Inflammatory Cell Responses to Vascular Regenerative Methacrylic Acid-Containing Materials

by

Kongyu David Zhang

A thesis submitted in conformity with the requirements for the degree of Masters of Applied Science

Institute of Biomaterials and Biomedical Engineering University of Toronto

© Copyright by Kongyu Zhang 2016

ii

Inflammatory Cell Responses to Vascular Regenerative

Methacrylic Acid-Containing Materials

Kongyu David Zhang

Master’s of Applied Science

Institute of Biomaterials and Biomedical Engineering

University of Toronto

2016

Abstract

Poly(methacrylic acid-co-methyl methacrylate) (MAA) beads improve vascularization when

applied to cutaneously-wounded diabetic mice. The aim of this thesis is to understand the

vascular regenerative properties of MAA at the cellular and molecular level. Subcutaneous

injection of MAA beads promoted the formation of a denser and perfusable network of blood

vessels at days 3 and 7 relative to poly(methyl methacrylate) (MM) control beads. MAA beads

modulated the host response; promoting more neutrophils at day 1 and more macrophages at day

7, relative to MM beads. A M2 macrophage polarization bias was observed in MAA-treated

animals but not in MM-treated animals. Additionally, complement was involved in the

mechanism of MAA; complement inhibition (at the C1 or C3 levels) diminished the M2

polarization bias at day 3, although no changes in vascularity were noted. These findings deepen

our understanding of MAA and benefit the development of MAA-based biomaterials for

applications in regenerative medicine.

iii

Acknowledgments

I owe my sincerest gratitude to the support and inspiration of many remarkable friends and

colleagues without which this work would not be possible. I am indebted to Prof. Michael Sefton

for giving me an opportunity to embark on this scientific journey over the last two years. I am

grateful for his unwavering support, continuous guidance, and astute criticism throughout the

entire course of my study. Thank you to my committee members, Prof. Warren Chan, Prof. Clint

Robbins, and Dr. Christoph Licht for their meticulous suggestions and for asking the important

questions to ensure that my research, presentation, and writing upholds the highest standard.

I am extremely fortunate and proud to be part of one of the greatest laboratories in the

world. Thank you to all the members of the Sefton Lab for supporting me, challenging me, and

helping to shape me into a better researcher and individual. Thank you to Sasha Lisovsky and

Dean Chamberlain for being excellent mentors, for pointing out the flaws in my experiments and

for making sure that I was asking the right questions. Thanks to Alexander Vlahos and Nicholas

Cober for the laughs, memories and helpful scientific discussions. Thank you also to Redouan

Mahou, Michael West, Gabrielle Lam, Ilana Talior-Volodarsky, Virginie Coindre, and Yarden

Gratch. A big thank you to Chuen Lo for his surgical wisdom and for all of our fascinating

discussions; this work would truly not be possible without him.

One of the best parts of working in such an interdisciplinary field is the opportunity to work

closely with individuals from many diverse fields. Thanks to the Chan, Wheeler, and Yip labs for

being such a fun crowd to work alongside. Thanks to Shrey Sindhwani for sharing his insight

and for the mentorship. Thank you to Wilson Poon for sharing his love of food and assimilating

me into the “foodie” culture. Some of our most productive scientific discussions occurred over

delicious (and sometimes, not so pleasant) meals.

Thank you to Dionne White for making flow cytometry enjoyable and to the PRP lab for sharing

all their histology-related expertise. Thank you to Prof. Penney Gilbert for giving me the

opportunity to teach and to Mohammad Saleh for teaching me how to become a better mentor.

Lastly, thank you to my parents and brother for encouraging me to pursue my passions,

regardless of what they are. This thesis is dedicated to my late grandfather, who instilled in me

the importance of higher education.

iv

Table of Contents

Acknowledgments.......................................................................................................................... iii

Table of Contents ........................................................................................................................... iv

List of Figures ............................................................................................................................... vii

List of Appendices ....................................................................................................................... viii

List of Abbreviations ..................................................................................................................... ix

Chapter 1 Inflammatory cell responses to methacrylic acid beads ..................................................1

Introduction .................................................................................................................................1

1.1 The need for vascularization in regenerative medicine .......................................................1

1.2 Host response to biomaterial implantation ..........................................................................2

1.2.1 Neutrophils ...............................................................................................................3

1.2.2 Monocytes/Macrophages .........................................................................................3

1.2.3 Foreign body giant cells ...........................................................................................4

1.3 Role of macrophages in healing and vascularization ...........................................................5

1.4 Vascularizing biomaterials...................................................................................................7

1.5 Methacrylic acid-containing materials .................................................................................7

1.6 Objectives ............................................................................................................................9

Materials and Methods ..............................................................................................................10

2.1 MAA and MM bead preparation........................................................................................10

2.2 Subcutaneous injection animal model ...............................................................................10

2.3 Histology and immunohistochemistry ...............................................................................11

2.4 Tissue explant and digestion ..............................................................................................12

2.5 Analysis of cellular infiltrate in explanted tissues .............................................................12

2.6 CLARITY preparation and imaging ..................................................................................13

2.7 Statistical Analysis .............................................................................................................14

Results .......................................................................................................................................15

v

3.1 Subcutaneous injection model ...........................................................................................15

3.2 Effect of MAA beads on vascularization ...........................................................................15

3.3 Cellular response to MAA beads .......................................................................................17

3.3.1 Effect of MAA beads on the inflammatory cell infiltrate ......................................17

3.3.2 Effect of MAA beads on macrophage polarization ...............................................19

3.4 Interrogating biomaterial-cell interactions in intact tissues ...............................................21

3.4.1 Effect of MAA beads on CD206 expression in surrounding macrophages ...........22

Discussion .................................................................................................................................24

4.1 Effect of MAA beads on vessel formation ........................................................................24

4.2 Effect of MAA beads on the inflammatory cell infiltrate ..................................................24

4.3 Effect of MAA beads on macrophage polarization ...........................................................26

4.4 Insights into MAA-mediated macrophage polarization using CLARITY .........................27

Conclusion ................................................................................................................................30

Chapter 2 Role of complement activation in MAA-mediated macrophage polarization ..............31

Introduction ...............................................................................................................................31

1.1 Protein-biomaterial interactions in the host response ........................................................31

1.2 Mechanisms of macrophage recruitment and polarization ................................................33

1.2.1 Neutrophils .............................................................................................................34

1.2.2 Complement proteins .............................................................................................34

1.2.3 IGF signaling pathway ...........................................................................................34

1.3 Biomaterial strategies for mediating macrophage polarization .........................................35

1.4 Complement modulating effects of MAA .........................................................................35

1.5 Objectives ..........................................................................................................................37

Methods .....................................................................................................................................38

2.1 Preparation of poly(methacrylic acid-co-isodecyl acrylate) films .....................................38

2.2 Isolation, culture, and characterization of bone marrow-derived monocytes ....................38

vi

2.3 Macrophage stimulation by biomaterials in vitro ..............................................................39

2.4 CH50 type hemolysis assays ..............................................................................................39

2.5 Complement drug inhibition study ....................................................................................40

2.6 Tissue explant and digestion ..............................................................................................40

2.7 Analysis of cellular infiltrate in explanted tissues .............................................................41

2.8 Statistical Analyses ............................................................................................................41

Results .......................................................................................................................................42

3.1 Investigating the mechanism of MAA-mediated macrophage polarization ......................42

3.1.1 In vitro analysis of BMDM treated with MAA beads and films ...........................42

3.2 Inhibition of serum-derived complement and its effect on MAA ......................................42

3.2.1 Effect of complement inhibition on the vascular regenerative properties of

MAA ......................................................................................................................43

3.2.2 Effect of complement inhibition on MAA-mediated inflammatory cell

infiltration ..............................................................................................................45

3.2.3 Effect of complement inhibition on MAA-mediated M2 macrophage

polarization ............................................................................................................46

Discussion .................................................................................................................................50

4.1 Role of complement inhibition in MAA-mediated vascularization ...................................50

4.2 Role of complement inhibition in MAA-mediated alternative host response ...................51

4.3 Role of complement inhibition in MAA-mediated M2 macrophage polarization .............52

4.4 Insight into the mechanism of vascular regenerative MAA beads ....................................54

Conclusion ................................................................................................................................55

References ......................................................................................................................................57

Appendices .....................................................................................................................................64

vii

List of Figures

Fig. 1. The host response to biomaterial implantation.

Fig. 2. Role of macrophages in the host response and vascularization.

Fig. 3. Subcutaneous injection mouse model.

Fig. 4. MAA beads induced formation of perfusable vessels when injected subcutaneously.

Fig. 5. No differences in the density of F4/80+ cells between MAA- or MM- treated animals.

Fig. 6. Treatment with MAA beads altered the inflammatory cell landscape.

Fig. 7. Treatment with MAA beads biased macrophages towards a M2 polarization state.

Fig. 8. More CD206+ macrophages are found in the vicinity of MAA beads relative to MM

beads.

Fig. 9. Drug-induced inhibition of complement activation.

Fig. 10. Administration of pentamidine and ATA inhibited complement activation.

Fig. 11. Inhibition of complement activation did not affect the vascular potency of MAA.

Fig. 12. Complement inhibition eliminated MAA’s neutrophil recruitment effect.

Fig. 13. Complement inhibition altered the MAA-mediated effects on M2 macrophage

polarization.

Fig. 14. Effect of MAA beads on macrophage polarization, vascularization and the role of

complement activation.

viii

List of Appendices

S1. Gating strategy for macrophages (day 3 shown; MAA beads).

S2. Markers used for immunohistochemistry and flow cytometry analyses and definitions.

S3. Explant mass, cell number, and normalized cell number for flow cytometry analyses.

S4. Leukocytes, endothelial and dendritic cell populations in explanted tissues.

S5. Expression of CD206, CD86, and MHCII in bone marrow-derived macrophages polarized by

IFNγ and IL-4.

S6. Macrophage polarization - single positive cells.

S7. Formation of giant-like cells in vitro.

S8. Gating strategy for validating dextran uptake in CD206+ macrophages.

S9. Bone marrow harvest, macrophage culture and treatment with MAA beads or films.

S10. Gating strategy for macrophages following complement inhibition (day 7 shown; MAA

beads).

S11. MAA beads increased CD206, but not MHCII expression in the presence of blood.

S12. MAA films stimulated M2 marker Arg1 in M0 and M(IFNγ) cells.

S13. Administration of 4 mg/kg pentamidine or 2.5- 10 mg/kg ATA did not inhibit complement

activation over time.

S14. Explant mass, cell number and normalized cell number in complement-inhibited animals.

S15. Leukocytes and macrophage populations in complement-inhibited animals.

S16. Published manuscript: Lisovsky A, Zhang, DKY, Sefton MV, Biomaterials 2016.

S17. Curriculum vitae.

ix

List of Abbreviations

ATA: aurin tricarboxylic acid

ATP: adenosine triphosphate

Cx,y: (number) complement component, subunit (number)

CDxx: cluster of differentiation (number)

CLARITY: Clear Lipid-exchanged Acrylamide-hybridized Rigid Imaging / Immunostaining /

in situ-hybridization-compatible Tissue hYdrogel

CSF: colony stimulating factor

DAMPS: danger associated molecular patterns

DNA: deoxyribonucleic acid

ECM: extracellular matrix

FBGC: foreign body giant cells

FBS: fetal bovine serum

GSL/BSL: Griffonia (Bandeiraea) Simplicifolia lectin

HUVEC: human umbilical vein endothelial cell

IFNγ: interferon gamma

IL-x: interleukin (number)

LAL: limulus amebocyte lysate

Ly6G: lymphocyte antigen 6 complex, class G

MAA: methacrylic acid

MAA beads: poly (methacrylic acid-co-methyl methacrylate) beads

MCP: monocyte chemoattractant protein

MHC: major histocompatibility complex

MM: methyl methacrylate

MM beads: poly (methyl methacrylate) beads

NOS: nitric oxide synthase

PBS: phosphate buffered saline

PDGF: platelet-derived growth factor

PEG: polyethylene glycol

TGF: transforming growth factor

TNF: tumor necrosis factor

VEGF: vascular endothelial growth factor

1

Chapter 1 Inflammatory cell responses to methacrylic acid beads

Introduction

Biomaterials are substances designed to interface with biological systems[1]. The past half-

century represents a “biomaterials revolution”; advancements in the development of biomaterials

for drug delivery (e.g., microcapsules, tablets), surgery (e.g., sutures, adhesives), and implants

(e.g., prostheses, vascular grafts) are promised to innovate modern medicine[2]. While

significant progress has been made in the development of “interesting” biomaterials[3], there

remains an incomplete understanding of the interactions between an implanted material and

biological tissues. The conventional interpretation of these interactions begins at the protein-

adsorption level. An adsorbed layer of proteins dictates changes in cell behavior and the

activation of blood-derived pathways. Together, these interactions translate into inflammation,

vascularization, and ultimately, tissue remodeling [4–6]. Here, we explore the biological

interactions between methacrylic acid (MAA)-based biomaterials, the host response and

vascularization. These materials have a vascular regenerative effect in vivo [7], through an

unclear mechanism. The present chapter evaluates the host response to MAA-containing

polymeric beads and aims to define a role for macrophage polarization in this context. Much of

this chapter has been published (See Appendix); the introduction, discussion, and parts of the

results have been expanded for the purpose of this thesis. Chapter 2 pursues mechanistic

questions, with a focus on complement activation, and aims to connect the events that occur

immediately following biomaterial implantation to the changes in the host response (i.e.,

inflammatory cell infiltration) and vascularization. Understanding of the mechanisms behind

MAA-mediated vascularization may afford the ability to control and dictate the biological

response to similar materials.

1.1 The need for vascularization in regenerative medicine

A perfusable network of blood vessels is vital for regenerative medicine[7,8]. In cell therapy, the

transplantation of therapeutic cells requires a vascularized network to ensure that nutrients and

oxygen are aptly delivered[8]. In the context of tissue regeneration, the development of a

vascular network stimulates endogenous repair by delivering growth factors to the site of

injury[9]. To meet this need for vascularization, a number of approaches have been devised[8].

2

However, most strategies employ the synergistic addition of vascular support cells (e.g.,

mesenchymal stromal cells)[10] or growth factors (e.g., VEGF)[11,12], leading to complicated

and costly treatments that are difficult to translate to the clinic[13]. This poses a unique

opportunity for the development of alternative, scalable and more cost-effective strategies (i.e.,

biomaterials) to address this need. Biomaterials that promote vascularization without the co-

delivery of cells or proteins would be highly advantageous, as they would be cost-effective and

easy to manufacture. Studies in wounded diabetic mice revealed that materials containing

methacrylic acid (MAA) have vascular regenerative properties[14–17]. The aim of this thesis is

to investigate the mechanism behind this beneficial effect at the cellular and molecular levels.

1.2 Host response to biomaterial implantation

One prominent issue with the use of biomaterials for improving vascularization is the host

response or foreign body response[6]. In the process of biomaterial implantation, cells and

tissues are inevitably damaged, setting the stage for the multitude of interactions collectively

known as the host response[18]. The host response is a generic biological response that begins

with inflammation and ends with fibrosis or tissue reconstitution and healing [5]. Following

biomaterial implantation, tissue-resident cells (e.g., tissue-resident macrophages) and blood-

derived proteins (e.g., complement) detect the presence of the foreign material indirectly via

damage-associated molecular patterns (DAMPS), such as cytoplasmic proteins (e.g., ATP, DNA,

uric acid, etc.) released from dying cells, or directly, via the non-specific adsorption of proteins

to the biomaterial itself[6,19,20]. The adsorbed proteins are dynamically changing and are

thought to dictate the outcome of the host response[5]. Concomitantly with protein adsorption,

damage to blood vessels initiates thrombosis and the formation of a fibrin clot; a process

involving platelets, the complement system, the fibrinolytic system, and others (reviewed in

[21]). Together, these processes facilitate the formation of a provisional matrix; a rich, dynamic

ecosystem of chemokines, cytokines, and growth factors that modulates cell activation and

proliferation in the inflammatory and healing phases of the host response[6]. Following

provisional matrix formation, inflammatory cues (e.g., IL-1β, TNF-α, and others) derived from

various sources including de-granulated mast cells, are released from the matrix into the blood

stream to signal the recruitment of innate immune cells (e.g., neutrophils, monocytes,

lymphocytes) to the site of the foreign material or the site of inflammation[5].

3

1.2.1 Neutrophils

The cell type dominating the host response is time-dependent[22]. Neutrophils are the hallmarks

of the inflammatory response and traditionally the first responders to a site of injury[23] (Fig. 1).

Neutrophils are short-lived cells (24-48h) whose primary function is to remove and contain the

spread of foreign particles (e.g., pathogens or debris, reviewed in [22]). Being myeloid-derived

cells (i.e., cells that express CD11b; cluster of differentiation 11b, an integrin associated with

leukocyte adhesion), they also express Ly6G (lymphocyte antigen 6 complex, class G), a GPI-

linked differentiation antigen at varying levels corresponding to their maturity[24]. Once the task

of “quarantining” foreign particles from the rest of the body is completed, neutrophils become

apoptotic, inhibiting further neutrophil recruitment while promoting monocyte recruitment and

their subsequent differentiation to macrophages (reviewed in [25–28]). This feedback loop

enables apoptotic neutrophils to be phagocytosed by growing numbers of macrophages that have

infiltrated the site of inflammation.

1.2.2 Monocytes/Macrophages

Monocytes responding to gradients of granulocyte (e.g., neutrophil, eosinophil, basophil) –

derived chemokines (e.g., MCP-1, IL-1, etc.) hone in to the site of the foreign material[22]. Once

these myeloid-derived innate immune cells leave the blood vessel, their differentiation to

macrophages is triggered, as noted by an upregulation in the expression of F4/80 (an adhesion G-

coupled protein receptor associated with peripheral T cell tolerance)[29]. A positive feedback

loop propagates further secretion of chemokines, such as granulocyte colony stimulating factor

(G-CSF), promoting more macrophage infiltration. At the site of the biomaterial, macrophages

serve multi-faceted roles and link the inflammatory and healing phases of the host

response[30,31]. The literature suggests that there are two distinct subsets of macrophages

(termed M1 and M2); however, this represents an oversimplification of a complex spectrum of

macrophage polarization states[32]. Macrophages are highly multi-functional and the M1/M2

classification is an in vitro artifact that represents the ends of a spectrum of phenotype and

function[33].

Following tissue infiltration and activation, macrophages promote inflammation by secreting

cytokines such as IL-1β and TNF-α – a property characteristic of the M1 polarization state (Fig.

2A). Later in the host response, macrophages secrete IL-10 and TGF-β, paving the road for the

4

resolution of inflammation – a property characteristic of the M2 polarization state (Fig. 2A). It is

accepted that macrophages shift from the M1 to the M2 phenotype 48-72 h post-injury,

coordinating the transition from the inflammatory phase to the healing phase of the host

response[34,35]. Consistent with this idea, studies involving fluorescently labeled macrophages

revealed that a part of the population of M2 macrophages that arises later in the host response is

derived directly from the original M1 macrophage population at the site of the foreign

material[36,37].

1.2.3 Foreign body giant cells

At later time points (several days), adherent macrophages on the surface of a biomaterial fuse to

form foreign body giant cells (FBGCs)[5]. The shift in macrophage polarization from M1 to M2

is expected; studies from J. Anderson et al indicated that FBGC formation requires IL-4 and IL-

13, agonists of the M2 phenotype[38]. However, the genomic and proteomic expression profile

of FBGC is distinct from M2 macrophages[38], highlighting the FBGC as a distinct phenotype.

FBGC formation is in a sense a cellular stress response to large foreign bodies and is a

conventional response to biomaterial implantation. Although macrophages are capable of

phagocytosing small particles (<5 μm), once they encounter a larger particle (>10 μm), they fuse

to increase their combined surface area and corresponding phagocytic potential[38]. If the

newly-formed FBGCs are unable to phagocytose the foreign material, they remain at the

biomaterial-host environment and attempt to degrade the foreign material instead via the

secretion of matrix metalloproteinases (MMPs), protons, and reactive oxygen species (ROS)[6].

Thus, FBGCs form an isolated degradative environment that may lead to 1) biomaterial

resorption and the resolution of the host response[6], if the biomaterial is degradable or 2)

persistent inflammation and evidence of chronic inflammation[5,39], if the biomaterial is not

degradable.

Successful tissue regeneration is associated with a milieu of anti-inflammatory mediators,

downregulation of inflammatory mediators and the apoptosis of immune cells; which

collectively mediates the resolution of inflammation[28]. FBGC that have failed to degrade the

foreign material secrete pro-fibrotic factors, such as TGF-β, recruiting and activating fibroblasts

to deposit collagen and remodel the extracellular matrix[5]; forming the underpinnings of the

5

fibrotic capsule[38]. Thus, biomaterial-adherent macrophages dictate the fibrotic response and

the formation of a fibrotic capsule, the conventional endpoint of biomaterial implantation.

Fig. 1. The host response to biomaterial implantation. Following implantation of a

biomaterial, a spike in neutrophils is observed at the site of the foreign material, followed by

macrophage infiltration and the beginnings of neovascularization (the formation of new vessels),

then the formation of foreign body giant cells, the recruitment of fibroblasts, and ultimately, the

formation of a fibrotic capsule. The y-axis (intensity) may be interpreted as the number of cells.

The scale of the x-axis (time) varies depending on the biomaterial; for most biomaterials, the

initial wave of neutrophils is resolved in 48h and fibrosis occurs several weeks after

implantation. Adapted from [18].

1.3 Role of macrophages in healing and vascularization

As an alternative to fibrosis, the host response can also prepare the ground for tissue regeneration

and vascularization[9]. Macrophages lie at the crossroads of inflammation, tissue regeneration

and vascularization[30]. The specific contributions of classically-activated (M1) and

alternatively-activated (M2) macrophages in the vascularization process are ill-defined; some

studies showed that lower ratios of M1/M2 macrophages improves vascularization[39], while

others have shown that increased M1/M2 ratios leads to more vascularization[40]. One

hypothesis claims that vascularization begins with classically-activated (M1) macrophages, a

potent source of vascular endothelial growth factor (VEGF), which initiates vessel sprouting in

responding endothelial cells[41,42] (Fig. 2B). Endothelial cells migrate and sprout outwards,

while secreting integrins and creating new extra-cellular matrix (ECM), forming a leaky and

immature vasculature.

6

Fig. 2. Role of macrophages in the host response and vascularization. (A) Blood-derived

monocytes infiltrate the site of inflammation, triggering their differentiation to macrophages.

Over time, macrophages shift from an inflammatory M1 to an anti-inflammatory M2 phenotype

directly at the site of inflammation. (B) Proposed roles of M1 and M2 macrophages in

vascularization. M1 macrophages secrete endothelial growth factors to initiate vessel sprouting

(via VEGF) while M2 macrophages promote vessel maturation by recruiting pericytes via

PDGF.

Eventually, as macrophages transition to the M2 polarization state, these alternatively-activated

(M2) macrophages aid in the degradation of the basal lamina by secreting metalloproteinases

(MMPs) to break down the surrounding extracellular matrix. These cells serve as chaperones for

endothelial cells by guiding tip cell anastomosis – the process of joining of the ends of two

newly formed vessels[11,43]. Additionally, M2 macrophages promote the maturation of vessels

through the recruitment of pericytes via the secretion of platelet-derived growth factor

(PDGF)[11]. Simply put, M2 macrophages mature the leaky vasculature formed by M1

macrophages, supporting the formation of a perfusable and mature vascular network[44].

7

Hematopoietic and mesenchymal-derived progenitor cells are also sources of VEGF and PDGF,

integrating themselves into vessels, among other tissue structures[45,46]. Fibroblasts also play a

role in laying down the foundation for vessels by synthesizing the extracellular matrix. In

summary, vascularization is a complex process that is intimately linked to the host response and

requires the synergistic contributions of both M1 and M2 macrophages.

1.4 Vascularizing biomaterials

Biomaterials have been used to deliver cells (e.g., vascular support cells; adMSC)[10] or growth

factors (e.g., VEGF). However, cell delivery strategies become increasingly complex with a

number of immunological barriers (e.g., inflammation, antigen-directed cytotoxic cell responses)

to overcome[47]. While the delivery of growth factors is simpler, it remains challenging to

temporally control growth factors to 1) initiate vessel formation and 2) mature the newly-formed

vessels. Attempts have been made to modulate the host response to benefit vascularization.

Madden et al showed that increasing porosity in acellular poly(2-hydroxyethl methacrylate)

[poly(HEMA)] scaffolds promoted cellular infiltration and facilitated vascularization[42]. They

attributed this beneficial effect to macrophage polarization; macrophages recruited to the

poly(HEMA) scaffold were polarized towards a “healing” phenotype, characterized by an

increase in the expression of CD206. Stupp et al used a bio-inspired approach to develop peptide

nanostructures that mimicked VEGF activity. The VEGF-like peptides bound to VEGF receptors

and initiated vessel sprouting in endothelial cells[48]. Biomaterials that have the ability to alter

the cellular landscape to promote regeneration have the potential to compete and replace

cell/protein-based strategies of vascularization.

1.5 Methacrylic acid-containing materials

Methacrylic acid (MAA)-based biomaterials were shown to have a vascular regenerative effect

in the absence of exogenous cells or growth factors[7]. These biomaterials promoted

vascularization[14,16,49], myocutaneous graft survival[14], and diabetic wound healing[16]. In

vitro studies[50,51] involving human endothelial cells (i.e., HUVEC) did not alter the expression

of classical angiogenic genes (i.e., VEGF)[51]. However, gene expression analysis revealed that

MAA modulated pleiotropic genes (i.e., Shh), pro-inflammatory genes (i.e., IL-1β, TNF-α), in

bone marrow-derived macrophages and macrophage-like cells (dTHP-1), as well as in diabetic

wounds and in an air pouch model[17,49,50]. More recently, a phosphoproteomics study with

8

dTHP-1 cells treated with MAA-based material highlighted a number of phosphorylated proteins

involved in macrophage polarization[52] among several hundred proteins that were differentially

phosphorylated between a MAA-based and a control (methyl methacrylate-based) material[53].

In the studies conducted to date, no change was observed in the number of infiltrating

macrophages between MAA-treated animals or the control MM-treated animals [17]. These data

led to the hypothesis that MAA elicited its vascular regenerative effect by modulating

inflammatory cell responses, specifically macrophage polarization.

Here, a subcutaneous injection model was devised to investigate the effects of MAA on the host

response and macrophage polarization. To this end, poly(methacrylic acid-co-methyl

methacrylate) (MAA) beads and control poly(methyl methacrylate) (MM) beads were injected

subcutaneously in male CD1 mice. The bead explants were processed for immunohistochemistry

and flow cytometry for the number of cells and the polarization state of macrophages. MAA

beads increased the density of neutrophils at day 1, macrophages at day 7 and biased

macrophages towards the MHCII-CD206+ state, representative of the “M2” phenotype.

