Advanced polymeric scaffolds for functional materials in...

Advanced polymeric scaffolds

for functional materials in biomedical applications

Kim Öberg Hed Doctoral Thesis Kungliga Tekniska högskolan, Stockholm 2013 AKADEMISK AVHANDLING som med tillstånd av Kungliga Tekniska högskolan i Stockholm framläggs till offentlig granskning för avläggande av teknisk doktorsexamen fredagen den 31 januari 2014, kl. 10:00 i sal F3, Lindstedtsvägen 26, KTH, Stockholm. Avhandlingen försvaras på engelska. Fakultetsopponent: Professor Eric Drockenmuller, University of Lyon, France.

Copyright © 2014 Kim Öberg Hed All rights reserved Paper I © 2013 Journal of Materials Chemistry A Paper II © 2011 Lab on a chip Paper III © 2013 Langmuir Paper IV © 2013 Chemical Communications Paper V © 2013 Journal of Materials Chemistry B TRITA-CHE Report 2014:1 ISSN 1654-1081 ISBN 978-91-7501-978-9

ABSTRACT

Advancements in the biomedical field are driven by the design of novel materials with controlled physical and bio-interactive properties. To develop such materials, researchers rely on the use of highly efficient reactions for the assembly of advanced polymeric scaffolds that meet the demands of a functional biomaterial. In this thesis two main strategies for such materials have been explored; these include the use of off-stoichiometric thiol-ene networks and dendritic polymer scaffolds. In the first case, the highly efficient UV-induced thiol-ene coupling (TEC) reaction was used to create crosslinked polymeric networks with a predetermined and tunable excess of thiol or ene functionality. These materials rely on the use of readily available commercial monomers. By adopting standard molding techniques and simple TEC surface modifications, patterned surfaces with tunable hydrophobicity could be obtained. Moreover, these materials are shown to have great potential for rapid prototyping of microfluidic devices. In the second case, dendritic polymer scaffolds were evaluated for their ability to increase surface interactions and produce functional 3D networks. More specifically, a self-assembled dendritic monolayer approach was explored for producing highly functional dendronized surfaces with specific interactions towards pathogenic E. coli bacteria. Furthermore, a library of heterofunctional dendritic scaffolds, with a controllable and exact number of dual-purpose azide and ene functional groups, has been synthesized. These scaffolds were explored for the production of cell interactive hydrogels and primers for bone adhesive implants. Dendritic hydrogels decorated with a selection of bio-relevant moieties and with Young’s moduli in the same range as several body tissues could be produced by facile UV-induced TEC crosslinking. These gels showed low cytotoxic response and relatively rapid rates of degradation when cultured with normal human dermal fibroblast cells. When used as primers for bone adhesive patches, heterofunctional dendrimers with high azide-group content led to a significant increase in the adhesion between a UV-cured hydrophobic matrix and the wet bone surface (compared to patches without primers).

SAMMANFATTNING Designandet av nya material med kontrollerade mekaniska och biointeraktiva egenskaper är centralt för framsteg inom biomedicinområdet. I jakten på funktionella biomaterial förlitar sig forskare på ytterst effektiva reaktioner och producerandet av avancerade polymera strukturer. Denna avhandling utforskar företrädesvis två strategier för framställandet av sådana material; den första baseras på användande av polymera tiol-ene nätverk som är icke-stökiometriskt tvärbundna och den andra på dendritiska polymera strukturer. I första fallet producerades polymera nätverk med överskott av tiol- eller ene-funktionella grupper genom att utnyttja den effektiva UV-initierade tiol-ene kopplingsreaktionen (TEK). Som råmaterial utnyttjades lättillgängliga kommersiella tiol och ene monomerer. Genom att använda standardiserade avgjutningstekniker och enkla TEK-baserade ytmodifikationer kunde fint mönstrade ytor med olika hydrofilicitet framställas. Dessa typer av material visade sig vidare vara väldigt lovande för att framställa nya mickrofluidiska prototyper på ett snabbt och effektivt sätt. I det andra fallet utvärderades dendritiska polymera strukturer för deras förmåga att både öka interaktioner vid biologiska ytor samt att bilda funktionella cellinteragerande 3D nätverk. Mer specifikt så utforskades initialt en metod för att producera dendritiska ytor med interaktion mot patogena E. coli bakterier. Vidare framställdes ett bibliotek av heterofunktionella dendritiska strukturer (HFD) innehållandes azid och ene funktionella grupper. Förmågan för dessa HFDs att bilda cellinteraktiva hydrogeler och primerar för benlim i implantat utforskades. Dendritiska hydrogeler dekorerade med utvalda biorelevanta grupper, och med elasticitetsmoduler i samma spann som flera kroppsvävnader, kunde framställas genom enkel UV-härdning. Gelerna visade låg cytotoxisk respons och en relativt snabb nedbrytning då de inkuberades med normala humana dermala fibroblaster. I fallet med benlimsadhesion visade det sig att HFD-primerar med högt azid-innehåll signifikant ökade adhesionen mellan det hydrofoba implantatet och den blöta hydrofila ben-ytan.

List of papers This thesis is a summary of the following papers:

I. “UV initiated thiol-ene chemistry: A facile and modular synthetic methodology for the construction of functional 3D networks with tunable properties.” Mongkhontreerat, S.; Öberg, K.; Erixon, L.; Löwenhielm, P.; Hult, A.; Malkoch, M.; Journal of Materials Chemistry A, 2013, 1(44), 13732-13737.

II. “Beyond PDMS: off-stoichiometry thiol–ene (OSTE) based soft lithography for rapid prototyping of microfluidic devices” Carlborg, C. F.; Haraldsson, T.; Öberg, K.; Malkoch, M.; Van der Wijngaart, W.; Lab on a chip, 2011, 11(18), 3136-3147.

III. “Templating gold surfaces with function – A self-assembled dendritic monolayer methodology based on monodisperse polyester scaffolds” Öberg, K.; Ropponen, J.; Kelly, J., Löwenhielm, P.; Berglin, M.; Malkoch, M.; Langmuir, 2013, 29(1), 456-465.

IV. “Dual-purpose PEG scaffolds for the preparation of soft and biofunctional hydrogels: the convergence between CuAAC and thiol-ene reactions” Öberg, K.; Hed, Y.; Joelsson Rahmn, I.; Kelly, J.; Löwenhielm, P.; Malkoch, M., Chemical Communications, 2013, 49(62), 6938-6940.

V. “Multipurpose heterofunctional dendritic scaffolds as crosslinkers towards functional soft hydrogels and implant adhesives in bone fracture applications” Hed, Y.; Öberg, K.; Berg, S.;

Nordberg, A.; Von Holst, H.; Malkoch, M.; Journal of Materials Chemistry B. 2013, 1, 6015-6019

My contribution to the appended papers:

I. Part of the experimental work, analysis, and part of the preparation of the manuscript.

II. Part of the experimental work, analysis, and part of the preparation of the manuscript.

III. Most of the experimental work, analysis, and preparation of the manuscript

IV. Most of the experimental work, analysis, and preparation of the manuscript.

V. Part of the experimental work, analysis, and preparation of the manuscript.

Scientific contributions not included in this thesis:

VI. Self-Assembled Arrays of Dendrimer-Gold-Nanoparticle Hybrids for Functional Cell Studies. Lundgren, A.; Hed, Y.; Öberg, K.; Sellborn, A.; Fink, H.; Löwenhielm, P.; Kelly, J.; Malkoch, M.; Berglin, M., Angewandte Chemie International Edition, 2011, 50, 3450-3453.

ABBREVIATIONS Bis-MPA 2,2-Bis(methylol)propionic acid CuAAC Copper(I) catalyzed azide-alkyne cycloaddition DCC N,N’-Dicyclohexylcarbodiimide DCM Dichloromethane DOPA L-dopa, 3,4-Dihydroxyphenylalanine DMAP 4-(Dimethylamino)pyridine DMF Dimethylformamide DMP 2,2-Dimethoxypropane DMPA 2,2-Dimethoxy-2-phenylacetophenone DMTA Dynamic mechanical thermal analysis DPTS 4-(Dimethylamino)pyridinium-p-toluenesulfonate DTT Dithiothreitol EDTA Ethylenediaminetetraacetic acid Ene Alkene, carbon-carbon double bond EtoAc Ethyl acetate EtOH Ethanol FE-SEM Field emission scanning electron microscope FRAP Fiber-reinforced adhesive patch GF-AAS Graphite furnace atomic absorption

spectroscopy HFD Heterofunctional dendrimer Hep Heptane LDH Lactate dehydrogenase Man D-Mannose MALDI-TOF MS Matrix-assisted laser desorption/ionization time of

flight mass spectrometry MeOH Methanol NHDF Normal human dermal fibroblasts NMR Nuclear magnetic resonance OSTE Off-stoichiometric thiol-ene p-TSA p-toluenesulfonic acid PEG Poly(ethylene glycol) RT Room temperature SADM Self-assembled dendritic monolayer SPR Surface plasmon resonance TEA Triethylamine TEC Thiol-ene coupling

Tetrathiol Pentaerythritol tetrakis(3-mercapto propionate)

Triallyl 1,3,5-Triallyl-1,3,5-triazine-2,4,6(1H,3H,5H)-trione

Trithiol Tris[2(3-mercaptopropionyloxy) ethyl]isocyanurate

TABLE OF CONTENTS 1. PURPOSE OF THE STUDY .................................................... 1

2. INTRODUCTION ..................................................................... 2

2.1 POLYMERS ........................................................................... 2 2.2 DENDRITIC POLYMERS ....................................................... 3

2.2.1 Dendritic polymer family ........................................... 3 2.2.1 Dendrimer synthesis .................................................. 4 2.2.2 Heterofunctional dendritic materials ......................... 6

2.3 CHEMICAL TOOLBOX: EFFICIENT COUPLING REACTIONS . 8 2.3.1 Esterification reactions .............................................. 8 2.3.2 Click chemistry .......................................................... 9 2.3.3 Copper(I) catalyzed azide-alkyne cycloaddition (CuAAC) ................................................................................. 10 2.3.4 Thiol-ene coupling (TEC) chemistry ........................ 11

2.4 APPLICATIONS .............................................................. 14 2.4.1 Functional materials for prototyping of microfluidic devices ................................................................................. 14 2.4.2 Dendritic surfaces for increased bacterial interactions ................................................................................. 15 2.4.3 PEG-based dendritic hydrogels for functional cell studies ................................................................................. 16 2.4.4 Dendritic primers in fiber reinforced adhesive patches (FRAPs) for complex bone fracture stabilization ...................... 18

3. EXPERIMENTAL .................................................................. 20

3.1 NOMENCLATURE ................................................................ 20 3.2 MATERIALS...................................................................... 21 3.3 INSTRUMENTATION AND METHODS ................................. 21 3.4 SYNTHETIC, PREPARATIVE AND ANALYTICAL PROCEDURES . 24

3.4.1 AB2C monomer synthesis ......................................... 24 3.4.2 Miscellaneous .......................................................... 25 3.4.3 Swelling and leaching .............................................. 26 3.4.4 Off-stoichiometric thiol-ene (OSTE) networks ......... 26 3.4.4.1 Postmodification of OSTE networks ........................ 27 3.4.5 Self-Assembled Dendritic Monolayers (SADMs) ..... 27

3.4.6 Bacterial binding to surfaces ................................... 28 3.4.7 Dendritic PEG hydrogels ........................................ 28 3.4.8 Determination of residual catalytic copper after CuAAC reaction ................................................................................. 29 3.4.9 Cell viability through lactate dehydrogenase (LDH) assay ................................................................................. 29 3.4.10 FRAP ....................................................................... 30

4. RESULTS AND DISCUSSION.............................................. 31

4.1 FUNCTIONAL OSTE POLYMERIC NETWORKS ........................ 31 4.1.1 Fabrication of OSTE networks ................................. 31 4.1.2 Mechanical tuning of OSTE networks ...................... 33 4.1.3 Functional availability and modification of OSTE networks ................................................................................. 35 4.1.4 OSTE-based microfluidic devices ............................ 37

4.2 SELF-ASSEMBLED DENDRITIC SURFACES ............................. 41 4.2.1 Synthesis of monodisperse sulfur-containing dendritic library ................................................................................. 42 4.2.2 Self-assembled dendritic monolayer (SADM) formation ................................................................................. 44 4.2.3 Bacterial scavenging of SADM surfaces ................. 45

4.3 FUNCTIONAL PEG HYDROGELS ........................................... 47 4.3.1 Synthesis of dendritic binders .................................. 47 4.3.2 Functional hydrogels ............................................... 49 4.3.3 Tunable mechanical properties ............................... 51 4.3.4 Cell viability and hydrolytic sensitivity of dendritic hydrogels ................................................................................. 53

4.4 FIBER REINFORCED ADHESIVE PATCHES ............................. 54 4.4.1 Dendritic dual-purpose primers .............................. 54 4.4.2 Shear strengths of FRAPs with dendritic primers ... 56

5. CONCLUSIONS ..................................................................... 58

6. FUTURE WORK .................................................................... 60

7. ACKNOWLEDGEMENTS .................................................... 61

8. REFERENCES ........................................................................ 64

PURPOSE OF THE STUDY

1

1. PURPOSE OF THE STUDY

Advances in the biomedical field are highly dependent on the design of new functional materials with tunable physical and biochemical properties. For the development of such materials, researchers have come to rely on highly robust and efficient chemistries for the synthesis of advanced polymeric scaffolds with the desired function(s). A prime example of such materials is the dendritic family of polymers which has attracted significant attention from the scientific community due to their unique properties. Their structural perfection, globular nature and high density of functional groups make these synthetic scaffolds highly interesting from a biomedical perspective. This thesis focuses on the design of functional polymeric networks and surfaces for biomedical applications. In pursuit of such materials two main approaches have been explored; these rely on the use of off-stoichiometric thiol-ene (OSTE) networks and dendritic polymer scaffolds, respectively. In the first case, UV-crosslinked thiol-ene networks were prepared from a range of commercial monomers with controlled excess of thiol or ene functionality. The ability to tune the surface chemistry and physical properties of these networks, through the monomer molecular structure and their off-stoichiometric feed ratio, was evaluated. Based on these examinations, selected OSTE materials were assessed for rapid prototyping of microfluidic devices. In the second approach, dendritic polymer scaffolds were evaluated for their ability to increase surface interactions and to impart function into 3D networks. More specifically, applications such as bacterial scavenging surfaces, bioactive hydrogels and primers for implant adhesives were examined.

INTRODUCTION

2

2. INTRODUCTION

2.1 POLYMERS

Natural polymers have been present since the dawn of life and they are an essential part of all organisms. For instance, the DNA that carries our genetic code is a polymer, as is the collagen that builds up our skin and the cellulose of plants. However, it was not until the early 1900s that the synthetic polymer era arose with the commercialization of the first synthetic polymer, Bakelite in 1907.1 Since then, the synthetic polymer field has expanded enormously and synthetic polymers are nowadays an essential part of our everyday life. The term polymer stems from the Greek words poly (many) and meros (part). Polymers are thus an assembly of many parts or monomers that together form a large molecule. By combining monomers in different ways a wide array of polymer compositions and architectures may be obtained (Figure 2.1). The polymer arrangement will in turn have a large impact on the final properties of the polymer. The design of novel polymer architectures is thus pivotal for the development of materials with desired properties for specific applications.

Figure 2.1. Illustration of possible polymer compositions and architectures.

