ВУЗ: Не указан
Категория: Не указан
Дисциплина: Не указана
Добавлен: 30.10.2019
Просмотров: 1134
Скачиваний: 2
The Kallikrein-Kinin System
11
logical effect in blood. Inhibitors of t-PA play an
important role in regulating fibrinolysis. Currently, four
distinct types of plasminogen activator inhibitors (PAI)
have been described: PAI-1, PAI-2, PAI-3, and protease
nexin. Of these, PAI-1 is the most important in inhibiting
t-PA in plasma. PAI-1 is present in molar excess over
t-PA; hence the majority of t-PA circulates bound to
PAI-1, thereby preventing its interaction with plasmino-
gen and thus a premature lysis of fibrin and a systemic
fibrinolysis (54).
II- Metabolism of Kinins
The nature and properties of the various peptidases
capable of metabolizing kinins have been extensively
reviewed (55). We have experimental evidences that 4
metallopeptidases are mainly responsible for the
metabolism of BK. These are angiotensin I-converting
enzyme (ACE), aminopeptidase P (APP), neutral endo-
peptidase 24.11 (NEP, neprilysin), and carboxy-
peptidases M and N (CPM, CPN). The importance of
these peptidases depends on the animal species, the
analytical approach, the biological milieu, and the
pathophysiological context. Interestingly, these pepti-
dases are zinc metallopeptidases, that is, they all require
zinc in their catalytic site to hydrolyse substrates, and
are membrane-bound glycoproteins, except CPN, which
is a soluble tetrameric glycoprotein. However, they are
all present in a soluble form in biologic fluids.
1. Angiotensin I-Converting Enzyme (ACE)
1.1 Definition
ACE (EC 3.4.15.1) is a well-characterized type I
ectoenzyme membrane anchored Zn
2+
-dependent dipep-
tidyl carboxypeptidase that regulates bioactivities of
vasoactive peptides such as angiotensin I (Ang I) and
BK, responsible for the control of blood pressure (56).
Two distinct forms of ACE are expressed in humans,
a larger one (150 – 180 kDa) usually referred to as
somatic ACE that is composed of approximately 1300
aminoacids and is present in most tissues (vascular
endothelial surface of the lungs and on brush-border
membranes of kidney, intestine, placenta, and choroid
plexus) and a smaller isoenzyme referred to as the
germinal form, testicular ACE (100 – 110 kDa), with
only 730 aminoacids and is found exclusively in the
testicles and appears to be involved in male fertility
(56, 57).
The somatic form has two homologous metallo-
proteinase domains (N- and C-terminal domains) with
an overall 60% homology in both nucleotide and
aminoacid sequence, each containing a canonical Zn
2+
-
binding sequence motif: HEXXH (His-Glu-X-X-His)
and bearing a functional active site (58, 59). The
stoichiometry value of 1:1 for the complete inhibition
of the enzyme indicates that both active sites would
work in a cooperative manner (60).
The testicular ACE has a single domain correspond-
ing to the C-terminal domain of somatic ACE (61)
together with the hydrophobic membrane-anchoring
domain and a small N-terminal region that has multiple
O
-linked oligosaccharides (62).
Although ACE is primarily a membrane-bound
protein, a soluble form does exist in many body fluids
and is the result of post-translational proteolytic
cleavage in the juxtamembrane stalk by a membrane
protein secretase or sheddase, itself a zinc metallo-
peptidase (63, 64). Although the ACE secretase has not
yet been identified, studies with a range of hydroxamic
acid-based inhibitors have shown that it has a remark-
ably similar inhibition profile to the amyloid precursor
protein
α
-secretase, leading to the conclusion that the
two secretases are, at the very least, closely related (65).
Human plasma ACE originates from endothelial cells,
while in other body fluids ACE originates from
epithelial, endothelial, or germinal cells (66).
A number of proteins with sequences related to ACE
have been described, but the only mammalian relative to
be found is ACE2. Similar to ACE, ACE2 is also a type I
integral membrane glycoprotein found on the surface of
endothelial and epithelial cells, although it has a more
limited tissue (testis, heart, kidney) distribution than
ACE. ACE2 consists of a single active site domain
that, by sequence comparison, more closely resembles
the N-domain than the C-domain of somatic ACE.
