Vasco Claro1, Albert Ferro1. 1. School of Cardiovascular Medicine and Sciences, British Heart Foundation Centre of Research Excellence, King's College London, London, UK.
Abstract
The netrins form a family of laminin-related proteins which were first described as modulators of cell migration and axonal guidance during fetal development. Netrin-1 is the most extensively studied member of this family and, since its discovery, non-neural roles have been associated with it. Together with its receptors, DCC/neogenin and UNC5, netrin-1 has been shown to be involved in the regulation of angiogenesis, organogenesis, cancer and inflammation. An NF-κB-dependent truncated isoform of netrin-1 has also been shown to be produced in endothelial and some types of cancer cells, which both accumulates in and affects the function of the nucleus. In atherosclerosis, conflicting roles for netrin-1 have been reported on plaque progression via its receptor UNC5b. Whereas endothelial-derived netrin-1 inhibits chemotaxis of leukocytes and reduces the migration of monocytes to the atherosclerotic plaque, netrin-1 expressed by macrophages within the plaque plays a pro-atherogenic role, promoting cell survival, recruiting smooth muscle cells and inhibiting foam cell egress to the lymphatic system. In contrast, there is evidence that netrin-1 promotes macrophage differentiation to an alternative activated phenotype and induces expression of IL-4 and IL-13, while downregulate expression of IL-6 and COX-2. Further work is needed to elucidate the precise roles of the two isoforms of netrin-1 in different cell types in the context of atherosclerosis, and its potential as a putative novel therapeutic target in this disease.
The netrins form a family of laminin-related proteins which were first described as modulators of cell migration and axonal guidance during fetal development. Netrin-1 is the most extensively studied member of this family and, since its discovery, non-neural roles have been associated with it. Together with its receptors, DCC/neogenin and UNC5, netrin-1 has been shown to be involved in the regulation of angiogenesis, organogenesis, cancer and inflammation. An NF-κB-dependent truncated isoform of netrin-1 has also been shown to be produced in endothelial and some types of cancer cells, which both accumulates in and affects the function of the nucleus. In atherosclerosis, conflicting roles for netrin-1 have been reported on plaque progression via its receptor UNC5b. Whereas endothelial-derived netrin-1 inhibits chemotaxis of leukocytes and reduces the migration of monocytes to the atherosclerotic plaque, netrin-1 expressed by macrophages within the plaque plays a pro-atherogenic role, promoting cell survival, recruiting smooth muscle cells and inhibiting foam cell egress to the lymphatic system. In contrast, there is evidence that netrin-1 promotes macrophage differentiation to an alternative activated phenotype and induces expression of IL-4 and IL-13, while downregulate expression of IL-6 and COX-2. Further work is needed to elucidate the precise roles of the two isoforms of netrin-1 in different cell types in the context of atherosclerosis, and its potential as a putative novel therapeutic target in this disease.
Atherosclerosis, the principal pathophysiology underlying ischaemic heart
disease and stroke, is caused by progressive accumulation of cholesterol in
the arterial wall. In recent years, it has become clear that there is an
important inflammatory component to atherosclerosis, which progresses over a
long period of time. Atherosclerosis is initially triggered by an
accumulation of low-density lipoproteins (LDL) in the arterial wall, between
the endothelium and the tunica intima, and mostly in artery segments where
endothelial integrity is compromised, such as curvatures or branching
points, due to local turbulence of blood flow.[1] Once inside the arterial wall, LDL undergoes a chain of chemical and
enzymatic reactions which ultimately results in an oxidized form of LDL,
changing the biological function of its membrane phospholipids and creating
degradation products, which promote oxidative stress.[2] Oxidized LDL (oxLDL) activates the endothelium and induces the
expression of chemokines and adhesion molecules, such as MCP-1 and VCAM-1,
leading to the recruitment of leukocytes, especially monocytes.[3] In the plaque, monocytes differentiate into macrophages and recognise
oxLDL as a pathogen-associated molecular pattern, through its scavenger and
toll-like receptors. This recognition leads to macrophage shifting into a
pro-inflammatory phenotype, increasing the expression of pro-inflammatory
cytokines and ultimately promoting atherogenesis.[4] Macrophages take up cholesterol from the sub-intimal layer and
transform into foam cells, losing their ability to egress to the lymphatic
system and contributing to the development of necrotic plaque core and
thinning of the fibrous cap, the plaque thereby becoming more vulnerable to
erosion or rupture with resultant superadded arterial thrombosis giving rise
to cardiovascular events.[5] Netrins and semaphorins have been studied as strong candidates for
causing the loss of mobility observed in foam cells.[6,7]
Nonetheless, more research is required to completely understand how these
molecules modulate atherosclerosis.Recent research has focussed on novel anti-inflammatory approaches which may
offer benefits over and above standard cardiovascular therapies such as
lipid-lowering and anti-platelet drugs. In this review, we will focus on the
role of netrin-1 in normal physiology and in the modulation of inflammation,
and we will discuss specifically how this may relate to the pathophysiology
of inflammation in atherosclerosis and how this may pave the way to novel
therapeutic approaches in this disease.
