathway regulates mTOR-dependent signaling.25 Aldose reductase inhibitors prevent increased flux through the sorbitol pathway, reduce PI3K/Akt activation, and mTOR-dependent signaling. PKCII is upregulated in many cell types in response to glucose26; exceptionally, one group has reported small decreases in PKCII activity in rat aorta,27,28 but cell cycle synchronization through serum starvation in those experiments meant that the S phaseeCdependent decrease in PKCII28 may have affected the results. In the present study in human aortic VSMCs, both PKC and PKCII were activated, but only PKCII demonstrated a major increase in the membrane fraction and was involved in glucose-potentiated chemotaxis.
We have previously shown that regulation of the PDGF- receptor depends on both the PI3K and MAPK pathways7 with mainly, but not exclusively, PI3K signaling to the MAPK pathway.12 Aldose reductase inhibition and blockade of PKCII prevent PDGF- receptor increases in rat VSMCs14; in this study, we confirm the effect in human aortic VSMCs and demonstrate that receptor protein expression is controlled at the transcriptional level. We demonstrated that PKCII inhibition suppresses, but does not abolish, activation of Akt and ERK in high glucose; PKCII inhibition has no effect on ERK activity in normal glucose (not shown). DAG can activate the MAPK pathway directly,17 and in our experiments, sorbitol pathway inhibition or MAPK inhibition (upstream and downstream of DAG, respectively) removed all of the high glucose effect on PDGF- receptor protein and chemotaxis. Under similar glucose conditions, migration of coronary VSMCs was also sorbitol pathway-dependent.13 Here, we show that high glucose induces both PKCII-independent and PKCII-dependent signaling to the MAPK pathway. PKCII-dependent signaling to MAPK occurs indirectly via PKCII-stimulated phosphorylation in the PI3K/Akt pathway with subsequent cross-talk to the ERK/MAPK pathway.12 The PKCII-independent signaling involves direct DAG activation of Ras upstream in the ERK/MAPK pathway bypassing both PKCII and the PI3K/Akt pathway.
Previously, we showed that MAPK pathway inhibition reduces filopodia but not lamellipodia formation12; in this study, PKCII inhibition affects both. The effect of PKCII on MAPK activation (via PI3K) may account for its effect on filopodia. To confirm that lamellipodia formation was PKCII/PI3K-dependent, we measured PKN activity, which is involved in lamellipodia formation,23 and demonstrated it was PKCII/PI3K-dependent and MAPK-independent. Speed of movement was higher in 5 mmol/L glucose (random movement) than in 25 mmol/L glucose (migration), provided cells were age- and source-matched. With PKCII inhibition (chemical inhibitor or microinjected antibody), migration was abolished and speed restored to that for 5 mmol/L glucose. Specific inhibitors of the MAPK pathway (PD98059 or microinjected anti-pERK antibody) prevented migration in high glucose without altering speed, indicating that speed is controlled at a higher point in the pathways. Cell adhesion in 25 mmol/L glucose was regulated in a similar manner to speed with PKCII inhibition reducing adhesion to that found with 5 mmol/L glucose levels, and MAPK inhibition having no effect. Focal adhesions, which control inside to outside cell signaling, contain PDGF- receptors and integrins linked via their cytoplasmic components to signaling molecules such as FAK.29 In this study, the phosphorylation site to which the FAK antibody was directed depends on PI3K signaling.30 We demonstrated that in 25 mmol/L glucose this signaling to FAK can be modified by input from PKCII. As demonstrated by phospho-FAK antibody microinjection, it is this PKCII/PI3K-dependent signaling to FAK that controls cell speed by altering adhesion (see Figure 6).
