I intend to focus on the pharmacology of the endothelial prostacyclin⁄ nitric oxide radical PGI2⁄ Keywords ACE-I; ASA; bradykinin; endothelial dysfunction; nitric oxide; prostacyclin; st
Trang 1Pharmacology of vascular endothelium
Delivered on 27 June 2004 at the 29th FEBS Congress in Warsaw
Ryszard J Gryglewski
Jagiellonian University, Cracow, Poland
Sir Hans Krebs was one of the most versatile
biochem-ists of the twentieth century Many of his sayings stay
as bright as his science My favourite quotation is: ‘‘…
we all are committed to correlating biochemical events
to function… the point I want to make is that it is not
always immediately clear what their relevance to
func-tion may be…’’ [1] Indeed, a burning desire for
imme-diate comprehension, amplified by the abomination of
being just another fact collector may overcome rational
cautiousness We pharmacologists know this only too
well Sir John Vane urged his young followers: ‘‘Do simple experiments and make simple hypotheses – there are plenty of others who will come along and show how much more complicated the answer really is …’’ [2] Keeping in mind the above advice, I present the vascular endothelium as a newly discovered target for the pharmacotherapy of arterial hypertension, athero-thrombosis and diabetic angiopathies
I intend to focus on the pharmacology of the endothelial prostacyclin⁄ nitric oxide radical (PGI2⁄
Keywords
ACE-I; ASA; bradykinin; endothelial
dysfunction; nitric oxide; prostacyclin;
statins; thienopyridines
Correspondence
R J Gryglewski, Jagiellonian University,
Kasztelan˜ska 30, 30-116 Cracow, Poland
E-mail: mfgrygle@cyf-kr.edu.pl
(Received 10 February 2005, revised
13 April 2005, accepted 20 April 2005)
doi:10.1111/j.1742-4658.2005.04725.x
Sir John Vane named vascular endothelium ‘the maestro of blood circula-tion’ Recently, ‘the maestro’ has become a target for pharmacotherapy of atherothrombotic and diabetic vasculopathies with well known cardio-vascular drugs belonging to the families of Angiotensin Converting Enzyme inhibitors, HMG CoA reductase inhibitors or b1-Adrenoceptor antagonists These drugs became upgraded to a position of the pleiotropic endothelial drugs It is not a simple verbal change in the nomenclature It means that these drugs apart from their well defined mechanisms of action, as indi-cated in their regular names, in addition they act in an unknown mechan-ism at the level of vascular endothelium preventing angina, myocardial infarction and stroke Many biochemical events take place in endothelial cells I chose for a closer inspection the nitric oxide/prostacyclin defensive system to explain the endothelial pleiotropism of the drugs in question I tried to examine the validity of this conception according to the general rule: in vitro cognitio sed in vivo veritas
Abbreviations
AA, arachidonic acid; ACE-I, angiotensin converting enzyme (and kininase 2) inhibitors; ADMA, asymmetric dimethylarginine; ASA,
acetylsalicylic acid; BH4, tetrahydrobiopterin; Bk, bradykinin; BPF, bradykinin potentiating factor; CAD, coronary heart disease; CaM, calmodulin; COX-1, constitutive cyclooxygenase 1; COX-2, inducible cyclooxygenase 2; EDHF, endothelium-derived hyperpolarizing factor; EDRF, endothelium-derived relaxing factor; EETs, cis-epoxyeicosatrienoic acids; eNOS, constitutive endothelial nitric oxide synthase; FAD, flavin adenine dinucleotide; FMD, flow mediated dilatation (of brachial artery in humans); FMN, flavin mononucleotide; HMG-CoA,
hydroxymethylglutaryl coenzyme A; HO-1, inducible heme oxygenase; 15-HPAA, 15-hydroperoxyarachidonic acid; HYHC, hyperhomo-cysteinemia; 6-keto-PGF1a, prostaglandin 6-keto-PGF1a, a stable product of decomposition of PGI2; LDL, low-density lipoproteins; L -NAME,
L -N(G)-nitroarginine methyl ester, a nonselective NOS inhibitor; NOHA, N ’ -hydroxy-Arg; ONOO – , peroxynitrite; ox-LDL, oxidized low-density lipoproteins; PARP, poly ADP ribosyl polymerase; PGE 2 , prostaglandin E 2 ; PGHS2, PGH 2 synthase; PGI 2 , prostacyclin; PGIS, prostacyclin synthase; RNS, reactive nitrogen species; ROS, reactive oxygen species; SDMA, symmetric dimethylarginine; TXA2, thromboxane A2; TXAS, thromboxane A 2 synthase; TXB 2 , thromboxane B 2
Trang 2) defence system Other aspects of endothelial
biology are reviewed by Nachman and Jaffe [3] with
a special attention being paid to the functioning of
Weibel–Palade bodies and their response to
proin-flammatory or prothrombotic agents as manifested by
the release of von Willebrand factor, P selectin and
interleukin-8 Readers interested in the endothelial
mitochondrion as a propagator of oxidative stress [4]
and the mitochondrion-oriented role of reactive
oxy-gen species (ROS) and hydrooxy-gen peroxide [5] are
directed to studies by Keaney and coworkers [4,5]
Mitochondrial oxidases along with NAD(P)H oxidase,
xanthine oxidase and uncoupled constitutive
endothel-ial nitric oxide synthase (eNOS) constitute the source
of endothelial ROS, which may act as modulators of
tone, growth and remodelling of the vascular wall It
may well be that inflammation plays a primary role
in atherogenesis, whereas oxidative stress is a
secon-dary phenomenon [6] At low concentrations, ROS
may protect endothelial cells against apoptotic
beha-viour [7] Long-term treatment with antioxidant
vita-mins does not influence the course of the disease
or correct endothelial dysfunction in patients with
atherosclerosis [8] The great expectations for the
therapeutic use of antioxidants in patients with
athero-sclerosis need to be re-examined
Endothelium as the endocrine organ
Why does blood not coagulate within healthy blood
vessels? This question has been addressed for centuries
The warmth of the body (Plato), the lack of contact
with air (James Hewson) and the vital power of blood
(John Hunter) have all been claimed as reasons The
truth is that vascular endothelium secretes a bunch of
antithrombotic and thrombolytic mediators that keep
blood fluid within an undamaged circulatory system
Vascular endothelium is neither a ‘primitive
mem-brane’, as claimed by Rudolph von Virchow, nor a
‘nucleated sheet of cellophane’, as Sir Howard Florey
stated [9] Sir John Vane named the endothelium ‘the
maestro of blood circulation’ [10], which should be
viewed as a peculiar dissipated endocrine organ (mass
1000 g, surface area 100 m2) Among others
sub-stances, endothelium releases into the passing blood –
labile, lipophilic and antithrombotic local hormones
