University of Tennessee, Knoxville TRACE: Tennessee Research and Creative Exchange
Doctoral Dissertations Graduate School
5-2008
Vertically Aligned Carbon Nanofiber Arrays as a Platform for Gene Delivery and Expression Analysis in Mammalian Cells
David George James Mann University of Tennessee - Knoxville
Follow this and additional works at: https://trace.tennessee.edu/utk_graddiss
Part of the Microbiology Commons
Recommended Citation Mann, David George James, "Vertically Aligned Carbon Nanofiber Arrays as a Platform for Gene Delivery and Expression Analysis in Mammalian Cells. " PhD diss., University of Tennessee, 2008. https://trace.tennessee.edu/utk_graddiss/362
This Dissertation is brought to you for free and open access by the Graduate School at TRACE: Tennessee Research and Creative Exchange. It has been accepted for inclusion in Doctoral Dissertations by an authorized administrator of TRACE: Tennessee Research and Creative Exchange. For more information, please contact [email protected]. To the Graduate Council:
I am submitting herewith a dissertation written by David George James Mann entitled "Vertically Aligned Carbon Nanofiber Arrays as a Platform for Gene Delivery and Expression Analysis in Mammalian Cells." I have examined the final electronic copy of this dissertation for form and content and recommend that it be accepted in partial fulfillment of the equirr ements for the degree of Doctor of Philosophy, with a major in Microbiology.
Gary S. Sayler, Major Professor
We have read this dissertation and recommend its acceptance:
Michael L. Simpson, Steven W. Wilhelm, Todd B. Reynolds
Accepted for the Council:
Carolyn R. Hodges
Vice Provost and Dean of the Graduate School
(Original signatures are on file with official studentecor r ds.) To the Graduate Council:
I am submitting herewith a dissertation written by David George James Mann entitled “Vertically Aligned Carbon Nanofiber Arrays as a Platform for Gene Delivery and Expression Analysis in Mammalian Cells.” I have examined the final electronic copy of this dissertation for form and content and recommend that it be accepted in partial fulfillment of the requirements for the degree of Doctor of Philosophy, with a major in Microbiology.
______
Gary S. Sayler, Major Professor
We have read this dissertation and recommend its acceptance:
______
Michael L. Simpson
______
Steven W. Wilhelm
______
Todd B. Reynolds
Accepted for the Council:
______
Carolyn R. Hodges, Vice Provost and Dean of the Graduate School
(Original signatures are on file with official student records.)
VERTICALLY ALIGNED CARBON NANOFIBER ARRAYS
AS A PLATFORM FOR GENE DELIVERY AND
EXPRESSION ANALYSIS IN MAMMALIAN CELLS
A Dissertation
Presented for the Doctor of Philosophy Degree
University of Tennessee, Knoxville
DAVID GEORGE JAMES MANN
May 2008
DEDICATION
I dedicate this dissertation to my wife and best friend, Laura, who has lovingly and
patiently stood by me through it all.
The tapestry gets more and more beautiful every day.
ii ACKNOWLEDGEMENTS
I would first like to thank my mentor on this project, Tim McKnight, for his seemingly limitless supply of ideas and for his ambition and energy. He has taught me to go beyond the boundary of specific disciplines and more importantly, that “there is always time for one more experiment.”
I want to thank my major professor, Dr. Gary Sayler, for his insight and guidance throughout this project and for teaching me to step into the role of an independent scientist. I would additionally like to thank the other members of my committee, Dr.
Michael Simpson, Dr. Steven Wilhelm and Dr. Todd Reynolds for their continual support and input.
I would like to thank Jack McPherson, the molecular biology wizard, who taught me much of what I know at the bench. His style of encouragement will always make me smile and not soon be forgotten.
I want to thank Dr. Guy Griffin, who has the most respectable cell culture technique I have ever observed and who took the time to train me on culturing the CHOsen ones in between stories of the Lost Dutchman Mine.
iii I would like to thank all my coworkers at the CEB over the years for their friendship.
Memories from the graduate office will not soon be forgotten.
I want to thank my family for their love and support and for encouraging me to pursue the biological sciences and to always set high goals.
And most importantly, I would like to thank my wife Laura for her joy, patience and comfort throughout this project. Thanks for holding my hand all along the way.
iv ABSTRACT
Vertically aligned carbon nanofiber (VACNF) arrays have been developed as a novel tool for direct physical introduction and expression of DNA in mammalian cells (termed impalefection). This study describes the optimization of impalefection, the quantification of immobilized DNA on VACNFs, and the application of VACNFs in analysing gene expression in mammalian cells. Mechanical, chemical and biological parameters were optimized for impalefection. Alterations in a majority of the parameters resulted in no significant difference in impalefection efficiency, including nanofiber composition, DNA precipitation, cell confluency, cell concentration and sodium butyrate. The optimal DNA concentration ranged between 100 nanograms and 1 microgram, and the optimal impalefection substrate proved to be a Durx filter pad on plastic surface. High levels of efficiency in a wide range of mammalian cell lines demonstrated the versatile applicability of the impalefection method. Polymerase chain reaction (PCR) and in-vitro transcription (IVT) were used to investigate the transcriptional accessibility of immobilized DNA on VACNF arrays by correlating the yields of both IVT and PCR to that of non-immobilized DNA. Quantitative PCR was used to quantify the number of accessible yfp reporter gene copies immobilized to nanofiber arrays. DNA yields decreased dramatically in the non-immobilized control over time, while the majority of immobilized DNA was retained on VACNF arrays. These data demonstrated the development of methods for monitoring DNA immobilization techniques. To validate the applicability of VACNF arrays for controlling and monitoring mammalian gene expression, a tetracycline-inducible shRNA vector system was designed for silencing
v CFP expression and was impalefected into mammalian cells. VACNF arrays provided simultaneous delivery of multiple genes, subsequent adherence and proliferation of cells, and repeated monitoring of single cells over time. Following impalefection and tetracycline induction, 53.1% ± 10.4% of impalefected cells were fully silenced by the inducible shRNA vector. Additionally, efficient CFP-silencing was observed in single cells among a population of cells that remained CFP-expressing. This effective transient expression system enabled rapid analysis of gene silencing effects using RNAi in single cells and cell populations.
vi TABLE OF CONTENTS
CHAPTER I...... 1 LITERATURE REVIEW ...... 1 Background...... 1 What is Nanotechnology?...... 1 Nanotoxicology: An Emerging Field...... 4 The Promise of Nanotechnology in the Biological Realm ...... 8 Vertically Aligned Carbon Nanofibers as a Tool of Nanobiotechnology...... 12 Methods and Chemistry of Covalent DNA Immobilization...... 15 Traditional Methods of Gene Delivery in Mammalian Cells ...... 21 Impalefection: Vertically Aligned Carbon Nanofiber-Mediated Gene Delivery .... 24 Potential Mammalian Cell Responses to Impalefection ...... 31 Inducible Gene Silencing and the RNA Interference Pathway...... 33 OBJECTIVES OF THIS STUDY...... 37
CHAPTER II...... 39 OPTIMIZATION OF THE VERTICALLY-ALIGNED CARBON NANOFIBER- MEDIATED TRANSFECTION METHOD IN MAMMALIAN CELLS ...... 39 Introduction...... 39 Materials and Methods ...... 43 Plasmid isolation and maintenance...... 43 Cell culture and reagents...... 43 Vertically aligned nanofiber-mediated impalefection ...... 44 Optimization of impalefection substrate...... 45 Optimization of cell confluency...... 45 Optimization of DNA precipitation and concentration...... 46 Optimization of Cell Concentration and Cell Washing ...... 47 Impalefection Efficiency Measurements ...... 48 Nanofiber synthesis...... 50 Statistical Analysis...... 51 Results and Discussion ...... 52 Optimization of mechanical parameters ...... 52 Optimization of chemical parameters ...... 55 Optimization of biological parameters...... 57 Evaluation of impalefection efficiency following optimization ...... 63 Acknowledgements...... 66
CHAPTER III ...... 68 QUANTITATIVE ANALYSIS OF IMMOBILIZED DNA ON VERTICALLY ALIGNED CARBON NANOFIBER GENE DELIVERY ARRAYS ...... 68 Introduction...... 69 Materials and Methods ...... 71 Synthesis of Vertically Aligned Carbon Nanofiber Arrays ...... 71
vii Covalent Attachment of DNA to Carboxylated Microspheres and Vertically Aligned Carbon Nanofiber Substrate...... 72 Primer and Probe Design for PCR and Quantitative PCR...... 74 PCR Amplification of pd2EYFP-N1 covalently-bound to Carboxylate Microspheres and Vertically Aligned Carbon Nanofibers ...... 76 In Vitro Transcription of pd2EYPF-N1 covalently tethered to Vertically Aligned Carbon Nanofibers...... 77 Quantitative PCR of pd2EYFP-N1 on Vertically Aligned Carbon Nanofibers...... 78 Results and Discussion ...... 79 PCR Amplification on Vertically Aligned Carbon Nanofibers ...... 79 In Vitro Transcription on Vertically Aligned Carbon Nanofibers...... 82 Quantitative PCR Amplification on Vertically Aligned Carbon Nanofibers ...... 82 Acknowledgements...... 93
CHAPTER IV ...... 95 INDUCIBLE RNAI-MEDIATED GENE SILENCING USING VERTICALLY ALIGNED CARBON NANOFIBER GENE DELIVERY ARRAYS ...... 95 Introduction...... 95 Materials and Methods ...... 99 Plasmids, construction and maintenance ...... 99 Cell culture and reagents...... 102 Cell transfection and generation of stable cell lines ...... 103 Multilabel counting analysis and tetracycline induction ...... 104 Fluorescence microscopy and cell counting ...... 105 Nanofiber-mediated impalefection ...... 106 Nanofiber Array Synthesis...... 107 Statistical analysis...... 108 Results and Discussion ...... 108 CFP temporal silencing in CHO-K1 cell populations by tetracycline-inducible shRNAs...... 108 Silencing of CFP by tetracycline-inducible shRNAs in cell populations on VACNF arrays...... 114 Gene silencing by tetracycline-inducible shRNAs monitored in a single cell on indexed VACNF arrays...... 118 Acknowledgements...... 125
CONCLUSIONS...... 126
VITA...... 167
viii
LIST OF TABLES
Table 1. Different parameters of the impalefection method...... 42
Table 2. Effects of cell confluency, impalement substrate, and sodium butyrate on
impalefection efficiency. * denotes a statistically significant difference (p <
0.05)...... 58
Table 3. Effects of cell concentration and cell washing on impalefection efficiency. .... 60
Table 4. Primers and probes used in this study...... 75
Table 5. Quantification of RNA after in vitro transcription of pd2EYFP-N1 on vertically
aligned carbon nanofiber arrays...... 83
Table 6. DNA vectors and cell lines used in this study...... 109
ix LIST OF FIGURES
Figure 1. EDC Condensation of DNA on a carbon nanofiber ...... 18
Figure 2. Effects of different DNA precipitation methods on the impalefection efficiency
of carbon and silicon nanofibers...... 56
Figure 3. Effects of the concentration of adsorbed and immobilized DNA on
impalefection efficiency...... 58
Figure 4. Impalefection efficiency in different mammalian cell types...... 64
Figure 5. Comparison of efficiencies of the non-optimized and optimized impalefection
method...... 65
Figure 6. Amplification of pd2EYFP-N1 on vertically aligned carbon nanofiber arrays 80
Figure 7. Illustration of the methods used for quantification of accessible DNA bound to
the nanofiber arrays...... 85
Figure 8. Quantitative PCR of pd2EYFP-N1 on vertically aligned carbon nanofiber
arrays...... 86
Figure 9. Illustration of potential immobilization of DNA on vertically aligned carbon
nanofiber arrays ...... 92
Figure 10. Illustration of the vertically aligned carbon nanofiber single-cell gene
silencing platform ...... 100
Figure 11. Tetracycline induction of silencing in CHO-K1 cells over time...... 110
Figure 12. Addition and removal of tetracycline to CHO-K1-CFP-TetR-shRNA cells
over time ...... 115
x Figure 13. Silencing of vertically aligned carbon nanofiber impalefected cells after
tetracycline induction over time...... 119
Figure 14. Tetracycline induction and tracking of YFP expression and shRNA-mediated
silencing of CFP expression in single impalefected cells on indexed vertically
aligned carbon nanofiber arrays over time...... 121
xi CHAPTER I
LITERATURE REVIEW Background
What is Nanotechnology?
Science and technology have made substantial strides of progress in the last few decades.
The number of technological advances is increasing exponentially, and the applications of these new technologies to scientific knowledge are making a significant impact worldwide. Nanotechnology is one example of a technology that has made a major contribution on the scientific community. Although the interdisciplinary field of nanotechnology has been widely recognized only in the last decade or so, science at the nanoscale has been addressed for almost half a century. Most notable is Richard
Feynman’s famous and somewhat prophetic invitation given at the annual meeting of the
American Physical Society in 1959, “There’s Plenty of Room at the Bottom,” in which he predicted that future scientists would have the ability to manipulate single molecules and atoms and arrange them into desirable materials (Feynman 1959). The term
“nanotechnology” was first coined by Norio Taniguchi at the Tokyo Science University in 1974 as consisting of “…the processing of separation, consolidation, and deformation of materials by one atom or one molecule…“ (Taniguchi 1974). Nanotechnology is defined by the National Research Council in the National Nanotechnology Initiative as “.
. . research and technology development at the atomic, molecular, or macromolecular
1 levels, in the length scale of approximately 1-100 nanometer range, to provide a
fundamental understanding of phenomena and materials properties at the nanoscale and
to model, create, characterize, manipulate, and use structures, devices, and systems that
have novel properties and functions because of their small or intermediate size” (NRC
2002). One advantage of nanofabrication is that devices are typically structured from the
“bottom-up” approach rather than the traditional method of “top-down”. As implied, this
means that construction occurs by beginning with a blank platform and assembling one
molecule at a time, rather than having to shape a device out of an already existing structure. This ordered assembly as opposed to controlled deconstruction can often be much more energy efficient (Masciangioli and Zhang 2003). Although the term nanotechnology is generally understood by scientists, it proves to be somewhat arbitrary.
In a survey conducted by 3i in association with the Economist Intelligence Unit and the
Institute of Nanotechnology, experts in the field of nanotechnology (e.g. corporate researchers, academic researchers, investors, venture capitalists, analysts, consultants) were asked how they would define nanotechnology. Although most individuals had a
general idea of defined nanotechnology, not all could agree on where the nanoscale
domain begins. Approximately half of those interviewed agreed that nanotechnology was
limited to less than 100 nm, while 25 percent of those interviewed restricted
nanotechnology to the atomic and molecular level (Pitkethly 2003). However, a much wider definition is warranted, where functionality is imparted or augmented by components of a system that exist at the nanoscale (<100 nm).
2 We are now in the middle of a new scientific age, where nanotechnology is exploding in
its stages of development and application. The application of nanotechnology has
facilitated a union between many fields of research, including, but not limited to:
computer science, materials science, engineering, physics, microscopy, medicine,
microbiology, biochemistry, and molecular biology. There are also a number of products with nanotechnological aspects to them that have entered the consumer market (Liu et al.
2006; Nohynek et al. 2007; Tsuji et al. 2006). Due to their efficiency at absorbing light – particularly in the ultraviolet (UV) range – nanoparticles have been successfully marketed in Australian sunscreen lotions (Wilkinson 2003). In fact, over 60% of the
Australian sunscreen industry has adapted to the addition of nanoparticles, and this same trend is expected to soon be seen in Europe and the United States as well (CSIRO 2004).
Other consumer-related products include nano-thin magnetic material layers in computer hard drives that result in increased storage capacity, protective paints and coatings that
resist corrosion and radiation, stain-free clothing, bonding agents for dentistry, and car
bumpers that are stronger - while at the same time - more flexible (NNI 2004). While the application of nanotechnology in consumer products has resulted in incremental advances, the truly dramatic impact of nanotechnology’s possibility lies in its more
significant manipulation of material at the molecular scale.
3 Nanotoxicology: An Emerging Field
With any new developing technology comes new concerns and cautions to shed light on
the environmental and toxicological impacts of the technology and to guarantee that it is
used responsibly. Although restrictions are often placed on research by scientists
themselves, sometimes more strict restraints are needed. This was seen in the 1970's, when legal and moral restraints were placed on biotechnology and genetic research by the
government in an effort to address potential and perceived harms of these fields on
society in general (OTA. Office of Technology Assessment 1984). It was a concern at
the time that cloning and genetic modification could do more harm than good in the long
run, and government intervention was used to keep the scientific community accountable.
While nanotechnology continues to gain more interest in the public eye, it is important to understand the impact it will have on society, and create more defined boundaries and regulations if needed. This has resulted in the emergence of a new field termed nanotoxicology.
It is worth mentioning that the toxicological effects of nanoparticles (which is a more recent term; also referred to as ultrafine particles or particulate matter) in emissions, from coal mines, and simply in the atmosphere, have been long studied (Davis et al.
1982; McClellan et al. 1982; Oberdorster et al. 1994). However, clarification needs to be
made to keep from grouping newly engineered nanoparticles into the same category as
nanoscaled particulate matter. Therefore, nanotoxicology in this section will address the
toxicological effects of engineered nanostructures, due to its specific relevance to this
4 work. Thorough reviews by Borm and Oberdorster discuss the links between the
historical foundation and progress of particle research with the rapidly growing field of
nanotechnology (Borm 2002; Oberdorster et al. 2005).
The biggest hurdle with discussing the potential risks of nanotechnology is the fact that
this area of technology is extremely novel and the impacts remain to be understood.
However, based on sparse preliminary experiments that suggest some nanostructures can
be harmful to biological systems, there has been an enormous call for research in this
area. In light of the potential market and practical applications of nanotechnology, it is
generally recognized that these environmental and health impacts must be addressed.
The economy of nanotechnology is projected to be valued at $1 trillion by 2012
(Hardman 2006), and the U.S. Congress has encouraged billions of dollars of funding on
research devoted to the medical and environmental impacts of the new technology (Giles
2003). There are many aspects to the possible risks that nanotechnology introduces, due
to the fact that nanotechnology is potentially useful to so many different disciplines.
There are medical concerns such as whether nanoparticles should be used for applications
in humans, and there are environmental issues such as the problem of releasing copious
amounts of nanostructures into nature. As Colvin emphasized in a recent review, it is
important to realize that any substance can be toxic at a given exposure level (Colvin
2003). However, with the growing demand for nanofabricated structures comes a
growing demand for protecting workers in the field of nanotechnology. For example,
Mitsubishi opened a fullerene-producing plant in Japan in May 2003, which produces
many tons of fullerene particles every year (Colvin 2003). These plant workers will all
5 potentially be exposed to nanomaterials, as will the consumers that buy products
containing an ever-increasing amount of nanomaterials. Furthermore, these
nanostructures will eventually find their way into the groundwater and soil, since the
increase of artificial substances in the environment is directly proportional to the use of
such substances by society (Colvin 2003).
The body of experimental evidence showing the cytotoxic effects of nanomaterials is
growing. Johnston et al. showed that Teflon (PTFE) nanoparticles result in pulmonary
toxicity when laboratory animals are exposed intratracheally (Johnston et al. 2000).
Interestingly, Teflon nanoparticles approximately 16 nanometers in diameter were toxic,
while nanoparticles >100 nanometers lost their toxic effect. This suggests the unique
toxicology that nanoparticles introduce. Early studies also demonstrated that carbon
nanotubes (2-50 nm) and nanoparticles (20-30 nm) can result in acute inflammatory
pulmonary effects including granuloma formation in rats (Warheit et al. 2004) and mice
(Lam et al. 2004), and can ultimately translocate to the liver (Oberdorster et al. 2002).
Similar toxicological responses have been seen in cultured cells using titanium dioxide
nanoparticles (Wang et al. 2007), C60 (Sayes et al. 2005), single wall carbon nanotubes
(SWNTs) (Shvedova et al. 2004) and multi-wall carbon nanotubes (MWNTs) (Monteiro-
Riviere et al. 2005). Oberdorster et al. demonstrated that carbon nanoparticles travel to
the brain upon inhalation in laboratory rats (Oberdorster et al. 2004). However, this study did not imply that the nanoparticles were toxic upon translocation to the brain. A different study, exposing largemouth bass to uncoated C60, suggested that these nanoparticles can not only travel to the brain upon inhalation, but can subsequently result
6 in oxidative stress and damage to the brain (Oberdorster 2004). A more recent in vitro
investigation showed that fullerene (C60 carbon nanoparticles) derivatives bind to beta-
lactoglobulin and are transferred to human serum albumin, suggesting a theoretical
pathway for how carbon nanomaterials interact on the molecular level and are transported
in biological systems (Belgorodsky et al. 2007). However, many early studies that
showed cytotoxic effects were using large doses of carbon nanostructures and could have
resulted in death not due to the nanostructure, but due to obstruction of the lungs
(Oberdorster et al. 2005). In addition, Henry et al. have recently demonstrated that C60 particles solubilized in tetrahydrofuran (THF), a common vehicle solvent used in nanotoxicology studies, may have cytotoxic effects in larval zebrafish not due to the carbon nanomaterials themselves, but strictly due to a THF degradation product, γ- butyrolactone (Henry et al. 2007). The aggregation state and metal impurities of preparations have also been proposed as contributions to the observed cytotoxicity of carbon nanostructures (Oberdorster et al. 2005). Very few studies have investigated the ecotoxicological effects of nanostructures. Zhu et al. demonstrated that C60-exposed
Daphnia showed an LC50 of >35 ppm, and fathead minnow showed no physical effects at 48 hours (Zhu et al. 2006). However, lipid peroxidation significantly increased in fathead minnows as a result of C60 exposure.
More conclusive data is needed to fully understand the toxic effects of nanostructured
materials. Most carbon nanostructures – single-walled nanotubes, fullerenes, and carbon
nanoparticles – and quantum dots are in the process of risk assessment. While substances
such as dendrimers and silica nanoparticles are considered relatively harmless due to their
7 biocompatible and biodegradable nature, there are many other engineered nanoparticles of differing compositions that need to be studied including iron and bimetallic nanoparticles, which are potential candidates for future in situ remediation. Likewise, no toxicological studies have been pursued for carbon nanofibers or vertically aligned carbon nanofibers (VACNFs). Nanotoxicology will continue to be a growing field, given the high demand for knowledge and literature on the toxicity and long-term exposure effects of nanostructured materials on medical fields and the environment.
The Promise of Nanotechnology in the Biological Realm
Currently, one of the primary scopes of nanotechnology is the arena of biological sciences. The biological applications of nanostructures have been as broad as the shape of the structures themselves. To highlight just a few applications, we will look at the application of nanomaterials to targeted drug delivery, molecular imaging, and gene delivery.
As stable, spectrally discrete, and chemically functional elements, nanoparticles provide unique characteristics for interfacing with biological systems at the molecular scale. The ability to generate particles that are spectrally discrete enables probing of the nanoparticles upon delivery, and through functionalization, they can be employed for material delivery applications. An elegant approach has been demonstrated by Salem et al. where a bimodal nanoparticle (nickel-gold) provided both material retention and nuclear targeting, enabling directed transport of genetic material to the nuclear domain
8 (Salem et al. 2003). In addition to subcellular scale targeting, such bimodality can also be used for targeting specific organs of the patient, thus carrying the drug to the area of treatment while also potentially masquerading the drug from degradation pathways which
can result in pharmaceutical break down and the commensurate requirement for larger
dosing (Shaffer 2005). Previously there has been limited success in efficiently delivering
drugs to the brain, due to the blood-brain barrier. However, nanoparticles have shown
promise to researchers in this area, due to the ability to coat the nanoparticle and
“disguise” the drugs being delivered. Lockman et al. injected 100 nm thiamine-coated
nanoparticles comprised of emulsifying wax and polyoxyl 20-stearyl ether into rats,
showing that the use of the thiamine ligand on the surface of the nanoparticles helped
facilitate binding with the blood-brain barrier transporters, thus successfully delivering
the nanoparticles to the brain (Lockman et al. 2003). The ultimate goal is to use these
nanoparticles as a tool for localized drug delivery to target and treat tumors in the brain.
Quantum dots continue to be the most widely applied nanoparticles in the biological
field. Quantum dots are engineered from heavy metals, such as cadmium selenide
(CdSe), and are often coated with a polymer that provides biocompatibility or molecular
targeting for use in cell labeling, cell-tracking, and in vivo imaging (Alivisatos 2004).
The core size of quantum dots is typically 5-6 nanometers. Once the CdSe core is layered with a polymer coating and then conjugated with specific molecules for biological recognition (proteins, DNA, antibodies), the typical size of the quantum dot while interacting with the aqueous solution is close to 25 nanometers (Zipfel 2004). The fluorescent capabilities of these semiconductor nanocrystals far outweigh any
9 conventional molecular fluorophores or fluorescent dyes (Alivisatos 2004). The emission
spectra of quantum dots can be finely tuned, providing spectrally discrete emission from individual nanoparticles and the ability to track multiple molecular events within biological systems simultaneously using a single source of excitation. Furthermore, unlike many traditional fluorophores, quantum dots are not easily photobleached and
provide long-term stability. They have recently been linked to antibodies and used for in
vitro detection of enteric pathogens such as Cryptosporidium parvum and Giardia lamblia (Zhu et al. 2004a), as well as in vivo targeting and detection of tumors in live
animals (Chan and Nie 1998). Additionally, Wiesner et al. are working on developing
silica particles that can be used for binding specific molecules, and then tracking their respective pathways in humans (Wiesner 2004). Due to their radiative properties, they have fluorescence levels similar to quantum dots. Additionally, the biocompatibility of silica particles gives them an advantage over quantum dots for human applications.
Likewise, they are easily and inexpensively made using well established synthesis techniques, including sol gel-based methodologies (Ow et al. 2004). However, typical silica nanoparticles are hundreds of nanometers in diameter, rendering them incapable for efficient molecular delivery to specific antibodies or protein receptors. Wiesner et al have
been able to construct silica nanoparticles down to the size of 30 nanometers, that are 20
times brighter and more photostable than traditional fluorescent dyes and molecular
fluorophores (Ow et al. 2005). Because 30 nanometers is still too large for silica
nanoparticles to be linked with specific antibodies and proteins (Garman et al. 2000),
progress is still underway to develop even smaller silica particles.
10 DNA delivery to mammalian cells has become a staple of biological studies in the last
few decades, and a crucial means of determining gene function and gene pathways. A
number of nanostructured platforms have been shown to effectively deliver plasmid DNA into multiple mammalian cell types. While a number of organic and biodegradable particles have been used for DNA complexation and delivery (Batard et al. 2001;
Chowdhury and Akaike 2005b; Hedley and Barman 2004; Nie and Wang 2007), inorganic nanostructures can offer additional stability to the DNA during nuclear transport (Chowdhury and Akaike 2005a). These include nanoparticles, nanoscaffolds, nanorods, nanofibers and nanotubes. Silica nanoparticles have been synthesized and complexed with DNA for efficient gene delivery. The surface of the silica nanoparticle can be modified with poly(L-lysine) (Zhu et al. 2004b) or aminoalkylsilanes (Kneuer et al. 2000) to electrostatically interact with the DNA and protect it from enzymatic degradation, and allowing controlled release of the DNA from the silica nanoparticle within the cell (Roy et al. 2005). Uchimura et al. used gold colloid nanoparticles as a negatively charged nanoscaffold for increasing the uptake of DNA in human mesenchymal stem cells (Uchimura et al. 2007). The 20 nanometer gold nanoparticles increased the transfection efficiency to levels 2.5-fold higher than the same technique in the absence of gold nanoparticles. Bifunctional nanorods were first introduced by Salem et al. (Salem et al. 2003). These particles were 200 nm in length, composed of discrete nickel and gold regions. DNA was linked to the nickel and thiol groups containing transferrin were linked to the gold. This resulted in increased efficiency of DNA uptake in HEK293 cells, and has potential for targeted DNA delivery in the future.
11 Recent demonstrations have shown that vertically aligned carbon nanofiber arrays
(VACNFs) grown on a silicon substrate can be used as a gene delivery platform in mammalian cells (McKnight et al. 2003a; McKnight et al. 2004b). This nanofiber- mediated gene delivery method, referred to as impalefection, will be discussed in more detail below. Cai et al. have demonstrated an approach similar to impalefection, termed nanospearing, in which carbon nanotubes impale mammalian cells due to magnetic agitation (Cai et al. 2005b). Nanospearing has proven to be highly efficient, resulting in
80%-100% efficiency of gene delivery and expression. Likewise, Cai et al. have recently shown that carbon nanotubes are biocompatible with cultured cells and can transfer genes into primary cell lines that are traditionally hard to transfect (Cai et al. 2007).
