Assembly and Post-Assembly Manipulation of
Polyelectrolyte Multilayers for Control of Bacterial
Attachment and Viability
By
Jenny A. Lichter
B.S. Materials Science and Engineering
Massachusetts Institute of Technology, 2004
SUBMITTED TO THE DEPARTMENT OF MATERIALS SCIENCE AND
ENGINEERING IN PARTIAL FULFILLMENT
OF THE REQUIREMENTS FOR THE DEGREE OF:
DOCTOR OF PHILOSOPHY
IN MATERIALS SCIENCE AND ENGINEERING
AT THE
MASSACHUSETTS INSTITUTE OF TECHNOLOGY
FEBRUARY 2009
©2009 Massachusetts Institute of Technology. All rights reserved.
Signature of Author: __________________________________________________
Department of Materials Science and Engineering
January 31, 2009
Certified by: __________________________________________________________
Michael F. Rubner
Professor of Materials Science and Engineering
Thesis Supervisor
Accepted by:__________________________________________________________
Christine Ortiz
Associate Professor, Department of Materials Science and Engineering
Chairman of the Department Graduate Committee
Assembly and Post-Assembly Manipulation of Polyelectrolyte Multilayers for
Control of Bacterial Attachment and Viability
By
Jenny A. Lichter
Submitted to the Department of Materials Science and Engineering
on January 30th 2009 in Partial Fulfillment of the
Requirements for the Degree of Doctor of Philosophy in
Materials Science and Engineering
ABSTRACT
The overall goal of this thesis was to exploit the versatility of the polyelectrolyte multilayer (PEM) platform to consider bacteria‐substrata interactions by varying multilayer assembly and post‐assembly conditions. We developed multiple PEM systems to probe the ability of substrata to resist bacteria attachment or act as contact‐killing antimicrobials. In the first study, by varying the pH of assembly, we developed PEMs of identical chemical composition (polyallylamine hydrochloride (PAH) and polyacrylic acid (PAA)) with distinct mechanical moduli (1‐100 MPa). Once characterized, these PEMs showed that, under certain conditions, bacterial attachment correlated with increasing modulus. Thus, substrata stiffness was found to be an additional parameter to consider when studying bacterial attachment. The next project focused on PEMs of PAH and poly(sodium‐4‐styrene sulfonate) (SPS) assembled at high pH that showed a reversible swelling transition upon immersion in a low pH solution. These acid‐treated PEMs presented high positive charge density and mobility, and were capable of killing bacteria on contact. SPS/PAH PEMs were used as a model system to enumerate the design parameters that should be considered to create a cationic killing surface. A third PEM system was employed to further illustrate the effects of multilayer assembly and post‐assembly conditions on bacteria. Cross‐linked hydrogen‐
bonded PAA and poly(acrylamide) (PAAm) multilayers were modified post‐assembly by the adsorption of PAH at various pH values. These multilayers underwent a variety of morphological transitions depending on the pH of PAH adsorption. At mid‐range pH values, the film stiffened and promoted aqueous bacterial attachment. At high pH values, PAH adsorbed onto the surface with many unbound uncharged amine groups. When the multilayer was exposed to physiological pH values for bacteria assays, the amine groups became protonated and participated in a cationic‐killing effect. Finally, biofilm control was examined by investigating initial biofilm formation on films of various mechanical stiffness and surface charge. No differences were visible via optical microscopy. An alternative approach to biofilm control was considered whereby a dissociating multilayer region lifted‐off a contaminated layer, exposing a clean, unfouled underlying surface. Thesis Supervisor: Michael F. Rubner
Title: TDK Professor of Materials Science and Engineering
2 | P a g e Table of Contents ABSTRACT ...................................................................................................................................................... 2 List of Figures ................................................................................................................................................ 6 List of Tables ............................................................................................................................................... 10 Acknowledgements ..................................................................................................................................... 11 1. Introduction ........................................................................................................................................ 12 1.1 Introductory Remarks ................................................................................................................. 12 1.2 General Introduction ......................................................................................................................... 13 1.2.1 Polyelectrolyte Multilayers ........................................................................................................ 13 1.2.2 Antibacterial Multilayers ............................................................................................................ 15 1.2.3 PEMs and Cells ........................................................................................................................... 25 1.3 2. Thesis overview ........................................................................................................................... 29 Regulation of Bacterial Adhesion by Manipulating Polyelectrolyte Multilayer Stiffness ................... 30 2.1 Introduction ................................................................................................................................ 30 2.2 Materials and Methods ............................................................................................................... 33 2.2.1 PEM Assembly. ........................................................................................................................... 33 2.2.2 PEM Elastic Moduli. ................................................................................................................... 34 2.2.3 Substrata Surface Energy and Interaction Energy. .................................................................... 35 2.2.4 Substrata Surface Charge Density. ............................................................................................. 37 2.2.5 Methylene Blue Staining ............................................................................................................ 38 2.2.6 Elimination of Free Carboxylate Charge .................................................................................... 38 2.2.7 Bacterial Attachment Assays...................................................................................................... 38 2.3 Results and Discussion ................................................................................................................ 40 2.3.1. Bacterial colonization can be reduced by material substrata modifications. ........................... 40 2.3.2. Weak polyelectrolyte multilayers modulate stable adhesion of s. epidermidis bacteria. ....... 41 2.3.3. Characterization of polymeric substrata properties. ................................................................ 42 2.3.4. S. epidermidis adhesion modulated chiefly by substrata mechanical compliance. .................. 46 2.3.5. Mechanoselective adhesion is independent of monovalent ion concentration. ..................... 50 3 | P a g e 5.3.6 Mechanoselective Adhesion is also Exhibited by Wild‐Type and Mutant E. coli ....................... 51 5.4. Conclusions ...................................................................................................................................... 54 3. Reversibly Antibacterial Polyelectrolyte Multilayers .......................................................................... 56 3.1 Introduction ................................................................................................................................ 56 3.2 Materials and Methods ............................................................................................................... 59 3.2.1 PEM Assembly: ........................................................................................................................... 59 3.2.2 Thickness and Swelling Measurements: .................................................................................... 60 3.2.3 QCM‐D Analysis .......................................................................................................................... 61 3.2.4 Rose Bengal Dye Absorption: ..................................................................................................... 61 3.2.5 Streaming Zeta Potential Measurements: ................................................................................. 61 3.2.6 Micron Particle Attachment: ...................................................................................................... 61 3.2.7 Aqueous Bacterial Attachment Assays (or waterborne bacterial assays): ................................ 62 3.2.8 Airborne Antibacterial Assays: ................................................................................................... 62 3.3 Results ......................................................................................................................................... 63 3.3.1 Airborne Antimicrobial Assays: .................................................................................................. 63 3.3.2 Waterborne antibacterial assays: .............................................................................................. 68 3.3.3 Multilayer characterization: ....................................................................................................... 70 3.3.4 Other antibacterial PEMs: .......................................................................................................... 74 3.4 4. Discussion and Conclusion .......................................................................................................... 76 Post‐assembly Alteration of Hydrogen Bonded Polyelectrolyte Mutlilayers ..................................... 80 4.1 Introduction ...................................................................................................................................... 80 4.2 Materials and Methods ..................................................................................................................... 85 4.2.1 Materials .................................................................................................................................... 85 4.2.2 Sample Preparation ................................................................................................................... 85 4.2.3 Polycation Adsorption onto PEMs ............................................................................................. 86 4.2.4 Surface Morphology Analysis ..................................................................................................... 86 4.2.5 Thickness measurements ........................................................................................................... 86 4 | P a g e 4.2.6 Fourier Transform Infrared (FT‐IR) Spectroscopy ...................................................................... 87 4.2.7 Fluorescent Labeling of PAA ...................................................................................................... 87 4.2.8 Fluorescence Microscopy ........................................................................................................... 88 4.2.9 Quartz Crystal Microbalance with Dissipation ........................................................................... 88 4.2.10 Waterborne Bacteria Attachment Assays ................................................................................ 88 4.2.11 Airborne Antibacterial Assays .................................................................................................. 89 4.3 Results and Discussion ...................................................................................................................... 89 4.3.1 Morphology dependence on pH ................................................................................................ 89 4.3.2 Fluorescence Microscopy ........................................................................................................... 91 4.3.3 Degree of Ionization of Weak Polyelectrolytes.......................................................................... 95 4.3.4 Adsorption of Other Polycations onto the PAA/PAAm Multilayer ............................................ 98 4.3.5 Bacterial Attachment and Viability on Different Morphologies. ............................................... 99 4.4 Conclusions ..................................................................................................................................... 101 5. Initial Biofilm Growth on Polyelectrolyte Multilayers ...................................................................... 103 5.1 Introduction .................................................................................................................................... 103 5.1.1 An Overview of Biofilms ........................................................................................................... 104 5.2 Biofilm growth on PEMs capable of resisting colonization of individual bacteria .......................... 106 5.2.1 Materials and Methods ............................................................................................................ 107 5.2.2 PEMs with Varying Mechanical Modulus ................................................................................. 110 5.2.3 Antibacterial PEMs ................................................................................................................... 112 5.3 Liftoff with Temperature and pH Triggerable Hydrogen Bonded PEMs ..................................... 113 5.4 Conclusions ..................................................................................................................................... 116 6. Summary and Future Work ............................................................................................................... 118 6.1 Thesis Summary .............................................................................................................................. 118 6.2 Future Research Directions ............................................................................................................. 120 APPENDIX A ............................................................................................................................................... 137 APPENDIX B ............................................................................................................................................... 139 5 | P a g e List of Figures Figure 1.1. (0.1) General setup of layer‐by‐layer dipping for PEM assembly. A substrate is sequentially dipped into aqueous solutions of oppositely charged species with rinse baths between each step. ....... 13 Figure 1.2. (0.2) Common polymers used in PEM synthesis: (a) weak polyelectrolyte poly(acrylic acid) (PAA), (b) polyacrylamide (PAAm), which can hydrogen‐ bond with PAA, (c) strong polyelectrolyte poly(sodium 4‐styrene sulfonate) (SPS) and (d) weak polyelectrolyte poly(allylamine hydrochloride) (PAH). .......................................................................................................................................................... 14 Figure 1.3. (0.3) Three general methods of creating antibacterial surfaces. .............................................. 15 Figure 1.4. (0.4) Proposed mechanisms of cationic contact‐killing: A) Cationic surface charge displaces membrane counterions increasing the permeability of the bacterial membrane. B) Cations hold negatively charged bacteria at the substrate surface and hydrophobic tail penetrates the membrane, causing leakage. .......................................................................................................................................... 19 Figure 2.1. (0.5) PEMs reduce bacterial adhesion on medical grade titanium. Adhesion of waterborne S. epidermidis is reduced by coating with a pH‐tunable polyelectrolyte multilayer (PEM) film of PAA and PAH assembled at pH 2.0, and is stable at both 2 h (inset; circle indicates one such colony) and 4 h incubation duration. Scale bars = 5 mm. .................................................................................................... 40 Figure 2.2. (0.6) Bacterial colonies observed for 103 – 108 S. epidermidis/mL in 150 mM NaCl PBS. (a) Average colony number per unit substrata area increased with increasing incubation concentration for greater than 105 cells/mL; for all concentrations, the density of colonies observed on the PEM substrata assembled at pH 6.5 (●) was significantly greater than that observed on the substrata assembled at pH 2.0 (▼). (b) For given initial concentration, colony number was greater and colony size was smaller on stiffer substrata, supporting a model whereby bacteria attachment is modulated in part by substrata stiffness, but subsequent growth is affected predominantly by available space and nutrients. Scalebars = 500 mm. ...................................................................................................................................................... 41 Figure 2.3. (0.7) Colony density as a function of various surface parameters. (a) Colony density varies directly with substrata elastic moduli E. All sample differences statistically significant (1‐way ANOVA, a = 0.05, P = 0.0059). (b) Colony density is independent of RMS surface roughness of the substrata. Scale bar = 5 mm. (c) Total interaction energy ΔGMWP for the microbe‐water‐PEM system is statistically indistinguishable among all substrata considered (1‐way ANOVA, a = 0.05, P =0.987). (d) Surface charge density Q, as measured via electrostatic repulsion of a carboxylated spherical probe in Milli‐Q water (see Methods), is within standard deviation for PEMs assembled at pH 2.0 (compliant) and pH 6.5 (stiff). Representative charge repulsion curve (solid) and constant‐surface‐charge model fit (dashed) are 6 | P a g e shown. Symbols refer to the following PEMs: PAA/PAH 2.0/2.0 (▼), 4.0/4.0 (x) in (a) to consider intermediate substrata stiffness, 6.5/6.5 (●), 3.5/7.5 (♦), and 3.5/8.6 (■). ............................................... 47 Figure 2.4. (0.8) Multilayer addition to modulate composite substrata stiffness. Addition of 0.5 and 1 bilayer of PAA/PAH at pH 2.0 onto a stiff PEM (pH 6.5) decreases the effective mechanical stiffness of the substrata (grey circles) and decreases the bacterial colony density (black columns). Addition of one bilayer of PAA/PAH at pH 6.5 to a compliant PEM (pH 2.0) increases effective stiffness (black triangles) and bacterial colony density (grey columns). We observed statistically significant differences in the colony densities among the masked PEM 6.5 substrata and among the masked PEM 2.0 substrata, respectively (1‐way ANOVA, a = 0.05 with P = 0.00027 and 0.0031, respectively). ................................... 49 Figure 2.5. (0.9) S. epidermidis colony density as a function of solution ion concentration. Colony density on compliant substrata (black, E ~ 1 MPa) is lower than that on stiff substrata (gray, E ~ 100 MPa), regardless of solution monovalent ion concentration in which 107 cells/mL were incubated with substrata. This suggests that activation of mechanosensitive monovalent ion channels is not required of mechanoselective adhesion in these bacteria. ........................................................................................... 50 Figure 2.6. (0.10) ) (a) Wild‐type E. coli exhibit colony density (bars) that varies directly with the stiffness (symbols) of the PEM substrata. (b) Viable, spherical ΔmreB E. coli that lack the actin analogue mreB also adhere more readily to the stiffest substrata (E ‫׽‬ 100 MPa) than to the most compliant substrata (E ‫׽‬ 1 MPa). Scalebars = 1 mm. ............................................................................................................................. 52 Figure 3.1. (0.11) Airborne S. epidermidis (14990) challenges showing no antibacterial activity on (a) PAA/PAH films (labeled with the pH of deposition and the identity of the top layer), (b) SPS/PAH films assembled at pH 4.0 with 0.1 M NaCl (PAH as top layer) and (c) PAA/PAAm films assembled at pH 3.0 and thermally crosslinked (PAA as top layer). All samples statistically identical to glass controls (student’s t‐test, p<0.05). ........................................................................................................................... 63 Figure 3.2. (0.12) Schematic of acid‐activated opening of hydrophobic clusters comprised of amine groups in SPS/PAH 9.3/9.3 PEMs. The hydrophobic clusters can be reformed with subsequent treatment at high pH .................................................................................................................................................... 65 Figure 3.3. (0.13) Spray (airborne) antibacterial test results on glass and SPS/PAH 9.3/9.3 as deposited and acid treated samples for non‐biofilm forming S. epidermidis (ATCC strain 14490), biofilm forming S. epidermidis (ATCC strain 35984), E. coli ATCC strain 14948 and E. coli ATCC strain 700728. All acid‐
treated films are statistically distinguishable from glass and as‐deposited samples (student’s t‐test, p<0.05). ....................................................................................................................................................... 66 Figure 3.5.(0.15) Airborne antibacterial activity of SPS/PAH films 9.3/9.3 as deposited and acid‐treated after one week (“1 Wk”) and one month (“1 Mon”). ................................................................................. 68 Figure 3.4. (0.14) Airborne antibacterial tests with S. epidermidis (non‐biofilm forming strain) with films that have been cyclically pH treated to create swollen films (acid‐treated) and unswollen films (pH 11‐
treated). ...................................................................................................................................................... 68 7 | P a g e Figure 3.6. (0.16) Waterborne bacteria experiments on glass (circles), SPS/PAH 9.3/9.3 as‐deposited (squares) and acid‐treated (triangles) with (A) S. epidermidis 14990, (B) E. coli 14948, (C) S. epidermidis 35984 and (D) E. coli 700728. Insets in (A) and (C) are enlarged to show the differences between the as‐
deposited and acid‐treated samples. For all strains, colony growth was most efficiently reduced on SPS/PAH 9.3/9.3 acid‐treated samples. ...................................................................................................... 69 Figure 3.7. (0.17) QCM‐D data showing changes in frequency (black line) and changes in dissipation (grey line) to qualitatively illustrate swelling transitions in SPS/PAH 9.3/9.3 PEMs. Downward arrows point to times when the pH of the flowing water was changed. ............................................................................. 71 Figure 3.8. (0.18) Airborne antibacterial study with S. epidermidis (non‐biofilm forming) on glass, PAA/PAH 3.5/7.5 and 3.5/8.6 with PAA on top (denoted 3.5/7.5 PAA and 3.5/8.6 PAA, respectively) and PAH on top (denoted 3.5/7.5 PAH and 3.5/8.6 PAH, respectively). ........................................................... 75 Figure 3.9. (0.19) Airborne antibacterial assays on PAA/PAH films assembled at pH 4.0, with the addition of one, two or three bilayers of (PAA/PAH 4.0/8.6). All PEMs are PAH topped. ....................................... 76 Figure 4.1. (0.20) Nominal elastic moduli E as measured by instrumented nanoindentation of surface modified PAA/PAAm polyelectrolyte multilayers. PEMs were indented to a depth of 20 nm using a scanning probe microscope in fluid (150 mM PBS, pH = 7.4) at room temperature. ................................ 83 Figure 4.2. (0.21) AFM images of (PAA/PAAm)10.5 multilayers (a) as prepared, and (b) after immersion in 20 mM PAH solutions adjusted to pH 1.5, (c) pH 3.0, (d) pH 5.0, (e) pH 7.0, (f) pH 8.6, (g) pH 10.0, and (h) pH 11.3. All images are of dry films, 15 μm x 15 μm x 0.2 μm, with the z‐axis color scale given on the right. ............................................................................................................................................................ 90 Figure 4.3. (0.22) Plot of (PAA/PAAm)10.5PAH dry film thicknesses and dry‐state root mean square roughness values, both in nanometers, as a function of the pH of the PAH immersion solution. The gray dashed line indicates the thickness of the cross‐linked PAA/PAAm film before PAH adsorption. ............ 91 Figure 4.4. (0.23) Confocal fluorescence microscopy images of (PAA/PAAm)60.5 multilayers fabricated using Rhodamine 123‐ labeled PAA and FITC‐labeled PAH. IR filters show (a) bright field (b) FITC fluorescence and (c) Rhodamine 123 fluorescence in the honeycomb ridges. Magnification is 63x. ....... 92 Figure 4.5. (0.24) QCM‐D analysis of PAA/PAAm film buildup. All polymers were adsorbed at pH 3.0 with a concentration of 0.01M and a flow rate of 0.1 mL/min. Between each polymer deposition, rinse water at pH 3.0 was flowed through the system at 0.1 mL/min .......................................................................... 94 Figure 4.6. (0.25) FT‐IR data from (PAA/PAAm)10.5 films on ZnSe crystals. (A) Offset spectra illustrate the protonation of PAA’s carboxylic acid groups with increasing pH. (B) S‐curves showing the DOI of PAA (squares) calculated using FT‐IR data and equation 1, and the DOI of PAH (triangles) from the literature[16]. .............................................................................................................................................. 96 8 | P a g e Figure 4.7. (0.26) AFM images of 10.5 bilayer PAA/PAAm multilayers immersed in 20 mM, pH 8.6 solutions of (a) linear polyethylenimine, and (b) poly(vinylamine hydrochloride), accompanied by schemes of the respective polymers. ......................................................................................................... 99 Figure 4.8. (0.27) Bacteria experiments with S. epidermidis on PAA/PAAm films with PAH absorbed at various pH values showing (A) waterborne bacterial attachment and (B) airborne antibacterial results.
.................................................................................................................................................................. 100 Figure 5.1. (0.28) Schematic of biofilm formation: attachment of bacteria Æ proliferation and excretion of polysaccharides Æ mature biofilm releasing planktonic cells. ............................................................ 104 Figure 5.2. (0.29) Optical microscopy images of initial biofilm formation on (a) highly compliant PAA/PAH 2.0/2.0 PEMs and (b) rigid PAA/PAH 6.5/6.5 PEMs. ................................................................................. 111 Figure 5.3. (0.30) AFM images of initial biofilm formation on (A) aminoalkylsilane treated glass slide, (B) PAA/PAH 2.0/2.0 PEMs and (C) PAA/PAH 6.5/6.5 PEMs. All images are 10μm x 10μm. The amount of adhered bacteria was not affected by the underlying substrate. ............................................................ 111 Figure 5.4. (0.31) Representative optical microscopy images of initial biofilm growth on (A) plain glass, (B) SPS/PAH 9.3/9.3 as deposited PEMs, and (C) SPS/PAH 9.3/9.3 acid‐treated PEMs. ......................... 113 Figure 5.5. (0.32) (A) S. epidermidis colony densities on glass and releasable PEM substrates. Samples were challenged with airborne bacteria, then immersed for 30 minutes in PBS at 4°C or 37°C. Except for PEM 4°C, all samples were statistically indistinguishable (student’s t‐test, p<0.05). Representative optical microscopy images of colony growth on (B) plain glass at 4°C, (C) surface after PEM release at 4°C, (D) plain glass at 37°C, and (E) non‐released PEM immersed at 37°C. .......................................................... 114 Figure 5.6. (0.33) Optical microscopy images of biofilm growth on (A) glass surface after immersion in PBS at 37°C, (B) PEM surface after immersion in PBS at 37°C, (C) glass surface after immersion in PBS at 4°C, and (D) PEM surface after immersion and lift‐off in PBS at 4°C. Insets show the same sample before PBS immersion. All scalebars = 100 μm. ............................................................................................... 116 9 | P a g e List of Tables Table 2.1. PEMs used to test physicochemical and mechanical properties affecting bacterial attachment.
.................................................................................................................................................................... 42 Table 2.2. Surface tension components for the microbe‐water‐polymer system used to test physicochemical and mechanical properties affecting bacterial attachment. ........................................... 44 Table 3.1. Streaming Zeta Potential and Number of Attached Charged Microparticles for Representative PEMs. .......................................................................................................................................................... 74 10 | P a g e Acknowledgements
Many thanks to Professor Michael Rubner, my advisor, for his encouragement and
confidence in my abilities. His love of science and research is contagious, and my discussions
with him were always reenergizing. Professor Krystyn Van Vliet has also been an invaluable
resource, and I often sought her counsel as my unofficial second advisor. I have also appreciated
Professor Robert Cohen’s advice and suggestions over the years.
The results and success of my research rest on the contributions of various individuals:
most notably, Todd Thompson, Maricela Delgadillo, Allison Kunz and Kathy Tompkins. They
have all been true collaborators and research partners. Thanks to my office-mates, Erik
Williamson and Gary Chia, for creating a great work atmosphere and answering my various
questions. Throughout the years, all the members of the Cohen and Rubner groups have been
supportive and helpful colleagues.
Over the course of my graduate studies, I have been blessed with several furry, fourlegged cheerleaders: Jakie, Bella and Sugi. My intelligent, eclectic, boisterous collection of
aunts, uncles, cousins and grandmothers has always celebrated my successes, and I value their
support. But above all else: Mom, Dad, Shari, Simon and Eliad – Thank you. Obviously, I could
not have reached this point without you.
This thesis is dedicated to my parents, Audrey and Arlen Lichter, in appreciation of their
love, support, encouragement and commitment to my education.
11 | P a g e CHAPTER 1
1. Introduction
1.1 Introductory Remarks
Understanding the interactions between bacteria and surfaces is an area of great interest
and extensive research. Whether determining the surface parameters that promote or resist
bacterial attachment, maximizing antibacterial efficiency or combating the problem of biofilm
production, researchers have exploited a number of different surface systems including
hydrogels1, 2, surface grafted molecules3-5 and nanoparticles6, 7. Polyelectrolyte multilayers
(PEMs) offer a simple, water-based, versatile platform for exploring biomaterials with cell and
bacterial control functionalities. Polyelectrolyte multilayers can incorporate a broad range of
molecules, nanoparticles6, proteins8 and even cells9. In addition, the layers can be controlled on
the nanometer scale providing precise control over many material properties such as layer
thickness 10, modulus 11, surface charge, porosity12, 13, surface roughness10, etc. Depending on
assembly and post-assembly conditions, the films can also be stable in physiological conditions.
Thus, PEMs are an ideal tool for exploring the many aspects of cell and bacterial control.
The first part of this introduction provides a brief overview of polyelectrolyte multilayer
(PEM) synthesis. The second part summarizes various PEMs that have been engineered for
antimicrobial applications. The major contributions to the field of eukaryotic cell-PEM
interactions are also highlighted, often paralleling the strategies used to create antimicrobial or
bacteria-resistant PEM surfaces. Finally, a brief overview of the topics covered in this thesis is
presented.
12 | P a g e 1.2 General Introduction
1.2.1 Polyelectrolyte Multilayers
First developed in the 1990s in the Decher lab 14, polyelectrolyte multilayers (PEMs) are
built by dipping a charged substrate sequentially into water baths containing oppositely charged
polymers with rinse steps in between to remove excess polymer [Fig 1.1]. Some researchers use
spin-coating or spraying as a variation to the dipping procedure. As the concept of
polyelectrolyte multilayers gained popularity, it was expanded to include incorporation of many
charged species, including nanoparticles15, enzymes8, dyes16 and cells17. Multilayers are now
also built with hydrogen-bonding layers instead of electrostatically bound layers1.
Figure 1.1. (0.1) General setup of layer‐by‐layer dipping for PEM assembly. A substrate is sequentially dipped into aqueous solutions of oppositely charged species with rinse baths between each step. Charged polymers used for PEM assembly are classified as strong or weak polyelectrolytes
based on their ionizable charged groups. [Those polymers that maintain a fixed charged over a
broad range of pH conditions are termed “strong polyelectrolytes.” Polymers that exhibit pH13 | P a g e dependent ionization are called “weak polyelectrolytes.”] Since charged polymers adsorb to the
surface due to electrostatic attraction, the pH dependence on degree of ionization can be
exploited to vary PEM morphologies and material properties depending on the pH of assembly.
This is a key concept for much of the research discussed throughout this thesis. Examples of
strong and weak polyelectrolytes, as well as hydrogen-bonding polymers, are given in Figure
1.2. For the most part, the work performed in this thesis utilizes only these four polymers.
Figure 1.2. (0.2) Common polymers used in PEM synthesis: (A) weak polyelectrolyte poly(acrylic acid) (PAA), (B) polyacrylamide (PAAm), which can hydrogen‐ bond with PAA, (C) strong polyelectrolyte poly(sodium 4‐styrene sulfonate) (SPS) and (D) weak polyelectrolyte poly(allylamine hydrochloride) (PAH). PEM film thickness is highly tunable. Multilayer buildup can proceed in a linear or
exponential fashion, depending on the polymer system chosen. Bilayer thickness can be varied
14 | P a g e by manipulating assembly conditions such as pH and ionic strength. Total film thickness can be
adjusted by controlling the number of bilayers.
Due to the nature of the PEM assembly process, PEMs can be adsorbed onto a variety of
substrates with differing geometries. Glass slides, polystyrene tissue culture dishes, fabrics, and
PDMS are common substrates used in our group. Once assembled, PEMs are stable with respect
to moderate pH and ionic strength and ethanol treatment for sterilization. The common notation
used to describe assembled PEMs is
(Poly1 / poly2 pH1/pH2) x
Where poly1 and poly2 are the polymers used and pH1 and pH2 are the respective pH
conditions at which those polymers were adsorbed. x is the number of bilayers where one
bilayer consists of one adsorption of poly1 and one adsorption of poly2.
1.2.2
Antibacterial Multilayers
The versatile layerby-layer assembly
procedure has been used for
many biological
applications, including the
creation of antibacterial
surfaces. Many studies
highlight adhesion of
Figure 1.3. (0.3) Three general methods of creating antibacterial surfaces. planktonic bacteria to a
surface as the first step in
15 | P a g e bacterial infection or biofilm formation18. As with many antibacterial surfaces, engineers of
layer-by-layer coatings have approached bacterial attachment prevention in three general ways:
creating bacteria resistant surfaces, creating surfaces that leach antibacterial species and creating
contact-killing surfaces [Figure 1.3].
ADHESION RESISTANT PEMS: During the first step of the proposed two-step model of
bacterial adhesion, bacteria form relatively weak and reversible van der Waals, electrostatic and
hydrophobic bonds with the surface18. Bacteria are generally considered to be easily removed
during this step by rinsing, although there have been documented instances where shear stress
enhanced binding18. In the second step, bacteria form stable attachments to the surface via
specific ligands (e.g., adhesions on pili or fimbriae), and bacteria are brought within nanometers
of the surface18, 19.
Bacteria-resistant surfaces seek to prevent stable bacterial adhesion. Poly(ethylene
glycol) (PEG) adsorbed or grafted onto surfaces forms a highly hydrated layer that entropically
and enthalpically prevents protein adsorption, platelet adhesion, bacterial adhesion and cellular
adhesion due to its strong affinity for water molecules. Egles et al. have incorporated PEG into
polyelectrolyte multilayers by attaching PEG to the backbone of poly(L-glutamic acid) (PGA), a
negatively charged biopolymer. Multilayers topped with a one to three PGA-g-PEG bilayers
significantly reduced bacterial attachment of Escherichia coli even in the presence of nutrient
media20. Heparin, another polymer associated with reduced bacterial adhesion21, is a strong
hydrophilic polyelectrolyte that can be assembled into multilayers to effectively reduce bacterial
adhesion. The anti-adhesiveness of heparin in multilayers has been attributed to its high
16 | P a g e hydrophilicity22. Picart and coworkers have found that chitosan/hyaluronan PEMs assembled at
higher ionic strength (0.15 M NaCl) resist 80% of E. coli attachment compared to the 20-40%
reduction seen on films assembled at lower ionic strength (10-2 M NaCl). The difference in
bacterial attachment was attributed to either the increased film thickness or the decreased film
rigidity at higher ionic strength23. Note, however, that this study does not directly address the
possible contact-killing effects of chitosan discussed below. Hyaluronan (HA), a polysaccharide
that has been shown to reduce bacterial adhesion due to its hydrophilicity24, has been
incorporated into many multilayers. However, the bacterial resistance of HA in multilayers has
not been directly tested since it is usually combined with other bacteria-killing species23, 25-27.
The Hammond group has developed another tactic for resisting bacterial attachment:
hydrolytically biodegradable coatings that shed their surfaces by top-down erosion, preventing
stable bacterial attachment28.
BIOCIDE LEACHING PEMS: The most common biocidal leaching materials used in PEMs are
silver and silver ions6, 7, 15, 25, 26, 29-37. Metallic silver, a known bactericide since ancient times,
slowly releases ions and can interact directly with cell membranes in nanoparticle form38. Silver
ions act by binding to thiol groups on bacterial membranes, increasing cell membrane
permeability, and entering the cell itself, binding to DNA and preventing replication38, 39. Silver
is effective against a broad spectrum of bacterial strains while minimally affecting human cells40.
One strategy for incorporating silver into PEMs has involved assembling silver nanoparticle
precursors directly during the assembly process followed by reduction to form silver
nanoparticles7, 30-32. Silver nanoparticles themselves have also been assembled into multilayers
during the dipping process34. Other groups have designed PEM “nanoreactors” in which pre17 | P a g e formed PEMs with abundant free carboxylic acids bind Ag+ ions or other precursors that are
subsequently reduced6, 15, 29, 35, 36. Reduction has been accomplished by a variety of methods:
chemical reduction, heating and UV-irradiation7, 29. The size of the resultant nanoparticles has
been controlled by modifying pH, silver-precursor concentration or number of Ag+ loading
cycles6, 7, 32. Surface density and arrangement of silver nanoparticles has also been tailored by
varying the number of PEM bilayers and the number of silver reduction cycles29. Others have
controlled the nanoparticle concentration by using copolymers of different ratios to produce
different numbers of binding sites35. Another approach to include silver has involved using
silver-ion containing liposome aggregates embedded in a PEM support. By raising the
temperature above the transition temperature of the vesicles (~34° C), silver ions were released
that effectively killed E. coli populations25.
In addition to silver, antibiotics have been incorporated into PEMs. The Hammond group
created degradable PEMs that controlled release of unmodified gentamicin, an aminoglycoside
antibiotic28. By modifying the number of bilayers and the chemistry of the degrading polymers,
both the dosage and the release rate could be controlled. PEMs have also been used to
encapsulate and control release of the antibiotic ciprofloxacin hydrochloride41, 42.
Triclosan, a biocide which has been shown to block fatty acid synthesis in bacteria and
cause cell death, is a hydrophobic drug that is not easily incorporated into multilayers43. By
encapsulating the triclosan in biodegradable polymeric micelles and assembling the micelles into
a hydrogen-bonded multilayer, triclosan has been included in multilayers that dissolve in
physiological conditions44. The rate of film deconstruction could be tuned by varying the degree
of cross-links within the film. Triclosan has also been loaded into linear-dendritic block
18 | P a g e copolymer micelles and assembled into PEMs for prolonged release of active drug over a period
of weeks45.
Leaching quaternary ammonium compounds like centrimide have been included in
PEMs37. These species act by the “cationic killing effect” in which a cationic molecule is able to
bind to and permeabilize or lyse the negatively charged bacterial membrane.
A number of layer-by-layer studies have evaluated the uptake/incorporation and release
of dyes and model drugs in PEMs, but have not performed bacteria studies practically testing the
efficacy of these systems. For instance, pH-dependent hydrolytically degradable thin films
comprised of degradable polyesters and model anionic drugs have been created for controllable
drug release46. Drugs have been loaded into porous multilayers and released over the course of
several days13. In addition, low molecular weight hydrophobic model drugs have been
encapsulated in and released from
PEM hollow capsule shells47.
CONTACT KILLING PEMS:
Contact killing surfaces are
engineered to kill bacteria that
interact with the surface. Many
Figure 1.4. (0.4) Proposed mechanisms of cationic contact‐killing: A) Cationic surface charge displaces membrane counterions increasing the permeability of the bacterial membrane. B) Cations hold negatively charged bacteria at the substrate surface and hydrophobic tail penetrates the membrane, causing leakage. researchers have focused on
incorporating chitosan, a
naturally occurring biocompatible
cationic antibacterial
19 | P a g e polysaccharide that has been shown to disrupt bacterial cell membranes. Although many
mechanisms of action have been proposed, chitosan and its derivatives have been shown to bind
to the negatively charged bacterial cell membrane and cause leakage both in solution and
immobilized on surfaces48. In a multilayer, chitosan’s cationic charges are able to interact with
bacteria that attempt to adhere to the surface22, 27, 49-51. Some multilayer systems clearly require
chitosan as the terminating layer for the PEM to perform as an antimicrobial50. Assembly
conditions such as pH influence chitosan’s antibacterial properties in multilayers, possibly due to
differences in number of chitosan chain segments at the film surface22, 23. It is important to note
that chitosan in multilayers has been shown to be more effective at reducing bacteria on surfaces
than either a single-layer of adsorbed chitosan49 or covalently grafted chitosan50. When
compared to chitosan adsorption onto a corona-treated polypropylene film, a chitosan multilayer
was able to present a higher density of chitosan on the surface, increasing the surface’s
antibacterial activity.
