Johanna_Edlund-Thesis-final

Development of a Cre Recombinase Inducible shRNAmiR System

Evaluating the feasibility of generating transgenic RNAi mouse models and CPPs potential as tools for inducing Cre recombination in ES cells

Master Thesis in Molecular Biology

University of Gothenburg

2009

Johanna Edlund

Supervisor:

John W. Wiseman

AstraZeneca Transgenics and

Comparative Genomics

Supervisor:

Jeanette Nilsson

Dep. of Cellular and Molecular Biology

University of Gothenburg


Johanna Edlund

ii

PREFACE

This report constitutes my thesis for receiving the M.Sc. degree (300 credits) in the field

of cell- and molecular biology at the University of Gothenburg, Sweden. The research

and work has been conducted at the Transgenics & Comparative Genomics centre at

AstraZeneca R&D, Mölndal, Sweden during a period of 40 weeks ending February

2009.

Supervisors for this project have been John. W. Wiseman at the Department of

Transgenics & Comparative Genomics at AstraZeneca R&D and Jeanette Nilsson at the

Department of Cell and Molecular Biology at the University of Gothenburg.

Johanna Edlund

February 2009


Johanna Edlund

iii

ABSTRACT

The finding that small interfering RNA’s (siRNAs) could mediate gene silencing in

mammalian cells has led to the emergence of RNA interference (RNAi) as the gold

standard tool for sequence-specific knockdown of gene expression. Stable silencing of

genes within any given tissue can be achieved by promoter-based expression of short

hairpin RNA’s (shRNAs). Many systems exist to facilitate knockdown but applying

these techniques in vivo is the greatest challenge in the field of RNAi.

Conventional transgenesis by homologous recombination can generate gene inactivated

mouse models. However, some biological processes cannot be accessed by gene

inactivation strategies. Many genes have roles in embryogenesis and knocking out these

genes may result in embryonic lethality. RNAi provides a powerful tool to circumvent

this problem by knocking-down gene expression in a controlled manner. Gene silencing

induced by intracellular silencing constructs, such as shRNAmiR expression vectors

which target and block expression of specific mRNAs, can be combined with Cre-loxP

technology to generate an inducible knock-down system which may overcome the risk

of lethal phenotypes. Induction of the model can be achieved by breeding with Cre-

expressing mouse lines or possibly by delivering Cre as a purified protein fused to a

cell-penetrating peptide (CPP) to allow the Cre to cross the plasma membrane and gain

entry into cells. The latter approach may be a time-saving alternative to breeding.

We have investigated using RNAi to inhibit the expression of KDR and NRP-1 in cells

having endogenous expression of these genes. Using commercially available

shRNAmiR systems, we designed and constructed different shRNA expression vectors

targeting KDR and NRP-1 in vitro. Knock-down efficiency was measured by

determining levels of KDR and NRP-1 mRNA using real-time RT-PCR and at the

protein level by quantitative western analysis. In addition we generated an inducible

shRNA expression system to evaluate different Cre-fused CPPs for their ability to

penetrate cells and induce recombination in both CHO and ES cells based systems using

luciferase and green fluorescent protein (GFP) reporters.

Real-time RT-PCR analysis indicated that specific shRNA sequences could knock down

expression of our genes of interest in LLC-1 cells. The constructs resulting in greatest

knock-down gave silencing rates of 58 % and 54 % compared to a negative control

shRNA in Open Biosystems and Invitrogen shRNAmiR systems respectively.

Quantitative western blot analysis indicated knock-down levels of 80 % in LLC-1 cells

using the Invitrogen system.

Reporter assays indicated successful penetration of cells and subsequent Cre

recombination following transduction of CHO and ES cells and highlighted major

differences in recombination levels dependent upon which CPP was used in the assays.

These data indicate knock-down of KDR and NRP-1 at mRNA and protein levels.

However, the study design needs to be revised to improve stability and reproducibility.

Also, CPPs have been found to be a promising application to facilitate effective Cre-

induced recombination.

Keywords: shRNAmiR, RNAi, CPPs, ES cells, Cre recombinase


Johanna Edlund

iv

ABSTRAKT

Upptäckten av att small interfering RNA’s (siRNAs) kan tysta ner gener i

däggdjursceller har lett till att RNA-interferens (RNAi) har vuxit fram som det

självklara verktygsvalet för sekvensspecifik nedreglering av genuttryck. Genom att

använda promotorbaserade uttryck av short hairpin RNA’s (shRNAs) kan man uppnå

stabilt tysta gener i utvald vävnad. Det existerar många system för att underlätta

nedreglering, men användandet av dessa tekniker in vivo är den största utmaningen

inom området för RNAi.

Konventionell transgenteknik genom homolog rekombination kan generera

geninaktiverade musmodeller. Det har dock funnits att vissa biologiska processer inte

kan nås via geninaktiveringsstrategier. Många gener är vitala för embryonal utveckling

och nedreglering av dessa kan resultera i embryonal dödlighet. Med RNAi finns

verktygen för att kringgå dessa problem genom en kontrollerad nedreglering av

genuttryck. Nedreglering av gener inducerad av intracellulära konstruktioner, som

shRNAmiR expressionsvektorer vilka attackerar och blockerar uttryck av specifika

mRNAs, kan kombineras med Cre-loxP-teknologi för att skapa ett kontrollerat

inducerbart nedregleringssystem. Inducering avsystemet kan åstadkommas genom avel

med Cre-uttryckande muslinjer eller eventuellt genom att administrera Cre som ett renat

protein kopplat till en cell-penetrerande peptid (CPP) som tillåter att Cre korsar

plasmamembranet och tillåts inträde till cellernas inre. Det senare alternativet kan vara

tidsbesparande jämfört med avel.

Vi har studerat hur RNAi inhiberar uttryck av KDR och NRP-1 i celler med endogent

uttryck av dessa gener. Med kommersiellt tillgängliga shRNAmiR-system

konstruerades olika shRNA-expressionsvektorer riktade mot KDR och NRP-1 in vitro.

Nedregleringseffektiviteten mättes genom att bestämma nivåerna av KDR och NRP-1

mRNA med realtids-RT-PCR och på proteinnivå genom kvantitativ western-analys. Vi

genererade även ett inducerbart shRNA-expressionssystem för att utvärdera olika Cre-

kopplade CPPs förmåga att penetrera celler och inducera rekombination i både CHO-

och ES-cellbaserade system genom att använda luciferase och grönflouroserande

protein (GFP) reportrar.

Realtids-RT-PCR-analys indikerade att specifika shRNA-sekvenser kan nedreglera

uttryck av våra gener i LLC-1-celler. Konstruktionerna som resulterade i kraftigast

nedreglering gav nedreglering med 58 % respektive 54 % jämfört mot en negativ

kontroll-shRNA i Open Biosystems och Invitrogen shRNAmiR-systemen. Kvantitativ

western blot analys indikerade nedreglering med 80 % i LLC-1-celler i Invitrogen-

systemet. Reporteranalyser indikerade vidare lyckad penetrering av celler och

efterföljande Cre-rekombination efter transduktion av CHO- och ES-celler, och

framhävde stora skillnader i rekombinationsnivåer beroende på använd CPP.

Resulterande data visar på nedreglering av KDR och NRP-1 på mRNA och likaså på

proteinnivå. Det behövs emellertid en kraftfullare försöksdesign för att uppnå stabilitet

och reproducerbarhet i studien. CPPs har dessutom funnits vara en lovande framtida

teknik för tidseffektiv Cre-inducerad rekombination.

Nyckelord: shRNAmiR, RNAi, CPPs, ES-celler, Cre-rekombinas


Johanna Edlund

v

ABBREVIATIONS

ATCG AstraZeneca Transgenics and Comparative Genomics

CMV cytomegalovirus

Cre cyclization recombination

Dox doxycycline

ES cell Embryonic stem cell

KDR Kinase insert Domain protein Receptor

loxP locus of X-over of bacteriophage P1

NRP-1 Neuropilin-1

RISC RNA-induced silencing complex

RNAi ribonucleic acid interference

shRNA short hairpin ribonucleic acid

siRNA short interfering ribonucleic acid

miRNA micro ribonucleic acid

shRNAmiR miRNA embedded shRNA

tetR tetracycline repressor

tetO tetracycline operator

tTA tetracycline transactivator protein

TRE tetracycline responsive element

CPP Cell penetrating peptides

GAPDH Glyceraldehyd-3-phosphate dehydrogenase

GFP Green fluorescence protein

NLS Nuclear Localization signal

PCR Polemyrase chain reaction

LB Luria-Bertani

SDS sodium dodecyl sulfat


Johanna Edlund

vi

TABLE OF CONTENTS

1 INTRODUCTION ................................................................................................. 1

1.1 TRANSGENIC ANIMALS ............................................................................................. 1

1.2 CONVENTIONAL TRANSGENIC SYSTEMS........................................................................ 1

1.3 CONDITIONAL TRANSGENIC SYSTEMS .......................................................................... 2 1.3.1 CRE-LoxP system ................................................................................................................... 2 1.3.2 Tet regulatable systems ....................................................................................................... 4

1.4 CELL PENETRATING PEPTIDES..................................................................................... 5 1.4.1 Neo removal on the fly ......................................................................................................... 5 1.4.2 Structure of available cell penetrating peptides ................................................................... 6

1.5 RNA INTERFERENCE LEADING TO GENE SILENCING ......................................................... 6

1.5.1 Short hairpin RNA ................................................................................................................. 8 1.5.2 Micro RNA ............................................................................................................................ 9 1.5.3 shRNAmiR technology ........................................................................................................ 10

1.5.3.1 Invitrogen shRNAmiR system.................................................................................................. 10 1.5.3.2 Open Biosystems shRNAmiR system ...................................................................................... 12

1.6 TARGET GENES FOR RNAI KNOCK-DOWN .................................................................. 14

1.7 AIMS ................................................................................................................. 14

1.8 STUDY DESIGN ..................................................................................................... 14

2 MATERIALS AND METHODS ............................................................................. 17

2.1 TISSUE CULTURE ................................................................................................... 17

2.2 TRANSFECTION OF PLASMID DNA INTO MAMMALIAN CELLS .......................................... 17 2.2.1 Transfection efficiency ........................................................................................................ 18 2.2.2 Transduction of mammalian cells with CRE-fusion proteins .............................................. 18 2.2.3 Luciferase Assay ................................................................................................................. 19

2.3 DNA PURIFICATION AND ANALYSIS ........................................................................... 19

2.3.1 Transformation of plasmid DNA ......................................................................................... 19 2.3.2 Purification of plasmid DNA ............................................................................................... 19 2.3.3 Extraction of ES cell genomic DNA ..................................................................................... 20 2.3.4 Ligation of DNA fragments ................................................................................................. 20

2.4 PCR CONDITIONS ................................................................................................. 20 2.4.1 TOPO cloning ...................................................................................................................... 20

2.5 PLASMIDS AND CLONING ........................................................................................ 22

2.6 PROTEIN PURIFICATION AND ELECTROPHORESIS .......................................................... 25 2.6.1 Quantitative western analysis ............................................................................................ 25

2.7 RNA ISOLATION AND ANALYSIS ............................................................................... 25

2.7.1 RT-PCR cDNA synthesis ....................................................................................................... 25 2.7.2 Quantitative mRNA detection by real-time PCR ................................................................. 26

2.8 SEQUENCING ....................................................................................................... 26


Johanna Edlund

vii

2.9 STATISTICAL ANALYSIS ........................................................................................... 27

2.10 SOFTWARE .......................................................................................................... 27

3 RESULTS .......................................................................................................... 28

3.1 CELL LINE EXPRESSION AND TRANSFECTION ANALYSIS ................................................... 28 3.1.1 Selection of model cell line ................................................................................................. 28 3.1.2 Optimization of mammalian cell line transfection ............................................................. 29 3.1.3 Establishing transfection efficiency .................................................................................... 33

3.2 SHRNAMIR EXPRESSION CONSTRUCTS ...................................................................... 35 3.2.1 Verification of shRNAmiR constructs .................................................................................. 35 3.2.2 Real time RT PCR mRNA analysis........................................................................................ 35 3.2.3 Quantitative western analysis ............................................................................................ 38

3.3 GENERATION OF A CRE INDUCIBLE SHRNAMIR EXPRESSION SYSTEM ............................... 39 3.3.1 Amplification of lox-CAT-lox stop cassette insert ............................................................... 39 3.3.2 PCR screen and sequencing to verify lox-CAT-lox insert in pGIPZ vector ............................ 41 3.3.3 TOPO TA cloning ................................................................................................................. 41 3.3.4 pcDNA 6.2-GW EmGFP miR destination vector .................................................................. 43 3.3.5 Amplification of loxP-CAT-loxP cassette ............................................................................. 44

3.3.5.1 Test of transcriptional inhibition ............................................................................................ 44

3.4 RECOMBINATION THROUGH CRE-FUSED CELL PENETRATING PEPTIDES ............................. 46

3.5 GENERATION OF A CHO REPORTER LINE FOR CRE-FUSED CPP EVALUATION ..................... 47 3.5.1 Transduction of CHO luciferase reporter cell line with Cre-fused CPP ................................ 49 3.5.2 Chloroquine enhances Cre-fused CPP transduction ............................................................ 50 3.5.3 Transduction of mouse embryonic stem cells with Cre-fused CPP ..................................... 52 3.5.4 Transduction of an ES cell GFP reporter line with Cre-fused CPP ....................................... 52 3.5.5 PCR screen for Cre mediated recombination in ES cells following transduction with Cre-fused CPP .......................................................................................................................................... 54

4 DISCUSSION .................................................................................................... 56

4.1 DEVELOPMENT OF A CRE RECOMBINASE INDUCIBLE SHRNAMIR SYSTEM ......................... 56 4.1.1 Concluding results .............................................................................................................. 57 4.1.2 Improvements .................................................................................................................... 57

4.1.2.1 Real time RT PCR ..................................................................................................................... 57 4.1.2.2 Western blot ........................................................................................................................... 58

4.1.3 Conclusions and future aspects .......................................................................................... 59

4.2 EVALUATING CPPS POTENTIAL AS TOOLS FOR INDUCING CRE RECOMBINATION IN ES CELLS . 59 4.2.1 Concluding results .............................................................................................................. 59 4.2.2 Improvements .................................................................................................................... 60 4.2.3 Conclusions and future aspects .......................................................................................... 60

5 ACKNOWLEDGMENTS ..................................................................................... 61

6 REFERENCES .................................................................................................... 62

APPENDIX 1: GENERAL OLIGO LIST ........................................................................... 64

APPENDIX 2: SEQUENCES DESIGNED FOR SHRNAMIR KNOCKDOWN OF NRP-I AND KDR USING INVITROGEN’S RNAI DESIGNER ALGORITHMS ................................................ 66

APPENDIX 3: CELL PENETRATING PEPTIDES .............................................................. 67

[This page is intentionally left blank]


Johanna Edlund

1

1 INTRODUCTION

1.1 Transgenic animals

An important milestone in biomedical research was the sequencing of several

mammalian genomes, including the human genome. This knowledge generated new

questions which further required new techniques to be answered.

Chromosomal-engineering technology has made possible the functional analysis of the

mammalian genome. By switching genes on and off an understanding of an individual

gene’s function is possible. This technology has impacted on the development of

genetically modified animals which are valuable tools in the drug discovery process. In

drug research target identification and target validation is crucial for the development of

candidate drugs leading to a therapeutic effect.

Creation of humanised disease models might in early phases of drug discovery give

answers regarding drug metabolism and toxicity. Tools for designing these systems are

found in the molecular engineering techniques, which allow for targeting specific

tissues and specific gene expression at any given time [12].

A transgenic animal is one that carries a foreign gene that has been deliberately inserted

into its genome. Examples are transgenic sheep and goats that express foreign proteins

in their milk. However, mice have proven to be the most widely used mammalian

species used in transgenesis for exploring multiple biological questions. Advantages of

using the mouse are its well characterized genome and even after extensive genetic

manipulation, mouse embryonic stem (ES) cells are able to reintegrate fully into viable

embryos when injected into a host blastocyst or aggregated with a host morula [10].

