00.16nrd-Targeting vascular and leukocyte communication in angiogenesis, inflammation and fibrosis

The discovery that the growth of solid tumours beyond a few millimetres in diameter is dependent on angio-genesis 1 has inspired the development of research fields related to blood vessel formation since the 1970s. Tumours were subsequently thought of as analogous to wounds that do not heal 2, and angiogenesis and com-ponents of the immune system have been shown to have crucial roles in tumour growth, wound healing and tissue restoration. Angiogenesis is clearly regu-lated by many different cell types that are present in the local vascular microenvironment, including specialized subpopulations of immune cells 3–6. During inflamma-tion, endothelial cells have a crucial role in recruiting circulating leukocytes to the tissue 7, and mural cells are also involved in the guidance of leukocytes to sites of inflammation 8. The local signals that regulate angio-genesis and inflammation are transmitted between cells by paracrine factors, secreted hormones and exosomes, homophilic and heterophilic cell–cell interactions, gap junctions and the recently described tunnelling nanotubes (TnTs)9. Importantly, the involvement of specific cell types in these processes can now be determined by high-resolution intravital microscopy . Cell migration, cell inter-actions and the presence and involvement of specific signalling molecules can thus be directly visualized in experimental in vivo models of development and disease, which, together with single cell gene profiling, will help to accelerate the identification of new drug targets 10,11. In this Review, we discuss some of the interactions between the different cells that regulate angiogenesis

and inflammation. Several new drug targets related to the communication between leukocytes and vascular cells — including interleukins, chemotactic cytokines (chemokines), and molecules within the Notch, WNT and transforming growth factor-β (TGFβ) signalling pathways — could be exploited for the treatment of diseases that involve angiogenesis or inflammation.

Microvascular cells, leukocytes and angiogenesis The microvasculature is the part of the vascular system that holds the smallest vessels: the terminal arterioles, the capillary network and the postcapillary venules (FIG. 1). Meticulous control of blood flow and vascular permeability are central to tissue and organ performance and are regulated by the microvasculature. Most of the gas, solute and hormone exchange between blood and tissue occurs in the microvasculature, which also has the highest regenerative capacity and is consequently the most active vascular compartment during angiogenesis. Leukocyte extravasation during inflammation also occurs predominantly in postcapillary venules.

Endothelial cells. Endothelial cells form the inner lining of all blood vessels and are in contact with each other through intercellular tight junctions and adherens junc-tions. Endothelial junctions control various key aspects of vascular homeostasis. For example, the regulated open-ing and closing of endothelial junctions occurs during leukocyte extravasation to inflamed tissue and also con-trols vascular permeability and affects communication

Department of Medical Cell Biology, Uppsala

University, Husargatan 3, Uppsala, 75123, Sweden.Correspondence to M.P . mia.phillipson@mcb.uu.se doi:10.1038/nrd.2015.2

Published online 27 Nov 2015

Leukocytes

White blood cells of the immune system. Different classes of leukocytes may fight infections or modulate

processes such as angiogenesis, inflammation and fibrosis.

Mural cells

The vascular smooth muscle cells and pericytes that surround and support endothelial cells.

T unnelling nanotubes

Very thin cytoplasmic

extensions containing actin that connect various cell types to allow direct exchange of intracellular components.

Intravital microscopy

A microscopy technique that enables direct observations of biological processes in vivo .

T argeting vascular and leukocyte communication in angiogenesis, inflammation and fibrosis

Johan Kreuger and Mia Phillipson

Abstract | Regulation of vascular permeability, recruitment of leukocytes from blood to tissue and angiogenesis are all processes that occur at the level of the microvasculature during both physiological and pathological conditions. The interplay between microvascular cells and leukocytes during inflammation, together with the emerging roles of leukocytes in the modulation of the angiogenic process, make leukocyte–vascular interactions prime targets for therapeutics to potentially treat a wide range of diseases, including pathological and dysfunctional vessel growth, chronic inflammation and fibrosis. In this Review, we discuss how the different cell types that are present in and around microvessels interact, cooperate and instruct each other, and in this context we highlight drug targets as well as emerging druggable processes that can be exploited to restore tissue homeostasis.

NATURE REVIEWS | DRUG DISCOVERY

VOLUME 15 | FEBRUARY 2016 | 125

|Drug Discovery

Extravasation

Emigration of immune cells from the vasculature into the surrounding tissue. Immunoliposomes Antibody-conjugated spheres of phospholipids that can be used to deliver drugs to tissues.between endothelial cells during angiogenesis12,13.

Increased vascular permeability is manifested during

many malignant and nonmalignant diseases. Oedema is

a cardinal sign of inflammation and is the consequence of

locally released inflammatory agents (such as bradykinin,

platelet-activating factor and histamine) that affect

adjacent blood vessels to increase junctional leakage.

The pathology of solid tumours as well as chronic

inflammatory conditions, age-related macular degenera-

tion, psoriasis, hereditary haemorrhagic telangiectasia and

the formation of atherosclerotic plaques all involve the for-

mation of microvascular networks with abnormal structure

and function. In these pathologies, excessive expres-

sion of angiogenic factors such as vascular endothelial

growth factor A (VEGFA) or of cytokines (such as brady-

kinin) and histamine destabilize vascular–endothelial

cell cadherin (VE-cadherin)-mediated endothelial cell–

cell interactions13. The resulting pathologically increased

vascular permeability leads to disease progression with

local oedema, elevated interstitial pressure and impaired

blood flow. Means to normalize the functions of the

microvasculature can both curb the development of dis-

ease and enhance the delivery of blood-borne pharma-

ceuticals to the target tissue, owing to reduced vascular

leakage with concomitantly reduced interstitial pressure

and increased blood flow14,15.

Endothelial cell heterogeneity. Endothelial cells are

equipped with sensors that measure the level of shear

stress generated by blood flow16,17. Shear stress induces

signals that lead to the adaptation of individual endothe-

lial cells and the development of situation-dependent

functional properties. Endothelial cells in high-flow

arterioles and arteries are elongated and narrow, whereas

endothelial cells in low-flow veins and venules have

cobblestone-like morphologies.Notably, the endothe-

lial cells of arteries, capillaries and veins have differential

gene expression and molecular profiles18 before the onset

of circulatory flow19, indicating the importance of local

microenvironmental factors to endothelial cell identity.

Some key regulators of arterial–venous specification

have been identified, including sonic hedgehog. Sonic

hedgehog induces the expression of VEGFA20, which, via

vascular endothelial growth factor receptor 2 (VEGFR2)

and the phospholipase Cγ (PLCγ)–RAF–extracellular

signal-regulated kinase (ERK) cascade, increases Notch

and ephrin B2 expression19,21. These molecules, together

with reduced ephrin type-B receptor 4, enhance the

arterial identity.

Delineation of the endothelial heterogeneity during

health and disease is important to the development of

strategies for the targeted delivery of pharmaceutical

compounds to organ-specific vascular beds, as this could

limit adverse effects. For example, leukocytes use dif-

ferent adhesion molecules to interact with the vascular

beds of different organs, which could be utilized to direct

drugs to a specific sites22. In a model of glomerulon-

ephritis, expression of the adhesion molecule E-selectin

by activated glomerular endothelium was used to direct

E-selectin-targeting immunoliposomes containing dexa-

methasone predominantly to the glomerular endothe-

lium23. As a result, pro-inflammatory gene expression

was strongly reduced in the inflamed glomeruli and

adverse effects normally associated with dexamethasone

were absent23,24.

The transcription factor SOX17 was recently shown

to be selectively expressed in arteries and suggested to

act upstream of Notch, but downstream of WNT25.

Canonical WNT signalling, which acts through

β-catenin, is activated by the binding of WNTs to Frizzled

receptor complexes, but also by cell–cell interactions

between cadherins. Abnormally high levels of WNT sig-

nalling have been observed both in cancer26 and during

wound healing and lead to the formation of abnormal

and dysfunctional vessels. Drugs that target WNTs or

β-catenin could inhibit the formation of pathological

blood vessels. WNTs are also involved in the commu-

nication between endothelium and leukocytes: WNTs

produced by the endothelium induce matrix metallopro-

teinase (MMP) expression in T lymphocytes that carry

Frizzled receptors, which accelerates extravasation27.

Both neutrophils and macrophages carry Frizzled recep-

tors and respond to WNT5a28,29. Thus, it is important to

keep in mind that targeting WNT and β-catenin may not

only affect communication within the endothelial cell

population, but will also probably affect the communi-

cation between the endothelium and pro-angiogenic

leukocytes.

the microvasculature. Mural cells and perivascular cells

that can influence the functions and growth of the

microvasculature are shown. The microvasculature is the

most active vascular compartment during angiogenesis.

Recent findings suggest that many cells have the capacity

to affect angiogenesis. Leukocytes, including neutrophils,

macrophages and mast cells, are potent regulators of

angiogenesis and can release growth factors that modulate

angiogenesis. In addition, it was recently suggested that

macrophages mediate the fusion of angiogenic sprouts.

Pericytes have been shown to mediate blood vessel

maturation and leukocyte trafficking. Furthermore, smooth

muscle cells, pericytes and mesenchymal stem cells (MSCs)

can give rise to fibroblasts and myofibroblasts that are

important regulators of angiogenesis, wound healing and

fibrosis. Nerves are often located close to vessels and can

modulate vascular tone as well as vessel formation.

Alternatively activated macrophages

These macrophages are less efficient at killing bacteria and are activated by different cytokines than are the classic M1 inflammatory macrophages. Alternatively activated macrophages are involved in tissue remodelling, wound healing, angiogenesis and tumour progression. Inside-out signalling Leukocyte integrin inside-out signalling occurs following chemokine receptor ligation and activates the ligand-binding function

of the integrins.

Outside-in signalling Integrin outside-in signalling occurs following the binding of leukocyte integrins to their counter-receptors on the endothelium and results in cytoskeletal rearrangements, strengthening of adhesion, cell spreading and migration.Mural cells. The traditional mural cells are vascular

smooth muscle cells and pericytes. Mural cells sur-

round and stabilize the endothelial cell tube and also

regulate angiogenesis, vascular function and leukocyte

recruitment. The identities, phenotypes and numbers of

perivascular cells differ between organs and vessel types

and depend on microenvironmental conditions30–34.

Smooth muscle cells form several layers around the

endothelial cells in the larger arteries and veins, whereas

pericytes partially cover the endothelial cell tube of

arteri o les, capillaries and postcapillary venules. Mural

cells may contract in response to stimuli to influence

organ perfusion and downstream blood flow. Pericytes

share a basement membrane with endothelial cells, inter-

act directly with the endothelium via cell surface recep-

tors and secrete signalling molecules to support vessel

maturation and function30,31,35,36.

Perivascular leukocytes. Many types of leukocytes are

found in perivascular positions. Tissue-resident mast cells

are often adjacent to blood vessels in skin and mucosa. In

response to inflammatory stimuli, the mast cells quickly

release permeability-increasing agents such as hista-

mine and VEGFA, which have potent paracrine effects

on endothelial cells37. Resident macrophages, located

perivascularly, sometimes have a pericyte-like appear-

ance. In addition to functioning as sentinel cells crucial for

host defence38, certain macrophages are pro-angiogenic

and modulate vascular branching morphogenesis during

embryonic development39.

Multiple macrophage subclasses are being character-

ized, and they differ in surface receptors, cytokine pro-

duction and phenotype. These macrophage subclasses

are believed to result from either different precursor cells

or from microenvironmental cues that lead to the activa-

tion of distinct macrophage signalling pathways, includ-

ing transcription factors, post-transcriptional regulators

and epigenetic regulators40. Classically activated macro-

phages (designated M1) are activated by interferon-γ

(IFNγ), express high levels of pro-inflammatory

cytokines (interleukin-1β (IL-1β), IL-6, IL-12, IL-23

and tumour necrosis factor (TNF)) as well as reactive

oxygen species (ROS) and nitric oxide (NO); inducible

nitric oxide synthase (iNOS) is an M1 hallmark. M1

macrophages are important inducers and effectors of

the T helper 1 (T

H

1) lymphocyte response40, and M1

macrophages clearly contribute to the development of

inflammatory diseases.

Other roles have been described for alternatively

activated macrophages, such as M2 macrophages. These

macrophages are associated with tissue remodelling,

wound healing, angiogenesis and tumour progres-

sion, and they are less efficient than M1 macrophages

at killing bacteria41–44. Alternatively activated macro-

phages are induced by IL-4 and IL-13 and produce

anti-inflammatory cytokines (IL-10 and TGFβ)40. The

signals and mechanisms that induce organ-resident

maintenance macrophages to become cells that either

actively fuel or inhibit the inflammatory process are

currently being delineated and are discussed below.

It was recently demonstrated that resident macrophages

are transcriptionally diverse, with very limited overlap

between different populations, suggesting that there are

probably additional classes of macrophages whose precise

function remains to be defined45.

Microvascular communication

The behaviour and function of the different cell types in

the microvascular compartment are regulated by inter-

cellular communication either through direct cell–cell

contact or by the transfer of secreted molecules. There

are four basic modalities for cell–cell communication at

this site: paracrine signalling; mechanosignalling, which

is mediated by direct interactions between surface mol-

ecules on adjacent cells; direct signal transduction via

gap junctions (connexins) and TnTs; and communica-

tion via the release and uptake of cell-derived vesicles

such as exosomes. The relative functional importance

of these modalities in different situations remains to be

fully established.