like PGI2and NO•
as well as a peptide – tissue plasmi-nogen activator These prevent the build up of thrombi
and disperse any thrombi at an early stage of their
for-mation This is why blood stays fluid within a healthy
vascular bed The inherent chemical instability of PGI2
and NO•
allows for the immediate transformation of
extravasated blood into a haemostatic plug
Unfortu-nately, the same transformation may occur locally inside the circulatory system of patients with athero-sclerotic plaques or diabetic angiopathies The endo-thelium then loses its protective properties and may even produce proinflammatory and thrombogenic agents (endothelial dysfunction)
Endothelium generates many biologically active sub-stances other than PGI2 or NO•
, to mention just four regioisomers of cis-epoxyeicosatrienoic acid (EETs) produced from AA by CYP2J2 epoxygenases [11] EETs are vasoprotective vasodilators Some may be responsible for the activity of endothelium-derived hyperpolarizing factor (EDHF) [11], and for prevent-ing platelet adhesion to endothelium [12] A potent vasoconstrictor, endothelin, is also produced [13], as are a vast number of mediators of haemostasis, growth factors and cytokines [14] The outer endothelial layer
of the glycocalyx houses the membrane sensors for shear stress and various types of endothelial receptors such as B2 for bradykinin (Bk), P2ysubtypes for ADP from platelets and ATP from erythrocytes, PAF-R for platelet activating factor (PAF) from leukocytes, and PAR for thrombin [15] The membrane-bound endo-thelial enzymes include kininase 2, also called angio-tensin 1-converting enzyme (ACE-I)
Prostacyclin Prostacyclin (PGI2) was discovered in 1976 during the search for biological systems that in addition to blood platelets might convert prostaglandin endoperoxides (PGG2 or PGH2) to thromboxane A2 (TXA2) [16,17] This search was possible because newly discovered PGG2 and PGH2 were kindly offered to John R Vane
by the discoverer of TXA2, Bengt Samuelsson of the Karolinska Institutet This search was not successful, except for the detection of minute amounts of TXA2 made from PGH2 by lung and spleen microsomes Instead we found that a microsomal fraction of pig aorta transformed prostaglandin endoperoxides into
an unknown, unstable substance (with a half-life of
4 min at 37C) that had vasodilator and platelet-suppressant properties in vitro This substance was later named prostacyclin (PGI2) Further studies revealed that PGI2, when administered intravenously, dissipated platelet-rich thrombi in arterial blood in vivo [18] and that this effect was augmented by theophyl-line This latter finding confirmed that a cyclic nucleo-tide (in this case cAMP) was the second messenger of PGI2in platelets [19]
The common precursor for prostanoids including PGI2 is the four double-bonds 2-carbon fatty acid – arachidonic acid (AA) It was found that the
Trang 3nonenzy-matic product of AA monooxygenation
(15-hydro-peroxy-arachidonic acid; 15-HPAA) and other linear
lipid peroxides are inhibitors of microsomal
prosta-cyclin synthase (PGIS) Therefore, we hypothesized
that PGI2 deficiency resulted from an excessive
non-enzymatic peroxidation of body lipids might contribute
to development of atherosclerosis [20] Consequently,
we hoped to use synthetic PGI2 as a replacement
ther-apy in patients with atherosclerosis
Actually, I was the first healthy volunteer to receive
an intravenous infusion of synthetic PGI2 sodium salt
I lost the noble position of an observer during the last
stage of this experiment Still, these early trials allowed
us to establish a range of therapeutic doses for PGI2
and to observe the side effects caused by its
overdos-age [21] Eventually, PGI2 was infused into patients
with atherosclerosis of the leg arteries [22] However,
like most other powerful biological mediators, both
PGI2 (epoprostenol) [23] and its stable analogues (e.g
iloprost) [24] never became first-line drugs for the
treatment of atherothrombosis, instead giving way to
drugs that act as releasers of endogenous endothelial
PGI2 [25] However, some PGI2 analogues (e.g
tre-prostinil) are still used to treat patients with
pulmon-ary arterial hypertension [26], including those with
connective tissue disease [27]
The lung is a rich source of eicosanoids including
leukotrienes Various prostanoids are generated within
different pulmonary compartments Tracheal smooth
muscles generate PGE2, contractile elements of lung
parenchyma generate TXA2 [28], whereas pulmonary
endothelium secrets PGI2 We hypothesized [29] that
pulmonary endothelium may serve as a source for
circulating PGI2 [30] This concept was not well
accepted What kind of a circulating hormone has a
half-life in blood of 3–4 min? Nonetheless, assuming
PGI2 is generated continuously by pulmonary
endo-thelium, the stability of PGI2 might be sufficient for
it to be transported within the blood from the lung
to atherosclerotic coronary or cerebral arteries with
dysfunctional endothelium, and to save them from
being occluded by platelet-rich thrombi Pulmonary
endothelium might be a good target for new specific
releasers of circulating PGI2, although the local
gen-eration of PGI2 by the endothelium lining the
vascu-lar tree is probably a more important therapeutic
target, at least up to the point when the efficacy of
peripheral endothelium in not seriously disturbed by
an advanced atherothrombosis Interestingly,
overex-pression of pulmonary PGIS decreases the incidence
of cancerogenesis in murine models of lung cancer [31]
The crude microsomal fraction of aortic
homogen-ates that allowed us to discover biosynthesis the of
PGI2from PGH2[16,17] contained PGIS This enzyme was purified and characterized as a member of cyto-chrome P450 family (CYP 8A1) [32] In endothelial cells it collaborates with a supplier of PGH2, i.e with PGH2 synthase (PGHS-2), commonly, but less pre-cisely, called cyclooxygenase 2 (COX-2) In endothelial cells COX-2 is induced by shear stress COX-2 seems
to be the major source of systemic PGI2 in healthy humans [33] In female mice oestrogens upregulate PGI2 production via COX-2, and subsequently offer protection against atherothrombosis [34] Also, intra-vascular thrombosis in rats next to hypoxia-induced hypertension is prevented by the upregulation of vascular COX-2 followed by increased generation of PGI2[35]
