Design and evaluation of drug delivery vehicles Caroline Palm Apergi

Design and evaluation of drug

delivery vehicles

Caroline Palm Apergi

©Caroline Palm Apergi, Stockholm 2008

ISBN 978-91-7155-754-4

Printed in Sweden by Universitetsservice US-AB, Stockholm 2008

Distributor: Department of Neurochemistry, Stockholm University

To Jannis, η αγάπη μοσ

List of publications

This thesis is based on the following publications, referred to in the text by

the following Roman numerals:

I Palm C, Netzereab S, Hällbrink M. Quantitatively determined

uptake of cell-penetrating peptides in non-mammalian cells with

an evaluation of degradation and antimicrobial effects. Peptides.

2006 Jul;27(7):1710-6.

II Palm C, Jayamanne M, Kjellander M, Hällbrink M. Peptide deg-

radation is a critical determinant for cell-penetrating peptide up-

take. Biochim Biophys Acta. 2007 Jul;1768(7):1769-76.

III Palm-Apergi C, Lorents A, Padari K, Pooga M, Hällbrink M.

The membrane repair response masks membrane disturbances

caused by cell-penetrating peptide uptake. FASEB J. In press

IV Palm-Apergi C, Hällbrink M. A new rapid cell-penetrating pep-

tide based strategy to produce bacterial ghosts for plasmid deliv-

ery. J Control Release. In press

Abstract

A crucial aspect of drug delivery is efficient transport to the site of action.

Thus, there is a need to design and evaluate new delivery vehicles. In this

thesis two delivery vehicles, cell-penetrating peptides and bacterial ghosts,

were evaluated. The understanding of the internalization and degradation

kinetics of cell-penetrating peptides is important for the practical aspects of

cargo delivery since peptides have a notorious reputation of being rapidly

degraded. If the cell-penetrating peptide remains intact inside the cellular

environment, there is a possibility that the peptide-cargo conjugate leaks

back to the extracellular environment. However, if it is degraded outside the

cell, the cargo will never be delivered. In order to improve uptake efficiency

and to be able to foresee side effects, the translocation mechanism needs to

be fully elucidated.

Data gathered from the first two papers led to the proposal of a new me-

chanism involved in cell-penetrating peptide uptake: the membrane repair

response, a resealing mechanism rapidly patching up broken membranes.

This mechanism could explain the divergence in perception concerning the

uptake pathways. Furthermore a new assay to produce the second delivery

vehicle, bacterial ghosts, was developed based on data from the cell-

penetrating peptide investigations. Bacterial ghosts are dead bacteria devoid

of cytoplasmic contents but still retaining their structural and morphological

characteristics, after protein E lysis of the bacterial cell membrane. By using

a cell-penetrating peptide with antimicrobial effects, a new rapid peptide-

based strategy to produce ghosts was developed and the capability to deliver

plasmid DNA into the cell for expression was evaluated.

Contents

Introduction .................................................................................................. 1 The cell membrane.................................................................................................... 2

Prokaryotes ........................................................................................................... 2 Eukaryotes ............................................................................................................ 4

Membrane transport ................................................................................................. 5 Prokaryotic transporter systems ....................................................................... 5 Eukaryotic transporter systems......................................................................... 6

Membrane repair response .................................................................................... 11 Target cell apoptosis ............................................................................................... 13

Perforin ................................................................................................................ 14 Granzymes .......................................................................................................... 15 Granulysin ........................................................................................................... 15

Cell-penetrating peptides as drug delivery vehicles ......................................... 16 Structural features ............................................................................................. 17 Mechanisms of uptake ....................................................................................... 18 Cargo delivery .................................................................................................... 20 Stability of cell-penetrating peptides.............................................................. 21 Cell-penetrating peptides versus antimicrobial peptides ............................ 23

Bacterial ghosts as drug delivery vectors ........................................................... 24

Aims of the thesis ...................................................................................... 27

Methodological considerations ................................................................ 28 Design and evaluation of peptides ....................................................................... 28 Synthesis (Paper I-IV) ............................................................................................ 29 Cell cultures .............................................................................................................. 30

Non-mammalian cells (Paper I and IV) ......................................................... 30 Mammalian cells (Paper II-III) ........................................................................ 30

Minimum inhibitory concentration (Paper I and IV) .......................................... 30 Optimization of ghost production ......................................................................... 31 Agarose gel analysis of DNA (Paper IV) .............................................................. 31 Uptake of cell-penetrating peptides ..................................................................... 31

High performance liquid chromatography (Paper I-III) .............................. 31 Fluorescence activated cell sorter (Paper III-IV) ......................................... 32 Fluorescence microscopy (Paper III-IV) ........................................................ 32 Electron microscopy (Paper III) ...................................................................... 33

Cytotoxicity assays ................................................................................................. 33 Lactate dehydrogenase leakage (Paper II-III) ............................................. 33 Calcium measurements (Paper III) ................................................................ 34 B-hexosaminidase leakage (Paper III)........................................................... 34

Results and discussion ............................................................................. 35 Evaluation of CPP uptake and antimicrobial effects in non-mammalian cells

(Paper I) .................................................................................................................... 35 Determining the importance of peptide degradation in CPP uptake (Paper II)

.................................................................................................................................... 37 The MRR masks membrane damages caused by CPPs (Paper III) ................ 40 Production of bacterial ghosts for plasmid delivery by CPPs (Paper IV) ....... 42

Conclusions ................................................................................................. 43 Populärvetenskaplig sammanfattning på svenska ............................................ 44

Acknowledgements ................................................................................... 46

References .................................................................................................. 48

Abbreviations

AAK Adaptor-associated kinase

ABC ATP-binding cassette

ADP Adenosine diphosphate

AM Acetoxymethyl

AMP Antimicrobial peptides

AP2 Adaptor protein 2

ARF ADP-ribosylation factor

Axh Aminohexanoic acid

Calu-3 Caucasian lung adenocarcinoma

CaMKII Calcium/calmodulin-dependent kinase II

CCV Clathrin coated vesicle

CHO Chinese hamster ovary

CME Clathrin-mediated endocytosis

CPPs Cell-penetrating peptides

CTB Cholera Toxin B

CTL Cytotoxic T lymphocyte

Da Dalton

DCC Dicyclohexylcarbodiimide

DiBAC3(4) bis-(1,3-dibarbituric acid)-trimethine oxanol

DIC Diisopropylcarbodiimide

DMF Dimethylformamide

DMSO Dimethylsulfoxide

EGF Epidermal growth factor

EGFP Enhanced green fluorescent protein

EM Electron microscopy

ER Endoplasmic reticulum

Fl Fluorescence

FACS Fluorescence-activated cell sorter

GAG Glycosaminoglycan

GTP Guanosine triphosphate

GPI Glycosyl-phosphatidylinositol

HA2 Hemagglutinin peptide 2

HIV-1 Human immunodeficiency virus type 1

HOBt N-Hydroxybenzotriazole

HPLC High Performance Liquid Chromatography

HS Heparan sulfate

HSPG Heparan sulfate proteoglycan

hCT Human calcitonin

IL Interleukin

LAP Leucine aminopeptidase

LDH Lactate dehydrogenase

LPS Lipopolysaccharide

MAP Model amphipathic peptide

MBHA 4-Methylbenzhydrylamine

MDCK Madin-Darby Canine Kidney

MFS Major facilitator superfamily

MMP Matrix metalloproteinase

MRR Membrane repair response

NAD Nicotinamide adenine dinucleotide

NK Natural killer

NLS Nuclear localization signal

Omp Outer membrane proteins

Pen Penetratin

PG Proteoglycan

PI Propidium iodide

PI3K Phosphatidylinositol-3 kinase

PTD Protein transduction domain

pVEC Vascular endothelial cadherin peptide

Sf9 Spodoptera frugiperda

SNAP Synaptosomal-associated protein

SPPS Solid phase peptide synthesis

SV40 Simian virus 40

TAP Transporter associated with antigen processing

TAT Trans-activator of transcription

t-Boc tert-Butyloxycarbonyl

TEM Transmission electron microscopy

TFA Trifluoroacetic acid

TLR-4 Toll-like-receptor-4

TOP Thimet oligopeptidase

TP10 Transportan 10

TPPII Tripeptidyl peptidase II

1

Introduction

One of the major obstacles in drug delivery today, is for the drug to reach

inside the cell to exert its biological effect. The difficulty lies in crossing the

hydrophobic plasma membrane since many drug candidates are hydrophilic.

Thus, there is a great need to design drug delivery vehicles, able to cross the

protective plasma membrane, without inducing toxicity. Twenty years ago, it

was shown that some proteins were able to internalize into cells. From these

proteins, a relatively new group of delivery vehicles, named cell-penetrating

peptides (CPPs), has evolved over the last fifteen years. Even though these

drug delivery vehicles have been shown to deliver different cargo into cells,

such as drugs and oligonucleotides, both in vivo and in vitro, the mechanism

of entry has not been fully elucidated. Another delivery system, with inhe-

rent adjuvant properties, is nonliving bacteria known as bacterial ghosts,

devoid of cytoplasmic contents but still retaining the native morphology.

These features make them an excellent tool in drug delivery and DNA vac-

cination.

The focus of the thesis is to evaluate CPPs as transporters, to better un-

derstand how these peptides are able to traverse the plasma membrane, not

only by themselves but also conjugated to cargo. The degradation kinetics of

CPPs, both inside and outside the cell, was investigated both in mammalian

and non-mammalian cells, in order to better understand the mechanism of

entry, determine the influence of membrane composition on uptake, and to

compare them to antimicrobial peptides. Shorter peptides were designed

from proteins known to internalize into cells, to explore if the new peptides

might have translocation ability. The new peptides, together with CPPs, were

analyzed to investigate if a membrane repair response (MRR) was induced in

cells during translocation, which could explain how CPPs translocate across

the plasma membrane into the cytoplasm. By taking advantage of the results

from the first paper in the thesis, CPPs were used to produce another deli-

very vehicle, bacterial ghosts, to investigate if the peptide-produced ghosts

were able to deliver plasmids into cancer cells.

2

The cell membrane

The cytoplasm of both prokaryotes and eukaryotes, are surrounded by a

plasma membrane, protecting it from the extracellular environment. The

membrane is a requirement for all living organisms for many reasons and its

structure was first described as a fluid mosaic model by Singer in 19721.

Membranes are not only needed to maintain intracellular compartments in a

highly ordered state but also for acquiring nutrients and eliminating wastes.

Independent of cell type, all membranes consist of a phospholipid bilayer,

about 5-10 nm thick2,3

. The protein, sugar, and lipid composition can vary

and determines the function of the cell. The phospholipid itself consists of a

hydrophilic polar head and a hydrophobic tail. These amphipathic properties

result in the formation of a bilayer. Inside the bilayer, the lipids are shielded

from the aqueous surrounding, in a hydrophobic interior approximately 4 nm

thick2. The bilayer can be divided into two leaflets and the major driving

force for the formation of the bilayer, is hydrophobic interactions between

the fatty acyl chains, and hydrogen bonding and electrostatic interactions

between polar head groups and water molecules, stabilizing the bilayer. Due

to the thermal motion, phospholipids and glycolipids can rotate freely around

their own axes and diffuse several micrometers per second laterally, within

the membrane leaflet. Most importantly, the plasma membrane has to re-

spond to extracellular changes and transfer external stimuli to the cell. Thus,

the membrane has an important role in cell function and the membrane com-

position of the analyzed cells will be described below.

Prokaryotes

The plasma membrane of most prokaryotes usually has higher protein con-

tent than eukaryotes, due to the many functions that need to be performed in

bacteria. Sterols like cholesterol are often not present in prokaryotic mem-

branes. Instead hopanoids, a pentacyclic sterol-like molecule, synthesized

from the same steroid precursor, is found in prokaryotic membranes (Ouris-

son87). Outside the cytoplasmic plasma membrane lies the periplasmic

space, surrounded by a complex cell wall of peptidoglycan, together forming

the cell envelope. There are two types of bacterium that differ in the struc-

tural composition of the membrane, gram-negative and gram-positive bacte-

ria. Gram-positive bacteria, have a plasma membrane surrounded by a thin

periplasmic space, followed by a single 20-80 nm thick peptidoglycan layer,

where acidic polysaccharides, such as teichoic acid, give the membrane a net

negative charge (Figure 1)4.

3

Figure 1. Structure of the gram-positive cell envelope. The plasma membrane is composed of phospholipids (brown) with hydrophobic tails (yellow area) and com-prises several proteins (blue). The periplasmic space (turquoise) is situated between the peptidoglycan layer (pink) and the plasma membrane. Anionic saccharides, such as teichoic acids (grey), protrude from the plasma membrane through the peptidog-lycan layer.

Peptidoglycan is a giant polymer composed of glycan strands consisting of

alternating N-acetylglucosamine and N-acetylmuramic acid, cross-linked by

short peptides, referred to as stem peptides. The stem peptides comprise

amino acids of both D and L configurations, shielding the peptidoglycan

from most peptidases5,6

. In comparison to gram-positive bacteria, the pepti-

doglycan layer of gram-negative bacteria, is approximately 2-7 nm and sur-

rounded by an outer membrane (Figure 2). The peptidoglycan layer and the

outer membrane, are firmly linked by a membrane protein named Braun’s

lipoprotein7. In addition, the outer part of the lipid bilayer, in the outer mem-

brane, is composed of lipopolysaccharides (LPSs), giving the membrane a

negative net charge. These LPS molecules are composed of three parts: lipid

A, comprising two glucosamine units and long fatty tails, the core polysac-

charide and an O side chain extending from the polysaccharide that can be

rapidly changed, to avoid detection by the immune system8. Thus, the cell

envelope of gram-negative bateria is much more complex than the cell

envelope of gram-positive bacteria.

4

Figure 2. Structure of the gram-negative cell envelope. . The inner and outer plasma membranes are composed of phospholipids (brown) with hydrophobic tails (yellow area) and comprise several proteins (blue) with different functions. The periplasmic space (turquoise), together with the peptidoglycan layer (pink), is situated between the inner and outer plasma membranes. The outer leaflet of the outer membrane consists of lipid A (green), and not phospholipids. Linked to lipid A, is a core poly-saccharide, with an extending O side chain (grey), together forming the LPS.

Eukaryotes

Prokaryotes and eukaryotes differ in several structural features. Eukaryotes

have membrane bound organelles and membrane enclosed nuclei, whereas

the size of prokaryotes is often the same as chloroplasts or mitochondria.

The much more structurally organized eukaryotes have only one lipid bilayer

and most of them do not have a cell wall. One exception is yeast, where the

plasma membrane is surrounded by a cell wall of chitin. The chitin layer

comprises repetitive units of glucose, N-acetylglucosamine and mannose

residues that function to maintain the structure and rigidity of the cell9,10

.

Solutes of lower mass than 600 Da, can diffuse freely through the cell wall,

whereas the plasma membrane forms a relatively impermeable barrier for

hydrophilic molecules11

. Nevertheless, the plasma membrane of the yeast S.

cerevisiae resembles the membrane of higher eukaryotes, with respect to

lipid composition. Like prokaryotes, specialized proteins mediate the uptake

5

and secretion of solutes, in and out of the cell. However, the lipid composi-

tion of the plasma membrane and organelle membrane is distinct. Therefore

special trafficking of lipids from internal membranes, especially to the en-

doplasmic reticulum (ER) and Golgi, is required12

.

Nearly all cells, from invertebrates, to humans, have the capability to pro-

duce glycoproteins, such as proteoglycans (PGs)13

. The PGs are composed of

a core protein, covalently linked to one or several polysaccharide chains. A

major constituent of the extracellular matrix, are the heparan sulfate PGs

(HSPGs), comprising repetitive disaccharide units of variable sugars14

. Un-

der physiological conditions, the polysaccharide chains, or glycosaminogly-

cans (GAGs), are anionic due to the alternating N-acetylated or N-sulfated

glucosamine and uronic acid units15

. In vivo, the GAG side chains are ap-

proximately 50-150 disaccharide unit repeats long, resulting in great struc-

tural diversity16

. The HSPGs can be divided into three subgroups: the mem-

brane spanning PGs, the glycophosphatidylinositol (GPI)-linked PGs, and

the secreted extracellular matrix PGs15,17

. Other variably sulfated sugars can

be found in the polysaccharide chains of PGs, such as chondroitin sulfate,

dermatan sulfate, and heparin, closely related to HSPGs.

Membrane transport

An enormous number of channels and transporter proteins have been

evolved by nature to control the translocation of molecules across the plasma

membrane and internal membranes. Some systems can be found in both pro-

karyotes and eukaryotes e.g. ATP-binding cassette (ABC) transporters. Oth-

er systems are specific for only eukaryotes, such as endocytosis, which can

not be found in prokaryotes. A brief introduction will follow, covering the

main transporter systems of prokaryotes and eukaryotes.