Paracrine signalling is the most well-studied mode

of communication between neighbouring cells. Classic

examples of paracrine signalling in the microvasculature

include mast cells releasing permeability-inducing fac-

tors, which then affect endothelial cells during inflam-

mation37, and the release of platelet-derived growth

factor (PDGF) from endothelial cells to recruit pericytes

that express PDGF receptor-β (PDGFRβ)46. Paracrine

signalling is also very relevant to stem cell therapies,

including those for tissue regeneration and angiogenesis,

that are now being evaluated in an increasing number of

clinical trials47,48. Interestingly, a substantial part of the

beneficial effects of stem cell therapies may be due to

their release of paracrine factors. Undifferentiated stem

cells secrete a range of factors that instruct nearby cells

in the microvascular compartment to participate in

angiogenesis. In addition, stem cells also secrete factors

that are immunosuppressive, anti-apoptotic or that may

stimulate proliferation49. Characterizing the secretome

of stem cells and circulating progenitor and/or myeloid

populations could identify alternatives to stem cell

therapy, as a locally administered cocktail of stem-cell-

secreted factors could be used to promote angiogenesis,

thereby circumventing some of the problems with sys-

temic injections of stem cells, such as nonspecific stem

cell homing and insufficient engraftment in the tissues

of interest50,51.

The recruitment of circulating leukocytes is directed

by mechanosignalling — between adhesion mol-

ecules on the endothelium and integrins on leuko-

cytes52. Integrin inside-out signalling occurs following

chemokine receptor ligation on leukocytes and results

in conformational changes in the extracellular integrin

domains, which then induces clustering of integrins

and increases the avidity of integrins for their ligands.

Following the binding of integrin to its counter-receptor

on the endothelium, integrin outside-in signalling occurs,

resulting in cytoskeletal rearrangements, strong cellu-

lar adhesion, cell spreading and leukocyte migration.

Distinct adhesion molecules are involved in the different

steps of the recruitment cascade, ultimately resulting in

leukocyte extravasation; this process is further discussed

Epithelial–mesenchymal transition

A process in which normally stationary epithelial cells lose cell–cell adhesions and cell polarity, thus becoming migratory.below. Notably, distinct leukocyte populations and

vascular beds express different combinations of adhe-

sion molecules that can enable tissue-specific therapeutic

targeting23,24.

Connexins are membrane proteins that form gap

junctions and enable direct signal transduction between

adjacent cells by allowing rapid transfer of ions, second

messengers and metabolites. Gap junction channels that

contain connexin 43 (CX43, also known as gap junc-

tion α-1) and CX45 (also known as gap junction γ-1)

and are between endothelial cells and mesenchymal cells

are important for mesenchymal cells to obtain a mural

cell fate during blood vessel formation53,54. CX43 chan-

nels are also involved in communication between leuko-

cytes and epithelial cells in both the lung and intestine,

resulting in immune suppression in the lung55 and oral

tolerance in the intestine56. In addition, in vitro studies

have suggested that connexin-dependent communica-

tion occurs between leukocytes and endothelial cells57,58.

Interestingly, the synthetic peptide ACT1, which mimics

the carboxy terminus of CX43, stabilized gap junctions

and decreased infiltration of inflammatory neutrophils

into wounds59. Application of Granexin gel containing

ACT1 also improved healing of diabetic foot ulcers in

clinical trials (Clinical Trials Registry India identifier:

CTRI/2011/09/001984)60. Targeting connexins to modu-

late the communication between epithelial or endothe-

lial cells and leukocytes could be an exciting approach to

improve wound healing60,61.

Cell–cell signalling also occurs via cell-derived

microvesicles, also called exosomes, that shuttle pro-

teins, lipids, microRNAs and mRNAs between cells62,63.

Exosomes are released from most cell types, including

leukocytes and endothelial cells, in response to various

stimuli. Platelets also constitute a rich source of exosomes

owing to their abundance in blood (≈ 1011 platelets per l).

Platelets are crucial for haemostasis and thrombosis but

are also involved in the regulation of both inflamma-

tion and vascular homeostasis64,65. Depending on the

stimulus, platelets can release exosomes containing

specific molecules; for example, exosomes containing

pro-angiogenic VEGFA can be released in response to

ADP or exosomes containing endostatin can be released

in response to thromboxane A

2

(REF. 66). In the United

States, many clinical trials investigating exosomes have

been initiated, and in one of these, exosomes will be

tested as an alternative vehicle to liposomes for drug

delivery (https://www.sodocs.net/doc/2e14426528.html, identifier: NCT01294072).

As drug delivery vehicles, exosomes could, in prin-

ciple, be injected into the target tissue; however, in

most studies conducted thus far the exosomes have

been administered systemically. An exciting possibility

is that exosomes could be designed to express surface

molecules that target them to specific vascular beds67.

For systemic delivery, an important question to address

is whether exosomes generally home to the activated

endothelium of inflamed tissues. Interestingly, exosomes

from rat pancreatic adenocarcinoma cells (the ASML cell

line) bind to and are taken up by all leukocyte subpopu-

lations in vivo to support leukocyte effector functions,

and it was suggested that exosomes could function as

adjuvants in immunotherapy68. Somewhat contradictory

to these findings, exosome removal has also been sug-

gested as a therapeutic adjuvant in cancer69. Clearly,

depending on the exosome content, removing or add-

ing exosomes will have different effects. Ongoing clinical

trials are also testing whether circulating exosomes can

be used as prognostic biomarkers, for example in gastric

cancer (the EXO–PPP Study; NCT01779583).

Recent findings link exocytosis mediated by exocyst

complex component 7 (EXO70) with both epithelial–mes-

enchymal transition (EMT)70 and the dynamic cell-shape

changes required for cell migration and morphogenesis71,

suggesting that selectively expressed exocyst compo-

nents, or isoforms thereof, could be targeted to modulate

EMT, cell migration and fibrosis. The exocyst complex

has been suggested to be involved in the formation of

exosomes72 and TnTs9. TnTs are F-actin-containing struc-

tures up to 200 nm in diameter that can be several cell

diameters long (over 70 μm in length)73, which, together

with gap junctions, are the only described way for cells

to directly exchange intracellular components. TnTs seem

to allow transport of larger cargo than gap junctions do.

Mitochondrial RHO GTPase 1 (MIRO1)-expressing mes-

enchymal stem cells (as well as smooth muscle cells and

fibroblasts) have very recently been shown to assist in

the repair of damaged epithelial cells by donating mito-

chondria through TnTs, suggesting that overexpression

of MIRO1 to induce TnT formation could be exploited

in stem cell therapy74. The actin-bundling protein fascin

has been identified as one of the protein components of

TnTs73, and there is interest in developing fascin-targeting

drugs, such as migrastatin analogues, that could impede

tumour growth by blocking cell–cell communication

between cancer cells and stromal cells (mesenchymal stem

cells (MSCs) or endothelial cells)75, as such interactions

may increase chemoresistance in cancer cells76. TnTs are

also utilized to transfer signalling proteins such as HRAS

from B cells to T cells77, and dying endothelial cells can

be rescued by TnT-mediated transfer of substances from

adjacent endothelial progenitor cells78 or mesenchymal

cells79. The roles of TnT s in leukocyte–endothelial exchange

have yet to be explored.

Angiogenesis

Several antibody-based drugs that inhibit angiogenesis

have reached the market for treating cancer or abnor-

mal blood vessel formation in wet age-related macular

degeneration (AMD)80. Despite its benefits in some forms

of cancer and in AMD, anti-angiogenic treatments have

also been associated with severe adverse effects, and this

has led to an increased interest in therapies aimed at

normalizing pathological vessels rather than eliminating

them81. Therapeutic means to induce the formation of

functional blood vessels are also helpful for the treatment

of local ischaemia during peripheral ischaemic disease,

stroke or engraftment following transplantation. A better

understanding of the different angiogenic mechanisms,

as well as the roles of microvascular cell–leukocyte com-

munication in regulating angiogenesis, is required to

develop new therapeutic agents with higher specificity

and efficacy.

| Drug Discovery

Mechanisms of angiogenesis. The main paradigm guid-ing the field of anti-angiogenic therapies has been that angiogenesis occurs predominantly via sprouting angio-genesis together with some degree of intussusceptive angiogenesis (vessel splitting)82,83 (FIG. 2). Notably, our understanding of sprouting angiogenesis is based mainly on data obtained from mouse and zebrafish models of developmental angiogenesis as well as on data from advanced 3D in vitro models 84,85. During developmental angiogenesis, reproducible patterning of the vasculature occurs in response to spatiotemporally regulated guid-ance cues (such as morphogen gradients) that induce expression of target genes to regulate the formation of angiogenic sprouts headed by a tip cell 84. Multiple extra-cellular concentration gradients of guidance cues help the vascular cells to know their position and guide the migrating tip cells 86. Comparatively less is known about the mechanisms of angiogenesis in adult tissues during wound healing and inflammation. In contrast to the

organized angiogenic process during development, a wound may appear in any location at any time. Acute and generic repair mechanisms are therefore crucial for vessel formation in association with wound healing.A rapid angiogenic response, termed looping angio-genesis, which is compatible with the special require-ments of wound healing, was recently proposed 87–89 (FIG. 2). During wound closure, looping angiogenesis occurs when contractile myofibroblasts pull adjacent intact tissue — including fully perfused vessels — into the wounded area 89, leading to vessel elongation and the formation of extended vascular loops. Interestingly, tumours have previously been described as wounds that do not heal 2, and data obtained by time-lapse intravital multiphoton microscopy indicate that physical forces in rapidly growing tumours expand the vasculature in a process that could be considered as looping angiogen-esis, which occurs together with sprouting and splitting angiogenesis (see Supplementary information S1 (video)

is thought to mainly occur through sprouting angiogenesis (top right). In this process, new vessel sprouts headed by tip

cells respond to highly regulated guidance cues often present in the form of concentration gradients of chemotactic and

pro-angiogenic factors. The conditions for angiogenesis during wound healing are very different from those during development. During wound healing, biomechanical forces elicited by myofibroblasts during the wound closure process induce looping and splitting angiogenesis, processes of vessel neoformation that are faster that sprouting (bottom left). Modulation and maturation of the primitive vascular network occurs through a combination of mechanisms, including sprout fusion, vessel splitting and pruning, as the tissue expands during development (bottom right). All mechanisms of vessel formation probably occur simultaneously, albeit to different extents in different situations. Whereas sprouting dominates early embryonic angiogenesis, looping and splitting may have more prominent roles during initial wound healing. Notably, the end result — irrespective of the degree and types of angiogenic mechanisms at play — would be the same, such that the final vessel density is ultimately regulated by the functional demands set by oxygen tension.

in REF. 90). The biomechanical hypothesis that explains the driving forces for looping angiogenesis is probably of general relevance to the development of both anti- and pro-angiogenic therapies, although the mechanism was first described in relation to wound healing87. The forma-tion of contractile myofibroblasts that generate the forces that lead to vessel elongation and looping is induced by TGFβ and can be inhibited by TGFβ-specific antibodies such as fresolimumab and 1D11 (REF. 91), small-molecule receptor serine/threonine kinase inhibitors such as LY2157299 (REF. 92) and blocking peptides such as P144 (a 14-mer peptide derived from the TGFβ1 proteoglycan receptor betaglycan (also known as TGFR3))93. Modulation of extracellular matrix production will also affect all modes of angiogenesis. Inhibition of col-lagen production can be achieved using small interfering RNA (siRNA)-mediated approaches, whereas inhibitors against cathepsin K, MMPs and other enzymes with collagen-degrading activity can increase levels of extra-cellular collagen. The precise contributions of different leukocyte populations to looping angiogenesis remains to be established, but the abilities of leukocytes to affect the extracellular matrix and to produce TGFβ suggest that they could have a role94,95.

Modulation of angiogenesis by leukocytes. In addition to accumulating in inflamed tissue, leukocytes are recruited in large numbers to sites of hypoxia and sterile injury. Tissue-resident macrophages and recruited neu-trophils and eosinophils clearly contribute to angiogen-esis as they alter the microenvironment by secreting factors such as VEGFA and MMPs that promote angio-genesis43,96. Interestingly, a direct association between perivascular macrophages and angiogenic blood vessels has also been reported during developmental angiogen-esis in the mouse hindbrain and in zebrafish, in which the macrophages were found to directly partake in the fusion of endothelial sprouts39.

Monocyte recruitment to damaged and hypoxic tissue involves chemokine receptors (including CXC-chemokine receptor 4 (CXCR4) and CC-chemokine receptor 2 (CCR2)) as well as sphingosine 1 phosphate receptor 3 (S1PR3), all of which are upregulated during ischaemia97–99. CXC chemokine ligand 12 (CXCL12), the ligand of CXCR4 (REF. 100), is upregulated around larger vessels in a VEGFA-dependent manner101, whereas sphingosine 1-phosphate, produced by platelets and endothelial cells, is blood-borne and crucial for vessel maturation31. At the affected site, the monocytes and/or macrophages acquire enhanced pro-angiogenic proper-ties following ‘on-site’ education by microenvironmen-tal cues102. Circulating pro-angiogenic neutrophils are also recruited to hypoxic tissue, where they are crucial for the initiation of angiogenesis because they deliver the pro-angiogenic enzyme MMP9 (REFS 4,43,103,104). MMP9 may promote angiogenesis both by degrading the extracellular matrix and by increasing the release of matrix-bound VEGFA. Thus, uncovering the underlying mechanisms regulating the recruitment and on-site edu-cation of leukocytes will probably provide new druggable targets to influence angiogenesis.Inhibiting angiogenesis. Anti-angiogenic therapy to combat cancer progression has been discussed since the discovery, in the 1970s, that inhibition of vascular growth can slow down the growth of solid tumours1. In 1991, Kerbel suggested that the genetically stable endothelial cells in tumours could be targeted to halt tumour growth105. To date, the best results for anti-angiogenic (that is, VEGF-specific) treatments are observed when treating macular degeneration, includ-ing proliferative diabetic retinopathy106. However, using a VEGF-specific treatment (TABLE 1), such as the humanized antibody bevacizumab (Avastin; Genentech/ Roche), to combat cancer through inhibition of tumour angiogenesis has yielded modest results relative to expectations. The addition of VEGF-specific therapy to radiotherapy and systemic chemotherapy has, in some studies, been bene f icial107–110, in agreement with the vascular normalization hypothesis81. This hypothesis states that normalization of pathological and disorgan-ized tumour vessels leads to decreased vascular leakage and normalization of interstitial pressure, and at least transiently improves drug uptake81,111. However, many adverse effects associ a ted with anti-angiogenic drugs have been reported, such as bleeding, protein u rea, hypertension, reduced wound healing and increased risk for thromboembolic events112,113.

