The Synthesis and in vitro Evaluation of

Chemically Modified siRNAs that Contain

Internal Azobenzene Derivative Spacers for

Selective and Tunable Photocontrol of Activity

Matthew Hammill

Faculty of Science, University of Ontario Institute of Technology

Oshawa, Ontario

August 2017

A thesis submitted to the University of Ontario Institute of Technology

In partial fulfillment of the requirements of the degree of Master of

Science in Applied Bioscience

© Matthew Hammill 2017

ii

Abstract

Short interfering RNAs (siRNAs) are biopolymers that are used as post-transcriptional

gene regulators and as a natural endogenous defense against attack from viruses. These

molecules hold a great deal of interest as pharmaceuticals for the knockdown of

unregulated genes, such as in cancers. Through chemical modification of the RNA

structure, inherent limitations can be overcome to make better pharmaceuticals. This study

investigated the incorporation of azobenzene derivative spacers into the RNA backbone

replacing 2 base pairs. A small library of siRNAs were synthesized and then characterized

bio-physically through several means. After bio-physical characterization, these siRNAs

were tested in vitro against firefly luciferase in order to determine gene silencing efficacy.

The ability of these azobenzene siRNAs to be controlled with light was then investigated.

Our results indicate that these modifications are well tolerated within the RNAi pathway

and have a robust ability to be activated and inactivated with UV (ultraviolet) and visible

light to control gene silencing.

iii

Acknowledgements

I would first like to acknowledge my supervisor Dr. Jean-Paul Desaulniers for

taking a chance on me even though I had been out of school for a little while, and

welcoming me into his lab. His guidance, support and patience has been invaluable over

the last 2 years, and has allowed me to grow as both a biological chemist and as a person.

I would also like to extend my gratitude to my committee members Dr. Marc Adler

and Dr. Olena Zenkina. Their helpful advice during my committee meetings was greatly

appreciated, and I thank them for all their help.

I would also like to thank Dr. Holly Jones-Taggart and Dr. Andrea Kirkwood for

taking the time and being involved in the defense of my thesis.

I would like to thank all the groups who allowed me to do work in their labs for my

research: Dr. Sean Forrester, Dr. Julia Green- Johnson and Dr. Janice Strap. I really

appreciate your support!

My thanks also go out to members past and present of Dr. Desaulniers’ research

group, for both their help, and they’re company. Firstly I would like to thank Cameron

Isaacs-Trepanier for his contribution helping with cell cultures and in vitro testing of the

siRNAs. I would also like to thank Lidya Salim whose company made the long days go

much faster, and whose knowledge and advice on various subjects was invaluable. Next I

would like to acknowledge all the undergraduates that worked with me during my time

here: Noor Bahsoun, Ben Jones, Dion Chang, Kevoy Lopez, Hafsa Hoda and Autumn

Collins. It was a pleasure working with you.

iv

A special mention goes out to Genevieve Barnes for always helping me with the

NMR when I needed a hand, and to Michael Allison whose expertise on HPLC purification

was invaluable.

Finally I would like to thank my parents, without whom none of this would be

possible, thank you for your support and encouragement during the years.

Statement of Originality

To the best of this authors knowledge, the study herein is considered to be a novel

contribution to the scientific community.

v

Table of Contents

Title

Abstract……………………………………………………………………………...…..ii

Acknowledgments………………………………………………………………………iii

Statement of Originality………………………………………………………..……….iv

Table of Contents…………………………………………………………………..……v

List of Tables…………………………………………………………………..……….vii

List of Schemes………………………………………………………………………..viii

List of Figures………………………………………………………………...………..ix

Abbreviations……………………………………………………………………….…..x

Chapter 1.0 Introduction and Literature Review

1.1 The difference between DNA and RNA…………………………………….1

1.2 The RNA Interference Pathway……………………………………………. 5

1.3 siRNA Therapeutic Problems……………………………………………… 8

1.4 Chemical Modifications of siRNAs………………………………………. 11

1.5 Azobenzene and its Role in Nucleic Acid Modification………………….. 15

1.6 The Benefits of using an Azobenzene Modification……………………… 23

1.7 Project definition …………………………………………………………..24

2.0 Experimental Procedures

2.1 General Methods............................................................................................26

2.2 Synthesis and Characterization of Organic Compounds…………………...27

2.3 Procedure for Oligonucleotide Synthesis and Purification………………....33

vi

2.4 Biophysical Characterization………………………………………………34

2.5 Maintaining Cell Cultures……………………………………………….…35

2.6 Cell Based Assays………………………………………………………….36

3.0 Results and Discussion

3.1 Organic synthesis of Phosphoramidites…………………………………….39

3.2 Photoswitching ability of Azobenzene derivatives, Lability and siRNAs....42

3.3 Thermal Stability of siRNAs……………………………………………….46

3.4 Circular Dichroism Conformation of siRNA Helical Structure....................50

3.5 Silencing Capability of Azobenzene Modified siRNAs of the Exogenous Firefly

Luciferase Gene………………………………………………………………..51

3.6 Inactivation of the Silencing Capability of Azobenzene Modified siRNAs of the

Exogenous Firefly Luciferase Gene with UV Light……………………...…….53

3.7 Reactivation of the Silencing Capability of Azobenzene Modified siRNAs of

the Exogenous Firefly Luciferase Gene with UV Light……………..………....55

4.0 Future Directions…………………………………………………………..……...61

5.0 Conclusions…………………………………………………………………….….63

References......................................................................................................................64

Appendix………………………………………………………………………...…….74

vii

List of Tables

Table 3.1. Sequences of anti-luciferase siRNAs, mass spec and Tm’s of siRNAs containing

the azobenzene spacers………………………………………………………………....47

Table 3.2. IC50 values for siRNAs 7-13 and wild type……………………….………..53

List of Schemes

Scheme 3.1. Synthesis of Azobenzene DMT-phosphoramidite Spacers ……………...42

viii

List of Figures

Figure 1.1. Comparison of DNA and RNA structures…………………………...……..2

Figure 1.2. Natural RNA base modifications …………………………………………..3

Figure 1.3. Left: G-tetrad. Right: G-quadruplex formed from 2 tetrads……………..…4

Figure 1.4. RNA interference pathway in humans…………………………………...…7

Figure 1.5. Degradation of native RNA with cleavage sites highlighted in red……....10

Figure 1.6. DMT-Phosphoramidite oligonucleotide solid-phase RNA synthesis……..12

Figure 1.7. Various chemically-modified RNA……………………………………….14

Figure 1.8. Conformation change of azobenzene when exposed to light, trans (left) and cis

(right)……………………………………………………………………………….…..16

Figure 1.9. Conformational change of azobenzene trans (left ~9Å) and cis (right ~5.5Å)

when inserted into RNA………………………………………………………………..17

Figure 1.10. o-nitrobenzyl derivative for use as a photocage…………………………19

Figure 1.11 Various central region modifications in siRNAs from Dr. Jean-Paul

Desaulniers’ research group…………………………………………………………....22

Figure 3.1. Azobenzene Derivatives ……………………………………………...…..39

Figure 3.2. Final Steps of Azobenzene Formation .........................................……...…40

Figure 3.3. Change in absorbance for Az1 when exposed to UV (right) and visible (left)

light…………………………………………………………………………….....…....43

Figure 3.4. Absorbance Profile of siRNA 7 when exposed to UV and Visible light.

Inset: Zoomed in portion of 320-380 nm highlighting Az1 changes…………..........…44

Figure 3.5. Photolability of siRNA 7 exposed to alternating UV and visible light...…44

Figure 3.6. HPLC chromatogram of siRNA 10 with cis and trans peaks indicated…..45

Figure 3.7. HPLC chromatogram of siRNA 10 after 60 min UV exposure with cis and

trans peaks indicated……………………………………………………………..…….46

ix

Figure 3.8. Proposed Structural Change……………………………………………….49

Figure 3.9. CD spectra of azobenzene modified spacers replacing two nucleobases

targeting firefly luciferase mRNAs. ………………………………………….......……50

Figure 3.10. Gene silencing capability of azobenzene spacer modified siRNAs 7-13 and

wildtype……………………………………………………………………......………52

Figure 3.11. Gene silencing capability of azobenzene spacer modified siRNA targeting

firefly luciferase mRNA and normalized to Renilla luciferase for

siRNAs 7, 11 and 12.......................................................................................................54

Figure 3.12. Reduction in normalized firefly luciferase expression for siRNA 7 at 160 and

800 pM in HeLa cells monitored 8 and 12 hours post-transfection................................56

Figure 3.13. Reduction in normalized firefly luciferase expression for siRNA 9 at 160 and

800 pM in HeLa cells monitored 8 and 12 hours post-transfection ………….………..57

Figure 3.14. Reduction in normalized firefly luciferase expression for siRNA 10 at 160

and 800 pM in HeLa cells monitored 8 and 12 hours post-transfection ……..........…..58

Figure 3.15. Reduction in normalized firefly luciferase expression for siRNA 11 at 160

and 800 pM in HeLa cells monitored 8 and 12 hours post-transfection ………...…….59

Figure 3.16. Reduction in normalized firefly luciferase expression for siRNA 12 at 160

and 800 pM in HeLa cells monitored 8 and 12 hours post-transfection.........................605

x

List of Abbreviations

A: Adenosine

ACN: Acetonitrile

Ago: Argonaute protein

APS: Ammonium persulfate

AS: Antisense

Az1: One carbon azobenzene derivative

Az2: Two carbon azobenzene derivative

BCL-2: B-cell lymphoma protein 2

BCL-3: B-cell lymphoma protein 3

bp: Base pair

C: Cytosine

CD: Circular dichroism

CDCl3: Deuterated Chloroform

CH2Cl2: Dichloromethane

CPG: Controlled pore glass

dFT: Difluorotoluene

DMEM: Dulbeccos modified eagle medium

DMSO: Dimethylsulfoxide

DMT: 4,4’-Dimethoxytrityl

DNA: Deoxyribonucleic acid

ds: Double stranded

dd: Doublet of doublets

xi

dt: Doublet of triplets

dT: Deoxythymidine

EDTA: Ethylenediaminetetraacetc acid

EMAM: Methylamine 40% wt. in H2O and Methylamine 33% wt. in EtOH 1:1

EtOAc: Ethyl acetate

EtOH: Ethanol

FBS: Fetal Bovine Serum

G: Guanine

J: Coupling constant (Hz)

HIV: Human immunodeficiency virus

HPLC: High pressure liquid chromatography

IC50: Half Maximal inhibitory concentration

IDT: Integrate DNA technologies

LCD: Liquid crystal display

m/z: mass to charge ratio

Me: Methyl

MeOH: Methanol

mRNA: Messenger RNA

NaOH: Sodium hydroxide

NaOAc: Sodium Acetate

NMR: Nuclear magnetic resonance

nt: nucleotide

OH: Hydroxyl

xii

O-Me: O-Methyl

PAGE: Polyacrylamide gel electrophoresis

PBS: Phosphate buffered saline

p/s: Penicillin-Streptomycin

PS: Phosphorothioate

RET: Rearranged during transfection

RISC: RNA induced silencing complex

RNA: Ribonucleic acid

RNAi: RNA interference

RNAse III: Ribonuclease III

RNAse H: Ribonuclease H

rt: Room temperature

siRNA: small interfering RNA

ss: Single stranded

t: Triplet

T: Thymine

TBE: Tris-boric acid EDTA

TBDMS: tert-Butyldimethylsilyl

TCA: Trichloroacetic acid

TEA: Triethylamine

TEAA: Triethylamine-acetic acid

TEMED: Tetramethylethylenediamine

THF: Tetrahydrofuran

xiii

TLC: Thin layer chromatography

Tm: Melting temperature

tRNA: Transfer RNA

U: Uracil

UV: Ultraviolet

vis: Visible

wt; Wild-type

1

1.0 Introduction and Literature Review

1.1 The difference between DNA and RNA

All of life on Earth uses the same basic building blocks in order to encode

information that can be passed onto the next generation. These building blocks are called

nucleotides, and they make up the genetic structure of all known living things. DNA

(deoxyribonucleic acid) is composed of four bases, and when the four bases pair with the

complement strand, they engage in Watson-Crick base pairing: Adenine (A), Guanine (G),

Thymine (T) and Cytosine (C). Adenine hydrogen bonds to thymine (two hydrogen bonds)

and guanine hydrogen bonds to cytosine (three hydrogen bonds forming a

double stranded duplex.1

RNA (ribonucleic acid) uses three of the same bases (A, G, C) and replaces T with

a new base, Uracil (U). In addition to the absence of thymine, RNA can have over 100

different base modifications, as a way to expand the functionality of RNA transcripts.

These modifications are present in all RNA types (messenger, transfer, ribosomal, small

nuclear).2 Both DNA and RNA are sugars bound together by a phosphate backbone, the

two biggest differences being that DNA is seldom single stranded whereas RNA often is,

and that the RNA sugar has a 2’ hydroxyl group that is absent from DNA. Figure 1.1 shows

how DNA and RNA compare:

2

Figure 1.1. Comparison of DNA and RNA structures.

The 2’ hydroxyl on the RNA forces a conformational change in the pucker of the

sugar, adopting a C3’-endo conformation. This occurs because of the gauche effect (two

vicinal functional groups separated by a torsional angle) between the 2’ OH and the 4’

oxygen on the sugar.3 RNA modifications such as methylation and acetylation can provide

the cell greater flexibility in how transcripts are processed and how proteins are made.

Figure 1.2 highlights some additional natural bases that can be utilized:

3

Figure 1.2. Natural RNA base modifications

Many types of RNAs exist, mRNA (messenger RNA used in translation to make proteins),

tRNA (transfer RNA used to bring amino acids to the ribosome during protein synthesis),

miRNA (micro RNAs which provide non-specific transcriptional control) and siRNA

(short interfering RNA which can provide highly specific knockdown of a target mRNA).4

Interestingly, RNAs can form structures made up of many small single stranded helices in

order to obtain the ability to chemically catalyze reactions. These are called ribozymes and

perform much like protein based enzymes.5 Another type of structure that can be formed

is a G-quadruplex.6 G-quadruplexes are naturally occurring structures at the ends of

telomeres, and consist of 4 guanine bases forming a tetrad which can then be stacked to

form a quadruplex (Figure 1.3). They are also implicated in multiple cancers since they

occur in transcriptional regulatory regions of oncogenes.7 These kinds of structures could

have been partially responsible for how life started spontaneously billions of years ago,

4

groups of RNAs catalyzing reactions and becoming ever more efficient, allowing

interactions that contributed to the building of the first cell to occur.8

Figure 1.3. Left: G-tetrad. Right: G-quadruplex formed from 2 tetrads

One of the central tenets of molecular biology is that the translation of the

information stored in DNA into functional molecules such as proteins is essential to all life.