Prokaryotic transporter systems

The term passive diffusion refers to concentration-dependent movement of

molecules, from one compartment to another, without requiring metabolic

energy. Prokaryotes, do not depend on passive diffusion since the concentra-

tion of most nutrients outside the cell, is lower than inside the cell. Thus,

nutrients need to be transported against the concentration gradient via a car-

rier protein, a process requiring energy. Three main transporters are engaged

by bacteria: 1) primary transporters, such as ABC transporters that use ATP

to drive substrates across membranes, 2) secondary transporters, that use the

free energy from the primary transporters, to translocate substrates, 3) group

translocation systems, which modify the substrates chemically during trans-

location18

.

6

The largest family of protein transport systems belonging to the group of

primary transporters, is the ABC transporters, present in both prokaryotes

eukaryotes19,20

. ABC transporters can be divided into three categories; im-

porters, exporters, and a group not involved in transport but DNA repair. A

common organization is found in importers and exporters, which are com-

posed of two hydrophobic membrane spanning or integral membrane do-

mains, forming a pore in the membrane. On the cytosolic side of the mem-

brane, two hydrophilic domains carrying the ABC, and two nucleotide bind-

ing domains, peripherally associated with the integral membrane domains,

can be found21

. ABC importers are dependent on special binding proteins,

located in the periplasmic space of gram-negative bacteria. In gram-positive

bacteria, they are predominantly lipoproteins, attached to the cytoplasmic

membrane. However, they can be fused to the membrane transporter itself,

in some gram-positive organisms22

. The first barrier, in translocation into the

cytoplasm of gram-negative bacteria, is the outer membrane. Molecules be-

low 650 Da, can pass through the nonspecific outer membrane proteins

(Omp), belonging to the class of porin proteins, e.g. OmpA in Escherichia

coli23

. However, specialized porins take over, when the size of the substrate

exceeds the limit of the generalized porins. After translocation through po-

rins to the periplasm, binding proteins in the periplasm can bind and trans-

port many different substrates, such as oligosaccharides and inorganic ions

but most importantly, amino acids and peptides24

.

The second largest family of transporters belonging to the family of sec-

ondary transporters, is the major facilitator superfamily (MFS) that covers

about 25% of all prokaryotic transport proteins18

. However, in some gram-

positive bacteria, MFS is more prevalent than ABC transporters25

. There are

three distinct mechanisms coupled to the MFS; uniporters, symporters, and

antiporters that transfer a tremendous variety of molecules such as sugars,

drugs and peptides. Thus, there are several transporter systems able to trans-

locate peptides in an energy dependent manner. Some transporters, e.g. the

ABC transporter OppA, in Lactococcus lactis, can handle peptides up to 18

amino acids in length26

. However, most of the MFS and ABC transporters

can only translocate shorter peptides, less than 10 amino acids.

Eukaryotic transporter systems

Small molecules, such as saccharides, amino acids, and short peptides, can

translocate across the membrane via protein transport systems, as mentioned

above. However, larger macromolecules have to be taken up by membrane

invagination or protrusion, forming a vesicle that can be pinched off from

the membrane, in order to enter the cell. This uptake process, called endocy-

tosis, is a characteristic of eukaryotic cells and can be divided into two sub-

groups: phagocytosis and pinocytosis. Phagocytosis occurs mostly in cells

involved in innate immunity, whereas pinocytosis, is found in most cells and

7

can be divided into 4 categories: macropinocytosis, clathrin-mediated endo-

cytosis, caveolae-mediated endocytosis, and clathrin- and caveolae-

independent endocytosis (Figure 3). An introduction to the different forms

of endocytosis will be given below.

Phagocytosis

Phagocytosis involves uptake of many large particles, e.g. pathogens or

apoptotic cells and is in most cases performed by specialized cells, such as

monocytes and macrophages27

. Other cells known to have evolved the recep-

tor-mediated, actin-driven process of phagocytosis, are epithelial28

and endo-

thelial cells29

. However, it should be noted that large particles and macromo-

lecules, can be internalized by cells via macropinocytosis30

. When a particle

has bound a receptor on the plasma membrane, transient actin rearrangement

is required for the membrane to engulf the particle. It is a highly regulated

process, involving signaling cascades mediated by Rho-family GTPases,

where Rac and Cdc42 are activated, inducing the formation of protrusions on

the cell surface that zipper up around the particle and engulf it31

.

Macropinocytosis

Several characteristics are shared between macropinocytosis (Figure 3)

and phagocytosis. They are distinct from other pinocytic pathways, in that

much larger vesicles are formed, which lack coating and can be several mi-

crometers in diameter. Both pathways are also morphologically more hete-

rogenic and induce protrusions, rather than invaginations of the plasma

membrane32

. However, one distinguishing feature is that the protrusions in

macropinocytosis, collapse onto the plasma membrane, forming a vesicle

that can be pinched off from the membrane, in comparison to phagocytosis,

where two protrusions come together like a zipper to form a vesicle33

. There

are little or no receptors involved in macropinocytosis. Instead, macropino-

somes form spontaneously from cell surface ruffles, or in response to

growth-factor-receptor stimulation34

. Ruffles are sheet-like extensions of the

cell surface that most of the time, recede into the cytoplasm, without forming

vesicles and only sometimes, lead to the formation of macropinosomes.

There are, however, cell lines with constitutive macropinocytosis, such as

dendritic cells but in most cell lines, it is transiently induced35

.

8

Figure 3. Different pathways of pinocytosis.

In fact, two bacterial pathogens, Salmonella typhimurium and Legionella

pneumophila, stimulate cell-ruffling and enter cells via macropinocyto-

sis36,37

. The route of macropinosomes, do not depend on dynamin, unlike

clathrin- and caveolae-mediated endocytosis and differs from other pinocytic

pathways in that the vesicles do not reach the early/late endosomes or lyso-

somes38

. Although much has happened since Warren Lewis first described

macropinocytosis, as waving sheets in 1931, there are still many questions to

be answered about the intracellular destination of macropinosomes.

Clathrin-mediated endocytosis

The major route for nutrient uptake and receptor recycling, is clathrin-

mediated endocytosis (CME), which occurs constitutively in all mammalian

cells 39

(Figure 3). Proteins, e.g. the cholesterol transporter low-density lipo-

protein and the iron transporter transferrin, are classical exploiters of

CME40,41

. The fundamental aspects of clathrin-coated pit formation, were

first described in 1964, after visualization in electron microscopy42

. CME is

a multistep process that begins with the assembly of soluble constituents,

such as clathrin, adaptors, and accessory proteins of the coat, to site in the

membrane where the vesicle will form around the cargo, i.e. the bud site43

.

9

Vesicle budding and package of cargo into vesicles, are triggered by inte-

racting subunits of the coat complex, forming the coated vesicles as the

emerging bud detaches from the membrane40

. Coat assembly may be enough

to promote vesicle detachment in vitro, but in vivo the packaging requires

other regulatory factors, like the dynamin family of GTPases40,44

. The major

coat proteins for clathrin-coated vesicles, are two oligomeric complexes,

clathrin and adaptor protein 2 (AP2). Clathrin is composed of three light

chains and three heavy chains, coming together as a heterohexameric three-

legged structure, named triskelion. Under non-physiologic conditions, the

triseklion self-assembles into cages45

. However, the recruitment and assem-

bly of clathrin on the plasma membrane into a polygonal lattice, surrounding

the endocytic vesicles, are coordinated by adaptor proteins such as AP2. The

invagination of the plasma membrane is thought to be driven by the assem-

bly of the clathrin lattice, to facilitate vesicle formation. CME begins with

the recruitment of AP2 complexes, to protease-sensitive sites on the plasma

membrane, followed by binding to tyrosine-based sorting motifs on the cy-

toplasmic domain of the receptor, to coordinate clathrin assembly and con-

centration into coated pits40

. When the coat is assembled and pinched off

from the plasma membrane, coat proteins are disassembled by the help of

ATPases, releasing the coat proteins for the cycle to begin again46,47

. Evi-

dence has been presented, indicating that CME is regulated via phosphoryla-

tion by adaptor-associated kinase 1 (AAK1), a binding partner of AP2 that

phosphorylates AP2. Together with an unidentified phosphatase, facilitating

AP2 recruitment to sorting signals, AAK1 was shown to regulate CME via a

phosphorylation/dephosphorylation cycle. Clathrin also plays a regulatory

role, by modulating the activity of AP2 complexes, through activation of

AAK145

. Moreover, AP2 has been shown to have a more prominent role in

cargo recruitment than previously believed48

. Thus, more information is

needed to fully understand the mechanistic order in which CME occurs and

the exact assignments for each protein involved.

Caveolae-mediated endocytosis

In the 1950s, caveolae were identified as flask shaped invaginations of the

plasma membrane, approximately 50-80 nm in diameter49

(Figure 3). Al-

though present in many cell lines, they are especially abundant in endothelial

cells, where they are involved in endocytosis and transcytosis of blood com-

ponents50

. Caveolae are subdomains of lipid rafts, or plasma membrane mi-

crodomains, rich in cholesterol and sphingolipids51

. Caveolae-mediated en-

docytosis is believed to be involved in intracellular cholesterol trafficking

and homeostasis but since caveolin null mice show no cholesterol imbalance

at the plasma membrane, it is believed not to be the only mechanism33

.

Caveolin, cholesterol and dynamin are known regulators of caveolae-

mediated endocytosis, as well as regulators of the actin cytoskeleton. The

shape and structural organization of caveolae is conferred by caveolin, a

10

dimeric protein that binds cholesterol and inserts as a loop into the inner

leaflet of the plasma membrane, followed by self-association to form a cave-

olin coated vesicle. There are thee types of caveolin; caveolin-1, 2 and 3,

where caveolin-1 is a scaffolding protein that oligomerizes at the plasma

membrane and is essential for the formation of caveolar vesicles. Cells that

do not express caveolin, are absent of caveolae. However, by reintroducing

it, the function is restored52

. Caveolin-2 is also involved in caveolae forma-

tion but is not essential, whereas caveolin-3 is specifically expressed in mus-

cle cells53-56

. The internalization via caveolae-mediated endocytosis is de-

pendent on dynamin and actin rearrangement, and uptake can be blocked by

overexpressing dominant negative mutants of dynamin or disassembling the

actin cytoskeleton (conner07, 15). Even though caveolae are highly immo-

bile at the plasma membrane, caveolin-1 mobility can be increased upon

activation, by e.g. the simian virus 40 (SV40) that uses a caveolae-mediated

pathway to enter cells and reach the smooth ER57-59

. However, in caveolin-1

knockout mice, the SV40 exploits a caveolin-independent pathway to inter-

nalize cells, indicating that if one endocytic route is blocked, other pathways

may be utilized60

. This is supported by another toxin, cholera toxin B (CTB)

that is internalized by cells via caveolae and CME, as well as dynamin-

dependent and independent raft pathways51

. The small caveolar vesicles are

internalized slowly, with a half-life longer than 20 minutes33

and similarly to

macropinocytosis and phagocytosis, caveolae-mediated endocytosis is trig-

gered by the cargo itself. Nevertheless, it seems to be highly regulated and

more investigations are needed to fully understand the mechanism.

Clathrin- and caveolae-independent endocytosis

A variety of nutrients, signaling receptors and pathogens use clathrin- and

caveolae-independent endocytosis to enter cells (Figure 3). Pathogens known

to be internalized by cells via caveolae or CME, e.g. SV40 and CTB, can

alternatively use this clathrin- and caveolae-independent pathway60,61

. Cho-

lesterol rich microdomains on the plasma membrane referred to as rafts, can

diffuse freely in the plasma membrane and are distinct from caveolae-

mediated endocytosis and CME, in that they lack caveolae and clathrin. It is

a constitutive internalization mechanism for the interleukin 2 (IL-2) receptor

and GPI-anchored proteins62,63

. Even though dynamin mutants have been

used to inhibit the uptake of IL-2, as well as caveolae and CME, they have

failed to inhibit fluid-phase uptake, indicating that there are other unidenti-

fied pinocytic pathways. One new pathway independent of clathrin, caveo-

lae, and dynamin, was shown to involve flotillin-164

. However, recently it

was found that a clathrin- and caveolae-independent but dynamin- and flotil-

lin-dependent route for PG-bound ligands, efficiently transported endocytic

vesicles to late endosomes (Payne07). The PG-dependent pathway was dis-

tinct from CME, in that it did not require phosphatidylinositol-3 kinase-

(PI3K) dependent sorting from early endosomes or microtubule-dependent

11

transport, to reach the late endosomes. There is also evidence for lipid-based

endocytosis in yeast since the yeast sterol, ergosterol, has a higher tendency

to form lipid rafts than cholesterol10,65

. In summary, the clathrin- and caveo-

lae-independent pathway may constitute a specialized high capacity endocyt-

ic pathway for both lipids and fluids, thus, the relationship of this pathway to

other forms of endocytosis remains elusive.

Membrane repair response

Although endocytosis and exocytosis allow macromolecules to enter and exit

cells via vesicular transport systems, without compromising the plasma

membrane, not all cellular import can be explained by this system66

. Disrup-

tion of the plasma membrane, in multicellular organisms, is not unusual.

Especially in mammalian cells, e.g. skeletal or cardiac muscle cells67,68

,

where the disruption frequency is directly correlated to the level of physical

activity69

. Thus, the plasma membrane constantly needs to be repaired.

One of the first studies to investigate cell membrane resealing in vivo,

used water-soluble markers, such as dextran labeled with fluorescein. It was

found that gut cells, wounded by mechanical forces, were able to reseal dis-

ruptions of the plasma membrane70

. However, resealing occurs in all cell

types with injured plasma membrane and the membrane repair process al-

lows the cells to survive injury71

. Due to the high extracellular calcium con-

centration, calcium ions leak into the cell when the plasma membrane is

injured and trigger a MRR mechanism. The MRR induces mobilization of

intracellular vesicles, to the disrupted site on the plasma membrane, to do-

nate their membrane and patch up the broken membrane (Figure 4). The

resealing is an active and complex structural modification, with endomem-

branes as the primary building block, together with cytoskeletal and mem-

brane fusion proteins as catalysts71

.

The MRR has been compared to neurotransmitter release since block of

calcium/calmodulin-dependent kinase II (CaMKII), a regulator of kinesin

and exocytic vesicles at the synapses, block both processes72

. Moreover, in

embryos, exocytosis and membrane resealing was blocked by neurotoxins

that selectively cleaved synaptobrevin, syntaxin, and synaptosomal-

associated protein-25 (SNAP-25), from the SNARE complex proteins. By

undocking cortical vesicles, resealing was blocked, indicating that internal

membranes are required and that a SNARE-like complex is important for

resealing, by membrane docking73

. It is believed that after injury the local

calcium level rises and activates CaMKII, which phosphorylates synapsin I,

thus releasing vesicles from actin filaments to be used for resealing74

.

12

Figure 4. MRR mechanism. Due to the disruption calcium ions can leak into the cell and trigger a MRR mechanism which induces the mobilization of intracellular ve-sicles to donate their membrane and patch up the broken membrane. Lysosomes are usually the main donator of membrane, resulting in the luminal side of the lysosome being exposed to the extracellular part of the plasma membrane.

This is followed by a kinesin driven delivery of vesicles, to the site of dis-

ruption, where vesicle docking and fusion is mediated by the SNARE pro-

teins75,76

. MRR driven exocytosis was visualized in confocal microscopy

during resealing upon wounding with laser and was shown to be calcium-

regulated, as a rapid burst of localized calcium-regulated exocytosis was

detected73,77

. The involvement of kinesin and myosin was further investi-

gated by blocking them with a kinesin antibody or with a myosin ATPase

inhibitor, respectively78

. Different temporal phases of exocytosis were found,

where the block of kinesin inhibited the slow phase and block of myosin

inhibited both the slow and fast phase. The blockage of CaMKII resulted in

inhibition of both phases, consistent with disruption of a myosin-actin-

dependent step of vesicle recruitment. Thus, kinesin and myosin motors may

mediate two sequential transport steps that recruit vesicles to the release sites

of calcium-regulated exocytosis. Furthermore it was shown that calpain was

required for the MRR as the calpain inhibitor, calpeptin, calpain knock-outs,

or the highly specific calpain inhibitor protein, calpastatin, decreased surviv-

al during plasma membrane disruption79

. Together, these results indicate that

13

calcium influx during membrane disruption activates calpain, which is ne-

cessary for remodeling of the cortical cytoskeleton, at the site of injury and

serves as a protector of the plasma membrane, against mechanical damage.

