Beneficial effects of statins on the microcirculation during sepsis: the role of nitric oxide
Academic Unit of Anaesthesia and Microcirculation Research Group, University of Sheffield, Royal Hallamshire Hospital, Sheffield S10 2JF, UK
* Corresponding author: Academic Unit of Anaesthesia and Microcirculation Research Group, University of Sheffield, K Floor, Royal Hallamshire Hospital, Sheffield S10 2JF, UK. E-mail: mdp04ccm{at}shef.ac.uk
 |
Abstract |
|---|
 Top Abstract IntroductionSepsisStatinsStatins and lipophilicitySepsis and NOStatins and changes in...Statins and endothelial...Statins and leucocyte...SummaryReferences |
|---|
This review describes the laboratory evidence and microvascular mechanisms responsible for the beneficial effects of statins in sepsis. During sepsis, changes occur within the microcirculation including alterations in arteriolar tone influencing blood pressure, adaptations to endothelial cell integrity causing leakage of proteins and macromolecules, and adhesion and migration of leucocytes through the vascular endothelium. Statins are widely used as cholesterol-lowering agents, but appear to have anti-inflammatory actions during sepsis. We have discussed the effects of statins on specific pathological processed within the microcirculation and focused on the role of nitric oxide (NO).
The main mechanism by which statins appear to be an effective treatment for sepsis is increased expression of endothelial nitric oxide synthase (eNOS), in conjunction with down-regulation of inducible nitric oxide synthase. Combined, this results in an increase in physiological concentrations of NO, thus restoring endothelial function. Laboratory studies have therefore suggested that enhancement of eNOS activity during sepsis may lead to restoration of microvascular tone, maintenance of microvascular integrity, and inhibition of cell adhesion molecules. However, other mechanisms independent of lipid-lowering effects, including antioxidant activity and alterations in the development of vascular atherosclerosis, may also contribute to the beneficial effects of statins. We have also addressed the influence on the effects of statins of lipid solubility and pre- and pro-phylactic administration.
Keywords: complications, sepsis; inhibitors, reductase, hydroxymethylglutaryl-CoA; microcirculation; nitric oxide; nitric oxide synthetase
|
Introduction |
|---|
|
TopAbstractIntroductionSepsisStatinsStatins and lipophilicitySepsis and NOStatins and changes in...Statins and endothelial...Statins and leucocyte...SummaryReferences |
|---|
The term microcirculation usually refers to capillaries, arterioles, and venules between 5 and 250 µm in diameter. Most endothelial cells are present in the microcirculation, as opposed to the larger arteries and veins, thus this system makes a significant contribution to the homeostasis of the cardiovascular system and body organs.83 The microcirculation serves as an important autocrine and paracrine organ, regulating vascular tone.
During sepsis, pathophysiological processes take place within the microcirculation, including changes in arteriolar tone influencing blood pressure, adaptations to endothelial cell integrity causing leakage of proteins and macromolecules, and adhesion and migration of leucocytes through the vascular endothelium. Such processes contribute to widespread tissue damage, multiple organ failure, and death and are potential therapeutic targets for treating sepsis. A number of microvascular mechanisms are responsible for these observed pathophysiological changes, of importance the altered balance between inducible nitric oxide synthase (iNOS) and endothelial nitric oxide synthase (eNOS): the two isoforms of NOS that catalyse nitric oxide (NO) production. This review will assess the laboratory evidence for the effects of statins on these microvascular adaptations occurring during sepsis.
|
Sepsis |
|---|
|
TopAbstractIntroductionSepsisStatinsStatins and lipophilicitySepsis and NOStatins and changes in...Statins and endothelial...Statins and leucocyte...SummaryReferences |
|---|
Animal models have been developed in the laboratory to induce the symptoms of sepsis. Cecal ligation and perfusion (CLP) is a surgical model activating multiple inflammatory mediators and producing a complex pathophysiological response, which is important as treatments that inactivate single inflammatory cascades are ineffective.21 However, studies utilizing CLP have limitations as severe complications occur, including decreased cardiac output and low peripheral resistance, causing all animals to die within hours or days.60 65 Other models include the administration of endotoxin or lipopolysaccharide (LPS) to animals.29 34 86 Rats and mice commonly receive a bolus dose of 5–15 mg kg–1 of LPS,1 7 68 inducing rapid hypotension and high mortality rates. In comparison, lower bolus doses of between 1 and 2 mg kg–1 LPS produce very low death rates in rodents, although hypotension is not demonstrated in all studies.29 85 86 Experiments in which rats are continually infused with a low dose of 150 µg kg–1h–1 LPS27 for many hours may be more clinically relevant as the pathophysiological response may represent the more insidious onset of sepsis. This model is ‘non-lethal’, but haemodynamic variables characteristic of sepsis such as hypotension and increase in haematocrit still occur.2 The ‘Schwarzman reaction’ is another laboratory model used in rodents to represent severe sepsis and involves administration of a low dose of LPS (5 µg).89 This is used as a ‘primer’ before a greater LPS challenge (300 µg) is given, inducing cardiovascular alterations such as platelet aggregation, vascular occlusion, and endothelial injury, representing disseminated intravascular coagulation. This process may complicate sepsis and is characterized by microvascular thrombosis and multiple organ failure.89
A less common sepsis model in rats involves the implantation of a sterile gauze swab ‘sponge’ into the subcutaneous space near the base of the tail.63 The swab is then injected with bacteria, commonly Escherichia coli, in order to establish a gradual onset septic model. The development of the continuous low-dose infusion provides a more controlled level of bacterial insult and although this model is less widely used, interesting findings have arisen.