In summary, we have demonstrated that in glucose conditions that occur intermittently in diabetes, human aortic VSMCs migrate in a concentration of PDGF-BB such as is present in atherosclerotic plaques. This glucose-potentiated migration is sorbitol pathway-dependent for activation of the ERK/MAPK pathway and for PKCII signaling to the PI3K pathway. We have shown in this study that in addition to PKCII-independent signaling to the MAPK pathway from increased flux through the sorbitol pathway, there is PKCII-dependent signaling to MAPK. Increases in PDGF- receptor level depend on PKCII signaling to both the PI3K and ERK/MAPK pathways in these conditions. PKCII influence on PI3K leads to alterations in adhesion, FAK phosphorylation, cell speed, and lamellipodia formation, but the ERK/MAPK pathway is not involved in these processes.
Acknowledgments
This work was supported by a project grant from the Research and Development Office, Department of Health, Social Services, and Public Safety, Northern Ireland. LY379196 was a generous gift from Eli Lilly. We thank P. Anderson (for practical assistance and manuscript comments) and W.E. Allen (for manuscript comments).
References
Beckman JA, Creager MA, Libby P. Diabetes and atherosclerosis: epidemiology, pathophysiology, and management. JAMA. 2002; 287: 2570eC2581.
Ceriello A. Impaired glucose tolerance and cardiovascular disease: the possible role of post-prandial hyperglycemia. Am Heart J. 2004; 147: 803eC807.
Schwartz SM. Smooth muscle migration in atherosclerosis and restenosis. J Clin Invest. 1997; 100: S87eCS89.
Waltenberger J. Modulation of growth factor action: implications for the treatment of cardiovascular diseases. Circulation. 1997; 96: 4083eC4094.
Kawano M, Koshikawa T, Kanzaki T, Morisaki N, Saito Y, Yoshida S. Diabetes mellitus induces accelerated growth of aortic smooth muscle cells: association with overexpression of PDGF beta-receptors. Eur J Clin Invest. 1993; 23: 84eC90.
Ross R. The pathogenesis of atherosclerosis: a perspective for the 1990s. Nature. 1993; 362: 801eC809.
Campbell M, Allen WE, Silversides JA, Trimble ER. Glucose-induced phosphatidylinositol 3-kinase and mitogen-activated protein kinase-dependent upregulation of the platelet-derived growth factor-beta receptor potentiates vascular smooth muscle cell chemotaxis. Diabetes. 2003; 52: 519eC526.
Sano H, Sudo T, Yokode M, Murayama T, Kataoka H, Takakura N, Nishikawa S, Nishikawa SI, Kita T. Functional blockade of platelet-derived growth factor receptor-beta but not of receptor-alpha prevents vascular smooth muscle cell accumulation in fibrous cap lesions in apolipoprotein EeCdeficient mice. Circulation. 2001; 103: 2955eC2960.
Servant G, Weiner OD, Herzmark P, Balla T, Sedat JW, Bourne HR. Polarization of chemoattractant receptor signaling during neutrophil chemotaxis. Science. 2000; 287: 1037eC1040.
Yamboliev IA, Gerthoffer WT. Modulatory role of ERK MAPK-caldesmon pathway in PDGF-stimulated migration of cultured pulmonary artery SMCs. Am J Physiol Cell Physiol. 2001; 280: C1680eCC1688.
Schneller M, Vuori K, Ruoslahti E. Alphavbeta3 integrin associates with activated insulin and PDGFbeta receptors and potentiates the biological activity of PDGF. EMBO J. 1997; 16: 5600eC5607.
Campbell M, Allen WE, Sawyer C, Vanhaesebroeck B, Trimble ER. Glucose-potentiated chemotaxis in human vascular smooth muscle is dependent on cross-talk between the PI3K and MAPK signaling pathways. Circ Res. 2004; 95: 380eC388.
Yasunari K, Kohno M, Kano H, Minami M, Yoshikawa J. Aldose reductase inhibitor improves insulin-mediated glucose uptake and prevents migration of human coronary artery smooth muscle cells induced by high glucose. Hypertension. 2000; 35: 1092eC1098.
Nakamura J, Kasuya Y, Hamada Y, Nakashima E, Naruse K, Yasuda Y, Kato K, Hotta N. Glucose-induced hyperproliferation of cultured rat aortic smooth muscle cells through polyol pathway hyperactivity. Diabetologia. 2001; 44: 480eC487.