The pathway starts when genomic DNA is unwound in preparation for one of two things:

Firstly cell division, which requires replication of the genome, and secondly, RNA

synthesis through transcription. After the mRNA is transcribed, the polymer leaves the cell

nucleus and is sent to the ribosome in the cytoplasm.9 The ribosome is then able through

translation to synthesize the specific protein from the coding the mRNA contains with the

help of tRNAs, which allow amino acids to from peptide bonds in the correct order. This

is done through a three nucleotide code, called codons, which are specific to certain amino

acids. Redundancies between codons/amino acids exist, and because of this the genome

can be slightly more flexible in the case of point mutations. A single point mutation may

show no effect in the protein, and therefore have no effect on the organism’s ability to

survive.10

5

Usually process only happens in one direction, DNA to protein, but there are

instances of the reverse occurring, where RNA can be reverse transcribed into DNA. One

notable example is the HIV virus,11 which possesses a reverse transcriptase enzyme that

can store its viral RNA as DNA in the host cell nucleus indefinitely, making it difficult to

eliminate. Once it is stored in the cell’s nucleus, it can be hard to target with therapeutics

because the nuclear membrane is also polyanionic and carrier molecules are often required

to shuttle them across.12. There is a lot of potential and interest using siRNAs to target

specific mRNAs, since it could be a facile way to knockdown specific gene targets with

little to no off target effects, particularly in the fields of cancer treatment and genetic

abnormalities.

1.2 The RNA Interference Pathway

The RNA interference pathway was first observed by plant biologists who tried to

design petunias with higher pigmentation and darker colours by adding extra copies of the

chalcone synthase gene, an important enzyme in flower pigmentation. Instead of the

predicted darker phenotypes, many plants instead had flowers with little or no pigment

expressed, producing mostly white flowers. They found both the endogenous and

transgenic genes were being suppressed.13 Further research a few years later found this

decrease in expression was due an increase in post-transcriptional mRNA degradation

through an unknown mechanism, and calling it “co-suppression of gene expression”.14

Seeing this phenomenon in plants prompted the study of higher order animals to find out

if the same mechanism could be observed. This would show that being highly conserved,

this mRNA degradation pathway would be important to all cell types, plant and animal.

The study of mRNA degradation continued in model organisms such as Drosophilia where

6

“co-suppression” of mRNA was first observed in animals.15 In 1998 Fire and Mello’s

research group found that the mechanism by which the mRNA was being degraded was

though the addition of double stranded RNA which was complementary to the mRNA

target. This was the beginning of the modern understanding RNA interference, and they

won the 2006 Nobel prize in physiology and medicine for this work.16

SiRNAs are natural defense mechanisms found in the cell that are a part of the RNA

interference pathway (RNAi). The cell uses the pathway as a defense mechanism against

viruses or other parasites that inject dsRNA into the cells to hijack the cellular machinery

to reproduce. They can also target specific mRNA molecules to inhibit gene expression

after transcription by preventing the mRNA translation by the ribosome via shRNAs that

are produced in the nucleus.17 The RNAi pathway uses an enzyme called Dicer which

targets long double stranded RNAs (dsRNA) and cleaves them into 20-25 base pair pieces

(siRNA). The siRNA is then unwound and the passenger strand is degraded while the guide

strand becomes part of the RISC (RNA-induced silencing) complex, acting as a template

for RISC to anneal to further RNA so they can be cleaved by the Argonaute 2 protein.18,19

Figure 1.4 below depicts the RISC pathway.

7

Figure 1.4. RNA interference pathway in humans

The RISC cascade can be activated through synthetic siRNAs transfected

into the target cells via a carrier molecule or transfection agent such as a virus,20,21 or native

long endogenous dsRNAs can be processed by Dicer into the active RISC complex through

cleavage at the active site. The synthetic siRNAs are often 19-22 bp long and thus bypass

the Dicer enzyme entirely, although evidence has shown that recruitment of Dicer into the

pathway via a 27 bp long siRNA can improve efficacy of the interference by up to ten

times.22 The Dicer enzyme itself is composed of two RNAse III motifs, a dsRNA binding

8

domain and a RNA helicase domain which catalyzes the cleavage of the long dsRNA. In

one example of Dicer’s importance in the RISC cascade, a Dicer knockdown study in

human HeLa cells showed an accumulation of long dsRNAs in the cells showing an

effective disabling of the pathway.23

The second part of the RISC pathway involves the endonuclease Argonaute 2

(Ago2) which is able to catalyze the destruction of many mRNAs by removing the

passenger strand and keeping the guide strand as a template. Ago2 is comprised of three

domains; PAZ on the carboxyl terminus, a PIWI domain at the amino terminus and a central

Mid section.19 The catalysis of the mRNAs is controlled by PIWI domain, and is also

responsible for binding to the 5’ region of the antisense strand.24 Mid interacts with the 5’

phosphorylated end while PAZ binds to the 3’ overhang of the antisense strand.25 Knowing

how these regions interact with each other and their mRNA substrates are incredibly

important for designing and testing chemically modified siRNA therapeutics.26

1.3 siRNA Therapeutic Problems

There is a great deal of promise in using siRNAs for therapeutic research, such as

a way to combat cancer, deadly viral infections such as Ebola, as a way to correct auto

immune disorders or even genetic defects.27 A few issues currently exist with siRNAs, one

is their stability as they are easily degraded, another is that an appropriate delivery system

needs to be developed in order ensure the siRNA finds its target.28 They also have poor cell

membrane permeability and can cause undesired off-target effects on partially homologous

targets.

Within all cells there are enzymes that exist solely to target and destroy nucleic

acids, especially foreign DNA/RNA which can cause a reduction in the efficacy of nucleic

9

acid based therapeutics. These enzymes are exo/endonucleases and are prevalent in blood

serum and in the cell’s cytoplasm which target the phosphodiester backbone of RNA.29

The phosphodiester bond is also subject to hydrolysis at physiological pH because the

phosphorous is oxophilic and water attacks on the electrophilic site, which is

thermodynamically favoured.30 The oxophilic nature of the phosphorous also allows

nucleophilic attack from the 2’ hydroxyl on the ribose in an intramolecular hydrolysis

(Figure 1.5).31 These are the most prevalent stability issues that should be overcome to

make an ideal therapeutic.

Deprotonation of the phosphodiester backbone occurs at physiological pH because

the hydroxyl group on the phosphorous has pKa of near zero.32 This gives the RNA a

polyanionic backbone that reduces cell membrane permeability because there is a repulsion

between the membrane, which also has a negative charge due to the phospholipid bilayer,

and the deprotonated siRNA backbone.33 This can negatively impact cell uptake.

10

Figure 1.5. Degradation of native RNA with cleavage sites highlighted in red

Off-target effects can also occur with siRNAs. Non-complementary base pairing

can occur and an mRNA with a near homologous structure could be taken up into the RISC

complex unintentionally, since mismatches can still allow the Ago2 protein to cleave its

target.34 Another way off target effects can occur is if RISC selects the guide strand of the

duplex to be cleaved in error, thus that Ago2 is selecting entirely the wrong targets, which

in the best cases would show no activity and in the worst case cause apoptosis of the cell.35

If siRNAs are going to be viable therapeutic options, these issues need to be

overcome with some creative design of delivery systems such a nanoparticles or viral

vectors; or chemical modifications could be used to improve the various flaws in the

oligonucleotides. Exploration of novel chemical modifications to the siRNAs will be the

main focus of this dissertation.

11

1.4 Chemical Modifications of siRNAs

Organic chemists have spent years trying to mediate these disadvantages on RNA

while keeping an appropriate amount of gene silencing activity. It matters little if the RNA

is stable if the activity is poor. Often these modifications have their own disadvantages,

such as being expensive, easily degraded, toxic, volatile or disrupt secondary structure and

hydrogen bonding. Many current modifications increase stability while retaining gene

knockdown, but no one modification seems to be able to accomplish all of these tasks

without some kind of drawback.

Chemical modification of oligonucleotides using commercially available protected

bases and DMT (dimethoxytrityl)-phosphoramidite chemical synthesis, makes adding

normally unavailable molecules very reliable.36 In solid phase synthesis, oligos are

synthesized one base pair at time on controlled pore glass (CPG) supports, and this is one

of the most commonly used methods of chemical synthesis of DNA/RNA (Figure 1.6).37

Using this method any kind of chemical modification can be incorporated into a RNA/DNA

molecule, as long as there is an alcohol on each end of our modification that can be

protected with DMT-Cl on one end, and a phosphite group on the other.38

12

Figure 1.6. DMT-Phosphoramidite oligonucleotide solid-phase RNA synthesis.

The CPG support can be purchased with either a dT or universal (no base) attached

to the solid support. The synthesis proceeds in the opposite direction to biological

synthesis, 3’→5’ direction and is entirely automated. The initial deprotection is acid

catalyzed removal of the DMT (i) using trichloroacetic acid which affords a primary

alcohol that can react with the 2’ phosphite of the next base in the sequence. As figure 1.6

shows, the trityl cation is stabilized by the methoxy groups allowing for easy removal.

Activation of the base via ethylthiotetrazole (ii) must occur before the coupling phase or

13

else no reaction will occur.39 This activation displaces the diisopropylamine group.

Because of the oxophilic nature of the phosphorus as previously discussed, the now

exposed primary alcohol can attack the phosphite center coupling the bases together (iii).

RNA has phosphate linkages as opposed to phosphite ones, so oxidation must occur which

is initiated by nucleophilic attack on the iodine (iv). Pyridine is then used to create a

hydroxide anion (v), which attacks the phosphorus (vi) and the resulting hydroxyl is easily

deprotonated by another pyridine molecule creating the phosphate (vii). At this stage the

cycle can continue to the next nucleotide to be added, or if the desired length has been

reached the oligos can be cleaved from the solid support, deprotected and then used in

whichever experiment was planned (viii).40

There are 3 kinds of chemical modifications that can be done to an oligonucleotide:

1) Backbone substitution/addition 2) modification of the sugar directly and 3) nucleobase

modifications. This dissertation will be concerned with modification of the phosphodiester

backbone.

An example of a backbone modification is to replace one more base pairs with non-

cleavable, neutral spacers that increase the stability in vivo by replacing the phosphodiester

linkages with something that endonucleases are unable to use as a substrate. One such

modification is the replacement of one of the non-bonding oxygen atoms with a sulfur.41

This phosphorothioate (PS) provides extra stability by imparting nuclease resistance.42 This

modification however can disrupt activity when placed in or near the AGO2 cleavage site,

since the PS substitution results in a chiral center and one of the stereoisomers is a poor

substrate for AGO2.43

14

As the second kind of modification, sugar modifications can be as simple as adding a

methyl group to the 2’ hydroxyl in order to eliminate self hydrolysis44 and nucleobases can

be modified with exotic additions such a fluorine atoms, or even a fluorescent marker to

allow for monitoring of the oligos such as 2,4-difluorotoluene (dFT)45,46 (Figure 1.7).

Figure 1.7. Various chemically-modified RNA

Even with these novel modifications, problems can arise. In order for RNAi to

function, the siRNA must be in the A-form α helix47 and therefore modifications to the

sugar must be done with care, especially important when trying to control the amount of

flexibility you want your nucleotide to have. The ribose conformation can be altered

through utilizing the gauche and anomeric effects. The favourable conformation of

wildtype dsRNA is the C2’-endo/C2’-exo conformation, and modifications such as the 2’-

fluoro are well tolerated,48 and even provide good RNAi activity in vivo.49

Finally, modification of the nitrogenous bases is the final path to creating a more

stable siRNA. Since modifications here can disrupt the Watson-Crick base pairing, and

thus distort the double helix, often substitutions are made that try to minimize the impact

of disrupting the hydrogen bonding that occurs naturally. For example the 2,4-

diflurotoluene (dFT) is a thymine (DNA) and uracil (RNA) derivative with fluorine

15

replacing the natural oxygens50 (Figure 1.7). This provides minimal disruption to the

normal structure when placed at the 5’ end to increase affinity to the target strand when

placed in the guide strand. As a testament to the challenges of nucleobase modification

however, when placed elsewhere in the sequence, a disruption of the duplex occurs as the

dFT is unable to hydrogen bond and destabilizes the helix.51

As mentioned above, adding new chemical properties to the siRNAs could alter

some of the biophysical characteristics of the duplex, the most important of which are the

melting temperature (Tm) which is a measure of thermal stability; and the helical

conformation which must be in the A-form to be active.52 Thermal stability is a factor

because an unstable siRNA will dissociate at a lower temperature and thus may prove

inactive at biological pH where many of these tests occur, since single stranded RNA is not

a substrate for the RISC complex.53 A reduction in Tm is most often the case since most

substitutions either distort or remove the hydrogen bonding from the base pairs.54

1.5 Azobenzene and its Role in Nucleic Acid Modification

One way to overcome these limitations is to develop a tunable controlled photo

switchable siRNA that can be easily turned off and on with UV and visible light

respectively.

Azobenzene is a compound composed of two phenyl rings linked together by a

nitrogen-nitrogen double bond (Figure 1.8).55

16

Figure 1.8. Conformation change of azobenzene when exposed to light, trans

(left) and cis (right)

This molecule is of particular interest because when exposed to certain wavelengths

of light, there is a large conformational change in the structure. The molecule can isomerize

at the N=N bond between the trans to cis stereoisomer (Figure 1.9).56 This photo-

isomerization is also incredibly labile and robust, and maintains its efficiency after several

isomerizations back and forth without becoming exhausted.57 The photoswitching occurs

for trans to cis when the molecule is exposed to light in the 360 nm range, and reverse

occurs when exposed to blue light around 430 nm. These wavelengths can be altered by

substitutions on the ring of the molecule, so it possible to fine tune the wavelengths at

which these isomerizations can occur.58 The thermal stability of the cis form is also

extremely important. Azobenzene is able to thermally relax back to the trans form from cis

just from the ambient thermal energy.59 This is especially problematic since higher

temperatures cause the thermal relaxation to occur faster than lower temperatures, and this

can become an issue if the experiment you are trying to run exceeds the functional half-life

of the molecule. For example at 37 °C the half-life is about 4 hours.60 Since many biological

assays are 24 hours long, that is about 3 half-lives and so the inactive form is becoming

active over time, which may not be ideal.

17

Figure 1.9. Conformational change of azobenzene trans (left ~9Å) and cis (right

~5.5Å) when inserted into RNA

Azobenzene’s unique properties have been utilized in many fields for optical

control. One such field is liquid crystals for image storage, such as a liquid crystal display

(LCD) in order to increase response times for better computer monitors.61 Another use is

to control with light the rigidity of polymers.62 There is also a great deal of interest in using

azobenzene as a control for biopolymers in vitro and in vivo in order to have some kind of

control over various biological functions. The photoswitching aspect has been shown to be

active in proteins, DNA and RNA as well as DNA/RNA hybrids and as a method for

delivery via irreversible photocaging.

In proteins, one end of the azobenzene is tethered to the protein, and at the other

end is the ligand for the protein. The conversion to cis from trans positions the ligand

18

within the protein’s active site and “activates” it, thus turning it on.63 The protein can then

relax back to cis normally, or a different wavelength of light can be used to turn it off.

Another use is to use the azobenzene as a control for the secondary structure of the protein,

such as allowing the isomerization to disrupt a helix, changing it from active to inactive or

the reverse.64 Although the options seem limitless, one more protein related option for

azobenzenes will be mentioned: as RET (rearranged during transfection) kinase inhibitors.

The RET gene is located on chromosome 10 and has many natural alternative splices

(single gene coding for many proteins). Transmembrane receptors like the RET kinase

inhibitors (proteins which span the entire membrane to which they are permanently

attached), such as those in neurons, are required for healthy cell function, but become

dysfunctional in several cancers (notably thyroid) which make them a good therapeutic

target with the on and off functionality azobenzene provides.65 Because these are

neurotrophic receptors, it may be possible to inhibit cancer growth by stopping the

neurotransmitters from binding by removing the receptor protein through the RNA

interference pathway.