|
Statins |
|---|
|
TopAbstractIntroductionSepsisStatinsStatins and lipophilicitySepsis and NOStatins and changes in...Statins and endothelial...Statins and leucocyte...SummaryReferences |
|---|
Statins or hydroxy methyl glutaryl-CoA (HMG-CoA) reductase inhibitors are widely used clinically as cholesterol-lowering agents because of their ability to block hepatic conversion of HMG-CoA to L-mevalonate and the production of isoprenoid geranylgeranylpyrophosphate (GGPP) (Fig. 1).52 55 57 90 102 108 Such administration results in decreased risk of coronary and cerebrovascular events and increased survival rates in patients with coronary artery disease.90
|
Several studies performed under low and normal cholesterol conditions have demonstrated that statins exert anti-inflammatory actions independent of their lipid-lowering effects.20 35 75 78 90 97 This has been the subject of many laboratory studies using statins to control inflammation during sepsis; many of which have demonstrated a decrease in mortality of septic animals, particularly after prior administration of statins.
|
Statins and lipophilicity |
|---|
|
TopAbstractIntroductionSepsisStatinsStatins and lipophilicitySepsis and NOStatins and changes in...Statins and endothelial...Statins and leucocyte...SummaryReferences |
|---|
Statins can be classified into lipophilic HMG-CoA reductase inhibitors (atorvastatin, simvastatin, cervastatin, fluvastatin, and lovastatin) and hydrophilic HMG-CoA reductase inhibitors (pravastatin and rosuvastatin). Hydrophilic statins display greatest uptake into hepatocytes compared with other cells.50 Thus, lipophilic statins may be more effective in their penetration of endothelial cells. However, this does not necessarily provide improved anti-inflammatory benefits, as increased permeability may have adverse effects because of the production of damaging reactive oxygen species.72 Thus, lipophilic statins may have adverse effects compared with hydrophilic varieties.72
|
Sepsis and NO |
|---|
|
TopAbstractIntroductionSepsisStatinsStatins and lipophilicitySepsis and NOStatins and changes in...Statins and endothelial...Statins and leucocyte...SummaryReferences |
|---|
Introduction to eNOS and iNOS
NO synthesis is mediated by the enzyme NOS that exists in several isoforms including the constitutive eNOS isozyme. eNOS is a calcium-dependent enzyme and forms NO in endothelial cells under physiological conditions. NO then diffuses into adjacent smooth muscle cells, activating guanyl cyclase, producing cyclic guanosine 5' monophosphate that mediates vascular relaxation via protein kinase G.101 Vascular inflammation and endothelial dysfunction in disease states such as sepsis are associated with a loss of eNOS-produced NO because of the vital role of this cascade in maintaining a non-thrombogenic endothelial cell surface.83
In sepsis, various mediators stimulate the overproduction of the inducible isoform, iNOS, which is inactive under normal physiological conditions. This overproduction has been well documented, including in vivo studies showing as increase in iNOS concentrations in tissues taken from septic animals.28 iNOS then acts to produce large amounts of NO. This high concentration of NO is thought to be beneficial because it is bactericidal66 and detrimental because it simultaneously causes vasodilation and acts as one of the key mediators of inflammation.29 This evidence also suggests that the homeostatic balance of iNOS and eNOS alters during the progression of sepsis, contributing to hypotension, multiple organ system failure, and mortality.1 For this reason, clinical trials that have used the non-specific NOS inhibitor NG-methyl-L-arginine to simply decrease the concentration of NOS have, to date, proved ineffective.33 107
The balance between eNOS and iNOS has been the subject of many in vivo studies, showing that endotoxin and cytokines stimulate iNOS expression, but decrease eNOS stability. Indeed, transgenic mice over-expressing eNOS in endothelial cells demonstrate a reduction in LPS-induced death.111 Such evidence, therefore, suggests that drugs acting to restore the original balance between the isoforms, rather than those that decrease or increase NOS, may have potential benefits for the treatment and prevention of sepsis.
