MASTER
Dynamic samples for STORM-Imaging
Pol, D.G.J.
Award date:
2016
Link to publication
Disclaimer
This document contains a student thesis (bachelor's or master's), as authored by a student at Eindhoven University of Technology. Student theses are made available in the TU/e repository upon obtaining the required degree. The grade received is not published on the document as presented in the repository. The required complexity or quality of research of student theses may vary by program, and the required minimum study period may vary in duration.
General rights
Copyright and moral rights for the publications made accessible in the public portal are retained by the authors and/or other copyright owners and it is a condition of accessing publications that users recognise and abide by the legal requirements associated with these rights.
Dynamic Samples for STORM-Imaging
Master’s Thesis
By
DANNYGERARD JAN POL
Department of Chemical Engineering and Chemistry EINDHOVENUNIVERSITY OFTECHNOLOGY
A dissertation submitted to the University of Eindhoven in accordance with the requirements of the degree of MASTER OF
SCIENCEin the Faculty of Chemistry under the supervision of D.VAN DERZWAAG, MSCandPROF.DR. E.W. MEIJER.
OCTOBER 2015
A
BSTRACTD
ynamic self-organizing and self-assembling systems are found all throughout nature.The kinetics of various (dis)assembly processes of these structures in biological systems are an interesting subject of study, because an improved understanding of these pro- cesses can help unravel the operational principles of living systems. The properties of dynamic self-assembling systems from nature can be mimicked by designing synthetic structures. One synthetic analogue is the supramolecular BTA-polymer.
In this project, the dynamic behaviour of supramolecular BTA-polymers was studied using super-resolution fluorescence microscopy. We are interested in imaging real-time exchange processes of BTA-fibers. For that, we need to prevent physisorption and induce selective binding of these dynamic BTA-fibers. We aim to do this by modifying glass microscope coverslip surfaces.
In chapter 2, we describe a method to synthesize BTA-monomers coupled with fluorescent dye molecules, which are necessary for imaging using fluorescence microscopy. This is a two-step synthesis starting from precursor BTA-triazide. The obtained BTA-triamine is coupled with a cyanine Cy3 or Cy5 dye in a statistical reaction.
In chapter 3, we describe how to change the properties of glass coverslips to obtain anti- fouling characteristics. This is done by chemical modification of the surface. We conclude that physisorption of BTA-fibers on the surface can be prevented by surface functionalization using silane chemistry approaches. Surfaces coated with a layer of polyethylene glycol (PEG) molecules show useful anti-fouling properties towards BTA-fibers, because no physisorption of the polymers is observed.
In chapter 4, we describe a procedure to obtain a reactivated surface that selectively binds to BTA-fibers from solution, maintaining their dynamic properties. This is done by incorporating linker molecules within an anti-fouling surface. These linker molecules can bind to BTA-fibers, so that their relative position to the surface is retained. Imaging the surfaces obtained using this method, revealed a reactivated surface where dynamic fiber-like aggregates were visible.
The research described in this report shows that it is possible to modify the properties of glass microscope coverslips used for imaging. Both the formation of anti-fouling surfaces and subsequent reactivation of those surfaces are successful. Next, we show that visualization of the dynamic behaviour of supramolecular structures or aggregates is possible. Real-time imaging of such aggregates in their native state had not been achieved before, because these structures had to be immobilized for imaging. With the method described in this thesis, a promising route towards imaging real-time exchange within supramolecular polymers is demonstrated.
D
EDICATIONI
n dedication to my dog Dio, who has been my best friend for over sixteen years. Sadly enough, I had to let him go during working on this thesis in the past year. Fortunately, my memories of him will be imprinted in my heart for the rest of my life.Dio
06/02/1999 – 20/05/2015
A
CKNOWLEDGEMENTSO
ne of the most intense years of my life. That is how I would describe the past year during my work on this project. Not only was it the year with the most valuable educational experience of my complete education, it also a year with personal difficulties. Nevertheless, I had a great time during my work and I can proudly say that both on a personal level as on subject-matter, I have enriched myself greatly by working on this project. I found that I like doing research as such, but also noticed what it takes to properly progress in this field. Research of this stand is a form of top-level sport.I would like to thank professor Bert Meijer for the opportunity to work on my graduation project in such a well-regarded department on a subject this interesting. Well thanks for the inspiring conversations and for supporting my decisions during the project.
Secondly, I am thankful to my direct supervisor Daan van der Zwaag. It has been an honour to work with someone that performs on such a high level in science and research. Thanks for the support on all levels, warm communication and for the immense amount of things I learned from you during the past year.
Next, I would like to thank both Lorenzo Albertazzi and Matthew Baker for their help. Both have played a significant role during different parts of my project. Lorenzo is a master in asking stimulating questions necessary to find the answers myself. This taught me a lot in changing the way of thinking and interpretation, which is often needed in research. The practical skills Matt showed me, opened my eyes and was the most instructive one in a practical sense I ever had. Also, lot of thanks for the personal help in difficult moments when, for example, synthesis behaved as it did, again.
Finally, I thank my family for their support during the past year. Things did not always look as promising and I wasn’t always the most cheerful one at home, but I’m proud to show you what I have become.
T
ABLE OFC
ONTENTSPage
List of Figures xi
1 Introduction 1
1.1 Self-organization . . . 1
1.1.1 History . . . 2
1.1.2 Emergence, self-organization and self-assembly . . . 2
1.1.3 Self-assembly in nature . . . 3
1.2 Supramolecular systems in chemistry . . . 3
1.2.1 The water-soluble BTA-molecule . . . 4
1.2.2 Applications of water-soluble BTA nanofibers . . . 6
1.3 Visualization techniques . . . 7
1.4 Super-resolution microscopy . . . 9
1.5 Surface chemistry . . . 11
1.5.1 Basic concepts of silanization . . . 11
1.5.2 Surface structure . . . 13
1.6 Aim and outline of this thesis . . . 15
2 BTA-synthesis and dye labeling 17 2.1 Reduction of azide . . . 17
2.2 Coupling with fluorescent dye . . . 19
2.3 Separation of desired product from statistical mixture . . . 19
2.4 Assembling BTA-fibers with fluorescently active monomers . . . 21
2.4.1 Calculating exact yield using calibration curve . . . 21
2.4.2 Stock solutions and stack-assembly . . . 22
2.5 Conclusion . . . 22
2.6 Materials and methods . . . 24
2.6.1 Materials . . . 24
2.6.2 Instrumentation . . . 24
2.6.3 Reduction of BTA-triazide to BTA-triamine . . . 24
2.6.4 Coupling of BTA-triamine with Cy3/Cy5 dyes . . . 24
2.6.5 Stack-assembly . . . 25
3 Anti-fouling surface 27 3.1 Current method . . . 27
3.2 Evaluating anti-fouling glass coverslips . . . 28
3.2.1 Slide manufacturing . . . 28
3.2.2 Anti-fouling assessment . . . 29
