Cardiovascular disease is still one main cause of death in the world . Coronary artery disease (CAD), myocardial infarction (MI), and atherosclerosis are of particular importance regarding cardiovascular events and death among patients suffering from cardiovascular disease [2–4]. The underlying pathophysiological processes causing cardiovascular diseases are mediated via several factors and signaling pathways in a variety of different cell types, such as endothelial cells, smooth muscle cells, platelets, monocytes, and cardiomyocytes [5–10]. Phosphoinositide 3-kinase (PI3K)-related signaling pathways play an important role in the pathogenesis of cardiovascular as well as other human diseases [11–14]. Therefore, the members of the PI3K family are promising therapeutic targets for the treatment of cardiovascular as well as other diseases.
A detailed description of the role of the PI3K family and therapeutic strategies targeting PI3Ks in cancer and immune diseases were given within several reviews before [15,16]. The question of the role of PI3K family members in cardiovascular disease and the therapeutic potential of pharmacologic PI3K inhibitors was rather neglected so far. Thus, this review will focus on the current state of information regarding the potential role of pharmacologic PI3K inhibitors in cardiovascular diseases.
About 20 years ago PI3Ks were initial discovered as lipid kinases associated with viral oncoproteins [17,18]. The association between PI3Ks and cancer was further confirmed during the next two decades [19,20] as well as the role of PI3Ks in other biologic processes, such as the regulation of cell survival, cell proliferation, migration, transcriptional, and translational responses and differentiation [21,22]. Dysregulation of the PI3K activity and, therefore, the phosphoinositide levels contributes to a variety of diseases including thrombosis, diabetes, inflammatory and autoimmune diseases, and cancer [16,23,24].
The enzyme activity to phosphorylate the 3′-hydroxyl group of membrane phosphatidylinositols (PtdIns) interconnects a group of at least eight proteins which differ in expression, regulation, structure, and substrate specificity with in the family of PI3Ks . The PI3K family have been divided into three classes. Class I PI3Ks form heterodimers, which consists of a catalytic (p110α, β, δ, and γ) and a regulatory subunit (p85 or p101 family). The members of class I PI3Ks phosphorylate PtdIns(4,5)P2 to PtdIns(3,4,5)P3 and are activated by a variety of cell surface receptors, such as growth factor receptors and G-protein-coupled receptors . Class II PI3Ks are capable to phosphorylate PtdIns to PtdIns(3)P and has been shown to contribute to the production of PtdIns(4,5)P2. This class of PI3Ks consists of three monomeric members PI3K-C2α, C2 β, and C2γ. This class is characterized by their Ca2+-insensitive C2 homology domain, which is responsible for the nomenclature of this PI3K class. Class II PI3Ks were shown to be associated to several receptors, such as EGF or PDGF receptors and were found to be induced by chemokines, cytokines, and insulin [26,27]. The single member of the class II of PI3K is vacuolar sorting protein 34 (Vsp34) which produced only PtdIns(3)P by phosphorylating PtdIns . This enzyme is ubiquitously expressed but little is known about the activation and role of this enzyme. Regarding the role in vascular biology class I PI3Ks are the best-characterized group of PI3Ks. In contrast to class I, little is known about the influence of class II and III PI3Ks in vascular (patho)physiology.
The (patho)Physiologic Role Class I PI3Ks
PI3K mediates key signaling pathways via several downstream molecules, such as protein kinase B (PKB/Akt), 3′-phosphoinositide-dependent kinase 1 (PDK1), mammalian target of rapamycin (mTOR), and glycogen synthase kinase 3 (GSK3) which modulate different biologic functions, such as cell proliferation, cell survival, cell differentiation, and chemotaxis (Figure 1) [21,25,28]. Activation of class I PI3Ks results in the generation of PtdIns(3,4,5)P3 and leads to the recruitment of PtdIns(3,4,5)P3-binding proteins to the plasma membrane . Many of these adaptor and effector proteins contain a pleckstrin homology (PH) domain . The prototype of PH domain-containing factors is PKB/Akt. The activation of PKB/Akt as well as other downstream proteins, such as mTOR leads to the modulation of essential cellular functions, such as cell proliferation, migration, transcriptional as well as translational responses, differentiation, and cell survival . The control of these cellular functions is important for the regulation of essential biologic processes, such as development, metabolism, vascular homeostasis, angiogenesis, and thrombogenicity [25,29,30]. These processes, and therefore the class I PI3Ks play a role under physiologic conditions but also in several human diseases including thrombosis, diabetes, inflammatory diseases, and cancer [16,23,24].