Nucleic acids are also good choices for an azobenzene modification. The ability to

control translation would be advantageous. Recent advances in this field have been made

in controlling DNA or RNA hybridization by utilizing the azobenzene as either a

replacement of the nitrogenous base so that the cis conformer disrupts hydrogen bonding

and the duplex becomes unstable and separates.66 Asunama et al. has shown that

azobenzene can be utilized as part of a threonine scaffold as a backbone replacement with

the azobenzene intercalating among the normal bases in both cases.67 Another interesting

use is in a hairpin at the loop position, allowing control over whether the hairpin can self-

19

anneal.68 Short hairpin RNAs can activate RISC,69 and could be used as a translational

control. An interesting study in 2015 by Wu et al. and their group showed that it was

possible to control the activity of RNase H (exclusively targets DNA/RNA hybrids) by

using a deoxynucleotide complement to an RNA strand, with azobenzene flanking either

side of the active site making a dumbbell. Exposure to light would allow the RNA/DNA

strands to anneal, and RNase H would be able to digest the hybrid.70 This kind of control

could allow for the targeting of specific RNA sequences at specific times, by activating

and deactivating it at will.

Lastly work with non-reversible photo cages has also shown some advancement. A

photo cage is a molecule or several molecules that when exposed to light either change

shape or are cleaved entirely from the molecule.71 One such example of a small molecule

delivery system is that of o-nitrobenzyl derivatives which when exposed to light goes

through a intramolecular hydrogen abstraction by the nitro group forms the aci-nitro and

goes through a rearrangement to nitroso derivatives.72

Figure 1.10. o-nitrobenzyl derivative for use as a photocage

20

Utilizing these kinds of photoactive molecules it is possible to attach them to small

molecules such as drugs,73 or enzyme substrates74 (as done by Hiraoka and Hamachi 2003)

and have them released upon exposure to light, keeping the inactive form available until

needed. In the case of enzyme substrates they are used to bulk the small molecule up so it

will no longer fit inside the active site of the enzyme, and when removed activity can

resume.75 Mikat and Heckel 2007 also showed that it is possible to control RNAi with

photo cage materials by surrounding the mRNA cleavage site between base pairs 10 and

11 in order to disable RNAi. Irradiation of UV light at 366 nm was shown to reactivate the

pathway and had similar activity to the unmodified siRNAs.76

Recent work from Dr. Jean-Paul Desaulniers’ research group has shown that

aromatic spacer linkages77 and non-aromatic alkyl spacer linkages54 are substrates for the

RNAi pathway when these modifications are placed within the central region of the sense

strand. This suggests that a large aromatic group may also function at the central region

and would make an appropriate bioactive siRNA bearing unique functionality. Desaulniers

et al. (2017) has shown siRNAs with biphenyl aromatic linkers did in fact have excellent

knockdown ability targeting the endogenous BCL-2 gene.78 Additional recent work from

his group has also revealed that large hydrophobic groups like cholesterol can cross the cell

membrane and inhibit gene expression without a transfection carrier79 (Figure 1.8 shows

the progression of modifications).

This suggests that azobenzene has the potential to offer some unique advantages

for a chemically modified siRNA. Despite this potential, siRNAs bearing azobenzene

within the central region has not been explored. This modification could prevent localized

off-target effects to surrounding healthy tissue, since potentially the siRNA is only being

21

activated by the light switch in the diseased tissue. Many cancer treatments employ systems

that harm healthy tissue, such as pharmaceuticals and excision, but some healthy tissue is

always destroyed in the process. Our azobenzene modified siRNA may be able to mitigate

some of these disadvantages. Homologous mRNA off-target effects may still occur,

causing undesirable phenotypes or unexpected toxicity but these are usually concentration

dependant and less likely to occur at the concentrations with which we will be testing these

siRNAs.80

22

Figure 1.11 Various central region modifications in siRNAs from Dr. Jean-Paul

Desaulniers’ research group54, 77, 78, 79

23

1.6 The Benefits of Using an Azobenzene Modification

If this modification is incorporated into a biological molecule like siRNA, this

conformational change could potentially allow for the controlled activation and

inactivation of the molecule with the application of a light source. Azobenzene as a

backbone substitution is also more stable, since enzymes that degrade RNA in the cell,

target the natural phosphodiester backbone as their substrate, and a different molecule

(azobenzene) in their active site is likely to show no degradation. Another benefit is that

since it has a neutral charge, cell membrane permeability should also increase, due to less

repulsion between the negatively charged siRNA and the cell membrane, which has a

negative charge as well.34

Incorporation of the photo-switchable azobenzene into a siRNA would allow

specific genes to be targeted without having to be inside the nucleus of the cell, which can

be difficult to get through with compounds that are not naturally occurring. It will also

allow potentially very good spacial and temporal control over translation. Two azobenzene

derivatives, Az1 and Az2 (Figure 1.8) were synthesized and their stability and activity was

tested. Potential targets for these siRNAs could be the BCL-2 or BCL-3 genes involved in

non-Hodgkins lymphoma81 or any other oncogene implicated in cancers. The genes that

code BCL-2 and BCL-3 are anti-apoptotic proteins that control cell death. Disruption of

this pathway prevents apoptosis through overexpression and can allow cancerous cells to

propagate. By targeting the mRNA, we prevent the protein from forming and possibly

mitigating or eliminating the actual effects of the mutation, without removing or disabling

the problematic DNA. This would allow therapeutic targets to be reached more easily, since

24

only one membrane would have to be crossed in order to silence the gene of interest which

would increase the effectiveness of the treatment.

1.7 Project Definition

Being able to silence genes pre translation provides many avenues of treatment for

various diseases. This promising future of siRNAs as therapeutic agents is limited only by

the kinds of chemical modifications we can impart in order to achieve our desired

characteristics. As this field expands, more siRNAs will be developed with better cell

permeability, resistance to nucleases and reduced off target effects and toxicity.

1.7.1 Objectives and Hypothesis

Objective 1. Chemical synthesis of azobenzene and azobenzene modified siRNAs.

constructs.

Objective 2. Biophysical Characterization of azobenzene modified siRNAs.

Objective 3. In vivo testing of azobenzene modified siRNAs against firefly

luciferase in HeLa cells.

Hypothesis: Inserting an azobenzene derivative into the siRNA backbone will

allow tunable photochemical control of the siRNA’s activity through the use of UV

(inactivation) and visible (reactivation) light.

Objective one of this project will involve the development of a small library of

central region and 3’ end azobenzene containing siRNA oligonucleotides. As seen in

Figure 1.8, and shown in the previous study by Desaulniers et al. (2017) the central region

25

will easily accommodate large aromatic spacers. The next step is to add functionality to the

spacer with N=N bond, through an azobenzene moiety. The modification will be

synthesized using organic chemistry and synthetically incorporated into RNA

oligonucleotides. Objective two will take the azobenzene containing sense strands and

anneal them to their complement sequence to form active siRNA duplexes. These annealed

siRNAs will undergo hybridization testing via melting temperature assays as well testing

of the α helix with circular dichroism to confirm the secondary structure. These bio-

physical characterizations will be tested under both visible and ultraviolet conditions in

order to determine the activity of the photo-switch. Further bio-physical characterization

of the siRNAs will be through absorbance spectra and how it changes, as well as HPLC

characterizations in order to quantify the trans to cis conversions immediately and

quantitatively. Objective three will test the siRNAs ability to silence gene expression using

the dual luciferase reporter system from Promega. This will allow for easy quantification

of the siRNAs knockdown ability. After gene silencing is confirmed the siRNAs will be

treated with UV light in order to inactivate them in vivo and if successful they will then be

treated with visible light to see if they can be reactivated after a period of inactivity.

26

2.0 Experimental Procedures

2.1 General Methods

Unless otherwise indicated all starting reagents used were obtained from

commercial sources without additional purification. Anhydrous CH2Cl2 and THF were

purchased from Sigma-Aldrich and run through a PureSolv 400 solvent purification system

to maintain purity. Flash column chromatography was performed with Silicycle Siliaflash

60 (230-400 mesh), using the procedure developed by Still, Kahn and Mitra.81 NMRs were

performed on a Varian 400 MHz spectrophotometer. All 1H NMRs were recorded for 64

transients at 400 MHz and all 13C NMRs were run for 1500 transients at 101 MHz and all

31P NMRs were recorded for 256 transients at 167 MHz. Spectra were processed and

integrated using ACD labs NMR Processor Academic Edition.

27

2.2 Synthesis and Characterization of Organic Compounds

2.2.1 Synthesis of 4,4’-bis(hydroxyethyl)-azobenzene (1)

(1)

To a solution of 90.0 mL of 5.70 M NaOH(aq) 4.00 g of 4-nitrophenethyl alcohol

(23.9 mmol, 1.00 equiv.) was added and stirred until fully dissolved. To this solution 6.00

g Zn powder (92.3 mmol, 3.85 equiv.) was slowly added to maintain stirring. After

refluxing overnight for 16 h it was then filtered on a Buchner Vacuum filter, and the crystals

were suspended in hot methanol. The crystals were collected and filtered again with a

gravity filter to remove residual salts and then the methanol solution was removed using a

rotary evaporator. Crystals were collected and purified on silica gel column using MeOH

(2% to 5%) in CH2Cl2 to afford compound 1 as orange crystals (2.34 g, 72%). 1H NMR

(400 MHz, d6-DMSO) δ 7.75 - 7.81 (d, J = 8.21 Hz 4H) 7.41 (d, J = 8.21, Hz 4H) 4.71 (t,

J = 5.28 Hz, 2H) 3.65 (td, J = 6.84, 5.47 Hz, 4H) 2.80 (t, J = 6.84 Hz, 4H). 13C NMR (101

MHz, d6-DMSO) 150.8, 144.0, 130.3, 122.7, 62.2, 38.8; ESI-HRMS (ES+) m/z calculated

for C16H18N2O2: 271.1441, found 271.1438 [M+H]+

28

2.2.2 Synthesis of 4-hydroxyethyl-4’-O-(4,4’dimethoxytrityl)-azobenzene (2)

(2)

To a solution of anhydrous of 30.0 mL of THF 1.50 g of compound 1 (5.55 mmol,

1.00 equiv.) was added and stirred until fully dissolved. To the solution 1.95 g of 4,4’-

dimethoxytrityl chloride (5.76 mmol, 1.00 equiv.) was added along with 2.50 mL

triethylamine (17.9 mmol, 3.00 equiv.) The reaction mixture was stirred vigorously

overnight and monitored by TLC. The crude reaction was then concentrated by rotovap

and purified on silica gel column using MeOH (2% to 5%) in CH2Cl2 to afford compound

2 as an orange oil (1.11 g, 35%). 1H NMR (400 MHz, d-CDCl3) δ 7.87 (td, J = 9.09, 1.37

Hz, 6H) 7.45 - 7.47 (m, 2H) 7.24-7.40 (m, 3 H) 7.18 - 7.22 (m, 5H) 6.80 - 6.87(m, 4H)

3.91(t, J = 6.45 Hz, 2H) 3.75 - 3.82(s, 6H) 3.36 (t, J = 6.64 Hz, 2H) 3.07 (s, 1H) 2.94 - 2.98

(m, 4H). 13C NMR (101 MHz, d-CDCl3) δ 158.3, 151.5, 151.4, 151.3, 151.2, 145.1, 143.1

142.9, 142.0, 141.8, 136.3, 130.0, 130.0, 129.9, 129.9, 129.7, 129.7, 129.3 129.2, 129.1,

128.1, 127.8, 127.8, 127.7, 127.7, 126.6, 123.0, 122.9, 122.7, 120.9, 120.5, 113.1 113.0,

113.0, 86.0, 77.4, 77.1, 76.7, 67.9, 64.4, 63.4, 63.3, 55.25, 55.20 53.4; ESI-HRMS (ES+)

m/z calculated for C37H36N2O4: 573.2748, found 573.2741 [M+H]+

29

2.2.3 Synthesis of 2-Cyanoethyl-4-O-{[4-hydroxyethyl-4’-O-(4,4’dimethoxytrityl)-O-

methyl-diazenyl]}-N,N’-diisopropylaminophosphoramidite (3)

(3)

To a solution of 4.00 mL of anhydrous DCM/ACN (1:1) 0.26 g of compound 2

(0.45 mmol, 1.00 equiv.) was added to a flame dried flask. To that solution 0.63 mL of

anhydrous triethylamine (4.50 mmol, 10.0 equiv.) was added along with 0.30 mL of 2-

cyanoethyl N,N-diisopropylchlorophosphoramidite with (1.35 mmol, 3.00 equiv.) and

stirred at room temperature until TLC showed starting materials were consumed (3 h).

After rotary evaporation, the compound was then purified on silica gel using a

68%:30%:2% hexanes/ethyl acetate/triethylamine mobile phase. This afforded compound

3 as an orange oil (0.22 g, 63%). 1H NMR (400 MHz, d-CDCl3) δ 7.82 - 7.88 (m, 4H) 7.32

- 7.41 (m, 6H) 7.27 - 7.30 (m, 4H) 7.21 - 7.25 (m, 3H) 6.84 (d, J = 8.99 Hz, 4H) 6.76 - 6.81

(m, 3H) 4.13 (q, J = 7.29 Hz, 2H) 3.84-3.94 (m, J = 7.42 Hz, 1H) 3.75 - 3.81 (m, 6 H) 3.55

- 3.67 (m, 1H) 3.35 (t, J = 6.64 Hz, 2H) 2.93 - 2.98 (t, J = 6.74 Hz, 2H) 1.50 (d, J = 6.64

Hz, 1H) 1.27 (t, J = 7.23 Hz, 6H) 1.17 (dd, J = 10.75, 6H).13C NMR (101 MHz, d-CDCl3)

δ 158.6, 158.3, 151.2, 145.1, 142.9, 136.3, 130.0, 129.9, 129.9, 129.7, 129.1, 128.1, 127.8,

127.7, 127.7, 127.0, 126.6, 122.7, 122.7, 122.6, 113.1 113.0 112.9, 86.0, 77.3, 77.0, 76.7,

30

64.3, 60.3, 55.2, 55.1, 43.1, 43.0, 36.6, 24.6, 24.5, 24.5, 21.0, 14.2. 31P NMR (162 MHz,

d-CDCl3) δ ppm 147.54, 147.13 ESI-HRMS (ES+) m/z calculated for C46H53N4O5P:

772.4797, found 703.3949 [M+H]+ (Hydrolyzed product).