Monomers

Linear polymers

Homopolymer

Branched polymers

Block copolymer

Alternatingcopolymer

Randomcopolymer

Comb polymer Star polymer Dendriticpolymer

INTRODUCTION

3

2.2 DENDRITIC POLYMERS

2.2.1 Dendritic polymer family

Dendritic polymers are a family of highly branched macromolecules. The name stems from the Greek word dendron which translates to tree. The tree-like or branched structure of dendritic polymers results in very different properties compared to their linear analogs. These differences include a lowered viscosity due to the lack of physical chain entanglements as well a more globular shape with a large number of concentrated end groups. Dendritic materials can be divided into monodisperse and polydisperse frameworks which in turn include sub-classes such as dendrimers, hyperbranched polymers and linear-dendritic hybrids (Figure 2.2).2 The monodisperse dendrimers have attracted special attention within the research community due to their structural perfection and globular nature with an exact and large number of multivalent presented end groups.

Figure 2.2. The dendritic family of polymers. Dendrimers consist of a single core which is capped with radial layers of repeating monomers (Figure 2.3). Each layer is called a generation and are comprised of ABn monomers where n ≥ 2. A and B are different functional groups that react with each other (but not with themselves) during the dendritic growth. The multiplicity of the core is usually 2, 3 or 4, and each substituting wedge of the dendrimer is called a dendron. The dendron consists of a focal point, interior branches and peripheral end groups.

HyperbranchedPolymers

Dendrimers Dendrons

Monodisperse frameworks Polydisperse frameworks

Linear-Dendritic Hybrids

Dendrigrafts

INTRODUCTION

4

Figure 2.3. Schematic representation of a dendrimer architecture.

2.2.1 Dendrimer synthesis

Even though dendrimer synthesis was theorized as early as the 1940s by Flory,3, 4 it was not until the late 1970s that the first synthetic report was published by Vögtle et al.5 This was followed in 1985 by two independent parallel reports on divergent synthesis of poly(amidoamine) (PAMAM) dendrimers by Tomalia et al.6 and poly(etheramide) arborols by Newkome et al.7 The divergent growth approach still remains the synthetic route of choice for commercial dendrimers.2 In 1990 the first example of a convergent growth approach to dendrimers was published by Hawker and Frechet.8 The convergent and divergent approaches to dendrimers are illustrated in Figure 2.4. In the divergent approach, the dendritic growth starts from the core by addition of one layer of AB2 monomers (growth step). This is then followed by activation of the B groups enabling reaction with new AB2 monomers. The growth is continued in a repetitive fashion by additional growth/activation steps until the desired generation is reached. The convergent approach instead starts by the synthesis of dendrons with desired generation. These are activated at the focal point and conjugated to the core, followed by a final activation step to obtain the corresponding dendrimer. Due to the rather tedious and expensive synthesis of dendrimers, the use of these materials have mainly been focused in cost tolerant

INTRODUCTION

5

Figure 2.4. Illustration of the divergent and convergent growth approaches to dendrimers. The growth step is denoted as “g”, “a” is activation of peripheral end groups and “a*” is activation of the core. applications such as targeted drug delivery, tissue engineering and antiviral agents.9 Furthermore, areas where relatively small amounts are required have also been explored including catalysis, electronics and sensing.10 The research of dendrimer material applications has benefitted from an increased availability of commercial dendrimers such as the PAMAMs (Dendritech®11 and Dendritic Nanotechnologies12), and Bis-MPA (2,2-bis(methylol) propionic acid) dendrimers (Polymer Factory Sweden AB13). Examples of other common dendrimer types include polypeptide,14 poly(aryl ether)8, phosphorous15 and carbohydrate based dendrimers16. Moreover, recent advancements in polymer conjugation chemistries have led to the emergence of a plethora of accelerated approaches to dendrimer synthesis.17 These focus on increasing the accessibility of dendrimers by reducing the amount of reaction and/or purification steps thus lowering the time and cost of production.

3x

g a

2x

g a*a*g,

g, a

Divergent growth approach

Convergent growth approach

g, a

a

INTRODUCTION

6

2.2.2 Heterofunctional dendritic materials

Traditional dendrimers commonly present one type of functional group at the periphery of the dendrimer in the form of external end groups. To meet the increasing demands for advanced materials that are able to perform multiple tasks, a range of heterofunctional dendrimers (HFDs) have been developed. Figure 2.5 illustrates some of the main approaches for introducing hetero- or bi-functional groups to dendrimers.

Figure 2.5. Example of traditional and heterofunctional dendrimers. In the first case, a bowtie dendrimer has been assembled by the conjugation of two dendron molecules bearing different end group functionalities. These two peripheral functionalities can be tailored to include cell targeting moieties, drug molecules or chromophores,18, 19 making them highly interesting from a therapeutic point of view. For instance, in 2006 Frechet et al.20 showed that a biodegradable bowtie dendrimer, loaded with the anticancer drug doxorubicin on one side and linear PEG chains on the other, was able to effectively cure mice bearing colon carcinomas in a single dose. Other strategies to HFDs include heterofunctionalization of the peripheral end groups21, 22 and internally functionalized dendrimers23, 24 (Figure 2.5) One facile approach to the internally functionalized dendrimers is to use AB2C-type monomers.25 These are used to synthesize dendrimers much like the traditional fashion by reaction of the A and B groups. However, unlike for the traditional AB2 dendrimers, each

Traditional Dendrimers

Heterofunctional Dendrimers

Bowtiedendrimers

External heterofunctionaldendrimers

Internal heterofunctional dendrimers

INTRODUCTION

7

monomer unit also includes an additional C group. This group is dormant during the dendrimer growth due to orthogonality towards the A and B groups, and can therefore be used for additional internal functionalization of the completed dendrimer. These types of dendrimers also experience a faster increase of the amount of available functional groups during the dendritic growth. For instance, the G4 internal HFD depicted in Figure 2.5 has a total of 48 peripheral groups and 45 internal groups leading to a total functionality of 93 groups. The corresponding traditional G4 dendrimer has only 48 external groups of one single type. Another type of heterofunctional dendritic materials is the linear dendritic hybrids (Figure 2.2), in which the properties of linear and dendritic polymers are married. These types of structures may be combined to form a vast number of different architectures including 4-armed linear polymers with terminal dendrons, linear polymers with dendritic side chains, linear dendritic (LD) block copolymers, LDL- and DLD-block copolymers and more. Such materials have been evaluated for applications ranging from micellar drug delivery vehicles26 and gene transfection agents27 to honeycomb sensor surfaces28 and hydrogels for tissue engineering.29

INTRODUCTION

8

2.3 CHEMICAL TOOLBOX: EFFICIENT COUPLING REACTIONS

2.3.1 Esterification reactions

Traditional esterification reactions are done using elevated temperatures in the presence of acid or base catalyst. Such harsh conditions are ill-suited for more advanced synthetic structures. A prime example is the synthesis of dendrimers which has a low tolerance for unwanted side reactions such as transesterifications or degradation. Instead, mild esterification protocols have been developed, such as the N,N’-dicyclohexylcarbodiimide (DCC) mediated coupling of alcohols and inactive carboxylic acids at room temperature in the presence of DPTS catalyst (1:1 complex of 4-(dimethy1amino)pyridine, DMAP, and p-toluenesulfonic acid, p-TSA).30 The DCC agent may also be used to activate carboxylic acids into their more reactive anhydride state. Scheme 2.1 shows the activation of a protected 2,2-bis(methylol)propionic acid (bis-MPA) into its corresponding anhydride. The protected bis-MPA anhydride has been frequently used as building block for the efficient divergent synthesis of polyester-based dendrimers under mild conditions.31, 32

O OH

O

O

NCNDCC

O O

O

O

NCN

H

OO

O

OCHN

N

O-acylisourea intermediate

O O

O

O

O

O

O

Protected Bis-MPA anhydride

HN C

HN

O

DCC-Urea

OHO

O

O

XX = H+X =

R O R

OO

N CHN

O

N-acylurea(byproduct)

RO

Bis-MPA

Scheme 2.1. DCC mediated coupling of acetonide protected bis-MPA.

INTRODUCTION

9

2.3.2 Click chemistry

The ‘click chemistry’ concept was coined in 2001 by Sharpless and coworkers33 for a set of highly efficient reactions exhibiting a “spring-loaded” nature. Such reactions are characterized by having a high thermodynamic driving force (commonly more than 20 kcal mol-1). Since their introduction, ‘click’ reactions have had a tremendous impact in many research fields ranging from drug discovery to polymer and material science.34-36 In order to be classified as ‘click’ a reaction needs to fulfill a number of criteria:33 they should have a high functional group tolerance and produce the desired product in very high yields with only inoffensive byproducts (possible to remove by non-chromatographic methods). Furthermore, they should proceed under simple reaction conditions; this includes being able to start with a wide range of different and readily available starting materials, possibility to perform the reaction with no solvent, an easily removed solvent or a benign solvent (such as water). Examples of reactions that are able to meet these demands are; cycloaddition reactions of unsaturated species (such as 1,3-dipolar cycloaddition reactions and Diels Alder transformations), nucleophilic substitution of heterocyclic electrophiles such as epoxides, and additions to carbon-carbon multiple bonds (such as Michael additions). When initially conceptualized, the main use of ‘click chemistry’ was foreseen to be in the drug discovery field. However, as previously mentioned their use is nowadays widespread in many research fields. In the polymer science field, their robust and efficient nature coupled with the easily removed byproducts has afforded the synthesis of many novel advanced polymeric structures, dendrimers being a prime example.37 Moreover, the initial constraints for ‘click’ chemistries have recently been revised to better suit the demands of the polymer synthetic field.38 For example, the criteria of simple purification, high selectivity, and high conversions, have been expanded to include equimolarity. In polymer synthesis “simple purification” procedures are

INTRODUCTION

10

commonly limited to selective precipitation or removal of volatile compounds. To enable the selective precipitation of polymer-polymer conjugates, equimolar ratios are a must. Even if many reactions fail to fulfill all the initial ‘click’ criteria in the context of polymer synthesis (such as non-chromatographic purification), these reactions are still highly effective and represent very useful synthetic methods.38

2.3.3 Copper(I) catalyzed azide-alkyne cycloaddition (CuAAC)

A prime example of ‘click’ reactions is the copper(I) catalyzed cycloaddition reaction between azides and alkynes to form 1,2,3-triazoles. The non-catalyzed thermally induced version of this reaction was introduced as early as 1893 by Michael et al. 39 This reaction was later extensively revisited in the mid-1900s by Huisgen40 for which it is also known as the Huisgen reaction. Although efficient, the thermally induced reaction produces a mixture of the 1,4 and 1,5 disubstituted products (Scheme 2.2). It was not until 2002 that reports on the stereospecific copper(I) catalyzed reaction, producing only the 1,4-disubstitued product (Scheme 2.2), were published in parallel by Meldal et al.41 and Fokin and Sharpless.42 The CuAAC reaction can be used to link a wide range of organic azides and terminal alkynes with very high efficiencies and high selectivity. It is furthermore insensitive (orthogonal) to most other functional groups and works with a

R2N3R1

N

N

NR1

R2

1

5

N

N

NR1 1

4R2

Cu(I)

N

N

NR1 1

4R2

Scheme 2.2. Thermally induced and Cu(I)-catalyzed 1,3 dipolar cycloaddition reactions of azides and alkynes.

INTRODUCTION

11

variety of solvents (including water).43 Although many different sources of copper(I) may be used, in situ reduction of Cu(II) salts by sodium ascorbate has proven to be a facile alternative in aqueous solutions.42 A high concentration of Cu(I) is crucial for the stepwise progression of the reaction (see mechanism, Scheme 2.3).44 However, the use of copper catalyst puts restraints on the use of the CuAAC reaction in biological applications due to potential copper induced toxicity. Thus, adequate purification procedures need to be adopted to reduce the copper levels to non-toxic levels before introduction into a biological environment. Another alternative which has been developed for use in biological applications is metal-free versions of the CuAAC reaction such as the strain-promoted azide-alkyne cycloaddition (SPAAC).45-47

LnCu

R1

[LnCu] 2

B

B-H

R1 CuLn

R1 Cu2Ln2

R2 N3

R1 CuLn

NN

N R2

N

CuLn

NN R2

R1

B-H

B

NN

N R2

R1

1,4-cycloadditionproduct

Scheme 2.3. Proposed mechanism for the CuAAC reaction.44

2.3.4 Thiol-ene coupling (TEC) chemistry

The thiol-ene coupling (TEC) reaction has become increasingly popular as a tool for efficient conjugations in polymer chemistry.48 In fact, its use was widespread already during the middle of the 20th century for making crosslinked polymer products on an industrial scale.49 The earliest report of the thiol-ene reaction was published already over one hundred years ago in 190550 and thiol-

INTRODUCTION

12

ene photopolymerization was reported as early as 1926.51 Since then, the two most common thiol-ene reactions used have been; (1) the free radical addition to electron-rich or -poor carbon-carbon double bonds (enes) and (2) the catalyzed thiol Michael addition reaction to electron deficient enes. For a period of time in the latter part of the 20th century thiol-ene chemistry was largely abandoned. This was a result of problems with smelly thiol monomers, initiator induced yellowing of thiol-ene based materials which pawed the way for the emerging inexpensive acrylate systems.52 However, the development of new photoinitiator systems (without yellowing), along with reduced toxicity compared with acrylate systems, has led to the revival of thiol-ene chemistry.48, 52 Moreover, due to its robust and highly efficient nature the TEC reaction has nowadays been classified as a ‘click’ reaction.53 The increase in popularity of the TEC reaction is reflected by the growing number of reviews on the subject.48, 49,

54-56 Two additional very important advantages for the thiol-ene reaction, as opposed to acrylate systems, lie in the step growth polymerization mechanism (Scheme 2.4 a49, 52) and the lack of oxygen inhibition. For the preparation of polymer networks/films, the step growth mechanism of thiol-ene systems leads to delayed gel points while the addition chain-growth polymerization of acrylates lead to early onset gelation. The delayed gelation results

R1SH Initiator R1S

R2

R1SR2

R1SHR1SR2

R2

R1S R1SR2

a)

b)R1SH

O2

R1SR1SR2

R1SR2

OO

R1SHR1S

R2

OHO

R1S

Scheme 2.4. a) Idealized mechanisms of the free radical thiol-ene reaction49, 52 and b) hydrogen abstraction and oxygen scavenging routes.