Incorporating other known antimicrobial agents as one of the layer components is another
effective means of killing bacteria. The cationic antimicrobial protein hen egg white lysozyme
(HEWL) displayed contact-killing properties when applied as the outermost layer of a PEM8.
Polyguanidines, another class of biocidal polycations, have also been used as the cationic species
in electrostatic layer-by-layer deposition52.
Quaternary ammonium compounds (QACs) have been widely used to create nonleaching biocidal surfaces4, 53. The antibacterial activity of QACs combined with fatty alkyl
chains has been attributed to a “hole-poking” mechanism whereby the positive charge anchors
the molecule to the bacterial surface, and the hydrophobic chain penetrates the membrane and
causes leakage5 (See Figure 1.4). QACs have been grafted to the surface of PEMs and have been
20 | P a g e grafted onto other charged polymers used in the PEM assembly process15, 33. In agreement with
other studies showing that surface-grafted QACs are more effective against Gram positive strains
than Gram negative strains54, QACs grafted onto charged polyelectrolytes and assembled into
multilayers show higher levels of Gram positive cytotoxicity33. The decreased Gram negative
killing is attributed to the more complicated cell membrane structure of Gram negative bacteria.
Antibacterial peptides, short (26-46) sequences of amino acids that participate in innate
immune defense against microorganisms, have been immobilized on surfaces using the layer-bylayer technique. Although the mechanism is still unclear, these peptides are able to form pores in
the bacterial membrane and cause the cells to lyse55, 56. One of the major benefits of surface
immobilization of antibacterial peptides is the ability to create localized doses that are sufficient
to effectively kill bacteria. Egles and coworkers used defensin’s positive charge to layer the
peptide on top of negatively charged polyelectrolytes57. The authors note the versatility of this
technique for immobilizing antibacterial peptides: multiple layers incorporating antimicrobial
peptides easily increase the peptide concentration, and the electrostatic nature of the deposition
allows more than one kind of peptide to be incorporated. In this study, the PEMs were only
antimicrobial when the final layer was positively charged, indicating that positive surface charge
was necessary for negatively charged bacteria to adhere to the surface and interact with the
incorporated antimicrobial peptide. Glinel et al. have incorporated the hydrophobic antibacterial
peptide gramicidin A by complexing the peptide with an anionic amphiphilic polysaccharide,
thereby creating a charged species that could be involved in electrostatic layer-by-layer
deposition58.
Various other contact-killing species have been incorporated into PEMs. Titania, known
to create biocidal radicals upon UV-irradiation, has been built directly into multilayers for long21 | P a g e term antimicrobial surfaces59. Reduction of certain polyoxometalates (POMs) (clusters of
transition metals) in PEMs has also been thought to kill bacteria by oxidizing the bacterial cell
membrane and inhibiting growth51. PEMs built with single-walled carbon nanotube-DNA
dispersions as the anionic component and single-walled carbon nanotube-lysozyme dispersions
as the cationic component exhibited excellent mechanical properties (Young’s Modulus ~ 22
GPa, hardness ~ 1 GPa) and antibacterial capability when the lysozyme layer was on top60.
Lysozyme is an antibacterial enzyme that lyses the cell walls of Gram positive bacteria, but its
antimicrobial activity can be extended to Gram negative bacteria with the addition of chelators
such as ethylenediaminetetraacetic acid (EDTA).
COMBINING APPROACHES: Two antimicrobial strategies are often combined to combat
bacteria more effectively. For example, although hen egg white lysozyme (HEWL) containing
PEMs can act as contact-killers with HEWL as the outer-most layer, HEWL is also released from
the film over time and acts as a leaching biocide8. Likewise, PEMs with the antibacterial peptide
gramicidin A display both contact-killing and peptide leaching58. Shen et.al combined
antibacterial polycationic chitosan with antiadhesive polyanionic heparin and controlled the
degree of layer interpenetration and resulting surface properties by manipulating the pH of film
assembly22. They also created chitosan/heparin films with silver nanoparticles to include a
leaching-biocide32. Likewise, van der Mei et al. found that chitosan assembled into a multilayer
with carrageenan (a polysaccharide extracted from seaweed) could combat bacteria by contact
killing and by preventing adhesion due to the highly swollen structure of the multilayer50. Silverloaded PEMs topped with covalently grafted antibacterial quaternary ammonium salts have also
proved to be an effective “two-prong” approach to antibacterial coatings15. Surface-sloughing
22 | P a g e films that erode top-down have been combined with antibiotic release, both preventing
attachment and killing bacteria in solution28. Shen et al. has even assembled silver and titanialoaded films of chitosan and heparin, thereby combining much functionality into one coating59.
PEM SUBSTRATES: The consequences of bacterial infections plague many fields from
medicine to oil, food and water processing. As a result, PEMs need to be effective antimicrobial
coatings on a variety of substrates for different applications. Wang et. al has successfully
produced an antibacterial coating on Ti, a major component in orthopedic implants27. Others
have proved that many of the antimicrobial PEMs discussed above can be applied to
poly(ethylene terephthalate) (PET) (cardiovascular implant material)22, 32, 59, polypropylene (PP)
(food packaging material)49, poly(L-lactic adic) membranes (biodegradable biomaterial)31,
poly(tetrafluoroethylene) (PTFE) (a common antiadhesive surface)37, cellulose fibers (paper and
packaging material)52, silicon wafers51, 52 and indium tin oxide (commonly used in
electrochemical applications)51. Stainless steel and glass have also been used30.
Coating particles with antibacterial PEMs has created additional applications. Rubner et
al. coated magnetic colloidal particles that could be directed with a magnetic field to specific
locations6.
Antibacterial substrate-less PEMs have also been constructed. The Kotov group
developed silver-containing antimicrobial PEMs with excellent mechanical properties similar to
nacre or lamellar bone that could be made into free-standing films34. The Hammond group
created free-standing triclosan-loaded antimicrobial PEMs that enabled the performance of
otherwise difficult bulk characterizations such as differential scanning calorimetry and
23 | P a g e transmission electron microscopy44. Micron-sized hollow PEM shells loaded with silver and
goethite nanocrystals have been prepared to create antibacterial shells that can be directed with
with an external magnetic field35. The interior of these shells could be used as reaction sites,
providing confined areas for organic or inorganic materials. Hollow PEM shells have also been
engineered to contain antibiotics such as ciprofloxacin hydrochloride that are capable of
sustained drug delivery. Controllable amounts of drug have been loaded both into the PEM shell
(leaving the capsule hollow) and into the inner region of the hollow shell41, 42.
RESPONSIVE ANTIMICROBIAL PEM COATINGS: Stimuli responsive multilayers offer the
advantage of delaying their functionality until a specific stimulus activates the multilayer. As
mentioned above, incorporating temperature responsive liposome aggregates into a multilayer is
one way of achieving a stimuli responsive antimicrobial coating25. Multilayer literature is replete
with pH-tunable systems, including hydrogen-bonded PEMs that leach antimicrobial species
when raised to neutral pH44. By incorporating photocatalytic titania into PEMs, UV-light has
been utilized as the trigger for antibacterial activity59. Antimicrobial polyguanidines have been
assembled with temperature responsive polyanions to create films that undergo a morphological
transition upon heating52. However, in this case, the effects of the morphological transitions on
the efficacy of antimicrobial activity were not tested.
PEMs have been engineered to include a wide array of antimicrobial strategies. Due to
the versatility of the layer-by-layer technique, many different approaches have been proven to
successfully inhibit bacteria. The variety of biocides and bacteria resistant materials, substrata
and other incorporated species make antimicrobial PEMs suitable for many different applications
and a promising area to look for new antimicrobial surfaces.
24 | P a g e 1.2.3
PEMs and Cells
To fully understand the versatility of polyelectrolyte multilayers in biological settings, it
is helpful to review some of the work studying the interactions of PEMs with eukaryotic cells.
Throughout this section, “cell” refers to a eukaryotic cell. Many of the general themes in PEMcell studies directly relate to approaches used to control bacteria. For example, the schemes for
preventing cell adhesion described below parallel the studies for preventing bacterial adhesion.
Although most researchers aim to prevent bacterial adhesion, there are some situations (e.g.,
nitrogen fixation and bioremediation of wastewater) in which bacterial attachment and biofilm
formation are encouraged61. Thus, it is relevant to discuss PEM strategies to encourage cell
adhesion. Other aspects of PEM-cell studies such as cell patterning and cell coating offer insight
into the versatility of the layer-by-layer technique for controlling and manipulating live
organisms. When relevant, related studies with bacteria are also mentioned.
Control of cell adhesion has been investigated by a number of groups and the discussion
below is by no means exhaustive. The Rubner group has shown that highly swellable (or low
modulus) films, such as PAA/PAH assembled at low pH and the hydrogen-bonded PAA/PAAm
system, resist fibroblast and endothelial cell attachment 11, 62, 63. The same polymers assembled
in a tightly ionically cross-stitched manner are highly fibroblast adherent (e.g., PAA/PAH
6.5/6.5). Poly-(lysine)/alginate multilayers that form gels on highly adherent and heterogeneous
protein-covered biological surfaces have also successfully resisted fibroblast attachment64.
Similarly, others have developed chitosan/hyaluronan films that resist chondrocyte adhesion23.
The Schlenoff group has also shown interesting cell-adhesion results with varying PEM surface
properties. They have proven that higher hydrophobicity encourages muscle cell attachment and
25 | P a g e that with surfaces of equal hydrophobicity, negative charge is more cytophilic 65. Although they
were only using a small number of layers (2.5 bilayers at most), they did not include modulus or
swelling in their analysis. Note that cell adherence is very cell-type dependent and other cell
types prefer a hydrophilic surface 66, 67. The Schlenoff group also showed that muscle cell
attachment to PEMs does not correlate completely with protein attachment 68.
Many cell-resistant or cytotoxic surfaces have been coated with PEMs to improve
cytophilicity. Chondrocytes have been grown on thin films of poly-(ethylenimine) (PEI)/alginate
or PEI/poly(l-lysine) deposited on a relatively bioinert biodegradable substrate of poly-(DLlactide) 69. Both neurons and endothelial cells have been grown on PEM-coated silicone after
biomacromolecules were incorporated into the PEM to present a hydrophilic and cytophilic
surface 66, 67. Van Vliet et al. adsorbed one polycation layer onto cytophobic PAA/PAAm PEMs
and saw a large increase in film modulus accompanied by increased cell adherence, probably due
to polycation diffusion and cross-linking in the bulk of the film36, 70. Other groups have also
used both synthetic polymer and non-synthetic polymer PEMs to encourage or modulate human
endothelial cell growth 71, 72. After being coated in collagen/PAA multilayers, cytotoxic heavy
metal nanoparticles were successfully incorporated into multilayers that showed high levels of
muscle cell attachment and survival 73.
PEMs have been used in many other interesting cell-adhesion applications. Free-standing
multilayer films incorporating single-walled carbon nanotubes have been shown to encourage
neuronal differentiation and attachment 74. PEM coatings in channels have been shown to align
smooth muscle cell growth 75.
26 | P a g e Initial cell coating research focused on using cells as templates to create uniquely-shaped
hollow microcapsules. Red blood cells and E. coli bacteria have been coated with PEMs for this
purpose76, 77. However, these studies all examined coatings on fixed (and therefore dead) cells.
PEM coating of live cells is of great interest for applications such as protecting cells from
an immunological response, functionalizing and directing the outer surface for drug delivery,
creating microenvironments for cell studies, tissue engineering, etc. Coating bacteria could be
relevant for bio-sensing and drug delivery. Germain et al. has directly coated live adherent
mammalian cells with polyelectrolytes and has proven that the coating preserves metabolic
functions and allows the passage of small molecules while isolating the cells from high
molecular weight molecules 78. Rajagopalan et al. created three layer stacks by sequentially
adsorbing one PEM bilayer and then culturing hepatocytes to create layered cell structures 17.
Living yeast cells and E. coli in solution have also been coated without affecting metabolic
activities or cell-division 79, 80. Small PEM patches have been attached to lymphocytes, creating
the possibility for attaching additional functionalities or tags to cells without completely
concealing their native surfaces81.
Recently, attempts to coat pancreatic islets with PEMs have revealed interesting
information about the linear charge density necessary to promote polycationic cytotoxicity82.
Limiting cation spacing along the chain allowed polycations to create pores in the cell membrane
and cause cell death. As will be discussed in later chapters, linear charge density is a valuable
parameter to consider when engineering polycationic antibacterial surfaces.
The above results generated developments in cell patterning for mammalian cells. The
soft lithography techniques developed by the Hammond lab are extremely popular 83. After
27 | P a g e creating a cytophobic surface, adherence promoting polymers or peptide units have been
polymer-on-polymer stamped to create a cytophilic pattern 65, 84, 85. After coating live yeast cells
in PEMs with specific outer layer charges, polymer-on-polymer stamped surfaces selectively
adhered the cells in patterns 86. Using microcontact printing, PEM-coated microspheres were
adsorbed onto bare silicone rubber to control cell patterns 66.
Other patterning methods for cell-attachment studies have been developed that do not
involve polymer-on-polymer stamping. McShane has used a lithography/lift-off technique
termed polymer surface micromachining (PSM) to create complex interdigitated patterns on a
cell-resistant background that selectively adhere neurons 87, 88. Microfluidic networks have also
been used to pattern PEMs on PDMS to create cell-adhesive mirco-networks to which retinal
cells adhere 89.
PEM patterns have led to the development of cell co-cultures which better mimic the invivo cellular environment. The Langer group has developed one method for co-cultures with the
following steps: 1) Hyaluronic acid (HA) micropatterns were deposited on glass. 2) Fibronectin
was bound to the exposed glass. 3) Cell type A bound preferentially to the fibronectin. 4) Poly-Llysine (PLL) was adsorbed onto the HA. 5) Cell type B bound to the PLL layer. In this way,
they successfully developed patterned co-cultures of hepatocytes, fibroblasts and stem cells 9.
As can be seen from the discussion above, many routes have been explored to manipulate
cell-PEM interactions. These approaches often parallel studies on antimicrobial PEMs. Taken
together, PEMs offer a rich and versatile platform for investigating and controlling living
organisms.
28 | P a g e 1.3 Thesis overview
The ultimate objective of this thesis is to exploit the versatility of polyelectrolyte
multilayers to gain insight into the surface properties and processing techniques that control
bacterial cell attachment and antimicrobial functionality. In Chapter 2, the use of PEMs of
varying mechanical modulus engineered by varying the pH of polymer assembly is explored.
Once characterized, these films were used to investigate the effects of surface mechanical
modulus on bacterial attachment. Chapter 3 details the use of a model PEM system to outline
design parameters to create contact-killing antimicrobial functionality in PEMs. Chapter 4
describes the mechanisms of post-assembly changes in a hydrogen-bonded multilayer. The
various surfaces are tested for antiadherent and antimicrobial properties, and reinforce the
concepts discussed in Chapters 2 and 3. In Chapter 5, the PEM surfaces developed in earlier
chapters are tested for their ability to resist initial biofilm formation under high nutrient
conditions. Also, a new surface-sloughing PEM system is introduced that may better combat
aggressive biofilm formation.
29 | P a g e CHAPTER 2
2. Regulation of Bacterial Adhesion by Manipulating Polyelectrolyte Multilayer
Stiffness
This work is the result of an extensive collaboration with the Van Vliet laboratory at MIT. M.
Todd Thompson, a graduate student in the Van Vliet group, was responsible for the
nanoindentation measurements to determine modulus and the force microscopy to determine
surface charge. With contact angle measurements that I supplied, Thompson also calculated
surface energy components. Bacteria experiments were done in partnership. Maricela
Delgadillo, an MIT undergraduate student, assisted in bacteria experiments and image analysis.
This chapter has been reproduced in part from:
Lichter, J. A., Thompson, M. T., Delgadillo, M., Nishikawa, T., Rubner, M. F., and Van Vliet, K.
J., Substrata stiffness can modulate adhesion of viable bacteria. Biomacromolecules, 9 (10),
1571-1578, 2008.
2.1 Introduction
The design of functional materials to control the formation of biofilms, structured
communities of bacteria protected by a polysaccharide matrix, has been the subject of numerous
research efforts. Hospital acquired infections represent an estimated $4.5 billion cost90 with an
associated annual mortality of 100,000 persons in the US alone.91 The commensal bacterial
species Staphylococcus epidermidis is the most common agent of infection,90, 92 exceeding the
infection rate of methicillin-resistant Staphylococcus aureus (MRSA), with virulence often
attributed to initial attachment of a viable bacterial population to the surface93 of a medical
device and subsequent formation of a mature biofilm.
Current approaches to limit bacterial colonization have focused on chemical degradation
of stably adhered bacteria, including surface functionalization with microbicidal agents;53, 94, 95
surface impregnation with slow releasing biocides such as gold or silver15, 95-97 and antibiotics;94,
98
or surface functionalization with specific antimicrobial peptides and polymers.53, 99 Because
30 | P a g e biofilm formation requires the initial attachment of a viable bacteria population on a surface,93
another promising approach to limiting microbial colonization is prevention of bacterial adhesion
to material substrata prior to colonization. Others have reported that poly(ethylene glycol)conjugated polypeptides confer adhesion resistance and suggested that such results may be due
to high degrees of substrata surface hydration20; this speculation was not systematically tested.
More generally, the development of a versatile and comprehensive approach to reduce stable
bacterial adhesion to surfaces has been limited by incomplete understanding of the regulating
physicochemical material properties. Physical characteristics such as surface roughness do not
appear to impact bacterial adhesion consistently, with some studies reporting reduced adhesion
of S. epidermidis to smoother surfaces100 and others finding no conclusive correlation.90, 100-104
Material surface charge and/or hydrophobicity have been reported to be crucial during the
primary, kinetic step of adhesion.18, 105-107 However, several studies have reported no correlation
between microbial adhesion and hydrophobicity,101, 102 and claimed the presentation of surface
functional groups capable of charge transfer (Lewis acid/base character of the surface) as the
critical surface factor governing bacterial adhesion.108-110 As the specific interactions among
bacteria, solvents, and substrata can each contribute independently to the efficiency of adhesion,
others have claimed the dominant factor to be total interaction energy between the microbe,
liquid media, and substrata material, usually expressed as the work of adhesion.110-112 Beyond the
potential strain-dependence and the convolution of competing factors such as surface roughness
and surface energy in the synthetic surfaces considered, such contradictory results may indicate
unrecognized surface properties that modulate bacterial attachment. Here, we engineer material
surfaces of measured surface roughness, charge density, interaction energy, and elastic moduli to
consider whether the mechanical compliance of the surface, now widely appreciated to modulate
31 | P a g e the adhesion and function of eukaryotic cells,62, 113, 114 may also regulate bacterial adhesion to
underlying substrata.
S. epidermidis, a spherical Gram positive (G+) microbe, was the
predominant bacterium type in our study. This microbe serves as an established model for
bacterial attachment,115 as well as a common cause of medical-device related and hospitalacquired infections.90, 92, 116 Strategies for prevention of initial colonization and infection with S.
epidermidis are clinically important, as these strains are increasingly resistant to antibiotics117. E.
coli, a rod-shaped Gram negative (G−) microbe, and its actin analogue mutant form ΔmreB E.
coli were also tested to consider whether our observations were limited to only specific strains,
cell shapes, or Gram-stain classes.
To vary the physicochemical and mechanical properties of the substrata, we employed a
class of synthetic polymer thin films termed weak polyelectrolyte multilayers (PEMs)
comprising the polyelectrolytes poly(allylamine) hydrochloride (PAH) and polyacrylic acid
(PAA). The chemical functionality and mechanical compliance of such films can be adjusted by
simple
variations
of
the
layer-by-layer
assembly
conditions
such
as
choice
of
polyanion/polycation or assembly pH,114 and can be applied to a wide range of surfaces requiring
biofilm prevention including polymers, glasses118, and metals.99 The effective elastic modulus E
or stiffness of such hydrated films under in vitro culture conditions can be varied over several
orders of magnitude by manipulation of assembly pH.114 We and others have shown that this
substrata stiffness modulates tissue cell adhesion independently of physicochemical
characteristics such as adhesive ligand density.62, 113, 114, 119 Through extensive characterization
of these tunable polymeric substrata, we demonstrate that S. epidermidis exhibit
mechanoselective adhesion. As a result, bacterial colonization can be significantly reduced by
32 | P a g e modulating the compliance of the material substrata, independently of short and long-range
physicochemical properties of the cell-material interface.
2.2 Materials and Methods
2.2.1 PEM Assembly.
Polyelectrolyte multilayers were assembled as previously described in a layer-by-layer
(LbL) automated assembly of alternate dipping into polycation/anion solutions.10 10-2 M
solutions in 18 MΩ Milli-Q water of PAA (poly(acrylic acid); Mw = >200 000 g/mol; 25%
aqueous solution; Polysciences (previously labeled as Mw = 90 000 g/mol)) or PAH
(poly(allylamine hydrochloride); Mw = 70 000 g/mol; Polysciences) were pH adjusted using 1M
HCl and NaOH.
Multilayers were assembled PAA first on aminoalkylsilane coated glass
(Sigma-Aldrich) or, for Fig. 2.1 only, on medical grade titanium (ASTM F67, President
Titanium, Hanson, MA). Notation refers to the assembly conditions with the PAA pH followed
by the PAH pH, i.e., a 3.5/8.6 PEM was assembled using PAA at pH 3.5 and PAH at pH 8.6. All
PEMs were prepared to dry thicknesses of ~50 nm, which required variation in the total number
of layering (dipping) cycles at each assembly pH. The samples included 2.0/2.0 (9.5 bilayers),
4.0/4.0 (7.5 bilayers), 6.5/6.5 (49.5 bilayers), 3.5/7.5 (5.5 bilayers) and 3.5/8.6 (5.5 bilayers).
The following samples studied the effect of masking underlying PEM substrata in Fig. 2.4:
6.5/6.5 (50 bilayers; PAH topped) plus 0.5 bilayer of pH 2.0 PAA; 6.5/6.5 (49.5 bilayers) plus
one bilayer of 2.0/2.0; and 2.0/2.0 (9.5 bilayers) plus pH 6.5 PAH. The propensity for assembly
pH-modulated extent of ionic crosslinking in these weak PEMs differs significantly from strong
PEMs used by others in studies for aspirated de-adhesion of eukaryotic cells120.
33 | P a g e 2.2.2 PEM Elastic Moduli.
The effective elastic moduli E, as determined from nanoindentation force-displacement
responses acquired from an atomic force microscope (3D Molecular Force Probe, Asylum
Research, Santa Barbara, CA), were quantified as previously described.114
Silicon nitride
cantilevers (MLCT-AUHW, Veeco Metrology Group, Sunnyvale, CA) were used to indent
PEMs to maximum depths of <20 nm with a threshold filter to maintain equal loads for each
indentation. The probe radius of curvature Rp was ~50 nm; cantilever spring constant k was
nominally 0.1 N/m and was experimentally determined for each cantilever.121 Nanoindentation
was performed in an acoustic isolation enclosure (Herzan, Inc.) at room temperature in 0.2 μmfiltered PBS or Milli-Q water. Nanoindentation force-depth data were analyzed in IGOR
(Wavemetrics, Lake Oswego, OR) and E was determined according to a modified Hertzian
contact model.114
Multiple batch runs of the pH assembled PAA/PAH samples were tested on different
days to identify any sample-to-sample variation and systematic experimental errors. Forcedisplacement data were processed prior to analysis using a 25 pass binomial smoothing filter to
eliminate random fluctuations. Each force curve was then visually inspected and aligned against
a representative force curve from the same experiment, whereby the force and separation data
were zeroed by overlaying the corresponding loading regions using a free-scale, non-rotatable
coordinate system. Since accurate determination of the initial contact point is a critical issue in
nanoindentation of compliant polymer films, an additional noise threshold was applied to the log
P – log Δ representation of smoothed curves to identify this (0,0) point objectively and
repeatedly. Linear least-square fits of the log P – log Δ representation of smoothed responses
34 | P a g e were conducted and yielded intercept values from which nominal E were calculated from Eq. (1),
as previously described 122
P = (4/3) Rp Er1/2 Δ3/2
(2.1)
where P is applied force, Rp is the radius of curvature of the cantilevered probe, Er is the reduced
elastic modulus, and Δ is the depth of penetration into the sample surface
123
. For samples of
compliance much greater than the indenter material (as in the present case), E = Er.
2.2.3 Substrata Surface Energy and Interaction Energy.
To determine the total interaction energy and surface tension components of the
substrata, liquid contact angles were measured for the polar solvents water and ethylene glycol;
and the apolar solvents hexadecane and diiodomethane.
Five to ten measurements were
performed on each sample using the sessile drop technique, and contact angles recorded for
static, advancing, and receding drops. Contact angles were measured using a camera-equipped
Advanced Surface Systems machine.
Liquid contact angles were used to determine
thermodynamic properties of the surface-bacterial cell-liquid interface according to the Lewis
acid-Lewis base theory of Van Oss.110 Using the Van Oss approach, liquid contact angles of
three or more test solvents are measured and then the nonlinear Van Oss-Young equations are
then solved simultaneously:
(1 + cosθ) γtotL = 2√(γLWS γLWL) + 2√(γ+S γ-L) +2√( γ-S γ+L)
35 | P a g e (2.2)
and
γtoti = γLWi + γABi
(2.3)
γABi = 2√(γ+i γ-i)
(2.4)
where γtoti is the total surface tension of a material i, γLWi is the apolar component of the surface
tension, γABi is the polar component, and γ+i, and γ-i are the electron acceptor (Lewis acid) and
electron donor (Lewis base) properties of the material. Van Oss-Young’s equation is solved
simultaneously to yield the apolar, Lewis acid, and Lewis base components of the interfacial
tension. The total interaction energy between the bacterial cell and the substrata material is then
given by:
ΔG = (γLW1 + γLW2)2 – (γLW1 + γLW2)2 –( γLW1 + γLW2)2 +2[ √γ-W (√γ+1+√γ+2 –√γ+W ) +
√γ+W (√γ-1 + √γ-2 – √γ-W ) – ( √(γ+1 γ-2)- √( γ-1γ+2) )]
(2.5)
To solve this nonlinear equation, total surface tension of the bacterial surface was
assumed from a representative slime producing strain of S. epidermidis (RP62A/ATCC 35894);
the bacterial components of this equation act as constants, and therefore do not affect
interpretation of trends. However, additional clinically relevant strains were examined using the
same technique and were qualitatively similar (data not shown), further indicating that interfacial
energetics as described by the current formulation of Young’s theory do not adequately address
the observed effects.
36 | P a g e 2.2.4 Substrata Surface Charge Density.
Surface charge density Q was analyzed for PEMs 2.0/2.0 and 6.5/6.5 via AFM force
spectroscopy (3DMFP, Asylum Research), using cantilevered carboxylic acid-functionalized
polystyrene spheres of approximately 3 μm-radius (BioForce Nanosciences, Ames, IA; nominal
k ~ 0.1 N/m) Force-distance curves were first acquired in deionized water using a test surface
comprised of mercapto-undecanoic acid (MUA) functionalized gold surface with calibrated Q124,
125
of Q = -18 mC/m2, from which the Q of the colloidal probe was calculated using models
adapted from Rixman et al.;
124, 125
Force-distance curves were acquired for experimental
samples in deionized water (ionic strength I = 0.0027 ) or 0.1 M NaCl (I =0.1) after an overnight
thermal equilibration of the surface and cantilevered probe within the AFM. The maximum
deflection of the cantilever on approach to the sample surface was maintained constant via a
closed-loop algorithm supplied by Asylum Research. All sample locations were measured a
minimum of twenty times per approach cycle, over 5-10 locations per surface.
Curves
representative of the data set were generated by alignment of the contact point, defined as the
beginning of the region of constant compliance, followed by statistical averaging of the
respective force and separation curves for a given approach cycle. The resultant curve for each
surface location was the average force detected by the average approach vector normal to the
sample surface. Measurements that did not possess a region of constant compliance were zeroed
by examining for a jump-to-contact region, followed immediately by cantilever deflection; data
acquired in I = 0.1 M solvent were compared to data acquired in Milli-Q water at the same
distance from the surface to determine where physical deflection occurred. Representative
curves were then used for modeling the electrostatic surface charge density to a distance within 5
37 | P a g e nm of the calculated contact point, and surface charge density calculated from a least squares fit
of the model to the data 124, 125.
2.2.5 Methylene Blue Staining
Prepared PEMs were immersed for 15 min in 0.005 M solutions of methylene blue in
deionized water (pH 7). The PEMs were then rinsed in two clean water baths for 2 min each and
dried with an air gun. Incorporation of methylene blue into the PEMs was measured by UV-Vis
spectroscopy (λ = 450 – 700 nm) with peak intensities at λ ~580 nm126.
2.2.6 Elimination of Free Carboxylate Charge
10 % (v/v) ethanol (VWR) added to 0.1 M 1-ethyl-3-(3-dimethylaminopropyl)-1carbodiimide hydrochloride (EDC) (Advanced ChemTech) was adjusted to pH 6.
PEMs
assembled at pH 4.0 were immersed in the solution overnight, followed by quenching of the
reaction with 0.1% (v/v) 2-mercaptoethanol (Sigma) for 3 hours. Samples were rinsed well with
deionized water and dried under compressed air.
2.2.7 Bacterial Attachment Assays.
Waterborne bacterial attachment assays were adapted from the protocol of Tiller et al 53.
Briefly, Miller Luria-Bertani or LB-Miller broth (VWR) was inoculated with a monoclonal strain
of Staphylococcus epidermidis (S. epidermidis, ATCC # 14990), E. coli (w3100 strain of welldocumented lineage ATTC #14948), or E. coli mutant strain ΔmreB, using a sterile loop and
incubated overnight at 37oC while shaking. Two 50 mL aliquots of culture were centrifuged at
2700 RPM for 10 min at 4oC, the LB broth decanted, and the bacterial cell pellets resuspended in
150 mM NaCl PBS (VWR). Following resuspension, the cells were centrifuged twice (5 min,
38 | P a g e 2700 RPM) to ensure complete removal of LB broth, with a final resuspension in 18MΩ
Millipore water. The resuspension was serially diluted with water from 109 cells/mL (measured
via optical density) to create suspensions of 103 - 108 cells/mL. Studies in water were conducted
at 107 cells/mL for S. epidermidis and 0.5 × 107 cells/mL for E. coli due to overall higher
adhesion efficiency of E. coli. Samples (in triplicate for each condition) were placed in the
bacterial solutions for 2 h at room temperature followed by agitation in three water bath rinses,
each for ‫׽‬5 s. Samples were incubated under 1% LB agar (VWR) gel overnight, and colonies
counted to determine the ability of viable bacteria to attach to each sample. Initial adhesion
assays in PBS (Figure 2.2) were identical to those in water, except that final resuspensions
occurred in PBS and incubation periods occurred at 37 °C with shaking; note that the salt
titration assay of Figure 2.5 was conducted at room temperature without agitation. Samples with
few colonies were hand counted. For densely populated slides, at least 10 digital images per
sample were acquired with a 4x objective using an inverted optical microscope (Leica), and
semi-automated image analysis was conducted. To determine colony size and density
(#colonies/cm2 of substrata), image analysis was conducted using ImageJ (Rasband, W.S.,
ImageJ, National Institutes of Health, Bethesda, Maryland, USA, http://rsb.info.nih.gov/ij/, 19972004) as bundled by the Wright Cell Imaging Facility (http://www.uhnresearch.ca/facilities
/wcif/download.php).
Images were converted to 8-bit binary format and flattened using a
pseudo-flatfield filter to normalize the luminescence across the image field. When necessary, the
ImageJ standard watershed algorithm was applied to separate intersecting colonies, thus
facilitating more accurate colony counts. Colony densities were determined by normalizing the
mean colony number per image by the calibrated total image area. Statistical analysis was
39 | P a g e conducted using the Tukey-Kramer multiple-comparisons post-test analysis of variance
(ANOVA).
2.3 Results and Discussion
2.3.1. Bacterial colonization can be reduced by
material substrata modifications.
Although the competing mechanisms remain
unclear, a large body of data suggests that both
physical and chemical modifications of a material
surface can be engineered to limit bacterial
Figure 2.1. (0.5) PEMs reduce bacterial adhesion on medical grade titanium. Adhesion of waterborne S. epidermidis is reduced by coating with a pH‐tunable polyelectrolyte multilayer (PEM) film of PAA and PAH assembled at pH 2.0, and is stable at both 2 h (inset; circle indicates one such colony) and 4 h incubation duration. Scale bars = 5 mm.
colonization.15, 53, 94-96, 98, 99, 104, 105 For example, as
shown in Fig. 2.1, coating a surgical-grade titanium
alloy with a synthetic PEM of PAA and PAH
reduced
colony
density
of
waterborne
S.
epidermidis bacteria by orders of magnitude after
immersion in 107 bacteria/mL. Reduced colonization over both 2 h and 4 h incubation timescales
is relevant to medical procedure durations such as cardiac assist and orthopedic implant
devices127. This PEM was ionically crosslinked through layer-by-layer dipping of the titanium
into polycation and polyanion solutions at pH 2.0 (see Methods) prior to full hydration and
equilibration in sterile deionized, distilled water.
40 | P a g e 2.3.2.
Weak
polyelectrolyte
multilayers
modulate stable adhesion of s. epidermidis
bacteria.
We varied the assembly pH of the PEM
substrata to consider how such modifications
might affect S. epidermidis colonization of these
surfaces. Assembly of these weak PEMs at pH
2.0 results in a substrata of much lower stiffness
(effective elastic modulus E ~ 1 MPa) than at pH
6.5 (E ~ 100 MPa).114 As shown in Fig. 2.2a, for
a 2 h incubation of substrata in seeding
concentrations
Figure 2.2. (0.6) Bacterial colonies observed for 103 –
108 S. epidermidis/mL in 150 mM NaCl PBS. (a) Average colony number per unit substrata area increased with increasing incubation concentration for greater than 105 cells/mL; for all concentrations, the density of colonies observed on the PEM substrata assembled at pH 6.5 (●) was significantly greater than that observed on the substrata assembled at pH 2.0 (▼). (b) For given initial concentration, colony number was greater and colony size was smaller on stiffer substrata, supporting a model whereby bacteria attachment is modulated in part by substrata stiffness, but subsequent growth is affected predominantly by available space and nutrients. Scalebars = 500 mm. ranging
from
103
to
108
bacteria/mL of 150 mM NaCl phosphate
buffered saline, average colony density (number
of colonies per unit substrata) was greater on
mechanically stiffer substrata.