3.3 Functionalization with fluorinated layer . . . 31
3.3.1 Surface modification . . . 32
3.3.2 Anti-fouling evaluation . . . 32
3.4 Functionalization with pegylated layer . . . 33
3.4.1 Functionalized surfaces using shorter PEG-linkers . . . 34
3.4.2 Functionalized surfaces using longer PEG-linkers . . . 35
3.4.3 Solvent effects . . . 36
3.5 Conclusion . . . 38
3.6 Materials and methods . . . 41
3.6.1 Materials . . . 41
3.6.2 Instrumentation . . . 41
3.6.3 Cleaning of glass coverslips . . . 41
3.6.4 Surface functionalization by vapor phase fluorination . . . 41
3.6.5 Surface functionalization by solution phase fluorination . . . 41
3.6.6 Surface functionalization by solution phase pegylation . . . 41
3.6.7 STORM-imaging of BTA-fibers on functionalized surfaces . . . 42
4 Linker molecules as anchoring points 43 4.1 Reactivation of surfaces . . . 43
4.2 Simultaneous functionalization . . . 44
4.3 Sequential functionalization . . . 45
4.4 Imaging and evaluation of surfaces . . . 45
4.5 Conclusion . . . 49
4.6 Materials and methods . . . 50
4.6.1 Materials . . . 50
4.6.2 Instrumentation . . . 50
4.6.3 Cleaning of glass coverslips . . . 50
4.6.4 Sequential surface functionalization for reactivated surface . . . 50
4.6.5 Sample preparation for imaging of BTA-fibers on reactivated surface . . . 50
5 Conclusion, discussion and outlook 51 5.1 Conclusion . . . 51
TABLE OF CONTENTS
5.2 Discussion . . . 52 5.3 Outlook . . . 54
A Appendix A 55
B Appendix B 61
Bibliography 69
L
IST OFF
IGURESFIGURE Page
1.1 Different systems of emergence and self-organization. . . 3
1.2 Schematic model of the cell cytoskeleton, consisting of a network of self-assembled fibers. . . 4
1.3 Examples of supramolecular systems. . . 5
1.4 Molecular structure of a water-soluble BTA-molecule. . . 5
1.5 Schematic representation of the helical structure of a self-assembled BTA-fiber. . . . 6
1.6 Examples of multiple different application areas for BTAs. . . 6
1.7 Imaging BTA-fibers using cryo-TEM. . . 7
1.8 Imaging BTA-fibers using fluorescence microscopy. . . 8
1.9 Difference between the wavelength of absorbed and emitted light; the Stokes shift. . 9
1.10 Evanescence field created due to TIRF-angle. . . 9
1.11 Proposed model for the dark state by addition of primary thiol. . . 10
1.12 Difference in conventional imaging and super-resolution imaging of BTA-fibers. . . . 11
1.13 Monomer exchange after mixing different BTA-fibers. . . 12
1.14 STORM-imaging of fiber-exchange at different time points. . . 13
1.15 Structure of a silane. . . 13
1.16 Reaction scheme for silanization. . . 14
1.17 Schematic view of growth of a monolayer. . . 14
1.18 Schematic drawing of silanes bound to a silicon oxide surface. . . 15
1.19 Schematic model of BTA-fibers attached to the surface via linker molecules. . . 16
2.1 Different end-groups of a BTA-monomer. . . 18
2.2 Reaction scheme of reduction of BTA-triazide to BTA-triamine. . . 18
2.3 Reaction mechanism of Staudinger reduction. . . 19
2.4 Reaction scheme of coupling of the BTA-triamine with a cyanine dye. . . 20
2.5 Coupling reaction of BTA-triamine with Cy5-NHS in DMF. . . 20
2.6 Dilution series of Cy3-NHS and Cy5-NHS esters in DMSO. . . 21
2.7 Calibration curves for Cy3-NHS and Cy5-NHS esters in DMSO. . . 22
3.1 BTA-fibers physisorbed to the glass surface after cleaning using piranha-etch. . . 28
3.2 Distinction between aggregates and background levels of fluorescence on the surface. 29
3.3 Reference values for background levels of fluorescence. . . 30
3.4 Surface overview before functionalization for reference values. . . 31
3.5 Used fluorinated compounds for surface functionalization. . . 31
3.6 Difference in surface energy between glass coverslips. . . 32
3.7 Glass surfaces after functionalization with fluorinated silanes . . . 33
3.8 Background levels of fluorescence for fluorinated slides. . . 34
3.9 Used pegylated compounds for surface functionalization. . . 34
3.10 Glass surfaces after functionalization with short pegylated silanes . . . 35
3.11 Background levels of fluorescence for pegylated slides using compound (3). . . 36
3.12 Glass surfaces after functionalization with longer pegylated silanes. . . 36
3.13 Background levels of fluorescence for pegylated slides using compound (4). . . 37
3.14 Solvent effects for functionalization with pegylated silanes . . . 37
3.15 Background levels of fluorescence using different solvents. . . 38
3.16 Background levels of fluorescence using vapor phase silanization. . . 39
3.17 Background levels of fluorescence of best slide compared to reference values. . . 39
3.18 Glass surface after functionalization with compound (4), 1% solution in THF. . . 40
4.1 Used linker molecule for reactivation, compound (5). . . 43
4.2 Schematic model of BTA-fibers attached to the surface via linker molecules. . . 44
4.3 Overview of dynamic BTA-fibers linked to a reactivated glass coverslip. . . 46
4.4 Close-up of a dynamic BTA-fiber linked to a reactivated glass coverslip. . . 46
4.5 Dynamics of two BTA-fibers linked to a reactivated glass coverslip. . . 47
4.6 Dynamic behaviour of BTA-fibers linked to a reactivated glass coverslip. . . 47
4.7 Three BTA-fibers linked to a reactivated glass coverslip. . . 48
4.8 Dynamic BTA-fiber linked to a reactivated glass coverslip. . . 48
A.1 IR-spectrum shows complete reduction, as seen by the absence of the azide-peak. . . . 56
A.2 1H-NMR analysis shows BTA-triamine as desired product without impurities. . . 57
A.3 Close-up of previous1H-NMR spectrum. . . 58
A.4 LCMS-spectrum after coupling of BTA-triamine to Cy3-dye. . . 59
A.5 LCMS-spectrum after coupling of BTA-triamine to Cy5-dye. . . 60
B.1 Overview of dynamic BTA-fibers linked to a reactivated glass coverslip. . . 62
B.2 Close-up of a dynamic BTA-fiber linked to a reactivated glass coverslip. . . 63
B.3 Dynamics of two BTA-fibers linked to a reactivated glass coverslip. . . 64
B.4 Dynamic behaviour of BTA-fibers linked to a reactivated glass coverslip. . . 65
B.5 Three BTA-fibers linked to a reactivated glass coverslip. . . 66
B.6 Dynamic BTA-fiber linked to a reactivated glass coverslip. . . 67
C
HAPTER1
I
NTRODUCTIONE
very researcher working in the field of supramolecular chemistry is likely to find in- spiration in nature and in mimicking its dynamic and adaptive behaviour in synthetic molecular systems. Is it possible to finally understand life if we are able to make living systems ourself? It is already possible to create synthetic analogues that resemble dynamic molecular structures from living cells. Such structures are under investigation and in this thesis I describe my research regarding one of these supramolecular polymers, the BTA-fiber.1.1 Self-organization
The second law of thermodynamics states that the total sum of entropy, a measure of disorder, is increased after every spontaneous process that occurs in nature. This means that every natural process tends to behave towards the most random and disordered state, equilibrium.
The introduction of extra components to an equilibrated state changes the situation so that the equilibration process towards disorder starts again.
When enough complex components capable of interaction are brought together, it is possible that a state of structure and order arises regionally. So, rather than absolute chaos, a local spontaneous form of order is found as a result of component interaction.[1] At the same time, the total sum of entropy of the entire system decreases, in accordance with the second law of thermodynamics. Instead of total disorder, the equilibrated state of such a regional system arrives at a local minimum of energy, an attractor state. This process is called self-organization.