9

1.6 Objectives

This thesis explores the interactions between methacrylic acid (MAA)-containing beads and the

host response. Chapter 1 investigates the effect of MAA beads on the inflammatory cell

response. As macrophages are known to be the orchestrators of vascularization, they were a

focus of the investigation. Macrophage polarization was studied using 1) flow cytometry to

evaluate global changes in macrophage phenotype in response to MAA bead implantation, and 2)

a 3D tissue imaging approach to interrogate local changes in macrophage polarization in the

immediate vicinity of MAA beads. Chapter 2 (Aim 2) investigates the role of complement

activation in MAA-mediated macrophage polarization and vascularization.

Aim 1A: Characterize the inflammatory cell infiltrate in animals injected subcutaneously with

MAA beads.

Hypothesis: Treatment with MAA beads alters the inflammatory cell landscape and alters

macrophage polarization relative to control MM beads.

Aim 1B: Interrogating MAA bead-cell interactions in intact tissues using CLARITY, a tissue

preparation protocol for 3D imaging of intact tissues.

Hypothesis: MAA beads polarize macrophages in its immediate vicinity (< 200 μm distance

from a cluster of MAA beads).

10

Materials and Methods

2.1 MAA and MM bead preparation

Poly(methacrylic acid-co-methyl methacrylate) (MAA-co-MMA or MAA) beads were composed

of 45 mol% methacrylic acid (Sigma-Aldrich Canada Ltd., Oakville, ON, Canada), 1 mol%

ethylene glycol dimethacrylate (Sigma-Aldrich Canada Ltd.) and 54 mol% methyl methacrylate

(Sigma-Aldrich Canada Ltd.). MAA beads were synthesized by suspension polymerization as

previously described and were sieved to obtain beads in the diameter range of 150-250 μm.

Methacrylic acid content of the synthesized beads was confirmed by titration. Control

poly(methyl methacrylate) (MM) beads (same diameter) were obtained from Polysciences

(Warrington, PA). Beads were washed in either 95% ethanol (MAA beads) or 1 N HCl (MM

beads) repeatedly and then rinsed five times in LAL reagent water (MJS Biolynx Inc.,

Brockville, ON, Canada) prior to use in vivo. Analysis with a limulus amebocyte lysate (LAL)

pyrochrome endotoxin test kit (Cape Cod Inc., Falmouth, MA) indicated that beads contained

<0.25 EU/100 mg. MAA beads had a rough, porous surface, were negatively charged and non-

degradable; MM beads were smooth and also not degradable.

For subcutaneous injections, a 1 mL syringe with an 18-gauge needle was loaded with either 5

mg MAA beads or 15 mg MM beads (or no beads, vehicle control) suspended in 250 μL of 50%

w/v polyethylene glycol (PEG, avg. mol. wt. 1450, sterile-filtered; Sigma-Aldrich Canada Ltd.)

in PBS. The 1:3 weight ratio (5 mg MAA: 15 mg MM) was used to account for MAA beads

swelling upon hydration at physiological pH to approximately equate implanted volumes. The

vehicle control was used only for flow cytometry analysis because the vehicle control implant

area could not be defined reproducibly for vessel and cell density analyses.

2.2 Subcutaneous injection animal model

Mice were anesthetized with 0.5% w/v isofluorane prior to surgery and an analgesic

(Ketoprofen, 5 mg/kg) was administered intraoperatively. The dorsal area of a mouse was shaved

and the remaining hair was removed by hair removal cream (Veet). The skin was sterilized with

70% ethanol and Betadine. An 18-gauge needle was used to inject MAA, control MM beads or

vehicle (PEG). Two injections on either side of the dorsum were performed for each mouse. A

small subcutaneous pocket was made with the needle on the side of the dorsum by moving the

11

syringe from side to side, while deliberately attempting to nick small blood vessels to promote

injury prior to injection (Fig. 3). Following surgery, mice were housed individually, fed chow

and water ad libitum, and monitored for any signs of discomfort. At 1 to 7 days post-injection,

the mice were sacrificed using CO2, followed by cervical dislocation. The implants were

removed surgically and processed for histology, imaging or flow cytometry. All animal work

was done with the approval of the University of Toronto Animal Care Committee. Animals were

housed under sterile conditions in the University of Toronto’s Department of Comparative

Medicine (AUP #20010994 and #20013994).

Fig. 3. Subcutaneous injection mouse model. CD1 mice were injected subcutaneously with

MAA or MM beads. At days 1, 3, and 7 post-implantation, the beads and surrounding tissues

were excised and processed for histological, imaging, and flow cytometry analyses. The

molecular analyses panel is included as it is a future possibility of this model; molecular analysis

was not conducted in the present study. In addition, this model may be used to explore biological

responses to other biomaterials.

2.3 Histology and immunohistochemistry

Immediately upon euthanizing mice, the bead implant and several mm of surrounding tissue was

excised from the right side of the dorsum and fixed in formalin. Tissue samples were embedded

in deep paraffin blocks, cut into sections, processed and stained with hematoxylin and eosin

(H&E), Masson's trichrome, CD31 and F4/80 (Appendix, S2). Histology slides were scanned

(20x) using an Aperio ScanScope XT (Leica Microsystems, Concord, ON, Canada) by the

Advanced Optical Microscopy Facility (AOMF, Toronto, ON, Canada).

12

The scanned slides were analyzed using Aperio ImageScope (Version 11) at 3 and 7 days post-

implantation. Vessel and cell quantitation was performed by first defining a region of interest

(ROI). For vessel counts, the ROI was defined by measuring a distance of 500 μm around a

hotspot (a clump of beads with CD31+ vessels in its ROI; some clumps of beads had no vessels

in its ROI). CD31+ vessel-like structures (criterion being the presence of a lumen) were counted

in the tissue within this defined region. The vessel density was calculated by dividing the total

number of vessels by area of the ROI. For F4/80+ cell counts, a distance of 200 μm around each

bead cluster was used to define the ROI.

2.4 Tissue explant and digestion

Subcutaneous tissue containing the injected beads was separated from the skin and muscle

layers. For PEG samples, subcutaneous tissue was explanted in the same manner using the

injection needle wound site as a guide. Tissues were weighed and then digested following a

previously described digestion protocol [54]. Briefly, samples were finely minced in 500 µL of 1

X HBSS containing 450 U/mL collagenase I (Sigma-Aldrich Canada Ltd.), 125 U/mL

collagenase XI (Sigma-Aldrich Canada Ltd.), 60 U/mL DNase I (Sigma-Aldrich Canada Ltd.),

60 U/mL hyaluronidase (Sigma-Aldrich Canada Ltd.) and 20 mM HEPES (Sigma-Aldrich

Canada Ltd.). The samples were homogenized using a gentleMACS Octo Dissociator (Miltenyi

Biotec Inc., San Diego, CA). The tissues were further digested for 60 min at 37 °C and 250 rpm.

The cell suspension was filtered using a 40 μm cell strainer (Fisher Scientific, Ottawa, ON,

Canada) to remove beads and debris. The remaining cells were washed in PBS supplemented

with 0.5% BSA and 2 mM EDTA, pelleted and stained with live/dead stain, CD11b, CD206,

CD11c, CD31, CD45, CD86, F4/80, Ly6G, and MHCII (Appendix, S2). All antibodies were

diluted according to the manufacturers’ recommendations and titrated in-house to optimize

staining.

2.5 Analysis of cellular infiltrate in explanted tissues

The gating strategy (Appendix, S1) was as follows: after isolating live single cells, CD45

distinguished leukocytes from non-leukocytes. Neutrophils were identified as Ly6G+ and

dendritic cells as CD11c+Ly6G-. Macrophages were first identified as CD11c-Ly6G-

F4/80+CD11b+ and then further characterized as MHCII+CD206- (“M1”) and MHCII-CD206+

(“M2”). Endothelial cells were identified as CD31+CD45-. Cells were gated according to

13

positive staining for each antibody using fluorescence minus one (FMO) controls. Cell

populations were expressed as either a percentage or as a normalized value (estimated total

number of cells divided by the weight of the explanted tissue). The number of cells was

estimated from the flow cytometry results with 123count eBeads (eBioscience) used to determine

cell recovery (~50%).

2.6 CLARITY preparation and imaging

Seven days following subcutaneous injection of MAA or MM beads, Cy5-conjugated dextran

(70kDa; 100 μg in 150 μL PBS; Chan lab) was injected via tail vein. Dextran is the ligand for the

CD206 scavenger receptor[55]. After 30 min of circulation, Alexa 555-conjugated lectin (GSL-

1: Griffonia (Bandeiraea) Simplicifolia; 100 μg in 150 μL PBS; Vector Laboratories, Burlington,

ON, Canada) was injected via tail vein prior to sacrifice and whole body perfusion with PBS-

heparin[56]. Fluorophore conjugation was performed in-house using Alexa 555 or Cy5 modified

with a NHS-ester chemistry[57]. GSL-1 is a lectin which binds to the galactosyl residues of

mouse endothelial cells, enabling labeling and visualization of the mouse vasculature[58]. Earlier

experiments were performed without the initial injection of Cy5-conjugated dextran. The

implants with the surrounding subcutaneous tissue were removed surgically and processed using

a modified CLARITY protocol developed by Sindwani, S. et al [56,59]. Briefly, explants were

fixed in a solution containing 2% acrylamide (Sigma-Aldrich Canada Ltd.), 4%

paraformaldehyde (Sigma-Aldrich Canada Ltd.) and 0.25% (w/v) VA-044 thermal initiator

(Sigma-Aldrich Canada Ltd). After one week of incubation, the acrylamide was polymerized at

37 °C for 1-3 h. Polyacrylamide-embedded explants were cleared for 14 days at 50°C in clearing

solution (8% SDS in borate buffer, pH 8.5; eBioscience, San Diego, CA), which was changed

every 2nd day. Post-clearing, the explants were counterstained with SYTOX green nucleic acid

stain (100 pmol/mg; Life Technologies, Burlington, ON, Canada) or DAPI nucleic acid stain

(200 pmol/mg; Life Technologies, Burlington, ON, Canada) for 48 hours. Refractive index

matching was performed by infusing explants with 70% 2,2’-thiodiethanol[56] in borate (Sigma-

Aldrich Canada Ltd.) for confocal microscopy. Explants were imaged using a Nikon A1 confocal

microscope (Nikon, Melville, NY) at the Center for Microfluidics Systems (University of

Toronto).

14

2.7 Statistical Analysis

Statistical analysis was performed using GraphPad PRISM 6.0. Data is represented as mean ±

standard error of mean (SEM). A two-way ANOVA was used to compare treatment groups over

the 3 time points. Tukey’s post hoc test was used to determine significance of multiple

comparisons. A p-value of less than 0.05 was considered significant.

15

Results

3.1 Subcutaneous injection model

Previously, the vascular potency of MAA was explored in cutaneous wounded diabetic db/db

mice, precluding the analysis of the inflammatory cell infiltrate, as recruited cells became

entrapped in scabs. The less-invasive subcutaneous injection model obviated this issue and

enabled the direct interrogation of cells that were associated with MAA beads (See Fig. 3).

Additionally, the subcutaneous model simplified the host response, as there was less of a

physiological need for wound healing and the complexities of diabetic wound healing were

removed. MAA beads were subcutaneously injected in the dorsal flank of male CD1 mice at two

sites; one implant was harvested for flow cytometry analysis while the other was processed for

histological analysis.

3.2 Effect of MAA beads on vascularization

MAA beads enhanced vascularization in the tissue directly surrounding the beads (< 500 μm

from a cluster of MAA beads) following subcutaneous injection. CD31+ vessel formation was

increased at days 3 and 7, relative to MM beads (Fig. 4A, B), validating the vascular

regenerative effect of MAA in the subcutaneous injection model. To confirm that the MAA-

induced vessels were perfusable, animals were injected with Alexa 647-conjugated mouse lectin

(GSL1), to visualize blood vessels in the immediate vicinity of the beads (Fig. 4C). A modified

CLARITY protocol was used to increase the depth of imaging. This was the first application of

CLARITY to visualize biomaterial-cell interactions, to our knowledge. MAA-treated animals

showed high levels of GSL1 staining around MAA beads, consistent with the greater CD31+

vascularity observed in the histological analysis (Fig. 4C). On the other hand, MM-treated

animals showed minimal or no lectin staining surround MM beads. Instead, lectin staining was

primarily concentrated to the panniculus carnosus of the skin layer (Fig. 4C).

16

Fig. 4. MAA beads induced formation of perfusable vessels when injected subcutaneously.

(A) Histology sections of animals treated with MAA or MM beads at day 7 stained with CD31

(left) and Masson’s trichrome (right). Arrows indicate examples of vessels. (B) Tissues treated

with MAA beads in mice had a significantly higher vessel density at day 7. (C) Confocal

microscopy image of CLARITY-processed tissues treated with MAA and MM beads from non-

transgenic CD1 mice stained with Alexa 647-GSL1, a lectin specific for the mouse endothelium,

and Sytox Green. Perfused vessels weaved around MAA beads but not MM beads. Most of the

vessels in MM-treated mice were found in the skin further away from the beads. Scale bars = 200

μm. n = 3-4.

In CLARITY processed tissues, a thick layer of cells was found surrounding MM, but not MAA

beads, suggesting a differential cellular response to the MAA beads. This dense layer of cells

was also observed in trichrome-stained sections (Fig. 4A), indicating the presence of a

conventional host response to biomaterial implantation. Together, these data suggested that the

MAA beads induced vascularization when injected subcutaneously and that the MAA-induced

vessels were perfusable.

17

3.3 Cellular response to MAA beads

As macrophages play a vital role in vascularization[39], the effect of MAA on macrophages was

investigated in histological sections using the pan macrophage marker F4/80. A dense ring of

F4/80+ cells was found surrounding MM beads; but rarely surrounding MAA beads (Fig. 5A),

similar to that seen in trichrome-stained sections and CLARITY-processed tissues (Fig. 4A, C).

The thick ring of cells resembled foreign body giant cells. No differences were observed in

F4/80+ cell density between MAA or MM beads at day 3 or day 7 (Fig. 5B).

Fig. 5. No difference in the

density of F4/80+ cells

between MAA- or MM-

treated animals. (A) Tissue

sections from MAA- and

MM-treated mice stained

with pan macrophage F4/80

marker at day 3. A dense ring

of F4/80+ cells

(macrophages) surrounded

control MM beads. Arrows

show examples of cells

positive for the marker of

interest. (B) Density of

F4/80+ cells in tissues

following treatment with

MAA and MM beads. Scale

bars = 200 μm. n = 3-4.

3.3.1 Effect of MAA beads on the inflammatory cell infiltrate

Flow cytometry was used to follow up on the histological analysis. An extra time point (day 1)

was added to investigate the host response (i.e., neutrophils) immediately following

subcutaneous injection of MAA and MM beads. Also, a PEG vehicle treatment was added. No

statistical difference was noted between the mass of the explanted tissues, or the estimated total

cell number, or the normalized cell numbers among the three treatment groups (MAA beads,

MM beads, and PEG vehicle) at the studied time points (Appendix, S3). The gating strategy

employed is illustrated in Appendix, S1. Higher densities of CD45+ cells were found in the

harvested MAA implants relative to the PEG vehicle control at day 1, with a corresponding

18

higher density of CD45- non-leukocytes in the PEG vehicle implant (Fig. 6A, B; Appendix, S4).

Interestingly, a progressive increase in CD45- cells were noted in tissues treated with MM

relative to MAA beads at day 7 (Fig. 6B).

Fig. 6. Treatment with MAA beads altered the inflammatory cell landscape. (A-D) Number

of CD45+ leukocytes (A), CD45- non-leukocytes (B), Ly6G+CD11b+CD45+ neutrophils (C)

and F4/80+CD11c-Ly6G-CD11b+CD45+ macrophages (D). MAA beads significantly increased

the number of CD45+ cells (A) and neutrophils (C) at day 1 and macrophages at day 7 (D), while

decreasing the number of CD45- cells at day 7 (B). (E) F4/80 and CD11b expression in CD45+

cells at day 1 and day 7; note the F4/80 mean fluorescent intensity increased over time. The

F4/80 and CD11b gate (black box) was set based on fluorescence minus one (FMO) negative

controls. n = 3-4.

In the biomaterial treatment groups (MAA and MM), neutrophils were most prevalent at day 1

post-injection and their numbers dwindled by days 3 and 7. Treatment with MAA beads

increased the number of neutrophils relative to MM and PEG controls at day 1 (Fig. 6C). More

macrophages were found in explants harvested from MAA-treated mice at day 7, relative to

MM- and PEG- treated animals. Indeed, the estimated number of macrophages decreased from

day 3 to day 7 in MM- and PEG-treated animals; while the number remained unchanged in

MAA-treated animals (Fig. 6D). Additionally, the intensity of F4/80 expression varied greatly

from day 1 to days 3 and 7, suggesting that macrophages “matured” in the subcutaneous

injection site. Endothelial and dendritic cells were also quantified by flow cytometry, although

19

no significant differences were noted for the density of endothelial cells among the three

treatment groups (Appendix, S4). The frequency (as a % of Ly6G-CD45+ cells) of dendritic

cells increased in the material treatment groups, relative to PEG vehicle at day 7 (Appendix, S4).

Together, the data suggests that MAA altered the inflammatory cell response compared to MM

and PEG controls, leading to an increase in the recruitment of neutrophils at day 1 and the

number of macrophages at day 7.

3.3.2 Effect of MAA beads on macrophage polarization

Next, the flow cytometry protocol was used to distinguish macrophage polarization states.

MHCII and CD86 were used as markers for “M1”, classically-activated macrophages, while

CD206 was used as a marker for “M2”, alternatively-activated macrophages. These markers

were validated with bone marrow-derived macrophages stimulated with IFNγ and IL-4 for “M1”

and “M2” macrophages, respectively (Appendix, S5). In the subcutaneous injection model, the

expression of CD86 did not change significantly between treatment groups at any of the studied

time points; it was dropped from further analysis. There were some significant differences in the

expression of MHCII and CD206, with a general trend involving more CD206+ expression in

MAA-treated animals and more MHCII+ expression in MM-treated animals (Appendix, S6).

20

Fig. 7. Treatment with MAA beads biased macrophages towards a M2 polarization state.

(A, B) Representative dot plot of F4/80+ cells (macrophages) at day 3 in mice treated with MAA

(A) and MM beads (B). (C, D) The number and frequency of the individual single positive,

double positive, and double negative MHCII or CD206 macrophage populations in mice treated

with MAA beads, MM beads or PEG vehicle control. (C) Normalized number and frequency of

MHCII-CD206+ (“M2”) macrophages. (D) Normalized number and frequency of

MHCII+CD206- (“M1”) macrophages. MAA beads biased macrophages towards a M2

polarization state; noted by a decrease in M1 macrophages and an increase in M2 macrophages,

compared to MM beads. (E, F) Distribution of polarized macrophages: normalized number (E)

and frequency (F) of macrophages that were MHCII-CD206+, MHCII+CD206+,

MHCII+CD206-, and MHCII-CD206-. n = 3-4.

The polarization bias was reflected in the representative dot plots for MAA vs MM beads (Fig.

7A, B). Treatment with MAA beads led to significantly more MHCII-CD206+ (M2)

macrophages and decreased MHCII+CD206- (M1) macrophages, relative to MM beads at day 7

21

(Fig. 7C). On the contrary, treatment with MM beads had the opposite effect, with significantly

more M1 and fewer M2 macrophages relative to MAA beads at day 7 (Fig. 7D). In PEG-treated

animals, the number of M1 and M2 macrophages remained steady over the three time points, as

expected. By day 7, the majority of macrophages in MM-treated mice were double-positive

(MHCII+CD206+) or double-negative (MHCII-CD206-) (Fig. 7E, F). Interestingly, the number

of double positive macrophages increased progressively from day 1 to day 7 in MM- but not

MAA-treated animals (Fig. 7E). Conversely, in MAA-treated mice, macrophages were

consistently MHCII-CD206+ (M2) from day 3 onwards to day 7 (Fig. 7E, F). In MM-treated

mice, the progressive increase in MHCII+CD206- (M1) macrophages suggested that

macrophages may have been fusing to form foreign body giant cells [38]. Consistent with this

observation, BMDM-induced fusion using IL-4 in vitro formed large cells that resembled foreign

body giant cells (FBGCs) (Appendix, S7); these cells were MHCII+ and MHCII+CD206+,

suggesting that FBGCs expressed MHCII.

3.4 Interrogating biomaterial-cell interactions in intact tissues

Next, the spatial orientation of M2 macrophages relative to MAA beads and blood vessels was

investigated. Alexa 647-conjugated dextran (70kDa), the ligand for the CD206 receptor[58], was

injected into MAA- or MM- treated animals 30 min prior to injection of the Alexa 555-

conjugated lectin (GSL1) to visualize cells that expressed the CD206 scavenger receptor and to

label blood vessels, respectively. A flow cytometry strategy was devised to evaluate the cells that

associated with the lectin (Appendix S8). Two gating strategies were employed: 1) Gating first

for macrophages, then dextran positive cells, and 2) gating first for dextran positive cells, then

dextran positive macrophages (Fig. 8A, Appendix S8). Both gating strategies produced

comparable results. Approximately 90% of dextran+ cells were macrophages

(F4/80+CD11b+Ly6G-CD45+ cells) (Fig. 8B). Of the dextran+ macrophages, approximately

90% were CD206+, while approximately 40% were MHCII+ (Fig. 8C). Thus, of all cells that

were associated with dextran, about 90% × 90% = 88% were CD206+ while 90% × 40% = 36%

were MHCII+. Hence, dextran+ cells were considered CD206+ (M2-like) macrophages. There

was no difference in the percentage of dextran+ cells between MAA- or MM- treated animals

(Fig. 8C). Unlike the previous definition of M1 (MHCII+CD206-) or M2 (MHCII-CD206+)

macrophages, the dextran label was unable to distinguish macrophages that were MHCII-

CD206+ or MHCII+CD206+.

22

3.4.1 Effect of MAA beads on CD206 expression in surrounding macrophages

Dextran+ macrophages were localized to the immediate vicinity (<200 μm) of vessels (Fig. 8D)

regardless of treatment with MAA or MM beads. However, in MAA-treated animals, dextran+

macrophages were found in the immediate vicinity (<200 μm) of MAA beads, even in the

absence of vessels (Fig. 8D, bottom). In MM-treated animals, dextran+ macrophages were

located further away from MM beads and in areas with lectin staining (vessels). In agreement

with previous observations (Fig. 4, 5), a dense layer of cells (presumably F4/80+ based on the

histological analyses, Fig. 5) were found surrounding MM, but not MAA beads.

Random slices were selected from each image stack and the density of dextran+ macrophages

were quantified (Fig. 8E). As expected, the number dextran+ macrophages surrounding MAA

beads was higher relative to MM beads. Notably, the cells adhered to MM beads were not

dextran+, suggesting that they were not M2 macrophages. Together, these results suggested that

MAA beads biased macrophages towards the M2 polarization state, supporting the flow

cytometry results. Additionally, these observations highlighted the potential of CLARITY to be

used for the direct interrogation of cell-biomaterial interactions in explanted tissues.

23

Fig. 8. More CD206+ macrophages are found in the vicinity of MAA beads relative to MM

beads. (A) Representative flow cytometry gating strategy to determine dextran uptake by

macrophages (dextran 70 kDa = dex70). Data from MM-treated mice shown. (B) Frequency of

all dextran+ cells that were also macrophages (F4/80+CD11b+Ly6G-CD45+). (C) Frequency of

dextran+ macrophages that were CD206+ or MHCII+. Approximately 90% of all dextran+ cells

were macrophages, regardless of treatment (MAA vs. MM) and ~90% dextran+ macrophages

were CD206+ (vs. ~40% MHCII+). No differences were noted between MAA or MM explants;

n = 2. (D) Representative slices of tissues explanted from animals treated with MM beads (top)

or MAA beads (bottom). The arrows indicate the cells of interest. (E) Number of dextran+ cells

in the vicinity of MAA or MM beads. The image slices ranged from 200 to 800 μm into the

tissue. n = 2. Scale bar = 200 μm.

24

Discussion This chapter investigated the inflammatory cell response to methacrylic acid-containing beads

and showed that MAA beads promoted vessel formation. Moreover, treatment with MAA beads

promoted what we propose was an “alternative foreign body response”.

4.1 Effect of MAA beads on vessel formation

Previous studies in diabetic mice (BKS.Cg-m+/+ Leprdb/J mice, db/db) showed that MAA beads

increased vascularization in cutaneous wounds[16,49]. Here, subcutaneous injection of MAA

beads increased vessel density in non-diabetic mice relative to control MM beads (Fig. 4),

highlighting the vascular potency of MAA beads even in the absence of the physiological need

during diabetic wound healing. MAA beads nearly doubled the number of vessels at day 7

(~90% increase) (Fig. 4B). Only a few other synthetic biomaterials improve vascularization, to a

similar or frequently lesser degree[42,48,60].

To investigate the perfusability of newly formed vessels following treatment with MAA beads,

the explants containing beads (MAA and MM) were processed using a modified CLARITY

protocol. During CLARITY processing, light-scattering fatty lipids were removed while

proteinaceous structures and morphology were retained, enabling deep imaging and 3D

visualization of these fragile tissues[56,59]. Alexa 647-GSL1 (via tail vein injection; GSL-1 is a

lectin specific to mouse endothelial cells) staining was only observed around MAA and not MM

beads (Fig. 4C) indicating that the MAA-induced vessels were perfusable.

4.2 Effect of MAA beads on the inflammatory cell infiltrate

Treatment with MAA beads did not alter the number of F4/80+ macrophages in its vicinity

relative to control MM beads; however, the distribution of macrophages was different (Fig. 5A).

Flow cytometry analysis revealed that MAA beads altered the inflammatory cell landscape

relative to controls (Fig. 6). As expected, the presence of MAA beads resulted in more CD45+

leukocytes relative to the PEG vehicle control (Fig. 6A, Appendix S4). A fourfold increase in

Ly6G+ neutrophils was evident at day 1 in mice injected with MAA beads relative to both

controls (Fig. 6C). The link between MAA beads and neutrophil infiltration is not well

understood but may have been a result of protein (e.g., complement) adsorption differences (to

be discussed in Chapter 2). The significance of the increase in CD11c+ dendritic cells

25

(Appendix S4) is unclear. One caveat with the data was that the total number of cells was

determined by flow cytometry, with calibration beads used to determine the ratio between the

number of events and the number of cells. Cell numbers were further normalized by the mass of

the explants, recognizing that the volume of tissue that was digested varied to a small extent

from sample to sample. Although not statistically significant, higher explant masses and cell

numbers were recovered from MAA-treated animals. The reported numbers were reasonable

estimates of cell numbers, recognizing that we were interested in differences in inflammatory

cell infiltration over the course of the study. The normalization protocol may account for the

apparent increase in CD45- cells seen with PEG at day 1 (Fig. 6B). Explant masses and total cell

numbers were low with the vehicle-only controls (Appendix S3) so that after normalization, the

normalized numbers were artificially high. Following PEG treatment, the numbers of CD45+

leukocytes and CD45- non-leukocytes were unchanged from day 1 to day 7, as expected.