Injection of genetically altered mouse ES cells into a blastocyst allows modification of a

gene of interest in a cell line prior to introducing the modification into the genome of

the mouse. Genetic alterations from the ES cells are incorporated into the mice genome

and are transmitted through the germ line [10].

1.2 Conventional transgenic systems

Conventional transgenic technologies include gene targeting by homologous

recombination, (e.g. gene knock-out or gene knock-in) 14. In a standard gene knock-

out a targeting vector is designed to alter genomic sequences by homologous

replacement. Homologous replacement is versatile, in that sequences can be directly

targeted and altered, inserted, or deleted. Positive and negative selection markers are

used to select clones where insertion of the targeting vector has occurred. The neomycin

gene (neo), which confers resistance to G418, is commonly used for positive selection.

To ensure that the insert of the vector has not occurred by non-homologous

recombination negative selection markers, such as the diphtheria toxin (DTA) or

thymidine kinase (TK), may be used. Positive selection cassettes can be removed from

the targeted allele by employing site specific recombinase technology. An example is

the Cre/LoxP system where loxp recombination sites which flank the selection cassette


Johanna Edlund

2

facilitate the removal of the cassette by introduction of Cre recombinase. Southern

analysis of clones is routinely used to confirm successful recombination [14]. Correctly

targeted ES clones are then injected into early stage mouse embryos. These embryos are

then implanted into pseudo-pregnant mice and the offspring can be tested for

chimerism. A chimera is an animal that has two or more different populations of

genetically distinct cells. Chimeric mice carry both wt allele and the transgenic allele.

The chimera can be bred to a wt mouse to establish the mutation in the germ line.

Standard knock-out mice lack targeted gene function in all cells throughout their lives.

Therefore genes which are vital during embryonic development cannot be studied in

this way as the effect of the knockout cannot be investigated in live mice. In response to

this there has been growing interest to generate tissue specific and inducible knock-outs

to overcome problems associated with early embryonic lethality and compensatory

effects of gene function [17].

In addition to targeted transgenesis by homologous recombination, transgenic mice can

be generated by random gene insertion by pronuclear injection of DNA in freshly

fertilized eggs. Embryos are then implanted into a pseudo-pregnant mouse. This results

in random integration of the DNA into the genome. Random integration may give rise

to problems such as insertion into an essential gene, variable transcription due to

positional effects, and an unpredictable copy number.

1.3 Conditional transgenic systems

The ability to turn genes on and off at our discretion is a powerful tool. However, there

are limitations. As mentioned earlier many mutants will lead to embryonic lethality or

compensatory effect may arise [17]. Therefore approaches to study gene function in

specific tissues and at specific times in the life of the mouse are desirable. In the mouse

this has been accomplished by using binary systems where gene activation or deletion is

dependent on the interaction of two components; tissue restricted expression can be

achieved by the use of specific promoters, and the use of inducible promoters that can

be turned on and off to give spatiotemporal control within such systems. Examples of

such approaches are the CRE/loxP recombinase and TetR based systems. These systems

are silent when not induced and active when an external stimulus is added. They both

resemble pathological conditions much better and therefore contribute to drug

interventions at a larger extent. Further, tissue restricted and inducible system in knock-

out mice often better mimic situation in certain human late-onset diseases.

1.3.1 CRE-LoxP system

A widely used system to achieve tissue specificity and inducibility is the Cre-LoxP

recombinase system which was originally discovered in bacteriophage P1. The

bacteriophage P1 gene cre which express a site specific DNA recombinase protein,

Cyclization recombination (Cre), catalyses recombination between two of its

recognition sites (loxP) which is a 34 bp consensus sequence consisting of a core spacer

sequence of 8 bp and two 13 bp palindromic flanking sequences. LoxP sites might be

located distally on a chromosome and even on different chromosomes [5]. If the loxP

sites are oriented in the same direction an excited fragment with one 13 bp inverted

repeat and a circular fragment with one 13 bp inverted repeat will be left after

recombination, see Figure 1.1. If the loxP sites are oriented in the opposite direction an


Johanna Edlund

3

inversion event will take place. Recombination is mediated by breaking and re-joining

of DNA strands by phosphotyrosine protein-DNA linkage. No accessory cofactors are

needed in this process. Cre recombinase contains a nuclear localization signal (nls) and

can therefore translocate to the nucleus of eukaryotic cells [22].

LoxP sites flank the target gene and Cre recombinase cleaves the flanked region. Mouse

lines expressing Cre from specific tissue specific promoters can be used to achieve

spatial control of gene expression where recombination results in inactivation of the

specific allele which in turn gives rise to a knock-out of the target gene [4].

Figure 1.1 Illustration of a Cre mediated recombination event at LoxP sites. LoxP sequence detailed, showing the two flanking inverted repeats and the spacer region between in bold.

Cre-loxP technology is also frequently used to remove selectable markers such as the

neomycin resistance gene from targeted ES cells as is may be important to remove the

marker to prevent its expression interfering with the expression at ones allele of interest

Another common use of Cre-loxP technology is in the generation of conditional

knockout mice where the modification to inactivate a gene is brought about by the

addition of Cre e.g. through breeding to a Cre expressing mouse line [5].


Johanna Edlund

4

Temporal control of one’s expression can be achieved by including a transcriptional

inhibitor (stop cassette) sequence flanked by loxP sites into ones construct. The gene to

be expressed is located downstream of the flanked transcriptional inhibitor cassette and

Cre recombination activates expression and the desired phenotype. Reporter gene

expression is commonly activated using such a system, see Figure 1.2.

Figure 1.2 Conditional Cre mediated gene activation. The stop cassette is excised by Cre administered through breeding with a Cre expressing mouse either globally or in a tissue specific manner based on promoter choice. Other administration routes are available, see text. The reporter gene is silent until removal of the stop cassette activates transcription.

The Cre-loxp system is though a non-reversible gene switch not allowing for

modulations of gene expression to gain a reversible system the Tetracycline technology

has to be added [4].

1.3.2 Tet regulatable systems

The tetracycline inducible transgenic system first developed by Bujard and Gossen is

controlled by administration of tetracycline (tet) or its derivative doxycycline (dox) [2].

Dox is often the inducer of choice due to its low cost, commercial availability, and the

fact that at the levels used to effectively induce the system low cytotoxic effects have

been found. There are two basic variants of this system that has been proven successful

in vivo; Tet-Off and Tet-On.

The Tet-Off system has a tissue specific promoter as the human cytomegalovirus

(CMV), which ubiquitous drives the expression of the gene of interest and the tet

transactivator protein tTA, which is a fusion protein of DNA-binding domain of the tet-

repressor in E.coli (Tetr) and a transcriptional activator domain of herpes simplex virus

(VP16). Expressed tTA binds specifically to a tet operator region and activates


Johanna Edlund

5

transcription of the target gene through the CMV promoter. The tet operator, Tet-O is

composed of seven tet operator sequences, including the specific site for tTA binding

called the tet responsive element (TRE).

In the Tet-Off system administrated tet or dox will prevent the binding due to a

conformational change of the DNA binding domain of tTA. This results in no CMV

promoter activity and hence no transcription of the downstream genes including the

gene for the tTA protein.

The Tet-On system is a modified Tet-Off system where a reverse transactivator protein

(rtTA) is present instead of the tTA protein. This system binds DNA only when

administrated tet or dox is present otherwise this system has the same features as the

Tet-Off system [3]. For an illustration of Tet-Off and Tet-On, see Figure 1.3.

Figure 1.3 The Tet-On and Tet-Off regulatable system. The Tet-On system express rtTA that binds to and activates expression from any TRE vector in the presence of Dox. The Tet-Off system express tTA which binds and activates in the absence of Dox.

1.4 Cell penetrating peptides

1.4.1 Neo removal on the fly

The lipid bi-layer of biological membranes may be crossed by naturally occurring

cationic low molecular weight peptides rich in lysine, proline, or basic amino acids,

such as arginine, even though the membranes usually are impermeable. These cell

penetrating peptides (CPPs) originate from naturally occurring proteins. An example of

a CPP is the HIV-1 TAT peptide, which is isolated from the Human Immunodeficiency

Virus (HIV) transcription activating factor.

Most eukaryotic cell types allow rapid uptake of these peptides through a mechanism

not yet fully understood. It seems to vary between different CPPs and also between


Johanna Edlund

6

target cells and proposed mechanisms include an endocytosis like event, entry via

electrostatic interaction or through pore formation [24, 25].

Conjugated to other peptides and proteins CPPs may serve as transporters if

administered in non-toxic concentrations. Therefore they have potential to be used as

tools for gene regulation.

Targeted ES cells may be transfected with a Cre expressing plasmid as an option for

removing loxP flanked target DNA sequences and could replace time-consuming

downstream breeding with a Cre line. However transfection efficiency of ES cells is

relatively low and transfection techniques may have toxic side effects or physiologically

deleterious consequences. Therefore CPPs may represent an alternate system for Cre

delivery into ES cells. Such protein transduction has exhibited high efficiency in

delivery of Cre into mammalian cells both in vivo and in vitro [26]. Injection of CPPs

may also be used in vivo in transgenic mice to generate recombination. This, for

example could dramatically reduce timelines associated with the generation of a

conditional knockout mouse model by circumventing conventional breeding to Cre

expressing lines.

A drawback to working with protein transduction is that high concentrations of CPPs

may arise in the cell by nonspecific fluid-phase endocytosis especially when working at

37˚C. At 4˚C this is reduced significantly [24]. Therefore uptake of CPP may be very

high but the biological availability may be low due accumulation in endocytotic

vesicles. Chloroquine and ExGEN500 are chemical agents which can disrupt

endosomes and have been reported to enhance CPP mediated delivery [11].

1.4.2 Structure of available cell penetrating peptides

Various Cre fused CPPs were obtained from ATCG stocks. A short peptide sequence,

e.g. Histone H1 or TAT, derived from protein transduction domains (PTDs) was

conjugated to a localization domain; the signal peptide (SP), the nuclear localization

signal (NLS) or the membrane translocation signal (MTS), which all target the Cre

proteins to various cellular destinations.

The proteins were produced using the E.coli strain BL21 (DE3) and are listed in

Appendix 3. Purification of proteins was performed using both IMAC and Source30

gels. In order to decrease the endotoxin levels, an extra purification step was performed.

1.5 RNA interference leading to gene silencing

RNA interference (RNAi) was first discovered by Mello and Fire in the model organism

Caenorhabditis Elegans (C. elegans), where dsRNA induced sequence specific gene

silencing [8]. RNAi is utilized by most eukaryotes in vivo for antiviral defence,

modulation of transposon activity and gene regulation [8].

RNAi occurs at the mRNA level and is sequence specific. The RNAi machinery and

pathway is composed of a RNA-induced silencing complex (RISC) and includes a

ribonuclease enzyme called Dicer. Dicer processes small interfering RNA (siRNA)

from double stranded RNAs (dsRNA) or short hairpin RNAs (shRNA). These siRNAs

are short RNA duplexes of 19-20 nucleotides with two nucleotide overhangs on each

strand which are incorporated into the RISC, which guides siRNA to the correct

location on the target mRNA. siRNAs may have different structures resulting in


Johanna Edlund

7

different mechanisms of gene silencing such that target mRNA can either be cleaved or

translationally repressed. [7].

Figure 1.4 Illustration of the different RNAi pathways. RNAi may be generated by three different sources; dsRNA, an siRNA expressing vector, or synthetic siRNA. All three strategies utilize the RNAi pathway machinery in different ways as indicated in the illustration, but all three results in gene regulation by silencing.

Obvious applications of gene silencing as a research tool are in drug discovery and for

the generation of animal disease models. However, RNAi is a powerful tool but it

results in knock-down of one’s target gene rather than a complete knock-out. In certain

circumstances this is an advantage since knock-down of a gene, rather than a complete

knock-out, may better mimic a particular disease and may also allow for threshold

studies when studying gene functions in development or disease.

One particular advantage of RNAi over knock-out technology is that in a transgenic

mouse model only one allele of the transgene is required to dominantly suppress the

endogenous gene of interest which thus eliminating the need for breeding together two

mutant alleles as in traditional knock-outs technology.

Moreover, considering embryonic lethality associated with knock-out of genes involved

in development, a regulatable and tissue specific knock-down would allow for studies

on adult and fully developed animals [17].

A range of strategies exists for modulating gene expression in vitro and in vivo by

RNAi. When delivered into the cytoplasm, either through a plasmid or direct injection,


Johanna Edlund

8

dsRNA mimics the endogenous RNAi system by generating siRNAs from Dicer

enzyme cleavage of the dsRNA. This system does not work well in mammals since

dsRNA may generate an interferon response often causing cell death. The use of

synthetic siRNA has been proved efficient at circumventing the interferon response [8].

Insertion of a gene cassette expressing siRNAs called shRNA in the nucleus mimicking

the micro interfering RNA system is a well-established strategy [7].

1.5.1 Short hairpin RNA

In RNAi there is considerable variation in the degree of silencing between genes, and

between functional gene target sequences. Not all siRNAs generated to target a gene are

equally effective in silencing the gene in mammalian cells. Thus it is important to

identify sequences that are effective inhibitors of gene expression by quantitative,

normalized, and internally controlled assays, to confirm the effectiveness of siRNAs.

In shRNA mediated silencing, siRNA expression from an expression vector gives rise

to a shRNA [15]. Early plasmid expression systems generating shRNA were based on

Pol III promoters, thus limiting the choice of promoter and making the construction of a

straight forward tissue specific system difficult. Given that most tissue and cell specific

transcripts are expressed from Pol II promoters more recent Pol II based systems are

preferable over those of Pol III when designing regulatable tissue specific silencing

systems. Transient or stable silencing is possible with shRNA expression vectors and

transgenic shRNA mice have been generated [10]. The stem loop of the hairpin is

important for function and potency. Within the single stranded RNA sequence there are

complementary antisense and sense sequences spaced by the loop region, this allows

complementary sequences, when transcribed, to fold back on each other forming a

hairpin loop [7, 15], see Figure 1.5.

Figure 1.5 Short hairpin expressing vector. Promoter driven expression of shRNA sequence. Antisense and sense strands are complementary and spaced by a loop region.

shRNA is transcribed in the nucleus and transported through the nuclear membrane to

the cytoplasm of the cell by exportin 5. In the cytoplasm Dicer processes the hairpin in


Johanna Edlund

9

the same way as with dsRNA generating siRNAs which are subsequently incorporated

into the RISC complex, see Figure 1.4 and Figure 1.6.

1.5.2 Micro RNA

The ability to control timing and location of the expression of shRNA results in a more

sophisticated system. This can be achieved by combining a micro RNA context and

shRNA technology coupled to inducible and tissue specific promoters.

Micro RNAs (miRNAs) are small endogenously expressed non coding single stranded

RNAs (ssRNA). They are expressed by RNA-polymerse II and exhibit high diversity in

their expression patterns and in their native environment they control and regulate gene

expression by sequence complementarity to mRNA [20].

The microRNA process starts with expression of a Pri-miRNA hairpin sequence which

is processed by a ribonuclease III enzyme called Drosha and a double-stranded RNA

binding protein Pasha, giving a pre-miRNA precursor forming an imperfect stem loop

structure containing a guide strand and a passenger strand. The pre-miRNA loop is

transported across the nuclear membrane into the cytoplasm by exportin 5 and a RAN-

GTP process. Pre-miRNA loops are processed by the ribonuclease III enzyme Dicer

into dsRNAs, called mature miRNA. By an unwinding mechanism siRNAs are

produced and the less stable 5´guide strand of the miRNA is subsequently loaded into

the RISC silencing complex [29]. When the guide strand finds its homologous

sequence it activates by the aid from Argonaute proteins the RNase activity of RISC,

which cleaves 10 nucleotides upstream of the target sequence. However if the miRNA

has an imperfect stem loop structure it binds complementary to the mRNA 3´

untranslated region (3ÚTR).

This leads to repression of translation of the mRNA by attenuating the translation

process. Most animal miRNA induce translational repression but several animal

miRNAs have been found that directs mRNA cleavage of their target so a clear

distinction cannot be done [20], see Figure 1.6.