This hypothesis, however, is contradicted by recent evidence postulating a pro-inflammatory role for eNOS, whereby studies in isolated rings of rat aorta and in macrophages have indicated that eNOS-derived NO is essential for maximal iNOS expression in the vasculature.14 101 Indeed, macrophages from eNOS knockout (KO) mice demonstrate reduced nuclear factor 
B (NFB) activity, iNOS expression, and NO production after LPS exposure, compared with wild-type mice,14 whilst blockade of eNOS prevented LPS-induced iNOS expression in endothelium- intact aortic rings.101 NOS activity and expression in tissue from mice receiving LPS and from CLP-operated rats have also shown that high concentrations of NO appear to ‘feedback’ and decrease eNOS levels.14 87 In contrast, in vitro studies in both human microglial cells and murine macrophages have indicated that NO inhibits iNOS levels through a negative feedback mechanism.13 69
Statins and the NO system: eNOS
In vitro studies using human endothelial cells have shown that simvastatin and lovastatin increase the half life of eNOS under low cholesterol conditions.49 51 Statins mediate such up-regulation by inhibition of HMG-CoA reductase and metabolites within the cholesterol biosynthetic pathway.51 83 The metabolite GGPP is involved in post-translational modifications of a variety of proteins, including the small GTPase Rho, causing them to become more lipophilic so that they can interact with the cell membrane.110 Geranylgeranylation by GGPP causes Rho to cycle between its inactive guanosine diphosphate (GDP)-bound state and its active guanosine triphosphate (GTP)-bound state (Fig. 2). This allows Rho to translocate from the cell cytoplasm, where it is inactive, to the membrane where it interacts with the cytoskeleton and other intracellular proteins to regulate endothelial integrity and gene expression. Bacterial activation of Rho appears to decrease eNOS expression.48 Therefore, as statins inhibit GGPP,55 109 down-regulation of Rho isoprenylation and activity may be the primary mechanism by which statins increase eNOS. It has also been reported that inhibition of Rho by bacterial toxins can increase eNOS concentrations.48 Furthermore, statins may block Caveolin-1, which interferes with the interaction between NO and calcium/calmodulin by sequestering eNOS into caveolae and inhibiting vascular relaxation.73
|
In vivo studies have demonstrated that the beneficial effects of statins on inflammation were absent in eNOS-deficient mice.20 An increase in the basal NO production in the picomolar range, in the case of eNOS, tends to restore endothelial function, whereas the large increase in the NO production in the nanomolar range, as is typical of iNOS activity, may promote cell damage. Therefore, an increase in eNOS by statins accompanied by a simultaneous reduction in iNOS activity could give endothelial protective effects.108
Statins and the NO system: iNOS
iNOS expression may be induced through the cytokine combination, tumour necrosis factor-
(TNF-) and interferon-
(IFN-), via the transcription factors NF-B and signal transducer/activator of transcription-1 (STAT-1).102 Studies in kidneys removed from cerivastatin-administered rats and rat myoblasts treated with simvastatin ex vivo have shown a statin-mediated reduction in iNOS expression.59 71 Further studies using segments of rat aorta demonstrated that this was because of inactivation of both of these transcription factors.103 In agreement, simvastatin decreases LPS-induced plasma nitrate levels in vivo, confirming that statins inhibit NO production because of decreased iNOS activity.29
Statins and the NO system: NO and post-translational modifications
Given the evidence for statin-mediated effects on eNOS and iNOS expression and activity, it follows that statins can decrease NO plasma concentrations in animals receiving LPS. Pre-treatment with statins before LPS administration appeared to down-regulate NO and cytokine synthesis in vivo, thus reducing mortality rate.1 70
In addition to alterations in eNOS and iNOS at the mRNA level, statins may modify proteins at the post-translational stage, leading to altered NO synthesis.35 110 Phosphatidylinositol-3 kinase (PI3-K) is an enzyme involved in several cellular processes related to cell growth, differentiation, and survival.67 PI3-K is attracted to phosphoinositides on the plasma membrane, which allows adenosine triphosphate transfer to the D-3 position of the phosphoinositide inositol ring.67 This transfer generates constitutive PI3P, present in all cells and PI3,4P and PI3,4,5P, which are usually absent in cells but are produced after PI3-K stimulation, recruiting a number of signalling molecules to the membrane, including 3-phosphatidylinositide-dependent kinase (PDK-1). PDK-1 phosphorylates and activates Akt, enabling it to act as an important effector of the PI3-K pathway. Akt further phosphorylates eNOS, leading to Ca2+-independent enzyme activation, thereby increasing NO synthesis.25
The role of Akt in sepsis is controversial, as mesenteric arteries from LPS-treated rabbits displayed decreased Akt phosphorylation and eNOS expression and phosphorylation.62 This was perhaps as a result of reduced translocation of PI3-K to the plasma membrane during sepsis, as PI3-K is an upstream step of the Akt pathway.62 Survival was improved in CLP mice over-expressing Akt.9 However, in vitro studies using human macrophages demonstrated that LPS activates Akt and in turn eNOS.67 This was again postulated as evidence for a pro-inflammatory role for eNOS, leading to an increase in NO production and iNOS expression.14
There is considerable evidence that statins may stimulate the activity of PI3-K, thereby phosphorylating downstream Akt and activating eNOS. Vascular tissue from fluvastatin-administered rabbits before LPS treatment demonstrated a 2.6- and 2-fold increase in phosphorylated Akt and eNOS, respectively, in comparison with control animals.62 This was partially because of attenuating the reduction in membrane translocation of PI3-K, as determined using the PI3-K inhibitor, wortmannin.62 Hence, the Akt pathway is a further mechanism by which statins may exert their endothelial-mediated effects.35 110
Statins: pre- or pro-phylactic treatment?
Several studies have investigated whether statins are most effective as a pre- or pro-phylactic treatment during sepsis. Pre-treatment with cerivastatin reduced mortality rate to 26.7% compared with 73.4% in mice administered LPS alone.1 Mice treated with simvastatin, 18 and 3 h before induction of CLP, also demonstrated extended survival time of 108 h in comparison with 28 h in the untreated group.65 Studies assessing pro-phylactic administration of statins in vivo have had less promising results, in that cerivastatin was shown to be ineffective in reducing mortality when administered after LPS injection.1 In addition, animals receiving statins after CLP demonstrated decreased mortality, but not to the same extent as pre-treatment.64 These findings may be dependent on the type of statin; interestingly, the greatest extension in survival time was seen with pravastatin (74%) compared with atorvastatin (70%), simvastatin (61%), and fluvastatin, which showed no significant change in survival time.65 The reason for this is unclear and requires further experimentation.