For a given
seeding concentration, the average colony size
observed after 24 h culture was also much
greater on the more compliant substrata; this
suggested that the properties of these substrata
affected
bacterial
adhesion
and/or
colony
growth. Figure 2.2b suggests that the observed differences in colony density occurred at the
adhesion step: colony size depended on colony density for both substrata. In other words, the
41 | P a g e initial bacterial attachment increased with increasing substrata stiffness, but the subsequent
colony growth was likely limited by available space and nutrients post-adhesion.
2.3.3. Characterization of polymeric substrata properties.
To consider the characteristics of the polymer substrata that directly affect attachment of
S. epidermidis, we conducted a larger study in deionized water to eliminate possible charge
shielding and reorganization of the ionic crosslinks within the PEM substrata in salt solutions.
For the substrata considered, we quantified the mechanical compliance and the physicochemical
surface properties considered to affect microbial adhesion. Table 2.1 indicates physicochemical
and mechanical characteristics of substrata employed in the larger study. PAA and PAH were
adjusted to the same pH (e.g., PAA/PAH 2.0/2.0) as well as to different pH (e.g., PAA/PAH
3.5/7.5) during assembly in order to increase the range of substrata properties. Atomic force
microscopy (AFM) imaging of hydrated substrata in tapping mode indicated a range of root
mean square (RMS) surface roughness from 3 to 30 nm. AFM-enabled nanoindentation of the
PEMs hydrated in deionized water indicated an average elastic modulus E
Table 2.1. PEMs used to test physicochemical and mechanical properties affecting bacterial attachment. Assembly pHa
(PAA/PAH)
2.0/2.0
6.5/6.5
3.5/7.5
3.5/8.6
a
Symbolb ΔGMWP (mJ/m2) c,d RMS Roughness (nm) d,e
▼
●
♦
■
29.0 ± 7.5
27.2 ± 8.95
27.2 ± 8.0
27.0 ± 6.9
30.2 ±29.5
2.7 ± 1.6
12.2 ± 9.0
18.5 ± 16.6
E (MPa) d,f
0.75 ± 0.05
80.4 ± 38.0
36.6 ± 5.7
73.2 ± 16.6
Assembly pH of polyanion and polycation indicated, respectively, for PEMs assembled to ~50 nm dry thickness (≥ 57 nm hydrated thickness) with PAA as the last layer. bSymbols used throughout to indicate the corresponding PEM in all figures. cTotal interaction energy ΔGMWP of the microbe‐water‐polymer system. dAll data expressed as average ± standard deviation. eRoot mean square (RMS) surface roughness. fNominal elastic moduli E. 42 | P a g e ranging over two orders of magnitude from the stiffest PEMs assembled at pH 6.5 (E = 80.4
MPa) to the most compliant PEMs assembled at pH 2.0 (E = 0.8 MPa), consistent with our
previously reported mechanical characterization of these PEMs in 150 mM NaCl phosphate
buffered saline (1x PBS).114
Surface energies of interaction were calculated according to Van Oss’ adaptation of
Young’s theory,110 which correlates the interfacial tension and surface energy of interaction
between materials in a solvent. Four solvents of disparate surface tension and polarity were used
(see Methods). The apolar and polar components of this surface tension relate to the Lifshitz-van
der Waals and Lewis acid-Lewis base (charge transfer) character of each sample, respectively;
both interactions are thought to influence bacterial adhesion.110 Thermodynamic properties at the
PEM-liquid interface have been characterized using the Van Oss approach to describe the
assembly process, but to our knowledge have not been applied in the context of microbe-waterPEM (MWP) interactions128. The surface interaction energy ΔGMWP for all PEM substrata
considered narrowly ranged from 26-29 mJ/m2 and were statistically indistinguishable (see Table
2.2 for the component determinants of ΔGMWP).
43 | P a g e Table 2.2. Surface tension components for the microbe‐water‐polymer system used to test physicochemical and mechanical properties affecting bacterial attachment. PEM
Assembly pH
(PAA/PAH)
Symbol
gTot
(mJ/m2)
gAB
gLW
g+
g-
2.0/2.0
▼
48.5
25.1
23.3
48.5
25.1
6.5/6.5
●
47.3
21.5
25.8
47.3
21.5
3.5/7.5
♦
47.4
20.5
26.9
47.4
20.5
3.5/8.6
■
47.3
20.1
27.2
47.3
20.1
s. epidermidis
52.7
17.2
35.5
52.7
17.2
Water
72.8
51.0
21.8
72.8
51.0
Ethylene glycol
48
19
29
48
19
diiodomethane
50.8
0
50.8
50.8
0
hexadecane
27.5
0
27.5
27.5
0
Components were determined by analyzing the contact angles of several solvents according to Young’s equation (see Methods); or were obtained from reported values in the literature. Data expressed as (mJ/m2) are relative to standard assumed values for water. Symbols are used to indicate PEMs corresponding to those found in Table 2.1. Initially, we attempted to assess the extent of negative charge present at the fluid-PEM
interface as indicated by free carboxylic acid groups by staining the PEM samples with the
common cationic aqueous dye, methylene blue (MB). In PAA based PEMs, MB incorporation
occurs due to interaction with “free” negatively charged carboxylate residues along the PAA
polymer chain that are readily accessible to soluble MB. PEMs assembled at pH 2.0 and 3.5/7.5
had maximum MB absorbance peaks of 0.6 and 0.24 absorbance units, respectively, while the
absorbance of PEMs assembled at pH 6.5 and 3.5/8.6 was indistinguishable from the
background. However, it is known that MB blue diffuses somewhat into the bulk of PEMs, and
44 | P a g e therefore, MB absorbance is not an ideal measure of the surface charge that interacts with
microbes at the outer PEM surface126.
A number of options were considered to isolate the effects of potential differences in
surface charge on bacterial attachment. In one case, the free carboxylic acids in the moderatelyswelling pH 4.0 system were “capped” with ethanol via a carbodiimide reaction. After capping,
the pH 4.0 system did not stain visibly with MB, similar to the pH 6.5 and pH 3.5/8.6 systems.
This implied that most of the free carboxylic acid groups had been reacted with ethanol. Initial
modulus measurements indicated that the modulus after capping increased only slightly (from
~10 MPa to ~15 MPa). Although possibly a viable avenue for eliminating surface charge,
capping the PEMs introduced new chemical species (ethanol groups and traces of reactants and
catalysts).
Because the multilayer system was chosen as a platform to examine surfaces
composed of identical chemical components, we preferred exploring more direct measurements
of surface charge to determine if it indeed varied in the different PEMs used here.
As an alternative, net surface charge density Q present at the fluid-PEM interfaces of the
substrata assembled at pH extremes of 2.0 and 6.5 was directly assessed by probing the PEMs in
deionized water using a carboxylated colloidal sphere approximately the size of a few bacteria (3
μm radius; see Methods). As PEM assembly relies on charge over-compensation to increase
substrata thickness, one might expect the observed net-negative Q because the anionic polymer,
PAA, was layered last. However, it is important to note that although these polymeric substrata
are termed multilayers due to the layer-by-layer assembly process, the structure is not striated
and the polyanion and polycation macromolecular chains are highly entangled. Charge densities
of PEMs assembled at pH 2.0 and 6.5 were well within one standard deviation (Q = -2.29 +/- 0.1
mC/m2 and -3.18 +/- 1.4 mC/m2, respectively); see Fig. 2.3d. Q was unchanged in solutions of
45 | P a g e higher ionic strength such as 150 mM NaCl PBS, although charge is effectively screened in such
ionic solutions. Thus, although MB staining indicates differences in carboxlate density
throughout the volume of the film, the net surface charge densities of the pH 2.0 and pH 6.5 were
identical within error.
In summary, the nominal elastic moduli of these substrata varied over nearly two orders
of magnitude, while the other physicochemical characteristics considered to regulate bacterial
adhesion varied to a known or statistically indistinguishable extent. We confirmed that these
surface properties were unchanged when the substrata were hydrated over the timescales of the
bacterial incubation assays discussed below.
2.3.4. S. epidermidis adhesion modulated chiefly by substrata mechanical compliance.
We employed the above ensemble of substrata in a 2 h incubation of 107 cells/mL in
deionized water and observed the average colony density following 24 h culture under 1% agar.
S. epidermidis remained viable in ion-free suspensions well in excess of the duration of the
attachment assays. Figure 2.3a demonstrates strong positive correlation between the substrata
elastic moduli and colony density, with an approximately 100-fold increase in colony density for
a 100-fold increase in substrata stiffness.
46 | P a g e Figure 2.3. (0.7) Colony density as a function of various surface parameters. (a) Colony density varies directly with substrata elastic moduli E. All sample differences statistically significant (1‐way ANOVA, a = 0.05, P = 0.0059). (b) Colony density is independent of RMS surface roughness of the substrata. Scale bar = 5 mm. (c) Total interaction energy ΔGMWP for the microbe‐water‐PEM system is statistically indistinguishable among all substrata considered (1‐way ANOVA, a = 0.05, P =0.987). (d) Surface charge density Q, as measured via electrostatic repulsion of a carboxylated spherical probe in Milli‐Q water (see Methods), is within standard deviation for PEMs assembled at pH 2.0 (compliant) and pH 6.5 (stiff). Representative charge repulsion curve (solid) and constant‐surface‐charge model fit (dashed) are shown. Symbols refer to the following PEMs: PAA/PAH 2.0/2.0 (▼), 4.0/4.0 (x) in (a) to consider intermediate substrata stiffness, 6.5/6.5 (●), 3.5/7.5 (♦), and 3.5/8.6 (■). 47 | P a g e As substrata stiffness may be correlative with physicochemical surface interactions that
more strongly or more directly affect this initial bacterial adhesion, we also considered
correlations with surface roughness, total interaction energy, and charge density. The RMS
surface roughness varied among the substrata from 3 to 30 nm, yet Fig. 2.3b indicates no
discernable effect on bacterial attachment over this range and distribution of surface roughness.
Figure 2.3c shows that the surface interaction energy of the S. epidermidis-water-PEM system
ΔGMWP was statistically indistinguishable (1-way ANOVA, a = 0.05, P=0.987) among these
mechanically dissimilar substrata. Finally, we found net surface charge density to be quite
similar for the two substrata that differed most in both surface roughness and mechanical
compliance (PEMs assembled at pH 2.0 and 6.5). In fact, Fig. 2.3d shows that the slight
interfacial electrostatic repulsion of these PEMs in deionized water (~ -3 mC/m2) extends less
than 20 nm from the PEM surface. This interaction distance is small compared to the projected
length of bacteria fimbriae or pili that extend 500 to 1000 nm from the bacterial cell surface18, 129,
130
, suggesting one mechanism by which bacteria overcome such electrostatic repulsion. Thus, at
least for substrata of comparable surface interaction energies and charge density, it appears that
adhesion of viable S. epidermidis can be modulated by the mechanical stiffness of the substrata.
For the physicochemical properties quantified here, S. epidermidis colony density increases with
increasing substrata stiffness over the range of 1 MPa < E < 100 MPa.
To further test this hypothesis, we leveraged the tunability of layer-by-layer assembly to
gradually alter effective compliance of the PEM surface. After assembling stiff substrata at pH
6.5, we then added 0.5 and 1 bilayer of the compliant PEM at pH 2.0; after assembling compliant
substrata at pH 2.0, we added 0.5 bilayer of the stiff PEM at pH 6.5 (see Methods). As expected,
E of the stiff PEM surface decreased upon addition of compliant layers from the extrema of E
48 | P a g e ~100 MPa (pH 6.5) to ~30 MPa
(pH 2.0, 0.5 bilayer), and ~1 MPa
(pH 2.0, 1 bilayer). Effective E of
the compliant PEM increased to
E~100 MPa when topped with
PAH
6.5,
polycation
due
ostensibly
interpenetration
crosslinking.131
Figure
to
and
2.4
demonstrates that by changing the
Figure 2.4. (0.8) Multilayer addition to modulate composite substrata stiffness. Addition of 0.5 and 1 bilayer of PAA/PAH at pH 2.0 onto a stiff PEM (pH 6.5) decreases the effective mechanical stiffness of the substrata (grey circles) and decreases the bacterial colony density (black columns). Addition of one bilayer of PAA/PAH at pH 6.5 to a compliant PEM (pH 2.0) increases effective stiffness (black triangles) and bacterial colony density (grey columns). We observed statistically significant differences in the colony densities among the masked PEM 6.5 substrata and among the masked PEM 2.0 substrata, respectively (1‐way ANOVA, a = 0.05 with P = 0.00027 and 0.0031, respectively). effective
substrata
through
this
compliance
approach,
epidermidis
colony
progressively
decreased
S.
density
with
increasing PEM compliance. The
assembly of such composite films
has the potential to alter other
surface characteristics within this substrata set, but the strong correlation between effective
substrata stiffness and colony density is retained. This gradual masking of mechano-selective
adhesion is consistent with previous studies on eukaryotic cells,62 but is observed here after
addition of just a single compliant polyelectrolyte layer; decreased adhesion of fibroblasts is not
observed until addition of at least five bilayers of the compliant PEM to the stiff PEM. This may
be attributed in part to the increased forces and distances over which eukaryotic cells can strain
49 | P a g e the underlying substrata through actomyosin traction
at focal adhesions of diameters comparable to a single
bacterium.132, 133
2.3.5. Mechanoselective adhesion is independent of
monovalent ion concentration.
To consider whether the presence of the
Figure 2.5. (0.9) S. epidermidis colony density as a function of solution ion concentration. Colony density on compliant substrata (black, E
~ 1 MPa) is lower than that on stiff substrata (gray, E ~ 100 MPa), regardless of solution monovalent ion concentration in which 107 cells/mL were incubated with substrata. This suggests that activation of mechanosensitive monovalent ion channels is not required of mechanoselective adhesion in these bacteria. monovalent ions in 150 mM NaCl phosphate buffered
saline (PBS) strongly affected the observed trends,
bacterial attachment to the most mechanically distinct
PEMs (assembled at pH 2.0 and 6.5) was monitored
over a titration of salt concentrations.
Solution
molarity of 150 mM approximates physiological ionic
strength, and the absence of Ca2+ and Mg2+ ions approximates low extracellular calcium levels
predominantly complexed with serum albumin or negative ions.134 Figure 2.5 shows that there is
no major change in colony density with increased solution ionic strength (pure water to 150 mM
NaCl PBS).
More generally, this suggests that the molecular agents involved in this
mechanosensation are not sensitive to monovalent ionic strength changes over this broad
spectrum. Additionally, the Debye screening length, the distance from the substrata surface over
which electrostatic effects extend through the aqueous media, is a function of the ionic strength
and is ~100 nm in water (Fig. 3d) and < 1 nm at the highest ionic strength assayed.110 One may
reasonably conclude that the effect of surface charge density and its associated free energy on
bacterial adhesion are negligible under all solution molarities in this system, because there is no
50 | P a g e significant change in the adhesion response as the screening length is modulated across different
length scales. This titration result is particularly interesting in light of recent hypotheses that
bacterial sensing of mechanical stimuli may occur through stretch-induced activation of transient
receptor potential (TRP) ion channels.135-137
Our results suggest that activation of
mechanosensitive TRP channels is not required. It is also now known that some bacteria possess
analogs to eukaryotic mechanoactive cytoskeletal components including actin analogues such as
mreB, molecular motors, and integrin analogues.138 However, although S. epidermidis have been
shown to express ftsZ (a tubulin analogue that is chiefly involved in cell division), these bacteria
have not been shown to express actin analogues presumed to be required of eukaryotic
mechanotransduction via macromolecular focal contacts with the substratum material.135
5.3.6 Mechanoselective Adhesion is also Exhibited by Wild-Type and Mutant E. coli
The above results raise the possibility that mechanoselective adhesion may be unique to this
particular strain of S. epidermidis. To address this issue and further probe the origins of this
mechanoselectivity, we also used the same suite of PEM substrata to assay the adhesion
efficiency of wild-type (wt) E. coli (K-12 w3100 strain) and a spherical mutant form, ΔmreB.
This mutant lacks the mreBCD operon responsible for the actin analogue mreB and, thus, the
rod-like shape of E. coli. Additionally, E. coli are G−, whereas S. epidermidis are G+, indicating
structural differences in the bacterial cell envelope. As wt E. coli adhesion efficiency to control
surfaces (e.g., to glass) exceeded that of S. epidermidis under these incubation conditions, the E.
coli seeding density in these experiments was reduced to 0.5 × 107 cells/mL; otherwise, all
experimental conditions were identical to those used in the S. epidermidis experiments reported
in Figure 2.3. Figure 2.6a shows that the colony density of wt E. coli increases directly with
51 | P a g e increasing PEM substrata stiffness over the previously detailed range of 1 MPa < E < 100 MPa.
This represents approximately a 1000-fold decrease in colony density over a 100-fold reduction
in E; on a proportional basis, this is a larger reduction in adhesion efficiency than observed for
the S. epidermidis. Thus, despite the differences between the biochemical compositions of
spherical S. epidermidis and rod-like E. coli, both cell types exhibit mechanoselective adhesion
over this range of substrata stiffness. As in the case of S. epidermidis cultured with the same
defined substrata, there was no correlation between colony density and substrata characteristics,
including surface roughness, total interaction
energy, or charge density. We note that the
relatively greater cell surface area of E. coli
(2−8-fold)139, 140 may contribute to the enhanced
adhesion efficiencies of this bacterium to these
surfaces; further studies are required to explore
this possibility.
To consider the effects of cell shape
conferred by protein expression, without the
concomitant effects of varied extracellular
envelope composition, we also assayed the
adhesion efficiency of spherical ΔmreB E. coli
Figure 2.6. (0.10) ) (a) Wild‐type E. coli exhibit colony density (bars) that varies directly with the stiffness (symbols) of the PEM substrata. (b) Viable, spherical ΔmreB E. coli that lack the actin analogue mreB also adhere more readily to the stiffest substrata (E ‫׽‬ 100 MPa) than to the most compliant substrata (E ‫׽‬ 1 MPa). Scalebars = 1 mm.
52 | P a g e to
the
substrata
of
extreme
mechanical
compliance. Figure 5.6b demonstrates that the
colony density of this actin analogue-mutated,
G− microbe also correlated directly with PEM
substrata stiffness. These ΔmreB E. coli exhibited a 100-fold reduction in colony density with a
100-fold decrease in E, a rate of change comparable to that observed for the spherical S.
epidermidis over the same range of conditions. Thus, although this represents an admittedly
incomplete survey of bacterial strains and types, these data demonstrate that the correlation of
viable bacteria adhesion efficiency with substrata stiffness is not unique to only the G+, spherical
S. epidermidis. Bakker et al. have noted mild, positive correlation of substrata stiffness with
adhesion of marine bacterial strains under certain flow chamber conditions, but not others, to
polyurethane surfaces141, 142; these authors considered much stiffer substrata over a considerably
narrower range of E than those of the present study (1500 MPa < E < 1900 MPa) and did not
report possible differences in surface charge density among those substrata. Thus, systematic
consideration of the generality of this dependence for other bacterial strains, incubation
conditions and substrata characteristics, including stiffness, remains an important topic for future
studies.
Although complete elucidation of the molecular mechanisms responsible for this new
observation is beyond the focus of this paper, our results indicate that neither stretch responsive
monovalent ion channels, cell envelope composition indicated by Gram-staining, nor cell shape
are required and/or causal elements. Alternatively, it is possible that bacterial fimbriae/pili,
which are constitutively expressed by the bacteria considered herein18,
129, 130
, mediate a
mechanoselective process similar to the so-called catch-bond mechanism posited to explain
effects of shear flow stress on cell adhesion dynamics: the lifetime of noncovalent interactions
can be increased under external mechanical force.135, 143 As bacterial pili collide with and sample
substrata during incubation, the mechanical resistance of the material to pili retraction would
increase with increasing substrata stiffness; this stabilization on stiffer substrata could increase
53 | P a g e the lifetime of pili-substrata interactions during the fast step of bacterial two-stage binding
kinetics.18 S. epidermidis and E. coli possess several glycosylated substructures at both the pili
and extracellular capsule18,
129, 130
known to form attachments to materials and capable of
complex interactions similar to those observed in other bacterial species that form pili catchbonds. Thorough consideration of this and other proposed mechanisms of this mechanoselective
adhesion is the focus of ongoing work.
Together, these results do not invalidate the
physicochemical effects reported to influence microbial adhesion. Clearly, several competing
surface features affect bacterial adhesion, viability, and subsequent colonization. Rather, the
current study demonstrates that mechanical compliance of the substrata presents an important
additional factor contributing to stable adhesion of viable bacteria.
5.4. Conclusions
We find that the adhesion of viable, colony-forming S. epidermidis and E. coli correlates
positively with increasing elastic modulus of weak polyelectrolyte multilayered substrata over
the range 1 MPa < E < 100 MPa. To our knowledge, this is the first demonstration that substrata
stiffness affects the adhesion of viable prokaryotes, such as bacteria, independently of other
surface characteristics. These observations were not attributable to differences in posited
physicochemical regulators of bacterial adhesion, including RMS surface roughness, surface
interaction energy, and surface charge density of the PEM thin films. For the bacteria
concentrations considered, neither divalent ions nor monovalent ions such as Na+ and Cl− are
required for this mechanosensory function, suggesting that activation of TRP ion channels is not
required for mechanoselective adhesion of S. epidermidis. Further, quantitatively similar trends
in wt and ΔmreB E. coli confirmed that this correlation is not limited to a single type or shape of
54 | P a g e bacteria. Although the underlying mechanisms require further study, it is clear that the
mechanical stiffness of nanoscale polymeric substrata can strongly modulate adhesion of viable
bacteria in aqueous suspensions, independently of several other interactions at the cell−material
interface. Thus, mechanical compliance of material surfaces represents an additional design
parameter by which colonization of both beneficial and potentially infectious bacteria can be
modulated.
55 | P a g e CHAPTER 3
3. Reversibly Antibacterial Polyelectrolyte Multilayers
3.1 Introduction
There is an emerging understanding of the material properties affecting the antimicrobial
properties of synthetic thin films. Prevention of bacterial attachment and subsequent bacterial
infections has been approached in three general ways. One method has been to create a surface
that, although not bactericidal, resists bacterial attachment. For example, anti-adhesive
polyelectrolyte multilayers (PEMs) incorporating poly(ethylene glycol)20 and other highly
swellable, low modulus materials have been shown to greatly reduce the attachment of
bacteria144. Others have found that, under certain conditions, hydrophobicity145 and surface
roughness146 can modulate bacterial attachment. Some argue that DVLO ((Derjaguin, Landau,
Verwey, and Overbeek) theory, evaluating a combination of Lifshitz–van der Waals and
electrostatic interactions, explains attachment of microbes to surfaces, and they have attempted
to optimize surfaces to minimize attachment147. Another method of biofilm prevention seeks to
kill bacteria before colonization by leaching antibacterial species like silver ions6, 148 or
antibiotics149. The third approach is the creation of contact-killing surfaces, usually containing
covalently attached cationic antimicrobials like quaternary ammonium compounds15, 150, 151, alkyl
pyridiniums4, 152, 153 or quaternary phosphonium154. Multiple mechanisms of action have been
proposed for how these cationic species are able to disrupt the bacterial cell membrane152, 155,
some requiring hydrophobic chains of certain lengths to penetrate and burst the bacterial
membrane. Nevertheless, it has been shown that high levels of positive charge are capable of
conferring antimicrobial properties irrespective of chain length, perhaps by an ion exchange
mechanism between the bacterial membrane and the charged surface152, 156. In addition, studies
56 | P a g e of cationic polymer cytotoxicity with eukaryotic cells have found that linear charge density and
chain flexibility also greatly determine the degree of cytotoxicity82, 157, and charge density and
mobility can also be factors in antimicrobial efficiency158. These findings create the possibility of
engineering a variety of positively charged surfaces to create a range of contact-killing materials.
The PEM assembly process distinguishes itself as a safe, water-based and versatile
technique for creating coatings. Glass, silicon, plastics, metals, colloids and even cells have been
coated with PEMs for a huge variety of applications. One of the most useful aspects of PEMs is
their sensitivity to assembly and post-assembly conditions. For example, addition of salt to a
polymer bath can change both the chain conformation and the layer thickness. Chemical,
thermal or UV crosslinking can significantly alter the stability of a given PEM159, 160 and
exposure to high ionic strength can etch away layers161.
Manipulating pH, both during and after assembly, is perhaps the richest method of PEM
architecture control. We and others have shown that with a given pair of polymers,
manipulation of pH conditions during and/or after assembly can significantly alter layer
thickness10, 126, swellability62, elastic modulus162, charge density126, surface roughness,
porosity163, 164, surface wettability126, index of refraction12, layer interpenetration126 and bulk and
surface composition10, 165. Adjustment of these material parameters can control, for example,
cytophilicity62, 162 and anti-reflection properties12. Certain PEMs are pH responsive and therefore
can be used for drug and dye uptake and release16, 166 and nanochannel gating167.
In this work, we build on previous research using pH to control key material parameters
on surfaces of identical chemical composition, tuning the layer-by-layer technique to create
surfaces with the necessary charge density and chain mobility to confer antimicrobial properties.
57 | P a g e We show that assembly and post-assembly pH manipulation of a model PEM system can turn an
adherent and non-toxic coating into a highly bacteria-resistant and antibacterial PEM. This
antibacterial and anti-adherent functionality can be again switched off with subsequent pH
treatment. Unlike most systems (such as grafted polymer surfaces) that require working at
different pH conditions to adjust degree of ionization, our system displays a “molecular memory
effect,” creating surfaces of identical chemical composition with very different properties at
neutral pH depending on the previous pH treatments168. The polyelectrolyte multilayers
discussed here are therefore an ideal platform for evaluating the effects of cationic surface charge
and polymer chain and charge presentation on antibacterial properties.
Our earlier research has shown that highly swellable, low moduli surfaces inhibit aqueous
bacterial attachment144. Others have shown that cationic polymers such as chitosan can be
incorporated into PEMs to create efficient contact-killing surfaces49, 50. Thus, we have the ability
to engineer a surface with two methods of colonization prevention: low modulus to resist
attachment and mobile cationic charges as biocides. The versatility of the PEM technique allows
for a degree of control difficult in other systems, and thus, PEMs are a useful model for studying
parameters affecting anti-adhesive and antibacterial properties.
Here, techniques were developed to create sufficiently positively charged surfaces with
the necessary chain mobility for antimicrobial functionality and bacteria-resistance. A model
system comprised of cationic poly(allylamine hydrochloride) (PAH) and anionic poly(sodium 4styrene sulfonate) (SPS) was used as illustration. Two strains of Staphylococcus epidermidis (S.
epidermidis), one biofilm forming and one non-biofilm forming, and two strains of Escherichia
coli (E. coli) were chosen as model gram positive and gram negative bacteria challenges. Both
airborne antibacterial capabilities and aqueous attachment were studied. In addition, other
58 | P a g e polymer systems were briefly investigated to expand the application of antibacterial PEMs.
Although the majority of the study focuses on the SPS and PAH PEM system, the general
concept, adjusting assembly and post-assembly conditions to create highly swellable systems
and/or high mobile cationic charge density, can be applied to tune other multilayers to produce
antibacterial capabilities.
3.2 Materials and Methods
3.2.1 PEM Assembly:
Polyelectrolyte multilayers were assembled in the manner previously described 10, 62. All polymer
solutions were prepared with deionized water (0.01 M, based on the repeating unit molecular
weight). SPS/PAH PEMs were assembled on plain glass slides (VWR) with SPS (poly(sodium
4-styrene sulfonate) Mw = 70,000 g/mol; Aldrich) and PAH (poly(allylamine hydrochloride); Mw
= 70 000 g/mol; Polysciences) that were pH adjusted using 1M HCl and NaOH. The
multilayers, assembled with the polymers adjusted to pH 4.0 or pH 9.3 (for pH 9.3 assembly,
rinse baths were also adjusted to pH 9.3), were 15.5 bilayers thick (with PAH on top). Note that
0.1M NaCl was added to the pH 4.0 polymer solutions to ensure good film growth. Acid treated
SPS/PAH films were immersed in water at pH 2.5 for 15 minutes, rinsed thoroughly in deionized
water and dried with compressed air. An additional multilayer system included PAH assembled
with the weak anionic polyelectrolyte PAA (poly(acrylic acid); Mw = >200 000 g/mol; 25%
aqueous solution; Polysciences (previously labeled as Mw = 90 000 g/mol)). These other
multilayers were assembled anion first on aminoalkylsilane coated glass (Sigma-Aldrich).
PAA/PAH samples were made with both polymers adjusted to pH = 2.0 (9.5 or 10 bilayers), 4.0
(7.5 or 8 bilayers) and 6.5 (49.5 or 50 bilayers), with both PAA and PAH on top. Samples with
PAA at pH 3.5 and PAH at pH 7.5 and 8.6 were also assembled with both PAA (5.5 bilayers)
59 | P a g e and PAH (6 bilayers) as the final layer. All PAA/PAH samples were assembled to have a final
dry thickness of ~ 50 nm. For Figure 1, a hydrogen-bonded system of PAA and
poly(acrylamide) (Mw = 600,000 – 1,000,000 g/mol; Polysciences) were assembled on
aminoalkylsilane-modified slides with 10.5 bilayers (PAA topped). All polymer baths and rinse
baths were adjusted to pH 3.0 during hydrogen-bonded PEM assembly. PAA/PAAm samples
were lightly crosslinked at 100°C for 24 hours so they would remain intact in physiological
conditions. For all ellispometry measurements, single-crystal polished silicon wafers (p-type, 150 M cm, [100] orientation, WaferNet) were used as substrates. Note that PAA/PAH films
built on silicon were 0.5 bilayer thicker due to the negative charge of the silicon versus the
positive charge of the aminoalkylsilane coated glass slides. For streaming zeta potential
measurements, the PEMs were assembled on 0.02” thick polycarbonate (Grainger). The notation
used to refer to PEMs throughout this paper is:
P1/P2 pH1/pH2
where P1 is the polyanion, P2 is the polycation, pH1 is the pH of deposition of the polyanion and
pH2 is the pH of deposition of the polycation.
3.2.2 Thickness and Swelling Measurements:
Spectroscopic ellipsometry (M-2000D, J. A. Woollam Co., Inc.) was used to determine the
thickness of PEMs. Swelling experiments, performed using in-situ ellipsometry in the manner
described previously169, were conducted after equilibrating the samples for 10 minutes in
deionized water. The PEMs assembled evenly on the Si for ellipsometry measurements and
were assumed to be equivalent to PEMs grown on glass after the initial few layers.
60 | P a g e 3.2.3 QCM-D Analysis
SPS/PAH 9.3/9.3 films with 6.5 bilayers (designated (SPS/PAH 9.3/9.3)6.5) were assembled on
SiO2 coated crystals (Q-Sense). Analysis was performed on an E4 Q-Sense Quartz Crystal
Microbalance with Dissipation. After a baseline was acquired in pH 9.3 water, aqueous solutions
at pH 2.5 or pH 11 were flowed over the crystal surface at a rate of 0.1 mL/min, and changes in
dissipation and frequency were measured.
3.2.4 Rose Bengal Dye Absorption:
Samples were immersed for 15 min in a 10-3 M solution of Rose Bengal, an anionic dye that
binds to positive charge, adjusted to pH 5.5. The samples were rinsed twice with deionized
water and dried with compressed air. Absorbance was measured with a Varian Cary 5E
spectrophotometer.
3.2.5 Streaming Zeta Potential Measurements:
PEMs were assembled on polycarbonate slides cut from large sheets and placed (two at a time)
in a 500 micron sample holder. Streaming zeta potential was measured at pH 6.8 with 1mM KCl
electrolyte using a ZetaCAD (Laval Lab, Inc.) streaming potential analyzer.
3.2.6 Micron Particle Attachment:
1.0 micron diameter carboxylate-modified fluorescent latex beads (2.5% aqueous suspension,
Sigma) were diluted 1:1000 in deionized water and adjusted to pH 6.5. Samples were immersed
in the suspension for 1 hour, rinsed twice with deionized water and dried with compressed air.
Samples were imaged in fluorescent mode at 50 X magnification with a Zeiss Axioplan 2
UV/visible microscope. At least five pictures of each sample were collected. Results are
presented as average number of particles per image. Zeta potential measurements on the latex
particles were acquired with a Brookhaven Instruments Zeta-PALS analyzer.
61 | P a g e 3.2.7 Aqueous Bacterial Attachment Assays (or waterborne bacterial assays):
Bacterial attachment was assayed as described previously144. Briefly, monoclonal
Staphylococcus epidermidis (S. epidermidis, ATCC # 14990 (non-biofilm forming) or ATCC #
35984 (biofilm forming)) or Escherichia coli (E. coli, ATCC # 700728 or ATCC # 14948) were
grown overnight in LB-Miller broth (VWR). The incubation was centrifuged at 2700 rpm,
decanted and resuspended in 150 mM NaCl PBS (VWR) three times (10 min, 5 min, 5 min).
The final resuspension was diluted to a density of 109 cells/mL measured via optical density.
This resuspension was then diluted in sterile deionized water to107 cells/mL. The PEMs were
placed in the diluted bacterial suspensioin for 2 hrs at room temperature and rinsed thoroughly in
3 water baths. Samples were incubated under 1% LB agar (VWR) gel overnight, and colonies
counted to determine the ability of viable bacteria to attach to each sample. Samples with few
colonies were hand counted. For densely populated slides, at least 10 digital images per sample
were acquired with a 4x objective using an inverted optical microscope (Leica) and the total
number of colonies per slide was extrapolated. Results were presented as either colony density /
area or as a normalized colony density calculated as follows: normalized colony density =
average colony density for sample / average colony density for glass control. By definition, the
normalized colony density of glass was equal to one. All experiments were performed in
triplicate, except for one experiment with biofilm-forming S. epidermidis in which four samples
were tested.
3.2.8 Airborne Antibacterial Assays:
Airborne assays were based on the protocol discussed by Klibanov et. al170. Overnight bacterial
cultures (~109 cells/mL) were diluted 1:100 in sterile deionized water unless otherwise noted.
The dilution was sprayed onto the sample surface using a gas chromatography sprayer (VWR
International, cat. no. 21428-350). Without rinsing, samples were immediately placed in 100mm
62 | P a g e Petri dishes (VWR) and covered with 1% LB agar gel overnight at 37C. Resulting colonies were
counted as described above. For airborne studies on PAA/PAAm and PAA/PAH samples, a
slightly different protocol was used. Bacterial suspensions of 109 cells/mL were prepared as
described for the waterborne attachment assays (centrifuged and measured for an optical density
of 1 in PBS) and diluted to 2 x 106 cells/mL in sterile deionized water. Fine mists (d~ 5-100 μm)
were sprayed at 1 μl/cm2•spray for a total of 6×103 bacterial cells/cm2. SPS/PAH samples
showed identical trends for both spray methods. Log kills were reported using the following
convention: log kill = log
(number of colonies on glass
control) – log (number of
colonies on sample). All
experiments were performed in
triplicate.