The netrin family of proteins
Netrins are a family of laminin-like proteins, first described in the early
1990s as axonal guidance cues during embryonic development. In humans, four
different netrins have been described (netrin-1, -3, -4 and -5: Figure 1),
presenting a secondary structure which is highly conserved between species.
Netrin-1 to -4 are composed of one N-terminal domain VI, a positively
charged C-terminal domain and three laminin V-type epidermal growth factor
(EGF) domains with the same characteristic number and spacing of Cys
residues as observed in laminins.[8,9] Unlike the
structures of netrin-1, -3, and -4, full length netrin-5 comprises three
V-type EGF and the C-terminal domains but lacks the N-terminal laminin VI
domain (Figure 1).[10] While netrin-1 and 3 present similar V- and VI-type domains to those
observed in laminin-γ1, netrin-4 amino-terminal domains are most similar to
the amino terminus of the laminin-β1 chain1.[11,12] The C-terminal
domain is the part which varies the greatest between species. The C-terminal
sequences are rich in basic amino acids, which are binding sites for
membrane glucolipids and proteoglycans, thereby allowing interaction with
cell surface components and extracellular matrix.[9]
Figure 1.
Schematic structure of the netrin family members.
Schematic structure of the netrin family members.Most of the members of the netrin family are secreted proteins and they are
bifunctional, acting as both short- and long-range signals.[13] They bind to different receptors due to the different homologies
between them.Although netrins were first defined due to their role in the development of the
central nervous system, more recently other physiological and pathological
roles have been ascribed to these proteins. There is evidence that netrins
play important roles in different types of cancer and in the morphogenesis
of various tissues;[14] nonetheless, the only netrin that has been described in the
cardiovascular system to date is netrin-1, and therefore we will concentrate
on this family member in this review.
Netrin-1
Netrin-1 is the most extensively studied of all the netrins. In humans, it is
encoded by the gene NTN1 in chromosome 17 and encodes a highly conserved 604
amino acid protein. It is expressed by different types of cells and can be
found in most tissues. Relative quantification of mRNA shows high expression
of netrin-1 in brain, heart, kidney and lungs, and low expression in the
liver, intestine and spleen.[15]Netrin-1 possesses two receptor-binding sites in the two ends of an
elongated flower-shaped structure (Figure 2), which interact with two
different receptor molecules.[16]
Figure 2.
Predicted unbound netrin-1 tertiary structure generated by Phyre2
software, using the human netrin-1 UniProt sequence. It has a
flower-shaped structure, formed by the N-terminus on the bottom
right hand-side of the Figure (green and blue motifs), followed
by three V-type epidermal growth factor domains and the
C-terminus at the top (red motif).