Additional nanomaterials that have incorporated into biology include biocompatible and robust nanolayers as coatings for orthopedic implants (Chun et al. 2004; Price et al. 2003;
Webster et al. 2004), silver nanomaterials for antibacterial wound dressings (Kim et al.
2007), iron oxide nanoparticles for MRI contrast agents (Pitkethly 2003), nanostructured surfaces for neuronal outgrowth and patterning (Dowell-Mesfin et al. 2004), and nanoscale aerosols for lung inhalation therapies (Courrier et al. 2002).
Vertically Aligned Carbon Nanofibers as a Tool of Nanobiotechnology
Carbon nanofibers have become an invaluable tool for nanobiotechnology in the last decade. Although carbon nanofibers have long been the product of interacting carbonaceous gas and metal catalysts, the deterministic synthesis of carbon nanofibers
12 has existed for less than 20 years. As Melechko pointed out in his excellent review, deterministic synthesis refers to the fact that the diameter, position, alignment, length and chemical composition of the nanofiber can be controlled (Melechko et al. 2005). Chen et al. were the first group to use catalytic plasma enhanced chemical vapor deposition (C-
PECVD) for the synthesis of carbon nanofibers (Chen et al. 1997). However, growth of nanofibers can be achieved using additional methods of chemical vapor deposition, such as catalytic thermal chemical vapor deposition (C-TCVD).
Carbon nanofibers (CNFs) are defined as conical structures that typically have a high aspect ratio, meaning that the diameter of the structure can be between tens and hundreds of nanometers while the length of the structure can range from hundreds of nanometers to a few millimeters (Meyyappan et al. 2003). The internal structure of the CNF consists of graphene sheets containing a hexagonal repeating pattern of carbon atoms covalently bound together (Melechko et al. 2005). The graphene sheets are arranged in a stacked- cone composition with exposed edges of unsaturated bonds. Nolan et al. note that hydrogen can be added during the growth process of CNFs to fulfill the valences at the graphene edges (Nolan et al. 1998). This composition differs widely from the more popular carbon nanotubes (CNTs). CNTs consist of graphene sheets rolled up into concentric cylinders with no exposed edges, which gives CNTs much less reactivity than
CNFs (Meyyappan et al. 2003).
Vertically aligned carbon nanofibers (VACNFs) refer to CNFs that have been grown perpendicular to the substrate surface. This can be achieved during PECVD, due to the
13 placement of the catalyst particle and the interaction with the electric field (Merkulov et
al. 2001). The growth and conditions of vertically aligned carbon nanofibers can vary
from process to process (Cojocaru et al. 2006), but the most common method of VACNF
growth is by PECVD using a catalyst on a substrate. In this growth process, the catalyst
can be patterned (layered, lined or spotted) on the substrate, and the substrate itself can be
silicon (Merkulov et al. 2000), quartz (Chen and Dai 2001) or glass (Ren et al. 1998).
The catalyst can be a variety of metals including, but not limited to, nickel, iron, cobalt
and a number of alloys (Melechko et al. 2005). Nickel (Ni) is a commonly used catalyst
particle and can be patterned on the silicon (Si) wafer by photo- or electron beam
lithography. The size of the Ni dots can be determined, and at 100 nm in diameter and 40 nm thick, the result is a single carbon nanofiber from each catalyst dot on the substrate
(Melechko et al. 2005). The spacing between each fiber can also be controlled, ranging from 2.5 to 20 µm pitch (pitch is defined as the distance between each nanofiber). A carbon source is required for the nanofiber formation, typically in the form of a carbonaceous gas such as acetylene, and growth takes place at temperatures ranging from
400 ºC to 1000 ºC. More recently, nanofiber growth has been reported at temperatures as low as room temperature, using methane gas as a carbon source and resulting in
“spaghetti-like” structures (Boskovic et al. 2002). In these studies, the nanofiber growth took place at standard temperatures of 600-650 ºC. The samples were pretreated with
NH3 plasma to form a Ni catalyst nanoparticle and acetylene (C2H2) was introduced as
the carbon source, instantly resulting in carbon nanofiber growth. The length of the
nanofibers is controlled by the length of plasma generation.
14 A number of VACNF applications can be found in the literature, including their use as an x-ray source (Matsumoto and Mimura 2003), a field-emission electron source (Baylor et al. 2004; Guillorn et al. 2001a; Guillorn et al. 2001b; Guillorn et al. 2004; Pirio et al.
2002; Teo et al. 2003), nanoporous membranes (Fletcher et al. 2004; Hinds et al. 2004;
Zhang et al. 2002), electrochemical probes (Fletcher et al. 2006a; McKnight et al. 2004a), atomic force microscopy tips (Ye et al. 2004) and charge storage devices (Chen et al.
2001), site-specific biochemical functionalization (Fletcher et al. 2006b; McKnight et al.
2006c), neuroelectrochemical probes in cells and tissue (McKnight et al. 2006a; Yu et al.
2007a; Yu et al. 2007b) and intracellular gene delivery in eukaryotic cell types
(McKnight et al. 2003b; McKnight et al. 2006b; McKnight et al. 2004b). This review will look in further detail at the biologically relevant applications of VACNF arrays, including biochemical functionalization and intracellular gene delivery.
Methods and Chemistry of Covalent DNA Immobilization
The chemistry of functionalization and immobilization of DNA has been used for a number of applications including nucleic acid hybridization assays (Nikiforov and Rogers
1995; Proudnikov et al. 1998; Wang et al. 1997), affinity chromatography (Bunemann
1982; Macdougall et al. 1980; Mykoniatis 1985; Poonian et al. 1971), gene delivery (Cai et al. 2005a; McKnight et al. 2003b; McKnight et al. 2004b) and microscopic
visualization (Allison et al. 1992; Morozov et al. 1996). While some of these
applications utilize noncovalent immobilization (Allison et al. 1992; Hirayama et al.
1996; Shchepinov et al. 1994), covalent bonding between DNA and the substrate is the
15 most common immobilization method used. Immobilization strategies resulting in
covalent attachment of nucleic acids include biotin-streptavidin (Crucifix et al. 2004;
Larsson et al. 2003; Su 2002), thiol-gold interaction (Ajore et al. 2007; Cho et al. 2004),
UV-radiation (Kalachikov et al. 1992; Yamada et al. 2001; Yamada et al. 1999), cyanogen bromide activation (Poonian et al. 1971), disulfide-mercaptosilane (Rogers et al. 1999) and 1-ethyl-3-(3-dimethylaminopropyl) carbodiimide (EDC) condensation (Ge et al. 2003; Mykoniatis 1985; Rasmussen et al. 1991). Among other methods, EDC condensation has an advantage of forming an amide linkage without the addition of a spacer molecule (Grabarek and Gergely 1990). This EDC-catalyzed approach has been effectively used for DNA immobilization on a variety of substrates in the past, including membranes (Okutucu and Telefoncu 2004), polymers (Taira and Yokoyama 2004; Taira and Yokoyama 2005), microwells (Rasmussen et al. 1991), glass slides (Zammatteo et al.
2000), glass beads (Walsh et al. 2001), diamond films (Christiaens et al. 2006), gold electrodes (Ge et al. 2003), carbon electrodes (Millan and Mikkelson 1993), carbon nanotubes (Dwyer et al. 2002; Nguyen et al. 2002) and carbon nanofibers (McKnight et al. 2003a). Due to the exposed graphene sheet edges at the surface, the functionality of
VACNF arrays is highly advantageous. Carbon nanofibers contain carboxyl groups
(COOH) that can be used for the covalent attachment of primary amine groups (NH2) using EDC condensation. For gene delivery in mammalian cells, plasmid DNA was tethered to VACNF arrays using the EDC-mediated condensation reaction between DNA base amines and carboxylic acid sites on nanofibers (McKnight et al. 2003a; McKnight et al. 2004b), and has also been used for immobilization of proteins (e.g., horseradish peroxidase) on VACNF arrays. The carboxyl groups are nonexistent on the silicon
16 substrate of the VACNF array, and are often limited in their location on the nanofibers as well. Sputtering of the silicon substrate onto the base of the nanofiber structure can eliminate accessible carboxyl groups in this region; likewise, the presence of a catalyst
(e.g. nickel) at the tip of the nanofiber restricts carboxyl group availability. However, the major portion of the nanofiber contains carboxyl groups that can be readily accessed by the primary amines present in three (adenine, guanine, cytosine) DNA bases. This covalent interaction is catalyzed by the presence of EDC.
In the EDC condensation, the EDC molecule reacts with a proton to form a carbocation.
The carbocation subsequently reacts with an ionized carboxyl group on the carbon nanofiber to form the highly reactive O-acylisourea derivative. O-acylisourea becomes reprotonated at the site of Schiff base and reacts again with an ionized carboxyl group to produce carboxylic anhydride which will rapidly form an amide bond in the presence of exposed primary amine groups in the DNA sequence. The ultimate product is a covalent linkage via a peptide bond between the DNA and the carbon nanofiber (Figure 1).
While the EDC-catalyzed reaction is easy applicable and straightforward for DNA binding strategies, a number of factors can affect the efficiency of the covalent immobilization and the functionality of the DNA after immobilization. N-acylurea is a product of the reaction between carboxylic acids and carbodiimides if the primary amines are not present or excess carbodiimide is present. In the presence of excess carbodiimide, the formation of N- acylurea from the O-acylisourea derivative can inhibit the production of amide linkages even upon addition of amines to the reaction (Nakajima and Ikada 1995).
17
Figure 1. EDC Condensation of DNA on a carbon nanofiber. In this reaction, the
EDC molecule reacts with a proton to form a carbocation. The carbocation subsequently reacts with an ionized carboxyl group on the carbon nanofiber to form a highly reactive ester intermediate that is an O-acylisourea derivative. The active ester intermediate becomes reprotonated at the site of Schiff base and reacts again with an ionized carboxyl group to produce carboxylic anhydride which will rapidly form an amide linkage in the presence of exposed primary amine groups in the DNA sequence. The ultimate product is a covalent linkage via a peptide bond between the DNA and the carbon nanofiber.
18 Therefore, it is important to refrain from using excessive amounts of EDC in the reaction.
The EDC reactivity with carboxyl groups is highly dependent on pH, with an optimum pH between 3.5 and 4.5 for short incubation times (Nakajima and Ikada 1995). EDC is highly stable at neutral to higher pH levels (pH 4 – pH 8) for up to 5 hours at 25 ºC.
However, very little carbodiimide activity remained after 2 hours at pH levels below 4
(Nakajima and Ikada 1995). Additionally, the presence of free amines, sulfhydryls or carboxyl groups in the buffer can compete with bond formation, rendering the DNA immobilization incapacitated (Hermanson et al. 1992). A well-characterized buffer like
2-(N-morpholino)ethanesulfonic acid (MES) buffer with pH 4.5 is often used
(Hermanson et al. 1992). For biological applications such as gene delivery of immobilized DNA, it is also important to limit the incubation time in acidic buffers, due to potential degradation of DNA. Millan demonstrated that when using EDC for DNA immobilization to carbon, the amide bond is formed specifically at exposed guanine and cytosine amines (Millan et al. 1992). Since DNA bound to carboxyl groups at these binding sites cannot be effectively directed, binding may occur at undesirable locations
within transcriptional regions of the DNA sequence and render the tethered DNA
template transcriptionally-inactive. Binding within the coding sequence may also
interfere with reading and complementation of the base during mRNA elongation.
Schaffer et al. found that the disassociation of noncovalent interactions between
polycations and DNA was required for efficient expression of the DNA by transcriptional
machinery (Schaffer et al. 2000). Thus, the transcriptional activity of template can be
significantly reduced, if not completely hindered by the interaction of the template with
large, covalently-bound molecules. Therefore, tethered gene strategies, where solid
19 scaffolding can present even larger hindrances, may present an additional variable with
respect to evaluation of efficiency (i.e. transgene expression per unit DNA).
Electrochemistry measurements have suggested that carbon nanofibers have a rough,
uneven surface. Although a non-geometric surface can create complications for
estimations of total surface area, electrochemistry measurements demonstrate that on
average the surface area of a carbon nanofiber mirrors its vertical height (e.g., a nanofiber
with a vertical height of 15 µm has a surface area of 15 µm2). Moreover, the percentage
of total surface area of carbon nanofibers on a VACNF array chip is dependant on the
pitch of the carbon nanofibers. For instance, a VACNF chip containing 15 µm
nanofibers at a 5 µm pitch has a total surface area per unit fiber of 40 µm2, of which the carbon nanofibers make up 15 µm2, or 37.5% of the total surface area. The respective
surface area of the nanofiber decreases to 13% if the pitch is increased to 10 µm between
each nanofiber. These percentages of the total surface area are a rough estimate, due to
fluctuations in actual surface area of the carbon nanofiber and fluctuations in nanofiber
height during growth in the PE-CVD process. In addition, predictions of DNA
adsorption on the surface of a VACNF chip become increasingly complicated, due to
differences in surface chemistry and potential preferential adsorption to either the carbon
nanofiber or silicon substrate. For example, if 100 ng of DNA was allowed to dry on the
VACNF chip described above, while the carbon nanofibers compose 37.5% of the total
surface area, more or less than 37.5% of the DNA may be initially adsorbed onto the
carbon nanofiber surface. Likewise, only a portion of the adsorbed DNA may be
transcriptionally active after the drying process. Therefore, precise theoretical
20 calculations of the percentage of DNA that is initially adsorbed to a VACNF chip and remains transcriptionally available to the cell are difficult to estimate without direct experimentation. It would be helpful for future applications of gene delivery to know how much DNA remains transcriptionally active after EDC-condensation immobilization on the VACNF arrays.
Traditional Methods of Gene Delivery in Mammalian Cells
Transfection is the process of introducing foreign DNA into eukaryotic cells and is often simply referred to as gene delivery in mammalian systems. Many different methods of transfection have become staples in the field of biological studies in the last few decades.
While some cell types exhibit the natural uptake of foreign DNA under certain circumstances (Kidwell 1993; Lehmann and Sczakiel 2005; Nevoigt et al. 2000), methods of transfection typically include some means of shielding the foreign DNA from cellular recognition and degradation, or evading the degradation pathways altogether.
Mammalian transfection can result in transient or stable expression. Transient expression refers to short-term gene expression and protein production of the introduced foreign
DNA that typically peaks between 12 and 72 hours (Colosimo et al. 2000), but on some occasions can last up to 10 days post-transfection (Baldi et al. 2007). Genes that are transiently expressed are not incorporated into the genome of the host cell, and expression occurs in the absence of a selection agent (i.e. blasticidin-S). This results in the loss of the introduced gene upon cell division. Transient expression assays are helpful for some applications, including recombinant protein production in large-scale
21 batch cultures for pharmaceuticals (Geisse and Henke 2005; Meissner et al. 2001; Shi et al. 2005). However, for the majority of applications, it is advantageous to have a stably- inserted copy of the gene internal to the host chromosomal DNA. This requires stable transfection of cells. Stable transfection occurs when the foreign gene is stably inserted into the host cell chromosome via recombination. Using microinjection in CHO-K1 cells, as high as 10% of the cells have been observed to insert foreign DNA by nonhomologous recombination when multiple copies are introduced into the nucleus
(Proctor 1992a). In order to resolve which cells have inserted the foreign gene into the chromosome during transfection, the gene of interest must be coupled with a selection agent. The selection agent typically imparts antibiotic resistance to the cell, allowing the cells possessing the gene of interest to survive after the addition of antibiotics to the cell culture medium. While stable transfection of mammalian cells gives the ability to perform long-term investigations on cell lines that are all genotypically identical, this method has the disadvantage of being costly and time consuming; it could easily take 2-3 weeks for the generation of a stable mammalian cell line, and often requires a number of months (Colosimo et al. 2000).
Mammalian transfection can be organized into three categories: chemical, electrical and physical. Chemical methods of transfection include, but are not limited to, calcium phosphate mediated (Graham and van der Eb 1973), DEAE-dextran mediated (Pagano and Vaheri 1965), peptide mediated (Wagner et al. 1992), dendrimer-mediated
(Kukowska-Latallo et al. 1996), liposome based (Felgner et al. 1987), and polyethyleneimene based techniques (Ferrari et al. 1997). Electrical means of
22 transfection are limited to electroporation (Andreason and Evans 1988) and sonic-
poration using ultrasound energy (Unger et al. 1997). However, chemical and electrical
methods will not be addressed in this review. The physical transfection methods of
microinjection and particle bombardment are the most similar methods to impalefection
and will be discussed in further detail.
Microinjection is one of the most versatile means of gene delivery to living cells.
Microinjection is a variation of conventional needle injection but differs in that it uses
ultrafine micropipets for the delivery of nucleic acids to a specific intracellular location.
Microinjection has been used for DNA delivery to various embryos (Guille 1999; Xu
1999), oocytes (Jaffe and Terasaki 2004), eggs (Cheers and Ettensohn 2004), bacteria
(Knoblauch et al. 1999), cultured cells (Graessmann and Graessmann 1983), and specific organelles (Kagawa et al. 2001). In fact, nearly any mammalian cell type can be successfully transfected using this method (Proctor 1992a). In microinjection, efficiency levels (the number of transgenic colonies generated per transfected cell) and expression
levels of the DNA are extremely high (Proctor 1992a). When microinjection in the
nucleus was compared with injection into the cytoplasm, the levels of gene expression in
the nucleus were significantly higher (Capecchi 1980). Another advantage is that the number of gene copies can be discretely controlled and delivered to specific individual cells (Mueller et al. 1980). The disadvantage with this method is its laborious nature; only one cell can be injected at a time and multiple injections are often needed before a successful transfection takes place (Luo and Saltzman 2000).
23 Biolistic bombardment (also termed ballistic bombardment or particle bombardment) is another physical transfection method similar to impalefection. While biolistic processes of transformation are relatively new, they have been commonly used for plant cells
(Sanford et al. 1987), bacteria (Shark et al. 1991), fungi (Armaleo et al. 1990), animal tissues (Yoshida et al. 1997), and mammalian cell cultures (Heiser 1994; Yang et al.
1990; Zelenin et al. 1989). Biolistic bombardment consists of delivering DNA internally to the target cell by coupling it with biocompatible accelerated particles such as gold or tungsten (Heiser 2004). Particles are accelerated by pressurized gas, and calibration of the gas pressure can determine the optimal level for individual cell types. Advantages of using biolistic bombardment include the bypass of specific membrane requirements for
DNA uptake, the low level of DNA required (submicrogram quantities), the delivery to multiple cells simultaneously, and the in vivo application of transfection
(Muangmoonchai et al. 2002; Tanigawa et al. 2002). The disadvantages of using biolistic bombardment include the high level of manipulation involved and the expensive cost of devices (gene gun) and reagents (gold particles) (Heiser 2004).
Impalefection: Vertically Aligned Carbon Nanofiber-Mediated Gene Delivery
The process of intracellular gene delivery to eukaryotic cells using VACNF arrays is referred to as impalefection (McKnight et al. 2004b). The term impalefection refers to the mammalian cells being transfected with DNA due to physical impalement with nanofibers. The success of impalefection can primarily be attributed to the physical and chemical properties of the VACNF. The nanoscale tip of the fiber (10-100 nm)
24 contributes to the impalement of the cell membrane and/or cell wall, and can be
compared to the use of ultrafine micropipets that have been used in microinjection for
DNA delivery since the early 1970s (Jaenisch and Mintz 1974; Lin 1971). Another attractive feature of the VACNF is its robust flexibility. While the tensile strength of carbon nanofibers is 100 times less than that of single-walled carbon nanotubes (Callister
2003; Hughes 1987), nanofibers have shown high levels of resilience during the impalefection process. It has been demonstrated that VACNFs are strong enough to penetrate the rigid structure of plant and yeast cell walls (McKnight et al. 2006b).
Further, the nanoscale diameter of VACNFs allows the plasma membrane of mammalian cells or the cell wall of plant cells to recover following nanofiber penetration, enabling the recovery and proliferation of the interfaced cell. This recovery includes the resealing of the membrane, which has been observed using propidium iodide. Immediately after impalefection, a large portion of mammalian cells were stained with propidium iodide, while mammalian cells in the presence of propidium iodide five minutes after impalefection were not stained (data not shown). This demonstrates that the membrane is impaled during impalefection, and successfully and rapidly reseals, either around the nanofiber or in the absence of the nanofiber.
VACNFs are also richly populated with chemical handles for post-synthesis modification of the nanofiber surface. Functional groups on the surface of the nanofiber can be adsorbed or covalently attached with a variety of macromolecules (i.e. plasmid DNA and soybean peroxidase) using simple chemistry techniques including 1-ethyl-3-(3- dimethylaminopropylcarbodiimide) (EDC) condensation or biotin-streptavidin binding
25 strategies (Fletcher et al. 2006b; McKnight et al. 2003b; McKnight et al. 2006c).
Following this immobilization process, at least some of the molecular species retain their
transcriptional and/or enzymatic activity. However, the percentage and amount of active
species remains unknown.
Impalefection was first demonstrated by McKnight et al. (McKnight et al. 2003b). In this
seminal work, VACNFs were adsorbed or covalently immobilized with plasmid DNA
and subsequently used as a platform for parallel DNA delivery and expression monitoring of the green fluorescent protein (gfp) gene in Chinese Hamster Ovary (CHO-K1) cells.
The protocol for this procedure is very straightforward: DNA is coated on the surface of the nanofibers and introduced to the cells by pressing the nanofibers into a settled cell suspension fluid from freshly passed CHO-K1 cultured cells. Once impaled on the nanofiber platform, the cells are then placed in fresh medium and allowed to recovery at
37 ºC and 5% CO2. More recent studies have determined that plasmid DNA encoding for
the gfp gene attached to the nanofiber can be stably inserted into the genome, resulting in
a colony of GFP-expressing cells. However, some impalefected cells receive and express
the DNA on the nanofiber, but do not pass it on to their progeny (McKnight et al. 2003b;
McKnight et al. 2004b). This could be a result of the DNA being immobilized on the carbon nanofiber and residing in the nucleus of the cell where it can be transcribed without difficulty, but not passed on to the daughter cells. This is more often observed after EDC-mediated immobilization of DNA on the nanofibers. Spatially indexed
VACNF arrays can also be used for convenient location and tracking of individually expressing cells without constant analysis under a microscope (McKnight et al. 2004b).
26
While the time allowed for recovery after impalefection is typically 18-24 hours,
transgene expression has been observed as early as 75 minutes after impalement (data not
shown). In other methods of transfection, a minimum of 24 hours is usually required for
recovery and expression. This significantly rapid lag time for gene expression could be ascribed to the direct penetration and introduction of DNA to the nucleus, thus being limited strictly by the amount of time required for transcription of the gene and translation of the resulting mRNA transcript into protein. Direct nuclear penetration has been implied in the impalefection process by a number of observations. Along with the
rapid gene expression after impalement, nuclear impalement has been experimentally
observed using a nuclear isolation method described by Butler (Butler 1984). Using this method, CHO-K1 cells were impalefected on a VACNF array platform, and all cellular components excluding the nuclei were lysed. Samples were then fixed and viewed under
SEM. Nuclei were frequently observed with nanofibers inserted in them, suggesting that
not only are impaled cells often recovering and remaining on the nanofibers themselves,
but nuclei are also directly impaled during the process (data not shown). Direct cellular
impalement has also been observed using confocal microscopy. Frequent impalement of
the nucleus in CHO-K1 cells may be due to the large volume of the nucleus inside this
cell type. Rounded cells have a diameter of 10-15 µm, while the nucleus of each cell can be from 5-10 µm inside the cell. This makes the nucleus a large target during impalefection, and highly likely to be penetrated. While this is the case, not all impalefected cells remain penetrated on the nanofiber. Their ability to express DNA is
27 probably similar to the uptake and expression of DNA after the “scrape-loading”
transfection method (Fechheimer et al. 1987; McNeil et al. 1984).
The efficiency of transfection and frequency of transgene expression following
impalefection have not been investigated. While the efficiency does not likely match the
high levels of efficiency of other transfection methods, there are a number of advantages
to the VACNF-mediated gene delivery method. First, expression of the delivered DNA
can take place very rapidly, due to the penetrant nanofiber and delivered DNA residing
directly in the nucleus. This bypasses the need for internalization of the DNA, evasion of
potential cytosolic degradation, localization of the DNA to the nuclear membrane,
decomplexation of the DNA from its carrier and facilitation of nuclear import, some or
all of which must occur prior to the initiation of transcription in other commonly used
transfection methods (Lechardeur and Lukacs 2006). Secondly, the immobilization of
DNA on the surface of the nanofiber using EDC condensation and other tethering
strategies allows for potential control over the fate of introduced genes. This could be
highly advantageous for environmental applications, where genetic manipulation of
species and cell types and control over these transgenic variants for in situ studies is
extremely desirable. Also, the spatially indexed feature of VACNF platforms allows for
convenient location and tracking of a multitude of individual cells over extended periods
of time without the need for expensive equipment. Traditionally, this would require a
microscope capable of CO2 and temperature regulation and an additional tracking system to monitor more than an individual cell per experiment. Using VACNF arrays, no external equipment is required and cells can simply be transported from the incubator to
28 the microscope time and time again. Another advantage of impalefection is that the simplicity of how the method works makes it potentially applicable to a number of cell types, including hard-to-transfect cell types such as neuronal cell lines and plant cells.
While the impalefection method is very straightforward and simple, there has been some degree of irreproducibility in the past. The cause of this variability has been difficult to identify, and warrants a need to identify and optimize the specific parameters of the impalefection method in order to gain consistent and reproducible results for future experiments. Optimization of physical transfection methods can involve a number of parameters. In microinjection, the manufacture of pipets, DNA concentration, viscosity and size of the target cell must all be taken into account (Brown and Fleming 1986;
Colosimo et al. 2000). No data has been collected on the importance of cell age and physiology, although it can be assumed that metabolic activity is an important factor for membrane recovery following the injection of DNA. In biolistic bombardment, gas pressure, target distance, DNA concentration, particle size and particle density are the main parameters involved in method optimization (Colosimo et al. 2000; Heiser 1994).
Sanford et al. have an extensive review on optimization of biolistic bombardment, noting that cell age and physiology can vary greatly depending on the organism or cell type
(Sanford et al. 1993). In B. megaterium strain 7A17, the highest efficiency was observed with cells transformed in early-log phase (Shark et al. 1991). In E. coli JA221 there appears to be no difference between midlog, late-log or stationary phase growth of cells for transformation efficiency (Smith et al. 1992). For biolistic bombardment in
Saccharomyces cerevisiae, the optimal efficiency of transformation occurs in mid-
29 stationary phase (Armaleo et al. 1990). Likewise, optimal growth phase of plant cell suspensions for transformation can differ greatly from species to species. In tobacco NT1
cells, 4 days after subculturing (early log-phase) results in the highest transformation
efficiency (Paszty and Lurquin 1987). However, in Oryza sativa L. Japonica cv. Taipei
309 rice cells, the most efficient method is to use cells 6-weeks after subculturing
(stationary phase) (Chen et al. 1998). There are no data available on the optimization of
mammalian cell physiology for biolistic bombardment of cell cultures. Conversely, a
number of studies have had success to varying degrees with biolistic transformation in
different mammalian cell lines. Novakovic et al. used biolistic bombardment in B-16
melanoma cells, breast adenocarcinoma MCF7 cells, and normal fibroblast L929 cells
(Novakovic et al. 1999). In this study, cells were cultured to high density before
bombardment. The proportion of transfected cells were 1%-38% in B-16 cells, 2%-27%
in MCF7 cells, and 1%-11% in L929 cells. O’Brien and Lummis had similar transfection
efficiency (up to 30%) in HEK293 cells when grown to 60%-80% confluency prior to
bombardment (O'Brien and Lummis 2004). For neuronal cell lines or astrocytes,
McAllister suggests transfecting cells near confluency, although this is an ambiguous
description (McAllister 2000). However, the expected transfection efficiency for random
physical methods like biolistic bombardment is typically low in neuronal cells, and are
often administered to slice preparations or disassociated cells as opposed to adhered
cultured cells (McAllister 2000). These studies demonstrate that cell physiology can play
a large role in measuring the efficiency and optimization of physical transfection
methods. Physiological factors of mammalian cells need to be further addressed in
30 relation to the efficiency of impalefection. These include cell concentration, cell
confluency, cell type and culturing conditions.
Potential Mammalian Cell Responses to Impalefection
After DNA is delivered to mammalian cells using VACNF arrays, the cells recover from
the nanofiber impalement and the transgene is expressed. So the question remains: How
do the internalized nanofibers affect the transcriptional activity of the cell? How do the
cells recover in response to nanofiber impalement and which gene pathways are up-
regulated to achieve this response? How does the membrane reseal after nanofiber
impalement? Investigations are currently underway in order to answer these questions
more directly, and insight on mammalian cell membrane repair can be found in the
literature. While the cellular response and genetic pathways of cell recovery after
nanofiber impalement will not be addressed in this study, understanding the potential
physiological response of the cell to intracellular nanofiber penetration is crucial in our
efforts to characterize and optimize the impalefection method in mammalian cells.