However, ACE2 differs from ACE in that it acts as a
carboxypeptidase removing single aminoacid residues
from its substrates, which include angiotensin II (Ang II)
with high catalytic efficiency (
k
cat
/ K
m
=
1.9
×
10
6
M
−
1
⋅
s
−
1
) (67). ACE2 efficiently hydrolyzes des-Arg
9
-BK
(
k
cat
/ K
m
=
1.3
×
10
5
M
−
1
⋅
s
−
1
), but fails to hydrolyze BK
itself (68). Also, ACE2 is not inhibited by some inhi-
bitors of ACE. Since the discovery of ACE2, some
authors have begun to refer to ACE as ACE1 (69, 70).
1.2 Synthesis, regulation, and localization
Somatic and germinal ACE are transcribed from the
same gene (17q23) using alternative promoters (71).
Cloning of the gene that encodes ACE revealed the
relationship between the two isoforms. Besides their
different tissue specificity, expression of the two ACE
transcripts is regulated by different developmental and
hormonal controls; for example, the endothelial enzyme
is induced by glucocorticoids, whereas the testicular
form is stimulated by androgens (71).
ME Moreau et al
12
1.3 Properties
1.3.1 Angiotensinase vs kininase
ACE catalyzes the conversion of the inactive
decapeptide Ang I to the potent vasopressor octapeptide
Ang II by the removal of the C-terminal dipeptide His
9
-
Leu
10
.
Ang I is originally considered the main physiological
substrate for ACE (
K
m
, approximately 16
µ
M). Because
of its higher affinity (
K
m
, approximately 0.18
µ
M) for
BK, ACE could also be now considered as a kininase
(kininase II) (72, 73). As a peptidyl dipeptidase, ACE
inactives BK by hydrolyzing two separate bonds on its
C-terminal end. It removes sequentially the dipeptide
Phe
8
-Arg
9
and next cleaves the Phe
5
-Ser
6
bond to
generate the second dipeptide Ser
6
-Pro
7
, transforming
BK into its inactive final BK
[1–5]
product (74, 75). ACE
also metabolizes des-Arg
9
-BK by removing the carboxy-
terminal tripeptide Ser
6
-Pro
7
-Phe
8
, yielding the same
final pentapeptide.
1.3.2 ACE GPIase activity
Recently, Kondoh et al. reported that ACE has a GPI-
ase activity which allows the release of a membrane
glycosylphosphatidylinositol (GPI)-anchored protein.
They have found that ACE-specific inhibitors, such as
captopril and lisinopril, which bind to the catalytic
center and completely inhibit the peptidase activity, had
only a minor inhibitory effect on this GPIase activity
which does not require the zinc ion (76). This new
activity of ACE which is different from the GPIase
activity of GPI-specific phospholipase D (GPI-PLD)
(76, 77) and of phosphatidylinositol-specific phos-
pholipase
C (PI-PLC)-like could be of physiological
importance in fertilization.
1.3.3 ACE: a signal transduction molecule
By its short cytoplasmic domain (29 aminoacids),
ACE could also play a role as a signal transduction
molecule. In fact, the cytoplasmic tail of ACE is phos-
phorylated in endothelial cells (78). ACE inhibitors as
well as BK elicit outside-in signalling in these cells.
They enhance the activity of ACE-associated protein
kinase CK2, increasing its phosphorylation of ACE, and
lead to the activation of c-Jun N-terminal kinase (JNK)
as well as the accumulation of phosphorylated c-Jun in
the nucleus (79).
1.3.4 ACE insertion / deletion polymorphism
Population studies indicated that the large inter-
individual variability in plasma ACE levels is geneti-
cally determined. Regulation, as well as tissue ACE
activity are under strong genetic control. An insertion
/
deletion (I
/
D) polymorphism located in the noncoding
region of the gene is associated with differences in the
level of ACE in plasma and cells. The insertion that
gives rise to the I allele is an
alu
repeat sequence
(287 bp) in intron 16 of the ACE gene; the D allele
results from the absence of the above insertion. There is
a relationship between the ACE I
/
D polymorphism and
circulating ACE activity, such that ACE activity is
highest in individuals homozygous for the D allele,
lowest in those homozygous for the I allele and inter-
mediate in heterozygotes.