Target cell apoptosis

The intracellular environment inside the body is constantly monitored by an

internal surveillance system that recognizes injury or defects of cells. Pro-

grammed cell death, apoptosis, is an innate mechanism, enabling organisms

to eliminate defected cells during e.g. development or the response of the

immune system to viral infection80,81

. When cells are infected with pathogens

or transformed into tumor cells, a group of key players in the immune re-

sponse, the killer lymphocytes, recognize the defected cells and eliminate

them. Natural killer (NK) cells and cytotoxic T cells (CTL) belong to the

group of killer lymphocytes and are involved in adapted and innate immuni-

ty. They both act by the same mechanism, although triggered by different

stimuli82

. Two pathways can be utilized by CTL and NK cells, the granule

exocytosis pathway (Figure 5), which is the dominant pathway in eliminat-

ing virus infected cells, or the involvement of cell-surface death receptors.

Figure 5. Simplified schematic of the mechanism during target cell apoptosis. Gra-nules are exocytosed from CTL and NK cells, containing proteins such as perforin, granzymes and granulysin. Inside the cytoplasm, granulysin and granzymes induce apoptosis.

14

Moreover, human NK cells are distinct from CTLs, in that they constitu-

tively express perforin and granzymes, whereas CTLs require pre-activation

by dendritic cells81

. Thus, NK cells are important in the early stages of infec-

tion, until the T-cell response is fully activated. The exocytosed granules are

secretory vesicles stored in killer lymphocytes, containing several proteins

such as perforin, a pore-forming protein and granzymes, a group of serine

proteases that activates the caspase pathway during target cells apoptosis.

Other proteins found in granules are chemokines and granulysin, a cytolytic

protein against microbes and tumors cells83

. Following interaction with the

target cell, granular proteins are exocytosed from the CTL, together with

structural elements, such as chondroitin 4-sulfate PGs, to form a lattice to

which the toxins bind through electrostatic interactions84

. The basic gran-

zymes, were shown to be non-covalently bound to the negatively charged

granule PG serglycin, which has a protein backbone of alternating serine-

glycine residues, modified by side-chains of chondroitin 4-sulfate GAGs85,86

.

Perforin has also been proposed to bind GAGs but not as strongly as gran-

zymes, due to the lower amount of basic residues87

. Well inside the cytop-

lasm, apoptosis is induced by granzyme A and granzyme B via caspase-

independent or dependent pathway, respectively88,89

.

Perforin

Perforin is synthesized in the rough ER inside CTL and NK cells, followed

by transportation through the Golgi apparatus, where it is post-translationally

modified and packed into lysosomal-like cytoplasmic granules90

. Its expres-

sion is regulated during lymphocyte differentiation, by receptor activation

signals and cytokines. Upon plasma membrane interaction, perforin can pro-

duce pores up to 20 nm in diameter91,92

. Perforin has a calcium-dependent

membrane binding domain that plays a crucial role in the first step of the

membranolytic activity of perforin93

. Its activity is exerted in the extracellu-

lar environment and require >100 µM calcium to bind efficiently to the tar-

get membrane. The low calcium affinity, probably protects the cytotoxic

cells from autolysis during synthesis of perforin and the low granular pH

protonates an aspartic residue in perforin, preventing calcium and membrane

binding during exocytosis92

. However, the pore-forming ability of perforin

has been questioned, since little or no leakage of cytoplasmic contents has

been detected at physiological concentrations. Therefore two hypotheses has

formed: 1) perforin induces pores in the plasma membrane of the cell, enabl-

ing granzymes to diffuse freely into the cytoplasm of the target cell and in-

duce apoptosis, or 2) perforin is endocytosed together with granzymes and

disrupts the endosomes, thus releasing them into the cytoplasm of the target

cell84

. One study proposed that at physiologically relevant concentrations,

perforin induced the MRR that resealed the broken plasma membrane rapid-

ly, after the delivery of granzymes94

. By triggering the MRR, membrane

15

disturbances caused by perforin were resealed, supporting the first hypothe-

sis. Thus, no leakage of endogenous molecules could be detected. Neverthe-

less, there are studies still claiming that perforin lyses the membrane of en-

dosomes and that perforin is not required for granzyme-induced apoptosis85

.

Granzymes

There are several groups within the family of granzymes. In humans and

mice, granzyme A and B, are most abundant. This group of highly specific

serine proteases, are processed either on route to, or in granules, from inac-

tive pro-enzymes into active enzymes by cathepsin C82

. In addition to induc-

ing apoptosis by cleaving caspases, granzymes have been proposed to have

other functions, as the presence of elevated numbers of circulating gran-

zymes in various inflammatory processes and cleavage of extracellular sub-

strates, suggest involvement in e.g. virus and tumor rejection95

. It was pre-

viously shown that granzyme B reduced the growth of adherent tumor cell

lines by preventing their adherence to extracellular matrix proteins96

. Inside

the granules, granzymes are bound to the chondroitin 4-sulfate PG, sergly-

cin, as mentioned above and due to the low pH they are inactive97

. After

exocytosis granzymes are proposed to undergo electrostatic exchange, from

serglycin to highly sulfated cell surface PGs98

. Two cationic sequences were

identified as important in uptake since replacing arginines within the se-

quence by alanines, inhibited the uptake, corroborating the exchange of

chondroitin 4-sulfate on serglycin, to anionic components on the plasma

membrane99

. In addition, the mutations and treatment with heparin reduced

the cytotoxicity. Two uptake models were proposed; receptor-dependent

endocytosis and nonselective adsorptive pinocytosis. However, the first

study did not find perforin to be involved in uptake, whereas the second

study found the granzyme internalization to be perforin-dependent. Nonethe-

less, serglycin of granules and PGs on the plasma membrane seem to be

involved in uptake. The question if perforin is involved, and how, remains

unanswered but a majority of the recent investigations find that perforin at

least enhances granzyme-dependent apoptosis81,82,84,92

.

Granulysin

As a member of the saposin-like family of lipid binding proteins, granulysin

is a lytic molecule active against e.g. tumor cells and microbes100

. During

target cell apoptosis two protein products of 15 and 9 kDa are detected in

CTL and NK cells100,101

. The proteins are exocytosed together with perforin

and granzymes. Recombinant 9 kDa granulysin was shown to activate cas-

pase-9 to induce apoptosis in nucleated cells but not red blood cells102

. How-

ever, shorter peptides of 9 kDa granulysin lyse erythrocytes and the hemoly-

sis was shown to depend on specific ion channels, as blockage of them inhi-

16

bit lysis103

. Peptides corresponding to the N or C terminus are not lytic, whe-

reas peptides from the central region lyse bacteria, human cells and synthetic

liposomes. The replacement of arginines by glutamine also reduces the ly-

sis104

. Upon binding to the cell surface, the intracellular calcium level in

human T cell tumor Jurkat increase102

. Furthermore, granulysin treatment

damage mitochondria, resulting in loss of electrostatic potential, release of

cytochrome c and apoptosis-inducing factor105

. It was shown that both cyto-

solic and mitochondrial calcium levels rise, whereas potassium levels de-

crease and, that this is required for granulysin-induced apoptosis since cell

death is prevented by calcium and potassium blockers106

.

Cell-penetrating peptides as drug delivery vehicles

The greatest hindrance in drug delivery of hydrophilic drugs today, is to

cross the hydrophobic plasma membrane, in order to reach the cytoplasm or

nucleus, to exert their therapeutic effect. Several transporters in the plasma

membrane are able to internalize smaller molecules but for larger macromo-

lecules such as peptides, proteins, and oligonucleotides, some kind of deli-

very vehicle is needed to deliver the cargo. Carriers have been developed for

delivery of different cargoes e.g. cationic lipids107

, nanotubes108

and poly-

mers109

. Two classes of delivery vehicles are assessed in this thesis namely,

CPPs (Table 1) and bacterial ghosts.

Twenty years ago, during the development of an activity assay for a pro-

tein from the human immunodeficiency virus type 1 (HIV-1), the trans-

activator of transcription (Tat) protein, was found to be internalized by tissue

culture cells110

. It localized to the nucleus and if lysosomotropic agents as

chloroquine were added, its activity increased dramatically.

Table 1. Names and sequences of some of the most common CPPs. All peptides have an amidated C terminus except MPG and Pep-1, which have a cysteamide.

Name Sequence Origin

Penetratin111 RQIKIWFQNRRMKWKK Protein derived

Tat(48-60)112 GRKKRRQRRRPPQ Protein derived

pVEC113 LLIILRRRIRKQAHAHSK Protein derived

bPrPp114 MVKSKIGSWI10LVLFVAMWSDVGLCKKRPKP Protein derived

MPG115 GALFLGFLGAAGSTMGAWSQPKKKRKV Chimera

Pep-1116 KETWWETWWTEWSQPKKKRKV Chimera

Transportan117 GWTLNSAGYLLGKINLKALAALAKKIL Chimera

TP10118 AGYLLGKINLKALAALAKKIL Chimera

MAP119 KLALKLALKALKAALKLA Non-protein derived

Poly Arginine120 (R)n Non-protein derived

17

In another laboratory, chemically synthesized Tat, together with two shorter

analogs of 21 and 41 amino acids with retained activity, was shown to enter

cells rapidly121

. About the same time, it was found that a 60 amino acids long

polypeptide from another transcription factor, corresponding to the Antenna-

pedia gene homeobox sequence in Drosophila, internalized into cells122

.

Some years later it was shown that amino acids 48-60 of Tat112

and 16 amino

acids from the third helix of the Antennapedia protein, named penetratin111

,

were sufficient for translocation. These two peptides, Tat (48-60) and pene-

tratin, initiated the field of CPPs which have expanded greatly over the last

two decades.

Structural features

Several attempts have been made to divide CPPs into different structural

categories i.e. cationic, amphipathic, chimeric. However, it has proven to be

difficult due to the high degree of overlap between the different categories.

Nevertheless, there is one feature that seems to be more important than oth-

ers; basic residues in the peptide sequence, due to the ability of arginines and

lysines to form hydrogen bonds with phosphate, carboxylate and sulfate

anions120,123,124

. Classical examples of CPPs containing arginines are penetra-

tin, Tat (48-60) and polyarginine120,125

. However, there are several peptides

without arginines e.g. MAP119

, transportan117

and TP10118

, containing lysines

that are also able to form hydrogen bonds, although not bidentate bonds as

arginines. The fact that both Tat (48-60) and penetratin originate from tran-

scription factors, suggests that there are natural sequences within some pro-

teins, enabling them to translocate across plasma membranes of cells, in

order to exert their effect. Other CPPs, e.g. MAP and polyarginines, are not

derived from proteins, although a high sequence resemblance with certain

proteins can be found. Some CPPs even originate from toxins, such as cro-

tamine126

originating from the venom of rattlesnake, and transportan and

TP10, which are chimeras of mastoparan from the wasp venom and the neu-

ropeptide galanin117,118

. Furthermore, CPPs are in many ways similar to an-

timicrobial peptides and a comparison will be described in an upcoming

section. Thus, the origin of the CPPs varies greatly and some of the structur-

al investigations are summarized below.

Many modifications and analyses of existing CPPs have been done in an

attempt to improve uptake efficiency and understand more about the uptake

mechanism. Introduction of a negative net charge, as well as reduction of the

amino acid sequence of MAP, by more than two amino acids, abolishes its

translocation127

. On the contrary a shorter analog of transportan, TP10, re-

tains its activity despite the truncation118

. In the presence of phospholipid

vesicles, transportan shows 60% helical structure and penetratin a β-sheet

like structure, whereas both peptides have random coiled structures in

aqueous solution128

. Another study showed that penetratin has an amphipath-

18

ic helical structure in SDS micelles and that the indole group of tryptophan

inserts into micelles and is important for uptake129

. Structure-activity rela-

tionship studies of truncated or mutated penetratin revealed that shorter se-

quences are be taken up, although not as efficiently as penetratin and replac-

ing the basic residues by alanines, reduces the uptake, whereas substitution

of hydrophobic residues with alanine is almost unnoticeable130

.

Extensive studies and modifications have been performed on several ana-

logs of Tat (48-60) and polyarginine. Unfortunately, it is difficult to compare

data from Tat studies since the analyzed sequence, referred to as Tat, varies

greatly in amino acid composition. Trucation of the 9-mer Tat (49-57), by

one amino acid from either terminus, is sufficient to reduce the uptake by

80% and additional truncation, reduces the uptake even further131,132

. Taken

together, the results indicate that arginine residues are more important for the

uptake of Tat (49-57) than lysine residues, as both nonaarginine and Tat (49-

57) were superior to nonalysine. After the discovery that the backbone ste-

reochemistry of the polyarginines is not necessary for uptake, a cavalcade of

non-peptide transporters has been designed. One potent non-peptide trans-

porter is the R-ahx-R, compromising a single ahx acid spacer between con-

secutive arginines, indicating that higher flexibility increases uptake. A

comparative study of transporters coupled to cargo corroborated that (R-ahx-

R)4 is the most efficient carrier out of Tat (48-60) and R9133

.

Many structural and conformational analyses have been performed with

the MPG peptide. MPG is a chimeric peptide derived from HIV-1 gp41 pro-

tein and the nuclear localization sequence (NLS) from SV40 large T anti-

gen115,134,135

. Its uptake is endocytosis-independent with or without cargo and

by replacing a lysine to a serine in the NLS sequence, targeting of the CPP is

changed from the nucleus to the cytoplasm, respectively136

. Both MPG and

its analog insert spontaneously into phospholipids membranes and strongly

disrupt lipid organization of the monolayers137

.

Mechanisms of uptake

Two decades ago the internalization of the Tat protein was believed to de-

pend on endocytosis138

. Nevertheless, shorter peptides of Tat and Antenna-

pedia, were interpreted to be independent of metabolic energy since uptake

still occurred at 4 °C111,112

and several direct penetration mechanisms were

proposed e.g. the inverted micelle139

and sinking raft models140

(Figure 6).

Two uptake mechanisms for antimicrobial peptides (AMPs), namely the

barrel-stave and carpet model, described in the following section (Figure 7),

were also proposed as potential mechanisms for CPP uptake. However, the

direct penetration mechanism had to be reevaluated since it was discovered

that fixation of cells caused artifacts141

and several well-characterized CPPs

were shown to enter cells by different forms of endocytosis.

19

Figure 6. Proposed mechanisms for CPP uptake. The inverted micelle model pro-poses that following peptide binding, the membrane is destabilized by tryptophan and the formation of a micelle is induced

139. In the sinking raft model, the peptide

helices aggregate parallel to the surface, forming a “raft” that sinks into the outer bilayer leaflet

142. The hydrophobic residues remain in contact with the lipids, as the

helices sink deeper, while the hydrophilic residues form a hole, similarly to the bar-rel-stave model.

Therefore the debate on translocation is still ongoing and highly active.

Nonetheless, the general opinion is that the first step of translocation in-

volves hydrogen bonding between guanidinium/amino groups with PGs on

the plasma membrane, although it is not crucial143,144

. Several endocytosis

routes have been suggested as major uptake pathways, e.g. macropinocyto-

sis124,145,146

, CME147

and caveolae-mediated endocytosis148-150

. It has also

been suggested that different forms of endocytosis function simultaneously

and by blocking one pathway, other pathways become more active151,152

,

which may explain ambiguous results in the application of endocytosis inhi-

bitors. Our investigations together with several other studies, suggest that

there are two uptake pathways involved in CPP uptake; endocytosis and

direct translocation through the plasma membrane (Paper I, II, III). One

study with D and L enantiomers of octaarginine, showed that the uptake is

concentration-dependent and that the CPPs use two pathways; endocytosis at

low concentrations and direct penetration at high concentrations153

. Another

study with Tat (47-57) indicated that at least two functionally distinct uptake

mechanisms are involved and that the uptake is dependent on the size of the

cargo154

. The direct translocation mechanism is supported by a molecular

dynamics simulation, indicating that arginine residues of Tat (47-57) insert

into the membrane, lowering the free energy by binding phospholipids and

carrying them along through a transient pore, as they insert into the mem-

brane155

.

20

Another CPP, Pep-1 change conformation upon increasing concentration,

from non-structured to helical and the helical conformation is also found in

contact with phospholipids116,156

. It was proposed that the peptide induces

formation of a transient transmembrane pore-like structure and that the con-

formational change occurs upon association with the plasma membrane,

independently of cargo presence. Moreover, the first step in the internaliza-

tion of the MPG peptide is also independent of cargo binding and involve

Rac-1 GTPase-dependent remodeling of the actin network157

. By binding

GAGs, MPG induces local membrane disturbances and actin rearrangement

that lead to uptake either by membrane fusion or endocytosis. An analytical

study found two kinetic models describing peptide-induced membrane lea-

kage of two prion-protein-derived peptides and melittin, a peptide from the

honey bee, as one pore-forming model and second membrane destabiliz-

ing/perturbing model158

.