INTRODUCTION

13

in higher monomer mobility and therefore higher conversion. This in turn results in the formation of networks that are more homogenous with more narrow glass transitions.52 Even if the thiol-ene reaction is not inhibited by oxygen, the presence of oxygen may still affect the structure of the final product (see oxygen scavenging route, Scheme 2.4 b). For film formation this side reaction is commonly neglected, whereas for more delicate structures, such as dendrimers, the oxygen needs to be removed prior to starting the reaction. Another significant side reaction for free-radical thiol-ene systems is the homopolymerization of ene monomers. However, this may essentially be avoided by selecting monomers which favor the thiol-ene addition over homopolymerization.52, 57 The following monomer types are ordered after decreasing reactivity for thiol-ene radical reaction (increasing likelihood of homopolymerization):

norbornene > vinyl ether > allyl ether > allyltriazine > acrylates

Due to the many above mentioned advantages, TEC chemistry has become a highly preferred method for the production of biomedical materials, for instance in artificial cell culturing scaffolds and tissue engineering scaffolds.57 For such applications, conditions suitable for the photo encapsulation of live cell have been explored by Anseth et al. in 2000.58 Out of the commercial UV-initiator systems tested (Irgacure 184, 651, 907 2959, and Camphorquinone), only Irgacure 2959 (I2959) met the demands of being non-toxic and providing sufficient curing. The established non-toxic conditions were; 0.05% (w/w) of photoinitiator dissolved in a small amount of ethanol (max 0.5% (w/w) of final solution), aqueous solvent with UV light at 365 nm, 8 mW cm-2 for 10 min. Since then, this has been the protocol of choice for photoencapsulation of live cells. However, the I2959 is still not entirely optimized for use at these conditions due to its relatively low adsorption at 365 nm and poor water solubility. More recently, photoinitiators which are better suited in terms of adsorption at 365 nm and water solubility have been explored.59

INTRODUCTION

14

2.4 APPLICATIONS

2.4.1 Functional materials for prototyping of microfluidic devices

Microfluidic devices have shown great potential for a wide array of biological applications including enzymatic assays,60 biomolecular separations,61 and pathogen diagnostics.62 However, despite the growing number of publications in the field, most research never make it beyond proof of concept prototypes.63 One of the reasons for this gap between research and the realization of commercial products is the lack of consistent fabrication, interfacing and material technology.64 The majority of conceptual development is based on the use of commercial polydimethylsiloxanes (PDMS, Sylgard 184) due to their attractive elastomeric properties and ease of use.65 The conversion of PDMS-based prototypes into commercial products is however hampered by several factors including; slow thermal curing processes,66 difficult plasmarization-based surface modification to introduce short term hydrophilicity,67 and low-pressure induced deformation of fine geometries.68 Therefore, the realization of commercially viable products from these prototypes usually requires complete re-development. A successful prototyping platform needs to be based on easily accessible materials that closely resemble the surface chemistry and physical properties of the final products. These properties can range from soft elastomers (Topaz Chip, Fluidigm69) to hard plastics (Triage point of care, Biosite70). Furthermore, the preparation of the materials needs to rely on inexpensive and highly efficient reactions to allow rapid redesign cycles. Two main alternatives to PDMS-based prototypes have emerged; (1) these use of off-the-shelf thermoplastic materials and (2) UV-curable thermosets.64 Out of the two, thermoset microfluidics may be prepared by rapid and inexpensive replica molding eliminating the need of expensive tools and equipment, which are needed for the preparation of

INTRODUCTION

15

thermoplastics. In this work a new material platform based on UV-curable off-stoichiometric thiol-ene (OSTE) thermosets has been evaluated for its use in rapid prototyping of microfluidic devices. This platform has the potential to overcome many of the above mentioned limitations of PDMS-based materials and compared to previous UV curable systems68, 71, 72, the OSTE materials enable direct photo-patternable surface modifications and low temperature dry-bonding of microfluidic layers.

2.4.2 Dendritic surfaces for increased bacterial interactions

One of the most interesting features of dendrimers is the multivalent presentation of functional ligands that enhance interactions with biological receptors and cell surfaces.73, 74 A prime example of this is the use of glycoconjugated dendrimers (in solution) as high affinity scavengers of E. coli bacteria in urinary tract infections.75-77 Dendrimers’ ability to act as interaction amplifiers has also attracted significant attention for their use in sensor and surface recognition applications. Examples include dendritic enzyme-linked immuneosorbent assays (ELISAs)78 and dendritic surfaces for functional cell studies.79, 80 However, surface conjugation of dendrimers has commonly been performed in a multiple-step fashion with several pre-activation steps. A more direct route is to use a self-assembled dendritic monolayer (SADM) approach (Figure 2.6).

Figure 2.6. Illustration of the spontaneous self-assembly of a disulfide-cored (red) dendrimer on a gold surface.

S

O

+

Gold surface

Self-assembleddendritic monolayer

(SADM)

INTRODUCTION

16

By preparing dendritic scaffolds with organosulfur core moieties, such as thiols (SH) or disulfides (SS), such molecules are able to spontaneously self-assemble on metal surfaces such as gold or silver.81-83 Even if it is recognized that dendritic structures do not reach the same molecular densities as monolayers from linear alkanethiols,84-86 the increased amount of functional end groups per dendron molecule may still yield highly functionalized surfaces. The assembly of such branched structures is influenced by several parameters including; core functionality (SH or SS), degree of branching (generation), chemical composition (monomer unit), temperature, co-solvent and more.87, 88 The dendritic surfaces investigated in this thesis are based on self-assembly of an organosulfur-cored dendritic library constructed from the bis-MPA monomer. Bis-MPA dendrimers have shown to be non-toxic89 and their attachment to a variety of surfaces has been reported such as gold,90 silicone91 and cellulose.92 The dendritic structures used in this work have been decorated with mannose or hydroxyl functional groups. The reason for including mannosylated SADMs was to explore if the previously mentioned high affinity bacterial binding of glycodendrimers (in solution) could be translated into surfaces with corresponding scavenging properties.

2.4.3 PEG-based dendritic hydrogels for functional cell studies

Poly(ethylene glycol) (PEG) hydrogels have a widespread use in biomedical applications such as artificial extracellular matrices for studying cell-matrix interactions, drug delivery and tissue engineering scaffolds.93-96 The hydrophilicity of PEG polymers result in gels that are soft and elastic with high water content, properties similar to many soft tissues. Unlike natural polymers, which include biologically interactive features such as cell adhesion and proteolytic degradation sites, PEG is known for its inherent inertness towards cell adhesion and protein binding.97, 98 Therefore, researchers commonly use PEG hydrogels as blank

INTRODUCTION

17

slates into which selected bio-relevant cues can be incorporated. Traditional PEG hydrogels have been made by radical induced chain growth polymerization of acrylic pre-polymers.99 However, due to the earlier discussed reasons of network homogeneity and crosslinking efficiency (Section 2.3.2.2), step-growth approaches, such as the UV initiated TEC reaction, are nowadays preferred.48,

100 To obtain efficient crosslinking of such networks, and in parallel insert bio-relevant moieties, the use bioorthogonal ‘click’ chemistries has emerged as a powerful tool.101 Moreover, the combination of such chemistries with dendrimer crosslinkers has proved facile for introducing a high degree of functionality into PEG networks.29, 102-104 In this work, an approach based on heterofunctional dendritic scaffolds combined with linear PEG polymers have been explored for producing functional hydrogels with tunable biochemical and mechanical properties (Figure 2.7). For these hydrogels, the decoration of prepolymer scaffolds with bio-relevant moieties was facilitated by the CuAAC reaction, while the subsequent crosslinking relied on the orthogonal UV-induced TEC reaction.

Figure 2.7. Illustration of hydrogel formation through combination of PEG-dendritic scaffold and linear PEG-dithiol. UV-irradiation is used to induce crosslinking of the thiol and enes.

= ene ( )

= thiol (-SH)

= CuAAC group

Hydrogel fabrication (UV)

Bioactive moietiesat crosslinks

INTRODUCTION

18

2.4.4 Dendritic primers in fiber reinforced adhesive patches (FRAPs) for complex bone fracture stabilization

Conventional methods for complex bone fracture stabilization include the use of metal implants, such as plates and screws. However, for fractures near sensitive tissue, or where there is insufficient bone for drilling, such as the upper cervical spine region, these methods are non-optimal. Instead, the use of percutaneous bone fixation adhesives has shown great promise.105 The investigation of bone adhesives began as early as the 1950s with the discovery of cyanoacrylate-based glues which showed excellent adhesion to wet bone.106 However due to toxic degradation products and numerous reports of adverse biocompatibility the use of these in implants has nowadays been restricted.107, 108 Since then, other more bio-friendly approaches have been explored. A recent study from our group reports on the use of topologically applied fiber reinforced adhesive patches (FRAPs, Figure 2.8).109

Figure 2.8. The fiber-reinforced adhesive patch (FRAP) system. These patches are applied on the outside of the bone, as oppose to inside the fracture. This has the benefit of providing a larger available bone surface area and also avoids interfering with the natural bone healing process within the fracture site.110 The FRAP system consists of mechanically reinforcing fibers embedded in a

INTRODUCTION

19

crosslinked matrix of rigid triazine-based thiol-ene monomers. The use of a benign UV-initiated TEC method for crosslinking and low cytotoxic response of this system111 makes it very promising for future applications. However, the hydrophobic character of the matrix results in poor adhesion to the wet bone necessitating the use of a primer layer to increase the interfacial adhesion between these. Therefore, the main interest in this thesis, with regards to the FRAP system, has been the evaluation of primers based on heterofunctional dendrimer scaffolds. The dual-functional nature of such primers has great potential for interacting with both the matrix and the bone.

EXPERIMENTAL

20

3. EXPERIMENTAL

3.1 NOMENCLATURE

To simplify the discussion of the OSTE materials (Section 4.1) the constituting monomers have been assigned abbreviations related to their functionality instead using their full chemical name. For instance, tris[2(3-mercaptopropionyloxy)ethyl]isocyanurate is referred to as “trithiol” and 1,3,5-triallyl-1,3,5-triazine-2,4,6-trinone is referred to as “triallyl”. The sulfur-containing dendrons and dendrimers used to create SADMs (Section 4.2) are written using the general formula Gn-XX-YY-sp where Gn is the dendron generation, XX is the core functionality (thiol, SH; disulfide, SS), YY is the peripheral end group functionality (hydroxyl, OH; mannose, Man), and sp is added for dendrimers with added spacer between the sulfur core and dendron wedges. For example, G1-SS-OH-sp refers to the first generation dendrimer with a disulfide core, hydroxyl end groups, and with added spacers between the disulfide core and dendron wedges. The dual-functional dendritic scaffolds (section 4.3 and 4.4.) are named according to the core moiety, generation and external/internal functionalities. Thus, TMP-G1-(ene)3-(N3)6 refers to a first generation dendrimer with TMP core bearing three internal ene groups and six external azide groups.

EXPERIMENTAL

21

3.2 MATERIALS

Tris[2(3-mercaptopropionyloxy)ethyl]isocyanurate, pentaery-thritol tetrakis(3-mercaptopropionate), 1,3,5-triallyl-1,3,5-triazine-2,4,6(1H,3H,5H)-trione were obtained from Sigma-Aldrich. All thiol- and vinyl-functional polydimethylsiloxanes were obtained from ABCR GmbH (AB109358, AB109359, AB109360, AB112958, AB113728, AB113729). Linear poly(ethylene glycol) (PEG) with different molecular weights were purchased from Fluka. 2,2-bis(methylol)propionic acid (bis-MPA) was kindly donated by Perstorp AB. Trimethylolpropane was obtained from Polymer Factory AB. 1-hydroxy-cyclohexyl-phenyl-ketone (Irgacure 184), 1-[4-(2-hydroxyethoxy)-phenyl]-2-hydroxy-2-methyl-1-propane-1-one (Irgacure 2959), 2,2-dimethoxy-2-phenylacetophenone (DMPA, Irgacure 651) were obtained from Ciba specialty chemicals. ITX photoinitiator from Firstcure® ITX, Albemarle Corp and Lucirin TPO-L from BASF GmbH. Silica gel (32-64 D 60 Å) for flash chromatography was purchased from ICN Silitech, GmbH. 90/0 E glass fibers style 106, were from Porcher industries. DPTS was made according to previously described procedures,112 as was the acetonide protected bis-MPA anhydride,113 and acetylene-cored bis-MPA dendrons used in for the convergent growth of disulfide dendrimers.18 All other chemicals and solvents were purchased from standard commercial sources (Sigma-Aldrich, Chemtronica, VWR, Fluka, Merck) and used as received unless otherwise noted.

3.3 INSTRUMENTATION AND METHODS

MALDI-TOF MS was conducted on a Bruker UltraFlex MALDI-TOF MS with SCOUT-MTP ion source (Bruker Daltonics) equipped with a N2-laser (337 nm), a gridless ion source, and a reflector. The instrument was calibrated with SperiCalTM

EXPERIMENTAL

22

calibrants purchased from Polymer Factory Sweden AB. The obtained spectra were analyzed using FlexAnalysis (Bruker Daltonics v. 2.2). Polymer dispersities (Ð) and average moleculear weights (Mn) (Paper IV) were obtained using Polytools v. 1.0. 1H NMR and 13C NMR were conducted on a Bruker Avance 400 MHz instrument. 1H NMR was acquired at 400 MHz and 13C NMR at 100 MHz using the residual solvent peak as an internal standard. UV irradiation was performed using a number of different radiation sources. The OSTE-networks in paper I were crosslinked using a Fusion curing system model F300 equipped with standard type BF9 Fusion electrodeless Hg-bulbs. All samples were exposed for a total time of 0.5 s at an intensity of 184 mW·cm-2 (measured in the wavelength interval 320-390 nm). The Fusion system was also used in the preparation of fiber reinforced adhesive patches (FRAPs, paper V). The OSTE formulations for microfluidic devices (Paper II) were crosslinked using a standard table-top UV-lamp (EFOS Lite, Efos) with a 365 nm band pass filter and an intensity of 4 mW·cm-2. For bonding of OSTE layers the band pass filter was removed exposing the materials to unfiltered medium pressure Hg spectrum including 254 nm. The dendritic hydrogels (paper IV and V) were cured using a 100W Hg table-top UV lamp (Blak Ray B 100AP) with a 365 nm band pass filter and an intensity of 9 mW·cm-2. Tensile testing was performed on an Instron5944 instrument. All measurements were done at 25 °C and 50% relative humidity (RH). DMTA was performed on a Q800 DMTA, TA instruments using a temperature ramp program in tensile mode. Rheological measurements were carried out on an ARES RDA-III rheometer, TA instruments at 23 °C using a Ø 25 mm parallel

EXPERIMENTAL

23

plate geometry with γ = 5% and ω = 10 rad/s for the linear viscoelastic regime. The storage modulus (G’) and loss modulus (G’’) of hydrogels (Paper IV) was recorded. For these type of networks G’ > G’’ resulting in G’ ≈ G. The Young’s moduli (E) were calculated according to rubber elasticity theory, where G = E/2(1+v), assuming a Poisson’s ratio (v) of 0.5 for bulk measurements of elastic hydrogel polymer networks . Static contact angle measurements were performed using a KSV Instrument CAM 2000 equipped with a Basler A602f camera. All measurements were done with using deionized water (5 and 10 µl drop size) at 25 °C at 50% RH. FE-SEM images were acquired using a Hitachi S-4300 FE-SEM. All samples were coated with 7 nm of gold/palladium using a Cressington208RH/Agar high resolution sputter coater prior to imaging. FT-Raman Spectroscopy was performed using a Perker-Elmer Spectrum 2000 NIR instrument. Spectra were obtained using 32 scans and 1500 mW laser power. GF-AAS measurements of copper concentration were done using a Perkin Elmer Analyst 800 GF-AAS. SPR measurements were performed using a BIAcore 200 System, GE Healthcare. Experiments were carried out at 22 °C with a flow rate of 5 µl·min-1. Mass calculations are based on well-established optical models with mSPR = CSPR ∆RU, where the constant CSPR = 6.5 × 10-2,114 and ∆RU is a dimensionless quantity proportional to the change in refractive index at the interfacial region.

EXPERIMENTAL

24

3.4 SYNTHETIC, PREPARATIVE AND ANALYTICAL PROCEDURES

This section describes some of the general synthetic, preparative and analytical procedures. For a more detailed description the reader is referred to the experimental sections of the respective articles/papers.

3.4.1 AB2C monomer synthesis

Dual-functional dendritic materials were synthesized using AB2C functional monomers. Scheme 3.1 shows the synthesis of the two AB2C monomers used in this thesis. Starting from Trizma® hydrochloride two of the hydroxyls (B-groups) were protected with an acetonide group (i). In the second step (ii) anhydride coupling to the amine facilitates introduction of a ‘click’-group (ene or azide) at the C-position (C-group). In the third step (iii) a carboxylic acid (A-group) is introduced by addition of succinic anhydride to the unreacted hydroxyl. In the fourth and final step (iv) the carboxylic acid is activated into an anhydride.

OH

OHH2N

Trizma ® hydrochloride

(i)

O

OH2N (ii)

O

OOC C O

OHN

O

OH OH OH

CO OO

O

OHNO

O

C

(iii)

(iv)

OHO

OO

OHNO

O

C

OO

O

2

C =

N3

ene

azide

Scheme 3.1. Synthesis of Trizma® based AB2C monomers. (i) DMP, pTSA, DMF, RT; (ii) TEA, DCM; (iii) DMAP, DCM, 0 °C; (iv) DCC, DCM, 0 °C.