1.1.1 History
The principle of self-organization has already been discussed in ancient Greek philosophy. Human understanding of the concept ’the whole is bigger than its parts’ started in ancient Greek times.[1]
In 1875, the English philosopher G.H. Lewes mentioned a dynamical construct arising over time in chemical reactions, that cannot be traced back to the individual combined agencies (chemical reactants).
The first appearance of the term self-organization is found in a paper from 1947 by W.R.
Ashby.[2] He defined a system to be self-organizing if the system changed its own organization, rather than being changed by an external entity. ’Autonomy’ and ’increase in structure’ are key words when trying to define the concept of self-organisation.[1]
1.1.2 Emergence, self-organization and self-assembly
The terms emergence and self-organization are usually incorrectly seen as synonyms. In dy- namic systems, often a combination of both phenomena occurs.[1] However, emergence and self-organization each emphasize different properties of a system and both phenomena can exist in isolation.
Emergence focuses on the presence of a new macro-level property as a result of the interactions between micro-level parts.[1] An example of this is the swarming movement of a flock of birds or a school of fish.
Self-organization focusses on the dynamic and adaptive increase in order or structure without external control.[1] An example of this are ad-hoc networks, that autonomously build their structure as network devices detect each others’ presence. An essential element is that self- organization is a process, meaning there is an increase of order over time.
In figure 1.1, the difference between the two concepts is schematically shown. Where self- organization emphasises the spontaneous increase in order (a), emergence emphasises the arise of novel coherent macro-level properties as a result from micro-level interactions (b). Also visualized is a situation where a combination of the two concepts at the same time is present. Often mistaken for being the same process, both can be involved in constructing the system (c). This type of process is most important for highly dynamic structures like living systems.
Some similarity is also found between the terms self-organization and self-assembly. Both processes are based on the development of collective order from small-scale interactions.[3] The difference is that self-organization is a non-equilibrium process. Self-assembly on the other hand is a spontaneous process that leads toward equilibrium and also requires the individual components to remain essentially unchanged throughout the process.
In complex dynamic systems that are usually found in nature, a combination of these phe- nomena is often present.
1.2. SUPRAMOLECULAR SYSTEMS IN CHEMISTRY
FIGURE1.1. Different systems of emergence and self-organization. Figure reproduced from [1].
1.1.3 Self-assembly in nature
All throughout nature, dynamic self-organizing and self-assembling systems are found. Examples include schools of fish, the flocking of birds, the spontaneous folding of proteins and the formation of lipid bilayer membranes. Self-assembling structures in biological systems are an interesting subject of study, because understanding the properties and behaviour of these structures can give us information about the formation and durability of living systems.
An inspiring example of a self-assembling structure found in living cells is the cytoskeleton.
This is a matrix-like structure in cells, that provides mechanical support and helps maintain the shape and internal organization of each cell. This enables cells to carry out their essential functions like division and movement. It consists of several types of one-dimensional fibers, which are formed from protein monomers by non-covalent interactions.[4] A schematic representation of the cytoskeleton is shown in figure 1.2. Clearly visible are the different types of microfibers, consisting of self-assembled proteins. As mentioned earlier, the cytoskeleton has important functions in intracellular transport, cell motility and mechanical response. In these functions, the cytoskeleton responds to chemical and physical cues by dynamic (de)polymerization of protein monomers within the long ordered filaments. Thus, the kinetics of various (dis)assembly processes within these fibers control vital cell functions. Functions from dynamic processes, like in this case, inspire us to investigate such systems more closely.
1.2 Supramolecular systems in chemistry
The properties of the cell cytoskeleton found in nature can be mimicked by designing synthetic chemical structures. Synthetic analogues of supramolecular polymer systems also consist of highly ordered filaments and display a rich dynamic behaviour, like for example reversible
FIGURE1.2. Schematic model of the cell cytoskeleton, consisting of a network of self- assembled fibers.
monomer-to-polymer transitions.[4]
The emergence of synthetic supramolecular assemblies involving non-covalent interactions like for example hydrogen bonding,π-stacking, charge transfer interactions and hydrophilic- hydrophobic interactions, has become a major subject of study over the last few decades. It has been shown that profound similarities exist between the kinetics of protein-based non-covalent fibers like the cytoskeleton and synthetic supramolecular polymers. Synthetic materials have been described in literature that have properties like adaptability, responsiveness and self-repair.[4]. As such, the latter provide a highly defined model system, suitable for thorough analysis of dynamic processes and aiming at mimicking natural functionality in synthetic materials. Different types of synthetic supramolecular systems have been developed and a number of examples are shown in figure 1.3. A variety of systems can be distinguished, each with their own construction, structure and function.
1.2.1 The water-soluble BTA-molecule
One specific example of a synthetically designed supramolecular system is based on the benzene- 1,3,5-tricarboxamide motif (BTA).[5] It is a supramolecular polymer that has the property to self-assemble in water.[6] When trying to mimic nature, one of the most important aspects to take into account is that water is the main component in living systems. Therefore, solubility of (supramolecular) compounds in water plays a crucial role.
The structure of a water-soluble BTA-molecule is given in figure 1.4. The molecule contains a core, capable of self-assembling into helical one-dimensional aggregates, that is stabilized
1.2. SUPRAMOLECULAR SYSTEMS IN CHEMISTRY
FIGURE1.3. Examples of supramolecular systems.
by threefold intermolecular hydrogen bonding.[7] A schematic representation of such a helical aggregate is shown in figure 1.5. Water solubility of this BTA is achieved due to the ethylene oxide based side chains, that are able to form hydrogen bonds with the solvent.[7] The core and the tail are connected through a C-12 hydrophobic spacer. Studies about self-assembly of BTA-fibers in water indicate that both hydrophobic effects and hydrogen bonding play a crucial role in the formation of fibrillar aggregates.[6] Recent experiments show that hydrophobic interactions dominate the assemblies, while hydrogen bonding provides directionality to the structures, resulting in the fiber-like BTA-structures.[8] Water-soluble supramolecular systems are predicted to show great promise in biomimetic applications.
FIGURE1.4. Molecular structure of a water-soluble BTA-molecule.
FIGURE 1.5. Schematic representation of the helical structure of a self-assembled BTA-fiber. Figure reproduced from [7].
1.2.2 Applications of water-soluble BTA nanofibers
Considerable research has been performed about application areas for BTA nanofibers. Some future applications in for example regenerative medicine are mentioned in literature, where arti- ficial structures can be placed in the human body and be safely eliminated via biodegredation.[4]
Another application could be found in the field of electronics, in the form of conducting nanofibers.
More typical applications can be seen in figure 1.6, and involve nucleating agents for polymers, liquid crystals, organogels, MRI contrast agents, and microcapsules for drug delivery.[7]
FIGURE1.6. Examples of multiple different application areas for BTAs. Figure repro- duced from [7].
1.3. VISUALIZATION TECHNIQUES
1.3 Visualization techniques
For possible applications, more understanding of the properties of BTA-fibers is needed. It is necessary to visualize the polymers to study these properties. There are different visualization techniques available for imaging BTA-fibers.