Lipophilic and hydrophilic properties of statins
Very few studies have formally distinguished between the effects of lipophilic and hydrophilic statins on the NO system and during sepsis. Both in vitro and in vivo experiments have demonstrated that hydrophilic rosuvastatin up-regulates eNOS expression at levels comparable with the lipophilic statins. This may be because of the well-documented higher potency and half life of the hydrophilic group.50
Ex vivo experimentation has also shown that lipophilic statins, but not hydrophilic varieties, increase iNOS expression and NO synthesis.40 However, given the body of evidence demonstrating that statins decrease iNOS levels, such a finding is controversial and must be investigated further. On the other hand, if decreased iNOS levels are beneficial in sepsis, hydrophilic statins may prove to be a more effective treatment.
Nitrostatins
Recently, a new class of statins known as nitrostatins has been developed. This includes nitropravastatin that comprises of a NO-donating moiety and was developed to enhance the anti-inflammatory actions of conventional statins,80 whose effectiveness are largely dependent on the concentration of NO being released.13 69 To the best of our knowledge, there are no studies assessing the effects of nitrostatins on sepsis, but given their possible anti-inflammatory effects these agents may provide exciting new treatments.
Statins and NO: summary
NO has a major role during sepsis and thus the balance of the enzymatic isoforms catalysing its synthesis is a major target for drugs in this inflammatory condition. Statins may cause concomitant up-regulation of eNOS and down-regulation of iNOS, although the evidence for such activity is controversial. Statins may therefore restore vascular responsiveness during sepsis via re-establishment of a favourable balance between eNOS and iNOS.29 There is also considerable evidence indicating that statins are more effective during sepsis when administered as a pre-treatment, although further in vivo and clinical studies are required to elucidate the mechanisms responsible. The differences between lipophilic and hydrophilic statins and the effects of new generations of statins on NO during sepsis also require clarification.
|
Statins and changes in vascular reactivity |
|---|
|
TopAbstractIntroductionSepsisStatinsStatins and lipophilicitySepsis and NOStatins and changes in...Statins and endothelial...Statins and leucocyte...SummaryReferences |
|---|
Cardiovascular alterations in sepsis include a change in arteriolar tone, which under physiological conditions is responsible for maintaining blood pressure and blood flow to organs. As arteriolar tone deteriorates, severe hypotension and inadequate organ perfusion are induced by endotoxin and cytokines, causing organ failure and in some cases, death.93 100 Increased endothelial-dependent production of NO by iNOS contributes to this hypotensive response, but also regulates adjacent vascular smooth muscle cell contractility. Thus, statins may also affect vascular reactivity secondary to alterations in eNOS and iNOS.
Both in vitro and in vivo studies have indicated that change in arteriolar tone only occurs in the later stages of sepsis (defined here > 4 h).30 31 93 Such alterations may be because of an imbalance between vasodilator and vasoconstrictor influences, rather than just an overproduction of vasodilators.100 Importantly, there may also be a failure of the endothelial-independent mechanisms regulating vascular smooth muscle contractility, but this review focuses on the endothelial-dependent mechanisms and NO pathways that are relevant to statins.
Rats rendered septic using the sponge model demonstrated impaired norepinephrine (NE)-induced vasoconstriction at 24 and 72 h.114 This decrease in sensitivity was further aggravated by a second LPS insult at 48 h. 2-adrenoceptor activity is also greatly attenuated during endotoxic shock.4 However, the role of an 1-adrenoceptor activity is controversial as both no change and an increase in responsiveness have been demonstrated during sepsis.4 27 Such differences may be because of differences in models, variations in dose, and the duration of sepsis. A change in the 1-adrenoceptor activity has been identified in the chronic, non-lethal LPS model, which may cause cardiovascular alterations more closely matched to those occurring in sepsis in patients, including hypotension.2 27
Consequently, agents that increase adrenergic responsiveness have great therapeutic potential, particularly as epinephrine and NE are administered to septic patients to support the failing myocardium and maintain blood pressure. In rats, it has also been determined that pre-treatment with simvastatin in vivo increased arterial constriction to phenylephrine during LPS administration.29
|
Statins and endothelial integrity |
|---|
|
TopAbstractIntroductionSepsisStatinsStatins and lipophilicitySepsis and NOStatins and changes in...Statins and endothelial...Statins and leucocyte...SummaryReferences |
|---|
The inflammatory processes associated with sepsis induce damage to the endothelial layer. Large gaps form between the cells resulting in a loss of barrier integrity to proteins and other macromolecules,83 known as macromolecular leak (Fig. 3). Several mechanisms may be responsible for such a phenomenon, including the release of pro-inflammatory mediators and alterations in the pathways/molecules maintaining tight and gap junctions,47 which will be discussed in greater detail.
|
Pro-inflammatory mediators
The pro-inflammatory cytokines involved in the compromise of endothelial integrity during sepsis include TNF-
, which includes disruption of the lung endothelium in patients leading to increase in permeability to protein.81 Sheep injected with recombinant human TNF- display a septic-like state, whereas antibodies against TNF- have reduced mortality in vivo primate models of sepsis.77 94 In addition, IL-1 type 1 receptor KO mice maintain endothelial integrity after the administration of endotoxin, compared with wild-type mice, which display extensive endothelial damage.91 IL-6 is also an important mediator of loss of barrier integrity during sepsis. In a murine CLP model of sepsis, IL-6 KO mice demonstrated decreased intestinal permeability in comparison with the wild-type group.105 Laboratory experiments investigating the role of IL-6 on endothelial integrity are limited and thus this may be an area deserving of both in vitro and in vivo investigation.