EXPERIMENTAL

25

3.4.2 Miscellaneous

The synthesis of other important compounds used in this thesis is presented in Scheme 3.2.

O O OOH

OOH

O

OO O

DMAPPyr

DCM

DCCDCM

O O

OO

O2

O O OO

ON N

N NO2

Cl

[1]

HO N NN NO2

Cl

DMAPTHF

[1]

HO O OH O O O O OO

O OO O

ODMAPTHF

[1]

NHHN

SHN

O

O

NHHN

SHO

O

OEDC

DMAPDMF:DCM

NH2

OH

OO O

DMAPDCM

O OHO

O

O OO

O

2DCCDCM

HO

HO

HN

O

O

O

HOHO

NH

TEADMSO

A.

B.

C.

D.

E.

HO O OHp-TSA

Toluene

HO

O

SHO O OHS SH

O

O

F.

OHOHO

HOHO

OH

OAcOAcO

AcOAcO

OAc

OAcOAcO

AcOAcO

OH

OAcOAcO

AcOAcO

O

CCl3

NH

HOCl

OAcOAcO

AcOAcO

O

Cl

BF3. Et2O4Å sieves

DCM

NaN3

DMSO80oC

MeOHNaOMe

Pyr

Ac2ODMAP

CCl3CNDBUDCM

BnNH2

THF

OAcOAcO

AcOAcO

O

HOOHO

HO

HOHO

O

BF3. Et2O4Å sieves

DCM

OAcOAcO

AcOAcO

O

N3

MeOH

NaOMe OHOHO

HOHO

O

N3

G.

Scheme 3.2. Synthetic overview of (A) allyloxy anhydride, (B) PEG-diallyl, (C) allyloxy-disperse red 13, (D) PEG-dithiol, (E) acetylene-dopamine, (F) acetylene-biotin, (G) acetylene- and azide-mannose.

EXPERIMENTAL

26

3.4.3 Swelling and leaching

All swelling and/or leaching samples were dried in vacuum at 50 °C for at least 12 h prior to testing. A minimum of three samples per material was used and the degree of swelling was calculated according to:

% swelling =𝑚𝑠𝑤𝑜𝑙𝑙𝑒𝑛 − 𝑚𝑑𝑟𝑦

𝑚𝑑𝑟𝑦× 100

The gel fractions after leaching were calculated as:

%gel fraction = (1 − (𝑚𝑖𝑛𝑖𝑡𝑖𝑎𝑙 − 𝑚𝑙𝑒𝑎𝑐ℎ𝑒𝑑

𝑚𝑖𝑛𝑖𝑡𝑖𝑎𝑙)) × 100

3.4.4 Off-stoichiometric thiol-ene (OSTE) networks

The off-stoichiometric thiol-ene (OSTE) formulations for tunable 3D networks (paper I) and microfluidic device prototyping (paper II) were prepared using two slightly different procedures. In the first case (paper I) prepolymer formulations, with predetermined off-stoichiometric ratios, were mixed thoroughly by heating and vortexing. This was followed by addition of 0.58 % w/w of photoinitiator 2,2-dimethoxy-2-phenylacetophenone (DMPA). No photoinitiator was used for PDMS based networks. Rigid triazine-based formulations and soft PDMS-based formulations were prepared as bulk mixtures while hydrogel formulations were prepared as 33 % w/w (dry content) solutions in EtOH. Curing was induced by passing the formulations on a fusion UV-belt subjecting the samples to a total dose of 372 mJ·cm-2. Hydrogel samples were dried for at least 12h under vacuum before subjected to any further testing. In the second case (paper II) prepolymer formulations were mixed in predetermined off-stoichiometric ratio (in bulk) and 0.5 % w/w of photoinitiator Lucirin TPO-L was added. Formulations were degassed in a vacuum chamber and irradiated with the table-top EFOS UV-lamp for 30 s (365 nm, 4 mW·cm-2).

EXPERIMENTAL

27

3.4.4.1 Postmodification of OSTE networks

The modification of OSTE networks with a red ene-functional dye (allyloxy-disperse red 13) was conducted by subjecting samples to dye solutions in chloroform (2 mg·mL-1) with a catalytic amount of DMPA photoinitiator. Samples were exposed to a dosage of 372 mJ·cm-2 to initiate the reaction between network-thiols and alkene-dye. For surface modification of rigid network with heneicosafluorododecyl acrylate a corresponding procedure was adapted using heptane instead of chloroform. For hydrophobic modifications of microfluidic device surfaces ene- and thiol-functional PDMS (vinylmethylsiloxane homopolymer and poly(mercaptopropyl)methylsiloxane) were diluted as 10 % w/w solutions with 0.5 % w/w ITX photoinitiator. The OSTE surfaces were covered with the PDMS solutions followed by induction of regiospecific modification using patterned stencil masks irradiating for 60 s (365 nm, 4 mW·cm-2).

3.4.5 Self-Assembled Dendritic Monolayers (SADMs)

Gold (Au) surfaces used for making self-assembling dendritic monolayers (SADMs) were prepared on polished silicon wafers by sputtering of 100 Å Ti followed by 1500Å Au. These were cut into 8x8 mm2 pieces and cleaned in a 1:1:5 solution of hydrogen peroxide (25%), ammonia (30%), and Milli-Q water at 85 °C for 10 min and dried in nitrogen gas. The adsorption was done in the SPR by 15 min long injections of dendrimer/dendron solutions (100 µg·mL-1) using a laminar flow with a flow rate of 5 µL·min-1. The dendritic solutions were prepared using either sodium acetate buffer (AcB; 0.1M, pH 5.5) or 70% EtOH as solvent. The adsorption process was monitored in situ through SPR measurements.

EXPERIMENTAL

28

3.4.6 Bacterial binding to surfaces

An Escherichia coli strain (MS7pPKL4) possessing type I fimbrae (Fim+) with specific interaction towards mannose115 was used for the attachment to SADM surfaces. Solutions with 3 × 107 cells·mL-1 in phosphate-buffered saline (PBS) were added to flow chambers containing SADM surfaces for 30 min under static conditions. Non-adhered or weakly adhered cells were removed by rinsing the flow chambers with 300 mL of PBS at a flow rate of 2.5 L·min-1. The attached bacteria were fixed and stained using a solution containing formaldehyde (2.5%) and the dye Acridine orange (0.5 mg·mL-1, Merck). Additional staining with 4’,6-diamino-2-phenylindole (DAPI, Vector Laboratories) was done before analysis by fluorescence microscopy. A minimum of four surfaces per SADM-type was analyzed.

3.4.7 Dendritic PEG hydrogels

The dendritic PEG hydrogels were prepared by dissolving the ene-functionalized dendritic precursors (either dendrimer or PEG-dendritic hybrids) with thiol-functional PEGs (equimolar ratio between thiol and enes). A small amount of photoinitiator Irgacure 2959 (I2959) was added to the solutions which were thoroughly mixed by vortexing and transferred to a suitable mold. UV-curing was induced by shining light (365nm) for 10 min at 9mW·cm-2. Gels for swelling and rheological measurements were prepared in EtOH solution with [I2959] = 0.5 wt%. These gels were cured between two glass coverslips spaced 0.5 mm apart to obtain flat hydrogel films. The hydrogels intended for the cell viability study were made in Milli-Q water solution with [I2959] = 0.05 wt%. These gels were cured directly in the bottom of cell culturing wells.

EXPERIMENTAL

29

3.4.8 Determination of residual catalytic copper after CuAAC reaction

After CuAAC-mediated conjugation of bioactive moieties to the heterofunctional dendritic scaffolds removal of the catalytic copper was done through a two-step purification procedure. This included initial purging through a neutral AL2O3 column and dialysis treatment against a 0.5 % (w/w) EDTA solution (pH 7.0). The efficiency of the purification process was monitored by graphite furnace atomic adsorption spectroscopy (GF-AAS). Briefly, lyophilized samples of precursor polymers were degraded by solubilizing in nitric acid (pH 2.0) for a minimum of 48h. Measurements were based on three replicate readings per sample. Calibration was done using ultrapure water with standards in known concentrations (10, 50 and 100 µg·L-1). All results were presented with the copper concentration of blank reference solutions subtracted. The detection limit was 0.37 µg·L-1.

3.4.9 Cell viability through lactate dehydrogenase (LDH) assay

The hydrogels cured in the bottom of cell culturing wells were washed twice with cell culture medium after which normal human dermal fibroblasts (NHDF; Karocell Tissue Engineering AB, Stockholm) were seeded topologically at 6000 cells per well. The culture plates were incubated at 37 °C in humidified 5% CO2 for 24h, 72h and 7 days (triplicate samples for each time-point and gel type). The samples were aspirated and the cell viability analyzed by the lactate dehydrogenase (LDH) content in the supernatants. LDH is a marker for cell membrane integrity and the levels were measured spectrophotometrically by evaluating LDH-mediated conversion of pyruvic acid to lactic acid. This was performed at the C-laboratory at Sahlgrenska University Hospital, Göteborg, Sweden.

EXPERIMENTAL

30

3.4.10 FRAP

Rectangular sample rods (50×10×5 mm3) were prepared of bovine femur bones received from an abattoir. All samples were wet sanded (grain size 80) until an even and smooth surface was obtained. Artificial fractures were created by sawing the rods in half. Fiber reinforced adhesive patches were applied topologically on the bone sides perpendicular to the induced fracture, thus bridging the fracture. In the first step, a thin layer of primer solution was applied to the bone surface followed by a thin layer of matrix. Primers were prepared as 50 mg·mL-1 solutions in ethanol or water:ethanol (1:1) depending on the solubility of the primer. The matrix was composed of thiol-ene formulations of the triallyl and trithiol monomers having a 1:1 ratio between thiol and enes. The first layer was cured by passing through the fusion UV belt for a dose of 350-700 mJ·cm-2 (320-390 nm). Thereafter an additional five layers of 90/0 E-glass fibers embedded in the thiol-ene matrix were applied and cured through an additional dose of 1050-1750 mJ·cm-2 (320-390 nm). A tip of a knife of Irgacure 184 (I184) was used as photoinitiator and once cured the samples were immersed in 0.9 % w/w solutions of sodium chloride in deionized water in order to mimic in vivo conditions. Tensile tests were performed by inserting metal pins through drilled holes in the ends of the bone, which were attached by metal wires to the load cell of the Instron. The maximum load(N) was measured when pulling the bone pieces apart calculating the shear strength by dividing the maximum load with the adhesive area. Only samples with adhesive failure were used.

RESULTS AND DISCUSSION

31

4. RESULTS AND DISCUSSION

4.1 FUNCTIONAL OSTE POLYMERIC NETWORKS

The first part of the results and discussion section concerns the creation and functionalization of off-stoichiometric thiol-ene networks (OSTEs). These materials constitute a versatile platform for creating functional crosslinked polymers in applications ranging from soft hydrogels to functional surfaces for microfluidic applications.

4.1.1 Fabrication of OSTE networks

Three general network types have been evaluated which included rigid triazine-based organic networks, aqueous PEG-based hydrogels and inorganic PDMS-based soft networks (Scheme 4.1).

Scheme 4.1. Three types of functional OSTE networks.

O O O O O O OO

O O

O

OO

OO

SHSHHS

HSO

OO

O

Tetrathiol

Thiol PDMS

S iO

S iO

S i

HS

Vinyl PDMS

S i O Si O S iN N

NO

O

O

Triallyl

N NNO O

OOHS O SH

OHSO

O

O

Trithiol

Rigid network Hydrogel network Soft network

++

+

Initi

ator U

V Initi

ator U

V

PEG-diallyl

UV


32

Table 4.1. Properties and compositions of OSTE materials

i Values for rigid, hydrogel, and soft networks were obtained by tensile testing while values for OSTE microfluidic device materials and reference materials were obtained using DMTA. ii Leaching was done in chloroform for all materials except hydrogel networks which were leached in water.


33

By combining readily available commercial monomers1 with thiol and ene functionalities, and varying their feed ratios, networks with excess of ene or thiol functional groups were obtained (Table 4.1). All networks were cured in timeframes ranging from sub-seconds to a few minutes depending on the UV-irradiation source. In the first case rigid triazine-based triallyl and trithiol monomers were mixed with a small amount of photoinitiator and cured in a solvent-free process. The ene:thiol ratios were varied between 1.9 excess of the enes to 2.5 thiol excess. For the hydrogels, PEG3k-diallyl and a pentaerythritol-based tetrathiol crosslinker were dissolved in ethanol and cured with a small amount of photoinitiator. The cured samples were solvent-exchanged from ethanol to water. The soft inorganic PDMS networks were cured by mixing vinyl-terminated polydimethylsiloxanes with mercaptopropyl-functional methylsiloxane crosslinkers. These networks were readily cured to high conversions without any addition of photoinitiator (using unfiltered light including 254 nm)116. The materials for OSTE microfluidic devices and reference material are also presented in Table 4.1, however they will be discussed in later sections.

4.1.2 Mechanical tuning of OSTE networks

The ability to tune the mechanical properties of the OSTE materials is critical for their versatility in various applications. For example, the modulus influences how a cell interacts with cell culturing scaffolds constructed from OSTE networks. Likewise, in other medical applications such as microfluidics and microarrays, there may be a critical toughness that is required for the handling and manufacturing properties of these devices. As can be seen in Table 4.1, the formed networks cover a wide range of elastic moduli stretching from very soft hydrogels of around 20 kPa to more rigid networks of around 800 MPa. Generally the materials

1 All except PEG3k-diallyl which was obtained by allyl functionalization of commercial PEG3k-OH2.


34

Figure 4.1. Tuning of Tg and E-modulus by changing the OSTE ratio of triallyl and tetrathiol monomers. exhibited high gel fractions of around 90 % or more and narrow glass transitions, which is indicative of effective thiol-ene conversions and formation of homogenous networks. The delayed gelation characteristics of the thiol-ene polymerization furthermore led to materials exhibiting low shrinkage as compared to acrylate system, which typically exhibit an early onset of gelation and deformation. As mentioned, the type of monomers used will affect the mechanical properties if the OSTE networks. Another very important factor is the off-stoichiometric ratio which also has large impact on the network moduli. An increase in the off-stoichiometric ratio typically brings about a decrease in both the elastic modulus and Tg of the materials, due to the introduction of more chain ends in the network. This effect is more pronounced for materials with thiol-excess (compared to the allyl-excess ones). This may be due to the more flexible structure of the trithiol monomer or the formation of small amounts of disulfide cyclization coupling products. As seen in Figure 4.1 both the E-modulus and the Tg of a triallyl-tetrathiol network can be effectively tuned by increasing the thiol-excess from 0% to 100%. The increased thiol excess results in a decreased E-modulus at room temperature from around 1300 MPa to 120 MPa and a drop in Tg from over 80 °C to 35 °C. By careful choice of monomer species and their off-stoichiometric ratio, materials with the desired mechanical properties may thus be obtained.

OSTE-thiol (0 to 100)

1 1.2 1.4 1.6 1.8 210

2

103

104

E-m

odul

usat

25

°C (M

Pa)

35

40

45

50

55

60

65

70

75

80

85

thiol:ene p roportion

T g (

°C)

E-modulusGlass transition


35

4.1.3 Functional availability and modification of OSTE networks

The underlying idea for introducing off-stoichiometric ratios was to create materials with a controlled amount of available functional groups which can be utilized for further postmodification. By implementing monomers with appropriate alkene moieties an almost perfectly alternating reaction between the thiols and enes may be obtained. In the case of the OSTE formulations this leads to materials with a controlled amount of leftover unreacted ene or thiol groups. In Figure 4.2 a) the presence of unreacted thiols (R5) and enes (R1) for the cured rigid networks is confirmed by FT-RAMAN stretches at 2575 cm-1 (S-H) and 1644 cm-1 (C=C), respectively.