Predicted unbound netrin-1 tertiary structure generated by Phyre2
software, using the humannetrin-1 UniProt sequence. It has a
flower-shaped structure, formed by the N-terminus on the bottom
right hand-side of the Figure (green and blue motifs), followed
by three V-type epidermal growth factor domains and the
C-terminus at the top (red motif).The functions of netrin-1 were initially identified as mediation of axonal
orientation, axonal outgrowth and neuronal migration.[17] The ability of netrin-1 to repel neuronal cells makes it a potential
candidate for the regulation of inflammatory cell migration. Netrin-1 has
been identified as having roles in angiogenesis,[18] tumorigenesis,[19] organogenesis[20] and inflammation, suggesting that it regulates cell migration in a
broad context, as described below.Netrin-1 has been reported as an angiogenic agent in both physiological and
pathological contexts, acting either as promoter or as inhibitor of
angiogenesis. It is well established that netrin-1 participates in both
angiogenesis and morphogenesis of the vasculature during fetal life and as
well as in adult homeostasis. Similar to what is observed in axon formation,
the specific effects of netrin-1 depend on which receptors it binds to. Park
et al. showed that netrin-1 not only promotes angiogenesis itself, but also
enhances the angiogenic activity of vascular endothelial growth factor
(VEGF) both in vitro and in vivo.[18]Pro-angiogenic effects were also observed in pathological
angiogenesis, where it promoted the creation of new vessels in conditions
such as cancer. Shimizu et al. and Akino et al. respectively showed that
netrin-1 promotes the creation of new vessels in glioblastoma and paediatric
medulloblastoma, leading to a more aggressive pathology.[21,22]Netrin-1 has also been classified as an oncogene due to its various roles in
tumorgenesis, including inhibition of macrophage recruitment, promotion of
cell survival, stimulation of invasiveness and of pathological angiogenesis.
Several groups have shown that netrin-1 is upregulated in glioblastoma,
medulloblastoma and other malignancies, and its overexpression leads to
poorer prognosis.[21,22] The ability of netrin-1 to inhibit apoptosis is
related to binding to its specific receptors, which trigger apoptosis in the
absence of the ligand. Cancers producing excessive amounts of netrin-1 will
therefore have their apoptosis program inhibited. This interaction between
netrin-1 and its receptors will be described in more detail in the following
section.In addition to axonal guidance and angiogenesis, the role of netrin-1 and its
receptors in organogenesis have been studied in terms of its regulation of
migration, differentiation and branching morphogenesis. The importance of
this molecule in the development of tissues, such as lungs, pancreas or
mammary glands, has been revealed in the last couple of decades.[20,23,24]
Dalvin et al. showed that netrin-1 controls epithelial-mesenchymal
interaction, which is a key process in lung organogenesis.[23] Furthermore, De Breuck et al. demonstrated that netrin-1 is involved
in pancreatic morphogenesis and tissue remodelling in vivo.[24] Together, these data show that netrin-1 and its receptors play an
important role in the morphogenesis of branched tissues.As mentioned above, there is evidence that netrin-1 mediates inflammation due
to its effect on immune cells, having a role in their recruitment and
motility, as well as in their stimulation and production of cytokines.
Nonetheless, it is controversial whether a high expression of netrin-1 is
beneficial or detrimental. Some studies show that netrin-1 has a protective
effect in pathological models. In vivo work performed by
Mirakaj et al. showed that expression of netrin-1 was repressed in acute
inflammation in lung injury and peritonitis. Moreover, this group showed
that treatment with exogenous netrin-1 dampened the inflammation and reduced
tissue damage.[25,26] In accordance with this, Grenz et al.
demonstrated that acute kidney injury was aggravated in a mouse model with
partial netrin-1deficiency, and treatment with exogenous netrin-1
attenuated the effects observed, reducing tissue injury.[27] In contrast, Mediero et al. have proposed a pro-inflammatory role for
netrin-1, which is required for the development of particle-induced
osteolysis, hypothesising that it leads to the recruitment, retention and
enhancement of osteoclast differentiation and activity.[28] Van Gils and colleagues have provided evidence that netrin-1 promotes
atherosclerosis due to its effect on inhibiting the emigration of
macrophages and increasing recruitment of smooth muscle cells to the plaque.[29] Ramkhelawon et al. showed that adipose tissue produces high amounts
of netrin-1, promoting macrophage retention, chronic inflammation and
insulin resistance.[30] The pro- and anti-inflammatory effects of netrin-1 therefore appear
to depend on the model being studied.