There is little doubt that, in humans and laboratory animals, the endothelial COX-2⁄ PGIS tandem is responsible for the generation of vasoprotective PGI2, whereas in blood platelets the constitutive cyclooxy-genase 1⁄ thromboxane A2 synthase (COX-1⁄ TXAS) tandem generates vasotoxic TXA2
Nitric oxide radical
In 1980, a series of in vitro experiments with acetyl-choline-treated aortic rings led Robert Furchgott to discover endothelium-derived relaxing factor (EDRF) [36] Robert Furchgott likes to say that his great dis-covery arose from a number of accidental findings Those in 1986 exploded in the grand finale, i.e in the discovery that EDRF is nitric oxide Actually, the idea that EDRF¼ NO was proposed by Robert Furchgott and Louis Ignarro, independently [37] Robert Furchg-ott is modest as only a great scholar can be His mod-esty provokes the quotation from Louis Pasteur: ‘‘… where observation is concerned, chance favours only the prepared mind’’
The fabulous story of the discovery of EDRF(NO) was presented by Robert Furchgott [37], Louis Ignarro [38] and Ferid Murad [39,40] – three 1998 Nobel prize laureates in medicine and physiology In vascular endo-thelium, NO•
is synthetized from Arg by eNOS, which competes for substrate with tissue arginases eNOS is a homodimeric oxidoreductase with NADPH, flavin mononucleotide (FMN), flavin adenine dinucleotide (FAD), calmodulin (CaM) and tetrahydrrobiopterin (BH4) acting as cofactors eNOS via N’-hydroxy-Arg (NOHA) generates NO•
and citrulline Physiologically, eNOS homodimer catalyses a five-electron oxidation of Arg, whereas BH4 plays a crucial role in the activation
of dioxygen In tissues, NO•
is a powerful endogenous stimulator of soluble cytosolic guanylate cyclase Thus made, cGMP is the second messenger for NO•
in the
Trang 4same way that cAMP is the second messenger for
PGI2 Both cyclic nucleotides mediate the vasodilator,
vasoprotective and platelet-suppressant activities of
NO•
and PGI2, respectively
However, when eNOS splits into monomers, the
eNOS monomer acts as a reductase, and one-electron
reduction of dioxygen leads to formation of a
super-oxide anion (O2) [41] Uncoupling of eNOS occurs
as a consequence of BH4 shortage resulting from
folate avitaminosis or from hyperhomocysteinaemia
(HYHC) [42] Apart from the uncoupling of eNOS,
the other source of vascular O2 might be NAD(P)H
oxidase (43) Dimerization of eNOS requires the
intracellular availability of the substrate, i.e Arg
This is ensured by the high-affinity cationic cell
mem-brane transporter for Arg Its functioning might be
invalidated by homocysteine or by asymmetric
dime-thylarginine (ADMA) (see below) Arg
supplementa-tion in patients with atherosclerosis may be also
desirable because Arg acts as a direct antioxidant In
addition, Arg promotes the secretion of insulin from
pancreatic b cells and the release of histamine from
mast cells – both being vasodilators Theoretically,
Arg may also produce unfavourable effects, such as
generation of S-adenosyl-homocysteine from
S-adeno-sylmethionine via the methylation-dependent
biosyn-thesis of creatinine from guanidine acetate (44) Yet,
the net therapeutic effect of Arg given orally to
patients with myocardial infarction is encouraging
(45) pointing to a favourable route of
biotransforma-tion in these patients
Patrick Valance discovered, in human plasma, the
presence of symmetric dimethylarginine (SDMA) and
ADMA Only ADMA is biologically active, i.e it acts
as endogenous inhibitor of eNOS and inhibitor of Arg
membrane transporter Clinical data on ADMA are
growing A high plasma level of ADMA is considered
a novel cardiovascular risk factor Nowadays, it is
clear that ADMA contributes to vascular pathology
in atherothrombotic and diabetic angiopathies,
pre-eclampsia and hypertension (46) Elevated plasma
lev-els of ADMA in those patients may also explain the
‘arginine paradox’, i.e that therapeutic
supplementa-tion with exogenous Arg is beneficial, although in
these patients plasma levels of endogenous Arg exceed
the Michaelis–Menten constant (Km) for purified eNOS
in vitroby 25-fold [47]
Prostacyclin and nitric oxide radicals
A complex relationship exists between these two
unsta-ble, lipophylic endothelial secretagogues At the time
when NO•
still was known as EDRF it was claimed
that porcine aorta endothelial cells cultured on cytodex beads, loaded into a heated column and perfused with Krebs’ buffer, when stimulated with Bk or calcium ionophore, released both PGI2 and EDRF in a cou-pled manner [48] Superoxide anions abolished the biological activity of the released EDRF from these cultured endothelial cells [49], and from native endo-thelium of perfused canine artery [50] These latter findings initiated a march towards the discovery of the product of the interaction between NO•
and O2 , i.e peroxynitrite (ONOO–) ONOO– is one of the most reactive nitrogen species (RNS) It arises most easily when the eNOS dimer coexists in the vicinity of a eNOS monomer – then both genders of labile free rad-icals, i.e NO•
and O2 arise side by side, and without any delay ONOO–is made
ONOO– is a powerful oxidant and nitrating agent that destroys the ‘macromolecules of life’, i.e proteins (e.g PGIS inactivation), lipids [e.g the generation of oxidized low-density lipoprotein (ox-LDL) and iso-prostanes], and nucleic acids [e.g DNA strand break-age with a subsequent activation of poly-ADP ribosyl polymerase (PARP)] [51]
The toxic properties of ONOO– play a major role in atherothrombotic and diabetic angiopathies In those endothelial cells, ONOO– oxidizes the four zinc thio-late centres of dimeric eNOS As a consequence, zinc atoms are removed and disulfide monomers of eNOS arise The coexistence of dimeric and monomeric forms
of eNOS is responsible for the further amplification of ONOO– generation by endothelial cells This newly made ONOO– selectively nitrates Tyr430 in the enzy-mic protein of endothelial PGIS When PGI2 is elimin-ated from the endothelial defence system TXA2 and PGH2gain the upper hand [52]
Endothelial NOS received the mischievous name of
‘the Cinderella of inflammation’ [53] The authors had
in mind that excessive stimulation of eNOS might lead
to increased vascular permeability by NO•
, and thus
to inflammation However, in light of the foul games played between homodimeric and monomeric forms of eNOS, ending with the generation of ONOO– which eliminates PGI2 – the best friend of NO•
– I would rather think of eNOS as ‘the Lady Macbeth of athero-thrombosis’
Endothelial pharmacology