3.3 Results
3.3.1 Airborne Antimicrobial
Assays:
The antibacterial
properties of cationic species
Figure 3.1. (0.11) Airborne S. epidermidis (14990) challenges showing no antibacterial activity on (a) PAA/PAH films (labeled with the pH of deposition and the identity of the top layer), (b) SPS/PAH films assembled at pH 4.0 with 0.1 M NaCl (PAH as top layer) and (c) PAA/PAAm films assembled at pH 3.0 and thermally crosslinked (PAA as top layer). All samples statistically identical to glass controls (student’s t‐test, p<0.05). 63 | P a g e have been extensively
documented4, 15, 54, 152, 156.
Cationic polymers, both with and
without hydrophobic chains,
have been used to effectively kill
a broad spectrum of bacteria. The cationic polymers present in PEMs have high linear charge
density, but their typical form in a multilayer, strongly coupled with anionic species, does not
produce any antibacterial activity even when the cationic species is the top layer. The cations are
not “free” or unbound, able to induce bacterial death. As shown in Figure 3.1, commonly used
polymer combinations (SPS/PAH and PAA/PAH) assembled under typical conditions (with both
polymers at the same pH) do not possess any airborne antibacterial properties against a model
non-biofilm forming S.epidermidis strain. The cross-linked hydrogen bonded system
PAA/PAAm, which likewise lacks free cationic charge, also does not exhibit any biocidal
activity (Figure 3.1c). However, as a surface modification process, PEM assembly offers a
unique advantage. The tunability of the PEM process can be exploited and surface parameters
can be adjusted to create desired functionalities. For example, adjusting assembly and postassembly pH can create surfaces with drastically different swellabilities and free cationic charge
(not coupled with an anionic polymer), two factors that greatly influence bacterial adhesion and
microbicidal abilities. One way to tune surface properties results from the pH-dependent degree
of ionization of weak polyelectrolytes (e.g., PAA and PAH). Our earlier research showed that
highly swellable, low modulus PEMs like PAA/PAH assembled at pH 2.0 decreased aqueous
bacterial attachment, but could not completely prevent bacterial growth. In that case, PAA, with
a solution pka of pH ~6.5171, was assembled into the film with a low degree of ionization, leaving
many uncharged carboxylic acid groups unbound to the cationic polymer. When exposed to
neutral physiological pH conditions, the carboxylic acids deprotonated and created a film with
excess negative charge that swelled due to charge-charge repulsion. To overcome the challenge
of creating antimicrobial surfaces, many researchers have incorporated antibacterial species,
64 | P a g e including silver ions6, 15, quaternary ammonium salts15, chitosan22, 27, 49, 50, antibiotics28,
enzymes8, etc.
PEMs assembled from SPS and PAH at high pH assembly conditions have been shown to
exhibit large, reversible swelling transitions with post-assembly acid treatment168. This
reversible change, illustrated in Figure 3.2, has been associated with the acid-activated opening
of hydrophobic PAH clusters and the
reforming of those clusters at high pH
(> pH 10)169. PAH, with a solution pka
~ 8.8171, contains hydrophobic clusters
comprised of uncharged amines along
the backbone in solution at pH 9.3.
Hydrophobic clusters of PAH
assembled into a film at high pH are
opened at low pH (<pH 4), creating an
excess of free ammonium groups that
leads to high film swellability. The asassembled film with hydrophobic
clusters is termed “as-deposited” or
Figure 3.2. (0.12) Schematic of acid‐activated opening of hydrophobic clusters comprised of amine groups in SPS/PAH 9.3/9.3 PEMs. The hydrophobic clusters can be reformed with subsequent treatment at high pH. “unswollen” throughout this work,
while the PEM that has been treated at
low pH is termed “acid-treated” or
“swollen.” Since the swelling
transition has a large hysteresis, acid
65 | P a g e treated PEMs remain highly
swellable up to ~pH 10. The
hydrophobic clusters of
deprotonated amines can be
reformed with subsequent high
pH treatment (greater than pH
10). Reformed clusters remain
intact until the pH is again
dropped below ~pH 4. The
film acts as a “molecular
Figure 3.3. (0.13) Spray (airborne) antibacterial test results on glass and SPS/PAH 9.3/9.3 as deposited and acid treated samples for non‐biofilm memory,” remaining highly
forming S. epidermidis (ATCC strain 14490), biofilm forming S. epidermidis
(ATCC strain 35984), E. coli ATCC strain 14948 and E. coli ATCC strain swollen or unswollen at neutral 700728. All acid‐treated films are statistically distinguishable from glass and as‐deposited samples (student’s t‐test, p<0.05). pH depending on previous pH
treatment. Although all bacteria experiments were performed at neutral pH, the films behaved
differently depending on whether they had been previously acid-treated to protonate the unbound
amine groups. Throughout the study, SPS/PAH films assembled at pH 4 were used as additional
control samples. These films were assembled with PAH containing fully protonated ammonium
groups and did not undergo any significant changes upon treatment in solution at low pH. Figure
3.3 shows that the free ammonium groups present in the swollen SPS/PAH 9.3/9.3 acid-treated
PEM led to cationic killing of bacteria in airborne antibacterial studies. After an acid-treated
SPS/PAH 9.3/9.3 PEM was immersed in a solution at high pH (pH 11), the hydrophobic clusters
reformed and antibacterial activity was lost (Figure 3.4). The cycle was repeated, with films
undergoing a second acid treatment once again demonstrating antibacterial capabilities. Note that
66 | P a g e SPS/PAH 4.0/4.0 control samples, both as-deposited and acid-treated, were statistically
indistinguishable from the glass controls (student’s t-test, p<0.05). Both biofilm-forming and
non-biofilm-forming strains of gram positive S. Epidermidis showed high levels of cell toxicity
on the SPS/PAH 9.3/9.3 acid treated samples. The biofilm-forming strain showed ~log 2.5 kill,
and the non-biofilm-forming strain showed ~log 3 kill. Gram negative E. coli showed much
lower levels of cell toxicity, with only ~50% kill on the acid treated samples. Many other
researchers have seen increased surface-contact killing efficiencies against Gram positive
strains2, 33, 54, 172. The different cell walls of gram positive and gram negative bacteria interact
differently with cationic charge, leading to different antibacterial efficiencies 173. Gram positive
bacteria contain a simple cell wall composed of a thick peptidoglycan outer layer. Gram negative
bacteria have a more complicated cell wall containing much less peptidoglycan and two
additional layers (the periplasmic space and the lipopolysaccharide layer (outer envelope)). In
particular, the lipopolysachharide layer protects the Gram negative cell, leading to lower cell
membrane permeabilities174. Gram negative cells can also change their outer envelope
composition in response to quaternary ammonium compounds for added protection175. For the
cationic surfaces that have successfully eliminated Gram negative bacteria, hydrophobic alkyl
chains of specific lengths seem to be necessary2-5. The SPS/PAH 9.3/9.3 system, lacking alkyl
chains, would therefore not be expected to be extremely effective against Gram negative
microbes. In future experiments, charged chains incorporating these alkyl segments (e.g., those
developed by Klibanov et al.3, 4) could be included in PEMs assembled under the correct
conditions to impart broad spectrum antibacterial functionality. 67 | P a g e It should be noted that, as found
previously, the swelling transition in the
SPS/PAH 9.3/9.3 system slowly diminished
over the course of several weeks in a nonvacuum environment, likely due to a new
molecular organization caused by increased
Figure 3.4. (0.14) Airborne antibacterial tests with S. chain mobility in a plasticizing humid
epidermidis (non‐biofilm forming strain) with films that have been cyclically pH treated to create swollen environment168, 169. Airborne antibacterial
films (acid‐treated) and unswollen films (pH 11‐
treated). activity against S. epidermidis likewise
decreased after one week and virtually
disappeared after one month (Figure 3.5).
Nevertheless, the SPS/PAH 9.3/9.3 system, with
antibacterial capabilities in the swollen state
resulting from mobile excess positive charge, is
an excellent model system to evaluate the charge
Figure 3.5.(0.15) Airborne antibacterial activity of SPS/PAH films 9.3/9.3 as deposited and acid‐treated after one week (“1 Wk”) and one month (“1 Mon”). and structural components necessary to kill and
resist bacteria.
3.3.2 Waterborne antibacterial assays:
To further test the capabilities of the SPS/PAH system, we tested bacterial growth after
immersion in aqueous solutions of bacteria at neutral pH. In this situation, two factors
influenced final colony density: the number of bacteria that were able to form stable attachments
with the surface and the number of bacteria that were killed. As shown previously, highly
68 | P a g e Figure 3.6. (0.16) Waterborne bacteria experiments on glass (circles), SPS/PAH 9.3/9.3 as‐deposited (squares) and acid‐treated (triangles) with (A) S. epidermidis 14990, (B) E. coli 14948, (C) S. epidermidis 35984 and (D) E. coli 700728. Insets in (A) and (C) are enlarged to show the differences between the as‐deposited and acid‐
treated samples. For all strains, colony growth was most efficiently reduced on SPS/PAH 9.3/9.3 acid‐treated samples. swellable PEMs correspond to highly compliant surfaces162. In earlier work, we showed that
surface stiffness correlated positively with bacterial attachment144. Low modulus surfaces
enabled less stable bacteria attachment. Thus, it was anticipated that the highly swellable and
highly antibacterial SPS/PAH 9.3/9.3 acid treated surface would show very little or no colony
growth in comparison to the control samples. As seen in Figure 3.6, no S. epidermidis colonies
and few E. coli colonies grew on the SPS/PAH 9.3/9.3 acid-treated surfaces. This indicated that
69 | P a g e the highly swellable surface repelled the majority of the bacteria, and those that did form stable
attachments with the surface were killed by the cationic effect. The as-deposited films showed
some bacterial resistance, perhaps due to protonation of some of the amine groups in the
hydrophobic clusters at the very surface of the film and/or increased chain mobility at the film
surface in an aqueous environment. However, the as-deposited system was not optimized to
present maximal free, mobile cationic charge, and most of the amine/ammonium groups were
involved in hydrophobic clustering or electrostatic binding with SPS. More reliable and higher
levels of bacterial resistance were seen on the highly swellable acid-treated system. Note that no
colony reduction was seen on SPS/PAH 4.0/4.0 as-deposited or acid-treated samples with any of
the bacteria strains. Thus, the unique charge density, charge presentation and film swellability of
the SPS/PAH 9.3/9.3 acid-treated system were essential for maximum aqueous bacterial
resistance. 3.3.3 Multilayer characterization:
To confirm that the SPS/PAH 9.3/9.3 surface contained sufficient cationic charge and
chain mobility for microbiocidal properties, we built on previous characterization of this PEM
system168, 169. As noted earlier, a large discontinuous swelling transition occurred when an
SPS/PAH film assembled at high pH (8.5 or greater) was immersed in a low pH solution. Here,
in-situ ellipsometry verified that the pH 2.5 acid-treated SPS/PAH 9.3/9.3 PEMs swelled more
than 700% in ~pH 6 water (similar to the waterborne bacteria test conditions) while the asdeposited SPS/PAH 9.3/9.3 PEMs only swelled about 30% . This large swelling was due to
electrostatic repulsion after the breakup of hydrophobic amine clusters and the creation of excess
positive charge in the acid-treated PEM. The 15.5 bilayer as-deposited SPS/PAH 9.3/9.3 PEMs
had a dry thickness of ~60nm. Quartz crystal microbalance with dissipation (QCM-D) assays,
70 | P a g e Figure 3.7. (0.17) QCM‐D data showing changes in frequency (black line) and changes in dissipation (grey line) to qualitatively illustrate swelling transitions in SPS/PAH 9.3/9.3 PEMs. Downward arrows point to times when the pH of the flowing water was changed.
shown in Figure 3.7, were performed with (SPS/PAH 9.3/9.3)6.5. Huge increases in dissipation,
indicative of a high swelling, and huge decreases in frequency, indicative of a large uptake of
water in the film, were seen when pH 2.5 water was flowed over the sample surface. This
swelling transition could be reversed by flowing pH 11 water over the surface. The cycle was
repeated a twice. Note that the swelling and de-swelling transitions did not reach equilibrium in
the time alloted, so the full extent of the transition is not visible. However, QCM-D data clearly
shows a reversible transition from a highly dissipative and hydrated state to a more rigid state. The presence of free cationic charges in the films was confirmed by staining with Rose
Bengal, an anionic dye. Unlike the as-deposited SPS/PAH 9.3/9.3 films, the large number of free
ammonium groups in the swollen films stained darkly with Rose Bengal at pH 5.5. After 15 min
incubation in the dye followed by rinsing, a maximum absorbance of 0.15 was measured for
glass slides coated on both sides with 60 nm as-deposited PEMs whereas a maximum absorbance
of 1.25 was measured in the acid-treated films. The high absorbance in the acid-treated films
was indicative of an extremely swollen structure with many unbound positive charges. The low,
71 | P a g e but existent staining on the as-deposited samples indicates the presence of a small amount of free
ammonium groups. This finding supports our other work describing the formation of gold
nanoparticles in SPS/PAH films via coupling of anionic gold precursors with positive charge and
subsequent reduction176. Although the SPS/PAH 9.3/9.3 acid-treated PEMs were capable of
forming many more gold nanoparticles, the as-deposited films showed a small amount of gold
formation. Consistent with the idea that the cationic charges in SPS/PAH 4.0/4.0 PEMs (both asdeposited and after acid-treatment) are strongly coupled to the polyanion, these films adsorbed
no Rose Bengal and were not capable of any nanoparticle formation.
The number of free ammonium groups per unit area in the acid treated film was
calculated with data from the literature indicating that an SPS/PAH 9.3/9.3 PEM contains 72
mol% PAH and that the PAH degree of ionization after acid treatment rises from 70% to 95%169.
With a dry film thickness of ~60 nm and an assumed PEM density of ~1.17 g/ cm3 (SPS
homopolyer = 1.18g/ cm3, PAH homopolyer = 1.15g/cm3)177, there are approximately 8 x 1015
free charged amines/cm2 in the acid-treated PEM. Others have found that a cationic chargedensity threshold of 1013 - 1014 charges/cm2 is necessary for complete kill of E. Coli and S.
Epidermidis152. Since acid-treated SPS/PAH 9.3/9.3 PEMs have ~1016 charged free amines/cm2
(throughout the whole volume of the film), it is probable that bacteria were exposed to the
threshold charge density.
Streaming zeta potential or streaming potential, E, was measured to further characterize
the PEM systems by evaluating the charge at the very surface of the film as opposed to
throughout the bulk of the film. Streaming potential measures the electric current generated as
an electrolyte is driven across a channel with charged walls (here, the PEM) and is related to the
surface zeta potential, ξ, as follows:
72 | P a g e ξ = ληE/εP
(Eq. 3.1)
where λ is the electrical conductivity of the solution, η is the viscosity of the solution, ε is the
permittivity and P is the pressure across the cell. Streaming potentials were measured using 1
mM KCl at pH 6.8 as the electrolyte. As can be seen in Table 3.1, the PEMs assembled at pH
9.3 (with PAH on top) had significantly higher streaming zeta potential than the PEMs
assembled at pH 4.0. As expected, the SPS topped systems assembled at pH 4.0 exhibited a
negative streaming zeta potential. The streaming zeta potentials for the SPS/PAH 4.0/4.0 system
generally agreed with the values for streaming potential of similar systems documented in the
literature177-180. Interestingly, acid treating the pH 9.3 samples did not produce a significant
change in streaming potential. Although the acid treated samples had more positive charge over
the whole volume of the film (as indicated by FTIR169 and Rose Bengal staining), they presented
similar charge at the very surface.
Typically, bacteria are negatively charged and a few microns in diameter. Thus, we
measured waterborne adsorption of 1 micron negatively charged microparticles onto the PEMs to
simulate PEM - bacteria interactions. Measurement of attachment of charged microspheres to
PEM surfaces has been used previously to characterize the surface charge of rigid PEMs171, 174,
175
. Here, the PEMs were immersed in a pH 6.5 aqueous suspension of 1 micron fluorescent
carboxylate-modified latex spheres (zeta potential = -68.68 +/- 3.16 mV). The average number
of adsorbed particles was counted and the results are shown in Table 3.1. As expected, the
SPS/PAH 4.0/4.0 PAH topped sample adhered many particles, while the SPS topped surface
adhered very few particles. The negatively charged particles were attracted to the positively
charged PAH topped surface and repelled from the negatively charged SPS topped surface. The
SPS/PAH 9.3/9.3 as-deposited samples adhered almost double the number of particles than the
73 | P a g e SPS/PAH 4.0/4.0 PAH topped surface, in agreement with the streaming potential measurements
that showed a significantly higher positive charge on the SPS/PAH 9.3/9.3 surface. Very few
particles bound to the SPS/PAH 9.3/9.3 acid-treated surface despite the fact that the as-deposited
and acid-treated PEMs displayed similar streaming potentials. The differences in particle
attachment in this case are thought to be a result of the differences in surface charge presentation:
charges on the acid-treated sample had significantly more mobility, creating steric hinderance
and making it harder for the particles to form stable bonds with the surface. Simulations of
particle adsorption onto polymer-coated surfaces support this hypothesis; particle adsorption was
reduced when the polymer was well-solvated181. Table 3.1. Streaming Zeta Potential and Number of Attached Charged Microparticles for Representative PEMs. pH 4.0 PAH
topped
pH 4.0 SPS
topped
pH 9.3 PAH
topped, as
deposited
pH 9.3 PAH
topped, acid
Streaming Zeta
Potential (mV)
+33.6 +/-11.7
-45.7 +/- 7.9
+48.4 +/- 6.4
+47.4 +/- 6.8
Microparticles/image
320 +/- 24.2
14.2 +/- 5.4
617 +/- 40.1
10.8 +/- 3.4
3.3.4 Other antibacterial PEMs:
The basic approach used to create antibacterial SPS/PAH films by manipulating assembly
and/or post-assembly conditions to expose free cationic charge and increase chain mobility can
be extended to other PEM systems. The PAA/PAH system assembled at various pHs has been
extensively studied10. As shown in Figure 3.1, the commonly used pH combinations (where both
polymers are at the same pH), showed no airborne antibacterial properties since all polycations
were strongly paired with polyanionic species. Studies of gold nanoparticle formation, requiring
74 | P a g e unpaired ammonium groups to bind gold
precurors, confirmed the total absence of free
amines in the case of PAA/PAH 3.5/3.5 and only
very low amounts of free amines above pH 2.5 in
the case of PAA/PAH 3.5/7.5176. Note that
Figure 3.8. (0.18) Airborne antibacterial study with antibacterial spray tests showed no killing in the
S. epidermidis (non‐biofilm forming) on glass, PAA/PAH 3.5/7.5 system with PAA on top and
PAA/PAH 3.5/7.5 and 3.5/8.6 with PAA on top (denoted 3.5/7.5 PAA and 3.5/8.6 PAA, respectively) only a ~50% reduction when PAH was on top
and PAH on top (denoted 3.5/7.5 PAH and 3.5/8.6 PAH, respectively). (Fig. 3.8). However, when we assembled PAH at
high pH (8.6) with anionic PAA at low pH (3.5), the results were somewhat different. Since the
pka of PAH is ~pH 8.8171, PAH deposited at pH 8.6 contains uncharged amines, but they are
mostly titrated and compensated for when the film is immersed in the low pH polyanion solution.
Thus, the PAA/PAH 3.5/8.6 system with PAA on top was also ineffective in killing bacteria.
When PAH is the top layer, there is one layer uncompensated, leading to excess positive charge
at physiological pH conditions. Rose Bengal staining indicated a clear difference between the
PAA-topped and the PAH-topped samples. Very little staining was visible on the PAA topped
samples (absorbance = 0.08 after 15 min), while a slightly darker stain was visible on the PAH
topped samples (absorbance = 0.14 after 15 min). Unpublished gold nanoparticle formation
studies showed that consistently higher amounts of gold precursors bound to the PEM when
PAH was the top layer in both the PAA/PAH 3.5/7.5 and PAA/PAH 3.5/8.6 PEMs, again
proving that more free amines are present in the PAH topped case. Swelling measurements in pH
6 water indicated that the PAH-topped system swelled slightly more (~42% swelling) than PAAtopped film (~30% swelling). This could be a result of increased swelling at the top layer in the
75 | P a g e PAH-topped film when the pH is lowered to neutral
pH ranges, an indication of chain mobility. See
Appendix A for additional characterization of the
PAA/PAH system. The PAH-topped film showed
airborne antibacterial activity (~log 2.5 kill) against
non-biofilm forming S. epidermidis without further
Figure 3.9. (0.19) Airborne antibacterial assays on modification or post-assembly processes (Figure
PAA/PAH films assembled at pH 4.0, with the 3.8). Further experimentation reinforced the
addition of one, two or three bilayers of (PAA/PAH 4.0/8.6). All PEMs are PAH topped.
hypothesis that the antibacterial activity in the
PAA/PAH 3.5/8.6 systems was a result of the high deposition of PAH in the last few layers.
PAA/PAH PEMs were assembled with both polymers at pH 4.0 and compared to films that
included one, two or three bilayers of PAA/PAH (4.0/8.6) at the surface. As can be seen in
Figure 3.9, the addition of even one layer of PAH at pH 8.6 drastically reduces non-biofilm
forming S. epidermidis airborne bacterial viability. After addition of two layers of PAH at pH
8.6, the film is essentially as antimicrobial as the film in which all bilayers were assembled with
PAH at pH 8.6.
3.4 Discussion and Conclusion
PEMs are emerging as convienent and versatile biomaterial coating. For instance, poly(Llactic acid), a biocompatible biodegradeable synthetic polymer, has been coated with silverloaded PEMs that improved surface hydrophilicity, hemocompatibility, cytocompatibilty and
antibacterial activity31. Stainless steel, an important orthopedic implant material, has been
76 | P a g e coated with PEMs containing quarternized polyethylamine-silver complex to create a cationic
contact-killing surface with additional silver-leaching capabilities33. Titanium, another
prevalent orthopedic implant material, has been functionalized with chitosan-containing PEMs
that confer cationic contact-killing properties27.
Chitosan, a natural biocompatible cationic polysaccharide, has been incorporated into
many PEMs for antimicrobial functionality27, 49-51. Although the innate cationic charge of
chitosan and its derivatives is known to be biocidal, differences in PEM assembly conditions
have shown to affect the film’s ultimate antimicrobial capabilities22, 23. This has been attributed
to a pH-dependence or ionic strength dependence on the amount of chitosan at the film surface,
differences in film thickness and variations in film rigidity. To our knowledge, no one has
investigated the effect of PEM assembly conditions on the actual mechanism of polycationic
antibacterial action.
Many studies have shown that sufficient free (unbound) cationic charge is capable of
disrupting bacterial membranes, leading to cell death152, 156. In addition, cation mobility has
proven to be an important factor to achieve cationic bacterial killing. In the case of antibacterial
cationic liposomes, a minimum zeta potential of +40 mV and a melted crystalline state (mobile
state) as opposed to a rigid phase were necessary for antibacterial functionality158. In studies of
polycationic cytotoxicity with mammalian cells, which procedes by a bacterial-disrupting
mechanism similar polycationic biocides, the union of high linear charge density and chain
mobility has led to higher levels of cell death. It has been concluded that multiple cation
attachment sites are necessary for membrane permeabilization, and as a result, increasing the
spacing between amino groups and/or rigidity of the chain should decrease the polycationic
bioactivity182. Specifically testing the effects of decreasing linear charge density of polycations
77 | P a g e has confirmed these conclusions; as linear charge density decreased, cytotoxicity decreased82, 157.
Likewise, the arrangement and flexibility of charges has proven to be important. Highly
branched chains or flexible backbones led to higher cytoxic effects than globular or rigid
chains157. Thus, it is clear that polycationic effects on membranes is dependent on the ability of
multiple charges to attach to and interact with the membrane.
As seen in Figure 3.1, the polycations in most PEMs do not exist in a state capable of
causing bacterial death. The cationic charges are mostly bound with anionic charges and rigidly
held in place. Assembly and post-assembly conditions needed to be modified to free sufficient
cationic charge to interact with the bacterial membranes.
The SPS/PAH 9.3/9.3 and the PAA/PAH 3.5/8.6 PAH topped systems illustrate methods
of freeing sufficient cationic charge. By adsorbing PAH at high pH, near or above its pka, the
chains are incorporated into the PEM with many uncharged amines. By acid-treating (in the case
of the SPS/PAH system) or simply lowering the pH to neutral pH ranges (in the case of the
PAA/PAH system), these uncharged amines can be protonated, leading to sufficient positive
charge to induce cell death.
Surface characterizations of the SPS/PAH 9.3/9.3 system supported the hypothesized
mechanism of antibacterial action in these films. The Rose Bengal staining, charge density
calculations and streaming potential measurements verified that the highly swollen acid-treated
film possessed significant positive charge. The swelling measurements. QCM-D analysis and
microparticle attachment experiments verified that the acid-treated film possessed considerable
chain mobility. The high positive charge density and high chain mobility created a film with
78 | P a g e substantial antibacterial and bacteria-resistant properties. Meanwhile, the as-deposited film’s
rigid charges were largely unable to efficiently disrupt the bacterial membrane.
High cationic charge and charge mobility can create bacteria contact-killing surfaces, and
high swelling or low modulus surfaces resist bacterial attachment. SPS/PAH PEMs assembled at
high pH and subsequently acid-treated present highly swollen positively charged surfaces that
efficiently kill Gram positive bacteria and reduce the viability of Gram negative bacteria. These
surfaces also effectively resist bacterial attachment in aqueous solutions. We have therefore
successfully developed antibacterial, bacteria-resistant PEMs with simple pH adjustments during
or after assembly. As shown with the PAA/PAH system, the general approach of manipulating
PEM conditions to create cationic charge and chain mobility can be applied to other polymer
systems. Thus, it is possible that many of the hundreds of layer-by-layer systems that have been
developed for a myriad of applications can be tweaked to create additional antibacterial
functionality.
79 | P a g e CHAPTER 4
4. Post-assembly Alteration of Hydrogen Bonded Polyelectrolyte Mutlilayers
This chapter is a combination of several separate projects dealing with post-assembly
treatments of hydrogen bonded PAA/PAAm multilayers. In the introductory section, a paper by
Van Vliet et al. (Biomacromolecules 7(6), 1990-1995, 2006) is discussed in detail. I am one of
the authors of that paper, contributing by preparing many of the PEM samples and
functionalizing the surfaces with specific peptide ligands. The full paper is included in Appendix
B. The bulk of the chapter addresses morphological changes associated with post-assembly
treatments of the PAA/PAAm system and their effect on bacterial adherence and viability.
Although I was extensively involved in experimental planning and interpretation of the data,
much of the characterization work was performed by an undergraduate student, Allison Kunz,
under my and Daeyeon Lee’s supervision. Kunz was also a helpful partner in the writing of this
section, and had great input into the results, discussion and conclusions sections below. Confocal
microscopy was performed by Albert Swiston. I performed some of the characterization work,
including profilometry measurements, contact angle measurements, QCM-D analysis and
fluorescent microscopy. I also synthesized the fluorescently-tagged PAA and conducted all the
bacteria work.
4.1 Introduction
Surface modification to control cell and bacterial attachment and viability is a rich and
dynamic field. In the area of cell biology, surfaces can also influence cell differentiation and
phenotype. To this end, researchers have specifically engineered surfaces of variable surface
topologies 183 and degrees of interchain crosslinking in a polymeric gel184. They have created
phase separated amphiphilic surfaces185 and functionalized surfaces with cell resistant materials
to direct cell attachment and create cell patterns186. Commonly, surfaces are altered to increase
cytophilicity with the addition of proteins or peptides containing the amino acid sequence ArgGly-Asp (RGD). RGD has been shown to attract and bind integrin receptors on the surfaces of
eukaryotic cells187-193.
80 | P a g e Polyelectrolyte multilayers (PEMs) have figured prominently in the field of surface
engineering for biological applications. For instance, the Rubner and Van Vliet groups have
shown that by adjustment of PEM assembly pH to control swellability and modulus, surfaces can
encourage (low swelling, high modulus films) or inhibit (high swelling, low modulus films) cell
and bacteria attachment1, 62, 70, 144. Likewise, many other groups have developed multilayers
capable of selectively resisting or promoting cell and bacterial attachment23, 27, 65, 68, 72.
Cytophobic surfaces of highly swellable cross-linked polyacrylic acid (PAA) and
polyacrylamide (PAAm) multilayers discourage cell attachment on both planar surfaces and
colloids1, 160. PAA, a weak polyelectrolyte, forms hydrogen bonds with neutral PAAm at low
pH values. If the pH is raised, the carboxylic acids in PAA become deprotonated, and the
hydrogen bonding is disrupted, destroying the film. By thermal annealing or carbodiimide
chemistry160, the film is cross-linked by the formation of imide bonds and can remain intact at
physiological pH ranges.
Berg et al. has shown that the cytophobicity of PAA/PAAm multilayers can be overcome
by surface functionalization with RGD85. Poly(allylamine hydrochloride) (PAH), cationic
polymer that binds to PAA, was attached in patterns to the surface by polymer-on-polymer
stamping. Via a protein cross-linker, RGD ligands were covalently attached to the surface,
enhancing fibroblast adherence. Control peptides of Arg-Gly-Glu (RGE) did not increase
fibroblast attachment and spreading. Throughout their experiments, it was assumed that the
mechanical properties of the underlying substrate were unchanged by the RGD functionalization
process and that the increase in cell adhesion and spreading were a result of biochemical, not
mechanical changes.
81 | P a g e A subsequent study by Van Vliet et al. investigated the effects of different modes of RGD
functionalization of PAA/PAAm on PEM modulus. A series of PAA/PAAm films were topped
with PAH at pH 9 by polymer-on-polymer stamping, adsorption for 30 seconds or adsorption for
15 minutes (the typical length of polymer adsorption during the layer-by-layer dipping process).
Stamping was performed as described in the work of Berg et al85. Patterned PDMS polymer
stamps were soaked in a solution containing 0.05 M PAH and brought into physical contact with
the PAA/PAAm PEM surface for 30 sec. The PEMs were then rinsed with 150 mM / pH = 7.4
phosphate buffered saline (PBS) several times under agitation, and allowed to dry in air.
Addition of RGD or RGE was accomplished as follows: (a) placing the PAH-treated PEMs in
0.5 mM Sulfo-LC-SPDP, a heterobifunctional crosslinker, for 30 min at room temperature, (b)
rinsing twice in PBS for 5 min, (c) placing the PEMs in 0.5 mM peptide solutions (GRGDSPC
or GRGESPC) in PBS for 8 hours at room temperature, and (d) rinsing the PEMs aggressively
with PBS.
82 | P a g e The
results
of
mechanical
testing
by
nanoindentation on the various samples are
presented in Fig 4.1.
Stamping PAA/PAAm
samples with PAH did not change the mechanical
modulus of the films (within error).
Thus, the
increased fibroblast attachment and spreading found
by Berg et al. on RGD functionalized PAH-stamped
samples resulted from chemical functionalization
Figure 4.1. (0.20) Nominal elastic moduli E as and was not the result of increased mechanical
measured by instrumented nanoindentation of surface modified PAA/PAAm polyelectrolyte modulus.
However, adsorption of PAH (30
multilayers. PEMs were indented to a depth of 20 nm using a scanning probe microscope in fluid seconds and 15 minutes) increased the Young’s
(150 mM PBS, pH = 7.4) at room temperature. modulus over two orders of magnitude. Subsequent addition of RGD and RGE to the PAHadsorbed samples did not change the modulus further, indicating that the increase in modulus
occurred at the PAH adsorption step. Further characterization found that the samples with
adsorbed PAH had slightly increased thickness and surface roughness (RMS). In addition,
adsorption of PAH led to an increase in film opacity.
An in-depth investigation of the
mechanisms of PEM stiffening upon PAH adsorption was outside the scope of this work. The
paper concluded by warning that differences in surface functionalization processes (e.g.,
stamping of PAH versus adsorption) can result in large difference in surface properties.
A further investigation of relevant work revealed additional information about the
underlying mechanisms of PEM stiffening. Other research has also studied the effects of
polycation adsorption on multilayers and related single-component hydrogels. Senger et al.
demonstrated that a single polycation layer could extensively interpenetrate an underlying
83 | P a g e electrostatically bound PEM, stiffening the originally compliant layers.131 The Decher group
reported that hydrogen bonded multilayers capped with electrostatic mutlilayers would only
dissolve and release a free standing film when the hydrogen bonded region was greater than a
minimum thickness. This indicated that the electrostatic layers participated in a certain amount
of interpenetration and cross-linking of the hydrogen bonded layers.194 Likewise, Caruso et al.
demonstrated that alternating layers of hydrogen and ionically bonded polymers could be
assembled, and the hydrogen bonded layers could be selectively dissociated by raising the pH.195
However, they reported that the hydrogen bonded layers needed to be a minimum number of
bilayers for the dissociation to occur because of interpenetration of electrostatically bound
polymer layers. Sukhishvili et al. has contributed extensively to the literature on hydrogen
bonded multilayers and derived hydrogels.196 They have developed a system similar to the
PAA/PAAm system, a cross-linked film composed of polymethacrylic acid (PMAA) and aminofunctionalized polyvinyl caprolactam (PVCL)197. Earlier work showed that polycations from
solution adsorbed into uncrosslinked hydrogen bonded films and replaced the neutral polymers,
and that the amount of polycation (or cationic molecules or proteins) correlated linearly with the
degree of ionization of the weak polyanion component.198 Also, adsorption of a polycation onto
single-component polyacid hydrogel capsules created a crosslinking effect that shrank the
capsules and rendered them impermeable to large molecules.199 With these single-component
hydrogels, polycations could be desorbed by either lowering the pH or raising the ionic
concentration of the solution. These studies depended largely on FTIR and confocal microscopy
data; the few surface images that were presented did not provide information regarding the
morphology of the surfaces.
84 | P a g e The following work presents a systematic investigation of the morphological transitions
in cross-linked PAA/PAAm multilayers induced by immersion in aqueous polycationic
solutions. To our knowledge, this is the first study addressing the adsorption of a polycation onto
a cross-linked two-component hydrogel. To uncover the underlying molecular mechanisms, we
study these changes over a large pH range using atomic force microscopy (AFM), fouriertransform infrared spectroscopy (FTIR) and confocal fluorescence microscopy. By exploring the
various results of polycation adsorption, we introduce a new category of controlled surface
modification techniques for hydrogen-bonded multilayers. To illustrate the effects of the
morphological transitions on biological systems, bacteria viability and attachment to a range of
PAA/PAAm films topped with PAH are evaluated and discussed.
4.2 Materials and Methods
4.2.1 Materials
Poly(acrylic acid) (MW 90,000; 25 wt % solution), polyacrylamide (MW 800,000; 10 wt %
solution), linear polyethylenimine (PEI; MW 25,000), and poly(vinylamine hydrochloride)
(PVAm; MW 25,000) were obtained from Polysciences. Poly(allylamine hydrochloride) (PAH;
MW 70,000 and MW 15,000) was obtained from Sigma Aldrich in both standard and FITClabeled forms. Silane-prep (aminoalkylsilane) glass slides were also obtained from Sigma
Aldrich. Deionized filtered water came from an 18.2 MΩ•cm Millipore Milli-Q system.