Netrin-1 receptors
The function of netrin-1 in regulating neuronal navigation has been associated
with two main types of receptors, termed dependence receptors: DCC (Deleted
in Colorectal Carcinoma) and its orthologue neogenin, and UNC5
(UNCoordinated protein 5) receptors (UNC5a-d). These are receptors that play
a dual role triggering different cellular responses in the presence or
absence of ligand. On the one hand cell proliferation, migration and
survival are promoted in the presence of netrin-1 whilst, on the other hand,
apoptosis is induced in the absence of the ligand. Moreover, relative
expression levels of both families of receptors will mediate the attraction
or repulsion effects observed in neuronal development, which will be further
explored below.[31,32]DCC and neogenin are receptor homologues which share nearly half of their amino
acid identity and display the same secondary structure.[33] They are single-pass transmembrane receptors with an extensive
extracellular region, which comprises four immunoglobulin (Ig)-like domains
in a horseshoe configuration,[34] followed by six fibronectin type III domains (FnIII). The
intracellular portion of the receptor contains three conserved motifs, which
are responsible for signalling (P1, P2 and P3) (Figure 3(a)).[35] Biochemical studies suggest that FnIII4, FnIII5 and FnIII6 are the
DCC domains responsible for the binding of netrins.[36,37]
Figure 3.
Netrin-1 dependence receptors. (a) Schematic structure of DCC and
UNC5 cytoplasmic and extracellular domains. (b) Netrin-1 binding
to DCC-DCC and DCC-UNC5 complexes, promoting neuronal attraction
and repulsion respectively. (c) Apoptosis promoted by DCC
receptor in the absence of netrin-1. (d) Apoptosis promoted by
UNC5 receptor in the absence of netrin-1.
Netrin-1 dependence receptors. (a) Schematic structure of DCC and
UNC5 cytoplasmic and extracellular domains. (b) Netrin-1 binding
to DCC-DCC and DCC-UNC5 complexes, promoting neuronal attraction
and repulsion respectively. (c) Apoptosis promoted by DCC
receptor in the absence of netrin-1. (d) Apoptosis promoted by
UNC5 receptor in the absence of netrin-1.UNC5 receptors are also single-pass transmembrane receptors with an
extracellular region consisting of two Ig-like motifs, followed by two
thrombospondin I modules (TSP).[38] On the cytosolic side, UNC5 receptors have a long tail which consists
of a ZU5 domain, a DCC-binding motif (UPA) and a death domain (DD) (Figure 3(a)).[39] Both Ig-like domains may be responsible for the primary binding of
netrins and it is suggested that the second motif (Ig2) contributes more to
the binding than Ig1.[36]Different signalling pathways have been proposed for these receptors and
receptor complexes. In the development of the central nervous system,
netrin-1 exerts a bifunctional effect, depending on which combination of
receptors it binds to. Netrin-1 can promote attraction if it binds to a
DCC-DCC complex, leading to the activation of several cascades of events
which promote directional axonal outgrowth. The opposite effect occurs,
namely repulsion, when netrin-1 binds exclusively to an UNC5 receptor or
UNC5 and DCC simultaneously, promoting short- and long-range repulsion
respectively (Figure
3(b)).[32,40]DCC was originally characterised as a suppressor of colorectal cancer when a
pronounced reduction of its expression was observed in several colorectal
carcinoma cell lines, due to a deletion in chromosome 18q21.[41] Many studies have established the importance of DCC/neogenin and
netrin-1 in different pathologies. For example, Reyes-Mugica et al. and
Koren et al. showed that loss of DCC expression is associated with poor
prognosis in patients with glioma and breast cancer respectively.[42,43] In
the immune system, the DCC/neogenin receptor family mediates a pro-migration
response to netrin-1. Boneschansker et al. reported that netrin-1 plays an
important role inducing migration of leukocytes via its receptor neogenin.