MAA beads increased the number of macrophages relative to MM beads at day 7 (Fig. 6D),

consistent with past observations of higher expression of TNFα and IL1β genes in diabetic

wounds at the same time point[49].The increase with histological analysis was not statistically

significant (Fig. 5B) presumably because of different regions of interest or a higher sensitivity of

flow cytometry to identify cells with lower levels of F4/80 expression. Although the analyses did

not distinguish between tissue-resident and bone-marrow derived macrophages, the mean

fluorescent intensity of F4/80 increased from day 1 to 7 for MAA and MM-treated animals (Fig.

6E), suggesting that macrophages were being recruited to the injection site, where they matured

over time. Macrophage maturation is associated with the expression of markers that are not

associated with blood-derived monocytes and changes in their transcriptome and proteome that

lead to fully-differentiated, tissue-resident cells[29,61].

In contrast to MAA beads, control MM beads were surrounded by a thick layer of F4/80+ cells

(i.e., macrophages) (Fig 5A); a common observation with implanted biomaterials (reviewed in

[5,6]). Similarly, a thick layer of cells was observed in Masson’s trichrome and CLARITY-

processed images around MM but not MAA beads (Fig. 4A, C). At day 7, MM-treated animals

had a higher density of CD45- cells (Fig. 6B), a majority of which were believed to be

fibroblasts. Overall, the distribution of F4/80 staining, the presence of a thick layer of

macrophages and a higher number of potentially fibroblasts suggested an increased fibrotic

response to control MM beads[41,62], a feature of a conventional foreign body

26

response[30,63,64]. On the other hand, the vascular regenerative MAA beads lacked a thick

layer of cells (Fig. 4A, C) and maintained low levels of CD45- cells (Fig. 6B); these are

indicative of what we have termed as an “alternative foreign body response”.

4.3 Effect of MAA beads on macrophage polarization

Macrophage phenotype varies depending on the conditions that have led to their

activation[32,65]; the M1/M2 distinction is an in vitro artifact and does not accurately reflect the

state of macrophages in vivo[32,66]. However, it is convenient to use the “M1” and “M2”

distinction as a simplification of the spectrum of polarization states. The importance of

macrophages has been readily tested; several groups have shown that elimination of

macrophages (via clodronate liposomes) detrimentally affects vessel formation[67] and that

addition of macrophages promotes neovascularization[68,69]. Despite the literature’s

considerable emphasis on the importance of M2 macrophages for vascularization, the extent of

their contribution remains unclear; the FBGC/ fibrotic qualities of M2 macrophages are often

ignored. While improved vascularization has been correlated with increased numbers of M2

macrophages[39], exogenous administration of M2 macrophages 1-3 days post-injury failed to

improve vascularization in a cutaneous wound model[70], although this may have reflected

changes that occur in pre-polarized macrophages upon implantation.

Macrophages, regardless of their polarization, have been shown to contribute to

vascularization[41]. Classically-activated, “M1” macrophages have a role in initiating vessel

formation[41] and alternatively-activated, “M2” macrophages that arise later in the foreign body

response are involved in promoting vessel maturation[41,71,72]. We hypothesized that MAA

beads orchestrated macrophage polarization towards the M2 state, consistent with the increased

vascularization. Treatment with MAA beads induced a M2 macrophage polarization bias (Fig.

7A, B). Flow cytometry analysis enabled quantification of macrophages that were CD206+, and

allowed for discrimination between those cells that were MHCII+ or MHCII-. Recognizing that

macrophage polarization is a complex spectrum, we designated MHCII+CD206- cells as M1

cells and MHCII-CD206+ cells as M2; double positive cells were also counted, although these

were neither M1 nor M2. At day 7, MAA beads increased the density of M2 cells fourfold

compared to MM beads (Fig. 7C), while MM beads induced a nearly nine-fold increase in M1

macrophages relative to MAA beads (Fig. 7D). Treatment with controls (MM beads) elicited a

27

more inflammatory macrophage response, with higher numbers of M1 and double positive

MHCII+CD206+ macrophages by day 7 (Fig. 7E, F). The latter are presumed to be cells in

transition from the initial inflammatory M1 cells to the later M2 cells, but additional research is

required to understand the role of these “hybrid” macrophages in vascularization. We think that

the increased numbers of M2 macrophages earlier in the host response promoted more vessel

maturation, resulting in a denser and perfusable vascular network.

The thick layer of cells (revealed to be F4/80+ macrophages, Fig. 5A) around MM but not MAA

beads, combined with the progressive increase in MHCII+CD206- (M1) macrophages in MM-

treated animals suggested the formation of FBGCs. Formation of large, MHCII+ cells that

resembled foreign body giant cells were noted in vitro in BMDM cultured with IL-4 (Appendix,

S7). Others have also noted the increased expression of MHCII in FBGCs[73]. MHCII and

CD206 are used as M1/M2 markers and do not accurate reflect the exact phenotype of the

labelled cells. Thus, the increase in MHCII+ and MHCII+CD206+ macrophages observed in

MM-treated animals may be an artifact of the markers used and may not be representative of M1

or hybrid macrophages.

4.4 Insights into MAA-mediated macrophage polarization using CLARITY

Macrophage polarization was further investigated using CLARITY to interrogate the spatial

distribution of macrophages in the context of MAA beads. Flow cytometry analysis revealed that

dextran (via tail vein injection) was associated with primarily CD206+ macrophages (~88% of

all dextran+ macrophages also expressed CD206+) (Fig. 8). Treatment with both MAA or MM

beads showed similar percentages, indicating that this observation was consistent between

treatment groups (Fig. 8C). Dextran+ macrophages were closely associated with vessels, but also

observed around MAA beads, in the presence and absence of vessels (Fig. 8D). In MM-treated

animals, the majority of the dextran+ macrophages were observed around blood vessels. The

increased expression of CD206 in cells surrounding MAA beads but not MM beads (Fig. 8E)

suggests that MAA beads may be interacting with macrophages and influencing their

polarization state directly. This data further supported our flow cytometry data and suggested

that MAA may be directly or indirectly (via various signaling pathways or neutrophils)

influencing M2 macrophage polarization.

28

Histological analysis on CD206-stained MAA and MM tissue sections in the same, albeit

transgenic mouse model produced a similar trend; although not statistically significant, MAA

treatment promoted more CD206+ cells in the vicinity of the beads at day 7 [15]. However, the

average CD206+ cell density was notably higher in the histological analyses relative to the

dextran-CLARITY method (Fig. 8E), which may have been a result of non-specific binding of

the CD206 antibody or the quantification strategy with histology. While it is unclear if the use of

dextran afforded higher specificity, what is clear is that the CLARITY protocol enabled the

direct interrogation of cells (i.e., dextran+ cells) whose labeling could potentially be validated

and quantified via flow cytometry simultaneously (from two tissue explants or one explant

divided in half). As the CLARITY technology advances and becomes more reliable and scalable

and strategies of labeling specific cells becomes available[74], it has the potential to rival

conventional histological protocols.

MAA beads promoted vascularization when implanted subcutaneously and the present data

suggests that M2 macrophages are one aspect of MAA’s vascular regenerative mechanism.

However, the extent of the contribution of M2 macrophages remains to be elucidated.

Macrophage depletion studies involving clodronate-liposomes conducted in a different system

(e.g., modules, grafts, etc.) revealed that macrophages are essential to vascularization; their

depletion leads to significantly reduced vessel formation[67]. It is necessary to conduct a

variation of a M2 macrophage knockdown to sufficiently demonstrate the importance of this

macrophage phenotype in the context of MAA. Yet, it may be difficult to design knockdown

studies that inhibit MAA-mediated M2 macrophage polarization, without adversely effecting

physiological vascularization as a whole. For example, a “M2 knockdown” may affect total

macrophage numbers, which would affect vessel formation[35,67]. Moreover, as macrophages

are polarized directly at the site of inflammation, it may be difficult to selectively knockdown

M2 macrophages, without adversely affecting M1 and other macrophage polarization states. To

this end, we propose to use the same subcutaneous injection model in Balb/c and C57BL/6 mice.

Balb/c mice are known to have a M2- biased inflammatory response while C57BL/6 have a M1-

biased inflammatory response; CD1 mice are in the middle of this spectrum and do not have a

M1-biased or M2-biased response [75]. Such experiments could shed light on the importance of

MAA-mediated macrophage polarization without affecting physiological functions.

29

While it is evident that MAA beads induced a bias in macrophage polarization towards M2, it is

unclear why this happens. Several interconnected mechanisms are likely involved and Chapter 2

aims to clarify these mechanisms. Knowledge of these mechanisms would translate to smarter

biomaterial designs for mediating vascularization and M2 macrophage polarization.

30

Conclusion

This chapter demonstrated that MAA beads promoted the formation of a denser and perfusable

network of blood vessels after subcutaneous injection, relative to control MM beads. Aim 1

revealed that the higher vessel density was accompanied by changes in inflammatory cell

infiltration (i.e., more neutrophils at day 1 and macrophages at day 7) and a macrophage

polarization bias towards the M2 state. Aim 2 explored this polarization bias further using

CLARITY, and revealed more M2-like macrophages in the immediate vicinity of MAA beads.

Together, these results suggest that MAA promoted an “alternative host response” (i.e., a foreign

body response distinct from the standard fibrosis) that is involved in MAA’s beneficial vascular

regenerative effect.

Chapter 2 Role of complement activation in MAA-mediated macrophage

polarization

Introduction

An altered inflammatory response involving M2 macrophage polarization is one element of a

complex network of pathways activated by MAA-based biomaterials to effect vascular

regeneration. The underlying mechanisms behind this polarization bias are the focus of this

chapter.

1.1 Protein-biomaterial interactions in the host response

The interactions between blood and a material is intimately associated with the inflammatory and

healing responses[76]. The inflammatory response is initiated by damaged tissues, but it is

modulated by the chemicals released from cells and those present in the plasma[5]. Within

milliseconds of contact between biomaterial and blood, a mixture of clotting factors (reviewed in

[21]) and complement proteins (reviewed in [77]) adsorb to the surface of the biomaterial,

initiating thrombosis, complement activation, and other blood-derived cascades[78]. The

acquisition of the layer of adsorbed proteins is an inevitable consequence of biomaterial

implantation[79]. One immediate and key mediator of this in vivo environment is the

complement system.

Complement is heralded as a major problem of biomaterial implantation; its activation promotes

adverse side-effects leading to poor biomaterial biocompatibility[6,63]. Blood-derived

complement proteins are the first arm of the innate immune system and are able to recognize

pathogens and foreign materials (reviewed in [80]). Upon contact with a biomaterial,

complement may be activated via three distinct pathways: 1) the classical pathway, 2) the lectin

pathway, and 3) the alternative pathway. The initiation of these pathways leads to the activation

of a cascade of proteases that converges at the level of C3 convertase, which mediates the

production of the anaphylatoxins C3a and later C5a, and ultimately the assembly of a terminal

complement complex (TCC) on a cellular surface, which forms pores in the cellular

membranes[80]. The anaphylatoxins propagate the inflammatory response by recruiting and

activating phagocytic cells (e.g., neutrophils and monocytes), triggering mast cell degranulation,

32

increasing vascular permeability, and inducing oxidative stress[6,19,63]. In the classical

pathway, adsorbed immunoglobulins (primarily IgG) are tagged by C1q, leading to the formation

of the C1 complex, which, via a cascade of proteolytic cleavages mediates the formation of C3

convertase. C3 can also adsorb directly on biomaterial surfaces and adopt an alternative

structure, initiating the alternative complement pathway[81]. Non-specifically adsorbed

carbohydrates and glycoproteins may also be recognized by the mannose binding lectin complex

(MBL), initiating the lectin pathway[80]. Dysregulated complement activation can lead to a

number of diseases, including kidney diseases and autoimmune diseases[82,83].

The potent effector functions of complement activation products have the potential to harm the

host. Thus, complement activation is a tightly-regulated process to prevent non-specific injury.

Initiators and inhibitors exist to regulate the location and activity of the complement system[81].

The C1 inhibitor (C1INH) binds to the C1 complex, as well as components of the lectin pathway,

inhibiting the activation of both the classical and lectin pathways[81]. Various plasma-derived

binding proteins and factors, such as C4 binding protein (C4BP), decay-accelerating factor

(DAF), Factor H, and others, accelerate the dissociation of the various complement proteins to

prevent hyperactivity. For example, DAF accelerates the dissociation of both the C3 and C5

complexes[81]. Various small-molecule drugs have been shown to mimic the function of some

of these regulators (Fig. 9). Pentamidine is a serine protease that inhibits C1s, analogous to

C1INH, inhibiting complement activation at the C1 level[84]. Aurin tricarboxylic acid (ATA)

inhibits the cleavage of C3b-Factor B to the active C3 convertase (C3b-Factor Bb) as well as the

attachment of C9 to C5b678; downstream of all three complement pathways.

33

Fig. 9. Drug-induced inhibition of complement activation. Pentamidine inhibits the classical

pathway. Aurin tricarboxylic acid (ATA) inhibits complement activation at the C3 and C5b-9

levels. A graphical depiction of the proteins adsorbed onto MAA beads is shown at the top of the

figure. Adapted from web.

1.2 Mechanisms of macrophage recruitment and polarization

It is traditionally thought that neutrophils promote the recruitment of monocytes to the site of

inflammation via the release of inflammatory markers (e.g., IL-1β, TNF-α, etc.); however, it has

since been demonstrated that monocyte recruitment may be independent of neutrophils[61].

Circulating monocytes respond to similar inflammatory signals as neutrophils and eventually

make their way to the site of inflammation. Neutrophils are primed to arrive faster due to

probability; there are significantly higher numbers of neutrophils circulating in the blood[22].

Therefore, complement activation products, such as C3a, are recognized by monocytes and

promote their recruitment to the site of inflammation[85]. Once differentiated into macrophages,

the inflammatory milieu of the implant site polarizes macrophage towards the M1 state, where

they play important roles in propagating inflammation and initiating vascularization. The

mechanisms that facilitate the transition from M1 to M2 macrophages are still not fully

understood; several mechanisms are proposed here.

34

1.2.1 Neutrophils

Neutrophils become apoptotic several hours after tissue infiltration. Apoptotic neutrophils have

been long known to be engulfed by macrophages[25]. More recently, it was shown that the

phagocytosis of apoptotic neutrophils promoted macrophages to adopt an M2 phenotype, seen by

an increase in the expression of anti-inflammatory genes [86] and prostaglandin E2, a key

mediator of vascularization[87]. Others also reported that the phagocytosis of apoptotic

neutrophils required macrophages to adopt a M2-like phenotype[88].

1.2.2 Complement proteins

Complement activation stimulates neutrophil infiltration, but their role in macrophage

polarization is not well defined. C1q, one of the key initiators of the classical pathway, was

shown to directly modulate macrophage polarization towards an alternatively activated “M2-

like” phenotype[89]. C1q was reported to interact with macrophages to stimulate the

upregulation of IL-10 and IL-33, polarizing macrophages towards the M2 phenotype,

accompanied by an upregulation in the apoptotic cell engulfment and a downregulation of the

NLRP3 inflammasome pathway[90]. On the contrary, others reported that C3-deficient mice

displayed increased neovascularization, albeit in a model of retinopathy, and that this effect was

mediated by macrophages[91]. Macrophages stimulated with C5a adopted an angiogenesis-

inhibitory phenotype, characterized by upregulated secretion of soluble VEGFR1, which

quenches VEGF, a potent initiator of vessel formation[91]. The disparity shown with these

examples highlight the diversity of mechanisms employed by complement proteins to influence

macrophage polarization.

1.2.3 IGF signaling pathway

IGF-1, a potent inducer of tissue regeneration, has been implicated in macrophage

polarization[92]. Expression of IGF-1 was increased in wounded tissues and

monocytes/macrophages were shown to be an initial source of IGF-1. Moreover, clonal deletion

of IGF-1 in specifically myeloid-derived cells impaired accumulation of CD206+ M2

macrophages [93]. Additionally, M(IL-4), “M2” macrophages were shown to upregulate IGF-1

expression [11] and macrophages from IGF-1 KO animals failed to generate a pro-healing

phenotype associated with M2 macrophage polarization – suggesting a role for IGF-1 in

promoting M2 macrophage polarization [93].

35

1.3 Biomaterial strategies for mediating macrophage polarization

Biomaterials are being used to effect M2 macrophage polarization. Mokarram et al developed

polymeric nerve guide channels that enabled the sustained release of IFNγ or IL-4 to promote

macrophage polarization towards the “M1” or “M2” state respectively, in a peripheral repair

model[94]. Polymers that released IL-4, but not IFNγ improved nerve repair while promoting a

higher ratio of CD206+/CCR7+ macrophages, indicating macrophage polarization towards the

M2 state. They attributed the increased nerve healing to the induction of M2 polarized

macrophages and the subsequent recruitment of Schwann cells. Others have attempted to take

this idea further in other models. Spiller et al developed a sequential-release strategy to deliver a

sequence of signals to enhance 1) M1 macrophage polarization, then 2) M2 macrophage

polarization[95]. Their bone scaffolds released IFNγ rapidly, then IL-4 over time to modulate the

shift from M1-to-M2 macrophages to promote vascularization. Interestingly, increased vessel

formation was only observed in animals implanted with scaffolds containing IFNγ and not

scaffolds containing IL-4 or the combination of IFNγ and IL-4. Although it is debatable if their

technique truly incorporates sequential IFNγ/IL-4 release, their observations highlight the

complexity of macrophage polarization. It is unclear why in one case IL-4 alone is able to

promote healing while in the other IFNγ alone was sufficient. Macrophage polarization is a

natural process of healing; M1 macrophage will inevitably dominate the early inflammatory host

response and M2 macrophages the late, healing response[28].

1.4 Complement modulating effects of MAA

MAA beads incubated with human serum showed preferential adsorption of various complement

proteins relative to MM beads (Wells, LA et al, Biomaterials, submitted). Among the adsorbed

proteins was the complement initiator C1q, and various inhibitor factors, such as factor H. The

presence of C4 and C3 suggested that complement was being activated on the surface of MAA

and MM beads. Interestingly, MAA beads incubated with human serum produced lower amounts

of C3a and C5b-9, suggesting that MAA beads inhibited complement activation. Protein

deposition on biomaterials is more complicated in vivo, involving the contribution of several

other blood-derived pathways which could alter the adsorbed layer of proteins over time[76].

The increased neutrophil infiltration following subcutaneous injection of MAA beads led us to

36

believe that MAA may indeed be activating complement and producing the anaphylatoxins C3a

and C5a to recruit more neutrophils and monocytes to the site of inflammation[15,96,97].

We sought to validate the in vitro findings and developed our hypothesis around the ideas that 1)

complement activation promotes neutrophil infiltration and 2) a less inflammatory response

involving lower numbers of apoptotic neutrophils reduces M2 macrophage polarization. We

hypothesized that inhibiting complement activation would 1) reduce neutrophil and monocyte

recruitment, 2) eliminate the M2 polarization bias induced by MAA, resulting in 3) lower vessel

densities. Our hypothesis aimed to link the events that occur immediate after biomaterial

implantation (i.e., complement protein adsorption) to the alternative host response and

vascularization response induced by MAA beads.

To investigate the role of complement in MAA-mediated macrophage polarization and

vascularization, we modified the subcutaneous injection model: CD1 mice were administered

either pentamidine or Aurin tricarboxylic acid (ATA) to inhibit complement activation at the C1

or C3 levels, respectively prior to the injection of MAA beads. Analogous to the previous study,

at day 1, 3, and 7 post-injection, the bead explants were processed for immunohistochemistry

and flow cytometry for the estimated number of cells and their polarization state (i.e.,

macrophages). Inhibition of complement activation eliminated the enhanced neutrophil

recruitment effect of MAA and reduced macrophage recruitment and the M2 macrophage

polarization bias at day 3; however, no notable changes in MAA-mediated vascularization were

observed.

37

1.5 Objectives

This chapter explores the role of complement activation on the MAA-mediated alternative host

response, with a focus on macrophage polarization, and vascularization.

Aim 2: Clarify the role of complement activation in MAA-mediated macrophage polarization

and vessel formation.

Hypothesis: Inhibition of complement activation reduces the number of infiltrating neutrophils

and monocytes and negates the effect of MAA beads on M2 macrophage polarization, leading to

lower numbers of CD31+ vessels.

38

Methods

2.1 Preparation of poly(methacrylic acid-co-isodecyl acrylate) films

Poly(methacrylic acid-co-isodecyl acrylate) (MAA-co-IDA) films or MAA films were composed

of 45 mol% methacrylic acid (Sigma-Aldrich Canada Ltd., Oakville, ON, Canada), 1 mol%

ethylene glycol dimethacrylate (Sigma-Aldrich Canada Ltd.) and 54 mol% IDA (Sigma-Aldrich

Canada Ltd.). These were used for some in vitro studies.

MAA films were synthesized by suspension polymerization as previously described[50] and

were dissolved in THF immediate prior to coating glass cover slips. The methacrylic acid content

of the films was confirmed by titration. Control poly(methyl methacrylate) (MM-co-IDA) films

or MM films were synthesized in a similar fashion to MAA films. MAA and MM films, post-

coating, were washed in 95% ethanol twice and then rinsed twice in sterile PBS prior to use in

vitro. Analysis with a limulus amebocyte lysate (LAL) pyrochrome endotoxin test kit (Cape Cod

Inc., Falmouth, MA) indicated that beads contained <0.25 EU/100 mg.

2.2 Isolation, culture, and characterization of bone marrow-derived monocytes

Bone marrow-derived macrophages (BMDM) were harvested from CD1 mice using

conventional techniques[11,45]. Briefly, hind limbs were removed from the hip joint and placed

in PBS on ice. Muscle was removed from the femur and tibia, washed and rinsed in RPMI 1640

and bone marrow was obtained by flushing the bone marrow contents using a 25G needle. The

acquired bone marrow cells and debris were filtered, washed, and cultured in RPMI 1640

supplemented with 10% heat-inactivated FBS, 1% Penn/Strep, and 20 ng/mL of M-CSF

(Invitrogen) on non-tissue culture-treated polystyrene plates for 5-7 days. Media was changed at

day 3 to remove non-adherent, non-macrophage cells. Macrophages (adherent) were

differentially polarized using 20 ng/mL IFNγ (Invitrogen) and 20 ng/mL IL-4 (Invitrogen) for 48

hours. Polarization was confirmed using CD11b, F4/80, along with either CD206 for

alternatively- activated (M2) macrophages or CD86 and MHCII for classically- activated (M1)

macrophages (Appendix, S9).

39

2.3 Macrophage stimulation by biomaterials in vitro

Cultured bone marrow-derived macrophages (BMDM) were stimulated with MAA (0.3g/cm2) or

MM (0.9g/cm2) beads or MAA-co-IDA/MM-co-IDA films for 24-72 hours, in the presence, or

absence of whole mouse CD1 blood (0.1% to 1% v/v, remainder RPMI 1640 medium + 10%

FBS + 1% P/S, minus M-CSF). MAA beads and films were equilibrated to pH 7.4 prior to

mixing with whole blood. The BMDM were treated in one of three ways: 1) Beads (2.85 mg

MAA or 8.55 mg MM), blood, and culture media mixture (totaling 1 mL) was directly

transferred into a 6 well plate containing adherent BMDM. 2) MAA-co-IDA or MM-co-IDA

films were added directly onto BMDM seeded onto 0.4 μm cell inserts and placed into 6 well

plates. 3) Beads, blood, and culture media (totaling 500 μL) were transferred onto 48 well plates

containing adherent BMDM.

After 24-72h of incubation, the beads or films were removed and the cells were washed 3 times

with PBS. For flow cytometry analysis, cells were harvested from the plate/cell insert,

resuspended in PBS supplemented with 0.5% BSA and 2 mM EDTA, and stained with F4/80,

MHCII and CD206. In other situations, involving 48 well plates, the BMDM were fixed, blocked

and stained with MHCII-PerCPe710, CD206-PE, and DAPI directly on the plate. The

fluorescence intensity of each well was acquired using a fluorescence plate reader (Tecan

M200Pro) (Appendix, S9).

2.4 CH50 type hemolysis assays

Mouse serum was prepared by collecting 20 μL of blood from the tail vein in normal tubes,

followed by fibrinization for 2h at 37°C, then centrifugation at 3000×g for 5 min. For hemolysis

assays, a CH50 type system was employed, as previously described[98,99]. The supernatant was

collected and diluted by adding 100 μL of complement buffer (20 mM HEPES, 0.5 mM MgCl2,

0.15 mM CaCl2, 141 mM NaCl and 0.1% gelatin). Separately, sheep red blood cells (QuadFive)

were washed 3 times in complement buffer followed by centrifugation at 3000×g for 5 min, until

no red color was observed in the supernatants. The RBCs were diluted to 5 × 107 cells/mL and

incubated with complement buffer containing 100 μg/mL of hemolysin (Rabbit Anti-Sheep IgGs

and IgM, CedarLane) to sensitize. The CH50 assay was performed using a reaction mixture of 50

μL sensitized RBCs and 50 μL of mouse serum, diluted serially 1- to 16- fold. The mixture was

incubated at 37°C for 30 min. The supernatants (80 μL) were collected and transferred to a 96-

40

well plate (Fischer) and the optical density (OD) at 414 nm was read with a plate reader (Tecan).

As a positive control, the RBCs were 100% lysed with water, and as a negative control, no serum

was added to the incubate. Hemolysis (as a % of the maximum level of hemolysis) of was

calculated using the following formula: (ODsample-ODpos)/(ODpos-ODneg) × 100%, where ODpos

and ODneg are the positive and negative controls, respectively. Inhibition of complement

activation was compared using the IC50, expressed as a dilution fold instead of as a

concentration.

2.5 Complement drug inhibition study

Prior to subcutaneous implantation of MAA or MM beads, animals were treated with either: 1)

pentamidine (intraperitoneal injection; 4 mg/kg – 20 mg/kg, Sigma-Aldrich), which inhibited the

initiation of the classical pathway by blocking complement activation at the C1 level or 2) Aurin

tricarboxylic acid (ATA; orally; 500 mg per kg of mash, Aurin Biotech), which inhibited

complement activation at the C3 and C5b-9 levels (see Fig. 9).

Animals were treated with either MAA beads, MM beads, or MAA beads following complement

inhibition. Pentamidine (4 mg/kg – 20 mg/kg) was injected intraperitoneally starting one day

prior to subcutaneous injection of MAA beads. Animals were treated with pentamidine once per

day until sacrifice at days 1, 3, or 7. ATA (500 mg/kg) was mixed with mash and fed to animals

in the absence of chow starting one day prior to subcutaneous injection of MAA beads. The

ATA-supplemented mash was changed once per day to avoid mold formation. Each animal was

estimated to receive an approximate dose of 100 mg/kg of ATA per day. During the ATA

optimization process, ATA was administered by subcutaneous injection (up to 2.5 - 10 mg/kg).

