Published online: November 27, 2015
News & Views
When endothelial cells go rogue
Pei-Yu Chen1 & Michael Simons1,2
Endothelial-to-mesenchymal
transition
(EndMT) is a poorly understood phenomenon that leads to endothelial cells acquiring a variety of different mesenchymal
fates. This results in a number of pathological consequences of considerable clinical
significance in diseases ranging from
cavernous malformations in the brain to
tissue fibrosis, atherosclerosis, and cancer.
Importantly, while there appears to be a
number of different triggers activating
EndMT, the final common pathway driving
the transition appears to be the same.
See also: R Cuttano et al (January 2016)
A
search for the Theory of Everything
has been as much a feature of biological research as it has been a singular
preoccupation of theoretical physicists.
Lately, endothelial-to-mesenchymal (EndMT)
transition has been making a concerted
push to visibility, respectability, medical
relevance, and, perhaps, a central place in a
host of critical biological and pathophysiological processes.
EndMT was initially discovered as an
essential step in heart development.
Endothelial cells lining atrio-ventricular
canal and the outflow tract undergo EndMT
and invade surrounding tissues to form the
valves and septa of the heart (van Meeteren
& ten Dijke, 2012). Further studies demonstrated EndMT occurrence in a broad spectrum of conditions including tissue fibrosis,
cancer, and heterotopic ossification (van
Meeteren & ten Dijke, 2012), as well as more
recently in neointima formation (Chen et al,
2012; Cooley et al, 2014), cerebral cavernous malformations (CCM) (Maddaluno
et al, 2013), and atherosclerosis (Chen et al,
2015) among others. The term “EndMT”
encompasses a range of transformations—
from the complete transdifferentiation of an
endothelial cell into a mesenchymal cell to a
number of intermediate states where cells
retain some of their endothelial characteristics while acquiring distinctly mesenchymal
features. This EndMT feature has a close
parallel with the related and better understood
process of epithelial-to-mesenchymal transition (EMT). Another parallel is that in both
cases replacement of VE-cadherin (EndMT) or
E-cadherin (EMT) with N-cadherin is a key
feature of mesenchymal transformation.
The direct pathological consequences of
EndMT (whether partial or complete) include
not only excessive production of mesenchymal cells (smooth muscle cells, fibroblasts,
osteoblasts, adipocytes, etc.) but also abundant deposition of the extracellular matrix
and increased thrombogenicity. Furthermore,
evidence is now emerging that links EndMT
directly to disease progression (Maddaluno
et al, 2013; Chen et al, 2015). While considerable effort has gone into establishing that
EndMT occurs and plays a role in a variety of
pathologic conditions, the regulation of this
process remains poorly understood.
One relatively well-accepted concept is
that the final common pathway leading to
EndMT involves TGF-b-driven activation of
transcription factors such as Snail, Slug,
Twist, and zinc finger E-box binding homeobox (ZEB) 1 and 2. These factors in turn
upregulate mesenchymal gene expression.
Yet molecular triggers of EndMT have not
been well defined. Inflammation and
mechanical stress are two conditions
frequently associated with EndMT and a
recent study demonstrated that both can
activate TGF-b signaling in the endothelium
via downregulation of FGFR1 expression
(Chen et al, 2015). The situation appears
somewhat different in CCM, a common
hemorrhagic vascular anomaly that can
present both in sporadic and in familial
autosomal dominant forms. Patients with
CCM have loss-of-function mutations in one
of the three CCM-encoding genes (CCM1,
CCM2, or CCM3) that result in formation of
enlarged and irregular blood vessels with
high propensity to bleeding leading to
seizures and strokes (Fischer et al, 2013).