Class I PI3K Isoforms in Cardiovascular Disease
The first observation indicating PI3Ks to be involved in cell and organ growth was made by Leevers et al. in 1996 in Drosophila. This raised the question whether PI3Ks are also involved in regulating the organ size of vertebrates. Four years later Shioi et al. found that overexpression of PI3Kα in hearts of mice resulting in an increase of heart size . They showed cardiac hypertrophy to be mediated by augmented cell size. Moreover, overexpressing PI3Kα in mouse hearts had no effect on function or constitution of these hearts. Two years later the same working group demonstrated the downstream effectors of PI3Ks, PKB/Akt as well as the mTOR to be involved in elevating cell size and leading to cardiac hypertrophy in mice , indicating the PI3Kα-PKB/Akt-mTOR signaling pathway to regulate heart size.
While PI3Kα seems to control heart size [32–34], PI3Kγ, another isoform of PI3Ks has emerged as an important modulator of cardiac contractility by modulating the cyclic adenosine monophosphate (cAMP) metabolism in cardiomyocytes [35–37]. Patrucco et al. also found the PI3Kγ to be involved in the modulation of cardiac contractility as well as cardiac remodeling in response to chronic pressure overload by influencing the cAMP metabolism via kinase-dependent as well as independent mechanisms . More recent data indicate PI3Kγ to be involved in the regulation of infarct size after ischemia/reperfusion injury, reparative neovascularization and healing of MI [11,39]. Moreover, this PI3K isoform was indicated to be involved in the homing of endothelial progenitor cells (EPC) affecting the neovascularization-promoting capacity of these cells in ischemic muscles . In this context Madeddu et al. also showed PI3Kγ to modulate proangiogenic processes—possibly—influencing the postischemic neovascularization by mechanisms independent of the kinase activity of this isoform . Thus, PI3Kγ may be a promising target for pharmaceutical treatment of patients with MI.
Another important aspect is the involvement of PI3K isoforms in thrombogenesis. In platelets PI3Kα, β, and γ are suggested to play a role in aggregation . Hirsch et al. found PI3Kγ-deficient mice to exhibit impaired platelet aggregation and reduced thromboembolism . Recently, Gilio et al. demonstrated that both, PI3Kα and β are required for glycoprotein VI-dependent platelet signaling and thrombus formation . These data indicates PI3Kα, β, and γ to be potential targets for antithrombotic therapy.
Finally, inflammatory processes play an important role in the different stages of atherosclerosis [45,46]. The most PI3K isoforms are involved in inflammatory processes [15,47]. Thus, targeting PI3Ks involved in inflammatory responses offers a therapeutic strategy for the treatment of atherosclerosis.
“First Generation” PI3K Inhibitors
The “first generation” of PI3K inhibitors includes the fungal metabolite wortmannin, its derivative demethoxyviridin and LY294002. In 1974 Wiesinger et al. described the antiinflammatory activity of wortmannin for the first time . About two decades later wortmannin was found to be a potent inhibitor of PI3Ks . Wortmannin interacts with the ATP-binding pocket by forming covalent interactions with the catalytic lysine residue. The high reactivity of wortmannin as well as the low selectivity may lead to potential hazardous off-target effects with other molecules . In 1994, a synthetic compound LY294002 developed by Lilly Research Laboratories was demonstrated to be a potential PI3K inhibitor . In contrast to wortmannin, LY294002 is an ATP competitive PI3K inhibitor. Thus, the PI3K inhibition is reversible. The sensitivity of LY294002 is about 500-fold lower that the sensitivity of wortmannin to PI3Ks [51,52]. LY294002 is like wortmannin not selective for PI3K isoforms or classes and both have several off-target effects on other enzymes [53,54]. It was shown that LY294002 inhibits K channels, thereby, affecting the contractility of cardiac myocytes independent of the blocking of PI3Ks . Also wortmannin was demonstrated to block the activity of other factors beside PI3Ks, such as mTOR . Therefore, these inhibitors were named “dirty” pan-PI3K inhibitors.” These “first generation” PI3K inhibitors have been used widely as pharmaceutical tools for elucidating the role of PI3Ks and associated signaling pathways in essential biologic processes [15,21,42]. These studies emerged PI3Ks also as important factors in several human diseases [16,29,45]. The insights into the role of the PI3Ks in human diseases offer a wide spectrum of therapeutic strategies. However because of the manifold side effects and toxicity of these “first generation” inhibitors [25,54] the development of PI3K inhibitors with lesser toxicity and higher selectivity for the PI3Ks or even specific PI3K isoforms was needed.