There are a number of observed effects in mammalian cells in response to acute cell
membrane injury. This infiltration of the cellular membrane can be achieved using fluid sheer stress, scrape-induced damage or laser irradiation (McNeil et al. 2006; Mellgren et al. 2007; Miyake et al. 2002; Ranjan et al. 1996). These methods result in membrane disruptions between 1 and 1000 µm2 (McNeil and Terasaki 2001; Steinhardt et al. 1994),
somewhat larger than the size of disruption that a 100 nm diameter VACNF would leave
31 in the cell membrane. However, the cellular processes involved may be very similar. An
increase in glucose uptake is involved in repair of membrane integrity, most likely due to
an increase in energy requirements (Ikari et al. 2003). It is well established that the resealing of membranes requires the influx of extracellular Ca2+ into the cell (McNeil and
Steinhardt 2003). However nucleated mammalian cells can reseal holes < 1 µm in the
absence of Ca2+ (McNeil and Kirchhausen 2005). For larger holes, Ca2+ influx to the cell
triggers the fusion of vesicles and the recruitment of vesicle-vesicle aggregates to the
damaged site, resulting in a “patch” of the membrane (McNeil 2002; McNeil et al. 2001;
McNeil et al. 2003; Miyake and McNeil 2003). While it was originally thought that
lysosomes were the vesicular candidates for membrane sealing , it has recently been
demonstrated that when lysosomal fusion is inhibited by vacuolin 1, the cellular membrane can still be repaired as normal (Cerny et al. 2004). This has lead to a hunt for
a new repair organelle candidate within the cell, with enlargeosomes being the current
focus (McNeil and Kirchhausen 2005). While these cellular processes have been
characterized, membrane recovery on a molecular scale is less understood. The family of
SNARE proteins has been linked with a number of roles in membrane fusion (Rothman
2002). The family of synaptotagmin proteins are also involved in Ca2+-mediated vesicle
fusion at the membrane surface, and inhibition of synaptotagmin III and synaptotagmin
VII results in a lack of membrane resealing in fibroblasts (Reddy et al. 2001). Fluid shear
stress on the cellular membrane of Chinese hamster ovary cells has been shown to induce
the expression of c-fos, a transcriptional activator involved in growth, differentiation and
stress (Ranjan et al. 1996). Annexin A1, a cytosolic protein that binds to phospholipids and increases aggregation and fusion of lipid bilayers, is recruited to the torn plasma
32 membrane after HeLa cells are scrape damaged (McNeil et al. 2006). This implies that
annexin A1 also plays a role in membrane repair machinery. More recently, calpains
have been associated with calcium-dependent membrane repair (Mellgren et al. 2007).
Calpains are cysteine proteinases found in all animal cells that target cytoskeletal proteins for proteolysis (Dayton et al. 1976; Wang and Yuen 1999). This suggests that cytoskeletal remodeling is immediately induced to assist in membrane repair.
Inducible Gene Silencing and the RNA Interference Pathway
It has been demonstrated that VACNF arrays can be used as electrodes for induction and
recording of electrochemical signals of mammalian cells (McKnight et al. 2006a; Yu et
al. 2007b), and it has additionally been demonstrated that VACNF arrays can be used for
gene delivery and the covalent immobilization of genes on the nanofiber surface
(McKnight et al. 2003b; McKnight et al. 2004b). At this point, it can be asked what the
limits of VACNF arrays may be for cellular gene expression and phenotypic response.
As an electrochemical probe that can stimulate cellular responses and is often residing
inside the nucleus of the cell, there is the potential that a resident nanofiber may
ultimately be used to electronically modulate the transcriptionally activity of DNA within
the nucleus, thus controlling gene expression or lack thereof. While the parameters
involved in achieving this function are still under investigation and this specific
application is out of reach at this point, steps can be taken towards this goal. One directly
addressable issue is whether endogenous genes within the cellular genome can be
controlled from the VACNF platform. Due to recent advances in molecular biology and
33 the discovery and exploitation of the RNA interference pathway, endogenous genes can
be silenced within a cell exogenously using post-transcriptional gene-silencing
technology. These processes are discussed in more detail below.
RNA interference (RNAi) is an evolutionarily conserved cellular process that operates to silence gene expression by the specific targeting of mRNA for degradation. While the mechanism of RNAi-mediated silencing was first described in Caenorhabditis elegans
(Fire et al. 1998a), the initial effects of RNAi were observed in the early 1990s in transgenic petunias (Napoli et al. 1990; van der Krol et al. 1990). The RNAi pathway has now been identified in most eukaryotic organisms including, but not limited to
Trypanosoma (Shi et al. 2000), Drosophila (Svoboda et al. 2000), Arabidopsis (An et al.
2003), Entamoeba (Kaur and Lohia 2004), yeast (Sigova et al. 2004) and humans
(Elbashir et al. 2001). In the RNAi pathway, long double-stranded RNAs (dsRNAs) are recognized by the RNase III enzyme Dicer which cleaves the dsRNAs into small interfering RNAs (siRNAs) that range from ~21 to 24 base pairs in length (Bernstein et al. 2001a). These siRNAs are subsequently joined by the RNA-induced silencing complex (RISC) and bind directly to the complementary mRNA transcript, resulting in cleavage and degradation of complementary mRNA transcripts (Hammond et al. 2000).
This targeting of mRNA for degradation and inhibition of translation is referred to as post-transcriptional gene silencing (PTGS) or co-suppression (Zamore et al. 2000).
While the major function of RNAi seems to be as a viral defense mechanism (Dalmay et al. 2001; Mourrain et al. 2000), it has been proposed that siRNAs are also used to silence
34 transposable elements and repetitive genes to help stabilize the genome (Novina and
Sharp 2004; Volpe et al. 2002).
The RNAi pathway has additionally been utilized as a therapeutic approach against human diseases (Kim and Rossi 2007), including age-related macular degeneration
(Check 2005; McFarland et al. 2004) and respiratory syncitial virus (RSV) infection.
Clinical trials are currently underway for other therapies which target viral diseases
(Dykxhoorn and Lieberman 2006), neurodegenerative disorders (Raoul et al. 2006) and cancer (Pai et al. 2006), and exciting recent advances have been confirmed in preventing
HIV-1 infection (Bagasra 2005; Jacque et al. 2002; Rossi 2006), although the safety of these therapies are possibly in question (Couzin 2006). The molecular mechanism of
RNAi has also been used for studying gene expression and gene function in mammalian cells. Highly specific and potent siRNAs have been repeatedly incorporated into methodologies for genetic studies (Brummelkamp et al. 2002; Kasim et al. 2004; Kaykas and Moon 2004; Mangeot et al. 2004; Matsukura et al. 2003a; Paddison et al. 2004). In
Caenorhabditis elegans, dsRNA has been introduced in a variety of ways (Fire et al.
1998a; Kennedy et al. 2004; Lau et al. 2001), and gene screening has been performed on a large scale (Berns et al. 2004; Konig et al. 2007; Maeda et al. 2001). The synthesis of short dsRNAs has opened the door for methodologies in reverse genetic studies of mammalian cells that only formerly existed for nematode and fruit fly platforms
(Matsukura et al. 2003a). In mammalian cells, the injection of long dsRNAs directly into mammalian cells can result in a cytotoxic reaction by triggering a type-I interferon response and terminating genome-wide translation (de Veer et al. 2005; Manche et al.
35 1992). However, introducing siRNAs directly to mammalian cells can bypass the
induction of a cytotoxic response, resulting in efficient silencing of the target genes
(Kariko et al. 2004; Kim et al. 2004). Likewise, vectors of DNA expressing siRNA
sequences as short inverted repeats (or short hairpin RNAs, also shRNAs) have been
developed and validated for stable down-regulation of targeted gene expression
(Brummelkamp et al. 2002; Kasim et al. 2004; Kaykas and Moon 2004; Mangeot et al.
2004; Matsukura et al. 2003a; Paddison et al. 2004). In mammalian cells, these shRNA
vectors include designs for constitutive and inducible gene silencing (Bantounas et al.
2004; Fewell and Schmitt 2006). The conditional gene silencing vectors can be induced
by a number of compounds (Wiznerowicz et al. 2006), but tetracycline or doxycycline (a
tetracycline derivative) induction are the most common choices in the literature (Ito et al.
2006; Kappel et al. 2007; Lin et al. 2004; Matsukura et al. 2003a). Polymerase III
promoters are highly advantageous for inducible shRNA vectors due to their high level of
activity (approximately 4 x 105 transcripts per cell), lack of a polyadenosine tail and well-
defined transcriptional start sites (Brummelkamp et al. 2002; Matthess et al. 2005).
These include the uridine-rich U6 and histone 1 (H1) gene promoters, which have both
been altered to contain two Tet operator (tetO2) sites in shRNA vectors (Wiznerowicz et
al. 2006). In tetracycline inducible expression, these TetO2 sites bind the tetracycline
repressor protein (TetR), which results in repression of transcription of downstream
shRNA sequences. However, upon the addition of tetracycline (Tc), Tc binds tightly to
TetR, generating a conformational change that displaces TetR from tetO2 sites (Gossen and Bujard 2002; Ramos et al. 2005). With TetR sequestered, the H1 promoter is no longer sterically hindered, resulting in transcription of the shRNA sequence. This results
36 in generation of “induced” siRNAs that can specifically silence a targeted gene. The simple concept and ease of use in RNAi technology makes it a very attractive and powerful tool in genetic studies when compared to traditional methods (e.g. knockout mice).
OBJECTIVES OF THIS STUDY
The major goals of this study were to experimentally investigate a series of hypotheses which have been chosen to further characterize the process of VACNF-mediated impalefection and evaluate the transcriptional activity of EDC-immobilized DNA on the
VACNF arrays. Lastly, impalefection was used for studying inducible gene silencing to test the range and applicability of the method for analyzing intracellular control and analysis of gene expression in Chinese hamster ovary cells. The hypotheses are listed below:
I. The parameters of impalefection can be characterized and the process of
impalefection can be optimized for higher efficiency of gene expression after
cellular impalement in Chinese hamster ovary cells.
II. DNA immobilized on VACNF arrays by EDC-condensation remains
transcriptionally active and can be quantified.
37 III. The transcription of impalefection-delivered DNA can be induced in Chinese
hamster ovary cells on VACNF arrays and can subsequently control the
expression of endogenous genes within these cells.
38 CHAPTER II
OPTIMIZATION OF THE VERTICALLY-ALIGNED CARBON NANOFIBER- MEDIATED TRANSFECTION METHOD IN MAMMALIAN CELLS
Introduction
The ability to insert a foreign gene into a mammalian cell line of choice is a pivotal tool in the current age of molecular biology. A variety of more traditional transfection methods, including chemical, electrical, physical and mechanical means, have become essential laboratory procedure for the generation of transient cell assays and stable- chromosomal insertions. More recently, nanostructured architectures have shown promise in bringing new application to the conventional discipline of mechanical
transfection. These nanosized structures include nanofibers (McKnight et al. 2003b;
McKnight et al. 2004b), nanotubes (Cai et al. 2005b; Pantarotto et al. 2004), nanorods
(Salem et al. 2003), nanoparticles (Bauer et al. 2004; Chowdhury et al. 2006; Zhi et al.
2006) and nanoscaffolds (Uchimura et al. 2007). Multifunctional nanorods and
nanoparticles offer a novel approach because they can be coupled with condensed DNA,
and can also be incorporated with ligands for specifically targeted sites of delivery
(Green et al. 2007; Salem et al. 2003; Tan et al. 2007). Likewise, vertically aligned
carbon nanofiber (VACNF) arrays possess unique characteristics for gene delivery in
mammalian cells. In a process termed impalefection, stacked-funnel type carbon
39 nanofibers of micron lengths and nanometer tips which have been deterministically
synthesized in a vertical fashion on a silicon substrate using plasma-enhanced chemical
vapor deposition (PECVD) are used as a platform to physically impale mammalian cells
(McKnight et al. 2004b). These nanofibers can be spotted or covalently immobilized
with transcriptionally-active plasmid DNA containing the gene of interest and allow for
parallel gene delivery into hundreds of mammalian cells (Mann et al. 2007; McKnight et
al. 2003b). Mammalian cells can then recover from the impalefection process and
proliferate on the nanofiber substrate, allowing for cell expression assays over extended
periods of time. Utilizing spatial indexing of the VACNF arrays, individual cells can be
tracked over time without having to remain under constant observation (McKnight et al.
2004b). This advantage allows the VACNF chips and impalefected cells to be monitored
in a relatively simple and inexpensive manner, due to the cells remaining under and
returning to the preferred conditions in the incubator except for times of analysis.
While the method of impalefection on VACNF arrays has been successfully used for gene delivery and subsequent experimentation, the process itself has never been optimized. The efficiency of gene delivery and expression of marker genes in cells using impalefection is often as high as 70%-80% in some regions of nanofiber array chips, while at other times no transfected cells have been observed. This level of variability can be seen in similar mechanical transfection methods, such as biolistic bombardment and microinjection. In his excellent review, Proctor states that microinjection “…is the least efficient method that has been used to transfer DNA into mammalian cells …. [and] is also the most technically difficult” (Proctor 1992b). However, these methods, along with
40 impalefection, are of great use due to their specialized applications. The difficulty and variability of mechanical transfection methods are largely in part from the numerous steps and components in the process. Therefore, like other mechanical transfection methods, it is vital to optimize the parameters involved in impalefection using VACNFs in order to gain consistent and reproducible results for future experiments.
Efficiency of impalefection is defined as the number of cells expressing the delivered gene of interest divided by the total number of viable cells after impalement. The parameters that can potentially affect the efficiency of impalefection are seemingly unlimited. Some mechanical parameters such as nanofiber length and tip diameter are easy to control during nanofiber synthesis. However, the effects of other parameters such as the cellular membrane composition are harder to evaluate, but nonetheless affect the efficiency to a similar degree. In this study, the method of impalefection was optimized by evaluating predetermined mechanical, chemical and biological parameters. A list of these impalefection parameters can be seen in Table 1. In order to measure efficiency consistently, the pd2EFYP-N1 vector, containing the gene sequence encoding for the enhanced yellow fluorescent protein (YFP) was introduced to a cyan fluorescent protein
(CFP) stably-expressing cell line or wild type cell line. The dual marker gene system was used for ease of cell counting and monitoring positive cell impalefections among non- impalefected cells using an inverted fluorescent microscope. Afterwards, the efficiency of the optimized technique was compared to the efficiency of impalefection prior to optimization.
41 Table 1. Different parameters of the impalefection method.
42 Materials and Methods
Plasmid isolation and maintenance
Plasmids were maintained in E. coli strain JM109 and stored at -80º C. E. coli strain
JM109 was grown in Luria-Bertani (LB) broth at 37° C with or without 100 µg
ampicillin/ml or 50 µg kanamycin/ml, depending on the requirements for plasmid
maintenance. Plasmid isolation was performed using Wizard mini- or midi-prep kits
(Promega, Madison, WI). The expression vectors pd2EYFP-N1 containing yellow
fluorescent protein (YFP) and pd2ECFP-N1 containing cyan fluorescent protein (CFP)
were purchased from Clontech (Mountain View, CA).
Cell culture and reagents
All cell types were maintained in plastic T25 and T75 flasks under 5 % CO2 and 95 % humidified air at 37º C. The cell types used are as follows: Chinese hamster ovary
(CHO-K1), human epithelial kidney (HEK293), human colorectal (HCT-116), rat liver epithelial (Clone 9), rat adrenal gland (PC12), rat aorta (A10) and mouse monocyte/macrophage (J774) cells. CHO-K1-CFP cells are a subclone of (CHO-K1 cells) (ATCC, Manassas, VA) containing a stabe insertion of the cyan fluorescent protein
(CFP) gene sequence. CHO-K1-CFP and Clone 9 cells were grown in Ham’s F12K medium supplemented with 2 mM L-glutamine (ATCC) and 10% fetal bovine serum
(Gibco-Invitrogen, Carlsbad, CA). HEK293 cells were grown in minimum essential
43 medium (Eagle) supplemented with 2 mM glutamine (ATCC) and 10% fetal bovine serum. PC12 cells were grown in Ham’s F12K medium supplemented with 2 mM L- glutamine, 15% horse serum and 2.5% fetal bovine serum. HCT-116 cells were grown in
McCoy’s 5a medium supplemented with 1.5 mM L-glutamine (ATCC) and 10% fetal
bovine serum. A10 and J774 cells were grown in Dulbecco's modified Eagle's medium
(DMEM) supplemented with 4 mM L-glutamine (ATCC) and 10% fetal bovine serum.
All cell media were additionally supplemented with 100 µg/ml of streptomycin and 100
U/ml penicillin (Gibco-Invitrogen). For experiments with sodium butyrate (NaB), cells
were allowed to recover in their respective medium supplemented with a final
concentration of 2 mM NaB (Sigma-Aldrich, St. Louis, MO).
Vertically aligned nanofiber-mediated impalefection
The method of impalefection has been previously described in detail (McKnight et al.
2003b). In this study, the technique was slightly modified and ultimately optimized in a
number of parameters. Briefly, cells were grown to near or full confluency in T25 or T75
flasks and disassociated from the plastic substrate using Cellstripper™ (Mediatech, Inc.;
Herndon, VA) and resuspended in 5 mls of the appropriate medium. Cells were centrifuged at 300 x g for 10 minutes, medium was aspirated and the cell pellet was resuspended in 0.2-1.0 ml, depending on the application. Cells were allowed to settle for
5 minutes and impaled using inverted nanofiber (carbon or silicon) array chips of 2.2 mm x 2.2 mm size adsorbed with pd2EYFP-N1 plasmid. The chip was then placed in inverted fashion in 200 µl of their respective, fully supplemented growth media in a
44 round-bottom well of a 96-well plate. The plate was incubated at 37 ºC with 5% CO2 and allowed to recover for 18-24 hours.
Optimization of impalefection substrate
For experiments involving the optimization of the nanofiber composition (carbon or silicon), nanofiber pitch or impalement substrate (glass slide, polydimethylsiloxane
(PDMS) pad, etc.), all other parameters were consistent. Impalefections were performed as described above and five to ten replicate chips were used for each experimental variable. For the impalement substrate experiments, 3” x 1” pre-cleaned glass micro slides were purchased from Corning (Corning, NY). Compliant silicone pads were fabricated of polydimethylsiloxane (PDMS) using a commercially available two part mix
(Sylgard 184, Dow Corning), mixed per manufacture’s specifications and poured on clean, 100-mm silicon wafers to ~3 mm thickness. Following 30 minute cure at 65 deg
C, PDMS pads were peeled from the silicon substrate and cut to individual pieces. For the plastic impalement substrate, Costar® clear, non-treated flat-bottom 96-well plates
(Corning Inc.; Corning, NY) were used with or without a 5 mm x 5 mm pad of Durx 770 clean room wipes (Berkshire Corp.; Great Barrington, MA).
Optimization of cell confluency
CHO-K1-CFP cells were grown over 48 hours to differing levels of confluency (60%, termed subconfluent, or 100%, termed confluent). To account for differences in cell
45 concentration, two T25 flasks of subconfluent cells were used alongside one T25 flask of confluent cells. Cell preparation for impalement transfection was described above. Cell pellets were similar in size, and cells were resuspended in 0.2 ml Ham’s F12K media.
For cell impalement, approximately 50 µl of the cell suspension was spotted onto a glass slide, and cells were allowed to settle for 5 minutes. For impalefection, 2 µl of DNA
(pd2EYFP-N1, 5 ng/µl) were spotted onto each 041906_2 (Sesha, 20 µm pitch) nanofiber chip and allowed to dry. Immediately after drying, each chip was wetted in Ham’s F12K media, then inverted and gently placed over the settled cells. After pressing, each chip was placed face down in 200 µl of Ham’s F12K media in a round-bottom 96 well plate and the cells were allowed to recover for 24 hours. Five duplicate samples were used for each variable.
Optimization of DNA precipitation and concentration
For optimization of DNA concentration, CHO-K1-CFP cells were grown to full confluency. Nanofiber chips were either spotted or EDC-condensed with varying DNA concentrations (1 µg, 100 ng, 10 ng, 1 ng, or 100 pg). EDC condensation was performed as previously described (Mann et al. 2007) with some modifications. Briefly, each DNA concentration was aliquoted into 100 µl of MES buffer, pH 4.5, with 10 mg/ml EDC in one well of a 96-well plate. The 96-well plate was rotated on a Lab Line 3-D Rotator for
2 hours. After 2 hours, each nanofiber chip was moved to a new well of the plate containing 200 µl of sterile H2O. The chips were incubated in H2O for 2 min followed by the replacement of H2O with 100 µl of 2 M NaCl in PBS and incubated for 2 min once
46 again. This washing step was repeated twice. Finally, the nanofiber chips were placed in
100 µl of PBS to prevent drying on the surface prior to cell impalement. Spotted DNA
chips were spotted with 2 µl of DNA (1 µg, 100 ng, 10 ng, 1 ng, or 100 pg) and allowed
to dry. Impalements were carried out described above, with five replicate carbon
nanofiber chips for each variable. As a negative control, three chips were incubated with
100 ng of pd2EYFP-N1 in MES buffer without EDC. For optimization of DNA
precipitation, pd2EYFP-N1 was spotted and allowed to dry on nanofiber chips under a
variety of conditions including DNA alone, DNA with 20 mM spermidine (Sigma-
Aldrich), DNA with 1M CaCl2 (Sigma-Aldrich), DNA with 20 mM spermidine and 1M
CaCl2. Additionally, pd2EYFP-N1 was dried and subsequently washed with ethanol
(EtOH) (Sigma-Aldrich) or isopropanol (IPA) (Sigma-Aldrich) prior to impalefection.
Impalefections were carried out as described above. When testing DNA precipitation, each variable was repeated with five individual nanofiber chips of each type. These
DNA precipitation experiments were done simultaneously using chips from carbon nanofiber and black silicon nanofiber arrays.
Optimization of Cell Concentration and Cell Washing
In order to optimize the cell concentration of impalefection, CHO-K1-CFP cells were
grown in T75 flasks to near confluency. After dissociation, centrifugation and
resuspension in fresh Ham’s F12K medium, cells were counted using a Beckman-Coulter
Cell Counter (Fullerton, CA). Cells were then pipetted into separate wells of a flat-
bottom 96-well plate containing Durx pad. Three different cell concentrations were used:
47 1 x 104 cells, 1 x 105 cells, and 1 x 106 cells well-1. Concurrently, the efficiency of a cell washing step immediately after impalefection was also investigated. Briefly, five replicated nanofiber chips of each differing cell concentration were prepared in duplicate
(for a total of ten nanofiber chips). One set of these duplicate chips were placed
immediately face down in 200 µl of fresh Ham’s F12K medium in a round-bottom 96-
well plate after the impalement step. For the other set of duplicate nanofiber chips, after
the impalement step each chip was placed in 200 µl of fresh Ham’s F12K medium and washed with a moderate degree of force for 5 seconds before being placed face down in
200 µl of fresh Ham’s F12K medium in a round-bottom 96-well plate and allowed to recover.
Impalefection Efficiency Measurements
Impalefection efficiency was determined by counting the total population of cells on each nanofiber chip, and dividing it by the number of YFP-expressing cells present on the same chip. Since CHO-K1 cells were stably expressing the CFP gene, cell counting could be done using direct observation using the inverted fluorescent microscope. For all other cell types, chips were first counted for all YFP-expressing cells, and then incubated in fluorescein diacetate at a final concentration of 10 µM for 5 minutes. This allowed for counting of the total population of cells on the chip. For comparison of non-optimal and optimal impalefection efficiency, efficiency data from 234 carbon nanofiber-mediated impalefections of CHO-K1 cells were used. Of these impalefection experiments, 117 sample chips were used under a variety of non-optimized conditions, while 117 of the
48 sample chips were used under optimized conditions. Optimized conditions are defined as
carbon nanofiber arrays with 5 µm pitch, spotted with between 100 ng and 1 µg of non-
complexed pd2EYFP-N1, and pressed on Durx 770 clean room wipes in flat-bottom 96-
well plates into cells at near or full confluency. Some cells were allowed to recover
under the presence of 2 mM NaB. These data were analyzed using SigmaPlot 10.0
(Systat Software Inc., San Jose, CA). For comparison of impalefection efficiency
between the whole nanofiber chip and regions of the nanofiber chip, 24 separate carbon
nanofiber-mediated impalefections of CHO-K1 were randomly chosen. Since digital
images of these impalefection experiments were previously recorded between 3/8/2007 and 8/6/2007, the impalefection efficiency of the whole chip was already known. In order to determine the regional efficiency, one region of the digital image was selected for each chip, containing no less than 20 total cells, and the impalefection efficiency of this region was calculated. The regional efficiency was averaged for all 24 impalefections, and then compared to the average whole chip efficiency of the same impalefection experiments.
Fluorescence microscopy, cell staining and cell counting
Cells were monitored and counted using a Nikon Diaphot 300 inverted fluorescent microscope (Nikon, Melville, NY) with filter sets for CFP (436 nm excitation, 480 nm emission) and YFP (500 nm excitation, 535 nm emission) purchased from Chroma
Technology Corp. (Rockingham, VT; Cat. No. 31044v2 and 41028). All nanofiber chips
49 were counted for total number of surviving cells on the chip and total number of cells
successfully impalefected (expressing YFP).
Nanofiber synthesis
For carbon nanofiber array synthesis, 4” silicon wafers (100) were photolithographically
patterned with 50 nm thick Ni thin films as discrete 500 nm diameter dots at either a 5,
10, or 20 μm pitch over the entire surface of the wafer. Nanofibers were synthesized in a
custom built DC-PECVD reactor at a temperature of 650 ºC, 10 torr, 2 A, using a mixture
of a carbonaceous source gas (acetylene) and an etch gas (ammonia). Growth time was
selected to provide fibers ranging from ~10-17 μm tall, with tip diameters of ~100 nm.
Spatially indexed arrays were subsequently spun with a 2 μm thick layer of SU-8 2002 photoresist (Microchem Corp., Newton, MA) and patterned with a grid and numerical indexing using contact photolithography and development (SU-8 Developer,
Microchem). Following growth and index patterning, wafers were spun in a protective
layer of photoresist (Microposit SPR220 CM 7.0, Shipley Corp., Marlborough, MA) and
diced into 2.2 mm x 2.2 mm square chips. Following dicing, the protective photoresist
was removed by soaking in Microposit Remover 1165 (Shipley), followed by copious
rinsing in water. For silicon nanofiber array synthesis, silicon wafers (Silicon Quest
International, Santa Clara, CA) were anisotropically etched in SF6/O2–based high density
plasma at cryogenic temperatures. Briefly, wafers were loaded in an Oxford 100
DRIE/RIE system capable of cryogenic processing. The wafers were cooled to -110 ºC
and SF6/O2 gases were introduced. The ratio of fluorine molecules to oxygen molecules 50 was selected by using the black silicon method (Jansen et al. 1995). An RF
electromagnetic field was then applied to the wafers, ionizing gas molecules and creating
a plasma. During the subsequent etch process, native oxides, dust particulates, organics, and other contaminants on the wafer surface acted as micromasks. Underlying silicon
was protected because of the anisotropic nature of the etch. This enabled the formation of
10 μm silicon fibers as the etch was allowed to proceed.
Statistical Analysis
All data for impalefection efficiency were calculated as means and standard deviation of the mean. For most experiments, determination of the statistical significance between different variables was made using Student’s t test in Microsoft Excel 2003 (Microsoft
Corp., Redmond, WA). For statistical comparisons of impalefection efficiency in
different cell types, a nonparametric one-way ANOVA on the ranked data, and Kruskal-
Wallis Multiple Comparison Z test with Tukey-Kramer adjustment was done in NCSS
(Number Crucher Statistical Systems, Kaysville, UT). For all analytical results, a p value
of less than 0.05 was considered to be statistically significant.
51 Results and Discussion
Optimization of mechanical parameters
Nanofiber composition. The vertically aligned nanofiber array platform was optimized
by experimenting with the nanofiber composition, the impalement substrate, and some
technique oriented variables. Traditionally, impalefection has been performed with
carbon nanofibers. However, recent synthesis of silicon nanofibers has shown potential
application for useful DNA immobilization chemistry (unpublished data). Carbon
nanofiber arrays were investigated parallel to silicon nanofiber arrays using spotted DNA
(Figure 2). While there was a significant difference between the silicon and carbon
nanofibers in the presence of CaCl2, spermidine and isopropanol (IPA), similar levels of
efficiency for impalefection were observed for most of the variables tested. While no
significant difference could be seen among the other variables individually, carbon
nanofibers did show a significantly higher impalefection efficiency over silicon
nanofibers when all variables were averaged (8.5% ± 4.4% vs. 5.7% ± 2.8%,
respectively. p value = 0.002). Therefore, carbon nanofibers were used for the remainder
of the optimization experiments. Worth noting is the fact that silicon nanofibers were not
examined using DNA immobilization techniques (e.g. EDC condensation) that have been
successful in the past on carbon nanofibers 1, 12. This could potentially be addressed in future studies.