It seems that the complexity of the angiogenic pro-cess was underestimated, at least initially, and a num-ber of compensatory mechanisms come into play when VEGF is inhibited. The modes of resistance to VEGFA pathway inhibitors have been discussed in detail in an expert review114, and proposed escape mechanisms typi-cally include a compensatory increase of the production of other pro-angiogenic compounds. It seems plausible that in order for anti-angiogenic therapy to be more effi-cient in the context of inhibiting tumour growth, multi-ple molecules expressed by endothelial cells or cells of the microvascular compartment (including angiomodu-latory leukocytes) should be simultaneously targeted by combination therapies. Examples include combinations of antibodies targeting VEGFR2 or VEGFA together with CXCR4 antagonists, which inhibit angiogenesis and reduce tumour growth in experimental mouse models115,116. Furthermore, a combination therapy using antibodies against both VEGFA and the VEGFR2 co-receptor neuropilin 1 (NRP1) reduced vessel for-mation and tumour growth in xenograft models, and blockade of NRP1 also resulted in reduced pericyte cov-erage117. Pericytes are crucial for blood vessel formation and function and are therefore prospective targets for anti-angiogenic therapies. As pericytes depend on PDGF signalling, many small multi-targeting kinase inhibitors of PDGF signalling have been investigated for their effects on pericytes and angiogenesis (BOX 1;TABLE 2).

Metabolic targets to inhibit angiogenesis. Several recent papers suggest that targeting endothelial cell metabolism could limit angiogenesis, as the rate of glycolysis can reg-ulate vessel branching118–120. Recently, it was shown that 6-phosphofructo-2-kinase/fructose-2,6-biphosphatase 3 (PFKFB3), an activator of glycolysis, promotes vascular

growth and that the PFKFB3 inhibitor 3-PO inhibits endothelial cell proliferation in mouse and zebrafish models121. Interestingly, 3-PO counteracts hyperbranch-ing induced by inhibition of Notch and augments the effects of VEGF-specific treatment. The effects were indeed partial and transient, but 3-PO could be a new type of anti-angiogenic compound, and the related molecule ACT-PFK-158 is now in a Phase I safety study of patients with cancer (NCT02044861).

Recently, the oxygen sensor prolyl hydroxylase 2 (PHD2), which promotes degradation of hypoxia-inducible factor 1α (HIF1α), has received a lot of attention. Haplodeficiency of the gene encoding PHD2 (PHD2, also known as EGLN1) resulted in normalization of tumour vasculature, which increased tumour blood per-fusion, decreased tumour interstitial pressure (caused by decreased vessel leakage) and inhibited metastasis122,123. In a model of muscle ischaemia, mice haplodeficient for Phd2 were less susceptible to ischaemia-induced tissue necrosis124. Interestingly, increased numbers of alterna-tively activated macrophages secreting pro-arteriogenic factors were found in the ischaemic muscle of Phd2+/?mice, and either chronic or acute deletion of Phd2 in macrophages skewed them to a pro-angiogenic pheno-type124. Furthermore, in situ education of pro-arteriogenic macrophages after vascular occlusion has been shown to rely on angiopoietin 1 (ANG1)–ANG1 receptor (TIE2, also known as TEK) signalling, and ANG1 was shown to downregulate PHD2 expression via activation of TIE2 in macrophages125. Thus, ANG1, which is stored in platelets and mural cells and released during tissue damage, can assist in macrophage reprogramming. Clearly, the homeo-static roles of the HIF pathway and factors regulating HIFs, such as PHD2, are of great interest as therapeutic targets, and screens to identify inhibitors and activators are ongoing126.

Table 1 |

A selection of anti-angiogenic therapies targeting VEGF–VEGFR signalling

50

FDA, US Food and Drug Administration; mTOR, mammalian target of rapamycin; PDGFR, platelet-derived growth factor receptor; PlGF, placenta growth factor; VEGF, vascular endothelial growth factor; VEGFR, VEGF receptor.

Targeting Notch signalling to inhibit angiogenesis. Notch signalling is central to many aspects of vascular develop-ment, including differentiation of endothelial and mural cells as well as regulation of sprouting. Tip cells that are positive for the Notch ligand delta-like protein 4 (DLL4) hold adjacent Notch 1-positive stalk cells in check by lateral inhibition127,128. Interestingly, inhibition of Notch signalling has recently been proposed to increase infiltra-tion of mononuclear cells into tissue, which is followed by intussusceptive angiogenesis129, and Notch signalling in macrophages is required for efficient interactions with DLL4-positive tip cells to mediate sprout fusion130. Whereas DLL4 in endothelial tip cells inhibits sprouting from adjacent stalk cells, DLL4 stimulates VEGFA production in pro-angiogenic macrophages. Pharmacological inhibition or activation of DLL4 may have very different net effects on inflammatory neovas-cularization and tumour vascularization, for example, depending on the balance of Notch signalling in endothelial cells and macrophages, both of which express Notch 1 and Notch 4 (REF. 131). Furthermore, blockade of the DLL4–Notch interaction leads to increased homing of leukocytes expressing CD45 (also known as receptor-type tyrosine-protein phosphatase C) and αM integrin (also known as CD11b) to ischaemic tissue, and IL-8 release by activated monocytes is strongly inhibited by endothelial cells through a DLL4–Notch-mediated mechanism132. Thus, it is conceivable that if the number of pro-angiogenic macrophages and other DLL4-responsive leukocytes is large, the effect of DLL4 on leukocytes will partially balance the inhibitory effect of DLL4 on endothelial cell activation. Notably, the regu-lation of the functional phenotypes of macrophages has also been reported to be partially downstream of Notch receptors133,134, and it is thus currently difficult to predict the effects of Notch pathway inhibitors on angiogenic events that are modified by the immune system.

Targeting macrophages to inhibit angiogenesis. Myeloid-derived immune cells, predominantly macrophages, supply tumours with VEGFA, which increases the num-ber of tumour vessels of an immature and leaky pheno-type135. Specific deletion of VEGFA in myeloid-derived cells resulted in large tumours with normalized vascula-ture, which were sensitive to chemotherapy135. Numerous clinical trials are evaluating anti-macrophage therapies, including antibodies and small-molecule inhibitors such as colony stimulating factor 1 (CSF1)-specific therapies, which reduce macrophage differentiation and survival, or CC-chemokine ligand 2 (CCL2)-specific therapies, which inhibit monocyte recruitment in patients with cancer136,137. However, a recent study revealed a hitherto unknown complexity of chemokine biology, as the with-drawal of CCL2-specific therapy in mice with breast

cancer tumours accelerated lung metastasis and death 138. The reason for this unexpected twist might be a feedback mechanism that results in increased CCL2 production upon the release of CCL2 inhibition, which then recruits monocytes to tumours. In addition, CCL2-specific therapy blocks monocyte recruitment from bone mar-row, resulting in an increased pool of CCR2-positive monocytes that, following cessation of CCL2-specific treatment, are released into the blood and recruited to metastatic tumours where they contribute to angiogenesis by releasing VEGFA.

Efforts are currently being made to develop treat-ments that shift tumour-associated and pro-angiogenic alternatively activated macrophages to an antitumour M1 subtype 40. The opposite shift (from M1 to alterna-tively activated macrophages) would be useful in situa-tions, such as peripheral ischaemia, in which increased angiogenesis would be beneficial. Several promising candidates for these opposing scenarios are listed in TABLE 3, some of which are under investigation and some of which are already on the market. As with all immuno-modulating therapies, systemic alterations of the immune system must be considered with great caution.

Macrophages have anti-angiogenic functions during retinal development that depend on non-canonical WNT signalling and the concomitant production of soluble VEGF receptor 1 (sVEGFR1, also known as sFLT1), which competitively inhibits VEGF signalling 139. This was recently demonstrated to be important during early phases of wound healing 140. Abnormally high levels of WNT signalling are observed during both wound healing and cancer, and the drugs that (at least in part) target WNT signalling and are currently in clinical trials for the treatment of malignancies and/or vascular con-ditions include the small molecular weight porcupine inhibitor LGK974 (for WNT-dependent malignancies; NCT01351103), as well as resveratrol (for coronary artery disease; NCT02137421), sulindac and curcumin (for acute kidney injury or abdominal aortic aneurysm; NCT01225094).

Inflammation

A regulated inflammatory response is vital for life after birth, and dysregulated inflammation is involved in the progression of many diseases. Similar cues drive inflam-mation irrespective of whether the underlying trigger is a bacterial infection or sterile tissue damage. Blunt and nonspecific anti-inflammatory therapies are associ-ated with serious adverse effects, such as increased risk for opportunistic infections and cancer 141. The fine-tuning mechanism that regulates the amplitude and prolong a tion of the local inflammatory process should be deciphered, to prevent chronic inflammation while maintaining a sufficient host defence.

Mechanisms of inflammation. The cellular ‘whistle-blowers’ that initiate inflammation following damage or bacterial detection are primarily the resident leukocyte populations (such as macrophages) that are found in most organs. Classic tasks of these sentinel cells are to maintain organ and system homeostasis by cleaning up after dying

Table 2 | Examples of PDGFR β

inhibitors exhibiting variable target selectivity

i d indicated. cMET , hepatocyte growth factor receptor; CSF1R, macrophage colony-stimulating factor 1 receptor; EGFR, epidermal growth factor receptor; FGFR, fibroblast growth factor receptor; FLT3, Fms-like tyrosine kinase 3; IC 50, half-maximal inhibitory concen t ration; PDGFR, platelet-derived growth factor receptor; RON, macrophage-stimulating protein receptor; VEGFR, vascular endothelial growth factor receptor.

cells and defeating potential threats. The macrophages have unique means to monitor their environment, and in response to certain alterations and challenges they release multiple paracrine factors to instruct cells in the micro-environment. In response to macrophage-produced cytokines and damage- or pathogen-associated molecu-lar pattern molecules (DAMPs or PAMPs) released by injured cells or the invaders, the postcapillary venular endothelium upregulates adhesion molecules involved in the leukocyte recruitment cascade. This cascade involves mechanosignalling between leukocytes in the circulation and endothelial cells, and it ultimately results in extrava-sation of leukocytes, of which neutrophils are the first to arrive to the affected site (FIG. 3) (reviewed in REFS 7, 142–144). Some leukocyte adhesion molecules — namely the β2 integrins: αLβ2 integrin (also known as lympho-cyte function-associated antigen 1 (LFA1)) and αMβ2 integrin (also known as macrophage-1 antigen (MAC1)) — switch from low-affinity to high-affinity states fol-lowing the binding of chemokines to G protein coupled receptors on leukocytes. As a consequence, the neutro-phils become activated and start to adhere to144,145 and crawl on the vessel wall along chemokine gradients146,147. The intravascular crawling continues until the neutro-phil reaches one of the optimal sites for transmigration, which are most often located at endothelial junctions147,148 and coincide with areas of low expression of laminin 10, collagen IV and nidogen 2 in the basement membrane between pericytes149,150.

Neutrophil extravasation requires bidirectional cross-talk between the neutrophil and endothelial cells and may involve the formation of endothelial protrusions that engulf the transmigrating neutrophils in dome-like structures151–153. This process involves ligation of numerous endothelial adhesion molecules (intracellular adhesion protein 1 (ICAM1), ICAM2 and vascular cell adhe-sion protein 1 (VCAM1)) and junctional adhesion molecules (platelet endothelial cell adhesion molecule (PECAM1), CD99 and junctional adhesion molecules (JAMs)) and ultimately results in the regulated opening of endothelial junctions, which probably limits vascular leakage143. Data suggest that the endothelium functions as a barrier regulating transendothelial migration, which has spurred new interest in the role of endothelial cells in leukocyte transmigration153–155. This gating could be regulated by targeting the junctional and cytoskeletal rearrangement of the endothelium, which would limit detrimental inflammation by reducing leukocyte trans-migration and/or vascular permeability154,155. In addition, intravascular neutrophils that adhere to the endothelium have the potential to modify the endothelial barrier.

A recent study demonstrated that neutrophils release TNF in close proximity to endothelial junctions in response to chemoattractants156 and locally increase microvascular permeability. This permeability increase was abolished in mice null for the TNF receptor (Tnfr?/?mice), even though leukocyte transmigration was not affected156. These observations highlight the bidirectional communication between leukocytes and endothelial cells during inflammation, an area in which further studies are warranted.

Pericytes support neutrophil transmigration, and outside the endothelial lining, abluminal crawling ulti-mately directs the neutrophils to gaps between adjacent pericytes157. These gaps become enlarged in response to pro-inflammatory chemokines, which also induce contractions and conformational changes in pericytes. Furthermore, extravasated neutrophils and monocytes chemotactically move towards and interact with arte-rial pericytes in a model of sterile skin inflammation8, a process that is dependent on the inflammation-induced pericyte presentation of macrophage migration inhibitory factor (MMIF, also known as MIF). These interactions result in increased expression of pro-migratory recep-tors as well as prosurvival molecules on the leukocytes, thereby increasing the efficiency with which they navi-gate to the site of injury. Furthermore, neutrophil trans-migration was recently demonstrated to occur at close proximity to perivascular macrophages in a skin infec-tion model using Staphylococcus aureus158. The virulence factor α-haemolysin (produced by the bacteria) lysed the perivascular macrophages, thus reducing neutrophil extravasation. These data demonstrate that the differ-ent cells of the microvascular compartment collaborate during leukocyte recruitment to inflamed tissue. T ogether, these cells — as well as their communication — comprise possible targets for anti-inflammatory therapies.