Figure 4.2: a) Presence of excess ene and thiol functional groups as confirmed by FT-RAMAN, and b) excess thiol groups detected by allyl-functional red dye incorporated to rigid, soft and hydrogel networks.


36

To further visualize the availability of excess thiol groups an allyloxy-functional red dye was reacted with the networks and used as a visual colorimetric agent (Figure 4.2 b)). When observing the soft PDMS networks (S1-S4) a clear increase of colorimetric agent with the increased thiol excess is visible. The PDMS and hydrogel networks with thiol excess were furthermore modified throughout the bulk of the materials since these swelled in the chloroform used for the modification. The rigid networks however do not swell in chloroform and were hence only functionalized at the surfaces. The ability to tailor material surfaces is a highly desired feature in many application areas. To demonstrate the versatility of the OSTE material platform for creating surfaces with altered surface structures and functionality, the tailoring of surfaces from rigid network R5 (1.9 thiol excess) is here chosen as an illustrative

Figure 4.3. The effect of changes in surface roughness and chemistry shown by contact angle measurements and SEM images. a-b) flat and microstructured surface of R5 rigid network (1.9 thiol excess); c) structured R5 surface postfunctionalized with heneicosafluorododecyl acrylate to create a superhydrophobic surface; d) snapshots (1s between) of a water droplet being pressed against the superhydrophobic substrate.


37

example (Figure 4.3). By introducing a micron-size pattern, through soft moulding of the R5 matrix, the wettability of the surface could be significantly decreased going from a contact angle of 60o to 120o (a-b)). Additional covalent modification with a heneicosafluorododecyl acrylate (c)) resulted in a superhydrophobic surface with a contact angle of 173o. The surface was so hydrophobic that the 5 µl size water droplet was unable to be transferred to the surface (figure d)). Conversely, the droplet volume had to be changed to 10 µl (figure c)). Superhydrophobic surfaces could be potentially interesting for a number of different applications including self-cleaning surfaces or microarrays requiring specific wetting of the surface.117 Furthermore, it is not hard to imagine the range of achievable surface properties considering the wide array of available commercial ene monomers and patterning possibilities.

4.1.4 OSTE-based microfluidic devices

In this section it will be shown how the versatile properties of OSTE-based materials may be used to bridge the gap between research prototyping and commercial production of microfluidic devices. Traditionally, conceptual microfluidic device development has been done using PDMS-based networks (Sylgard 184) because of their ease of use and attractive elastomeric properties. However, the translation of these prototypes into commercial products has been hampered by factors such as variable and hard to tailor surface chemistry, poor solvent barrier properties, and inadequate mechanical strength. Another factor is the un-scalable or time consuming assembly steps, such as slow thermal curing or manual handling of thin films. The vast majority of commercial devices are made from medical grade thermoplastic materials such as PMMA or polystyrene. However these commonly require expensive machines and tools for injection molding or hot embossing along with complicated bonding of device layers and time consuming surface modifications. The novel OSTE-material platform has the


38

Figure 4.4. Dynamic mechanical thermal analysis of OSTE microfluidic device demonstrator materials. a) Thiol(90); b) Allyl(30); c) PDMS-thiol(20). potential to overcome all the above listed limitations by providing important features such as: Tunable mechanical properties that can mimic both PDMS and commercial thermoplastics, simple one-step surface modifications, and leakage free sealing through direct UV-bonding. It enables fast prototyping through replica molding via a fast UV-curing process, and has low shrinkage stress which prevents deformation of the thin geometries common in microfluidic devices. To demonstrate the feasibility of the OSTE materials towards microfluidic prototyping three demonstrator materials were developed (Table 4.1). The first material, Thiol (90), is composed of tetrathiol and triallyl monomers (Scheme 4.1) and has a large excess of thiol functionality (90%). The large off-stoichiometric ratio results in a lowered crosslinking density and a material that is pliable and rubberlike. Here it is used as a polymer that can be easily applied for micromolding which seals well to other surfaces


39

when heated to its Tg at 35 °C (E-modulus drop from 250 MPa to 10 MPa, Figure 4.4 a). In the second sample material, Allyl(30), the more rigid trithiol monomer is used in combination with the triallyl (30% excess of allyl groups). This material has an E-modulus close to commercial PMMA (1740 and 1800 MPa, respectively, Table 4.1), but it too softens to about 10 MPa at the Tg 68 °C (Figure 4.4 b)) enabling simple release after molding processes. It is used as a hard substrate which can be covalently bonded to the Thiol(90) material and machine drilled without cracking. The final material, PDMS-thiol(20), is composed of a thiol-PDMS crosslinker (as oppose to silyl hydride in Sylgard 184) that crosslinks with vinyl-PDMS. It is a soft material with an E-modulus of 350 kPa at room temperature (Figure 4.4 c)), which is lower, but comparable, to that of Sylgard 184 (Table 4.1). The PDMS-Thiol(20) is used for creating valves or soft membranes. In Figure 4.5 selected examples of how these materials can be applied in the fabrication of key features in microfluidic devices is demonstrated. In the first case (Figure 4.5 A), hydrophobic retention valves are created by first creating a top channel layer by standard molding, using Thiol(90) matrix and a patterned silicon wafer (SU-8). This is then combined with a flat bottom layer of Allyl(30) matrix. In the second step, hydrophopic vinyl-PDMS and thiol-PDMS were grafted using a patterned stencil mask. In the third and final step the layers were aligned and heated above 35 degrees to maximize the contact area between the layers. The layers were then dry bonded by exposure to unfiltered UV-light (including 254 nm) initiating polymerization between the excess thiols and enes at the interphase. Upon filling the drilled inlet holes with green colored water (candy dye) the liquid slowly filled the channels until the hydrophobic areas where it stopped. In the second example (Figure 4.5 B) pneumatic valves (controlled by pressurized gas) are made my laminating a thin membrane of PDMS-thiol(20) between two crossed-channeled Allyl(30) pieces using a similar bonding procedure. The pressure required to close of a flow of 10 µl·min-1 at different membrane thicknesses is plotted in the inset


40

Figure 4.5. OSTE-based microfluidics. A) Hydrophobic retention valves; B) Pneumatic valves; and C) Chip-to-world connectors. graph. The performance is comparable to valves made from Sylgard 184.118 Finally, in Figure 4.5 C it is demonstrated how the Allyl(30) material can be used to make chip-to-world connectors by first molding fluidic ports (based on NanoPortTM) using a PDMS mold. These were threaded to fit standard coned fittings and finally aligned to the drilled inlet holes and dry bonded using the same method as before (Figure 4.5 A). The fittings withstood pressure exceeding 4 bars before rupturing which is above the needs of most microfluidic applications.

Alignmentmarks

UV-patterned graft

OSTE-Thiol (channel layer)

OSTE-Allyl (bottom layer)

Alignmentmarks

UV-patterned graft

S S S

S S S

grafted PDMSlayer

grafted PDMSlayer

ii) UV-bonding

A. Hydrophobic stops

PDMSmold

Threaded holeOSTE-Allylprepolymer

Polycarbon ate

UV curing UV curing (unfiltered)

35 °C

OSTE-Allyl

OSTE-AllylconnectorOSTE-Thiol

400 µm

400

µm

OPEN CLOSED

OSTE-Allyl (30)

OSTE-Allyl (30)OSTE-PDMS (20)

50 150 250 350 450010203040506070

membrane thickness (µm)

clos

ing

pres

sure

(kPa

)

flow channelB. Pneumatic valves C. Connectors in OSTE

Closing pressure vs. membrane thickness

pneumatic control channel

i) Heat aboveTg 35˚C


41

4.2 SELF-ASSEMBLED DENDRITIC SURFACES

In the previous sections, off-stoichiometric thiol-ene networks were used to create functional surfaces. In this section, multivalent sensor surfaces are produced by relying on self-assembled dendritic monolayers (SADMs, Scheme 4.2). These surfaces are further evaluated for their ability to attract and scavenge pathogenic E. coli bacteria. The multivalently presented

Scheme 4.2. Assembly and bacterial interactions of SADMs.

ligands in dendrimers have the potential to enhance target interactions with biological receptors and cell surfaces,73, 74 which has spurred the interest for using them as interaction amplifiers in surface-recognition applications. By including an organo-sulfur core moiety, the produced dendritic scaffolds were able to spontaneously self-assemble onto gold sensor surfaces. The assembly process is significantly affected by the structure of the scaffolds as well as the employed assembly conditions.87, 88 Therefore, a library of dendritic molecules was synthesized and their assembly process evaluated with respect to the following structural parameters: (1) impact of thiol (SH) or disulfide (SS) core functionalities, (2) dendron-size dependency (branching), (3) effect of distance between sulfur core and main dendritic wedge; and (4) peripheral end-group functionality. Finally, hydroxyl- and mannose-functionalized surfaces are evaluated for their scavenging properties of uropathogenic E. coli Fim+ bacteria.

SADMs (Self-AssembledDendritic Monolayers)

1.Assess assembling propertiesof dendritic library

2.Study bacterial adhesionof dendritic surfaces

S

O


42

4.2.1 Synthesis of monodisperse sulfur-containing dendritic library

The assembly of the dendritic bis-MPA library relied on two different routes. In the first case (Scheme 4.3 A) a divergent growth approach was used by starting from a bis(2-hydroxyethyl)disulfide core that was reacted by sequential growth and activation steps to produce the disulfide dendrimers G1-SS-OH4 and G3-SS-OH16. By a cleavage reaction using a DTT agent the corresponding dendrons G1-SH-OH2 and G3-SH-OH8 could also be produced. This set of dendritic precursors will self-assemble to identical dendrons on the surface (within corresponding generations), enabling a direct comparison of the disulfide and thiol core effect. Furthermore, they are spaced with a small ethyl linker of 3.9 Å between the sulfur units and the main dendritic wedge (calculated using CambridgeSoft Chem3D Pro). In the second case (Scheme 4.3 B) the spacer distance was increased to 13.9 Å (1 nm increase). This was accomplished by first reacting the disulfide core with an azide functional spacer (4-azidobutanoic acid). This was then followed by CuAAC-mediated efficient coupling of acetylene cored dendrons (G1 and G3) to obtain the spacer dendrimers G1-SS-OH4-sp and G3-SS-OH16-sp, through a convergent growth approach. In order to demonstrate the potential of using this dendritic platform to introduce specific cell-interactive moieties to a surface, the spacer disulfide dendrimers were further decorated with multiple carbohydrate (mannose) ligands by a facile two-step modification procedure. In this case the peripheral hydroxyl groups of G1-SS-OH4-sp and G3-SS-OH16-sp were first converted in to acetylenes using mild anhydride chemistry119 followed by efficient conjugation of a azide-functional carbohydrate (azido a-D-mannopyranoside) to produce the dendrimers G1-SS-Mannose4-sp and G3-SS-Mannose16-sp. The effect of the represented structural variations on the SADMs will be demonstrated in following sections.


43

OH

OH

O

OH

O

HO

O

O

NN

N

NNN

O

O

O

OS

S

N3

N3

O

O

O

OS

S

OH

HO

SS

N3

OH

O

G1-SS-O

H4 -sp

(and G3-SS-O

H16 -sp)

OO

SS

O

O O

OO O

OO

O OO

O O

DM

AP

PyridineD

CM

Grow

th (g)

DO

WEX

(R) (H

+)A

ctivation (a)

OO

SS

O

HO

HO

OO

H

OH

(g), (a), (g)

SO

O

OO OO

OOO

O

O

OO

HO

H

OH

OH

O

OH

OH

OO

HO

H

SO

O

OO OO

OOO

O

O

OH

OH

O

HO H

O

O

HOHO

OH

OH

O

OH

8

DPTS

DC

CD

CM

CuSO

4N

aAsc

THF:H

2 O

OH

2

OSH

O

O OD

TTTEAD

CM

Ar(g)

Cleav(c)

OSH

O

HO

HO

G3-SS-A

n8

(a)

(a)

G3-SS-A

n8SH

OO

OO OO

OOO

O

O

OH

OH

O

HO H

O

O

HOHO

OH

OH

O

(a)

OO

O

O2

O

HO

OH

OH

OO

HN

3

OOO

OOO

O

O

NN

N

NNN

O

O

O

OS

S

O

OH

HO

HO

OH

O

NNN

OOO

O

OH

HO

HO

OH

O

N NN

O OO

OHO

OH O

HO

OH

NNN

OOO

O HO

OHO

HO

OH

N NN

O OO

G1-SS-M

an4 -sp

(and G3-SS-M

an16 -sp)

CuSO

4N

aAsc

THF:H

2 O

DM

AP

PyridineD

CM

G1-SS-O

H4

G1-SH

-OH

4

G3-SS-O

H16

G3-SH

-OH

8

A. D

ivergent growth:

B. C

onvergent growth:

Scheme 4.3. A

ssembly of sulfur cored dendritic library

G3 G1(c)


44

4.2.2 Self-assembled dendritic monolayer (SADM) formation

The SADM mechanism was monitored by surface plasmon resonance (SPR) technique which enables in situ high resolution monitoring of the adsorption process. 100 µM solutions of the hydroxyl functional dendritic scaffolds in sodium acetate buffer (AcB, 0.1 M, pH 5.5) were adsorbed during 15 min long injections. In Figure 4.6 a) the adsorption profiles of the short spacer thiol and disulfide compounds are shown. As can be seen, the G1 compounds yield equally dense layers, a behavior similar to linear disulfides and thiols.120 However unlike the linear counterparts, which commonly exhibit identical binding profiles, the G1-disulfide has a slower packaging profile compared to the more instant G1-thiol. This difference is even more profound for the more branched G3 compounds where the absorption of the G3-disulfide is 54% lower than the G3-thiol. The final adsorption level is even lower than the two G1 compounds indicating severe hampering of the binding process. This behavior has also been observed for branched alkanethiols and disulfides,87 where unfavorable intermolecular interactions between the substituting chains of disulfides prohibit close packing.

Figure 4.6. The self-assembly process of hydroxyl-functional dendrimers and dendrons followed by SPR. a) Injection curves of thiol-dendrons and disulfide dendrimers; b) Surface density of chains and hydroxyl groups.


45

The G3 compounds also show a drop upon at the end of the injection indicating presence of low affinity binding (non-covalent) which is disrupted during the purging with pure AcB. For the compounds with the longer spacer (G1-SS-OH4-sp and G3-SS-OH16-sp) an increased adsorption (mass) was observed compared to the shorter spacer counterparts (data not shown). However, due to the different sizes of the adsorbed dendrons, the actual density of dendrons and end-groups on the surface was obtained by normalization against the molecular weight (Figure 4.6 b). As can be seen, the G1-SS-OH4-sp has a lower functional density compared to the short G1 compounds. In contrast, the G3-SS-OH16-sp, with a functional density of 5.5 OH·nm-2, has the highest density of all the scaffolds. This represents a 70 % increase in density compared to the G3-SS-OH16, and it is in the same range as a well-packed layer of linear alkanethiols.121

4.2.3 Bacterial scavenging of SADM surfaces

A critical part of the process by which urinary tract infections occur is the attachment of E. coli Fim+ bacterial cells to the bladder epithelium.122 This attachment is facilitated by the bacterial FimH lectin which strongly binds to mannosylated proteins located in the bladder epithelium. Glycoconjugated dendrimers have shown great promise as multivalent high-affinity antagonists for the bacterial lectin FimH when applied in the bulk phase.75-77 Therefore, the scavenging ability of the SADM glycoconjugate surfaces for E. coli Fim+ bacteria was evaluated. Four different surfaces were prepared from the four spacer dendritic compounds and their adsorption was monitored by SPR (Figure 4.7 a)). In order to fully solubilize the dendritic scaffolds (mannose in particular) a 70% ethanol (EtOH) solution was used, which has the benefit of keeping the surfaces sterile before the bacterial attachment. Interestingly, this change of solvent resulted in a significant 47% decrease in adsorption for the G3-SS-OH16-sp (figure 4.6 and 4.7), while the G1-SS-OH4-sp remained unchanged.