2.2.4 Synthesis of 4,4’-bis(hydroxymethyl)-azobenzene (4)

(4)

To a solution of 90.0 mL of 5.70 M NaOH(aq) 4.00 g of 4-nitrobenzyl alcohol (26.1

mmol, 1.00 equiv.) was added and stirred until fully dissolved. To this solution 6.00 g of

Zn powder (92.3 mmol, 3.53 equiv.) was slowly added to maintain stirring. After refluxing

overnight for 16 h it was then filtered on a Buchner Vacuum filter, and the crystals were

suspended in hot methanol. The crystals were collected and filtered again with a gravity

filter to remove residual salts and then the methanol solution was removed using a rotary

evaporator. Crystals were collected and purified on silica gel column using MeOH (2% to

5%) in CH2Cl2 to afford compound 4 as orange crystals (2.13 g, 67%). 1H NMR (400 MHz,

d6-DMSO) δ 7.85 (d, J = 8.60 Hz, 4H) 7.51 (d, J = 8.21 Hz, 4H) 5.35 (s,2 H) 4.59 (d, J =

5.47 Hz, 4H). 13C NMR (101 MHz, d6-DMSO) δ 146.2, 127.0, 122.3, 105.8, 63.4. ESI-

HRMS (ES+) m/z calculated for C14H14N2O2: 243.1128, found 243.1126 [M+H]+

31

2.2.5 Synthesis of 4-hydroxymethyl-4’-O-(4,4’dimethoxytrityl)-azobenzene (5)

(5)

To a solution of 30.0 mL of anhydrous THF 1.5 g of compound 4 (6.19 mmol, 1.00

equiv.) was added and stirred until fully dissolved. To the solution 2.09 g of 4,4’-

dimethoxytrityl chloride (6.19 mmol, 1.00 equiv.) was added along with 2.60 mL of

triethylamine (18.6 mmol, 3.00 equiv.) The reaction mixture was stirred vigorously

overnight and monitored by TLC. The crude reaction was then concentrated by rotovap

and purified on silica gel column using MeOH (2% to 5%) in CH2Cl2 to afford compound

5 as an orange oil (1.11 g, 33%). 1H NMR (400 MHz, d-CDCl3): δ 7.87 (m, 4H) 7.51 (m,

6H) 7.44 (m, 4 H) 7.32 (t, J = 7.62 Hz, 3H) 6.82 (m, 4H) 4.59 (s, 2H) 4.18 (s, 2H) 3.8 (s,

6H).13C NMR (101 MHz, d-CDCl3) δ 158.6, 158.2, 151.3, 146.6, 145.2, 140.6, 135.9,

130.1, 129.3, 128.4, 128.0, 127.9, 127.8, 127.5, 123.0, 122.8, 113.7, 113.2, 86.5, 62.9, 55.5,

55.4; ESI-HRMS (ES+) m/z calculated for C35H32N2O4: 545.2435, found 545.2430

[M+H]+

32

2.2.6 Synthesis of 2-Cyanoethyl-4-O-{[4-hydroxymethyl-4’-O-(4,4’dimethoxytrityl)-

O-methyl-diazenyl]}-N,N’-diisopropylaminophosphoramidite (6)

(6)

To a solution of 4.00 mL of anhydrous DCM/ACN (1:1) 247 mg of compound 5 (0.45

mmol, 1.00 equiv.) was added to a flame dried flask. To that solution 0.63 mL of anhydrous

triethylamine (4.50 mmol, 10.0 equiv.) was added along with 0.30 mL of 2-cyanoethyl

N,N-diisopropylchlorophosphoramidite with (1.35 mmol, 3.00 equiv.) and stirred at room

temperature until TLC showed starting materials were consumed (3 h). After rotary

evaporation, the compound was then purified on silica gel using a 68%:30%:2%

hexanes/ethyl acetate/triethylamine mobile phase. This afforded compound 6 as an orange

oil (0.23 g, 66%). 1H NMR (400 MHz, d-CDCl3) δ 7.88 - 7.94 (m, 4H) 7.50 - 7.56 (m, 5H)

7.39 - 7.45 (m, 3H) 7.29 - 7.33 (m, 1H) 6.83 - 6.89 (m, 4 H) 4.80 (dd, J = 19.73, 8.40 Hz,

1 H) 4.26 (s, 2 H) 4.09 - 4.16 (m, 1H) 3.82 - 3.93 (m, 2H) 3.79 - 3.81 (m, 6H) 3.59 - 3.73

(m, 2H) 2.62 - 2.68 (m, 2H) 1.18 - 1.25 (m, 12H). 13C NMR (101 MHz, d-CDCl3) ppm

158.3, 152.0, 150.7, 144.8, 141.9, 141.5, 136.1, 130.0, 129.7, 128.2, 127.7, 127.5, 127.4,

126.6, 122.8, 122.7, 122.6, 117.6, 113.1, 113.0, 86.4, 77.3, 77.0, 76.7, 65.2, 65.0, 58.5,

58.3, 58.3, 55.1, 51.8, 50.7, 50.5, 46.3, 43.2, 43.1, 43.0, 42.8, 24.7, 24.6, 24.6, 24.6, 24.5,

33

22.8, 21.5, 20.4, 20.3, 20.3. 31P NMR (162 MHz, d-CDCl3) δ ppm 149.37, 148.72.

C46H53N4O5P: 744.87, found 660.34 [M+H]+ (Hydrolyzed product).

2.3 Procedure for Oligonucleotide Synthesis and Purification

All standard β-cyanoethyl 2'-O-TBDMS protected phosphoramidites, reagents and

solid supports were purchased from Chemgenes Corporation and Glen Research. Wild-

type luciferase strands including the sense and 5'-phosphorylated antisense strand were

synthesized. All commercial phosphoramidites were dissolved in anhydrous acetonitrile to

a concentration of 0.10 M. The chemically synthesized (azobenzene derivative)

phosphoramidites were dissolved in 3:1 (v/v) acetonitrile:THF (anhydrous) to a

concentration of 0.10 M. The reagents that were used for the phosphoramidite coupling

cycle were: acetic anhydride/pyridine/THF (Cap A), 16% N-methylimidazole in THF (Cap

B), 0.25 M 5-ethylthio tetrazole in ACN (activator), 0.02 M iodine/pyridine/H2O/THF

(oxidation solution), and 3% trichloroacetic acid/dichloromethane. All sequences were

synthesized on 0.20 μM or 1.00 μM dT solid supports except for sequences that were 3'-

modified, which were synthesized on 1.00 μM Universal III solid supports. The entire

synthesis ran on an Applied Biosystems 394 DNA/RNA synthesizer using 0.20 μM or 1.00

μM cycles kept under argon at 55 psi. Standard and synthetic phosphoramidites ran with

coupling times of 999 seconds.

Antisense sequences were chemically phosphorylated at the 5'-end by using 2-[2-

(4,4’-dimethoxytrityloxy)ethylsulfonyl]ethyl-(2-cyanoethyl)-(N,N-diisopropyl)-

phosphoramidite. At the end of every cycle, the columns were removed from the

synthesizer, dried with a stream of argon gas, sealed and stored at 4 °C. Cleavage of

34

oligonucleotides from their solid supports was performed through on-column exposure to

1.50 mL of EMAM (methylamine 40% wt. in H2O and methylamine 33% wt. in ethanol,

1:1 (Sigma-Aldrich)) for 1 hour at room temperature with the solution in full contact with

the controlled pore glass. The oligonucleotides were then incubated overnight at room

temperature in EMAM to deprotect the bases. On the following day, the samples were

concentrated on a Speedvac evaporator overnight, resuspended in a solution of

DMSO:3HF/TEA (100 μL:125 μL) and incubated at 65 °C for 3 hours in order to remove

the 2'-O-TDBMS protecting groups. Crude oligonucleotides were precipitated in EtOH and

desalted through Millipore Amicon Ultra 3000 MW cellulose. Oligonucleotides were

separated on a 20% acrylamide gel and were used without further purification for annealing

and transfection. Equimolar amounts of complimentary RNAs were annealed at 95 °C for

2 min in a binding buffer (75.0 mM KCl, 50.0 mM Tris-HCl, 3.00 mM MgCl2, pH 8.30)

and this solution was cooled slowly to room temperature to generate siRNAs used for

biological assays. A sodium phosphate buffer (90.0 mM NaCl, 10.0 mM Na2HPO4, 1.00

mM EDTA, pH 7.00) was used to anneal strands for biophysical measurements.

2.4 Biophysical Characterization

2.4.1 Procedure for Performing CD Experiments

Circular Dichroism (CD) spectroscopy was performed on a Jasco J-815 CD

equipped with temperature controller. Equimolar amounts of each siRNA (10 μM) were

annealed to their compliment in 500 μL of a sodium phosphate buffer. CD measurement of

each duplex were recorded in triplicate from 200-500 nm at 25 °C with a screening rate of

20.0 nm/min and a 0.20 nm data pitch. The average of the three replicates was calculated

35

using Jasco’s Spectra Manager version 2 software and adjusted against the baseline

measurement of the sodium phosphate buffer.

2.4.2 Procedure for Absorbance Spectra Experiments

All absorbance spectra measurements were done on a Jasco J-815 CD with

temperature controller. Measurement was recorded from 200 -500 nm at 10 °C at least 3

times. UV treated samples were placed under a UVP UVL-23RW Compact UV lamp 4.00

W 365 nm for the indicated time at 25 °C . Visible light treated samples were placed under

a 60.0 W daylight bulb from NOMA in standard desk lamp.

2.4.3 Procedure for HPLC Characterization

HPLC chromatograms were obtained on a Waters 1525 binary HPLC pump with a

Waters 2489 UV/Vis detector using the Empower 3 software. A C18 4.6 mm x 150 mm

reverse phase column was used. Conditions were 5% acetonitrile in 95% 0.1 M TEAA

(Triethylamine-Acetic Acid) buffer up to 100% acetonitrile over 40 min.

2.5 Maintaining Cell Cultures

For biological analysis of these siRNAs in a live environment, human epithelial

cervix carcinoma cells were used (HeLa cells). They were kept in 250 mL vented culture

flasks using 25.0 mL of DMEM with 10% fetal bovine serum and 1% penicillin-

streptomycin (Sigma) in an incubator set for 37 °C @ 5% CO2 humidified atmosphere.

36

Once cell lines became confluent (80-90%) they were passaged by washing 3 times

with 10 mL of phosphate buffered saline (NaCl 137 mM, KCl 2.70 mM, PO43-

10.0 mM,

pH 7.40) and incubated with 3.00 mL of 0.25% trypsin (SAFC bioscience) for 4 min @ 37

°C to detach the cells. The cells were transferred to a 50.0 mL centrifuge tube after the

addition of 10.0 mL of DMEM solution and pelleted at 2000 rpm for 5 minutes. The

supernatant was discarded and the pellet resuspended in 5.0 mL DMEM with 10% FBS.

A standard haemocytometer was used to obtain cell counts, after which the cells

were diluted to a final concentration of 1.00 x 106 cells/mL for subsequent assays. To

continue the cell line 1.00 mL of freshly passaged cells was added to 24.0 mL of

DMEM/10% FBS and 1% penicillin-streptomycin at 37 °C in a new culture flask while the

rest were used for assays.

2.6 Cell Based Assays

2.6.1 Procedure for siRNA Transfections

100 μL of cells (total 1.00 x 105) were transferred into 12 well plates (Falcon®)

with 1 mL of DMEM (10% FBS, 1% penicillin-streptomycin) and incubated at 37 °C with

5% CO2. After 24 hours the cells were transfected with various concentrations of siRNAs,

along with both pGL3 (Promega) and pRLSV40 luciferase plasmids using Lipofectamine

2000 (Invitrogen) in Gibco’s 1X Opti-Mem reduced serum media (Invitrogen) according

to the manufacturer’s instructions. 1.00 μL of siRNA was added along with 2.00 μL (pGL3

200 ng) and 0.50 μL pRLSV40 (50.0 ng) to 100 μL of 1X Opti-Mem in a microcentrifuge

tube and kept on ice for 5 min. In a different microcentrifuge tube 1.00 μL of Lipofectamine

2000 (Invitrogen) was mixed with 100 μL of Gibco’s 1X Opti-Mem reduced serum media

37

(Invitrogen) and incubated at room temperature for 5 min. After 5 minutes the tubes were

mixed and incubated at room temperature for 20 min and then the entire contents

transferred to the wells of the 12 well plate.

2.6.2 Procedure for in vitro Dual-Reporter Luciferase Assay

100 μL of cells (total of 1.00 x 105 cells) were added to 12 well plates (Falcon®)

with 1 mL of growth media (DMEM 10% FBS, 1% penicillin-streptomycin) and incubated

at 37 °C with 5% CO2. After 24 hours the cells were transfected with 8.00, 20.0, 40.0, 160,

400 and 800 pM concentrations of siRNAs, co-transfected with both the pGL3 (Promega)

coding for the firefly luciferase protein and the pRLSV40 Renilla luciferase plasmids using

Lipofectamine 2000 (Invitrogen) in Gibco’s 1X Opti-Mem reduced serum media

(Invitrogen) according to the manufacturer’s instructions. After a set amount of time (8, 12

or 22 h) the cells were incubated at room temperature in 1X passive lysis buffer (Promega)

for 20 minutes. The lysates were collected and loaded onto a 96 well, opaque plate (Costar).

With a Dual-Luciferase reporter Assay kit (Promega), Lar II enzyme solution was added

which produces the photochemical reaction with the firefly luciferin protein. Stop & Glo®

with the Renilla luciferase enzyme was sequentially added to the same lysate and enzyme

activity was measured through luminescence of both firefly and Renilla luciferin on a

Synergy HT (Bio-Tek) plate luminometer. The ratio of firefly and Renilla luminescence is

expressed as a percentage to untreated controls for siRNA efficacy. Each value is the

average of at least 3 different experiments with standard deviation indicated.

38

2.6.3 Procedure for Light Inactivation of Azobenzene Modified siRNA (trans to cis)

1.00 μL of the desired siRNA was added to a microcentrifuge tube and exposed to

a 4.00 W 365 nm UV lamp (UVP) at 25 °C and was exposed for at least 4 hours. Longer

exposure times required the addition of a small amount of RNAse free water to prevent

evaporation.

After UV exposure the siRNA can be used in the transfection procedure above, but

the transfection must be done in the dark room, at 25 °C to prevent the cis to trans

conversion back into the active form.

2.6.4 Procedure for Light Reactivation of Azobenzene Modified siRNA (cis to trans)

4 hours after the transfection procedure, the plate was exposed to a 60.0 W daylight

bulb (NOMA) and left under the visible lamp for 4 h (8 h assay) or 8 h (12 h assay) after

which the cells were lysed and the plate read as above.

39

3.0 Results and Discussion

3.1 Organic synthesis of Phosphoramidites

In this study a small library of azobenzene modified siRNAs were synthesized in

order to increase the scope of possible modifications of the central region of siRNAs. This

was done using azobenzene spacer phosphoramidites which were then incorporated into

oligonucleotides using solid phase synthesis. Two azobenzene derivatives were

synthesized for use as DMT-phosphoramidite spacers as shown in Figure 3.1:

Figure 3.1. Azobenzene Derivatives

In order to obtain the nitrogen double bond functionality of the azobenzene, the

synthesis begins with either 4-nitropenethyl alcohol (Az2) or 4-nitrobenzyl alcohol (Az1).

The nitro-alcohols were added to a concentrated basic solution of 5.7 M NaOH in order to

reduce the nitro group through a nitrosobenzene (R-N=O) intermediate and becoming an

aniline (R-NH2) functional group when fully reduced. For this to occur an electron source

is required and Zn powder was used to accomplish this.

One possible pathway for this reduction where the nitro group becomes protonated

and then an electron is able to attack the positive nitrogen. Subsequent to this attack the

40

hydroxyl group is protonated after the N=O bond reforms and leaves as water forming a

nitrosobenzene group. The lone electron on the nitrogen is then protonated and

consequently a new electron from a Zn atom can attack the nitrogen. Since this is done

under basic conditions, the formation of a hydroxyl ion is favourable and addition of an

electron can allow it to form and leave a negatively charged nitrogen behind. Protonation

of the nitrogen gives the 4-aminobenzyl alcohol which is required for the next step.