Samuel Beckett (1906–1989) wrote: ‘‘we need new par-adigms to accommodate the mess’ The paradigm of
‘pleiotropic action’ for some of cardiovascular drugs was coined to accommodate a discrepancy between their officially accepted modes of action and their
Trang 5additional therapeutic properties, as reported
unexpect-edly, but repeatunexpect-edly, by clinicians For example, statins
were introduced to the clinic with the aim of lowering
blood levels of low-density lipoprotein (LDL)
cho-lesterol, however, they were also found to correct
symptoms of myocardial and cerebral ischaemia,
inde-pendent of their capacity to inhibit
hydroxymethyl-glutaryl coenzyme A (HMG CoA) reductase [54–56]
Further support for the existence of the ‘pleiotropic
action’ of cardiovascular drugs was offered by the
efficacy of ACE-I to protect against myocardial
isch-aemia, stroke and diabetic angiopathies, as confirmed
in multicentre trials that included over 25 000 patients
[57], whereas the classic indication for ACE-I was the
treatment of patients with arterial hypertension The
phrase ‘pleiotropic action’ is not a cognitive
descrip-tion of reality Rather, it is an attempt ‘to
accommo-date the mess’ Our experimental data [58,59] pointed
to the possibility that the pleiotropic action of ACE-I
and statins might be explained by their stimulatory
effect on the endothelial generation of PGI2 and NO•
There are other propositions concerning the
mechan-ism of endothelial actions of statins, e.g the induction
of heme oxygenase (HO-1) [60] with a subsequent
anti-oxidant effect of biliverdin and CO mediation Here, I
take the opportunity to present our conception of the
endothelial pharmacology emerging from clinical
observations on the unexpected therapeutic effects of
known cardiovascular drugs This conception embraces
not only ACE-I and statins, but also other
cardiovas-cular drugs, e.g nebivolol and carvedilol (b-adrenergic
receptor antagonists) as well as ticlopidine and
clopi-dogrel (antiplatelet thienopyridines)
In vivo assay of endothelial function
Clinicians have developed an excellent noninvasive
method to measure endothelial function in humans
The method is based on the ultrasound scanning of
the flow-mediated dilatation (FMD) of the brachial
artery after its occlusion and reopening [61] In
prin-ciple, the FMD response is proportional to the
amount of NO•
released from endothelium of the vas-cular bed in question, however, an additional
bio-chemical assay pointed to the release of PGI2, along
with NO•
, from the endothelium during FMD [62]
No wonder – in vitro cultured endothelial cells
released EDRF(NO) and PGI2 in a coupled manner
[48] The FMD method allowed the detection of
endo-thelial dysfunction in patients with arterial
hyperten-sion [62], in patients with atherosclerosis undergoing
percutaneous coronary intervention with stenting [63],
and in patients with type 2 diabetes [64] In patients
with chest pain, a depressed FMD of the brachial artery was a sensitive indicator of coronary heart dis-ease (CAD) [65] FMD is impaired in tobacco smokers and in smokeless tobacco users compared with tobacco nonusers [66] There is ample evidence for the state-ment that endothelial dysfunction occurs in patients with hypertension, atherosclerosis and type 2 diabetes,
as well as in tobacco users
In vitro cognitio sed in vivo veritas(in vitro one may look for meaning, however, only in vivo is the truth to
be found) This motto stimulated us to develop our own experimental model for the in vivo assay of endothelial function [18–20,29,59,67–70] In our in vivo method it is not the vasodilator response (as in the case of FMD in humans) but rather the thrombolytic response that is used to assess endothelial capacity Therefore, it is the endothelial release of PGI2 that is appreciated at the first place, whereas the release of NO•
remains in the background Heparinized cats, rabbits and, most fre-quently, Wistar rats under general anaesthesia with extracorporeal circulation are used The arterial blood superfuses (2–3 mLÆmin)1) a collagen strip attached to a balance Blood returns to the venous system Thrombus mass is recorded continuously along with arterial blood pressure (Fig 1) Platelet-rich thrombus [70] gains a maximum mass of 100 mg within 30 min and stays unchanged for at least 4 h, unless a stimulator of vas-cular endothelium (e.g Ach, Bk or an endotheliotropic drug) is injected intravenously Then thrombolysis occurs (Fig 1) Its intensity and duration correlate with plasma levels of prostaglandin 6-keto-PGF1a (6-keto-PGF1a), whereas the levels of other stable prostanoids
do not (Fig 2) The participation of endothelial NO•
in thrombolytic response is checked by the pretreatment of animals with l-NAME or with any other NOS inhib-itor The participation of endogenous bradykinin in this response was checked by pretreatment with Icatibant,
an antagonist of B2 receptors (Fig 1A) In this system, thrombi were dissipated by intravenous administration
of PGI2 sodium salt or by its stable analogue (e.g ilo-prost) NO-donors (glyceryl trinitrate, molsidomine, sodium nitropusside, NONOates) also produced throm-bolysis but their effective doses were at a range of three orders of magnitude higher than those required for PGI2 or for its analogues Unlike PGI2, NO-donors at thrombolytic doses were highly hypotensive
Angiotensin-converting enzyme inhibitors
ACE-I, this name does not do justice to this class of drugs, namely captopril, enalapril, and especially per-indopril, quinapril, ramipril and many other lipophylic
Trang 6ACE-I There is no doubt that the pharmacological
activity of ACE-I is associated with the elimination of
cytotoxic and vasoconstrictor angiotensin 2, however,
the endothelial action of those ACE-I is also executed
via the local vascular accumulation of Bk, as our data
clearly show (Fig 1A) [59,67–69]
In 1965 a young Brazilian researcher, Sergio Ferreira
discovered the ‘bradykinin potentiating factor’ (BPF)
in the venom of Brazilian viper Bothrops jararaca [71]
At John Vane’s laboratory in London (where Sergio
Ferreira was a visitor) his discovery was appreciated
than it should have been At the time, Bk was
per-ceived as a mediator of pain and inflammation
respon-sible for paralytic vasodilatation in the course of acute pancreatitis The reasoning was as follows: BPF might
be good for this particular viper for swift killing of its victims but for us humans – it is no good at all So, why should we care about BPF?