The “Next Generation” of PI3K Inhibitors
Most protein kinase inhibitors developed so far target the ATP-binding site . The ATP-binding site of the most protein kinases are often very similar structured. Thus, local variation in the mode of ATP binding of protein kinase inhibitors allowed the development of protein kinase-specific inhibitors, leading to promising clinical applications . X-ray crystallography of PI3Kγ bound to various PI3K inhibitors, such as wortmannin or LY294002 revealed how these components fit into the ATP-binding site . These studies facilitated the development of a new generation of PI3K inhibitors with less toxicity and higher selectivity for the PI3Ks, specific subgroups or even distinct catalytic isoforms [15,16,28].
As mentioned above, the high similarity in the amino acid sequence and structure of the ATP-binding site of PI3K isoforms the development of isoform specific inhibitors was difficult. The ICOS Corporation developed the first PI3Kδ specific inhibitors, and one of them was the compound IC87114 . PI3Kδ inhibition by IC87114 led to a reduction of joint inflammation through effects on neutrophils in mice . These first achievements pushed the development of further isoform selective PI3K inhibitors. The effort of the following years of research in this field resulted in the development of several isoform-specific and isoform-prefering inhibitors. Examples for isoform-specific inhibitors are PI3Kγ inhibitors, such as AS-604850 and AS-605240, developed by Serono International. Both components were shown to reduce the chemotaxis of neutrophils and monocytes and to reduce the progression of inflammation and damage in a mouse model of rheumatoid arthritis . Another example is thrieno [2,3] pyrimidine, compound 15e. This compound was found to be a selective inhibitor of PI3Kα and to suppress tumor cell proliferation in vitro.
Also several isoform-preferring inhibitors, which inhibit a subset of PI3K isoforms were developed and tested in vitro and in vivo. One example is the imidazoquinazoline PIK-90 from Bayer, which targets in favour PI3Kα, γ, and δ isoforms . The compound TG100–115, developed by TargeGen Inc., San Diego, CA, is a further isoform-preferring inhibitor, shown to inhibit PI3Kγ and δ[39,64].
The development of PI3K isoform-specific inhibitors with a safe and therapeutic potent profile is difficult . Therefore, the selectivity as well as specificity of pharmacologic inhibitors is a critical point and should be viewed carefully. Moreover, human diseases are in most cases not the result of one pathological process [15,44]. Therefore, it could be argued that inhibitors targeting all PI3K isoforms, but exhibit a safer profile and lesser side effects would be useful for the treatment of multifactorial caused diseases, such as inflammatory diseases . In 1996, Yaguchi et al. described such a selective “clean” pan-PI3K inhibitor for the first time and showed this compound (ZSTK474) to exhibit strong antitumor activity in vitro and in vivo without toxic effects on critical organs . Recently, Kong and Yamori demonstrated that ZSTK474 has a potent antiangiogenic effect due to the inhibition of PI3Ks and subsequently VEGF secretion in endothelial cells .
Therapeutic Potential of PI3K Inhibitors in Cardiovascular Disease
Currently, ADP-receptor blockers or antiplatelet therapy represents the standard of care for the therapeutic treatment of several cardiovascular diseases . But in some cases intolerance to antiplatelet drugs reduce the applicability and effectiveness of these standard therapies . As mentioned above little is known about the therapeutic potential of pharmaceutical PI3K inhibitors in cardiovascular diseases so far. It was shown that PI3K isoforms control a variety of important biological processes, such as cell growth, cell migration or translational responses [21,30,32], subsequently modulating essential physiologic functions during health and disease, such as inflammation, vascular homeostasis, angiogenesis and thrombogenicity [25,29,30]. Thus, the therapeutic potential of pharmacological PI3K inhibitors is enormous (Table 1). Moreover, it represents a possible therapeutic option to standard therapies when these therapies were ineffective.