Mean serum ACE level in DD homozygotes are
nearly twice the value measured in II plasma, while DI
heterozygous have intermediate activities (80 – 82).
Although some association between the I
/
D poly-
morphism with the incidence of cardiovascular and
Alzheimer diseases, the results remain inconsistent
(83 – 87). However, ACE genotype determines BK
degradation and suggests another mechanism whereby
the ACE D allele could be associated with deleterious
cardiovascular effects (82).
2. Neprilysin (NEP)
2.1 Definition
NEP (EC 3.4.24.11, neprilysin) is a type II surface
protein with a short membrane-proximal stalk region
that is not sensitive to proteolytic activity of any parti-
cular secretase (88). It is the prototype zinc peptidase
of the M13 membrane metalloendopeptidase (MME)
family. This family includes other peptidases of poten-
tial pathophysiological interest: ECE-1 and ECE-2,
which generate the vasoconstrictor endothelins from
big endothelin; the erythrocyte cell-surface antigen
KELL; PHEX, associated with congenital X-linked
hypophosphatemic ricket; ECEL1, present in CNS; SEP
(MML1 or NL1) present in testis; and MMEL2 (89 – 91).
2.2 Synthesis, localization, and properties
Human NEP gene (
MME
) is localized to 3q21 – q27
(92). First identified in the brush border of the kidney
epithelial cells, where it represents 4 – 5% of the protein
content, its immunoreactivity or activity has also been
detected in cells or tissues as different as the CNS, the
endothelium, testis, lungs, salivary glands, and bone
marrow (93). NEP preferentially cleaves bonds on
the amino-side of hydrophobic aminoacids residues.
However, the physiological role of NEP depends on its
tissue localization and the presence of the substrate.
Like ACE, NEP inactivates BK by sequentially
removing dipeptide Phe
8
-Arg
9
and tripeptide Phe
5
-Ser
6
-
Pro
7
to yield BK
[1–4]
, an inactive metabolite. However,
a prolonged incubation also results in the hydrolysis of
the Gly
4
-Phe
5
bond (94). In contrast to ACE, it is likely
that NEP cleaves the Gly
4
-Phe
5
bond of des-Arg
9
-BK
The Kallikrein-Kinin System
13
to generate the same BK
[1–4]
-inactive peptides. NEP is
the main enzyme responsible for the metabolism of
kinins in the kidney and plays an important role in the
metabolism of BK at the endothelium. In plasma, unlike
ACE, NEP does not play a significant role in the meta-
bolism of kinins (95).
Besides kinins, NEP also inactivates other peptides
like enkephalins, neurokinins, and amyloid-
β
peptide,
a marker for Alzheimer disease of the CNS (96). In
the kidney, it regulates the degradation of natriuretic
peptides (97).
3. Aminopeptidase P (APP)
3.1 Definition
Human APP (X-prolyl aminopeptidase, EC 3.4.11.9)
exists at least in two forms: a soluble cytosolic (cAPP)
and membrane-bound (mAPP) APP forms. Both forms
have an X-prolyl aminopeptidase activity (98). cAPP
and mAPP share 43% of homology. cAPP is a homo-
dimer of 70 kDa subunits and the mature form APP has
a predicted subunit molecular weight of 71 kDa (99,
100). mAPP is a heavily N-glycosylated, Zn
2+
-contain-
ing, peptidase and contains a C-terminally attached GPI
membrane anchor that increases the overall subunit
molecular weight to approximately 90 kDa. A secreted
form of mAPP has recently been produced and charac-
terized in vitro in a transfected HEK-293 cell line.
This protein exhibits a molecular mass of 85 kDa (101).
The effects of divalent metal ions on the activity of
mammalian APPs appear to differ between the cytosolic
and membrane-bound form of the enzyme (102). Indeed,
whereas mAPP uses Zn
2+
at the active site, human cAPP
contains Mn
2+
and is activated in vitro by the presence of
Mn
2+
, but not Zn
2+
, which is inhibitory (99, 103).
There is evidence for the existence of another form of
APP in mammals. The third isoform is hypothetical and
has been identified on the basis of sequence homology.
The chromosomal location of this isoform is 22q13.31 –
q13.33. This isoform is predicted to have a subunit
molecular weight of 57 kDa and is therefore smaller than
cAPP and mAPP (104).