Perfusate from isolated guinea-pig lungs dripping over Vane’s bioassay cascade was used to study Bk [71] and angiotensin 1 [72] metabolism Fortunately, it was soon found that various fractions of BPF given via the lungs inhibited the conversion of angiotensin 1
to angiotensin 2, and thus BPF was proved to act also
as an ACE-I [73] Inhibiting the conversion of biolo-gically inactive angiotensin 1 to hypertensive angioten-sin 2 – yes, it was an excellent principle on which to develop a new class of antihypertensive drugs [74] Indeed, at the request of John Vane, the top industrial chemists eventually did [75], and the first orally active ACE-I (a proline derivative – captopril) was intro-duced for the treatment of arterial hypertension The TREND trial [76] offered the first direct clinical evidence of improvement, by an ACE-I (quinapril), in endothelium-dependent vasorelaxation in patients with CAD There then appeared a number of clinical trials pointing to the same mechanism of vascular protection
by various ACE-I in patients at high risk of athero-thrombotic and diabetic vasculopathies [57]
B
A
BP
THR
THR
Dose-dependent thrombolysis by perindopril µg/kg i.v
Thrombolysis by QUINAPRIL depends on the release of
endogenous bradykinin and PGl 2 – only partially on NO
THR
30 min mg mg
100
0
100
0 THR
icatibant 100 µg/kg
indomethacin 5 mg/kg L-NAME 5 mg/kg
quinapril 30 µg/kg
quinapril 30 µg/kg
quinapril 30 µg/kg
quinapril 30 µg/kg
0
100 mg Thrombogenesis thrombus weight
Thrombolysis
pressure transducer
weight transducer
carotid artery
arterial blood
collagen THROMBUS
i.v drug injection
jugular vein
PERINDOPRIL
30.
10.
3.
30 min
100
mg 0
Fig 1 In vivo bioassay of endothelial secretory function.
Fig 2 Effect of quinapril on prostanoid plasma levels in Wistar rats
(n ¼ 7).
Trang 7Bk is the most potent releaser of PGI2from cultured
endothelial cells [48], and the most potent thrombolytic
agent acting via endothelial B2 receptors in vivo [67]
In Wistar rats, exogenous Bk at thrombolytic doses
is strongly hypotensive In contrast, endogenous Bk
released from vascular endothelium by low doses of
ACE-I (quinapril > perindopril > captopril) evokes
thrombolysis, but not a fall in blood pressure [59,68]
The principal mechanism of the thrombolytic action of
ACE-I stems from their secondary nature (or rather
their primary nature) of being BPF [71] Moreover,
there exist other Bk-potentiating effects of exogenous
ACE-I, such as the upregulation of B2 receptors, the
induction and activation of B1 receptors in the
endo-thelium and the stimulation of biosynthesis of
angio-tensin (1–7), which acts as an endogenous ACE-I (that
is BPF) [77] It should be added that in cultured
endo-thelial cells Bk acts as a ‘minicytokin’, inducing
mRNA for HO-1 and COX-2 [67] The interaction
between these two enzyme systems was claimed to
amplify the generation of PGI2 [78] It may well be
that, in addition to the immediate thrombolytic effects
of ACE-I, chronic treatment with ACE-I offers an
additional advantage of increasing the efficacy of the
endothelial enzymic raft (COX-2⁄ PGIS) responsible
for the biosynthesis of vascular prostacyclin along with
increasing local levels of CO and biliverdin – the
defensive products of endothelial HO-1
In our in vivo model for studying
endothelial-medi-ated thrombolysis in Wistar rats [59,68,69] it was
found that ACE-I (captopril < perindopril <
quina-pril) at low nonhypotensive intravenous doses of 10–
60 lgÆkg)1 dissipated thrombi that were superfused
with arterial blood The intensity and duration of this
thrombolysis were paralleled by an increase in arterial
plasma levels of 6-keto-PGF1a, and no change in
plasma levels of TXB2 and PGE2 (Figs 1 and 2)
Thrombolysis and prostacyclinaemia by ACE-I were
blunted or abolished by pretreatment with icatibant (a
B2 Bk receptor antagonist), by acetylsalicylic acid
(ASA) at a high dose of 50 mgÆkg)1 (Fig 3), and by
the coxibs (rofecoxib > celecoxib > nimesulide) at
low doses of 30–300 lgÆkg)1 Thrombolysis by ACE-I
was augmented by pretreatment with ASA at a dose of
1 mgÆkg)1 (Fig 3) or by acetaminophenen
Pretreat-ment with l-NAME delayed and flattened the
throm-bolytic response to ACE-I only slightly (Fig 1)
Pharmacological analysis of the above data led us to
conclude that ACE-I evoked thrombolysis by
pre-venting endothelial Bk from being destroyed by cell
membrane-bound ACE Bk that appeared at the
endothelial cell surface stimulated B2 receptors, which
triggered the COX-2⁄ PGIS system to generate PGI2,
and e-NOS to generate NO The final thrombolytic response to ACE-I depended mainly on PGI2, whereas
NO•
served as a helper with a permissive action The endothelial release of NO•
did not appear as the conditio sine qua non for thrombolytic response to ACE-I (Fig 1A)
There is another conclusion that derives from these studies It is as follows: effective endothelial COX-2 inhibition might be followed by thrombogenesis, whereas preferential COX-1 inhibition in platelets rein-forced the vasoprotective action of ACE-I (Fig 3) Our data cannot be considered as a good prognostic for the clinical use of high doses of coxibs in patients with cardiovascular disorders, but they do support the idea of administrating of low doses of ASA along with ACE-I (Fig 3)
Statins
In our in vivo model statins (e.g atorvastatin and simvastatin) produce endothelium-mediated, PGI2 -dependent thrombolysis when administered intraven-ously at doses 2–3 orders of magnitude higher than those for ACE-I [59] In Langendorff’s preparation of guinea-pig heart, statins produce NO•
-dependent vaso-dilatation of coronary vascular bed [59] The precon-tracted bovine coronary artery rings with endothelium are relaxed by statins, partially via a NO•
⁄ PGI2 -dependent mechanism [79] In cultured bovine aortic endothelial cells lipophylic statins, i.e atorvastatin, simvastatin and lovastatin (but not a hydrophilic pravastatin) at a concentration of 30 lm mobilize free cytoplasmic calcium [Ca2+]i to 30–50% of that induced by Bk at a concentration of 10 nm In the case of simvastatin and lovastatin, this effect disap-pears if their lactone rings are hydrolysed [80] The above endotheliotropic properties of statins are hardly
Fig 3 Dose-dependent effect of aspirin (ASA) on perindopril-induced thrombolysis.