Johanna Edlund

10

Figure 1.6 Illustration of the micro RNA pathway. Transcription of a miRNA gene within the nucleus creates an immature miRNA which passes through the micro RNA pathway’s intermediate stages and is transported to the cytoplasm. The different outcomes at translational level depends on the structure of the ssRNA.

1.5.3 shRNAmiR technology

Double stranded shRNAs can be produced in a way that mimics the structure of

miRNAs and can therefore function as a substrate for Dicer. This is the basis of miR-

based shRNA technology. miRNA embedded shRNA (shRNAmiR) has been shown to

be more effective than the simpler shRNA design. Studies have shown increased

Drosha and Dicer processing resulting in greater levels of siRNA leading to greater

knock-down [23].

Several commercial systems exploit this technology including systems from Invitrogen

and Open Biosystems.

1.5.3.1 Invitrogen shRNAmiR system

The Invitrogen BLOCK-iT™ system is a so called empty system where the guide

sequence having homology to the target of interest is cloned in to a Pol II driven

expression vector generating a synthetic shRNAmiR system. This system is based upon

mouse miR-155. The native stem loop precursor for the mouse miR-155 miRNA is

located within the third exon of the mouse BIC gene. BIC is an evolutionary conserved

region coding for this noncoding RNA having features useful for effective RNAi

silencing. The specific structure includes flanking mir-155 regions surrounding the stem


Johanna Edlund

11

loop precursor and two extra nucleotides in the guide strand compared to the

complementary passenger strand, see Figure 1.7. These features have improved RNase

III (Dicer) enzyme recognition [28].

The Invitrogen system has the 5ánd 3´flanking regions of native miR-155 transcripts

incorporated into the expression vector. The stem loop has been optimized from the

native miR-155 by introducing an MscI site allowing for linearization. Further, by using

a 2-nucleotide loop instead of a 5-nucleotide loop as in the native form even higher

knock-down efficiency is accomplished [18], see Figure 1.7.

Figure 1.7 The engineered pre-miRNA sequence and structure. The 5’ and 3’ flanking regions are derived from miR-155. The MscI linearization site is shown within the terminal loop and is marked in red. The targeting sequence containing the internal loop of two nucleotides is in bold.

Optimally designed oligonucleotides encoding specific pre-miRNA against the target

gene are generated using Invitrogen’s RNAi designer program. This designer software

automatically applies rules for designing the oligos in an optimal way. Sequences are

checked for specificity to minimise interference with expression of other genes. The

engineered sequence are cloned into the cloning site of the shRNAmiR expression

vector, pcDNA 6.2-GW EmGFP-miR, supplied from Invitrogen, see Figure 1.8. It is a

Pol II hCMV (human cytomegalovirus) promoter driven vector with an EmGFP

(Emerald Green Flourescent Protein) sequence incorporated, allowing for tracking of

shRNAmiR expression.

The miR sequence and the flanking sequences of miR-155 constitute the expression

cassette, which may be transferred between vectors. Once inserted into mammalian


Johanna Edlund

12

cells this expression cassette is thought to be expressed at high levels and further form

an intra-molecular stem loop further processed by Dicer enzyme into a functional

microRNA [18].

As a negative control plasmid the BLOCK-iT system contains a vector, pcDNA 6.2-

GW EmGFP-negmiR, see Figure 1.8. This vector contains an insert able to form a

hairpin structure in the similar way as described before; this hairpin has been evaluated

and found not to target any vertebrate gene [18]. The vector contains several restriction

sites to facilitate easy sub cloning.

Figure 1.8 Vector maps of pcDNA 6.2-GW-EmGFP-miR and pcDNA 6.2-GW-EmGFP-miR negative vectors. The vector in the left panel shows the double stranded oligo cloning site between 5’ and 3’ miR flanking regions. The vector in the right panel shows pre-inserted negative control miR cloned at the same location. Both vectors also contain an EmGFP reporter gene.

1.5.3.2 Open Biosystems shRNAmiR system

Open Biosystems has developed a similar system which incorporates the human

primary microRNA-30 (miR-30). The designed hairpin consists of 22 nucleotides of

dsRNA in the stem structure and a single stranded loop of 19 nucleotides from the

human miR-30, see Figure 1.9. Addition of the miR-30 loop and 125 extra nucleotides

flanking the hairpin results in a 10-fold increase of Drosha and Dicer processing of the

loop [27]. Open Biosystems provide mir-30 hairpins cloned into their Pol II CMV

promoter driven pGIPz vector , which contains an GFP expression marker to allow for

easy tracking of transfected cells, see Figure 1.9 and Figure 1.10.


Johanna Edlund

13

Figure 1.9 Illustration of the native and the optimized miR-30 hairpin. The Drosha and Dicer processing sites are indicated and further also the miR-30 loop. The antisense sequence of the optimized shRNA is shown together with complementary sense sequence.

pGIP z (empty vec tor)

11688 bp

5´mir30 reg

IRES reg

3´ mir30 reg

SnaB1 Seq For2

SnaB1 seq For1

ZeoR Marker

AmpR Marker

PuroR Marker

BGH- polyA

SnaB I seq Rev1

SnaB I seq Rev2

Not I seq For 1

Not I seq For 2

Not I seq Rev 1

Not I seq Rev2

CMV-IE-Promoter-Enhancher

Lac Promoter

fi Origin

pUC Origin

delta U3

turbo GFP tag

Sna BI (3071)

Not I (4101)

KpnI (4561)

KpnI (6392)

Figure 1.10 Vector map of pGIPZ empty vector. Map showing locations of 5’ and 3’ miR-30 regions, EmGFP reporter, and the ampicillin resistance gene. The shRNAmiR sequence is cloned in between of 5’ and 3’ miR-30 flanking regions. Primers used for insert verification are also indicated.


Johanna Edlund

14

1.6 Target genes for RNAi knock-down

RNAi is a powerful tool but it results in knock-down of one’s target gene rather than a

complete knock-out. Therefore given the possibility that relatively low levels of

expression from incomplete silencing may still result in wild type activity of one’s

protein of interest we chose to focus on genes where published knock-out mouse

models exhibited strong, embryonically lethal phenotypes. Blood vessel development

requires many key factors which promote proliferation and migration of vessels into a

mature vascular network. During mouse embryo development absence of a number of

genes result in embryonic lethality due to impaired vascular development [1].

Neuropilin 1 (NRP-1) is a co-receptor for VEGFR-1 (Flt-1 is the analogue in mouse)

and when expressed in endothelial cells it has been shown to be involved in the

regulation of angiogenesis and endothelial cell migration. Functional inactivation of

NRP-1 results in defects leading to embryonic lethality, thus implicating NRP-1 as

having a crucial role in vascular development [1, 2].

KDR (VEGFR-2 receptor) has been shown to be crucial for the development of

endothelial cells. Expression level studies at different developmental stages have shown

an increase during embryonic vasculogenesis and also in tumour angiogenesis. Mice

homozygous for deletion of Flk-1die during embryonic development due to impaired

vasculogenesis [2, 21].

1.7 Aims

To evaluate the feasibility of generating transgenic RNAi mouse models we have

chosen genes where knock-down could result in strong readily identifiable phenotypes.

However, given reports of fatality in mice due to oversaturation of RNAi related

molecular pathways the development of an inducible system to investigate knock-down

in adult mice would be of benefit [3].

The first aim was to design efficient shRNAmiR constructs to target and silence the

genes KDR and NRP-1 by using two shRNAmir systems (Invitrogen and Open

Biosystems) as comparison and to develop a system allowing inducible knock-down

based on Cre/loxP technology by inserting a loxP flanked transcriptional inhibiting

cassette to prevent shRNAmiR expression.

The second aim was to investigate the potential of Cre-fused CPPs as tools for inducing

Cre recombination and activation in vitro of our inducible system and further to this an

evaluation of Cre-fused CPP potency in inducing Cre recombination in mouse ES cells

with the purpose of assessing the wider use of these fusion peptides in influencing the

generation of transgenic mouse models including the costs and time involved.

1.8 Study design

When choosing a study design, factors giving variation in the information gathering

process should be taken into account. In biology there are both factors that are under the

experimenter’s control and those that are not. Since we were interested in the effect of

an induced process in cells we used both positive and negative controls as well as

comparison to a standard functioning as a baseline for our experiments. To reduce as


Johanna Edlund

15

much variation as possible multiple replicates of each experiment were conducted and

within each experiment triplicates were used to calculate mean.

First a model cell line having endogenous expression of our genes of interest has to be

selected. Expression of these genes in a range of cancer cell lines will be tested by RT-

PCR and as efficient transfection of the cell line we use is important, optimal

transfection conditions for suitable lines need to be established.

A range of shRNAmiR expression vectors based on miR-155 and miR-30, potentially

knocking down our genes of interest, will be generated. Validation of knock-down by

RNAi for each construct will be at both the RNA and protein level, using quantitative

Real time RT PCR PCR and quantitative western blot respectively.

Potent shRNAmiR’s will be cloned into a conditional shRNAmiR expression vector in

order to introduce inducibility into the system. A conditional shRNAmiR vector will be

developed by introducing a transcriptional inhibiting cassette (loxP-CAT-loxP) at sites

within the promoter region of the miR-155 and miR-30 expression vector system.

Potent sequences in both conditional and non-conditional vector systems would be used

to generate transgenic knock-down mice to evaluate the efficacy of RNAi in mice by

attempting to silence genes whose knock-outs give rise to strong phenotypes.

Various Cre-fused CPPs will be evaluated by reporter assays and PCR screening for

their ability to effect recombination in vitro. Initial studies in CHO cells will be

performed to identify protocols to attempt to mediate recombination in more

challenging mouse ES cells.

See Figure 1.11 for a flowchart illustration of the study design.


Johanna Edlund

16

Selection

of model

cell line

Construction

of shRNAmiR

expression

vectors

Transfection

RNA analysis Protein analysis

Selection of

shRNAmiR

vector

Construction

of conditional

destination

vector

Find optimal

transfection

conditions

Costructions

of conditional

shRNAmiR

expression

vector

Transfection

Cre-administration

RNA analysis Protein analysis

Verified in vitro

system

Selection

CPPs

Verification of

recombination

ability (CHO cells)

Selection of

CPP

Construction

of stable luc-

expressing

CHOs

Verification of cell

lines

Verification of

recombination

ability (ES cells)

Figure 1.11 Flowchart of the major steps in the study design for the thesis project. Paths display the relationship between each analysis step. Black arrows indicate the main project outline whereas grey arrows indicate the sub project outline. The dashed arrow indicates the intersection between main and sub projects.


Johanna Edlund

17

2 MATERIALS AND METHODS

2.1 Tissue culture

Bl6.F10 mouse melanoma, MC.38 mouse colon adenocarcinoma, LCC-1 Lewis lung

mouse carcinoma, CHO-KI Chinese hamster ovary, and 293 T human embryonic

kidney cell lines were provided from AstraZeneca tissue culture unit (TCU).

Unless stated all cell culture reagents were supplied by Invitrogen and plastic ware from

Costar. All the above cell lines were cultured in Dulbecco’s Modified Eagle’s Medium

(DMEM) supplemented with 10 % fetal calf serum (Hyclone), 1 %

penicillin/streptomycin and 1% non-essential amino acids in a humidified incubator at

37C and 5 % CO2. Cells were passaged and maintained as sub-confluent monolayers.

The embryonic stem cell lines, IDG3:2R26 CAG CAT emGFP shRNA ERG (3:4) and

IDG3.2R26 CAGemGFP shRNAmiR LacZ (Lac8), were obtained from ATCG stocks.

Undifferentiated stem cells were grown on mitotically inactive mouse embryonic

fibroblasts (MEFs) in DMEM media supplemented with G418 at 300 μg/ml, LIF

(Chemicon 10xE7 IU), 2-mercaptoethanol, 1 % non-essential amino acids, 1 %

penicillin/streptomycin and 10 % fetal calf serum. MEFs were cultured in DMEM

supplemented with 1 % non-essential amino acids 1 % penicillin/streptomycin and 10 %

fetal calf serum. Mitotically inactive MEFs were generated by a standard mitomycin C

treatment of 10 μg/ml for 2,5 hours followed by PBS wash.

2.2 Transfection of plasmid DNA into mammalian cells

Transient transfections were performed by separately using FuGENE 6 (Roche) or

ExGen500 (Fermentas Life Science) according to respective manufacturer’s

instructions. FuGENE 6 is a non-liposomal reagent known to usually transfect most cell

types. It can be used in the presence or absence of serum, and previous studies have

shown minimal cytotoxic effects and minimal effects on cell physiology [19].

ExGen500 is a linear polyethylenimine molecule (22 kDa) exhibiting high transfection

efficiency in most cell types, and having minimal cytotoxic effects on cells. It may be

used with our without serum present in cell media [19].

Optimal transfection conditions relating to DNA:transfection reagent ratio were

determined empirically. We tried to optimize the ratio of reagent to DNA to get the

transfection levels as high as possible in the cell population, thereby minimizing the

number of non-transfected cells within the same population. Further, we also observed

the cell viability to assess if any of the ratios resulted in a high number of cell deaths.

We analyzed the cell viability by ocularly observing the amount of cells in tissue culture

dishes 72 hours after transfection. GFP expression analysis revealed the transfection

efficiency. Ratios of FuGENE 6:DNA ExGen500 at 6:1, 1:2, 4:1 3:2 and 3:1 were

evaluated using the GFP expressing plasmid pcDNA6.2GW/EmGFP-miR as a reporter

of transfection efficiency at these ratios. No assay on changes in cell physiology was


Johanna Edlund

18

conducted, as we here chose to rely on results from previous studies performed by the

supplier [19].

Cells were seeded 24 hours prior to transfection to attain 60 % confluency at time of

transfection. Cells were then transfected at these conditions with a FuGENE 6 (μl):DNA

(μg) ratio of 6:1 for CHO cells, and 3:2 for LLC-1. The transfection complex was

incubated at RT for 1 hour then added to the cells. Fresh medium was added after 24

hours. Cells were harvested 72 hours post-transfection. If required, stably transfected

cells were selected over a 14 day period by addition of G418 at 200 μg/ml to the culture

medium.

2.2.1 Transfection efficiency

β-galactosidase activity was used to determine transfection efficiency post-transfection

by using Pierce All-In-One β-galactosidase assay system (Fisher Scientific) according

to manufacturer’s instructions [13]. To determine the transfection efficiency in LLC-1

cells with shRNAmiR vectors we conducted a co-transfection with a CMV-β-

galactosidase plasmid as an internal control at a ratio of 3:1 of experimental plasmid to

reporter plasmid. This ratio had previously been reported to work well [15]. Co-

transfection was followed by a β-galactosidase assay where luminescence was used to

determine transfection efficiency.

Due to the nature of the β-galactosidase assay where the plate has to be discarded upon

completion, we seeded cells on two identical plates and conducted the co-transfection

on the control plate which otherwise had the same conditions as the master plate

following standard transfection protocols for FuGENE6.

2.2.2 Transduction of mammalian cells with CRE-fusion proteins

CHO cells were incubated for 3 hours with various cell penetrating peptides diluted in

PBS, PBS/ExGen500, Optimem or Optimem/ExGen500. In some transductions

chloroquine was added at 100 m. After incubation the peptide mix was replaced with

complete medium for 48 hours prior to harvest for luciferase assay.

Prior to transduction of ES cells with cell penetrating peptides cell viability was

established following PBS incubation at intervals between 0 and 120 minutes. ES cells

were centrifuged and resuspended in PBS and at each time point 10 μl was removed and

mixed with 10 μl of Trypan Blue stain and counted in either an automated cell counter

(Invitrogen) or manually. Following established cell viability interval, ES cells were

incubated in suspension with various cell penetrating peptides for 5, 10, 20 and 30

minutes before seeding ES cells on MEF’s. Cells were harvested 4 days after

transduction and assessed by GFP expression analysis of transduced cells or PCR

analysis of ES cell DNA.