General mechanisms to inhibit inflammation. Anti-inflammatory drugs on the market span from general immune-suppressing therapies to highly specific mono-clonal antibodies targeting distinct adhesion molecules or downstream intracellular signalling proteins. The major problems with the general anti-inflammatory

colony-stimulating factor 1; CSF1R, CSF1 receptor; HRG, histidine-rich glycoprotein; IFNγ, interferon-γ; IL-33, interleukin 33; GLUT1, glucose transporter 1; PlGF, placenta growth factor; PPARγ, peroxisome proliferator-activated receptor γ; STAT, signal transducer and activator of transcription; TLR, Toll-like receptor.

| Drug Discovery

therapies are the adverse effects associated with global immunosuppression, which render the patient more prone to infections and also increase the risk of acquiring other pathological conditions associated with impaired function of the immune system, such as cancer. Clinically available approaches to reduce inflammation include inhibiting the cell-specific effects triggered by inflamma-tory molecules, such as nonsteroidal anti-inflammatory drugs (NSAIDs) that inhibit cyclooxygenase (COX), neutralizing antibodies that target chemokines and cytokines and inhibitory antibodies that target adhe-sion molecules 159. In recent years, the treatment of auto-immune diseases such as colitis and arthritis has been revolutionized by the introduction of TNF-specific thera p ies 159. Other means to reduce the induction of inflammation that are currently under clinical evaluation include the inhibition of pattern recognition receptors, such as toll-like receptor 2 (TLR2), which is targeted by

the monoclonal antibody OPN-305 (NCT01794663); and inhibition of the master regulator nuclear factor κB (NF-κB) by small molecule inhibitors of upstream acti-vators such as the IκB kinases (IKKs), JUN N-terminal kinases (JNKs), Janus kinases (JAKs)160 or mitogen-activated protein kinases (MAPKs)159. However, as long as the targeted pathways are also crucial for the efficient host response that is required for restoration of homeo-stasis, the general immune-suppressing effects will result in unpleasant adverse effects and thus reinforce the need to identify site-specific drug targets.

Targeting adhesion molecules. The adhesion molecules involved in leukocyte–endothelial cell interactions during leukocyte recruitment are obvious drug targets that have been explored over the past decade, and several on g oing clinical trials are currently examining this strategy 161 (TABLE 4). However, the broad cellular expression of many

Figure 3 | The different stages of the leukocyte recruitment cascade and the potential drug targets central to leukocyte recruitment. Leukocyte rolling, adhesion, crawling and transmigration involve distinct adhesion molecules on

leukocytes and endothelium and result in leukocytes being recruited from the circulation to inflamed tissues. Some key molecules that mediate leukocyte–endothelial cell interactions in each step of the recruitment cascade are listed. Drugs and compounds that bind to these targets could prevent interactions between leukocytes and the endothelium to impede or block leukocyte extravasation during inflammation. ESAM, endothelial cell-selective adhesion molecule; ESL1, E-selectin ligand 1; ICAM1, intercellular adhesion molecule 1; JAMA, junctional adhesion molecule A; MAdCAM1, mucosal addressin cell adhesion molecule 1; PECAM1, platelet endothelial cell adhesion molecule; PSGL1, P-selectin glycoprotein ligand 1; PTX3, pentraxin-related protein; VCAM1, vascular cell adhesion protein 1; VE-cadherin, vascular–endothelial cell cadherin; VEGFA, vascular endothelial growth factor A.

NATURE REVIEWS DRUG DISCOVERY

VOLUME 15 FEBRUARY 2016 135

of the targeted adhesion molecules prevents specific inhibition and hence these strategies are associated with severe adverse effects. For instance, the life threatening condition progressive multifocal leukoencephalo p athy (PML) was observed during inhibition of either α4 or αL integrin, which was due to reactivation of John Cunningham (JC) virus in the immunosuppressed patient161–164. However, certain endothelial adhesion molecules are expressed in an organ-specific manner, and strategies to block these adhesion molecules could provide localized solutions with fewer adverse effects. For example, in recent clinical trials (the Gemini stud-ies)165,166, treatment with humanized monoclonal anti-bodies targeting α4β7 integrin showed promising results in the management of both Crohn disease and ulcerative colitis. The α4β7 ligand mucosal addressin cell adhe-sion molecule 1 (MAdCAM1) is specifically expressed in blood vessels of the intestinal tract167–169. As a con-sequence, α4β7 inhibition results in local effects in the intestine and does not alter the phenotype of cerebro-spinal fluid T lymphocytes that are believed to underlie the development of PML in response to treatment with the α4 integrin-specific antibody natalizumab170–172. No case of PML has yet been reported during α4β7 inhibi-tion, further stressing the importance of developing drugs that only target selected organs.

Targeting chemokines and their receptors. Circulating leukocytes are directed to inflammatory sites by the release of chemokines that bind to extracellular or cell-surface associated glycosaminoglycans (GAGs), mainly through electrostatic interactions. The chemokines are released by cells in the affected tissue, including endothe-lial cells, which also present the chemokines to leuko-cytes to recruit them from the blood stream. Targeting the activity of these chemokines and/or their cognate G protein-coupled chemokine receptors on leukocytes are attractive approaches to limit leukocyte influx. Several ways to bind and inactivate chemokines in the circulation or in tissue are currently being evaluated experimentally and include the cytokine receptor domain fused to the crystallizable fragment (Fc) region of immunogluobu-lin G (IgG)173, nanobodies174 and GAG-mimetics175,176. Several therapies targeting chemokines or their recep-tors have been tested in clinical trials, including the small molecule CCX282-B, which targets CCR9 and has been investigated for the treatment of Crohn dis-ease; eldelumab, a CXCL10-specific antibody that has been investigated for the treatment of ulcerative colitis; and MLN1202, a CCR2-specific antibody that has been investigated for the treatment of cardiovascular disease177. Small molecular receptor antagonists that have made it to the market include the CCR5 antagonist maraviroc (Pfizer; for HIV infection178) and a CXCR4 antagonist AMD3100, which is a stem cell mobilizer179. Another possible approach to limit chemokine activity could be to increase inhibitory post-translational modi-fications. In this context, the enzyme protein-arginine deiminase (PAD) changes peptide-bound arginine into citrulline and leads to citrullination of chemokines such as CXCL8 (IL-8), CXCL10 and TNF, which reduces their activities180,181. Interestingly, neutrophils — as well as certain bacterial pathogens such as Porphyromonas gingivalis that also express PAD — might thus be able to inactivate chemokines to impede leukocyte infiltra-tion, and in the case of bacteria this could allow them to escape immune recognition181,182. Whether citrullina-tion can be induced pharmacologically remains to be established.

Accelerating resolution of inflammation. An attractive but relatively unexplored approach to limit inflamma-tion is to accelerate the inherent mechanisms underlying resolution of inflammation. In the best-case scenario, the inflammatory process is self-limiting and tissue homeo-stasis is completely restored. Inflammation is often resolved by expanding the population of alternatively acti-vated macrophages, which are crucial for the resolution phase183. The mechanisms underlying the re-education of macrophages and their subsequent differentiation into distinct phenotypes are now considered potential tar-gets for a wide range of therapies owing to the divergent effector functions of these cells (reviewed in REF. 40) and numerous therapies are being evaluated or are already in the clinic (TABLE 3). Efferocytosis, the removal of dying cells by phagocytosis, is a key event during resolution of inflammation that is executed by resident macrophages; during this process these cells acquire an alternatively activated phenotype184. Simultaneously, the macrophages switch their production of arachidonic acid-derived eicosanoids from prostaglandins and leukotrienes to anti-inflammatory lipoxins (leukotriene A

4

and leuko-triene B

4

)185. Lipoxins, together with the omega-3-derived resolvins, protectins and maresins, were recently con-firmed to be expressed in human tissues186 and are spe-cialized pro-resolving mediators that inhibit leukocyte recruitment and enhance macrophage efferocytosis (reviewed in REF. 187).

In animal models, addition of resolvin E1 suppresses the recruitment of leukocytes and the expression of inflammatory mediators (such as TNF and IL-1β188) during sterile inflammation, reduces herpes simplex virus-induced ocular inflammatory lesions189 and accel-erates healing of diabetic wounds190. In a double-blind clinical trial of infants with eczema, topical application of 15(R/S)-methyl-lipoxin 4 relieved symptoms and improved quality of life191. Additional clinical applica-tions are now being tested: mimetics of resolvin E1 are being examined for ocular indications (NCT01639846 and NCT01675570; also see the Auven Therapeutics website) and protectin D1 is being examined for neuro-degenerative disorders (see the Anida Pharma web-site). The existence of these specialized pro-resolving mediators in humans as well as their ability to enhance resolution of inflammation without concomitant immuno s uppression are indeed promising, even though their specific anti-inflammatory mechanisms of action and their functional importance in human health need to be better understood. Altering the microenvironmental cues provided by cells in microvessels and their close sur-roundings to re-educate resident and infiltrating leuko-cytes is an attractive approach to limit inflammation.

A recently developed zebrafish model of sterile injury might be helpful for screening for new compounds that limit chronic inflammation and stimulate resolution192.

Combatting the development of fibrosis Vasculitis, the inflammation of blood vessels, may lead to vessel damage, tissue destruction (necrosis) and even-tually fibrosis. The diagnosis of many microvascular pathologies is based on symptoms and relatively little is known regarding the precise underlying mechanisms and molecular aetiologies. The pathogenesis of vasculitis is considered to be predominantly autoimmune, but infec-tion by bacteria or viruses may also spark its progression. Fibrosis is generally the result of prolonged injury and deregulated wound healing (reviewed in REF. 193).

Mechanisms of fibrosis. Fibrosis is a protective mechanism that follows inflammation and tissue injury whereby large quantities of extracellular matrix (mainly type I colla g en) are secreted by cells that invade the wounded area in order to reconstitute and strengthen the damaged tissue and accelerate the healing process (FIG. 4). Dysregulated fibrosis that results from repeated tissue damage leads to continued activation of matrix-secreting fibroblasts and myofibroblasts, which eventually alters tissue function and can ultimately lead to organ dysfunction. Leukocytes have key roles as modulators of the fibrotic process and are thus targets for reducing uncontrolled fibrosis. Continued wounding or irritant-elicited responses can lead to over-active leukocytes that damage the affected tissue, whereas a fine-tuned and self-regulated leukocyte response that is gradually turned off promotes tissue repair.

The fibrotic process is activated by the destruction of cell membranes, which releases cell contents and DAMPs. The DAMPs activate mesenchymal cells and mural cells located in the proximity of the damaged site, followed by the recruitment of monocytes and macro-phages via chemokine-induced chemotaxis (for example through activation of the CCL2–CCR2 signalling axis). The recruited macrophages phagocytose dead cells and secrete additional chemokines to recruit other types of leukocytes, including neutrophils.

Inhibition of fibrosis. EMT occurs during fibrosis, and TGFβ and related signalling proteins are major media-tors of both EMT and fibrosis in many organs194. Activated macrophages release TGFβ and other pro-inflammatory mediators, which leads to the generation of fibroblasts from several cell sources, as well as to transdifferentia-tion of fibroblasts into myofibroblasts195,196. Interestingly, αV integrins on myofibroblasts are involved in the activa-tion of extracellular TGFβ, and pharmacological block-ade of αV-containing integrins using the small-molecule inhibitor CWHM 12 can reduce fibrosis in both the lung and liver197. Furthermore, the EMT transcriptional pro-gramme is anti-apoptotic and generally promotes cell migration, such that cells undergoing EMT often move away from their original position over time. Thus, reduc-ing EMT-dependent formation and migration of myo-fibroblasts (FIG. 4) could be of interest to target fibrosis and also to target angiogenesis198. The anti-fibrotic prop-erties of the tyrosine-kinase inhibitor nintedanib (Ofev; Boehringer Ingelheim; approved by for treatment of idio-pathic pulmonary fibrosis) were recently described199. Nintedanib was shown to have anti-angiogenic effects and to inhibit tumour growth in vivo, and it did not promote EMT, despite inducing hypoxia.

Attempts are ongoing to identify targets within pro-fibrotic signalling cascades such as the TGFβ path-way197,200. Of interest, signalling via TLR4 can down-regulate BMP and activin membrane-bound inhibitor homolog (BAMBI), the competitive and non-signalling pseudoreceptor for TGFβ201. Thus, it is possible that inhibition of TLRs or activation of BAMBI could be novel ways to modulate or counteract TGFβ-dependent fibrosis. Accordingly, the TLR4-specific monoclonal anti-body NI-0101 has entered clinical trials (NCT01808469) for chronic inflammation and fibrosis. Additionally,

Table 4 |

Therapies targeting adhesion molecules involved in leukocyte recruitment to tissue

|Drug Discovery

targeting of 5′-AMP-activated protein kinase (AMPK) by metformin leads to increased BAMBI expression, which could explain some of the reported anti-fibrotic effects of this small organic compound, which is normally used to manage type 2 diabetes202.

Furthermore, release of PDGF subunit B (PDGFB) from platelets and activated macrophages in the wounded area will also attract and activate myofibro-blasts, and indeed a PDGFB-specific monoclonal antibody reduces the development of liver fibrosis in experimental mouse models203.