In addition to the release of pro-inflammatory cytokines, the septic response is also characterized by the release of anti-inflammatory mediators, including IL-10. IL-10 appeared to be protective during LPS administration in mice.638 A CLP model of sepsis in IL-6 KO mice demonstrated both increased IL-10 levels and reduced mucosal cell permeability.105 Rats administered simvastatin displayed an increase in IL-10 expression, whilst expression of the pro-inflammatory cytokines IL-1ß, IL-6, and TNF- was decreased in rat hearts.113 These are very promising findings given the anti-inflammatory properties of IL-10. Clinical trials have shown that agents acting to down-regulate TNF- and IL-1ß are ineffective in reducing mortality in sepsis22 23 and therefore the down-regulatory effects of statins on cytokines could contribute to their beneficial effects during sepsis.
Tight junctions
Vascular endothelial cells are ‘held together’ mainly by tight junctions and gap junctions.24 Tight junctions are complex structures composed of integral membrane proteins that interact with intracellular proteins close to the cytoplasmic surface5 and tight junctions in the membranes of adjacent cells.82 The maintenance of a close association between tight junctions and the endothelial actin cytoskeleton is vital in maintaining the structural integrity of the endothelium.47 During sepsis, pro-inflammatory mediators instigate a contraction at the margins of the cell, causing gaps to form, thus increasing the permeability of the endothelium. The integral membrane proteins composing tight junctions include occludin, claudin, and junctional adhesion molecule (JAM).
Occludin is a 65-kDa protein, which is expressed specifically in tight junctions.95 The protein comprises four transmembrane domains with two extracellular loops and intracellular termini.5 95 The COOH terminus of occludin interacts with several cytosolic proteins, including zona occludens (ZO). ZO bonds with actin to form a linkage between occludin and the cytoskeleton (Fig. 4). Rho GTPases exert their effects via the cytoskeleton and, therefore, Rho is an important mediator of tight junction function and endothelial cell permeability via this pathway.
|
The claudin protein family has 20 members, each with a molecular mass of 20–25 kDa. Claudins localize to the cell membrane and form tight junction-like structures in fibroblasts that physiologically lack tight junctions.26 Interactions between occludin and claudin may modulate endothelial permeability, as indicated by the incorporation of occludin into claudin strands after transfection of the DNA for the two proteins into mouse fibroblasts.26
JAM-A is a 32-kDa glycoprotein that belongs to a family of cell adhesion molecules, including JAM-B and JAM-C. These proteins localize close to tight junctions at the intracellular junctions of endothelial cells and comprise a single transmembrane segment with an extracellular region and a cytoplasmic tail. The JAM family interacts with dimers expressed on the surface of adjacent cells, but it remains to be determined if these proposed interactions have an impact on endothelial cell permeability.
To the best of our knowledge, all information on the effects of statins on tight junctions has been obtained within neural tissues. Thus, studies assessing statin administration on tight junction protein expression and function may have less relevance to macromolecular leak. However, simvastatin decreased LPS-induced albumin extravasation into the lungs in vivo via Rho.43 In vitro studies have also demonstrated that simvastatin reduced endothelial permeability to Evans Blue, a finding attributed to blockade of the GGPP-dependent isoprenylation of Rho.97 Hence, perhaps surprisingly, the use of statins during sepsis may contribute to reduced macromolecular leakage via tight junctions, but their effects are not fully understood, nor is the role of NO.
Gap junctions
Gap junctions form aqueous pores between the membranes of two adjacent cells, permitting the movement of small molecules from one cell to another. The term ‘gap’ refers to the space between the plasma membrane of two adjacent cells, formed because of the presence of gap junctions embedded in the membranes. Studies examining tight junctional protein expression have indicated that arterial endothelial cells express 18-fold more occludin protein compared with venous endothelial cells.45 Hence, as venules are the site of macromolecular leak, gap junctions may have more importance for studying the mechanisms by which statins potentially maintain vascular integrity during sepsis.
Each gap junction is formed by a hexamer of the protein connexin, called a connexon (Fig. 5). There are approximately 20 different subunits of connexin expressed in mammalian cells;104 at least 4 of which are expressed in the vascular endothelium [Connexin 37 (C x 37), C x 40, C x 43, and C x 45].41
|
In vivo studies have shown that adaptations occur in gap junction protein expression and function in both LPS and CLP models of sepsis.17 79 88 Interestingly, KO mice for C x 43 display an altered phenotype compared with wild-type animals, including bradycardia and hypotension,56 whereas C x 40 KO and double KO mice for C x 43 and C x 40 display hypertension.18 19 42 Modulation of connexin expression and gap junction function could also have a role in the loss of endothelial integrity during sepsis. Studies investigating the activity of statins in this area are limited, but there is in vivo evidence that C x 40 and C x 43 levels were increased after arteriolar injury, an effect which was decreased by lovastatin and fluvastatin administration.104
Gap junction formation is regulated via interaction with adherens junctional cell adhesion molecules including vascular endothelial (VE)-cadherin and catenin (Fig. 6). VE-cadherin is a transmembrane protein that forms homodimers with VE-cadherin proteins from neighbouring cells. ß-Catenins interact with the intracellular segment of the VE-cadherin protein and -catenin, which in turn binds to the cell cytoskeleton.99 Catenins may also interact with the cytoskeleton via ZO. Such interactions appear to increase cellular adhesion whereas disruption of these relationships appears to weaken VE-cadherin-mediated cellular interactions.99 The dependency of gap junctions on these proteins was demonstrated by a reduction in gap junctions in cardiomyocytes from cadherin KO mice.58 However, VE-cadherin dominant negative constructs displayed disrupted gap junction formation in rat cardiomyocytes.36 NO donors significantly reduced the amount of VE-cadherin and catenin expressed in murine microvascular endothelial cells in vitro.32 This finding was correlated with an increase in vascular permeability, both in vitro and in vivo, as quantified by Evans Blue leakage from cutaneous vessels in mice.32 This study has relevance to sepsis, given the large release of NO in this disease state, but the interactions of statins are uncertain. Of note, TNF-, a key pro-inflammatory mediator during sepsis, also influences VE-cadherin via mediation of tyrosine phosphorylation.3
|
In summary, there are multiple processes contributing to endothelial layer damage during sepsis. Whilst studies into the effect of statins on pro-inflammatory mediators and the small GTPases are fairly well documented, the effects of statins on the modification of gap junctions during sepsis are yet to be elucidated.