3.2 Synthesis, regulation, and localization
The chromosomal location of the human cAPP gene
(
XPNPEPL
) is 10q25.3 (105) and the
XPNPEP2
gene
of mAPP localizes to chromosome Xq25-26.1 (106).
mAPP is located on the external side of the plasma
membrane of vascular endothelial cells and on brush
border membranes of epithelial cells in the intestine and
the renal proximal tubule (107 – 112).
3.3 Properties
cAPP exhibits broad substrate specificity and can
cleave X-Pro dipeptides as well as longer peptides of
the form X-Pro-Y-, where X and Y are of the common
aminoacids (113 – 115). Compared to cAPP, mAPP
has much more restricted substrate specificity; it fails to
hydrolyze X-Pro dipeptides and cleaves X-Pro-Y-
peptides poorly when X is Pro or Gly or when Y is an
aminoacid with a bulky side chain (116 – 118).
Kinins are the best substrates for mAPP. In fact, the
secreted form of mAPP exhibits a
K
m
of 75
±
15
µ
M
for BK and a
K
m
of 56
±
13
µ
M for des-Arg
9
-BK. In
human plasma, APP transforms BK and des-Arg
9
-BK
into the inactive peptide BK
[2–9]
and BK
[2–8]
, respectively,
and is the major inactivating pathway for des-Arg
9
-BK
in plasma (55, 95, 119).
4. Carboxypeptidase N and M (CPN, CPM)
Currently known as kininase I, both CPN and CPM
are zinc metallopeptidases that exhibit a 41% sequence
identity (120).
CPN (kininase I, EC 3.4.17.3) is a tetrameric protein
synthetized in the liver and secreted in blood, although
CPM (membrane-bound kininase I) is a GPI peptidase
anchored at the membrane of lung and kidney epithelial
cells. They cleave a variety of peptides containing a
carboxy-terminal Arg or Lys. CPN inactivates comple-
ment anaphylatoxins (C3a, C4a, and C5a) by cleaving
their carboxy-terminal Arg residue (121). Although
likely, definitive evidence for such a metabolic role
does not exist for CPM. Both carboxypeptidases trans-
form BK and KD into des-Arg
9
-BK and des-Arg
10
-KD,
which are active metabolites via B
1
receptor (B1R). This
kininase I activity constitutes a minor metabolic path-
way unless ACE is inhibited (55, 121).
5. Other peptidases
The properties of dipeptidyl peptidase IV (DPP IV)
have been extensively reviewed (122). DPP IV (CD26;
EC 3.4.14.5) is a 110 kDa plasma membrane glyco-
protein ectopeptidase (119, 122). The most important
physiological actions of DPP IV is on regulatory
peptides in mammals. DPP IV degrades the inactive
metabolite BK
[2–9]
generated by APP and leads to the
final BK
[4–9]
product (122). Substance
P is another
example of a regulatory peptide for which DPP IV plays
an additional role in the inactivation (123).
Aminopeptidase N (APN, CD13) is an ectoenzyme
(EC 3.4.11.2) of the superfamily of zinc metallo-
proteases widely used as a marker of myelomonocytic
cells in the diagnosis of hematopoietic, malignant dis-
orders (88, 124). Because of its ectopeptidase activity,
APN has also been implicated in the regulation of vaso-
active peptides, neuropeptide hormones, and immuno-
modulating peptides such as interleukin-6 and -8 (IL-6
ME Moreau et al
14
and IL-8, respectively) (125, 126). Its importance in the
field of kinins is related to its ability to hydrolyze the N-
terminal Lys residue in KD and des-Arg
10
-KD; this reac-
tion is functionally significant for the latter peptide
which is the optimal agonist of the B1R in humans and
other species. Indeed, the reaction product des-Arg
9
-BK
has a much lower affinity at the human, rabbit, porcine,
and bovine B1R. APN mediates the major inactivation
pathway for des-Arg
10
-KD in the human isolated umbil-
ical artery (127) and rabbit aorta (F. Marceau and
A. Adam, unpublished observations). APN blockade
with amastatin potentiates the hypotensive effect of the
B1R agonist des-Arg
10
-KD in the LPS-pretreated rabbit
(128).