One type of technique is cryo-TEM, a form of transmission electron microscopy, where samples are studied at cryogenic temperatures. When imaging BTA-fibers, the highly dynamic polymers are frozen during sample preparation, resulting in an image shown in figure 1.7. It is a very efficient method to look at the structure of BTA-fibers in high resolution. Long thin fibers in the order of micrometers in length and with a diameter of approximately 5 nanometers are observed.[6] However, due to the use of cryogenic conditions it is not possible to look at the fiber in its native state.
FIGURE1.7. Imaging BTA-fibers using cryo-TEM. Figure reproduced from [6].
An alternative method for imaging BTA-fibers, is using fluorescence microscopy. This is a common approach for studying dynamic events, compared to other methods that only provide a static snapshot view.[9] It is an accessible technique that does not require extreme imaging conditions. On the other hand, it makes use of a fluorescent dye molecule that needs to be coupled to the sample. Because of its high selectivity towards labeled fluorescent molecules, fluorescence imaging has become one of the most important imaging tools in biology.[10] An example of BTA-fibers imaged using fluorescence microscopy is shown in figure 1.8, where a cluster of fibers is visible.
FIGURE1.8. Imaging BTA-fibers using fluorescence microscopy.
The specificity of fluorescence microscopy is based on the difference between the wavelengths of absorbed light from a photon source and emitted light from the fluorophore. This is called the Stokes shift and it is schematically depicted in figure 1.9. The photon absorption process consists of a transition to an excited state of a fluorophore. After vibrational relaxation, the fluorophore returns to its ground state by fluorescent emission that can be measured. Using this principle, single fluorescent molecules are visible if the background has no autofluorescence.[10]
Total internal reflection fluorescent microscopy (TIRF), makes direct imaging of processes within very close proximity to the coverslip possible.[ 11] An evanescent field of excitation light, see figure 1.10, is created after excitation at a critical angle of the sample.[9] What follows is that the excitation depth is limited to approximately 100 nm, because of rapid decay of the excitation light intensity into the sample. This allows selective imaging of samples in close proximity to the surface. Combining the use of TIRF-illumination with state-of-the-art equipment, gives the ability to image single molecules.[12]
Fundamental limitations impede imaging at great resolution. The Abbe diffraction limit restricts the distinction between two objects separated by a distance less than approximately half the wavelength of light used to image the sample. This limit is roughly 250 nm, which is substantial compared to the size of BTA-fibers. In order to improve the resolution and to be able to study dynamic behaviour inside these fibers, we need to use an alternative visualization technique.
1.4. SUPER-RESOLUTION MICROSCOPY
FIGURE 1.9. Difference between the wavelength of absorbed and emitted light; the Stokes shift.
FIGURE1.10. Evanescence field created due to TIRF-angle.
1.4 Super-resolution microscopy
Recently, super-resolution microscopy techniques have been developed that are able to achieve a resolution up to 20 nm using visible light.[13] In these methods, the position of single fluo- rescent molecules is determined by fitting a two-dimensional Gaussian profile to the individual point-spread function of the excited molecule. Because this visualization technique is based on the detection of single molecules, it offers great improvement in resolution when compared to conventional microscopy.[14]
A particular technique called stochastic optical reconstruction microscopy (STORM), can
be used for high precision imaging. It makes use of sequential and stochastic reconstruction, resulting in precise localization of multiple single molecules.[14] The working principle of STORM entails the reconstruction of images after localization of a single photoswitchable dye. When only a fraction of the fluorophores are switched on, detection of a single fluorescent molecule is possible.[15] Using STORM, only a small, optically resolvable fraction of the photoswitchable dye is detected every snapshot. A photoswitchable dye is a fluorescent molecule that can be switched on and off. Cyanine dyes are an example of such fluorophores. They can be turned on and off reliably for hundreds of cycles before photobleaching.[15]
The switching between on and off state is a reversible photoconversion between a fluorescent and a dark state.[16] The exact mechanism of this photoconversion process is not known, but observations suggest that this switching is facilitated by a primary thiol.[17] For the commonly used Cy5 cyanine dye, the proposed model is the formation of a dark state, because of a thiol- addition to the polymethine bridge of the molecule.[16] This is a reversible chemical reaction, as can be seen in figure 1.11. Experiments showed that adding a primary thiol to the buffer with oxygen scavenging system improves on and off switching of the dyes.[18] Reducing the level of oxygen is important to prevent photobleaching.[9]
FIGURE1.11. Proposed model for the dark state by addition of primary thiol.
As explained before, during imaging, a Gaussian fit is performed for determining the exact place of emission after detection of a single fluorophore. Subsequently, many of such localizations can be combined to reconstruct a final image of high resolution.[15] Labeling BTA-monomers with a fluorophore like Cy5 thereby gives us the ability to look at BTA-polymer fibers. A comparison in achieved resolution between conventional and super-resolution techniques can be seen in figure 1.12, in which BTA-fibers imaged using conventional techniques look blurred compared to the super-resolution images.
In previous experiments, it was already found that monomers dynamically exchange position between different BTA-polymers, as shown schematically in figure 1.13, in contrast to covalently bonded polymers.[20] STORM was used to further examine this process. Experiments showed a homogeneous exchange of monomers along the polymer backbone.[19] By imaging the migration of Cy3-labeled and Cy5-labeled monomers between BTA-fibers, dynamic exchange between solvent and polymer-backbones was demonstrated. This investigation was performed using
1.5. SURFACE CHEMISTRY
FIGURE1.12. Difference in conventional imaging (top) and super-resolution imaging (down) of BTA-fibers. Figure reproduced from [19]
snapshot images at different time points after mixing, as can be seen in figure 1.14. Limitations in sample preparation prevent real-time imaging of these dynamics. The properties of the microscope glass coverslips favour fiber attachment and thus prevent us from imaging kinetic behaviour. Modification of these surfaces using surface chemistry could prove useful for looking at these dynamics, since anti-fouling coatings can be used to prevent settlement of organic residues.[21][22][23]
1.5 Surface chemistry
1.5.1 Basic concepts of silanization
Modification of the chemical composition of surfaces can be performed to obtain surface coatings with altered physical properties. This can be done by incorporating specifically selected functional molecules.
One type of organic molecules that can chemisorb to silicon oxide substrates (glass), are silanes. They contain a chemical group that can bind spontaneously and covalently to the surface and an organic rest group, schematically shown in figure 1.15. The silane part consists of a silicon atom with three leaving groups Xiand a linker part, which often consists of a long hydrophobic alkyl chain. Leaving groups are often hydroxy (−OH), chlorine (−Cl), methoxy (−OCH3) or ethoxy (−OCH2CH3).
Silanes can react with silanol groups ( −SiOH) on silicate substrates like glass surfaces.[ 25]
FIGURE1.13. Monomer exchange after mixing different BTA-fibers. Figure reproduced from [20]
This reaction is called silanization and is shown in figure 1.16. It contains a crucial condensation step, in which small molecules are released. This release prevents the reverse reaction. For example, when a trifunctional silane binds to the silanol groups on the surface, it releases three HX molecules per silane. When X is a hydroxyl group, three molecules of water are released.