4.2.2 Sample Preparation
All PEM films were prepared by layer-by-layer deposition with an automated dipping machine
(Zeiss HMS series Programmable Slide Stainer). Each bilayer was deposited by immersion of a
silane-prep glass slide into a 20 mM solution of PAA, followed by three deionized water (DIW)
rinse baths of 2, 1, and 1 minutes each, and then a reiteration of these steps for PAAm. Both the
85 | P a g e polymer solutions and rinse baths were pH adjusted to 3.0. All pH adjustments were made with
1.0 M solutions of hydrochloric acid or sodium hydroxide, and all polymer solutions had a
concentration of 20 mM. After deposition, the samples were thermally crosslinked at 100°C for
24 hours. The notation (PAA/PAAm)n indicates a PAA/PAAm multilayer consisting of n
bilayers.
4.2.3 Polycation Adsorption onto PEMs
The prepared (PAA/PAAm)n multilayer films were immersed in a 20 mM polycationic treatment
solution for 15 minutes. This polycationic adsorption step was immediately followed by two
DIW rinse baths of 5 and 2 minutes each. The rinse baths were pH-adjusted to match the pH of
the treatment solution. After treatment, all samples were allowed to air dry in normal laboratory
conditions.
4.2.4 Surface Morphology Analysis
The surface morphologies of the multilayers were studied in dry state ambient conditions.
Images of all multilayer surfaces were taken using an atomic force microscope (AFM, Digital
Instruments D3100) in tapping mode. The AFM’s software (v5.31r1) was used to find root mean
square (RMS) roughness for each image. (AFM data should be considered accurate to within +/3 nm.)
4.2.5 Thickness measurements
For thickness measurements, the PEMs were scored with the tip of a steel razor blade and a
profilometer (Tencor P-10 surface profiler) scanned across the resulting trench. A typical scan
length was approximately 600 μm. The profilometer software calculated the average step height.
86 | P a g e 4.2.6 Fourier Transform Infrared (FT-IR) Spectroscopy
PAA/PAAm multilayers with 10.5 bilayers were built onto ZnSe crystal substrates for FTIR
analysis. The only difference between these samples and those assembled on silane-prep glass
slides was the necessity of a PAH adhesion layer. Measurements were taken with a Nicolet
Magna 860 Fourier Transform Infrared Spectrometer in transmission mode. Samples were
immersed in pH-adjusted phosphate buffer solutions (BHR) containing less than 10 mM ionic
concentrations for 10 minutes and then allowed to air dry. The instrument’s Omnic software
provided atmospheric and baseline corrections. A previous FT-IR study by Rubner et al.171
identified the peaks at 1711 and 1554 cm-1 as the carboxylic acid groups and carboxylate groups,
respectively, belonging to PAA, and we use a modified form of their equation for calculating the
degree of ionization for weak polyelectrolytes:
4.2.7 Fluorescent Labeling of PAA
Fluorescent tagging of PAA was based on the protocol of Laguecir et al200. Briefly, 0.1 g
Rhodamine 123 (Sigma) was mixed with 10 mL of N,N-Dimethylformamide (DMF) (VWR). In
a separate flask, a total of 50 mL of 50 mM 1-Ethyl-3-(3-dimethylaminopropyl)carbodiimide
Hydrochloride (EDC) (Thermo Scientific) and 100 mM N-hydroxysuccinimide (NHS) (Thermo
Scientific) in water was prepared. PAA (8 g) was added to the EDC/NHS solution, and the
mixture was adjusted to pH 5.5. The Rhodamine/DMF mixture was added and left under
agitation in dark conditions overnight. The mixture was completely dried in a vacuum and
resuspended in water at pH 9.0. Five extractions with CH2Cl2 were performed. The solution
87 | P a g e was then dialyzed three times for at least 8 hours each with sterile deionized water (once at pH 3
and twice at pH 5.5) and dried under vacuum.
4.2.8 Fluorescence Microscopy
Confocal fluorescence microscopy was performed with a Zeiss LSM laser scanning microscope
with a W Plan-Apochromat 63x objective and 488 nm excitation wavelength. Detailed
specifications can be found in the original image files; please see the Supplemental Information
for image file location.
4.2.9 Quartz Crystal Microbalance with Dissipation
PAA/PAAm films were assembled in-situ on SiO2 coated crystals (Q-Sense) in an E4 Q-Sense
Quartz Crystal Microbalance with Dissipation (QCM-D). One layer of PAH at pH 3.0 was
adsorbed before film build-up to improve adhesion with the substrate. All polymers were
adsorbed at pH 3.0 with a concentration of 0.01M and a flow rate of 0.1 mL/min. Between each
polymer deposition, rinse water at pH 3.0 was flowed through the system at 0.1 mL/min.
4.2.10 Waterborne Bacteria Attachment Assays
Bacterial attachment was assayed as described previously144. Monoclonal Staphylococcus
epidermidis (S. epidermidis, ATCC # 14990) was grown overnight in LB-Miller broth (VWR).
The incubation was centrifuged at 2700 rpm, decanted and resuspended in 150 mM NaCl PBS
(VWR) three times (10 min, 5 min, 5 min). The final resuspension was diluted to a density of
109 cells/mL measured via optical density. This resuspension was then diluted in sterile
deionized water to 107 cells/mL. The PEMs were placed in the water dilution for 2 hrs at room
temperature and rinsed thoroughly in 3 water baths. Samples were incubated under 1% LB agar
(VWR) gel overnight, and colonies counted to determine the ability of viable bacteria to attach to
each sample. Samples with few colonies were hand counted. For densely populated slides, at
88 | P a g e least 10 digital images per sample were acquired with a 4x objective using an inverted optical
microscope (Leica) and the total number of colonies per slide was extrapolated.
4.2.11 Airborne Antibacterial Assays
Airborne assays were based on the protocol discussed by Klibanov et. al170. Bacterial
suspensions of 109 cells/mL were prepared as described for the waterborne attachment assays
(centrifuged and optical density measured in PBS) and diluted to 2 x 106 cells/mL in sterile
deionized water. Fine mists (d~ 5-100 μm) were sprayed at 1 μl/cm2•spray for a total of 6×103
bacterial cells/cm2. Samples were immediately placed in 100mm Petri dishes (VWR) and
covered with 1% LB agar gel overnight at 37C. Resulting colonies were counted as described
above.
4.3 Results and Discussion
4.3.1 Morphology dependence on pH
The work by Van Vliet et al. suggested that PAA/PAAm multilayers undergo significant
changes when immersed in a solution containing PAH at pH 9. AFM images of the surface
morphologies induced by the adsorption of PAH onto PAA/PAAm thin films over a large pH
range revealed a number of interesting morphologies (Figure 4.2). Thermally crosslinked
(PAA/PAAm)10.5 multilayers were immersed for 15 minutes in PAH solutions at pH 1.5, 3.0, 5.0,
7.0, 8.6, 10.0, and 11.3, and allowed to air dry. Root mean square (RMS) roughness and overall
film thickness data measured by profilometry are provided in Figure 4.3. Ellipsometry
measurements of this system were hindered by the opacity of the films and were therefore
unattainable. Under acidic conditions (pH 1.5), no morphological differences are seen between
treated and untreated films. From pH 3.0 to 8.6, there is a gradual transition from a wrinkled
texture with low RMS surface roughness to a dramatic honeycomb texture with higher
89 | P a g e roughness. Film thickness also increases, as expected from the increasing ionization and
expanded coil nature of PAA and from the adsorption of more PAH. For highly basic solutions,
pH 10.0 and 11.3, the films show increasing coalesced regions of hydrophobic PAH. Overall
roughness decreases significantly, but film thickness spikes at pH 10.0 and shows huge variance
at pH 11.3, perhaps as a result of large amounts of hydrophobically associated PAH depositing
on the surface.
Samples immersed in either deionized water or a 10 mM NaCl solution that had been
adjusted to pH 8.6 remained as smooth as the untreated films; these control samples confirmed
that the adsorption of PAH induced the surface textures. In additional experiments, we also
adsorbed PAH at pH 8.6, but varied the temperature (44°C, room temperature (~25°C) and 3°C),
time (5 sec, 30 sec, 15 min, 30 min, 3 h, 8 h, 17 h and 98 h), or molecular weight of the PAH
Figure 4.2. (0.21) AFM images of (PAA/PAAm)10.5 multilayers (a) as prepared, and (b) after immersion in 20 mM PAH solutions adjusted to pH 1.5, (c) pH 3.0, (d) pH 5.0, (e) pH 7.0, (f) pH 8.6, (g) pH 10.0, and (h) pH 11.3. All images are of dry films, 15 μm x 15 μm x 0.2 μm, with the z‐axis color scale given on the right. 90 | P a g e immersion solution (MW= 15,000 or 70,000). None of these variations resulted in significantly
different surface morphologies than sample in Figure 4.2f. The honeycomb morphology was
evident even after immersion in a pH 8.6 PAH solution for only five seconds, agreeing with the
report of rapid film changes by Van Vliet et al.
Figure 4.3. (0.22) Plot of (PAA/PAAm)10.5PAH dry film thicknesses and dry‐state root mean square roughness values, both in nanometers, as a function of the pH of the PAH immersion solution. The gray dashed line indicates the thickness of the cross‐linked PAA/PAAm film before PAH adsorption.
4.3.2 Fluorescence Microscopy
Fluorescent labeling of polymers allows for simultaneous spatial and compositional
analysis of PEMs. To determine the composition of honeycomb ridges in PAH topped
PAA/PAAm multilayers, thick multilayer samples were prepared using the same procedure as
before (LbL assembly, thermal crosslinking, and adsorption of PAH) with Rhodamine 123labeled PAA and FITC-labeled PAH. Confocal microscopy images of these fluorescently labeled
(PAA/PAAm)60.5 multilayers after polycation adsorption at pH 8.6 are presented below. The
bright field image in figure 4.4a exhibits the honeycomb features seen previously in Figure 4.2f.
91 | P a g e Figure 4.4. (0.23) Confocal fluorescence microscopy images of (PAA/PAAm)60.5 multilayers fabricated using Rhodamine 123‐ labeled PAA and FITC‐labeled PAH. IR filters show (a) bright field (b) FITC fluorescence and (c) Rhodamine 123 fluorescence in the honeycomb ridges. Magnification is 63x. (It should be noted that the length scale of the honeycomb features correlate linearly with film
thickness.) Figures 4.4b and 4.4c, which were acquired using specific wavelength filters, clearly
indicate that PAH and PAA are concentrated in the honeycomb ridges. Control samples with
fluorescently tagged PAA or PAH, but not both, confirmed these observations.
The fluorescent images clearly indicate that PAH complexes with PAA via electrostatic
attraction upon adsorption. This result agrees with previous work by Sukhishvili et al. in which a
polycation was adsorbed onto capsules of uncrosslinked hydrogen bonded multilayers198. In
their work, the strong electrostatic attraction between the weak polyelectrolytes resulted in
replacement of the neutral polymer with the polycation (as opposed to this work where the
neutral polymer is retained – discussed below). In a similar study, polycations were adsorbed
onto single-component cross-linked polyacid hydrogel capsules, resulting in a decrease in
capsule diameter and impermeability to dextran molecules199. These observations suggested that
the polyelectrolytes formed a dense network of ionic bonds that effectively crosslinked the
multilayer. In the PEMs studied here, the initial hydrogen bonded multilayers were covalently
92 | P a g e cross-linked, limiting the mobility of the PAA molecules and preventing the removal of PAAm.
The retained PAAm does not interact with PAH and remains in a highly swollen, hydrophilic
state. As a result, regions of hydrophobic ionically complexed PAA and PAH phase separated to
form the honeycomb ridges seen in Figure 4.4. As seen in the side-view images, the PAA-PAH
ridges span the entire thickness of the film. Contact angle measurements confirmed that the
hydrophobic character of the multilayers increased significantly after PAH asdsorption at pH 8.6.
The unmodified cross-linked PAA/PAAm film had an advancing contact angle of ~61 and an
advancing contact angle of ~96 after adsorption of PAH at pH 8.6. It should be noted that
previous quartz crystal microbalance (QCM) work found that the ratio of PAA to PAAm in these
multilayers is approximately 23:771. In the course of these studies, in-situ QCM with dissipation
(QCM-D) experiments of the assembly process verified that, compared to PAA, significantly
more PAAm is incorporated into the multilayer [Figure 4.5]. The large amount of PAAm helps
explain the presence of large hydrophilic regions that do not contain PAA or PAH (as seen with
fluorescent microscopy) in the middle of the honeycomb structures seen here.
93 | P a g e Figure 4.5. (0.24) QCM‐D analysis of PAA/PAAm film buildup. All polymers were adsorbed at pH 3.0 with a concentration of 0.01M and a flow rate of 0.1 mL/min. Between each polymer deposition, rinse water at pH 3.0 was flowed through the system at 0.1 mL/min Both optical and confocal images of 100.5 bilayer films verified the presence of PAH in
multilayers regardless of immersion pH. The stacked side-view confocal microscopy images in
Figure 4.5 also suggested that PAH incorporates throughout the thickness of the films. However,
interpretation of the confocal data is complicated by the observation of an exponential-to-linear
transition in multilayer growth between 10 and 20 bilayers. Such a transition indicates a
fundamental change in the molecular assembly of the film, and images of 100.5 bilayer films
after polycation adsorption did not always exhibit the same surface morphologies as their 10.5
bilayer counterparts. Our observations suggest that the film thickness also influences surface
morphology development, perhaps due to the overall amount of charge presented to adsorbing
polycations.
94 | P a g e 4.3.3 Degree of Ionization of Weak Polyelectrolytes.
Solution pH controls the degree of ionization (DOI) of weak polyelectrolytes, and it
therefore dictates their conformation and interactions. To understand the molecular mechanisms
behind the polycation induced changes in multilayer surface morphology, a familiarity with the
pH-dependent behavior of the polyelectrolytes involved is necessary.
Much of the relevant information describing PAH in solution and in multilayers can be
found in the literature. The amine group of PAH has a reported pKa value of 8.8 in an aqueous
solution171, but incorporation into an electrostatically bonded multilayer can raise the pKa to
above 10.0 169. PAH and similar polycations undergo conformational changes at different
degrees of ionization201, 202. Well below the pKa, all of the amine groups are protonated and
carry a positive charge. The electrostatic repulsion caused by like charges results in an expanded
chain conformation to maximize the distance between ammonium groups. When the
polyelectrolyte is only partially ionized at medium pH values, the charged ammonium groups
continue to repel each other while the neutral amines begin to agglomerate into hydrophobic
beads along the polymer chain. The size of the hydrophobic beads in this “pearl necklace”
structure varies inversely with the degree of ionization. At sufficiently high pH, the degree of
ionization falls below a critical point, and the polycations collapse into a globular conformation.
Previous work by Park et al. documents the pearl necklace behavior of PAH in solution and
confirms the presence of globular conformations at pH 9202. The AFM images in Figure 4.2 also
indicate a transition in that pH region, specifically between pH 8.6 and 10.0. The lower RMS
roughness and increasingly variable thickness measurements of these samples in Figure 4.3 are
consistent with the presence of large regions of coalesced PAH globules adhered to the surface
of the film. PAH at high pH minimally interpenetrates and cross-links the PAA/PAAm hydrogel.
95 | P a g e Instead, large “clumps” of hydrophobic polymer are being deposited on the surface. The
apparent coil-to-globule transition point between pH 8.6 and 10.0 suggests that
Figure 4.6. (0.25) FT‐IR data from (PAA/PAAm)10.5 films on ZnSe crystals. (A) Offset spectra illustrate the protonation of PAA’s carboxylic acid groups with increasing pH. (B) S‐curves showing the DOI of PAA (squares) calculated using FT‐IR data and equation 1, and the DOI of PAH (triangles) from the literature[16]. 96 | P a g e the adsorbed PAH behaves much more like a polymer in aqueous solution than in an ionically
bonded multilayer system; otherwise the transition from pearl necklace to globular
conformations would have occurred at a higher pH. Therefore, we can extend this
conformational knowledge of PAH in solution, as well as the degree of ionization curve provided
by Rubner et al.171, to this research.
The carboxylic acid groups of PAA have a pKa of approximately 6.5 in solution and less
than 3.0 in an ionically bonded multilayer system such as PAA/PAH171. To our knowledge,
characterization of the pKa of PAA in the PAA/PAAm hydrogen bonded system is reported for
the first time here. FTIR analysis of the (PAA/PAAm)10.5 multilayer over a wide pH range,
shown in Figure 4.6a, tracks the characteristic peaks of the charged and uncharged carboxylic
acid groups in PAA.
The degree of ionization (DOI) of PAA at different pH ranges was calculated using
Equation 1. These data, along with the literature data for PAH171, are plotted in Figure 4.6b. The
PAA DOI calculations assume full protonation at pH 1.5 and full ionization at pH 11.3, and give
an approximate pKa of 7.8 for PAA in this hydrogen-bonded system. These findings are
consistent with previous work by Sukhishvili et al. on similar hydrogen-bonded systems, where
incorporation of a polyacid into a hydrogen bonded multilayer significantly increased its pKa203.
The PEM stiffening and increased film opacity observed by Van Vliet et al. can now be
explained. Immersion of the multilayer into a pH 9.0 solution of PAH enabled the polycation to
adsorb and incorporate into the multilayer. Although the film was already lightly covalently
cross-linked, the adsorption of PAH created additional ionic crosslinks (Note that Figure 4.6b
indicates that both polyelectrolytes are significantly ionized at pH 9.0). This additional
97 | P a g e crosslinking raised the effective Young’s modulus of the film and lead to phase separation. We
hypothesize that the application of PAH via stamping onto a dry-state film lacked the same effect
due to the lower mobility of the polyelectrolytes in the dry state and the lower DOI of PAA (the
films were assembled at pH 3.0, where PAA is less than one percent ionized).
4.3.4 Adsorption of Other Polycations onto the PAA/PAAm Multilayer
Adsorption onto PAA/PAAm PEMs was repeated with different polycations, revealing
additional information about the mechanism of the morphological transitions.
Figure 4.7
presents AFM images of (PAA/PAAm)10.5 multilayers that had been immersed in pH 8.6
solutions of either LPEI and PVAm, whose schema are provided to the right of their respective
images.
These polycations are structurally dissimilar to PAH: PVAm lacks a methyl group
between the amine and the polymer backbone, and LPEI has nitrogen incorporated directly into
the polymer backbone. The PVAm-adsorbed multilayer data found in Figure 4.7b exhibits both
honeycomb- and globular-like surface morphologies similar to Figure 4.2f and 4.2g (PAH
adsorbed at pH 8.6 and 10.0). These results correlate well with comparisons of DOI between
these polymers and PAH at various pH values: PVAm is roughly 40% ionized at pH 8.6201, a
DOI that lies between that of Figure 4.2f (approximately 55% ionized) and 4.2g (30%). Similar
to PAH, PVAm has been observed in extended, pearl necklace, and globular structures, and both
polymers near their coil-to-globule transition point at approximately the same degree of
ionization. LPEI is roughly 20% ionized at pH 8.6204, and adsorption onto the PAA/PAAm
multilayer resulted in a slightly wrinkled surface morphology similar to PAH-adsorbed samples
at low pH (Figures 4.2c and 4.2d), which have DOI’s of less than 5%. LPEI has not been
observed in the pearl necklace conformation and is a much less hydrophobic than PAH and
98 | P a g e PVAm. The difference in DOI between
the PAH and PEI samples with similar
appearance suggests that
hydrophobicity of the adsorbing
polycation also has some influence on
the resulting surface morphology. The
results of these experiments indicate
Figure 4.7. (0.26) AFM images of 10.5 bilayer PAA/PAAm multilayers immersed in 20 mM, pH 8.6 solutions of (a) linear polyethylenimine, and (b) poly(vinylamine hydrochloride), accompanied by schemes of the respective polymers.
that changes in the surface morphology
of PAA/PAAm multilayers can be
induced by many different polycations,
and the phase separation induced by ionic crosslinking appears to be commensurate with the
degree of ionization of the adsorbing polycation.
4.3.5 Bacterial Attachment and Viability on Different Morphologies.
The adsorption of polycations onto a PAA/PAAm multilayer significantly changes the
properties of the resulting film. An analysis of bacterial attachment and viability on PAA/PAAm
film with PAH adsorbed at various pH values offers insight into the effect these changes have on
surface interactions with biological systems. Airborne antibacterial experiments assayed the
biocidal nature of a particular surface. Waterborne experiments added information about
aqueous bacterial attachment.
Native cross-linked PAA/PAAm films are highly swellable, swelling about 350% in pH
7.4 phosphate buffered saline63. This swelling corresponds to a low modulus film of about 2.4 x
105 Pa, which, based on the studies detailed in Chapter 2, was expected to limit bacterial
attachment205. As seen in Figure 4.8a, waterborne attachment studies of S. epidermidis on
99 | P a g e PAA/PAAm films showed markedly
fewer bacteria colonies than glass
controls. Airborne antibacterial studies
showed that as-assembled PAA/PAAm
and PAA/PAAm with PAH adsorbed up
to pH 8.6 do not exhibit any bacteriakilling capabilities (Figure 4.8b). Thus,
in the waterborne studies, the reduction
in colony density on the PAA/PAAm
films and those with PAH deposited at
low pH is a result of low film modulus
discouraging stable bacterial attachment
and not a result of any biocidal film
properties. As the pH of the adsorbed
PAH is raised, the polycation begins to
Figure 4.8. (0.27) Bacteria experiments with S. epidermidis on PAA/PAAm films with PAH absorbed at various pH values showing (A) waterborne bacterial attachment and (B) airborne antibacterial results.
extensively crosslink and stiffen the
film. This was measured directly in the
work of Van Vliet et al., where the
modulus increased over two orders of magnitude when PAH was adsorbed at pH 9.0 onto a
PAA/PAAm film206. As seen in Chapter 2, stiffer films lead to higher levels of bacterial
attachment 205. In addition, although some see no correlation, many researchers claim that
increased roughness encourages bacterial adhesion207. Thus, the increase in bacterial attachment
with increasing pH of adsorbed PAH is a result of film stiffening and possibly increased surface
100 | P a g e roughness. As can be seen in Figures 4.8a and 4.8b, in both airborne and waterborne assays, no
bacterial growth was visible on PAA/PAAm films after PAH was adsorbed at extremely high pH
ranges. The airborne antibacterial tests indicated that these films display a high level of
antibacterial activity. As seen from the AFM images in Figures 4.2g and 4.2h, PAH adsorbed at
high pH has a very low degree of ionization and deposits globularly on the film surface. When
these films are exposed to bacteria suspensions at neutral pH ranges, many of the unbound
ammonium groups become protonated and create a high degree of positive charge at the surface.
Other studies have shown that high cationic charge density can disrupt the counterions on
negatively charged bacterial cell membranes, leading to cell death 152, 156. Thus, in the
waterborne attachment experiments, no attachment was seen with high PAH deposition pH
values because bacteria that interacted with the surface were killed. The results here support the
research in Chapter 3 describing polycationic antimicrobial properties created by adjusting
assembly and post-assembly pH conditions. In Chapter 3, post-assembly acid-treatment of an
SPS/PAH PEM assembled at high pH created high density, mobile cationic charge that disrupted
bacterial cell membranes on contact. Here, a similar effect is created by a different postassembly process: adsorption of PAH at high pH onto a PAA/PAAm multilayer.
4.4 Conclusions
Our work has investigated a polycation-induced morphological transition in hydrogen
bonded multilayers. We have shown that the effects of polycation adsorption depend on the pH
of the polycation immersion solution: pH controls the degree of ionization of the
polyelectrolytes, and therefore, their conformations and degree of ionic cross-linking. The
hydrogen bonding in the initial PAA/PAAm multilayer increases the pKa of PAA relative to an
aqueous solution. The initial multilayer is smooth, complaint, and highly hydrated, but the
101 | P a g e incorporation of a polycation can drastically change these properties. The degree of ionic
crosslinking, which induces new morphologies and increases stiffness, depends on the degree of
ionization of the two weak polyelectrolytes. Electrostatic complexing of PAA and PAH can
create hydrophobically ionically bonded regions that phase separate from the hydrophilic PAAm.
At very basic pH, coalesced hydrophobic clusters of PAH adsorb and create a surface that may
display strong positive charge.
The practical effects of these changes were investigated with a series of bacteria studies.
Van Vliet et al. and others have demonstrated that PAH-stiffened PAA/PAAm films encourage
higher cell attachment36, 206. Here, we show that the same attachment phenomenon occurs with
bacteria: as the multilayer becomes stiffer, more bacteria form stable attachments to the surface.
We also demonstrate that PAH deposited at high pH creates an extremely effective antibacterial
surface. The bactericidal properties probably result from increased cationic charge at the film
surface that is able to disrupt the negatively charged bacterial cell membrane.
This research provides several new avenues for future investigation. For example,
bioinert films made of simple PAA/PAAm multilayers may face problems in vivo due to
interactions with polycations naturally present in the body. A biomedical device coated with a
compliant multilayer could adsorb polycations and stiffen, promoting infection rather than
inhibiting it. Also, the bactericidal surfaces merit further exploration: it would be useful to
determine how long the antibacterial properties last in different environments, as well as observe
the adhesion and behavior of eukaryotic cells when PAH is adsorbed at high pH. Nevertheless, it
is clear that polycation adsorption onto a highly hydrated polymer surface can create films with
vastly different properties that can be exploited in a biological setting.
102 | P a g e CHAPTER 5
5. Initial Biofilm Growth on Polyelectrolyte Multilayers
The majority of the experiments in this section were performed by various undergraduates under
my supervision. As part of her undergraduate thesis project, Maricela Delgadillo created a
biofilm assay protocol and tested PAA/PAH samples for biofilm reduction. Kathleen Tompkins,
as an undergraduate research assistant, performed most of the biofilm assays on the SPS/PAH
system. With the help of me and Albert Swiston, Tompkins also assembled the PEMs and
conducted the majority of the bacteria experiments with the lift-off PEM system.
5.1 Introduction
The ultimate intended goal of many antibacterial surfaces is to aid in the prevention of
biofilm formation by inhibiting initial attachment of bacteria to a surface20, 27, 50, 144. The bacteria
challenges described in previous chapters mainly involved isolated bacteria suspensions in sterile
water or ionic solutions (e.g., phosphate buffered saline). There are some known instances in
which biofilms develop in these low nutrient conditions. For example, biofilms have been
known to slowly grow on the inner surface of water pipes208. Thus, the bacteria challenges
described above may prove to be good models for some biofilms.
However, many examples of unwanted biofilm formation occur in high nutrient
conditions, including during food processing and the implantation of biomedical devices. In this
work, we increased the bacteria challenge by testing the ability of the bacteria-resistant and
bacteria-killing PEMs developed in chapters 2 and 3 to resist biofilm formation in highly nutrient
rich environments. In addition, we examined the possibility of creating PEMs capable of
sloughing an entire surface if a biofilm successfully formed, thereby exposing a new unfouled
103 | P a g e surface. Before discussing our biofilm-formation protocol and PEM assays, it is important to
begin with a brief overview of biofilm structure, development and impact on mankind.
5.1.1 An Overview of Biofilms
Biofilms are a major component of world-wide bacterial biomass,209, 210 and their
existence often has deleterious effects. In the industrial setting, biofilms on metallic surfaces can
cause biocorrosion, leading to clogging and damage costing millions of dollars in repairs19. In
medicine, biofilms cause extensive medical device and implant failure, growing on catheters,
prosthetic heart valves, pacemakers, shunts, contact lenses, orthopedic devices, etc. They are
also the source of many chronic infections not related to medical devices: tooth decay, cystic
fibrosis pneumonia, prostate gland inflammation and more210.
There are, however, applications that take advantage of biofilm growth. The ability of the
biofilm polymer matrix to entrap minerals and nutrients is exploited in wastewater treatment61.
Biofilms are also used in the fermentation of ethanol and vinegar and the biosynthesis of
polymers211. Dense layers of bacteria and biofilms have also been useful in converting organic
waste and renewable biomass to electricity212.
Figure 5.1. (0.28) Schematic of biofilm formation: attachment of bacteria Æ proliferation and excretion of polysaccharides Æ mature biofilm releasing planktonic cells.
104 | P a g e OVERVIEW OF BIOFILM STRUCTURE: Bacteria exist in two general populations: planktonic
(freely floating) or sessile (attached to a surface). Bacteria in the sessile state can exist as
individuals or small colonies of identical bacteria or as part of complex structures of biofilms.
The highly hydrated matrix or “slime” in which the bacteria microcolonies in biofilms are
embedded is composed of polysaccharides excreted from sessile bacteria, components of dead
bacteria and molecules incorporated from the environment. In fact, 73-98% of the biofilm is
composed of noncellular material61. Individual biofilms can be comprised of many different
bacterial strains, and different phenotypes within a particular strain. They are heterogeneous and
dynamic well-organized communities with channels and pores to distribute nutrients throughout
the biofilm and dispose of waste. Once formed, the biofilm can slough pieces of biofilm or
individual cells that can start new biofilms or infections in different locations.
BIOFILM DEVELOPMENT: Bacterial attachment to a surface is the first step in biofilm
formation, followed by proliferation or recruiting of additional planktonic bacteria18. As
described in previous chapters, bacterial adhesion is mediated by a number of factors and is not
totally understood. Depending on environmental conditions and bacterial strain, many
parameters can promote or inhibit bacterial attachment, including surface roughness,
hydrophobicity, surface conditioning, surface modulus, etc19. In some cases, a first layer of
bacteria and excreted polymers can serve as a substrate for additional bacterial adhesion.
Once attached to the surface, gene expression is modified and bacteria excrete
polysaccharides that protect and support the biofilm and trap components from the environment,
thereby incorporating additional species into the matrix. The continual development of a biofilm
is directed by a quorum-sensing mechanism, a cell-to-cell signaling process by which bacteria
105 | P a g e are aware of the cell density in their vicinity. This may occur through critical concentrations of
species normally released by cells (e.g., lactones in many Gram negative bacteria and peptides in
many Gram positive bacteria) or by sensing the relative proportion of dead cells in the biofilm18.
The ultimate structure of the biofilm is also determined by local conditions. For instance,
biofilms have been found to be flat or mushroom- shaped depending on the availability of
nutrients210.
Fully formed biofilms can enter a death phase in which enzymes are produced by the
bacteria themselves to break down the polymers holding the biofilm together. At the same time,
genes expressing flagella and other necessary machinery for mobility are up-regulated213.
BIOFILMS LEAD TO BACTERIA PROTECTION: The biofilm structure offers powerful
protection for prokaryotes. For instance, the polymer matrix acts as a barrier to antibiotics,
disinfectants and changing environmental conditions213. In some cases, the decreased oxygen
availability in the biofilm slows metabolic activity, creating metabolically inactive bacteria that
are less susceptible to antibacterial agents214. Many other protective mechanisms exist in biofilm
communities, but an extensive discussion is outside the scope of this brief overview210. The end
result, however, is that our current methods of combating bacteria are often ineffective against
biofilms, and therefore, we tend to kill the sloughed planktonic bacteria without killing the
source210. Thus, the need to prevent early stages of biofilm formation (e.g., bacterial attachment
to a substrate) becomes evident.
5.2 Biofilm growth on PEMs capable of resisting colonization of individual bacteria
As stated above, the first phase of biofilm analysis involved assessing the resistance of
two previously studied PEMs to bacterial attachment in aggressive biofilm-forming conditions.
106 | P a g e The potentially biofilm resistant PEMs tested in these experiments included bacteria resistant
PEMs comprised of poly(acrylic acid) (PAA) and poly(allylamine hydrochloride) (PAH)
assembled at pH 2.0, and antibacterial and bacterial-resistant PEMs comprised of (sodium 4styrene sulfonate) (SPS) and PAH assembled at high pH and subsequently acid-treated. These
PEMs successfully reduced or eliminated bacterial attachment in low-nutrient aqueous solutions.
In particular, the SPS/PAH PEM completely inhibited bacterial attachment of a biofilm-forming
strain of S. epidermidis, albeit under conditions that would not form a biofilm during the timelengths tested. Thus, these films were prime candidates to evaluate biofilm prevention or
reduction in high nutrient conditions. As control samples, plain glass slides, the rigid and
bacteria adherent PAA/PAH system assembled at pH 6.5 and the non-acid treated SPS/PAH
system assembled at high pH were included.
5.2.1 Materials and Methods
PEM assembly: PEMs were assembled as discussed above (Chapters 2 and 3). 10-2 M solutions
in 18 MΩ Milli-Q water of PAA (poly(acrylic acid); Mw = >200 000 g/mol; 25% aqueous
solution; Polysciences), PAH (poly(allylamine hydrochloride); Mw = 70 000 g/mol;
Polysciences) or SPS (sodium 4-styrene sulfonate, Mw = 70 000 g/mol, Aldrich) were pH
adjusted using 1M HCl and NaOH. PAA/PAH multilayers, assembled with both polymers at pH
2.0 (9.5 bilayers) or pH 6.5 (49.5 bilayers) were assembled PAA first on aminoalkylsilane coated
glass slides (sigma). SPS/PAH PEMs were assembled PAH first on plain glass slides (VWR).
Between each sequential immersion in a polymer bath, the samples were rinsed in three water
baths at ~ pH 5.5 for the PAA/PAH films and pH 9.3 for the SPS/PAH films. PAA was the final
layer in all PAA/PAH films, and PAH was the top layer in all SPS/PAH films. For acid-treated
107 | P a g e SPS/PAH films, samples were immersed in a solution at pH 2.5 for 15 minutes, followed by a
water rinse at pH 5.5.
PEM assembly for the lift-off systems followed the procedure outlined by Swiston et al81.
PEMs were assembled on aminoalkylsilane coated slides (Aldrich). An adhesive region was
deposited consisting of 15.5 bilayers of poly(diallyldimethylammonium chloride) (PDAC,
Aldrich, M=200-350kDa in 20% aqueous solution) and SPS at pH 4.0 with 0.1 M NaCl in the
polymers baths. Next, an 80.5 bilayer release region was constructed of hydrogen bonded
poly(methylacrylic acid) (PMAA, PolySciences, M=100kDa) and poly(Nisopropylacrylamide)
(PNIPAAm, Polymer Source, M=258kDa) with polymers and rinse baths adjusted to pH 3.0.
The substrates were topped with an additional supportive film of fluorescein-labeled PAH
(FITC-PAH, Aldrich, M=70kDa) and iron oxide magnetic nanoparticles (Fe3O4 NP, 10nm
diameter, Ferrotec EMG 705) stabilized with an anionic surfactant (10 bilayers, FITC-PAH
solution at pH 3.0, the nanoparticle suspension at pH 4.0 and the rinse baths at pH 3.0).
Initial Biofilm Growth: The experimental procedures for biofilm growth were modeled on
those described by Stepanovic et al215. Trypticase Soy Broth (TSB) (BD) was inoculated with a
biofilm forming monoclonal strain of Staphylococcus epidermidis (S. epidermidis, ATCC #
35894), and incubated for 18 ± 1 hour at 37°C with shaking agitation. Polyelectrolyte
multilayers were sterilized with 70% ethanol and left to air dry. Two 20 mL aliquots of the
primary culture were centrifuged at 2700 RPM for ten minutes at 4°C (Centrifuge 5804 R). The
TSB was decanted, and the bacteria pellet was re-suspended into two 15 mL aliquots of TSB.