Figure 3.2. Final Steps of Azobenzene Formation

In Figure 3.2, the azobenzene functional group finally begins to take shape. This

begins with an attack by the lone pair of electrons on the –NH2 group of the 4-aminobenzyl

alcohol on the 4-nitrosobenzyl alcohol’s –N=O group. This forms a σ bond between each

nitrogen. This allows the negatively charged oxygen to be protonated, after which a proton

can be removed from the positive nitrogen as well. Once this occurs, a condensation

reaction where a water molecule is removed and allows the π bond to form between the

nitrogens. This has the added effect of making the molecule planar, indicative of trans

azobenzene. These steps resulted in the formation of compound 1 and compound 2 in a

41

70% yield. This kind of chemistry results in a symmetrical molecule, whereas the often

used diazonium groups may not have given the symmetry required. These diazonium

coupling reactions usually occur at the para position, but by-products could be formed as

well depending on the substituents present on the benzene ring.83

As shown in Scheme 3.1, both compounds were then protected with dimethoxytrityl

chloride. This is group is often used for the protection of primary alcohols in

oligonucleotide synthesis because it is easy to remove with weak acidic conditions, and the

formation of the carbocation results in a bright orange colour that make coupling easy to

monitor.83 Attaching a DMT group to a di-alcohol (diol) such as this can prove to be

problematic, since it possible to get double protection of both alcohols, rendering it useless

for the next step. This is reflected in the yields, which proved to be low for both compounds

3 and 4 at 35%. To try and mitigate this, one equivalent (or less) of DMT-Cl is used to try

and prevent double protection. The final step is the attachment of a phosphite group. This

can be synthetically challenging because the phosphite is sensitive to both heat and

moisture, and the ambient amount of water present in the atmosphere can be enough to

degrade it.84 The phosphite was attached to the other end of the spacer using a weak amine

base, triethylamine, at room temperature under an Argon atmosphere. This produced

compounds 5 and 6 in yields of 63% and 33% respectively. During oligonucleotide

synthesis the phosphite will be oxidized into a phosphate, the natural backbone of RNA.

42

Scheme 3.1. Synthesis of Azobenzene DMT-phosphoramidite Spacers

3.2 Photoswitching Ability of Azobenzene Derivatives, Lability and siRNAs

In order to confirm the azobenzene’s ability to change conformation the first

experiments that were run were absorbance spectra. As seen in Figure 3.4, exposure to UV

light causes a decrease in the maximum absorbance around 330 nm, and exposure to visible

light causes the absorbance to increase. This agrees with the literature showing that as the

trans isomer of azobenzene absorbs UV light, and converts to the cis isomer, absorbance

decreases. The opposite effect also occurs upon exposure to visible light with the cis form

absorbing the visible wavelength light and reconverting back to the more stable trans

form.64-67

43

Figure 3.3. Change in absorbance for Az1 when exposed to UV (right) and visible (left)

light

After photoswitching ability was confirmed, we then proceeded to synthesize the siRNAs

shown in Table 3.1 with Az1 and Az2 in various locations. After synthesis, we ran a new

absorbance spectra experiment with the expectation that there would be large maxima in 2

locations: at 260 nm since that is where nucleobases absorb most strongly, and at 330 nm,

where the azobenzene derivatives absorb. This is exactly what was found, and Figure 3.4

shows how the maxima change when exposed to UV (decreases) and visible (increases)

light. This is reflected in the literature as well69-74 and all siRNAs shared a similar profile

and ability to return to the trans conformer (See Appendix 1). A lability test was performed

to make sure the N=N double bond was robust and would not fail after multiple conversions

from trans to cis to trans.

44

Figure 3.4. Absorbance Profile of siRNA 7 when exposed to UV and Visible light.

Inset: Zoomed in portion of 300-380 nm highlighting Az1 changes

Using siRNA 7, the same sample was exposed to multiple iterations of alternating visible

and UV light and the relative absorbance was measured and is shown in Figure 3.5.

Figure 3.5. Photolability of siRNA 7 exposed to alternating visible and UV light

-0.003

-0.001

0.001

0.003

0.005

0.007

0.009

0.011

0 5 10 15 20 25 30 35 40

Rel

ativ

e C

han

ge

in A

bso

rban

ce (

Δ A

bs)

Time (min)

siRNA 7

Δ Abs

45

Each time point is 5 min of alternating visible UV exposure, and highlights the robust

nature of the conformational change. As can clearly be seen, there is no loss in the ability

of the azobenzene to interconvert even after several exposures.54,56 All of our siRNA

constructs had similar results (Appendix 1). In a paper from 1995, Ikeda et al. attempted to

exploit this lability to try and make more efficient liquid crystal films. They found not only

does the isomerization occur on the order of 10 ns, but that several laser pulses show no

loss of the ability to transform.90

Figure 3.6. HPLC chromatogram of siRNA 10 with cis and trans peaks indicated

To further characterize the trans and cis functionality, siRNA 10 was purified using

HPLC (High Pressure Liquid Chromotography). The chromatogram in Figure 3.7 clearly

shows two peaks, one at 16.8 min and one at 17.7. The trans peak is much larger, which

would be expected since at room temperature, the trans is the dominant form.59,92

46

Interestingly, exposure to UV light causes the peaks to shift not in retention time, but in

abundance and the cis peak grows and the trans peaks recedes ( Figure 3.8).

Figure 3.7. HPLC chromatogram of siRNA 10 measured immediately after 60 min UV

exposure with cis and trans peaks indicated

These results indicate that there are in fact two different conformers coexisting at

room temperature, and exposure to UV light causes the interconversion from trans to cis

but also confirms the cis conformer has a lower retention time than trans and agrees well

with the body of literature.92 Appendix 1 has the chromatograms for all central region

siRNAs avavilable, and show similar results.

3.3 Thermal Stability of siRNAs

The siRNA helix is stable because of two main types of interactions; hydrogen

bonding and base stacking.85 The melting temperature (Tm), is the temperature that there is

an equal amount of double stranded and single stranded RNA present, since as the

temperature increases, the two strands become unstable and dissociate. This is a useful

47

property when looking to measure how stable a RNA duplex is. More stable duplexes will

have higher Tm temperatures. Nucleobases have a maximum absorption at 260 nm, and this

makes it easy to measure the Tm since exposed nucleobases will have an increased

absorbance, much higher than their double stranded counterparts.87 Comparing thermal

stabilities to wild type siRNA can determine whether these azobenzene modified strands

are more or less stable and how that can affect activity of the siRNA. Table 3.1 displays

the sequences, modifications, Tm’s and the mass spectrometry results of the siRNA sense

strands.

Table 3.1. Sequences of anti-luciferase siRNAs, mass spec and Tm’s of siRNAs

containing the azobenzene spacers.

siRNA siRNA Duplex Predicted

Mass

Actual

Mass

αTm

(°C)

ΔTm

(°C)

wt 5’- C UUA CGC UGA GUA CUU CGAtt -3’ (s)

3’- ttG AAU GCG ACU CAU GAA GCU – 5’(as)

73.99 --

7 5’- C UUA CGC UAz1 GUA CUU CGAtt -3’ (s)

3’- ttG AAU GCG ACU CAU GAA GCU – 5’(as)

6316.81

6315.85

52.68 - 21.31

8 5’- C UUA CGC UGA GUA CUU CGA Az1 -3’ (s)

3’- ttG AAU GCG ACU CAU GAA GCU – 5’(as)

6382.84

6380.86

82.21

+ 8.22

9 5’- C UUA CGC Az2 A GUA CUU CGAtt -3’ (s)

3’- ttG AAU GCG ACU CAU GAA GCU – 5’(as)

6367.91

6366.89

62.75

- 11.24

10 5’- C UUA CGC U Az2 GUA CUU CGAtt -3’ (s)

3’- ttG AAU GCG ACU CAU GAA GCU – 5’(as)

6344.86

6344.91

64.78

- 9.21

11 5’- C UUA CGC UG Az2 UA CUU CGAtt -3’ (s)

3’- ttG AAU GCG ACU CAU GAA GCU – 5’(as)

6344.86

6344.86

54.56

- 19.43

12 5’- C UUA CGC UGA Az2 A CUU CGAtt -3’ (s)

3’- ttG AAU GCG ACU CAU GAA GCU – 5’(as)

6367.91

6366.95

57.71

- 16.28

13 5’- C UUA CGC UGA GUA CUU CGA Az2 -3’ (s)

3’- ttG AAU GCG ACU CAU GAA GCU – 5’(as)

6410.89

6410.86

82.72

+ 8.73

The azobenzene modification locations are Az1

for the 1 carbon species and Az2 for the 2 carbon

species

48

It should come as no surprise that replacing two natural nucleobases with an

unnatural molecule would cause disruption in both the base stacking and hydrogen bonding

of the duplex.88 This destabilization is reflected in the Tm measurements above, where every

case of a central modification resulted in a decrease of Tm compared to wild-type siRNA.

Interesting to note is that Az1 (siRNA 7) is twice as destabilizing as Az2

(siRNA 10) when replacing the Ago2 cleavage site at base pairs 9 and 10. Also interesting

is that moving the Az2 spacer towards the 3’-end causes it to be more destabilizing than

when replacing the cleavage site (siRNA 11 and 12). In the study done

by Desaulniers et al. (2017) their biphenyl modification had similar destabilization, 15 °C

less than wildtype.77 Depending on the position, the chemical modifications to the siRNA

can be more or less stabilizing than the biphenyl linker perhaps due to their supra position

in the major or minor groove. The major groove would likely be more accommodating than

the minor, and this could explain some of the difference.89 In all cases the same number of

hydrogen bonds are being replaced, one A-U and and C-G base pair, so that cannot be the

source of the variable destabilization. When exposed to UV light, there was found to be no

difference in Tm which was unexpected. If the proposed change from trans to cis prevents

activity in vivo then one would expect a resulting change in the Tm. This was not observed

however

despite the literature showing it occurring in several other types of nucleic acids.67,69 Figure

3.8 shows a proposed structural change when going from trans to cis.

As seen in Figure 3.8, a large amount hydrogen bonding is still occurring, and this

could help to explain why the melting temperatures remain stable even after exposure to

UV light. One more interesting thing was that both Az1 and Az2 (siRNAs 8 and 13

49

respectively) when placed at the 3’-end actually caused an increase in Tm of 8.22 - 8.73 °C.

This could simply be due to the azobenzene derivatives having a favourable interaction

Figure 3.8. Proposed Structural Change

with the anti sense strand as compared to the double thymine endcaps on each of the other

sense strands. A study by Perez-Rentero et al. 2013 indicates that pyrenes and other π-

conjugated compounds did increase the Tm by 2-5 °C.104 Another study by Brotschi et al.

in 2005 found that increasing biphenyl substituted deoxynucleotides centrally could

increase Tm by up to 10 °C .105 These 3’-end azobenzenes also have a free hydroxyl group

that may be able to hydrogen bond, as well being free to intercalate in between the bases

more favourably than the thymines.63,86

50

3.4 Circular Dichroism Conformation of siRNA Helical Structure

In order to determine whether the siRNAs maintained their A-form helical

conformation, CD (Circular Dichroism) spectra were measured. If the A-form helix is not

present, it is not able to be a tolerable substrate for RISC.93 Interactions between

asymmetrical molecules and circularly plane polarized light causes distinct absorption

patterns to emerge based on the secondary structure of nucleic acids and proteins.94

Figure 3.9. CD spectra of azobenzene modified spacers replacing two nucleobases

targeting firefly luciferase mRNAs. Wildtype and modified anti-firefly luciferase siRNAs

(10 μM/duplex) were suspended in 500 μL of a sodium phosphate buffer (90.0 mM NaCl,

10.0 mM Na2HPO4, 1.00 mM EDTA, pH 7.00) and scanned from 200-300 nm at 25 °C

with a screening rate of 20.0 nm/min and a 0.20 nm data pitch. All scans were performed

in triplicate and averaged using Jasco’s Spectra Manager version 2.

Figure 3.9 shows the CD spectra of the azobenzene modified duplexes

demonstrating a typical absorption pattern for an A- form helix. Each kind of helix (A, B

and Z) have similar but distinct absorption profiles. A-form helices as shown above shallow

-10

-8

-6

-4

-2

0

2

4

6

8

200 220 240 260 280 300

CD

[m

deg

]

Wavelength

siRNAs 7-13

siRNA 7

siRNA 8

siRNA 9

siRNA 10

siRNA 11

siRNA 12

siRNA 13

wt

51

troughs at around 210 nm and a large absorption maxima at 260 nm. B-form helices are

similar but have a 10 and 20 nm shift to the right (to 220 nm and 280 nm respectively). B-

form helices are most often exhibited by DNA duplexes.96 The Z-form has a distinct sharp

trough at 205 nm, a large peak at 260 nm and smaller trough around 290 nm, but is less

common than the other two forms.97 While there appears to be some small deviations in

the helices compared to the wildtype, these siRNAs are all confirmed as A-form helices

and should therefore be an active substrate for RISC.93

3.5 Silencing Capability of Azobenzene Modified siRNAs of the Exogenous Firefly

Luciferase Gene

After the battery of biophysical tests performed, it was determined that the modified

siRNAs would not only be compatible with RNAi, but that temporal and spacial control of

the siRNAs should be possible with light. To determine this the dual luciferase assay was

used. The dual luciferase assay is widely used as a way to test potential gene knockdown

of siRNAs on an easily detected exogenous target. These siRNAs target the firefly

luciferase mRNA (expressed with the pGL3 plasmid) while Renilla luciferase (SV40

plasmid) is used as an internal control to normalize the signal. When luciferase enzymes

from the assay kit are added to the substrate luciferin, they catalyze a light producing

reaction98 which is measured as luminescence.99,100 A control well was utilized on each

plate by co-transfecting a well with both plasmids only with no siRNA present. The control

was then used to express luciferase luminescence as a percent of the control. As shown in

Figure 3.10, the entire library of novel siRNAs were tested for activity at seven different

concentrations.

52

Figure 3.10. Gene silencing capability of azobenzene spacer modified siRNAs 7-13 and

wildtype targeting firefly luciferase mRNA at 8, 20, 40, 80, 160, 400 and 800 pM

concentrations and normalized to Renilla luciferase in HeLa cells and lysed 22 h post

transfection. Untreated control was co-transfected with both plasmids with no siRNA

present.

Each siRNA operates in a dose dependent manner, and in some cases, the 3’-end

modified siRNAs 8 and 13, is comparable to wild type. Despite high efficiency at large

doses (800 pM) decreasing the concentration of the siRNAs causes a fold decrease in

activity compared to wild type. This shows that even with high destabilization of the

duplex, these azobenzene modified siRNAs are well tolerated

within RISC as a substrate. This agrees with Addepalli et al.’s 2010 study in which they

presented findings that position based destabilization of the duplex can increase potency of

the RNAi substrate.101

Further investigation into the dose dependency produced the IC50 (Inhibitory

Concentration) values shown in table 3.2. They were calculated using Graphpad Prism 6

software.