The process of silanization consists of several underlying steps. To react, the molecules have to diffuse towards the substrate and at a certain point adsorb to the surface. After adsorption the molecules migrate laterally until they either desorb again or form a chemical bond with the substrate by binding in a potential minimum. The groups X of the silane will first react with water to form silanol groups (SiOH). This is called hydrolysis and it determines the overall reaction speed. The reaction rate depends on the identity of groups X and increases in the series OEt < OMe < Cl. After hydrolysis, the second step is a condensation reaction where the silanol group of the reactant binds to the silanol group on the surface, releasing a water molecule. The quality of deposition is strongly influenced by migration, possible nucleation to the substrate and desorption of the reactant.[24]
These processes are heavily influenced by the experimental steps of the functionalization protocol. Silane chemistry applied to silicate surfaces is usually performed using either vapor phase deposition or solution phase deposition of reactant silanes. Reactions take place at the gas-solid interface with deposition of thin films using chemical vapor deposition.[26] Especially at low pressure, surfaces can be coated with a homogeneous layer. A stable coating is obtained when a gas can react with the substrate surface, also forming gaseous by-products. Using this method, layer thicknesses of the order of 5 - 10µm can be obtained, depending on the exact system and
1.5. SURFACE CHEMISTRY
FIGURE1.14. STORM-imaging of fiber-exchange at different time points. Figure repro- duced from [19]
FIGURE1.15. Structure of a silane. Figure reproduced from [24].
protocol used.[24] The reaction can also be performed in solution, using solution phase deposition, in which we look at the liquid-solid interface. Surface coating can be obtained by immersing glass slides in a solution of reactant silanes. An incubation period of 10 - 60 minutes allows the dissolved silanes to react with the surface, resulting in a more densely functionalized surface.
1.5.2 Surface structure
The surfaces that are formed after reaction of silanes with solid substrates are often so-called monolayers. Because the binding of these molecules occurs spontaneously, these layers are called self-assembled monolayers (SAM).[27]
The growth of a self-assembled monolayer can take place in multiple ways and the quality of
FIGURE1.16. Reaction scheme for silanization. Figure reproduced from [24].
a SAM is highly dependent on the formation mechanism. Different growth mechanisms can be envisioned for these SAMs, because the formation of a SAM is governed by the aforementioned diffusion and adsorption processes. Those two processes occur at different rates, causing SAM formation to be either diffusion or adsorption controlled.[24] Different possible scenarios of monolayer growth are shown in figure 1.17. One is that reactants prefer to adsorb to aggregates already present on the substrate surface until a closed film is formed, scenario A. Another scenario is that first a closed film is formed, either ordered or disordered, scenarios B and C, after which the molecules in the film reorganize to form the final SAM in a second step.[24]
FIGURE1.17. Schematic view of growth of a monolayer by different scenarios. Figure reproduced from [24].
1.6. AIM AND OUTLINE OF THIS THESIS
The final structure of this monolayer is determined by three types of interaction.[24] The first one is the strength of the bond between the reactive head group and the substrate surface.
Further interactions are the attractive van der Waals forces between the linker groups of the silanes. The third and final interaction is the often repulsive interaction between the tail groups of the silanes due to, for example, steric hindrance.
A perfect monolayer is very difficult to form, because of the complex interplay between aforementioned interactions. It is found that it is sterically impossible to form a 1:1 stoichiometric saturation of all silanol groups on the substrate surface.[ 24] The hypothesis is that the surface layer will consist of a polysiloxane network anchored to the surface on some points, where approximately one in five silanes will bind to the surface and the other four are somewhat cross-linked above it, following scenario A, as can be seen schematically in figure 1.18.
FIGURE1.18. Schematic drawing of silanes bound to a silicon oxide surface. Figure reproduced from [24].
To prevent polymerization, water needs to be excluded from the reaction, hence the fact that usually this reaction is performed in dry organic solvents. However, a trace of water is needed for activating both the surface and the silanes. Water can react to the silanes in solution, forming polysiloxane networks that will precipitate, because the water molecules can compete with the hydroxyl groups on the substrate.[24] This could result in a rough surface, while the polysiloxanes can still, partly, bind to the substrate surface. Suitable silane molecules and reaction conditions should therefore be chosen to obtain the desired surface properties.
1.6 Aim and outline of this thesis
In this project, we aim to use surface chemistry approaches to create functionalized glass coverslips suitable for dynamic STORM experiments. The exchange of monomers between solution and aggregate state of a BTA-fiber in an aqueous environment can then be investigated.
In earlier experiments, BTA-fibers were physically adsorbed to glass microscope coverslips before imaging.[19][20] Once the fibers become physisorbed, no dynamic behaviour and monomer
exchange can be studied. This inhibits us from looking in real-time at changes in molecular composition of BTA-fibers. In order to understand how this exchange occurs, physisorption of BTA-fibers needs to be prevented. The approach towards imaging dynamic exchange of monomers consists of several experiments.
The first step is modification of the glass coverslip surface to prevent physisorption of BTA- fibers. The second step involves mixing in a bi-functional linker molecule, that at certain points links the BTA-fiber to the surface to prevent drift, while still maintaining its dynamic properties.
A schematic visualization of this approach is shown in figure 1.19. It shows that BTA-fibers are not completely bonded to the surface, but merely a loose flexible connection via linker molecules is formed, resulting in a fiber that still has the capacity to exchange monomers between solution and aggregate-state. The final step then consists of studying real-time dynamic exchange. In order to study this mechanism at the single-aggregate level, the super-resolution optical microscopy technique STORM is utilized.
FIGURE 1.19. Schematic model of BTA-fibers attached to the surface via linker molecules.
Modification of glass surfaces will be performed using silane chemistry. For anti-fouling behaviour, perfluorinated molecules will be used to create glass with a very high surface energy, and pegylated compounds will be used for their known anti-fouling characteristics in biomimetic anti-fouling coatings. After functionalization of the glass surfaces, the anti-fouling properties will be analysed using fluorescence microscopy and quantitative STORM measurements.
Subsequently, a commercially available silane linker molecule with an NHS-ester functional- ized end-group will be incorporated in the obtained anti-fouling surface for reactivation of this surface. Analysis using fluorescence microscopy will show if BTA-fibers indeed bind to these linker molecules, while maintaining their fiber-like shape and dynamic behaviour.
The final goal is to come up with a general method of producing glass coverslips that can be used for real-time imaging of the dynamics of supramolecular fibers in solution. If surface treatment is promising and successful, also other compounds and structures can be studied via selective binding to these glass coverslips.
C
HAPTER2
BTA-
SYNTHESIS AND DYE LABELINGL
ately, state-of-the-art experiments aim to investigate the dynamics of BTA-fibers using fluorescence microscopy, and for that it is necessary to acquire BTA-fibers with fluorescent properties. These are obtained by assembling fibers in which a small percentage of the monomers are labeled with a fluorescent dye molecule. This chapter describes how these BTA- molecules are obtained after reduction of BTA-triazide and subsequent coupling of BTA-triamine to a fluorescent cyanine dye (Cy3 and Cy5).2.1 Reduction of azide
A BTA-monomer has three arms with functional end-groups, schematically depicted in figure 2.1. The commercially available fluorescent dye molecule we use to couple with the BTA contains an NHS-ester reactive group. To react with our BTA, we need to obtain a BTA-monomer with primary amines at the ends of the three arms.
The BTA-triamine cannot be stored for too long, because it is unstable over time. The BTA- triazide does not degrade appreciably over time when stored at minus 20 degrees Celsius, and was therefore used as the starting material for the synthesis. Thus, the first reaction step is a reduction from precursor BTA-triazide to BTA-triamine. The reaction scheme of this reduction is shown in figure 2.2.