Netrin-1 is an inducer of CD4+ T cells, and its interactions with
neogenin augmented the migration of these leukocytes in an in
vitro model. The same group also reported that
neogenin-expressing T cells are recruited in pro-inflammatory environments
where netrin-1 is expressed. Using a humanized SCIDmouse model engrafted
with human skin, injection of netrin-1 led to an increase in T cell
infiltration compared to vehicle injection.[44]Apoptosis and cell survival have also been identified as downstream effects
dependent on netrin-1 and its receptors. Mehlen et al. showed that, in the
absence of netrin-1, DCC induces apoptosis. Moreover, this receptor is a
caspase substrate which loses its pro-apoptotic effect if there is a
mutation at the site where caspase-3 cleaves DCC (Figure 3(c)).[45] Cleavage of UNC5 by caspase-3 similarly induces cell apoptosis,
through its intracellular death domain (Figure 3(d)). However, contrary to
that which is observed in DCC, mutation of the caspase-3 cleavage site of
UNC5 does not inhibit apoptosis; on the other hand, deletion of the
myristoylation signal peptide, which is localized in the C-terminus of the
receptor, prevents its induction of apoptosis.[31]Netrin-1 has been shown to play a key role in modulating the migration of
leukocytes through its UNC5b receptor.[15] Although the precise signalling pathway is not yet understood, there
is strong evidence that this event is an important modulator in the
recruitment of leukocytes and inflammation; this will be described further
below, in the context of the role of netrin-1 in atherosclerosis.CD146, an adhesion molecule expressed by endothelial cells, mostly in the
vasculature, has also been described as a netrin-1 receptor. The molecular
mechanisms associated with this receptor are angiogenesis, vessel
permeability and leukocyte transendothelial migration.[46] CD146 has been described as a pro-angiogenic mediator,
counterbalancing the UNC5b anti-angiogenic signal. Further studies are
needed to clarify which signalling pathways it utilises in leukocytes to
modulate their migration.Down syndrome cell adhesion molecule (DSCAM) has more recently been proposed as
a novel receptor for netrin-1. Its collaboration with DCC has been shown to
be essential for commissural axons to grow towards and across the midline.[47] There is no evidence to date that this receptor plays a role in
cardiovascular disease, and therefore it will not be explored further in
this review.
Netrin-1 isoforms
In the last decade, a number of studies have shown the existence of two
isoforms of netrin-1 (Figure 4), which may help to explain some of the apparently
conflicting effects reported for this protein.[19,48,49]Delloye-Bourgeois
et al. first described a new netrin-1 isoform which is truncated and
abundantly expressed in a neuroblastoma cell line model.[19] This isoform is produced by an alternative promoter, localised in
intron I of the netrin-1 DNA sequence, which is NF-κB dependent.[48] They reported that the C-terminus of netrin-1 mediates its nucleolar
localisation and the N-terminus prevents it, the latter instead promoting
its secretion from the cell. As a result, the truncated netrin-1, which
lacks part of the N-terminal domain, stays localised within the nuclei.[19] Truncated netrin-1 has been shown to be expressed in a range of types
of cancer, and high expression of this isoform appears to associate with
poorer patient prognosis.[19,49] This truncated
isoform affects nucleolar function; for instance, in cancer cells,
overexpression of this isoform promotes ribosome synthesis, tumour growth
and cell proliferation.[19]
Figure 4.
Schematic structure of netrin-1 isoforms.
Schematic structure of netrin-1 isoforms.Passacquale et al. showed that endothelial cells also express the truncated
isoform of netrin-1. TNF-α treatment disturbs the balance of expression
between the two isoforms, promoting upregulation of the truncated one and
downregulation of the secreted (full length) isoform. They also demonstrated
that the two isoforms are regulated differently in endothelial cells. As
previously described by Delloye-Bourgeois et al. in neuroblastoma cells, the
truncated isoform promoter was found to be NF-κB dependent, so that treating
endothelial cells with an NF-κB inhibitor supressed TNF-α-induced expression
of the truncated isoform, both at the transcriptional and protein levels.
They also showed that the induction of secreted full-length netrin-1 is
NF-κB-independent, but on the other hand is dependent on histone H3
acetylation, which in turn could be induced by aspirin treatment of
endothelial cells.[48]The effects of the nucleolar isoform in other cell types, such as leukocytes,
has not been determined as yet. Further work is required to clarify whether
this isoform is expressed in inflammatory conditions, such as
atherosclerosis, and what its contribution to the modulation of such
inflammatory pathologies might be.