III- Kinin Receptors
Two types of G-protein-coupled receptors (GPCRs)
mediate the cellular effects of kinins, the B1R and B
2
receptor (B2R) (129, 130). The various pharmacological
effects of the BK-related peptides, such as vasodilata-
tion, increased vascular permeability, stimulation of
sensorial and sympathetic nervous connections, and
smooth muscle contraction, derive from the presence of
these receptors on various cell types such as the vascular
endothelium, primary sensory afferent neurons, vascular
and nonvascular smooth muscle, epithelial cells, and
perhaps some types of leukocytes. The two types of
receptors were initially defined using pharmacological
criteria, before their molecular characterization.
1. Pharmacological classification
The potency order of agonists and the affinity of
antagonists and their properties are summarized in
Table 1 for B1R and Table 2 for B2R.
1.1 Potency order of agonists
The B2R has a high affinity for “native” kinins
(generated by either plasma or tissue kallikreins), BK,
and KD in all mammalian species. None of the frag-
ments of BK retain a significant affinity for the B2R. KD
has also a significant activity for the human and rabbit
B1R (e.g., in radioligand binding assays), but in
complex bioassay systems, BK or KD effects at B1R
are often mediated by their des-Arg metabolites
generated in situ (131). Indeed, the B1R is specialized
across species to respond to kinin metabolites generated
by arginine carboxypeptidases (either des-Arg
9
-BK,
Lys-des-Arg
9
-BK or des-Arg
10
-KD) (129, 130). Des-
Arg
9
-BK is a highly selective agonist of the mammalian
B1R, but of high affinity (nanomolar) only in rodents
(the rat and mouse); the only natural kinin sequence
with a subnanomolar affinity for the human, rabbit,
porcine, and bovine B1R is Lys-des-Arg
9
-BK, suggest-
ing that the B1R works in concert with the tissue
kallikrein that generates its parent native peptide, KD
(130).
1.2 Affinity of antagonists
Antagonists for the B1R were discovered almost
10 years before the antagonists for the B2R; thus, the
receptor nomenclature is justified by the fact that it
was the first to be pharmacologically fully defined.
The first series of compounds capable of antagonizing
BK and des-Arg
9
-BK with specificity for B1R included
the prototype [Leu
8
]des-Arg
9
-BK (132). However, this
prototype exhibits fairly high partial agonist behavior in
some species, especially in the rat and mouse (130). This
has practical implications because one of the emerging
therapeutic applications of B1R antagonists, analgesia,
is commonly evaluated using models involving these
species.
The first generation of B2R antagonists was based on
[
D
-Phe
7
]-BK (133), but these early peptidic compounds
showed an antagonist
/
partial agonist activity and a low
potency. This problem was progressively solved with
the second generation of B2R antagonists in which
rigidity was added to the peptide backbone by intro-
ducing non-natural aminoacid residues; thus, the spatial
orientation of the C-terminal region of the peptide
molecule critical for antagonism was more precise.
Optimal peptide B2R antagonists retain both Arg
1
and
Arg
9
residues and B1R agonists or antagonists typically
lack Arg
9
(as in [Leu
8
]des-Arg
9
-BK), just like agonists;
furthermore, the N-terminal Lys residue present in KD
also confers high affinity to peptide B1R antagonists in
human, rabbit, and other species (as in Lys-[Leu
8
]des-
Arg
9
-BK).
HOE 140 (
D
-Arg-[Hyp
3
, Thi
5
,
D
-Tic
7
, Oic
8
]-BK, Icati-
bant) is a representative of the second generation of
peptide B2R antagonists (134) and has been exploited in
more than 700 research papers and several clinical
studies. Significant species-related differences in potency
and competitive behavior are typically seen with kinin
receptor antagonists; thus HOE 140 is potent and com-
petitive at the human B2R, but insurmontable and
atypically induces a slow B2R endocytosis of the rabbit
B2R (135). The selectivity of HOE 140 for the human
B2R is fair, but not complete; the fragment des-Arg
10
-
HOE 140 is predominantly a B1R antagonist (130).
Second generation B1R antagonists include the N-
terminal Lys residue or its analog Orn that affords
high affinity for the human B1R (as in B-9858
=
Lys-
Lys-[Hyp
3
, Igl
5
,
D
-Igl
7
, Oic
8
]des-Arg
9
-BK or R-954
=
Ac-
Orn-[Oic
2
,
α
-MePhe
5
,
D
-
β
Nal
7
, Ile
8
]des-Arg
9
-BK) (136).