In summary, there are several studies indicating that endocytosis are not

the only uptake pathway and by changing only a single amino acid, the des-

tination of the CPP can be changed136

.

Cargo delivery

The capacity of CPPs to deliver cargo molecules into the cell and induce a

biological response, has opened up new possibilities in drug delivery re-

search159-161

. In addition, the CPP alone, can induce biological effects, as

shown with a shorter sequence from the ADP-ribosylation factor (ARF) pro-

tein that translocates into cells efficiently and induce apoptosis162

. The two

main strategies in CPP delivery are: 1) to couple the cargo covalently to the

CPP or 2) to bind the cargo electrostatically to the CPP.

The first strategy most often involves disulfide bridges that are used to

couple the cargo covalently to the peptide, as in the delivery of oligonucleo-

tides to induce splice-switching both in vivo163

and in vitro164

. The peptide-

cargo conjugate can also be coupled via a native peptide bond. However, this

will most probably affect the binding of the cargo to the target negatively,

due to steric hindrance. Inside the cell, the disulfide bridge is reduced by

glutathione and the cargo is released. To be able to quantitatively determine

how much cargo is internalized, a screening system was developed based on

the conversion of luciferin in the cytoplasm of cells, from an inactive to an

active state165,166

. Thus, successful delivery into the cytoplasm is ascertained.

Another study used a Tat-Cre (48-57) fusion peptide to investigate both the

biological effect and the translocation mechanism. The fusion peptide was

found to be internalized via lipid raft-dependent macropinocytosis and by

adding a sequence from the influenza virus protein, hemagglutinin (HA2),

known to disrupt endosomes upon acidification, the biological effect was

enhanced167

. Similarly, another study took advantage of the endosomal acidi-

fication by inserting histidines into the sequence of penetratin, in an attempt

21

to construct a lysosomal disrupting peptide168

. The new peptide, named EB1,

was found to be more efficient in cargo delivery than penetratin. Impor-

tantly, the luciferin assay, the splice correction assay, and the Tat-Cre (48-

57) fusion peptide, eliminate the risk of fixation artifacts. In addition, the

covalent strategy is advantageous since it is possible to measure the effective

concentration of the construct. However, it includes several purification

steps in which a certain amount of material will be lost.

The second strategy used for delivery of oligonucleotides and proteins, is

coincubation, where the peptide binds the cargo electrostatically and/or by

hydrophobic interactions116,169

. This method does not require covalent cou-

pling or purification of the constructs and is therefore time-saving and results

in minor loss of material. CPPs have even been conjugated to already estab-

lished carriers, such as liposomes, to improve the efficiency of the delivery

vehicle even further170,171

. In summary, numerous investigations have been

performed to analyze internalization and biological functions of CPPs. How-

ever, due to the high quantity of studies, further information on the diversity

of cargo delivery by CPPs can be found in several excellent reviews172-175

.

Stability of cell-penetrating peptides

After proteins are processed in the proteasome, nearly all generated peptides

are rapidly cleaved into amino acids, either in the cytoplasm or nucleus176

.

Today, there are only a few known cytoplasmic peptidases, e.g. tripeptidyl

peptidase II (TPPII)177

, leucine aminopeptidase (LAP)178

, and thimet oligo-

peptidase (TOP)179

that are able to degrade peptides with different specifici-

ties. Almost independently of sequence, aminopeptidases degrade peptides

in the cytoplasm within seconds180,181

. LAP represents one of the more spe-

cific peptidases, cleaving single hydrophobic amino acids from the N termi-

nus. Peptides comprising more than 15 amino acids are trimmed by TPPII, a

serine protease removing tripeptides from the N terminus but also exhibiting

low endopeptidase activity182

. Sequences of 6-17 amino acids are preferen-

tially cleaved by the zinc-containing metallopeptidase, TOP, in addition to

TPPII, by endoproteolytic cleavage that can be inhibited by phenanthro-

line176

. TOP is a highly conserved enzyme and a homolog can be can be

found in both bacteria183

and yeast184

. Furthermore, peptides in the cytosol

can be transferred to the ER by a transporter associated with antigen

processing (TAP), belonging to the family of ABC transporters, to be further

degraded by ER-aminopeptidases185,186

. There may even be a competition

between cytosolic peptidases and TAP, in binding of peptides that may result

in delayed degradation, if peptides are bound to cytoplasmic/nuclear chape-

rons187

. Although TOP belongs to the family of matrix metalloproteinase

(MMP), it is predominantly situated in the cytoplasm. Otherwise most

MMPs are capable of degrading all constituents of the extracellular matrix

and members of this family are believed to be zinc proteinases188-190

. MMPs

22

are either secreted from the cell or anchored to the plasma membrane and up

to date, 24 different vertebrate MMPs have been identified191

. There are sev-

eral other groups of transmembrane extracellular proteases, such as the β-

secretace and a disintegrin and metalloproteinase (ADAM) families, which

can be transmembrane or secreted and play important roles in cell adhesion,

migration, proteolysis and signaling192,193

.

A crucial aspect of drug delivery is, of course, efficient transport of the

pharmaceuticals to the site of action, but also the following clearance of the

transporting moiety from the cell. The CPP needs to translocate its cargo

before it is extracellularly degraded, although, inside the cell it has to be

eliminated, in order not to cause toxicity176

. In addition, if the CPPs remain

intact in the cellular environment, there is a possibility that the peptide-cargo

conjugate leaks back to the extracellular environment. The understanding of

the internalization and degradation kinetics of CPPs is important for the

practical aspects of cargo delivery. In uptake experiments, with incubation

time points of one hour or longer, the peptide needs to be stabile outside the

membrane and not degraded within that time frame. Hence, evaluation of

peptide stability is an important parameter. These studies are also essential in

order to achieve a better understanding of the mechanism by which CPPs

pass through membranes and enter cells. Recently, it was shown that the

main mechanism by which CPP-cargo accumulate into the cellular interior,

is by proteolytic processing of the CPP-cargo conjugate, into membrane

impermeable products194

. Thus, it is important to characterize the degrada-

tion kinetics and products of CPPs, in order to investigate how they relate to

cargo-delivery efficiency.

A few studies have investigated the metabolic stability and/or the pattern

of degradation of CPPs. Enzymatic degradation of pVEC and its all-D analog

was investigated in buffer containing physiological concentration of trypsin

or carboxypeptidase A and B and the half-lives were found to be 10.5 and

44.6 min in respective buffer sample195

. Another study investigating trans-

portan, TP10 and penetratin in contact with Caco-2 cells, found that the sta-

bility of the peptides was in the order of transportan>TP10>penetratin196

.

Moreover, a comparative study of Tat (47-57), penetratin and several human

calcitonin (hCT) analogs in MDCK (Madin-Darby Canine Kidney), Calu-3

(Caucasian, lung, adenocarcinoma) and TR146 cells, showed the levels of

proteolytic activity varies highly among cell lines197

and that the extracellu-

lar half-life of the individual peptides depend on cell type and cell density. In

an attempt to increase CPP stability, two amino acids were replaced with N-

methylphenylalanine or D-phenylalanine in hCT(9-32) and the proteolytic

resistance in human blood plasma and HEK 293T cell culture supernatants

was found to increase198

. However, these studies did not include intracellular

degradation and how it affects the uptake. A similar study investigated the

possibility to increase the stability of the peptide by incorporating non-α-

amino acids into CPPs coupled to oligonucleotides. Both intra- and extracel-

23

lular stability was increased199

and time-dependent CPP degradation was

observed in both serum and tissue lysates200

.

Cell-penetrating peptides versus antimicrobial peptides

AMPs serve as natural antibiotics and can be found in all organisms includ-

ing humans and plants201-203

. In vitro and in vivo studies have illustrated the

importance of AMPs in innate immunity. Similarly to CPPs, the high diver-

sity of AMPs makes it difficult to categorize them, except on structural fea-

tures. Several structural characteristics are shared between AMPs and CPPs,

e.g. the high degree of cationic amino acids in combination with hydrophob-

ic amino acids. Still, most AMPs are characterized by high amphipathicity

that often results in helical conformations. Many AMPs have random coil

structure in solution but adopt helical conformation upon binding to the

membrane, such as cecropin and magainin, originating from the silk moth

and the African clawed frog, respectively204,205

. It is believed that the net

positive charge of the AMPs facilitates binding to the negatively charged

LPS on gram-negative bacteria and teichoic acid on gram-positive bacteria.

Proposed uptake mechanisms have been e.g. the carpet and barrel-stave

model206

(Figure 7). Although all AMPs are derived from larger precursors,

the diversity of peptides is such that the same peptide sequence seldom is

found in two different species203

. However, there are exceptions, e.g. buforin

II discovered in the stomach of the Asian toad Bufo bufo gargarizans207

.

Figure 7. Proposed mechanisms of AMP uptake. The barrel-stave model is a clas-sical transmembrane pore, where amphipathic α-helices form bundles. The hydro-phobic residues interact with the lipid membrane and the hydrophilic residues form a water-filled pore

208. In the carpet model, peptides adsorb to the membrane until a

critical concentration, where the peptides act like a detergent, causing disintegration of the membrane

209.

24

There are also differences in how the peptides kill the bacteria i.e. some

AMPs permeabilize the bacterial membrane, others target intracellular com-

pounds and some utilize both procedures203,210,211

.

Recently, we showed that CPPs translocate across non-mammalian cells

and act as AMPs against gram-positive and gram-negative bacteria (Paper I).

Similarly, an analog of the CPP Pep-1, is antimicrobial against gram-positive

and negative bacteria already at 2 µM, whereas no toxicity is found in hu-

man erythrocytes even at 200 µM212

. AMPs like magainin 2 and lactoferricin

B, have been shown to translocate across the bacterial membrane and reside

in the cytoplasm213

. Furthermore, analogs of magainin 2 and buforin 2, inter-

nalize mammalian HeLa cells and the buforin analog is able to deliver green

fluorescent protein (GFP) in a comparative study with TAT (47-57)214

. Al-

though the peptides internalize cells, the respective mechanisms seem to be

distinct. Another AMP belonging to the class of cathelicidins, LL-37 is able

to transfer extracellular DNA plasmids to the nuclear compartment of mam-

malian cells215

. LL-37 expression is known to be induced by bacterial infec-

tions resulting in efficient lysis of the microbes. Interestingly, when LL-37 is

incubated with plasmid DNA at concentrations lethal to bacteria but without

cytotoxicity to mammalian cells, the plasmid DNA is transferred and ex-

pressed in the nucleus. The uptake was found to be caveolae-independent

raft-mediated endocytosis, dependent on PGs. In addition, LL-37 protects

the DNA from serum nuclease degradation. Taken together, there are a great

number of studies illustrating the similarities between CPPs and AMPs.

However, due to previous proposals of non-endocytic uptake for AMPs206

,

together with the opinion that CPPs do not enter cells by membrane disturb-

ing pathways i.e. direct penetration, the similarities have not been well-

accepted.

Bacterial ghosts as drug delivery vectors

A major breakthrough in the field of vaccinology is the use of naked DNA

for immunization. However, DNA vaccines today are restrained by low im-

munogenicity and the requirement of high doses of plasmids. An explanation

for the poor immunogenicity may be that antigens are not delivered in the

context of an optimal danger signal and since the immune system rather rec-

ognizes injury, than foreign entities, it is not activated. Thus, proper activa-

tion of antigen presenting cells can only occur by pathologically altered

cells216

. A new field exploring the ability of dead but morphologically intact

bacteria, known as bacterial ghosts, to act as potential delivery vehicles of

DNA, has emerged217,218

. Bacterial ghosts are considered to be a new and

exciting non-viral tool in gene delivery and vaccination since they efficiently

target antigen-presenting cells and other eukaryotic cells. The ghosts are

produced by protein E-mediated lysis of gram-negative bacteria, carrying the

25

plasmid encoding lysis gene E of bacteriophage PhiX174219,220

. The lytic

activity of the membrane protein comprising 91 amino acids, has been local-

ized to 29 amino acids in the N terminus and is believed to be positioned in

the membrane spanning region of the protein220,221

. The protein induces a

transmembrane tunnel in the bacteria by fusing the inner and outer mem-

brane, releasing the cytoplasmic contents from the bacteria, leaving it empty

but structurally intact222

. Thus, the surface structures and antigenic features,

including bioadhesive properties of the natural cell, remain unaltered. Lysis

of bacteria has also been shown to depend on the rate of gene E translation

and on the bacterial growth phase223

. The protein E lysis is a time consuming

process since the bacteria have to be loaded with the lysis plasmid in order to

be lysed and then reloaded with new plasmids or DNA of interest.

As macrophages localize to sites of inflammation and tumors, as well as

adhere to endothelium and transmigrate to the focus of injury, they have

been proposed as cellular delivery vehicles for adoptive immunotherapy.

Applications as gene-dependent enzyme prodrug therapy and expression of

cytokines for the stimulation of macrophage tumoricidal activity, have been

proposed224

. Thus, increased targeting of macrophages by DNA delivery

vehicles and expression of transgenes in macrophages is needed. Bacterial

ghosts are known to efficiently target antigen presenting cells as well as

other eukaryotic cells225-228

. Transfection of macrophages has otherwise

proven to be difficult due to the low size limit and efficiency of transfection

agents. Thus, bacterial ghosts are potential delivery vehicles of DNA as they

efficiently internalize macrophages.

Another application is to use bacterial ghosts for gene delivery into cancer

cells. Recently, melanoma cells were shown to internalize ghosts containing

DNA via phagocytosis, with much higher efficacy than the non-liposomal

lipid transfection reagent Effectene217

. It was suggested that the LPS on the

ghosts activated the melanoma cells through the toll-like-receptor-4 (TLR-

4), which is constitutively expressed in these cells, resulting in an increased

production of IL-8 and cell adhesion229

. Thus, the native function of the cell

could potentially be restored by delivering genes encoding the appropriate

proteins to tumor cells with known mutations. Potentially, even the immune

system could be induced by delivering genes encoding cytokines that are

known to activate and recruit antigen presenting cells, further activating the

immune response230,231

. During the activation and development of the im-

mune response, cytokines are crucial factors, including the response against

tumors, which was shown in patients given IL-2, a T-cell growth factor232

.

Even the anticancer drug doxorubicin exhibited antiproliferative activities

against cancer cells after delivery by bacterial ghost in vitro233

. Furthermore,

the significance of which bacterial strain to use for each cell line was illus-

trated in the same study as the ghosts of M. haemolytica were more potent

delivery vectors than E. coli in Caco-2 cells. Moreover, ghosts produced

from M. haemolytica induced maturation and activation of dendritic cells in

26

vitro and β-galactosidase-specific immune response was detected after intra-

venous immunization of mice with dendritic cells transfected ex vivo with

bacterial ghosts loaded with a β-galactosidase-coding plasmid234

. Thus, the

combination of natural adjuvant properties together with the potential carrier

function, make ghosts an interesting novel DNA delivery vehicle in DNA

vaccination.

27

Aims of the thesis

Understanding how CPPs enter cells and what uptake pathways are involved

is of outmost importance in order to produce more efficient CPPs and to be

able to foresee side-effects. If these peptides are to be used in drug delivery

in vivo, it is necessary to know the uptake and degradation kinetics. If not,

the peptides might be degraded before they reach their target. Thus, the aim

of the thesis has been to investigate how CPPs translocate across cell mem-

branes and to evaluate the importance of the plasma membrane from differ-

ent species to study how it affects uptake and degradation kinetics. Similari-

ties between CPPs and antimicrobial peptides were investigated and used for

development of new CPP applications. Moreover, the correlation between

native proteins, able to translocate across cells in target cell apoptosis and

shorter sequences from these proteins, was investigated and compared to

CPPs, to establish the involvement of the MRR in CPP uptake. The main

goals are presented below.

Study how membrane composition in different cell lines affects up-

take and investigate if CPPs are antimicrobial. (Paper I)

Investigate degradation and uptake kinetics of structurally different

CPPs in mammalian and non-mammalian. (Paper I and II)

Elucidate uptake pathways involved in CPP internalization in both

mammalian and non-mammalian cells. (Paper I-III)

Examine if the MRR is responsible for masking membrane distur-

bances caused by CPP uptake. (Paper III)

Develop a new CPP based strategy to produce bacterial ghosts.

(Paper IV)

28

Methodological considerations

Design and evaluation of peptides

Throughout the thesis two well-known CPPs have been studied, namely

MAP and penetratin, in order to be able to compare the effect of the peptides

in each paper. These peptides are structurally different both in secondary

structure and amino acid composition. The non-protein derived peptide

MAP, is highly amphipathic and forms a perfect α-helix with one hydropho-

bic and one hydrophilic side. Unlike most CPPs, it contains no arginine resi-

dues but consists of repetitive sections of lysine, alanine and leucine. Pene-

tratin on the other hand has several arginine residues and is derived from a

non-mammalian transcription factor from the Antennapedia protein in Dro-

sophila. The secondary structure of penetratin depends on the surrounding

environment and it has been shown to change structure depending on the

environment128,129

.