46

Figure 4.7. Assembly of OH- and Mannose-functional dendrons of generations 1 and 3 (a)) and attachment of E. coli Fim+ cells (b)). This could be an effect of that EtOH is a better solvent for the G3 bisMPA dendritic scaffold, resulting in a more extended geometry that occupies a larger surface area upon attachment; thus, affecting the binding affinity negatively. When comparing the adsorption of the hydroxyl- and mannose-functional scaffolds, it is clear that the resulting surface density of hydroxyl groups is higher than the corresponding mannose group densities. This is probably a size-effect since the G3-SS-Mannose16-sp molecular weight is over five times larger than the G3-SS-OH16-sp (Mn = 4604 g·mol-1 and 890 g·mol-1, respectively). Nevertheless, the density of attached G3-Mannose dendrons is still over twice as high than the reference (linear 5kDa PEG-SH) with a similar molecular weight. After the binding process the sensor surfaces with the four different SADMs were incubated with E. coli Fim+ cells for 30 minutes under static conditions followed by a purge with 300 mL of PBS buffer to remove any poorly bound cells. The two hydroxyl-functional SADMs showed low binding affinity similar to that of an octanethiol reference surface(Figure 4.7 b)). The G3-Mannnose surfaces displayed a superior scavenging capability, compared to both the hydroxyl functional surfaces as well as the G1-Mannose surface. In spite of its lower end-group density, the binding still corresponds to a 2.5-fold increase compared to the G3-OH counterpart. Hence, mannosylated SADMs represent a facile methodology to increase bacterial-surface interactions.


47

4.3 FUNCTIONAL PEG HYDROGELS

PEG-based hydrogels are very attractive materials for producing high water content reservoirs that mimic natural extracellular matrices, due to their biocompatible, hydrophilic and elastic nature. However, the inherent inertness of PEG commonly requires introduction of additional biorelevant cues to the hydrogel to enhance the cell interaction capability. One way to introduce such additional functionality is to use the previously presented OSTE-strategy for hydrogel networks. In this section, another approach based on heterofunctional dendritic scaffolds is introduced. It combines the advantageous properties of thiol-ene networks with the possibility to introduce additional biorelevant moieties through the presence of a second orthogonal ‘click’ group. The ability to precisely tune the architecture of such dendritic scaffolds enables fabrication of hydrogel networks with highly controlled mechanical and biochemical properties.

4.3.1 Synthesis of dendritic binders

The synthesis of the heterofunctional dendritic scaffolds relied on the use of AB2C-type dendritic monomers, where A is an anhydride-activated carboxylic acid, B are acetonide protected hydroxyl groups, and C is an orthogonal azide or ene ‘click’ group (Scheme 4.4). In the first case (Scheme 4.4 A) two G1 dendrimers were produced starting from a trimethylolpropane (TMP) core adapting a three step modification procedure. The first step (dendritic growth, (i)) involved efficient and mild anhydride coupling of AB2C monomers, presenting either ene or azide C-functionality, to the TMP core. This was then followed by mild acidic deprotection (activation (ii), a THF:H2O mixture was required to fully solubilize the hydrophobic dendrimer scaffolds) of the acetal protecting groups, and in a final step the deprotected hydroxyl groups were activated (iii) using either azide or ene functional anhydrides obtaining the dendrimers TMP-G1-(ene)3-(N3)6 and TMP-G1-(ene)6-(N3)3. These two scaffolds display an


48

HO OHHO

O O

O O OO

NHO

OOO

ONH

O

O O

O O

OO

OO

NH

OO

OO

N3O

N3

OO

ON3

ON3

HNO

OO

ON3

O N3

O O

O OO

ONHO

N3 OOO

ONH

O

O O

O O

OO

OO

NH

OO

OO

O

OO

O

O

HNO

OO

O

O

N3

N3N3

2

2

i ii

TMP

A.

TMP-G1-(ene)3-(N3)6

TMP-G1-(N3)3-(ene)6

O

O OO

HO

HOOH

OH

On

O OOO

O

NH

ON3

OO O

O

O

OO

OO

O

O OO

O

O

O

O

O

OHN

OO

N3

O

ONH

OO

N3

O

OOO

HNO

N3

O OO

NHO

N3

O OOO

O

O OOO

O

OOO OO

OOO OO

OOO

O

O

OOO

OO

OO O

OO

OO O

OO

HO Hn

OO

O

HN

ON3

OOO

O

O

OO

OO

O

n

n

PEG

O O

O OO

ONHO

N3

2

O O

O OO

ONHO

N3

2

i ii

PEG-G1-(N3)2-(ene)4

PEG-G2-(N3)4-(ene)8

i ii

i ii

B.

i

iv

O

O

2

OO

N3

2

OO

OO

O

2

OO

OO

2

(PEG Mn = 6kDa and 10kDa)

(PEG Mn = 6kDa)

iii

iii

iii

iiiO

OOO

O

2

Scheme 4.4. Synthesis of heterofunctional dendritic scaffolds. i) DMAP, Pyr, DCM; ii) pTSA, H2O:THF; iii) DMAP, Pyr, DCM iv) DOWEX® (H+), MeOH.


49

inversed and exact number of ene and azide groups available for crosslinking reactions and functional modification via TEC and CuAAC chemistries. In the second case (Scheme 4.4 B) a similar procedure was used, but this time starting from a linear PEG core, to produce linear-dendritic hybrids PEG-G1-(N3)2-(ene)4 with PEG molecular weights of 6 kDa and 10 kDa, and with a functionality of two azides and four peripheral enes. This concept was further explored by equivalent modification of a produced PEG-G1-(OH)4 (Scheme 4.4 B), obtaining PEG-G2-(N3)4-(ene)8 with twice the number of orthogonal functionalities, i.e. four azides and eight enes. While the focus here lies on the use of these heterofunctional scaffolds as hydrogel crosslinkers, it is important to point out that their dual functionalities can accurately be tailored for other multipurpose studies.

4.3.2 Functional hydrogels

In order to activate the dendritic scaffolds the highly efficient CuAAC reaction, between acetylene-functional biorelevant moieties and the scaffold azide groups, was chosen as a facile route. The activated scaffolds could thereafter be crosslinked through the UV-induced TEC reaction with linear PEG-dithiols. In Figure 4.8 this concept is illustrated by the activation and subsequent crosslinking of the PEG-G1-(N3)2-(ene)4 scaffold forming hydrogels that include either biotin, L-dopamine (DOPA) or D-mannose (Man). By performing the functionalization step before the hydrogel formation, a high level of control is maintained over the amount of groups included as well as the integrity of the scaffolds. In Figure 4.9 the activation of PEG(6 kDa)-G1-(N3)2-(ene)4 with acetylene functional DOPA is followed by MALDI-TOF and 1H-NMR. As seen by the MALDI-TOF (a) the formation of the DOPA-coupled product is accompanied by a corresponding increase in molecular weight. The success of the reaction is furthermore confirmed in the 1H-NMR (b) through the disappearance of the –CH2N3 proton peaks at 3.24 ppm and the


50

O OOO

O

NH

ON3

OO O

O

O

OO

OO

O

OO

O

HNO

N3

OOO

O

O

OO

OO

O

n

PEG-G1-(N3)2-(ene)4

CuAAC

O OOO

O

NHO

N

OO O

O

O

OO

OO

O

OO

O

HNO

N

OOO

O

O

OO

OO

O

n

PEG-G1-(N3)2-(ene)4NN

NN

O OHSO

OSH

PEG2k-(SH)2

TEC

=HN NH

S

HNOO O O

OHOH

HOHOHO

HO

HNO

O

OBiotin DOPA Man

Dendritic hydrogel

Figure 4.8. Functionalization of PEG-dendritic scaffold via CuAAC ‘click’ reaction and subsequent crosslinking into dendritic hydrogel using TEC.

Figure 4.9. CuAAC activation of PEG-G1-(N3)2-(ene)4 with DOPA-acetylene followed by a) MALDI-TOF and b) 1H-NMR. Dispersities (D) were obtained using Polytools v. 1.0.

b)

a)

P6k-G1-(N3)2-(ene)4

P6k-G1-(N3)2-(ene)4

P6k-(OH)2

P6k-G1-(DOPA)2-(ene)4

P6k-G1-(DOPA)2-(ene)4

(ppm)

No -N3left“enes”

preserved

P6k-[G1]

P6k


51

appearance of the characteristic triazole peak at 7.65 ppm. The spectra also confirms the preservation of the orthogonal enes which are still available for the subsequent crosslinking. Another very important aspect to consider, when using the CuAAC reaction to produce biomaterials, is the removal of catalytic copper to avoid any copper induced toxicity. This was done by adopting a simple purification protocol including passing the dendritic scaffolds through a neutral Al2O3 column followed by dialysis against a 0.5 % w/w EDTA solution with an adjusted pH of 7.0. Graphite furnace atomic absorption spectroscopy (GF-AAS) revealed a [Cu] reduction from about 35000 ppm to concentrations of around 3.5-11.5 ppm for the PEG(6 kDa) scaffolds. These values are well below toxic-dose-50 (TD50) values found for in vitro culturing of hepatocyte cells. 121

4.3.3 Tunable mechanical properties

As previously mentioned, the mechanical properties will have a huge impact on how cells interact and proliferate within artificial extracellular matrices (aECMs). We therefore wanted to explore the ability of using the heterofunctional dendritic scaffolds for making PEG hydrogels with tunable softness. Some potentially relevant tissue, in terms of Young’s moduli, include soft brain tissue (0.1-1 kPa),123 muscle tissue (8-17 kPa),124 and the osteoid collagen of bone (25-40 kPa).125 As can be seen in Figure 4.11 a) hydrogels with Young’s moduli in the desired ranges could successfully be produced. These were prepared by crosslinking of the azide-functional dendritic scaffolds presented in Scheme 4.4 with PEG-dithiols in EtOH solution and 0.5 % w/w of the photoinitiator I2959. For the smaller TMP-cored dendrimers, a 6 kDa PEG-dithiol was used and for the larger PEG-cored dendritic hybrids a shorter 2 kDa PEG-dithiol was used. Altering the amount of crosslinking groups and the polymer length between crosslinks showed a large impact on the mechanical and swelling properties of the hydrogels. For instance the PEG(10 kDa)-G1 hydrogel with four crosslinking enes is soft with a modulus of 3.5 kPa, while the PEG(6 kDa)-G2 hydrogel with eight crosslinking


52

Figure 4.11. Mechanical and swelling properties of dendritic hydrogels. a) Youngs moduli obtained by rheological measurements and b) visualization of swelling properties (after 72 h). The gels were cured in EtOH solution with 0.5 wt% of photoinitioator I2959 with UV (365 nm, 10 min, 9 mW cm-2). The PEG-dendritic hybrids were crosslinked with a PEG2k-(SH)2 and the TMP-scaffolds by a PEG6k-(SH)2. * Conditions adapted to those suitable for cell encapsulation; H20, [I2959] = 0.05 wt%.

enes is much more firm with a modulus of 27 kPa. This difference in crosslinking density is also reflected by the different swelling profiles of the two gels corresponding to 2000 % and 800 % (w/w), respectively (Figure 4.11 b). Interestingly, changing the curing conditions to those suitable for in situ encapsulation of live cells (365 nm, 9 mW cm-2, water solvent, [I2959] = 0.05 % w/w),58 led to a decrease in modulus from 16 kPa to 11 kPa for the PEG(6 kDa)-G1 hydrogel. This indicates that the curing efficiency is slightly decreased by the changed conditions. Such a decrease may be explained both by the lower amount of radicals produced (lower concentration of photoinitiator), as well as the limited solubility of I2959 in water compared to ethanol.59 Finally, the hydrogels crosslinked with TMP-cored dendrimers have moduli in the same range as the OSTE hydrogels H2 and H3 (20 and 40 kPa, Table 4.1). The H1 gel without off-stoichiometric ratio (1:1) is however relatively rigid in comparison with a moduli of 70 kPa.

a) b)Mold size: Ø 8 mm

H10k-G1-(N3)2 H6k-G1-(N3)2 H6k-G2-(N3)4


53

4.3.4 Cell viability and hydrolytic sensitivity of dendritic hydrogels

In order to evaluate the feasibility of using these dendritic scaffolds for in vitro cell culturing experiments initial cell viability tests were performed. Normal human dermal fibroblast (NHDF) cells were cultured on five types of dendritic hydrogels and their viability evaluated using a lactate dehydrogenase (LDH) assay (Figure 4.12). The hydrogels included were made from the PEG(6 kDa) G1-scaffolds with no (neutral), azide (N3), DOPA, and Man (G1 and G2) functionalities. The hydrogels were cured at the bottom of cell culturing wells with the PEG2k-(SH)2 crosslinker at 30 % w/w dry content using benign conditions (Milli-Q water, 365 nm, 9 mW cm-2,[I2959] = 0.05 % (w/w), 10 min). NDHF cells were seeded at 6k cells per well and the viability measured over seven days. The LDH levels were all in the same range of the control well with cell culture medium, indicating low cytotoxic response. The incubation with cells did however lead to full structural degradation of G1-gels within three days. The G2-Man gels were slightly more stable, but they too were fully degraded after seven days. This did not occur while incubating gels without cells under equivalent conditions, where the corresponding degradation took weeks.

Figure 4.12. Viability of normal human dermal fibroblasts incubated with five dendritic hydrogels measured by LDH assay. The detection limit was 0.17 µkat L-1.


54

4.4 FIBER REINFORCED ADHESIVE PATCHES

In this section the properties of the heterofunctional dendrimer scaffolds from the hydrogel section (Scheme 4.4 A) will be married with the rigid triazine based matrix from the OSTE section (Scheme 4.1) in the formation of fiber reinforced adhesive patches (FRAPs) for complex bone fracture stabilization.

4.4.1 Dendritic dual-purpose primers

The FRAP system (Figure 4.13), based on a topologically applied patch with fibers embedded in a triazine thiol-ene matrix, has shown to be mechanically strong with promising bone stabilizing features.111 However, the hydrophobic character of the matrix results in poor adhesion to the wet bone surface necessitating the use of a primer layer that will increase this interaction. In that context, the heterofunctional dendrimers were foreseen as promising scaffolds with available enes able to crosslink with the matrix, as well as azides for introduction of groups to increase the interaction with bone. The two G1 scaffolds were successfully modified to include both 3,4-dihydroxyphenylalanine (DOPA) and carboxyl groups (COOH) by facile CuAAC modifications which were followed by NMR and MALDI-TOF (Table 4.2, Figure 4.13)

Table 4.2. Dual-functional dendritic primers.

Primerm/z calcd. [M + Na+]

m/z obtained [M + Na+]

TMP-G1-(ene)3-(An)3 1303.64 1303.7

TMP-G1-(ene)3-(N3)6 1846.89 1847.03

TMP-G1-(N3)3-(ene)6 1675.80 1675.59

TMP-G1-(ene)3-(DOPA)6 3593.56 3594.67

TMP-G1-(DOPA)3-(ene)6 2549.13 2550.67

TMP-G1-(ene)3-(COOH)6 2435.12 2435.47

TMP-G1-(COOH)3-(ene)6 1969.91 1970.14


55

Figure 4.13. The buildup of fiber reinforced adhesive patches with fibers embedded in a triazine matrix and dual-purpose dendritic primers.


56

4.4.2 Shear strengths of FRAPs with dendritic primers

Primers were applied as 5 % w/w solutions in EtOH or EtOH:H2O (1:1) together with the triazine matrix and subsequently UV-cured on the bone surrounding the fracture (Figure 4.13). Additional layers of matrix and fibers were applied to form a stabilizing patch for the artificially fractured bovine femur bone. The relative strength of the patches with different primers was thereafter tested through tensile testing to obtain the tensile shear strengths of the FRAPs (Figure 4.14).