We used Staudinger conditions for the reduction that was performed. This reaction was named after Hermann Staudinger, who in 1919 reported that azides could react with triphenylphosphine to form phosphazide.[28] The reaction produces a high phosphazide yield and is nowadays commonly referred to as the Staudinger reduction. The exact reaction mechanism is shown in figure 2.3. It entails a nucleophilic attack of the phosphine at the terminal nitrogen atom of the azide, that eventually produces an iminophosphorane intermediate. Nitrogen gas is evolved
FIGURE2.1. Different end-groups of a BTA-monomer.
FIGURE2.2. Reaction scheme of reduction of BTA-triazide to BTA-triamine.
during this reaction, making it irreversible. In the second step, this intermediate is hydrolysed to form the amine and triphenylphosphine oxide. The reaction conditions are very mild and are here preferred over hydrogenation using palladium on carbon.
Exact experimental details can be found in the materials and methods section at the end of this chapter. Analysis using1H-NMR and IR-spectroscopy showed complete reduction of the starting material. The spectra can be found in appendix A. The desired product was obtained with a yield of 70 mg (55µmol), or 74%.
2.2. COUPLING WITH FLUORESCENT DYE
FIGURE2.3. Reaction mechanism of Staudinger reduction.
2.2 Coupling with fluorescent dye
Next step was coupling the obtained BTA-triamine with a fluorescent dye. Commercially available NHS-esters of the dye were available for this. In this case, two different dyes were used, namely the Cy3 and Cy5 cyanine dyes, because of their high extinction coefficient and their fluorophore emission maximum in the green and red region. These dyes have an activated ester that can react with primary amines.
The coupling of this dye with an amine is strongly pH-dependent reaction. At a low pH-value, the amino group is protonated, so that no reaction takes place. The optimal pH-value for reaction is around 8.3-8.5. At a higher value, NHS-ester hydrolysis is fast, so that modification yield diminishes.
The coupling reaction is performed as a statistical reaction, because the starting material is a BTA-monomer with three primary amines, that can all react with the dye-ester. The coupling reaction is shown in figure 2.4 and figure 2.5 for the case of Cy5. The preferred product is the mono-labeled BTA-dye, but during the reaction also the di- and tri-labeled BTA-dye molecules are formed. A stoichiometric ratio of 0.8 was used in the reaction, maximizing the theoretical relative yield of the mono-labeled BTA-dye.
2.3 Separation of desired product from statistical mixture
The desired product is the mono-substituted BTA-dye, because for detecting a BTA-monomer, only a single fluorophore is required. Thus, it was necessary to separate it from the mixture with unreacted BTA-triamine, di-substituted and tri-substituted BTA-dye. Separation of the multiple components has been performed using reversed-phase high-pressure liquid chromatography. The crude mixture of dry product was dissolved in a 30:70 mixture of acetonitrile and demineralized water. Next, the solution mixture was injected into the column and then run with a gradient of
FIGURE2.4. Reaction scheme of coupling of the BTA-triamine with a cyanine dye.
(a) (b) (c)
FIGURE2.5. Coupling reaction of BTA-triamine (a) after adding 0.8 eq of Cy5-NHS (b) and dissolving in DMF (c).
acetonitrile (MeCN) and demineralized water with 0.1% TFA from 40% MeCN to 50% MeCN in 10 minutes.
After performing these steps, multiple compound fractions were found, separated and charac- terized by mass. The mass-spectra can be found in appendix A. Some unidentified and unwanted side products were present, which are indicated by the multiple peaks in the spectra. These other compounds competed with the BTA-triamine in the coupling reaction, resulting in a mixture of different BTA-based compounds, each with or without the dye-molecule. The desired BTA- diamine-monoCy3 and BTA-diamine-monoCy5 were identified in the spectra and separated from the mixture.
2.4. ASSEMBLING BTA-FIBERS WITH FLUORESCENTLY ACTIVE MONOMERS
Each fraction with the mono-labelled BTA-dye was lyophilized and subsequently redissolved in a small volume of 30:70 acetonitrile and demineralized water. After recombination of these small solutions and another lyophilization step, the final product was obtained. The final yield was determined to be 1.6 mg (5%) for BTA-Cy3 and 1.2 mg (4%) for BTA-Cy5 as a rough estimate after weighing. The dried products obtained after final purification were dissolved as a stock solution in DMSO to an approximate concentration of 1 mM.
2.4 Assembling BTA-fibers with fluorescently active monomers
2.4.1 Calculating exact yield using calibration curve
After obtaining BTA-Cy3 and BTA-Cy5 monomers, the exact yield was determined by measuring absorbance spectra and constructing a calibration curve. Dilution series of both the Cy3-NHS and Cy5-NHS ester dyes in DMSO were made (shown in figure 2.6), after which absorption spectra were recorded. The absorption coefficient was determined by a linear fit through the origin. The absorption values at 554 nm for the Cy3-NHS ester solutions and values at 649 nm for the Cy5-NHS ester solutions were used for plotting the calibration curve, which are shown in figure 2.7. The absorption coefficient of Cy3-NHS ester in DMSO was found to be 101·103 M−1·cm−1 and the absorption coefficient of Cy5-NHS ester in DMSO was calculated at 158·103 M−1·cm−1.
(a) (b)
FIGURE 2.6. Dilution series of Cy3-NHS ester (a) and Cy5-NHS ester (b) in DMSO, series range: 10 - 5 - 2.5 - 1.25 - 0.625µM.
Subsequently, the absorption of diluted samples of the synthesized BTA-Cy3 and BTA-Cy5 stock solutions were determined. These measurements showed that the exact concentrations of these solutions were 0.69 mM for the BTA-Cy3 solution and 0.43 mM for the BTA-Cy5 solution.
Using these values, a more precise estimation of the reaction yield could be calculated than weighing. It was found that the exact yield of the BTA-Cy3 was 0.35 mg and of the BTA-Cy5
(a) (b)
FIGURE2.7. Calibration curve for Cy3-NHS ester (a) and Cy5-NHS ester (b) in DMSO.
was 0.22 mg. After this final determination, the stock-solutions could be used for assembling BTA-fibers to desired specification.
2.4.2 Stock solutions and stack-assembly
Two stock solutions of mono-labelled BTA-Cy3 and BTA-Cy5 in DMSO were prepared and used for the assembly of supramolecular polymer samples. The mono-labelled BTA-Cy3 and BTA- Cy5 were mixed with BTA-3OH monomers (see figure 2.1, R = OH), after which milliQ water was added to obtain two stock-solutions of 25µM total BTA concentration. The ratio between BTA-Cy3/BTA-Cy5 and BTA-3OH was chosen, so that 2% of the monomers were labeled with a dye molecule. The samples were then equilibrated for 24 hours before experiments, crucial for reliable self-assembly of BTA-fibers. Exact experimental details can be found in the materials and methods section at the end of this chapter. The obtained stock-solutions in milliQ are used throughout all further experiments mentioned in this thesis.
2.5 Conclusion
BTA-triamine was synthesized using the Staudinger reaction. The reduction from BTA-triazide towards BTA-triamine was successful, shown by analysis using1H-NMR and IR-spectroscopy.
The desired BTA-triamine was obtained with a yield of 70 mg (55µmol), or 74%.
After a coupling reaction, both the mono-labeled BTA-Cy3 and BTA-Cy5 were obtained. The final yield turned out to be relatively low, because of undesired side-products that competed in the coupling reaction. We managed to separate and purify the final product, resulting in a yield of 0.35 mg of the BTA-diamine-monoCy3 and 0.22 mg of the BTA-diamine-monoCy5.