In paper I uptake and antimicrobial effects of CPPs were investigated in

different non-mammalian cell lines. The idea was to compare peptides with

different origins. Therefore a third peptide was introduced, pVEC, derived

from vascular endothelial cadherin (VEC) which is an integral membrane

protein from murine and involved in cell adhesion113

. The negative control in

this paper was taken from the same protein to show that not any peptide can

enter cells235

. In addition a D-analog of pVEC was studied to compare deg-

radation and uptake kinetics. The degradation investigations of MAP and

penetratin were continued in paper II, and in paper III 4 new peptides from 3

different proteins involved in target cell apoptosis; perforin, granulysin and

granzyme B, were synthesized and investigated. The sequences were chosen

based on their basic and hydrophobic amino acid residue contents. The novel

peptides were named pPrF82, pPrF338, pGrL and pGrB (Table 2). Shorter

sequences of the last two peptides were recently shown to be important in

uptake and toxicity of the native proteins99,103

(Li and Bird) therefore these

sequences were extended in order to make them more CPP-like. Two more

peptides, previously reported to cause calcium influx and cytotoxicity,

pPrF1a and the shorter version pPrF1b236

. However, in our hands the pep-

tides did not display any of these features and were therefore used as nega-

tive controls (Table 2).

29

Synthesis (Paper I-IV)

All peptides were synthesized by solid phase peptide synthesis (SPPS)237

.

The advantage of this method is that the amino acids are coupled to a solid

support and not in solution, which simplifies the washing steps. Amino acids

were coupled with dicyclohexylcarbodiimide (DCC) and N-

hydroxybenzotriazole (HOBt). Each residue has a tert-butyloxycarbonyl (t-

Boc) protected α-amino group which is easily removed by trifluoroacetic

acid (TFA) before the next amino acid is coupled. After synthesis the peptide

and the side-chain protecting groups are cleaved from the resin by hydrogen

fluoride, an even stronger acid. Cresol scavengers are used to protect the

side-chains from the newly formed carbocations, arisen from the protecting

groups. The peptide is then extracted from the scavengers in a mixture of

ether and 10% acetic acid, filtered from the resin and lyophilized to remove

the solvent.

5,6-Carboxyfluorescein was activated with HOBt and diisopropylcar-

bodiimide (DIC) and coupled to the N terminus of the peptide prior to pep-

tide cleavage. Following cleavage, the peptides were lyophilized and puri-

fied on a C18 column in reverse-phase (RP)-HPLC using a gradient of ace-

tonitrile and water containing 0.1% TFA. The corresponding peptides were

identified by matrix-assisted laser desorption time-of-flight (MALDI-ToF)

mass spectrometry. For electron microscopy performed in paper IV, cysteine

was coupled to the N terminus of the peptide instead of fluorescein, followed

by coupling to a 1.4 nm nanogold particle via a maleimide bond.

Table 2. Name and sequence of the synthesized peptides utilized in this thesis with their respective reference. All peptides were amidated at the C terminus.

Name Sequence

Penetratin111 RQIKIWFQNRRMKWKK

MAP119 KLALKLALKALKAALKLA

pVEC113 LLIILRRRIRKQAHAHSK

pVEC(28-35)235 YDEEGGGE

pPrF82238 QRHVTRAKVSSTEAVAR

pPrF338238 ALRRALSQYLTDRARWR

pGrB238 IMLLQLERKAKRTRAV

pGrL238 NAATRVCRTGRSRWR

pPrF1a236 PCHTAARSECKRSHKFVPGAWLAGEGVDVTSLRR

pPrF1b236 PCHTAARSECKRSHKF

30

Cell cultures

Non-mammalian cells (Paper I and IV)

Four non-mammalian cell lines were analyzed in paper I to investigate how

important the composition of the plasma membrane is for uptake and if CPPs

are able to traverse non-mammalian cell barriers. In addition to internaliza-

tion, the antimicrobial effects of CPPs were analyzed. Bacillus megaterium

was chosen to represent the class of gram-positive bacteria. It has been used

in industry for over 50 years due to its high capacity to produce exoenzymes.

The size of B. megaterium is about a hundred times that of E. coli and was

therefore named the big beast by De Bary. It has been used as a genetic tool

for cloning since it can maintain a variety of plasmids and does not express

external alkaline proteases. Next E. coli was chosen to represent gram-

negative bacteria in paper I. E. coli was first discovered in 1885 by the Ger-

man pediatrician and bacteriologist Theodor Escherich and is widely used as

a genetic tool. Moreover, in paper IV bacterial ghosts produced from E. coli

were used as delivery vehicles of plasmids into mammalian cells. S. cere-

visiae, representing yeast in paper I, is often used as a tool in cell cycle regu-

lation and is one of the most widely studied eukaryotes. The last non-

mammalian cell line in paper I was insect cells derived from the pupal ovar-

ian tissue of the Fall armyworm, Spodoptera frugiperda (Sf9), a moth larva

to compare with bacteria and yeast.

Mammalian cells (Paper II-III)

Chinese hamster ovary (CHO-K1) cells are an epithelial cell line taken from

a biopsy of an adult Chinese hamster. The CHO cells are extensively used in

biological and medical research and were therefore used in paper II and III

for ease of comparison.

Henrietta Lacks (HeLa) is involuntary a well-known name in the research

community. Cervical adenocarcinoma cells were taken from her without her

knowledge after her death in 1951 and have since then been used all over the

world in cancer research. HeLa cells are a rapidly growing immortalized

epithelial cell line known to be easily transfected and were previously uti-

lized to internalize ghost217

. The transfection efficacy of ghosts containing

the GFP plasmid239

was analyzed in paper IV.

Minimum inhibitory concentration (Paper I and IV)

The minimum inhibitory concentration (MIC) is used to investigate if a pep-

tide is antimicrobial or not. By performing the broth microdilution method it

31

is possible to measure at what concentration the peptide is able to kill bacte-

ria or yeast240

. The principle is to grow cultures over night, transfer aliquots

to fresh culture medium the next day and incubate for an additional 3–5 h to

reach mid-logarithmic phase organisms. The cultures are then diluted in

medium together with peptides in a 96-well plate or spread onto agar plate

and incubated for 16 h. The antimicrobial activity of the peptide is then ana-

lyzed at 595 nm in a microplate reader. The aim was to optimize the MIC

values and time points of incubation. Therefore bacteria were incubated in

medium or 10 mM NaH2PO4 pH 7.4 for one or two hours, since incubation

with buffer gave the lowest MIC value.

Optimization of ghost production

A variant of the MIC assay was used to determine peptide concentrations

lethal to E. coli. The same procedure was followed except that after reaching

mid-logarithmic phase the bacteria was centrifuged for 5 min at 1000 g and

resuspended in broth or buffer. Bacteria of different optical densities (ODs)

were incubated with peptides in broth or buffer for 1 or 2 h and then a small

aliquot of the peptide-bacteria incubation solution was transferred to fresh

broth in a 96-well plate and onto agar plates followed by 16 h incubation to

determine the lethal dose of peptide at a certain time point. Survival was

measured as absorbance at 595 nm on a Digiscan absorbance reader where

zero absorbance was interpreted as a lethal dose.

Agarose gel analysis of DNA (Paper IV)

Bacteria with different ODs were treated with 10 or 100 µM MAP in 10 mM

Na2HPO4 buffer (pH 7.4) or LB for 1 h to perforate the plasma membrane

and release the cytoplasmic contents of the bacteria. To confirm that the

plasmid still was inside the bacterial ghosts plasmid DNA was extracted and

prepared as described previously241

and analyzed on an agarose gel.

Uptake of cell-penetrating peptides

High performance liquid chromatography (Paper I-III)

Quantitative data of CPP uptake was obtained by using an HPLC coupled to

a fluorescence detector. The advantage of this method is that the degradation

pattern of each CPP can be examined and thus the half-lives can be calcu-

lated. In all studies relating to peptide uptake it is critically important to

32

separate membrane attached peptide from internalized peptide. To achieve

this, cells were treated, prior to cell lysis, with diazotized 2-nitroaniline,

which is a chemical agent that reacts with primary amino groups119

. Since

the diazotized 2-nitroaniline is a small molecule compared to the more

commonly used trypsin, it should lead to less over-estimation of internalized

peptide. Primary amino groups of lysine residues will attack the diazo com-

pound and form a highly unstable triazene which decomposes and the carbon

stretch of lysine transforms to either a ring structure or an alcohol. Thus the

modified peptide will become more hydrophobic and have a longer retention

time in the HPLC. Internalized and extracellular peptides are separated since

the reagent cannot cross the plasma membrane and modify internalized pep-

tides. The disadvantage with this method is that it requires lysine residues in

the peptide since these are the only amino acids with primary amino groups.

Intact and degraded peptides were identified in the HPLC by analyzing 10

pmol of the parent peptide and comparing the area under the curve.

Fluorescence activated cell sorter (Paper III-IV)

Fluorescence activated cell sorter (FACS) analysis is a sensitive tool to study

uptake of e.g. different macromolecules. Concurrently toxicity can be inves-

tigated either by adding dyes such as propidium iodide (PI), that fluoresces

when it binds DNA in cells with damaged plasma membrane, due to the

several fluorescence filters situated in the FACS, or by investigating if the

cell population is homogenous. Thus it is possible to investigate if the inter-

nalization of molecules is due to toxicity. In paper III FACS was used to

corroborate the uptake established by HPLC and fluorescence microscopy.

However, it is not possible to use the diazo method in the FACS studies;

therefore cells were treated with trypsin. In addition, trypan blue was added

to quench any membrane bound fluorescing peptide. In paper IV, FACS was

used to analyze the morphology of peptide-produced bacterial ghosts and

compare the morphology to live untreated bacteria. Moreover, FACS was

used to investigate the transduction of the ghosts into HeLa cells and to show

that GFP plasmids in the internalized bacterial ghosts were expressed.

Fluorescence microscopy (Paper III-IV)

Quantitative analysis of CPPs is important in order to estimate half-lives of

peptides and to investigate degradation and uptake kinetics. However, it does

not say anything about the localization inside the cell. In paper III fluores-

cence microscopy was used to investigate how the new peptides distributed

inside the cell. Both diffuse cytoplasmic staining of peptides was found to-

gether with endosomal vesicles. Confocal laser scanning microscopy

(CLSM) was used to photograph sections in the vertical plane of cells to

prove that the peptides did not reside in the plasma membrane but were actu-

33

ally inside the cells. Moreover, it was used to visualize mobilization of ly-

sosomes to the plasma membrane which would occur during the MRR. This

was done by detecting the lysosomal luminal membrane protein LAMP-2 on

the plasma membrane with an antibody indicating that the lysosomes had

translocated to the plasma membrane to donate their membrane, thus expos-

ing the luminal lysosomal membrane protein LAMP-2 on the surface of the

plasma membrane. Both CLSM and PI were used to confirm that the anti-

bodies were not inside the cell but on the plasma membrane and that the

membrane was intact. In paper IV uptake of GFP plasmid carrying ghosts

was visualized with fluorescence microscopy.

Electron microscopy (Paper III)

The high resolution of transmission electron microscopy (TEM) makes it

possible to deduce more specifically where inside the subcellular compart-

ments peptides are localized and how they are affecting the membrane. In

TEM a beam of electrons is transmitted through an ultrathin section of the

cell and recognizes different densities in the material. As described above

the peptides were labeled with 1.4 nm nanogold coupled via a maleimide

bond. Undecagold with a size of 0.8 nm was also investigated but without

success. Cells incubated with gold labeled peptides were fixed and prepared

by silver enhancement to be able to visualize the particles in TEM. After the

enhancement processes the cells were mounted into an epoxy resin followed

by thin sectioning with a diamond knife into approximately 40 nm thick

slices. The main drawback with TEM is the extensive sample preparation it

requires to produce a specimen thin enough to be electron transparent and

still able to withstand the high vacuum present inside the instrument.

Cytotoxicity assays

Lactate dehydrogenase leakage (Paper II-III)

Cytotoxicity can be determined by measuring the leakage of the enzyme

lactate dehydrogenase (LDH) which catalyzes the interconversion of lactate

to pyruvate at the same time as NAD+

is reduced to NADH. If the cell mem-

brane is compromised, leakage of LDH can be measured by the increase in

NADH which catalyzes the reaction of resazurin to resorufin which can be

measured in a fluorometer. Although LDH is a large molecule, the test has

been shown to correlate with leakage of smaller molecules such deoxyglu-

cose162,242

.

34

Calcium measurements (Paper III)

The MRR is activated by calcium influx and therefore calcium measure-

ments were performed in paper IV. The cell permeable compound FURA-2

acetoxymethyl (AM), which can enter cells in the form of an AM ester, was

used. Well inside the cell the ester is hydrolyzed, thus FURA-2 can not es-

cape the cell. A ratio is measured between unbound (380 nm) and calcium

bound (340 nm) FURA-2 with an emission of 520 nm. Even though there

was no leakage of LDH a clear concentration dependent increase of calcium

was seen with all CPPs. To test that not any peptide caused a calcium influx

two non-penetrating peptides were used as negative controls. Ionomycin was

used as positive control, which is a pore-forming calcium ionophore.

B-hexosaminidase leakage (Paper III)

If the plasma membrane is damaged lysosomes are mobilized by the MRR to

the site of disruption to patch the broken membrane. When the lysosomes

donate their membranes, lysosomal enzymes are released to the extracellular

milieu. One of those enzymes is β-hexosaminidase which activity can be

measured colorimetrically by the conversion of 4-methylumbelliferyl-2-

acetamido-2-deoxy-β-D-glucopyranoside to 4-methylumbelliferone.

35

Results and discussion

The aim of the thesis was to understand how and why CPPs internalize into

cells. Throughout the thesis two structurally different peptides, MAP and

penetratin, have been analyzed in order to be able to compare results. In

paper I, translocation of CPPs into non-mammalian cells were investigated

to understand the significance of the composition of different biological

membranes and the antimicrobial activity of the analyzed CPPs. The impor-

tance of peptide degradation both inside and outside cells was assessed in

both non-mammalian and mammalian cells in paper I and II, respectively.

The first two papers indicated that there might be a direct penetration me-

chanism involved in CPP uptake. Paper III investigates this possibility and

offers a possible explanation to why there is no leakage of endogenous mo-

lecules during this direct penetration. Finally, in paper IV, the antimicrobial

effects of CPPs were used to produce delivery vectors, known as bacterial

ghosts, to deliver plasmids.

Evaluation of CPP uptake and antimicrobial effects in non-mammalian cells (Paper I)

There are many similarities between CPPs and antimicrobial peptides as

aforementioned. Several antimicrobial peptides have been shown to translo-

cate across cell membranes214

. Therefore it was necessary to investigate if

CPPs showed any antimicrobial characteristics and if they could translocate

across non-mammalian plasma membranes. Four different non-mammalian

cell lines were assessed, namely B. megaterium (gram-positive bacteria), E.

coli (gram-negative bacteria), S. cerevisisae (yeast) and Sf9 cells (insect)

together with four CPPs of different origins. MAP, a model amphipathic,

non-protein-derived peptide, shares many characteristics with antimicrobial

peptides. Penetratin is a non-mammalian protein-derived peptide from Dro-

sophila, composed of both hydrophilic and hydrophobic patches, however

not as regularly spread as MAP. The third CPP, pVEC is derived from a

vascular endothelial mammalian protein cadherin (VE-cadherin), found in

blood vessels of murine and consist of a hydrophobic N terminus, shown to

be important in uptake, and a hydrophilic C terminus243

. In order to evaluate

36

the effect of degradation, a D-analog of pVEC was analyzed. To ascertain

that not any peptide could translocate across the cells, a sequence from the

VE-cadherin protein, was used as a negative control235

.

The antimicrobial effects of the CPPs were assessed by the broth microdi-

lution method. All peptides were antibacterial against gram-positive and

gram-negative bacteria but at different MICs. B. megaterium showed the

highest sensitivity with a MIC of 1 µM for each peptide. In E. coli, the MIC

was 25 µM for each peptide. No antimicrobial activity was found in S. cere-

visiae at the highest concentration tested (25 µM), indicating that the yeast

membrane is more resistant to peptide activity, followed by gram-negative

bacteria and gram-positive bacteria. Nevertheless the four CPPs translocated

all four cell lines, although with different efficacy and degradation patterns.