Control animals were given either an equal volume IP injection of 0.85% saline or were fed an

ATA-free diet. CH50-type assays were performed prior to injection of MAA or MM beads to

validate inhibition of complement activation. All animal work was done with the approval of the

University of Toronto Animal Care Committee. Animals were housed under sterile conditions in

the University of Toronto’s Department of Comparative Medicine (AUP #20011359).

2.6 Tissue explant and digestion

Tissue explant and digestion were performed as described in Chapter 1, with the following

exceptions: 1) No PEG vehicle control was included, and 2) cell suspensions were stained with

41

live/dead stain, CD45, CD11b, F4/80, Ly6G, CD86, MHCII, and CD206. CD206 conjugated to

PE was used instead of CD206 conjugated to BV650. The markers CD11c and CD31 were not

used.

2.7 Analysis of cellular infiltrate in explanted tissues

A modified gating strategy was devised from the one reported in Chapter 1 (Appendix, S10).

After isolating live single cells, CD45 distinguished leukocytes from non-leukocytes.

Neutrophils were identified as Ly6G+. Macrophages were first identified as Ly6G-

F4/80+CD11b+ and then further characterized as MHCII+ CD206- (“M1”) and MHCII-CD206+

(“M2”). Cells were gated according to positive staining for each antibody using fluorescence

minus one (FMO) controls. Cell populations were expressed as either a percentage or as a

normalized value (estimated total number of cells divided by the weight of the explanted tissue).

2.8 Statistical Analyses

All values are reported as mean ± SEM, unless indicated otherwise. No statistical analysis was

conducted to compare complement-inhibited (n = 2) animals to non-complement-inhibited (n =

2-3) animals due to low n values. The reported changes in the host response only represent

trends.

42

Results

3.1 Investigating the mechanism of MAA-mediated macrophage polarization

In the subcutaneous injection model, it was noted that deliberate nicking of small vessels

enhanced the vascular regenerative effect of MAA beads (but not MM beads), suggesting a role

for whole blood or one of its components. Motivated by these results, an attempt was made to

study the effect of MAA on primary macrophages in vitro. As bone marrow-derived

macrophages (BMDMs) are a widely-accepted model used for macrophage polarization[32], we

opted to use BMDMs in our in vitro experiments.

3.1.1 In vitro analysis of BMDM treated with MAA beads and films

BMDMs treated with MAA beads or MM beads showed similar levels of CD206, CD86, or

MHCII, as analyzed by flow cytometry and fluorescence imaging (Appendix, S11). However,

when a small quantity of whole murine blood (0.5-1% v/v) was added with the MAA beads prior

to its addition to BMDM, the expression of CD206, but not MHCII increased; treatment with

MM beads had the opposite effect (Appendix, S11). Additionally, stimulation of BMDM with

MAA films (which enabled more material-BMDM contact relative to MAA beads) upregulated

arginase 1 (Arg1; an M2 marker) mRNA expression and did not significantly change iNOS (an

M1 marker) mRNA expression (Appendix, S12). These trends, albeit small, suggested that

MAA-containing materials were directly polarizing macrophages towards the M2 state.

Moreover, these data suggested a role for blood or its components (i.e., serum) in MAA-

mediated macrophage polarization.

3.2 Inhibition of serum-derived complement and its effect on MAA

Separately, we have seen that incubating beads with plasma or serum resulted in more

complement proteins (e.g., C1q, Factor H) adsorbed to MAA relative to MM beads yet

complement was activated to a lower degree with MAA beads (Wells, L.A. et al, Biomaterials,

submitted). The in vitro situation does not completely reflect the in vivo situation, which involves

a number of additional pathways that add to the complexity of complement activation[21,77].

We sought to shed light on the in vitro findings by investigating the effect of complement

43

activation in the context of MAA in vivo. Pentamidine and Aurin tricarboxylic acid (ATA) were

used to inhibit complement activation and the resulting effects on MAA-mediated macrophage

polarization and vascularization were evaluated. Pentamidine inhibited the proteolytic activity of

C1, preventing the cleavage of C4 and C2 and the classical arm of the complement pathway [84].

Aurin tricarboxylic acid (ATA) inhibited the cleavage of C3b- Factor B to the active C3

convertase (C3b-Factor Bb) as well as the attachment of C9 to C5b678 (See Fig. 9); inhibiting

complement activation further downstream than pentamidine, at the C3 and C5b-9 levels

[100,101].

3.2.1 Effect of complement inhibition on the vascular regenerative properties of MAA

The results of the CH50-type assays are shown in Fig. 10. Diluted serum (1- to 16- fold) was

incubated with sensitized red blood cells (RBCs). Serum from pentamidine-and ATA-treated

animals required less dilution to reach baseline (i.e., negative control) than control animals,

indicating that complement activation was inhibited. Indeed, one single intraperitoneal (IP)

administration of 20 mg/kg pentamidine inhibited complement-mediated red blood cell

hemolysis with minor changes in inhibition over 24h (Fig. 10A); the effect of IP administration

of 4 mg/kg pentamidine was shorter-lived (Appendix, S13). Subcutaneous injection of up to 10

mg/kg ATA inhibited hemolysis, but the effect was short-lived (Appendix, S13). Oral

administration (by mixing ATA with mash) of 500 mg/kg ATA sufficiently inhibited hemolysis

24h following administration (Fig. 10B). The fecal matter of ATA-fed animals was a distinct red

color relative to normal, non-treated animals (Fig. 10B), suggesting that some ATA was not

adsorbed by the body [100,102]. The sera of animals administered pentamidine or ATA had their

lowest IC50 values of 0.373- and 0.366-fold, respectively, indicating a ~5-fold enhanced

protection compared to control animals (IC50 1.891-fold) (Fig. 10C). Animals were administered

pentamidine or ATA once daily to ensure that complement was inhibited over the entire duration

of each experiment.

44

Fig. 10. Administration of pentamidine and ATA inhibited complement activation. (A, B)

Serum from complement-inhibited mice were compared to control mice administered 0.85%

saline or normal diets (indicated by Ctrl). CH50-type assay of CD1 mice treated with 20 mg/kg

pentamidine (A) or 500 mg/kg ATA (B). Data points represent means only. (C) Efficiency of

complement inhibition, presented as the IC50. The IC50 at 2h following 20mg/kg pentamidine

treatment was 0.373-fold and 24h following 500mg/kg ATA treatment was 0.366-fold, while the

control animals were 1.891-fold. A lower IC50 indicated better inhibition of complement

activation. Pooled data from n=3. The dotted line represents the IC50 of control animals not given

pentamidine or ATA.

Animals treated with pentamidine or ATA showed no dramatic difference in the number of

CD31+ vessels relative to controls (Fig. 11A, B). Similar vessel densities were noted in

pentamidine- or ATA- treated animals at day 3 (Fig. 11C) relative to animals treated with MAA

alone. A similar observation was noted in pentamidine-treated animals at day 7. The day 7 time-

point was not evaluated in ATA-treated animals. A thick layer of cells was not seen around

MAA beads, regardless of complement inhibition. Together, these data suggested that the

complement inhibition at the C1 or C3 levels did not dramatically affect MAA-mediated

vascularization.

45

Fig. 11. Inhibition of complement activation did not affect the vascular potency of MAA.

(A, B) Histology sections of animals treated with MAA (A-left) or MM beads (A-right) or MAA

with either 20 mg/kg pentamidine (B-left) or 500 mg/kg ATA (B-right) at day 3. Arrows indicate

examples of vessels. (C) Inhibition of complement activation did not dramatically affect vessel

densities at day 3 or day 7. Scale bar = 200 μm. Pentamidine or ATA-treated animals; n = 2.

MAA- and MM-treated animals, n = 2-3.

3.2.2 Effect of complement inhibition on MAA-mediated inflammatory cell infiltration

Administration of pentamidine and ATA prior to subcutaneous injection of MAA beads modified

the MAA-mediated alternative host response. Using the modified gating strategy outlined in

Appendix, S10, no difference was noted in the mass of the tissue explants, the estimated total

number of cells, and the normalized cell numbers (Appendix, S14). The PEG vehicle control

was removed. No difference was noted in the densities of CD45+ or CD45- cells in complement-

inhibited animals relative to animals treated with MAA beads alone (Fig. 12A, B). Although a

progressive increase in CD45- cells (i.e., non-leukocytes) in pentamidine-treated animals was

observed (Fig. 12B), analogous to MM-treated animals. The frequencies of CD45+ and CD45-

cells (as a percentage of live cells) followed similar trends (Appendix, S15).

46

Fig. 12. Complement inhibition eliminated MAA’s enhanced neutrophil recruitment effect.

(A-C) Number of CD45+ leukocytes (A), CD45- non-leukocytes (B), and

Ly6G+CD11b+CD45+ neutrophils (C) in animals treated with 20 mg/kg pentamidine or 500

mg/kg ATA or MAA or MM beads alone. Administration of pentamidine lowered the numbers

of neutrophils (C) at day 1. (D) Representative dot plot of Ly6G+ neutrophils compared between

MAA-treated animals and pentamidine-treated animals. (E) Number of F4/80+ Ly6G-

CD11b+CD45+ macrophages. Inhibition of complement activation decreased macrophage

numbers at day 3 and day 7. MAA- and MM-treated animals, n = 2-3. Pentamidine/ATA-treated

animals; n = 2.

Inhibition of complement activation dramatically reduced the estimated number of neutrophils

recruited to the site of MAA beads at day 1 (Fig. 12C); a prominent decrease in neutrophil

density was observed in the representative dot plots (Fig. 12D). Inhibition of complement

activation decreased the number of macrophages at day 3 relative to treatment with MAA or MM

beads alone (Fig. 12E). Additionally, in pentamidine-treated animals, fewer macrophages were

observed at day 7. Overall, these results indicated that inhibition of complement activation

modified the MAA-driven host response; resulting in lower levels of neutrophil and macrophage

infiltration.

3.2.3 Effect of complement inhibition on MAA-mediated M2 macrophage polarization

Next, flow cytometry was used to distinguish macrophages between the “M1 (MHCII+CD206-)”

and “M2 (MHCII-CD206+)” polarization states, as described previously. There were some small

47

differences in macrophages expressing MHCII and CD206 following complement inhibition,

with a general trend involving less CD206+ expression in complement-inhibited animals at day 3

(Appendix, S15).

Treatment with pentamidine or ATA altered MAA-mediated macrophage polarization at day 3

(Fig. 13). The shift in polarization was reflected in the representative dot plots for pentamidine

or ATA-treated animals, relative to controls (Fig. 13A-C). Animals treated with pentamidine or

ATA prior to subcutaneous injection of MAA beads had lower numbers and frequencies of

MHCII-CD206+ (M2) macrophages at day 3, relative to animals treated with MAA beads alone

(Fig. 13D). However, by day 7, the number of M2 macrophages in pentamidine-treated animals

was restored to levels comparable to animals treated with MAA beads alone (Fig. 13D).

Conversely, the number of MHCII+CD206- (M1) macrophages remained at levels comparable to

animals treated with MAA beads only for the duration of the experiments (Fig. 13E), although a

small spike in these M1 macrophages was noted in pentamidine-treated animals at day 3.

48

Fig. 13. Complement inhibition altered the MAA-mediated effects on M2 macrophage

polarization. (A-C) Representative dot plot of F4/80+ cells (macrophages) in mice treated with

MAA beads (A) or pentamidine (20 mg/kg) (B) and ATA (500 mg/kg) (C) prior to subcutaneous

injection of MAA beads at day 3. (D, E) The number and frequency of the individual single

positive, double positive, and double negative MHCII or CD206 macrophage populations in

mice treated with MAA beads, MM beads, MAA + pentamidine or MAA + ATA. (D)

Normalized number and frequency of MHCII-CD206+ (“M2”) macrophages. (E) Normalized

number and frequency of MHCII+CD206- (“M1”) macrophages. Inhibition of complement

activation eliminated the M2 macrophage polarization bias observed in animals treated with

MAA beads alone. (F, G) Distribution of polarized macrophages: normalized number (F) and

frequency (G) of macrophages that were MHCII-CD206+, MHCII+CD206+, MHCII+CD206-,

and MHCII-CD206-. MAA- and MM-treated animals, n = 2-3. Pentamidine/ATA-treated

animals; n = 2.

49

Although by day 7, pentamidine-treated animals showed lower numbers of macrophages relative

to MAA alone controls (Fig. 13F), the polarization of macrophages was similar to MAA-treated

animals (Fig. 13G). Interestingly, the number of double positive (MHCII+CD206+)

macrophages in pentamidine-treated animals increased progressively over time, mimicking that

of MM-treated animals (Fig. 13F, G). The population of double-positive macrophages was small

in ATA-treated animals; instead, the majority of macrophages were double-negative (MHCII-

CD206-) at day 3 (Fig. 13G). Overall, these results indicated that complement inhibition

eliminated the M2 macrophage polarization bias mediated by MAA beads at day 3, but not at

day 7.

50

Discussion This chapter aimed to connect the protein-adsorption events that occur immediately following

MAA biomaterial implantation to the later changes in the host response. Previous studies showed

that more complement proteins (e.g., C1q, factor H, C4) were adsorbed onto MAA beads relative

to MM beads, although in vitro experiments showed inhibition of complement activation (Wells,

L.A. et al, Biomaterials, submitted). Recognizing the more complex nature of the in vivo

situation, and motivated by in vitro findings suggesting the involvement of serum in MAA-

mediated macrophage polarization (Appendix, S11, S12), this study explored the role of

complement activation in the context of MAA in vivo. The increased deposition of C1q and C4

suggested activation of the classical pathway; thus, pentamidine was used to inhibit complement

activation at the C1 level. Aurin tricarboxylic acid (ATA) was used to inhibit complement

activation at the C3 level as other complement pathways (i.e., the alternative pathway) could

potentially compensate for inhibition at the C1 level. The sera of pentamidine- or ATA- treated

animals gave a maximum ~5- fold protection compared to control animals (based on IC50) (Fig.

10), as expected. Complement inhibition at the C1 and C3 stages altered the MAA-mediated

alternative host response and macrophage polarization, but did not affect vascularization.

4.1 Role of complement inhibition in MAA-mediated vascularization

Treatment with pentamidine or ATA prior to subcutaneous injection of MAA beads resulted in a

marginal decrease in CD31+ vessels relative to animals treated with MAA alone at day 3. At day

7, similar numbers of CD31+ vessels were noted in pentamidine-inhibited and control animals

(Fig. 11). C1qa-/- knockout mice (a knockout of subcomponent a of the larger C1q complex)

exhibit impaired vascularization following injury; C1q is thought to directly interact with

endothelial cells to effect vascularization[103]. In this C1qa-deficient model, a lack of C3 and C4

deposition indicates marginal classical complement pathway activation, suggesting that C1q

independently mediates vessel formation without complement activation. On the contrary, C3-/-

knockout mice exhibit enhanced neovascularization, albeit in a model of retinopathy [91]. These

studies suggest that aspects of MAA-mediated activation may occur independently of the host

response and directly via C1q and endothelial cells, for example. Additionally, complement-

deficient animal models (knockout or knockdown models) likely inhibit complement activation

51

better (lower IC50) than Pentamidine or ATA. Thus, in our study, incomplete complement

inhibition may have contributed to the lack of change in vessel densities.

Although complement inhibition reduced MAA-mediated M2 macrophage polarization (to be

discussed below), there was no correlation to vessel density. The marginal drop in vascularity in

pentamidine and ATA-treated animals at day 3 was likely an artifact of low n and could be

explained by the lower number of macrophages in complement-inhibited animals at day 3 (Fig.

12E). We did not investigate vessel maturity, which may have been a more specific metric for

functional M2 macrophages than vessel density. One important factor secreted by M2

macrophages is PDGF, which recruits pericytes to the blood vessels, facilitating vessel

maturation[72].

4.2 Role of complement inhibition in MAA-mediated alternative host response

Complement has been implicated as an important mediator of inflammatory cell infiltration (e.g.,

neutrophils and macrophages) [85,104]. Adsorption of complement proteins onto biomaterials

leads to the activation of the classical, lectin, or alternative complement pathways, resulting in

the production of the anaphylatoxins C3a and C5a, which in turn promotes neutrophil and

monocyte recruitment[5]. Interestingly, our previous data suggested that MAA beads had

complement-inhibiting properties in vitro (Wells, L.A. et al). However, the increased neutrophil

infiltration observed in our most recent dataset was not consistent with reduced complement

activation, although complement is not the only driver for neutrophil infiltration[15]. We sought

to clarify these observations and for the first time, connect the events that occur immediately

following biomaterial implantation (i.e., complement protein deposition) to the alternative host

response mediated by MAA.

Administration of pentamidine and ATA prior to subcutaneous injection of MAA beads modified

the MAA-mediated alternative host response. A dramatic drop in Ly6G+ neutrophils was

observed in complement-inhibited animals relative to MAA controls at day 1 (Fig. 12C, D).

Furthermore, a decrease in the number of macrophages was noted at day 3 in complement-

inhibited animals relative to MAA and MM beads (Fig. 12E). It was likely that complement

inhibition led to lower levels of C3a and C5a, promoting less neutrophil and monocyte

infiltration to the site of biomaterial injection. The small, but progressive, increase in CD45-

52

cells in pentamidine-treated animals may have indicated more fibroblast recruitment and a more

conventional host response (Fig. 12B).

4.3 Role of complement inhibition in MAA-mediated M2 macrophage polarization

As apoptotic neutrophils are engulfed by macrophages, a shift in macrophage polarization from

M1 to M2 is induced[28,88]. C1q, a component of the larger C1 complex and an initiator of the

classical pathway, has been implicated in macrophage polarization[89]. C1q and other members

of the opsonin family (e.g., MBL and adiponectin) downregulate inflammasome activation in

macrophages while upregulating the phagocytosis of apoptotic cells, consistent with the M2

polarization state. Thus, complement may directly and indirectly affect macrophage polarization

through C1q and apoptotic neutrophils, respectively. We hypothesized that the complement-

induced dampening of inflammatory cell infiltration would lead to fewer apoptotic neutrophils

and consequently less M2 macrophage polarization.

Indeed, complement inhibition reduced the MAA-mediated M2 macrophage polarization (Fig.

13A-C). A notable drop in the number and frequency of M2 (MHCII-CD206+) macrophages

was observed at day 3 in both pentamidine and ATA-treated animals; however, by day 7, this

effect was nullified; the number and frequency of M2 macrophages in pentamidine-treated

animals were comparable to animals treated with MAA alone (Fig. 13D). The number of M1

(MHCII+CD206-) macrophages in complement-inhibited animals followed a similar trend to

MAA only animals (Fig. 13E). A small spike in M1 macrophages was noted in pentamidine-

treated animals at day 3 but lowered to levels similar to animals treated with MAA only at day 7.

(Fig. 13E). Together with the restored M2 macrophage polarization in pentamidine-treated

animals at day 7, these data suggest that there may be other pathways compensating for the effect

of complement activation. Alternatively, it may suggest that the MAA-mediated M2 polarization

bias was delayed due to the less robust inflammatory response.

Indeed, MAA beads activated the sonic hedgehog (Shh) signaling pathway, as observed by the

upregulation of Shh and its receptor Ptch1 (patched 1) in surrounding cells in the same

subcutaneous injection model[15]. Shh expression was upregulated in MAA film-treated BMDM

in vitro, suggesting a link between Shh and M2 macrophage polarization. The expression of Shh

has been associated with increased macrophage infiltration and expression of M2-related genes,

53

such as IL-10, suggesting that Shh may act as a M2 polarization signal[105]. The insulin-like

growth factor (IGF) signaling pathway has also been implicated in healing and macrophage

polarization. Clonal deletion of IGF-1 in myeloid-derived cells impaired accumulation of

CD206+ M2 macrophages in wounded muscle; suggesting a role for IGF-1 in promoting M2

macrophage polarization[93]. Indeed, past genomic and phosphoproteomics screens identified

the IGF signaling pathway as a prominent pathway that was engaged following MAA

treatment[51,53]. More recently, BMDM stimulated with MAA films were shown to secrete

more IGF-1. Endothelial cells treated with MAA-stimulated BMDM conditioned medium

improved their proliferative and migratory capabilities; BMDM-derived IGF was revealed to be

a main mediator of this effect. In the same study, BMDM treated with MAA films adopted an

M2-like phenotype (Appendix, S12), further suggesting that IGF-1 modulates M2 macrophage

polarization. These alternative pathways of M2 macrophage polarization may have compensated

for the inhibition of complement activation.

In our inhibition experiments, an important control (MM + drug) was missing; it was unclear

whether the effect of complement inhibition was affecting MAA-mediated macrophage

polarization or simply macrophage polarization in general. This being said, since the addition of

pentamidine or ATA did not dramatically affect vascularization, it was unlikely that pentamidine

or ATA had adverse effects on physiological macrophage polarization and may have only

influenced MAA-mediated effects. Complement protein knockout models (e.g., C1qa-/- or C3-/-

mice) may inhibit complement activation better than pentamidine or ATA (and are likely more

specific), but there are challenges associated with these models. If lower vessel densities are seen

in C1qa-/- animals, it may be difficult to distinguish whether this phenomenon is a result of the

knockout affecting MAA-mediated effects or rather, physiological vascularization as a whole.

Since MAA-based biomaterials are thought to act as agonists of biological responses, it is

necessary to tease out which responses are the most important. The data from the previous

phosphoproteomics[53] and current mass spectrometry studies (Wells, L.A. et al, Biomaterials,

submitted) could be employed here. A correlation analysis could be employed to correlate the

proteins and ligands most readily adsorbed to MAA (Wells, L.A. et al, Biomaterials, submitted)

to the cellular signaling pathways that are activated [53], rather than evaluating each component

individually. This would provide a more rigorous method of identifying signaling pathways to

follow up on.

54

4.4 Insight into the mechanism of vascular regenerative MAA beads

The results described here and previously (reviewed in [7]) revealed that MAA-based

biomaterials elicit its regenerative properties by modulating several aspects of vascular biology,

including macrophage polarization. We think that upon interaction with tissues, MAA-based

biomaterials adsorb proteins, such as complement and numerous others (Wells, LA. et al,

Biomaterials, submitted), which then modulate phosphorylation pathways within minutes of

contact between the biomaterial and cells[53]. These interactions modulate the subsequent host

response; differential expression of mRNA and proteins promote an alternative foreign body

response, characterized by increased neutrophil infiltration and monocytes. This in turn results in

the activation of other signaling pathways (e.g., Shh[15], IGF signaling, etc.), neutrophil

apoptosis, and the modulation of cells involved in vascularization (i.e., macrophages and

endothelial cells). It is clear that macrophages and complement activation are two small pieces to

a much larger and complex network of events engaged by MAA-based biomaterials to effect

vascular regeneration.

What is unclear is how all of this from protein adsorption to neutrophil recruitment, to

macrophage polarization, to vascularization is determined by the properties of the biomaterial.

This thesis is a first step towards linking the pathways that are activated immediately following

protein adsorption (i.e., complement) to neutrophils, macrophage polarization and

vascularization. Follow up studies will need to link these components using more sophisticated

models. We attribute the vascular regenerating effect to the methacrylic acid and its strong

charge but how the properties of this charge (e.g., pKa) influence the adsorbed protein and

whether other similar anionic polymers would also promote an alternative host response and are

vascular regenerating is an as of yet, unanswered question.

55

Conclusion

This chapter demonstrated that inhibition of complement activation dampened some aspects of

the MAA-mediated host response; reduced neutrophil and macrophage infiltration was noted at

day 1 and at day 3, respectively. Furthermore, inhibition of complement activation eliminated the

MAA-mediated M2 macrophage polarization bias at day 3, but not at day 7. However, the

reduced MAA-mediated M2 macrophage polarization did not translate to reduced vessel density,

suggesting that complement activation and M2 macrophage polarization bias may not play vital

roles in MAA’s vascular regenerative mechanism (Fig. 14). However, more detailed studies are

needed to fully understand the role of complement and M2 macrophages in the context of MAA.

Finally, the subcutaneous model and techniques devised here may also prove to be a useful tool

in understanding the immediate host response to the implantation of various biomaterials.

56

Fig. 14. Effect of MAA beads on macrophage polarization, vascularization and the role of

complement activation. MAA beads promoted the formation of a denser and perfusable

vascular network when implanted subcutaneously, accompanied by an M2 macrophage

polarization bias, relative to controls. Inhibition of complement activation abated the M2

polarization bias at day 3, but not at day 7; no effect on MAA-mediated vascularization was

noted. While complement appeared to play some role in MAA-mediated vascularization, further

studies need to be conducted to fully understand its role.

57

References

[1] Hench LL. Bioactive Materials 1996;8842:493–507.

[2] Peppas NA, Langer R. New challenges in biomaterials. Science (80- ) 1994;263:1715–20.

doi:10.1126/science.8134835.

[3] Michael V. Sefton. Interesting biomaterials keep biomaterials interesting. World

Biomater. Congr., Montreal: WBC; 2016.

[4] Ziats NP, Miller KM, Anderson JM. In vitro and in vivo interactions of cells with

biomaterials. Biomaterials 1988;9:5–13. doi:10.1016/0142-9612(88)90063-4.

[5] James M. Anderson1, 2,*, Analiz Rodriguez1,* and DTC. Foreign Body Reaction To

Biomaterials. Changes 2012;29:997–1003.

doi:10.1016/j.biotechadv.2011.08.021.Secreted.

[6] Franz S, Rammelt S, Scharnweber D, Simon JC. Immune responses to implants - A

review of the implications for the design of immunomodulatory biomaterials. Biomaterials

2011;32:6692–709. doi:10.1016/j.biomaterials.2011.05.078.

[7] Lisovsky A, Chamberlain MD, Wells LA, Sefton MV. Cell Interactions with Vascular

Regenerative MAA-Based Materials in the Context of Wound Healing. Adv Healthc

Mater 2015:n/a – n/a. doi:10.1002/adhm.201500192.

[8] Rouwkema J, Rivron NC, van Blitterswijk CA. Vascularization in tissue engineering.

Trends Biotechnol 2008;26:434–41. doi:10.1016/j.tibtech.2008.04.009.

[9] Forbes SJ, Rosenthal N. Preparing the ground for tissue regeneration: from mechanism to

therapy. Nat Med 2014;20:857–69. doi:10.1038/nm.3653.

[10] Butler M, Sefton M. Cotransplantation of adipose-derived mesenchymal stromal cells and

endothelial cells in a modular construct drives vascularization in SCID/bg mice. Tissue

Eng Part A 2012;18:1628–41. doi:10.1089/ten.tea.2011.0467.

[11] Jetten N, Verbruggen S, Gijbels MJ, Post MJ, De Winther MPJ, Donners MMPC. Anti-

inflammatory M2, but not pro-inflammatory M1 macrophages promote angiogenesis in

vivo. Angiogenesis 2014;17:109–18. doi:10.1007/s10456-013-9381-6.

[12] Adini A, Fainaru O, Udagawa T, Connor KM, Folkman J, D’Amato RJ. Matrigel

cytometry: a novel method for quantifying angiogenesis in vivo. J Immunol Methods

2009;342:78–81. doi:10.1016/j.jim.2008.11.016.