|Cardiovascular disease||Involved Pi3K isoform||Inhibition by||Reference|
PI3K Inhibition in Atherosclerosis
Inflammatory processes play an important role in the pathogenesis of atherosclerosis and other cardiovascular diseases [5,8,30,69]. Over the last years, many potential inflammatory targets were identified, including chemokines, integrins and adhesion molecules [70–72]. Furthermore, several pathways influencing inflammatory processes in atherosclerosis were identified, such as the coagulation cascade and Toll-like receptor-mediated signaling [5,30,73]. PI3K isoforms are involved in the regulation and transmission of these proinflammatory pathways and signals [30,74,75]. The use of “dirty” pan-PI3K inhibitors results in a deeper understanding how PI3Ks influence these processes [30,75,76]. These nonselective inhibitors not allowed discriminating between PI3K isoforms to characterize the function of specific PI3K isoforms. Through the use of isoform-specific inhibitor, PI3Kγ has emerged as an important factor and attractive target in inflammatory diseases and atherosclerosis (Table 1) [61,77,78]. Fougerat et al. showed that PI3Kγ is highly expressed in human as well as murine atherosclerotic lesions . Moreover they found the pharmacological PI3Kγ-selective inhibitor AS605240 to significantly reduce early atherosclerotic lesions in apolipoprotein E (ApoE)-null mice. The same effect was observed on advanced atherosclerotic lesions of LDL receptor (LDLR)-deficient mice in response to this PI3Kγ inhibitor  suggesting this isoform to be a promising target for the treatment of atherosclerosis. These observations were substantiated by results published by Chang et al. showing ApoE-null mice lacking PI3Kγ to possess reduced atherosclerotic lesions than those observed in control ApoE-null mice . Moreover, they found no detectable activation of PI3K/PKB/Akt in macrophages as well as in atherosclerotic lesions of PI3Kγ-lacking ApoE-null mice . This is in line with the results of Fougerat et al. showing AS605240 to decrease the PKB/Akt phosphorylation by inhibiting PI3Kγ activity in atherosclerotic lesions of LDLR-deficient mice compared to nontreated controls . The selectivity as well as specificity of such a “new generation” PI3K inhibitor is a critically discussed point. Therefore, it is essential to confirm observations obtained from pharmacologic inhibition by data from corresponding specific knockout models or transgenic models.
PI3Kγ was also demonstrated to positively regulate glucose-induced insulin secretion in vivo. Therefore, in the context of methabolic diseases, such as diabetes or glucose intolerance, the pharmacologic inhibition of PI3Kγ could—possibly—contribute to metabolic-related complications under those conditions. In such cases a critical monitoring should be self-evident and absolutely necessary.
Taken together, these data show PI3Kγ to play an important role in regulating inflammatory processes involved in several stages of atherosclerosis, such as cellular composition and final fibrous cap establishment. Thus, pharmacologic inhibition by selective and “clean” PI3Kγ inhibitors is a potentially important therapeutic tool for the treatment of atherosclerosis.
PI3K Isoform-Specific Inhibition in MI
As mentioned above the development of PI3K isoform-selective inhibitors offered the opportunity to analyze the role of specific PI3K isoforms in health as well as under pathophysiologic conditions [16,77]. With regard to MI, PI3Kγ, and δ were found to be of particular importance (Table 1) [11,39,64,81]. Both isoforms are involved in inflammatory processes, maybe contributing to tissue damage upon reperfusion [28,47,61]. In 2006, Doukas et al. showed that pharmacologic inhibition of PI3Kγ and δ by compound TG100–115 in functional animal MI models reduced infarct development after ischemia/reperfusion injury while preserving myocardial function . In human endothelial cells the inhibition of PI3Kγ and δ blocked VE-catherin-mediated proedema VEGF signaling pathways, whereas, VEGF-induced ERK signaling pathways regulating cell proliferation were not affected . In vivo the inhibition of PI3Kγ/δ by TG100–115 reduced edema and inflammation in a hindpaw model substantiating an antiedema effect of this compound . Moreover, TG100–115 was shown to limit infarct size in both rodent- and porcine-based models of MI up to 3 hours after reperfusion, which corresponds to the time period when patients are most accessible for therapeutic intervention . These data emerge the selective PI3Kγ/δ inhibitor TG100–115 as an attractive compound for a cardioprotective therapy in the postreperfusion phase of MI. TG100–115 has entered clinical phase I and II trials for acute MI [39,64,82].
In contrast to the observation of Doukas et al., the selective inhibition of PI3Kγ by the pharmaceutical inhibitor AS605240 as well as specific siRNAs impaired survival and proliferation of endothelial cells in the periinfarct zone . Moreover, PI3Kγ inhibition by AS605240 resulted in a defective reparative neovascularization leading to an increased infarct size and to impair recovery of left ventricular function in a mouse model of MI . In vitro, the lack of PI3Kγ led to an increased apoptosis rate of cardiomyocytes as well as impaired reparative neovascularization and reduced the cardiac function post MI as also observed after pharmacological inhibition . The authors discussed these differences between both studies by emphasizing that Doukas et al. only analyzed VEGF-induced proangiogenic signaling acting via tyrosine kinase receptors while leaving open a possible interaction between the inhibitor and G-protein coupled receptor (GPCR)-dependent angiogenesis [11,39]. This may suggest an incomplete inhibition of PI3K activity of the γ and/or δ isoform by TG100–115.