Accumulating evidence supports the notion that pathological angiogenesis contributes to fibrosis. For example, blockade of VEGFR2 attenuates the steatosis and inflammation that can precede fibrosis in mice204. Interestingly, fibrotic scars, mainly composed of type I collagen, can be degraded by MMPs that are regulated by endogenous tissue inhibitors of metalloproteinases (TIMPs). Notably, neutrophils are the only cell type that release MMP9 without concomitant release of its specific inhibitor TIMP1, and this unique property enables them to deliver highly active MMP9 to sites of hypoxia205. Targeting MMPs is generally of great interest to control cell invasiveness, matrix remodelling and pathological angiogenesis. An interesting concept was recently pre-sented in which injectable and bioresponsive hydrogels were used to locally deliver TIMPs: in response to MMP activity, MMP-degradable hydrogels released TIMPs locally ‘on-demand’ to limit unwanted effects of the inhibitor in other tissues206. Monocytes may also medi-ate fibrosis-associated angiogenesis: inhibiting infiltra-tion of monocytes expressing increased levels of VEGF

Figure 4 |Processes related to vascular and leukocyte communication that can be targeted to combat inflammation and fibrosis. Repeated tissue damage can lead to high levels of matrix deposition following myofibroblast activation and the continued secretion of pro-inflammatory factors in the wounded tissue. Recruitment of M1 macrophages, mast cells and neutrophils, together with on-site formation of myofibroblasts, supports fibrosis that will eventually lead to tissue dysfunction. Different classes of leukocytes, as well as myofibroblasts, are potential targets

to reduce uncontrolled fibrosis. Elimination of the inflammatory triggers together with on-site reprogramming of macrophages and myofibroblasts as well as inhibition of recruitment of inflammatory leukocytes can be used to promote fibrosis resolution. Processes that can be targeted therapeutically are indicated. DAMPs, damage-associated molecular pattern molecules; EMT, epithelial–mesenchymal transition; EndMT, endothelial–mesenchymal transition; MSC, mesenchymal stem cell; PDGF, platelet-derived growth factor; rM, resolution-phase macrophage; TGFβ, transforming growth factor-β; VEGF, vascular endothelial growth factor.

and MMP9 using the novel inhibitor mNOX-E36 that targets CCL2 was shown to prevent fibrosis-associated angiogenesis6.

An additional macrophage subtype — resolution-phase macrophages (rMs) — that has increased expression of the M1 markers iNOS and COX2 is also present during resolu-tion of inflammation207. Differentiation into this subtype is controlled by cAMP levels; M1 macrophages shift to rMs when cAMP levels are high and the opposite occurs when cAMP levels are low. Furthermore, in a model of liver fibrosis, regression of fibrosis was dependent on restorative post-phagocytic macrophages that express high levels of MMPs, growth factors and phagocytosis-related genes208. The restorative macrophage phenotype and concomitant reduction in scar formation could be induced by liposome administration, which promoted macrophage phagocytic behaviour208. Inducing the rM phenotype could thus be a potential therapeutic option. These results further dem-onstrate the wide range of effector functions exerted by different types of macrophages.Conclusions

As of now, information about the various cell types in and around microvessels that influence angiogenesis, inflammation and the development of fibrosis have been compiled, but the integrative effects of these different cells remain to be fully established, and comprehensive molecular and cellular interaction maps have not yet been built. This ‘knowledge integration’ is now taking place at an increasingly rapid pace, facilitated by direct in vivo observations of how cells of the microvascula-ture move and interact during angiogenesis and inflam-mation. A full appreciation of how these cells interact will not only tell us more about how angiogenesis and inflammation are regulated, but should also unveil many new potential methods to influence the microenviron-ment in order to steer cellular functions towards the restoration of homeostasis. New indications will also probably be found for existing drugs as a result of a more detailed molecular understanding of the communication between leukocytes and vascular cells.

1. Folkman, J. T umor angiogenesis: therapeutic

implications. N. Engl. J. Med. 285, 1182–1186 (1971).

2. Dvorak, H. F. T umors: wounds that do not heal.

Similarities between tumor stroma generation and

wound healing. N. Engl. J. Med. 315, 1650–1659

(1986).

3. Casazza, A.

et al. Impeding macrophage entry into hypoxic tumor areas by Sema3A/Nrp1 signaling

blockade inhibits angiogenesis and restores antitumor immunity. Cancer Cell 24, 695–709 (2013).

4. Christoffersson, G.

et al. VEGF-A recruits a

proangiogenic MMP-9-delivering neutrophil subset

that induces angiogenesis in transplanted hypoxic

tissue. Blood 120, 4653–4662 (2012).

5. Chung, A. S.

et al. An interleukin-17-mediated paracrine network promotes tumor resistance to

anti-angiogenic therapy. Nat. Med. 19, 1114–1123

(2013).

6. Ehling, J.

et al. CCL2-dependent infiltrating

macrophages promote angiogenesis in progressive

liver fibrosis. Gut 63, 1960–1971 (2014).

7. Kolaczkowska, E. & Kubes, P. Neutrophil recruitment

and function in health and inflammation. Nat. Rev.

Immunol. 13, 159–175 (2013).

8. Stark, K.

et al. Capillary and arteriolar pericytes attract innate leukocytes exiting through venules and

‘instruct’ them with pattern-recognition and motility

programs. Nat. Immunol. 14, 41–51 (2013).

9. Hase, K.

et al. M-Sec promotes membrane nanotube formation by interacting with Ral and the exocyst

complex. Nat. Cell Biol. 11, 1427–1432 (2009).

10. Kristensen, V. N. et al. Principles and methods of

integrative genomic analyses in cancer. Nat. Rev.

Cancer 14, 299–313 (2014).

11. Islam, S. et al. Quantitative single-cell RNA-seq with

unique molecular identifiers. Nat. Meth. 11, 163–166 (2014).

12. Bentley, K. et al. The role of differential VE-cadherin

dynamics in cell rearrangement during angiogenesis.

Nat. Cell Biol. 16, 309–321 (2014).

13. Dejana, E., T ournier-Lasserve, E. & Weinstein, B. M.

The control of vascular integrity by endothelial cell

junctions: molecular basis and pathological

implications. Dev. Cell 16, 209–221 (2009).

14. Chauhan, V. P., Stylianopoulos, T., Boucher, Y. &

Jain, R. K. Delivery of molecular and nanoscale

medicine to tumors: transport barriers and strategies.

Annu. Rev. Chem. Biomol. Eng. 2, 281–298 (2011). 15. Heldin, C. H., Rubin, K., Pietras, K. & Ostman, A.

High interstitial fluid pressure — an obstacle in cancer therapy. Nat. Rev. Cancer 4, 806–813 (2004).

16. Conway, D. E. et al. Fluid shear stress on endothelial

cells modulates mechanical tension across VE-cadherin and PECAM-1. Curr. Biol. 23, 1024–1030 (2013). 17. Tzima, E. et al. A mechanosensory complex that

mediates the endothelial cell response to fluid shear

stress. Nature 437, 426–431 (2005).18. Langenkamp, E. & Molema, G. Microvascular

endothelial cell heterogeneity: general concepts and

pharmacological consequences for anti-angiogenic

therapy of cancer. Cell Tissue Res. 335, 205–222

(2009).

19. Swift, M. R. & Weinstein, B. M. Arterial-venous

specification during development. Circ. Res. 104,

576–588 (2009).

20. Lawson, N. D., Vogel, A. M. & Weinstein, B. M.

Sonic hedgehog and vascular endothelial growth

factor act upstream of the Notch pathway during

arterial endothelial differentiation. Dev. Cell 3,

127–136 (2002).

21. Wang, H. U., Chen, Z. F. & Anderson, D. J.

Molecular distinction and angiogenic interaction

between embryonic arteries and veins revealed by

ephrin-B2 and its receptor Eph-B4. Cell 93,

741–753 (1998).

22. D’Onofrio, N. et al. Vascular-homing peptides for

targeted drug delivery and molecular imaging:

meeting the clinical challenges. Biochim. Biophys. Acta

1846, 1–12 (2014).

23. Ageirsdottir, S. A. et al. Site-specific inhibition

of glomerulonephritis progression by targeted

delivery of dexamethasone to glomerular

endothelium. Mol. Pharmacol. 72, 121–131 (2007).

This study demonstrates means to direct drug

delivery to specific organs and thereby limit

systemic adverse effects by decorating exosomes

with relevant adhesion molecules.

24. Ageirsdottir, S. A. et al. Inhibition of proinflammatory

genes in anti-GB; glomerulonephritis by targeted

dexamethasone-loaded AbEsel liposomes.

Am. J. Renal Physiol. 294, F554–F561 (2008).

25. Corada, M. et al. Sox17 is indispensable for

acquisition and maintenance of arterial identity.

Nat. Commun. 4, 2609 (2013).

26. Fancy, S. P. et al. Parallel states of pathological Wnt

signaling in neonatal brain injury and colon cancer.

Nat. Neurosci. 17, 506–512 (2014).

27. Wu, B., Crampton, S. P. & Hughes, C. C. Wnt signaling

induces matrix metalloproteinase expression and

regulates T cell migration. Immunity 26, 227–239

(2007).

28. J ung, Y. S. et al. Wnt5a stimulates chemotactic

migration and chemokine production in human

neutrophils. Exp. Mol. Med. 45, e27 (2013).

29. Maiti, G., Naskar, D. & Sen, M. The Wingless homolog

Wnt5a stimulates phagocytosis but not bacterial

killing. Proc. Natl Acad. Sci. USA 109, 16600–16605

(2012).

30. Armulik, A. et al. Pericytes regulate the blood-brain

barrier. Nature 468, 557–561 (2010).

31. Gaengel, K. et al. The sphingosine-1-phosphate

receptor S1PR1 restricts sprouting angiogenesis by

regulating the interplay between VE-cadherin and

VEGFR2. Dev. Cell 23, 587–599 (2012).

This study shows for the first time that sphingosine

1?phosphate (produced by, for example, platelets)

is required for vessel maturation via the inhibition

of VEGFA signalling and stabilization of

VE?cadherin?based junctions.

32. Krueger, M. & Bechmann, I. CNS pericytes: concepts,

misconceptions, and a way out. Glia 58, 1–10 (2010).

33. Sims, D. E. The pericyte — a review. Tissue Cell 18,

153–174 (1986).

34. Sims, D. E. Recent advances in pericyte biology —

implications for health and disease. Can. J. Cardiol.

7, 431–443 (1991).

35. Falcón, B. L. et al. Contrasting actions of selective

inhibitors of angiopoietin-1 and angiopoietin-2 on the

normalization of tumor blood vessels. Am. J. Pathol.

175, 2159–2170 (2009).

36. Gerhardt, H., Wolburg, H. & Redies, C. N-cadherin

mediates pericytic-endothelial interaction during brain

angiogenesis in the chicken. Dev. Dyn. 218, 472–479

(2000).

37. Wernersson, S. & Pejler, G. Mast cell secretory

granules: armed for battle. Nat. Rev. Immunol.

14, 478–494 (2014).

38. Pollard, A. J., Perrett, K. P. & Beverley, P. C.

Maintaining protection against invasive bacteria with

protein-polysaccharide conjugate vaccines. Nat. Rev.

Immunol. 9, 213–220 (2009).

39. Fantin, A. et al. Tissue macrophages act as cellular

chaperones for vascular anastomosis downstream of

VEGF-mediated endothelial tip cell induction. Blood

116, 829–840 (2010).

This study demonstrates for the first time the role

of tissue resident macrophages in vessel fusion in

the developing brain.

40. Sica, A. & Mantovani, A. Macrophage plasticity and

polarization: in vivo veritas. J. Clin. Invest. 122,

787–795 (2012).

41. Mantovani, A. Molecular pathways linking inflammation

and cancer. Curr. Mol. Med. 10, 369–373 (2010).

42. Mantovani, A. & Sica, A. Macrophages, innate

immunity and cancer: balance, tolerance, and

diversity. Curr. Opin. Immunol. 22, 231–237 (2010).

43. Nozawa, H., Chiu, C. & Hanahan, D. Infiltrating

neutrophils mediate the initial angiogenic switch in a

mouse model of multistage carcinogenesis. Proc. Natl

Acad. Sci. USA 103, 12493–12498 (2006).

44. Wynn, T. A., Chawla, A. & Pollard, J. W.

Macrophage biology and health and disease.

Nature 496, 445–455 (2013).

45. Gautier, E. L. et al. Gene-expression profiles and

transcriptional regulatory pathways that underlie the

identity and diversity of mouse tissue macrophages.

Nat. Immunol. 13, 1118–1128 (2012).

46. Armulik, A., Genové, G. & Betsholtz, C.

Pericytes: development, physiological, and

pathological perspectives, problems, and promises.

Dev. Cell 21, 193–215 (2011).

47. Wei, X. et al. Mesenchymal stem cells: a new trend

for cell therapy. Acta Pharmacol. Sin. 34, 747–754

(2013).

48. Leeper, N. J., Hunter, A. L. & Cooke, J. P. Stem cell

therapy for vascular regeneration. Circulation 122,

517–526 (2010).

49. Schink?the, T., Bloch, W. & Schmidt, A. In vitro

secreting profile of human mesenchymal stem cells.

Stem Cells Dev. 17, 199–206 (2008).

50. Hsiao, S. T. et al. Comparative analysis of paracrine

factor expression in human adult mesenchymal stem

cells derived from bone marrow, adipose tissue, and

dermal tissue. Stem Cells Dev. 21, 2189–2203 (2012).

51. Rustad, K. C. & Gurtner, G. C. Mesenchymal stem cells

home to sites of injury and inflammation. Adv. Wound

Care 1, 147–152 (2012).

52. Abram, C. L. & Lowell, C. A. The ins and outs of

leukocyte integrin signaling. Annu. Rev. Immunol.

27, 339–362 (2009).

53. Fang, J. S., Dai, C., Kurjiaka, D. T., Burt, J. M. &

Hirschi, K. K. Connexin45 regulates endothelial-

induced mesenchymal cell differentiation toward a

mural cell phenotype. Arterioscler. Thromb. Vasc. Biol.

33, 362–368 (2013).

54. Hirschi, K. K., Burt, J. M., Hirschi, K. D. & Dai, C.

Gap junction communication mediates transforming

growth factor-β activation and endothelial-induced

mural cell differentiation. Circ. Res. 93, 429–437

(2003).

This study shows that CX43 gap junction

communication between endothelial cells and

mesenchymal progenitors is necessary for the

differentiation of progenitors into mural cells.