This solution was centrifuged a second time for five minutes, the TSB was decanted, and the
bacteria pellet was re-suspended into 10 mL of TSB. The bacterial solution was diluted to an
108 | P a g e optical density of one at a wavelength of 540 nm, corresponding to a cell density of 109 cells/mL.
10 mL of bacterial solution was added to 90 mL of Trypticase Soy Broth corresponding to a final
bacterial density of 108 cells/mL. Glucose monohydrate (Biochemika) was added to the bacteria
solution for a total concentration of 2% (v/v) glucose. The samples were placed in 100mm petri
dishes (VWR), coated with 15 mL of the bacterial solution and incubated at 37°C for 2 h. The
short incubation time did not give the bacteria enough time to develop into a mature biofilm, but
was sufficient for initial biofilm formation. After initial biofilm formation, the excess bacterial
solution was removed, and the slides were rinsed twice in Phosphate Buffered Saline (PBS, from
VWR). The bacteria were fixed with 15 mL of Methanol (Mallinckrodt Chemicals) at room
temperature for 20 minutes. Finally, the excess methanol was drained and the samples were left
to dry overnight. All samples were tested in triplicate.
Biofilm Analysis: Fixed biofilm samples were stained with 20 mL 2% Gram Crystal Violet
(BD) for 20 minutes, followed by careful rinsing with deionized water. Not all the samples were
stained, however, because it was found that the stain was adsorbed onto the PEM, making it
difficult to distinguish between the polymers and the bacteria. Optical microscopy images were
acquired (Axiovert 200 Zeiss microscope). Atomic Force Microscopy (Nano Scope Control,
Digital Instruments) was performed on the PAA/PAH samples.
Airborne Bacteria Application and Liftoff: For initial tests with the lift-off PEM systems,
samples were sprayed with non-biofilm forming S. epidermidis (ATCC #: 14990) with a gas
chromatography sprayer as described in Chapter 3 (see methods section). Films were then placed
in phosphate buffered saline (PBS, VWR) at 4°C for 30 minutes to allow the hydrogen bonded
region to dissociate and release the top film. The PEMs were placed in 100mm Petri dishes
109 | P a g e (VWR) and coated with a slice of 1% LB agar (VWR) gel overnight. Resultant colony density
was determined as described in Chapter 3. All samples were tested in triplicate.
Biofilm Growth and Liftoff: Trypticase Soy Broth (TSB) (BD) was inoculated with a biofilm
forming monoclonal strain of Staphylococcus epidermidis (S. epidermidis, ATCC # 35894) and
incubated for 18 ± 1 hour at 37°C with shaking agitation. PEMs were sterilized with 70%
ethanol at pH 3.0 and left to air dry. The bacterial culture was diluted 1:100 in TSB adjusted to
pH 5.5 with 0.1 M NaCl. Samples were placed in 100mm Petri dishes (VWR) and covered with
15mL of the diluted bacterial suspension and 200 μL of glucose. The biofilm was grown for 24
hours at 37°C. The samples were rinsed twice with 15mL MES buffer (2-(Nmorpholino)ethanesulfonic acid) adjusted to pH 5.5. Samples were then placed in PBS at either
37°C or 4°C for 30 minutes. Each sample was tested in triplicate.
5.2.2 PEMs with Varying Mechanical Modulus
As discussed in Chapter 2, PAA/PAH films assembled at low pH greatly reduced
bacterial attachment in comparison to control glass slides or PEMs assembled at pH 6.5. In the
case of low pH assembly, sparsely ionized PAA chains were incorporated into the PEM. At
neutral pH, the uncharged PAA carboxylic acid groups were deprotonated, creating a large
degree of charge-charge repulsion that swelled the film. It was found that this highly swollen,
low modulus film resisted attachment of non-biofilm forming S. epidermidis and E. coli in low
nutrient conditions.
Figure 5.2 shows the bacterial attachment and initial biofilm growth of a biofilm-forming
S. epidermidis strain in high nutrient conditions on PAA/PAH PEMs assembled at pH 2.0 and
pH 6.5. At periods of 24 hours or more, dense biofilms formed under these conditions. In this
case, short times were tested to measure differences in initial biofilm formation. After a two
110 | P a g e hour incubation, there was no clear differences in bacterial attachment between the two PEMs.
The AFM images of these initial biofilms, shown in Figure 5.3, confirmed that there were no
differences between the control aminoalkylsilane-treated glass slide and the PEMs. The AFM
images clearly show clumping of bacteria, which is typical in staphylococci strains. The
different morphologies of the PEM surfaces are also evident in the images, reinforcing the
conclusion that assembly conditions of PAA/PAH films did not affect initial biofilm formation.
Figure 5.2. (0.29) Optical microscopy images of initial biofilm formation on (a) highly compliant PAA/PAH 2.0/2.0 PEMs and (b) rigid PAA/PAH 6.5/6.5 PEMs. Figure 5.3. (0.30) AFM images of initial biofilm formation on (A) aminoalkylsilane treated glass slide, (B) PAA/PAH 2.0/2.0 PEMs and (C) PAA/PAH 6.5/6.5 PEMs. All images are 10μm x 10μm. The amount of adhered bacteria was not affected by the underlying substrate. 111 | P a g e 5.2.3 Antibacterial PEMs
Whereas PAA/PAH 2.0/2.0 films only reduced waterborne bacterial attachment,
SPS/PAH 9.3/9.3 acid-treated films were able to completely resist attachment under the
conditions tested for both biofilm-forming and non-biofilm forming strains of S. epidermidis
(See Chapter 3). As a result, it was hypothesized that these PEMs may be able to slow biofilm
formation even if the PAA/PAH 2.0/2.0 systems did not.
The mechanism of antibacterial action in the SPS/PAH 9.3/9.3 acid-treated system was
discussed at length in Chapter 3. Briefly, at high pH, PAH was assembled into the multilayer
with regions of hydrophobic clusters comprised of uncharged amines. After acid-treatment, the
amines were protonated and the PEM swelled greatly. This created a coating with both high
positive charge and low modulus, a combination capable of both killing and repelling bacteria.
The results of initial biofilm formation on SPS/PAH 9.3/9.3 acid-treated films are shown
in figure 5.4. Unfortunately, despite the increased antibacterial functionality in these films, no
differences were apparent between the glass, SPS/PAH 9.3/9.3 as deposited and acid-treated
films.
It is important to note that these preliminary investigations do not completely rule out
differences in the biofilms on various surfaces. For instance, biofilms and bacteria adhered to
various PEMs could have different strengths of attachment. In addition, since we did not allow
the biofilms to mature, we do not know if there would be any difference in the structures of
mature biofilms formed on different substrata.
112 | P a g e Figure 5.4. (0.31) Representative optical microscopy images of initial biofilm growth on (A) plain glass, (B) SPS/PAH 9.3/9.3 as deposited PEMs, and (C) SPS/PAH 9.3/9.3 acid‐treated PEMs. 5.3 Liftoff with Temperature and pH Triggerable Hydrogen Bonded PEMs
Due to the failure of our antibacterial PEMs to show significant difference in initial
biofilm attachment, we considered a more aggressive method of preventing bacterial adhesion
and biofilm growth on surfaces. By utilizing the thermal and pH-responsive hydrogen-bonded
PEMs developed by Swiston et al. for cell patches81, we developed films that could potentially
slough away a contaminated surface upon exposure to a certain stimulus to expose an unfouled
surface. Although biofilm release coatings have been investigated, to our knowledge the
versatility of the layer-by-layer technique has not been exploited for this application216. The
“lift-off” PEMs were designed with a hydrogen-bonded region topped with a supportive
electrostatic PEM comprised of FITC-PAH and iron oxide nanoparticles. The hydrogen-bonded
region, consisting of PMAA and temperature-sensitive PNIPAAm, was not chemically or
thermally covalently cross-linked. At high pH, the carboxylic acids of PMAA become charged,
disrupting the hydrogen-bonding. However, PNIPAAm exhibits a lower critical solution
temperature (LCST) behavior (32°C in pure water) and at high temperature, it is insoluble in
water and entraps the film components (preventing film dissociation). At low temperature,
PNIPAAm undergoes a hydrophobic to hydrophilic transition leading to energetically favorable
113 | P a g e Figure 5.5. (0.32) (A) S. epidermidis colony densities on glass and releasable PEM substrates. Samples were challenged with airborne bacteria, then immersed for 30 minutes in PBS at 4°C or 37°C. Except for PEM 4°C, all samples were statistically indistinguishable (student’s t‐test, p<0.05). Representative optical microscopy images of colony growth on (B) plain glass at 4°C, (C) surface after PEM release at 4°C, (D) plain glass at 37°C, and (E) non‐released PEM immersed at 37°C.
interactions between PNIPAAm and water. If the temperature is lowered without raising the pH,
the hydrogen-bonds between PNIPAAm and PMAA maintain the integrity of the film. Thus,
both high pH and low temperature are necessary for the hydrogen bonded film to dissociate,
leading to lift-off of the supportive PEM81.
As an initial test of the lift-off system, non-biofilm forming S. epidermidis was sprayed
on the surface of the PEMs and control glass slides and placed in PBS (pH 7.4) at 4°C or 37°C
for 30 min. Lift-off of the low temperature PEM was clearly visible since the FITC-PAH is
slightly reddish in color. After coating the surfaces with agar and incubating overnight, the
resultant colony densities were determined. As can be seen in Figure 5.5, the releasable PEM
immersed at 4°C successfully removed many of the bacteria applied to the surface and reduced
the number of viable attached bacteria as compared to the PEM at 37°C and the glass controls at
4°C and 37°C. The few colonies that were visible on the PEM immersed at 4°C probably rinsed
114 | P a g e off the film as it was releasing. Since the release was done in 50 mL tubes, any bacteria in
solution had the opportunity to reattach to the clean surface. In a setting in which the released
film is washed downstream, this would not occur. Note that the cytotoxicity of the FITCPAH/iron oxide nanoparticle PEM was previously tested and showed no killing abilities. In
addition, bacteria grew successfully on the PEMs immersed in PBS at 37°C. Thus, the reduction
in colony density could not be attributed to contact-killing on the PEM surface and was only a
result of the PEM lift-off. To test mature biofilm lift-off, biofilms were grown for 24 hours on releasable PEMs and
glass controls. The samples were then placed either in PBS at 37°C or 4°C for 30 minutes and
viewed with an optical microscope. Once again, lift-off at low temperature was visible due to
the reddish color of the FITC-labeled PAH. As can be seen in Figure 5.6, biofilm growth was
apparent on all samples except the low-temperature PEM. Thus, a contaminated surface was
successfully lifted-off to reveal an unfouled surface underneath. Since the new surface could
again become contaminated, the lift-off strategy demonstrated here would only be valuable for
applications requiring a single cleaning. To address this limitation, stacks of releasable films
with different dissociation stimuli (different pH values, temperature, electric field, etc) could be
constructed to create multiple lift-off surfaces capable of multiple cleaning cycles.
115 | P a g e Figure 5.6. (0.33) Optical microscopy images of biofilm growth on (A) glass surface after immersion in PBS at 37°C, (B) PEM surface after immersion in PBS at 37°C, (C) glass surface after immersion in PBS at 4°C, and (D) PEM surface after immersion and lift‐off in PBS at 4°C. Insets show the same sample before PBS immersion. All scalebars = 100 μm. 5.4 Conclusions
Biofilm formation leads to negative consequences in many situations: medical implants,
metal piping, membranes, etc. Preventing biofilm formation on surfaces under optimal biofilm
conditions (little mechanical shear, high nutrient environment, no biocides in solution) is
extremely challenging. In this work, we investigated differences in initial biofilm formation
(first 2 hours) under these conditions on previously examined multilayers that decreased or
116 | P a g e eliminated bacteria in low nutrient conditions. Both the highly compliant surface, which resisted
bacterial attachment, and the swellable cationic-killing surface, which both resisted and killed
bacteria, showed no reduction in initial biofilm formation with the assays used here.
As an alternative, releasable PEMs were investigated to examine the possibility of
removing a contaminated surface to expose a clean surface. The hydrogen-bonded release
system required low temperature and physiological pH ranges to dissociate. Release of a
contaminated surface was successfully demonstrated when surfaces sprayed with bacteria were
immersed in PBS. In addition, biofilm formation and release was shown. Thus, the lift-off
concept was validated and may be an interesting avenue to explore in future biofilm prevention
studies. Although we only demonstrated single layer lift-off, the strategy can be expanded to
include stacks of releasable materials that dissociate upon application of different stimuli,
creating a long-lasting surface that could be cleaned several times.
117 | P a g e CHAPTER 6
6. Summary and Future Work
6.1 Thesis Summary
The main objective of this thesis was to investigate the effects of PEM manipulation on
bacteria-surface interactions. A deeper understanding of the relationship between bacteria and
PEMs and substrata in general is an area of increasing interest. Many studies have previously
established the value of PEMs as antimicrobial coatings, and this work added additional surface
factors and design techniques that could be considered to maximize their efficacy.
After reviewing previously published studies on antimicrobial PEMs and summarizing
the highlights of PEM-eukaryotic cell interactions in the introduction, Chapter 2 evaluated the
effects of surface modulus on bacterial attachment. PEMs were chosen as a model substrate due
to their unique tunability. By simply adjusting assembly pH, surfaces of identical composition
with varying mechanical properties were produced. Bacterial attachment studies on these
surfaces showed that compliant surfaces resisted both Gram negative and Gram positive bacterial
attachment in aqueous and salt solutions. This study was the first to systematically evaluate
surface stiffness effects on bacterial attachment and unequivocally add it to the roster of surface
parameters that can influence bacterial attachment.
Chapter 3 dealt with another aim of many antimicrobial coatings: contact-killing. PEM
engineers have incorporated many biocidal species into PEMs (e.g., surface-grafted quaternary
ammonium compounds, antimicrobial peptides, chitosan, lysozymes, etc.), but have never
considered studying the variation of assembly and post-assembly conditions of typically non118 | P a g e biocidal PEMs to produce surfaces with the necessary cationic charge to kill bacteria upon
contact. In Chapter 3, a model PEM system was chosen to illustrate the importance of PEM
assembly and post-assembly conditions on antibacterial properties. Although most PEMs do not
present sufficiently mobile and dense cationic charge to disrupt bacterial membranes, we
demonstrated that PEM production could be modified to create antibacterial coatings without
complicated chemistry or addition of other species.
In Chapter 4, we characterized new morphologies present in hydrogen-bonded
multilayers capped with a single layer of polycation. By varying the adsorption pH of the
polycation layer, the film displayed a variety of properties and surface morphologies. These
different films were tested for their ability to resist or kill bacteria in air or solution. The pH of
polycation adsorption affected the properties studied in Chapters 2 and 3. That is, with certain
PAH adsorption pH values, we were able to modulate mechanical modulus to create films that
either inhibited or promoted bacterial attachment. In addition, the pH of PAH could be adjusted
to create a surface with the necessary cationic charge to kill bacteria on contact. The polycationtopped hydrogen bonded system therefore further validated the concepts established in earlier
chapters.
The final project discussed in this thesis tested the ability of antimicrobial films to resist
biofilm formation in high nutrient environments. The films in these experiments were challenged
to a greater extent because of the ideal conditions for biofilm formation. For both the low
modulus and cationic-killing surfaces, no differences in initial biofilm formation were visible
with optical microscopy. Although this was not definitive proof that the surfaces had no effect
on biofilm formation, it indicated that a more aggressive biofilm prevention strategy might be
necessary. As a result, another PEM system, a multi-stack PEM with a releasable region, was
119 | P a g e investigated. The releasable PEM successfully lifted-off a layer of non-biofilm forming bacteria
and a biofilm grown in nutrient-rich conditions for 24 hours. The concept of a lift-off PEM to
remove biofilms was thereby established and could possibly prove helpful in the future.
6.2 Future Research Directions
Using this work as starting point, there are many avenues that can be explored to gain a
deeper understanding of the relationship between bacteria and substrata. Also, additional
experiments may further illuminate the usefulness of PEMs in controlling bacterial and
eukaryotic cell attachment and viability.
Although Chapter 2 illustrated how PEM stiffness affected bacterial attachment densities,
little is known about the strength of bacterial adhesion to different PEM substrates with different
mechanical properties. With advances in atomic force microscopy (AFM), bacterial adhesion to
different substrata can be directly measured with clumps of bacteria attached to an AFM tip
approaching a specific surface or a modified tip approaching a lawn of bacteria217, 218. Likewise,
parallel plate flow chambers have been used to measure relative strength of adhesion by flowing
solution or air through a system with adhered bacteria and measuring number of detached
bacteria219. These techniques could help determine the forces and methods necessary to remove
individual bacteria attached to various substrates.
The research in Chapter 2 also leaves unsolved the question of the exact mechanism
leading to a decrease in bacterial attachment with increasing film compliance. Catch-bonds were
posited as a possible explanation, but a full study with genetically modified bacteria with specific
120 | P a g e knock-out genes would help prove the mechanism and shed light on the varieties of bacteria that
are especially susceptible to adherence control with substrata stiffness.
As discussed in Chapter 3, the antimicrobial cationic PEM surfaces we created lacked
powerful Gram-negative killing capabilities. However, the versatility of the layer-by-layer
technique allows for many modifications to create a potent broad-spectrum biocidal surface. This
may be accomplished with the selection of other charged polycations, such as those with alkyl
side-chains, assembled into PEMs under the right conditions to expose dense, mobile cationic
charge.
Another interesting area to explore is the influence of the different surface morphologies
discussed in Chapter 4 on eukaryotic cell attachment, morphology and viability. To date,
unmodified cross-linked PAA/PAAm hydrogels have been shown to be cell resistant, while
stiffer PAA/PAAm films with PAH adsorbed at pH 9 have been shown to be cell adherent.
Bacteria attachment and viability was explored across the entire PAH adsorption pH range, but
the effects of PAH adsorption at lower and higher pH values on eukaryotic cells has not been
considered. PAH adsorption at different pH values could prove to be an easy and useful tool to
modify surfaces for cell studies.
Optical microscopy images of initial biofilm formation showed identical biofilm growth
on both compliant and rigid surfaces, and cationic killing and non-biocidal surfaces. However, a
more sophisticated analysis of biofilm growth would provide additional information. For
instance, biofilms on various surfaces could have different strength of adhesion or dissimilar
structures. The strength of biofilm adhesion has been measured by tensile force applied by
centrifugation and by shear force via collision of biofilm attached plates by gravity220.
121 | P a g e Micromanipulation has also been used to measure adhesion force by physically pulling biofilms
away from substrata221. Physical biofilm structure has been analyzed in a variety of ways
involving fluorescent stains, electron microscopy, confocal imaging, molecular transport
analyses, and modeling and simulation222. In addition, chemical components and gradients
within biofilms have been evaluated with many techniques including two-photon excitation
microscopy223, infrared spectroscopy, microelectrodes, electron microscopy, atomic force
microscopy, etc224. These additional assays could determine the effects of substrata mechanical
compliance and cationic charge on biofilm properties.
Finally, to expand the biofilm lift-off concept to other stimuli responsive multilayers (and
to possibly create a multi-stack surface capable of multiple lift-off cleanings), the
electrochemically sensitive PEMs composed of Prussian Blue (PB) nanoparticles developed by
the Hammond group could be used as the releasable region for biofilm liftoff225, 226. Hammond
et al. has constructed PEMs comprised of negatively charged PB particles and positively charged
linear poly(ethylene imine) (LPEI) on indium tin oxide (ITO) coated glass slides. Application of
an external voltage in an electrobath with excess K+ ions oxidizes the PB. With enough
oxidation, the charge can be fully removed from the PB. Since the films are held together by
electrostatic forces, once the PB is neutral the PEMs begin to dissolve. Note that PB
nanoparticles released in this process are non-cytotoxic. Thus, LPEI/PB PEMs can be used as
sacrificial layers that dissolve upon application of an external stimulus in physiological
conditions.
122 | P a g e REFERENCES
1. Yang, S. Y.; Mendelsohn, J. D.; Rubner, M. F., New class of ultrathin, highly cell‐adhesion‐
resistant polyelectrolyte multilayers with micropatterning capabilities. Biomacromolecules 2003, 4, (4), 987‐994. 2. Roy, S.; Das, P. K., Antibacterial hydrogels of amino acid‐based cationic amphiphiles. Biotechnol Bioeng 2008, 100, (4), 756‐64. 3. Tiller, J. C.; Liao, C. J.; Lewis, K.; Klibanov, A. M., Designing surfaces that kill bacteria on contact. Proceedings of the National Academy of Sciences of the United States of America 2001, 98, (11), 5981‐5985. 4. Tiller, J. C.; Lee, S. B.; Lewis, K.; Klibanov, A. M., Polymer surfaces derivatized with poly(vinyl‐
N‐hexylpyridinium) kill airborne and waterborne bacteria. Biotechnology and Bioengineering 2002, 79, (4), 465‐471. 5. Lewis, K.; Klibanov, A. M., Surpassing nature: rational design of sterile‐surface materials. Trends in Biotechnology 2005, 23, (7), 343‐348. 6. Lee, D.; Cohen, R. E.; Rubner, M. F., Antibacterial properties of Ag nanoparticle loaded multilayers and formation of magnetically directed antibacterial microparticles. Langmuir 2005, 21, (21), 9651‐9659. 7. Dai, J. H.; Bruening, M. L., Catalytic nanoparticles formed by reduction of metal ions in multilayered polyelectrolyte films. Nano Letters 2002, 2, (5), 497‐501. 8. Rudra, J. S.; Dave, K.; Haynie, D. T., Antimicrobial polypeptide multilayer nanocoatings. Journal of Biomaterials Science‐Polymer Edition 2006, 17, (11), 1301‐1315. 9. Khademhosseini, A.; Suh, K. Y.; Yang, J. M.; Eng, G.; Yeh, J.; Levenberg, S.; Langer, R., Layer‐by‐
layer deposition of hyaluronic acid and poly‐L‐lysine for patterned cell co‐cultures. Biomaterials 2004, 25, (17), 3583‐3592. 10. Shiratori, S. S.; Rubner, M. F., pH‐dependent thickness behavior of sequentially adsorbed layers of weak polyelectrolytes. Macromolecules 2000, 33, (11), 4213‐4219. 11. Thompson, M. T.; Berg, M. C.; Tobias, I. S.; Rubner, M. F.; Van Vliet, K. J., Tuning compliance of nanoscale polyelectrolyte multilayers to modulate cell adhesion. Biomaterials 2005, 26, (34), 6836‐45. 12. Hiller, J.; Mendelsohn, J. D.; Rubner, M. F., Reversibly erasable nanoporous anti‐reflection coatings from polyelectrolyte multilayers. Nature Materials 2002, 1, (1), 59‐63. 13. Berg, M. C.; Zhai, L.; Cohen, R. E.; Rubner, M. F., Controlled drug release from porous polyelectrolyte multilayers. Biomacromolecules 2006, 7, (1), 357‐64. 14. Decher, G.; Hong, J. D., Buildup of Ultrathin Multilayer Films by a Self‐Assembly Process .2. Consecutive Adsorption of Anionic and Cationic Bipolar Amphiphiles and Polyelectrolytes on Charged Surfaces. Berichte Der Bunsen‐Gesellschaft‐Physical Chemistry Chemical Physics 1991, 95, (11), 1430‐
1434. 15. Li, Z.; Lee, D.; Sheng, X. X.; Cohen, R. E.; Rubner, M. F., Two‐level antibacterial coating with both release‐killing and contact‐killing capabilities. Langmuir 2006, 22, (24), 9820‐9823. 16. Chung, A. J.; Rubner, M. F., Methods of loading and releasing low molecular weight cationic molecules in weak polyelectrolyte multilayer films. Langmuir 2002, 18, (4), 1176‐1183. 17. Rajagopalan, P.; Shen, C. J.; Berthiaume, F.; Tilles, A. W.; Toner, M.; Yarmush, M. L., Polyelectrolyte Nano‐scaffolds for the Design of Layered Cellular Architectures. Tissue Eng 2006. 18. Ofek, I.; Hasty, D. L.; Doyle, R. J., Bacterial Adhesion. American Society for Microbiology: 2003. 19. Palmer, J.; Flint, S.; Brooks, J., Bacterial cell attachment, the beginning of a biofilm. J Ind Microbiol Biotechnol 2007, 34, (9), 577‐88. 123 | P a g e 20. Boulmedais, F.; Frisch, B.; Etienne, O.; Lavalle, P.; Picart, C.; Ogier, J.; Voegel, J. C.; Schaaf, P.; Egles, C., Polyelectrolyte multilayer films with pegylated polypeptides as a new type of anti‐microbial protection for biomaterials. Biomaterials 2004, 25, (11), 2003‐2011. 21. Riedl, C. R.; Witkowski, M.; Plas, E.; Pflueger, H., Heparin coating reduces encrustation of ureteral stents: a preliminary report. International Journal of Antimicrobial Agents 2002, 19, (6), 507‐
510. 22. Fu, J. H.; Ji, J.; Yuan, W. Y.; Shen, J. C., Construction of anti‐adhesive and antibacterial multilayer films via layer‐by‐layer assembly of heparin and chitosan. Biomaterials 2005, 26, (33), 6684‐
6692. 23. Richert, L.; Lavalle, P.; Payan, E.; Shu, X. Z.; Prestwich, G. D.; Stoltz, J. F.; Schaaf, P.; Voegel, J. C.; Picart, C., Layer by layer buildup of polysaccharide films: Physical chemistry and cellular adhesion aspects. Langmuir 2004, 20, (2), 448‐458. 24. Cassinelli, C.; Morra, M.; Pavesio, A.; Renier, D., Evaluation of interfacial properties of hyaluronan coated poly(methylmethacrylate) intraocular lenses. Journal of Biomaterials Science‐
Polymer Edition 2000, 11, (9), 961‐977. 25. Malcher, M.; Volodkin, D.; Heurtault, B.; Andre, P.; Schaaf, P.; Mohwald, H.; Voegel, J. C.; Sokolowski, A.; Ball, V.; Boulmedais, F.; Frisch, B., Embedded silver ions‐containing liposomes in polyelectrolyte multilayers: Cargos films for antibacterial agents. Langmuir 2008, 24, (18), 10209‐
10215. 26. Lee, H.; Lee, Y.; Statz, A. R.; Rho, J.; Park, T. G.; Messersmith, P. B., Substrate‐independent layer‐by‐layer assembly by using mussel‐adhesive‐inspired polymers. Advanced Materials 2008, 20, (9), 1619‐+. 27. Chua, P. H.; Neoh, K. G.; Kang, E. T.; Wang, W., Surface functionalization of titanium with hyaluronic acid/chitosan polyelectrolyte multilayers and RGD for promoting osteoblast functions and inhibiting bacterial adhesion. Biomaterials 2008, 29, (10), 1412‐1421. 28. Chuang, H. F.; Smith, R. C.; Hammond, P. T., Polyelectrolyte multilayers for tunable release of antibiotics. Biomacromolecules 2008, 9, (6), 1660‐1668. 29. Cui, X. Q.; Li, C. M.; Bao, H. F.; Zheng, X. T.; Lu, Z. S., In situ fabrication of silver nanoarrays in hyaluronan/PDDA layer‐by‐layer assembled structure. Journal of Colloid and Interface Science 2008, 327, (2), 459‐465. 30. Yliniemi, K.; Vahvaselka, M.; Van Ingelgem, Y.; Baert, K.; Wilson, B. P.; Terryn, H.; Kontturi, K., The formation and characterisation of ultra‐thin films containing Ag nanoparticles. Journal of Materials Chemistry 2008, 18, (2), 199‐206. 31. Yu, D. G.; Lin, W. C.; Yang, M. C., Surface modification of poly(L‐lactic acid) membrane via layer‐by‐layer assembly of silver nanoparticle‐embedded polyelectrolyte multilayer. Bioconjugate Chemistry 2007, 18, (5), 1521‐1529. 32. Fu, J. H.; Ji, J.; Fan, D. Z.; Shen, J. C., Construction of antibacterial multilayer films containing nanosilver via layer‐by‐layer assembly of heparin and chitosan‐silver ions complex. Journal of Biomedical Materials Research Part A 2006, 79A, (3), 665‐674. 33. Shi, Z.; Neoh, K. G.; Zhong, S. P.; Yung, L. Y. L.; Kang, E. T.; Wang, W., In vitro antibacterial and cytotoxicity assay of multilayered polyelectrolyte‐functionalized stainless steel. Journal of Biomedical Materials Research Part A 2006, 76A, (4), 826‐834. 34. Podsiadlo, P.; Paternel, S.; Rouillard, J. M.; Zhang, Z. F.; Lee, J.; Lee, J. W.; Gulari, L.; Kotov, N. A., Layer‐by‐layer assembly of nacre‐like nanostructured composites with antimicrobial properties. Langmuir 2005, 21, (25), 11915‐11921. 35. Choi, W. S.; Koo, H. Y.; Park, J. H.; Kim, D. Y., Synthesis of two types of nanoparticles in polyelectrolyte capsule nanoreactors and their dual functionality. Journal of the American Chemical Society 2005, 127, (46), 16136‐16142. 124 | P a g e 36. Yang, S. Y.; Seo, J. Y., Cellular interactions on nano‐structured polyelectrolyte multilayers. Colloids and Surfaces a‐Physicochemical and Engineering Aspects 2008, 313, 526‐529 704. 37. Grunlan, J. C.; Choi, J. K.; Lin, A., Antimicrobial behavior of polyelectrolyte multilayer films containing cetrimide and silver. Biomacromolecules 2005, 6, (2), 1149‐1153. 38. Krutyakov, Y. A.; Kudrinskiy, A. A.; Olenin, A. Y.; Lisichkin, G. V., Synthesis and properties of silver nanoparticles: Achievements and prospects. Uspekhi Khimii 2008, 77, (3), 242‐269. 39. Feng, Q. L.; Wu, J.; Chen, G. Q.; Cui, F. Z.; Kim, T. N.; Kim, J. O., A mechanistic study of the antibacterial effect of silver ions on Escherichia coli and Staphylococcus aureus. Journal of Biomedical Materials Research 2000, 52, (4), 662‐668. 40. Berger, T. J.; Spadaro, J. A.; Chapin, S. E.; Becker, R. O., Electrically Generated Silver Ions ‐ Quantitative Effects on Bacterial and Mammalian‐Cells. Antimicrobial Agents and Chemotherapy 1976, 9, (2), 357‐358. 41. Mao, Z. W.; Ma, L.; Gao, C. Y.; Shen, J. C., Preformed microcapsules for loading and sustained release of ciprofloxacin hydrochloride. Journal of Controlled Release 2005, 104, (1), 193‐202. 42. Bhadra, D.; Gupta, G.; Bhadra, S.; Umamaheshwari, R. B.; Jain, N. K., Multicomposite ultrathin capsules for sustained ocular delivery of ciprofloxacin hydrochloride. Journal of Pharmacy and Pharmaceutical Sciences 2004, 7, (2), 241‐251. 43. McMurry, L. M.; Oethinger, M.; Levy, S. B., Triclosan targets lipid synthesis. Nature 1998, 394, (6693), 531‐532. 44. Kim, B. S.; Park, S. W.; Hammond, P. T., Hydrogen‐bonding layer‐by‐layer assembled biodegradable polymeric micelles as drug delivery vehicles from surfaces. Acs Nano 2008, 2, (2), 386‐
392. 45. Nguyen, P. M.; Zacharia, N. S.; Verploegen, E.; Hammond, P. T., Extended release antibacterial layer‐by‐layer films incorporating linear‐dendritic block copolymer micelles. Chemistry of Materials 2007, 19, (23), 5524‐5530. 46. Wood, K. C.; Boedicker, J. Q.; Lynn, D. M.; Hammond, P. T., Tunable drug release from hydrolytically degradable layer‐by‐layer thin films. Langmuir 2005, 21, (4), 1603‐9. 47. Caruso, F.; Yang, W. J.; Trau, D.; Renneberg, R., Microencapsulation of uncharged low molecular weight organic materials by polyelectrolyte multilayer self‐assembly. Langmuir 2000, 16, (23), 8932‐8936. 48. Rabea, E. I.; Badawy, M. E. T.; Stevens, C. V.; Smagghe, G.; Steurbaut, W., Chitosan as antimicrobial agent: Applications and mode of action. Biomacromolecules 2003, 4, (6), 1457‐1465. 49. Elsabee, M. Z.; Abdou, E. S.; Nagy, K. S. A.; Eweis, M., Surface modification of polypropylene films by chitosan and chitosan/pectin multilayer. Carbohydrate Polymers 2008, 71, (2), 187‐195. 50. Bratskaya, S.; Marinin, D.; Simon, F.; Synytska, A.; Zschoche, S.; Busscher, H. J.; Jager, D.; van der Mei, H. C., Adhesion and viability of two enterococcal strains on covalently grafted chitosan and chitosan/kappa‐carrageenan multilayers. Biomacromolecules 2007, 8, (9), 2960‐2968. 51. Feng, Y. H.; Han, Z. G.; Peng, J.; Lu, J.; Xue, B.; Li, L.; Ma, H. Y.; Wang, E. B., Fabrication and characterization of multilayer films based on Keggin‐type polyoxometalate and chitosan. Materials Letters 2006, 60, (13‐14), 1588‐1593. 52. Pan, Y. F.; Xiao, H. N.; Zhao, G. L.; He, B. H., Antimicrobial and Thermal‐Responsive Layer‐by‐