52 Impalement substrate. The substrate used for impalefection has been a variety of flat
surfaces in past studies including PDMS and wetted glass slides (McKnight et al. 2003b;
McKnight et al. 2004b). Additionally CHO-K1 cells have been centrifuged onto
nanofiber arrays, although this step can be replaced with simply allowing the cells to settle onto the nanofiber chip prior to impalement, which was the method used throughout this study. Different impalement substrates were investigated to test for differences and optimal qualities. Impalements were performed on 1) a glass slide, 2) a
PDMS pad, 3) a flat-bottom well from a 96-well plate, and 4) a flat-bottom well from a
96-well plate containing Durx clean-wipe pad. Each impalement substrate was examined and the efficiency of impalefection was calculated by dividing the total number of cells by the number of transfected cells expressing YFP (Table 2). The glass slide and PDMS pad showed very low efficiency values (0.88% ± 1.91% and 1.71% ± 2.39%, respectively) when compared to the other variables. On the other hand, the flat- bottom plate wells with and without Durx pad showed significantly higher impalefection efficiencies (6.44% ± 2.05% and 6.06% ± 1.30%, respectively). This could be explained by the nature of the plastic flat-bottom plate with and without the Durx pad. In the flat- bottom plate, the cells are limited in movement due to the small size of the well.
Additionally, the plastic surface may allow for cellular adherence, resulting in cells becoming immobilized to some degree prior to impalement. The pore size and composition of the Durx pad allows for cells to settle in the crevices of the pad, similarly immobilizing them for the pressing step during impalefection. This is not the case with the glass surface and PDMS pad. The cells are not spatially limited and nanofibers can
53 Table 2. Effects of cell confluency, impalement substrate, and sodium butyrate on impalefection efficiency. * denotes a statistically significant difference (p < 0.05).
54 often break off the chip when pressing against hard surfaces like glass. Likewise,
nanofibers can become embedded in plastic surfaces during impalement, making the
Durx pad even more convenient as an impalement substrate.
Optimization of chemical parameters
DNA precipitation. In traditional methods of gene delivery, the DNA is often
precipitated or condensed using divalent or trivalent cations or alcohols (Behr et al. 1989;
Knight and Adami 2003). The enhancement of impalefection by the addition of a
divalent cation (CaCl2), a trivalent cation (spermidine), and alcohol – ethanol (EtOH) or
isopropanol (IPA) – were investigated (Figure 2). Strikingly, no means of DNA
precipitation had an enhancing effect on impalefection efficiency, while spermidine and
CaCl2 or any combination of both resulted in decreased efficiency. The addition of
EtOH or IPA prior to impalement appeared to increase efficiency, but not by any
significant degree. This data suggests that DNA alone is sufficient for high impalefection
efficiency, and was used for all further experimentation. While this result was surprising,
it also revealed the simplicity of the concept behind the impalefection method: DNA is
simply being inserted mechanically into cells by physical means, often directly to the
nucleus, and cells are subsequently allowed to recover and express the inserted genes.
This explains why complexed DNA, often used to “disguise” the DNA for passage through the cell membrane or to protect the DNA from degradation once inside the cell, had no effect on the efficiency of impalefection.
55
Figure 2. Effects of different DNA precipitation methods on the impalefection efficiency of carbon and silicon nanofibers. Experimental variables: 1) DNA, 2) DNA
and CaCl2, 3) DNA and spermidine, 4) DNA and CaCl2 and spermidine, 5) DNA and
CaCl2 and spermidine and IPA wash, 6) DNA and IPA wash, 7) DNA and EtOH wash. *
denotes a statistically significant difference (p < 0.05). Error bars denote standard
deviation with n = 5 for each variable.
56 DNA concentration. The efficiency of expression increased with increasing concentrations of DNA (Figure 3). The efficiency of impalefection was significantly
higher at DNA concentrations of 100 ng chip-1 and 1 µg chip-1 (15.06% ± 4.06% and
12.92% ± 5.52%, respectively). While impalefection efficiency increased at these DNA
concentrations, successfully impalefected cells have been observed with as little as 20 pg
of DNA chip-1. Plasmid DNA was freshly prepared prior to these experiments, although the quality and purity of the DNA as well as the size of the plasmid could have differing effects on the impalefection efficiency. These factors were not quantitatively examined during this study. In addition, Figure 2 reveals that adsorbed DNA is impalefected into
CHO cells at a higher efficiency than immobilized DNA. This result was not surprising, due to the likely decreased amount of DNA that remains on nanofiber arrays after EDC- condensation and successive washing steps. Any DNA that is not covalently immobilized to the nanofibers is washed off the nanofiber array, resulting in a decreased amount of DNA when compared to nanofiber arrays that have been adsorbed with DNA.
Quantitation and analysis of transcriptional availability of EDC-immobilized DNA will be addressed in future studies.
Optimization of biological parameters
Cell concentration and cell washing. While washing can be a vital step in the impalefection process, decreasing the number of non-impalefected cells on the chip during the recovery period, extensive washing can also result in the dissociation and loss
57
Figure 3. Effects of the concentration of adsorbed and immobilized DNA on impalefection efficiency. * denotes a statistically significant difference (p < 0.05). Error bars denote standard deviation with n = 5 for each variable.
58 of many loosely-attached impalefected cells from the array. However, if the chip is not washed after impalement and prior to recovery, it can become overgrown with a population of non-impalefected cells that remained on the nanofiber array post- impalement, depending on the initial cell concentration used for impalefection. In order to experimentally address these issues, impalefections were carried out on different cell concentrations with and without the presence of a wash step. While variations in cell concentration did not seem to have any effect on the impalefection efficiency (Table 3), the initial cell concentration did expectedly affect the resulting number of cells left on the nanofiber array after impalefection and recovery. Therefore, if fewer impalefected cells or less confluent populations are desired for a particular study (i.e., longer assays requiring more space for cell growth over time), lower cell concentrations or short cell settling times (<5 minutes) should be used. Likewise, if a high number of impalefected cells or confluent cell populations are desired, higher cell concentrations and longer cell settling times (>5 minutes) should be used. Table 3 demonstrates that introducing a wash step immediately after the impalement step resulted in the loss of a substantial portion of cells, impalefected and non-impalefected. In past studies, chips have been oriented upright in 30 mm or 60 mm dishes to allow access to nutrients in the medium during incubation. Orientation of the chip during recovery can also affect the long-term viability and ability to track cells due to overgrown cell populations and agitation of the cells while inverting the chips for microscopic observation. To overcome these difficulties, a new technique has been introduced. Currently, the washing step has been removed and chips are placed in an inverted orientation in round-bottom 96-well plates for recovery.
59 Table 3. Effects of cell concentration and cell washing on impalefection efficiency.
* denotes a statistically significant difference (p < 0.05), and n is equal to 5 for each variable.
60 The round-bottom of the plate allows cells access to nutrients, while minimizing the
settling and growth of non-impaled cells during the first 24 hours of recovery.
Cell confluency. In recent experiments, Noll et. al demonstrated that standard
transfection methods (e.g. lipofection) resulted in higher levels of gene expression and
protein production when cells were cultured to confluency prior to transfection than when
cells were cultured to subconfluency (Noll et al. 2002). Therefore, the effect of cell
confluency on impalefection efficiency was investigated in CHO-K1 cells. While
differences in cell confluency did not have significant effects on impalement efficiencies
(Table 2), slightly higher efficiency was seen with fully confluent cells when compared to
subconfluent cells. Whether this increase is due to cell cycle, cell physiology or simply
cell number was not further investigated and remains unknown, but fully confluent cells
were used for future experiments.
Sodium butyrate. Sodium butyrate (NaB) was investigated as an additive during the recovery of impalefected cells, due to its known ability to enhance specific gene
expression and prolong transient gene expression post-transfection in mammalian cells
(De Leon Gatti et al. 2007; Gorman et al. 1983). Table 2 shows the impalefection efficiency of CHO-K1 cells in the presence and absence of NaB. No significant
difference with or without NaB was observed in the percentage of cells that expressed the
transgene after impalefection. However, the fluorescence of impalefected cells
expressing YFP in the presence of NaB was visibly brighter when compared to the
impalefected cells in the absence of NaB (data not shown). While the impalefection
61 efficiency was not affected by its presence, NaB has still proven to be useful for extended
transgene expression after impalefection, and will continue to be used for experiments
where longer transient expression assays (72 hrs+) are desired.
Cell type. CHO-K1 cells have been used as a model cell line in the past due to their robust nature in cell culture and the ease with which they can be transfected. In this
study, numerous other cell types were tested using the optimized impalefection method.
The impalefection efficiency in these different cell types are shown in Figure 4, and
included a variety of human (HEK-293, HCT-116), rat (A10, Clone 9, PC12) and mouse
(J774) cell types. HEK-293, PC12 and HCT-116 cells showed no significant difference
in impalefection efficiency when compared with CHO-K1 (Anova analysis; d.f. = 70, F-
ratio = 16.21, p = 0.00001), possibly due to the robust nature of all these cell types. A
slight increase was observed when PC12 cells were impalefected using VACNF arrays
with a nanofiber pitch of 2.5 µm rather than 5 µm (data not shown). This was likely due to the smaller diameter of PC12 cells. A similar increase in impalefection efficiency was not observed when CHO cells were impalefected with 2.5 µm pitch nanofiber arrays.
A10, Clone 9 and J774 cells showed significantly lower impalefection efficiency. Very few viable cells could be observed on the nanofiber arrays 24 hours after impalement in these cell types. Due to the physical nature of the impalefection technique, the variability
of efficiency in different cell types is most likely not due to difference in cell surface, membrane composition or uptake pathways. However, delivery to the cell nucleus is required for successful DNA expression following impalefection. Therefore, differences
62 in nucleus size may contribute to the variability of efficiency observed. Likewise, Clone
9 cells have also shown resistance to liposome-based transfection methods used in our lab
(data not shown), suggesting that this cell line may be more resistance to transfection in
general. Nonetheless, the results in Figure 4 demonstrate the applicability of the
impalefection technique over a diverse range of cell lines.
Evaluation of impalefection efficiency following optimization
Once the technique was optimized, impalefection efficiencies were compared between
the optimized parameters and a variety of non-optimized parameters. Efficiencies were
measured for 234 impalefection experiments of CHO-K1 (117 non-optimized samples
and 117 optimized samples) and compared (Figure 5). The distribution of efficiency
values for the two methods illustrates an increased efficiency of the new optimized
method that is statistically significant (p = 0.0004) when compared to the non-optimized
method. Additionally, these data demonstrate that the reproducibility has increased, as no efficiency values between 0.0% and 1.0% were observed for the optimized impalefection method.
In biolistic bombardment or other physical or mechanical transfection methods, regions of cells rather than the entire cell culture population are counted for efficiency, due to non-uniform parameters for the entire cell culture. This is rarely observed in chemical transfection methods like lipofection, where the entire cell culture receives consistent
63
Figure 4. Impalefection efficiency in different mammalian cell types. * denotes a
statistically significant difference (p < 0.05). Error bars denote standard deviation with n
= 5 for each variable.
64
Figure 5. Comparison of efficiencies of the non-optimized and optimized impalefection method. This box-whisker plot compares the efficiency values of impalefection samples using non-optimized parameters against impalefection samples using optimized parameters (n = 117 for each variable). Black dots correspond to outliers. There was a significant difference between the two values (p = 0.0004), determined using a p value < 0.05.
65 amounts of the transfection agent. However, in biolistic processes often only a local region of the cell culture dish is successfully transfected, and the central region of the cell culture dish is often killed due to exposure to gas pressure (Novakovic et al. 1999).
Therefore, the proportion of transfected cells is often determined only in regions of the cell culture. This concept of regional efficiency can be applied to VACNF-mediated impalefection as well, due to the frequent observation of many cells on the nanofiber array chips, but only small regions of the chip containing impalefected cells. This could be due to a number of factors, including loss of nanofibers on other regions of the chip, the pressing step taking place on an uneven surface, cells settling unevenly on the substrate, and misorientation of the nanofiber chip during impalement. In order to evaluate the regional impalefection efficiency, 24 chips were selected and counted for regional proportions of expressing cells. The results confirmed that the efficiency increased to 37.5% ± 11.5% in regions of the chip under optimal circumstances. This is a significant increase from 9.6% ± 6.1% observed in optimized impalefections, and would be expected since the majority of non-impalefected cells have been removed from the calculations.
Acknowledgements
The authors are grateful to T. Subich, D. Hensley, D. Thomas, S. Randolph, and P.
Fleming for carbon nanofiber fabrication, and to B. Fletcher and S. Retterer for black silicon nanofibers. This study was supported in part by grant R01EB006316 (NBIB) and by the Material Sciences and Engineering Division Program of the DOE Office of
Science (DE-AC05-00OR22725) with UT-Battelle, LLC and through the Laboratory
66 Directed Research and Development funding program of the Oak Ridge National
Laboratory, which is managed for the U.S. Department of Energy by UT-Battelle, LLC.
M.L.S. acknowledges support from the Material Sciences and Engineering Division
Program of the DOE Office of Science. A portion of this research was conducted at the
Center for Nanophase Materials Sciences, which is sponsored at Oak Ridge National
Laboratory by the Division of Scientific User Facilities (DOE).
67 CHAPTER III
QUANTITATIVE ANALYSIS OF IMMOBILIZED DNA ON VERTICALLY ALIGNED CARBON NANOFIBER GENE DELIVERY ARRAYS
This chapter is a modified version of a paper published with the same title in journal
Biotechnology and Bioengineering in July 2007 by David Mann, Timothy McKnight,
Anatoli Melechko, Michael Simpson and Gary Sayler.
Mann, D.G.J., McKnight, T.E., Melechko, A.V., Simpson, M.L., Sayler, G.S. 2007.
Quantitative Analysis of Immobilized DNA on Vertically Aligned Carbon Nanofiber
Gene Delivery Arrays. Biotechnology and Bioengineering, 97(4): 680-688.
The use of the term “we” in this chapter refers to my co-authors and myself. My primary
contributions in this chapter were the following: 1) assisting in the experimental design
of the PCR, IVT and qPCR investigations, 2) developing the PCR and qPCR primer and
probe sets, 3) performing all experiments and assisting in data analysis, 4) gathering and
reviewing the background literature and 5) writing the majority of the manuscript and
responding to the reviewer and editorial comments.
68 Introduction
There has recently been a heightened interest in the integration of synthetic, nanostructured materials with biological systems. Use of such materials to deliver tethered, transcriptionally-active DNA into mammalian cells has enabled novel approaches to genetic manipulation. This includes nuclear targeting of delivered DNA via heterofunctional metallic nanorods (Salem et al. 2003) and controlled, non-segregated transgene expression from arrays of nuclear-penetrant, vertically-aligned carbon nanofiber (VACNF) arrays (McKnight et al. 2003b; McKnight et al. 2004b). This latter method, termed parallel impalefection, is a technique similar to microinjection, differing in that it can be conducted on a highly parallel basis to manipulate many cells simultaneously. Vertically-aligned carbon nanofibers with micron lengths and sub 100 nm diameters are arrayed in parallel normal to the surface of a solid substrate. Pressing this array into a cell or tissue matrix results in cellular penetration and ‘microinjection’ of surface-bound material into a large number of target cells. Due to the nanoscale diameter of VACNFs, the plasma membranes of mammalian cells can recover following fiber penetration which enables the proliferation of the interfaced cell (McKnight et al. 2003b).
Often, impalefection appears to result in direct nuclear delivery, particularly for suspended or unattached mammalian cells with large nuclear cross-sections as compared to the overall cross sectional area of the cell.
Although DNA has been successfully delivered to mammalian cells using vertically aligned carbon nanofibers, the details of this process are still being investigated. One
69 element of this process that has not been previously addressed is a quantitative determination of the amount of DNA immobilized on the nanofiber platform. In previous work, plasmid DNA was tethered to impalefection arrays using a carbodiimide mediated condensation reaction between DNA base amines and carboxylic acid sites on nanofibers
(McKnight et al. 2003b; McKnight et al. 2004b). This approach has been effectively used for DNA immobilization onto a variety of substrates, including polymers (Taira and
Yokoyama 2004), microwells (Rasmussen et al. 1991), glass slides (Zammatteo et al.
2000), glass beads (Walsh et al. 2001), gold electrodes (Ge et al. 2003), carbon electrodes
(Millan et al. 1992), carbon nanotubes (Dwyer et al. 2002; Nguyen et al. 2002) and carbon nanofibers (McKnight et al. 2003b). Millan demonstrated that when using 1- ethyl-3-(3-dimethylaminopropyl)-carbodiimide (EDC) for DNA immobilization to carbon, the amide bond is formed specifically at exposed guanine and cytosine amines
(Millan et al. 1992). Since DNA bound to carboxyl groups at these binding sites cannot be effectively directed, binding may occur at undesirable locations and might render the tethered DNA template transcriptionally-inactive.
Covalent binding within or near the promoter region may also limit or eliminate transcription initiation due to steric hindrance of polymerase and/or transcription factor binding. In the eukaryotic cell, transcription initiation is similarly impacted by steric hindrances imposed by ionic interactions between histones and DNA template, with, for example, acetylation of lysine residues of these chromatin proteins being a significant regulatory mechanism of gene expression (Latchman 2004). Binding within the coding sequence may also interfere with reading and complementation of the base during mRNA
70 elongation. Schaffer et al. found that the disassociation of noncovalent interactions between polycations and DNA was required for efficient expression of the DNA by transcriptional machinery (Schaffer et al. 2000). In either case, the transcriptional activity of template can be significantly reduced, if not completely hindered by the interaction of the template with large molecules. Therefore, tethered gene strategies, where solid scaffolding can present even larger hindrances, may present an additional variable with respect to evaluation of efficiency (i.e. transgene expression per unit DNA).
Having knowledge of the amount of transcriptionally-active DNA that is successfully being delivered to cells will be helpful for future studies where the number of gene copies per cell will play a significant role. In this manuscript, using impalefection-based nanofiber arrays as our substrate, we explore cell-free methods to obtain a first order approximation of the accessibility and transcriptional activity of immobilized DNA template by correlating the yields of PCR, quantitative PCR and in-vitro transcription
(IVT) against that of unbound, free DNA.
Materials and Methods
Synthesis of Vertically Aligned Carbon Nanofiber Arrays
Arrays of vertically-aligned carbon nanofibers were fabricated as previously described
(Melechko et al. 2003). In brief, 100 mm silicon wafers (n-type, <100>) were spun with photoresist [SPR 955 CM 0.7] and patterned with 500 nm diameter holes on a 2.5 μm
71 pitch using projection photolithography [GCA Autostep 200] and development
[Microposit, CD26]. A 30 second reactive ion etch (RIE) [Trion Oracle, 150 torr, 50 sccm oxygen, and 150 W of power] in oxygen plasma was used to remove residual resist from the developed regions. A Ni layer [500 Å] was deposited onto a wafer using electron-gun evaporation at 10-6 torr. The excess metallization and photoresist was lifted off using a 1 hour soak in acetone, followed by rinsing in a spray of acetone, followed by
2-propanol. Vertically-aligned carbon nanofibers were then grown from the patterned Ni dots using dc catalytic plasma enhanced chemical vapor deposition (C-PECVD)
(Melechko et al. 2003). Typically, nanofibers were grown to a length of approximately 7
μm with a conical shape featuring a tip diameter of < 100 nm and a base diameter of 200-
300 nm. Following synthesis, wafers were coated with a protective layer of photoresist
(SPR 220 CM 7.0) and cut with a dicing saw into sized pieces. Prior to use, each piece was cleaned of the protective photoresist with a 30 minute soak in acetone, followed by rinse in acetone, 2-propanol, and water.
Covalent Attachment of DNA to Carboxylated Microspheres and Vertically Aligned
Carbon Nanofiber Substrate
The DNA vector pd2EYFP-N1 (BD Biosciences) was transformed into DH5α E.coli cells and isolated using the Wizard Plus Miniprep Kit (Promega). DNA was covalently attached to carboxylated beads using 100 ng of pd2EYFP-N1 vector (BD Biosciences) placed in 200 uL of 100mM 2-[N-morpholino]ethane sulfonic acid buffer (MES) (pH
72 4.7) containing 1 mg of 1-ethyl-3-(3-dimethylaminopropyl)-carbodiimide (EDC). 1.0 μm
polybead carboxylate microspheres (Polysciences, Inc., Warrington, PA) were added at a
final concentration of 2.28 x 104 particles ml-1. For VACNFs, 100 ng of pd2EYFP-N1
were placed in 200 uL of 100mM 2-[N-morpholino]ethane sulfonic acid buffer (MES)
(pH 4.7) containing 1 mg of EDC along with one carbon nanofiber chip (2 mm x 2 mm
dimensions). The carboxylate microspheres and nanofiber chips were incubated for 18
hours at 25°C on an orbital shaker overnight to condense primary amines of the DNA to
the carboxylic acid sites of the solid substrates (Dwyer et al. 2002; McKnight et al.
2003b). The carboxylate microspheres and nanofiber chips were then incubated in 1 ml
of HPLC grade sterile H2O for 15 minutes followed by vortexing. This washing step was
repeated in 1 ml of 2M NaCl in PBS and once again in 1 ml of HPLC grade sterile H2O.
This stepwise wash procedure was elected due to heterogeneity of the silicon and
VACNF surfaces. This heterogeneity presents difficulty with effectively removing nonspecifically bound DNA during washing. Oxidized silicon behaves as conventional glass or glass-milk, requiring low salt solutions for DNA elution. The heterogeneous charge and high surface area of carbon nanofibers appears to present both ionic as well as physical adsorption. Thus, conventional approaches at optimizing salt concentrations are confounded. Our methods were tested using three different salt concentrations, and a sequential water/2M NaCl/water was selected for this study. For carboxylate microspheres, the samples were centrifuged after each wash in 1.5 ml microcentrifuge tubes (Eppendorf) for 2 min. at 16,000 x g in order to remove the microspheres from suspension. Carboxylate microspheres and carbon nanofiber chips were also prepared in the absence of EDC as a negative control. Sham silicon chips, without nanofibers, were
73 also prepared with and without EDC as additional negative controls. These silicon chips without fibers result on each wafer in areas without Ni catalyst sites, as defined by the
pregrowth Ni photolithographic patterning step. Without Ni catalyst, these regions are unpopulated with nanofibers, but remain otherwise exposed to all other processing steps, including high temperature plasma, photoresist protection and acetone stripping.
Primer and Probe Design for PCR and Quantitative PCR
Three sets of primers (Sigma Genosys, St. Louis, MO) were constructed specifically for qualitative determination of amplifiable regions of the pd2EYFP-N1 vector when tethered to the carboxylate microspheres or carbon nanofiber chips (Table 4). This 4.9 kb plasmid contains a CMV promoter upstream of the enhanced yellow fluorescent protein gene (eYFP). The forward primer (YFP F1) was identical for all three primer sets. The three reverse primers (YFP R1, YFP R2, YFP R3) were made to amplify 1) the CMV promoter region (amplicon = 648 bp), 2) the CMV promoter and eYFP gene (amplicon =
1603 bp) and 3) a large fragment containing over 80% of the plasmid length (amplicon =
4053 bp). Additionally, two primer and probe sets (Biosearch Technologies, Novato,
CA) were designed for quantitative determination of the pd2EYFP-N1 vector when tethered to the carbon nanofiber chips. The first primer and probe set (YFP FQ, YFP RQ,
YFP Probe1) amplifies a 72 bp region internal to the CMV promoter upstream of the
EYFP gene. The second primer and probe set (YFP FV, YFP RV, YFP Probe2)
74 Table 4. Primers and probes used in this study.
Name Sequence Area of Amplicon Amplification Size YFP F1 5’-CCT GAT TCT GTG GAT AAC CGT AT-3’ CMV Promoter 648 bp YFP R1 5’-ATC TGA GTC CGG TAG CGC TA-3’
YFP F1 5’-CCT GAT TCT GTG GAT AAC CGT AT-3’ CMV Promoter 1603 bp YFP R2 5’-AAA TGT GGT ATG GCT GAT TAT GAT C-3’ and eYFP Gene
YFP F1 5’-CCT GAT TCT GTG GAT AAC CGT AT-3’ 80% of Plasmid 4053 bp YFP R3 5’-TAT ATA TGA GTA ACC TGA GGC TAT G-3’ Length
YFP FQ 5’-CAC CAA AAT CAA CG-3’ Internal to CMV 72 bp YFP RQ 5’-ACG CCT ACC GCC CAT TT-3’ Promoter YFP Probe1 5’-6-FAM d(AAT GTC GTA ACA ACT CCG CCC CA)BHQ-1-3’
YFP FV 5’-CAA TTA GTC AGC AAC CAG G TG TG-3’ Internal to SV40 97 bp YFP RV 5’-CGG GAC TAT GGT TGC TGA CTA A-3’ Promoter YFP Probe2 5’-6-FAM d(CAG GCT CCC CAG CAG GCA GAA GTA T)BHQ-1-3’
75 amplifies a 97 bp region of the SV40 promoter, 200 bp upstream of the kanamycin
resistance gene and 938 bp downstream of the CMV promoter and EYFP gene.
PCR Amplification of pd2EYFP-N1 covalently-bound to Carboxylate Microspheres and
Vertically Aligned Carbon Nanofibers
The PCR reaction for the three amplicons was optimized by using temperature gradient
protocols (45-60°C), different concentrations of primers and template DNA, as well as the
Invitrogen PCR Optimization Kit. The final mix of each PCR tube contained 1 Ready-
To-Go PCR Bead (Amersham Biosciences), 400 nM forward primer, 400 nM reverse
primer, and HPLC grade sterile H2O. For PCR amplification, the PTC-200 Peltier
Thermal Cycler (MJ Research) was used. The optimal temperature for all primer sets
was 56°C. Negative controls in the presence and absence of EDC were performed
in duplicate via PCR amplification of the 648 bp CMV promoter region using chips
selected from regions of the silicon growth wafer that were unpopulated with carbon
nanofibers. Negative control chips generated no detectable bands, thereby indicating that
DNA binding is occurring to nanofiber sites and not the underlying Si substrate (data not shown). For positive controls of pd2EYFP-N1, 10 ng of the vector was added to each reaction tube. For amplification of pd2EYFP-N1 on carboxylate microspheres, 5 ul of the 2.28 x 104 particles ml-1 microsphere suspension were added to the reaction tube. For
amplification of pd2EYFP-N1 on a carbon nanofiber chip, the chip was placed directly
into the reaction tube containing the Ready-To-Go PCR Bead, primers and HPLC grade
76 sterile H2O. The final volume for all reaction tubes was 25 uL. For gel electrophoresis, a
1% agarose gel of molecular biology grade agarose (Fisher Science) was run at 70 volts
for 45 minutes. For the DNA ladder mix, 1 KB+ Ladder (Invitrogen) was mixed with
HPLC grade sterile H2O and 5X bromophenol blue.
In Vitro Transcription of pd2EYPF-N1 covalently tethered to Vertically Aligned Carbon
Nanofibers
In vitro transcription was performed using the HeLa Scribe Kit (Promega), following the
protocol from the manufacturer. Briefly, each nanofiber chip was added to 10 mM rNTP
mix, 50 mM MgCl2, 20 mM HEPES (pH 7.9 at 25°C), 100 mM KCl, 0.2 mM EDTA, 0.5
mM DTT, 20% glycerol and HeLa nuclear extract and incubated for 1 hour at 30°C.
After the reaction was terminated by the addition of 0.3 M Tris-HCl (pH 7.4 at 25°C),
0.3M sodium acetate, 0.5% SDS, 2 mM EDTA and 3µg/ml tRNA, the generated RNA
was isolated using phenol chloroform extraction as described in the HeLa Scribe Kit.
Samples were stored at -80°C. Quantification of RNA in each sample was determined
using the Ribogreen RNA Quantitation Kit (Molecular Probes, R-11490). Samples were
measured in triplicate on the VersaFluor Fluorometer (BioRad) using VersaFluor
Cuvettes (BioRad).