[13] Mason C, Dunnill P. The strong financial case for regenerative medicine and the regen

industry. Regen Med 2008;3:351–63. doi:10.2217/17460751.3.3.351.

[14] Eckhaus A a, Fish JS, Skarja G, Semple JL, Sefton M V. A preliminary study of the effect

of poly(methacrylic acid-co-methyl methacrylate) beads on angiogenesis in rodent skin

grafts and the quality of the panniculus carnosus. Plast Reconstr Surg 2008;122:1361–70.

doi:10.1097/PRS.0b013e3181882082.

[15] Lisovsky A, Zhang DKY, Sefton M V. Effect of methacrylic acid beads on the sonic

hedgehog signaling pathway and macrophage polarization in a subcutaneous injection

mouse model. Biomaterials 2016;98:203–14. doi:10.1016/j.biomaterials.2016.04.033.

[16] Martin DC, Semple JL, Sefton M V. Poly(methacrylic acid-co-methyl methacrylate) beads

promote vascularization and wound repair in diabetic mice. J Biomed Mater Res A

2010;93:484–92. doi:10.1002/jbm.a.32528.

[17] Patel RJ, Sefton M V. Some aspects of the host response to methacrylic acid containing

beads in a mouse air pouch. J Biomed Mater Res A 2012;100:2054–62.

doi:10.1002/jbm.a.34101.

[18] Anderson JM. Biological responses to materials. Annu Rev Mater Res 2001;31:81–110.

doi:10.1146/annurev.matsci.31.1.81.

58

[19] Nilsson B, Ekdahl KN, Mollnes TE, Lambris JD. The role of complement in biomaterial-

induced inflammation. Mol Immunol 2007;44:82–94. doi:10.1016/j.molimm.2006.06.020.

[20] Bianchi ME, Manfredi A a. How macrophages ring the inflammation alarm. Proc Natl

Acad Sci U S A 2014;111:2866–7. doi:10.1073/pnas.1324285111.

[21] Gorbet MB, Sefton M V. Biomaterial-associated thrombosis: Roles of coagulation factors,

complement, platelets and leukocytes. Biomater Silver Jubil Compend 2006;25:219–41.

doi:10.1016/B978-008045154-1.50025-3.

[22] Mak TW, Saunders ME. Ch 4 Innate Immunity. Elsevier; 2006. doi:10.1016/B978-

012088451-3.50006-5.

[23] Deryugina EI, Zajac E, Juncker-Jensen A, Kupriyanova T a, Welter L, Quigley JP. Tissue-

infiltrating neutrophils constitute the major in vivo source of angiogenesis-inducing

MMP-9 in the tumor microenvironment. Neoplasia 2014;16:771–88.

doi:10.1016/j.neo.2014.08.013.

[24] Rose S, Misharin A, Perlman H. A novel Ly6C/Ly6G-based strategy to analyze the mouse

splenic myeloid compartment. Cytom Part A 2012;81A:343–50.

doi:10.1002/cyto.a.22012.

[25] Savill JS, Wyllie AH, Henson JE, Walport MJ, Henson PM, Haslett C. Macrophage

phagocytosis of aging neutrophils in inflammation. Programmed cell death in the

neutrophil leads to its recognition by macrophages. J Clin Invest 1989;83:865–75.

doi:10.1172/JCI113970.

[26] Shi C, Pamer EG. Monocyte recruitment during infection and inflammation. Nat Rev

Immunol 2011;11:762–74. doi:10.1038/nri3070.

[27] Mantovani A, Biswas SK, Galdiero MR, Sica A, Locati M. Macrophage plasticity and

polarization in tissue repair and remodelling. J Pathol 2013;229:176–85.

doi:10.1002/path.4133.

[28] Ortega-Gómez A, Perretti M, Soehnlein O. Resolution of inflammation: an integrated

view. EMBO Mol Med 2013;5:661–74. doi:10.1002/emmm.201202382.

[29] Austyn JM, Gordon S. F4/80, a monoclonal antibody directed specifically against the

mouse macrophage. Eur J Immunol 1981;11:805–15. doi:10.1002/eji.1830111013.

[30] Brancato SK, Albina JE. Wound macrophages as key regulators of repair: origin,

phenotype, and function. Am J Pathol 2011;178:19–25. doi:10.1016/j.ajpath.2010.08.003.

[31] Mills CD. M1 and M2 Macrophages: Oracles of Health and Disease. Crit Rev Immunol

2012;32:463–88. doi:03b91cfa635b86ee,7acd81f604dc8b0c [pii].

[32] Murray PJ, Allen JE, Biswas SK, Fisher EA, Gilroy DW, Goerdt S, et al. Macrophage

Activation and Polarization: Nomenclature and Experimental Guidelines. Immunity

2014;41:14–20. doi:10.1016/j.immuni.2014.06.008.

[33] Lawrence T, Natoli G. Transcriptional regulation of macrophage polarization: enabling

diversity with identity. Nat Rev Immunol 2011;11:750–61. doi:10.1038/nri3088.

[34] Troidl C, Möllmann H, Nef H, Masseli F, Voss S, Szardien S, et al. Classically and

alternatively activated macrophages contribute to tissue remodelling after myocardial

infarction. J Cell Mol Med 2009;13:3485–96. doi:10.1111/j.1582-4934.2009.00707.x.

[35] Arnold L, Henry A, Poron F, Baba-Amer Y, van Rooijen N, Plonquet A, et al.

Inflammatory monocytes recruited after skeletal muscle injury switch into

antiinflammatory macrophages to support myogenesis. J Exp Med 2007;204:1057–69.

doi:10.1084/jem.20070075.

[36] Arnold L, Henry A, Poron F, Baba-Amer Y, van Rooijen N, Plonquet A, et al.

Inflammatory monocytes recruited after skeletal muscle injury switch into

antiinflammatory macrophages to support myogenesis. J Exp Med 2007;204:1057–69.

59

doi:10.1084/jem.20070075.

[37] Ramachandran P, Pellicoro A, Vernon MA, Boulter L, Aucott RL, Ali A, et al.

Differential Ly-6C expression identifies the recruited macrophage phenotype, which

orchestrates the regression of murine liver fibrosis. Proc Natl Acad Sci 2012;109:E3186–

95. doi:10.1073/pnas.1119964109.

[38] Brodbeck WG, Anderson JM. Giant cell formation and function. Curr Opin Hematol

2009;16:53–7. doi:10.1097/MOH.0b013e32831ac52e.

[39] Badylak SF, Valentin JE, Ravindra AK, McCabe GP, Stewart-Akers AM. Macrophage

Phenotype as a Determinant of Biologic Scaffold Remodeling. Tissue Eng Part A

2008;14:1835–42. doi:10.1089/ten.tea.2007.0264.

[40] Tous E, Weber HM, Lee MH, Koomalsingh KJ, Shuto T, Kondo N, et al. Tunable

hydrogel-microsphere composites that modulate local inflammation and collagen bulking.

Acta Biomater 2012;8:3218–27. doi:10.1016/j.actbio.2012.05.027.

[41] Spiller KL, Anfang RR, Spiller KJ, Ng J, Nakazawa KR, Daulton JW, et al. The role of

macrophage phenotype in vascularization of tissue engineering scaffolds. Biomaterials

2014;35:4477–88. doi:10.1016/j.biomaterials.2014.02.012.

[42] Madden LR, Mortisen DJ, Sussman EM, Dupras SK, Fugate JA, Cuy JL, et al.

Proangiogenic scaffolds as functional templates for cardiac tissue engineering. Proc Natl

Acad Sci 2010;107:15211–6. doi:10.1073/pnas.1006442107.

[43] Fantin A, Vieira JM, Gestri G, Denti L, Schwarz Q, Prykhozhij S, et al. Tissue

macrophages act as cellular chaperones for vascular anastomosis downstream of VEGF-

mediated endothelial tip cell induction. Blood 2010;116:829–40. doi:10.1182/blood-2009-

12-257832.

[44] Spiller KL, Freytes DO, Vunjak-Novakovic G. Macrophages Modulate Engineered

Human Tissues for Enhanced Vascularization and Healing. Ann Biomed Eng 2014.

doi:10.1007/s10439-014-1156-8.

[45] Cho D-I, Kim MR, Jeong H, Jeong HC, Jeong MH, Yoon SH, et al. Mesenchymal stem

cells reciprocally regulate the M1/M2 balance in mouse bone marrow-derived

macrophages. Exp Mol Med 2014;46:e70. doi:10.1038/emm.2013.135.

[46] Sasaki M, Abe R, Fujita Y, Ando S, Inokuma D, Shimizu H. Mesenchymal Stem Cells

Are Recruited into Wounded Skin and Contribute to Wound Repair by

Transdifferentiation into Multiple Skin Cell Type. J Immunol 2008;180:2581–7.

doi:10.4049/jimmunol.180.4.2581.

[47] Hubbell J a, Thomas SN, Swartz M a. Materials engineering for immunomodulation.

Nature 2009;462:449–60. doi:10.1038/nature08604.

[48] Rajangam K, Arnold MS, Rocco MA, Stupp SI. Peptide amphiphile nanostructure-heparin

interactions and their relationship to bioactivity. Biomaterials 2008;29:3298–305.

doi:10.1016/j.biomaterials.2008.04.008.

[49] Fitzpatrick LE, Lisovsky A, Sefton M V. The expression of sonic hedgehog in diabetic

wounds following treatment with poly(methacrylic acid-co-methyl methacrylate) beads.

Biomaterials 2012;33:5297–307. doi:10.1016/j.biomaterials.2012.04.008.

[50] Wells LA, Sefton M V. The effect of methacrylic acid in smooth coatings on dTHP1 and

HUVEC gene expression. Biomater Sci 2014;2:1768–78. doi:10.1039/C4BM00159A.

[51] Fitzpatrick LE, Chan JWY, Sefton M V. On the mechanism of poly(methacrylic acid -co-

methyl methacrylate)-induced angiogenesis: gene expression analysis of dTHP-1 cells.

Biomaterials 2011;32:8957–67. doi:10.1016/j.biomaterials.2011.08.021.

[52] Tan Z, Xie N, Banerjee S, Cui H, Fu M, Thannickal VJ, et al. The monocarboxylate

transporter 4 is required for glycolytic reprogramming and inflammatory response in

60

macrophages. J Biol Chem 2015;290:46–55. doi:10.1074/jbc.M114.603589.

[53] Chamberlain MD, Wells LA, Lisovsky A, Guo H, Isserlin R, Talior-Volodarsky I, et al.

Unbiased phosphoproteomic method identifies the initial effects of a methacrylic acid

copolymer on macrophages. Proc Natl Acad Sci U S A 2015;112:10673–8.

doi:10.1073/pnas.1508826112.

[54] Robbins CS, Hilgendorf I, Weber GF, Theurl I, Iwamoto Y, Figueiredo J-L, et al. Local

proliferation dominates lesional macrophage accumulation in atherosclerosis. Nat Med

2013;19:1166–72. doi:10.1038/nm.3258.

[55] Egeblad M, Nakasone ES, Werb Z. Tumors as organs: complex tissues that interface with

the entire organism. Dev Cell 2010;18:884–901. doi:10.1016/j.devcel.2010.05.012.

[56] Sindhwani S, Syed AM, Wilhelm S, Glancy DR, Chen YY, Dobosz M, et al. Three-

Dimensional Optical Mapping of Nanoparticle Distribution in Intact Tissues. ACS Nano

2016:Article ASAP. doi:10.1021/acsnano.6b01879.

[57] Nanda JS, Lorsch JR. Labeling a protein with fluorophores using NHS ester derivitization.

Methods Enzymol 2014;536:87–94. doi:10.1016/B978-0-12-420070-8.00008-8.

[58] Alroy J, Goyal V, Skutelsky E. Lectin histochemistry of mammalian endothelium.

Histochemistry 1987;86:603–7.

[59] Tomer R, Ye L, Hsueh B, Deisseroth K. Advanced CLARITY for rapid and high-

resolution imaging of intact tissues. Nat Protoc 2014;9:1682–97.

doi:10.1038/nprot.2014.123.

[60] Cho H, Balaji S, Sheikh AQ, Hurley JR, Tian YF, Collier JH, et al. Regulation of

endothelial cell activation and angiogenesis by injectable peptide nanofibers. Acta

Biomater 2012;8:154–64. doi:10.1016/j.actbio.2011.08.029.

[61] Henderson RB, Hobbs JAR, Mathies M, Hogg N. Rapid recruitment of inflammatory

monocytes is independent of neutrophil migration. Blood 2003;102:328–35.

doi:10.1182/blood-2002-10-3228.

[62] Higgins DM, Basaraba RJ, Hohnbaum AC, Lee EJ, Grainger DW, Gonzalez-Juarrero M.

Localized immunosuppressive environment in the foreign body response to implanted

biomaterials. Am J Pathol 2009;175:161–70. doi:10.2353/ajpath.2009.080962.

[63] Thevenot P, Hu W, Tang L. Surface chemistry influences implant biocompatibility. Curr

Top Med Chem 2008;8:270–80. doi:10.2174/156802608783790901.

[64] Nishi O, Nishi K, Sakka Y, Sakuraba T, Maeda S. Intercapsular cataract surgery with lens

epithelial cell removal. Part IV: Capsular fibrosis induced by poly(methyl methacrylate). J

Cataract Refract Surg 1991;17:471–7.

[65] Murray PJ, Wynn TA. Protective and pathogenic functions of macrophage subsets. Nat

Rev Immunol 2011;11:723–37. doi:10.1038/nri3073.

[66] Xue J, Schmidt S V, Sander J, Draffehn A, Krebs W, Quester I, et al. Transcriptome-based

network analysis reveals a spectrum model of human macrophage activation. Immunity

2014;40:274–88. doi:10.1016/j.immuni.2014.01.006.

[67] Hibino N, Yi T, Duncan DR, Rathore A, Dean E, Naito Y, et al. A critical role for

macrophages in neovessel formation and the development of stenosis in tissue-engineered

vascular grafts. FASEB J 2011;25:4253–63. doi:10.1096/fj.11-186585.

[68] Hirose N, Maeda H, Yamamoto M, Hayashi Y, Lee G-H, Chen L, et al. The local

injection of peritoneal macrophages induces neovascularization in rat ischemic hind limb

muscles. Cell Transplant 2008;17:211–22.

[69] Hisatome T, Yasunaga Y, Yanada S, Tabata Y, Ikada Y, Ochi M. Neovascularization and

bone regeneration by implantation of autologous bone marrow mononuclear cells.

Biomaterials 2005;26:4550–6. doi:10.1016/j.biomaterials.2004.11.032.

61

[70] Jetten N, Roumans N, Gijbels MJ, Romano A, Post MJ, De Winther MPJ, et al. Wound

administration of M2-polarized macrophages does not improve murine cutaneous healing

responses. PLoS One 2014;9:1–9. doi:10.1371/journal.pone.0102994.

[71] Ding Y, Song N, Luo Y. Role of bone marrow-derived cells in angiogenesis: focus on

macrophages and pericytes. Cancer Microenviron 2012;5:225–36. doi:10.1007/s12307-

012-0106-y.

[72] Hellberg C, Ostman A, Heldin C-H. PDGF and vessel maturation. Recent Results Cancer

Res Fortschritte Der Krebsforsch Progres Dans Les Rech Sur Le Cancer 2010;180:103–

14. doi:10.1007/978-3-540-78281-0_7.

[73] Shishido-Hara Y, Kurata A, Fujiwara M, Itoh H, Imoto S, Kamma H. Two cases of breast

carcinoma with osteoclastic giant cells: are the osteoclastic giant cells pro-tumoural

differentiation of macrophages? Diagn Pathol 2010;5:55. doi:10.1186/1746-1596-5-55.

[74] Murray E, Cho JH, Goodwin D, Ku T, Swaney J, Kim SY, et al. Simple, Scalable

Proteomic Imaging for High-Dimensional Profiling of Intact Systems. Cell

2015;163:1500–14. doi:10.1016/j.cell.2015.11.025.

[75] Mills CD, Kincaid K, Alt JM, Heilman MJ, Hill AM. M-1/M-2 Macrophages and the

Th1/Th2 Paradigm. J Immunol 2000;164:6166–73. doi:10.4049/jimmunol.164.12.6166.

[76] Engberg AE, Rosengren-Holmberg JP, Chen H, Nilsson B, Lambris JD, Nicholls IA, et al.

Blood protein-polymer adsorption: Implications for understanding complement-mediated

hemoincompatibility. J Biomed Mater Res - Part A 2011;97 A:74–84.

doi:10.1002/jbm.a.33030.

[77] Cazander G, Jukema GN, Nibbering PH. Complement activation and inhibition in wound

healing. Clin Dev Immunol 2012;2012. doi:10.1155/2012/534291.

[78] Vogler E a., Siedlecki C a. Contact activation of blood-plasma coagulation. Biomaterials

2009;30:1857–69. doi:10.1016/j.biomaterials.2008.12.041.

[79] Hu WJ, Eaton JW, Ugarova TP, Tang LP. Molecular basis of biomaterial-mediated

foreign body reactions (vol 98, pg 1231, 2001). Blood 2002;99:3908.

[80] Mak TW, Saunders ME. Ch 19 Complement. The Immune Response 2006:553–81.

[81] Noris M, Remuzzi G. Overview of complement activation and regulation. Semin Nephrol

2013;33:479–92. doi:10.1016/j.semnephrol.2013.08.001.

[82] Vieyra MB, Heeger PS. Novel aspects of complement in kidney injury. Kidney Int

2010;77:495–9. doi:10.1038/ki.2009.491.

[83] Lalli PN, Strainic MG, Yang M, Lin F, Medof ME, Heeger PS. Locally produced C5a

binds to T cell-expressed C5aR to enhance effector T-cell expansion by limiting antigen-

induced apoptosis. Blood 2008;112:1759–66. doi:10.1182/blood-2008-04-151068.

[84] Hauptmann J MF. Inhibition of the haemolytic complement activity by derivatives of

benzamidine. Allergy n.d.

[85] Zwirner J, Götze O, Moser A, Sieber A, Begemann G, Kapp A, et al. Blood- and skin-

derived monocytes/macrophages respond to C3a but not to C3a(desArg) with a transient

release of calcium via a pertussis toxin-sensitive signal transduction pathway. Eur J

Immunol 1997;27:2317–22. doi:10.1002/eji.1830270928.

[86] El Kebir D, Filep JG. Targeting neutrophil apoptosis for enhancing the resolution of

inflammation. Cells 2013;2:330–48. doi:10.3390/cells2020330.

[87] Brecht K, Weigert A, Hu J, Popp R, Fisslthaler B, Korff T, et al. Macrophages

programmed by apoptotic cells promote angiogenesis via prostaglandin E2. FASEB J

2011;25:2408–17. doi:10.1096/fj.10-179473.

[88] Zizzo G, Hilliard B a, Monestier M, Cohen PL. Efficient clearance of early apoptotic cells

by human macrophages requires M2c polarization and MerTK induction. J Immunol

62

2012;189:3508–20. doi:10.4049/jimmunol.1200662.

[89] Bohlson SS, Oâ€TMConner SD, Hulsebus HJ, Ho M-M, Fraser D a. Complement, C1q,

and C1q-Related Molecules Regulate Macrophage Polarization. Front Immunol 2014;5:1–

7. doi:10.3389/fimmu.2014.00402.

[90] Benoit ME, Clarke E V, Morgado P, Fraser D a, Tenner AJ. Complement protein C1q

directs macrophage polarization and limits inflammasome activity during the uptake of

apoptotic cells. J Immunol 2012;188:5682–93. doi:10.4049/jimmunol.1103760.

[91] Langer HF, Chung K-J, Orlova V V, Choi EY, Kaul S, Kruhlak MJ, et al. Complement-

mediated inhibition of neovascularization reveals a point of convergence between innate

immunity and angiogenesis. Blood 2010;116:4395–403. doi:10.1182/blood-2010-01-

261503.

[92] Tidball JG, Welc SS. Macrophage-Derived IGF-1 Is a Potent Coordinator of Myogenesis

and Inflammation in Regenerating Muscle. Mol Ther 2015;23:1134–5.

doi:10.1038/mt.2015.97.

[93] Tonkin J, Temmerman L, Sampson RD, Gallego-Colon E, Barberi L, Bilbao D, et al.

Monocyte/macrophage-derived IGF-1 orchestrates murine skeletal muscle regeneration

and modulates autocrine polarization. Mol Ther 2015:1–12. doi:10.1038/mt.2015.66.

[94] Mokarram N, Merchant A, Mukhatyar V, Patel G, Bellamkonda R V. Effect of

modulating macrophage phenotype on peripheral nerve repair. Biomaterials

2012;33:8793–801. doi:10.1016/j.biomaterials.2012.08.050.

[95] Spiller KL, Nassiri S, Witherel CE, Anfang RR, Ng J, Nakazawa KR, et al. Sequential

delivery of immunomodulatory cytokines to facilitate the M1-to-M2 transition of

macrophages and enhance vascularization of bone scaffolds. Biomaterials 2015;37:194–

207. doi:10.1016/j.biomaterials.2014.10.017.

[96] Rinder CS, Rinder HM, Smith BR, Fitch JCK, Smith MJ, Tracey JB, et al. Blockade of

C5a and C5b-9 generation inhibits leukocyte and platelet activation during extracorporeal

circulation. J Clin Invest 1995;96:1564–72. doi:10.1172/JCI118195.

[97] Guo R-F, Ward PA. Role of C5a in inflammatory responses. Annu Rev Immunol

2005;23:821–52. doi:10.1146/annurev.immunol.23.021704.115835.

[98] Lee M, Narayanan S, McGeer EG, McGeer PL. Aurin tricarboxylic acid protects against

red blood cell hemolysis in patients with paroxysmal nocturnal hemoglobinemia. PLoS

One 2014;9:1–7. doi:10.1371/journal.pone.0087316.

[99] Bing DH. Inhibition of Guinea Pig Complement by Aromatic Amidine and Guanidine

Compounds. J Immunol 1972;108:1563–9.

[100] Lee M, Guo J, Schwab C, McGeer EG, McGeer PL. Selective inhibition of the membrane

attack complex of complement by low molecular weight components of the aurin

tricarboxylic acid synthetic complex. Neurobiol Aging 2012;33:2237–46.

doi:10.1016/j.neurobiolaging.2011.12.005.

[101] Bicknell R, Zhuang X. ( 12 ) Patent Application Publication 2012;1.

[102] Lee M, Guo JP, McGeer EG, McGeer PL. Aurin tricarboxylic acid self-protects by

inhibiting aberrant complement activation at the C3 convertase and C9 binding stages.

Neurobiol Aging 2013;34:1451–61. doi:10.1016/j.neurobiolaging.2012.10.023.

[103] Bossi F, Tripodo C, Rizzi L, Bulla R, Agostinis C, Guarnotta C, et al. C1q as a unique

player in angiogenesis with therapeutic implication in wound healing. Proc Natl Acad Sci

U S A 2014;111:10–5. doi:10.1073/pnas.1311968111.

[104] Snyderman R, Pike MC. Chemoattractant receptors on phagocytic cells. Annu Rev

Immunol 1984;2:257–81. doi:10.1146/annurev.iy.02.040184.001353.

[105] Zhou X, Liu Z, Jang F, Xiang C, Li Y, He Y. Autocrine Sonic hedgehog attenuates

63

inflammation in cerulein-induced acute pancreatitis in mice via upregulation of IL-10.

PLoS One 2012;7:e44121. doi:10.1371/journal.pone.0044121.

Appendices

S1. Gating Strategy for macrophages (day 3 shown; MAA beads). Single cells and live cells

were gated from all the cells. Leukocytes (CD45+) and non-leukocytes (CD45-) cells were gated

from the live, single cell population. From the CD45- non-leukocytes, endothelial (CD31+) cells

were gated. From the CD45+ gate, neutrophils (Ly6G+) were removed. Then, from the Ly6G-

gate, dendritic cells (CD11c+) were removed. Macrophages were gated on the Ly6G-CD11c-

population as F4/80+ and CD11b+. From the macrophage population, MHCII and CD206 were

used to distinguish between “M1” and “M2” macrophages, respectively.

65

Analysis Stain or antigen Type or fluorophore

Source (Clone/product code) Function

Histology CD31 Rabbit

polyclonal

Santa Cruz Biotechnology

(SC-1506)

Endothelial cell marker of

intercellular junctions.

F4/80 Rat

monoclonal

AbD Serotec (MCA497GA

Cl: A3-1)

Surface marker specific to

macrophages; involved in

peripheral T cell tolerance

Flow Cytometry Live/Dead Blue UV450 Invitrogen Discerns live cells from

dead cells.

CD45 (Leukocytes) BV711 BD Biosciences (30-F11) Especially expressed in

hematopoietic cells;

phosphatase involved in

cell growth and

differentiation.

CD11b (Macrophages) PE-eF594 BD Biosciences (M1/70) Regulator of leukocyte

adhesion, migration,

phagocytosis, and others.

F4/80 (Macrophages) APC-e710 eBioscience (BM8) See above.

Ly6G (Neutrophils) V450 BD Biosciences (1A8) A maturation marker

specific to neutrophils.

CD11c (Dendritic cells) BV511 BD Biosciences (HL3) An integrin associated

with adherence and

phagocytosis.

CD86 BUV395 BD Biosciences (B7-2) Provides “signal 2” for T

cell activation and

survival

CD206 BV650 or PE Biolegend (C068C2)

Biolegend (C068C2)

A mannose scavenger

receptor.

MHCII PerCP-e780 eBioscience (M5/114.15.2) Involved in antigen

presentation.

CD31 (Endothelial Cells) PE-Cy7 eBioscience (390) See above.

Unused, but mentioned

Arg1 arginase 1 N/A Enzyme upregulated in

M2 macrophages that

converts arginine to

ornithine

IFNγ interferon

gamma

N/A Potent inducer of M1

macrophage activation

IL-4 interleukin-4 N/A Primary inducer of M2

macrophage activation.

IL-10 interleukin-10 N/A An anti-inflammatory

cytokine that promotes

M2 macrophage activation

IL-13 interleukin-13 N/A An anti-inflammatory

cytokine that promotes

M2 macrophage

activation; shares same

receptor as IL-4.

S2. Markers used for immunohistochemistry and flow cytometry analyses and definitions.

List of markers used for analyses in addition to proteins and markers mentioned throughout the

document.

66

S3. Explant mass, cell number and normalized cell number for flow cytometry analyses.

(A) Mass of explants. (B) Numbers of cells collected by flow cytometry. (C) Normalized cell

numbers, obtained by dividing the cell number by the explant mass.

67

S4. Leukocytes, endothelial and dendritic cell populations in explanted tissues. Frequency of

CD45+ leukocytes (A) and CD45- non-leukocytes (B) expressed as a percentage of live cells.

(C) Normalized cell number and frequency of CD31+ endothelial cells, the latter expressed as a

percentage of CD45- leukocytes. (D) CD11c+ dendritic cells, expressed as a number (per mg of

explant) or as a percentage of CD45+ leukocytes.