|
Statins and leucocyte–endothelial interactions |
|---|
|
TopAbstractIntroductionSepsisStatinsStatins and lipophilicitySepsis and NOStatins and changes in...Statins and endothelial...Statins and leucocyte...SummaryReferences |
|---|
The presence of endotoxin in the blood acts to up-regulate leucocyte–endothelial interactions. This is a step-wise process requiring specific adhesion molecules to be expressed on the surface of endothelial cells and leucocytes: first leucocytes roll along the vascular endothelium (Fig. 7), followed by firm adhesion and migration into tissue (Fig. 8).
|
|
Once migration has occurred, tissue damage can result because of release of superoxide and trans-endothelial leucocyte proteases.15 The release of pro-inflammatory cytokines including TNF-
, IL-1, and IL-6 then further increase expression of adhesion molecules. Rolling involves the selectin family of adhesion molecules, which includes three proteins designated by the prefixes E (endothelial), P (platelet) and L (leucocyte).11 P-selectin and E-selectin are expressed on the endothelial surface, whereas L-selectin is expressed on the leucocytes.11 75 The counter receptors for the selectins are believed to be carbohydrate-containing molecules such as P-selectin glycoprotein ligand-1 (PSGL-1).61 Rolling is initiated as cytokines up-regulate synthesis of E-selectin and ligands for L-selectin that are responsible for leucocyte recruitment. As rolling begins, the leucocytes change shape, shed L-selectin, and up-regulate ß2 integrins that are responsible for tight adhesion with endothelial cells.75
Firm adhesion involves the ß2 integrins CD11b and CD18 (LFA-1; lymphocyte function antigen-1), which are expressed on leucocytes and bind to intracellular adhesion molecule-1 (ICAM-1) on the vascular endothelium.54 ICAM-1 is responsible for firm attachment and may be followed by trans-endothelial migration of leucocytes into inflamed tissue via platelet/endothelial cell adhesion molecule-1 (PECAM-1).75 106 ICAM-1 KO mice show a significant reduction in CLP-induced mortality compared with the wild-type mice.96 However, P-selectin/ICAM-1 KO mice did not show a reduction in leucocyte migration in comparison with wild-type mice after CLP. Leucocyte–endothelial interactions are therefore crucial in the occurrence of detrimental microvascular responses to sepsis.
Rho has also been implicated in the expression of leucocyte–endothelial adhesion molecules, as inhibition of Rho decreases LPS-induced ICAM-1 expression on endothelial cells.10 Rho, in its activate conformation, interacts with the cytoskeleton, activating NF-B, thus allowing it to migrate to the nucleus where it promotes expression of proteins involved in leucocyte expression, including ICAM-1, VCAM-1, and E-selectin.10 Eight hours of simvastatin administration inhibited transmigration of THP-1 monocytic cells in vitro, via inhibition of Rho as a result of upstream blockade of mevalonate and GGPP.74 Short-term treatment with simvastatin, however, did not cause inhibition,74 perhaps because of the extended periods of time required for post-translational modification of the geranylgeranylated proteins. Nevertheless, this study indicated that statin administration increased expression of ICAM-1 and VCAM-1 on endothelial cells;74 an unexpected result given that decreased transmigration often correlates with decreased adhesion molecule expression. Several studies contradicted this finding, as in vitro administration of statins down-regulated expression of ICAM-1 and VCAM-1.76 92 112 This evidence suggests that statins inhibit Rho because of the upstream blockade of mevalonate and, in doing so, may have down-regulatory effects on the adhesion molecules mediating leucocyte rolling, firm adhesion, and transmigration.