CAGCATemGFPshRNAmiR ERG (3:4) colonies were selected based upon having

GFP expression or not. Colonies were picked from tissue culture plates from

transductions with respective CPP, 2 GFP positive and 2 GFP negative for each.

Selected ES cell Clones were lysed by Proteinas K lysis buffer at 37˚C for 1 hour

followed by incubation at 85˚C for 10 minutes. The PCR mastermix used for

amplification was Extensor HI-Fi (Thermo scientific) and primers used were; number

18 and 19. For primer ID, name, sequence and Tm of the primers, see Appendix 1.


Johanna Edlund

19

A NTC negative control was used and the CAGemGFPshRNAmiR LacZ (Lac 8) cell

line was used as control for unrecombined CAGCATemGFPshRNAmiR ERG (3:4)

cells.

2.2.3 Luciferase Assay

Transfected cells were assayed for luciferase using Steady Lite HTS Luciferase system

(Perkin Elmer) according to manufacturer’s instructions. Relative light units (RLU)

were measured using a 1420 VICTOR Light luminometer (Perkin Elmer), after a 1 s

delay, over a 10 s integration period. Luciferase activity was standardized to the protein

content of each sample determined using a BioRad protein assay (BioRad). Protein

concentrations were calculated from a BSA standard curve and luciferase activity was

normalized to protein content and expressed as relative light units/g protein.

2.3 DNA purification and analysis

2.3.1 Transformation of plasmid DNA

Plasmids used for cloning and expression were obtained from ATCG or purchased from

Invitrogen or Open Biosystems. Plasmids were transformed into chemically competent

E. coli (DH5α) or One shot TOP10 (Invitrogen). Plasmid DNA was routinely added to

100l E.coli and then incubated on ice for 30 min. The cells were heat-shocked by

incubating at 42C for 45 s, and then incubated on ice for 2 min. 900 l Luria Bertani

(LB) medium was added before incubation at 37C for 1 hour with gentle shaking (300

rpm). After incubation, 200 l of the cell suspension was spread on pre-warmed LB

agar-plates containing the appropriate antibiotic. Plates were incubated at 37C over

night.

2.3.2 Purification of plasmid DNA

Single colonies were transferred to 3 ml LB medium containing antibiotic, and

incubated at 37C with gentle shaking (300 rpm) over night. 2 ml of the bacterial

suspension was harvested by centrifugation (13 000 rpm for 1 min). Plasmid DNA was

prepared using QIAprep Spin Miniprep Kit (Qiagen), according to manufacturer’s

instructions. DNA yield was determined spectrophotometrically by using a NanoDrop

ND-1000 Spectrophotometer (NanoDrop Technologies). Plasmid DNA was verified by

restriction enzyme digestion. Digested plasmid DNA was analysed on a 1% ethidium

bromide agarose gel and visualized under ultraviolet light using standard techniques.

Fragment sizes were verified with DNA markers of known length.

Large-scale plasmid DNA preparations were performed using the QIAGEN Plasmid

Maxi Kit (Qiagen) according to manufacturer’s instructions.

DNA fragments to be recovered from agarose gels were excised in a minimum of gel

using a sterile scalpel blade. DNA was isolated from the gel slice using the QIAquick

Gel Extraction Kit (Qiagen), according to manufacturer’s instructions.

If necessary terminal phosphates were removed following restriction enzyme digest

using Antarctic Phosphatase (Roche), according manufacturer’s instructions.


Johanna Edlund

20

2.3.3 Extraction of ES cell genomic DNA

Single ES cell clones were picked in PBS and centrifuged to at 3000 RPM for 5 minutes

to pellet cells. Cells were resuspended in lysis buffer (17 M SDS, 1X MGB, 0.1mg/ml

ProteinaseK) and incubated for 60 minutes at 37C followed by incubation at 85C for

10 minutes.

2.3.4 Ligation of DNA fragments

Linearized DNA fragments were ligated using the Rapid DNA ligation Kit (Roche)

according to manufacturer’s instructions. Ligation reactions were transformed into

DH5 cells and minipreps were performed and analysed as described above.

2.4 PCR conditions

PCR reactions were performed on a PTC-200 Peltier Thermal Cycler (MJ Research)

and optimised depending on amplicon size and the Tm (˚C) for each primer. Unless

indicated Extensor High Fidelity PCR MasterMix (Thermo Scientific) was routinely

used for amplifications according to manufacturer’s instructions.

QIAquick Nucleotide Removal Kit (Qiagen) was used to purify PCR reactions before

performing downstream applications. Following agarose gel purification, extraction of

PCR products was performed using the QIAquick Gel Purification Kit, according to

manufacturer’s instructions.

2.4.1 TOPO cloning

If necessary, PCR fragments were cloned into a TA cloning vector using TOPO TA

Cloning KIT (Invitrogen).

The TOPO TA cloning procedure adds 3À overhangs to the PCR product by TAQ

Polymerase and dNTPs action. This aid cloning into the pCR 4-TOPO (3956 bp) vector

which carries an ampicilin resistance marker and sites for M13 priming allowing for

sequencing with M13 primers. Therefore it may be used for transformation into E.coli

and further selection of resistant colonies and at the end sequencing for the PCR

product, see Figure 2.1.


Johanna Edlund

21

Figure 2.1 Illustration of pCR®4-TOPO vector. The localization of the cloning site were the

generated PCR product having A overhangs are cloned are Illustrated. The M13 priming sites and the antibiotic resistance markers location are also indicated.

The TOPO cloning reaction uses Topoisomerases for its reaction. The Topoisomerase

binds to duplex DNA at specific sites and cleaves after 5ĆCCTT in one of the strands.

This cleavage produces energy which is conserved by a covalent bond between the

3´phosphate of cleaved strand and Tyr-274 residue of the Topoisomerase. The

5´hydroxyl remaining from the cleaved strand may attack the phosphor-tyrosyl bond

and thereby releasing the Topoisomerase, see Figure 2.2.

Figure 2.2 Illustration of the TOPO cloning reaction. The Tyr.274 residues and the Topoisomearse binding site are indicated in the illustration. An attack by the 5´hydroxyl group releases the topoisomerase after energy conservation produce covalent bonds between the 3´phosphate of cleaved strand and Tyr-274 residue of the Topoisomerase.


Johanna Edlund

22

2.5 Plasmids and cloning

Five shRNAmiRs were designed as perfect matches for both mouse KDR or NRP-1

using RNAi Express, the RNAi designer software provided from Invitrogen

(http://rnaidesigner.invitrogen.com/rnaiexpress/). Target sequences for mRNAs were

chosen using manufacturer´s documented rules. Double stranded oligos (see Appendix

2) were designed accordingly with overhangs allowing directional cloning using the

BLOCK-iT Pol II miR RNAi expression vector kit (Invitrogen) using standard

molecular biology techniques. Cloned inserts were verified by DNA sequencing, as

described. A negative control shRNAmiR vector from Invitrogen was also purchased.

The two single-stranded oligos were annealed giving a double-stranded oligo which

were cloned into the cloning site (nucleotides 763 and 764) of linearized pcDNA6.2-

GW-miR expression vector using T4-Ligase, see Figure 2.3. The vector was later

transformed into One shot Competent E.Coli cells (Invitrogen) and subsequently

plated onto selective Spectinomycin LB-plates. Positive clones were analysed by

culture and isolation of plasmids by method described previously. Sequencing and

restriction enzyme digest were used to confirm the presence and correct orientation of

ds oligo insert.

Figure 2.3 Flowchart of major steps for producing a pcDNA 6.2-GW/EmGFP-miR expression clone. The generated ds oligo are ligated in between of the miR flanking regions of the pcDNA 6.2-GW/EmGFP-miR expression vector as indicated in the figure.

http://rnaidesigner.invitrogen.com/rnaiexpress/


Johanna Edlund

23

Plasmids containing pre-designed shRNAmir sequences against KDR and NRP-1

situated in the shRNAmiR-30 expressing pGIPZ vector from Open Biosystems were

purchased. The following accession numbers for 3 vectors expressing shRNAmirs

against each gene are as follows. KDR; V2LMM_81666, V2LMM_76432 and

V2LHS_76971, NRP-1; V2LMM_217599, V2LMM_22108 and V2LMM_20817.

Expression vectors for NRP-1 (Accession Number 6409596) and KDR (Accession

Number 4238984) cDNA were also purchased from Open Biosystems. See Appendix 2

for accessory information.

The shRNAmiR expression vectors pcDNA 6.2-GW EmGFP negmiR and the pGIPZ

were modified to include a transcriptional inhibition cassette. A loxP-CAT-loxP

cassette was amplified from ATCG plasmid 497, see Figure 2.4, using Phusion Hot

Start polymerase (Finnzymes). Primers were designed to incorporate SnaBI, NotI or

SacI restriction sites and extra nucleotides (indicated by bold letters) to the 5ènd of the

generated strand for efficient digestion by respective enzyme at the ends of the PCR

product to allow cloning into pcDNA 6.2-GW EmGFP negmiR or pGIPZ as described

above. Correct orientation of the loxP-CAT-loxP was confirmed by PCR and DNA

sequencing, as described.

For the SnaBI and the SacI amplified amplicon a direct amplification was done but for

the NotI amplification an overlapping PCR by an intermediate step hade to be

performed since a direct PCR did not generate any bands. Primers 105 KMB and 106

KMB were used for the intermediate step. The expected size of bands would be 1700 bp

after analyzing the Vector NTI file of the 497 plasmid used, see Figure 2.4.

p165LCAGCATLo

8640 bp

LoxP

LoxP

CAGprom

CAT

Amp

Luc

PolyA Signal

PolyA Signal

105 KMB

SnaB loxCATlox Forw

SnaB loxCATlox Rev

Not loxCATlox Forw

Not I loxCATlox Rev

106 KMB

Figure 2.4 Vector NTI file of the plasmid number 497 from ATCG plasmid bank. The loxCATlox sequence location is indicated in the illustration. Locations of primers used for amplification of loxCATlox stop cassette insert are also indicated.


Johanna Edlund

24

Expression cassettes containing a CAG promoter driving luciferase expression with or

without a loxP flanked transcriptional inhibiting chloramphenicol acetyl transferase

(CAT) expression cassette between promoter and luciferase were digested from their

parent plasmid (number 497) with Mlu I and BamHI. pCDNA3.1 was also digested

with Mlu I and BamHI and the cassettes were cloned into pCDNA3.1 and verified by

restriction analysis, as described above. This allowed the expression from the resulting

plasmids shown in Figure 2.5, CAG-loxP-CAT-loxP-luciferase and CAG-loxP-

luciferase, to be assessed in stably transfected mammalian cell lines, as described.

pCDNA3.1(-)CAGLoxCATLoxLuc

10187 bpneo cassette

Amp

CAG-Lox-CAT-Lox Luciferase

BGH PolyA

BGH Rev erse Primng Site

BamHI (5738)

MluI (229)

pCDNA3.1(-)CAG lox luc

8491 bp

neo cassette

Amp

CAG-Lox-Luciferase

BGH PolyA

BGH Rev erse Primng Site

BamHI (4042)

MluI (229)

Figure 2.5 Vector NTI file of the pCDNA3.1(-)CAGloxCATloxLuc and pCDNA3.1(-)CAGloxLuc. The illustration indicates the location of the ampicillin resistance marker and further also the sites for CagloxCATloxLuc and GagloxLuc.


Johanna Edlund

25

2.6 Protein purification and electrophoresis

Protein was purified from cells 72 hours post-transfection. Culture medium was

removed and cells washed with PBS. RIPA protein extraction buffer (Tris-HCL 50 mM

pH 7,4, 1 % NP-40, 0,25 % Na-deoxycholate, NaCl 150 mM and EDTA 1 mM,

proteinase inhibitors) was used to lyse the cells. Lysis was performed on ice and 200 l

of buffer was added to each 3 cm plate. Plates were rocked gently for 15 minutes to aid

lysis. Samples were transferred to microtubes and incubated at 70oC for 10 minutes

prior to loading on a SDS-PAGE gel.

Synthesized fusion proteins were thawed and incubated at 70oC for 10 minutes prior to

loading on a SDS-PAGE gel.

Proteins were analyzed by denaturing dodecylsulphate-polyacryalimide gel

electrophoresis (SDS-PAGE). SDS-PAGE loading buffer (Invitrogen) was added to the

samples prior to loading on to a NuPAGE 4-12% Bis-Tris gel (Invitrogen). Samples

were subjected to electrophoresis at 120V for 1 hour in an X-Cell Sure Lock (Novex).

Molecular weights of proteins of interest were determined using pre-stained molecular

weight standards, Standard Sea Blue 2 ladder (Invitrogen). If necessary, gels were

stained with a coomassie blue solution to directly visualize proteins using a standard

protocol.

2.6.1 Quantitative western analysis

To correct for protein concentration following β-galactosidase assay, reducing agent

and loading buffer were added to protein lysates and samples boiled for 10 minutes

followed by SDS-PAGE gel electrophoresis, as described above. β-actin was used as

endogenous control to normalize loading. Gels were transferred to PVDF membranes

by electroblotting for 1 hour at 250 V. Membranes were placed in blocking buffer for 2

hours and incubated overnight with primary antibodies against β-actin and NRP-1

(Invitrogen) at a dilution of 1:300. Blots were washed 4 times in PBS followed by

incubation with alkaline phosfatase (AP) secondary goat anti-rabbit and rabbit anti-goat

antibodies at a dilution of 1:3000 for 2 hours.

After PBS washing the alkaline phosfatase (AP) antibody signal was detected by

chemiluminiscence using a ChemDoc camera and software (Applied Biosystems).

Protein band densities were analyzed using Quantity One spot density software

provided with the ChemDoc apparatus.

2.7 RNA isolation and analysis

Total RNA was isolated 72 hours post-transfection. Cells were lysed directly in the

culture plates by adding RNASTAT 60 (TLT-Tet Inc) and homogenized by passing the

lysate several times through a pipette. RNA was prepared according to manufacturer’s

instructions. RNA yield was determined spectrophotometrically by using a NanoDrop

ND-1000 Spectrophotometer. RNA was stored at –80oC until analysis.

2.7.1 RT-PCR cDNA synthesis

Total RNA was reverse transcribed using an oligo (dT) primer and SuperScript first-

strand synthesis kit for RT-PCR (Invitrogen). To ensure that there was no genomic


Johanna Edlund

26

DNA contamination, reverse transcription of the samples was also conducted without

the use of reverse transcriptase. The resulting cDNA was subjected to PCR analysis

using specific primers using either standard PCR amplification or Real time RT PCR

PCR with primers designed for the relevant target, see Appendix 1.

2.7.2 Quantitative mRNA detection by real-time PCR

Quantitative PCR was performed on a Taqman 7500 Real Time PCR system (Applied

Biosystems) with SYBRgreen mastermix (Applied Biosystems). The average threshold

cycle for each gene was determined from triplicate reactions and the level of gene

expression was normalized to the constitutively expressed internal reference gene

Glyceraldehyde-3-phosphate dehydrogenase (GAPDH). The Real time RT PCR

conditions were set to 40 cycles of 5 seconds at 95˚C and 35 seconds at 60˚C. This was

followed by a final temperature gradient to profile the melting curve. Conditions for the

melting curve program were 95˚C for 15 seconds, 60˚C for 1 minute followed by 95˚C

for 15 seconds. Target mRNA levels of shRNAmiR treated samples were calculated

relative to their respective untreated control template using the delta Ct method and

results normalized to GAPDH [9].

2.8 Sequencing

All sequencing was performed on a 3730 DNA Genetic analyser (Applied Biosystems)

using ABI BigDye Terminator Cycle sequencing Kit (Applied Biosystems).

Analysis of sequences was performed with software from DNASTAR Inc, Seqman II.

PCR amplification for sequencing was performed in a PTC-200 Peltier Thermal Cycler

(MJ Research) according to a standard protocol with 25 cycles of 95˚C for 10 seconds

followed by 20 seconds at 50˚C. For primers used for sequencing see Appendix 1.