55. Westphalen, K. et al. Sessile alveolar macrophages

communicate with alveolar epithelium to modulate

immunity. Nature 506, 503–506 (2014).

56. Mazzini, E., Massimiliano, L., Penna, G. &

Rescigno, M. Oral tolerance can be established via

gap junction transfer of fed antigens from CX3CR1+

macrophages to CD103+ dendritic cells. Immunity

40, 248–261 (2014).

57. Oviedo-Orta, E., Errington, R. J. & Evans, W. H.

Gap junction intercellular communication during

lymphocyte transendothelial migration. Cell. Biol. Int.

26, 253–263 (2002).

58. Zahler, S. et al. Gap-junctional coupling between

neutrophils and endothelial cells: a novel modulator of

transendothelial migration. J. Leuk. Biol. 73, 118–126 (2003).

59. Ghatnekar, G. S. et al. Connexin43 carboxyl-terminal

peptides reduce scar progenitor and promote

regenerative healing following skin wounding.

Regen. Med. 4, 205–223 (2009).

60. Grek, C. S., Rhett, J. M. & Ghatnekar, G. S. Cardiac to

cancer: connecting connexins to clinical opportunity.

FEBS Lett. 588, 1349–1364 (2014).

61. Ghatnekar, G. S., Grek, C. L., Armstrong, D. G.,

Desai, S. C. & Gourdie, R. G. The effect of a

connexin43-based peptide on the healing of chronic

venous leg ulcers: a multicenter, randomized trial.

J. Invest. Dermatol. 135, 289–298 (2015).

62. van Balkom, B. W. et al. Endothelial cells require

miR-214 to secrete exosomes that suppress senescence and induce angiogenesis in human and mouse

endothelial cells. Blood 121, 3997–4006 (2013).

63. Valadi, H. et al. Exosome-mediated transfer of mRNAs

and microRNAs is a novel mechanism of genetic

exchange between cells. Nat. Cell Biol. 9, 654–659

(2007).

64. Patzelt, J. & Langer, H. F. Platelets in angiogenesis.

Curr. Vasc. Pharmacol. 10, 570–577 (2012).

65. Sreeramkumar, V. et al. Neutrophils scan for activated

platelets to initiate inflammation. Science 346,

1234–1238 (2014).

66. Battinelli, E. M., Markens, B. A. & Italiano, J. E. Jr.

Release of angiogenesis regulatory proteins from

platelet alpha granules: modulation of physiologic and

pathologic angiogenesis. Blood 118, 1359–1369

(2011).

67. Alvarez-Erviti, L. et al. Delivery of siRNA to the mouse

brain by systemic injection of targeted exosomes.

Nat. Biotech. 29, 341–345 (2011).

This study presents evidence that exosomes from

self?derived dendritic cells, engineered to express

the neuron?specific rabies viral glycoprotein (RVG)

peptide, can selectively deliver functional siRNA to

distinct cell populations in the brain.

68. Zech, D., Sanyukta, R., Büchler, M. W. & X?ller, M.

T umor-crosstalk and leukocyte activation: an

ambivalent crosstalk. Cell Comm. Signal. 10, 37

(2012).69. Marleau, A. M., Chen, C. S., Joyce, J. A. & T ullis, R. H.

Exosome removal as a therapeutic adjuvant in cancer.

J. T ransl. Med. 10, 134 (2012).

70. Lu, H. et al. Exo70 isoform switching upon epithelial–

mesenchymal transition mediates cancer cell invasion.

Dev. Cell 27, 560–573 (2013).

71. Zhao, Y. et al. Exo70 generates membrane curvature

for morphogenesis and cell migration. Dev. Cell 26,

266–278 (2013).

72. Chacon-Heszele, M. F. et al. The exocyst and regulatory

GTPases in urinary exosomes. Physiol. Rep. 2, e12116

(2014).

73. Lou, E. et al. T unneling nanotubes provide a unique

conduit for intercellular transfer of cellular contents in

human malignant pleural mesothelioma. PLoS ONE

7, e33093 (2012).

74. Ahmad, T. et al. Miro1 regulates intercellular

mitochondrial transport and enhances mesenchymal

stem cell rescue efficacy. EMBO J. 33, 994–1010

(2014).

75. Chen, L., Yang, S., Jakoncic, J., Zhang, J. J. &

Huang, X. Y. Migrastatin analogues target fascin to

block tumour metastasis. Nature 464, 1062–1066

(2010).

76. Pasquier, J. et al. Preferential transfer of mitochondria

from endothelial to cancer cells through tunneling

nanotubes modulates chemoresistance. J. T ransl.

Med. 11, 94 (2013).

77. Rainy, N. et al. H-Ras transfers from B to T cells via

tunneling nanotubes. Cell Death Dis. 4, e726 (2013).

78. Yasuda, K. et al. T unneling nanotubes mediate rescue

of prematurely senescent endothelial cells by

endothelial progenitors: exchange of lysosomal pool.

Aging 3, 597–608 (2011).

79. Liu, K. et al. Mesenchymal stem cells rescue injured

endothelial cells in an in vitro ischemia–reperfusion

model via tunneling nanotube like structure-mediated

mitochondrial transfer. Microvasc. Res. 92, 10–18

(2014).

80. Claesson-Welsh, L. & Welsh, M. VEGFA and tumour

angiogenesis. J. Intern. Med. 273, 114–127 (2013).

81. Jain, R. K. Normalizing tumor vasculature with anti-

angiogenic therapy: a new paradigm for combination

therapy. Nat. Med. 7, 987–989 (2001).

82. Adams, R. H. & Alitalo, K. Molecular regulation of

angiogenesis and lymphangiogenesis. Nat. Rev. Mol.

Cell. Biol. 8, 464–478 (2007).

83. Styp-Rekowska, B., Hlushchuk, R., Pries, A. R. &

Djonov, V. Intussusceptive angiogenesis: pillars

against the blood flow. Acta Physiol. (Oxf.) 202,

213–223 (2011).

84. Adams, R. H. & Eichmann, A. Axon guidance

molecules in vascular patterning. Cold Spring Harb.

Perspect. Biol.2, a001875 (2010).

85. Jakobsson, L., Kreuger, J. & Claesson-Welsh, L.

Building blood vessels-stem cell models in vascular

biology. J. Cell Biol. 177, 751–755 (2007).

86. Lander, A. D. How cells know where they are. Science

339, 923–927 (2013).

87. Benest, A. V. & Augustin, H. G. T ension in the

vasculature. Nat. Med. 15, 608–610 (2009).

88. Boerckel, J. D., Uhrig, B. A., Willett, N. J., Huebsch, N.

& Guldberg, R. E. Mechanical regulation of vascular

growth and tissue regeneration in vivo. Proc. Natl

Acad. Sci. USA 108, E674–E680 (2011).

89. Kilarski, W. W., Samolov, B., Petersson, L., Kvanta, A.

& Gerwins, P. Biomechanical regulation of blood

vessel growth during tissue vascularization. Nat. Med.

15, 657–664 (2009).

This study demonstrates that during wound healing

activated myofibroblasts generate mechanical

forces that rapidly pull and elongate vessels from

intact vascular beds into the wounded area.

The vascular network is thereafter modulated

by sprouting, splitting and regression.

90. Vakoc, B. J. et al. Three-dimensional microscopy of the

tumor microenvironment in vivo using optical frequency

domain imaging. Nat. Med. 15, 1219–1223 (2009).

91. T abe, Y. et al. TGF-β-neutralizing antibody 1D11

enhances cytarabine-induced apoptosis in AML cells

in the bone marrow microenvironment. PLoS ONE 8,

e62785 (2011).

92. Dituri, F. et al. Differential inhibition of the TGF-β

signaling pathway in HCC cells using the small

molecule inhibitor LY2157299 and the D10

monoclonal antibody against TGF-β receptor type II.

PLoS ONE 8, e67109 (2013).

93. Santiago, B. et al. T opical application of a peptide

inhibitor of transforming growth factor-β1 ameliorates

bleomycin-induced skin fibrosis. J. Invest. Dermatol.

125, 450–455 (2005).

94. Khalil, N., Bereznay, O., Sporn, M. & Greenberg, A. H.

Macrophage production of transforming growth factor β

and fibroblast collagen synthesis in chronic pulmonary

inflammation. J. Exp. Med. 170, 727–737 (1989).

95. Nacu, N. et al. Macrophages produce TGF-β-induced

(β-ig-h3) following ingestion of apoptotic cells and

regulate MMP14 levels and collagen turnover in

fibroblasts. J. Immunol. 180, 5036–5044 (2008).

96. T aichman, N. S., Young, S., Cruchley, A. T., T aylor, P. &

Paleoplog, E. Human neutrophils secrete vascular

endothelial growth factor. J. Leuk. Biol. 62, 397–400

(1997).

97. Auffray, C. et al. Monitoring of blood vessels and

tissue by a population of monocytes with patrolling

behavior. Science 317, 666–670 (2007).

98. Awojoodu, A. O. et al. Sphingosine 1-phosphate

receptor 3 regulates recruitment of anti-inflammatory

monocytes to microvessels during implant

arteriogenesis. Proc. Natl Acad. Sci. USA 110,

13785–13790 (2013).

99. Carlin, L. M. et al. Nr4a1-dependent Ly6C(low)

monocytes monitor endothelial cells and orchestrate

their disposal. Cell 153, 362–375 (2013).

100. Ganju, R. K. et al. The α-chemokine, stromal cell-

derived factor-1α, binds to the transmembrane

G-protein-coupled CXCR-4 receptor and activates

multiple signal transduction pathways. J. Biol. Chem.

273, 23169–23175 (1998).

101. Grunewald, M. et al. VEGF-induced adult

neovascuarisation, recruitment, retention and role

of accessory cells. Cell 124, 175–189 (2006).

102. Avraham-Davidi, I. et al. On-site education of

VEGF-recruited monocytes improves their performance

as angiogenesis and arteriogenic accessory cells.

J. Exp. Med. 210, 2611–2625 (2013).

This study demonstrates that local VEGF

‘re?educates’ monocytes to attain a pro?angiogenic

phenotype.

103. Christoffersson, G. et al. Clinical and experimental

pancreatic islet transplantation to striated muscle:

establishment of a vascular system similar to that

in native islets. Diabetes 59, 2569–2578 (2010).

104. Massena, S. et al. Identification and characterization

of VEGF-A-responsive neutrophils expressing CD49d,

VEGFR1 and CXCR4 in mice and humans. Blood126,

2016–2026 (2015).

105. Kerbel, R. S. Inhibition of tumor angiogenesis as a

strategy to circumvent acquired resistance to anti-

cancer therapeutic agents. Bioessays 13, 31–36

(1991).

106. Lim, L. S., Mitchell, P., Seddon, J. M., Holz, F. G. &

Wong, T. Y. Age-related macular degeneration. Lancet

379, 1728–1738 (2012).

107. Reck, M. et al. Phase III trial of cisplatin plus

gemcitabine with either placebo or bevacizumab as

first-line therapy for nonsquamous non-small-cell lung

cancer: AVAiL. J. Clin. Oncol. 27, 1227–1234 (2009).

108. Saltz, L. B. et al. Bevacizumab in combination with

oxaliplatin-based chemotherapy as first-line therapy in

metastatic colorectal cancer: a randomized Phase III

study. J. Clin. Oncol. 26, 2013–2019 (2008).

109. Sandler, A. et al. Paclitaxel–carboplatin alone or with

bevacizumab for non-small-cell lung cancer. N. Engl.

J. Med. 355, 2542–2550 (2006).

110. Willett, C. G. et al. Direct evidence that the VEGF-

specific antibody bevacizumab has antivascular effects

in human renal cancer. Nat. Med. 10, 145–147 (2004).

111. Batchelor, T. T. et al. Improved tumor oxygenation and

survival in glioblastoma patients who show increased

blood perfusion after cediranib and chemoradiation.

Proc. Natl Acad. Sci. USA 47, 19059–19064 (2013).

112. Armstrong, T. S., Wen, P. Y., Gilbert, M. R. & Schiff, D.

Management of treatment-associated toxicites of

anti-angiogenic therapy in patients with brain tumors.

Neuro. Oncol. 14, 1203–1214 (2012).

113. Wu, J. M. & Staton, C. A. Anti-angiogenic drug

discovery: lessons from the past and thoughts for the

future. Expert Opin. Drug Discov. 7, 723–743 (2012).

114. Bergers, G. & Hanahan, D. Modes of resistance

to anti-angiogenic therapy. Nat. Rev. Cancer 8,

592–603 (2008).

115. Barone, A. et al. Combined VEGF and CXCR4

antagonism targets the GBM stem cell population

and synergistically improves survival in an

intracranial mouse model of glioblastoma.

Oncotarget 5, 9811–9822 (2014).

In a mouse model of glioblastoma, this study shows

that angiogenesis, tumour growth and metastasis

are reduced by treatment with the CXCR4 inhibitor

CTCE?9908 in combination with docetaxel or a

VEGF?specific antibody.

116. Hassan, S. et al. CXCR4 peptide antagonist inhibits primary breast tumor growth, metastasis and

enhances the efficacy of anti-VEGF treatment or

docetaxel in a transgenic mouse model. Int. J. Cancer

129, 225–232 (2011).

117. Pan, Q. et al. Blocking neuropilin-1 function has an additive effect with anti-VEGF to inhibit tumor growth.

Cancer Cell 11, 53–67 (2007).

118. Ruan, G.-X. & Kazlauskas, A. Lactate engages receptor tyrosine kinases Axl, Tie2, and vascular endothelial

growth factor receptor 2 to activate phosphoinositide

3-kinase/Akt and promote angiogenesis. J. Biol. Chem.

288, 21161–21172 (2013).