77 Quantitative PCR of pd2EYFP-N1 on Vertically Aligned Carbon Nanofibers.
DNA was immobilized on the VACNFs as described above. Once pd2EYFP-N1 had been tethered to the carbon nanofiber arrays, the YFP F1 and YFP R2 primer set was used to amplify the CMV promoter and EYFP gene for three cycles of PCR. Volumes containing carbon nanofiber arrays cannot be measured directly in quantitative PCR reactions due to the opacity of the nanofiber substrate. This initial step allowed amplification of pd2EYFP-N1 that was present on the carbon nanofibers to be amplified into solution, such that the chip could be removed for subsequent q-PCR. Primers designed to amplify the CMV promoter and EYFP gene region were used for these initial
PCR cycles because the promoter and gene region are what need to remain accessible to the cellular machinery following impalefection. After the initial three rounds of PCR, the nanofiber chips were removed from the tube and a 5 µl aliquot of the PCR solution was then added to the quantitative PCR mix in place of the template DNA. The quantitative PCR mix consisted of 400 nM of each primer (YFP FQ, YFP RQ), 600 nM of probe (YFP Probe1), and 1X Taq PCR Master Mix (Qiagen). The pd2eYFP-N1 plasmid was run as a standard in known amounts using 10-fold dilutions to generate a standard curve. Positive controls of known amounts of template were run against the standard curve to verify the accuracy of this technique. Quantitative real time PCR assays were performed on the DNA Engine Opticon system (MJ Research, Watertown,
MA).
78 Results and Discussion
PCR Amplification on Vertically Aligned Carbon Nanofibers
PCR was used to determine the presence or absence as well as the accessibility of DNA on the nanofiber arrays after extensive washing steps. Figure 6a shows that the CMV promoter region of pd2EYFP-N1 can be amplified when the DNA vector is putatively covalently tethered to nanofibers (lane 6). Light bands can also be seen in the lane 5, where no EDC was present for covalent attachment, indicating retention of non- specifically adsorbed DNA on the nanofibers. In Figure 6b, it is shown that the larger region of the CMV promoter and EYFP gene sequence of pd2EYFP-N1 is also accessible for amplification after EDC condensation and extensive washing of the nanofiber chip
(lane 6). It is worth noting that the nanofiber chips incubated with DNA in the absence of
EDC showed undetectable levels of amplifiable CMV promoter and EYFP gene (lane 5).
Figure 6c shows the results of using primers (YFP F1, YFP R3) for amplification of approximately 80% of the pd2EYFP-N1 vector sequence. No detectable levels of this length of the vector could be seen from the nanofiber arrays with DNA incubated in the presence or absence of EDC (lanes 5 and 6). Similar results were observed with the carboxylated microspheres using all three primer sets in Figure 6a, b and c (lanes 3 and
4).
79
Figure 6. Amplification of pd2EYFP-N1 on vertically aligned carbon nanofiber arrays. Amplified regions of pd2EYFP-N1 on both carbon nanofiber arrays and carboxylated latex beads using different primer sets. Amplification of A) the CMV promoter region (648 bp), B) the CMV promoter and EYFP gene region (1603 bp) and C)
80% of the vector length (4051 bp) of pd2EYFP-N1. Lanes are identical for each PCR gel. LANES: 1) 1 KB+ Ladder (Invitrogen), 2) 10 ng of non-tethered DNA, 3)
Carboxylated microspheres without EDC, 4) Carboxylated microspheres with EDC, 5)
Carbon nanofiber chip without EDC, 6) Carbon nanofiber chip with EDC, 7) non- tethered DNA without primers, 8) Primers without DNA.
80
Figure 6. (continued…)
81 In Vitro Transcription on Vertically Aligned Carbon Nanofibers
Regardless of the evidence that DNA immobilized on carbon nanofibers is available for amplification, this does not mean that the DNA sequence of interest is also available for transcription in the cell. To evaluate the transcriptional efficiency of pd2EYFP-N1 covalently bound to the nanofiber arrays, in vitro transcription was performed. Table 2 shows a comparison of in vitro transcription of pd2EYFP-N1 in solution and on nanofibers in the presence or absence of EDC. There was no significant difference between the transcription of RNA from nanofiber arrays bound with pd2EYFP-N1 in the presence of EDC (135 ± 44µg/µl) and the transcription of RNA from pd2EYFP-N1 in solution (177 ± 40 µg/µl). Conversely, the transcriptional levels of RNA from pd2EYFP-
N1 on nanofibers bound in the absence of EDC were significantly lower (70 ± 4 µg/µl).
These data are consistent with the results from the PCR assays. Likewise, these results confirm previously reported data that tethered DNA is available to the transcriptional machinery of mammalian cells and can be expressed by these cells when introduced by means of impalefection (McKnight et al. 2003b; McKnight et al. 2004b).
Quantitative PCR Amplification on Vertically Aligned Carbon Nanofibers
Figure 6 indicates that pd2EYFP-N1 is covalently tethered to the carbon nanofiber arrays when incubated in the presence of EDC. However, similar to a previous examination of the accessibility of immobilized DNA using polymerase chain reaction, the amount of immobilized DNA cannot be accurately estimated via conventional PCR
82 Table 5. Quantification of RNA after in vitro transcription of pd2EYFP-N1 on vertically aligned carbon nanofiber arrays.
Sample RNA (µg/µl) st. dev.
Positive Control (No nanofibers) 176.7 ± 40.3
Nanofibers with EDC 134.8 ± 44.2
Nanofibers without EDC 69.9 ± 4.3
83 (Bulyk et al. 1999). These data only provide a qualitative indication of DNA present on
the nanofibers. In order to assess how much DNA (e.g. copies of accessible CMV
promoter and EYFP genes) remains on the nanofibers, we used quantitative PCR
methods coupled to an initial three cycles of PCR as illustrated in Figure 7 and described in the methods. These results are presented in Figure 8a. Amplification yields of the
CMV promoter and EYFP gene region of tethered DNA were quantified and compared to controls which were incubated in the absence of EDC. During the first round of quantification, approximately 1.6 x 109 gene copies were amplified from each nanofiber chip in the presence of EDC, while 7.9 x 108 gene copies were amplified from nanofiber
chips in the absence of EDC. In subsequent reactions of the same nanofiber chips, DNA
yields decreased dramatically (1.2 x 106 gene copies) on chips without EDC, while chips
incubated in the presence of EDC retained DNA (5.6 x 108 gene copies) up to 10 days
and 3 qPCR reaction cycles after initial covalent attachment. The subsequent decrease in
gene copies from these samples suggests that non-tethered DNA can remain non-
specifically adsorbed to the nanofibers during washing steps, but can be removed during thermal cycling.
The results from Figure 8a also suggest that some portion of the pd2EYFP-N1 measured was the result of unbound plasmid, not amplification product. Therefore, a set of primers and probe were designed which bound external to the CMV promoter and EYFP gene region that was being amplified during the three initial PCR cycles illustrated in Figure 7.
This method was used to determine if the DNA measurements from Figure 8a were in fact products of the preliminary PCR amplification, or of pd2EYFP-N1 plasmid released
84
Figure 7. Illustration of the methods used for quantification of accessible DNA bound to the nanofiber arrays.
85
Figure 8. Quantitative PCR of pd2EYFP-N1 on vertically aligned carbon nanofiber
arrays. These assays were used to determine how much DNA remains on the carbon
nanofibers over time in the presence or absence of EDC. The subsequent decrease in
gene copies from carbon nanofibers incubated in the absence of EDC suggests that this
DNA remained non-specifically adsorbed to the nanofibers after the wash steps and was removed during thermal cycling. A) shows the number of EYFP gene copies amplified from the nanofiber using the internal primer and probe set (YFP FQ, YFP RQ and YFP
Probe1) and B) shows the number of DNA copies released from the nanofiber arrays into solution and amplified using the external primer and probe set (YFP FV, YFP RV and
YFP Probe 2). Error bars denote standard deviation.
86
Figure 8. (continued…)
87 from the nanofibers during the reaction. After these initial PCR cycles, this primer and probe set (YFP FV, YFP RV, YFP Probe2) were used for Taqman assays of real time quantitative PCR on the same samples used in Figure 8a. The results are shown in Figure
8b. During the first reaction, there were high levels of DNA copies for both samples
(EDC and no EDC). These copies of DNA cannot be attributed to the first three PCR cycles, because the primer and probe set used bind to a region of the vector sequence found well outside the CMV promoter and EYFP gene region that were initially amplified. These data would imply that those copies of DNA being quantified are equivalent to the number of pd2EYFP-N1 vectors being entirely released from the nanofiber arrays during the first three cycles of PCR. These copies of the vector were not removed during the extensive washing steps. The subsequent data from reaction 3 are consistent with the data in Figure 8a, showing that no significant amounts of DNA vector were amplified outside of the CMV promoter and EYFP gene region for either sample
(EDC or no EDC), suggesting that the majority of DNA being amplified in reaction 3 remained covalently tethered to the nanofiber arrays.
Nanofibers have been previously used for DNA delivery into Chinese Hamster ovary cells (McKnight et al. 2003b; McKnight et al. 2004b). Penetration and residence of
DNA-modified nanofibers within the nucleus offers numerous possibilities for both gene delivery applications and the fundamental study of gene expression and transcriptional phenomenon. For example, delivery of nanofiber-tethered DNA into a cell can offer a higher level of control over the fate of introduced genes, including the potential to remove these genes from a system after a period of transient expression simply by
88 removing the cells from the nanofiber array. Nuclear delivery of tethered DNA on a parallel basis may also provide more efficient methods for studying the impact of template length and topology on transcriptional activity, which has traditionally been investigated through application of the serial method of microinjection (Harland et al.
1983; Krebs and Dunaway 1996; Weintraub et al. 1986). Similarly, nuclear-penetrant nanofiber arrays might be used for in-cell transcriptional assays based on immobilized- template methods that have been developed for the study of in-vitro transcription using nuclear extracts (Adamson et al. 2003), thereby providing new levels of insight into the fundamental processes of transcription and transcriptional regulation.
In prior work, DNA was typically bound to the nanofiber arrays using EDC condensation. While this method for DNA attachment is rapid and simple, it is also relatively random in where it will attach the amine groups of the individual DNA strand to the carboxyl groups available on the carbon nanofiber. This study shows that simple molecular techniques such as PCR, in vitro transcription and quantitative PCR can be used to efficiently evaluate EDC condensation and other DNA immobilization methods on nanoscale substrates. The PCR results provided a qualitative evaluation of how accessible the DNA remains after condensation. The IVT results were consistent with
PCR, showing that after extensive washing, high levels of DNA remained accessible and could be utilized by both DNA and RNA polymerase. The results of the quantitative
PCR showed that in the presence of EDC, large quantities of accessible DNA remained tethered to the VACNF arrays for at least 10 days and through 3 subsequent qPCR thermal cycling sessions. While 5.6 x 108 gene copies per chip is only a fraction of the
89 DNA that was originally attached (approximately 2.98 ng of the initial 100 ng), it is still a substantial amount. For example, the VACNF array samples used in this study contained approximately 1 million fibers. Therefore every fiber likely contained more than 500 accessible gene copies.
The PCR results in Figure 6 show that amplification decreases as the amplicon becomes larger. While tethered pd2EYPF-N1 vector on these nanofiber chips is evidenced by the yields of shorter amplicons, long stretches of DNA cannot be amplified, indicating that they are not accessible to the polymerase enzyme during PCR. It is possible that this is due to steric hindrance or stalling of bound RNA and DNA polymerases. Stalling of
RNA polymerase II can occur in vivo during transcription due to DNA lesions (Yu et al.
2003) or nucleotide-specific binding of proteins (Bertin et al. 1992). Likewise, DNA polymerase has been shown to stall during DNA amplification due to stable secondary structures from base repeats (Krasilnikova et al. 1998) or bulky DNA lesions (Yan et al.
2004). Although no studies have examined the integrity of DNA for polymerization or transcription following covalent immobilization on a substrate, it can be assumed that stalling of the polymerase enzyme does occur under these circumstances. This explanation would be consistent with the results of Figure 6, where amplification of the
DNA template decreased as the amplicon became larger. As the amplicon increases in size and more nucleotide bases are required during elongation, there is a higher probability that these base amines will be randomly immobilized to the available carboxyl groups on the VACNFs due to stochastic fluctuations in the helical structure of the DNA. Figure 9 illustrates various binding modalities of DNA with VACNFs. In
90 Figure 9a, the template DNA is initially tethered to the nanofiber, but will be removed during the washing steps or subsequent experimentation. Figure 9b illustrates DNA template that is tethered to the nanofiber, but the gene region is not fully accessible for primer annealing or polymerase binding. In this case, fragments of the gene may have base amines bound to carboxyl sites of the carbon nanofiber, restricting access to the
DNA sequence. Large stretches of DNA (i.e. the large 80% plasmid amplicon of pd2eyfp-n1) may not be amplified or transcribed due to the limited binding to the nanofiber that is required for these sequences to still be accessible. Figure 9c shows the desired scenario, when DNA is covalently bound to the nanofiber in multiple locations, giving it resilience during the experimentation, while retaining an available gene sequence that can be accessed by the cellular machinery.
The results of the real time quantitative PCR show that over a period of 10 days and 3 subsequent thermal cycling sessions, there are three orders of magnitude more DNA copies remaining on the nanofiber chips in the presence of EDC compared to thenanofiber chips incubated without EDC. This is consistent with the qualitative data from the PCR gels, verifying that more accessible DNA is retained on the nanofibers when incubated in the presence of EDC. However, the qPCR results from day 1 show that a large amount of DNA is initially adsorbed to the nanofiber chips without EDC, even after extensive washing. These results were not observed in Figure 6, although they have been confirmed from multiple experimental reactions (data not shown). These
91
Figure 9. Illustration of potential immobilization of DNA on vertically aligned carbon nanofiber arrays. A) The template DNA is weakly tethered to the nanofiber at only one or two binding sites, and will be removed during the washing steps or subsequent experimentation. B) The DNA is tethered to the nanofiber, but the gene region is not fully accessible for primer annealing or polymerase binding. C) This is the desired scenario, where DNA template is covalently bound to the nanofiber in multiple locations, giving it resilience during the experimentation, while retaining an accessible gene sequence.
92 results could be a result of differences in salt concentrations and oligonucleotide content between the PCR and qPCR mixes, both of which have been shown to affect the binding efficiency of DNA (Castelino et al. 2005; Huang et al. 1996).
Other DNA binding strategies are currently being evaluated, including the use of short linear DNA constructs featuring biotin, thiol, and amine terminations. Enhancing the binding strategies could result in fewer DNA molecules being needed and less DNA being released from the nanofiber arrays, as well as yielding a higher percentage of successful delivery and expression inside the cell. It is anticipated that the combination of PCR, IVT, and qPCR assays presented in this manuscript along with the evaluation of engineered linear constructs with well defined site-specific binding properties, will provide insight into steric hindrances and other interfering properties of the solid scaffolding. In addition to their evaluation with nanofiber based DNA delivery systems, it is anticipated that these same analysis methods will prove useful for other nanostructured systems used for DNA delivery.
Acknowledgements
This work was supported in part by the National Institute for Biomedical Imaging and
Bioengineering under assignments 1-R01EB006316-01 and 1-R21EB004066 and through the Laboratory Directed Research and Development funding program of the Oak
Ridge National Laboratory, which is managed for the U.S. Department of Energy by UT-
Battelle, LLC. AVM and MLS acknowledge support from the Material Sciences and
93 Engineering Division Program of the DOE Office of Science under contract DE-AC05-
00OR22725 with UT-Battelle, LLC. A portion of this research was conducted at the
Center for Nanophase Materials Sciences, which is sponsored at Oak Ridge National
Laboratory by the Division of Scientific User Facilities, U.S. Department of Energy.
94 CHAPTER IV
INDUCIBLE RNAI-MEDIATED GENE SILENCING USING VERTICALLY ALIGNED CARBON NANOFIBER GENE DELIVERY ARRAYS
This chapter is a modified version of a paper published with the same title in journal ACS
Nano in January 2008 by David Mann, Timothy McKnight, Jackson McPherson, Peter
Hoyt, Anatoli Melechko, Michael Simpson and Gary Sayler.
Mann D.G.J., McKnight T.E., McPherson J.T., Hoyt P.R., Melechko A.V., Simpson
M.L., Sayler G.S. 2008. Inducible RNA Interference-Mediated Gene Silencing Using
Nanostructured Gene Delivery Arrays. ACS Nano, 2(1): 69-76.
The use of the term “we” in this chapter refers to my co-authors and myself. My primary contributions in this chapter were the following: 1) assisting in the experimental design of the investigations, 2) performing all experiments and assisting in data analysis, 3)
gathering and reviewing the background literature and 4) writing the majority of the
manuscript.
Introduction
Gene delivery to mammalian cells has become a staple of biological studies in the last
few decades. A number of nanostructured platforms have been shown to effectively
95 deliver plasmid DNA into mammalian cells, including nanofibers (McKnight et al.
2003a; McKnight et al. 2004b), nanotubes (Cai et al. 2005b; Pantarotto et al. 2004),
nanorods (Salem et al. 2003) and nanoparticles (Bauer et al. 2004; Zhi et al. 2006).
Recent demonstrations have shown that vertically aligned carbon nanofiber arrays
(VACNFs) grown on a silicon substrate can be used as a parallel gene delivery platform in mammalian cells (McKnight et al. 2003a). In this process, termed impalefection, plasmid DNA is spotted or covalently immobilized on the nanoscale carbon fibers.
Subsequently, mammalian cells are impaled onto the VACNFs, recover, and express the introduced genes. During impalefection, single mammalian cells attach and proliferate on the nanostructured platform and can be tracked using spatial indexing of VACNF arrays (McKnight et al. 2004b). Highly parallel introduction of DNA directly into mammalian cells and facile monitoring of the gene expression within individual cells
over time are distinct advantages of impalefection when compared to traditional
transfection methods. Thus, the VACNF platform also allows for rapid and efficient transient assays of gene expression while tracking individual cells over time.
Additionally, DNA immobilized on VACNFs may minimize the potential for incorporation of foreign genes into the chromosomes of manipulated cells (Mann et al.
2007; McKnight et al. 2003a).
Coupled with nucleotide delivery methods, RNA interference (RNAi) has become a powerful tool for controlling gene expression in many different cell types. First discovered in Caenorhabditis elegans, the RNAi pathway is an evolutionarily conserved cellular mechanism that specifically down-regulates gene products post-transcriptionally
96 (Fire et al. 1998b). The pathway has now been identified in most eukaryotes (Hannon
2002). The RNAi pathway has been extensively studied and well reviewed in the literature (Hammond 2005; Hannon and Rossi 2004; Karagiannis and El-Osta 2005).
RNAi targets long double-stranded RNAs (dsRNAs), which are cleaved into ~21 to 24 base pair small-interfering RNAs (siRNAs) by the RNase III enzyme Dicer (Bernstein et al. 2001b). The siRNAs are coupled with an enzyme complex referred to as the RNA- induced silencing complex (RISC) (Hammond et al. 2000), and effectively target the
RISC complex to complementary mRNA transcripts which are cleaved and degraded by a process called post-transcriptional gene silencing (PTGS) (Zamore et al. 2000).
Although the major function of RNAi seems to be as a viral defense mechanism, siRNAs may also silence transposable elements and repetitive genes to help stabilize the genome
(Novina and Sharp 2004). More recently, the RNAi pathway has been exploited for therapeutic approaches in human diseases (Kim and Rossi 2007), including age-related macular degeneration (Check 2005; McFarland et al. 2004).
Introducing long dsRNAs into mammalian cells can cause a cytotoxic reaction by triggering a type-I interferon response and shutting down genome-wide translation, resulting in cell death (de Veer et al. 2005; Manche et al. 1992). The interferon response can be avoided by introducing siRNAs directly to the cells, resulting in efficient silencing of the target genes (Kariko et al. 2004; Kim et al. 2004). Additionally, DNA vectors expressing siRNA sequences as short inverted repeats (termed short hairpin RNAs, or shRNAs) have been developed and shown to stably down-regulate gene expression of targeted genes (Brummelkamp et al. 2002). These shRNA vectors include designs for
97 constitutive and inducible gene silencing in mammalian cells (Bantounas et al. 2004;
Fewell and Schmitt 2006). For inducible shRNA expression, a tetracycline induction
system is most commonly used (Ito et al. 2006; Kappel et al. 2007; Lin et al. 2004;
Matsukura et al. 2003b) and is well reviewed in the literature (Berens and Hillen 2003;
Ramos et al. 2005).
For tetracycline inducible expression, the RNA polymerase III promoter for the histone 1
(H1) gene can be altered to contain two Tet operator (tetO2) sites, and used to express the
shRNA sequence. TetO2 sites bind the tetracycline repressor protein (TetR), which
represses the transcription of downstream shRNA sequences. However, when
tetracycline is administered to the cells, it binds tightly to TetR generating a
conformational change that displaces TetR from tetO2 sites. With TetR gone, the H1
promoter is no longer repressed, allowing transcription of the shRNA sequence. This
results in generation of “induced” siRNAs that can specifically silence a targeted gene.
Conventional tetracycline inducible approaches involve generating stable cell lines which
express one or more elements of the induction pathway. In addition to a cell line
expressing the protein to be silenced, subclones of this line transgenically expressing the
TetR protein and the shRNA construct are needed. For negative controls, additional lines
expressing non-sense or mismatched shRNA elements are prepared. Positive controls include cells which do not express the repressor protein. The generation and screening of
these stable cell lines is not a trivial process, requiring significant time (4+ weeks) and
effort prior to evaluating the effects of any inducible RNAi on a cell system.
98 In this study, we investigated the application of VACNF arrays as a platform for the rapid assay of tetracycline-inducible RNAi-mediated gene silencing using the cyan fluorescent protein (cfp) gene expressed in Chinese hamster ovary (CHO-K1) cells. Using the parallel gene-delivery capabilities of VACNFs, the yellow fluorescent protein (yfp) gene was simultaneously delivered with an inducible shRNA sequence to allow observation of successfully impalefected cells (Figure 10). Using the VACNF platform for transient assays of gene expression and silencing we followed both nanofiber-mediated gene delivery and tetracycline induction of gene silencing within individual CHO-K1 cells.
Cells were tracked using spatially indexed patterns on VACNF arrays. This is the first demonstration of co-transfection of multiple DNA vectors using carbon nanofibers as a delivery tool, as well as the first manipulation of shRNA-mediated gene silencing on nanostructured architectures.
Materials and Methods
Plasmids, construction and maintenance
Strains and plasmids used in this study are listed in Table S1. Plasmids were maintained in E. coli strain JM109 and stored at -80ºC. E. coli strain JM109 was grown in Luria-
Bertani (LB) broth at 37°C with or without ampicillin (100 µg/ml) or kanamycin (50
µg/ml), depending on the requirements for plasmid maintenance. Plasmids were transformed into chemically competent E. coli strain JM109. Briefly, 2-4 µl of plasmid
99
Figure 10. Illustration of the vertically aligned carbon nanofiber single-cell gene silencing platform. (A) Nanofiber-mediated impalefection was used to deliver the yfp gene and CFP-silencing shRNA sequence into CHO-K1 cells that constitutively express the cfp and tetR genes (CHO-K1-CFP-TetR). TetR represses the expression of the CFP- silencing shRNA in the absence of tetracycline. Once tetracycline is added to the system, it binds TetR, allowing expression of the shRNA and induced silencing of CFP expression. Expression of the yfp gene allows for direct observation of impalefected cells. (B) This VACNF platform allows for monitored silencing in single cells among a non-silenced cell population.
100 DNA (~100 ng/µl) was added to 100 µl of freshly thawed chemically competent E. coli
strain JM109 cells on ice. After 30 minutes, the cells were heat shocked at 42ºC for 45
seconds and immediately placed on ice for 2 minutes. Following transformation, 1 ml of
2XYT medium was added and cells were allowed to recover at 37°C with shaking at 200
rpm for at least 1 hour. Cells were then plated on LB-agar plates with ampicillin or
kanamycin and colonies were selected and screened for the correct insert 18-20 hours
later. Plasmid isolation was performed using Wizard mini- or midi-prep kits (Promega,
Madison, WI). The expression vectors pd2EYFP-N1 containing destabilized enhance
yellow fluorescent protein (yfp) and pd2ECFP-N1 containing destabilized enhanced cyan
fluorescent protein (cfp) were purchased from Clontech (Mountain View, CA). The
expression vector for the tetR gene (pcDNA6/TR) was purchased from Invitrogen
(Carlsbad, CA). pCFPQuiet was constructed using the pENTR/H1/TO vector from
Invitrogen. Using the BLOCK-iT™ Inducible H1 RNAi Entry Vector Kit from
Invitrogen, a 50-bp oligonucleotide (5’-
CACCGCATCAAGGCCAACTTCAAGACGAATCTTGAAGTTGGCCTTGATGC-3’)
containing a complementary sequence to the cfp gene from pd2ECFP-N1 was ligated into
pENTR/H1/TO and resulted in the pCFPQuiet expression vector. This DNA sequence
was selected using the Invitrogen BLOCK-iT™ RNAi Designer website
(https://rnaidesigner.invitrogen.com/sirna/). In addition to the shRNA sequence specific
for the cfp gene sequence, the pCFPQuiet vector also contains the Zeocin resistance gene
for mammalian selection. A negative control expression vector (pLacZQuiet) was
constructed from the pENTR/H1/TO vector which contained a 49-bp oligonucleotide (5’-
101 CACCAAATCGCTGATTTGTGTAGTCGGAGACGACTACACAAATCAGCGA-3’)
containing a complementary sequence to the lacZ gene. This DNA sequence was
included in the BLOCK-iT™ Inducible H1 RNAi Entry Vector Kit. All constructed
vectors were confirmed for the correct DNA sequence using M13 forward and reverse
primers. DNA sequencing was performed with an ABI Big Dye Terminator cycle
sequencing reaction kit on an ABI 3100 DNA sequencer (Perkin-Elmer, Inc., Foster City,
CA) at UT-MBRF (Knoxville, TN).
Cell culture and reagents
Chinese hamster ovary (CHO-K1BH4) cells were maintained in Ham’s F-12 (Gibco-
Invitrogen) and Ham’s F-12K (ATCC, Manassas, VA) media supplemented with 5% fetal bovine serum (FBS), 100 µg/ml of streptomycin, 100 U/ml penicillin and 250 ng/ml amphotericin B under an atmosphere of 5% CO2 at a temperature of 37°C. Cells were
grown to confluence and passed every 4-5 days by trypsinization at a ratio of 1:9. The
selection agents blasticidin-s (5 µg/ml) and Zeocin (300 µg/ml) (Invitrogen, Carlsbad,
CA) were used on appropriate cell lines. Sodium butyrate was used in the media at a final concentration of 2 mM where mentioned below.
102 Cell transfection and generation of stable cell lines
Cells grown in 35 mm wells were transfected with 1 µg of the pd2ECFP-N1 expression vector. All cell transfections were achieved using the Fugene 6 transfection reagent
(Roche Diagnostics, Indianapolis, IN) in accordance with the manufacturer’s instructions.
The cells were then cultured in the presence of G 418 sulfate (EMD Biosciences Inc., San
Diego, CA) and resistant colonies were passed after 14 days of treatment. 20 separate single clones from the population were obtained through isolation by limiting dilution in
96 well plates. The cell lines from each clone were examined for CFP expression by fluorescent microscopy. The brightest fluorescent cell line was chosen (termed CHO-K1-
CFP) and stocks were made and stored at -80°C. CHO-K1-CFP was then used to create the cell line CHO-K1-CFP-TetR by transfecting the cells with 1 µg of pcDNA6/TR. The cells were cultured in the presence of 5 µg/ml blasticidin-S, and a single clone was isolated as described above. The cell line (CHO-K1-CFP-TetR) was examined for TetR expression by RT-PCR and the results were positive (data not shown). CHO-K1-CFP-
TetR was then used to create CHO-K1-CFP-TetR-CFPQuiet by transfecting the cells with 1 µg of pCFPQuiet. These cells were cultured in the presence of 5 µg/ml blasticidin-S and 300 µg/ml of Zeocin, and a single clone was isolated as described above. In similar fashion, the CHO-K1-CFP-CFPQuiet cell line was generated from
CHO-K1-CFP. An additional cell line (CHO-K1-CFP-LacZQuiet) was generated as a negative control to CHO-K1-CFP-CFPQuiet.