Figure 4.14. Shear strengths of FRAPs with dual-purpose dendritic primers. As can be seen, all primers lead to an increased shear strength compared to the system without any primer, which had a low value of 0.4 MPa. Contrary to what was envisioned, the two most polar primers with six DOPA- or COOH-groups did not give rise to any significant adhesion enhancement and actually noted the lowest shear strength values of all primers (0.6 and 0.8 MPa


57

respectively). This may be due to stronger cohesive forces between the dendritic primers in comparison to the adhesive forces towards the wet bone. Switching the functionalities resulted in the less polar TMP-G1-(DOPA)3-(ene)6 and TMP-G1-(COOH)3-(ene)6, which had increased shear strengths of 3.2 and 1.8 MPa. Interestingly, the nonpolar reference primer, TMP-G1-(N3)3-(An)3, with acetonide(An)-protected hydroxyl groups and three azides, noted a relatively high shear strength of 2.8 MPa. Converting the protected hydroxyls of this scaffold into enes (TMP-G1-(N3)3-(ene)6) only led to a marginal increase in shear strength (3.0 MPa), indicating a minor influence of the ene groups. Instead the adhesion seemed to be strongly influenced by the presence of azide groups in the primers. This is furthermore corroborated by the excellent shear strength of 4.2 MPa obtained for the TMP-G1-(ene)3-(N3)6 with a total number of six azides. This value is over 10-folds higher than for the FRAP with no primer and exceeds the adhesion obtained from a commercial wound closure adhesive, Histoacryl®, which is known for its good adhesion to wet substrates.126 The explanation for the increased adhesion most likely lies in the combination of azides with the UV-curing process. Alkyl azides are known to form reactive imines upon irradiation127 that may participate in intermolecular crosslinking reactions with other imines.128 This could potentially lead to the formation of high molecular weight primers, which have been shown to increase the shear strengths of FRAPs.111 Furthermore, in the presence of water and ethanol imines will react to form aldehydes, hydrates and/or acetals. Aldehydes have been evaluated for their advantageous adhesion to wet bone through reactions with bone surface amines.129 The increased adhesion for primers with azides may thus be explained by one or a combination of these pathways.

CONCLUSIONS

58

5. CONCLUSIONS

In this thesis two main strategies were explored for the preparation of functional polymeric materials targeting biomedical applications. These relied on the use of off-stoichiometric thiol-ene (OSTE) networks and highly functional dendritic polymeric scaffolds.

For the OSTE networks, a facile modular approach based on UV-induced thiol-ene coupling (TEC) was adopted for the manufacturing of functional polymeric materials. By carefully choosing the monomer molecular structure and altering their off-stoichiometric ratios, materials with a range of different Young’s moduli and available functional groups were produced. Standard molding techniques and simple TEC surface modifications proved to be powerful tools for tailoring the surface characteristics, such as introducing micron-size patterns and/or altering the surface hydrophobicity. Moreover, the OSTE material platform showed great promise for bridging the gap between research prototyping and commercial production of microfluidic devices. Materials mimicking the thermoplastic properties of commercial PMMA and PDMS networks could be prepared by rapid curing, with simple one-step surface modifications, and by the use of an inexpensive table-top UV-lamp setup. Their versatile nature was demonstrated by the creation of important microfluidic features such as hydrophobic stops, pneumatic valves and dry-bonded chip-to-world connections.

Dendronized gold surfaces were successfully obtained through a self-assembled dendritic monolayer (SADM) approach using a library of sulfur-cored dendritic polyesters. SPR revealed rapid assembly of all scaffolds upon injection. The resulting binding density and availability of functional groups on the surface were significantly affected by factors such as; the nature of the core sulfur group (SH or SS), the dendron generation, the distance

CONCLUSIONS

59

between the core sulfur and dendritic wedge, and the dendron end groups. Adsorption of the G3-SS-OH16-sp scaffold in acetate buffer resulted in the highest surface functional density (5.5 OH groups·nm-2), which is in the same range as a well-packed layer of linear alkanethiols. Furthermore, comparison between mannosylated and hydroxylated SADM surfaces revealed a 2.5-fold improvement for mannose surfaces in the scavenging capacity for uropathogenic E. coli Fim+ bacteria.

Heterofunctional dendritic scaffolds (HFDs) were successfully synthesized and their potential use as crosslinkers for functional hydrogels and primers for bone adhesive patches was explored. The hydrogel precursor materials were efficiently decorated with biorelevant moieties via CuAAC reactions. GF-AAS measurements revealed that the adopted purification protocol successfully brought the concentration of copper catalyst down to non-toxic levels. UV-induced TEC crosslinking of the structurally different HFDs with PEG-dithiols yielded hydrogels with Young’s moduli in the same range as several body tissues. The LDH assay indicated a low cytotoxic response for the tested hydrogels. However, in vitro culturing led to full structural degradation of all tested hydrogels within seven days.

The heterofunctional dendrimer scaffolds evaluated as primers for fiber-reinforced adhesive patches (FRAPs) in bone fracture stabilization all resulted in increased shear strength when compared to FRAPs without any primer. The adhesion strength was strongly influenced by the presence of azide groups in the primers. This was demonstrated by the excellent shear strength of 4.2 MPa for the TMP-G1-(ene)3-(N3)6 primer with a total of six azides. This value is over 10-fold higher than for FRAPs without any primers and in the same range as the commercial wound closure adhesive Histoacryl®.

FUTURE WORK

60

6. FUTURE WORK

The OSTE material platform showed to be very promising for prototyping of microfluidic devices. The facile preparation and simple modifications of these materials combined with readily available commercial starting materials make them highly interesting from a commercial point of view. This technology is currently being further developed for commercial purposes by the company Mercene Lab AB.130 Concerning the dendritic polymer approaches, it would be very interesting to combine the strategies used for dendronized surfaces and dendritic hydrogels in a single evaluation platform. This way the cellular response of a specific dendritic polymer could be evaluated both in a simplified 2D environment as well as in a more complex 3D network. The trends observed for the dendritic architectural impact on the SADM mechanism suggest that thiol-cored dendrons with long spacers between the core and dendritic wedge (not tested) could potentially lead to even higher surface functionalization. The hydrolytic sensitivity found for the trizma-based heterofunctional dendritic hydrogels, during cell culturing experiments, suggests that a more stable monomer should be explored. The effects of increasing the dendrimer generation and incorporating extracellular matrix adhesive peptides to these hydrogels would also be highly interesting. Furthermore, the method for preparing these hydrogels would benefit from using a photoinitiator system better suited for the conditions used (365 nm, water solvent). Finally, for the heterofunctional dendritic primers for FRAPs, a more thorough investigation of the mechanisms behind the increased adhesion of azide-functional primers would be of high interest.

ACKNOWLEDGEMENTS

61

7. ACKNOWLEDGEMENTS

First and foremost I would like to thank my main supervisor Associate Professor Michael Malkoch for all your help and support during these years. I am impressed of your dedication and how you always find time to help and guide all of us who are working for you despite your many obligations. Your passion and scientific knowledge are an inspiration and I have truly learnt a lot from working with you. My co-supervisor Professor Anders Hult is greatly thanked for accepting me as a PhD student in the coating group, for sharing your knowledge about photochemistry and for always having entertaining stories in stock. Professor Mats Johansson is thanked for all the interesting discussions, both private and work-related, especially our recent discussions concerning the paints and coating industry. Professor Eva Malmström is thanked for her pinpoint-accuracy when giving scientific feedback and for always spreading positive energy despite your very hectic schedule. Associate Professor Anna Carlmark is thanked for her kindness and positive attitude and for being my temporary roommate between the parental leaves. Sonny Jönsson is thanked for sharing your knowledge on photochemistry and for all the nice personal discussions in the kitchen. All other Professors and staff at the Department of Fibre and Polymer technology are thanked for contributing to a very nice working atmosphere. The administrative and technical personnel are greatly thanked for making sure that the FPT machinery works, in particular Inga, Inger, Mia, Vera, Barbara, Karin, Kjell and Bosse. My sincere gratitude to all my co-authors and collaborators: Jarmo Ropponen, Jonathan Kelly, Mattias Berglin, Anders Selborn, Helen Fink, Anders Lundgren, Thomas Fabo, Carl Fredrik Carlborg, Tommy Haraldsson, Professor Wouter van der Wijngaart, Sandra Berg, Isabella Joelsson Rahmn, Axel Nordberg, Professor Hans von Holst. An extra special thank you goes to

ACKNOWLEDGEMENTS

62

Peter Löwenhielm who introduced me to the world of silicone adhesives and inspired me to pursuit a PhD in polymer science. I really enjoyed working with you and the choices you inspired me to make has truly changed my life for the better. For financial support, Vinnova Sambio and the Swedish research council are greatly acknowledged. All former and present members of “ytgruppen” are thanked for contributing to an absolutely amazing work place. Lina for having the courage to follow your heart and making a choice that also affected me in an extremely positive way. Thank you Robban for your kind and helpful nature and for all the fun trips, past and future. Pontus for playing beach-volleyball with a heart-shaped t-shirt and for being my future “neighbor” in Norway. Linn for being with me since the beginning and for being the other part of our tag team “Lim”. Thank you Markus for your awesome friendship and for being my partner in the Swedish “klassiker”. Susanne for giving a face to the expression “Freudian slip” and thereby putting a smile on all of our faces. Carl for laughing the loudest and for all the great food. Christian is thanked for living the good life with me in Athens and for being my roomie in Bagis. Both the before mentioned are also thanked for being part of the “identical” curly-haired triplet. Linda is thanked for being my constant office-roomie and for throwing a magical dissertation party. Maribel for helping me in the lab when I was new and for partying in Berlin. Oliver is thanked for being such a nice guy and for taking over the SEC-responsibility and my lab project. Jan for making me smile and for taking such good care of all of my belongings in the lab =) Kristina for being a really thorough proofreader and an equally un-thorough holder of things (such as food or drinks). Sam is thanked for being a stand-up guy and helping me find my dream-job. Marie for being an inspiration in the lab and always being so friendly. Martin for our bike discussions and for shaking it out in the London night scene. Emelie is thanked for clarifying the difference between Hofors’ many pizzerias. Susana for her injection of positive energy to the group and for taking care of our apartment when we leave.

ACKNOWLEDGEMENTS

63

Mauro and Alireza for taking care of the FTIR instrument. Hui for being my lab-neighbor and testing the Tg of NMR-caps. Ting for all her hilarious comments. Assya for being super-nice. Viktor for replacing me as the calm guy of the group. Ville for all interesting discussions and for your superb impersonation of a gym-owner. Emma L. for exploring the “hand-grenades” in New Orleans. Carmen for being an early bird in the office. Danne and Sandra for shouldering my AB2C project. Axel for the midnight hydrogel-sessions and playing “rundpingis”. Hanna L for intense “GRIS”-kicking sessions. I would like to also thank all the students I have worked with; Bella, Marco, Mattia, Jocke, Tobbe, Jesper, Partiban, Kambiz, Jessica, Andrea, Patrick. I have learned a lot from working with you all. The PF-crew; Jonas for all your interesting anecdotes and for the fun climbing sessions, Mason for spreading positive energy and traveling to more countries in a year than I have in my life. I would also like to thank Stacy, Emma Ö, Eric, Sara O, Sara K, Petra, Camilla, Pelle, Niklas and everyone else that I forgot to mention. Till alla mina föräldrar och många syskon, er kärlek och ert stöd har betytt otroligt mycket genom åren och det har burit mig dit jag är idag. Sist men lång ifrån minst; Yvonne, du är utan tvekan det bästa som har hänt mig under den här tiden och i hela mitt liv. Att du, lillskrot och jag nu har blivit en familj känns helt fantastiskt och jag längtar med glädje efter att få spendera resten av mitt liv med er!

REFERENCES

64

8. REFERENCES

1. G. Patterson, A prehistory of polymer science, 2012. 2. D. A. Tomalia and J. M. J. Frechet, Journal of Polymer

Science Part a-Polymer Chemistry, 2002, 40, 2719-2728. 3. P. J. Flory, J. Am. Chem. Soc., 1941, 63, 3083-3090. 4. P. J. Flory, J. Am. Chem. Soc., 1941, 63, 3091-3096. 5. E. Buhleier, W. Wehner and F. Vogtle, Synthesis-Stuttgart,

1978, 155-158. 6. D. A. Tomalia, H. Baker, J. Dewald, M. Hall, G. Kallos, S.

Martin, J. Roeck, J. Ryder and P. Smith, Polymer Journal, 1985, 17, 117-132.

7. G. R. Newkome, Z. Q. Yao, G. R. Baker and V. K. Gupta, Journal of Organic Chemistry, 1985, 50, 2003-2004.

8. C. J. Hawker and J. M. J. Frechet, Journal of the American Chemical Society, 1990, 112, 7638-7647.

9. C. C. Lee, J. A. MacKay, J. M. J. Frechet and F. C. Szoka, Nature Biotechnology, 2005, 23, 1517-1526.

10. D. Astruc, E. Boisselier and C. Ornelas, Chemical Reviews, 2010, 110, 1857-1959.

11. Dendritech Inc, Midland, USA www.dendritech.com. 12. Dendritic Nanotechnologies, Inc., www.dnanotech.com. 13. Polymer Factory Sweden AB, Stockholm, Sweden,

www.polymerfactory.com. 14. K. Sadler and J. P. Tam, Journal of biotechnology, 2002, 90,

195-229. 15. B. Klajnert, M. Cortijo-Arellano, J. Cladera, J.-P. Majoral,

A.-M. Caminade and M. Bryszewska, Biochemical and Biophysical Research Communications, 2007, 364, 20-25.

16. W. B. Turnbull and J. F. Stoddart, Journal of biotechnology, 2002, 90, 231-255.

17. M. V. Walter and M. Malkoch, Chemical Society Reviews, 2012, 41, 4593-4609.

http://www.dendritech.com/

http://www.dnanotech.com/

http://www.polymerfactory.com/

REFERENCES

65

18. P. Wu, M. Malkoch, J. N. Hunt, R. Vestberg, E. Kaltgrad, M. G. Finn, V. V. Fokin, K. B. Sharpless and C. J. Hawker, Chem. Commun., 2005, 5775-5777.

19. E. R. Gillies and J. M. J. Frechet, Journal of the American Chemical Society, 2002, 124, 14137-14146.

20. C. C. Lee, E. R. Gillies, M. E. Fox, S. J. Guillaudeu, J. M. J. Frechet, E. E. Dy and F. C. Szoka, Proceedings of the National Academy of Sciences of the United States of America, 2006, 103, 16649-16654.

21. M. Slany, A.-M. Caminade and J. P. Majoral, Tetrahedron Lett., 1996, 37, 9053-9056.

22. A. P. Goodwin, S. S. Lam and J. M. J. Frechet, Journal of the American Chemical Society, 2007, 129, 6994-+.

23. L. Lochmann, K. L. Wooley, P. T. Ivanova and J. M. J. Frechet, Journal of the American Chemical Society, 1993, 115, 7043-7044.

24. R. J. Amir, L. Albertazzi, J. Willis, A. Khan, T. Kang and C. J. Hawker, Angewandte Chemie-International Edition, 2011, 50, 3425-3429.

25. P. Antoni, Y. Hed, A. Nordberg, D. Nystrom, H. von Holst, A. Hult and M. Malkoch, Angewandte Chemie-International Edition, 2009, 48, 2126-2130.

26. E. R. Gillies, T. B. Jonsson and J. M. J. Frechet, Journal of the American Chemical Society, 2004, 126, 11936-11943.

27. L. Albertazzi, F. M. Mickler, G. M. Pavan, F. Salomone, G. Bardi, M. Panniello, E. Amir, T. Kang, K. L. Killops, C. Braeuchle, R. J. Amir and C. J. Hawker, Biomacromolecules, 2012, 13, 4089-4097.