The generated expression vector, pcDNA 6.2-GW EmGFP miR, was sequenced to

confirm the sequence of the dsoligo insert and further also for the orientation and

presence of the loxCATlox stop cassette insert. The vector was digested within the

miRNA loop by MscI enzyme and subsequently purified by method previously

described. Primers used were EmGFP forward sequencing primer for forward

sequencing and miRNA reverse sequencing primer for reverse sequencing, both

provided from Invitrogen, see Appendix 1.

The generated pGIPZ destination vector was sequenced in an attempt to verify the

loxCATlox stop cassette insert. The vector was digested with restriction enzyme and

subsequently purified by method previously described. Primers used have ID 28, 29, 30

and 31 see Appendix 1 for name, sequence and Tm.

The TOPO TA generated PCR amplicon were sequenced to confirm orientation and

presence of product. M13 Forward (-20) and M13 Reverse primers were used, see

Appendix 1 for details.


Johanna Edlund

27

2.9 Statistical analysis

All values are given as group means ± standard error of the mean (SEM). Error bars are

used to display the SEM, which is calculated according to the formula below where s is

the standard deviation and n is the sample size.

A two tailed Students t-test was used to compare and calculate shRNAmiR constructs

relative to the negative miR, p-values of p < 0,05 were considered significant.

2.10 Software

DNAstar software was used for bioinformatic manipulations. The Vector-NTI

bioinformatics tool was used to construct plasmid maps and to plan and analyze results

from restriction digests and PCR amplifications. Primers were designed using the

AstraZeneca bioinformatics portal, e-lab, see Appendix 1.


Johanna Edlund

28

3 RESULTS

3.1 Cell line expression and transfection analysis

3.1.1 Selection of model cell line

Initial experiments were performed to select a cell line with endogenous expression of

our target genes and to determine empirically optimal transfection conditions for that

cell line.

Cell lines chosen for analysis were MC.38, LLC-1, and BL6.F10, all three from

AstraZeneca tissue unit. These mouse cancer cell lines were picked as potential

candidates for having endogenous expression of KDR and NRP-1. In addition, CHO KI

cells from AstraZeneca tissue unit were amended to the selection since it is possible to

transfect them with full length cDNA and enable silencing studies despite their lack of

endogenous expression of target genes.

RNA from each cell line was reverse transcribed and cDNA was analyzed by PCR

using primers specific for NRP-1 or KDR.

Figure 3.1 KDR and NRP-1 specific PCR products from amplification of cDNA from LLC-1 and B16.F10. Primer pair numbering corresponds to that shown in Appendix 1. Arrows indicate bands of expected size. -RT samples were used as controls. Abbreviations: +RT, reverse transcriptase reaction, -RT, without reverse transcriptase,* indicates primer ID, see Appendix 1.

As shown in Figure 3.1, NRP-1 specific primers gave bands of expected size (482 bp

and 535 bp from primer pairs 1,3 and 2,4 respectively) indicating that NRP-1 is

endogenously expressed in the LLC-1 and B16.F10 cells. KDR specific primers gave a


Johanna Edlund

29

band of expected size from only one of the primer pairs (456 bp from primer pair 5,6).

The negative controls (-Reverse Transcriptase samples) verified the reliability of the

analysis.

Figure 3.2 KDR and NRP-1 specific PCR products from amplification of cDNA from MC.38. Primer pair numbering corresponds to that shown in Appendix 1. Arrows indicate bands of expected size. -RT samples and positive control RNA samples were used as controls. Abbreviations: +RT, reverse transcriptase reaction, -RT, without reverse transcriptase,* indicates primer ID, see Appendix 1.

The expresstion analysis results in Figure 3.2 reveal that KDR specific primers failed to

generate any of expected bands from primer pairs 5,6 or 5,7 from MC.38 cDNA.

However MC.38 cells were shown to have endogenous expression of NRP-1 as both

primer pairs 1,3 and 2,4 used for the NRP-1 generated bands of expected size (482 bp

and 535 bp respectively). The negative controls (-Reverse Transcriptase samples) and

the positive controls (control RNA) verified the reliability of the analysis.

Given these results, LLC-1 and B16.F10 cell lines were selected for continued studies.

3.1.2 Optimization of mammalian cell line transfection

Optimization of the transfection protocol for cells used in this study was determined

empirically. Monlayers of LLC-1, Bl6.F10 and CHO cells were transfected with

FuGENE 6:DNA and ExGen500:DNA complexes where the amount of DNA was held

constant and the amount of transfection reagent was increased. Ratios of transfection

reagent:DNA at 6:1, 4:1, 3:1, 2:1 and 3:2 were evaluated using the GFP expressing

plasmid pcDNA6.2GW/EmGFP-miR as a reporter of transfection efficiency at these

ratios. Analysis of cell viability and cytotoxic effects was done by ocularly observing

the amount of cells in tissue culture dishes 72 hours after transfection.


Johanna Edlund

30

GFP expression indicated that FuGENE 6 was chosen over ExGen500 as the

transfection reagent best suitable for the continued studies. Results from FuGENE 6

GFP analysis are shown in Table 3.1 through Table 3.4.

CHO cells were the easiest to transfect, and most efficiently transfected at the FuGENE

6:DNA ratio of 6:1. LLC-1 cells were more efficiently transfected than Bl6.F10 cells by

exhibiting the highest levels of transfected cells and the transfection ratio 4:1 was the

most efficient for transfecting LLC-1 cells while still maintaining low cytotoxic effect.

As expected no GFP signal could be detected for untransfected samples.

Concentration Phase contrast GFP

6:1

3:1

3:2

Table 3.1 Photo images of the different transfection ratios used in CHO KI cells. The phase contrast images show the distribution and amount of cells and the corresponding GFP images show the efficiency of the respective transfection ratio.


Johanna Edlund

31


6:1

4:1

3:1

2:1

3:2

Table 3.2 Photo images of the different transfection ratios used in LLC-1 cells. The phase contrast images show the distribution and amount of cells and the corresponding GFP images show the efficiency of the respective transfection ratio.


Johanna Edlund

32


6:1

3:1

3:2

Table 3.3 Photo images of the different transfection ratios used in B16.F10 cells. The phase contrast images show the distribution and amount of cells and the corresponding GFP images show the efficiency of the respective transfection ratio.


Johanna Edlund

33

Wild type Phase contrast GFP

LLC-1

B16.F10

CHO

Table 3.4 Photo image of wild type controls for LLC-1 and B16.F10 cells. The phase contrast images show the distribution and amount of cells and the corresponding GFP images show wt cells being untransfected.

3.1.3 Establishing transfection efficiency

Determination of transfection efficiency in LLC-1 cells for the western blot analysis

was evaluated by co-transfecting 5 NRP-1 specific shRNAmiR-155 vectors with a

control β-galactosidase expression plasmid. Co-transfections were run on replica plates

to allow GFP expression analysis.


Johanna Edlund

34

Figure 3.3 β-galactosidase assay of transfection efficiency. Graphs showing the absorbance level in duplicate β-galactosidase assays (a and b), indicating transfection efficiency for each of 5 NRP-1 specific shRNAmiR-155 constructs and also for a negative shRNAmiR-155 construct. All data points are mean values with error bars showing the standard error of the mean (SEM) (n=3).

Differences were seen within and between transfections in the β-galactosidase assay,

but GFP expression analysis revealed similar GFP signal strengths in all wells

indicating equal transfection efficiency throughout all wells. Therefore results from the

β-galactosidase assay revealed that GFP observation fails to reveal minor differences in

transfection efficiency. Results from the β-galactosidase assay were used to normalize

lysates from each well before initiating downstream assays. However, the variation

within each shRNAmiR triplet, denoted by the error bars in Figure 3.3, indicated that

this normalization may contain deviations, especially for the triplets shRNAmiR4 in run

a and shRNAmiR1 in run b.

a)

b)


Johanna Edlund

35

We also tried to use x-gal staining to measure the transfection efficiency but did not

find an optimal protocol generating useful results. After two attempts, this approach was

therefore abandoned.

3.2 shRNAmiR expression constructs

shRNAmiRs expression vectors designed using Invitrogen’s miR-155 based BLOCK-

iT system or Open Biosystems miR-30 based systems for shRNAmiR design to target

the genes KDR and NRP-1 were tested to identify shRNAmiR constructs giving high

levels of knock-down of these genes.

Transient transfections in LLC-1 cells were performed to assess and compare the gene

silencing effect of these constructs upon endogenous expression of these genes.

Analysis of exogenous NRP-1 and KDR expression by the same constructs was

performed in CHO cells co-transfected with full length cDNA’s for either NRP-1 or

KDR. Transfected cells were analyzed at the mRNA and protein level to estimate

degree of gene silencing.

3.2.1 Verification of shRNAmiR constructs

To verify the insert of the correct shRNAmiR sequence into the expression vector,

sequencing with provided sequencing primers were conducted. For primers, see

Appendix 1. The generated sequences were assembled and compared to respective

shRNAmiR ds oligo insert, which sequences were known. Contiques were aligned and

analyzed by using SeqMan II software from DNAstar. All aligned sequences verified

correct insert of respective shRNAmiR into the vector. Since the information from the

results is difficult to illustrate due to large amount of information data, no images of

alignment are included in this report.

The Open Biosystems shRNAmiR vectors aiming for KDR and NRP-1 mRNA were

ordered ready to use from the manufacturer, therefore no sequencing was done to verify

the vectors. Selection from ampicillin containing agar plates were the only verification

of vectors, since they should contain an ampicillin resistant marker, see Figure 1.10.

3.2.2 Real time RT PCR mRNA analysis

Real time RT PCR analysis was used to measure NRP-1 and KDR mRNA levels of

shRNAmiR transfected cells compared to a negative shRNAmiR transfected control.

Quantitative expression analysis results were normalized to the constitutively expressed

reference gene GAPDH. For each run, a melting curve analysis was performed to assess

the accuracy of the SYBR green measurement and standard curves for both internal

control and the negative shRNAmiR sample were included. Standard curves were used

to verify the reliability of the reaction conditions.

All cDNA and internal control samples were run in triplicate. Mean Ct values were

calculated and samples were normalized against GAPDH Ct values using the

comparative Ct method.

Assuming 100 % PCR efficiency, each Ct represents an amplification of the product by

a factor of two, resulting in the 2-ΔCt

formula which gives the possibility to compare

expression levels of knock-down relative to the negative shRNAmiR control.


Johanna Edlund

36

Samples where reverse transcriptase was not added (-RT) and no template control

(NTC) were used to verify cDNA synthesis and to assess the purity of the samples,

respectively.

Figure 3.4 Dissociation curves for Real time RT PCR assay of shRNAmiR knock-down of NRP-1 gene. Melting points for the amplified internal control GAPDH product and the NRP-1 product generated from a dissociation program included in each Real time RT PCR run. The panel on the left represents results from transfections using the miR-155 shRNAmiR vectors and the right panel represents results from transfections using miR30 shRNAmir vectors.

Figure 3.5 Standard curves for Real time RT PCR assay of shRNAmiR knock-down levels of NRP-1 gene. Linear equations of internal control GAPDH and negative shRNAmiR controls.


Johanna Edlund

37

Figure 3.6 RNAi knock-down of NRP-1 using shRNAmir miR155 specific constructs. Levels of NRP-1 mRNA after knock-down by shRNAmiR-155 constructs (1 to 5) expressed relative to the negative shRNAmiR-155 control construct. All data points are mean values with error bars showing the standard error of the mean (SEM) (n=5). ** indicates significant difference in Student’s t-test with threshold p < 0.05.

Figure 3.7 RNAi knock-down of NRP-1 using shRNAmir miR30 specific constructs. Levels of NRP-1 mRNA after knock-down by shRNAmiR-155 constructs (1 to 5) expressed relative to the negative shRNAmiR-30 control construct. All data points are mean values with error bars showing the standard error of the mean (SEM) (n=5). No significant differences between groups were seen.

Melting curves in Figure 3.4 showed satisfactory results, amplification of products and

intercalation of SYBRgreen were specific. Figure 3.4 displays results from one Real

time RT PCR run for each shRNAmir expression system, subsequent Real time RT

PCR runs displayed similar melting curves.

The standard curves in Figure 3.5 show a coefficient of determination, R2, which is

close to one indicating reliability of primer pair and efficiency of the model. Also, the

internal control GAPDH and the negative shRNAmiR both show similar k-values and

R2-values which indicate that the satisfactory choice of internal control.


Johanna Edlund

38

Real time RT PCR analysis was run five times for both miR-155 and miR-30 constructs

and average knock-down levels for each shRNAmiR construct were calculated along

with corresponding standard error of the mean. Student’s t-test with threshold p < 0.05

was used to test significance.

Results show that the construct most effective in silencing NRP-1 expression was

shRNAmiR-155 construct 2, which resulted in 54% less expression compared to the

negative control shRNAmiR shRNAmiR with a significance of p < 0.05. ShRNAmiR-

155 constructs 1, 3, 4,and 5 gave reductions in expression of 29 %, 38 %, 21 % and 17

% respectively, relative to the negative control construct, see Figure 3.6.

Real time RT PCR analysis of miR30 mediated silencing showed that the construct

most effective in silencing NRP-1 expression was shRNAmiR-30 construct 1, which

resulted in 58% less expression compared to the negative control shRNAmiR.

ShRNAmiR-30 constructs 2 gave a reduction in expression of 34 %, whereas construct

3 gave increased expression relative to the negative control construct, see Figure 3.7.

Therefore RNAi is indicated in our results, but the large variations in NRP-1 silencing

seen with only one construct exhibiting a statistically significant reduction in expression

compared to control (p < 0.05) infers that the silencing level of each construct is

difficult to assess.

Real time RT PCR studies of mir-155 and miR-30 mediated silencing of KDR

expression failed to show gene silencing compared to controls (data not shown).

3.2.3 Quantitative western analysis

The efficacy of shRNAmiR-155 silencing of NRP-1 gene expression was also

investigated at the protein level using quantitative western analysis. β-galactosidase

assays described above established the efficiency for each shRNAmiR transfection and

results were used to normalize each protein sample following transfection by diluting

each lysate based on individual concentration prior to loading. When studying blots

incubated with NRP-1 specific antibodies the generated bands shown in Figure 3.8

indicates as if we have an isoform of NRP-1 since we have two bands relatively close to

each other. This isoform may be due to alternative transcript splicing.

Figure 3.8 Western analysis of RNAi knock-down of NRP-1 using shRNAmir miR155 specific constructs. Upper panels show detection using a goat NRP-1 primary antibody and an alkaline phosphatase labeled rabbit anti-goat secondary antibody. Lower panels show detection using a rabbit β-actin primary antibody and an alkaline phosphatase labeled goat anti-rabbit secondary antibody. Images a and b show results from two separate experiments.


Johanna Edlund

39

Figure 3.9 Quantification of western analysis of RNAi knock-down of NRP-1 using shRNAmir miR155 specific constructs. Levels of Nrp-1 protein after knock-down by shRNAmiR-155 constructs (1 to 5) expressed relative to shRNAmiR-155 negative control. Results were quantified by densitometric analysis of the autoradiogram derived from the upper NRP-1 panel after normalization to the lower β-actin control panel. All data points are mean values with error bars showing the standard error of the mean (SEM) (n=2). *For shRNAmiR3 no variation estimate could be determined due to insufficient readings.

The qquantitative western analysis shown in Figure 3.9 indicate that the constructs most

effective in silencing NRP-1 expression were shRNAmiR-155 constructs 1 and 3, both

resulting in 80 % less expression compared to the negative shRNAmiR.

However, as only one of the runs for construct 3 generated valid values the results for

this construct should be used cautiously, therefore construct 1 was selected from the two

most effective constructs. Constructs 2, 4 and 5 gave reductions in expression of 62 %,

52 % and 13 % respectively, compared to the negative shRNAmiR construct.

3.3 Generation of a Cre inducible shRNAmiR expression system

3.3.1 Amplification of lox-CAT-lox stop cassette insert

To verify the pGIPZ vector an analytical cleavage was conducted which confirmed that

the starting vector was correct. After verification of the correct band sizes an antharctic

phosphatase treatment was conducted to prevent self-ligation of the vector at the

ligation step.