103 Multilabel counting analysis and tetracycline induction
Cells were seeded at 1.0 x 105 in 6-well plates and allowed to attach overnight at 37°C in
an incubator with atmosphere of 5% CO2. After 24 hours of growth, tetracycline was
added to cells every 3 hours for a 27 hour time period. At 27 hrs, all wells were
thoroughly washed with PBS three times. 200 µl of Cell Stripper (Mediatech Inc.,
Herndon, VA) were added to each well and allowed to incubate for 30-35 minutes. After
dissociation, cells were resuspended and pipetted into each well of a 96-well flat-bottom
plate. The plate was covered with a Breathe-Easy™ sealing membrane (Fisher Scientific
Co. L.L.C., Pittsburgh, PA) and centrifuged at 400 x g for 15-20 seconds. The multiwell plates were assayed as previously described with slight modifications (Green and Rasko
2002). Briefly, 96-well black clear-bottom plates were read using a Wallac Victor2 1420
Multilabel Counter (Perkin Elmer Life Sciences, Waltham, MA). Absorbance was
measured at 450 nm for 1.0 second, followed by a fluorescence reading using a CFP filter
set with an excitation of 436 nm and emission of 480 nm (Chroma Technology Corp.,
Rockingham, VT). Readings were taken 8 mm from the bottom of the plates with a 0.5 second measuring time. The lamp control was on a stabilized energy setting, and all reads were obtained at 35,000 volts of lamp energy. The relative fluorescence units described in this study were determined as follows: Fluorescence in each cell line could not be measured on the plate reader from monolayers at optimal levels of confluence (60-
80%), due to the low level of CFP fluorescence. Likewise, the cell fluorescence could not accurately be determined at higher levels of confluency (~100%), due to changes in
104 cellular expression that resulted in lower fluorescence of the stably inserted cfp. Due to these limitations and variations, cells were maintained at 50-60% confluence in each well for the tetracycline induction experiments, and prepared as described above. Standard curves of cell number versus absorbance were generated for each individual cell line between 1 x 104 and 1 x 106 cells, and the slope from each of these standard curves was used to extrapolate the estimated number of cells in each well based on the absorbance measured. Additionally, the fluorescence was measured for each well, and normalized using the absorbance to determine the estimated fluorescence per cell, measured in relative fluorescent units (RFU).
Fluorescence microscopy and cell counting
Cells were observed and counted on chips using a Nikon Diaphot 300 or Nikon Eclipse
TE300 inverted fluorescent microscope, depending on the experiment. Filter sets on both microscopes for CFP (436 nm excitation, 480 nm emission) and YFP (500 nm excitation,
535 nm emission) were purchased from Chroma Technology Corp. (Rockingham, VT;
Cat. No. 31044v2 and 41028). Digital imaging was performed using a MicroPublisher
3.3 CCD camera integrated with QCapture 2.60 imaging software (QImaging Corp.,
Surrey, BC, Canada), and a Nikon Coolpix 4500 digital camera with a Scopetronix
MaxView Plus attachment on the microscope eyepiece lens. For the QImaging system, all images of CFP were obtained using an exposure time of 5 seconds. Bright field images were obtained using an exposure time of 6.49 milliseconds. For the Nikon
105 Coolpix system, all images of CFP were obtained using an exposure time of 2 seconds.
Images of YFP were obtained using exposure times between 2 and 8 seconds, due to the varying levels of YFP fluorescence expressed by the cells. This variation did not affect the statistical data of CFP expression, since YFP was only used as a biomarker of gene delivery. For population counting of tetracycline-induction of shRNA silencing in CHO-
K1-CFP-TetR cells, ten replicate VACNF array chips were used for each variable. For each chip, four images were taken and the total YFP-expressing cells were counted by hand. Each YFP-expressing cell was then confirmed for the presence or absence of CFP.
The presence of CFP was defined as any fluorescence that could be detected in the CFP filter image. At minimum, 250 cells were counted for each variable.
Nanofiber-mediated impalefection
Cells were trypsinized and resuspended in 5 ml of Ham’s F12K media. Cells were centrifuged at 300 x g for 10 minutes. The media was aspirated and the cell pellet was resuspended in 0.2-1.0 ml, depending on the size of the pellet and the specific application. For each test sample, 100 microliters of the cell suspension was allowed to settle for 5 minutes on a 5 mm x 5 mm pad of clean room wipes (Durx 770, Berkshire
Corp) in multiple wells of a 96-well plate. Once the cells were settled, the nanofiber chip was slowly lowered face down and lightly placed on top of the settled cells. The chip was then sharply tapped with flame-sterilized tweezers, and placed face down in 200 µl of Ham’s F12K with 2 mM sodium butyrate in a round-bottom well of a 96-well plate.
106 The 96-well plate was placed in an incubator at 37ºC with 5% CO2 and allowed to recover for 24 hours.
Nanofiber Array Synthesis
4” silicon wafers (100) were photolithographically patterned with 50 nm thick Ni thin- films as discrete 500 nm diameter dots at either 5, 10, or 20 μm pitch over the entire surface of the wafer. Nanofibers were synthesized in a custom built DC-PECVD reactor at a temperature of 650 ºC, 10 torr, 2 A, using a mixture of a carbonaceous source gas (acetylene) and an etch gas (ammonia). Growth time was selected to provide fibers ranging from approximately 10 μm to 17 μm tall, with tip diameters of approximately
100 nm. Spatially indexed arrays were subsequently spun with a 2 μm thick layer of SU-
8 2002 (Microchem Corp., Newton, MA) and patterned with a grid and numerical indexing using contact photolithography and development (SU-8 Developer,
Microchem). Following growth and index patterning, wafers were spun in a protective layer of photoresist (Microposit SPR220 CM 7.0, Shipley Corp., Marlborough, MA) and diced into 2.2 mm square pieces. Following dicing, the protective photoresist was removed by soaking in Microposit Remover 1165 (Shipley), followed by copious rinsing in water.
107 Statistical analysis
All data were analyzed using Microsoft Excel. The student t-test was used to determine statistical significance.
Results and Discussion
CFP temporal silencing in CHO-K1 cell populations by tetracycline-inducible shRNAs
Multiple DNA vectors and cell lines were constructed to investigate the tetracycline- inducible gene silencing capacity of shRNAs against the cyan fluorescent protein (CFP) mRNA in CHO-K1 cells (Table 6). The shRNA sequence had high specificity for CFP, and did not silence YFP or GFP. This specificity is due to a two base pair difference within the targeted sequence of these fluorescent variants (data not shown) and is important because the yfp gene was used later in this study as a marker alongside the
CFP-silencing shRNA vector (pCFPQuiet). The specificity of the pCFPQuiet vector in silencing CFP was also tested by comparing it to pLacZQuiet, a shRNA vector targeting the lacZ gene sequence. pCFPQuiet significantly decreased CFP expression levels when transfected into CHO-K1 cells while pLacZQuiet did not (data not shown).
Next, the inducibility of pCFPQuiet was determined. Figure 11a and b show pCFPQuiet is induced in CHO-K1 cells (CHO-K1-CFP-shRNA-TetR) using tetracycline (1 µg/ml)
108 Table 6. DNA vectors and cell lines used in this study.
Vector/Cell Characteristics Source Name pd2EYFP-N1 Mammalian expression vector containing the BD destabilized yellow variant (d2EYFP) of the enhanced Biosciences gfp gene, constitutive CMV promoter, contains a (discontinued) Neomycin G418 resistance gene and pUC origin and Kan resistance for E. coli pd2ECFP-N1 Mammalian expression vector containing the BD destabilized cyan variant (d2ECFP) of the enhanced gfp Biosciences gene, constitutive CMV promoter, contains a Neomycin (discontinued) G418 resistance gene and pUC origin and Kan resistance for E. coli pENTR/H1/TO Mammalian vector for shRNA expression, inducible Invitrogen polymerase III H1/TO promoter, contains the Zeocin resistance gene and pUC origin and Kan resistance for E. coli pcDNA6/TR Mammalian expression vector containing the tetR gene, Invitrogen constitutive CMV promoter, contains the Blasticidin resistance gene and pUC origin and Amp resistance for E. coli pCFPQuiet pENTR/H1/TO vector harboring a 50 bp sequence from This study the cfp gene downstream of the H1/TO promoter pLacZQuiet pENTR/H1/TO vector harboring a 50 bp sequence from This study the lacZ gene downstream of the H1/TO promoter CHO-K1 Derived as a subclone from the parental CHO cell line ATCC initiated from a biopsy of an ovary of an adult Chinese hamster ATCC# CCL-61 CHO-K1-CFP CHO-K1 with stable chromosomal insertion of cfp gene This study from pd2ECFP-N1 vector, phenotype is fluorescent cyan color CHO-K1-CFP- CHO-K1-CFP with stable chromosomal insertion of This study shRNA CFP-silencing shRNA sequence from pCFPQuiet, phenotype is nonfluorescent CHO-K1-CFP- CHO-K1-CFP with stable chromosomal insertion of This study TetR tetR gene from pcDNA6/TR vector, phenotype is fluorescent cyan color CHO-K1-CFP- CHO-K1-CFP-shRNA with stable chromosomal This study shRNA-TetR insertion of tetR gene from pcDNA6/TR vector, phenotype is fluorescent cyan color in the absence of tetracycline CHO-K1-CFP- CHO-K1-CFP with stable chromosomal insertion of This study LacZshRNA pLacZQuiet vector, phenotype is fluorescent cyan color
109
Figure 11. Tetracycline induction of silencing in CHO-K1 cells over time in the
absence (A, B) or presence (C, D) of sodium butyrate (NaB). Cell lines used are represented as: CHO-K1-CFP-TetR-shRNA (■), CHO-K1-CFP-TetR (▲), CHO-K1-
CFP (●) and CHO-K1-CFP-shRNA (♦) in A and C. The average was determined from
five individual replicates at each time point. RFU = Relative Fluorescence Units.
Standard deviation is shown in bars. Measurements were normalized as described in the
Supplementary Information. Fluorescent micrographs of cells in the absence (B) or
presence (D) of NaB are also shown at 0, 12 and 24 time intervals. Bar represents 100
µm. Error bars denote standard deviation.
110
Figure 11. (continued…)
111
Figure 11. (continued…)
112
Figure 11. (continued…)
113 and efficiently silences CFP over time. Data were quantified using multilabel counting and inverted fluorescence microscopy. At 24 hours, approximately 89.8% of the CFP fluorescence was silenced when compared to the negative controls. When tetracycline was removed, CFP expression did not return to wild type levels again until an additional
120 hours after tetracycline removal, which suggests a high level of potency of the siRNAs cleaved from the shRNAs (Figure 12).
In addition, tetracycline induction of shRNA-mediated silencing was carried out in the presence of sodium butyrate (NaB). NaB is a short-chain fatty acid that has been shown to inhibit histone deacetylase activity, and can result in enhancement of specific gene expression in CHO-K1 cells (De Leon Gatti et al. 2007), including genes under the control of the cytomegalovirus immediate early (CMV IE) promoter (Palermo et al.
1991). In this study, NaB was used to enhance the expression of the cfp gene on VACNF arrays and to increase the longevity of transient expression after transfection in mammalian cells (Gorman et al. 1983). NaB temporarily enhanced and prolonged transgene expression resulting in a 4- to 8-fold increase in CFP fluorescence, but did not affect the time constants of CFP-silencing (Figure 11c and d). Therefore, NaB was used during all subsequent impalefection experiments.
Silencing of CFP by tetracycline-inducible shRNAs in cell populations on VACNF arrays
To examine the use of VACNF arrays as a platform for regulated expression of CFP- silencing shRNAs, pCFPQuiet was impalefected into CHO-K1-CFP-TetR cells, along
114
Figure 12. Addition and removal of tetracycline to CHO-K1-CFP-TetR-shRNA cells over time. A) CHO-K1-CFP-TetR-shRNA cells in the absence of tetracycline over time. B) CHO-K1-CFP-TetR-shRNA cells in the presence of tetracycline. Tetracycline
(1 µg/ml) was added at time 0 hours and cells were monitored for CFP silencing. Once
CFP silencing was fully induced at 24 hours, tetracycline was removed from the media and cells were allowed to recover CFP expression. Expression was not fully recovered until time 120 hours, 96 hours after tetracycline had been removed from the media.
115
Figure 12. (continued…)
116 with pd2EYFP-N1, a vector containing the destabilized enhanced yellow fluorescent
protein (yfp) gene. yfp was used as a marker gene, to designate which cells were successfully impaled and received DNA. Codelivery of more than one DNA vector using impalefection had never been addressed prior to this study, thus the optimal DNA concentration of coexpression after impalement were determined. Gene expression was studied by codelivery of either 10 ng or 100 ng of pd2EYFP-N1 and varying amounts
(0.1, 1, 10 and 100 ng) of pd2ECFP-N1. Wild type CHO-K1 cells were impalefected with both plasmids, allowed to recover for 24 hours, and counted using inverted fluorescence microscopy. By comparing the number of cells expressing one or both genes to the total number of impalefected cells, the optimal concentration was established as 100 ng of each vector, resulting in 76.7% ± 3.9% codelivery of both genes (yfp and cfp) (data not shown). This was significantly higher than the co-transfection rate using
10 ng of each DNA plasmid (61.87% ± 10.80%).
While successful impalefections have been seen with DNA concentrations as low as 20 pg and some successful multi-plasmid impalefections were seen with DNA concentrations as low as 1 ng (data not shown), our current study shows that higher DNA concentrations will significantly improve the proportion of multiple-vector impalefected cells when using VACNFs. The optimized DNA concentration of 100 ng each was used for codelivery of pCFPQuiet and pd2EYFP-N1 into CHO-K1-CFP-TetR cells, containing the stably inserted cfp and tetR genes. Ten replicates of VACNF array chips (3 mm x 3 mm) were used for each variable. pCFPQuiet and pd2EYFP-N1 were impalefected into the cells. pLacZQuiet, was used as a negative control. After 24 hours of recovery,
117 tetracycline was added and fluorescent cell counts began (time = 0 hours). Significant levels of silencing (53.1% ± 10.4%) were observed 48 hours after addition of tetracycline
(Figure 13). These data show that VACNF arrays can be used for highly efficient simultaneous delivery of multiple DNA plasmids to mammalian cells. Notably, these levels were not as substantial compared with the results from the multilabel counting method used in Figure 11a and c. This is due to the counting methods used. Using multilabel counting, the estimates of shRNA-mediated silencing were based upon whole population fluorescence measurements. In this experiment, each chip expressing YFP was counted, and then each cell was observed for CFP expression. This method results in a conservative estimate of shRNA-mediated silencing, because a number of the cells exhibiting low level silencing effects and lower levels of CFP expression may still be counted as CFP positive. In addition, some cells were still expressing DNA up to 7 days after impalefection, which is longer than a typical transient transfection will last, suggesting stable chromosome insertions. Importantly, these data show that inducible shRNA vectors can be incorporated into VACNF-mediated gene delivery systems.
Gene silencing by tetracycline-inducible shRNAs monitored in a single cell on indexed
VACNF arrays
We additionally investigated whether tetracycline-inducible shRNA regulation of CFP expression could be monitored in single cells over time. Using indexed VACNF arrays,
CHO-K1-CFP-TetR cells were impalefected with pCFPQuiet and pd2EYFP-N1 (10 ng of each) using VACNF arrays with nanofibers at a 20 µm pitch and an XY indexing pattern.
118
Figure 13. Silencing of vertically aligned carbon nanofiber impalefected cells after tetracycline induction over time. CHO-K1-CFP-TetR cells were impalefected with pd2EYFP-N1 (black), pd2EYFP-N1 and pLacZQuiet (gray), or pd2EYFP-N1 and pCFPQuiet (white). Methods used for calculating cell counts are described in the discussion. The average was determined from ten individual replicates at each time point. Standard deviation is shown in bars. (*) denotes statistical significance (p < 0.05).
119 Controls were performed by impalefecting CHO-K1-CFP-shRNA-TetR with 10 ng each of pCFPQuiet and pd2EYFP-N1 (positive control) and CHO-K1-CFP-TetR with 10 ng of pd2EYFP-N1 (negative control). After 24 hours of recovery, tetracycline was added
(time = 0 hours) and at least 50 individual cell trajectories were photographed and tracked every 24 hours. As expected, single CHO-K1-CFP-TetR cells expressing the shRNA and yfp marker gene were silenced for CFP expression at 24 and 48 hours, while the same cells expressing only the yfp marker gene were not (Figure 14). Figure 14 shows representative images where virtually every CHO-K1-CFP-shRNA-TetR cell was silenced at 24 and 48 hours. Cells appear at different positions in each image, because
CHO-K1 cells adhere to the surface of the substrate, but continue to move over time. The relatively long time interval between pictures and the small plane of view (100µm x
100µm) give the impression that the cells are moving large distances. Some cells are lost during handling of the VACNF chips between incubators and the microscope.
Currently, VACNF array chips are incubated inverted in round-bottom 96-well plates, and require transfer to flat-bottom wells for microscopic observation. Due to the high surface area of the carbon nanofibers and poor diffusion of media, the cells cannot survive on VACNF array chips inverted in flat-bottom wells. Efforts are underway to develop better substrates for cell culture and microscopic observations.
This study is the first to establish the ability with VACNFs to simultaneously introduce multiple foreign genes into a single cell among a population of cells, and track the exogenously controlled expression of genes over time within a 72 hour period. This contrasts with other studies that are limited to observing phenotypic changes in a single
120
Figure 14. Tetracycline induction and tracking of YFP expression and shRNA- mediated silencing of CFP expression in single impalefected cells on indexed vertically aligned carbon nanofiber arrays over time. CHO-K1-CFP-TetR cells were impalefected with a) pd2EYFP-N1, b) pd2EYFP-N1 and pCFPQuiet and c) CHO-K1-
CFP-TetR-shRNA cells were impalefected with pd2EYFP-N1. The top panel shows cells viewed using the CFP filter set; the bottom panel shows cells viewed using the YFP filter set. Tetracycline (1 µg/ml) was added at time 0 hours and cells were monitored for CFP silencing at 0, 24 and 48 hours. White arrows denote the original CFP-expressing cell, and subsequent silencing of CFP expression. Bar represents 50 µm.
121
Figure 14. (continued…)
122
Figure 14. (continued…)
123 cell over time, or a few studies that observed phenotypic changes in a single cell among a
population of genotypically different cells over time (Valiunas et al. 2005). Additionally, impalefection is a rapid method to generate and track stably transfected cells en masse.
Using traditional methods, the generation of stable cell lines alone would require a
minimum of 10-14 days. While microinjection could be used for this same approach, it
would be much more tedious to microinject multiple cells to be tracked over time. In the
current impalefection study, approximately 25-30 discrete, single cell trajectories were
tracked for each VACNF chip (data not shown).
We are currently investigating observations that yfp gene expression occurs as early as 75 minutes after impalefection in CHO-K1 cells. This is remarkable when considering the maturation time for wild type GFP is ~3 hours at room temperature (Rekas et al. 2002).
Even when newer fluorescent protein genes with shorter maturation times are considered
(Tsien 1998), our data suggest that the introduction and expression of the yfp gene in our system is extremely rapid and efficient. We hypothesize that the impalefection process delivers high gene copy numbers directly to the nucleus, similar to microinjection, allowing for nearly immediate transcription and translation. This may be the fastest method available for assaying gene silencing, and could be an important advancement for assays using siRNAs or micro RNAs (miRNA) expressed from shRNA or similar vectors. Other transfection methods have significantly longer periods of time prior to gene expression, due to several factors including the need for internalization of the DNA, localization to the nuclear membrane, decomplexation of the DNA from its carrier, and nuclear import, some or all of which must occur prior to the initiation of transcription. In
124 contrast, we have shown that the fast transcriptional initiation of impalefection, along
with the specific tracking of individual cells and longevity of transient expression in the
presence of NaB, allow for very rapid, and persistent assays of regulated gene expression
in mammalian cells without the generation of stable clones.
Acknowledgements
The authors are grateful to T. Subich, D. Hensley, D. Thomas, S. Randolph, and P.
Fleming for nanofiber fabrication. We thank S. Ripp and J. Fleming for valuable input in
editing the manuscript. This study was supported in part by grant R01EB006316
(NIBIB) and by the Material Sciences and Engineering Division Program of the DOE
Office of Science (DE-AC05-00OR22725) with UT-Battelle, LLC and through the
Laboratory Directed Research and Development funding program of the Oak Ridge
National Laboratory, which is managed for the U.S. Department of Energy by UT-
Battelle, LLC. M.L.S. acknowledges support from the Material Sciences and
Engineering Division Program of the DOE Office of Science. A portion of this research
was conducted at the Center for Nanophase Materials Sciences, which is sponsored at
Oak Ridge National Laboratory by the Division of Scientific User Facilities (DOE).
125 CONCLUSIONS
Through this series of investigations mechanical, chemical and biological parameters of the VACNF-mediated transfection method were optimized for higher efficiency, the transcriptional activity of EDC-immobilized DNA on VACNF arrays was analyzed and quantified, and VACNF arrays were successfully used as nanoscale probes for gene delivery and modulation of gene expression in a single cell as well as populations of cells. Based on the findings of this study, the following conclusions were drawn:
• For mechanical parameters, the optimal impalement substrate was demonstrated
to be plastic flat-bottom well containing a Durx™ 770 filter pad for cells to settle
onto prior to impalement. Carbon nanofibers had a higher overall impalefection
efficiency than silicon nanofibers.
• For chemical parameters, the optimal amount of adsorbed or immobilized DNA
on carbon nanofiber arrays was between 100 ng and 1 µg per chip. Adsorbed
DNA showed significantly higher impalefection efficiencies when compared to
EDC-immobilized DNA.
• For biological parameters, the washing of cells after impalefection significantly
decreased the impalefection efficiency. The impalefection efficiency in HEK293,
HCT-116 and PC12 cells showed no significant difference when compared to the
standard cell line CHO-K1. The impalefection efficiency of A10, Clone 9 and
J774 cells was significantly lower than CHO-K1.
126 • Multiple parameters including DNA precipitation, cell confluency, cell
concentration, and presence or absence of sodium butyrate demonstrated no
significant impact on the efficiency of impalefection in CHO-K1 cells.
• DNA covalently bound to carbon nanofibers can be amplified in PCR reactions.
The CMV promoter and EYFP gene region of tethered pd2eYFP-N1 were readily
amplified, while the longer sequence consisting of 80% of the vector length could
not be amplified from the carbon nanofiber arrays.
• In vitro transcription assay showed that DNA bound to carbon nanofibers can be
accessed by transcriptional machinery such as RNA polymerase.
• Using quantitative real time PCR assays, the promoter and gene region of tethered
pd2eYFP-N1 were easily amplified and the number of DNA copies bound to
carbon nanofibers in the presence of EDC (5.6 x 108 gene copies) were
significantly higher than in the absence of EDC (1.2 x 106 gene copies).
• DNA can adsorb non-specifically to carbon nanofibers in the absence of EDC and
can remain adsorbed even after extensive washing. However, this non-
specifically adsorbed DNA is rapidly removed over time.
• A tetracycline-inducible shRNA vector (pCFPQuiet) can efficiently knockdown
the expression of a stably expressed reporter gene (cfp) in Chinese Hamster Ovary
(CHO-K1-CFP-TetR) cells in less than 24 hours. After removal of tetracycline,
the expression of the cfp gene returns to constitutive levels at 120 hours.
127 • Following impalefection and tetracycline induction of populations of CHO-K1-
CFP-TetR cells, 53.1% ± 10.4% of impalefected cells were fully silenced by the
inducible CFP-silencing shRNA vector.
• The delivery of the pCFPQuiet vector to CHO-K1-CFP-TetR cells on indexed
VACNF arrays via impalefection provides the ability to not only track specific
gene delivery events in discrete, single cells over extended periods of time
without constant observation under the microscope, but it additionally provides
the ability to control the expression of the delivered genes by turning them “on or
off” using tetracycline induction among a population of cells that remain
uninduced.
Using the optimized impalefection method, the reproducibility and efficiency have significantly increased. Furthermore, efficiency is much higher if specific regions of cells on the VACNF chip arrays are counted, as is often done for biolistic bombardment efficiency counts. During the optimization process, very few parameters seemed to have a significant impact on the resulting impalefection efficiency. This result could be due to the nature of the impalefection method. Impalefection is not limited by cell confluency or cell concentration. Impalefection efficiency may potentially increase in cell lines other than CHO-K1 under cell-line specific optimization of parameters. Additional parameters such as nanofiber pitch may also play a role in impalefection efficiency of specific cell lines and should be addressed in future studies. These results, along with the success of impalefection in numerous cell types, demonstrate that nanofiber-mediated impalefection is a versatile means of gene delivery under a broad range of experimental circumstances.
128
PCR, IVT and qPCR have proven to be highly beneficial methods for analyzing the presence and amount of DNA on VACNF arrays. Additionally, these techniques provide a means of efficiently evaluating the site specific activity of tethered DNA constructs and
thus provide a means for optimizing DNA attachment methods such as EDC
condensation or biotin labeling, which are both currently being used in biological
applications. Optimization of current protocols and investigations of other binding
strategies are presently being assessed to create more efficient DNA delivery systems.
The surface chemistry of DNA on nanofibers composed of silicon should be further
characterized in future studies as well, due to the ease of synthesizing vertically aligned
silicon nanofiber arrays. This could result in potentially increased ease of fabrication
with no significant impact on the efficiency of impalefection.
While the initial studies helped characterize the method of impalefection, the RNAi-
mediated gene silencing indexed array platform demonstrated the applicability of the
VACNF impalefection process. In this study, the specific target gene suppression of the
RNAi pathway was combined with the gene delivery platform of VACNFs to investigate
the silencing efficacy in nanostructure-immobilized and spatially indexed cell culture.
VACNF arrays provided a mechanism for extremely rapid expression, silencing, and
tracking in single cells of interest. This VACNF-mediated technique can be used for a
variety of applications as nanoscale probes that provide for gene delivery and modulation
of gene expression in a single cell to overcome limitations and help address new types of
experimental questions in many fields of study.
129
LIST OF REFERENCES
130 Adamson TE, Shore SM, Price DH. 2003. Analysis of RNA polymerase II elongation in
vitro. Methods in Enzymology 371:264-275.
Ajore R, Kumar R, Kaur I, Sobti RC, Bharadwaj LM. 2007. DNA immobilization
chemical interference due to aggregates study by Dip and Drop approach. J
Biochem Biophys Methods 70(5):779-85.
Alivisatos P. 2004. The use of nanocrystals in biological detection. Nature Biotechnology
22(1):47-52.
Allison DP, Bottomley LA, Thundat T, Brown GM, Woychik RP, Schrick JJ, Jacobson
KB, Warmack RJ. 1992. Immobilization of DNA for scanning probe microscopy.
Proc Natl Acad Sci U S A 89(21):10129-33.
An CI, Sawada A, Fukusaki E, Kobayashi A. 2003. A transient RNA interference assay
system using Arabidopsis protoplasts. Bioscience, Biotechnology and
Biochemistry 67:2674-2677.
Andreason GL, Evans GA. 1988. Introduction and expression of DNA molecules in
eukaryotic cells by electroporation. Biotechniques 6:650-660.
Armaleo D, Ye G-N, Klein TM, Shark KB, Sanford JC, Johnston SA. 1990. Biolistic
nuclear transformation of Saccharomyces cerevisiae and other fungi. Current
Genetics 17:97-103.
Bagasra O. 2005. RNAi as an antiviral therapy. Expert Opin Biol Ther 5(11):1463-74.
Baldi L, Hacker DL, Adam M, Wurm FM. 2007. Recombinant protein production by
large-scale transient gene expression in mammalian cells: state of the art and
future perspectives. Biotechnology Letters 29:677-684.
131 Bantounas I, Phylactou LA, Uney JB. 2004. RNA interference and the use of small
interfering RNA to study gene function in mammalian systems. Journal of
Molecular Endocrinology 33(3):545-557.
Batard P, Jordan M, Wurm F. 2001. Transfer of high copy number plasmid into
mammalian cells by calcium phosphate transfection. Gene 270:61-68.
Bauer LA, Birenbaum NS, Meyer GJ. 2004. Biological applications of high aspect ratio
nanoparticles. Journal of Materials Chemistry 14(4):517-526.
Baylor LR, Gardner WL, Yang X, Kasica RJ, Guillorn MA, Blalock B, Cui H, Hensley
DK, Islam S, Lowndes DH and others. 2004. Initial lithography results from the
digital electrostatic e-beam array lithography concept. Journal of Vacuum Science
& Technology B 22(6):3021-3024.
Behr JP, Demeneix B, Loeffler JP, Perez-Mutul J. 1989. Efficient gene transfer into
mammalian primary endocrine cells with lipopolyamine-coated DNA.
Proceedings of the National Academy of Sciences USA 86(18):6982-6986.
Belgorodsky B, Fadeev L, Kolsenik J, Gozin M. 2007. Biodelivery of a fullerene
derivative. Bioconjugate Chemistry 18(4):1095-1100.
Berens C, Hillen W. 2003. Constraints of resistance regulation in bacteria shape TetR for
application in eukaryotes. European Journal of Biochemistry 270:3109-3121.
Berns K, Hijmans EM, Mullenders J, Brummelkamp TR, Velds A, Heimerikx M,
Kerkhoven RM, Madiredjo M, Nijkamp W, Weigelt B and others. 2004. A large-
scale RNAi screen in human cells identifies new components of the p53 pathway.
Nature 428(6981):431-7.