Netrin-1 in atherosclerosis
As described earlier, atherosclerosis is a chronic inflammatory disease that
progresses over many years and often is only detected when the patient
develops symptoms such as myocardial ischaemia. Monocytes are recruited to
the arterial wall in areas of developing atherosclerotic plaque, where they
differentiate into macrophages and take up oxLDL, becoming immobile foam
cells, increasing the cellular burden and instability within the plaque.The reported effects of netrin-1 in atherosclerosis have been conflicting,
having been shown to play both pro- and anti-inflammatory roles in the
development of the disease. This conflicting evidence may be explained by
the temporal and spatial expression of netrin-1: endothelial-derived
netrin-1 has been reported as playing a protective role, whilst netrin-1
secreted within the atherosclerotic plaque by macrophages appears to exert a
pro-atherogenic effect. These are described in detail below.
Endothelial-derived netrin-1
Endothelial cells secrete full length netrin-1, leading to a reduction of
monocyte chemotaxis into the arterial wall (Figure 5(a)), as first
demonstrated by Passacquale et al. In ApoE–/– mice, they
found that UNC5b blockade inhibited the effects of netrin-1 with a
consequent increase in accumulation of monocytes in the
brachiocephalic artery.[48] Furthermore, Lin et al. and Bruikman et al. showed that
netrin-1 reduces the expression of adhesion molecules induced by the
pro-inflammatory cytokine TNF-α. In an in vitro model
of human aortic endothelial cells stimulated with TNF-α, a significant
reduction in the expression of VCAM-1, ICAM-1 and E-selectin was
observed when cells were co-treated with netrin-1, and high
concentrations of netrin-1 completely abolished the expression of
these adhesion molecules.[50] Significant downregulation of ICAM-1, IL-6 and MCP-1 has also
been observed in endothelial cells from patients with underlying
atherosclerosis, this effect being abolished by UNC5b blockade.[51] Furthermore, Lin et al. also showed that netrin-1 supresses
endothelial-derived cytokine production induced by TNF-α in a
concentration-dependent manner (Figures 5(b) and 6). They
demonstrated that netrin-1 selectively suppresses toll-like receptor 4
(TLR4) and inhibits the NF- κB pathway by suppressing TNF-α-induced
IKK and IκBα activation in the cytoplasm. They also showed that
netrin-1 affects the nuclear NF- κB subunit p65 by reducing its
accumulation in the nuclei. While stimulation with TNF-α led to an
increase of approximately 3.4 fold of p65 in the nuclei compared with
basal levels, treatment with 200 and 400 ng/ml of netrin-1 reduced its
nuclear accumulation by 25% and 50%, respectively. Moreover, netrin-1
suppressed NF-κB promoter activation in endothelial cells after TNF-α
stimulation: TNF-α increased the promoter activity by 70-fold, and
this was supressed by 30% and 60% after treatment with 200 and
400 ng/ml of netrin-1 respectively.[50]
Figure 5.
Overview of netrin-1 effects in the context of normal
physiology and atherosclerosis. (a) Endothelial-derived
netrin-1 interacts with UNC5b receptors on the monocyte
surface, causing chemorepulsion. (b) In atherosclerosis,
pro-inflammatory cytokines act on endothelial cells to
downregulate expression of endothelial-derived netrin-1,
as well as upregulate endothelial cell adhesion molecules,
the resultant effect of both being increased monocyte
recruitment to the plaque.
Figure 6.