The amplification of the stop cassette insert having SnaBI and NotI ends were expected

to give a PCR product of 1700 bp. The NotI PCR product amplification required an

intermediate PCR step to generate the final stop cassette insert, see Figure 3.10. The

resulting products were then verified by agarose gel electrophoresis after digest with

SnaBI or NotI restriction enzyme, see Figure 3.11 and Figure 3.12 respectively.


Johanna Edlund

40

Figure 3.10 Photo image of NotI intermediate PCR products. For generation of PCR product of 1327 bp primers 11 and 16 were used, and for generation of PCR product of 783 bp primers 10 and 17 were used. For detailed information of primers, see Appendix 1.

Figure 3.11 Photo image of SnaBI digested PCR product. PCR amplification of loxCATlox with primers adding SnaBI nucleotides to the ends generated an expected PCR product of 1700 bp. This product was digested with SnaBI restriction enzyme to facilitate cloning into the expression vector.

Figure 3.12 Photo image of NotI digested PCR product. PCR amplification of loxCATlox with primers adding NotI nucleotides to the ends generated an expected PCR product of 1700 bp. This product was digested with NotI restriction enzyme to facilitate cloning into the expression vector.


Johanna Edlund

41

The ligations of the antharctic phosphatase treated pGIPZ vector and the stop cassette

insert were performed by using T4 ligase at a ratio of insert (35 ng/μl) to vector (50

ng/μl) of 1:3. Selection of colonies containing pGIPZ vector was aided by an ampicillin

resistance marker. After the transformation into TOP 10 E.coli (Invitrogen) eight

colonies on the NotI ligation plate were found. However, no colonies were growing on

the SnaBI ligation plate. A control plasmid provided from Invitrogen and a vector-only

control ligation were used as controls, and they both displayed satisfactory ligation and

transformation results.

3.3.2 PCR screen and sequencing to verify lox-CAT-lox insert in pGIPZ vector

A PCR screen was performed on the individually selected colonies to verify the cloning

of loxCATlox stop cassette inserts. However, the PCR did not generate a product which

indicated that no cloning of loxCATlox stop cassette insert into pGIPZ vector had

occurred, see Figure 3.13.

Figure 3.13 Photo image of PCR screen of colonies from ligation of pGIPZ vector and stop cassette insert NotI. Primers used have ID 16 and 17, see Appendix 1. Hef511 plasmid was used as positive control. The negative control is a no template control (NTC) sample.

Further sequencing was conducted on minipreps from the eight NotI colonies. Primers

used were NotI Sequencing Forward and NotI Sequencing Reverse, see primer details

in Appendix 1. However, no alignment of contiques against known sequence to

generated sequence could be found. This supports previous results indicating that no

cloning of loxCATlox stop cassette inserts into pGIPZ vector at the SnaBI or the NotI

site had occurred.

3.3.3 TOPO TA cloning

In an attempt to verify the loxCATlox stop cassette PCR product a TOPO TA cloning

and subsequent sequencing was conducted.

An Eco RI restriction enzyme digest was performed on four selected colonies from

SnaBI transformation plates and three colonies from NotI transformation plates. This


Johanna Edlund

42

gave correct band sizes for two of the NotI samples but for none of the SnaBI samples,

see Figure 3.14. These two NotI samples were selected for subsequent sequencing.

Since the digested SnaBI colonies did not show any expected band sizes another

selection of individual colonies from SnaBI transformation plates was done. Additional

NotI colonies were also selected. A PCR screen was run on these colonies to search for

loxCATlox stop cassette insert and bands of correct sizes were generated, see Figure

3.15.

A number of samples were chosen from the screen along with the two previously

selected NotI samples for sequencing with M13 forward and reverse primers to verify

loxCATlox stop cassette insert. The generated sequences were assembled and compared

to a created Vector NTI file of our pCR 4-TOPO vector containing our stop cassette

insert. For vector design see section 0. Contiques were aligned and analyzed by using

SeqMan II software from DNAstar. Sequencing did not verify the correct stop cassette

sequence, therefore the conclusion was that the amplified PCR product was not

correctly inserted at the NotI or the SnaBI site.

Figure 3.14 Photo image of pCR®4-TOPO vector digested with Eco RI. The digest was

expected to generate bands of sizes 3921 bp, 1396 bp, and 266 bp. The NotI lanes indicated with a * displayed all three band sizes and were chosen for subsequent sequencing.


Johanna Edlund

43

Figure 3.15 Photo image of PCR screen of pCR®4-TOPO vector for loxCATlox stop

cassette insert. Primers used have ID 16 and 17, see Appendix 1. Hef511 plasmid was used as positive control. The negative control is a no template control (NTC) sample. Lanes indicated with * were selected for sequencing.

3.3.4 pcDNA 6.2-GW EmGFP miR destination vector

Since the stop cassette insert of the pGIPZ destination vector could not be confirmed a

second attempt to generate a Cre regulatable shRNAmiR destination vector was done in

the pcDNA 6.2-GW EmGFP miR vector. A transcriptional inhibiting cassette

containing a loxP flanked CAT gene with either SnaBI or SacI restriction sites was

cloned as previously described in section 2.5.


Johanna Edlund

44

3.3.5 Amplification of loxP-CAT-loxP cassette

A PCR screen was run to verify insertion of the loxP-CAT-loxP cassette into the

pcDNA 6.2-GW EmGFP miR vector. Agarose gel electrophoresis indicated bands of

expected size, see Figure 3.16. Correct orientation following cloning of the cassette was

confirmed by sequencing.

Figure 3.16 PCR screen for loxP-CAT-loxP cassette insertion into pcDNA 6.2-GW EmGFP miR vector. The amplified SnaBI and SacI PCR products showed a band of 2000 bp. Primer pairs used are numbered, see Appendix 1. A no template control (NTC) was used as negative control.

3.3.5.1 Test of transcriptional inhibition

GFP expression from the pcDNA6.2 GW EmGFP miR vector was used to verify the

cloning and efficacy of the transcriptional inhibiting cassette. Loss of GFP expression

indicates successful insertion and promoter silencing of the loxP-CAT-loxP cassette.

Restoration of GFP expression would indicate successful Cre mediated recombination.

Transient transfections of CHO cells, followed by GFP expression analysis, with

vectors containing loxP-CAT-loxP cloned into the pcDNA6.2 GW EmGFP miR vector

at either SnaBI or SacI sites GFP analysis revealed no GFP expression when the

cassette was inserted at the SacI site, but when the cassette was inserted at the SnaBI

site GFP expression was unaffected, see Table 3.5.

Transient transfections of Cre expressing CHO cells, followed by GFP expression

analysis, of the vector containing loxP-CAT-loxP cloned into the pcDNA6.2 GW

EmGFP miR vector at the SacI site restored GFP expression, see Table 3.6.


Johanna Edlund

45

Phase contrast GFP

SnaBI loxP-CAT-

loxP Insertion

SacI loxP-CAT-

loxP Insertion

Untransfected

cells

Table 3.5 CHO cells transfected with vectors containing a loxP-CAT-loxP cassette inserted at either SnaBI or SacI sites of pcDNA6.2 GW EmGFP miR. The phase contrast images show the distribution and amount of cells and the corresponding GFP images show GFP signal in transfected cells. Untransfected cells are used as a control.


Johanna Edlund

46

Phase contrast GFP

CHO Cre cells

transfected with

loxP-CAT-loxP

SacI pcDNA6.2

GW EmGFP miR

CHO Cre cells

transfected with

loxP-CAT-loxP

SacI pcDNA6.2

GW EmGFP miR

Untransfected

cells

Table 3.6 Cre expressing CHO cells transfected with a vector containing a loxP-CAT-loxP cassette inserted at the SacI site of pcDNA6.2 GW EmGFP miR. The phase contrast images show the distribution and amount of cells and the corresponding GFP images show the restored GFP signal. Untransfected CHO cells were used as control.

3.4 Recombination through Cre-fused cell penetrating peptides

Cre fused cell penetrating peptides (CPP) were evaluated for their ability to penetrate

cells and generate recombination at loxP sites in vitro. A number of Cre-fused CPP

were available for testing and as a first step these proteins were verified for size and

purity by SDS PAGE and Coomassie blue staining to visualize the proteins, see Figure

3.17.


Johanna Edlund

47

Figure 3.17 SDS PAGE electrophoresis and coomassie blue staining of various Cre-fused CPPs. Protein preparations on the right membrane display no sign of degradation and proteins were found to be of correct size. On the left membrane the HI-4f, TNC and HNC protein preparations were not intact enough, the other proteins displayed satisfactory results regarding degradation and size.

Proteins suitable for further study were HHNC iösi, TC, TNC sdcf, TNC g25cf, HNC,

HHNC dg25, HHNC s30, HHNC deae, HHNC pd10, HNCM dg25, HNCM refold and

HTNCM. Further details on these proteins can be found in Appendix 3.

3.5 Generation of a CHO reporter line for Cre-Fused CPP evaluation

To generate a reporter cell line for working with Cre-fused CPP, the expression plasmid

pCDNA3.1 was modified to include either CAG-loxP-CAT-loxP-luciferase or CAG-

loxP-luciferase cassettes, allowing expression of luciferase to be assessed in stably

transfected mammalian cell lines, see Figure 2.5.

To identify stably transfected CAG-loxP-luciferase clones with high expression of

luciferase individual clones were picked following G418 selection and a luciferase

assay was performed. A Bradford Protein assay was first conducted to determine the

protein concentration in each clone sample. Standard curves were established, see

Figure 3.18, and used to translate measured absorbance for each CAGloxLuc and

CAGloxCATloxLuc clone sample into protein concentrations.

Individual stably transfected CAG-loxP-CAT-loxP-luciferase clones were transfected

with a Cre expression plasmid, CAGKoznlsCre pHarrie followed 48 hours later by

luciferase assay. Results are displayed in Figure 3.19 and Figure 3.20.

CAG-loxP-Luciferase clones 18 and 20 and CAG-loxP-CAT-loxP-Luciferase clones 6

and 19 were selected for further studies as these showed highest levels of luciferase

activity.


Johanna Edlund

48

Figure 3.18 Standard curves for Bradford Protein assay. Standard curves used to calculate protein concentration for each CAGloxLuc and CAGloxCATloxLuc clone sample.


Johanna Edlund

49

Figure 3.19 Luciferase assay on CAG-loxP-Luciferase transfected CHO cell clones. Luciferase activity, standardized to the protein content of each sample, of individual clones measured as lum 10s per total protein concentration (μg/10μl). All data points are mean values with error bars showing the standard error of the mean (SEM) (n=2).

Figure 3.20 Luciferase assay on CAG-loxP-CAT-loxP-Luciferase transfected CHO cell clones. Luciferase activity, standardized to the protein content of each sample, of individual clones measured as lum 10s per total protein concentration (μg/10μl). All data points are mean values with error bars showing the standard error of the mean (SEM) (n=2).

3.5.1 Transduction of CHO luciferase reporter cell line with Cre-fused CPP

To assess Cre-fused CPPs for efficacy of recombination, luciferase assays were

performed following treatment of the CHO reporter line CAG-loxP-CAT-loxP-

Luciferase with a range of Cre-fused CPPs. The CAG-loxP-Luciferase line was used as


Johanna Edlund

50

a positive control for luciferase expression in these studies. Also tested was the effect of

different diluents for the Cre-fused CPPs by using either PBS or Optimem for the

transductions. Addition of the chemical ExGen500 to evaluate its ability to rupture

endosomes and increase levels of recombination from these proteins was also

investigated.

The luminescence levels in Figure 3.21 indicate that the Cre-fused CPPs giving the

highest levels of luciferase were TNCg25cf, HHNCiösi, HHNCpd10, and HTNCM.

These were chosen for further studies. Comparison of different diluents showed PBS to

be the most effective.

For this set of experiments no correction against protein concentration by Bradford

Protein Assay was done since it was difficult to perform in the 96-well study setup

chosen.

Figure 3.21 Effect of Cre-fused CPP on recombinase mediated luciferase expression following transduction of CAG-loxP-CAT-loxP-Luciferase transfected CHO cells. Max level of luminescence (lum 10s) for each CPP diluted in either PBS, Optimem, PBS/ExGen500 or Optimem/ExGen500. Negative control samples contained no peptide.

3.5.2 Chloroquine enhances Cre-fused CPP transduction

To determine the optimal concentration for each of the four Cre-fused CPPs highlighted

above, proteins were added to the CAG-loxP-CAT-loxP-Luciferase-CHO cell line at

varying concentrations using PBS as diluent. Chloroquine, a chemical reported to

increase the efficiency of transduction experiments, was also tested. Results are

presented in Figure 3.22 and Figure 3.23.

Luciferase assay on CAGloxCATloxLuc


Johanna Edlund

51

Figure 3.22 Dose response curve for Cre-fused CPP addition to CAG-loxP-CAT-loxP-Luciferase-CHO cells in the absence of chloroquine. Luminescence (lum 10s) measured following the addition of proteins at varying concentrations.

Figure 3.23 Dose response curve for Cre-fused CPP addition to CAG-loxP-CAT-loxP-Luciferase-CHO cells in the presence of chloroquine. Luminescence (lum 10s) measured following the addition of proteins at varying concentrations.

The dose-response analysis revealed that for the TNC CPP recombination levels were

highest at a concentration of approximately 45 μg/100 μl. However, the optimal

concentration was seen to vary between different Cre-fused CPPs upon transduction of


Johanna Edlund

52

this line. The data suggests that the optimal recombination levels for HHNC, HTNCM

and HHNCiösi CPPs are likely to be achieved at concentrations greater than those

tested in these experiments. As we were limited in supply of CPPs further

experimentation to determine this empirically were not possible.

The introduction of chloroquine into the transduction complex results in relatively

modest increases in luciferase levels and thus recombination.

3.5.3 Transduction of mouse embryonic stem cells with Cre-fused CPP

Cre-fused CPPs were tested for their ability to mediate recombination in a mouse

embryonic stem (ES) cell line. Given the possibility of increased sensitivity of an ES

cell line upon transduction compared to CHO cells we first evaluated the ES cell lines

ability to withstand prolonged incubation in PBS, our diluent for Cre-fused CPPs. ES

cells were maintained in PBS for up to 120 minutes. Viability of the ES cells was

calculated by counting the number of live and dead cells at each time interval. ES cell

viability, as measured in this study, was found to be relatively constant at all time points

up to 120 minutes, see Figure 3.24.

There were minor differences in manual counting of cells in Buckner chamber

compared to automated counting in cell counter. However the time needed for manual

counting was greatly reduced.

Figure 3.24 Viability of ES cells maintained in PBS. ES cells were placed in PBS for up to 120 minutes. Cells were counted at varying time points and percentage of live and dead cells calculated using Trypan blue stain as well as automated counting (Invitrogen).

3.5.4 Transduction of an ES cell GFP reporter line with Cre-fused CPP

Cre-fused CPPs were tested for their ability to mediate recombination in ES cell lines

where the following cassettes have been targeted in the Rosa26 locus by homologous

recombination, CAG-loxP-CAT-loxP-EmGFP-shRNAmiR-ERG and CAG-loxP-

EmGFP-shRNAmiR-LacZ. The CAG-loxP-CAT-loxP-EmGFP-shRNAmiR-ERG line

will only express GFP upon Cre mediated recombination whereas the CAG-loxP-

EmGFP-shRNAmiR-LacZ line is a positive control for constitutive GFP expression, see

Figure 3.25.


Johanna Edlund

53

CAG-loxP-CAT-loxP-EmGFP-shRNAmiR-ERG was transduced by the four Cre-fused

CPP highlighted above at varying concentrations (250 μg/ml, 500 μg/ml and 700 μg/ml)

and incubation times (5, 10, 20 and 30 minutes). Microscopic analysis of CAG-loxP-

CAT-loxP-EmGFP-shRNAmiR-ERG ES cells treated with Cre-fused CPP revealed

GFP expressing cells, indicating that transduction and Cre mediated recombination had

occurred. However not all cells were recombined within the same transduction

experiment as mixtures of GFP positive and negative cells were seen. The positive

control CAG-loxP-EmGFP-shRNAmiR-LacZ ES cells consistently showed GFP

expression, see Table 3.7.