2.5. CONCLUSION
Even though the yield was low, it was enough for assembling the desired BTA-fibers, for which the mono-labelled BTA-Cy3 and BTA-Cy5 were mixed with BTA-3OH monomers. Finally, two stock-solutions of 25µM were obtained in which 2% of the monomers were labeled with a dye molecule.
2.6 Materials and methods
2.6.1 Materials
All commercial reagents were purchased from Aldrich and used as received unless stated otherwise. All solvents were purchased from Biosolve and used without further purification unless stated otherwise. Water was purified on an EMD Milipore Mili-Q Integral Water Purification System. Cy3- and Cy5-NHS esters were obtained from Lumiprobe.
Reactions were followed by thin-layer chromatography (precoated 0.25 mm, 60-F254 silica gel plates from Merck).
2.6.2 Instrumentation
Infrared spectra were recorded on a Perkin-Elmer Spectrum One 1600 FT-IR spectrometer.1H-NMR spectra were recorded on a Varian Mercury Vx 400MHz NMR spectrometer. Ultraviolet-visible absorbance spectra were recorded on a Jasco V-650 UV-vis spectrometer with a Jasco ETCT-762 temperature controller. Flash chromatography was performed on a Biotage flash chromatography system using 200-425 mesh silica gel (Type 60A Grade 633). Preparative reversed-phase high-pressure liquid chromatography (prep-HPLC) was performed on a system consisting of the following components: Shimadzu LC-8A preparative liquid chromatography pumps (with an Alltima C185 u (125x20 mm) preparative reversed-phase column and gradients of water-acetonitrile, supplemented with 0.1% trifluoroacetic acid), a Shimadzu CBM-20A prominence communications bus module and Shimadzu DGU 20A3 prominence degasser, Thermo Finnigan Surveyor PDA detector, Finigan LCQ Deca XP and Thermo Finnigan surveyor auto sampler.
Reversed-phase high-pressure liquid chromatography-mass spectrometry (RP-HPLC-MS) was performed on a system consisting of the following components: Shimadzu SCL-10A VP system controller with Shimadzu LC-10AD VP liquid chromatography pumps (with an Alltima C18 3 u (50x2.1 mm) reversed-phase column and gradients of water- acetonitrile supplemented with 0.1% formic acid, a Shimadzu DGU 20A3 prominence degasser, a Thermo Finnigan surveyor auto sampler, a Thermo Finnigan surveyor PDA detector and a Thermo Scientific LCQ Fleet.
2.6.3 Reduction of BTA-triazide to BTA-triamine
For the reduction, 102 mg (74 µmol) of the BTA-triazide and 106 mg (401 µmol) of triphenyl phosphine were dissolved in 13 ml of THF and 3 ml of H2O. The reaction was stirred overnight. The reaction was monitored using TLC (5%
MeOH/CHCl3). Another 107 mg (404 µmol) of triphenyl phosphine was added to the reaction mixture to promote complete reduction. After stirring at room temperature for another night, and completion of hydrolysis, analysis using IR-spectroscopy showed complete reduction of the starting material. The reaction mixture was separated into the different components using a Biotage silica column. Triphenyl phospine and its oxide could be removed by eluting with 5% MeOH/CHCl3. The product was obtained after addition of 2% i-PrNH2, yielding 70 mg (55 µmol) of the BTA-triamine as a colorless solid, resulting in a 74% yield. Analysis using1H-NMR and IR-spectroscopy showed complete reduction of the starting material.
2.6.4 Coupling of BTA-triamine with Cy3/Cy5 dyes
For dye labeling with Cy3, 25 mg (20 µmol) of BTA-triamine together with 9 mg (15 µmol) of the Cy3-NHS ester were dissolved in 1 ml of dry dimethylformamide (DMF). After adding a drop of TEA, the reaction mixture was allowed to stir overnight. After evaporation of the solvent, the mixture was purified via preparative-HPLC to isolate only the monosubstituted product, using a gradient from 40-50% MeCN. After purification and lyophilization, 1.6 mg (by weighing) of the desired product was isolated as a red powder-like solid. LC-MS analysis of the product revealed a single peak corresponding to the expected mass. The same procedure was followed for labelling with Cy5, which was isolated as a blue powder-like solid, where a yield of 1.2 mg (by weighing) was obtained.
2.6. MATERIALS AND METHODS
2.6.5 Stack-assembly
The assembly of supramolecular polymer samples was achieved through a dilution protocol. Stock solutions of BTA- 3OH (10 mM in DMSO), and single labeled BTA-Cy3 (0.69 mM in DMSO) and BTA-Cy5 (0.43 mM in DMSO) were prepared. The stock solutions of dye labeled BTAs were standardized based on the absorbance of free dye in DMSO.
The stock solutions were combined and mixed to provide the correct concentration and dye ratio for the desired sample, and finally diluted with filtered Milli-Q water. The preparation involved mixing 10 µl of BTA-3OH solution with 2.9 µl of BTA-Cy3 solution and a separate batch that involved mixing 10 µl of BTA-3OH solution with 4.7 µl of BTA-Cy5 solution. Both were followed by dilution with 4 ml of water, producing 25µM BTA-solution concentrations, each with 2% dye labeling. The samples were then allowed to equilibrate for 24 hours before experiments, crucial for reliable stack formation.
C
HAPTER3
A
NTI-
FOULING SURFACEI
n order to overcome physisorption and the corresponding freezing of kinetics in BTA-fibers, we focus on the realization of an anti-fouling surface. In this approach we test a selection of different functional molecules that can react with the glass surface and change its physical properties, thus aiming to create a surface with anti-fouling properties.[29][30]3.1 Current method
In previous experiments involving the imaging of BTA-fibers, fibers were physisorbed to a glass coverslip surface used for microscopy. For sample preparation, a solution of BTA-fibers was flushed over such a glass slide. During a period of incubation, the polymers collide with the glass surface and a small proportion then remains adsorbed. This immobilization on the one hand allows us to image these fibers, but on the other hand it freezes the kinetics and prevents us from looking at the highly dynamic properties these BTA-fibers possess when they are in solution.
Imaging dynamic exchange of monomers along the chain of a single fiber is impossible with this method. Additionally, physisorption can be used as a general procedure for fiber sample preparation, the adsorption protocol needs to be optimized for different supramolecular polymers.
In the current protocol, samples with physisorbed BTA-fibers are measured to provide snap- shot information about their structure and composition. In order to achieve coverslips with excellent properties for BTA-imaging, the fiber-surface interaction is of great importance. Un- treated commercial slides suffer from surface contaminations and don’t always have a smooth surface. Intact BTA-fibers will not physisorb to these slides. Therefore, the glass coverslips are thoroughly cleaned using a selection of chemicals and solvents to remove contaminations.[25]
With this procedure, a smooth glass surface is obtained with active hydroxyl groups on the
surface. After incubation with a BTA-solution, these rinsed slides become covered BTA-fibers. An example of a surface obtained using this method can be seen in figure 3.1. This is an image of the surface of a cleaned glass coverslip and the physisorbed fiber-like aggregates are clearly visible.
To prevent fibers from adsorbing to the surface, the glass surface needs to be modified to avert physisorption.
FIGURE3.1. BTA-fibers physisorbed to the glass surface after cleaning using piranha- etch.