28. M. V. Walter, P. Lundberg, D. Hult, A. Hult and M. Malkoch, Polymer Chemistry, 2013, 4, 2680-2690.

29. L. Degoricija, P. N. Bansal, S. H. M. Sontjens, N. S. Joshi, M. Takahashi, B. Snyder and M. W. Grinstaff, Biomacromolecules, 2008, 9, 2863-2872.

30. J. S. Moore and S. I. Stupp, Macromolecules, 1990, 23, 65-70.

REFERENCES

66

31. M. Malkoch, E. Malmström and A. Hult, Macromolecules, 2002, 35, 8307-8314.

32. A. Carlmark, E. Malmstrom and M. Malkoch, Chemical Society Reviews, 2013, 42, 5858-5879.

33. H. C. Kolb, M. G. Finn and K. B. Sharpless, Angewandte Chemie-International Edition, 2001, 40, 2004-+.

34. H. C. Kolb and K. B. Sharpless, Drug Discovery Today, 2003, 8, 1128-1137.

35. J. E. Moses and A. D. Moorhouse, Chemical Society Reviews, 2007, 36, 1249-1262.

36. W. H. Binder and R. Sachsenhofer, Macromolecular Rapid Communications, 2007, 28, 15-54.

37. G. Franc and A. K. Kakkar, Chemical Society Reviews, 2010, 39, 1536-1544.

38. C. Barner-Kowollik, F. E. Du Prez, P. Espeel, C. J. Hawker, T. Junkers, H. Schlaad and W. Van Camp, Angewandte Chemie-International Edition, 2011, 50, 60-62.

39. A. Michael, Journal für Praktische Chemie, 1893, 48, 94-95. 40. R. Huisgen, Angew. Chem., 1963, 75, 742-754. 41. C. W. Tornoe, C. Christensen and M. Meldal, Journal of

Organic Chemistry, 2002, 67, 3057-3064. 42. V. V. Rostovtsev, L. G. Green, V. V. Fokin and K. B.

Sharpless, Angewandte Chemie-International Edition, 2002, 41, 2596-+.

43. J. E. Hein and V. V. Fokin, Chemical Society Reviews, 2010, 39, 1302-1315.

44. V. O. Rodionov, S. I. Presolski, D. D. Diaz, V. V. Fokin and M. G. Finn, Journal of the American Chemical Society, 2007, 129, 12705-12712.

45. J. C. Jewett and C. R. Bertozzi, Chemical Society Reviews, 2010, 39, 1272-1279.

46. C. A. DeForest and K. S. Anseth, Nature Chemistry, 2011, 3, 925-931.

47. C. R. Becer, R. Hoogenboom and U. S. Schubert, Angewandte Chemie-International Edition, 2009, 48, 4900-4908.

REFERENCES

67

48. A. B. Lowe, Polymer Chemistry, 2010, 1, 17-36. 49. C. E. Hoyle and C. N. Bowman, Angewandte Chemie-

International Edition, 2010, 49, 1540-1573. 50. T. Posner, Ber. Dtsch. chem. Ges., 1905, 38, 646-657. 51. B. J. v and R. Murjahn, Ber. Dtsch. Chem. Ges. B, 1926, 59B,

1202-1209. 52. C. E. Hoyle, T. Y. Lee and T. Roper, Journal of Polymer

Science Part a-Polymer Chemistry, 2004, 42, 5301-5338. 53. K. L. Killops, L. M. Campos and C. J. Hawker, Journal of

the American Chemical Society, 2008, 130, 5062-+. 54. M. J. Kade, D. J. Burke and C. J. Hawker, Journal of

Polymer Science Part a-Polymer Chemistry, 2010, 48, 743-750.

55. A. Dondoni and A. Marra, Chemical Society Reviews, 2012, 41, 573-586.

56. B. S. Sumerlin and A. P. Vogt, Macromolecules, 2010, 43, 1-13.

57. M. W. Tibbitt and K. S. Anseth, Biotechnology and Bioengineering, 2009, 103, 655-663.

58. S. J. Bryant, C. R. Nuttelman and K. S. Anseth, Journal of Biomaterials Science-Polymer Edition, 2000, 11, 439-457.

59. B. D. Fairbanks, M. P. Schwartz, C. N. Bowman and K. S. Anseth, Biomaterials, 2009, 30, 6702-6707.

60. E. T. Lagally, I. Medintz and R. A. Mathies, Analytical Chemistry, 2001, 73, 565-570.

61. D. J. Harrison, K. Fluri, K. Seiler, Z. H. Fan, C. S. Effenhauser and A. Manz, Science, 1993, 261, 895-897.

62. E. T. Lagally, J. R. Scherer, R. G. Blazej, N. M. Toriello, B. A. Diep, M. Ramchandani, G. F. Sensabaugh, L. W. Riley and R. A. Mathies, Analytical Chemistry, 2004, 76, 3162-3170.

63. H. Becker, Lab on a Chip, 2010, 10, 271-273. 64. D. Mark, S. Haeberle, G. Roth, F. von Stetten and R.

Zengerle, Chemical Society Reviews, 2010, 39, 1153-1182. 65. S. K. Sia and G. M. Whitesides, Electrophoresis, 2003, 24,

3563-3576.

REFERENCES

68

66. J. M. K. Ng, I. Gitlin, A. D. Stroock and G. M. Whitesides, Electrophoresis, 2002, 23, 3461-3473.

67. H. Hillborg and U. W. Gedde, Ieee Transactions on Dielectrics and Electrical Insulation, 1999, 6, 703-717.

68. D. Bartolo, G. Degre, P. Nghe and V. Studer, Lab on a Chip, 2008, 8, 274-279.

69. Fluidigm, San Fransisco, USA, www.fluidigm.com. 70. Biosite Inc., San Diego, USA, www.biosite.com. 71. J. P. Rolland, R. M. Van Dam, D. A. Schorzman, S. R.

Quake and J. M. DeSimone, Journal of the American Chemical Society, 2004, 126, 2322-2323.

72. M. Natali, S. Begolo, T. Carofiglio and G. Mistura, Lab on a Chip, 2008, 8, 492-494.

73. M. Mammen, S. K. Choi and G. M. Whitesides, Angewandte Chemie-International Edition, 1998, 37, 2755-2794.

74. L. Roglin, E. H. M. Lempens and E. W. Meijer, Angewandte Chemie-International Edition, 2011, 50, 102-112.

75. S. G. Gouin, A. Wellens, J. Bouckaert and J. Kovensky, Chemmedchem, 2009, 4, 749-755.

76. M. Touaibia, A. Wellens, T. C. Shiao, Q. Wang, S. Sirois, J. Bouckaert and R. Roy, Chemmedchem, 2007, 2, 1190-1201.

77. M. Touaibia and R. Roy, Mini-Reviews in Medicinal Chemistry, 2007, 7, 1270-1283.

78. H. J. Han, R. M. Kannan, S. Wang, G. Mao, J. P. Kusanovic and R. Romero, Advanced Functional Materials, 2010, 20, 409-421.

79. A. Lundgren, Y. Hed, K. Oeberg, A. Sellborn, H. Fink, P. Loewenhielm, J. Kelly, M. Malkoch and M. Berglin, Angewandte Chemie-International Edition, 2011, 50, 3450-3453.

80. S. R. Benhabbour, H. Sheardown and A. Adronov, Biomaterials, 2008, 29, 4177-4186.

http://www.fluidigm.com/

http://www.biosite.com/

REFERENCES

69

81. A. Friggeri, H. Schonherr, H. J. van Manen, B. H. Huisman, G. J. Vancso, J. Huskens, F. van Veggel and D. N. Reinhoudt, Langmuir, 2000, 16, 7757-7763.

82. N. Satoh and K. Yamamoto, Organic Letters, 2009, 11, 1729-1732.

83. L. Zhang, B. Zou, D. Dong, F. W. Huo, X. Zhang, L. F. Chi and L. Jiang, Chem. Commun., 2001, 1906-1907.

84. L. Zhang, F. W. Huo, Z. Q. Wang, L. X. Wu, X. Zhang, S. Hoppener, L. F. Chi, H. Fuchs, J. W. Zhao, L. Niu and S. J. Dong, Langmuir, 2000, 16, 3813-3817.

85. Z. S. Bo, L. Zhang, B. Zhao, X. Zhang, J. C. Shen, S. Hoppener, L. F. Chi and H. Fuchs, Chemistry Letters, 1998, 1197-1198.

86. C. B. Gorman, R. L. Miller, K. Y. Chen, A. R. Bishop, R. T. Haasch and R. G. Nuzzo, Langmuir, 1998, 14, 3312-3319.

87. V. Chechik, H. Schonherr, G. J. Vancso and C. J. M. Stirling, Langmuir, 1998, 14, 3003-3010.

88. G. Jiang, S. Deng, A. Baba, C. Huang and R. C. Advincula, Macromolecular Chemistry and Physics, 2010, 211, 2562-2572.

89. N. Feliu, M. V. Walter, M. I. Montanez, A. Kunzmann, A. Hult, A. Nystrom, M. Malkoch and B. Fadeel, Biomaterials, 2012, 33, 1970-1981.

90. E. Ostmark, L. Macakova, T. Auletta, M. Malkoch, E. Malmstrom and E. Blomberg, Langmuir, 2005, 21, 4512-4519.

91. R. Vestberg, M. Malkoch, M. Kade, P. Wu, V. V. Fokin, K. B. Sharpless, E. Drockenmuller and C. J. Hawker, Journal of Polymer Science Part a-Polymer Chemistry, 2007, 45, 2835-2846.

92. M. I. Montanez, Y. Hed, S. Utsel, J. Ropponen, E. Malmstrom, L. Wagberg, A. Hult and M. Malkoch, Biomacromolecules, 2011, 12, 2114-2125.

93. J. L. Drury and D. J. Mooney, Biomaterials, 2003, 24, 4337-4351.

REFERENCES

70

94. B. V. Slaughter, S. S. Khurshid, O. Z. Fisher, A. Khademhosseini and N. A. Peppas, Advanced Materials, 2009, 21, 3307-3329.

95. X. Wei, C. Gong, M. Gou, S. Fu, Q. Guo, S. Shi, F. Luo, G. Guo, L. Qiu and Z. Qian, International Journal of Pharmaceutics, 2009, 381, 1-18.

96. J. Zhu, Biomaterials, 2010, 31, 4639-4656. 97. P. K. Working, M. S. Newman, J. Johnson and J. B.

Cornacoff, in Poly(Ethylene Glycol): Chemistry and Biological Applications, eds. J. M. Harris and S. Zalipsky, Amer Chemical Soc, Washington, 1997, vol. 680, pp. 45-57.

98. R. Gref, M. Luck, P. Quellec, M. Marchand, E. Dellacherie, S. Harnisch, T. Blunk and R. H. Muller, Colloids and Surfaces B-Biointerfaces, 2000, 18, 301-313.

99. A. M. Kloxin, C. J. Kloxin, C. N. Bowman and K. S. Anseth, Advanced Materials, 2010, 22, 3484-3494.

100. T. Yang, M. Malkoch and A. Hult, Journal of Polymer Science Part a-Polymer Chemistry, 2013, 51, 363-371.

101. M. A. Azagarsamy and K. S. Anseth, Acs Macro Letters, 2013, 2, 5-9.

102. M. W. Grinstaff, Journal of Polymer Science Part a-Polymer Chemistry, 2008, 46, 383-400.

103. N. Joshi and M. Grinstaff, Current Topics in Medicinal Chemistry, 2008, 8, 1225-1236.

104. M. Malkoch, R. Vestberg, N. Gupta, L. Mespouille, P. Dubois, A. F. Mason, J. L. Hedrick, Q. Liao, C. W. Frank, K. Kingsbury and C. J. Hawker, Chem. Commun., 2006, 2774-2776.

105. A. K. Gosain and D. C. Plastic Surg Ed Fdn, Plastic and Reconstructive Surgery, 2002, 109, 2581-2583.

106. M. Donkerwolcke, F. Burny and D. Muster, Biomaterials, 1998, 19, 1461-1466.

107. S. E. Olsson and K. A. Rietz, Acta Chir Scand Suppl, 1966, 367, 4-19.

REFERENCES

71

108. S. C. Weber and M. W. Chapman, Clinical Orthopaedics and Related Research, 1984, 249-261.

109. A. Nordberg, H. von Holst, K. Brolin and A. Beckman, Bio-Medical Materials and Engineering, 2007, 17, 299-308.

110. S. C. Woodward, Annals of the New York Academy of Sciences, 1968, 146, 225-250.

111. A. Nordberg, P. Antoni, M. I. Montanez, A. Hult, H. Von Holst and M. Malkoch, Acs Applied Materials & Interfaces, 2010, 2, 654-657.

112. L. Feng, H. Fang, S. Zhou and L. Wu, Macromolecular Chemistry and Physics, 2006, 207, 1575-1583.

113. M. Malkoch, E. Malmstrom and A. Hult, Macromolecules, 2002, 35, 8307-8314.

114. W. D. Wilson, Science, 2002, 295, 2103-+. 115. K. A. Krogfelt, H. Bergmans and P. Klemm, Infection and

Immunity, 1990, 58, 1995-1998. 116. N. B. Cramer, J. P. Scott and C. N. Bowman,

Macromolecules, 2002, 35, 5361-5365. 117. I. P. Parkin and R. G. Palgrave, Journal of Materials

Chemistry, 2005, 15, 1689-1695. 118. M. A. Unger, H. P. Chou, T. Thorsen, A. Scherer and S. R.

Quake, Science, 2000, 288, 113-116. 119. H. Ihre, O. L. P. De Jesus and J. M. J. Frechet, Journal of

the American Chemical Society, 2001, 123, 5908-5917. 120. H. A. Biebuyck, C. D. Bian and G. M. Whitesides,

Langmuir, 1994, 10, 1825-1831. 121. S. Flink, F. van Veggel and D. N. Reinhoudt, Advanced

Materials, 2000, 12, 1315-1328. 122. S. L. Chen, C. S. Hung, J. S. Pinkner, J. N. Walker, C. K.

Cusumano, Z. Li, J. Bouckaert, J. I. Gordon and S. J. Hultgren, Proceedings of the National Academy of Sciences of the United States of America, 2009, 106, 22439-22444.

123. L. A. Flanagan, Y. E. Ju, B. Marg, M. Osterfield and P. A. Janmey, Neuroreport, 2002, 13, 2411-2415.

REFERENCES

72

124. A. J. Engler, M. A. Griffin, S. Sen, C. G. Bonnetnann, H. L. Sweeney and D. E. Discher, Journal of Cell Biology, 2004, 166, 877-887.

125. A. J. Engler, S. Sen, H. L. Sweeney and D. E. Discher, Cell, 2006, 126, 677-689.

126. www.tissueseal.com, Histoacryl ® topical skin adhesive, www.TissueSeal.com, Accessed 2013-11-19.

127. N. Gritsan and M. Platz, in Organic Azides, John Wiley & Sons, Ltd, 2010, pp. 311-372.

128. P. Ling and C. A. Wight, Journal of Physical Chemistry B, 1997, 101, 2126-2131.

129. B. Hoffmann, E. Volkmer, A. Kokott, P. Augat, M. Ohnmacht, N. Sedlmayr, M. Schieker, L. Claes, W. Mutschler and G. Ziegler, Journal of Materials Science-Materials in Medicine, 2009, 20, 2001-2009.

130. Mercene Labs AB, www.mercenelabs.com.

http://www.tissueseal.com/

http://www.tissueseal.com/

http://www.mercenelabs.com/

Advanced polymeric scaffolds for functional materials in...

Documents

Transcript of Advanced polymeric scaffolds for functional materials in...