Idris I, Gray S, Donnelly R. Protein kinase C activation: isozyme-specific effects on metabolism and cardiovascular complications in diabetes. Diabetologia. 2001; 44: 659eC673.
Williamson JR, Chang K, Frangos M, Hasan KS, Ido Y, Kawamura T, Nyengaard JR, van den EM, Kilo C, Tilton RG. Hyperglycemic pseudohypoxia and diabetic complications. Diabetes. 1993; 42: 801eC813.
Ebinu JO, Bottorff DA, Chan EY, Stang SL, Dunn RJ, Stone JC. RasGRP, a Ras guanyl nucleotide-releasing protein with calcium- and diacylglycerol-binding motifs. Science. 1998; 280: 1082eC1086.
Lee IK, Koya D, Ishi H, Kanoh H, King GL. d-Alpha-tocopherol prevents the hyperglycemia induced activation of diacylglycerol (DAG)-protein kinase C (PKC) pathway in vascular smooth muscle cell by an increase of DAG kinase activity. Diabetes Res Clin Pract. 1999; 45: 183eC190.
Collier A, Small M. The role of the polyol pathway in diabetes mellitus. Br J Hosp Med. 1991; 45: 38eC40.
Oka M, Kato N. Aldose reductase inhibitors. J Enzyme Inhib. 2001; 16: 465eC473.
Aronson D, Rayfield EJ. How hyperglycemia promotes atherosclerosis: molecular mechanisms. Cardiovasc Diabetol. 2002; 1: 1eC10.
Edwards JL, Apicella MA. The Molecular Mechanisms Used by Neisseria gonorrhoeae To Initiate Infection Differ between Men and Women. Clin Microbiol Rev. 2004; 17: 965eC981.
Dong LQ, Landa LR, Wick MJ, Zhu L, Mukai H, Ono Y, Liu F. Phosphorylation of protein kinase N by phosphoinositide-dependent protein kinase-1 mediates insulin signals to the actin cytoskeleton. Proc Natl Acad Sci U S A. 2000; 97: 5089eC5094.
Patel NA, Chalfant CE, Watson JE, Wyatt JR, Dean NM, Eichler DC, Cooper DR. Insulin regulates alternative splicing of protein kinase C beta II through a phosphatidylinositol 3-kinase-dependent pathway involving the nuclear serine/arginine-rich splicing factor, SRp40, in skeletal muscle cells. J Biol Chem. 2001; 276: 22648eC22654.
Oudit GY, Sun H, Kerfant BG, Crackower MA, Penninger JM, Backx PH. The role of phosphoinositide-3 kinase and PTEN in cardiovascular physiology and disease. J Mol Cell Cardiol. 2004; 37: 449eC471.
Way KJ, Katai N, King GL. Protein kinase C and the development of diabetic vascular complications. Diabet Med. 2001; 18: 945eC959.
Patel NA, Chalfant CE, Yamamoto M, Watson JE, Eichler DC, Cooper DR. Acute hyperglycemia regulates transcription and posttranscriptional stability of PKCbetaII mRNA in vascular smooth muscle cells. FASEB J. 1999; 13: 103eC113.
Yamamoto M, Acevedo-Duncan M, Chalfant CE, Patel NA, Watson JE, Cooper DR. Acute glucose-induced downregulation of PKC-betaII accelerates cultured VSMC proliferation. Am J Physiol Cell Physiol. 2000; 279: C587eCC595.
Schaller MD. Biochemical signals and biological responses elicited by the focal adhesion kinase. Biochim Biophys Acta. 2001; 1540: 1eC21.
Sundberg LJ, Galante LM, Bill HM, Mack CP, Taylor JM. An endogenous inhibitor of focal adhesion kinase blocks Rac1/JNK but not Ras/ERK-dependent signaling in vascular smooth muscle cells. J Biol Chem. 2003; 278: 29783eC29791.(Malcolm Campbell, Elisabe)