The presence of
D
-Arg instead of Lys, as in des-Arg
10
-
The Kallikrein-Kinin System
15
HOE 140 and
D
-Arg-[Hyp
3
, Igl
5
,
D
-Igl
7
, Oic
8
]-BK, is
associated with a strong decrease in affinity in human
and rabbit, but not in the mouse B1R, corroborating
the previous interpretation.
Efforts to create a third generation of kinin receptor
antagonists have evolved towards conventional non-
peptide drug development programs, with oral bio-
availability, higher lipophilicity, and lower molecular
weight increasingly represented. Various chemical sub-
classes of nonpeptide B2R antagonists have been dis-
covered: phosphonium compounds, as in WIN64338,
WIN62318 (136 – 138); quinoline and imidazol [1,2-
a
]
pyridine family, such as FR165649, FR167344,
FR173657, and FR184280; compound 38; substituted
1,4-dihydropyridines; CP2522; and CP0597 (138).
Several nonpeptide B1R antagonists have also been
synthesized: PS020990 (139), benzodiazepine-based
compounds, such as compound 11 and 12 (140); power-
ful oxo-sulfonyl agents (141 – 143).
Some kinin antagonists have also been discovered
as natural compounds: martinelline, a pyrroloquinoline
alkaloid isolated from the plant
Martinella iquitosensis
,
is the most remarkable example (144).
Table 1.
Pharmacological and clinical application of kinin B
1
-receptor ligands
Ligands
Application
Studies
References
Agonists
R-838 (Sar-[
D
-Phe
8
]des-Arg
9
-BK)
Metabolically stable
Rabbit
128
High affinity and selectivity
Hypertension
Stimulation of vasculature formation
(following ischemia)
Rodent
130, 281
Antagonists
[Leu
8
]des-Arg
9
-BK
Pain
Rat
132
Mice
302, 317
Ischemic vascular disease
Mice
130, 281
Lys-[Leu
8
]des-Arg
9
-BK
Optimal B1R antagonist
Human B1R
130
Ac-Lys-[MeAla
6
, Leu
8
]des-Arg
9
-BK
Metabolically stable (not very potent
compared with the affinity of the
reference compound Lys-[Leu
8
]des-Arg
9
-BK)
Rabbit
318
R-715 (Ac-Lys-[
β
D
-Nal
7
, Ile
8
]des-Arg
9
-BK)
High affinity
Human and rabbit B1R
130
Allergic lung inflammation
Mice
319
B9858 (Lys-Lys-[Hyp
3
, Igl
5
,
D
-Igl
7
,
Oic
8
]des-Arg
9
-BK)
Fairly high selectivity for B1R due to Lys
0
Human and rabbit B1R
320
Metabolically stable residue
des-Arg
10
-HOE 140
Residual antagonistic effects on B2R
Rabbit jugular vein, guinea pig ileum,
rabbit aorta
321
Moderate affinity
B9430 (
D
-Arg-[Hyp
3
, Igl
5
,
D
-Igl
7
, Oic
8
]-BK)
Mixed B1R and B2R antagonist even if
desArg
9
fragment has substantial
selectivity for B1R
Demonstration of compatibility of B1R
and B2R structure by the accommodation
of a single antagonist pharmacophore
322
R-954 (Ac-Orn-[Oic
2
,
α
-MePhe
5
,
D
-
β
Nal
7
,
Ile
8
]des-Arg
9
-BK)
Allergic lung inflammation
Mice
319, 323
Airway allergy
Rat model, speculative on human
324
PS020990
Potent and competitive B1R antagonist
High affinity
Human receptor (no in vivo data)
139
Compound 12 (benzodiazepine-based
structure)
Selective antagonist
Human and rat B1R in vitro (equal activity)
140
Benzo-sulfonylamide componds
Powerful and selective antagonists
Compound 12
Hyperalgesia
Rat, dog, orally bioavailable
142
Compound 11
Speculative on pain, inflammation and sepsis
Rabbit aortic preparations,
rabbit jugular vein
141
SSR240612
Inflammation and hyperalgesia
Mice, rat, oral activity
143