In an elegant series of studies in this issue
of EMBO Molecular Medicine, Cuttano et al
(2016) show that the loss of CCM1 activates
the MEKK3-MEK5-ERK5 cascade that, in
turn, upregulates KLF4. KLF4 then promotes
TGF-b/BMP signaling via production of
BMP6. A deletion of KLF4 in the endothelial
cell-specific CCM1 knockout reduced the size
and number of the lesions, prevented vascular
leakage, and restored endothelial/ astrocyte
proximity in mice (Cuttano et al, 2016). Importantly, in a previous study, the same group
demonstrated changes consistent with EndMT
such as increased pSmad3 and N-cadherin levels
in human lesions due to mutations in CCM1
and CCM2 genes (Maddaluno et al, 2013).
The finding of the central role played by
KLF4 in this disease setting is certainly intriguing. KLF4 is a zinc finger protein that functions
as one of the reprogramming “Yamanaka
factors” in pluripotent stem cell induction
cocktails. It is also a shear stress-inducible
gene, a regulator of endothelial activation in
response to pro-inflammatory stimuli, a
phenotypic modulator of smooth muscle cells
and a protein intimately, if confusingly,
involved with TGF-b signaling (Yoshida &
Hayashi, 2014). Similar to the strong induction
of its expression seen in CCM lesions, KLF4 is
also upregulated following the loss of endothelial FGF signaling input (Chen et al, 2014),
another EndMT trigger, although its role in this
form of EndMT has not been established. We
thus have at least two different paradigms
leading to EndMT activation: One is the loss of
1 Section of Cardiovascular Medicine, Department of Internal Medicine, Yale Cardiovascular Research Center, New Haven, CT, USA. E-mail: [email protected]
2 Department of Cell Biology, Yale University School of Medicine, New Haven, CT, USA. E-mail: [email protected]
DOI 10.15252/emmm.201505943 | Published online 27 November 2015
ª 2015 The Authors. Published under the terms of the CC BY 4.0 license
EMBO Molecular Medicine
Vol 8 | No 1 | 2016
1
Published online: November 27, 2015
EMBO Molecular Medicine
Endothelial-to-mesenchymal transition
Pei-Yu Chen & Michael Simons
tory is supported by the NIH Grant R01 HL053793,
HL084619, HL062289, and P01 HL107205 (M.S.)
FGF
CELL MEMBRANE
II I
II I
References
CCM1
BMPR
Bravi L, Rudini N, Cuttano R, Giampietro C,
V
T
TGFβR
FRS2α
FGFR1
MEK5 P
Sulindac metabolites decrease cerebrovascular
malformations in CCM3-knockout mice. Proc
ERK5 P
SIGNALLING
CASCADE
Maddaluno L, Ferrarini L, Adams RH, Corada M,
Boulday G, Tournier-Lasserve E et al (2015)
MEKK3 P
let-7
• INFLAMMATION
• SHEAR STRESS
Natl Acad Sci USA 112: 8421 – 8426
Chen PY, Qin L, Barnes C, Charisse K, Yi T, Zhang X,
Ali R, Medina PP, Yu J, Slack FJ et al (2012) FGF
KLF4
regulates TGF-beta signaling and endothelial-toBMP6
P
Smad2
Smad1
NUCLEAR
TRANSCRIPTION
FACTORS
mesenchymal transition via control of let-7
KLF4
P
Smad3
P
miRNA expression. Cell Rep 2: 1684 – 1696
Chen PY, Qin L, Tellides G, Simons M (2014)
Fibroblast growth factor receptor 1 is a
Snail
Slug
Twist
ZEB
key inhibitor of TGFbeta signaling in the
endothelium. Sci Signal 7: ra90
EndMT genes
Chen P-Y, Qin L, Baeyens N, Li G, Afolabi T, Budatha
M, Tellides G, Schwartz MA, Simons M (2015)
Endothelial-to-mesenchymal transition drives
atherosclerosis progression. J Clin Invest
doi:10.1172/JCI82719
Cooley BC, Nevado J, Mellad J, Yang D, St Hilaire C,
CELL PHENOTYPE
TYPE
Mesenchymal
‘stem’ cell
• FSP1
• Sca1
Fibroblast
• Collagen
• Fibronectin
• N-cadherin
• Vimentin
Smooth muscle cell
• SM α-actin
• SM22α
• SM-calponin
• Notch3
Negro A, Fang F, Chen G, San H, Walts AD et al
(2014) TGF-beta signaling mediates endothelialto-mesenchymal transition (EndMT) during vein
graft remodeling. Sci Transl Med 6: 227ra234
Cuttano R, Rudini N, Bravi L, Corada M, Giampietro
C, Papa E, Morini MF, Maddaluno L, Baeyens N,
Figure 1. Two different signaling pathways leading to EndMT.