Layer Assembly Based on Ionic‐Modified Guanidine Polymer and PVA. Polymer Bulletin 2008, 61, (5), 541‐551. 53. Tiller, J. C.; Liao, C. J.; Lewis, K.; Klibanov, A. M., Designing surfaces that kill bacteria on contact. Proc Natl Acad Sci U S A 2001, 98, (11), 5981‐5. 125 | P a g e 54. Gottenbos, B.; van der Mei, H. C.; Klatter, F.; Nieuwenhuis, P.; Busscher, H. J., In vitro and in vivo antimicrobial activity of covalently coupled quaternary ammonium silane coatings on silicone rubber. Biomaterials 2002, 23, (6), 1417‐1423. 55. Bechinger, B.; Lohner, K., Detergent‐like actions of linear amphipathic cationic antimicrobial peptides. Biochimica Et Biophysica Acta‐Biomembranes 2006, 1758, (9), 1529‐1539. 56. Brogden, K. A., Antimicrobial peptides: Pore formers or metabolic inhibitors in bacteria? Nature Reviews Microbiology 2005, 3, (3), 238‐250. 57. Etienne, O.; Picart, C.; Taddei, C.; Haikel, Y.; Dimarcq, J. L.; Schaaf, P.; Voegel, J. C.; Ogier, J. A.; Egles, C., Multilayer polyelectrolyte films functionalized by insertion of defensin: A new approach to protection of implants from bacterial colonization. Antimicrobial Agents and Chemotherapy 2004, 48, (10), 3662‐3669. 58. Guyomard, A.; De, E.; Jouenne, T.; Malandain, J. J.; Muller, G.; Glinel, K., Incorporation of a hydrophobic antibacterial peptide into amphiphilic polyelectrolyte multilayers: A bioinspired approach to prepare biocidal thin coatings. Advanced Functional Materials 2008, 18, (5), 758‐765. 59. Yuan, W. Y.; Ji, J.; Fu, J. H.; Shen, J. C., A facile method to construct hybrid multilayered films as a strong and multifunctional antibacterial coating. Journal of Biomedical Materials Research Part B‐
Applied Biomaterials 2008, 85B, (2), 556‐563. 60. Nepal, D.; Balasubramanian, S.; Simonian, A. L.; Davis, V. A., Strong antimicrobial coatings: single‐walled carbon nanotubes armored with biopolymers. Nano Lett 2008, 8, (7), 1896‐901. 61. Dunne, W. M., Bacterial adhesion: Seen any good biofilms lately? Clinical Microbiology Reviews 2002, 15, (2), 155‐+. 62. Mendelsohn, J. D.; Yang, S. Y.; Hiller, J.; Hochbaum, A. I.; Rubner, M. F., Rational design of cytophilic and cytophobic polyelectrolyte multilayer thin films. Biomacromolecules 2003, 4, (1), 96‐
106. 63. Yang, S. Y.; Mendelsohn, J. D.; Rubner, M. F., New class of ultrathin, highly cell‐adhesion‐
resistant polyelectrolyte multilayers with micropatterning capabilities. Biomacromolecules 2003, 4, (4), 987‐94. 64. Elbert, D. L.; Herbert, C. B.; Hubbell, J. A., Thin polymer layers formed by polyelectrolyte multilayer techniques on biological surfaces. Langmuir 1999, 15, (16), 5355‐5362. 65. Salloum, D. S.; Olenych, S. G.; Keller, T. C. S.; Schlenoff, J. B., Vascular smooth muscle cells on polyelectrolyte multilayers: Hydrophobicity‐directed adhesion and growth. Biomacromolecules 2005, 6, (1), 161‐167. 66. Ai, H.; Lvov, Y. M.; Mills, D. K.; Jennings, M.; Alexander, J. S.; Jones, S. A., Coating and selective deposition of nanofilm on silicone rubber for cell adhesion and growth. Cell Biochemistry and Biophysics 2003, 38, (2), 103‐114. 67. Ai, H.; Meng, H. D.; Ichinose, I.; Jones, S. A.; Mills, D. K.; Lvov, Y. M.; Qiao, X. X., Biocompatibility of layer‐by‐layer self‐assembled nanofilm on silicone rubber for neurons. Journal of Neuroscience Methods 2003, 128, (1‐2), 1‐8. 68. Olenych, S. G.; Moussallem, M. D.; Salloum, D. S.; Schlenoff, J. B.; Keller, T. C. S., Fibronectin and cell attachment to cell and protein resistant polyelectrolyte surfaces. Biomacromolecules 2005, 6, (6), 3252‐3258. 69. Zhu, H. G.; Ji, J.; Shen, J. C., Construction of multilayer coating onto poly‐(DL‐lactide) to promote cytocompatibility. Biomaterials 2004, 25, (1), 109‐117. 70. Thompson, M. T.; Berg, M. C.; Tobias, I. S.; Lichter, J. A.; Rubner, M. F.; Van Vliet, K. J., Biochemical functionalization of polymeric cell substrata can alter mechanical compliance. Biomacromolecules 2006, 7, (6), 1990‐5. 71. Zhu, Y. B.; Sun, Y., The influence of polyelectrolyte charges of polyurethane membrane surface on the growth of human endothelial cells. Colloids and Surfaces B‐Biointerfaces 2004, 36, (1), 49‐55. 126 | P a g e 72. Boura, C.; Menu, P.; Payan, E.; Picart, C.; Voegel, J. C.; Muller, S.; Stoltz, J. F., Endothelial cells grown on thin polyelectrolyte mutlilayered films: an evaluation of a new versatile surface modification. Biomaterials 2003, 24, (20), 3521‐3530. 73. Sinani, V. A.; Koktysh, D. S.; Yun, B. G.; Matts, R. L.; Pappas, T. C.; Motamedi, M.; Thomas, S. N.; Kotov, N. A., Collagen coating promotes biocompatibility of semiconductor nanoparticles in stratified LBL films. Nano Letters 2003, 3, (9), 1177‐1182. 74. Gheith, M. K.; Sinani, V. A.; Wicksted, J. P.; Matts, R. L.; Kotov, N. A., Single‐walled carbon nanotube polyelectrolyte multilayers and freestanding films as a biocompatible platformfor neuroprosthetic implants. Advanced Materials 2005, 17, (22), 2663‐+. 75. Glawe, J. D.; Hill, J. B.; Mills, D. K.; McShane, M. J., Influence of channel width on alignment of smooth muscle cells by high‐aspect‐ratio microfabricated elastomeric cell culture scaffolds. Journal of Biomedical Materials Research Part A 2005, 75A, (1), 106‐114. 76. Donath, E.; Moya, S.; Neu, B.; Sukhorukov, G. B.; Georgieva, R.; Voigt, A.; Baumler, H.; Kiesewetter, H.; Mohwald, H., Hollow polymer shells from biological templates: Fabrication and potential applications. Chemistry‐a European Journal 2002, 8, (23), 5481‐5485. 77. Neu, B.; Voigt, A.; Mitlohner, R.; Leporatti, S.; Gao, C. Y.; Donath, E.; Kiesewetter, H.; Mohwald, H.; Meiselman, H. J.; Baumler, H., Biological cells as templates for hollow microcapsules. Journal of Microencapsulation 2001, 18, (3), 385‐395. 78. Germain, M.; Balaguer, P.; Nicolas, J. C.; Lopez, F.; Esteve, J. P.; Sukhorukov, G. B.; Winterhalter, M.; Richard‐Foy, H.; Fournier, D., Protection of mammalian cell used in biosensors by coating with a polyelectrolyte shell. Biosens Bioelectron 2006, 21, (8), 1566‐73. 79. Diaspro, A.; Silvano, D.; Krol, S.; Cavalleri, O.; Gliozzi, A., Single living cell encapsulation in nano‐organized polyelectrolyte shells. Langmuir 2002, 18, (13), 5047‐5050. 80. Hillberg, A. L.; Tabrizian, M., Biorecognition through layer‐by‐layer polyelectrolyte assembly: in‐situ hybridization on living cells. Biomacromolecules 2006, 7, (10), 2742‐50. 81. Swiston, A. J.; Cheng, C.; Um, S. H.; Irvine, D. J.; Cohen, R. E.; Rubner, M. F., Surface Functionalization of Living Cells with Multilayer Patches. Nano Lett 2008. 82. Wilson, J. T.; Cui, W. X.; Chaikof, E. L., Layer‐by‐layer assembly of a conformal nanothin PEG coating for intraportal islet transplantation. Nano Letters 2008, 8, (7), 1940‐1948. 83. Jiang, X. P.; Zheng, H. P.; Gourdin, S.; Hammond, P. T., Polymer‐on‐polymer stamping: Universal approaches to chemically patterned surfaces. Langmuir 2002, 18, (7), 2607‐2615. 84. Zheng, H.; Berg, M. C.; Rubner, M. F.; Hammond, P. T., Controlling cell attachment selectively onto biological polymer‐colloid templates using polymer‐on‐polymer stamping. Langmuir 2004, 20, (17), 7215‐22. 85. Berg, M. C.; Yang, S. Y.; Hammond, P. T.; Rubner, M. F., Controlling mammalian cell interactions on patterned polyelectrolyte multilayer surfaces. Langmuir 2004, 20, (4), 1362‐8. 86. Krol, S.; Nolte, M.; Diaspro, A.; Mazza, D.; Magrassi, R.; Gliozzi, A.; Fery, A., Encapsulated living cells on microstructured surfaces. Langmuir 2005, 21, (2), 705‐9. 87. Mohammed, J. S.; DeCoster, M. A.; McShane, M. J., Fabrication of interdigitated micropatterns of self‐assembled polymer nanofilms containing cell‐adhesive materials. Langmuir 2006, 22, (6), 2738‐
2746. 88. Mohammed, J. S.; DeCoster, M. A.; McShane, M. J., Micropatterning of nanoengineered surfaces to study neuronal cell attachment in vitro. Biomacromolecules 2004, 5, (5), 1745‐1755. 89. Reyes, D. R.; Perruccio, E. M.; Becerra, S. P.; Locascio, L. E.; Gaitan, M., Micropatterning neuronal cells on polyelectrolyte multilayers. Langmuir 2004, 20, (20), 8805‐8811. 90. Katsikogianni, M.; Missirlis, Y. F., Concise review of mechanisms of bacterial adhesion to biomaterials and of techniques used in estimating bacteria‐material interactions. Eur Cell Mater 2004, 8, 37‐57. 127 | P a g e 91. Klevens, R. M.; Edwards, J. R.; Richards, C. L.; Horan, T. C.; Gaynes, R. P.; Pollock, D. A.; Cardo, D. M., Estimating health care‐associated infections and deaths in US hospitals, 2002. Public Health Reports 2007, 122, (2), 160‐166. 92. Donlan, R. M., Biofilms and device‐associated infections. Emerg Infect Dis 2001, 7, (2), 277‐81. 93. Costerton, J. W.; Stewart, P. S.; Greenberg, E. P., Bacterial biofilms: a common cause of persistent infections. Science 1999, 284, (5418), 1318‐22. 94. Chang, C. C.; Merritt, K., Microbial adherence on poly(methyl methacrylate) (PMMA) surfaces. J Biomed Mater Res 1992, 26, (2), 197‐207. 95. Tiller, J. C.; Lee, S. B.; Lewis, K.; Klibanov, A. M., Polymer surfaces derivatized with poly(vinyl‐
N‐hexylpyridinium) kill airborne and waterborne bacteria. Biotechnol Bioeng 2002, 79, (4), 465‐71. 96. Saygun, O.; Agalar, C.; Aydinuraz, K.; Agalar, F.; Daphan, C.; Saygun, M.; Ceken, S.; Akkus, A.; Denkbas, E. B., Gold and gold‐palladium coated polypropylene grafts in a S. epidermidis wound infection model. J Surg Res 2006, 131, (1), 73‐9. 97. Hetrick, E. M.; Schoenfisch, M. H., Reducing implant‐related infections: active release strategies. Chemical Society Reviews 2006, 35, (9), 780‐789. 98. Kohnen, W.; Kolbenschlag, C.; Teske‐Keiser, S.; Jansen, B., Development of a long‐lasting ventricular catheter impregnated with a combination of antibiotics. Biomaterials 2003, 24, (26), 4865‐
9. 99. Ignatova, M.; Voccia, S.; Gilbert, B.; Markova, N.; Cossement, D.; Gouttebaron, R.; Jerome, R.; Jerome, C., Combination of electrografting and atom‐transfer radical polymerization for making the stainless steel surface antibacterial and protein antiadhesive. Langmuir 2006, 22, (1), 255‐62. 100. Teixeira, P.; Trindade, A. C.; Godinho, M. H.; Azeredo, J.; Oliveira, R.; Fonseca, J. G., Staphylococcus epidermidis adhesion on modified urea/urethane elastomers. J Biomater Sci Polym Ed 2006, 17, (1‐2), 239‐46. 101. Cao, T.; Tang, H.; Liang, X.; Wang, A.; Auner, G. W.; Salley, S. O.; Ng, K. Y., Nanoscale investigation on adhesion of E. coli to surface modified silicone using atomic force microscopy. Biotechnol Bioeng 2006, 94, (1), 167‐76. 102. Tang, H.; Cao, T.; Wang, A.; Liang, X.; Salley, S. O.; McAllister, J. P., 2nd; Ng, K. Y., Effect of surface modification of siliconeon Staphylococcus epidermidis adhesion and colonization. J Biomed Mater Res A 2007, 80, (4), 885‐94. 103. Katsikogianni, M.; Spiliopoulou, I.; Dowling, D. P.; Missirlis, Y. F., Adhesion of slime producing Staphylococcus epidermidis strains to PVC and diamond‐like carbon/silver/fluorinated coatings. J Mater Sci Mater Med 2006, 17, (8), 679‐89. 104. An, Y. H.; Friedman, R. J., Concise review of mechanisms of bacterial adhesion to biomaterial surfaces. Journal of Biomedical Materials Research 1998, 43, (3), 338‐348. 105. Ilker, M. F.; Nusslein, K.; Tew, G. N.; Coughlin, E. B., Tuning the hemolytic and antibacterial activities of amphiphilic polynorbornene derivatives. J Am Chem Soc 2004, 126, (48), 15870‐5. 106. MacKintosh, E. E.; Patel, J. D.; Marchant, R. E.; Anderson, J. M., Effects of biomaterial surface chemistry on the adhesion and biofilm formation of Staphylococcus epidermidis in vitro. J Biomed Mater Res A 2006, 78, (4), 836‐42. 107. Triandafillu, K.; Balazs, D. J.; Aronsson, B. O.; Descouts, P.; Tu Quoc, P.; van Delden, C.; Mathieu, H. J.; Harms, H., Adhesion of Pseudomonas aeruginosa strains to untreated and oxygen‐
plasma treated poly(vinyl chloride) (PVC) from endotracheal intubation devices. Biomaterials 2003, 24, (8), 1507‐18. 108. Abu‐Lail, N. I.; Camesano, T. A., Specific and nonspecific interaction forces between Escherichia coli and silicon nitride, determined by poisson statistical analysis. Langmuir 2006, 22, (17), 7296‐301. 128 | P a g e 109. Speranza, G.; Gotterdi, G.; Pederzolli, C.; Lunelli, L.; Canteri, R.; Pasquardiini, L.; Carli, E.; Lui, A.; Maniglio, D.; Brugnara, M.; Anderle, M., Role of chemical interactions in bacterial adhesion to polymer surfaces. Biomaterials 2004, 25, 2029–2037. 110. van Oss, C. J., Long‐range and short‐range mechanisms of hydrophobic attraction and hydrophilic repulsion in specific and aspecific interactions. Journal of Molecular Recognition 2003, 16, (4), 177‐190. 111. Kang, S.; Choi, H., Effect of surface hydrophobicity on the adhesion of S‐cerevisiae onto modified surfaces by poly(styrene‐ran‐sulfonic acid) random copolymers. Colloids and Surfaces B‐
Biointerfaces 2005, 46, (2), 70‐77. 112. Liu, Y.; Strauss, J.; Camesano, T. A., Thermodynamic Investigation of Staphylococcus epidermidis interactions with protein‐coated substrata. Langmuir 2007, 23, (13), 7134‐42. 113. Richert, L.; Engler, A. J.; Discher, D. E.; Picart, C., Elasticity of native and cross‐linked polyelectrolyte multilayer films. Biomacromolecules 2004, 5, (5), 1908‐16. 114. Thompson, M. T.; Berg, M. C.; Tobias, I. S.; Rubner, M. F.; Van Vliet, K. J., Tuning compliance of nanoscale polyelectrolyte multilayers to modulate cell adhesion. Biomaterials 2005, (In press). 115. Mack, D.; Rohde, H.; Harris, L. G.; Davies, A. P.; Horstkotte, M. A.; Knobloch, J. K., Biofilm formation in medical device‐related infection. Int J Artif Organs 2006, 29, (4), 343‐59. 116. Danese, P. N., Antibiofilm approaches: prevention of catheter colonization. Chem Biol 2002, 9, (8), 873‐80. 117. Arciola, C. R.; Campoccia, D.; Gamberini, S.; Donati, M. E.; Pirini, V.; Visai, L.; Speziale, P.; Montanaro, L., Antibiotic resistance in exopolysaccharide‐forming Staphylococcus epidermidis clinical isolates from orthopaedic implant infections. Biomaterials 2005, 26, (33), 6530‐6535. 118. van der Kooij, D.; Veenendaal, H. R.; Scheffer, W. J., Biofilm formation and multiplication of Legionella in a model warm water system with pipes of copper, stainless steel and cross‐linked polyethylene. Water Res 2005, 39, (13), 2789‐98. 119. Janmey, P. A.; McCulloch, C. A., Cell mechanics: integrating cell responses to mechanical stimuli. Annu Rev Biomed Eng 2007, 9, 1‐34. 120. Richert, L.; Lavalle, P.; Vautier, D.; Senger, B.; Stoltz, J. F.; Schaaf, P.; Voegel, J. C.; Picart, C., Cell interactions with polyelectrolyte multilayer films. Biomacromolecules 2002, 3, (6), 1170‐1178. 121. Hutter, J.; Bechhoefer, J., Calibration of atomic‐force microscope tips. Rev. Sci.Instrum. 1993, 64, 1868–1873. 122. Thompson, M. T.; Berg, M. C.; Tobias, I. S.; Rubner, M. F.; Van Vliet, K. J., Tuning compliance of nanoscale polyelectrolyte multilayers to modulate cell adhesion. Biomaterials 2005, 26, 6836‐6845 123. Johnson, K. L., Contact Mechanics. Cambridge University Press: Cambridge, 1994. 124. Dean, D.; Han, L.; Grodzinsky, A. J.; Ortiz, C., Compressive nanomechanics of opposing aggrecan macromolecules. J Biomech 2006, 39, (14), 2555‐65. 125. Rixman, M.; Dean, D.; Macais, C.; Ortiz, C., Nanoscale Intermolecular Interactions between Human Serum Albumin and Alkanethiol Self‐assembled Monolayers. Langmuir 2003, 19, 6202‐6218. 126. Yoo, D.; Shiratori, S. S.; Rubner, M. F., Controlling bilayer composition and surface wettability of sequentially adsorbed multilayers of weak polyelectrolytes. Macromolecules 1998, 31, (13), 4309‐
4318. 127. Arnsdorf, M. F.; Knight, B. P., Patient Information: Implantable cardio‐defibrillators. In: UpToDate. 2006. 128. Kostler, S.; Delgado, A. V.; Ribitsch, V., Surface thermodynamic properties of polyelectrolyte multilayers. Journal of Colloid and Interface Science 2005, 286, (1), 339‐348. 129. Banner, M. A.; Cunniffe, J. G.; Macintosh, R. L.; Foster, T. J.; Rohde, H.; Mack, D.; Hoyes, E.; Derrick, J.; Upton, M.; Handley, P. S., Localized tufts of fibrils on Staphylococcus epidermidis NCTC 11047 are comprised of the accumulation‐associated protein. J Bacteriol 2007, 189, (7), 2793‐804. 129 | P a g e 130. Veenstra, G. J.; Cremers, F. F.; van Dijk, H.; Fleer, A., Ultrastructural organization and regulation of a biomaterial adhesin of Staphylococcus epidermidis. J Bacteriol 1996, 178, (2), 537‐41. 131. Francius, G.; Hemmerle, J.; Ball, V.; Lavalle, P.; Picart, C.; Voegel, J. C.; Schaaf, P.; Senger, B., Stiffening of soft polyelectrolyte architectures by multilayer capping evidenced by viscoelastic analysis of AFM indentation measurements. Journal of Physical Chemistry C 2007, 111, (23), 8299‐8306. 132. Oommen, B.; Van Vliet, K. J., Effects of nanoscale thickness and elastic nonlinearity on measured mechanical properties of polymeric films. Thin Solid Films 2006, 25, 235‐242. 133. Walton, E. B.; Oommen, B.; Van Vliet, K. J., How stiff and thin can an extracellular matrix be? Effects of finite film thickness and compliance on eukaryotic cell adhesion. In Engineering in Medicine and Biology, Layton, B., Ed. Lyon, France, 2007. 134. Greenspan, F. S.; Gardner, D. G., Basic and Clinical Endocrinology. 7th ed.; Lange Medical Books/McGraw‐Hill Publishing: 2004. 135. Vogel, V., Mechanotransduction involving multimodular proteins: converting force into biochemical signals. Annu Rev Biophys Biomol Struct 2006, 35, 459‐88. 136. Anishkin, A.; Kung, C., Microbial mechanosensation. Curr Opin Neurobiol 2005, 15, (4), 397‐
405. 137. Lin, S. Y.; Corey, D. P., TRP channels in mechanosensation. Curr Opin Neurobiol 2005, 15, (3), 350‐7. 138. Shih, Y. L.; Rothfield, L., The bacterial cytoskeleton. Microbiology and Molecular Biology Reviews 2006, 70, 729‐754. 139. Koch, A. L.; Woeste, S., Elasticity of the Sacculus of Escherichia‐Coli. Journal of Bacteriology 1992, 174, (14), 4811‐4819. 140. Sundararaj, S.; Guo, A.; Habibi‐Nazhad, B.; Rouani, M.; Stothard, P.; Ellison, M.; Wishart, D. S., The CyberCell Database (CCDB): a comprehensive, self‐updating, relational database to coordinate and facilitate in silico modeling of Escherichia coli. Nucleic Acids Research 2004, 32, D293‐D295. 141. Bakker, D. P.; Huijs, F. M.; de Vries, J.; Klijnstra, J. W.; Busscher, H. J.; van der Mei, H. C., Bacterial deposition to fluoridated and non‐fluoridated polyurethane coatings with different elastic modulus and surface tension in a parallel plate and a stagnation point flow chamber. Colloids and Surfaces B‐Biointerfaces 2003, 32, (3), 179‐190. 142. Bakker, D. P.; Busscher, H. J.; van Zanten, J.; de Vries, J.; Klijnstra, J. W.; van der Mei, H. C., Multiple linear regression analysis of bacterial deposition to polyurethane coating after conditioning film formation in the marine environment. Microbiology‐Sgm 2004, 150, 1779‐1784. 143. Thomas, W.; Forero, M.; Yakovenko, O.; Nilsson, L.; Vicini, P.; Sokurenko, E.; Vogel, V., Catch‐
bond model derived from allostery explains force‐activated bacterial adhesion. Biophys J 2006, 90, (3), 753‐64. 144. Lichter, J. A.; Thompson, M. T.; Delgadillo, M.; Nishikawa, T.; Rubner, M. F.; Van Vliet, K. J., Substrata mechanical stiffness can regulate adhesion of viable bacteria. Biomacromolecules 2008, 9, (6), 1571‐8. 145. Ciston, S.; Lueptow, R. M.; Gray, K. A., Bacterial attachment on reactive ceramic ultrafiltration membranes. Journal of Membrane Science 2008, 320, (1‐2), 101‐107. 146. Whitehead, K. A.; Verran, J., The effect of surface topography on the retention of microorganisms. Food and Bioproducts Processing 2006, 84, (C4), 253‐259. 147. Hermansson, M., The DLVO theory in microbial adhesion. Colloids and Surfaces B‐
Biointerfaces 1999, 14, (1‐4), 105‐119. 148. Jeon, H. J.; Yi, S. C.; Oh, S. G., Preparation and antibacterial effects of Ag‐SiO2 thin films by sol‐
gel method. Biomaterials 2003, 24, (27), 4921‐4928. 130 | P a g e 149. Kohnen, W.; Kolbenschlag, C.; Teske‐Keiser, S.; Jansen, B., Development of a long‐lasting ventricular catheter impregnated with a combination of antibiotics. Biomaterials 2003, 24, (26), 4865‐
4869. 150. Lee, S. B.; Koepsel, R. R.; Morley, S. W.; Matyjaszewski, K.; Sun, Y. J.; Russell, A. J., Permanent, nonleaching antibacterial surfaces. 1. Synthesis by atom transfer radical polymerization. Biomacromolecules 2004, 5, (3), 877‐882. 151. Kurt, P.; Wood, L.; Ohman, D. E.; Wynne, K. J., Highly effective contact antimicrobial surfaces via polymer surface modifiers. Langmuir 2007, 23, (9), 4719‐4723. 152. Kugler, R.; Bouloussa, O.; Rondelez, F., Evidence of a charge‐density threshold for optimum efficiency of biocidal cationic surfaces. Microbiology‐Sgm 2005, 151, 1341‐1348. 153. Cen, L.; Neoh, K. G.; Kang, E. T., Surface functionalization technique for conferring antibacterial properties to polymeric and cellulosic surfaces. Langmuir 2003, 19, (24), 10295‐10303. 154. Popa, A.; Davidescu, C. M.; Trif, R.; Ilia, G.; Iliescu, S.; Dehelean, G., Study of quaternary 'onium' salts grafted on polymers: antibacterial activity of quaternary phosphonium salts grafted on 4 gel‐type' styrene‐divinylbenzene copolymers. Reactive & Functional Polymers 2003, 55, (2), 151‐158. 155. Gilbert, P.; Moore, L. E., Cationic antiseptics: diversity of action under a common epithet. Journal of Applied Microbiology 2005, 99, (4), 703‐715. 156. Murata, H.; Koepsel, R. R.; Matyjaszewski, K.; Russell, A. J., Permanent, non‐leaching antibacterial surfaces ‐ 2: How high density cationic surfaces kill bacterial cells. Biomaterials 2007, 28, (32), 4870‐4879. 157. Fischer, D.; Li, Y. X.; Ahlemeyer, B.; Krieglstein, J.; Kissel, T., In vitro cytotoxicity testing of polycations: influence of polymer structure on cell viability and hemolysis. Biomaterials 2003, 24, (7), 1121‐1131. 158. Bombelli, C.; Bordi, F.; Ferro, S.; Giansanti, L.; Jori, G.; Mancini, G.; Mazzuca, C.; Monti, D.; Ricchelli, F.; Sennato, S.; Venanzi, M., New cationic liposomes as vehicles of m‐
tetrahydroxyphenylchlorin in photodynamic therapy of infectious diseases. Molecular Pharmaceutics 2008, 5, (4), 672‐679. 159. Yang, S. Y.; Rubner, M. F., Micropatterning of polymer thin films with pH‐sensitive and cross‐
linkable hydrogen‐bonded polyelectrolyte multilayers. Journal of the American Chemical Society 2002, 124, (10), 2100‐2101. 160. Yang, S. Y.; Lee, D.; Cohen, R. E.; Rubner, M. F., Bioinert solution‐cross‐linked hydrogen‐
bonded multilayers on colloidal particles. Langmuir 2004, 20, (14), 5978‐5981. 161. Nolte, A. J.; Takane, N.; Hindman, E.; Gaynor, W.; Rubner, M. F.; Cohen, R. E., Thin film thickness gradients and spatial patterning via salt etching of polyelectrolyte multilayers. Macromolecules 2007, 40, (15), 5479‐5486. 162. Thompson, M. T.; Berg, M. C.; Tobias, I. S.; Rubner, M. F.; Van Vliet, K. J., Tuning compliance of nanoscale polyelectrolyte multilayers to modulate cell adhesion. Biomaterials 2005, 26, (34), 6836‐
6845. 163. Mendelsohn, J. D.; Barrett, C. J.; Chan, V. V.; Pal, A. J.; Mayes, A. M.; Rubner, M. F., Fabrication of microporous thin films from polyelectrolyte multilayers. Langmuir 2000, 16, (11), 5017‐5023. 164. Tjipto, E.; Quinn, J. F.; Caruso, F., Layer‐by‐layer assembly of weak‐strong copolymer polyelectrolytes: A route to morphological control of thin films. Journal of Polymer Science Part a‐
Polymer Chemistry 2007, 45, (18), 4341‐4351. 165. Yuan, W. Y.; Dong, H.; Li, C. M.; Cui, X. Q.; Yu, L.; Lu, Z. S.; Zhou, Q., pH‐Controlled construction of chitosan/alginate multilayer film: Characterization and application for antibody immobilization. Langmuir 2007, 23, (26), 13046‐13052. 166. Zhu, Y. F.; Shi, J. L., A mesoporous core‐shell structure for pH‐controlled storage and release of water‐soluble drug. Microporous and Mesoporous Materials 2007, 103, (1‐3), 243‐249. 131 | P a g e 167. Lee, D.; Nolte, A. J.; Kunz, A. L.; Rubner, M. F.; Cohen, R. E., pH‐induced hysteretic gating of track‐etched polycarbonate membranes: Swelling/deswelling behavior of polyelectrolyte multilayers in confined geometry. Journal of the American Chemical Society 2006, 128, (26), 8521‐8529. 168. Hiller, J.; Rubner, M. F., Reversible molecular memory and pH‐switchable swelling transitions in polyelectrolyte multilayers. Macromolecules 2003, 36, (11), 4078‐4083. 169. Itano, K.; Choi, J. Y.; Rubner, M. F., Mechanism of the pH‐induced discontinuous swelling/deswelling transitions of poly(allylamine hydrochloride)‐containing polyelectrolyte multilayer films. Macromolecules 2005, 38, (8), 3450‐3460. 170. Haldar, J.; Weight, A. K.; Klibanov, A. M., Preparation, application and testing of permanent antibacterial and antiviral coatings. Nature Protocols 2007, 2, (10), 2412‐2417. 171. Choi, J.; Rubner, M. F., Influence of the degree of ionization on weak polyelectrolyte multilayer assembly. Macromolecules 2005, 38, (1), 116‐124. 172. Adamopoulos, L.; Montegna, J.; Hampikian, G.; Argyropoulos, D. S.; Heitmann, J.; Lucia, L. A., A simple method to tune the gross antibacterial activity of cellulosic biomaterials. Carbohydrate Polymers 2007, 69, (4), 805‐810. 173. Murguia, M. C.; Vaillard, V. A.; Sanchez, V. G.; Conza, J. D.; Grau, R. J., Synthesis, surface‐active properties, and antimicrobial activities of new double‐chain gemini surfactants. J Oleo Sci 2008, 57, (5), 301‐8. 174. Denyer, S. P.; Maillard, J. Y., Cellular impermeability and uptake of biocides and antibiotics in Gram‐negative bacteria. Journal of Applied Microbiology 2002, 92, 35s‐45s. 175. Guerin‐Mechin, L.; Dubois‐Brissonnet, F.; Heyd, B.; Leveau, J. Y., Quaternary ammonium compound stresses induce specific variations in fatty acid composition of Pseudomonas aeruginosa. International Journal of Food Microbiology 2000, 55, (1‐3), 157‐159. 176. Chia, K., Cohen, R. E., Rubner, M. F., Amine‐Rich Polyelectrolyte Multilayer Nanoreactors for in Situ Gold Nanoparticle Synthesis. Chemistry of Materials 2008, 20, (21), 6756–6763. 177. Adamczyk, Z.; Zembala, M.; Warszynski, P.; Jachimska, B., Characterization of polyelectrolyte multilayers by the streaming potential method. Langmuir 2004, 20, (24), 10517‐25. 178. Michna, A.; Adamczyk, Z.; Zembala, M., Deposition of colloid particles on polyelectrolyte multilayers. Colloids and Surfaces a‐Physicochemical and Engineering Aspects 2007, 302, (1‐3), 467‐
472. 179. Caruso, F.; Donath, E.; Mohwald, H., Influence of polyelectrolyte multilayer coatings on Forster resonance energy transfer between 6‐carboxyfluorescein and rhodamine B‐labeled particles in aqueous solution. Journal of Physical Chemistry B 1998, 102, (11), 2011‐2016. 180. Ladam, G.; Schaad, P.; Voegel, J. C.; Schaaf, P.; Decher, G.; Cuisinier, F., In situ determination of the structural properties of initially deposited polyelectrolyte multilayers. Langmuir 2000, 16, (3), 1249‐1255. 181. Gibson, J. B.; Chen, K.; Chynoweth, S., Simulation of Particle Adsorption onto a Polymer‐
Coated Surface Using the Dissipative Particle Dynamics Method. J Colloid Interface Sci 1998, 206, (2), 464‐474. 182. Ryser, H. J., A membrane effect of basic polymers dependent on molecular size. Nature 1967, 215, (5104), 934‐6. 183. Andersson, A. S.; Backhed, F.; von Euler, A.; Richter‐Dahlfors, A.; Sutherland, D.; Kasemo, B., Nanoscale features influence epithelial cell morphology and cytokine production. Biomaterials 2003, 24, (20), 3427‐36. 184. Pelham, R. J.; Wang, Y. L., Cell locomotion and focal adhesions are regulated by substrate flexibility. Proceedings of the National Academy of Sciences of the United States of America 1997, 94, (25), 13661‐13665. 132 | P a g e 185. Rimmer, S.; German, M. J.; Maughan, J.; Sun, Y.; Fullwood, N.; Ebdon, J.; MacNeil, S., Synthesis and properties of amphiphilic networks 3: preparation and characterization of block conetworks of poly(butyl methacrylate‐block‐(2,3 propandiol‐1‐methacrylate‐stat‐ethandiol dimethacrylate)). Biomaterials 2005, 26, (15), 2219‐30. 186. Kumar, G.; Wang, C. W.; Co, C.; Ho, C., Spatially controlled cell engineering on biomaterials using polyelectrolytes. Langmuir 2003, 19, (25), 10550‐10556. 187. McGregor, W. C., Interfacial Phenomena and Bioproducts. Marcel Dekker, Inc.: Madison, 1996; Vol. 23, p 510. 188. Hersel, U.; Dahmen, C.; Kessler, H., RGD modified polymers: biomaterials for stimulated cell adhesion and beyond. Biomaterials 2003, 24, (24), 4385‐415. 189. Gailit, J.; Clarke, C.; Newman, D.; Tonnesen, M. G.; Mosesson, M. W.; Clark, R. A., Human fibroblasts bind directly to fibrinogen at RGD sites through integrin alpha(v)beta3. Exp Cell Res 1997, 232, (1), 118‐26. 190. Berg, M. C.; Yang, S. Y.; Hammond, P. T.; Rubner, M. F., Controlling mammalian cell interactions on patterned polyelectrolyte multilayer surfaces. Langmuir 2004, 20, (4), 1362‐1368. 191. Lieb, E.; Hacker, M.; Tessmar, J.; Kunz‐Schughart, L. A.; Fiedler, J.; Dahmen, C.; Hersel, U.; Kessler, H.; Schulz, M. B.; Gopferich, A., Mediating specific cell adhesion to low‐adhesive diblock copolymers by instant modification with cyclic RGD peptides. Biomaterials 2005, 26, (15), 2333‐41. 192. Holland, J.; Hersh, L.; Bryhan, M.; Onyiriuka, E.; Ziegler, L., Culture of human vascular endothelial cells on an RGD‐containing synthetic peptide attached to a starch‐coated polystyrene surface: comparison with fibronectin‐coated tissue grade polystyrene. Biomaterials 1996, 17, (22), 2147‐56. 193. Drumheller, P. D.; Hubbell, J. A., Polymer networks with grafted cell adhesion peptides for highly biospecific cell adhesive substrates. Anal Biochem 1994, 222, (2), 380‐8. 194. Ono, S. S.; Decher, G., Preparation of ultrathin self‐standing polyelectrolyte multilayer membranes at physiological conditions using pH‐responsive film segments as sacrificial layers. Nano Letters 2006, 6, (4), 592‐598. 195. Cho, J.; Caruso, F., Polymeric multilayer films comprising deconstructible hydrogen‐bonded stacks confined between electrostatically assembled layers. Macromolecules 2003, 36, (8), 2845‐2851. 196. Kharlampieva, E.; Sukhishvili, S. A., Hydrogen‐bonded layer‐by‐layer polymer films. Polymer Reviews 2006, 46, (4), 377‐395. 197. Kozlovskaya, V.; Shamaev, A.; Sukhishvili, S. A., Tuning swelling pH and permeability of hydrogel multilayer capsules. Soft Matter 2008, 4, (7), 1499‐1507. 198. Kharlampieva, E.; Sukhishvili, S. A., Competition of hydrogen‐bonding and electrostatic interactions within hybrid polymer multilayers. Langmuir 2004, 20, (24), 10712‐7. 199. Kozlovskaya, V.; Kharlampieva, E.; Mansfield, M. L.; Sukhishvili, S. A., Poly(methacrylic acid) hydrogel films and capsules: Response to pH and ionic strength, and encapsulation of macromolecules. Chemistry of Materials 2006, 18, (2), 328‐336. 200. Laguecir, A.; Ulrich, S.; Labille, J.; Fatin‐Rouge, N.; Stoll, S.; Buffle, J., Size and pH effect on electrical and conformational behavior of poly(acrylic acid): Simulation and experiment. European Polymer Journal 2006, 42, (5), 1135‐1144. 201. Kirwan, L. J.; Papastavrou, G.; Borkovec, M.; Behrens, S. H., Imaging the coil‐to‐globule conformational transition of a weak polyelectrolyte by tuning the polyelectrolyte charge density. Nano Letters 2004, 4, (1), 149‐152. 202. Park, J.; Hammond, P. T., Polyelectrolyte multilayer formation on neutral hydrophobic surfaces. Macromolecules 2005, 38, (25), 10542‐10550. 203. Sukhishvili, S. A.; Granick, S., Layered, erasable polymer multilayers formed by hydrogen‐
bonded sequential self‐assembly. Macromolecules 2002, 35, (1), 301‐310. 133 | P a g e 204. Weyts, K. F.; Goethals, E. J., Back Titration of Linear Polyethylenimine ‐ Decrease of Ph by Addition of Sodium‐Hydroxide. Makromolekulare Chemie‐Rapid Communications 1989, 10, (6), 299‐