CAG EmGFP shRNA miR LacZ pA in CB9

7379 bp

FRT

NeobpA

attB

attB

m13U

M13R

ampR

emGFP miR dattB2 polyA

cloning site

CAGprom

LoxP

EmGFP

CAG CAT EmGFP shRNA miR3 pA in CB9

9100 bp

FRT

Neo

bpA

attB

attB

m13U

M13R

ampR

emGFP miR dattB2 polyA

CAG loxp CAT loxp

cloning site

CAGprom

CAT+SV40pA

LoxP

LoxP

M13R

CAG upper

emGFP lower

Figure 3.25 Plasmid maps of CAG-loxP-EmGFP-shRNAmiR-LacZ and CAG-loxP-CAT-loxP-EmGFP-shRNAmiR-ERG. The loxP sites, GFP expression cassette are indicated in both vectors. The transcriptional inhibiting CAT cassette is shown in CAG-loxP-CAT-loxP-EmGFP-

shRNAmiR-ERG.


Johanna Edlund

54

Phase contrast GFP

CAG-loxP-

CAT-loxP-

EmGFP-

shRNAmiR-

ERG, 500

μg/ml TNC

CAG-loxP-

CAT-loxP-

EmGFP-

shRNAmiR-

ERG, 500

μg/ml HHNC

CAG-loxP-

CAT-loxP-

EmGFP-

shRNAmiR-

ERG, 500

μg/ml

HTNCM

CAG-loxP-

EmGFP-

shRNAmiR-

LacZ.

Table 3.7 ES cell line CAG-loxP-CAT-loxP-EmGFP shRNAmiR ERG following Cre-fusion CPP transduction. Phase contrast and corresponding GFP images for ES cell clones. GFP expression in the CAG-loxP-CAT-loxP-EmGFP-shRNAmiR-ERG treated ES cells is indicative of successful Cre mediated recombination. CAG-loxP-EmGFP-shRNAmiR-LacZ ES cells are a positive control for GFP expression.

3.5.5 PCR screen for Cre mediated recombination in ES cells following transduction with Cre-fused CPP

Cre-mediated recombination at the genomic DNA level in both GFP positive and GFP

negative CAG-loxP-CAT-loxP-EmGFP shRNAmiR ERG transduced ES cell clones

was investigated by PCR screening. A screen with primers distinguishing between

recombined and unrecombined cells was designed. This screen is shown in Figure 3.26

and the primers used have ID 18 and 19, see Appendix 1.

Clones transduced with Cre-fused CPP and selected based on GFP expression generated

a recombined band of 341 bp compared to clones selected as GFP negative which gave


Johanna Edlund

55

an unrecombined band of 2063 bp. Some colonies displayed both bands indicating a

mixture of recombined and unrecombined cells in that clone. Some samples failed to

give a PCR product, possibly due to lysate preparation. Positive control samples CAG-

loxP-EmGFP-shRNAmiR-LacZ and untransduced CAG-loxP-CAT-loxP-EmGFP-

shRNAmiR-ERG, gave band size of 341 bp and 2063 bp respectively.

These results illustrate the potential for application of CPPs in vitro to perform

recombination. An interesting question is if these peptides have possible applications

also in vivo. However further studies have to be done to confirm and strengthen the

results generated in this first study with limited resources.

Figure 3.26 PCR screen for Cre mediated recombination in ES cells following transduction with Cre-fused CPP. Recombined clones give a band of 341 bp and unrecombined colonies generate a band of 2063 bp. Some colonies display both band sizes indicating a mixture of recombined and unrecombined cells within that clone.


Johanna Edlund

56

4 DISCUSSION

4.1 Development of a Cre recombinase inducible shRNAmiR system

Since gene targeting technologies is an important tool in the process of drug discovery,

improvements that speed up and refine the process while decreasing the costs and

efforts are of high value for the pharmaceuptical industry.

One of these gene targeting technologies, vector based RNAi, has become a popular

tool for knocking down the expression of genes. There are today several suppliers of

RNAi systems and they all use the same technology, but they have all optimized their

design in different ways. Invitrogen and Open Biosystems have RNAi vector designs

optimized upon modified miR-155 and miR-30 miRNA precursor respectively. These

vectors have been proven to provide more effective RNAi studies compared to

conventional RNAi [18, 29].

The success of in vitro studies of vector based RNAi systems is dependent on several

variables including the target cell line where the study takes place, transfections

efficiencies, abundance of the mRNA or protein under study, and robust experimental

protocols. To ensure that conclusions drawn from RNAi studies are of any value,

running measurable controls are of high importance to ensure accuracy in the

experimental setup. It is also important to have in mind that gene studies need the

experimenter to recognize every factor that may interfere or supplement the specific

part under study. As an example, when studying RNAi, knock-out of a gene and

depletion of a protein may answer questions regarding the genes’ specific function, but

there is a possibility that the protein participates in a multi-molecular complex and

therefore give a phenotype not only responsible of the desired knock-out misleading the

experimenter’s interpretations.

By using a RNAi system, BLOCK-iT from Invitrogen, we developed the pcDNA 6.2-

GWEmGFPshRNAmiR.SacI. vector, which carries a short hairpin embedded within a

miRNA transcript from the naturally occurring miR-155. To establish a miR-155 based

CRE regulatable RNAi system we inserted a lox-CAT-lox stop cassette at the SacI site.

The PGIPZ vectors from Open Biosystems having a miR-30 embedded shRNAmiR

were also evaluated since different miR scaffolds may vary in their efficiency. The

pGIPZ vector may be cloned into the conditional destination vector produced from the

Invitrogen system giving a conditional system even for this shRNAmiR.

The shRNAmiR vectors from Invitrogen and Open Biosystems were evaluated in vitro

and if any of them would have proven efficient enough in knocking down our genes of

interest and also showed a robust conditional nature it could have been used in an in

vivo study.

First we evaluated the knock-down levels of our shRNAmiR expression constructs by

assays on protein and on mRNA levels. The mRNA levels give a direct answer of the

success of the shRNAmiRs whereas the protein level tells most about the loss of

function of the gene. The knock-down level required to be able to trace the functional


Johanna Edlund

57

loss may vary between genes and cell systems, therefore an estimation of both the

knock-down level on protein and RNA level is of high value. There might also be a

chance that a to high knock-down may end up with a lethal phenotype not answering

functional questions regarding the gene. A fine tuning of each system depending on the

question and the desired results have to be done.

Invitrogen assure 70 % knock-down from at least one of four of their shRNAmiRs

provided for the gene of interest provided that the transfection efficiency in the cells

used are at least 80 %. The Open Biosystems system was thought to give knock-down

levels of 80 % from manufacturers indications. Both manufacturers further claim that

their shRNAmiRs have a specific homology to the target sequence requested resulting

in mRNA degradation at a high rate.

4.1.1 Concluding results

Our study did not indicate the high knock-down efficiencies assured by Invitrogen and

Open Biosystems. Rather, our study indicated knock down levels of 54 % and 80 % for

the Real time RT PCR and quantitative Western Blot respectively for the Invitrogen

system and knock down levels of 58 % for the Real time RT PCR of Open Biosystems

system. Comparing Real time RT PCR results for Invitrogen and Open Biosystems

systems revealed that the shRNAmiR having the best knock-down for each system

displayed a similar knock-down level of approximately 55 %.

In the Invitrogen system, both Real time RT PCR and quantitative Western blot analysis

indicated the same three shRHAmiR constructs to result in the highest knock-down

levels. However, when comparing knock-down levels within these three constructs the

individual order was not the same for Real time RT PCR and Western blot assays. For

Real time RT PCR shRNAmiR2 was found most efficient and also significant, whereas

for Western blot shRNAmiR1 showed the highest knock-down level while not

displaying significance.

We further had a high variance between all different runs, and also poor reproducibility

between runs indicated by the variance showed by error bars calculated as the standard

error of the mean (SEM). The high variance gave cause to logarithmize the normalized

measured values to produce a normal distribution. Even so, only one shRNAmiR was

found to be statistically significant having p-value < 0.05.

4.1.2 Improvements

Obtaining reproducible and reliable data from a multi-step technique is a challenge, and

in this study we were not able to ensure reproducible and reliable results between the

different transfection occasions. Since the knock-down effects were unstable and the

degree of variation was high no selection of a specific shRNAmiR construct for cloning

into the conditional destination vector was done. However, by evaluating the parameters

discussed below a higher and more stable knock-down efficiency as well as less

variance between runs could be achieved with our constructs.

4.1.2.1 Real time RT PCR

The transfection efficiency is of high importance for a successful RNAi study [30]. We

therefore conducted experiments in an attempt to find optimal transfection conditions

for the construct and cell line used. We found transfection ratios and conditions giving

good transfections efficiency but we might have had to increase the transfection


Johanna Edlund

58

efficiency even more to gain reliable results. As the subsequent transfections of LLC-1

cells never displayed the same level of transfected cells as the first round, it is also

important to consider the risk of this uncertainty throughout the entire project. Further,

it might have been a good idea to reduce the number of cells not being transfected

within the population by using a selection marker. Cells not being transfected will give

background expression of our gene of interest and this may therefore introduce variance

in our results. By optimizing the conditions for the transfection even more this problem

may also be reduced. One way to increase the transfection efficiency that we did not try

in this study is to perform multiple transfections.

Since only an estimation by ocular observation and no exact determination of cell

density was done before transfection, there might be a risk that this may have led to

variable results in transfection efficiencies since the correct window for cell density to

vessel size may vary between cell lines [30]. This fact should perhaps also be taken into

account when interpreting the transfection efficiency test in section 3.1.3. Since the

ratio of vessel size to density of cells is of high value for the transfection efficiency,

using vessels of different sizes for correlation of transfection efficiency could lead to

misinterpretation of the efficiency. For future studies ensuring minimum variance in

vessel size should be of importance. It is also of high importance that the same promoter

is used for the cotransfected plasmids since different promoters may elicit different

strengths in expressions levels. Therefore, when trying to determine transfection

efficiencies using the same promoter giving same level of strength would reduce this

problem. In this study this problem would not have affected the result since the plasmid

chosen has the same driving promoter. Is the plasmid used really unbiased as a

measurement for transfection efficiency [13]? There is always a risk when using a

promotor driven plasmid as transfection control since the measured vector or plasmid

could be affected by the other, or by the media components used. A solution could be to

use β-gal protein instead of β-gal plasmid. β-gal protein has been shown to be

internalized by cells when incorporated into calcium phosphate precipitate [13].

It might also have increased consistency downstream if a normalization of RNA

concentration had been done to the transfection efficiency in the same way as for the

Western blot quantification. However, it is worth taking into account that it has also

been shown that cotransfection of different plasmids may alter the expression of any of

them [13].

The non-silencing controls, negative miR vector and empty pGIPZ vector, function as a

reference for target gene expression. An evaluation of these vectors’ effect on the

expression of the gene under study has not been included here, although the suppliers

both claim that no such effects exist. There might, however, be a risk that these negative

controls have a negative effect. As an answer to this question including wild type cells

not transfected at all might reveal if that is the case or not. Further no evaluation on

whether the GAPDH gene expression is effected by the shRNAmiR was done.

4.1.2.2 Western blot

One issue when interpreting the results from Western blotting is the initial levels of the

protein under study. There might be a risk that the initial levels of protein may vary

between cell samples and that measurements are actually done on initially lower levels

of proteins and not specifically showing knock down effects only. Proteins have half-

life times that vary when comparing proteins relative to each other and depending on

this the results may be misinterpreted if assays are performed to early after transfection


Johanna Edlund

59

with the shRNAmiR. This could be the case in this study since we harvested transfected

cells 72 hours after transfection. It might have been wise to determine the half-life for

the protein under study before protein harvest. Information on the half-life of the

proteins under study has not been taken into account in this study.

A pre-evaluated shRNAmiR LacZ was provided with the BLOCK-iT kit from

Invitrogen. This ShRNAmiR could have been used as an positive control if

cotransfections with a LAC Z gene expressing plasmid producing β-galactosidase had

been conducted together with the shRNAmiR LacZ. Unfortunately this notion of the

vector was discovered too late, and if this study would be repeated this positive control

could be added to the study design since this pre-evaluated vector would give an answer

whether the setup had been successful or not.

4.1.3 Conclusions and future aspects

For the shRNAmiR constructs used the manufacturers both claim that their systems will

lead to mRNA degradation since 100 % homology against target gene is guaranteed.

However, if the constructs do not have the promised 100 % homology the miRNA

could switch to a translational inhibition mood instead, see Figure 1.6, which then

perhaps could explain the fact that we had less knockdown at RNA level compared to

protein level.

Finally, the system reported here may potentially be adapted to enable inducible RNAi

in vivo by introduction into ES cells stably expressing the conditional shRNAmiR. This

would generate a knock-down ES cell which when injected into mouse blastocysts

potentially will have a chimera and further germline transmission as results. This would

allow for studies on genes of interest even if they are lethal when knocked out.

To be able to ask more sophisticated questions regarding the functionality and “co-

work” of the human gene systems, the future challenge is to develop and design systems

that allow regulation and control of expressions for two or more genes within the same

mouse line.

4.2 Evaluating CPPs potential as tools for inducing Cre recombination in ES cells

Using mouse models at an early stage in drug discovery process will enable more

efficient selection of promising leads. It is therefore important to be able to develop

mouse models in as short amount of time as possible since this could avoid long

running high-cost projects. Cell penetrating peptides can make the time consuming

process of breeding transgene mouse lines to Cre-expressing mouse lines unnecessary

thereby reducing time lines for constructing transgenic mice lowering overall

development times and costs for new drugs. Further, methods used today to introduce

substance into cells, including electroporation, transfection, and microinjection, are

harsh and impractical to use in vivo as cell membranes need to be disrupted before

introduction of the substance resulting in limited amount of possible cells .

4.2.1 Concluding results

In summary, the system reported here represents a possible improvement over the use of

traditional Cre-induced recombination. Cells transduced with CPPs showed positive


Johanna Edlund

60

penetration and recombination events as cells transduced exhibited GFP signals and

positive luceferase expression. Also, cells sorted for positive GFP signal generated after

PCR-amplification a decreased product size compared to cells sorted GFP negative,

which indicates successful recombination events.

4.2.2 Improvements

We could also detect colonies displaying both product sizes indicating on a mixture of

recombined and unrecombined cells in the colony. To eliminate this mixture an

additional purification step could be added where the individual colonies are trypsinated

and resuspended before seeded onto new plates.

For the luceferase assay on CHO transducted cell described in section 3.5.1 no

correction for protein concentration by Bradford protein assay was performed. This

could mean that the information used to choose CPPs might have been biased as the

actual initial protein concentration affects the measured luceferase results.

Another important consideration is whether selection of CPPs from CHO-based results

really can be considered wise since CPPs have been found to have different penetration

mechanisms and recombination efficiency in different cell types [24, 25, 26]. This

imposes an imminent risk that the CPPs discarded in the first pilot study in CHO cells

might in fact perform better in ES-cells.

4.2.3 Conclusions and future aspects

Before continued studies there are also some other aspects to consider. When

constructing CPPs their intra-cellular stability was enhanced, but this improvement of

CPPs is a two-edged sword with potential negative side effects. The increased stability

could introduce an accumulation of CPPs within the cells having an unknown effect on

the cells.

Also, the side effects of CPPs are not yet studied in detail even though there are some

results indicating that effects are minor [24]. However, other studies have shown toxic

side effects on cell membranes and organelles, as well as toxic effects resulting from

specific interactions of CPPs with cell components [26].

A final interesting note is that some CPPs are found to have a structure similar to

antibacterial peptides which kill microbes through cell lyses which could indicate CPPs

to have harmful properties. The same study compared different CPPs where the TAT

protein, which is used in our study, was shown to be least harmful. The parent protein

of TAT, the HIV transcription factor, is involved in a number of processes including

angiogenesis, and TAT has been shown to inhibit angeogenesis and induce endothelial

cell apoptosis [25]. These factors should be taken into account when selecting specific

CPPs for studies, where it is important to use a CPP which does not affect the system to

be studied.