68

S5. Expression of CD206, CD86, and MHCII in bone marrow-derived macrophages

polarized by IFNγ and IL-4. (A) Cells recovered from bone marrow cultured in M-CSF were

primarily macrophages. Black box designates F4/80+CD11b+ cells, gated on fluorescence-

minus-one (FMO) controls. (B) BMDM treated with IFNγ or IL-4. Note the shift in macrophage

expression of MHCII and CD206. Gating based on FMO controls. (C) Histograms depicting

CD86+, MHCII+, and CD206+ cells in BMDM polarized with IFNγ or IL-4. The numbers

represent the cells positive for the marker of interest, presented as a fraction of the parent

population (live cells (A) or macrophages (C)).

69

S6. Macrophage polarization - single positive cells. (A) Frequency of macrophages expressed

as a frequency of CD45+ leukocytes. (B-D) Normalized number and frequency (as a percentage

of macrophages) of (B) CD86+ cells, (C) MHCII+ cells and (D) CD206+ cells.

70

S7. Formation of giant-like cells in vitro. (A) Macrophages cultured on flat, non-tissue culture-

treated polystyrene and polarized with IFNγ or IL-4. (B) Expression of CD206 and MHCII in

macrophages recovered from flat, non-tissue culture-treated polystyrene. As expected, M(IFNγ)

“M1” macrophages expressed MHCII but not CD206 and M(IL-4) “M2” macrophages expressed

CD206 but not MHCII. (C) Macrophages cultured on tissue culture-treated polyethylene

terephthalate (PET) inserts over 72h in the presence of IL-4. Note the formation of larger, giant-

like cells that expressed MHCII+. The arrows indicate examples of the giant-like cells. (D)

Expression of CD206+ and MHCII+ in macrophages recovered from PET cell inserts. Note that

with IL-4 treatment, macrophages were primarily MHCII+ or MHC+CD206+, instead of the

conventional MHCII-CD206+. Scale bars = 100 μm. Giant cell fluorescence image courtesy of M.

Saleh.

S8. Gating strategy for validating dextran uptake in CD206+ macrophages. (A) Doublets

were removed, followed by dead cells. Two gating strategies were used to look at dextran+

macrophages: (B) Macrophages were identified first, then dextran+ macrophages, and lastly

CD206+ and MHCII+ dextran+ macrophages. (C) Dextran+ cells were identified first, then

macrophages, and lastly CD206+ and MHCII+ dextran+ macrophages. Note that not all

macrophages were associated with the dextran, but the majority of macrophages that were

associated with the dextran were CD206+, as identified in (B) and (C).

72

S9. Bone marrow harvest, macrophage culture and treatment with MAA beads or films. An in vitro

assay designed to investigate the effect of MAA on BMDM. (A) Bone marrow cells were harvested from

murine bone marrow, differentiated with M-CSF, and polarized with IFNγ or IL-4 on 10 cm petri dishes.

These cells were processed for flow cytometry to validate macrophage polarization (See Appendix, S5).

Additionally, non-polarized (M0) BMDM were re-plated onto 48 well plates, where they were treated

with MAA or MM beads, in the presence or absence of whole blood (0.5% v/v – 5 % v/v). These cells

were washed, fixed directly on the plate, and stained with DAPI, PE conjugated CD206 or PerCPe710

conjugated MHCII. The fluorescence signal was determined using a fluorescent plate reader. (B)

Schematic depicting the treatment of BMDM with MAA or MM films, instead of beads.

73

S10. Gating strategy for macrophages following inhibition of complement activation (day 7

shown; MAA beads). Single cells and live cells were gated from all the cells. Leukocytes

(CD45+) and non-leukocytes (CD45-) cells were gated from the live, single cell population.

From the CD45+ gate, neutrophils (Ly6G+) were removed. Then, from the Ly6G- gate,

macrophages were gated as F4/80+ and CD11b+. From the macrophage population, MHCII and

CD206 were used to distinguish between “M1” and “M2” macrophages, respectively.

74

S11. MAA beads modulated CD206, but not MHCII expression in the presence of blood.

(A) Macrophages cultured in IFNγ or IL-4 in 48 well plates and processed for fluorescence

scanning. As expected, M(IL-4) macrophages highly expressed CD206 but not MHCII while

M(IFNγ) highly expressed MHCII but not CD206. (B) BMDM treated with MAA or MM beads

for 24h in the presence or absence of 1% v/v whole mouse blood. Representative dot plots of

macrophages treated with MAA beads (C) or MM beads (D) and 0.5% whole blood (v/v) for

24h. Expression of CD206 (E) and MHCII (F) of macrophages treated with MAA or MM beads

in the presence of 0.5% v/v whole mouse blood. Treatment with MAA beads led to a small

increase in the expression of CD206, but not MHCII in BMDM. Conversely, treatment with MM

beads led to a small increase in the expression of MHCII, but not CD206. Uns = unstained.

75

S12. MAA films stimulated M2 marker Arg1 in M0 and M(IFNγ) cells. BMDM were

polarized, then treated with MAA or MM films for 72h. RNA from the cells were extracted,

reverse transcribed, then quantified. mRNA expression is expressed relative to NT controls and

ribosomal protein S18. (A) Arg-1 mRNA expression in BMDM polarized with IFNγ or IL-4 and

treated with MAA or MM films. (B) iNOS mRNA expression level in BMDM polarized with

IFNγ or IL-4 and treated with MAA or MM films. Treatment with MAA films increased the

expression of Arg-1, an M2 marker in M0 and M(IFNγ) cells. NT= no treatment. PCR

experiments conducted by I. Talior-Volodarsky.

76

S13. Administration of 4 mg/kg pentamidine or 2.5- 10 mg/kg ATA did not inhibit

complement activation over time. Serum from complement-inhibited mice were compared to

control mice administered 0.85% saline or normal chow (Ctrl). Complement-mediated red blood

cell hemolysis for animals administered 4 mg/kg pentamidine IP (A) or 2.5 – 10 mg/kg ATA SC

(B). (C) Efficiency of complement inhibition in ATA-treated animals, presented as the IC50. The

IC50 increased rapidly over time, indicating that the complement-inhibition effect of ATA was

short lived.

77

S14. Explant mass, cell number and normalized cell number in complement-inhibited

animals. (A) Mass of explants. (B) Numbers of cells collected by flow cytometry. (C)

Normalized cell numbers, obtained by dividing the cell number by the explant mass.

78

S15. Leukocytes and macrophage populations in complement-inhibited animals. (A-B)

Frequency (as a percentage of live cells) of CD45+ cells (A) and CD45-cells (B). (C) Frequency

(as a percentage of CD45+ cells) of Ly6G+ neutrophils. (D-F) Frequency (as a percentage of

F4/80+ macrophages) of CD206+ cells (D), CD86+ cells (E), and MHCII+ cells (F).

79

S16. Published manuscript: Lisovsky A, Zhang, DKY, Sefton MV, Biomaterials 2016. Reprinted

from [15] with permission from Elsevier.

lable at ScienceDirect

Biomaterials 98 (2016) 203e214

Contents lists avai

Biomaterials

journal homepage: www.elsevier .com/locate/biomater ia ls

Effect of methacrylic acid beads on the sonic hedgehog signalingpathway and macrophage polarization in a subcutaneous injectionmouse model

Alexandra Lisovsky a, 1, David K.Y. Zhang a, 1, Michael V. Sefton a, b, *

a Institute of Biomaterials and Biomedical Engineering, University of Toronto, 164 College Street, Suite 407, Toronto, Ontario, Canada M5S 3G9b Department of Chemical Engineering and Applied Chemistry, University of Toronto, 164 College Street, Suite 407, Toronto, Ontario, Canada M5S 3G9

a r t i c l e i n f o

Article history:Received 12 February 2016Received in revised form14 April 2016Accepted 20 April 2016Available online 4 May 2016

Keywords:Methacrylic acidSonic hedgehogMacrophage polarization

* Corresponding author. Institute of BiomaterialsUniversity of Toronto, 164 College Street, Suite 407, T3G9.

E-mail address: michael.sefton@utoronto.ca (M.V.1 These authors contributed equally.

http://dx.doi.org/10.1016/j.biomaterials.2016.04.0330142-9612/© 2016 Elsevier Ltd. All rights reserved.

a b s t r a c t

Poly(methacrylic acid-co-methyl methacrylate) (MAA) beads promote a vascular regenerative responsewhen used in diabetic wound healing. Previous studies reported that MAA beads modulated theexpression of sonic hedgehog (Shh) and inflammation related genes in diabetic wounds. The aim of thiswork was to follow up on these observations in a subcutaneous injection model to study the hostresponse in the absence of the confounding factors of diabetic wound healing. In this model, MAA beadsimproved vascularization in healthy mice of both sexes compared to control poly(methyl methacrylate)(MM) beads, with a stronger effect seen in males than females. MAA-induced vessels were perfusable, asevidenced from the CLARITY-processed images. In Shh-Cre-eGFP/Ptch1-LacZ non-diabetic transgenicmice, the increased vessel formation was accompanied by a higher density of cells expressing GFP (Shh)and b-Gal (patched 1, Ptch1) suggesting MAA enhanced the activation of the Shh pathway. Ptch1 is theShh receptor and a target of the pathway. MAA beads also modulated the inflammatory cell infiltrate inCD1 mice: more neutrophils and more macrophages were noted with MAA relative to MM beads at days1 and 7, respectively. In addition, MAA beads biased macrophages towards a MHCII�CD206þ (“M2”)polarization state. This study suggests that the Shh pathway and an altered inflammatory response aretwo elements of the complex mechanism whereby MAA-based biomaterials effect vascular regeneration.

© 2016 Elsevier Ltd. All rights reserved.

1. Introduction

Methacrylic acid (MAA)-based biomaterials have a vascularregenerative effect in the absence of exogenous cells or growthfactors [reviewed in Refs. [1,2]]. These biomaterials were previouslyshown to promote vascularization [3e6], and improve myocuta-neous graft survival [4] and diabetic wound healing [5]. Togetherwith past work [6e9], the present study was aimed at further un-derstanding the mechanism behind this effect. We presume thatsuch biomaterials drive an “alternative foreign body response” thatis distinct from the fibrosis associated with the classical foreignbody response.

Gene expression analysis of diabetic wounds treated with

and Biomedical Engineering,oronto, Ontario, Canada M5S

Sefton).

poly(methacrylic acid-co-methyl methacrylate) (MAA) beadsshowed an over fourfold upregulation in the expression of the sonichedgehog (Shh) gene [6], which has been implicated in adultvascularization [10,11]. The Shh pathway is also activated duringinflammation [12] and has been shown to polarize macrophagestowards an alternatively-activated, wound healing (“M2”) pheno-type [13]. Although, MAA-based biomaterials did not modulate theexpression of classical angiogenic genes (e.g., vascular endothelialgrowth factor (VEGF)) [6e8], MAA beads modulated inflammation-associated genes (e.g., interleukin 1b (IL1b)) in diabetic wounds [6],in an air pouch model [9] and macrophage-like cells (dTHP1 cells)in vitro [7,8]. A phosphoproteomics study of dTHP1 cells treated fora few minutes with a MAA-based biomaterial distinguished anumber of phosphorylated proteins involved in macrophage po-larization (e.g., solute transporter monocarboxylate transporter 4)among the many phosphorylated proteins that were differentiallyregulated between a MAA-based and a control biomaterial [14].Overall, these results led to the hypothesis that MAA-based bio-materials modulate the Shh signaling pathway and inflammatory

A. Lisovsky et al. / Biomaterials 98 (2016) 203e214204

cell responses, including macrophage polarization. Hence, a sub-cutaneous injection model was devised to investigate the effects ofMAA beads on the host response without the confounding factorsof diabetic wound healing. To investigate the activation of the Shhpathway specifically, MAA beads and control poly(methyl meth-acrylate) (MM) beads were injected subcutaneously in transgenicShh-Cre-eGFP/Ptch1-LacZ mice. In these mice the expression of thereporters, GFP and b-Gal, was shown to be consistent with thepattern of Shh and Ptch1 mRNA expression, respectively [15,16].The activation of the Shh signaling pathway was suggested as thedensities of both GFPþ (Shh) and b-Galþ (patched 1, Ptch1, thetarget of the pathway) cells were upregulated by treatment withMAA beads in this model. To investigate inflammatory cell re-sponses, bead implants were analyzed by immunohistochemistryand flow cytometry for the number and polarization state of infil-trating inflammatory cells. MAA beads increased the density ofneutrophils at day 1 and macrophages at day 7 and biased mac-rophages towards the MHCII�CD206þ state representative of the“M2” phenotype. We also compared the therapeutic effect of abiomaterial in both males and females illustrating a difference inresponse between sexes.

2. Methods

2.1. MAA and MM bead preparation

Poly(methacrylic acid-co-methyl methacrylate) (MAA-co-MMAor MAA) beads were composed of 45 mol% methacrylic acid(Sigma-Aldrich Canada Ltd., Oakville, ON, Canada), 1 mol% ethyleneglycol dimethacrylate (Sigma-Aldrich Canada Ltd.) and 64 mol%methyl methacrylate (Sigma-Aldrich Canada Ltd.). MAA beads weresynthesized by suspension polymerization as previously described[4] and were sieved to obtain beads in the diameter range of150e250 mm. Methacrylic acid content of the synthesized beadswas confirmed by titration. Control poly(methyl methacrylate)(MM) beads (same diameter) were obtained from Polysciences(Warrington, PA). Beads were washed in either 95% ethanol (MAAbeads) or 1 N HCl (MM beads) repeatedly and then rinsed five timesin LAL reagentwater (MJS Biolynx Inc., Brockville, ON, Canada) priorto use in vivo. Analysis with a limulus amebocyte lysate (LAL)pyrochrome endotoxin test kit (Cape Cod Inc., Falmouth, MA)indicated that beads contained <0.25 EU/100mg. Elemental surfacecomposition analysis (ThermoFisher XPS, Surface-InterfaceOntario, University of Toronto) showed minimal Si contamination(~0.07%) and that measured surface composition (atom%) was closeto the theoretical expectation. MAA beads had a rough, poroussurface, were negatively charged and did not degrade over timein vivo; MM beads were smooth [4,5].

For subcutaneous injections, a 1 mL syringe with an 18 gaugeneedlewas loadedwith either 5mgMAA beads or 15mgMMbeads(or no beads, vehicle control) suspended in 250 mL of 50% w/vpolyethylene glycol (PEG, avg. mol. wt. 1450, sterile-filtered; Sigma-Aldrich Canada Ltd.) in PBS. The 1:3 wt ratio (5 mg MAA:15 mg MM) was used to account for MAA bead swelling upon hy-dration at physiological pH [4] to approximately equate implantedvolumes. Vehicle control was used only for the flow cytometrystudy because the vehicle control implant area could not be definedreproducibly for vessel and cell density analyses.

2.2. Animals

All animal work was done with the approval of the University ofToronto Animal Care Committee. Animals were housed understerile conditions in the University of Toronto's Department ofComparative Medicine. The experiments were done with CD1 mice

(6e8 week old, males, Charles River Laboratories, MA) and Shh-Cre-eGFP/Ptch1-LacZ mice (10e12 week old, males and females).Shh-Cre-eGFP/Ptch1-LacZ heterozygous mice of CD1 backgroundwere bred in house by crossing CD1 females (Charles River Labo-ratories, MA or bred in house) with Shh-Cre-eGFP/Ptch1-LacZheterozygous males. The original Shh-Cre-eGFP/Ptch1-LacZ malewas donated by Professor Chi-chung Hui (Hospital for Sick Chil-dren, Toronto, ON, Canada) and created by crossing Shh-Cre-eGFP[15] with Ptch1-LacZ [16] mice. Shh-Cre-eGFP mice were createdby inserting a gfpcre cassette at the ATG of Shh; expression of GFPprotein was reported to colocalize with Shh mRNA [15]. Ptch1-LacZmice were developed by inserting the LacZ gene into Ptch1; theexpression of the reporter was consistent with the pattern of Ptch1transcription [16].

Mice were genotyped to detect the presence of Shh-Cre-eGFPand Ptch1-LacZ mutations. DNA from ear notches was extractedby alkaline lysis. Primers (Supplementary Information, Table 1)were synthesized by Sigma Genosys (Sigma-Aldrich Canada Ltd.)and prepared by resuspension in RNase/DNase free water. PCR re-actions are detailed in the Supplementary Methods [80].

2.3. Subcutaneous injection

Mice were anesthetized with 0.5% w/v isofluorane prior tosurgery and an analgesic (Ketoprofen, 5 mg/kg) was administeredintraoperatively. The dorsal area of a mouse was shaved and theremaining hair was removed either by waxing (Nair wax strips) orby hair removal cream (Veet). The skin was sterilized with 70%ethanol and Betadine. An 18-gauge needle was used to inject MAA,control MM beads or vehicle (PEG). Two injections on either side ofthe dorsumwere performed for each mouse. A small subcutaneouspocket was made with the needle on the side of the dorsum bymoving the syringe from side to side, while deliberately attemptingto nick small blood vessels to promote injury prior to injection(Fig.1). Following surgery, mice were housed individually, fed chowand water ad libitum, and monitored for any signs of discomfort. At1e7 days post-injection, the mice were sacrificed using CO2, fol-lowed by cervical dislocation. The implants were removed surgi-cally and processed for histology, imaging or flow cytometry.

2.4. Histology and immunohistochemistry

Immediately upon euthanizing mice, the bead implant andseveral mm of surrounding tissue was excised from the right side ofthe dorsum and fixed in formalin. Tissue samples were embeddedin deep paraffin blocks, cut into sections, processed and stainedwith hematoxylin and eosin (H&E), Masson's trichrome, CD31, GFP,b-galactosidase (b-Gal), F4/80 and CD206 (SupplementaryInformation, Table 2). Histology slides were scanned (20�) usingan Aperio ScanScope XT (LeicaMicrosystems, Concord, ON, Canada)by the Advanced Optical Microscopy Facility (AOMF, Toronto, ON,Canada).

The scanned slides were analyzed using Aperio ImageScope(Version 11) at 4 and 7 days post-implantation. For vessel counts, aregion of interest (ROI) was defined by measuring a distance of500 mm around each cluster of beads. CD31þ vessel-like structures(criterion being the presence of a lumen)were counted in the tissuewithin this defined region. The vessel density was calculated bydividing the total number of vessels by area of the ROI. For othercounts (GFP, b-Gal, F4/80 and CD206), a distance of 200 mm aroundbead clusters was used to define the ROI. The cell density wascalculated by dividing the total number of positive cells by the areaof the ROI.

Fig. 1. Subcutaneous injection model for investigating cellular and molecular mechanisms of vascular regenerative MAA-based biomaterials. MAA beads, control MM beads orvehicle (PEG) were injected on both sides of the dorsum of mice. At 1e7 days post-injection, the tissues were removed and processed for histology, imaging and flow cytometry. Inthe future, the model will also be used for molecular analysis and inhibition studies. The image is a scanning electron micrograph of MAA beads.

A. Lisovsky et al. / Biomaterials 98 (2016) 203e214 205

2.5. CLARITY preparation and imaging

Seven days following subcutaneous injection of MAA or MMbeads, Alexa 647-conjugated lectin (GSL-1: Griffonia (Bandeiraea)Simplicifolia; 100 mg in 150 mL PBS; Vector Laboratories, Burlington,ON, Canada) was injected via tail vein 10 min prior to sacrifice.Fluorophore conjugation was performed in-house using Alexa 647modified with a NHS-ester chemistry [17]. GSL-1 is a mouse-specific lectin which binds to the galactosyl residues of mouseendothelial cells, enabling labeling and visualization of the mousevasculature [18]. The implants with the surrounding subcutaneoustissue were removed surgically and processed using a modifiedCLARITYprotocol for confocal imaging [19]. Explants were fixed in asolution containing 4% acrylamide (Sigma-Aldrich Canada Ltd.), 4%paraformaldehyde, 0.05% bis-acrylamide (Sigma-Aldrich CanadaLtd.) and 0.25% (w/v) VA-044 thermal initiator (Sigma-AldrichCanada Ltd). After one week of incubation, the acrylamide waspolymerized at 37 �C for 3 h. Polyacrylamide-embedded explantswere cleared for 14 days at 50 �C in the clearing solution (8% SDS inborate buffer, pH 8.5; eBioscience, San Diego, CA), which waschanged every 2nd day. Post-clearing, the explants were counter-stained with SYTOX green nucleic acid stain (100 pmol/mg; LifeTechnologies, Burlington, ON, Canada) for 48 h. Refractive indexmatching was performed by infusing explants with 70e75% 2,20-thiodiethanol in borate (adjusted to pH 10; Sigma-Aldrich CanadaLtd.) [21]. Explants were imaged using a Nikon A1 confocal mi-croscope (Nikon, Melville, NY) at the Center for Microfluidics Sys-tems (University of Toronto).

2.6. Digestion of tissue explants

Subcutaneous tissue containing injected beads was separatedfrom the skin and muscle layers. For PEG samples, subcutaneoustissuewas explanted in the samemanner using the injection needlewound site as a guide. Tissues were weighed and then digestedfollowing a previously described digestion protocol [22]. Briefly,samples were finely minced in 500 mL of 1 X HBSS containing450 U/mL collagenase I (Sigma-Aldrich Canada Ltd.), 125 U/mLcollagenase XI (Sigma-Aldrich Canada Ltd.), 60 U/mL DNase I

(Sigma-Aldrich Canada Ltd.), 60 U/mL hyaluronidase (Sigma-Aldrich Canada Ltd.) and 1 M HEPES (Sigma-Aldrich Canada Ltd.).The sample was incubated in this solution for 1 h at 37 �C andhomogenized using a gentleMACS Octo Dissociator (Miltenyi BiotecInc., San Diego, CA). Tissue was further digested for 60 min at 37 �Cand 250 rpm. The cell suspension was filtered using a 40 mm cellstrainer (Fisher Scientific, Ottawa, ON, Canada) to remove beadsand debris. The remaining cells were washed in PBS supplementedwith 0.5% BSA and 2 mM EDTA, pelleted and stained with live/deadstain, CD11b, CD11c, CD206, CD31, CD45, CD86, F4/80, Ly6G, MHCII(Supplementary Information, Table 2). All antibodies were dilutedaccording to the manufacturers' recommendations and titrated inhouse to optimize staining.

2.7. Flow cytometry analysis of tissue explants

The gating strategy (Supplementary Information, Fig. 1) was asfollows: after isolating live single cells, CD45 distinguished leuko-cytes from non-leukocytes. Neutrophils were identified as Ly6Gþand dendritic cells as CD11cþLy6G�. Macrophages were firstidentified as CD11c�Ly6G�F4/80þCD11bþ and then further char-acterized as MHCIIþ CD206� (“M1”) and MHCII�CD206þ (“M2”).Endothelial cells were identified as CD31þCD45�. Cells were gatedaccording to positive staining for each antibody using fluorescenceminus one (FMO) controls. Cell populations were expressed eitheras a percentage or as a normalized value: estimated total number ofcells divided by the weight of the explanted tissue. The number ofcells was estimated from the flow cytometry results with 123counteBeads (eBioscience) used to determine cell recovery (~50%).

2.8. Statistical analysis

All data are presented as mean ± standard error of the mean(SEM). Statistical analysis was performed using IBM SPSS Statistics(Version 22). Levene's test was used to test for equality of varianceamong samples. Statistical comparison between two treatmentgroups (MAA, MM) was performed using independent t-test forsamples with equal variances (Levene's test: p > 0.05) and Mann-Whitney test for samples with significantly different variances

A. Lisovsky et al. / Biomaterials 98 (2016) 203e214206

(Levene's test: p < 0.05). Statistical comparison among threetreatment groups (MAA, MM, PEG) was performed using two-wayANOVA followed by Tukey's post hoc test for significance. The levelof statistical significance was set at p < 0.05.

3. Results

3.1. Subcutaneous injection model

Previously, the vascular regenerative MAA beads were investi-gated in vivo in wounds of diabetic db/db mice [5,6]. During woundhealing, MAA beads became entrapped in a scab, which at times felloff prior to the studied time points, preventing the identification ofthe cells that were associated with the beads. A subcutaneous in-jection model was developed to directly interrogate cellular re-sponses and vascularization, while simplifying the host response;injections were less invasive compared to full-thickness wounds.An additional advantage of this model was the ability to studyMAA-mediated changes in the cellular infiltrate. Consistent withthe previous studies with MAA-based biomaterials that had onlyused males, flow cytometry studies were also done only in males.However, since females have been shown to respond differently tovarious diseases and treatments [reviewed in Refs. [23,24]], histo-logical analyses were done in bothmale and female transgenicmiceand sex differences were noted.

The earlier study showed that Shh gene expression was upre-gulated more than fourfold compared to the controls (MM beadsand no treatment) in diabetic wounds [6], but this did not showthat the pathway was indeed engaged. Motivated by this result, weinvestigated the expression of the Shh pathway in non-diabeticShh-eGFP-Cre/Ptch1-LacZ CD1 mice. In these double heterozy-gous mice, the expression of reporters (GFP and b-Gal) was shownto be consistent with the patterns of Shh and Ptch1 mRNA expres-sion [15,16], thus allowing the investigation of cells expressing GFP(Shh) and b-Gal (Ptch1) in the tissue surrounding the beads. Weused the corresponding CD1 mice to investigate the effect of MAAbeads on inflammatory cells, with a focus on macrophages.

3.2. The effect of MAA beads on vessel density in males and females

As expected, MAA beads promoted vascularization in the tissuesurrounding the beads in non-diabetic mice (Fig. 2A,Supplementary Information, Fig. 2). MAA beads increased CD31þvessel formation at day 7 in both males (p ¼ 0.013, Fig. 2B) andfemales (p ¼ 0.009, Fig. 2C), although the effect in females was notas pronounced as in males. At day 4, MAA beads appeared to in-crease the CD31þ vessel density in males, but this difference(relative toMM) was not statistically significant (p¼ 0.085, Fig. 2B).Compared to females, males had an increased level of vascularity atday 4 (p ¼ 0.024) and similar vessel densities at day 7 in MAA-treated mice. No differences were noted between sexes in theirvascular response to control MM beads.

CD1 male mice were injected (via the tail vein) at day 7 with amouse Alexa 647-conjugated lectin (GSL1, a lectin that labels themouse endothelium) to visualize perfused blood vessels in the vi-cinity of the beads (Fig. 2D). To increase the depth of imaging, ex-plants were processed using a CLARITY protocol [20]. To ourknowledge, this is the first use of the technique to investigate host-biomaterial interactions. Explanted tissues fromMAA-treated miceshowed high levels of Alexa 647-GSL1 staining surrounding MAAbeads, consistent with the greater levels of vascularizationobserved with the histological analysis. In explanted tissues fromMM-treated mice, lectin staining was not seen around the beadsand was primarily seen in the skin far from the beads. A thick layerof cells surrounded MM (Fig. 2D) but not MAA beads, suggested an

alternative cellular response. Similarly, a dense layer of cells wasseen in Masson's trichrome images (Fig. 2A). Taken together, theseresults suggested that MAA beads were effective in promotingvascularization in both sexes (in non-diabetic mice) and that theMAA-induced vessels were perfusable.