132 Bernstein E, Caudy A, Hammond S, Hannon G. 2001a. Role for a bidentate ribonuclease
in the initiation step of RNA interference. Nature 409(6818):363-366.
Bernstein E, Caudy AA, Hammond SM, Hannon GJ. 2001b. Role for a bidentate
ribonuclease in the initiation step of RNA interference. Nature 409(6818):363-
366.
Bertin J, Sunstrom NA, Jain P, Acheson NH. 1992. Stalling by RNA polymerase II in the
polyomavirus intergenic region is dependent on functional large T antigen.
Virology 189(2):715-724.
Borm PJA. 2002. Particle Toxicology: From Coal Mining to Nanotechnology. Inhalation
Toxicology 14:311-324.
Boskovic BO, Stolojan V, Khan RUA, Haq S, Silva SRP. 2002. Large-area synthesis of
carbon nanofibres at room temperature. Nature Materials 1(3):165-168.
Brown KT, Fleming DG. 1986. Advanced Micropipette Techniques for Cell Physiology.
New York: John Wiley & Sons.
Brummelkamp TR, Bernards R, Agami R. 2002. A System for Stable Expression of Short
Interfering RNAs in Mammalian Cells. Science 296:550-553.
Bulyk ML, Gentalen E, Lockhart DJ, Church GM. 1999. Quantifying DNA-protein
interactions by double-stranded DNA arrays. Nature Biotechnology 17:573-577.
Bunemann H. 1982. Immobilization of denatured DNA to macroporous supports: II.
Steric and kinetic parameters of heterogeneous hybridization reactions. Nucleic
Acids Res 10(22):7181-96.
133 Butler WB. 1984. Preparing nuclei from cells in monolayer cultures suitable for counting
and for following synchronized cells through the cell cycle. Analytical
Biochemistry 141(1):70-73.
Cai D, Doughty CA, Potocky TB, Dufort FJ, Huang Z, Blair D, Kempa K, Ren ZF,
Chiles TC. 2007. Carbon nanotube-mediated delivery of nucleic acids does not
result in non-specific activation of B lymphocytes. Nanotechnology 18:365101
(10pp).
Cai D, Mataraza JM, Qin Z-H, Huang Z, Huang J, Chiles TC, Carnahan D, Kempa K,
Ren Z. 2005a. Highly efficient molecular delivery into mammalian cells using
carbon nanotube spearing. Nature Methods 2:449 - 454.
Cai D, Mataraza JM, Qin ZH, Huang ZP, Huang JY, Chiles TC, Carnahan D, Kempa K,
Ren ZF. 2005b. Highly efficient molecular delivery into mammalian cells using
carbon nanotube spearing. Nature Methods 2(6):449-454.
Callister WDJ. 2003. Materials Science and Engineering an Introduction. New York:
Wiley.
Capecchi MR. 1980. High efficiency transformation by direct microinjection of DNA
into cultured mammalian cells. Cell 22:479-488.
Castelino K, Kannan B, Majumdar A. 2005. Characterization of Grafting Density and
Binding Efficiency of DNA and Proteins on Gold Surfaces. Langmuir 21:1956-
1961.
Cerny J, Feng Y, Yu A, Miyake K, Borgonovo B, Klumperman J, Meldolesi J, McNeil
PL, Kirchhausen T. 2004. The small chemical vacuolin-1 inhibits Ca2+-
134 dependent lysosomal exocytosis but not cell resealing. Embo Reports 5(9):883-
888.
Chan WC, Nie S. 1998. Quantum dot bioconjugates for ultrasensitive nonisotopic
detection. Science 5385:2016-2018.
Check E. 2005. A crucial test. Nature Medicine 11:243–244.
Cheers MS, Ettensohn CA. 2004. Rapid microinjection of fertilized eggs. Methods in
Cell Biology 74:287-310.
Chen JH, Huang ZP, Wang DZ, Yang SX, Li WZ, Wen JG, Ren ZF. 2001.
Electrochemical synthesis of polypyrrole films over each of well-aligned carbon
nanotubes. Synthetic Metals 125(3):289-294.
Chen L, Marmey P, Taylor NJ, Brizard J-P, Espinoza C, D’Cruz P, Huet H, Zhang S,
Kochko Ad, Beachy RN and others. 1998. Expression and inheritance of multiple
transgenes in rice plants. Nature Biotechnology 16:1060-1064.
Chen Q, Dai L. 2001. Three-dimensional micropatterns of well-aligned carbon nanotubes
produced by photolithography. Journal of Nanoscienc and Nanotechnology
1(1):43-47.
Chen Y, Wang ZL, Yin JS, Johnson DJ, Prince RH. 1997. Well-aligned graphitic
nanofibers synthesized by plasma-assisted chemical vapor deposition. Chemical
Physics Letters 272(3-4):178-182.
Cho YK, Kim S, Kim YA, Lim HK, Lee K, Yoon D, Lim G, Pak YE, Ha TH, Kim K.
2004. Characterization of DNA immobilization and subsequent hybridization
using in situ quartz crystal microbalance, fluorescence spectroscopy, and surface
plasmon resonance. J Colloid Interface Sci 278(1):44-52.
135 Chowdhury EH, Akaike T. 2005a. Bio-Functional Inorganic Materials: An Attractive
Branch of Gene-Based Nano-Medicine Delivery for 21st Century. Current Gene
Therapy 5:669-676.
Chowdhury EH, Akaike T. 2005b. A bio-recognition device developed onto nano-
crystals of carbonate apatite for cell-targeted gene delivery. Biotechnology and
Bioengineering 90(4).
Chowdhury EH, Maruyama A, Kano A, Nagaoka M, Kotaka M, Hirose S, Kunou M,
Akaike T. 2006. pH-sensing nano-crystals of carbonate apatite: effects on
intracellular delivery and release of DNA for efficient expression into mammalian
cells. Gene 376(1):87-94.
Christiaens P, Vermeeren V, Wenmackers S, Daenen M, Haenen K, Nesladek M,
vandeVen M, Ameloot M, Michiels L, Wagner P. 2006. EDC-mediated DNA
attachment to nanocrystalline CVD diamond films. Biosensors and Bioelectronics
22:170-177.
Chun AL, Moralez JG, Fenniri H, Webster TJ. 2004. Helical rosette nanotubes: a more
effective orthopedic implant material. Nanotechnology 15:S1-S6.
Cojocaru CS, Senger A, Le Normand F. 2006. A Nucleation and Growth Model of
Vertically-Oriented Carbon Nanofibers or Nanotubes by Plasma-Enhanced
Catalytic Chemical Vapor Deposition. Journal of Nanoscience and
Nanotechnology 6:1331-1338.
Colosimo A, Goncz KK, Holmes AR, Kunzelmann K, Novelli G, Malone RW, Bennett
MJ, Gruenert DC. 2000. Transfer and expression of foreign genes in mammalian
cells. Biotechniques 29(2):314-324.
136 Colvin VL. 2003. The potential environmental impact of engineered nanomaterials.
Nature Biotechnology 21(10):1166-1170.
Courrier HM, Butz N, Vandamme TF. 2002. Pulmonary drug delivery systems: recent
developments and prospects. Critical reviews in therapeutic drug carrier systems
19(4-5):425-498.
Couzin J. 2006. Molecular biology. RNAi safety comes under scrutiny. Science
312(5777):1121.
Crucifix C, Uhring M, Schultz P. 2004. Immobilization of biotinylated DNA on 2-D
streptavidin crystals. J Struct Biol 146(3):441-51.
CSIRO. 2004. Commonwealth Scientific Industrial Research Organization.
www.csiro.au/index.asp?type=mediaRelease&id=particles.
Dalmay T, Horsefield R, Braunstein TH, Baulcombe DC. 2001. SDE3 encodes an RNA
helicase required for post-transcriptional gene silencing in Arabidopsis. The
EMBO Journal 20:2069-2077.
Davis JM, Addison J, Bruch J, Bruyere S, Daniel H, Degueldre G. 1982. Variations in
Cytotoxicity and Mineral Content Between Respirable Mine Dusts from the
Belgium, British, French and German Coalfields. Annals of Occupational
Hygiene 26:541-549.
Dayton WR, Reville WJ, Goll DE, Stromer MH. 1976. A calcium(2+) ion-activated
protease possibly involved in myofibrillar protein turnover. Partial
characterization of the purified enzyme. Biochemistry 15:2159-2167.
De Leon Gatti M, Wlaschin KF, Nissom PM, Yap M, Hu WS. 2007. Comparative
Transcriptional Analysis of Mouse Hybridoma and Recombinant Chinese
137 Hamster Ovary Cells Undergoing Butyrate Treatment. Journal of Bioscience and
Bioengineering 103(1):82-91. de Veer MJ, Sledz CA, Williams BR. 2005. Detection of foreign RNA: Implications for
RNAi. Immunology and Cell Biology 83:224-228.
Dowell-Mesfin NM, Abdul-Karim M-A, Turner AMP, Schanz S, Craighead HG, Roysam
B, Turner JN, Shain W. 2004. Topographically Modified Surfaces Affect
Orientation and Growth of Hippocampal Neurons. Journal of Neural Engineering
1:78-90.
Dwyer C, Guthold M, Falvo M, Washburn S, Superfine R, Erie D. 2002. DNA-
functionalized single-walled carbon nanotubes. Nanotechnology 13:601-604.
Dykxhoorn DM, Lieberman J. 2006. Silencing viral infection. PLoS Med 3(7):e242.
Elbashir S, Harborth J, Lendeckel W, Yalcin A, Weber K, Tuschi T. 2001. Duplexes of
21-nucleotide RNAs mediate RNA interference in cultured mammalian cells.
Nature 411:494-498.
Fechheimer M, Boylan JF, Parker S, Sisken JE, Patel GL, Zimmer SG. 1987.
Transfection of mammalian cells with plasmid DNA by scrape loading and
sonication loading. Proceedings of the National Academy of Sciences, USA
84:8463-8467.
Felgner PL, Gadek TR, Holm M, Roman R, Chan HW, Wenz M. 1987. Lipofection: a
highly efficient, lipid-mediated DNA-transfection procedure. Proceedings of the
National Academy of Sciences, USA 84:7413-7417.
138 Ferrari S, Moro E, Pettenazzo A, Behr JP, Zacchello F, Scarpa M. 1997. ExGen 500 is an
efficient vector for gene delivery to lung epithelial cells in vitro and in vivo. Gene
Therapy 4:1100-1106.
Fewell GD, Schmitt K. 2006. Vector based RNAi approaches for stable, inducible and
genome-wide screens. Drug Discovery Today 11(21-22):975-982.
Feynman RP. There's Plenty of Room at the Bottom; 1959; California Institute of
Technology.
Fire A, Xu S, Montgomery M, Kostas S, Driver S, Mello C. 1998a. Potent and specific
genetic interference by double-stranded RNA in Caenorhabditis elegans. Nature
391(6669):806-811.
Fire A, Xu S, Montgomery MK, Kostas SA, Driver SE, Mello CC. 1998b. Potent and
specific genetic interference by double-stranded RNA in Caenorhabditis elegans.
Nature 391(6669):806.
Fletcher BL, Hullander ED, Melechko AV, McKnight TE, Klein KL, Hensley DK,
Morrell JL, Simpson ML, Doktycz MJ. 2004. Microarrays of biomimetic cells
formed by the controlled synthesis of carbon nanofiber membranes. Nano Letters
4(10):1809-1814.
Fletcher BL, McKnight TE, Melechko AV, Hensley DK, Thomas DK, Ericson MN,
Simpson ML. 2006a. Transfer of flexible arrays of vertically aligned carbon
nanofiber electrodes to temperature-sensitive substrates. Advanced Materials
18(13):1689-+.
139 Fletcher BL, McKnight TE, Melechko AV, Simpson ML, Doktycz MJ. 2006b.
Biochemical functionalization of vertically aligned carbon nanofibres.
Nanotechnology 17(8):2032-2039.
Garman SC, Wurzburg BA, Tarchevskaya SS, Kinet JP, Jardetzky TS. 2000. Structure of
the Fc fragment of human IgE bound to its high-affinity receptor Fc epsilonRI
alpha. Nature 406:259-266.
Ge C, Liao J, Yu W, Gu N. 2003. Electric potential control of DNA immobilization on
gold electrode. Biosens Bioelectron 18(1):53-8.
Geisse S, Henke M. 2005. Large-scale transient transfection of mammalian cells: a
newly emerging attractive option for recombinant protein production. Journal of
Structural and Functional Genomics 6(2-3):165-170.
Giles J. 2003. Nanotechnology: what is there to fear from something so small? Nature
426:750.
Gorman CM, Howard BH, Reeves R. 1983. Expression of recombinant plasmids in
mammalian cells is enhanced by sodium butyrate. Nucleic Acids Research
11(21):7631-7648.
Gossen M, Bujard H. 2002. Studying Gene Function in Eukaryotes by Conditional Gene
Inactivation. Annual Review of Genetics 36:153-173.
Grabarek Z, Gergely J. 1990. Zero-length crosslinking procedure with the use of active
esters. Analytical Biochemistry 185:131-135.
Graessmann M, Graessmann A. 1983. Microinjection of tissue culture cells. Methods in
Enzymology 101:482-492.
140 Graham FL, van der Eb AJ. 1973. Transformation of rat cells by DNA of human
adenovirus 5. Virology 54:536-539.
Green BJ, Rasko JEJ. 2002. Rapid Screening for High-Titer Retroviral Packaging Cell
Lines Using an In Situ Fluorescence Assay. Human Gene Therapy 13:1005-1013.
Green JJ, Chiu E, Leshchiner ES, Shi J, Langer R, Anderson DG. 2007. Electrostatic
ligand coatings of nanoparticles enable ligand-specific gene delivery to human
primary cells. Nano Letters 7(4):874-879.
Guille M. 1999. Microinjection into Xenopus oocytes and embryos. Methods in
Molecular Biology 127:111-123.
Guillorn MA, Melechko AV, Merkulov VI, Ellis ED, Britton CL, Simpson ML, Lowndes
DH, Baylor LR. 2001a. Operation of a gated field emitter using an individual
carbon nanofiber cathode. Applied Physics Letters 79(21):3506-3508.
Guillorn MA, Simpson ML, Bordonaro GJ, Merkulov VI, Baylor LR, Lowndes DH.
2001b. Fabrication of gated cathode structures using an in situ grown vertically
aligned carbon nanofiber as a field emission element. Journal of Vacuum Science
& Technology B 19(2):573-578.
Guillorn MA, Yang X, Melechko AV, Hensley DK, Hale MD, Merkulov VI, Simpson
ML, Baylor LR, Gardner WL, Lowndes DH. 2004. Vertically aligned carbon
nanofiber-based field emission electron sources with an integrated focusing
electrode. Journal of Vacuum Science & Technology B 22(1):35-39.
Hammond SM. 2005. Dicing and slicing: The core machinery of the RNA interference
pathway. FEBS Letters 579(26):5822-5829.
141 Hammond SM, Bernstein E, Beach D, Hannon GJ. 2000. An RNA-directed nuclease
mediates post-transcriptional gene silencing in Drosophila cells. Nature 404:293-
296.
Hannon GJ. 2002. RNA interference. Nature 418(6894):244.
Hannon GJ, Rossi JJ. 2004. Unlocking the potential of the human genome with RNA
interference. Nature 431(7006):371-378.
Hardman R. 2006. A toxicologic review of quantum dots: toxicity depends on
physicochemical and environmental factors. Environmental Health Perspectives
114(2):165-172.
Harland RM, Weintraub H, McKnight SL. 1983. Transcription of DNA injected into
Xenopus oocytes is influenced by template topology. Nature 301:38-43.
Hedley ML, Barman SP. 2004. Microparticle delivery of plasmid DNA to mammalian
cells. Methods in Molecular Biology (Clifton, N.J.) 245:265-286.
Heiser WC. 1994. Gene transfer into mammalian cells by particle bombardment.
Analytical Biochemistry 217:185-196.
Heiser WC. 2004. Gene Delivery to Mammalian Cells. Walker JM, editor. Totowa, New
Jersey: Humana Press. 3-301 p.
Henry TB, Menn F-M, Fleming JT, Wilgus J, Compton RN, Sayler GS. 2007. Attributing
Effects of Aqueous C60 Nano-Aggregates to Tetrahydrofuran Decomposition
Products in Larval Zebrafish by Assessment of Gene Expression. Environmental
Health Perspectives 115(7):1059-1065.
Hermanson GT, Mallia AK, Smith PK. 1992. Immobilized Affinity Ligand Techniques.
San Diego: Harcourt Brace Jovanovich.
142 Hinds BJ, Chopra N, Rantell T, Andrews R, Gavalas V, Bachas LG. 2004. Aligned
Multiwalled Carbon Nanotube Membranes. Science 303(5654):62-65.
Hirayama H, Tamaoka J, Horikoshi K. 1996. Improved immobilization of DNA to
microwell plates for DNA-DNA hybridization. Nucleic Acids Res 24(20):4098-9.
Huang SC, Stump MD, Weiss R, Caldwell KD. 1996. Binding of biotinylated DNA to
streptavidin-coated polystyrene latex: effects of chain length and particle size.
Analytical Biochemistry 237(1):115-122.
Hughes JDH. 1987. The evaluation of current carbon fibres. Journal of Physics D
20(3):276-285.
Ikari A, Nakano M, Ishibashi M, Kawano K, Suketa Y, Harada H, Takagi K. 2003.
Recovery from heat shock injury by activation of Na+-glucose cotransporter in
renal epithelial cells. Biochimica et Biophysica Acta 1643:47-53.
Ito T, Hashimoto Y, Tanaka E, Kan T, Tsunoda S, Sato F, Higashiyama M, Okumura T,
Shimada Y. 2006. An inducible short-hairpin RNA vector against osteopontin
reduces metastatic potential of human esophageal squamous cell carcinoma in
vitro and in vivo. Clinical Cancer Research 12(4):1308-1316.
Jacque JM, Triques K, Stevenson M. 2002. Modulation of HIV-1 replication by RNA
interference. Nature 418(6896):435-8.
Jaenisch R, Mintz B. 1974. Simian Virus 40 DNA Sequences in DNA of Healthly Adult
Mice Derived from Preimplantation Blastocysts Injected with Viral DNA.
Proceedings of the National Academy of Sciences, USA 71(4):1250-1254.
Jaffe LA, Terasaki M. 2004. Quantitative microinjection of oocytes, eggs, and embryos.
Methods in Cell Biology 74:219-242
143 Jansen H, de Boer M, Legtenberg R, Elwenspoek M. 1995. The black silicon method: a
universal method for determining the parameter setting of a fluorine-based
reactive ion etcher in deep silicon trench etching with profile control. Journal of
Micromechanics and Microengineering 5(2):115-120.
Johnston CJ, Finklestein JN, Mercer P, Corson N, Gelein R, Oberdorster G. 2000.
Pulmonary effects induced by ultrafine PTFE particles. Toxicology and Applied
Pharmacology 168(3):208-215.
Kagawa Y, Inoki Y, Endo H. 2001. Gene therapy by mitochondrial transfer. Advanced
Drug Delivery Reviews 49:107-119.
Kalachikov SM, Adarichev BA, Dymshits GM. 1992. [Immobilization of DNA on
microporous membranes using UV-irradiation]. Bioorg Khim 18(1):52-62.
Kappel S, Matthess Y, Zimmer B, Kaufmann M, Strebhardt K. 2007. Tumor inhibition
by genomically integrated inducible RNAi-cassettes. Nucleic Acids Research
34(16):4527-4536.
Karagiannis TC, El-Osta A. 2005. RNA interference and potential therapeutic
applications of short interfering RNAs. Cancer Gene Therapy 12(10):787-795.
Kariko K, Bhuyan P, Capodici J, Weissman D. 2004. Small interfering RNAs mediate
sequence-independent gene suppression and induce immune activation by
signaling through Toll-like receptor 3. Journal of Immunology 172:6545-6549.
Kasim V, Miyagishi M, Taira K. 2004. Control of siRNA expression using the Cre-loxP
recombination system. Nucleic Acids Res 32(7):e66.
144 Kaur G, Lohia A. 2004. Inhibition of gene expression with double strand RNA
interference in Entamoeba histolytica. Biochemical and Biophysical Research
Communications 320:1118-1122.
Kaykas A, Moon RT. 2004. A plasmid-based system for expressing small interfering
RNA libraries in mammalian cells. BMC Cell Biology 5:16-26.
Kennedy S, Wang D, Ruvkun G. 2004. A conserved siRNA-degrading RNase negatively
regulates RNA interference in C. elegans. Nature 427(6975):645-659.
Kidwell MG. 1993. Lateral Transfer in Natural Populations of Eukaryotes. Annual
Review of Genetics 27:235-256.
Kim D-H, Behlke MA, Rose SD, Chang M-S, Choi S, Rossi JJ. 2004. Synthetic dsRNA
Dicer substrates enhance RNAi potency and efficacy. Nature Biotechnology
23:222-226.
Kim DH, Rossi JJ. 2007. Strategies for silencing human disease using RNA interference.
Nature Reviews Genetics 8:173-184.
Kim JS, Kuk E, Yu KN, Kim JH, Park SJ, Lee HJ, Kim SH, Park YK, Park YH, Hwang
CY and others. 2007. Antimicrobial effects of silver nanoparticles.
Nanomedicine-Nanotechnology Biology and Medicine 3(1):95-101.
Kneuer C, Sameti M, Haltner EG, Schiestel T, Schirra H, Schmidt H, Lehr CM. 2000.
Silica nanoparticles modified with aminosilanes as carriers for plasmid DNA.
International Journal of Pharmaceutics 196:257-261.
Knight JD, Adami RC. 2003. Stabilization of DNA utilizing divalent cations and alcohol.
International Journal of Pharmaceutics 264:15-24.
145 Knoblauch M, Hibberd JM, Gray JC, van Bel AJE. 1999. A galinstan expansion
femtosyringe for microinjection of eukaryotic organelles and prokaryotes. Nature
Biotechnology 17:906-909.
Konig R, Chiang CY, Tu BP, Yan SF, Dejesus PD, Romero A, Bergauer T, Orth A,
Krueger U, Zhou Y and others. 2007. A probability-based approach for the
analysis of large-scale RNAi screens. Nat Methods 4(10):847-849.
Krasilnikova MM, Samadashwily GM, Krasilnikov AS, Mirkin SM. 1998. Transcription
through a simple DNA repeat blocks replication elongation. EMBO J
17(17):5095-5102.
Krebs JE, Dunaway M. 1996. DNA Length is a critical parameter for eukaryotic
transcription in vivo. Molecular and Cellular Biology 16(10):5821-5829.
Kukowska-Latallo JF, Bielinska AU, Johnson J, Spindler R, Tomalia DA, Baker JJR.
1996. Efficient transfer of genetic material into mammalian cells using Starburst
polyamidoamine dendrimers. Proceedings of the National Academy of Sciences,
USA 93:4897-4902.
Lam CW, James JT, McCluskey R, Hunter RL. 2004. Pulmonary toxicity of single-wall
carbon nanotubes in mice 7 and 90 days after intratracheal instillation.
Toxicological Sciences 77:126-134.
Larsson C, Rodahl M, Hook F. 2003. Characterization of DNA immobilization and
subsequent hybridization on a 2D arrangement of streptavidin on a biotin-
modified lipid bilayer supported on SiO2. Anal Chem 75(19):5080-7.
Latchman DS. 2004. Eukaryotic transcription factors. London: Elsevier Academic Press.
146 Lau N, Lim L, Weinstein E, Bartel D. 2001. An abundant class of tiny RNAs with
probable regulatory roles in Caenorhabditis elegans. Science 294(5543):858-862.
Lechardeur D, Lukacs GL. 2006. Nucleocytoplasmic Transport of Plasmid DNA: A
Perilous Journey from the Cytoplasm to the Nucleus. Human Gene Therapy
17:882-889.
Lehmann MJ, Sczakiel G. 2005. Spontaneous uptake of biologically active recombinant
DNA by mammalian cells via a slected DNA segment. Gene Therapy 12(5):446-
451.
Lin P. 1971. Methods in Mammalian Embryology. Daniel J, editor. San Francisco, CA:
Freeman Co. 186-214 p.
Lin X, Yang J, Chen J, Gunasekera A, Fesik SW, Shen Y. 2004. Development of a
tightly regulated U6 promoter for shRNA expression. FEBS Letters 577(3):376-
380.
Liu HK, Wang GX, Guo Z, Wang J, Konstantinov K. 2006. Nanomaterials for lithium-
ion rechargeable batteries. Journal of Nanoscience and Nanotechnology 6(1):1-15.
Lockman PR, Oyewumi MO, Koziara JM, Roder KE, Mumper RJ, Allen DD. 2003.
Brain uptake of thiamine-coated nanoparticles. Journal of Controlled Release
93(3):271-282.
Luo D, Saltzman WM. 2000. Synthetic DNA delivery systems. Nature Biotechnology
18:33-37.
Macdougall AJ, Brown JR, Plumbridge TW. 1980. Immobilization of DNA for affinity
chromatography and drug-binding studies. Biochem J 191(3):855-8.
147 Maeda I, Kohara Y, Yamamoto M, Sugimoto A. 2001. Large-scale analysis of gene
function in Caenorhabditis elegans by high-throughput RNAi. Curr Biol
11(3):171-6.
Manche L, Green SR, Schmedt C, Mathews MB. 1992. Interaction between double-
stranded RNA regulators and the protein kinase DAI. Molecular and Cellular
Biology 12:5238-5248.
Mangeot P, Cosset F, Colas P, Mikaelian I. 2004. A universal transgene silencing method
based on RNA interference. Nucleic Acids Res 32(12):e102.
Mann DGJ, McKnight TE, Melechko AV, Simpson ML, Sayler GS. 2007. Quantitative
Analysis of EDC-Condensed DNA on Vertically Aligned Carbon Nanofiber Gene
Delivery Arrays. Biotechnology and Bioengineering 97(4):680-687.
Masciangioli T, Zhang WX. 2003. Environmental technologies at the nanoscale.
Environmental Science and Technology 37:102A-108A.
Matsukura S, Jones P, Takai D. 2003a. Establishment of conditional vectors for hairpin
siRNA knockdowns. Nucleic Acids Res 31(15):e77.
Matsukura S, Jones PA, Takai D. 2003b. Establishment of conditional vectors for hairpin
siRNA knockdowns. Nucleic Acids Research 31(15):e77.
Matsumoto T, Mimura H. 2003. Point x-ray source using graphite nanofibers and its
application to x-ray radiography. Applied Physics Letters 82(10):1637–1639.
Matthess Y, Kappel S, Spankuch B, Zimmer B, Kaufmann M, Strebhardt K. 2005.
Conditional inhibition of cancer cell proliferation by tetracycline-responsive, H1
promoter-driven silencing of PLK1. Oncogene 24:2973-2980.
McAllister AK. 2000. Biolistic Transfection of Neurons. Science STKE 51:PL1.
148 McClellan RO, Brooks AL, Cuddihy RG, Jones RK, Mauderly JL, Wolff RK. 1982.
Inhalation toxicology of diesel exhaust particles. Development in Toxicology and
Environmental Science 10:99-120.
McFarland TJ, Zhang Y, Appukuttan B, Stout JT. 2004. Gene therapy for proliferative
ocular diseases. Expert Opinion on Biological Therapy 4:1053–1058.
McKnight TE, Melechko AV, Austin DW, Sims T, Guillorn MA, Simpson ML. 2004a.
Microarrays of Vertically-Aligned Carbon Nanofiber Electrodes in an Open
Fluidic Channel. Journal of Physical Chemistry B 108:7115-7125.
McKnight TE, Melechko AV, Fletcher BL, Jones SW, Hensley DK, Peckys DB, Griffin
GD, Simpson ML, Ericson MN. 2006a. Resident Neuroelectrochemical
Interfacing Using Carbon Nanofiber Arrays. Journal of Physical Chemistry B
110:15317-15327.
McKnight TE, Melechko AV, Griffin GD, Guillorn MA, Merkulov VI, Serna F, Hensley
DK, Doktycz MJ, Lowndes DH, Simpson ML. 2003a. Intracellular integration of
synthetic nanostructures with viable cells for controlled biochemical
manipulation. Nanotechnology 14(5):551-556.
McKnight TE, Melechko AV, Griffin GD, Guillorn MA, Merkulov VI, Serna F, Hensley
DK, Doktycz MJ, Lowndes DH, Simpson ML. 2003b. Intracellular integration of
synthetic nanostructures with viable cells for controlled biochemical
manipulation. Nanotechnology 14:551-556.
McKnight TE, Melechko AV, Griffin GD, Simpson ML. 2006b. Plant tissue
transformation using periodic arrays of vertically aligned carbon nanofibers. In
Vitro Cellular & Developmental Biology-Animal 42:17A-17A.