302. 205. Lichter, J. A.; Thompson, M. T.; Delgadillo, M.; Nishikawa, T.; Rubner, M. F.; Van Vliet, K. J., Substrata mechanical stiffness can regulate adhesion of viable bacteria. Biomacromolecules 2008, 9, (6), 1571‐1578. 206. Thompson, M. T.; Berg, M. C.; Tobias, I. S.; Lichter, J. A.; Rubner, M. F.; Van Vliet, K. J., Biochemical functionalization of polymeric cell substrata can alter mechanical compliance. Biomacromolecules 2006, 7, (6), 1990‐1995. 207. Palmer, J.; Flint, S.; Brooks, J., Bacterial cell attachment, the beginning of a biofilm. Journal of Industrial Microbiology & Biotechnology 2007, 34, (9), 577‐588. 208. Boe‐Hansen, R.; Albrechtsen, H. J.; Arvin, E.; Jorgensen, C., Bulk water phase and biofilm growth in drinking water at low nutrient conditions. Water Research 2002, 36, (18), 4477‐4486. 209. Costerton, J. W.; Stewart, P. S.; Greenberg, E. P., Bacterial biofilms: A common cause of persistent infections. Science 1999, 284, (5418), 1318‐1322. 210. Costerton, J. W., The Biofilm Primer. Springer‐Verlag Berlin Heidelberg: 2007. 211. Li, X. Z.; Webb, J. S.; Kjelleberg, S.; Rosche, B., Enhanced benzaldehyde tolerance in Zymomonas mobilis biofilms and the potential of biofilm applications in fine‐chemical production. Applied and Environmental Microbiology 2006, 72, (2), 1639‐1644. 212. Reguera, G.; Nevin, K. P.; Nicoll, J. S.; Covalla, S. F.; Woodard, T. L.; Lovley, D. R., Biofilm and nanowire production leads to increased current in Geobacter sulfurreducens fuel cells. Applied and Environmental Microbiology 2006, 72, (11), 7345‐7348. 213. Garrett, T. R.; Bhakoo, M.; Zhang, Z. B., Bacterial adhesion and biofilms on surfaces. Progress in Natural Science 2008, 18, (9), 1049‐1056. 214. Pavithra, D.; Doble, M., Biofilm formation, bacterial adhesion and host response on polymeric implants‐‐issues and prevention. Biomed Mater 2008, 3, (3), 034003. 215. Stepanovic, S.; Vukovic, D.; Hola, V.; Di Bonaventura, G.; Djukic, S.; Cirkovic, I.; Ruzicka, F., Quantification of biofilm in microtiter plates: overview of testing conditions and practical recommendations for assessment of biofilm production by staphylococci. Apmis 2007, 115, (8), 891‐
899. 216. Ista, L. K.; Perez‐Luna, V. H.; Lopez, G. P., Surface‐grafted, environmentally sensitive polymers for biofilm release. Applied and Environmental Microbiology 1999, 65, (4), 1603‐1609. 217. Burks, G. A.; Velegol, S. B.; Paramonova, E.; Lindenmuth, B. E.; Feick, J. D.; Logan, B. E., Macroscopic and nanoscale measurements of the adhesion of bacteria with varying outer layer surface composition. Langmuir 2003, 19, (6), 2366‐2371. 218. Lower, S. K.; Tadanier, C. J.; Hochella, M. F., Measuring interfacial and adhesion forces between bacteria and mineral surfaces with biological force microscopy. Geochimica Et Cosmochimica Acta 2000, 64, (18), 3133‐3139. 219. Poortinga, A. T.; Bos, R.; Busscher, H. J., Charge transfer during staphylococcal adhesion to TiNOX (R) coatings with different specific resistivity. Biophysical Chemistry 2001, 91, (3), 273‐279. 220. Ohashi, A.; Harada, H., A novel concept for evaluation of biofilm adhesion strength by applying tensile force and shear force. Water Science and Technology 1996, 34, (5‐6), 201‐211. 221. Chen, M. J.; Zhang, Z.; Bott, T. R., Direct measurement of the adhesive strength of biofilms in pipes by micromanipulation. Biotechnology Techniques 1998, 12, (12), 875‐880. 222. Singleton, S.; Treloar, R.; Warren, P.; Watson, G. K.; Hodgson, R.; Allison, C., Methods for microscopic characterization of oral biofilms: analysis of colonization, microstructure, and molecular transport phenomena. Adv Dent Res 1997, 11, (1), 133‐49. 134 | P a g e 223. Vroom, J. M.; De Grauw, K. J.; Gerritsen, H. C.; Bradshaw, D. J.; Marsh, P. D.; Watson, G. K.; Birmingham, J. J.; Allison, C., Depth penetration and detection of pH gradients in biofilms by two‐
photon excitation microscopy. Appl Environ Microbiol 1999, 65, (8), 3502‐11. 224. Wingender, J., Neu, T.R., and Flemming, H.C., Microbial Extracellular Polymeric Substances. Springer: 1999. 225. DeLongchamp, D. M.; Hammond, P. T., High‐contrast electrochromism and controllable dissolution of assembled Prussian blue/polymer nanocomposites. Advanced Functional Materials 2004, 14, (3), 224‐232. 226. DeLongchamp, D. M.; Hammond, P. T., Multiple‐color electrochromism from layer‐by‐layer‐
assembled polyaniline/Prussian Blue nanocomposite thin films. Chemistry of Materials 2004, 16, (23), 4799‐4805. 227. Linder, A.; Weiland, U.; Apell, H. J., Novel polymer substrates for SFM investigations of living cells, biological membranes, and proteins. J Struct Biol 1999, 126, (1), 16‐26. 228. Lee, K. Y.; Mooney, D. J., Hydrogels for Tissue Engineering. Chemical Reviews 2001, 101, (7), 1869‐1879. 229. Keselowsky, B. G.; Collard, D. M.; Garcia, A. J., Integrin binding specificity regulates biomaterial surface chemistry effects on cell differentiation. Proc Natl Acad Sci U S A 2005, 102, (17), 5953‐7. 230. Liao, H.; Andersson, A. S.; Sutherland, D.; Petronis, S.; Kasemo, B.; Thomsen, P., Response of rat osteoblast‐like cells to microstructured model surfaces in vitro. Biomaterials 2003, 24, (4), 649‐54. 231. Ignatius, A.; Peraus, M.; Schorlemmer, S.; Augat, P.; Burger, W.; Leyen, S.; Claes, L., Osseointegration of alumina with a bioactive coating under load‐bearing and unloaded conditions. Biomaterials 2005, 26, (15), 2325‐32. 232. Gailit, J.; Clark, R. A., Wound repair in the context of extracellular matrix. Curr Opin Cell Biol 1994, 6, (5), 717‐25. 233. Feng, X.; Clark, R. A.; Galanakis, D.; Tonnesen, M. G., Fibrin and collagen differentially regulate human dermal microvascular endothelial cell integrins: stabilization of alphav/beta3 mRNA by fibrin1. J Invest Dermatol 1999, 113, (6), 913‐9. 234. Boura, C.; Muller, S.; Vautier, D.; Dumas, D.; Schaaf, P.; Voegel, J. C.; Stoltz, J. F.; Menu, P., Endothelial cell‐interactions with polyelectrolyte multilayer films. Biomaterials 2005, 26, 4568‐4575. 235. Engler, A. J.; Richert, L.; Wong, J. Y.; Picart, C.; Discher, D. E., Surface probe measurements of the elasticity of sectioned tissue, thin gels and polyelectrolyte multilayer films: Correlations between substrate stiffness and cell adhesion. Surface Science 2004, 570, 142‐154. 236. Ngankam, A.; Mao, G.; Tassel, P. V., Fibronectin adsorption onto polyelectrolyte multilayer films. Langmuir 2004, 20, 3362‐3370. 237. Salloum, D.; Olenych, S.; Keller, T.; Schlenoff, J., Vascular smooth muscle cells on polyelectrolyte multilayers: hydrophobicity‐directed adhesion and growth. Biomacromolecules 2005, 6, 161‐167. 238. Brown, X.; Ookawa, K.; Wong, J., Evaluation of polydimethylsiloxane scaffolds with physiologically‐relevant elastic moduli: interplay of substrate mechanics and surface chemistry effects on vascular smooth muscle cell response. Biomaterials 2005, 26, 3123‐3129. 239. Marchand, P.; Marmet, L., Binomial Smoothing Filter ‐ a Way to Avoid Some Pitfalls of Least‐
Squares Polynomial Smoothing. Review of Scientific Instruments 1983, 54, (8), 1034‐1041. 240. Dimitriadis, E.; Horkay, F.; Maresca, J.; Kachar, B.; Chadwick, R., Determination of elastic moduli of thin layers of soft material using the atomic force microscope. Biophysical Journal 2002, 82, 2798‐2810. 241. Domke, J.; Radmacher, M., Measuring the elastic properties of thin polymer films with the atomic force microscope. Langmuir 1998, 14, (12), 3320‐3325. 135 | P a g e 242. Yang, S. Y.; Berg, M. C.; Hammond, P. T.; Rubner, M. F. In Primary hepatocytes and fibroblasts response to polyelectrolyte multilayers with polyacrylamide containing polymers, American Chemical Society, 2003; 2003. 136 | P a g e APPENDIX A
Additional Characterization of PAA/PAH PEMs with PAA at pH 3.5
Dry-state atomic force microscopy (AFM) images of the PAA/PAH system assembled
with PAA at pH 3.5 and PAH at either pH 7.5 or pH 8.6 are shown below. Images were acquired
in tapping mode with a Veeco/Digital Instruments Nanoscope IIIa Scanned Probe Microscope
Controller with Dimension 3000 SPM. Roughness values were calculated using the Nanoscope
IIIa software.
Figure 1. AFM images of PAA/PAH (a) 3.5/7.5 (z‐height = 50 nm) and (b) 3.5/8.6 (z‐height = 90 nm). Root‐mean‐
square (RMS) roughness values are given below the images. 137 | P a g e In-situ images were acquired by M. Todd Thompson of the Van Vliet Laboratory at MIT
with a 3D Molecular Force Probe (Asylum Research, Santa Barbara, CA). Samples were
hydrated in sterile, deionized water overnight before imaging.
Figure 2. In‐situ SPM images of PAA/PAH (a) 3.5/7.5 and (b) 3.5/8.6. 138 | P a g e APPENDIX B
Note: References for Appendix B are included in main bibliography
Biochemical functionalization of polymeric cell substrata can
alter mechanical compliance
M. Todd Thompson*, #, Michael C. Berg§, Irene S. Tobias*, Jenny A. Lichter*,
Michael F. Rubner*, Krystyn J. Van Vliet*,¶
*Department of Materials Science and Engineering, Massachusetts Institute of Technology, 77
Massachusetts Avenue, Cambridge, MA 02139 USA
§
Department of Chemical Engineering, Massachusetts Institute of Technology,
77 Massachusetts Avenue, Cambridge, MA 02139 USA
#
Harvard-MIT Division of Health Sciences and Technology, 77 Massachusetts Avenue,
Cambridge, MA 02139 USA
¶
Corresponding Author Information:
Krystyn J. Van Vliet
MIT; Room 8-237
77 Massachusetts Avenue
Cambridge, MA 02139 USA
Phone:
617.253.3315
Fax:
617.253.8745
Email: krystyn@mit.edu
URL:
http://web.mit.edu/vvgroup
Keywords: contact mechanics, mechanical compliance, polyelectrolyte multilayer, RGD, surface
functionalization
139 | P a g e ABSTRACT:
Biochemical functionalization of surfaces is an increasingly utilized mechanism to promote or
inhibit adhesion of cells. To promote cell-substrata adhesion, one common example of this
approach is surface conjugation of adhesion peptide sequences such as Arg-Gly-Asp (RGD)
which bind specifically to transmembrane integrin receptors of the cell. In the context of cell
mechanotransduction, it is assumed that such functionalization does not alter the local
mechanical properties of the functionalized surface. Here, we examine this assumption
systematically, through nanomechanical measurement of the nominal elastic modulus of
polyelectrolyte multilayer (PEM) thin films functionalized with RGD through different
processing routes. We find that the method of biochemical functionalization can significantly
alter mechanical compliance of polymeric substrata such as PEMs. In particular, immersed
adsorption of intermediate functionalization reagents significantly decreases compliance of the
PEMs considered herein, whereas polymer-on-polymer stamping of these same reagents does not
alter compliance. This finding points to the potential, unintended alteration of mechanical
properties via surface functionalization, and also suggests functionalization methods by which
chemical and mechanical properties of cell substrata can be controlled independently for
applications including cell mechanotransduction studies.
INTRODUCTION
Surface functionalization to promote cellular adhesion to biomaterials used as cellular growth
substrata is an important component of many biological research efforts and bioengineering
applications. High resolution imaging of cytoskeletal substructure and dynamics is critically
dependent on the ability to successfully immobilize cells through formation of tight adhesive
contacts 227. In addition, in vitro culture of adherent cell types, whether for tissue engineering or
cell biology studies, also depends on the quality and strength of adhesion events 114, 187, 228. In the
field of medical implants, precise control of cellular attachment is necessary to prevent
140 | P a g e microbiological contamination and promote proper graft response, a topic of particular interest in
the area of osteogenic implantable devices 229-231.
Indeed, interfacial biology is a well-developed and rich field, and many types of
biointerfacial modifications exist to promote the attachment and proliferation of cells on given
growth substrates 187. Techniques to induce phenotypic change and control spatial distribution in
various cell types include alteration of surface topology
crosslinking in a polymeric gel
184
183
and/or degree of interchain
, creation of phase separated amphiphilic surfaces
185
, and
functionalization with cell resistant materials that restrict cell growth and enforce patterning
186
.
With increasing frequency, cytophilic surface modifications are employed via adsorption of
extracellular matrix proteins or related derivatives onto a rigid or semi-rigid support to
reconstitute aspects of the in vivo extracellular environment. One widely used approach involves
the incorporation of proteins or peptides containing the sequence Arg-Gly-Asp (RGD), which
recruits and binds to integrin receptors on the surfaces of eukaryotic cells
187-193
.
This is
particularly significant because differential integrin binding alters specific cellular behaviors
such as differentiation in human unmbilical vascular endothelial cells
229
.
Conversely,
differential integrin expression is known to be an important marker of cell state during
angiogenesis and capillary invasion during wound healing 232, 233.
Increasingly, polyelectrolyte multilayers (PEMs) are used as bioactive substrata for the
study of cell adhesion or phenotype
1, 62, 113, 190, 234-237
. PEMs are polyelectrolyte complexes
assembled via a layer-by-layer assembly process with dilute solutions of positively and
negatively charged polymers. Because the physical properties and film thickness of weak (pHsensitive) PEMs can be controlled with high precision through assembly conditions such as
solution pH, these materials find utility in a range of applications including but not limited to
141 | P a g e cytophilic substrata and cytophobic coatings. Importantly, these materials effectively modulate
cell behavior when assembled to only nanoscale thicknesses
114
, and are thus amenable to high
resolution optical imaging approaches desirable for a range of in vitro cell experiments. Berg et
al. have demonstrated that the cytophobic properties of a PEM comprising polyacrylic acid
(PAA) and polyacrylamide (PAAm) can be reversed via surface functionalization with RGD 190.
In these studies, it is assumed but not demonstrated that biochemical functionalization of such
surfaces does not alter the mechanical properties of that surface, such that the mechanical and
chemical characteristics of substrata can be modulated independently to evaluate cell response.
That is, if surface modifications such as RGD incorporation alter only the biochemical interface
between the substrata and adhered cells, then cellular processes such as adhesion, spreading,
proliferation, and differentiation on those surfaces could be attributed unambiguously to
biochemical rather than mechanical characteristics of the substrata.
Thompson et al. have shown that mechanical compliance of nanoscale PEM films can be
modulated directly via assembly conditions
114
.
For weak PEMs comprising PAA and
poly(allylamine hydrochloride) or PAH, nominal elastic modulus E varies by orders of
magnitude for mod-2 changes in assembly pH.
Further, Thompson et al. found that this
mechanical compliance correlated directly with the capacity of mammalian (microvascular
endothelial) cells to attach to and proliferate on unfunctionalized PEMs under in vitro culture,
and others have demonstrated similar effects of mechanical compliance for other PEM or
hydrogel systems on different adherent mammalian cell types
113, 235, 238
. In light of these
previous findings on biochemical and mechanical modulation of cell-substrate adhesion, here we
sought to confirm that the mechanical properties of PEMs were unaffected by a particular
biochemical surface functionalization process.
142 | P a g e To that end, we employed scanning probe
microscope-enabled nanoindentation to measure the nominal elastic modulus E of PEMs
functionalized through various processing routes with a synthetic peptide containing the integrin
binding sequence RGD.
MATERIALS AND METHODS
Materials
Poly(acrylic acid) (PAA) (MW = 90 000; 25% aqueous solution and polyacrylamide (PAAm)
(MW = 5 000 000; 1% aqueous solution) were purchased from Polysciences. Poly(allylamine
hydrochloride) (PAH) (MW = 70 000) was purchased from Sigma-Aldrich. Peptides GRGDSPC
and GRGESPC were provided by the MIT Biopolymers Lab.
Sulfosuccinimidyl 6-[3’-(2-
pyridyldithio)-proprionamido] hexanoate (Sulfo-LC-SPDP) was purchased from Pierce
Biotechnology. Poly(dimethylsiloxane) (PDMS) stamps were made obtained according to the
previously described protocol 190.
Polymer electrolyte multilayer assembly
PEM films were assembled as previously described 1, 62, 159, 190. Briefly, dilute solutions (0.01 M)
of PAA, PAAm, and PAH were prepared in deionized water and the solution pH adjusted to 3.0
using HCl. The multilayers were assembled on standard glass slides, silicon wafers, and in 60
mm-diameter polystyrene petri dishes using an automated layer-by-layer dipping method. Each
sample was assembled with one layer of PAH to promote strong adhesion of the PAA/PAAm
PEM, followed by 5.5 bilayers of PAA/PAAm. The PEMs were then covalently crosslinked, as
143 | P a g e required for stability at neutral pH conditions required for cell culture, via elevated temperature
in vacuum (180oC, 2 hrs for glass and silicon, 90oC, overnight for polystyrene).
For surface-modified samples, PAH was first added to the surface by one of two routes.
In the first case, surface modification was achieved via incubation of the PEM sample in a 0.01
M / pH = 9.0 polyelectrolyte solution at room temperature for 15 min or 30 sec (hereafter termed
adsorbed PAH). In the second case, surface modification was achieved via polymer-on-polymer
transfer with a patterned PDMS stamp inked with 0.05 M PAH / pH = 9.0, as described
previously (hereafter termed stamped PAH) 190. Briefly, PDMS polymer stamps were soaked in
a solution containing PAH at the aforementioned concentration and then allowed to physically
contact the PAA/PAAm PEM surface for 30 sec before removal. The PEMs were then rinsed
with 150 mM / pH = 7.4 phosphate buffered saline (PBS) several times under agitation, and
allowed to dry in air for subsequent rehydration and use.
Modification of PAH-treated PEMs with RGD or Arg-Gly-Glu (RGE) was accomplished
first by incubation of 0.5 mM Sulfo-LC-SPDP in the presence of PAH treated PEMs for 30 min
at room temperature. Following the addition of this heterobifunctional crosslinker, the samples
were washed with PBS twice for 5 min. Incubation of 0.5 mM peptide solution (GRGDSPC or
GRGESPC) in PBS for 8 hours at room temperature yielded RGD and RGE modified
PAA/PAAm samples, ostensibly conjugated to the heterobifunctional crosslinker via a disulfide
linkage. PEMs were rinsed several times in PBS under agitation, and allowed to dry in air for
subsequent rehydration and use.
144 | P a g e Mechanical testing of PAA/PAH multilayers
Mechanical testing was performed using a scanning probe microscope well-suited for the forcedepth acquisition required of nanoindentation (Molecular Force Probe 3D, Asylum Research,
Santa Barbara, CA).
Commercially available unsharpened silicon nitride cantilevers (MLCT-
AUHW, Veeco Metrology Group, Sunnyvale, CA) were used to indent PEMs to maximum
depths of ~20 nm under constant force. The probe tip radius of curvature Rp was ~50 nm, and
the cantilever spring constant k was experimentally determined for each cantilever via the
thermal power spectral density approach to be within the factory specifications of 0.1 N/m by
factor of two 114.
Nanoindentation was performed in an acoustic isolation enclosure (Herzan, Inc.) at room
temperature in 0.2 μm-filtered PBS. At least 50 nanoindentation experiments were conductued
on each polymer sample, and each indentation was performed at a unique point on the sample
surface to rule out effects due to cyclic loading and/or plastic deformation. To ensure that
indentation occurred at sites of PAH patterning in PMDS-stamped samples, the sample surface
was first imaged in contact mode using the MFP3D (90 μm x 90 μm) . Furthermore, multiple
regions were scanned over a sample area that spanned ~50% the total stamped region of the
PEM. Finally, multiple samples of the PAH-stamped samples were tested on different days, to
identify any sample-to-sample variations and systematic experimental errors.
145 | P a g e Nanomechanical data analysis
Nanoindentation force-depth data were analyzed in IGOR (Wavemetrics, Lake Oswego, OR) on
a windows platform PC using the method previously described
114
. The nominal elastic moduli
of the PEM samples were calculated using a spherical Hertzian model of the form
P = (4/3) Er1/2 Rp Δ3/2
(1)
where P is applied force, Er is the reduced elastic modulus, Rp is the radius of curvature of the
cantilevered probe, and Δ is the depth of penetration into the sample surface 123.
Force-displacement datawere processed prior to analysis using a 25 pass binomial
smoothing filter to eliminate random fluctuations 239. Since accurate determination of the initial
contact point is a critical issue in nanoindentation of compliant polymer films240,
241
, an
additional noise threshold was applied to the log P – log h representation of smoothed curves to
identify this (0,0) point objectively and repeatably. Linear least-square fits of the log P – log h
representation of smoothed responses were conducted and yielded intercept values from which
nominal E were calculated from Eq. (1), as previously described 114.
PEM Film Thickness Measurement
In order to determine whether any experimentally observed differences in PEM mechanical
compliance could be attributed to differences in hydrated film thickness t, hydrated samples were
assembled on glass substrate and imaged via scanning probe microscopy (SPM) over regions
146 | P a g e including through-thickness scratches. Unmodifiied PAA/PAAm, PAA/PAAm/adsorbed PAH,
PAA/PAAm/stamped PAH, and PAA/PAAm/stamped PAH/RGE PEMs of nanoscale thickness
were prepared on glass slides as described above. Sample slides were cleaned by dipping in
sterile 0.2
m filtered PBS, rehydrated in PBS, and scratched with a standard razor blade. PEMs
were imaged in contact mode (MFP3D) using a Si3N4 probe of k = 0.06 N/m over regions
including the scratch site at both 0o and 90o scan angles. Height measurements were calculated
by measuring ΔZ at six different randomly selected regions, where,
ΔZ = ZPEM surface - Ztrough
(2)
Standard deviation of the mean sample height was significantly smaller than the associated error
in the surface roughness across the trough in individual image cross-sections, which can be
attributed to slight damage of the underlying glass substrate and/or limited residual PEM within
the scratch trough. Root mean square (RMS) surface roughness was determined directly from
contact images via MFP3D IGOR subroutines. Average +/- standard deviation RMS roughness
among six cross-sections within a given sample image is reported. In addition, in situ
ellipsometry (ISE) was employed to validate SPM measurements of water-hydrated film
thickness t for the same PEMs assembled on silicon (reflective) substrates. ISE determines t as a
function of changes in indices of refraction n measured via light reflected from the material
surface, and samples mm2-scale surface areas 62.
147 | P a g e RESULTS AND DISCUSSION
Previous studies have demonstrated that the unmodified PAA/PAAm system is completely
cytophobic to both hepatocytes and human microvascular endothelial cells
114, 190
. However,
using a patterned polymer-on-polymer stamping technique, Berg et al. demonstrated that PAHstamping followed by covalent conjugation of RGD-containing peptides at the multilayer surface
could switch the cytophobicity of PAA/PAAm to that of a cytophilic substrate in a PAH
concentration-dependent manner (See Fig. 1). This response was not reproduced via conjugation
of the dummy peptide sequence RGE, and thus attributed to specific chemical interactions
between this particular adhesive ligand and the mammalian cell surfaces 190.
Figure 1: Wild-type NR6 fibroblast attachment as a function of RGD concentration on PAA/PAH PEMs,
surface modified via polymer-on-polymer stamping of PAH in a vertical line pattern followed by RGD
conjugation via a heterobifunctional crosslinker. Cells do not adhere readily on PAA/PAAm PEM lines
functionalized with low RGD concentrations of ~53,000 molecules/μm2 (A), but do adhere readily to the same
PAA/PAAm lines functionalized with a higher RGD concentration of 152,000 molecules/μm2 (B). Scalebars =
50 μm. These materials and cell adhesion results are detailed in Ref. 85.
148 | P a g e Effect of surface functionalization on mechanical compliance
To ascertain any changes in mechanical properties of PAA/PAAm PEMs that such
surface engineering may engender, instrumented nanoindentation was performed on samples
representing each processing step during surface modification of PAA/PAAm with RGD or
RGE. Representative P – ∆  responses for each PAA/PAAm sample is shown in Fig. 2, and
nominal elastic moduli E calculated from such data are shown as a function of surface
modification in Fig. 3. Unmodified PAA/PAAm exhibits the lowest nominal E (2.4 x 105 Pa),
consistent with the high swelling capacity and low crosslinking density of this PEM, as well as
with previous mechanical analysis of this polymer film
1, 114, 190, 242
. The second step in the
process of surface engineering involves the addition of PAH as a base for conjugation of RGD.
This can be readily accomplished by adsorption of the polymer chain from a dilute solution of
PAH, or by polymer-on-polymer stamping as described by Berg et al.
190
. Samples prepared
with PAH according to this stamping protocol exhibited a slightly lower mean E with respect to
the unmodified PEM (1.6 x 105 Pa), although this difference was found to be within the margin
of error of the nanoindentation approach. Together with the mechanical characterization and
cytophobicity results of the unmodified PAA/PAAm PEM from Thompson et al.
114
, these
findings suggest the reversal of cytophobicity in RGD-modified PAA/PAAm via polymer-onpolymer stamping is due chiefly to changes in surface chemistry that alter cellular attachment,
most likely through an integrin mediated mechanism, and not to changes in mechanical
compliance of the polymer substrata.
In contrast, PEMs modified by adsorption of PAH for 15 min exhibited E = 4.16 x 107
Pa, an increase in mechanical stiffness by more than two orders of magnitude. PEMs modified
149 | P a g e Figure 2: Representative force-depth responses acquired during nanoindentation of PAA/PAAm PEMs in
PBS PAA/PAAm/adsorbed PAH, 15 min (solid black); PAA/PAAm/adsorbed PAH + RGD (solid gray);
unmodified PAA/PAAm (dashed black); PAA/PAAm/stamped PAH, 30 sec (dashed gray).
Figure 3: Nominal elastic moduli E as measured by instrumented nanoindentation of surface modified
PAA/PAAm polyelectrolyte multilayers. PEMs were indented to a depth of 20 nm using a scanning probe
microscope in fluid (150 mM PBS, pH = 7.4) at room temperature.
via adsorbed PAH followed by either RGD or RGE peptide conjugation showed similar,
dramatic increases in stiffness (E = 1.67 x 107 Pa and 6.74 x 106 Pa, respectively) with respect to
150 | P a g e the unmodified PEM or the stamped PAH modification. Therefore, it is demonstrated that the
transition from a mechanically compliant PEM to a mechanically stiff PEM occurs at the point of
PAH adsorption, and not through the addition of the Sulfo-LC-SPDP heterobifunctional
crosslinker or the RGD/RGE heptamers. Reducing the PAH incubation time to 30 sec, the time
the time scale of PDMS stamping, showed only a modest reduction in the stiffness (E = 6.15 x
106 Pa), suggesting that this material modification occurs rapidly.
Consideration of PEM film thickness
It is not immediately apparent why the compliance of PAA/PAAm/adsorbed PAH PEMs is so
dramatically affected by adsorption of this polycation. One possible explanation is that the
sample thickness decreases significantly after adsorption of PAH (e.g., by increased interchain
crosslinking), such that mechanical probing of all samples to the same depth (∆ ~ 20 nm) induces
artifacts associated with proximity to the rigid polystyrene substrate on which the PEMs were
assembled. To address this possibility, PEM film thickness was determined via scanning probe
microscopy contact-mode imaging for all samples. As shown in Table 1, surface modifications
did not decrease PEM thickness. In fact, the nanoscale thickness and RMS surface roughness of
PAA/PAAm with an adsorbed layer of PAH is slightly greater than that of unmodified
PAA/PAAm, which is consistent with the increased deposition of more material in the modified
film. ISE results for the same PEMs assembled on silicon and hydrated with water were
consistent with these SPM measurements of hydrated film thickness t, and are representative of a
much larger surface area than considered via SPM. Hydrated t of unmodified and adsorbed PAH
PEMs measured via ISE was ~100 nm, and that of stamped PAH PEMs with and without RGE
151 | P a g e heptamer was ~230 nm. Thus, the observed change in E between unmodified PAA/PAAm and
the associated PAH adsorbed derivative cannot be attributed to a change in sample thickness.
TABLE 1 Properties of PAA/PAAm polyelectrolyte multilayer derivatives*
Sample
PAA/PAAm bilayers
Hydrated thickness
[nm]
Surface roughness
[nm]
E
[105 Pa]
PAA/PAAm
5.5
87.8 ± 19.3
34.1 ± 23.5
2.4 ± 1.7
PAA/PAAm,
Adsorbed PAH†
5.5
99.8 ± 16.6
52.0 ± 37.8
416.0 ± 89.2
PAA/PAAm,
Stamped PAH
5.5
213.1 ± 59.5
130.2 ± 95.7
15.9 ± 0.6
PAA/PAAm,
RGD modified‡
5.5
--167.0 ± 60.0
PAA/PAAm,
5.5
214.2 ± 48.0
94.5 ± 69.6
67.4 ± 19.9
RGE modified‡
*Young’s moduli E are measured via nanoindentation. Hydrated thickness and surface roughness were acquired
separately through scanning probe microscopy imaging of a surface area including through-thickness scratch.
† PAH adsorbed for 30 sec
‡ PAH stamped, followed by Sulfo-LC-SPDP and RGD or RGE heptamer, as indicated. RGD modified samples
were not analyzed for hydrated thickness and surface roughness to conserve peptide, but the difference of only one
amino acid between the RGD and RGE samples would not be predictive of any differences between these samples.
Neither the amount of total PAH adsorbed onto the surface nor the amount of PAH
transferred via stamping were quantified rigorously. Therefore, it remains possible that observed
increases in the E upon PAH adsorption are related to differences in the amount of PAH
integrated within the PEM surface in each deposition protocol or to potential phase transitions /
separations that would be consistent with the slightly increased opacity of the PEM upon PAH
adsorption. However, the central finding remains clear: Mechanical properties of weak PEMs
can be significantly and unintentionally altered via certain biochemical surface modification
routes, and these effects are independent of PEM thickness.
152 | P a g e CONCLUSIONS
Biochemical surface modification of polymeric growth substrates to enhance or inhibit cellular
attachment is important for a wide range of biological and bioengineering problems. It is tacitly
assumed that these modifications, including incorporation of adhesion proteins and peptides such
as RGD, alter only the local chemical environment and leave the mechanical properties of the
surface unaffected. Here, the effect of RGD incorporation on the mechanical compliance of a
specific polyelectrolyte multilayer system comprising poly(acrylic acid) and poly(acrylamide)
has been characterized systematically. Significant, processing dependent changes in nominal
elastic modulus E have been demonstrated: for the weak PEM considered herein, surface
functionalization with RGD via polymer-on-polymer stamping of dilute PAH does not alter
mechanical compliance, whereas functionalization via incubation in dilute PAH over the same
duration dramatically increases E.
Thus, for weak polyelectrolyte multilayers of nanoscale
thickness such as PAA/PAAm, the method by which the cellular interface is modified can have
unintended and profound consequences on mechanical compliance of the substrata and thereby
the mechanical environment of attached cells.
Furthermore, the changes in substratum
mechanical compliance demonstrated herein cannot be attributed to changes in sample thickness
or surface roughness. It is an open and important question whether these results are generally
true for other polyelectrolyte multilayer systems and/or polymeric hydrogels. These findings
serve both as a caution in the design of surfaces and experiments for which only chemical
modification is desired, and as an opportunity to choose surface modification routes that alter
mechanical and biochemical interfaces independently.
153 | P a g e M.T. Thompson and K.J. Van Vliet acknowledge use of the MFP3D (Asylum Research, Inc.) within the MIT
NanoMechanical Technology Laboratory. J.A. Lichter and M.C. Berg acknowledge use of shared experimental
facilities in the Institute for Soldier Nanotechnologies.
M.T. Thompson acknowledges support through the National Institutes of Health Bioinformatics Training Grant.
M.F. Rubner and J.A. Lichter acknowledge financial support from the DuPont-MIT Alliance. This work was
supported in part by and made use of the Shared Experimental Facilities supported by the MRSEC Program of the
National Science Foundation under award number DMR 02-13282.
154 | P a g e