Interestingly, degradation of pVEC was only detected in E. coli and Sf9

cells. In B. megaterium and S. cerevisiae all CPPs were highly degraded

except the D-analog of pVEC. Moreover, in cells that could degrade pVEC

to a high degree, pVEC was also the most efficient CPP. In cells not degrad-

ing pVEC, MAP was most efficient. This indicated that the degradation of

the CPP is important in uptake. To investigate this further we compared the

uptake of pVEC with its D-analog. Both the ratio of intact pVEC versus the

D-analog and total uptake of pVEC versus the D-analog was compared. The

comparison showed that at low peptide concentrations, higher uptake of

pVEC was found and at high peptide concentrations, the D-analog was taken

up more efficiently, suggesting that two uptake mechanisms are involved.

The first mechanism, is dominating at low concentrations, where pVEC is

most efficient, thus dependent on chirality but with low capacity. The second

mechanism, is independent of chirality but has high capacity, dominating at

high concentrations, where the D-analog is most efficient.

In conclusion, the CPPs were antibacterial and the uptake and degradation

varied widely with pVEC illustrating the greatest difference in degradation

between the cell lines. The uptake mechanism dependent on chirality, sug-

gests the involvement of a transporter protein with high specificity and low

capacity. The other mechanism with high capacity, low specificity and inde-

pendent of chiralty, may be a direct penetration through the membrane.

37

Determining the importance of peptide degradation in CPP uptake (Paper II)

Due to the rapid degradation of CPPs in non-mammalian cells, it was natural

to continue to investigate the degradation of CPPs in mammalian cells. Most

in vitro and in vivo studies, have not taken into account, how quickly pep-

tides are degraded both inside and outside cells. If CPPs shall be used to

transport drugs or oligonucleotides in vivo, it is critically important to know

the extracellular half-life of the peptide. Therefore the degradation of the

structurally different CPPs, MAP and penetratin, was analyzed in CHO-K1

cells, extracellularly and intracellularly. An important difference, in addition

to amphipathicity and origin, is that MAP does not have any arginine resi-

dues, only lysines, whereas penetratin contains both. This is an important

feature because arginine residues are often given as a characteristic of CPPs.

Moreover, in this study detailed Fl-HPLC examination was used to obtain

quantitative data on uptake and degradation of CPPs. In all studies relating to

peptide uptake, it is critically important to separate membrane attached pep-

tide from internalized peptide. To achieve this, cells were treated prior to cell

lysis, with diazotized 2-nitroaniline, a chemical agent that primary amino

groups react with244,245

. After the modification, the peptide will be more hy-

drophobic and have a longer retention time in the HPLC. Internalized and

membrane bound peptides are separated since the reagent cannot cross the

plasma membrane and modify internalized peptides. However, it might be

possible that the detergents, used for cell lysis or components in the cell lys-

ates, form strong complexes that cannot be separated even in the HPLC.

These complexes may be eluted with differing retention times, compared to

the parent peptide, which would decrease the amount of the intact peptide

deceptively. Therefore intact diazotized labeled peptides were mixed with

cell lysate, to confirm that this was not the case.

The extracellular degradation of MAP and penetratin was rapid (Figure

8). Shorter degradation fragments of the CPPs could be found inside the cell

within minutes. Surprisingly, intact MAP and penetratin could still be found

inside the cell after one hour, even though the extracellular peptides should

have been degraded by that time, indicating that the extracellular degradation

might be determining the CPP uptake. This was corroborated by comparing

the curve corresponding to uptake, with the curve corresponding to the in-

crease in extracellular degradation products. Moreover, the fact that there

was degraded peptide within minutes inside the cell but intact peptide still

remaining after one hour, indicates that there might be at least two uptake

processes. One pathway resulting in intact peptide, possibly corresponding to

endocytosis, shielding the peptide from proteases at least until the peptide

reaches the lysosomes and another pathway that results in degraded peptides,

possibly corresponding to direct translocation.

38

Table 3. Protease and endocytosis inhibitors used in the study.

Name Target/Activity

Phenylmethylsulfonyl fluoride (PMSF)

Serine proteases

Phenanthroline

Matrix metalloproteinases (MMP)

Leupeptin

Serine and Cysteine proteases

Bacitracin

Most proteases

Bovine serum albumin (BSA )

Saturation of proteases

Cytochalasin B

Inhibiting the assembly of actin filaments

Wortmannin

Inhibits PI-3K, fusions of early endosomes

Nocodazole

Depolymerizing microtubules

NaN3/Deoxy-glucose

Energy depletion

Chloroquine

Endosomolytic reagent, prevents formation and maturation of endosomes

Next an attempt was made to decrease the extracellular degradation with

different protease inhibitors and substrates, to investigate if it influenced the

uptake (Table 3). Phenanthroline, which is a MMP inhibitor, decreased the

intra- and extracellular degradation of MAP and penetratin. A corresponding

increase in uptake was found for penetratin, indicating that TOP could be

involved in the intracellular degradation176

. On the contrary, the uptake of

MAP was decreased, which was later correlated to toxicity. MAP and phe-

nanthroline alone did not induce any LDH leakage but together they had a

synergistic effect that could lead to peptides leaking out of the cell, resulting

in a decreased uptake. Moreover, BSA was added to the peptide solution in

an attempt to saturate extracellular proteases and thereby increase the extra-

cellular half-lives of the peptides. BSA, together with MAP, resulted in more

intact peptide on the outside but surprisingly, there was no increase in up-

take. To investigate if this was a result of electrostatic interactions between

cationic MAP and anionic BSA, shielding the peptide from proteases but

preventing it from internalizing cells, MAP was preincubated with BSA or

bacitracin for thirty minutes before incubation with cells. Penetratin was

unaffected at the utilized concentrations but the uptake of MAP was reduced

by 50%, indicating that electrostatic interactions were involved. A similar

study with polyarginine peptides, found that the peptides bound to BSA,

which reduced the uptake246

. Since the data suggested both endocytosis and

direct translocation, the effect of several endocytosis inhibitors was eva-

39

luated (Table 3). Chloroquine, which in addition to buffering endosomes,

prevents maturation and formation of endosomal vesicles, wortmannin,

which inhibits PI3K thus, preventing the formation of early endosomes to

late endosomes, cytochalasin B, which inhibits the assembly of actin fila-

ments and thereby block intracellular transport and, nocodazole, which depo-

lymerizes microtubules, were analyzed. Chloroquine and cytochalasin B

decreased the uptake significantly, indicating that endocytosis is involved

and that it requires actin filaments. Nocodazole and wortmannin decreased

the uptake of penetratin, although not significantly at these concentrations.

In summary, the data indicated two uptake mechanisms, resulting in rapid

or delayed degradation of the peptide, which could be interpreted as direct

penetration versus endocytosis. Alternatively, the rapid degradation may

correspond to fast endocytosis, with an even faster endosomal escape. The

rapid extracellular degradation is an important parameter that should be tak-

en into account and determined the uptake at least in this system.

Figure 8. Uptake pathways and corresponding half-lives derived from Gepasi simu-lation program. The extracellular degradation half-lives of penetratin (Pen) and MAP were 5 and 10 min, respectively. Intact Pen and MAP were found inside the cell after about 2 and 6 min, respectively, whereas degraded peptides were found within about 4 and 3 min approximately. There was a delayed degradation of the peptides corresponding to 20 and 86 min for Pen and MAP, respectively.

40

The MRR masks membrane damages caused by CPPs (Paper III)

Our previous studies indicated that CPPs could translocate across the mem-

brane via a direct penetration mechanism. In paper II, the kinetics indicated

that one mechanism lead to almost instant degradation of the peptides, whe-

reas the other mechanism resulted in a delayed degradation. Therefore the

focus was on investigating if such a direct penetration mechanism existed,

how was it possible, and why was there no leakage of endogenous molecules

out of the cell? Moreover, if CPPs coupled to cargo are supposedly trapped

in endosomes, how can they induce a biological effect and if some cargo-

coupled CPPs leak out from the endosomes, why are not all CPPs able to

disrupt the endosomes? Since paper II suggested that there was a direct pe-

netration concurrently with endocytosis, perhaps the CPPs are trapped in the

endosomes and the biological effect comes from the fraction of peptides that

end up in the cytoplasm, after the direct penetration. With that in mind, we

read about the protein perforin, involved in target cell apoptosis, which is

exocytosed in granules together with granzymes and granulysin during target

cell apoptosis94

. Perforin, known to be a pore forming protein91

, caused ne-

crosis in cells at high concentrations, whereas at physiologically relevant

concentrations, it promoted granzymes to induce apoptosis. During the re-

lease of granzymes, it was found that perforin activated the MRR, which was

triggered by the calcium influx caused by the membrane disturbing effects of

perforin, resulting in translocation of intracellular vesicles to patch up the

membrane at the site of entry. Altogether, the characteristics and mechanism

of perforin, granzymes and granulysin resembled the actions of CPPs. Thus,

the MRR could be involved in CPP uptake.

Potential CPPs from the three native proteins were synthesized together

with two well-known CPPs, MAP and penetratin, to investigate if the pep-

tides alone were able to induce the MRR, as the native protein perforin. To

confirm that not any peptide could induce the MRR, two control peptides

from perforin were synthesized and utilized as negative controls. The novel

peptides translocated across cells although not as efficiently as MAP and

penetratin. Nevertheless, all peptides were found to induce calcium influx,

supporting the involvement of the MRR. Furthermore, the peptides caused

translocation of a lysosomal protein LAMP-2, as the protein perforin, from

inside the cell to the plasma membrane, corroborating the activity of the

MRR. The discovery in EM, that all CPPs were found in the cytoplasm with

and without vesicular structure surrounding them was an important finding

(Figure 9), as well as the insertion into the membrane without vesicle forma-

tion, corroborating that the direct penetration mechanism could exist.

41

Figure 9. EM photos of CPP distribution in HeLa cells. The black dots correspond to CPPs, coupled to 1.4 nm nanogold, magnified to approximately 8-12 nm with silver-enhancement. Penetratin (A, B), pGrB (C) and MAP (D) were found in both vesicular and non-vesicular structures. In some vesicles the conjugate was inserted into the endosomal membrane. Interestingly, penetratin seemed to affect the struc-ture of the nuclear membrane (B).

42

Production of bacterial ghosts for plasmid delivery by CPPs (Paper IV)

The information that some CPPs had an antimicrobial activity, found in pa-

per I, was used in this study. Instead of acting as a drug delivery vector, the

CPP MAP was used to produce a drug delivery vector by using a new strate-

gy. The novelty of the method was to use the antimicrobial activity of MAP,

to produce bacterial ghosts by disrupting the membrane of E. coli, causing

an empty bacterial envelope that could be used to deliver plasmids into cells.

Usually bacteria are loaded with a plasmid encoding lysis gene E, expressing

a 91 amino acid membrane protein that disrupts the membrane of the bacte-

rium, resulting in an empty shell that can be loaded with DNA and used for

delivery to other cells223,247

. In the new strategy, MAP was utilized to perfo-

rate the membrane and produce a ghost, containing the desired plasmid. To

further investigate the potential of the ghosts to act as delivery vehicles,

HeLa cells were incubated with the ghosts in an attempt to deliver the plas-

mid by simply letting the HeLa cells phagocytose or endocytose the ghosts.

Recently melanoma cells were shown to internalize ghosts containing DNA

via phagocytosis217

, with much higher efficacy than a non-liposomal lipid

transfection reagent, and it was suggested that the LPS of the ghosts, acti-

vated the melanoma cells through TLR-4 that is constitutively expressed in

these cells, resulting in an increased production of IL-8 and cell adhe-

sion229,248

.

Remarkably our strategy worked as the CPP MAP, was able to kill the

bacteria without distorting the structural features of the bacteria, which still

retained the plasmid after treatment. After optimization, a rapid method was

produced comprising incubation of bacteria with peptide for one hour in

buffer, significantly reducing the experimental steps and excluding lysis

gene E. The ghosts were then incubated with HeLa cells, followed by 45

hour incubation in complete medium and assessed by FACS. It is difficult to

explain how the DNA escapes the phagoendolysosome in which the bacterial

ghosts should be entrapped after internalization. However, it has been sug-

gested that the phosphatidylethanolamine, present in the membrane of the

ghosts, mediate the DNA escape as previously reported for dioleoylphospha-

tidylethanolamine delivery of DNA218,249

.

43

Conclusions

Bacterial ghosts and CPPs are two relatively new and exciting delivery ve-

hicles. However, peptides are believed to be rapidly degraded and only a few

studies have analyzed the degradation pattern of CPPs. Although CPPs have

been under investigation for about 15 years, the proposed mechanism has

gone from direct penetration to endocytosis and back again. Therefore one of

the purposes with the thesis was to study CPP entry in different organisms,

i.e. mammalian and non-mammalian cells to investigate how the difference

in membrane composition affected uptake. Furthermore, degradation of

CPPs in various cell lines was investigated to understand how it affected

internalization. The other aim was to design and evaluate a new peptide-

based strategy to produce bacterial ghosts, to be utilized as delivery vehicles

in gene delivery and DNA vaccination.

The conclusion of paper I and II were that CPPs indeed were rapidly de-

graded both in non-mammalian and mammalian cells and that it had a high

impact on the CPP uptake. While analyzing the difference in membrane

composition, the analyzed CPPs appeared to be antibacterial further corrobo-

rating the similarities between CPPs and AMPs. In addition, data suggested

that CPPs utilized at least two uptake pathways, which could correspond to

direct penetration and endocytosis in eukaryotic cells, and direct penetration

and transporter-mediated uptake in prokaryotic cells. With these results in

mind the MRR mechanism suddenly appeared and, as suggested in paper III,

it was the perfect explanation to the controversial direct penetration mechan-

ism. Furthermore, the conclusion that some CPPs were lytic against bacteria,

led to the development of the new CPP-based strategy to produce bacterial

ghost and as illustrated in paper IV, the ghosts proved to be successful deli-

very vehicles of plasmids.

44

Populärvetenskaplig sammanfattning på svenska

Många potentiella läkemedel för sjukdomar, som till exempel cancer, är av

vattenlöslig karaktär. Detta är ett problem om man har en aktiv substans som

behöver komma in några av kroppens celler, för att utöva sin effekt. Runt

cellen finns ett fettlösligt plasmamembran som skyddar cellen mot många

främmande substanser. En del stora, vattenlösliga molekyler kan tas upp

med hjälp av cellens olika transportsystem, men de flesta kan inte passera

membranet. Det är därför av stor vikt att hitta olika transportörer för dessa

vattenlösliga, potentiella läkemedel. Transportörernas uppgift är således att

ta sig in i cellen, genom plasmamebranet, tillsammans med den aktiva sub-

stansen och frigöra den inne i cellen, för att substansen ska kunna utöva sin

effekt. Det finns många varianter av dessa transportörer, varav två relativt

nya klasser undersöks och utvärderas i denna avhandling: cellpenetrerande

peptider och bakteriella spöken.

För 20 år sen upptäcktes det att vattenlösliga proteiner kunde ta sig in i

celler, genom plasmamembranet, för att sedan utföra olika funktioner inne i

cellen. Några år senare fann man att även kortare sekvenser av dessa protei-

ner tog sig in i cellerna. Dessa sekvenser döptes till cellpenetrerande peptider

och har använts för att transportera en mängd olika gensekvenser, proteiner

och substanser in i celler. I den här avhandlingen utvärderas och jämförs

både nya och gamla cellpenetrerande peptider. Om dessa peptider ska kunna

användas för att transportera läkemedel in i kroppen, måste hastigheten med

vilken de bryts ned utanför och inuti cellen undersökas. Annars hinner läke-

medlet inte transporteras in i cellen, innan peptiden bryts ned. Dessutom är

det viktigt att i detalj förstå, hur dessa peptider tar sig in i cellerna, för att

kunna göra dem ännu mer effektiva och kunna förutse bieffekter. Därför

undersöks både mammala och ickemammala celler för att förstå betydelsen

av sammansättning av olika plasmamembran. I avhandlingen presenteras

även en ny mekanism, med vilken membran repareras, som kan förklara hur

dessa peptider kan ta sig igenom cellmembranet utan att skada det.

När bakterier kommer in i kroppen, detekteras de av kroppens immunför-

svar, som har till uppgift att hitta och oskadliggöra bakterierna. Immunför-

svarets cirkulerande celler, känner igen bakterierna på grund av att de har

speciella sockermolekyler på plasmamebranet, så kallade lipopolysacchari-

der. När immunförsvarets celler åker runt i kroppen, känner de igen lipopo-

lysacchariderna och kan då binda till dessa, för att sedan äta upp bakterierna

och på så sätt oskadliggöra dem. Jämfört med mammala celler, är det lätt att

införa olika gensekvenser i bakterier, vilka sen kan uttryckas i bakterierna

till olika proteiner. Proteinerna kan sedan, likt läkemedel, ha olika funktioner

i cellen som motverkar sjukdomsförloppet. Alternativt kan gensekvenserna

användas för vaccinering.