At physiological concentrations, NO is a potent inhibitor of leucocyte rolling and adhesion during inflammation in vivo.90 Indeed, eNOS KO mice do not exhibit leucocyte–endothelial interactions.90 Furthermore, statin protection against leucocyte–endothelial interactions is also completely abolished after inhibition of eNOS with N-nitro-L-arginine methyl ester (L-NAME).49 iNOS may also inhibit leucocyte–endothelial interactions, as iNOS KO mice display increased LPS-mediated leucocyte rolling and adhesion in comparison with the wild-type mice.37 However, iNOS KO mice treated with LPS demonstrated no differences in microvascular expression in P-selectin, E-selection, or ICAM-1 level,37 indicating that alterations in adhesion are not because of the effect of iNOS on adhesion molecule expression. This disagrees with in vitro studies whereby NO inhibited ICAM-1, VCAM-1, and E-selectin expression in both LPS- and cytokine-stimulated human endothelial cells.16 46 103 This anomaly may be because of differences between adhesion molecules within the microvasculature vs those on circulating leucocytes or endothelial cells in vitro. iNOS may therefore have limited relevance to leucocyte–endothelial expression in the microvasculature, but this remains uncertain.8
Simvastatin also inhibits leucocyte–endothelial rolling, adherence, and transmigration in post-capillary venules in response to inflammatory stimuli: an effect which increased with increased time of pre-treatment.75 Pre-phylactic statin administration may be of relevance for surgical procedures where postoperative sepsis is common, such as coronary artery bypass grafting.12 39 In vitro studies have demonstrated that statin-mediated inhibition of leucocyte–endothelial interactions may be because of LFA-1 and ICAM-1.44 84 90 LFA-1 deficient mice, the statin-derived LFA-1 inhibitor LFA703, and antibodies against LFA-1 have all been used in in vivo models to demonstrate a reduction in leucocyte adhesion in comparison with control groups.53 54 65
In summary, statins have potential for preventing leucocyte–endothelial interactions during sepsis, including inhibition of Rho, leading to down-regulation and interference with adhesive cell surface molecules, and modulation of NO, but the precise mechanisms by which statins may exert such effects have not been fully elucidated.
|
Summary |
|---|
|
TopAbstractIntroductionSepsisStatinsStatins and lipophilicitySepsis and NOStatins and changes in...Statins and endothelial...Statins and leucocyte...SummaryReferences |
|---|
We have discussed a number of in vitro and in vivo laboratory studies documenting that statins may be effective in treating sepsis via three important pathways implicated in inflammation (Fig. 9). We have also discussed how the beneficial effects of statins may be mediated via NO. We have focused on mechanisms that are relevant at the microvascular level. The laboratory studies we describe provide optimism that the use of statins will become effective in clinical treatment. We have also highlighted the benefits of pre- vs pro-phylactic treatment and the importance of choice of statin. Benefits of hydrophilicity vs lipophilicity are yet to be fully elucidated during sepsis.
|
The main mechanism by which statins appear to be an effective treatment for sepsis is increased expression of eNOS, in conjunction with down-regulation of iNOS. Combined, this results in an increase in physiological concentrations of NO, thus restoring endothelial function. Laboratory studies have therefore suggested that enhancement of eNOS activity during sepsis may lead to restoration of microvascular tone, maintenance of microvascular integrity, and inhibition of cell adhesion molecules (Fig. 9). However, other mechanisms independent of lipid-lowering effects may also contribute to the beneficial effects of statins. These include antioxidant activity and alterations in the development of vascular atherosclerosis.98 Further laboratory studies are now required to uncover the full potential of this group of drugs during sepsis.
|
References |
|---|
|
TopAbstractIntroductionSepsisStatinsStatins and lipophilicitySepsis and NOStatins and changes in...Statins and endothelial...Statins and leucocyte...SummaryReferences |
|---|
1 Ando H, Takamura T, Ota T, Nagai Y, Kobayashi K. (2000) Cerivastatin improves survival of mice with lipopolysaccharide-induced sepsis. J Pharmacol Exp Ther 294:1043–6.
2 Andrew PS and Kaufman S. (2001) Splenic denervation worsens lipopolysaccharide-induced hypotension, hemoconcentration, and hypovolemia. Am J Physiol Regul Integr Comp Physiol 280:R1564–72.
3 Angelini DJ, Hyun SW, Grigoryev DN, et al. (2006) TNF{alpha} increases tyrosine phosphorylation of vascular endothelial-cadherin opens the paracellular pathway through fyn activation in human lung endothelia. Am J Physiol Lung Cell Mol Physiol 291:L1232–45.
4 Baker CH, Sutton ET, Price JM, Ortiz-Tweed M, Nessellroth S. (1994) Attenuation of arteriolar alpha2-adrenoceptor sensitivity during endotoxemia. Am J Physiol 267:H2171–8.
5 Bazzoni G. (2006) Endothelial tight junctions: permeable barriers of the vessel wall. Thromb Haemost 95:36–42.[ISI][Medline]
6 Berg DJ, Kuhn R, Rajewsky K, et al. (1995) Interleukin-10 is a central regulator of the response to LPS in murine models of endotoxic shock the Shwartzman reaction but not endotoxin tolerance. J Clin Invest 96:2339–47.[ISI][Medline]
7 Bernard C, Merval R, Esposito B, Tedgui A. (1998) Resistance to endotoxin shock in spontaneously hypertensive rats. Hypertension 31:1350–6.
8 Binion DG, Fu S, Ramanujam KS, et al. (1998) iNOS expression in human intestinal microvascular endothelial cells inhibits leukocyte adhesion. Am J Physiol 275:.
9 Bommhardt U, Chang KC, Swanson PE, et al. (2004) Akt decreases lymphocyte apoptosis and improves survival in sepsis. J Immunol 172:7583–91.
10 Cernuda-Morollon E and Ridley AJ. (2006) Rho GTPases and leukocyte adhesion receptor expression and function in endothelial cells. Circ Res 98:757–67.