Recently, it was indicated that the PI3Kγ isoform is involved in postischemic neovascularization by modifying the homing of EPC [40,41]. Therefore, it is possible that the pharmacologic inhibition of PI3Kγ in patients with MI could lead to a reduction of EPC migration and their neovascularization-promoting capacity in these patients. Whether this is scenario plays a role under (patho)physiologic conditions is not known and should be investigated in further studies.
In summary, these results indicate the pharmacological inhibition of PI3Kγ and/or δ to be of potential interest for therapeutic treatment of MI patients, at least, after reperfusion. However, possible side effects on angiogenesis or other physiologic relevant processes should be critically examined.
PI3K Inhibition in Thrombosis
Recently, Gilio et al. demonstrated PI3Kα and β to be involved into glycoprotein VI (GPVI)-induced platelet activation and thrombus formation . Treatment of platelets with specific pharmacologic inhibitors against PI3K isoforms α (PIK-75) and β (TGX-221) reduced GPVI-induced downstream signaling events, such as the Ca2+ mobilization, the production of inositol 1,4,5-trisphosphate as well as phosphorylation of PKB/Akt and substantially decreased the platelet deposition on collagen. In contrast to PI3Kα and β, the selective pharmacologic inhibition of PI3Kδ, by IC-87114 and PI3Kγ by AS-252424 had no effect on these platelet signaling . These observations were substantiated by additional experiments demonstrating platelets from PI3Kα-deficient mice to show lower Ca2+ signal and decreased thrombus formation under dynamic flow conditions, compared to platelets isolated from wild type mice as well as mice lacking PI3Kγ. As mentioned earlier, the selectivity of pharmacologic inhibitors is a crucial point, wherefore, it the validation by using corresponding specific knockout or transgenic models is absolutely essential. The role of PI3Kα and β in GPVI-induced platelet signaling and thrombosis was further substantiated by the findings of Mangin et al. showing the selective inhibition of PI3Kα or β by PIK-75 or TGX-221, respectively to reduce the GPVI-induced Akt phosphorylation, Ca2+ mobilization and platelet aggregation . They found TGX-221-mediated blocking of PI3Kβ to exhibit a stronger inhibitory effect on these processes than blocking of PI3Kα by PIK-75. In accordance to the observation of Gilio et al. , the pharmacologic inhibition of PI3Kδ and PI3Kγ had no effect on GPVI-induced platelets . Thus, PI3Kα and β play seem to be essential in GPVI-mediated platelet aggregation and are interesting targets for antiplatelet therapy.
Contrary to the observation made in GPVI-induced platelets [44,83], Hisch et al. showed ADP-stimulation of platelets from PI3Kγ-deficient mice to result in a decreased level of PKB/Akt phosphorylation and integrin αIIbβ3 activation as well as impaired aggregation . It was demonstrated that the lack of PI3Kγ protected mice from ADP-induced thromboembolic vascular occlusion, indicating PI3Kγ to be a potential target for antithrombotic drugs without causing major adverse effects . This study had some elementary limitations with regard to the used model of thromboembolism. The authors found that other stimuli, such as collagen or thrombin caused a normal aggregation of platelets from PI3Kγ-null mice. Thus, a lack of PI3Kγ prevented mice exclusively from death caused by ADP-induced platelet-dependent thromboembolic vascular occlusion. Therefore, the authors admitted that the ADP-induced thromboembolism model is not representative of any thrombotic process . This could be a possible explanation for the lack of effect of pharmacological PI3Kγ inhibition on GPVI-induced aggregation and thrombus formation.
Further evidence for the relevance of PI3Kβ in the development of thrombosis were provided by Sturgeon et al. using a Folts-like model of arterial thrombosis in rats . Selective PI3Kβ inhibition by TGX-221 reduced thrombus formation in the rat carotid artery by abolition of cyclic flow reductions, without affecting bleeding time or blood pressure . In contrast to antagonizing PI3Kβ activity, the pharmacologic inhibition of PI3Kδ by IC87114 had no antithrombotic effect . These observations are in line with the results of Gilio et al. and Mangin et al. showing PI3Kβ to play an important role in the development of thrombosis [44,83] and indicating this isoform to be a potential therapeutic target for the treatment of thrombosis by isoform selective “clean” PI3K inhibitors.
Taken together, these observations indicate the pharmacological inhibition of PI3Kα, β, and γ to be a new therapeutic strategy for preventing thrombotic complications during cardiovascular diseases. However, a better understanding of the activities of specific PI3K isoforms as well as a reduction of possible side effects is necessary to ensure the efficiency and safety of pharmacologic PI3K inhibitors in cardiovascular diseases.