119. De Bock, K. et al. Role of PFKFB3-driven glycolysis in vessel sprouting. Cell 154, 651–663 (2013).

This study demonstrates that loss of the glycolytic

activator PFKFB3 in endothelial cells results in

impaired angiogenesis. PFKFB3 is required for

proliferation but also regulates filopodia formation

and chemotaxis.

120. Leopold, J. A. et al. Glucose-6-phosphate

dehydrogenase modulates vascular endothelial growth factor-mediated angiogenesis. J. Biol. Chem. 278,

32100–32106 (2003).

121. Schoors, S. et al. Partial and transient reduction of glycolysis by PFKFB3 blockade reduces pathological

angiogenesis. Cell. Metab. 19, 37–48 (2014).

122. Leite de Oliveira, R. et al. Gene-targeting of Phd2 improves tumor response to chemotherapy and

prevents side-toxicity. Cancer Cell 22, 263–277 (2012). 123. Mazzone, M. et al. Heterozygous deficiency of PHD2 restores tumor oxygenation and inhibits metastasis via endothelial normalization. Cell 136, 839–851 (2009). 124. T akeda, Y. et al. Macrophage skewing by Phd2 haplodeficiency prevents ischaemia by inducing

arteriogenesis. Nature 479, 122–126 (2011).

125. Hamm, A. et al. PHD2 regulates arteriogenic macrophages through TIE2 signalling. EMBO Mol. Med.

5, 843–857 (2013).

126. Flagg, S. C., Martin, C. B., T aabazuing, C. Y.,

Holmes, B. E. & Knapp, M. J. Screening chelating

inhibitors of HIF–prolyl hydroxylase domain 2 (PHD2)

and factor inhibiting HIF (FIH). J. Inorg. Biochem. 113,

25–30 (2012).

127. Hellstr?m, M. et al. Dll4 signalling through Notch1 regulates formation of tip cells during angiogenesis.

Nature 445, 776–780 (2007).

128. Phng, L. K. et al. Nrarp coordinates endothelial Notch and Wnt signaling to control vessel density in

angiogenesis. Dev. Cell 16, 70–82 (2009).

129. Dimova, I. et al. Inhibition of Notch signaling induces extensive intussusceptive neo-angiogenesis by

recruitment of mononuclear cells. Angiogenesis

16, 921–937 (2013).

130. Outtz, H. H., T attersall, I. W., Kofler, N. M.,

Steinbach, N. & Kitajewski, J. Notch1 controls

macrophage recruitment and Notch signaling is

activated at sites of endothelial cell anastomosis

during retinal angiogenesis in mice. Blood 118,

3436–3439 (2011).

131. Camelo, S. et al. Delta-like 4 inhibits choroidal neovascularization despite opposing effects on

vascular endothelium and macrophages. Angiogenesis 15, 609–622 (2012).

132. Haj Zen, A. A. et al. Inhibition of delta-like-4-mediated signaling impairs reparative angiogenesis after

ischemia. Circ. Res. 107, 283–293 (2010).

133. Wang, Y. C. et al. Notch signaling determines the M1 versus M2 polarization of macrophages in antitumor

immune responses. Cancer Res. 70, 4840–4849

(2010).

134. Xu, H. et al. Notch–RBP-J signaling regulates the transcription factor IRF8 to promote inflammatory

macrophage polarization. Nat. Immunol. 13,

642–650 (2012).

135. Stockmann, C. et al. Deletion of vascular endothelial growth factor in myeloid cells accelerates

tumorigenesis. Nature 456, 814–819 (2008).

136. DeNardo, D. G. et al. Leukocyte complexity predicts breast cancer survival and functionally regulates

response to chemotherapy. Cancer Discov. 1, 54–67

(2011).

137. Ries, C. H. et al. T argeting tumor-associated macro-phages with anti-CSF-1R antibody reveals a strategy

for cancer therapy. Cancer Cell 25,846–859 (2014). 138. Bonapace, L. et al. Cessation of CCL2 inhibition accelerates breast cancer metastasis by promoting

angiogenesis. Nature 515, 130–133 (2014).

139. Stefater, J. A. 3rd et al. Regulation of angiogenesis by

a non-canonical Wnt–Flt1 pathway in myeloid cells.

Nature 474, 511–515 (2011).140. Stefater, J. A. 3rd et al. Macrophage Wnt–Calcineurin–

Flt1 signaling regulates mouse wound angiogenesis

and repair. Blood 121, 2574–2578 (2013).

141. Ross, K. For organ transplant recipients, cancer

threatens long-term survival. J. Nat. Cancer Inst.

99, 421–422 (2007).

142. Ley, K., Laudanna, C., Cybulsky, M. I. &

Nourshargh, S. Getting to the site of inflammation:

the leukocyte adhesion cascade updated. Nat. Rev.

Immunol. 7, 678–689 (2007).

143. Nourshargh, S., Hordijk, P. L. & Sixt, M. Breaching

multiple barriers: leukocyte motility through venular

walls and the interstitium. Nat. Rev. Mol. Cell. Biol.

11, 366–378 (2010).

144. Phillipson, M. & Kubes, P. The neutrophil in vascular

inflammation. Nat. Med. 17, 1381–1390 (2011).

145. Luo, B. H., Carman, C. V. & Springer, T. A. Structural

basis of integrin regulation and signaling. Annu. Rev.

Immunol. 25, 619–647 (2007).

146. Massena, S. et al. A chemotactic gradient sequestered

on endothelial heparan sulfate induces directional

intraluminal crawling of neutrophils. Blood 116,

1924–1931 (2010).

This study demonstrates that an intravascular

chemo k ine gradient on endothelial heparan

sulfate directs crawling neutrophils towards the

site of chemokine origin and accelerates their

transmigration.

147. Phillipson, M. et al. Intraluminal crawling of

neutrophils to emigration sites: a molecularly distinct

process from adhesion in the recruitment cascade.

J. Exp. Med. 203, 2569–2575 (2006).

148. Schenkel, A. R., Mamdouh, Z. & Muller, W. A.

Locomotion of monocytes on endothelium is a critical

step during extravasation. Nat. Immunol. 5, 393–400

(2004).

149. Voisin, M. B., Pr?bstl, D. & Nourshargh, S.

Venular basement membranes ubiquitously express

matrix protein low-expression regions: characterization

in multiple tissues and remodeling during inflammation.

Am. J. Pathol. 176, 482–495 (2010).

150. Wang, S. et al. Venular basement membranes contain

specific matrix protein low expression regions that act

as exit points for emigrating neutrophils. J. Exp. Med.

203, 1519–1532 (2006).

151. Carman, C. V. & Springer, T. A. A transmigratory cup

in leukocyte diapedesis both through individual

vascular endothelial cells and between them.

J. Cell Biol. 167, 377–388 (2004).

152. Petri, B. et al. Endothelial LSP1 is involved in

endothelial dome formation, minimizing vascular

permeability changes during neutrophil

transmigration in vivo. Blood 117, 942–952

(2011).

153. Phillipson, M., Kaur, J., Colarusso, P.,

Ballantyne, C. M. & Kubes, P. Endothelial domes

encapsulate adherent neutrophils and minimize

increases in vascular permeability in paracellular

and transcellular emigration. PLoS ONE 3, e1649

(2008).

154. Liu, L. et al. LSP1 is an endothelial gatekeeper of

leukocyte transendothelial migration. J. Exp. Med.

201, 409–418 (2005).

155. Broermann, A. et al. Dissociation of VE-PTP from

VE-cadherin is required for leukocyte extravasation

and for VEGF-induced vascular permeability in vivo.

J. Exp. Med. 208, 2393–2401 (2011).

156. Finsterbusch, M., Voisin, M. B., Beyrau, M.,

Williams, T. J. & Nourshargh, S. Neutrophils recruited

by chemoattractants in vivo induce microvascular

plasma protein leakage through secretion of TNF.

J. Exp. Med. 211, 1307–1314 (2014).

157. Proebstl, D. et al. Pericytes support neutrophil

subendothelial cell crawling and breaching of venular

walls in vivo. J. Exp. Med. 209, 1219–1234 (2012).

This study demonstrates that pericytes support

neutrophil migration to inflamed tissue by

providing a surface for neutrophil crawling and by

enlarging the inter?pericyte spaces by contraction.

158. Abtin, A. et al. Perivascular macrophages mediate

neutrophil recruitment during bacterial skin infection.

Nat. Immunol. 15, 45–53 (2014).

This study shows that perivascular macrophages

guide neutrophil extravasation and comprise

targets during bacterial immune evasion.

159. T abas, I. & Glass, C. K. Anti-inflammatory therapy in

chronic disease: challenges and opportunities. Science

339, 166–172 (2013).

160. Fox, D. A. Kinase inhibition — a new approach to the

treatment of rheumatoid arthritis. N. Engl. J. Med.

367, 565–567 (2012).

161. Hoepner, R., Faissner, S., Salmen, A., Gold, R. &

Chan, A. Efficacy and side effects of natalizumab

therapy in patients with multiple sclerosis.

J. Cent. Nerv. Syst. Dis. 6, 41–49 (2014).

162. Kothary, N. et al. Progressive multifocal

leukoencephalopathy associated with efalizumab use

in psoriasis patients. J. Am. Acad. Dermatol. 65,

546–551 (2011).

163. Schwab, N. et al. Fatal PML associated with

efalizumab therapy: insights into integrin αLβ2 in

JC virus control. Neurology 78, 458–467 (2012).

164. Van Assche, G. et al. Progressive multifocal

leukoencephalopathy after natalizumab therapy for

Crohn’s disease. N. Engl. J. Med. 353, 362–368

(2005).

165. Feagan, B. G. et al. Vedolizumab as induction and

maintenance therapy for ulcerative colitis. N. Eng.

J. Med. 369, 699–710 (2013).

In this clinical study, the beneficial effect of

integrin?specific antibody therapy targeting α4β7

in patients with ulcerative colitis is demonstrated.

166. Sandborn, W. J. et al. Vedolizumab as induction and

maintenance therapy for Crohn’s disease. N. Eng.

J. Med. 369, 711–721 (2013).

167. Briskin, M. et al. Human mucosal addressin cell

adhesion molecule-1 is preferentially expressed in

intestinal tract and associated lymphoid tissue.

Am. J. Pathol. 151, 97–110 (1997).

168. Danese, S. & Panés, J. Development of drugs to target

interactions between leukocytes and endothelial cells

and treatment algorithms for inflammatory bowel

diseases. Gastroenterology 147, 981–989 (2014).

169. Fedyk, E. R. et al. Exclusive antagonism of the α4β7

integrin by vedolizumab confirms the gut-selectivity of

this pathway in primates. Inflamm. Bowel. Dis. 18,

2107–2119 (2012).

170. Haanstra, K. G. et al. Antagonizing the α4β1 integrin,

but not α4β7, inhibits leukocytic infiltration of the

central nervous system in rhesus monkey

experimental autoimmune encephalomyelitis.

J. Immunol. 190, 1961–1973 (2013).

171. Milch, C. et al. Vedolizumab, a monoclonal antibody

to the gut homing α4β7 integrin, does not affect

cerebrospinal fluid T-lymphocyte immunophenotype.

J. Neuroimmunol. 264, 123–126 (2013).

172. Watanabe, M. et al. 370 AJM300, an oral α4 integrin

antagonist, for active ulcerative colitis: a multicenter,

randomized, double-blind, placebo-controlled

Phase 2A study. Gastroenterology146, S-82 (2014).

173. Economides, A. N. et al. Cytokine traps: multi-

component, high-affinity blockers of cytokine action.

Nat. Med. 9, 47–52 (2003).

174. Blanchetot, C. et al. Neutralizing nanobodies

targeting diverse chemokines effectively inhibit

chemokine function. J. Biol. Chem. 288,

25173–25182 (2013).

175. Berghmans, N. et al. Rescue from acute neuro-

inflammation by pharmacological chemokine-mediated

deviation of leukocytes. J. Neuroinfl. 9, 243 (2012).

176. Li, S. et al. Interference with glycosaminoglycan-

chemokine interactions with a probe to alter leukocyte

recruitment and inflammation in vivo. PLoS ONE 9,

e104107 (2014).

177. Klarenbeek, A. et al. T argeting chemokines and

chemokine receptors with antibodies. Drug Discov.

T oday T ech. 9, 237–244 (2012).

178. Perry, C. M. Maraviroc: a review of its use in the

management of CCR5-tropic HIV-1 infection. Drugs

70, 1189–1213 (2010).

179. De Clercq, E. The AMD3100 story: the path to the

discovery of a stem cell mobilizer (Mozobil). Biochem.

Pharmacol. 77, 1655–1664 (2009).

180. Moelants, E. A. et al. Citrullination of TNF-α by

peptidylarginine deiminases reduces its capacity to

stimulate the production of inflammatory chemokines.

Cytokine 61, 161–167 (2013).

181. Proost, P. et al. Citrullination of CXCL8 by peptidyl-

arginine deiminase alters receptor usage, prevents

proteolysis, and dampens tissue inflammation.

J. Exp. Med. 205, 2085–2097 (2008).

This study demonstrates that inflammation can be

regulated by post?translational modifications of

chemokines to greatly affect chemokine activity

and stability.

182. Moelants, E. A. et al. Citrullination and proteolytic

processing of chemokines by Porphyromonas

gingivalis. Infect. Immun. 82, 2511–2519 (2014).

183. Mantovani, A., Biswas, S. K., Galdiero, M. R., Sica, A.

& Locati, M. Macrophage plasticity and polarization

in tissue repair and remodelling. J. Pathol. 229,

176–185 (2013).

184. Korns, D., Frasch, S. C., Fernandez-Boyanapalli, R.,

Henson, P . M. & Bratton, D. L. Modulation of macrophage efferocytosis in inflammation. Front. Immunol. 2, 57 (2011).

185. Markworth, J. F . et al. Human inflammatory and

resolving lipid mediator responses to resistance exercise and ibuprofen treatment. Am. J. Physiol. Regul. Integr. Comp. Physiol. 305, R1281–R1296 (2013).