Trang 8associated with their inhibitory action on HMG CoA
reductase In genetic and pharmacological models of
rat hypertension, rosuvastatin, another lipophilic
sta-tin, was found to exert a beneficial pleiotropic
endo-thelial effect [81] Patients with acute coronary
syndromes benefit from statin therapy [82] Statins
mobilize bone-marrow-derived endothelial progenitor
cells [83] and exert a vast number of other
pharmaco-logical effects that are not associated with modulation
of the lipoprotein profile by statins These unexpected
effects of statins are generally described as ‘pleiotropic
effects’ [84], and one of them is the endotheliotropic
action of statins described by us [59,79,80] The mode
of activation of the endothelial PGI2⁄ NO•
system by statins is not clear An interesting proposal was put
forward by Bill Sessa [85]
Thienopyridines and some of
b1-adrenergic receptor antagonists
Here we present two groups of highly effective
cardio-vascular drugs, the efficiency of which may or may not
depend on their additional stimulatory action of
vas-cular endothelium
Thienopyridines (ticlopidine and clopidogrel) belong
to a family of antiplatelet drugs, however, in vitro they
do not inhibit platelet aggregation Their in vivo
plate-let-suppressant action is executed by their labile
meta-bolites Therefore, a substantial lag period is required
for the appearance of the antiplatelet action of
thieno-pyridines Only unstable metabolites of theirs are
cap-able of antagonizing endogenous ADP on P2y12
purinergic platelet receptors, which when activated by
ADP induce platelet release and platelet aggregation
[86] The clopidogrel metabolite exerts its antiplatelet
action at IC50¼ 1.8 lm [87] There exists ample
evi-dence for the high efficacy of thienopyridines
(especi-ally clopidogrel) in the treatment of patients with
advanced atherothrombosis of coronary or cerebral
arteries, to mention only the following megatrials:
clopidogrel vs aspirin in patients at risk of ischemic
events (CAPRIE) [88], clopidogrel in unstable angina
to prevent recurrent events (CURE) and management
of atherothrombosis with clopidogrel in high risk
patients with recent transient ischaemic attacks or
isch-aemic stroke (MATCH) [89]
In 1996 [90] we demonstrated that ticlopidine
(10 mgÆkg)1) given intravenously to cats with
extracor-poreal circulation evoked immediate dissipation of the
platelet-rich clots superfused with their arterial blood
[18] This thrombolytic effect of ticlopidine was
com-parable with that induced by PGI2 at 0.3 lgÆkg)1
These and other data [90] prompted us to postulate
that the therapeutic efficacy of ticlopidine might be associated not only with the delayed platelet-suppres-sant effect of its unstable metabolite via blockade of P2y12 platelet receptors, but also with the instan-taneous endothelial action of the native molecule
of ticlopidine showing up as an immediate, endo-thelium-mediated thrombolysis of platelet-rich clots
in vivo[90]
In rats, these ‘immediate thrombolytic effects’ of thienopyridines were rather weak (EC30¼ 15–30 mgÆkg)1) Jean-Pierre Dupin of the Bordeaux II Uni-versity decided to synthetize a series of thienopyrimi-dinones under the guidance of our pharmacological assay of their endothelium-dependent thrombolytic effects in vivo Assessment of their structure–activity relationship revealed that the most active compound, i.e 3[(2-trifluoromethyl-phenyl)-methyl] 1,2-dihydro-benzo[b]thieno[2,3-d]pyrimidinone-4(3H)one dissipated platelet clots in rats in vivo at a dose of IC30¼
8 lgÆkg)1[91]
We conclude that in addition to in vivo endothelial PGI2-mediated thrombolysis, thienopyrimidinones and thienopyridines exert endothelial NO•
-mediated coron-ary vasodilatation in perfused guinea-pig heart [92] Mechanisms of endotheliotropic actions of these strongly lipophylic compounds remain unknown Nebivolol and carvedilol – two b1-adrenoceptor antagonists – founded the ‘third generation’ of selective b-adrenolytic drugs, which are endowed with endothe-liotropic properties Eleven years ago, Bowman et al [93] proposed that the antihypertensive effects of nebiv-olol in man might be partially associated with endo-thelium-dependent, NO•
-mediated vasodilatation In two interesting studies Ignarro et al [94,95] clearly demonstrated that relaxation of vascular smooth muscle by nebivolol is partially mediated by endothe-lium-dependent release of NO•
and the subsequent accumulation of cGMP in smooth muscle [94], how-ever, nebivolol also inhibits vascular smooth muscle proliferation by a mechanism involving NO•
but not cGMP [95] Various routes were proposed by which the ‘third generation’ of b1-adrenoceptor antagonists may release endothelial NO•
Certainly adrenergic and serotoninergic receptors are not involved [96] A fascin-ating hypothesis has been proposed [97] Nebivolol and carvedilol stimulate the renal efflux of ATP, that releases NO•
via activation of P2Y purinoceptors
in glomerular endothelium On top of the regular
b1-adrenoceptor blockade there appears NO•