3.2 Evaluating anti-fouling glass coverslips
3.2.1 Slide manufacturing
The glass coverslips are chemically functionalized to obtain an anti-fouling surface that prevents physisorption of BTA-fibers. The functionalization of silicate surfaces like glass is carried out us- ing the following procedure. To remove organic residues and create a high hydroxyl group density, the substrates are first hydrophilized using a strong oxidizing etch like piranha, consisting of
3.2. EVALUATING ANTI-FOULING GLASS COVERSLIPS
sulfuric acid (H2SO4) mixed with hydrogen peroxide (H2O2).[25][31] Afterwards, the substrates are washed with different organic solvents in a series that becomes less and less polar, to remove as much water as possible. Subsequently, reactant silanes that are used to functionalize the surface can be dissolved in a dry organic solvent, creating a solution in which the coverslips can be immersed. After the reaction, the substrates are taken out of the silane solution and are washed with solvent to remove unreacted and physisorbed silanes.[24] Different types of reactant silanes have been have been tested to obtain an anti-fouling surface. Subsequently, anti-fouling characteristics of these modified surfaces have been evaluated.
3.2.2 Anti-fouling assessment
Evaluating anti-fouling characteristics of coverslips starts with checking for aggregates on the surface using fluorescence microscopy.[32] For an anti-fouling surface, the resulting slide should be free of any extended aggregates, like those seen in figure 3.2. The presence of aggregates is evaluated by visual inspection of the surface after incubation with a solution of BTA-fibers.
FIGURE3.2. Distinction between aggregates and background levels of fluorescence on the surface.
The second evaluation step was the investigation of the amount of background fluorescence that is present on the surface. In addition to large aggregates, small particles and monomers could also settle on the surface. An optimal anti-fouling surface has been obtained when the background levels of fluorescence are low, thus indicating that smaller fluorescent moieties, monomers and contamination do not stick to the surface. In figure 3.2, an arbitrarily chosen point on the surface is indicated with ’background’ showing the distinction in evaluation between aggregates and background levels of fluorescence.
To measure the background levels of fluorescence, we used quantitative STORM experiments.
Glass coverslip surfaces were examined by determining the number of localizations on three random areas on the surface using a STORM-analysis, during which all fluorophores present on that surface area are located and counted. Two different coverslips were analysed for each type of surface, resulting in a total of six background measurements. The average number of localizations was calculated, which is a measure of the amount of fluorescent moieties present per surface area. The lower this number, the better the anti-fouling properties of the glass coverslip.
Normally, the slides are flushed with a solution of BTA-fibers that contains fluorescently active dye-molecules. However, contaminations that are present in the solvent could also be fluorescently active. A control experiment is therefore performed by incubating untreated slides with the pure solvent to obtain reference values, which are presented in figure 3.3. For untreated commercial slides, the average number of localizations was found to be 5,20 perµm2when flushed with a 5µM solution of BTA-fibers and 0,19 perµm2 when flushed with milliQ water.
FIGURE3.3. Reference values for background levels of fluorescence.
An overview of the surfaces of these slides, that also gives an impression of the type of aggregates that are present on the surface, is given in figure 3.4. Clearly visible are fiber-like aggregates on the surface of the piranha-etched slide, while the aggregates on the untreated slides are less defined. We need to modify the surface properties to prevent this physisorption.
3.3. FUNCTIONALIZATION WITH FLUORINATED LAYER
(a) (b)
FIGURE 3.4. Surface overview before functionalization for reference values with (a) untreated slide and (b) piranha-etched slide.
3.3 Functionalization with fluorinated layer
Fluorinated molecules have been previously used to create anti-fouling surfaces.[33][34][35] They are prepared by functionalizing glass coverslips with a compound that has a large fluorinated end-tail, giving a surface that has hydrophobic properties. The high surface energy could cause BTA-fibers to remain in solution instead of physisorbing to the surface. Two different fluorinated compounds, shown in figure 3.5, were investigated for these effects.
(1) (2)
FIGURE3.5. Used fluorinated compounds for surface functionalization.
Compounds (1) and (2) display structural variations in two respects; reactive group and functional tail. Compound (1) has a shorter fluorinated tail than compound (2), which could affect the degree of hydrophobicity and amount of steric effects. Compound (2) could be more prone to cross-linking because of the two extra chlorines, which are better leaving groups, leading to faster reactions and a higher probability of polymerization in solution. The consequences are that a difference in reactivity and in final morphology between the two compounds may be observed.
3.3.1 Surface modification
Both vapor phase and solution phase silanization approaches were used to create fluorinated surfaces. Compound (1), a liquid at room temperature, was tested using vapor deposition methods.
Compound (2) is a solid at room temperature and was applied using solution phase silanization.
For this compound, the solvophobic interactions with the different tested solvents appeared to be high, causing poor solubility. In the end, concentrations as low as 0.01% w/v have been used for this compound to obtain a clear solution in dichloromethane (DCM), that could be used for functionalization.
3.3.2 Anti-fouling evaluation
After functionalization with the fluorinated compounds, the anti-fouling characteristics of the coverslips were investigated. A first visual inspection of the surface obtained after chemical functionalization using the vapor method with compound (1), showed that the modification was successful. According to contact angle assessment, shown in figure 3.6, we can conclude that the surface has been functionalized, as the surface energy for the fluorinated slide is much higher than the reference slide, as was expected.
FIGURE3.6. Difference in surface energy visually observed between glass coverslips using a water droplet; left = untreated, right = fluorinated.
Further evaluation was performed by flushing a 5µM solution of 2% dye-labeled BTA-fibers over the glass coverslip, as described for previous experiments in paragraph 3.2. Images of these fluorinated surfaces can be seen in figure 3.7. The visible aggregates on these images are still mobile and thus not physisorbed. By visual inspection of the surface using fluorescence microscopy, it appears that on a timescale of seconds, BTA-fibers from solution approach the surface, giving a small spherical dot of fluorescent activity. After a few seconds, these fluorescent signatures spread out, becoming larger spherical blobs of fluorescence before eventually vanishing into the background. Our hypothesis based on these observations is that a BTA-fiber hits the surface and, instead of returning into solution, falls apart, resulting in monomers spreading
3.4. FUNCTIONALIZATION WITH PEGYLATED LAYER
across the surface due to the strong hydrophobic interaction between the surface and the C-12 spacer of the BTA.
When looking at levels of background fluorescence using quantitative STORM, the fluorinated slides provide no improvement when compared to the reference values of the untreated and piranha-etched slides, as can be seen in figure 3.8. Thus, we conclude that fluorinated coverslips do not provide the desired surface characteristics. An anti-fouling surface for BTA-fibers should be obtained using functionalization with a different molecule.
(a) (b)
FIGURE3.7. Glass surfaces after functionalization with (a) compound (1) using vapor phase silanization and (b) compound (2) using solution phase silanization.
3.4 Functionalization with pegylated layer
Surfaces coated with water-soluble polyethylene-glycol (PEG) brushes are known from literature for their anti-fouling behaviour.[36][37][31] In this work, two different types of compounds were investigated, a silane with a short ethylene-glycol chain of three units and a silane with a slightly longer ethylene-glycol of six to nine units. These silane-compounds (3) and (4) are shown in figure 3.9. A difference with respect to resulting functionalized surface can be expected between these compounds. On the one hand, the longer compound (4) may provide better anti-fouling characteristics, because it creates a larger distance between the actual surface and the moieties in solution. On the other hand however, the shorter compound (3) may provide a better monolayer surface, because there is less steric hindrance during functionalization, making the actual surface more accessible for reaction.