11 Chandra A, Enkhbaatar P, Nakano Y, Traber LD, Traber DL. (2006) Sepsis: emerging role of nitric oxide and selectins. Clinics 61:71–6.[Medline]
12 Chello M, Goffredo C, Patti G, et al. (2005) Effects of atorvastatin on arterial endothelial function in coronary bypass surgery. Eur J Cardiothorac Surg 28:805–10.
13 Colasanti M, Persichini T, Menegazzi M, et al. (1995) Induction of nitric oxide synthase mRNA expression. Suppression by exogenous nitric oxide. J Biol Chem 270:26731–3.
14 Connelly L, Madhani M, Hobbs AJ. (2005) Resistance to endotoxic shock in endothelial nitric-oxide synthase (eNOS) knock-out mice: a pro-inflammatory role for eNOS-derived no in vivo. J Biol Chem 280:10040–6.
15 Croner RS, Kulu Y, Hoerer E, et al. (2005) Intravenous glycine after cecal ligation and puncture has no effect on impaired hepatic microperfusion, leukocyte adhesion, and mortality in septic rats. Microvasc Res 69:71–8.[CrossRef][ISI][Medline]
16 De Caterina R, Libby P, Peng HB, et al. (1995) Nitric oxide decreases cytokine-induced endothelial activation. Nitric oxide selectively reduces endothelial expression of adhesion molecules and proinflammatory cytokines. J Clin Invest 96:60–8.[ISI][Medline]
17 De Maio A, Gingalewski C, Theodorakis NG, Clemens MG. (2000) Interruption of hepatic gap junctional communication in the rat during inflammation induced by bacterial lipopolysaccharide. Shock 14:53–9.[ISI][Medline]
18 De Wit C, Hoepfl B, Wolfle SE. (2006) Endothelial mediators and communication through vascular gap junctions. Biol Chem 387:3–9.[CrossRef][ISI][Medline]
19 De Wit C, Wolfle SE, Hopfl B. (2006) Connexin-dependent communication within the vascular wall: contribution to the control of arteriolar diameter. Adv Cardiol 42:268–83.[ISI][Medline]
20 Endres M, Laufs U, Huang Z, et al. (1998) Stroke protection by 3-hydroxy-3-methylglutaryl (HMG)-CoA reductase inhibitors (statins) unrelated to cholesterol lowering: new concepts for cardiovascular disease. Proc Natl Acad Sci USA 95:8880–5.
21 Eskandari MK, Bolgos G, Miller C, Nguyen DT, DeForge LE, Remick DG. (1992) Anti-tumor necrosis factor antibody therapy fails to prevent lethality after cecal ligation and puncture or endotoxemia. J Immunol 148:2724–30.[Abstract]
22 Fisher C, Opal S, Lowry S, et al. (1994) Role of interleukin-1 and the therapeutic potential of interleukin-1 receptor antagonist in sepsis. Circ Shock 44:1–8.[Medline]
23 Fisher CJ Jr, Opal SM, Dhainaut JF, et al. (1993) Influence of an anti-tumor necrosis factor monoclonal antibody on cytokine levels in patients with sepsis. The CB0006 Sepsis Syndrome Study Group. Crit Care Med 21:318–27.[ISI][Medline]
24 Fukuhra S, Sakurai A, Yamagishi A, Sako K, Mochizuki N. (2006) Vascular endothelial cadherin-mediated cell-cell adhesion regulated by a small GTPase, Rap1. J Biochem Mol Biol 39:132–9.[ISI][Medline]
25 Fulton D, Gratton JP, McCabe TJ, et al. (1999) Regulation of endothelium-derived nitric oxide production by the protein kinase Akt. Nature 399:597–601.[CrossRef][Medline]
26 Furuse M, Sasaki H, Fujimoto K, Tsukita S. (1998) A single gene product, claudin-1 or -2, reconstitutes tight junction strands and recruits occludin in fibroblasts. J Cell Biol 143:391–401.
27 Gardiner SM, Kemp PA, March JE, Bennett T. (1995) Cardiac and regional haemodynamics, inducible nitric oxide synthase (NOS) activity, and the effects of NOS inhibitors in conscious, endotoxaemic rats. Br J Pharmacol 116:2005–16.[ISI][Medline]
28 Gardiner SM, Kemp PA, March JE, Bennett T. (1996) Temporal differences between the involvement of angiotensin II and endothelin in the cardiovascular responses to endotoxaemia in conscious rats. Br J Pharmacol 119:1619–27.[ISI][Medline]
29 Giusti-Paiva A, Martinez MR, Felix JV, et al. (2004) Simvastatin decreases nitric oxide overproduction and reverts the impaired vascular responsiveness induced by endotoxic shock in rats. Shock 21:271–5.[CrossRef][ISI][Medline]
30 Glembot TM, Britt LD, Hill MA. (1995) Lack of direct endotoxin-induced vasoactive effects on isolated skeletal muscle arterioles. Shock 3:216–23.[ISI][Medline]
31 Glembot TM, Britt LD, Hill MA. (1996) Endotoxin interacts with tumour necrosis factor-alpha to induce vasodilation of isolated rat skeletal muscle arterioles. Shock 5:251–7.[ISI][Medline]
32 Gonzalez D, Herrera B, Otero BK, Quintero G, Rojas A. (2003) Nitric oxide disrupts VE-cadherin complex in murine microvascular endothelial cells. Biochem Biophys Res Commun 304:113–8.[CrossRef][ISI][Medline]
33 Grover R, Zaccardelli D, Colice G, Guntupalli K, Watson D, Vinc