3.3. The effect of MAA beads on the Shh signaling pathway

After the subcutaneous injection of beads in transgenic mice,GFPþ and b-Galþ cells were found in close proximity to theinjected beads (Fig. 3A and B, Supplementary Information, Fig. 3).At day 4, MAA beads increased the expression of GFPþ (p ¼ 0.001)(Fig. 3C) but not b-Galþ cells relative to control MM beads (Fig. 3D).At this time, both males and females showed similar numbers ofGFPþ and b-Galþ cells, whether animals were treated with MAA orMMbeads; thus, for day 4 the data for both sexes were combined toachieve a higher n. At day 7, the MAA-induced upregulation of GFPwas seen in females (and not males); the increase (p ¼ 0.051) justmissed our criterion for statistical significance (Fig. 3C). MAA-treated mice showed greater b-Gal protein expression at day 7 inmales (p ¼ 0.026) but not females (Fig. 3D). There was a significantincrease in b-Gal (p < 0.001) from day 4 to 7 and no change wasnoted for GFP over time (Fig. 3C and D). Together, these data sug-gested thatMAA beads enhanced the activation of the Shh signalingpathway, as Ptch1 is the target of the Shh signaling pathway[reviewed in Ref. [25]].

Some of the GFPþ and b-Galþ cells exhibited a rounded shape(Fig. 3A and B), suggesting that macrophages may have beeninvolved in the Shh pathway modulation with MAA beads.Consistent with this observation, in vitro mouse bone marrow-derived macrophages treated with MAA-based films increasedthe expression of the Shh gene more than fourfold compared tomacrophages left untreated (Supplementary Information, Fig. 4,p ¼ 0.035); control MM films did not upregulate Shh.

3.4. The effect of MAA beads on inflammatory cell recruitment

To further investigate the effect of MAA beads on macrophages,the pan macrophage marker F4/80 and M2 marker CD206 wereused to quantify macrophages surrounding beads after subcu-taneous injection (Fig. 4A and B). A dense ring of F4/80þ cells wasseen surrounding MM beads (Fig. 4A), similar to that present in theMasson's trichrome and CLARITY-processed images (Fig. 2A and D);this dense ring was rarely seenwith MAA beads. Nonetheless, therewas no difference in F4/80þ cell density between MAA and MMbeads at either time point (Fig. 4C) consistent with a differentdistribution of cells; no differences were seen in comparing malesand females. CD206þ cells were closely associated with both beadtypes (Fig. 4B). Treatment with MAA beads resulted in a higher celldensity of CD206þ cells in females (and not males) at day 4(p ¼ 0.013) compared to MM beads (Fig. 4D). There was an increasein CD206þ cells in males at day 7 with MAA beads (p ¼ 0.066), butthe increase was not statistically significant (Fig. 4D); there was nodifference at day 7 in females.

Flow cytometry was used to follow up on these observations. Anearlier time point (day 1) was added to investigate neutrophilrecruitment. There was no statistical difference in the mass of tis-sue explants (Supplementary Information, Fig. 5A), estimated totalnumbers of collected cells (Supplementary Information, Fig. 5B)and normalized cell numbers (Supplementary Information, Fig. 5C)among treatment groups (MM beads, MAA beads, and PEG vehicle)at all studied time points. Using the gating strategy illustrated inthe Supplementary Fig. 1, higher densities of CD45þ cells werenoted in both bead explants compared to the PEG vehicle control atday 1 (Fig. 5A; Supplementary Information, Fig. 6A) with a

Fig. 2. Vessel formation in mice injected with vascular regenerative MAA beads. (A) Histology sections of Shh-eGFP-Cre/Ptch1-LacZ CD1 mice treated with MAA or MM beads at day7 stained with CD31 (left) and Masson's trichrome (right). Arrows indicate examples of vessels. (B, C) Tissues treated with MAA beads in transgenic mice had a significantly highervessel density in males (B) and females (C) at day 7. (D) Confocal microscopy image of CLARITY-processed tissues treated with MAA and MM beads from non-transgenic CD1 micestained with Alexa 647-GSL1, a lectin specific for mouse vasculature, and Sytox Green. Perfused vessels weaved around MAA beads but not MM beads. Most of the vessels in MM-treated mice were found in the skin further away from the beads. Scale bars ¼ 200 mm. n ¼ 3 (day 4), n ¼ 6 (day 7); *p < 0.05; **p < 0.01. (For interpretation of the references tocolour in this figure legend, the reader is referred to the web version of this article.)

A. Lisovsky et al. / Biomaterials 98 (2016) 203e214 207

corresponding higher number of CD45� (cells other than leuko-cytes) in vehicle explants (Fig. 5B; Supplementary Information,Fig. 6B). More CD45� cells were noted in tissues treated withMM compared to MAA beads at day 7 (Fig. 5B).

For the biomaterial treatment groups, neutrophils were mostprevalent at day 1 post-injection and as expected, their estimatednumbers declined dramatically at days 3 and 7 (Fig. 5C). MAA beadssignificantly increased neutrophil recruitment (p < 0.0001)compared to both controls at day 1 (Fig. 5C). More macrophageswere found in MAA-treated mice compared to both MM and PEGcontrols at day 7 (p < 0.05) (Fig. 5D). The estimated number ofmacrophages in theMM- and PEG-treatedmice decreased from day3 to day 7; this effect was not observed with MAA beads (Fig. 5D).The intensity of F4/80 expression by macrophages varied dramat-ically from day 1 to days 3 and 7 (Fig. 5E). Endothelial and dendriticcells were also quantified by flow cytometry. No statistically sig-nificant differences were noted in the density of endothelial cellsamong the treatment groups (Supplementary Information, Fig. 6C).At day 7, the frequency of dendritic cells increased in explants forboth bead types (Supplementary Information, Fig. 6D) relative tovehicle control. Overall, these results indicated a differential in-flammatory cell response to MAA beads compared to biomaterialand vehicle controls. Treatment with MAA beads increased therecruitment of neutrophils at day 1 and the number of

macrophages at day 7.

3.5. The effect of MAA beads on macrophage polarization

The flow cytometry protocol was used to distinguish macro-phage polarization states: “M1” using MHCII and CD86 as markersand “M2” using CD206 [reviewed in Refs. [26e28]] (SupplementaryInformation, Fig. 7). While the expression of CD86 was similaramong all three treatment groups (Supplementary Information,Fig. 7B), there were some significant differences in the numberand frequency of macrophages expressing MHCII (SupplementaryInformation, Fig. 7C) and CD206 (Supplementary Information,Fig. 7D); CD86 was excluded from further analyses.

The polarization bias was reflected in the representative dotplots for both biomaterials (Fig. 6A and B). Treatment with MAAbeads significantly increased MHCII-CD206þ (M2) macrophagesand decreased MHCIIþCD206� (M1) macrophages compared toboth MM and PEG controls at day 7 (p < 0.01) (Fig. 6C). Conversely,treatment with control MM beads had an opposite effect withsignificantly more M1 and fewer M2 macrophages relative to MAAbeads at day 7 (p < 0.01) (Fig. 6D). By day 7, the majority of mac-rophages in MM-treated mice expressed either both markers(MHCIIþCD206þ) or neither marker (MHCII�CD206�) (Fig. 6E andF). The frequency of cells expressing a double-positive phenotype

Fig. 3. Modulation of the GFPþ and b-Galþ cell density after subcutaneous injection of MAA beads in Shh-eGFP-Cre/Ptch1-LacZ CD1 mice. (A, B) Serial sections of day 4 tissuestreated with MAA beads, stained for GFP (A) or b-Gal (B). Arrows show examples of cells positive for the marker of interest. (C, D) Density of GFPþ (C) and b-Galþ (D) cells at days 4and 7. Treatment with MAA beads increased the density of GFP-expressing cells but not b-Gal-expressing cells at day 4; there was no difference between males and females at day 4.At day 7, the density of cells expressing b-Galþ was upregulated in males (but not females) and the density of cells expressing GFP was greater in MAA-treated females but thedifference was not significant (p ¼ 0.051). There were more b-Galþ cells at day 7 than at day 4 in both sexes and with both materials; there was no substantive effect of time on thenumber of GFPþ cells. Scale bars ¼ 200 mm. n ¼ 6; *p < 0.05; **p < 0.01.

A. Lisovsky et al. / Biomaterials 98 (2016) 203e214208

increased progressively from day 1 to day 7 in MM but not in MAA-treated mice (Fig. 6F). On the other hand, in the MAA-treated mice,the majority of macrophages were consistently MHCII�CD206þ(M2) from day 3 onwards to day 7 (Fig. 6E and F). Together, theseresults suggested that MAA beads biased macrophages towards aM2 polarization state.

4. Discussion

4.1. The effect of MAA beads on vascularization

Previous studies in diabetic male mice (BKS.Cg-mþ/þ Leprdb/Jmice, db/db) showed that MAA beads increased vascularization incutaneous wounds [5,6]. In agreement with these studies, subcu-taneous injection of MAA beads increased vessel density in non-diabetic mice compared to control MM beads (Fig. 2), high-lighting the vascular potency of MAA beads even in the absence ofthe physiological need present during diabetic wound healing.MAA beads nearly doubled the number of vessels at day 7 (~90%increase) (Fig. 2B) in males. Other synthetic biomaterials have alsobeen shown to improve vascularization without the addition ofexogenous factors [29e31]. For example, changing porosity(decrease from 60 mm to 30 mm) resulted in ~40% increase in vesselformation with poly(2-hydroxyethyl methacrylate, poly(HEMA))scaffolds implanted into the myocardium of male rats [29].

In females, the effect of MAAwas more modest with an increaseof only about 30% at the same time point (Fig. 2C). The strongerMAA-mediated vascular response seen in males was most likelydue to sex-based hormone differences [32,33]. The Shh pathway isregulated by estrogen [34,35] and hence differences in the extent ofShh pathway activation between males and females may accountfor the higher vascularization induced by MAA beads. Males had ahigher density of b-Galþ (Ptch1) cells than females at day 7(Fig. 3D). The inclusion of females in this study increased ourappreciation for the effect of sex on the host response to MAA-based biomaterials, consistent with efforts to decrease sex biasand improving the quality and reproducibility of preclinical

research [reviewed in Refs. [36,37]], here in the context of thera-peutic biomaterials design.

To investigate the perfusability of newly formed vesselsfollowing the MAA bead treatment, the bead explants were pre-pared using the CLARITY protocol (Fig. 2D). During CLARITY pro-cessing, fatty lipids were removed while proteinaceous structuresand morphology were retained, enabling deep imaging and 3Dvisualization of these fragile tissues [20,38]. Alexa 647-GSL1 (viatail vein injection) staining was only observed aroundMAA and notMM beads (Fig. 2D) indicating that the MAA-induced vessels wereperfusable.

4.2. The effect of MAA beads on Shh signaling

During Shh pathway activation, binding of Shh to its receptorPtch1 relieves the inhibitory effect of Ptch1 on Smoothenedresulting in the activation of the Gli transcription factors respon-sible for downstream modulation of Shh target genes [reviewed inRef. [25]]. Previously, MAA beads were shown to upregulate Shh(and perhaps Gli3) gene expression in diabetic wounds at day 4(although the latter was not significant, p ¼ 0.052) [6]; Gli3 is atranscription factor associated with the angiogenic effect of the Shhpathway [39]. Because the diabetic wound healing milieu wascomplex, it was challenging to investigate the differential expres-sion of other genes associated with the Shh pathway in thesewounds. The transgenic Shh-Cre-eGFP/Ptch1-LacZ mouse modelwas used to enable quantification of cells expressing GFP or b-Galproteins (Fig. 3A and B), rather than just gene expression within awound as a whole. This study suggested, for the first time, that theShh pathway was indeed modulated by MAA beads: MAA-treatedmice had a significant increase in the expression of both GFP(Shh) and b-Gal (the target of Shh signaling) (Fig. 3C and D). Theactivation of the Shh signaling pathway is presumed to be oneaspect of the mechanism of the vascular regenerative effect ofMAA-based biomaterials.

Shh is well known for its role in embryogenesis, carcinogenesis[reviewed in Refs. [25,40]] and more recently for its involvement in

Fig. 4. Macrophage infiltration after treatment with MAA beads, by histology. (A) Tissue section from MAA- and MM-treated mice stained with pan macrophage F4/80 marker atday 4. A dense ring of F4/80þ cells (macrophages) surrounded control MM beads but not MAA beads. (B) Serial sections stained with F4/80 and CD206 (M2 macrophage marker)fromMAA-treated mice at day 4. CD206þ cells were closely associated with the beads. Arrows show examples of cells positive for the marker of interest. (C, D) Density of F4/80þ (C)and CD206þ (D) cells in tissues following treatment with MAA and MM beads at days 4 and 7 in males and females. CD206þ cell density was increased in females at day 4. Therewas no substantive effect of time on the number of F4/80þ and CD206þ cells in both sexes in MAA-treated mice. Scale bars ¼ 200 mm. n ¼ 6 (except for CD206 at day 4, n ¼ 3);*p < 0.05.

A. Lisovsky et al. / Biomaterials 98 (2016) 203e214 209

Fig. 5. Analysis of inflammatory cells in tissues treated with MAA beads, MM beads or PEG vehicle (CD1 mice). (AeD) Estimated number of CD45þ leukocytes (A), CD45� cells otherthan leukocytes (B), Ly6GþCD11bþCD45þ neutrophils (C) and F4/80þ CD11c-Ly6G-CD11bþCD45þ macrophages (D). MAA beads significantly increased the number of CD45þ cells(A) and neutrophils (C) at day 1 and macrophages at day 7 (D), while decreasing the number of CD45� cells at day 7 (B). (E) F4/80 and CD11b expression in CD45þ cells at day 1 andday 7; note the F4/80 mean fluorescent intensity increased over time. The F4/80 and CD11b gate (black box) was set based on fluorescence minus one (FMO) negative controls. n ¼ 4(except MM and PEG for day 1, n ¼ 3); *p < 0.05, ***p < 0.001.

A. Lisovsky et al. / Biomaterials 98 (2016) 203e214210

adult vascularization, inflammation andwound healing [10,41e43].Intramuscular delivery of Shh improved blood flow in a limb[10,44] while inhibition of endogenous Shh abolished angiogenesis[45] in mouse models of hind limb ischemia. The Shh pathway wasalso shown to be activated in skeletal muscle post-injury and itsinhibition resulted in decreased vascularity and increased fibrosisand inflammation [46]. During ischemic injuries, onemechanism ofShh-induced vascularization is via upregulation of angiogenic andvasculogenic factors (e.g., VEGF, angiopoietins) [44,46e48]. Inter-estingly, our previous studies indicated that MAA-based bio-materials did not alter the expression of classical angiogenic genes(e.g., VEGF) in diabetic wounds in vivo [6] and in macrophages(dTHP1) [7,8] and endothelial cells (HUVEC) [7] in vitro. At the sametime, MAA-based biomaterials upregulated inflammation-associated genes [6e9] and earlier in vivo work in diabeticwounds [6] and an air pouch model [9] suggested that MAA beadsaltered the foreign body response.

As such, the MAA-mediated Shh vascular effect may haveinvolved inflammation [11,12]. Qualitative observations of GFPþ(Shh) and b-Galþ (Ptch1) cells indicated that some of the cellspositive for these markers had round macrophage-like cellmorphology (Fig. 3A and B). Additionally, in vitro treatment of bonemarrow derived macrophages with MAA-based films upregulatedthe Shh gene fourfold compared to no treatment (no film) control(Supplementary Information, Fig. 4). The Shh signaling pathwayhas been shown to be activated in inflammatory cells in response toinflammation during injury [43] and host-pathogen interactions[12,13]. LPS upregulated Shh expression in brain astrocytes [43] anda monocyte cell line (THP-1) [49], while infection with a pathogeninduced upregulation of Shh, Ptch1 and Gli transcription factors inmacrophages [12].

This study demonstrated that MAA beads increased the numberof inflammatory cells (Fig. 5A) at day 1 compared to biomaterial andvehicle controls indicative of a stronger initial inflammatory

response. In addition, previous work showed that MAA beads [8]and films [7] upregulated two important mediators of inflamma-tion e tissue necrosis factor a (TNFa) [reviewed in Ref. [50]] andinterleukin 1b (IL1b) [reviewed in Ref. [51]], both of which havebeen implicated in modulation of Shh signaling [12,43,52,53]. Shhpathway activation was decreased in TNFa-null macrophages [12]and increased with exogenous TNFa treatment [12,53]. In amouse model with depleted CD11bþ macrophages, the Shhresponse was reduced but partially rescued by the injection of IL1bat the time of injury [43]. Thus, Shh signaling may have beenupregulated by MAA-mediated inflammatory signals. At the sametime, once activated, the Shh pathway has also been implicated inmodulating inflammation [12,54,55], specifically macrophage po-larization towards the M2 state [13] (discussed in Section 4.4).

4.3. The effect of MAA beads on inflammatory cell infiltration

MAA beads altered the inflammatory cell landscape relative tocontrols (Fig. 5). As expected, the presence of a biomaterial (eitherMAA or MM beads) resulted in more CD45þ leukocytes relative tothe PEG vehicle control (Supplementary Information, Fig. 6A). Afourfold increase in Ly6Gþ neutrophils was evident at day 1 inmiceinjected with MAA beads relative to both controls (Fig. 5C). The linkbetween MAA beads and neutrophil infiltration is not well under-stood but may have been a result of protein (e.g., complement)adsorption differences [56]. The significance of the increase inCD11cþ dendritic cells (Supplementary Information, Fig. 6D) is notclear. One caveat with the data is that the total number of cells wasdetermined by flow cytometry, with calibration beads used todetermine the ratio between number of events and number of cells.Cell numbers were further normalized by the mass of the explants,recognizing that the volume of tissue that was digested varied fromsample to sample. The reported numbers are reasonable estimatesof cell numbers, recognizing that we are interested in differences in

Fig. 6. Analysis of polarization states in recovered explants (CD1 mice). (A, B) Representative dot plots of F4/80þ cells (macrophages) at day 3 in mice treated with MAA (A) and MMbeads (B). (C, D) The number and frequency of the individual single positive, double positive, and double negative MHCII or CD206 macrophage populations in mice treated withMAA beads, MM beads or PEG vehicle control. (C) Normalized number and frequency of MHCII-CD206þ (“M2”) macrophages. (D) Normalized number and frequency ofMHCIIþ CD206- (“M1”) macrophages. MAA beads biased macrophages towards a M2 polarization state; noted by a progressive increase in M2 macrophages, relative to MM beads.(E, F) Distribution of polarized macrophages: normalized number (E) and frequency (F) of macrophages that were MHCII-CD206þ, MHCIIþCD206þ, MHCIIþCD206�, andMHCII�CD206�. n ¼ 4 (except MM and PEG for day 1, n ¼ 3); *p < 0.05, **p < 0.01, ***p < 0.001.

A. Lisovsky et al. / Biomaterials 98 (2016) 203e214 211

inflammatory cell infiltration over the course of the study. Thenormalization protocol may account for the apparent increase inCD45� cells seen with PEG at day 1 (Fig. 5B). Explant masses andtotal cell numbers were low (Supplementary Information Fig. 5)with the vehicle-only controls so that after normalization, thenormalized numbers were artificially high. Following PEG treat-ment, the numbers of CD45þ leukocytes and CD45� non-leuko-cytes were unchanged from day 1 to day 7, as expected.

Most importantly for this analysis, MAA beads increased thenumber of macrophages relative toMMbeads at day 7 (Figs. 4D and5D), consistent with the earlier observations of higher expression ofTNFa and IL1b genes in diabetic wounds at the same time point [6].The increase with histological analysis was not statistically signif-icant, presumably because of different regions of interest or highersensitivity of flow cytometry to identify cells with low F4/80expression. For all treatment groups, themean fluorescent intensityof F4/80 increased from day 1 to 7 (Fig. 5E), suggesting increasedmaturation [57].

In contrast to MAA beads, control MM beads were surroundedby a thick layer of macrophages (Fig. 4A); a common observationwith implanted biomaterials [reviewed in Refs. [58,59]]. Similarly, athick layer of cells was observed in Mason's trichrome andCLARITY-processed images around MM but not MAA beads (Fig. 2Aand D). At day 7, MM beads also had a higher density of CD45� cells(Fig. 5B), a majority of which were believed to be fibroblasts.Overall, the distribution of F4/80 staining, the presence of a thicklayer of cells and a higher number of CD45� cells suggested anincreased fibrotic response to control MM beads [60,61], a featureof a conventional foreign body response [62e64]. On the otherhand, the vascular regenerative MAA beads lacked a thick layer ofcells (Fig. 2A and D) and maintained low levels of CD45� cells(Fig. 5B); these are indicative of what we have termed as an alter-native foreign body response.

Since macrophages play an important role in vascularization[65e67], the results presented here supported our premise thatmacrophages were one of the orchestrators of vessel formation

A. Lisovsky et al. / Biomaterials 98 (2016) 203e214212

with MAA beads.

4.4. The effect of MAA beads on macrophage polarization

Macrophage phenotype varies depending on the conditions thathave led to their activation [68]. However, it is still convenient touse a distinction between “M1” and “M2” cells as a simplification ofthe spectrum of polarization states. While primarily considered asclassically activated inflammatory macrophages, M1 macrophageshave a role in initiating vessel formation [60] and M2 macrophagesthat arise later in the foreign body response are involved in pro-moting vessel maturation [60,69,70]. Improved vascularizationcorrelated with increased numbers of M2 macrophages [71];however, exogenous administration of M2 macrophages 1e3 dayspost-injury failed to improve vascularization in a cutaneous woundmodel [72], although this may have reflected changes that occur inpre-polarized macrophages upon implantation. We hypothesizedthat MAA beads polarized macrophages to a M2 state, consistentwith the increased vascularization.

Indeed, treatment with MAA beads induced a M2 macrophagepolarization bias: therewas an increased density of CD206þ cells inproximity to the beads (within 200 mm) in females but not males atday 4 (Fig. 4CeE). An increase in CD206þ cell density was sug-gested in males at day 7 but no difference was noted in females.These results highlighted the sex differences in macrophage po-larization observed by others [73,74] and may have been an un-derlying reason for differential MAA-mediated vascularizationresponse between males and females (Fig. 2B and C).

Flow cytometry analysis allowed quantification of macrophagesthat were CD206þ, but additionally allowed for discriminationbetween those that were also MHCIIþ or MHCII�. TermingMHCIIþCD206� cells as M1 cells and MHCII�CD206þ cells as M2,at day 7, MAA beads increased the density of M2 cells fourfoldcompared to MM beads (Fig. 6C), while MM beads induced a nearlynine-fold increase in M1 macrophages relative to MAA beads(Fig. 6D). Treatment with both controls (MM beads and PEGvehicle) elicited a more inflammatory macrophage response, withhigher numbers of M1 or double positive MHCIIþCD206þ macro-phages by day 7, relative to MAA beads (Fig. 6E and F). We presumethat these double positive cells are those in transition from theinitial inflammatory M1 cells to the later M2 cells, but additionalresearch is required to understand the role of these “hybrid”macrophages in vascularization.

While it is evident that MAA beads induced a bias in macro-phage polarization towards M2, it is unclear why this happens.Several interconnected mechanisms are likely involved and studiesare underway to clarify these mechanisms. Several hypotheses areproposed here. In this subcutaneous injection model, it was notedthat deliberate nicking of small vessels enhanced the vascularregenerative effect of MAA beads (but not MM beads), suggesting arole for whole blood or one of its components. Separately, we haveseen that incubating beads with plasma or serum resulted in morecomplement proteins (e.g., C1q, Factor H) adsorbed to MAA than toMM beads; yet complement was activated to a lower degree withMAA beads [55]. C1q has been implicated in macrophage polari-zation [75,76] and its differential adsorption to MAA beads may inpart explain the polarization bias. Alternatively, the increasedneutrophil density in MAA-treated mice (Fig. 5C) likely led to moreapoptotic neutrophils, which has also been implicated in M2 po-larization [77,78].

Finally, Shh expression was increased by MAA beads in macro-phages (Supplementary Information, Fig. 4), suggesting a link be-tween Shh and M2 polarization with MAA beads. Exposure ofprimary macrophages and a macrophage cell line (RAW264.7 cells)to a pathogen upregulated markers of alternatively activated M2

macrophages (Arg1, Fizz1 and Ym1) and this response was inhibitedwith Shh pathway specific antagonists [13]. Primary brain tumorsexpressing Shh were characterized by increased infiltration ofmacrophages and expression of genes implicated in polarization ofmacrophages towards M2 [55]. Thus, it may be that the Shhpathway has been activated during the initial wave of inflammationwhich then served as a M2 polarization signal [54,79].

4.5. The insight into mechanism of vascular regenerative MAAbeads

The results described here and previously [reviewed in Ref. [2]]revealed that MAA-based biomaterials elicit its regenerativeproperties by modulating several aspects of vascular biology. Wehypothesize that upon interaction with tissue, MAA-based bio-materials differentially adsorb proteins including complement (e.g.,C1q) [56], which then modulate phosphorylation pathways withinminutes of contact between the biomaterial and cells [14]. Subse-quently, differential expression of mRNA [6e8] and proteins lead toa modified initial inflammatory response characterized byincreased neutrophil infiltration. This in turn results in the activa-tion of other signaling pathways (i.e., Shh) and the modulation ofcells involved in vascularization (i.e., macrophages and endothelialcells). Follow up studies will need to link these components. Forexample, whether complement, neutrophils or Shh signaling leadsto vascularization by modulating yet other cells and pathways re-mains to be elucidated. It is clear that MAA-based biomaterialsactivate a complex network of events to effect vascularregeneration.

What is not clear is how all of this from protein adsorptionthrough Shh pathway modulation to vascularization is determinedby the properties of the biomaterial. We attribute the vascularregenerating effect to themethacrylic acid and its strong charge buthow the properties of this charge (e.g., pKa) influence the adsorbedprotein and whether other similar anionic polymers would also bevascular regenerating is an as of yet unanswered question.

5. Conclusion

MAA beads improved vascularization in healthy mice of bothsexes after subcutaneous injection. The higher vessel density wasaccompanied by an increase in the expression of GFP (Shh) and b-Gal (Ptch1), differences in inflammatory cell infiltration (includingmore neutrophils at day 1 and macrophages at day 7) and amacrophage polarization bias towards M2. These results suggestthat the Shh signaling pathway and an altered inflammatoryresponse are associated, and together may be involved in the MAA-mediated modulation of the host response and its beneficialvascular regenerative effect. This subcutaneousmodel may prove tobe a useful tool for further understanding of the host response tobiomaterials.