Bakteriella spöken är döda bakterier som fortfarande har kvar sin ur-

sprungliga struktur, som till exempel lipopolysacchariderna på plasmamem-

45

branet. Detta gör att immunförsvaret fortfarande känner igen spökena, vilka

därför kan användas för att introducera gensekvenser i kroppen. Det finns

flera celler i kroppen som kan äta upp bakterierna och därför kan spökena

användas för att leverera gensekvenser, antingen för att uttrycka proteiner,

eller för vaccinering. I avhandlingen presenteras en ny metod att tillverka

dessa spöken. Metoden är snabbare och lättare än tidigare tillvägagångssätt

och visar att spöken med en specifik gensekvens, som tillverkats med hjälp

av den nya metoden, tas upp av mammala celler, vilka sedan uttrycker prote-

inet från gensekvensen som befann sig i spökena.

46

Acknowledgements

När jag mådde som sämst och mamma körde mig till Södertörn för sista gången

innan min sjukskrivning på nästan 3 år, trodde jag helt ärligt, inte att jag någonsin

skulle sitta här med en avhandling i handen. Därför går mitt största tack till Mia och

Inger som gjorde det möjligt för mig att sitta här idag. Tack Mia, för att du hjälpte

mig hitta tillbaka igen, jag tänker på dig ofta. Tack Inger för att du gjort livet så

mycket lättare. Jag hoppas vi alltid behåller kontakten.

Därefter vill jag självklart tacka min handledare Mattias, du har varit en väldigt

speciell handledare. De första ord jag sa till dig var, ”Jag har hört rykten om att du

behöver en doktorand”. Och du svarade, ”Dom rykten har jag också hört”. Dessa

repliker sammanfattar ganska bra våran relation och kommunikation under mina år

som doktorand. Du har gett mig så mycket frihet att jag ibland blandat ihop vem av

oss som är handledare. Jag vill tacka dig för att du gjort mig extremt självständig

och fått mig att lita på mig själv otroligt mycket, samt alla dina underbara och smar-

ta idéer som du förmedlat under åren. Men framförallt, för att du alltid stöttat mig i

mina beslut. Jag vill också tacka min bihandledare Ülo för att du granskat mig med

kritiska ögon genom åren, vilket betytt mycket för mig och bidragit stort till min

utveckling. Tack alla lärare, Anna för att du är en sån underbar människa, Kerstin,

för all din hjälp och Anders, dina föreläsningar är bäst. All administrativ personal för

allt ni hjälpt mig med, Marie-Louise för stipendietips, Ulla för att du alltid gläds

med mig, Birgitta, för att du fick vem som helst att känna sig hemma här på institu-

tionen och speciellt Siv för att du stöttat och pushat mig att bli klar i förtid.

Tack till alla doktorander som gör den här arbetsplatsen till ett så roligt ställe, jag

kommer att sakna er jätte mycket. Jag vill tacka supertjejerna på mitt rum, Johanna

och Veronica, för alla middagar, öl och pratstunder. Det har varit så tråkigt när ni

varit mammalediga. Maarja för att du är så underbar på alla sätt. Tina för att du

stöttat och lyssnat så många gånger. Karolina för att du är du, vi syns i USA! Samme

för att du hela tiden kommer in och babblar hela dagarna och uppdaterar om allt och

alla. Jessica du är en solstråle. Peter, du är en frisk fläkt och det har varit jätte roligt

att lära känna dig. Mats, jag saknar våra pratstunder. Henke, för att du är så rolig.

Malin och Linda L för era härliga skratt som har hörts genom korridorerna. Ulla och

Johan för att ni är en så lugna och coola. Yang, good luck in Oslo! Nina för att du är

så omtänksam. Anna-Lena för alla USA tips. Maria för att du alltid är så positiv och

hjälpsam. Katri, för att du hjälper till när som helst. Ett stort tack till Anders, Sandra,

Katarina, Jonas, Emelia, Per, Marie, Linda, Sofia, Kristin, Tom, Heléne, Helena för

alla skratt och pratstunder.

47

Tack till alla på Södertörn, speciellt Pontus och Peter, för att ni är såna sköna kil-

lar och för allt kul både på Södertörn och på Neurokemi. Ina, Lotta, Marianna och

Ullis för alla fester och roliga stunder. Tack Birgitta för att du gjorde min återkomst

till Södertörn så lätt och rolig.

Thank you, Margus, for being so positive and for a wonderful collaboration and,

for inviting me to come and work in your lab. Kärt, Annely, Pille, Heline and Kaida,

thank you for taking care of me and making my short stay in Tartu so much fun.

Thank you Steve for being so inspiring and for encouraging me to work 80 hours

per week and go out and socialize with all the professors, it really paid off!

Det är så väldigt många kompisar kvar att tacka… Mest av allt mina älsklingstje-

jer, Sara och Lotta, för att ni verkligen alltid finns och har funnits där, var ni än har

bott. Jag älskar er så!

Alla hoppkompisar från frifallsskolan för alla roliga fester och hopp (och golf!).

Alla golfkompisar som får komma och hälsa på i världens bästa golfstad: San Di-

ego!

Alla från AF, Lina (min gamla bästis, ska du börja golfa nu när jag ska åka?!),

Fredde, Danne, Jessica och från gymnasiet, speciellt Fredde (igen, som jag aldrig

verkar bli av med, inte ens efter 12 års pluggande) och världens roligaste PM och

Filip.

Jag vill tacka min familj för att ni orkat med mig under alla år jag var sjuk. Jag

kommer längta ihjäl mig efter er! Mormor och Günter jag kommer att sakna er och

Skåne så mycket. Lotta du är min älsklingsfaster och jag är så glad att du är frisk.

Tack Ia, du är världens bästa älsklingssyrra som alltid ställer upp. Pappa för att du är

världens gulligaste, gladaste och roligaste pappa, som älskar golf och förstår mycket

mer än jag någonsin anat. Mamma, min klippa, du har gett mig så mycket kraft. Att

du orkat gå igenom allt med mig är helt fantastiskt och jag kan inte med ord beskriva

hur tacksam jag är för att du ställt upp för mig hela tiden. Στιλιανό, είςαι ςτιν καρδιά

μοσ όλι τιν ώρα.

Slutligen vill jag tack den viktigaste personen i mitt liv, som skämmer bort mig med

så mycket kärlek varje dag. Den person som jag ser upp till mest av alla, som är helt

orädd, bara kan lämna sin firma och flytta till USA, bara för sin frus skull och är den

enda person som fått mig att skratta varje dag, sju år i sträck. Jannis, tack för att du

hela tiden visar hur underbart livet är och hur man njuter av det på mesta möjliga

sätt.

48

References

1. Singer, S. J. & Nicolson, G. L. The fluid mosaic model of the struc-

ture of cell membranes. Science 175, 720-31 (1972). 2. Singer, S. J. & Nicolson, G. L. The structure and chemistry of

mammalian cell membranes. Am J Pathol 65, 427-37 (1971). 3. Zhang, Y. M. & Rock, C. O. Membrane lipid homeostasis in bacte-

ria. Nat Rev Microbiol 6, 222-33 (2008). 4. Yamaguchi, T., Tamura, G. & Arima, K. Substructure of the cytop-

lasmic membrane of Bacillus megaterium. I. Method for the fractio-nation of "Ghosts". J Bacteriol 93, 483-9 (1967).

5. Vollmer, W., Blanot, D. & de Pedro, M. A. Peptidoglycan structure and architecture. FEMS Microbiol Rev 32, 149-67 (2008).

6. Park, J. T. & Uehara, T. How bacteria consume their own exoskele-tons (turnover and recycling of cell wall peptidoglycan). Microbiol Mol Biol Rev 72, 211-27, table of contents (2008).

7. Hayashi, S. & Wu, H. C. Lipoproteins in bacteria. J Bioenerg Bio-membr 22, 451-71 (1990).

8. Gronow, S. & Brade, H. Lipopolysaccharide biosynthesis: which steps do bacteria need to survive? J Endotoxin Res 7, 3-23 (2001).

9. Lesage, G. & Bussey, H. Cell wall assembly in Saccharomyces ce-revisiae. Microbiol Mol Biol Rev 70, 317-43 (2006).

10. van der Rest, M. E. et al. The plasma membrane of Saccharomyces cerevisiae: structure, function, and biogenesis. Microbiol Rev 59, 304-22 (1995).

11. Scherrer, R., Louden, L. & Gerhardt, P. Porosity of the yeast cell wall and membrane. J Bacteriol 118, 534-40 (1974).

12. Schnabl, M., Daum, G. & Pichler, H. Multiple lipid transport path-ways to the plasma membrane in yeast. Biochim Biophys Acta 1687, 130-40 (2005).

13. Esko, J. D. & Lindahl, U. Molecular diversity of heparan sulfate. J Clin Invest 108, 169-73 (2001).

14. Esko, J. D. & Selleck, S. B. Order out of chaos: assembly of ligand binding sites in heparan sulfate. Annu Rev Biochem 71, 435-71 (2002).

15. Bishop, J. R., Schuksz, M. & Esko, J. D. Heparan sulphate proteog-lycans fine-tune mammalian physiology. Nature 446, 1030-7 (2007).

49

16. Bülow, H. E. & Hobert, O. The molecular diversity of glycosami-noglycans shapes animal development. Annu Rev Cell Dev Biol 22, 375-407 (2006).

17. Belting, M. Heparan sulfate proteoglycan as a plasma membrane carrier. Trends Biochem Sci 28, 145-51 (2003).

18. Law, C. J., Maloney, P. C. & Wang, D. N. Ins and Outs of Major Facilitator Superfamily Antiporters. Annu Rev Microbiol (2008).

19. Davidson, A. L., Dassa, E., Orelle, C. & Chen, J. Structure, function, and evolution of bacterial ATP-binding cassette systems. Microbiol Mol Biol Rev 72, 317-64, table of contents (2008).

20. Detmers, F. J., Lanfermeijer, F. C. & Poolman, B. Peptides and ATP binding cassette peptide transporters. Res Microbiol 152, 245-58 (2001).

21. Biemans-Oldehinkel, E., Doeven, M. K. & Poolman, B. ABC trans-porter architecture and regulatory roles of accessory domains. FEBS Lett 580, 1023-35 (2006).

22. van der Heide, T. & Poolman, B. ABC transporters: one, two or four extracytoplasmic substrate-binding sites? EMBO Rep 3, 938-43 (2002).

23. Smith, S. G., Mahon, V., Lambert, M. A. & Fagan, R. P. A molecu-lar Swiss army knife: OmpA structure, function and expression. FEMS Microbiol Lett 273, 1-11 (2007).

24. Doeven, M. K., Kok, J. & Poolman, B. Specificity and selectivity determinants of peptide transport in Lactococcus lactis and other mi-croorganisms. Mol Microbiol 57, 640-9 (2005).

25. Lorca, G. L. et al. Transport capabilities of eleven gram-positive bacteria: comparative genomic analyses. Biochim Biophys Acta 1768, 1342-66 (2007).

26. Detmers, F. J. et al. Combinatorial peptide libraries reveal the li-gand-binding mechanism of the oligopeptide receptor OppA of Lac-tococcus lactis. Proc Natl Acad Sci U S A 97, 12487-92 (2000).

27. Hart, S. P., Dransfield, I. & Rossi, A. G. Phagocytosis of apoptotic cells. Methods 44, 280-5 (2008).

28. Walsh, G. M., Sexton, D. W., Blaylock, M. G. & Convery, C. M. Resting and cytokine-stimulated human small airway epithelial cells recognize and engulf apoptotic eosinophils. Blood 94, 2827-35 (1999).

29. Dini, L. et al. Phagocytosis of apoptotic bodies by liver endothelial cells. J Cell Sci 108 ( Pt 3), 967-73 (1995).

30. Groves, E., Dart, A. E., Covarelli, V. & Caron, E. Molecular me-chanisms of phagocytic uptake in mammalian cells. Cell Mol Life Sci 65, 1957-76 (2008).

31. Hall, A. & Nobes, C. D. Rho GTPases: molecular switches that con-trol the organization and dynamics of the actin cytoskeleton. Philos Trans R Soc Lond B Biol Sci 355, 965-70 (2000).

50

32. Jones, A. T. Macropinocytosis: searching for an endocytic identity and role in the uptake of cell penetrating peptides. J Cell Mol Med 11, 670-84 (2007).

33. Conner, S. D. & Schmid, S. L. Regulated portals of entry into the cell. Nature 422, 37-44 (2003).

34. Swanson, J. A. Shaping cups into phagosomes and macropinosomes. Nat Rev Mol Cell Biol 9, 639-49 (2008).

35. Norbury, C. C., Chambers, B. J., Prescott, A. R., Ljunggren, H. G. & Watts, C. Constitutive macropinocytosis allows TAP-dependent ma-jor histocompatibility complex class I presentation of exogenous so-luble antigen by bone marrow-derived dendritic cells. Eur J Immu-nol 27, 280-8 (1997).

36. Alpuche-Aranda, C. M., Racoosin, E. L., Swanson, J. A. & Miller, S. I. Salmonella stimulate macrophage macropinocytosis and persist within spacious phagosomes. J Exp Med 179, 601-8 (1994).

37. Watarai, M. et al. Legionella pneumophila is internalized by a ma-cropinocytotic uptake pathway controlled by the Dot/Icm system and the mouse Lgn1 locus. J Exp Med 194, 1081-96 (2001).

38. Hewlett, L. J., Prescott, A. R. & Watts, C. The coated pit and ma-cropinocytic pathways serve distinct endosome populations. J Cell Biol 124, 689-703 (1994).

39. Ungewickell, E. J. & Hinrichsen, L. Endocytosis: clathrin-mediated membrane budding. Curr Opin Cell Biol 19, 417-25 (2007).

40. Schmid, S. L. Clathrin-coated vesicle formation and protein sorting: an integrated process. Annu Rev Biochem 66, 511-48 (1997).

41. Brodsky, F. M., Chen, C. Y., Knuehl, C., Towler, M. C. & Wake-ham, D. E. Biological basket weaving: formation and function of clathrin-coated vesicles. Annu Rev Cell Dev Biol 17, 517-68 (2001).

42. Roth, T. F. & Porter, K. R. Yolk Protein Uptake in the Oocyte of the Mosquito Aedes Aegypti. L. J Cell Biol 20, 313-32 (1964).

43. Heuser, J. Three-dimensional visualization of coated vesicle forma-tion in fibroblasts. J Cell Biol 84, 560-83 (1980).

44. Conner, S. D. & Schmid, S. L. Identification of an adaptor-associated kinase, AAK1, as a regulator of clathrin-mediated endo-cytosis. J Cell Biol 156, 921-9 (2002).

45. Conner, S. D., Schroter, T. & Schmid, S. L. AAK1-mediated micro2 phosphorylation is stimulated by assembled clathrin. Traffic 4, 885-90 (2003).

46. Hannan, L. A., Newmyer, S. L. & Schmid, S. L. ATP- and cytosol-dependent release of adaptor proteins from clathrin-coated vesicles: A dual role for Hsc70. Mol Biol Cell 9, 2217-29 (1998).

47. Newmyer, S. L. & Schmid, S. L. Dominant-interfering Hsc70 mu-tants disrupt multiple stages of the clathrin-coated vesicle cycle in vivo. J Cell Biol 152, 607-20 (2001).

48. Conner, S. D. & Schmid, S. L. Differential requirements for AP-2 in clathrin-mediated endocytosis. J Cell Biol 162, 773-9 (2003).

51

49. Palade, G. Intracellular Aspects of the Process of Protein Synthesis. Science 189, 867 (1975).

50. Nichols, B. J. & Lippincott-Schwartz, J. Endocytosis without clath-rin coats. Trends Cell Biol 11, 406-12 (2001).

51. Lajoie, P. & Nabi, I. R. Regulation of raft-dependent endocytosis. J Cell Mol Med 11, 644-53 (2007).

52. Fra, A. M., Williamson, E., Simons, K. & Parton, R. G. De novo formation of caveolae in lymphocytes by expression of VIP21-caveolin. Proc Natl Acad Sci U S A 92, 8655-9 (1995).

53. Scherer, P. E. et al. Cell-type and tissue-specific expression of cave-olin-2. Caveolins 1 and 2 co-localize and form a stable hetero-oligomeric complex in vivo. J Biol Chem 272, 29337-46 (1997).

54. Lahtinen, U., Honsho, M., Parton, R. G., Simons, K. & Verkade, P. Involvement of caveolin-2 in caveolar biogenesis in MDCK cells. FEBS Lett 538, 85-8 (2003).

55. Razani, B. et al. Caveolin-2-deficient mice show evidence of severe pulmonary dysfunction without disruption of caveolae. Mol Cell Bi-ol 22, 2329-44 (2002).