0

20

40

60

80

100

120

140

Control 7 8 9 10 11 12 13 wtRel

ativ

e N

orm

aliz

ed E

xp

ress

ion o

f fi

refl

y

Luci

fera

se (

%)

siRNA

siRNAs 7-13 and wt

800 pM

400 pM

160 pM

80 pM

40 pM

20 pM

8 pM

53

Table 3.2. IC50 values for siRNAs 7-13 and wild type

siRNA IC50

(pM)

7 99.21

8 5.122

9 235.5

10 150.9

11 28.4

12 45.76

13 7.415

wt 4.685

Looking at Table 3.2, siRNAs 8 and 13 have the lowest IC50 values only slightly

above wild type. This would be expected since no destabilization occurred in these two

oligonucleotides. Interestingly, the next lowest values were for siRNAs 11 and 12 at 28.4

and 45.76 pM respectively. Both of these modifications were closer to the 5’-end of the

antisense strand, and that makes it likely the dose was lower since it was easier to unwind

the duplex100 and allowed RISC to select the guide strand because of the destabilizing

nature of Az2. The remaining oligonucleotide strands were from 99-235 pM either because

it was too destabilizing, or it was interfering with the selection of the guide strand

somehow.

3.6 Inactivation of the Silencing Capability of Azobenzene Modified siRNAs of the

Exogenous Firefly Luciferase Gene with UV Light

Having established good dose dependency of the siRNAs and its ability to function

supported, the next experiment was designed to inactive the siRNAs with UV light. This

proved to be challenging, since exposure to UV light is detrimental to cell cultures. UV

light is damaging to DNA and to other cellular functions, especially in cell cultures.102 To

bypass the difficulty of exposing cell culture to UV light, the siRNAs were pre-inactivated

54

prior to transfection with the UV lamp. The subsequent transfection was then carried out

in the dark, to prevent reactivation by visible light. A second difficulty also had to be

overcome and that was the rate of thermal relaxation at 37 °C.60 Since the cis form is not

thermally stable at physiological temperatures, these inactivation assays had to be

performed at 8 h time points instead of the normal 24 h time points. Time points past 8 h

showed reduced effectiveness of the inactivation of the siRNA as the azobenzene slowly

reactivated itself through thermal conversion of cis to trans isoforms. Despite these two

major hurdles, pre-inactivation was found to be effective, Figure 3.12 supports inactivation

of the siRNA was successful, in some cases to near control levels. The data for siRNAs 7,

11 and 12 with concentrations of 800, 400 and 80.0 pM were selected to showcase this

effect because the other siRNAs were ineffective at this time point, and lower

concentrations below 80 pM showed no difference between UV exposure and untreated

samples.

Figure 3.11. Gene silencing capability of azobenzene spacer modified siRNA targeting

firefly luciferase mRNA and normalized to Renilla luciferase for siRNAs 7, 11 and 12 at a

80, 400 and 800 pM concentrations in HeLa cells and lysed 16 h post transfection. UV

corresponds to the siRNA being exposed under a 365nm UV lamp for inactivation prior to

transfection. No light corresponds to siRNAs being transfected in HeLa cells in the absence

0

20

40

60

80

100

120

control 7 800

pM

7 400

pM

7 80

pM

11 800

pM

11 400

pM

11 80

pM

12 800

pM

12 400

pM

12 80

pM

Rel

ativ

e N

orm

aliz

ed E

xp

ress

ion o

f

fire

fly

Luci

fera

se (

%)

Concentration

siRNAs 7, 11 and 12

UV

No light

55

of both UV and visible light. Untreated control was co-transfected with both plasmids with

no siRNA present.

Showcasing Figure 3.11, the trend is easy to see, exposure to UV light causes the

siRNA to become ineffective. This is shown across 3 different siRNAs (7, 11, 12) and 3

concentrations (80, 400 and 800 pM). The dose dependency is harder to see with UV

inactivated wells, but generally lower doses resulted in higher luciferase levels after UV

treatment. Interesting to note is that siRNA 7 with the Az1 spacer was able to get near

control levels after inactivation even at 800 pM. This may be due to the fact that spacer

itself is 2 carbons shorter, and so has less flexibility when in the cis conformer resulting in

an impressive inactivation.103 These results support the inactivation of the siRNAs after

UV light exposure, and the next step is reactivation with visible light

3.6 Reactivation of the Silencing Capability of Azobenzene Modified siRNAs of the

Exogenous Firefly Luciferase Gene with UV Light

In order to test the reactivation of the siRNAs, the experiment was set up as

previous, but with one major change: one plate was exposed to visible light in the incubator

4 h after transfection, and then all plates were lysed at the 8 h time point and compared.

56

Figure 3.12. Reduction in normalized firefly luciferase expression for siRNA 7 at

160 and 800 pM in HeLa cells monitored 8 and 12 hours post-transfection. 1) UV

corresponds to the siRNA being exposed under a 365nm UV lamp for inactivation prior to

transfection. 2) Vis corresponds to the siRNA being exposed under a 365nm UV lamp for

inactivation prior to transfection, however the transfected cells were exposed to a 13 W

daylight lamp 4 hours post-transfection for the remainder of the assay. 3) No light

corresponds to siRNAs being transfected in HeLa cells in the absence of both UV and

visible light. Untreated control was co-transfected with both plasmids with no siRNA

present.

In figure 3.12, which shows the data for siRNA 7, the only Az1 internal spacer

were able to synthesize, UV corresponds to the previous treatment for inactivation, and Vis

is the visible light treated plate 4 h post transfection while no light is the untreated plate

with baseline levels of activity. The concentrations shown were selected in order to

demonstrate that high doses (800 pM) of siRNA does not cause significant cell death, and

a lower concentration is able to be inactivated and reactivated successfully. The time points

were chosen because 8 h was the earliest time point that showed significant accumulation

of the luciferin proteins, and the 12 h ( 1 half life) time point was selected to show the effect

of the thermal relaxation on UV inactivation . Starting with the 800 pM concentration, both

0

20

40

60

80

100

120

140

160

control 8h 800 pM 12h 800 pM 8h 160 pM 12h 160 pM

Rel

ativ

e N

orm

aliz

ed E

xp

ress

ion o

f

fire

fly

Luci

fera

se (

%)

Concentration and Time

siRNA 7

UV

Vis

No light

57

time points exhibit clear trends where UV inactivation has higher levels of luciferin than

both the reactivated visible treated cells and the untreated no light cells. In both cases the

Vis treated cells have levels in between UV and no light, but in the case of the 12 h time

point all activity levels are much lower than the 8 h time point, and this is due to the thermal

relaxation mentioned previously. With a half life of 4 h the 12 h time point could have as

much as twice as much active siRNA resulting in the lower expression levels.60 In contrast

the 160 pM concentration exhibits its own trends. While UV inactivation was higher than

both visible and untreated cells, the visible and no light treatments gave comparable levels

of activity. This could have been due to having less siRNA present to be deactivated, or

that it was being degraded at a similar rate for both treatments.

Figure 3.13. Reduction in normalized firefly luciferase expression for siRNA 9 at

160 and 800 pM in HeLa cells monitored 8 and 12 hours post-transfection. 1) UV

corresponds to the siRNA being exposed under a 365nm UV lamp for inactivation prior to

transfection. 2) Vis corresponds to the siRNA being exposed under a 365nm UV lamp for

inactivation prior to transfection, however the transfected cells were exposed to a 13 W

daylight lamp 4 hours post-transfection for the remainder of the assay. 3) No light

corresponds to siRNAs being transfected in HeLa cells in the absence of both UV and

visible light. Untreated control was co-transfected with both plasmids with no siRNA

present.

0

20

40

60

80

100

120

control 8h 800 pM 12h 800 pM 8h 160 pM 12h 160 pM

Rel

ativ

e N

orm

aliz

ed E

xp

ress

ion o

f

fire

fly

Luci

fera

se (

%)

Concentration and Time

siRNA 9

UV

Vis

No light

58

Results for siRNA 9 shown in Figure 3.13, follow the same trends discussed

previously, but with better results at the 160 pM, 8 h time point. This could be due to the

the Az2 spacer, or that the location of the spacer was more favourable than in siRNA 7.

Showing the same trend in a second siRNA, a clear inactivation effect under UV light, is

observed and a clear reactivation when exposed to visible light.

Figure 3.14. Reduction in normalized firefly luciferase expression for siRNA 10 at

160 and 800 pM in HeLa cells monitored 8 and 12 hours post-transfection. 1) UV

corresponds to the siRNA being exposed under a 365nm UV lamp for inactivation prior to

transfection. 2) Vis corresponds to the siRNA being exposed under a 365nm UV lamp for

inactivation prior to transfection, however the transfected cells were exposed to a 13 W

daylight lamp 4 hours post-transfection for the remainder of the assay. 3) No light

corresponds to siRNAs being transfected in HeLa cells in the absence of both UV and

visible light. Untreated control was co-transfected with both plasmids with no siRNA

present.

0

20

40

60

80

100

120

control 8h 800 pM 12h 800 pM 8h 160 pM 12h 160 pM

Rel

ativ

e N

orm

aliz

ed E

xp

ress

ion o

f

fire

fly

Luci

fera

se (

%)

Concentration and Time

siRNA 10

UV

Vis

No light

59

Figure 3.15..Reduction in normalized firefly luciferase expression for siRNA 11 at

160 and 800 pM in HeLa cells monitored 8 and 12 hours post-transfection. 1) UV

corresponds to the siRNA being exposed under a 365nm UV lamp for inactivation prior to

transfection. 2) Vis corresponds to the siRNA being exposed under a 365nm UV lamp for

inactivation prior to transfection, however the transfected cells were exposed to a 13 W

daylight lamp 4 hours post-transfection for the remainder of the assay. 3) No light

corresponds to siRNAs being transfected in HeLa cells in the absence of both UV and

visible light. Untreated control was co-transfected with both plasmids with no siRNA

present.

0

20

40

60

80

100

120

control 8h 800 pM 12h 800 pM 8h 160 pM 12h 160 pM

Rel

ativ

e N

orm

aliz

ed E

xp

ress

ion o

f

fire

fly

Luci

fera

se (

%)

Concentration and Time

siRNA 11

UV

Vis

No light

60

Figure 3.16. Reduction in normalized firefly luciferase expression for siRNA 12 at

160 and 800 pM in HeLa cells monitored 8 and 12 hours post-transfection. 1) UV

corresponds to the siRNA being exposed under a 365nm UV lamp for inactivation prior to

transfection. 2) Vis corresponds to the siRNA being exposed under a 365nm UV lamp for

inactivation prior to transfection, however the transfected cells were exposed to a 13 W

daylight lamp 4 hours post-transfection for the remainder of the assay. 3) No light

corresponds to siRNAs being transfected in HeLa cells in the absence of both UV and

visible light. Untreated control was co-transfected with both plasmids with no siRNA

present.

Figures 3.14-3.16 shows the remaining siRNAs with internally located Az2 spacer

molecules.It appears as the Az2 linker moves towards the 3‘-end of the sense strand, that

the trends start to normalize until siRNA 12, where each time point and each concentration

makes have the same trend, where UV > Vis > No light treated cells.

Introduction of azobenzene spacers causes significant destabilization in the siRNA

duplex and the data supports that this destabilization can be utilized to create a spacially

and temporally controlled siRNA through the azobenzene moiety in and around the central

region of the sense strand. These results look to be very promising, and work will likely

continue down this path for some time to come.

0

20

40

60

80

100

120

control 8h 800 pM 12h 800 pM 8h 160 pM 12h 160 pM

Rel

ativ

e N

orm

aliz

ed E

xp

ress

ion o

f

fire

fly

Luci

fera

se (

%)

Concentration and Time

siRNA 12

UV

Vis

No light

61

4.0 Future Directions

These azobenzene siRNAs have proven to be effective RISC substrates, and further

more exhibit excellent photochemical properties when exposed to light. These

photochemical properties could be further explored in order make the siRNAs better

pharmaceuticals. One such way would be to red-shift the trans to cis photo-isomerization

out of the UV portion of the spectrum. UV light is toxic to cells and does not penetrate very

far into biological tissues, which makes for a poor inactivation method in living subjects.

Recent research has shown that it is possible to control the wavelengths that the N=N bond

utilizes for photo-isomerization for in vivo applications by shifting the band from 330 nm

to 530 nm resulting in thermally stable isomers.106 In a different study, Dong et al. 2015

synthesized several azobenzene derivatives with peak absorbance well past the UV portion

of the spectrum into the near red 600 nm range.107 They accomplished this by adding

methoxy substituents at the four ortho positions flanking the N=N bond. This could be

useful in transfection assays, since we would be able to expose the cell culture to the light

indefinitely with little to no toxic effects. This would greatly improve temporal control of

the siRNAs, but also open other avenues to explore that would not be possible with a UV

light isomerization.

A second pathway of investigation would be to further test these azobenzene

modifications for stability inside the cell. Many kinds of end cap provide nuclease

resistance to siRNAs,3,42 but this one has the added functionality of being photo-

isomerizable. The effect of UV and visible light could be tested on these siRNAs to see if

silencing efficacy is affected, or whether central region or 3’-end modifications show

differences in their nuclease stability.

62

Other possible investigative paths include adding multiple azobenzene

modifications into a single sense strand to make it more effective at lower concentrations

by causing multiple disruptions in the duplex, or cytotoxicity assays to determine the

effects of high doses of siRNA on the cell population. Modification of the anti-sense strand

instead of the sense strand to determine if incorporation into RISC is still possible. More

interesting possibilities include molecular modelling to see how these modifications affect

the helix in both the cis and trans form, but also how they interact with bases and how

favourable base stacking is, especially once ortho substitutions become a factor, bulking

the molecule up and perhaps making it a poor substrate for RISC. Finally, testing these

siRNAs in a clinically relevant way, such as with the BCL-2 gene would open the door to

not only treating cancer, but also many other types of endogenous targets to prevent or cure

diseases.

Some challenges yet remain however. Because of the thermal instability of cis

conformers, it can be difficult to find a switch that remains stable at physiological

temperature. Azobenzenes can also be reduced under physiological conditions, so even if

thermal stability is good, this issue remains a constant hurdle.108

63

5.0 Conclusions

This study presents a logical next step into the realm of photo-controlled siRNAs

through central region azobenzene modifications. Shown here is the ability to inactivate

and then re-activate siRNA activity through UV and visible light in order to provide a great

deal of temporal control over mRNA expression through RISC. In addition to this, several

siRNAs were shown to have IC50 very close to wild-type showing it could be possible to

construct siRNAs with a low dose potential, reducing toxicity and off-target effects.

Incorporation of these photo-switchable azobenzenes have incorporated several beneficial

properties on these siRNAs: not just a photo tunable control for activity, but also

applications in introducing site specific destabilization to strategically control which part

of the siRNA we would like to inactivate. With several different azobenzene modifications

that are controlled with different, specific wavelengths of light oligonucleotides could be

the subject of site specific activation or inactivation at different times or in different

locations. This would have widespread applicability in gene expression, transcriptional or

translational controls and ribosomal RNA modification potential.

64

References

1. Alberts B.(2008). Molecular biology of the cell. New York: Garland Science.

ISBN 0-8153-4105-9

2. Li, S. Mason, C. E., The Pivotal Regulatory Landscape of RNA Modifications. In

Annual Review of Genomics and Human Genetics, Vol 15, Chakravarti, A.;

Green, E., Eds. Annual Reviews: Palo Alto, 2014; Vol. 15, pp 127-150.

3. Egli, M. Minasov, G. Tereshko, V. Pallan, P. S. Teplova, M. Inamati, G. B.

Lesnik, E. A. Owens, S. R. Ross, B. S. Prakash, T. P. Manoharan, M., Probing the

influence of stereoelectronic effects on the biophysical properties of

oligonucleotides: Comprehensive analysis of the RNA affinity, nuclease

resistance, and crystal structure of ten 2 '-O-ribonucleic acid modifications.