149 McKnight TE, Melechko AV, Hensley DK, Mann DGJ, Griffin GD, Simpson ML.
2004b. Tracking Gene Expression after DNA Delivery Using Spatially Indexed
Nanofiber Arrays. Nano Letters 4(7):1213-1219.
McKnight TE, Peeraphatdit C, Jones SW, Fowlkes JD, Fletcher BL, Klein KL, Melechko
AV, Doktycz MJ, Simpson ML. 2006c. Site-specific biochemical
functionalization along the height of vertically aligned carbon nanofiber arrays.
Chemistry of Materials 18(14):3203-3211.
McNeil AK, Rescher U, Gerke V, McNeil PL. 2006. Requirement for annexin A1 in
plasma membrane repair. Journal of Biological Chemistry 281(46):35202-35207.
McNeil PL. 2002. Repairing a torn cell surface: make way, lysosomes to the rescue.
Journal of Cell Science 115(5):873-879.
McNeil PL, Kirchhausen T. 2005. An emergency response team for membrane repair.
Nature Reviews Molecular Cell Biology 6(6):499-505.
McNeil PL, Miyake K, Vogel SS. 2001. Rapid resealing requirement for internal
membrane. Molecular Biology of the Cell 12:75A-75A.
McNeil PL, Miyake K, Vogel SS. 2003. The endomembrane requirement for cell surface
repair. Proceedings of the National Academy of Sciences of the United States of
America 100(8):4592-4597.
McNeil PL, Murphy RF, Lanni F, Taylor DL. 1984. A Method for Incorporating
Macromolcules into Adherent Cells. The Journal of Cell Biology 98:1556-1564.
McNeil PL, Steinhardt RA. 2003. Plasma membrane disruption: Repair, prevention,
adaptation. Annual Review of Cell and Developmental Biology 19:697-731.
150 McNeil PL, Terasaki M. 2001. Coping with the inevitable: how cells repair a torn surface
membrane. Nature Cell Biology 3(5):E124-E129.
Meissner P, Pick H, Kulangara A, Chatellard P, Friedrich K, Wurm FM. 2001. Transient
gene expression: recombinant protein production with suspension-adapted
HEK293-EBNA cells. Biotechnology and Bioengineering 75(2):197-203.
Melechko AV, McKnight TE, Hensley DK, Guillorn MA, Borisevich AY, Merkulov VI,
Lowndes DH, Simpson ML. 2003. Large-scale synthesis of arrays of high-aspect-
ratio rigid vertically aligned carbon nanofibres. Nanotechnology 14(9):1029-
1035.
Melechko AV, Merkulov VI, McKnight TE, Guillorn MA, Klein KL, Lowndes DH,
Simpson ML. 2005. Vertically aligned carbon nanofibers and related structures:
Controlled synthesis and directed assembly. Journal of Applied Physics 97(4).
Mellgren RL, Zhang WL, Miyake K, McNeil PL. 2007. Calpain is required for the rapid,
calcium-dependent repair of wounded plasma membrane. Journal of Biological
Chemistry 282(4):2567-2575.
Merkulov VI, Lowndes DH, Wei YY, Eres G, Voelkl E. 2000. Patterned growth of
individual and multiple vertically aligned carbon nanofibers. Applied Physics
Letters 76(24):3555-3557.
Merkulov VI, Melechko AV, Guillorn MA, Lowndes DH, Simpson ML. 2001.
Alignment mechanism of carbon nanofibers produced by plasma-enhanced
chemical-vapor deposition. Applied Physics Letters 79(18):2970–2972.
Meyyappan M, Delzeit L, Cassell A, Hash D. 2003. Carbon nanotube growth by PECVD:
a review. Plasma Sources Science & Technology 12(2):205-216.
151 Millan KM, Mikkelson SR. 1993. Sequence-selective biosensor for DNA based on
electroactive hybridization indicators. Analytical Chemistry 65(17):2317-2323.
Millan KM, Spurmanis AJ, Mikkelsen SR. 1992. Covalent immobilization of DNA onto
glassy-carbon electrodes. Electroanalysis 4(10):929-932.
Miyake K, Godt RE, Vogel SS, McNeil PL. 2002. Surface tears in mouse skeletal
muscle: Lysosomes to the rescue. Biophysical Journal 82(1):271A-271A.
Miyake K, McNeil PL. 2003. Mechanical injury and repair of cells. Critical Care
Medicine 31(8):S496-S501.
Monteiro-Riviere N, Nemanich R, Inman A, Wang Y, Riviere J. 2005. Multi-walled
carbon nanotube interactions with human epidermal keratinocytes. Toxicology
Letters 155:377-384.
Morozov VN, Morozova TY, Hiort C, Schwartz DC. 1996. New polyacrylamide gel-
based methods of sample preparation for optical microscopy: immobilization of
DNA molecules for optical mapping. J Microsc 183(Pt 3):205-14.
Mourrain P, Beclin C, Elmayan T, Feuerbach F, Godon C, Morel J, Jouette D, Lacombe
A, Nikic S, Picault N. 2000. Arabidopsis SGS2 and SGS3 Genes are Required for
Posttranscriptional Gene Silencing and Natural Virus Resistance. Cell
101(5):533-542.
Muangmoonchai R, Wong SC, Smirlis D, Phillips IR, Shephardl EA. 2002. Transfection
of liver in vivo by biolistic particle delivery: its use in the investigation of
cytochrome P450 gene regulation. Molecular Biotechnology 20(2):145-151.
Mueller C, Graessmann A, Graessmann M. 1980. Microinjection: Turning Living Cells
Into Test Tubes. Trends in Biochemical Sciences 5:60-62.
152 Mykoniatis MG. 1985. Immobilization of native and denatured DNA on Sephadex G200.
J Biochem Biophys Methods 10(5-6):321-8.
Nakajima N, Ikada Y. 1995. Mechanism of Amide Formation by Carbodiimide for
Bioconjugation in Aqueous Media. Bioconjugate Chemistry 6:123-130.
Napoli C, Lemieux C, Jorgensen R. 1990. Introduction of a Chimeric Chalcone Synthase
Gene into Petunia Results in Reversible Co-Suppression of Homologous Genes in
trans. Plant Cell 2:279-289.
Nevoigt E, Fassbender A, Stahl U. 2000. Cells of the yeast Saccharomyces cerevisiae are
transformable by DNA under non-artificial conditions. Yeast 16(12):1107-1110.
Nguyen CV, Delzeit L, Cassell AM, Li J, Han J, Meyyappan M. 2002. Preparation of
Nucleic Acid Functionalized Carbon Nanotube Arrays. Nano Letters 2(10):1079-
1081.
Nie H, Wang CH. 2007. Fabrication and characterization of PLGA/HAp composite
scaffolds for delivery of BMP-2 plasmid DNA. Journal of Controlled Release
120(1-2):111-121.
Nikiforov TT, Rogers YH. 1995. The use of 96-well polystyrene plates for DNA
hybridization-based assays: an evaluation of different approaches to
oligonucleotide immobilization. Anal Biochem 227(1):201-9.
NNI. 2004. National Nanotechnology Initiative.
www.nano.gov/html/facts/appsprod.html.
Nohynek GJ, Lademann J, Ribaud C, Roberts MS. 2007. Grey goo on the skin?
Nanotechnology, cosmetic and sunscreen safety. Critical Reviews in Toxicology
37(3):251-277.
153 Nolan PE, Lynch DC, Cutler AH. 1998. Carbon Deposition and Hydrocarbon Formation
on Group VIII Metal Catalysts. J. Phys. Chem. B 102(21):4165-4175.
Noll T, Eisenkratzer D, Kiesewetter A, Dinter A, Zeng S, Wandrey C, Biselli M, Berger
EG. 2002. Improved product formation in high density Chinese hamster ovary
cell cultures transfected at confluency. Biotechnology Letters 24:861-866.
Novakovic S, Knezevic M, Golouh R, Jezersek B. 1999. Transfection of Mammalian
Cells by the Methods of Receptor Mediated Gene Transfer and Particle
Bombardment. Journal of Experimental and Clinical Cancer Research 18(4):531-
536.
Novina CD, Sharp PA. 2004. The RNAi Revolution. Nature 430:161-164.
NRC. 2002. Small Wonders, Endless Frontiers: A Review of the National
Nanotechnology Initiative. In: Council NR, editor: National Academy Press,
Washington, D.C.
O'Brien J, Lummis SCR. 2004. Biolistic and diolistic transfection: using the gene gun to
deliver DNA and lipophilic dyes into mammalian cells. Methods 33:121-125.
Oberdorster E. 2004. Manufactured Nanomaterials (Fullerenes, C60) Induce Oxidative
Stress in Brain of Juvenile Largemouth Bass. Environmental Health Perspectives
112(10):1058–1062.
Oberdorster G, Galein RN, Ferin J, Weiss B. 1994. Association of Acute Air Pollution
and Acute Mortality: Involvement of Ultrafines? Inhalation Toxicology 7:111-
124.
154 Oberdorster G, Oberdorster E, Oberdorster J. 2005. Nanotoxicology: An Emerging
Discipline Evolving from Studies of Ultrafine Particles. Environmental Health
Perspectives 113(7):823-839.
Oberdorster G, Sharp Z, Atudorei V, Elder A, Gelein R, Kreyling W, Cox C. 2004.
Translocation of inhaled ultrafine particles to the brain. Inhalation Toxicology
16(6-7):437-445.
Oberdorster G, Sharp Z, Atudorei V, Elder A, Gelein R, Lunts A, Kreyling W, Cox C.
2002. Extrapulmonary translocation of ultrafine carbon particles following whole-
body inhalation exposure of rats. Journal of Toxicological and Environmental
Health Part A 65(20):1531-1543.
Okutucu B, Telefoncu A. 2004. Covalent attachment of oligonucleotides to cellulose
acetate membranes. Artif Cells Blood Substit Immobil Biotechnol 32(4):599-608.
OTA. Office of Technology Assessment USC. 1984. Chapter 21: Public Perception. In:
OTA-BA-218, editor. Commercial Biotechnology: An International Analysis.
Washington, D.C. p 489-500.
Ow H, Larson D, Sengupta P, Baird B, Webb W, Wiesner U. 2004. CU-dots: A novel
radiative materials platform for nanobiotechnology. Nanobiotechnology News,
NBTC 5:1-6.
Ow H, Larson DR, Srivastava M, Baird BA, Webb WW, Wiesner U. 2005. Bright and
Stable Core-Shell Fluorescent Silica Nanoparticles. Nano Letters 5(1):113-117.
Paddison PJ, Silva JM, Conklin DS, Schlabach M, Li M, Aruleba S, Balija V,
O'Shaughnessy A, Gnoj L, Scobie K and others. 2004. A resource for large-scale
RNA-interference-based screens in mammals. Nature 428(6981):427-431.
155 Pagano JS, Vaheri A. 1965. Enhancement of infectivity of poliovirus RNA with
diethylaminoethyl-dextran (DEAE-D). Archiv fur die Gesamte Virusforschung
17:456-464.
Pai SI, Lin Y-Y, Macaes B, Meneshian A, Hung C-F, Wu T-C. 2006. Prospects of RNA
interference therapy for cancer. Gene Therapy 13(464-477).
Palermo DP, DeGraaf ME, Marotti KR, Rehberg E, Post LE. 1991. Production of
analytical quantities of recombinant proteins in Chinese hamster ovary cells using
sodium butyrate to elevate gene expression. Journal of Biotechnology 19(1):35-
47.
Pantarotto D, Singh R, McCarthy D, Erhardt M, Briand JP, Prato M, Kostarelos K,
Bianco A. 2004. Functionalized carbon nanotubes for plasmid DNA gene
delivery. Angewandte Chemie-International Edition 43(39):5242-5246.
Paszty C, Lurquin PF. 1987. Improved Plant Protoplast Plating/Selection Technique for
Quantitation of Transformation Frequencies. BioTechniques 5(8):716-718.
Pirio G, Legagneux P, Pribat D, Teo KBK, Chhowalla M, Amaratunga GAJ, Milne WI.
2002. Fabrication and electrical characteristics of carbon nanotube field emission
microcathodes with an integrated gate electrode. Nanotechnology 13(1):1-4.
Pitkethly MJ. 2003. Nanoparticles as building blocks? Nanotoday:36-42.
Poonian MS, Schlabach A, Weissbach A. 1971. Covalent attachment of nucleic acids to
agarose for affinity chromatography. Biochemistry 10:424-427.
Price RL, Waid MC, Haberstroh KM, Webster TJ. 2003. Selective bone cell adhesion on
forumulations containing carbon nanofibers. Biomaterials 23(1877-1887).
156 Proctor GN. 1992a. Microinjection of DNA into mammalian cell in culture: Theory and
practise. Methods in Molecular and Cellular Biology 3:209-231.
Proctor GN. 1992b. Microinjection of DNA into Mammalian Cells in Culture: Theory
and Practice. Methods in Molecular and Cellular Biology 3:209-231.
Proudnikov D, Timofeev E, Mirzabekov A. 1998. Immobilization of DNA in
polyacrylamide gel for the manufacture of DNA and DNA-oligonucleotide
microchips. Anal Biochem 259(1):34-41.
Ramos JL, Martinez-Bueno M, Molina-Henares AJ, Teran W, Watanabe K, Zhang X,
Gallegos MT, Brennan R, Tobes R. 2005. The TetR Family of Transcriptional
Repressors. Microbiology and Molecular Biology Reviews 69(2):326–356.
Ranjan V, Waterbury R, Xiao Z, Diamond SL. 1996. Fluid Shear Stress Induction of the
Transcriptional Activator c-fos in Human and Bovine Endothelial Cells, HeLa,
and Chinese Hamster Ovary Cells. Biotechnology and Bioengineering 49:383-
390.
Raoul C, Barker SD, Aebischer P. 2006. Viral-based modeling and correction of
neurodegenerative diseases by RNA interference. Gene Therapy 13:487-495.
Rasmussen SR, Larsen MR, Rasmussen SE. 1991. Covalent immobilization of DNA onto
polystyrene microwells: the molecules are only bound at the 5' end. Anal
Biochem 198(1):138-42.
Reddy A, Caler EV, Andrews NW. 2001. Plasma membrane repair is mediated by Ca2+-
regulated exocytosis of lysosomes. Cell 106:157-169.
Rekas A, Alattia JR, Nagai T, Miyawaki A, Ikura M. 2002. Crystal Structure of Venus, a
Yellow Fluorescent Protein with Improved Maturation and Reduced
157 Environmental Sensitivity. Journal of Biological Chemistry 277(52):50573-
50578.
Ren ZF, Huang ZP, Xu JW, Wang JH, Bush P, Siegal MP, Provencio PN. 1998.
Synthesis of large arrays of well-aligned carbon nanotubes on glass. Science
282(5391):1105-1107.
Rogers YH, Jiang-Baucom P, Huang ZJ, Bogdanov V, Anderson S, Boyce-Jacino MT.
1999. Immobilization of oligonucleotides onto a glass support via disulfide bonds:
A method for preparation of DNA microarrays. Anal Biochem 266(1):23-30.
Rossi JJ. 2006. RNAi as a treatment for HIV-1 infection. Biotechniques Suppl:25-9.
Rothman JE. 2002. Lasker basic medical research award. The machinery and principles
of vesicle transport in the cell. Nature Medicine 8:1059-1062.
Roy I, Ohulchanskyy TY, Bharali DJ, Pudavar HE, Mistretta RA, Kaur N, Prasad PN.
2005. Optical tracking of organically modified silica nanoparticles as DNA
carriers: A nonviral, nanomedicine approach for gene delivery. Proceedings of the
National Academy of Sciences, USA 102:279-284.
Salem AK, Searson PC, Leong KW. 2003. Multifunctional nanorods for gene delivery.
Nature Materials 2(10):668-671.
Sanford JC, Klein TM, Wolf ED, Allen. N. 1987. Delivery of substances into cells and
tissues using a particle bombardment process. Journal of Particulate Science and
Technology 5:27-37.
Sanford JC, Smith FD, Russell JA. 1993. Optimizing the Biolistic Process for Different
Biological Applications. Methods in Enzymology 217:483-509.
158 Sayes CM, Gobin AM, Ausman KD, Mendez J, West JL, Colvin VL. 2005. Nano-C60
cytotoxicity is due to lipid peroxidation. Biomaterials 26(36):7587-7595.
Schaffer DV, Fidelman NA, Dan N, Lauffenburger DA. 2000. Vector unpacking as a
potential barrier for receptor-mediated polyplex gene delivery. Biotechnology and
Bioengineering 67(5):598-606.
Shaffer C. 2005. Nanomedicine transforms drug delivery. Drug Discovery Today
10:1581-1582.
Shark KB, Smith FD, Harpending PR, Rasmussen JL, Sanford JC. 1991. Biolistic
transformation of a procaryote, Bacillus megaterium. Applied and Environmental
Microbiology 57:480-485.
Shchepinov MS, Esipov DS, Korobko VG, Dobrynin VN. 1994. [Selectively cleaved
synthetic oligodeoxyribonucleotides for reverse immobilization of DNA]. Bioorg
Khim 20(8-9):955-66.
Shi C, Shin YO, Hanson J, Cass B, Loewen MC, Durocher Y. 2005. Purification and
characterization of a recombinant G-protein-coupled receptor, Saccharomyces
cerevisiae Ste2p, transiently expressed in HEK293 EBNA1 cells. Biochemistry
44(48):15705-15714.
Shi H, Djikeng A, Mark T, Wirtz E, Tschudi C, Ullu E. 2000. Genetic interference in
Trypanosoma brucei by heritable and inducible double-stranded RNA. RNA
6(7):1069-1076.
Shvedova AA, Kisin E, Keshava N, Murray AR, Gorelik O, Arepalli S. Cytotoxic and
genotoxic effects of single wall carbon nanotube exposure on human
159 keratinocytes and bronchial epithelial cells; 2004; Anaheim, CA. American
Chemical Society, IEC.
Sigova A, Rhind N, Zamore PD. 2004. A single Argonaute protein mediates both
transcriptional and posttranscriptional silencing in Schizosaccharomyces pombe.
Genes Dev 18:2359-2367.
Smith FD, Harpending PR, Sanford JC. 1992. Biolistic transformation of prokaryotes:
factors that affect biolistic transformation of very small cells. Journal of General
Microbiology 138:239-248.
Steinhardt RA, Bi G, Alderton JM. 1994. Cell Membrane resealing by a vesicular
mechanism simiilar to neurotransmitter release. Science 263:390-393.
Su X. 2002. Covalent DNA immobilization on polymer-shielded silver-coated quartz
crystal microbalance using photobiotin-based UV irradiation. Biochem Biophys
Res Commun 290(3):962-6.
Svoboda P, Stein P, Hayashi H, Schultz R. 2000. Selective reduction of dormant maternal
mRNAs in mouse oocytes by RNA interference. Development 127(19):4147-
4156.
Taira S, Yokoyama K. 2004. DNA-conjugated polymers for self-assembled DNA chip
fabrication. Anal Sci 20(2):267-71.
Taira S, Yokoyama K. 2005. Immobilization of single-stranded DNA by self-assembled
polymer on gold substrate for a DNA chip. Biotechnol Bioeng 89(7):835-8.
Tan K, Cheang P, Ho IA, Lam PY, Hui KM. 2007. Nanosized bioceramic particles could
function as efficient gene delivery vehicles with target specificity for the spleen.
Gene Therapy 14(10):828-835.
160 Tanigawa K, Tanaka K, Nagase H, Miyake H, Kjniwa M, Ikizawa K. 2002. In vivo
analysis of glucocorticoid-induced reporter gene expression using gene gun DNA
delivery. Biological and Pharmaceutical Bulletin 25(8):1115-1118.
Taniguchi N. On the Basic Concept of 'Nano-Technology'; 1974; Tokyo, Japan. Japan
Society of Precision Engineering.
Teo KBK, Chhowalla M, Amaratunga GAJ, Milne WI, Legagneux P, Pirio G, Gangloff
L, Pribat D, Semet V, Binh VT and others. 2003. Fabrication and electrical
characteristics of carbon nanotube-based microcathodes for use in a parallel
electron-beam lithography system. Journal of Vacuum Science & Technology B
21(2):693-697.
Tsien RY. 1998. The Green Fluorescent Protein. Annual Review of Biochemistry 67:509-
544.
Tsuji JS, Maynard AD, Howard PC, James JT, Lam CW, Warheit DB, Santamaria AB.
2006. Research strategies for safety evaluation of nanomaterials, part IV: risk
assessment of nanoparticles. Toxicological Sciences 89(1):42-50.
Uchimura E, Yamada S, Uebersax L, Fujita S, Miyake M, Miyake J. 2007. Method for
reverse transfection using gold colloid as a nano-scaffold. Journal of Bioscience
and Bioengineering 103(1):101-103.
Unger EC, McCreery TP, Sweitzer RH. 1997. Ultrasound enhances gene expression of
liposomal transfection. Investigative Radiology 32:723-772.
Valiunas V, Polosina YY, Miller H, Potapova IA, Valiuniene L, Doronin S, Mathias RT,
Robinson RB, Rosen MR, Cohen IS and others. 2005. Connexin-specific cell-to-
161 cell transfer of short interfering RNA by gap junctions. Journal of Physiology
568(2):459–468.
van der Krol A, Mur L, Beld M, Mol J, Stuitje A. 1990. Flavonoid Genes in Petunia:
Addition of a Limited Number of Gene Copies May Lead to a Suppression of
Gene Expression. Plant Cell 2:291-299.
Volpe TA, Kidner C, Hall IM, Teng G, Grewal SIS, Martienssen RA. 2002. Regulation
of Heterochromatin Silencing and Histone H3 Lysine-9 Methylation by RNAi.
Science 297:1833-1837.
Wagner E, Plank C, Zatloukal K, Cotten M, Birnstiel ML. 1992. Influenza virus
hemaglutinin HA-2 N-terminal fusogenic peptides augment gene transfer by
transferrin-polylysine-DNA complexes: toward a synthetic virus-like gene-
transfer vehicle. Proceedings of the National Academy of Sciences, USA
89:7934-7938.
Walsh MK, Wang X, Weimer BC. 2001. Optimizing the immobilization of single-
stranded DNA onto glass beads. J Biochem Biophys Methods 47(3):221-31.
Wang J, Cai X, Rivas G, Shiraishi H, Dontha N. 1997. Nucleic-acid immobilization,
recognition and detection at chronopotentiometric DNA chips. Biosens
Bioelectron 12(7):587-99.
Wang JJ, Sanderson BJ, Wang H. 2007. Cyto- and genotoxicity of ultrafine TiO2
particles in cultured human lymphoblastoid cells. Mutation Research 628(2):99-
106.
162 Wang KKW, Yuen P-W. 1999. Calpains: Pharmacology and Toxicology of Calcium-
dependent Protease. Wang KKW, Yuen P-W, editors. Philadelphia: Taylor and
Francis. 77-101 p.
Warheit DB, Laurence BR, Reed KL, Roach DH, Reynolds GAM, Webb TR. 2004.
Comparative pulmonary toxicity assessment of single-wall carbon nanotubes in
rats. Toxicological Sciences 77:117-125.
Webster TJ, Waid MC, McKenzie JL, Price RL, Ejiofor JU. 2004. Nanobiotechnology:
carbon nanofibers as improved neural and orthopedic implants. Nanotechnology
15:48-54.
Weintraub H, Cheng PF, Conrad K. 1986. Expression of Transfected DNA depends on
DNA topology. Cell 46:115-122.
Wiesner U. Nanostructured Hybrids for Sensing and Separation; 2004; Ithaca, N.Y.
Cornell University.
Wilkinson JM. 2003. Nanotechnology applications in medicine. Medical Device
Technology 14(29-31).
Wiznerowicz M, Szulc J, Trono D. 2006. Tuning silence: conditional systems for RNA
interference. Nat Methods 3(9):682-8.
Xu Q. 1999. Microinjection into zebrafish embryos. Methods in Molecular Biology
127:125-132.
Yamada M, Kato K, Shindo K, Nomizu M, Haruki M, Sakairi N, Ohkawa K, Yamamoto
H, Nishi N. 2001. UV-irradiation-induced DNA immobilization and functional
utilization of DNA on nonwoven cellulose fabric. Biomaterials 22(23):3121-6.
163 Yamada M, Kato K, Shindo K, Nomizu M, Sakairi N, Yamamoto H, Nishi N. 1999.
Immobilization of DNA by UV irradiation and its utilization as functional
materials. Nucleic Acids Symp Ser(42):103-4.
Yan SF, Wu M, Geacintov NE, Broyde S. 2004. Altering DNA polymerase incorporation
fidelity by distorting the dNTP binding pocket with a bulky carcinogen-damaged
template. Biochemistry 43(24):7750-7765.
Yang NS, Burkholder J, Roberts B, Martinelli B, McCabe D. 1990. In vivo and in vitro
gene transfer to mammalian somatic cells by particle bombardment. Proceedings
of the National Academy of Sciences, USA 87:9568-9572.
Ye Q, Cassell AM, Liu H, Chao K-J, Han J, Meyyappan M. 2004. Large-Scale
Fabrication of Carbon Nanotube Probe Tips for Atomic Force Microscopy
Critical Dimension Imaging Applications. Nano Letters 4(7):1301-1308.
Yoshida Y, Kobayashi E, Endo H, Hamamoto T, Yamanaka T, Fujimura A, Kagawa Y.
1997. Introduction of DNA into Rat Liver with a Hand-Held Gene Gun:
Distribution of the Expressed Enzyme, [32P]DNA, and Ca2+ Flux. Biochemical
and Biophysical Research Communications 234:695-700.
Yu SL, Lee SK, Johnson RE, Prakash L, Prakash S. 2003. The stalling of transcription at
abasic sites is highly mutagenic. Molecular and Cellular Biology 23(1):382-388.
Yu Z, McKnight TE, Ericson MN, Melechko AV, Simpson ML, Morrison B. 2007a.
Vertically aligned carbon nanofiber array: A new type of microelectrode array for
electrophysiological recording. Journal of Neurotrauma 24(7):1235-1235.
164 Yu Z, McKnight TE, Ericson MN, Melechko AV, Simpson ML, Morrison B. 2007b.
Vertically aligned carbon nanofiber arrays record electrophysiological signals
from hippocampal slices. Nano Letters 7(8):2188-2195.
Zammatteo N, Jeanmart L, Hamels S, Courtois S, Louette P, Hevesi L, Remacle J. 2000.
Comparison between different strategies of covalent attachment of DNA to glass
surfaces to build DNA microarrays. Anal Biochem 280(1):143-50.
Zamore PD, Tuschl T, Sharp PA, Bartel DP. 2000. RNAi: double-stranded RNA directs
the ATPdependent cleavage of mRNA at 21 to 23 nucleotide intervals. Cell
101:25-33.
Zelenin AV, Titomirov AV, Kolesnikov VA. 1989. Genetic transformation of mouse
cultured cells with the help of high-velocity mechanical DNA injection. FEBS
Letters 244:65-67.
Zhang L, Melechko AV, Merkulov VI, Guillorn MA, Simpson ML, Lowndes DH,
Doktycz MJ. 2002. Controlled transport of latex beads through vertically aligned
carbon nanofiber membranes. Applied Physics Letters 81(1):135-137.
Zhi PX, Qing HZ, Gao QL, Ai BY. 2006. Inorganic nanoparticles as carriers for efficient
cellular delivery. Chemical Engineering Science 61(3):1027-1040.
Zhu L, Ang S, Liu WT. 2004a. Quantum dots as a novel immunofluorescent detection
system for Cryptosporidium parvum and Giardia lamblia. Applied and
Environmental Microbiology 70(1):597-598.
Zhu S, Oberdorster E, Haasch ML. 2006. Toxicity of an engineered nanoparticle
(fullerene, C60) in two aquatic species, Daphnia and fathead minnow. Marine
Environmental Research 62:S5-S9.
165 Zhu SG, Xiang JJ, Li XL, Shen SR, Lu HB, Zhou J, Xiong W, Zhang BC, Nie XM, Zhou
M and others. 2004b. Poly(L-lysine)-modified silica nanoparticles for the delivery
of antisense oligonucleotides. Biotechnology and Applied Biochemistry 39:179-
187.
Zipfel W. Optical Microscopy and Fluorescence Methods; 2004; Ithaca, N.Y. Cornell
University.
166 VITA
David George James Mann was born in Abbotsford, B.C., Canada in 1980. He graduated
with a Bachelor of Science degree in Biology from Bryan College in Dayton, T.N. in
2002. After internships at the U.S. Army Medical Research Institute of Chemical
Defense (USAMRICD) and Oak Ridge National Laboratory in 2002 and 2003, he began
the pursuit of his Ph.D. in Microbiology at the Center for Environmental Biotechnology and the University of Tennessee in Knoxville, T.N. in August 2003. He completed the requirements for his Ph.D. in February 2008.
167