186. Colas, R. A., Shinohara, M., Dalli, J., Chiang, N. &

Serhan, C. N. Identification and signature profiles for pro-resolving and inflammatory lipid mediators in human tissue. Am. J. Physiol. Cell Physiol. 307, C39–C54 (2014).

This study identifies the mediators involved in resolution of acute inflammation, including lipoxins, resolvins, protectins and maresins in human plasma and lymphoid organs.

187. Serhan, C. N. Pro-resolving lipid mediators are leads

for resolution physiology. Nature 510, 92–101 (2014).188. Jin, Y. et al. Anti-angiogenesis effect of the novel

anti-inflammatory and pro-resolving lipid mediators. Invest. Ophthalmol. Vis. Sci. 50, 4743–4752 (2009).189. Rajasagi, N. K. et al. Controlling herpes simplex

virus-induced ocular inflammatory lesions with the lipid-derived mediator resolvin E1. J. Immunol. 186, 1735–1746 (2011).

190. T ang, Y. et al. Proresolution therapy for the treatment

of delayed healing of diabetic wounds. Diabetes 62, 618–627 (2013).

191. Wu, S. H., Chen, X. Q., Liu, B., Wu, H. J. & Dong, L.

Efficacy and safety of 15(R/S)-methyl-lipoxin A 4 in topical treatment of infantile eczema. Br. J. Dermatol. 168, 172–178 (2013).

192. Robertson, A. L. et al. A zebrafish compound screen

reveals modulation of neutrophil reverse migration as an anti-inflammatory mechanism. Sci. T ransl. Med. 6, 225ra29 (2014).

193. Pellicoro, A., Ramachandran, P ., Iredale, J. P . &

Fallowfield, J. A. Liver fibrosis and repair: immune regulation of wound heling in solid organs. Nat. Rev. Immunol. 14, 181–194 (2014).

194. Moustakas, A. & Heldin, P . TGF β and matrix regulated

epithelial to mesenchymal transition. Biochim. Biophys. Acta 1840, 2621–2634 (2014).

195. Brenner, D. A. et al. Origin of myofibroblasts in liver

fibrosis. Fibrogen. Tissue Repair 5, S17 (2012).

196. Hinz, B. et al. The myofibroblast: one function, multiple

origins. Am. J. Pathol. 170, 1807–1816 (2007).

197. Henderson, N. C. et al. T argeting of αV integrin identifies

a core molecular pathway that regulates fibrosis in several organs. Nat. Med. 19, 1617–1624 (2013).Depletion of αV integrin in mouse hepatic stellate cells was shown to protect against fibrosis in several organs. The authors suggest that

pharmacological targeting of αV integrin may be of broad clinical utility.

198. Thiery, J. P ., Acloque, H., Huang, R. Y. J. &

Nieto, M. A. Epithelial-mesenchymal transitions in development and disease. Cell 139, 871–890 (2009).199. Cenik, B. K., Ostapoff, K. T ., Gerber, D. E. &

Brekken, R. A. BIBF 1120 (nintedanib), a triple angio-kinase inhibitor , induces hypoxia but not EMT and blocks progression of preclinical models of lung and pancreatic cancer. Mol. Cancer Ther. 12, 992–1001 (2013).200. Mehal, W. Z., Iredale, J. & Friedman, S. L. Scraping

fibrosis: expressway to the core of fibrosis. Nat. Med. 17, 552–553 (2011).

201. Seki, E. et al. TLR4 enhances TGF-β signaling and

hepatic fibrosis. Nat. Med. 13, 1324–1332 (2007).202. Subramaniam, N. et al. Metformin-mediated BAMBI

expression in hepatic stellate cells induces prosurvival Wnt/β catenin signaling. Cancer Prev. Res. 5, 553–561 (2012).

203. Ogawa, S. et al. Anti-PDGF-B monoclonal antibody

reduces liver fibrosis development. Hepatol. Res. 40, 1128–1141 (2010).

204. Coulon, S. et al. Role of vascular endothelial growth

factor in the pathophysiology of nonalcoholic

steatohepatitis in two rodent models. Hepatology 57, 1793–19805 (2013).

205. Ardi, V. C., Kupriyanova, T . A., Deryugina, E. I. &

Quigley, J. P . Human neutrophils uniquely release TIMP-free MMP-9 to provide a potent catalytic

stimulator of angiogenesis. Proc. Natl Acad. Sci. USA 104, 20262–20267 (2007).

206. Purcell, B. P . et al. Injectable and bioresponsive

hydrogels for on-demand matrix metalloproteinase inhibition. Nat. Mater. 13, 653–661 (2014).207. Bystrom, J. et al. Resolution-phase macrophages

possess a unique inflammatory phenotype that is controlled by cAMP . Blood 112, 4117–4127 (2008).

208. Ramachandran, P . et al. Differential Ly-6C expression

identifies the recruited macrophage phenotype, which orchestrates the regression of murine liver fibrosis. Proc. Natl Acad. Sci. USA 109, E3186–E3195 (2012).

This paper describes the first identification and characterization of the ‘restorative macrophage’, which can promote tissue remodelling and resolution of fibrosis.

209. Hellstr?m, M. et al. Lack of pericytes leads to

endothelial hyperplasia and abnormal vascular

morphogenesis. J. Cell. Biol. 153, 543–553 (2001).210. Lindblom, P . et al. Endothelial PDGF-B retention is

required for proper investment of pericytes in the microvessel wall. Genes Dev. 17, 1835–1840 (2003).211. Lindahl, P ., Johansson, B. R., Leveen, P . &

Betsholtz, C. Pericyte loss and microaneurysm formation in PDGF-B-deficient mice. Science 277, 242–245 (1997).

212. Kelly-Goss, M. R., Sweat, R. S., Stapor, P . C.,

Peirce, S. M. & Murfee, W. L. T argeting pericytes for angiogenic therapies. Microcirculation 21, 345–357 (2014).

213. Ayres-Sander, C. E. et al. T ransendothelial migration

enables subsequent transmigration of neutrophils through underlying pericytes. PLoS ONE 8, e60025 (2013).

214. Ren, S. et al. LRP-6 is a coreceptor for multiple

fibrogenic signaling pathways in pericytes and

myofibroblasts that are inhibited by DKK-1. Proc. Natl Acad. Sci. USA 110, 1440–1445 (2013).

215. Humphreys, B. D. et al. Fate tracing reveals the

pericyte and not epithelial origin of myofibroblasts in kidney fibrosis. Am. J. Pathol. 176, 85–97 (2010).216. Lee, W. Y. et al. An intravascular immune response to

Borrelia burgdorferi involves Kupffer cells and iNKT cells. Nat. Immunol. 11, 295–302 (2010).217. Chintalgattu, V. et al. Coronary microvascular

pericytes are the cellular target of sunitinib malate-induced cardiotoxicity. Sci. T ransl. Med. 5, 187ra69 (2013).

218. Ehnman, M. et al. Distinct effects of ligand-induced

PDGFR α and PDGFR β signaling in the human rhabdomyosarcoma tumor cell and stroma cell

compartments. Cancer Res. 73, 2139–2149 (2013).219. Ruan, J. et al. Imatinin disrupts lymphoma

angiogenesis by targeting vascular pericytes. Blood 121, 5192–5202 (2013).

220. Reck, M. et al. Docetaxel plus nintedanib versus

docetaxel plus placebo in patients with previously treated non-small-cell lung cancer (LUME-Lung 1): a phase 3, double-blind, randomised controlled trial. Lancet Oncol. 15, 143–155 (2014).

221. Purow, B. Notch inhibition as a promising new

approach to cancer therapy. Adv. Exp. Med. Biol. 727, 305–319 (2012).

222. Sun, L. et al. Discovery of 5-[5-fluoro-2-oxo-1,2-

dihydroindol-(3Z)-ylidenemethyl]-2,4- dimethyl- 1H-pyrrole-3-carboxylic acid (2-diethyl a minoethyl)amide, a novel tyrosine kinase inhibitor targeting vascular endothelial and platelet-derived growth factor receptor tyrosine kinase. J. Med. Chem. 46, 1116–1119 (2003).

223. Roberts, W. G. et al. Antiangiogenic and antitumor

activity of a selective PDGFR tyrosine kinase inhibitor, CP-673,451. Cancer Res. 65, 957–966 (2005).224. Wilhelm, S. M. et al. BAY 43–9006 exhibits broad

spectrum oral antitumor activity and targets the

RAF/MEK/ERK pathway and receptor tyrosine kinases involved in tumor progression and angiogenesis. Cancer Res. 64, 7099–7109 (2004).

225. Laird, A. D. et al. SU6668 is a potent antiangiogenic

and antitumor agent that induces regression of established tumors. Cancer Res. 60, 4152–4160 (2000).

226. Pan, B. S. et al. MK-2461, a novel multitargeted

kinase inhibitor, preferentially inhibits the activated c-Met receptor. Cancer Res. 70, 1524–1533 (2010).227. Albert, D. H. et al. Preclinical activity of ABT -869,

a multitargeted receptor tyrosine kinase inhibitor. Mol. Can. Therap. 5, 995–1006 (2006).

228. Hu-Lowe, D. D. et al. Nonclinical antiangiogenesis and

antitumor activities of axitinib (AG-013736), an oral, potent, and selective inhibitor of vascular endothelial growth factor receptor tyrosine kinases 1, 2, 3. Clin. Cancer Res. 14, 7272–7283 (2008).

229. Buchdunger, E. et al. Selective inhibition of the

platelet-derived growth factor signal transduction pathway by a protein-tyrosine kinase inhibitor of the 2-phenylaminopyrimidine class. Proc. Natl Acad. Sci. USA 92, 2558–2562 (1995).

230. Heinrich, M. C. et al. Inhibition of c-KIT receptor

tyrosine kinase activity by STI 571, a selective tyrosine kinase inhibitor. Blood 96, 925–932 (2000).231. Heinrich, M. C. et al. Crenolanib inhibits the

drug-resistant PDGFRA D842V mutation associated with imatinib-resistant gastrointestinal stromal tumors. Clin. Cancer Res. 18, 4375–4384 (2012).232. T rudel, S. et al. CHIR-258, a novel, multitargeted

tyrosine kinase inhibitor for the potential treatment of t(4;14) multiple myeloma. Blood 105, 2941–2948 (2005).

233. Hilberg, F . et al. BIBF 1120: triple angiokinase

inhibitor with sustained receptor blockade and good antitumor efficacy. Cancer Res. 68, 4774–4782 (2008).

234. Pyonteck, S. M. et al. CSF-1R inhibition alters

macrophage polarization and blocks glioma

progression. Nat. Med. 19, 1264–1272 (2013).235. Qian, B. Z. et al. CCL2 recruits inflammatory

monocytes to facilitate breast-tumour metastasis. Nature 475, 222–225 (2011).

236. Allavena, P . et al. Intraperitoneal recombinant

gamma-interferon in patients with recurrent ascitic ovarian carcinoma: modulation of cytotoxicity and cytokine production in tumor-associated effectors and of major histocompatibility antigen expression on tumor cells. Cancer Res. 50, 7318–7323 (1990).

237. Krieg, A. M. Therapeutic potential of T oll-like receptor 9

activation. Nat. Rev. Drug Discov. 5, 471–484 (2006).238. Sun, Z., Yao, Z., Liu, S., T ang, H. & Yan, X. An

oligonucleotide decoy for Stat3 activates the immune response of macrophages to breast cancer. Immunobiology 211, 199–209 (2006).

239. Hemmi, H. et al. Small anti-viral compounds activate

the immune cells via the TLR7 MyD88-dependent signaling pathway. Nat. Immunol. 3, 196–200 (2002).240. Rolny, C. et al. HRG inhibits tumor growth and

metastasis by inducing macrophage polarization and vessel normalization through downregulation of PlGF . Cancer Cell 19, 31–44 (2011).

241. Freemerman, A. J. et al. Metabolic reprogramming

of macrophages: glucose transporter 1

(GLUT1)-mediated glucose metabolism drives a pro i nflammatory phenotype. J. Biol. Chem. 289, 7884–7896 (2014).

242. Beatty, G. L. et al. CD40 agonists alter tumor stroma

and show efficacy against pancreatic carcinoma in mice and humans. Science 331, 1612–1616 (2011).243. Odegaard, J. I. et al. Macrophage-specific PPAR γ

controls alternative activation and improves insulin resistance. Nature 447, 1116–1120 (2007).

244. Lutgens, E. et al. Deficient CD40–TRAF6 signaling in

leukocytes prevents atherosclerosis by skewing the immune response toward an antiinflammatory profile. J. Exp. Med. 207, 391–404 (2010).

245. Kurowska-Stolarska, M. et al. IL-33 amplifies the

polarization of alternatively activated macrophages that contribute to airway inflammation. J. Immunol. 183, 6469–6499 (2009).

246. Germano, G. et al. Antitumor and anti-inflammatory

effects of trabectedin on human myxoid liposarcoma cells. Cancer Res. 70, 2235–2244 (2010).

Acknowledgements

The authors would like to thank the members of their labora-tories and many of their colleagues, especially L. Claesson-Welsh, A. Moustakas and M. Welsh for sharing their insights and for stimulating discussions. The authors apologize to all colleagues whose relevant work has not been cited owing to space limitations. The authors are supported by the Swedish Medical Research Council, the Swedish Cancer Society, the Royal Swedish Academy of Sciences, the Swedish Diabetes Foundation, the Swedish Foundation for Strategic Research, the Novo Nordisk Foundation, the Ragnar S?derberg founda-tion, the Knut and Alice Wallenberg Foundation, the Ruth and Nils-Erik Stenb?cks Foundation, the Foundations for Proteoglycan Research, the Marie Sklodowska-Curie Innovative T raining Network InCeM (Integrated Component Cycling in Epithelial Cell Motility) and Uppsala University.

Competing interests statement

The authors declare no competing financial interests.