Acknowledgments

The authors acknowledge financial support from the OntarioResearch Foundation and the Natural Sciences and EngineeringResearch Council (NSERC). A. Lisovsky acknowledges scholarshipsupport from the Province of Ontario, the University of Toronto andthe NSERC Collaborative Research and Training Experience(CREATE) in Manufacturing Materials and Mimetics (M3) trainingprogram. D. K. Y. Zhang acknowledges scholarship support from theProvince of Ontario, the University of Toronto and the CanadianInstitutes for Health Research. The authors acknowledge help fromI. Talior-Volodarsky and R. Mahou in the setup of the in vitromacrophage experiment and C. Lo for his surgical expertise.

A. Lisovsky et al. / Biomaterials 98 (2016) 203e214 213

Appendix A. Supplementary data

Supplementary data related to this article can be found at http://dx.doi.org/10.1016/j.biomaterials.2016.04.033.

References

[1] L.A. Wells, M.S. Valic, A. Lisovsky, M.V. Sefton, Angiogenic biomaterials topromote tissue vascularization and integration, Israel J. Chem. 53 (2013)637e645.

[2] A. Lisovsky, M.D. Chamberlain, L.A. Wells, M.V. Sefton, Cell interactions withvascular regenerative MAA-based materials in the context of wound healing,Adv. Healthc. Mater. 16 (2015) 2375e2387.

[3] M.J. Butler, M.V. Sefton, Poly(butyl methacrylate-co-methacrylic acid) tissueengineering scaffold with pro-angiogenic potential in vivo, J. Biomed. Mater.Res. Part A 82 (2007) 265e273.

[4] A.A. Eckhaus, J.S. Fish, G. Skarja, J.L. Semple, M.V. Sefton, A preliminary studyof the effect of poly(methacrylic acid-co-methyl methacrylate) beads onangiogenesis in rodent skin grafts and the quality of the panniculus carnosus,Plastic Reconstr. Surg. 122 (2008) 1361e1370.

[5] D.C. Martin, J.L. Semple, M.V. Sefton, Poly(methacrylic acid-co-methyl meth-acrylate) beads promote vascularization and wound repair in diabetic mice,J. Biomed. Mater. Res. Part A 93 (2010) 484e492.

[6] L.E. Fitzpatrick, A. Lisovsky, M.V. Sefton, The expression of sonic hedgehog indiabetic wounds following treatment with poly(methacrylic acid-co-methylmethacrylate) beads, Biomaterials 33 (2012) 5297e5307.

[7] L.A. Wells, M.V. Sefton, The effect of methacrylic acid in smooth coatings ondTHP1 and HUVEC gene expression, Biomaterial Sci. 2 (2014) 1768e1778.

[8] L.E. Fitzpatrick, J.W. Chan, M.V. Sefton, On the mechanism of poly(methacrylicacid-co-methyl methacrylate)-induced angiogenesis: gene expression anal-ysis of dTHP-1 cells, Biomaterials 32 (2011) 8957e8967.

[9] R.J. Patel, M.V. Sefton, Some aspects of the host response to methacrylic acidcontaining beads in a mouse air pouch, J. Biomed. Mater. Res. Part A 100(2012) 2054e2062.

[10] R. Pola, L.E. Ling, M. Silver, M.J. Corbley, M. Kearney, R. Blake Pepinsky, et al.,The morphogen Sonic hedgehog is an indirect angiogenic agent upregulatingtwo families of angiogenic growth factors, Nat. Med. 7 (2001) 706e711.

[11] I.V. Yang, S. Alper, B. Lackford, H. Rutledge, L.A. Warg, L.H. Burch, et al., Novelregulators of the systemic response to lipopolysaccharide, Am. J. Respir. CellMol. Biol. 45 (2011) 393e402.

[12] D.S. Ghorpade, S. Holla, S.V. Kaveri, J. Bayry, S.A. Patil, K.N. Balaji, Sonichedgehog-dependent induction of microRNA 31 and microRNA 150 regulatesMycobacterium bovis BCG-driven toll-like receptor 2 signaling, Mol. Cell. Biol.33 (2013) 543e556.

[13] T.A. Pereira, G. Xie, S.S. Choi, W.K. Syn, I. Voieta, J. Lu, et al., Macrophage-derived Hedgehog ligands promotes fibrogenic and angiogenic responses inhuman Schistosomiasis mansoni, Liver Int. 33 (2013) 149e161.

[14] M.D. Chamberlain, L.A. Wells, A. Lisovsky, H. Guo, R. Isserlin, I. Talior-Volo-darsky, et al., Unbiased phosphoproteomic method identifies the initial effectsof a methacrylic acid copolymer on macrophages, Proc. Natl. Acad. Sci. U. S. A.34 (2015) 10673e10678.

[15] B.D. Harfe, P.J. Scherz, S. Nissim, H. Tian, A.P. McMahon, C.J. Tabin, Evidence foran expansion-based temporal Shh gradient in specifying vertebrate digitidentities, Cell 118 (2004) 517e528.

[16] L.V. Goodrich, L. Milenkovic, K.M. Higgins, M.P. Scott, Altered neural cell fatesand medulloblastoma in mouse patched mutants, Science 277 (1997)1109e1113.

[17] J.S. Nanda, J.R. Lorsch, Labeling a protein with fluorophores using NHS esterderivitization, Methods Enzym. 536 (2014) 87e94.

[18] J. Alroy, V. Goyal, E. Skutelsky, Lectin histochemistry of mammalian endo-thelium, Histochemistry 86 (1987) 603e607.

[19] S. Sindhwani, A.M. Syed, S. Wilhelm, D.R. Glancy, Y.Y. Chen, M. Dobosz, et al.,Three-dimensional optical mapping of nanoparticle distribution in intactyissues, ACS Nano (2016), http://dx.doi.org/10.1021/acsnano.6b01879. ArticleASAP.

[20] R. Tomer, L. Ye, B. Hsueh, K. Deisseroth, Advanced CLARITY for rapid and high-resolution imaging of intact tissues, Nat. Protoc. 9 (2014) 1682e1697.

[21] T. Staudt, M.C. Lang, R. Medda, J. Engelhardt, S.W. Hell, 2,20-thiodiethanol: anew water soluble mounting medium for high resolution optical microscopy,Microsc. Res. Tech. 70 (2007) 1e9.

[22] C.S. Robbins, I. Hilgendorf, G.F. Weber, I. Theurl, Y. Iwamoto, J.L. Figueiredo, etal., Local proliferation dominates lesional macrophage accumulation inatherosclerosis, Nat. Med. 19 (2013) 1166e1172.

[23] F. Franconi, S. Brunelleschi, L. Steardo, V. Cuomo, Gender differences in drugresponses, Pharmacol. Res. 55 (2007) 81e95.

[24] R.R. Voskuhl, K. Palaszynski, Sex hormones in experimental autoimmuneencephalomyelitis: implications for multiple sclerosis, Neuroscientist 7 (2001)258e270.

[25] J. Briscoe, P.P. Therond, The mechanisms of Hedgehog signalling and its rolesin development and disease, Nat. Rev. Mol. Cell Biol. 14 (2013) 416e429.

[26] T. Lawrence, G. Natoli, Transcriptional regulation of macrophage polarization:enabling diversity with identity, Nat. Rev. Immunol. 11 (2011) 750e761.

[27] F.O. Martinez, S. Gordon, The M1 and M2 paradigm of macrophage activation:

time for reassessment, F1000prime Rep. 6 (2014) 13.[28] M. Weiss, K. Blazek, A.J. Byrne, D.P. Perocheau, I.A. Udalova, IRF5 is a specific

marker of inflammatory macrophages in vivo, Mediat. Inflamm. 2013 (2013)245804.

[29] L.R. Madden, D.J. Mortisen, E.M. Sussman, S.K. Dupras, J.A. Fugate, J.L. Cuy, etal., Proangiogenic scaffolds as functional templates for cardiac tissue engi-neering, Proc. Natl. Acad. Sci. U. S. A. 107 (2010) 15211e15216.

[30] H. Cho, S. Balaji, A.Q. Sheikh, J.R. Hurley, Y.F. Tian, J.H. Collier, et al., Regulationof endothelial cell activation and angiogenesis by injectable peptide nano-fibers, Acta Biomater. 8 (2012) 154e164.

[31] K. Rajangam, M.S. Arnold, M.A. Rocco, S.I. Stupp, Peptide amphiphilenanostructure-heparin interactions and their relationship to bioactivity, Bio-materials 29 (2008) 3298e3305.

[32] D.P. Sieveking, P. Lim, R.W. Chow, L.L. Dunn, S. Bao, K.C. McGrath, et al., A sex-specific role for androgens in angiogenesis, J. Exp. Med. 207 (2010) 345e352.

[33] D.W. Losordo, J.M. Isner, Estrogen and angiogenesis: a review, Arterioscler.Thromb. Vasc. Biol. 21 (2001) 6e12.

[34] K. Koga, M. Nakamura, H. Nakashima, T. Akiyoshi, M. Kubo, N. Sato, et al.,Novel link between estrogen receptor alpha and hedgehog pathway in breastcancer, Anticancer Res. 28 (2008) 731e740.

[35] Y. Sun, Y. Wang, C. Fan, P. Gao, X. Wang, G. Wei, et al., Estrogen promotesstemness and invasiveness of ER-positive breast cancer cells through Gli1activation, Mol. Cancer 13 (2014) 137.

[36] A.K. Beery, I. Zucker, Sex bias in neuroscience and biomedical research,Neurosci. Biobehav. Rev. 35 (2011) 565e572.

[37] J.A. Clayton, F.S. Collins, Policy: NIH to balance sex in cell and animal studies,Nature 509 (2014) 282e283.

[38] B. Yang, J.B. Treweek, R.P. Kulkarni, B.E. Deverman, C.K. Chen, E. Lubeck, et al.,Single-cell phenotyping within transparent intact tissue through whole-bodyclearing, Cell 158 (2014) 945e958.

[39] M.A. Renault, J. Roncalli, J. Tongers, S. Misener, T. Thorne, K. Jujo, et al., TheHedgehog transcription factor Gli3 modulates angiogenesis, Circ. Res. 105(2009) 818e826.

[40] P.W. Ingham, Transducing Hedgehog: the story so far, EMBO J. 17 (1998)3505e3511.

[41] J. Asai, H. Takenaka, K.F. Kusano, M. Ii, C. Luedemann, C. Curry, et al., Topicalsonic hedgehog gene therapy accelerates wound healing in diabetes byenhancing endothelial progenitor cell-mediated microvascular remodeling,Circulation 113 (2006) 2413e2424.

[42] J.D. Luo, T.P. Hu, L. Wang, M.S. Chen, S.M. Liu, A.F. Chen, Sonic hedgehogimproves delayed wound healing via enhancing cutaneous nitric oxidefunction in diabetes, Am. J. Physiol. Endocrinol. Metab. 297 (2009)E525eE531.

[43] N.M. Amankulor, D. Hambardzumyan, S.M. Pyonteck, O.J. Becher, J.A. Joyce,E.C. Holland, Sonic hedgehog pathway activation is induced by acute braininjury and regulated by injury-related inflammation, J. Neurosci. 29 (2009)10299e10308.

[44] M. Palladino, I. Gatto, V. Neri, S. Straino, M. Silver, A. Tritarelli, et al., Pleio-tropic beneficial effects of sonic hedgehog gene therapy in an experimentalmodel of peripheral limb ischemia, Mol. Ther. 19 (2011) 658e666.

[45] R. Pola, L.E. Ling, T.R. Aprahamian, E. Barban, M. Bosch-Marce, C. Curry, et al.,Postnatal recapitulation of embryonic hedgehog pathway in response toskeletal muscle ischemia, Circulation 108 (2003) 479e485.

[46] G. Straface, T. Aprahamian, A. Flex, E. Gaetani, F. Biscetti, R.C. Smith, et al.,Sonic hedgehog regulates angiogenesis and myogenesis during post-natalskeletal muscle regeneration, J. Cell. Mol. Med. 13 (2009) 2424e2435.

[47] K.F. Kusano, K.L. Allendoerfer, W. Munger, R. Pola, M. Bosch-Marce,R. Kirchmair, et al., Sonic hedgehog induces arteriogenesis in diabetic vasanervorum and restores function in diabetic neuropathy, Arterioscler. Thromb.Vasc. Biol. 24 (2004) 2102e2107.

[48] K.F. Kusano, R. Pola, T. Murayama, C. Curry, A. Kawamoto, A. Iwakura, et al.,Sonic hedgehog myocardial gene therapy: tissue repair through transientreconstitution of embryonic signaling, Nat. Med. 11 (2005) 1197e1204.

[49] A. Yamasaki, C. Kameda, R. Xu, H. Tanaka, T. Tasaka, N. Chikazawa, et al.,Nuclear factor kappaB-activated monocytes contribute to pancreatic cancerprogression through the production of Shh, Cancer Immunol. Immunother.CII. 59 (2010) 675e686.

[50] K.J. Tracey, A. Cerami, Tumor necrosis factor: a pleiotropic cytokine andtherapeutic target, Annu. Rev. Med. 45 (1994) 491e503.

[51] C.A. Dinarello, Immunological and inflammatory functions of the interleukin-1 family, Annu. Rev. Immunol. 27 (2009) 519e550.

[52] M. Waghray, Y. Zavros, M. Saqui-Salces, M. El-Zaatari,C.B. Alamelumangapuram, A. Todisco, et al., Interleukin-1beta promotesgastric atrophy through suppression of Sonic Hedgehog, Gastroenterology138 (2010) 562e572, 572.e1e2.

[53] H. Kasperczyk, B. Baumann, K.M. Debatin, S. Fulda, Characterization of sonichedgehog as a novel NF-kappaB target gene that promotes NF-kappaB-mediated apoptosis resistance and tumor growth in vivo, FASEB J. 23 (2009)21e33.

[54] X. Zhou, Z. Liu, F. Jang, C. Xiang, Y. Li, Y. He, Autocrine Sonic hedgehog at-tenuates inflammation in cerulein-induced acute pancreatitis in mice viaupregulation of IL-10, PloS One 7 (2012) e44121.

[55] A.S. Margol, N.J. Robison, J. Gnanachandran, L.T. Hung, R.J. Kennedy, M. Vali, etal., Tumor-associated macrophages in SHH subgroup of medulloblastomas,Clin. Cancer Res. 21 (2015) 1457e1465.

A. Lisovsky et al. / Biomaterials 98 (2016) 203e214214

[56] L.A. Wells, H. Guo, A. Emili, M.V. Sefton, Angiogenic Methacrylic Acid Co-polymers Adsorb Different Proteins and Result in Lower Complement Acti-vation in Comparison to Poly (Methyl Methacrylate), Society for BiomaterialsConference, 2015.

[57] A. Francke, J. Herold, S. Weinert, R.H. Strasser, R.C. Braun-Dullaeus, Generationof mature murine monocytes from heterogeneous bone marrow anddescription of their properties, J. Histochem. Cytochem. 59 (2011) 813e825.

[58] S. Franz, S. Rammelt, D. Scharnweber, J.C. Simon, Immune responses to im-plants - a review of the implications for the design of immunomodulatorybiomaterials, Biomaterials 32 (2011) 6692e6709.

[59] J.M. Anderson, A. Rodriguez, D.T. Chang, Foreign body reaction to bio-materials, Semin. Immunol. 20 (2008) 86e100.

[60] K.L. Spiller, R.R. Anfang, K.J. Spiller, J. Ng, K.R. Nakazawa, J.W. Daulton, et al.,The role of macrophage phenotype in vascularization of tissue engineeringscaffolds, Biomaterials 35 (2014) 4477e4488.

[61] D.M. Higgins, R.J. Basaraba, A.C. Hohnbaum, E.J. Lee, D.W. Grainger,M. Gonzalez-Juarrero, Localized immunosuppressive environment in theforeign body response to implanted biomaterials, Am. J. Pathol. 175 (2009)161e170.

[62] S.K. Brancato, J.E. Albina, Wound macrophages as key regulators of repair:origin, phenotype, and function, Am. J. Pathol. 178 (2011) 19e25.

[63] P. Thevenot, W. Hu, L. Tang, Surface chemistry influences implant biocom-patibility, Curr. Top. Med. Chem. 8 (2008) 270e280.

[64] O. Nishi, K. Nishi, Y. Sakka, T. Sakuraba, S. Maeda, Intercapsular cataract sur-gery with lens epithelial cell removal. Part IV: Capsular fibrosis induced bypoly(methyl methacrylate), J. Cataract Refract. Surg. 17 (1991) 471e477.

[65] E. Sakurai, A. Anand, B.K. Ambati, N. van Rooijen, J. Ambati, Macrophagedepletion inhibits experimental choroidal neovascularization, InvestigativeOphthalmol. Vis. Sci. 44 (2003) 3578e3585.

[66] T. Hisatome, Y. Yasunaga, S. Yanada, Y. Tabata, Y. Ikada, M. Ochi, Neo-vascularization and bone regeneration by implantation of autologous bonemarrow mononuclear cells, Biomaterials 26 (2005) 4550e4556.

[67] N. Hirose, H. Maeda, M. Yamamoto, Y. Hayashi, G.H. Lee, L. Chen, et al., Thelocal injection of peritoneal macrophages induces neovascularization in ratischemic hind limb muscles, Cell Transplant. 17 (2008) 211e222.

[68] D.M. Mosser, J.P. Edwards, Exploring the full spectrum of macrophage

activation, Nat. Rev. Immunol. 8 (2008) 958e969.[69] Y. Ding, N. Song, Y. Luo, Role of bone marrow-derived cells in angiogenesis:

focus on macrophages and pericytes, Cancer Microenviron. 5 (2012) 225e236.[70] C. Hellberg, A. Ostman, C.H. Heldin, PDGF and vessel maturation, Recent Re-

sults Cancer Res. 180 (2010) 103e114.[71] S.F. Badylak, J.E. Valentin, A.K. Ravindra, G.P. McCabe, A.M. Stewart-Akers,

Macrophage phenotype as a determinant of biologic scaffold remodeling,Tissue Eng. Part A 14 (2008) 1835e1842.

[72] N. Jetten, N. Roumans, M.J. Gijbels, A. Romano, M.J. Post, M.P. de Winther, etal., Wound administration of M2-polarized macrophages does not improvemurine cutaneous healing responses, PloS One 9 (2014) e102994.

[73] C. Garcia-Morales, S. Nandi, D. Zhao, K.A. Sauter, L. Vervelde, D. McBride, et al.,Cell-autonomous sex differences in gene expression in chicken bone marrow-derived macrophages, J. Immunol. 194 (2015) 2338e2344.

[74] R.S. Scotland, M.J. Stables, S. Madalli, P. Watson, D.W. Gilroy, Sex differences inresident immune cell phenotype underlie more efficient acute inflammatoryresponses in female mice, Blood 118 (2011) 5918e5927.

[75] M.E. Benoit, E.V. Clarke, P. Morgado, D.A. Fraser, A.J. Tenner, Complementprotein C1q directs macrophage polarization and limits inflammasome ac-tivity during the uptake of apoptotic cells, J. Immunol. 188 (2012) 5682e5693.

[76] S.S. Bohlson, S.D. O'Conner, H.J. Hulsebus, M.M. Ho, D.A. Fraser, Complement,c1q, and c1q-related molecules regulate macrophage polarization, Front.Immunol. 5 (2014) 402.

[77] M.J. Stables, S. Shah, E.B. Camon, R.C. Lovering, J. Newson, J. Bystrom, et al.,Transcriptomic analyses of murine resolution-phase macrophages, Blood 118(2011) e192e208.

[78] G. Zizzo, B.A. Hilliard, M. Monestier, P.L. Cohen, Efficient clearance of earlyapoptotic cells by human macrophages requires M2c polarization and MerTKinduction, J. Immunol. 189 (2012) 3508e3520.

[79] W.J. Zacharias, X. Li, B.B. Madison, K. Kretovich, J.Y. Kao, J.L. Merchant, et al.,Hedgehog is an anti-inflammatory epithelial signal for the intestinal laminapropria, Gastroenterology 138 (2010) 2368e2377.

[80] K. Kroeger, M. Collins, L. Ugozzoli, The preparation of primary hematopoieticcell cultures from murine bone marrow for electroporation, J. Vis. Exp. 23(2009) 1026.

80

S17. Curriculum vitae.

Page 1 of 4

DAVID K.Y. ZHANG

Phone

+1 (647) 974-4510

Email dk.zhang@mail.utoronto.ca

EDUCATION

University of Toronto, Toronto, Ontario, Canada

Institute of Biomaterials and Biomedical Engineering (IBBME)

MASc in Biomedical Engineering

Advisor: Prof. Michael V. Sefton

GPA: 4.0/4.0

Sept 2014 – Jun

2016

McGill University, Montréal, Canada

BSc in Honours Immunology

Advisor: Prof. David Juncker

GPA: 3.93/4.0

Sept 2011 – Apr

2014

RESEARCH

MASc Thesis

Institute of Biomaterials and Biomedical Engineering, University of Toronto,

Canada

Prof. Michael Sefton

Inflammatory cell responses to vascular regenerative methacrylic acid-containing

materials.

Sept 2014 – Jun

2016

Undergraduate Thesis Department of Biomedical Engineering, McGill University, Montréal, Canada

Prof. David Juncker

Assessing cytokine responses in stressed microglia cells.

Sept 2013 – Apr

2014

Research Assistant Department of Biomedical Engineering, McGill University, Montréal, Canada

Prof. David Juncker

Development of an automated microfluidic device for ultrahigh-sensitive

quantification of biomarkers.

May 2013 –

Aug 2013

Research Assistant Faculty of Medicine, McGill University, Montréal, Canada

Dr. Ivan Rohan, MD

Development and implementation of an E-learning oncology course for primary

care physicians.

May 2012 – Jan

2013

D.K.Y. Zhang

Page 2 of 4

Research Assistant Ecology Lab, Faculty of Environment, University of Waterloo, Canada

Kari Olsen

Monitoring soil content, water quality, aquatic specimens in the Region of

Waterloo.

Sept 2009 – Jun

2011

AWARDS

SGS Conference Grant. University of Toronto. $880.

May 2016

Canadian Graduate Scholarship- Master’s. Canadian Institute of Health

Research (CIHR). $17,500.

Sept 2015

Institute of Biomaterials and Biomedical Engineering (IBBME) Graduate

Fellowship. University of Toronto. $2,000.

Sept 2015

NSERC CREATE Trainee in Manufacturing, Materials, and Mimetics (M3).

University of Toronto. $15,000. Stipend declined.

Aug 2015

Barbara and Frank Milligan Graduate Fellowship. University of Toronto.

$5,311.

Aug 2015

Queen Elizabeth II & Thomas Noakes Scholarship in Science and Technology.

University of Toronto. $15,000.

Aug 2014

Canadian Graduate Scholarship- Master’s. Natural Sciences and Engineering

Research Council of Canada (NSERC). $17,500. Declined.

Aug 2014

Dean’s Honour List. McGill University (All three years).

May 2014

Undergraduate Research Award. Natural Sciences and Engineering Research

Council of Canada (NSERC). $5600.

May 2013

Alma Mater Scholarship. McGill University. $3000.

Quebec University Support Bursary. McGill University. $1000.

National AP Scholar. Canada. $50.

Sept 2011

Sept 2011

Sept 2011

PUBLICATIONS

1. A. Lisovsky*, D.K.Y. Zhang*, M.V. Sefton, “Effect of methacrylic acid beads on the sonic

hedgehog signaling pathway and macrophage polarization in a subcutaneous injection mouse

model”, Biomaterials, 2016. doi:10.1016/j.biomaterials.2016.04.033

* These authors contributed equally.

D.K.Y. Zhang

Page 3 of 4

PRESENTATIONS

I. Tailor-Volodorsky, D.K.Y. Zhang, R. Mahou, M.V. Sefton. The effect of methacrylic acid-

stimulated macrophages on endothelial cells. Poster. World Biomaterials Congress. Montreal, Canada.

May, 2016. DOI: 10.3389/conf.FBIOE.2016.01.02635

D.K.Y. Zhang, M.V. Sefton. Methacrylic acid- containing beads modulate macrophage polarization in

a vascularizing subcutaneous mouse model. Oral Presentation. World Biomaterials Congress.

Montreal, Canada. May, 2016. DOI: 10.3389/conf.FBIOE.2016.01.01325

K. Zhang, M.V. Sefton. Macrophages are polarized to a CD206+MHCII- phenotype by methacrylic

acid-containing beads in a subcutaneous injection model. Poster. Institute of Biomaterials and

Biomedical Engineering Scientific Day, University of Toronto. Toronto, Canada. May, 2015.

K. Zhang, G. Zhou, A. Ng, D. Juncker. Optimizing gold nanoparticle generation for quantitative

plasmonic ELISA. Poster. Biomedical Engineering Symposium, McGill University. Quebec, Canada.

September, 2013.

TEACHING

Head Lab Teaching Assistant

BME 205, Biomolecules and Cells.

Division of Engineering Science, Faculty of Arts & Science, University of Toronto.

Jan 2016 – Apr

2016

Teaching Assistant BME 395, Biomedical Systems Engineering II: Cells and Tissues.

Division of Engineering Science, Faculty of Arts & Science, University of Toronto.

Sept 2015 –

Dec 2015

Lab Teaching Assistant BME 205, Biomolecules and Cells.

Division of Engineering Science, Faculty of Arts & Science, University of Toronto.

Jan 2015 – Apr

2015

MENTORSHIP

Mohammad Saleh, Thesis Student.

Division of Engineering Science, Faculty of Arts & Science, University of Toronto.

Sept 2015 – Apr

2016

Robert Brais, Thesis Student.

Department of Electrical Engineering, Faculty of Engineering, McGill University.

Sept 2013 –

Aug 2014

COMMUNITY

Student mentor. Saturday Day Program, University of Toronto.

Content development committee. Stem Cell Talks 2016, Let’s Talk Science.

Graphics director. IBBME Scientific Day, University of Toronto.

Jan 2016 – Apr 2016

Sept 2015 – Mar 2016

Sept 2015 – Mar 2016

D.K.Y. Zhang

Page 4 of 4

Judge recruitment committee. IBBME Scientific Day, University of Toronto.

Science Rendezvous volunteer. Toronto, Ontario.

Tech Staff. Canadian Biomaterials Society Conference. Toronto, Ontario.

Student mentor. Saturday Day Program, University of Toronto.

Graphic design committee. IBBME Scientific Day, University of Toronto.

Content development committee. Stem Cell Talks 2015, Let’s Talk Science.

Let’s Talk Science volunteer. Toronto, Ontario.

University of Toronto peer tutor (chemistry, biochemistry, immunology).

McGill Sketching Club VP. McGill University.

CaPS tutor (calculus, immunology, organic chemistry). McGill University.

Volunteer at the Montreal General Hospital

Peer tutor (physics, chemistry, calculus), Waterloo, Canada

Art instructor at Little Artists Workshop, Waterloo, Canada

Sept 2015 – Mar 2016

May 2015

May 2015

Jan 2015 – May 2015

Sept 2014 – Mar 2015

Sept 2014 – Mar 2015

Sept 2014 – Jun 2016

Sept 2014 – Jun 2016

Sept 2012 – May 2014

Sept 2013 – May 2014

Jan 2012 – Apr 2012

2007 – 2011

2007 – 2011