Biochemistry 2005, 44 (25), 9045-9057.

4. Mack, G. S., MicroRNA gets down to business. Nature Biotechnology 2007, 25

(6), 631-638.

5. Higgs, P. G., RNA secondary structure: physical and computational aspects.

Quarterly Reviews of Biophysics 2000, 33 (3), 199-253.

6. Parkinson, G. N. Lee, M. P. H. Neidle, S., Crystal structure of parallel

quadruplexes from human telomeric DNA. Nature 2002, 417 (6891), 876-880.

7. Burge, S. Parkinson, G. N. Hazel, P. Todd, A. K. Neidle, S., Quadruplex DNA:

sequence, topology and structure. Nucleic Acids Research 2006, 34 (19), 5402-

5415.

8. Frank, T. D., Front waves in the early RNA world: The Schlogl model and the

logistic growth model. Journal of Theoretical Biology 2016, 392, 62-68.

9. Allison,L. (2007). Fundamental Molecular Biology. London: Blackwell

Publishing. ISBN 978-1-4051-0379-4.

10. Griffiths, A. (2008). Introduction to Genetic Analysis (9th ed.). New York: W.H.

Freeman and Company. ISBN 978-0-7167-6887-6.

11. Wei, X. P. Ghosh, S. K. Taylor, M. E. Johnson, V. A. Emini, E. A. Deutsch, P.

Lifson, J. D. Bonhoeffer, S. Nowak, M. A. Hahn, B. H. Saag, M. S. Shaw, G. M.,

65

Viral dynamics in human immunodeficiency-virus type-1 infection. Nature 1995,

373 (6510), 117-122.

12. Mattaj, I. W. Englmeier, L., Nucleocytoplasmic transport: The soluble phase.

Annual Review of Biochemistry 1998, 67, 265-306.

13. Napoli, C. Lemieux, C. Jorgensen, R., Introduction of a chimeric chalcone

Synthase Gene into Petunia results in reversible co-suppression of homologous

genes in trans. Plant Cell 1990, 2 (4), 279-289.

14. Vanblokland, R. Vandergeest, N. Mol, J. N. M. Kooter, J. M., Transgene-

mediated suppression of chalcone synthase expression in petunia-hybrida results

from and increase in RNA turnover. Plant Journal 1994, 6 (6), 861-877.

15. PalBhadra, M. Bhadra, U. Birchler, J. A., Cosuppression in Drosophila: Gene

silencing of Alcohol dehydrogenase by white-Adh transgenes is Polycomb

dependent. Cell 1997, 90 (3), 479-490.

16. Fire, A. Xu, S. Q. Montgomery, M. K. Kostas, S. A. Driver, S. E. Mello, C. C.,

Potent and specific genetic interference by double-stranded RNA in

Caenorhabditis elegans. Nature 1998, 391 (6669), 806-811.

17. Elbashir, S. M. Harborth, J. Lendeckel, W. Yalcin, A. Weber, K. Tuschl, T.,

Duplexes of 21-nucleotide RNAs mediate RNA interference in cultured

mammalian cells. Nature 2001, 411 (6836), 494-498.

18. Daniels, S. M. Gatignol, A., The Multiple Functions of TRBP, at the hub of cell

responses to viruses, stress, and cancer. Microbiology and Molecular Biology

Reviews 2012, 76 (3), 652.

19. Schirle, N. T. MacRae, I. J., The crystal structure of human argonaute2. Science

2012, 336 (6084), 1037-1040.

20. Zhuang, X. W., Visualizing infection of individual influenza viruses. Abstracts of

Papers of the American Chemical Society 2003, 226, U299-U299.

21. Joo, K. I. Wang, P., Visualization of targeted transduction by engineered lentiviral

vectors. Gene Therapy 2008, 15 (20), 1384-1396.

22. Amarzguioui, M. Lundberg, P. Cantin, E. Hagstrom, J. Behlke, M. A. Rossi, J. J.,

Rational design and in vitro and in vivo delivery of Dicer substrate siRNA.

Nature Protocols 2006, 1 (2), 508-517.

66

23. Hutvagner, G. McLachlan, J. Pasquinelli, A. E. Balint, E. Tuschl, T. Zamore, P.

D., A cellular function for the RNA-interference enzyme Dicer in the maturation

of the let-7 small temporal RNA. Science 2001, 293 (5531), 834-838.

24. Liu, J. D. Carmell, M. A. Rivas, F. V. Marsden, C. G. Thomson, J. M. Song, J. J.

Hammond, S. M. Joshua-Tor, L. Hannon, G. J., Argonaute2 is the catalytic engine

of mammalian RNAi. Science 2004, 305 (5689), 1437-1441.

25. Wang, Y. L. Juranek, S. Li, H. T. Sheng, G. Wardle, G. S. Tuschl, T. Patel, D. J.,

Nucleation, propagation and cleavage of target RNAs in Ago silencing

complexes. Nature 2009, 461 (7265), 754-U3

26. Prakash, T. P. Allerson, C. R. Dande, P. Vickers, T. A. Sioufi, N. Jarres, R. Baker,

B. F. Swayze, E. E. Griffey, R. H. Bhat, B., Positional effect of chemical

modifications on short interference RNA activity in mammalian cells. Journal of

Medicinal Chemistry 2005, 48 (13), 4247-4253.

27. Aagaard, L. Rossi, J. J., RNAi therapeutics: Principles, prospects and challenges.

Advanced Drug Delivery Reviews 2007, 59 (2-3), 75-86.

28. Wilner, S. E. Levy, M., Synthesis and Characterization of Aptamer-Targeted

SNALPs for the Delivery of siRNA. Methods in molecular biology (Clifton, N.J.)

2016, 1380, 211-24.

29. Collingwood, M. A. Rose, S. D. Huang, L. Y. Hillier, C. Amarzguioui, M.

Wiiger, M. T. Soifer, H. S. Rossi, J. J. Behlke, M. A., Chemical modification

patterns compatible with high potency Dicer-substrate small interfering RNAs.

Oligonucleotides 2008, 18 (2), 187-199.

30. Oivanen, M. Kuusela, S. Lonnberg, H., Kinetics and mechanisms for the cleavage

and isomerization of the phosphodiester bonds of RNA by Bronsted acids and

bases. Chemical Reviews 1998, 98 (3), 961-990.

31. Bibillo, A. Figlerowicz, M. Kierzek, R., The non-enzymatic hydrolysis of

oligoribonucleotides VI. The role of biogenic polyamines. Nucleic Acids Research

1999, 27 (19), 3931-3937.

32. Kaukinen, U. Lyytikainen, S. Mikkola, S. Lonnberg, H., The reactivity of

phosphodiester bonds within linear single-stranded oligoribonucleotides is

strongly dependent on the base sequence. Nucleic Acids Research 2002, 30 (2),

468-474.

33. Singer, S. J. Nicolson, G. L., Fluid mosaic model of structure of cell-membranes.

Science 1972, 175 (4023), 720-&.

67

34. Deleavey, G. F. Damha, M. J., Designing chemically modified Oligonucleotides

for Targeted Gene Silencing. Chemistry & Biology 2012, 19 (8), 937-954.

35. Jackson, A. L. Burchard, J. Schelter, J. Chau, B. N. Cleary, M. Lim, L. Linsley, P.

S., Widespread siRNA "off-target'' transcript silencing mediated by seed region

sequence complementarity. Rna-a Publication of the Rna Society 2006, 12 (7),

1179-1187.

36. Vanboom, J. H. Dahl, O. Marugg, J. E. Nielsen, J., Side products in nucleotide

chemistry. Phosphorus Sulfur and Silicon and the Related Elements 1987, 30 (3-

4), 819-819.

37. Ohtsuka, E. Ikehara, M. Soll, D., Recent developments in the chemical synthesis

of polynucleotides. Nucleic Acids Research 1982, 10 (21), 6553-6570.

38. Sinha, N. D. Biernat, J. McManus, J. Koster, H., Polymer support oligonucleotide

synthesis. 18. Use of beta-cyanoethyl-N,N-dialkylamino-N-morpholino

phosphoramidite of deoxynucleosides for the synthesis of DNA fragments

simplifying deprotection and isolation of the final product. Nucleic Acids

Research 1984, 12 (11), 4539-4557.

39. Iyer, R. P. Kuchimanchi, S. N. Pandey, R. K., RNA interference: an exciting new

approach for target validation, gene expression analysis and therapeutics. Drugs of

the Future 2003, 28 (1), 51-59.

40. Beaucage, S. L., Solid-phase synthesis of siRNA oligonucleotides. Current

Opinion in Drug Discovery & Development 2008, 11 (2), 203-216.

41. Stein, C. A. Cheng, Y. C., Antisense oligonucleotides as therapeutic agents-is the

bullet really magical. Science 1993, 261 (5124), 1004-1012.

42. Eckstein, F., Phosphorothioate oligodeoxynucleotides: What is their origin and

what is unique about them? Antisense & Nucleic Acid Drug Development 2000,

10 (2), 117-121.

43. Bennett, C. F. Swayze, E. E., RNA Targeting Therapeutics: Molecular

Mechanisms of Antisense Oligonucleotides as a Therapeutic Platform. In Annual

Review of Pharmacology and Toxicology, Annual Reviews: Palo Alto, 2010, 259-

293.

44. Inoue, H. Hayase, Y. Imura, A. Iwai, S. Miura, K. Ohtsuka, E., Synthesis and

hybridization studies on 2 complementary nona(2’-O-methyl) ribonucleotides.

Nucleic Acids Research 1987, 15 (15), 6131-6148.

68

45. Guckian, K. M. Krugh, T. R. Kool, E. T., Solution structure of a DNA duplex

containing a replicable difluorotoluene-adenine pair. Nature Structural Biology

1998, 5 (11), 954-959.

46. Moran, S. Ren, R. X. F. Rumney, S. Kool, E. T., Difluorotoluene, a nonpolar

isostere for thymine, codes specifically and efficiently for adenine in DNA

replication. Journal of the American Chemical Society 1997, 119 (8), 2056-2057.

47. Fire, A. Xu, S. Q. Montgomery, M. K. Kostas, S. A. Driver, S. E. Mello, C. C.,

Potent and specific genetic interference by double-stranded RNA in

Caenorhabditis elegans. Nature 1998, 391 (6669), 806-811.

48. Siddiqi, S. M. Jacobson, K. A. Esker, J. L. Olah, M. E. Ji, X. D. Melman, N.

Tiwari, K. N. Secrist, J. A. Schneller, S. W. Cristalli, G. Stiles, G. L. Johnson, C.

R. Ijzerman, A. P., Search for new purine-modified and ribose-modified

adenosine-analogs as selective agonists and antagonists at adenosine receptors.

Journal of Medicinal Chemistry 1995, 38 (7), 1174-1188.

49. Pallan, P. S. Greene, E. M. Jicman, P. A. Pandey, R. K. Manoharan, M. Rozners,

E. Egli, M., Unexpected origins of the enhanced pairing affinity of 2 '-fluoro-

modified RNA. Nucleic Acids Research 2011, 39 (8), 3482-3495.

50. Xia, J. Noronha, A. Toudjarska, I. Li, F. Akinc, A. Braich, R. Frank-Kamenetsky,

M. Rajeev, K. G. Egli, M. Manoharan, M., Gene silencing activity of siRNAs

with a ribo-difluorotoluyl nucteotide. Acs Chemical Biology 2006, 1 (3), 176-183.

51. Efthymiou, T. Gong, W. Desaulniers, J. P., Chemical Architecture and

Applications of Nucleic Acid Derivatives Containing 1,2,3-Triazole

Functionalities Synthesized via Click Chemistry. Molecules 2012, 17 (11), 12665-

12703.

52. Singh, S. K. Nielsen, P. Koshkin, A. A. Wengel, J., LNA (locked nucleic acids):

synthesis and high-affinity nucleic acid recognition. Chemical Communications

1998, (4), 455-456.

53. Filipowicz, W., RNAi: The nuts and bolts of the RISC machine. Cell 2005, 122

(1), 17-20.

54. Efthymiou, T. C. Peel, B. Huynh, V. Desaulniers, J. P., Evaluation of siRNAs that

contain internal variable-length spacer linkages. Bioorganic & Medicinal

Chemistry Letters 2012, 22 (17), 5590-5594.

55. Zimmerman, G. Chow, L. Y. Paik, U. J., The photochemical isomerization of

azobenzene. Journal of the American Chemical Society 1958, 80 (14), 3528-3531.

69

56. Tait, K. M. Parkinson, J. A. Bates, S. P. Ebenezer, W. J. Jones, A. C., The novel

use of NMR spectroscopy with in situ laser irradiation to study azo

photoisomerisation. Journal of Photochemistry and Photobiology a-Chemistry

2003, 154 (2-3), 179-188.

57. Bandara, H. M. D. Burdette, S. C., Photoisomerization in different classes of

azobenzene. Chemical Society Reviews 2012, 41 (5), 1809-1825.

58. Beharry, A. A. Woolley, G. A., Azobenzene photoswitches for biomolecules.

Chemical Society Reviews 2011, 40 (8), 4422-4437.

59. Hartley, G. S., 113 The cis-form of azobenzene and the velocity of the thermal cis

-> trans-conversion of azobenzene and some derivatives. Journal of the Chemical

Society 1938, 633-642.

60. Yamana, K. Kan, K. Nakano, H., Synthesis of oligonucleotides containing a new

azobenzene fragment with efficient photoisomerizability. Bioorganic & Medicinal

Chemistry 1999, 7 (12), 2977-2983.

61. Ikeda, T. Tsutsumi, O., Optical switching and image storage by means of

azobenzene liquid-crystal films. Science 1995, 268 (5219), 1873-1875.

62. Natansohn, A. Rochon, P., Photoinduced motions in azo-containing polymers.

Chemical Reviews 2002, 102 (11), 4139-4175.

63. Bartels, E. Wasserma.Nh Erlanger, B. F., Studies Of Acetylcholine Receptor

using Photochromic Activators. Abstracts of Papers of the American Chemical

Society 1971, (NSEP), 303-&.

64. Woolley, G. A., Photocontrolling peptide alpha helices. Accounts of Chemical

Research 2005, 38 (6), 486-493.

65. Ferreira, R. Nilsson, J. R. Solano, C. Andreasson, J. Grotli, M., Design, Synthesis

and Inhibitory Activity of Photoswitchable RET Kinase Inhibitors. Scientific

Reports 2015, 5, 8.

66. Goldau, T. Murayama, K. Brieke, C. Asanuma, H. Heckel, A., Azobenzene C-

Nucleosides for Photocontrolled Hybridization of DNA at Room Temperature.

Chemistry-a European Journal 2015, 21 (49), 17870-17876.

67. Asanuma, H. Liang, X. G. Yoshida, T. Komiyama, M., Photocontrol of DNA

duplex formation by using azobenzene-bearing oligonucleotides. Chembiochem

2001, 2 (1), 39-44.

70

68. Wu, L. Wu, Y. Jin, H. W. Zhang, L. R. He, Y. J. Tang, X. J., Photoswitching

properties of hairpin ODNs with azobenzene derivatives at the loop position.

Medchemcomm 2015, 6 (3), 461-468.

69. Paddison, P. J. Caudy, A. A. Bernstein, E. Hannon, G. J. Conklin, D. S., Short

hairpin RNAs (shRNAs) induce sequence-specific silencing in mammalian cells.