Johanna Edlund

61

5 ACKNOWLEDGMENTS

First I would like to express my sincere appreciation and gratitude to the ATCG centre

for giving me the opportunity and support to deepen my knowledge within the field of

molecular biology.

I especially wish to thank the following people for helping me and supporting me

during the work with my Master of Science thesis:

John W. Wiseman

For accepting me as a student and guiding me through this project.

Jeanette Nilsson

For beeing my supervisor at the university.

Sandra Rodrigo Blomqvist

For all your RNAi help and your never-failing patience.

Katja Madeyski Bengtsson

For sharing your knowledge about enzymes and cloning.

Anette Persson Kry

For all your smiles and great help in the tissue culture lab.

Gunilla Kanter Smooler

For guidance through the Luceferase assays.

Therese Admyre

For instructions on Real time RT PCR analysis and for the time spent biking.

Mikael Bjursell

For help with the mathematical statistics

Alexandra Viden

For all chats across the lab bench and for beeing such a good friend.

Pernilla Grundevik

For your encouragements and our talks.

Camilla Bernhardsson

For beeing a true star at all our singstar events .

The entire staff at ATCG for providing a friendly atmosphere, many laughs, and

memories.

My parents, my brother, and Stefan for all their love and support.


Johanna Edlund

62

6 REFERENCES

1. J.E. Ferguson, Rusty W. Kelley and Cam Patterson. Mechanisms of Endothelial

Differentation in Embryonic Vasculogenesis. Arterioscler. Thromb. Vasc. Biol.

2005;25;2246-2254.

2. M. J. Gray, J. S. Wey. Neurophilin-1 Supresses Tumorigenic Properties in a

Human Pancreatic Adenocarcinoma Cell LineLacking Neurophilin-1 Coreceptors.

Cancer Research 2005;65(9):p 3664-3670.

3. M. Gossen and H. Bujard. Tight control of gene expression in mammalian cells by

tetracycline responsive element. Proc.Natl. Acad. Sci USA, 1992(89):p. 5547-5551.

4. Y. Sun, X. Chen and D. Xiao. Tetracycline-inducible Expression Systems: New

Strategies and Practices in the transgenic Mouse Modeling. Acta Biochimica et

Biophysica Sinica 2007, 39(4):235-246.

5. B. Sauer. Inducible Gene Targeting in Mice Using the Cre/lox System. Methods: A

companion to methods in Enzymology 14, 381-392(1998).

6. Li. Yiwen, et al. Active Immunization Against the Vascular Endothelial Growth

Factor Receptor flk-1 Inhibits Tumor Angionesis and Metastasis. J. Exp. Med. 12,

1575-1584 (2002).

7. P.J Paddison et al. Short hairpin RNAs (shRNA) induce sequence-specific silencing

in mammalian cells. Genes and Dev. 2002 16: 948-958.

8. A. Fire, C.C. Mello. Potent and specific genetic interference by double-stranded

RNA in Caenorhabditis elegans. Nature. 1998 Feb 19;391(6669):744-5.

9. V. Ullmannova, C. Haskovec. The use of Housekeeping genes (HKG) as an

internal control for the Detection of Gene expression by Quantitative Real-time RT-

PCR. Folia Biologica (Praha) 49, 211-216 (2003).

10. M. A. Carmell et al. Germline transmission of RNAi in mice. Nature structural

biology. Volume 10 number 2. Feb 2003.

11. Paul A. Shaw, Ian R. Catchpole, Comparison of Protein Transduction Domains in

Mediating Cell Delivery of a Secreted CRE Protein. Biochemistry 47, 1157-1166,

(2008).

12. Michael R. Snaith, Jan Törnell. The use of transgenic systems in pharmaceutical

research. Functional Genomics and Proteomics. Vol 1. No 2. 119-130. July 2002.

13. T.K Howcraft, S.L Kirshner and D.S Singer. Measure of transient transfection

efficiency using β-Galactosidase Protein. Analytical Biochemistry, Vol 244, Nr 1,

January 1997, pp.22-27 (6).

14. S.K Bronson, O. Smithies. Altering mice by homologous recombination using

embryonic stem cells. Journal of Biological Chemistry. Vol. 269, No. 44, Issue of

November 4, pp. 27155-27158, 1994

15. A. Bergström. RNA interferance and Off-target effects. Master Thesis in Molecular

Biology, pp. 16, 2007.


Johanna Edlund

63

16. G.J McIntyre. Design and cloning strategies for constructing shRNA expression

vectors. BMC. Biotechnol. January 5. 2006.

17. W. Cao et al. DNA constructs designed to produce short hairpin, interfering RNAs

in transgenic mice sometimes show early lethality and an interference responce. J

Appel Genet. 46(2), pp. 217-225, 2005.

18. Invitrogen, product and user manual. BLOCK-iT Pol II miR RNAi Expression

vector kits. Version E, June 22, 2006.

19. FuGENE 6, product and user manual.

20. L. He, G J. Hannon. Micro RNAs: Small RNAs with a big role in gene regulation.

Nature. July, Vol 5, 2004.

21. F. Shalaby et al. Failure of blood-island formation and vascologenesis in Flk-1

deficent mice. Nature 376, pp 62 – 66,06 July 2002.

22. N. L. Craig et al. Mobile DNA II, Chapter 4 Chromosome Manipulation by Cre-lox

Recombination.

23. J. Wang et al. Site directed, virus-free, and inducible RNAi in embryonic stem cells.

PNAS. Vol. 104, No. 52, Issue of December 26, pp. 20850-220855, 2007

24. M. Zorko, U. Langel. Cell penetrating peptides: mechanism and kinetics of cargo

delivery. Advanced drug delivery Reviews 57, 529-545, 2005.

25. H. Brooks et al. Tat peptide-mediated cellular delivery: back to basics. Advanced

drug delivery Reviews 57, 559-577, 2005.

26. P. Järver, U. Langel. The use of cell penetrating peptides as a tool for gene

regulation. DTT Vol. 9, No. 9 May 2004.

27. http://www.openbiosystems.com/RNAi/shRNAmirLibraries/pSM2Retroviral/shRNA

mirValidation/

28. Kwan-Ho Chung et al. Polycistronic RNA polemyrase II expression vectors for

RNA interference based on BIC/miR-155. Nucleic Acids Research.

34(7):e53;doi:10.1093/nar/gkl 143. 2006.

29. J. M. Silva. Second-generation shRNA libraries covering the mouse and human

genomes. Nature genetics. Vol 37, No 11, Nov 5. 2005.

30. Open Biosystems, product and user manual. Expression Arrest GIPZ lentiviral

shRNAmiR. 2005-2007.


Johanna Edlund

64

APPENDIX 1: General Oligo List

Oligos for this work were routinely designed using the AstraZeneca bioinformatics

portal, e-lab. Oligos marked with an * were obtained from existing ATCG stocks. The

ID number relates to references made throughout this report. Tm denotes the melting

temperature of each oligo.

ID Name Sequence (5’-3’) Tm

1 NRP-1 Forward 1 GGAGCTACTGGGCTGTGAAG 63,9

2 NRP-1 Forward 2 GACTTCCAGCTCACAGGAGG 64,0

3 NRP-1 Reverse 1 TGACCCTCAGTGTACCCACA 64,2

4 NRP-1 Reverse 2 CAGCAATTCCACCAAGGTTT 63,7

5 KDR Forward 1 GACGGAGAAGGAGTCTGTGC 64,0

6 KDR Reverse 1 AAGATACTGTCACCACCGCC 63,9

7 KDR Reverse 2 GTCACTGACAGAGGCGATGA 64,2

8 SnaBI lox-cat-lox Forward ATATATTACGTAATAACTTCGTATAGCATACAT 59,7

9 SnaBI lox-cat-loxReverse ATATATTACGTAATAACTTCGTATAATGTATGC 60,0

10 NotI lox-cat-lox Forward ATATATGCGGCCGCATAACTTCGTATAGCATACAT 73,7

11 NotI lox-cat-lox Reverse ATATATGCGGCCGCATAACTTCGTATAATGTATGC 73,8

12 KDR SYBR F1 CTGCTAGCTGTCGCTCTGTG 64,1

13 KDR SYBR R1 TTTCTGTGTGCTGAGCTTGG 64,2

14 NRP-1 SYBR F1 ATGAGTGTGACGACGACCAG 64,0

15 NRP-1 SYBR R2 AATGGTTGGCTTCTCTGTGG 64,0

16* 105 KMB CGACGATTTCCGGCAGTTTCTACA 71,5

17* 106 KMB TATTGGTGCCCTTAAACGCCTGGT 71,3

18* 82U GAG Upper CTCCTGGGCAACGTGCTGGT 68,0

19* 82L2 GFP Lower GCACTGCACGCCGTAGGTGA 68,0

20* GAPDH Forward AACGACCCCTTCATTGAC 59,9

21* GAPDH Reverse TCCACGACATACTCAGCA 58,8

22* M13 Forward GTAAAACGACGGCCAG -


Johanna Edlund

65

23* M13 Reverse CAGGAAACAGCTATGAC -

24 SnaBI Sequencing Forward 1

TGACGTCAATGGGTGGAGTA 64,0

25 SnaBI sequencing Forward 2

ATCATATGCCAAGTACGCCC 63,5

26 SnaBI sequencing Reverse 1

GGCGGAGTTGTTACGACATT 63,7

27 SnaBI sequencing Reverse 2

GTGCCAAAACAAACTCCCAT 63,6

28 NotI Sequencing Forward 1 AGGAGGATCACAGCAACACC 64,1

29 NotI Sequencing Forward 2 ACTACAGCTCCGTGGTGGAC 64,2

30 NotI Sequencing Reverse 1 AGACCCCTAGGAATGCTCGT 63,8

31 NotI Sequencing Reverse 2 AAAAGACGGCAATATGGTGG 63,5

32 SnaBI loxCATlox Forward 2 ATATATTACGTAATAACTTCGTATAGCATACATTATACGAAGTTATATTAAG 67,0

33 SnaBI loxCATlox Reverse 2 ATATATTACGTAATAACTTCGTATAATGTATGCTATACGAAGTTATTAGG

68,0

34 EmGFP forward sequencing primer

GGCATGGACGAGCTGTACAA -

35 miRNA reverse sequencing primer

CTCTAGATCAACCACTTTGT -

36 SacI loxCATlox Forward 3 ATATATGAGCTCATAACTTCGTATAGCATACATTATACGAAGTTATATTAAG 69,6

37 SacI loxCATlox Reverse 3 ATATATGAGCTCATAACTTGCTATAATGTATGCTATACGAAGTTATTAGG 70,8


Johanna Edlund

66

APPENDIX 2: Sequences designed for shRNAmiR knockdown of NRP-I and KDR using Invitrogen’s RNAi designer algorithms

List of single stranded oligos which when annealed generated double stranded

shRNAmiR fragments for cloning into the BLOCK-iT Pol II miR RNAi expression

vector kit (Invitrogen). Top and bottom strands are indicated by the letters T and B in

the name column. Tm denotes the melting temperature of each oligo.

Name Sequence (5’-3’) Tm

NRP-1 T1 TGCTGTTTATGGTCCCGCCACATTTGGTTTTGGCCACTGACTGACCAAATGTGGGGACCATAAA 92,3

NRP-1 B1 CCTGTTTATGGTCCCCACATTTGGTCAGTCAGTGGCCAAAACCAAATGTGGCGGGACCATAAAC 91,5

NRP-1 T2 TGCTGTAAGAATGAGGGTAACCGGGAGTTTTGGCCACTGACTGACTCCCGGTTCCTCATTCTTA 89,7

NRP-1 B2 CCTGTAAGAATGAGGAACCGGGAGTCAGTCAGTGGCCAAAACTCCCGGTTACCCTCATTCTTAC 88,9

NRP-1 T3 TGCTGTATAGTTCTGAGAACATTCGGGTTTTGGCCACTGACTGACCCGAATGTTCAGAACTATA 86,6

NRP-1 B3 CCTGTATAGTTCTGAACATTCGGGTCAGTCAGTGGCCAAAACCCGAATGTTCTCAGAACTATAC 85,8

NRP-1 T4 TGCTGTGTAGGTGCACTCCAAGCAGTGTTTTGGCCACTGACTGACACTGCTTGGTGCACCTACA 93,1

NRP-1 B4 CCTGTGTAGGTGCACCAAGCAGTGTCAGTCAGTGGCCAAAACACTGCTTGGAGTGCACCTACAC 92,2

NRP-1 T5 TGCTGCTCTGTAGCACACTGTAGTTGGTTTTGGCCACTGACTGACCAACTACAGTGCTACAGAG 88,1

NRP-1 B5 CCTGCTCTGTAGCACTGTAGTTGGTCAGTCAGTGGCCAAAACCAACTACAGTGTGCTACAGAGC 88,1

KDR T1 TGCTGATCTCTTTCAGTCACTTCCATGTTTTGGCCACTGACTGACATGGAAGTCTGAAAGAGAT 88,1

KDR B1 CCTGATCTCTTTCAGACTTCCATGTCAGTCAGTGGCCAAAACATGGAAGTGACTGAAAGAGATC 87,7

KDR T2 TGCTGTGTAGGAGCAGGCTTCTTCTAGTTTTGGCCACTGACTGACTAGAAGAACTGCTCCTACA 86,7

KDR B2 CCTGTGTAGGAGCAGTTCTTCTAGTCAGTCAGTGGCCAAAACTAGAAGAAGCCTGCTCCTACAC 85,8

KDR T3 TGCTGTTTACAAGCATACGGGCTTGTGTTTTGGCCACTGACTGACACAAGCCCATGCTTGTAAA 90,5

KDR B3 CCTGTTTACAAGCATGGGCTTGTGTCAGTCAGTGGCCAAAACACAAGCCCGTATGCTTGTAAAC 89,7

KDR T4 TGCTGACAACAGGGACACACTCTCCTGTTTTGGCCACTGACTGACAGGAGAGTGTCCCTGTTGT 92,3

KDR B4 CCTGACAACAGGGACACTCTCCTGTCAGTCAGTGGCCAAAACAGGAGAGTGTGTCCCTGTTGTC 91,9

KDR T5 TGCTGACAAGAAGGAGCCAGAAGAACGTTTTGGCCACTGACTGACGTTCTTCTCTCCTTCTTGT 89,6

KDR B5 CCTGACAAGAAGGAGAGAAGAACGTCAGTCAGTGGCCAAAACGTTCTTCTGGCTCCTTCTTGTC 89,2


Johanna Edlund

67

APPENDIX 3: Cell Penetrating Peptides

List of proteins produced using the E.coli strain BL21 (DE3). Each name indicates the

peptide component and if present the histidine purification tag. Each CPP is abbreviated

in the appendix. The column to the right shows the assesory information for each CPP.

Abbreviation Name

TC Tat-Cre Method A

HHNC Histidine tag-Histone H1- Nuclear localised Cre DG25

HHNC Histidine tag-Histone H1- Nuclear localised Cre SourceS30

HHNC Histidine tag-Histone H1- Nuclear localised Cre DEAE

HHNC Histidine tag-Histone H1- Nuclear localised Cre PD10

HHNC Histidine tag-Histone H1- Nuclear localised Cre IöSI

HNCM Histidine tag-Nuclear localiced Cre-Membrane translocating hydrophobic sequence

DG25

HNCM Histidine tag-Nuclear localiced Cre-Membrane translocating hydrophobic sequence

Refold

HTNCM Histidine tag-Tat-Nuclear localised Cre- Membrane translocating hydrophobic sequence

Refold

TNC Tat- Nuclear localised Cre Sdcf

TNC Tat- Nuclear localised Cre G25cf

TNC Tat- Nuclear localised Cre Unknown

HI-4F Histone polypeptide Unknown

Johanna_Edlund-Thesis-final

Documents

expression of kdr

shrnamir expression

gene silencing

endogenous expression

inducing cre recombination

inducible knock

field of rnai

johanna edlund supervisor