The loss of either CCM (left) or FGFR1 (right) leads to EndMT via two distinctly different pathways converging on
activation of KLF4/Smad-dependent transcription of the mesenchymal program.
Adams RH et al (2016) KLF4 is a key determinant
in the development and progression of cerebral
cavernous malformations. EMBO Mol Med 8:
6 – 24
Fischer A, Zalvide J, Faurobert E, Albiges-Rizo C,
CCM gene-mediated suppression of activation
of the ERK5/KLF4 pathway, and the other is
the inflammation- and shear stress-driven loss
of FGF signaling input that leads to increased
expression of TGF-b-related genes. In both
cases, the end result is Smad-dependent activation of EndMT transcriptional program (Fig 1).
It is very likely that other means of TGF-b/
BMP/Smad signaling activation also exist.
Indeed Wnt/b-catenin and Notch signaling
pathways may well be involved as well.
This presents an interesting therapeutic
dilemma. Presently, there are no treatments
able to prevent development or reduce the
size of existing CCMs nor are there effective
approaches for prevention of EndMT in other
disease settings, albeit TGF-b/BMP inhibitors
and drugs affecting signaling pathways
potentiating TGF-b/BMP such as Wnt/
b-catenin show promise in mice models
(Maddaluno et al, 2013; Bravi et al, 2015;
Chen et al, 2015). Yet, TGF-b signaling has a
2
EMBO Molecular Medicine Vol 8 | No 1 | 2016
number of beneficial effects such that
systemic TGF-b inhibition may not be
desirable. Thus, the ability to selectively
target various forms of EndMT, based on their
specific signaling signatures and specfic ways
of activating the common TGF-b/BMP/Smad
pathway, may be of considerable interest.
EndMT is now emerging as a grand unifying concept bringing together a variety of
initiating impulses that make endothelial
cells “go rogue” and adapt mesenchymal
features that play a central role in disease
progression. The study from Cuttano et al
(2016) is an important advance that further
cements the central role of EndMT in a
human disease entity and lays the groundwork for novel therapeutic developments.
Tournier-Lasserve E (2013) Cerebral cavernous
malformations: from CCM genes to endothelial
cell homeostasis. Trends Mol Med 19: 302 – 308
Maddaluno L, Rudini N, Cuttano R, Bravi L,
Giampietro C, Corada M, Ferrarini L, Orsenigo F,
Papa E, Boulday G et al (2013) EndMT contributes
to the onset and progression of cerebral
cavernous malformations. Nature 498: 492 – 496
van Meeteren LA, ten Dijke P (2012) Regulation of
endothelial cell plasticity by TGF-beta. Cell
Tissue Res 347: 177 – 186
Yoshida T, Hayashi M (2014) Role of Kruppellike factor 4 and its binding proteins in vascular
disease. J Atheroscler Thromb 21: 402 – 413
License: This is an open access article under the
Acknowledgements
terms of the Creative Commons Attribution 4.0
The authors wish to apologize for not having been able
License, which permits use, distribution and repro-
to individually cite all the relevant original papers, due
duction in any medium, provided the original work
to space limits. Research activity in the authors’ labora-
is properly cited.
ª 2015 The Authors