Genes & Development 2002, 16 (8), 948-958.

70. Wu, L. He, Y. J. Tang, X. J., Photoregulating RNA Digestion Using Azobenzene

Linked Dumbbell Antisense Oligodeoxynucleotides. Bioconjugate Chemistry

2015, 26 (6), 1070-1079.

71. Yu, H. T. Li, J. B. Wu, D. D. Qiu, Z. J. Zhang, Y., Chemistry and biological

applications of photo-labile organic molecules. Chemical Society Reviews 2010,

39 (2), 464-473.

72. Pelliccioli, A. P. Wirz, J., Photoremovable protecting groups: reaction

mechanisms and applications. Photochemical & Photobiological Sciences 2002, 1

(7), 441-458.

73. Zhang, Q. Ko, N. R. Oh, J. K., Recent advances in stimuli-responsive degradable

block copolymer micelles: synthesis and controlled drug delivery applications.

Chemical Communications 2012, 48 (61), 7542-7552.

74. Hiraoka, T. Hamachi, I., Caged RNase: Photoactivation of the enzyme from

perfect off-state by site-specific incorporation of 2-nitrobenzyl moiety.

Bioorganic & Medicinal Chemistry Letters 2003, 13 (1), 13-15.

75. Anstaett, P. Pierroz, V. Ferrari, S. Gasser, G., Two-photon uncageable enzyme

inhibitors bearing targeting vectors. Photochemical & Photobiological Sciences

2015, 14 (10), 1821-1825.

76. Mikat, V. Heckel, A., Light-dependent RNA interference with nucleobase-caged

siRNAs. Rna-a Publication of the Rna Society 2007, 13 (12), 2341-2347.

77. McKim, Chris. 2014 Msc thesis, University of Ontario Institute of Technology.

78. Desaulniers, J. P. Hagen, G. Anderson, J. McKim, C. Roberts, B., Effective gene-

silencing of siRNAs that contain functionalized spacer linkages within the central

region. Rsc Advances 2017, 7 (6), 3450-3454.

71

79. Peel, B. J. Hagen, G. Krishnamurthy, K. Desaulniers, J. P., Conjugation and

Evaluation of Small Hydrophobic Molecules to Triazole-Linked siRNAs. Acs

Medicinal Chemistry Letters 2015, 6 (2), 117-122.

80. Jackson, A. L.; Linsley, P. S., Recognizing and avoiding siRNA off-target effects

for target identification and therapeutic application. Nature Reviews Drug

Discovery 2010, 9 (1), 57-67.

81. Vogler, M. Dinsdale, D. Dyer, M. J. S. Cohen, G. M., Bcl-2 inhibitors: small

molecules with a big impact on cancer therapy. Cell Death and Differentiation

2009, 16 (3), 360-367.

82. Still, W. C. Kahn, M. Mitra, A., Rapid Chromatographic Technique for

Preparative Separations with Moderate Resolution. Journal of Organic Chemistry

1978, 43 (14), 2923-2925.

83. McMurry, J.(2012). Oragnic Chemistry. Brooks/Cole, Cengage Learning. ISBN

13: 978-0-8400-5444-0

84. Bessodes, M. Komiotis, D. Antonakis, K., Rapid and Selective Detritylation of

Primary Alcohols using Formic-acid. Tetrahedron Letters 1986, 27 (5), 579-580.

85. Kupihar, Z. Varadi, G. Monostori, E. Toth, G. K., Preparation of an

asymmetrically protected phosphoramidite and its application in solid-phase

synthesis of phosphopeptides. Tetrahedron Letters 2000, 41 (22), 4457-4461.

86. Guckian, K. M. Schweitzer, B. A. Ren, R. X. F. Sheils, C. J. Paris, P. L.

Tahmassebi, D. C. Kool, E. T., Experimental measurement of aromatic stacking

affinities in the context of duplex DNA. Journal of the American Chemical

Society 1996, 118 (34), 8182-8183.

87. Nygaard, A. P. Hall, B. D., A Method for Detection of RNA-DNA Complexes.

Biochemical and Biophysical Research Communications 1963, 12 (2), 98.

88. Schweitzer, B. A. Kool, E. T., Hydrophobic, Non-Hydrogen-Bonding Bases and

Base-Pairs in DNA. Journal of the American Chemical Society 1995, 117 (7),

1863-1872.

89. Bramsen, J. B. Pakula, M. M. Hansen, T. B. Bus, C. Langkjaer, N. Odadzic, D.

Smicius, R. Wengel, S. L. Chattopadhyaya, J. Engels, J. W. Herdewijn, P.

Wengel, J. Kjems, J., A screen of chemical modifications identifies position-

specific modification by UNA to most potently reduce siRNA off-target effects.

Nucleic Acids Research 2010, 38 (17), 5761-5773.

72

90. Asanuma, H. Kashida, H. Kamiya, Y., De Novo Design of Functional

Oligonucleotides with Acyclic Scaffolds. Chemical Record 2014, 14 (6), 1055-

1069.

91. Ikeda, T. Tsutsumi, O., Optical Switching and Image Storage by Means of

Azobenzene Liquid-Crystal Films.Science 1995, 268 (5219), 1873-1875.

92. Tamai, N. Miyasaka, H., Ultrafast dynamics of photochromic systems. Chemical

Reviews 2000, 100 (5), 1875-1890.

93. Morihiro, K. Hasegawa, O. Mori, S. Tsunoda, S. Obika, S., C5-azobenzene-

functionalized locked nucleic acid uridine: isomerization properties, hybridization

ability, and enzymatic stability. Organic & Biomolecular Chemistry 2015, 13

(18), 5209-5214.

94. Chendrimada, T. P. Gregory, R. I. Kumaraswamy, E. Norman, J. Cooch, N.

Nishikura, K. Shiekhattar, R., TRBP recruits the Dicer complex to Ago2 for

microRNA processing and gene silencing. Nature 2005, 436 (7051), 740-744.

95. F. Wold. (1971). Macomolecules: Structure and Function. New York: Garland

Science.

96. J.C. Maurizot. (2000). Circular dichroism of nucleic acids: nonclassical

conformations and modified oligonucleotides. Circular Dichroism: Principals

and Applications. New York: Wiley-VCH.

97. Kypr, J. Kejnovska, I. Renciuk, D. Vorlickova, M., Circular dichroism and

conformational polymorphism of DNA. Nucleic Acids Research 2009, 37 (6),

1713-1725.

98. Dewet, J. R. Wood, K. V. Deluca, M. Helinski, D. R. Subramani, S., Firefly

Luciferase Gene – Structure and Expression In Mammalian-cells. Molecular and

Cellular Biology 1987, 7 (2), 725-737.

99. Promega Technical Manual: Dual Luciferase Reporter Assay. Revised June 2015.

www.promega.com

100. Keller, G. A. Gould, S. Deluca, M. Subramani, S., Firefly Luciferase Is

Targeted to Peroxisomes in Mammalian-cells. Proceedings of the National

Academy of Sciences of the United States of America 1987, 84 (10), 3264-3268.

100. Addepalli, H. Meena Peng, C. G. Wang, G. Fan, Y. P. Charisse, K. Jayaprakash,

K. N. Rajeev, K. G. Pandey, R. K. Lavine, G. Zhang, L. G. Jahn-Hofmann, K.

Hadwiger, P. Manoharan, M. Maier, M. A., Modulation of thermal stability can

enhance the potency of siRNA. Nucleic Acids Research 2010, 38 (20), 7320-7331.

73

101.Stoien, J. D. Wang, R. J., Effect Of Near-Ultraviolet and Visible Light on

Mammalian-cells in Culture.2. Formation of Toxic Photoproducts in Tissue-

Culture Medium by Blacklight. Proceedings of the National Academy of Sciences

of the United States of America 1974, 71 (10), 3961-3965.

102. Lubbe, A. S. Szymanski, W. Feringa, B. L., Recent developments in

reversible photoregulation of oligonucleotide structure and function. Chemical

Society Reviews 2017, 46 (4), 1052-1079.

103.Perez-Rentero, S. Somoza, A. Grijalvo, S. Janousek, J. Belohradsky, M.

Stara, I. G. Stary, I. Eritja, R., Biophysical and RNA Interference Inhibitory

Properties of Oligonucleotides Carrying Tetrathiafulvalene Groups at

Terminal Positions. Journal of Chemistry 2013, 11.

104.Brotschi, C. Mathis, G. Leumann, C. J., Bipyridyl- and biphenyl-DNA: A

recognition motif based on interstrand aromatic stacking. Chemistry-a

European Journal 2005, 11 (6), 1911-1923.

105.Beharry, A. A. Sadovski, O. Woolley, G. A., Azobenzene Photoswitching

without Ultraviolet Light. Journal of the American Chemical Society 2011,

133 (49), 19684-19687.

106.Dong, M. X. Babalhavaeji, A. Samanta, S. Beharry, A. A. Woolley, G. A.,

Red-Shifting Azobenzene Photoswitches for in Vivo Use. Accounts of

Chemical Research 2015, 48 (10), 2662-2670.

107.Holder, G. M. Willox, S., Nitrobenzene Reduction and Reductive Cleavage of

Azobenzenes in 2 Species of Arachnida. Life Sciences 1973, 13 (4), 391-400.

74

Appendix

A1.1. 1H NMR Spectrum 4,4’-bis(hydroxyethyl)-azobenzene Compound (1) in DMSO-d6

at 400 MHz

A1.2. 13C NMR Spectrum 4,4’-bis(hydroxyethyl)-azobenzene Compound (1) in DMSO-

d6 at 400 MHz

75

A1.3. 1H NMR Spectrum 4,4’-bis(hydroxymethyl)-azobenzene Compound (2) in DMSO-

d6 at 400 MHz

A1.4. 13C NMR Spectrum 4,4’-bis(hydroxymethyl)-azobenzene Compound (2) in

DMSO-d6 at 400 MHz

76

A1.5. 1H NMR Spectrum 4-hydroxyethyl-4’-O-(4,4’dimethoxytrityl)-azobenzene

Compound (3) in CDCl3 at 400 MHz

A1.6. 13C NMR Spectrum 4-hydroxyethyl-4’-O-(4,4’dimethoxytrityl)-azobenzene

Compound (3) in CDCl3 at 400 MHz

77

A1.7. 1H NMR Spectrum 4-hydroxymethyl-4’-O-(4,4’dimethoxytrityl)-azobenzene

Compound (4) in CDCl3 at 400 MHz

A1.8. 13C NMR Spectrum 4-hydroxymethyl-4’-O-(4,4’dimethoxytrityl)-azobenzene

Compound (4) in CDCl3 at 400 MHz

78

A1.9. 1H NMR Spectrum of 2-Cyanoethyl-4-O-{[4-hydroxyethyl-4’-O-

(4,4’dimethoxytrityl)-O-methyl-diazenyl]}-N,N’-diisopropylaminophosphoramidite

Compound (5) in CDCl3 at 400 MHz

A1.10. 13C NMR Spectrum of 2-Cyanoethyl-4-O-{[4-hydroxyethyl-4’-O-

(4,4’dimethoxytrityl)-O-methyl-diazenyl]}-N,N’-diisopropylaminophosphoramidite

Compound (5) in CDCl3 at 400 MHz

79

A1.11. 31P NMR Spectrum of 2-Cyanoethyl-4-O-{[4-hydroxyethyl-4’-O-

(4,4’dimethoxytrityl)-O-methyl-diazenyl]}-N,N’-diisopropylaminophosphoramidite

Compound (5) in CDCl3 at 400 MHz

A1.12. 1H NMR Spectrum of 2-Cyanoethyl-4-O-{[4-hydroxymethyl-4’-O-

(4,4’dimethoxytrityl)-O-methyl-diazenyl]}-N,N’-diisopropylaminophosphoramidite

Compound (6) in CDCl3 at 400 MHz

80

A1.13. 13C NMR Spectrum of 2-Cyanoethyl-4-O-{[4-hydroxymethyl-4’-O-

(4,4’dimethoxytrityl)-O-methyl-diazenyl]}-N,N’-diisopropylaminophosphoramidite

Compound (6) in CDCl3 at 400 MHz

A1.14. 31P NMR Spectrum of 2-Cyanoethyl-4-O-{[4-hydroxymethyl-4’-O-

(4,4’dimethoxytrityl)-O-methyl-diazenyl]}-N,N’-diisopropylaminophosphoramidite

Compound (6) in CDCl3 at 400 MHz

81

s iR N A IC 5 0 D o s e R e s p o n s e

C o n c e n tr a tio n (p M )

1 1 0 1 0 0 1 0 0 0 1 0 0 0 0

0

5 0

1 0 0

1 5 0

7

8

9

1 0

1 1

1 2

1 3

w t

In

hib

itio

n (

% m

ax

imu

m)

A1.15. Inhibitory response curves of siRNAs 7-13 and wt.

A1.16. Photolability of siRNAs 7, 9, 10 11 and12 exposed to alternating visible and UV

light

-0.1

-0.08

-0.06

-0.04

-0.02

0

0.02

0.04

0 5 10 15 20 25 30 35 40

Rel

ativ

e C

han

ge

in A

bso

rban

ce (

Δ A

bs)

Time (min)

Photolability of sense strands

siRNA 7

siRNA 9

siRNA 10

siRNA 11

siRNA 12

82

A1.17. Deconvolution results for siRNA 7. ESI-HRMS (ES+) m/z calculated for siRNA 7

6316.81, found 6315.85 [M+H]+

A1.18. Deconvolution results for siRNA 8. ESI-HRMS (ES+) m/z calculated for siRNA 8

6382.84, found 6380.85 [M+H]+

83

A1.19. Deconvolution results for siRNA 9. ESI-HRMS (ES+) m/z calculated for siRNA 9

6367.91, found 6366.89 [M+H]+

A1.20. Deconvolution results for siRNA 10. ESI-HRMS (ES+) m/z calculated for siRNA

10 6344.86, found 6344.91 [M+H]+

84

A1.21. Deconvolution results for siRNA 11. ESI-HRMS (ES+) m/z calculated for siRNA

11 6344.86, found 6344.86 [M+H]+

A1.22. Deconvolution results for siRNA 12. ESI-HRMS (ES+) m/z calculated for siRNA

12 6367.91, found 6366.95 [M+H]+

85

A1.23. Deconvolution results for siRNA 13. ESI-HRMS (ES+) m/z calculated for siRNA

13 6410.89, found 6410.86 [M+H]+

A1.24. HPLC chromatogram for siRNA 7. Arrows indicate trans and cis peaks and

retention times.

cis

rt:13.6

trans

rt: 14.5

86

A1.25. HPLC chromatogram for siRNA 9. Red no treatment, blue 60 min UV exposure.

Arrows indicate trans and cis peaks and retention times.

A1.26. HPLC chromatogram for siRNA 10. Red no treatment, blue 60 min UV exposure.

Arrows indicate trans and cis peaks and retention times.

cis

rt: 15.5

trans

rt: 16.1

cis

rt: 16.7

trans

rt: 17.4

87

A1.27. HPLC chromatogram for siRNA 11. Red no treatment, blue 60 min UV exposure.

Arrows indicate trans and cis peaks and retention times.

A1.28. HPLC chromatogram for siRNA 12. Red no treatment, blue 60 min UV exposure.

Arrows indicate trans and cis peaks and retention times.

cis

rt: 15.3

trans

rt: 16.0

cis

rt: 15.6

trans

rt: 16.3

88