Proposed model for the protective role of netrin-1 in
endothelium. In the absence of netrin-1, pro-inflammatory
stimuli induce the TLR4 cascade leading to transmigration
of NF-κB to the nucleus, promoting expression of
pro-inflammatory cytokines and adhesion molecules. The
protective effects of netrin-1 are believed to be exerted
through inhibition of IKK activation and IκB
phosphorylation and degradation, via the UNC5b receptor,
thereby preventing nuclear translocation. Image adapted
from Mehlen et al.[45]
Overview of netrin-1 effects in the context of normal
physiology and atherosclerosis. (a) Endothelial-derived
netrin-1 interacts with UNC5b receptors on the monocyte
surface, causing chemorepulsion. (b) In atherosclerosis,
pro-inflammatory cytokines act on endothelial cells to
downregulate expression of endothelial-derived netrin-1,
as well as upregulate endothelial cell adhesion molecules,
the resultant effect of both being increased monocyte
recruitment to the plaque.Proposed model for the protective role of netrin-1 in
endothelium. In the absence of netrin-1, pro-inflammatory
stimuli induce the TLR4 cascade leading to transmigration
of NF-κB to the nucleus, promoting expression of
pro-inflammatory cytokines and adhesion molecules. The
protective effects of netrin-1 are believed to be exerted
through inhibition of IKK activation and IκB
phosphorylation and degradation, via the UNC5b receptor,
thereby preventing nuclear translocation. Image adapted
from Mehlen et al.[45]From a different perspective, Passacquale et al. showed that
pro-inflammatory stimuli cause downregulation of endothelial-derived
netrin-1 expression.[48] Thus, it would be expected that the protective effect of
netrin-1 will decrease as the atherosclerotic process proceeds,
thereby further accelerating its progression. Indeed, Bruikman and
colleagues showed that netrin-1 plasma levels are inversely correlated
with arterial wall inflammation as well as total plaque volume.[51]
Macrophage-derived netrin-1
Macrophages can also express netrin-1.[15,29,30,52] Nonetheless, what effects macrophage-derived
netrin-1 may exert on the ability of macrophages to migrate in
response to chemokines or on modulation of their phenotype remain
entirely unclear. Tadagavadi et al. and Komatsuzaki et al. reported,
in terms of netrin-1 receptors, that only UNC5b is expressed in
monocytes, granulocytes and lymphocytes, using immunohistochemistry
and quantitative RT-PCR.[15,53] We will therefore concentrate on the possible
effects of netrin-1 on macrophages on the basis that it interacts with
UNC5b only.Ly et al. showed that, in infection and pro-inflammatory models, the
levels of netrin-1 decrease and the recruitment of leukocytes to the
site is promoted. They hypothesised that netrin-1 is an
immunomodulator, which keeps leukocyte influx in check, preventing
aberrant tissue destruction. This inhibitory effect on migration was
verified in vitro and in vivo in
response to a range of chemoattractive stimuli, whilst not affecting
other functions, for example the production of superoxide.[15] Moreover, Taylor et al. demonstrated that netrin-1 is also
capable of inhibiting macrophage chemotaxis to non-chemokine attractants.[52] This group showed that the chemotaxis generated by the
complement component C5a can be partially inhibited by netrin-1,
through UNC5b.In contrast, other workers describe netrin-1 as a pro-inflammatory
mediator. Van Gils and colleagues showed that netrin-1 can promote
atherosclerosis, inhibiting the migration of the foam cells to the
lymph nodes, becoming trapped in the atherosclerotic plaque and
thereby contributing to its development and instability. In accordance
to what was reported in cancer cell models, they also reported that
netrin-1 promotes cell survival in the atherosclerotic plaque.
Deficiency of netrin-1 in the macrophages reduced atherosclerosis and
allowed egress of foam cells to the lymphatic system.[29]Similarly, Ramkhelawon et al. showed that netrin-1 promotes the retention
of macrophages in adipose tissue, enhancing metabolic dysfunction and
chronic inflammation. In an experiment where mutant mice, whose
macrophages do not express netrin-1, and wild-type mice were fed a
high fat diet, they found that recruitment of macrophages to the
adipose tissue was similar in both groups. However, on the 14th day,
the retention of macrophages was significantly lower in the mutant
mice compared with the wild-types, consistent with the hypothesis that
netrin-1 traps macrophages in the site and promotes chronic
inflammation. Furthermore, adipose tissue macrophages isolated from
the mutant mice exhibited lower mRNA expression of pro-inflammatory
markers and cytokines and a higher expression of M2 macrophage
markers, compared with the wild-type.[30]Regarding macrophage phenotypes, Ranganathan et al. and Mao et al. showed
the opposite to what was observed in the Ramkhelawon et al. study. In
their experiments, they found that higher expression of netrin-1 is
protective, enhancing the shifting of macrophage phenotype towards M2,
with concomitant upregulation of anti-inflammatory markers such as
IL-4 and IL-13 and downregulation of IL-6 and cyxlooxygenase-2
(COX-2).[54,55] Both groups showed that netrin-1 activates
anti-inflammatory pathways that are known to regulate macrophage
polarization, such as peroxisome proliferator-activated receptors
(PPAR). PPAR mediate a variety of inflammatory mechanisms, which
include repressing the activities of pro-inflammatory transcription
factors such as NF- κB or STATs.[56]Netrin-1 led to a decrease of in M1 polarization and cytokine
production induced by IFN-γ and suppressed ischaemia-reperfusion
injury, effects which were inhibited by PPAR antagonists.[54] Further work by Ranganathan et al. suggests that netrin-1
regulates inflammation and cell migration through suppression of NF-
κB, leading to downregulation of COX-2-mediated PGE2 and thromboxane
A2 production, both in vitro and in
vivo.[57]The current data suggest that the immunomodulatory effects of netrin-1
depend on which cell type expresses it and in which environment.