-mediated relaxation of renal glomerular microvasculature This
is why nebivolol and carvedilol are so efficient in controlling arterial hypertension and improving renal circulation
Trang 9In endothelial pharmacology everything is new: (a) the
idea that vascular endothelium may be looked upon as
an organ with a secretory function; (b) considering
pulmonary endothelium as a separate endocrine organ
that supplies prostacyclin to the coronary and cerebral
circulations; (c) a complex relationship between two
endothelial mediators – NO and PGI2 – a role for
ROS and RNS in it; (d) discovering new
endothelio-tropic mechanisms for old cardiovascular drugs like
for ACE-I, statins or nebivolol; (e) planning new
endotheliotropic chemical structures, e.g
thienopiry-midodiones; (f) discovering new biochemical
mecha-nisms of action for drugs affecting endothelial function
like in case of nebivolol; and (g) the interaction
between basic and clinical researchers, probably one of
the most efficient in the field of medicine Old, known
roads are safe, but the newly discovered roads are
interesting
References
1 Krebs HA (1980) Excerpts from an introductory address
Int J Biochem 12, 1–8
2 Gryglewski RJ & McGiff JC (2005) Eulogy John R
Vane (1927–2004) Hypertension 45, 319–320
3 Nachman RL & Jaffe EA (2004) Endothelial cell
cul-ture: beginning of modern vascular biology J Clin
Invest 114, 1037–1040
4 Schulz F, Antor F & Keaney JF (2004) Oxidative stress,
antioxidants and endothelium function Curr Med Chem
11, 1093–1104
5 Chen K, Thomas SR, Albano M, Murphy MP &
Kea-ney JF (2004) Mitochondrial function is required for
hydrogen peroxide-induced growth factor receptor
transactivation and downstream signalling J Biol Chem
279, 35079–35086
6 Stocker R & Keaney JF (2004) Role of oxidative
modifications in atherosclerosis Physiol Rev 84, 1381–
1478
7 Haendeler J, Tischler V, Hoffman J, Zeiher AM &
Dimmeler S (2004) Low doses of reactive oxygen species
protect endothelial cells from apoptosis by increasing
thioredoxin-1 expression FEBS Lett 577, 427–433
8 Kinlay S, Behrendt D, Fang JC, Delagrange D,
Mor-row J, Witztum JL, Ganz P, Rifai N, Selwyn AP &
Creager MA (2004) Long-term effect of combined
vita-min E and C on coronary and peripheral endothelial
function J Am Coll Cardiol 43, 629–634
9 Florey H (1966) The endothelial cell Br Med J 2, 487–
490
10 Vane JR (1994) The endothelium: maestro of the blood
circulation Philos Trans R Soc Lond 343, 225–246
11 Spiecker M & Liao JK (2005) Vascular protective effects of cytochrome P450-derived eicosanoids Arch Biochem Biophys 433, 413–420
12 Krotz F, Riexinger T, Buerkle MA, Nithipatikom K, Gloe T, Sohn HY, Campbell WB & Pohk U (2004) Membrane potential-dependent inhibition of platelet adhesion to endothelial cells by epoxteicosatrienoic acids Atheroscler Thromb Vasc Biol 24, 595–600
13 Masaki T (2004) Historical review: endothelin Trends Pharmacol Sci 25, 219–224
14 Chłopicki S & Gryglewski RJ (2002) Pharmacology of endothelium (in Polish) Kardiologia Polska 57, 5–15
15 Garcia-Cardena G, Comander J, Loscalzo J, Anderson
KR & Gimbrone MA Jr, (2001) Biochemical activation
of vascular endothelium as a determinant of its func-tional phenotype Proc Natl Acad Sci USA 98, 4478– 4485
16 Gryglewski RJ, Bunting S, Moncada S, Flower RJ & Vane JR (1976) Arterial walls are protected against deposition of platelet thrombi by a substance (prosta-glandin X) which they make from prosta(prosta-glandin endo-peroxides Prostaglandins 12, 685–713
17 Moncada S, Gryglewski RJ, Bunting S & Vane JR (1976) An enzyme isolated from arteries transforms prostaglandin endoperoxides to an unstable substance that inhibits platelet aggregation Nature 263, 663–665
18 Gryglewski RJ, Korbut R, Ocetkiewicz A & Stachura J (1978) In vivo method for quantitation of anti-platelet potency of drugs Naunyn Schmiedebergs Arch Pharma-col 302, 25–30
19 Gryglewski RJ, Korbut R & Ocetkiewicz A (1978) De-aggregatory action of prostacyclin in vivo and its enhancement by theophyline Prostaglandins 15, 637– 644
20 Gryglewski RJ (1980) Prostacyclin, platelets and athero-sclerosis In Reviews on Biochemistry, Vol 7 (Fasman
GD, ed), pp 291–338 CRC Press, New York
21 Gryglewski RJ, Szczeklik A & Ni_zankowski R (1978) Anti-platelet effects of intravenous infusion of prosta-cyclin in man Thromb Res 13, 153–163
22 Szczeklik A, Ni_zankowski R, Skawin´ski S, Głuszko P & Gryglewski RJ (1979) Successful therapy of advanced arteriosclerosis obliterans with prostacyclin Lancet 1, 1111–1114
23 Gryglewski RJ, Szczeklik A & McGiff JC, eds (1983) Prostacyclin – Clinical Trials Raven Press, New York
24 Gryglewski RJ & Stock G, eds (1987) Prostacyclin and its Stable Analogue Iloprost Springer-Verlag, Berlin
25 Chłopicki S & Gryglewski RJ (2004) Endothelial secre-tory function and atherothrombosis In The Eicosanoids (Curtis-Prior P, ed), pp 267–276 Wiley, Chichester
26 Vachiery JI & Naeije R (2004) Treprostinil for pulmon-ary hypertension Expert Rev Cardiovasc Ther 2, 183– 191
Trang 1027 Oudiz RJ, Schilz RJ, Barst RJ, Galie N, Rich S, Rubin
LJ & Simonneau G (2004) Treprostinil, a prostacyclin
analogue in pulmonary arterial hypertension associated
with connective tissue disease Chest 126, 420–427
28 Gryglewski RJ, Dembinska-Kiec A, Grodzinska L &
Panczenko B (1976) Differential generation of
sub-stances with prostaglandin-like and thromboxane-like
activities by the guinea pig trachea and lung strips In
Lung Cells in Disease (Bouhuys A, ed), pp 289–307
Elsevier, Amsterdam
29 Gryglewski RJ, Korbut R & Ocetkiewicz A (1978)
Gen-eration of prostacyclin by lungs in vivo and its release
into the arterial circulation Nature 273, 765–767
30 Gryglewski RJ (1980) The lung as a generator of
pro-stacyclin In Ciba Foundation Symposium No 78:
Meta-bolic Activities of the Lung, pp 147–164 Excerpta
Medica, Amsterdam
31 Keith RL, Miller YE, Hudish TM, Girod CE,
Sotto-Santiago S, Franklin WA, Nemenoff RA, March FH,
Nana-Sinkam SP & Geraci MW (2004) Pulmonary
prostacyclin synthase overexpression chemoprevents
tobacco smoke lung cancerogenesis in mice Cancer Res
64, 5897–5904
32 Shimonishi M, Nakamura M, Imai Y, Ullrich V &
Tanabe T (2004) Purification and characterisation of
recombinant human prostacyclin synthase J Biochem
(Tokyo) 135, 455–463
33 McAdam BF, Catella-Lawson F, Mardini IA, Kapoor
S, Lawson JA & FitzGerald GA (1999) Systemic
bio-synthesis of prostacyclin by cycloxygenase (COX)-2: the
human pharmacology of a selective inhibitor of COX-2
Proc Natl Acad Sci USA 96, 272–277
34 Egan KM, Lawson JA, Fries S, Koller B, Rader DJ,
Smyth EM & FitzGerald GA (2004) COX-2 derived
prostacyclin confers atheroprotection on female mice
Science 306, 1954–1957
35 Pidgeon GP, Tamosiuniana R, Chen G, Leonard I,
Bel-ton O, Bradford A & Fitzgerald DJ (2004) Intravascular
thrombosis after hypoxia-induced pulmonary
hyperten-sion: regulation by cycloxygenase-2 Circulation 110,
2701–2707
36 Furchgott RF & Zawadzki JV (1980) The obligatory
role of endothelial cells in the relaxation of arterial
smooth muscle by acetylcholine Nature 288, 373–376
37 Furchgott RF (2001) Research leading to nitric oxide:
the importance of accidental findings In Nitric Oxide
(Gryglewski RJ & Minuz P, eds), NATO Life Sciences
Series A 317, 1–4 IOS Press, Amsterdam
38 Ignarro LJ (2002) Nitric oxide as a unique signalling
molecule in the vascular system: a historical overview
J Physiol Pharmacol 53, 504–514
39 Murad F (2004) Ferid Murad MD, PhD: a conversation
with the Editor Am J Cardiol 94, 75–91
40 Seminara AR, Krumenacker JS & Murad F (2001)
Signal transduction with nitric oxide, guanylyl cyclase
and cyclic guanosine monophosphate In Nitric Oxide (Gryglewski RJ & Minuz P, eds), NATO Life Sciences Series A 317, pp 5–22 IOS Press, Amsterdam
41 Stuehr DJ, Adak S, Santolini J, Wei CC & Wang Z (2001) Mechanism of oxygen activation in NO synthases, and a kinetic model for catalysis In Nitric Oxide (Gry-glewski RJ & Minuz P, eds), NATO Life Sciences Series
A 317, pp 23–26 IOS Press, Amsterdam
42 Topal G, Brunet A, Millanvoye E, Boucher JL, Rebdu
F, Devyck MA & David-Dufilho M (2004) Homocys-teine induces oxidative stress by uncoupling of NO synthase activity through reduction of tetrahydrobiop-terin Free Radical Biol Med 36, 1532–1541
43 Channon KM & Guzik TJ (2002) Mechanisms of super-oxide production in human blood vessels Relationship
to endothelial dysfunction, clinical and genetic factors
J Physiol Pharmacol 53, 515–524
44 Loscalzo J (2004) l-Arginine and atherothrombosis
J Nutr 134 (10 Suppl.), 2798S–2800S
45 Chamiec T, Ceremuzynski L & Bednarz B (2004) Effects
of l-arginine in myocardial infarction: double-blind, placebo-controlled multicenter pilot study J Am College Cardiol 43 (Suppl A), (abstract, p 1)
46 Vallance P & Leiper J (2004) Cardiovascular biology of the asymmetric dimethylarginine: dimethylarginine dimethylaminohydrolase pathway Arterioscler Thromb Vasc Biol 24, 1023–1030
47 Boger RH (2004) Asymmetric dimethylarginine, an endogenous inhibitor of nitric oxide synthase explains
‘the arginine paradox’ and acts as a novel cardio-vascular risk factor J Nutr 134 (10 Suppl.), 2842S– 2847S
48 Gryglewski RJ, Moncada S & Palmer RM (1986) Bioas-say of prostacyclin and endothelium-derived relaxing factor (EDRF) from cultured porcine aortic endothelial cells Br J Pharmacol 87, 685–694
49 Gryglewski RJ, Palmer RM & Moncada S (1986) Superoxide anion is involved in the breakdown of endothelium-derived vascular relaxing factor Nature
320, 454–456
50 Rubanyi GM & Vanhoutte PM (1986) Superoxide anions and hyperoxia inactivate endothelium-derived relaxing factor Am J Physiol 250, H822– H827
51 Szabo C (2003) Multiple pathways of peroxynitrite cyto-toxicity Toxicol Lett 140, 105–112
52 Zou MH, Cohen R & Ullrich V (2004) Peroxynitrite and vascular endothelial dysfunction in diabetes melli-tus Endothelium 11, 89–97
53 Cirino G, Fiorucci S & Sessa WC (2003) Endothelial nitric oxide synthase: the Cinderella of inflammation? Trends Pharmacol Sci 24, 91–95
54 Davignion J (2004) Beneficial cardiovascular pleiotropic effects of statins Circulation 109 (Suppl 1), 39–43