Macrophage-derived netrin-1 in the plaque may be detrimental because
it stops macrophage egress to lymph nodes, whereas endothelial-derived
(systemic) netrin-1 may be essential to suppress levels of
inflammation.
Conclusion
Netrin-1 plays different roles in a range of pathologies, and its effects vary
depending on which type of cells it acts upon and which receptors it binds
to. In the context of atherosclerosis, its effects may be both pro- and
anti-inflammatory, the balance depending on where it is produced
(endothelial versus macrophage-derived), their relative production, and the
relative expression of the truncated versus the full length isoform. Despite
netrin-1 isoforms having been studied and characterised in endothelial
cells, information is currently lacking as to how their expression varies in
macrophages under different inflammatory conditions. To fully understand the
relevance of netrin-1 in atherosclerosis, it will be important for
expression of its isoforms to also be characterised in the different
macrophage phenotypes. As described previously, endothelial-derived netrin-1
expression is suppressed under inflammatory conditons,[48] but further work is needed to establish what effect restoration of
endothelial-derived netrin-1 expression would have on foam cell function and
motility within the plaque.Whether the netrin-1 system presents a worthwhile therapeutic target in
atherosclerosis remains unclear. It may be that netrin-1 delivery would have
to be targeted in order to exert a useful therapeutic effect. Further work
is needed to address this question and to further elucidate the precise
roles played by the different isoforms of netrin-1 in the pathophysiology of
atherosclerosis.
Authors: Kye Won Park; Dana Crouse; Mark Lee; Satyajit K Karnik; Lise K Sorensen; Kelly J Murphy; Calvin J Kuo; Dean Y Li Journal: Proc Natl Acad Sci U S A Date: 2004-11-01 Impact factor: 11.205
Authors: Valbona Mirakaj; Cyril A Thix; Stefanie Laucher; Carina Mielke; Julio C Morote-Garcia; Marthe A Schmit; Janek Henes; Klaus E Unertl; David Köhler; Peter Rosenberger Journal: Am J Respir Crit Care Med Date: 2010-01-14 Impact factor: 21.405
Authors: Kai Xu; Zhuhao Wu; Nicolas Renier; Alexander Antipenko; Dorothea Tzvetkova-Robev; Yan Xu; Maria Minchenko; Vincenzo Nardi-Dei; Kanagalaghatta R Rajashankar; Juha Himanen; Marc Tessier-Lavigne; Dimitar B Nikolov Journal: Science Date: 2014-05-29 Impact factor: 47.728
Authors: Janine M van Gils; Merran C Derby; Luciana R Fernandes; Bhama Ramkhelawon; Tathagat D Ray; Katey J Rayner; Sajesh Parathath; Emilie Distel; Jessica L Feig; Jacqueline I Alvarez-Leite; Alistair J Rayner; Thomas O McDonald; Kevin D O'Brien; Lynda M Stuart; Edward A Fisher; Adam Lacy-Hulbert; Kathryn J Moore Journal: Nat Immunol Date: 2012-01-08 Impact factor: 25.606