56. Way, M. & Parton, R. G. M-caveolin, a muscle-specific caveolin-related protein. FEBS Lett 376, 108-12 (1995).

57. Thomsen, P., Roepstorff, K., Stahlhut, M. & van Deurs, B. Caveolae are highly immobile plasma membrane microdomains, which are not involved in constitutive endocytic trafficking. Mol Biol Cell 13, 238-50 (2002).

58. Tagawa, A. et al. Assembly and trafficking of caveolar domains in the cell: caveolae as stable, cargo-triggered, vesicular transporters. J Cell Biol 170, 769-79 (2005).

59. Pelkmans, L., Kartenbeck, J. & Helenius, A. Caveolar endocytosis of simian virus 40 reveals a new two-step vesicular-transport path-way to the ER. Nat Cell Biol 3, 473-83 (2001).

60. Damm, E. M. et al. Clathrin- and caveolin-1-independent endocyto-sis: entry of simian virus 40 into cells devoid of caveolae. J Cell Biol 168, 477-88 (2005).

61. Kirkham, M. et al. Ultrastructural identification of uncoated cave-olin-independent early endocytic vehicles. J Cell Biol 168, 465-76 (2005).

62. Lamaze, C. et al. Interleukin 2 receptors and detergent-resistant membrane domains define a clathrin-independent endocytic path-way. Mol Cell 7, 661-71 (2001).

63. Sabharanjak, S., Sharma, P., Parton, R. G. & Mayor, S. GPI-anchored proteins are delivered to recycling endosomes via a distinct cdc42-regulated, clathrin-independent pinocytic pathway. Dev Cell 2, 411-23 (2002).

64. Glebov, O. O., Bright, N. A. & Nichols, B. J. Flotillin-1 defines a clathrin-independent endocytic pathway in mammalian cells. Nat Cell Biol 8, 46-54 (2006).

52

65. Xu, X. et al. Effect of the structure of natural sterols and sphingoli-pids on the formation of ordered sphingolipid/sterol domains (rafts). Comparison of cholesterol to plant, fungal, and disease-associated sterols and comparison of sphingomyelin, cerebrosides, and cera-mide. J Biol Chem 276, 33540-6 (2001).

66. McNeil, P. L. & Ito, S. Molecular traffic through plasma membrane disruptions of cells in vivo. J Cell Sci 96 ( Pt 3), 549-56 (1990).

67. McNeil, P. L. & Khakee, R. Disruptions of muscle fiber plasma membranes. Role in exercise-induced damage. Am J Pathol 140, 1097-109 (1992).

68. Clarke, M. S., Caldwell, R. W., Chiao, H., Miyake, K. & McNeil, P. L. Contraction-induced cell wounding and release of fibroblast growth factor in heart. Circ Res 76, 927-34 (1995).

69. McNeil, P. L. & Kirchhausen, T. An emergency response team for membrane repair. Nat Rev Mol Cell Biol 6, 499-505 (2005).

70. McNeil, P. L. & Ito, S. Gastrointestinal cell plasma membrane wounding and resealing in vivo. Gastroenterology 96, 1238-48 (1989).

71. McNeil, P. L. & Steinhardt, R. A. Plasma membrane disruption: repair, prevention, adaptation. Annu Rev Cell Dev Biol 19, 697-731 (2003).

72. Steinhardt, R. A., Bi, G. & Alderton, J. M. Cell membrane resealing by a vesicular mechanism similar to neurotransmitter release. Science 263, 390-3 (1994).

73. Bi, G. Q., Alderton, J. M. & Steinhardt, R. A. Calcium-regulated exocytosis is required for cell membrane resealing. J Cell Biol 131, 1747-58 (1995).

74. Llinas, R., Gruner, J. A., Sugimori, M., McGuinness, T. L. & Greengard, P. Regulation by synapsin I and Ca(2+)-calmodulin-dependent protein kinase II of the transmitter release in squid giant synapse. J Physiol 436, 257-82 (1991).

75. Söllner, T. & Rothman, J. E. Neurotransmission: harnessing fusion machinery at the synapse. Trends Neurosci 17, 344-8 (1994).

76. Rizo, J. & Rosenmund, C. Synaptic vesicle fusion. Nat Struct Mol Biol 15, 665-74 (2008).

77. Miyake, K. & McNeil, P. L. Vesicle accumulation and exocytosis at sites of plasma membrane disruption. J Cell Biol 131, 1737-45 (1995).

78. Bi, G. Q. et al. Kinesin- and myosin-driven steps of vesicle recruit-ment for Ca2+-regulated exocytosis. J Cell Biol 138, 999-1008 (1997).

79. Mellgren, R. L., Zhang, W., Miyake, K. & McNeil, P. L. Calpain is required for the rapid, calcium-dependent repair of wounded plasma membrane. J Biol Chem 282, 2567-75 (2007).

80. Jacobson, M. D., Weil, M. & Raff, M. C. Programmed cell death in animal development. Cell 88, 347-54 (1997).

53

81. Cullen, S. P. & Martin, S. J. Mechanisms of granule-dependent kill-ing. Cell Death Differ 15, 251-62 (2008).

82. Lieberman, J. The ABCs of granule-mediated cytotoxicity: new weapons in the arsenal. Nat Rev Immunol 3, 361-70 (2003).

83. Ernst, W. A. et al. Granulysin, a T cell product, kills bacteria by altering membrane permeability. J Immunol 165, 7102-8 (2000).

84. Voskoboinik, I., Smyth, M. J. & Trapani, J. A. Perforin-mediated target-cell death and immune homeostasis. Nat Rev Immunol 6, 940-52 (2006).

85. Metkar, S. S. et al. Cytotoxic cell granule-mediated apoptosis: perfo-rin delivers granzyme B-serglycin complexes into target cells with-out plasma membrane pore formation. Immunity 16, 417-28 (2002).

86. Raja, S. M. et al. Cytotoxic cell granule-mediated apoptosis. Charac-terization of the macromolecular complex of granzyme B with ser-glycin. J Biol Chem 277, 49523-30 (2002).

87. Brunner, G., Simon, M. M. & Kramer, M. D. Activation of pro-urokinase by the human T cell-associated serine proteinase HuTSP-1. FEBS Lett 260, 141-4 (1990).

88. Lieberman, J. & Fan, Z. Nuclear war: the granzyme A-bomb. Curr Opin Immunol 15, 553-9 (2003).

89. Trapani, J. A. & Sutton, V. R. Granzyme B: pro-apoptotic, antiviral and antitumor functions. Curr Opin Immunol 15, 533-43 (2003).

90. Peters, P. J. et al. Cytotoxic T lymphocyte granules are secretory lysosomes, containing both perforin and granzymes. J Exp Med 173, 1099-109 (1991).

91. Liu, C. C., Walsh, C. M. & Young, J. D. Perforin: structure and function. Immunol Today 16, 194-201 (1995).

92. Bolitho, P., Voskoboinik, I., Trapani, J. A. & Smyth, M. J. Apopto-sis induced by the lymphocyte effector molecule perforin. Curr Opin Immunol 19, 339-47 (2007).

93. Voskoboinik, I. et al. Calcium-dependent plasma membrane binding and cell lysis by perforin are mediated through its C2 domain: A critical role for aspartate residues 429, 435, 483, and 485 but not 491. J Biol Chem 280, 8426-34 (2005).

94. Keefe, D. et al. Perforin triggers a plasma membrane-repair response that facilitates CTL induction of apoptosis. Immunity 23, 249-62 (2005).

95. Romero, V. & Andrade, F. Non-apoptotic functions of granzymes. Tissue Antigens 71, 409-16 (2008).

96. Sayers, T. J. et al. Purification of a factor from the granules of a rat natural killer cell line (RNK) that reduces tumor cell growth and changes tumor morphology. Molecular identity with a granule serine protease (RNKP-1). J Immunol 148, 292-300 (1992).

97. Raja, S. M., Metkar, S. S. & Froelich, C. J. Cytotoxic granule-mediated apoptosis: unraveling the complex mechanism. Curr Opin Immunol 15, 528-32 (2003).

54

98. Raja, S. M. et al. A novel mechanism for protein delivery: granzyme B undergoes electrostatic exchange from serglycin to target cells. J Biol Chem 280, 20752-61 (2005).

99. Bird, C. H. et al. Cationic sites on granzyme B contribute to cytotox-icity by promoting its uptake into target cells. Mol Cell Biol 25, 7854-67 (2005).

100. Pena, S. V., Hanson, D. A., Carr, B. A., Goralski, T. J. & Krensky, A. M. Processing, subcellular localization, and function of 519 (gra-nulysin), a human late T cell activation molecule with homology to small, lytic, granule proteins. J Immunol 158, 2680-8 (1997).

101. Pena, S. V. & Krensky, A. M. Granulysin, a new human cytolytic granule-associated protein with possible involvement in cell-mediated cytotoxicity. Semin Immunol 9, 117-25 (1997).

102. Kaspar, A. A. et al. A distinct pathway of cell-mediated apoptosis initiated by granulysin. J Immunol 167, 350-6 (2001).

103. Li, Q. et al. Hemolysis of erythrocytes by granulysin-derived pep-tides but not by granulysin. Antimicrob Agents Chemother 49, 388-97 (2005).

104. Wang, Z. et al. Bactericidal and tumoricidal activities of synthetic peptides derived from granulysin. J Immunol 165, 1486-90 (2000).

105. Pardo, J. et al. A role of the mitochondrial apoptosis-inducing factor in granulysin-induced apoptosis. J Immunol 167, 1222-9 (2001).

106. Okada, S., Li, Q., Whitin, J. C., Clayberger, C. & Krensky, A. M. Intracellular mediators of granulysin-induced cell death. J Immunol 171, 2556-62 (2003).

107. Eichhorn, M. E. et al. Cationic lipid complexed camptothecin (En-doTAG-2) improves antitumoral efficacy by tumor vascular target-ing. Cancer Biol Ther 6, 920-9 (2007).

108. Kam, N. W., Liu, Z. & Dai, H. Carbon nanotubes as intracellular transporters for proteins and DNA: an investigation of the uptake mechanism and pathway. Angew Chem Int Ed Engl 45, 577-81 (2006).

109. Asane, G. S. et al. Polymers for Mucoadhesive Drug Delivery Sys-tem: A Current Status. Drug Dev Ind Pharm, 1-21 (2008).

110. Frankel, A. D. & Pabo, C. O. Cellular uptake of the tat protein from human immunodeficiency virus. Cell 55, 1189-93 (1988).

111. Derossi, D., Joliot, A. H., Chassaing, G. & Prochiantz, A. The third helix of the Antennapedia homeodomain translocates through bio-logical membranes. J Biol Chem 269, 10444-50 (1994).

112. Vives, E., Brodin, P. & Lebleu, B. A truncated HIV-1 Tat protein basic domain rapidly translocates through the plasma membrane and accumulates in the cell nucleus. J Biol Chem 272, 16010-7 (1997).

113. Elmquist, A., Lindgren, M., Bartfai, T. & Langel, Ü. VE-cadherin-derived cell-penetrating peptide, pVEC, with carrier functions. Exp Cell Res 269, 237-44 (2001).

55

114. Magzoub, M. et al. N-terminal peptides from unprocessed prion proteins enter cells by macropinocytosis. Biochem Biophys Res Commun 348, 379-85 (2006).

115. Morris, M. C., Vidal, P., Chaloin, L., Heitz, F. & Divita, G. A new peptide vector for efficient delivery of oligonucleotides into mam-malian cells. Nucleic Acids Res 25, 2730-6 (1997).

116. Morris, M. C., Depollier, J., Mery, J., Heitz, F. & Divita, G. A pep-tide carrier for the delivery of biologically active proteins into mammalian cells. Nat Biotechnol 19, 1173-6 (2001).

117. Pooga, M., Hällbrink, M., Zorko, M. & Langel, Ü. Cell penetration by transportan. Faseb J 12, 67-77 (1998).

118. Soomets, U. et al. Deletion analogues of transportan. Biochim Bio-phys Acta 1467, 165-76 (2000).

119. Oehlke, J. et al. Cellular uptake of an alpha-helical amphipathic model peptide with the potential to deliver polar compounds into the cell interior non-endocytically. Biochim Biophys Acta 1414, 127-39 (1998).

120. Wender, P. A. et al. The design, synthesis, and evaluation of mole-cules that enable or enhance cellular uptake: peptoid molecular transporters. Proc Natl Acad Sci U S A 97, 13003-8 (2000).

121. Green, M. & Loewenstein, P. M. Autonomous functional domains of chemically synthesized human immunodeficiency virus tat trans-activator protein. Cell 55, 1179-88 (1988).

122. Joliot, A., Pernelle, C., Deagostini-Bazin, H. & Prochiantz, A. An-tennapedia homeobox peptide regulates neural morphogenesis. Proc Natl Acad Sci U S A 88, 1864-8 (1991).

123. Rothbard, J. B., Jessop, T. C., Lewis, R. S., Murray, B. A. & Wend-er, P. A. Role of membrane potential and hydrogen bonding in the mechanism of translocation of guanidinium-rich peptides into cells. J Am Chem Soc 126, 9506-7 (2004).

124. Nakase, I. et al. Interaction of arginine-rich peptides with mem-brane-associated proteoglycans is crucial for induction of actin or-ganization and macropinocytosis. Biochemistry 46, 492-501 (2007).

125. Futaki, S. et al. Arginine-rich peptides. An abundant source of membrane-permeable peptides having potential as carriers for intra-cellular protein delivery. J Biol Chem 276, 5836-40 (2001).

126. Kerkis, A. et al. Crotamine is a novel cell-penetrating protein from the venom of rattlesnake Crotalus durissus terrificus. Faseb J 18, 1407-9 (2004).

127. Scheller, A. et al. Structural requirements for cellular uptake of al-pha-helical amphipathic peptides. J Pept Sci 5, 185-94 (1999).

128. Magzoub, M., Kilk, K., Eriksson, L. E., Langel, Ü. & Gräslund, A. Interaction and structure induction of cell-penetrating peptides in the presence of phospholipid vesicles. Biochim Biophys Acta 1512, 77-89 (2001).

129. Berlose, J. P., Convert, O., Derossi, D., Brunissen, A. & Chassaing, G. Conformational and associative behaviours of the third helix of

56

antennapedia homeodomain in membrane-mimetic environments. Eur J Biochem 242, 372-86 (1996).

130. Fischer, P. M. et al. Structure-activity relationship of truncated and substituted analogues of the intracellular delivery vector Penetratin. J Pept Res 55, 163-72 (2000).

131. Goun, E. A., Pillow, T. H., Jones, L. R., Rothbard, J. B. & Wender, P. A. Molecular transporters: synthesis of oligoguanidinium trans-porters and their application to drug delivery and real-time imaging. Chembiochem 7, 1497-515 (2006).

132. Wender, P. A., Galliher, W. C., Goun, E. A., Jones, L. R. & Pillow, T. H. The design of guanidinium-rich transporters and their interna-lization mechanisms. Adv Drug Deliv Rev 60, 452-72 (2008).

133. Abes, S. et al. Vectorization of morpholino oligomers by the (R-Ahx-R)4 peptide allows efficient splicing correction in the absence of endosomolytic agents. J Control Release 116, 304-13 (2006).

134. Vidal, P., Morris, M. C., Chaloin, L., Heitz, F. & Divita, G. [New strategy for RNA vectorization in mammalian cells. Use of a peptide vector]. C R Acad Sci III 320, 279-87 (1997).

135. Morris, M. C., Chaloin, L., Mery, J., Heitz, F. & Divita, G. A novel potent strategy for gene delivery using a single peptide vector as a carrier. Nucleic Acids Res 27, 3510-7 (1999).

136. Simeoni, F., Morris, M. C., Heitz, F. & Divita, G. Insight into the mechanism of the peptide-based gene delivery system MPG: impli-cations for delivery of siRNA into mammalian cells. Nucleic Acids Res 31, 2717-24 (2003).

137. Deshayes, S. et al. Primary amphipathic cell-penetrating peptides: structural requirements and interactions with model membranes. Bi-ochemistry 43, 7698-706 (2004).

138. Mann, D. A. & Frankel, A. D. Endocytosis and targeting of exogen-ous HIV-1 Tat protein. Embo J 10, 1733-9 (1991).

139. Prochiantz, A. Messenger proteins: homeoproteins, TAT and others. Curr Opin Cell Biol 12, 400-6 (2000).

140. Pokorny, A. & Almeida, P. F. Kinetics of dye efflux and lipid flip-flop induced by delta-lysin in phosphatidylcholine vesicles and the mechanism of graded release by amphipathic, alpha-helical peptides. Biochemistry 43, 8846-57 (2004).

141. Richard, J. P. et al. Cell-penetrating peptides. A reevaluation of the mechanism of cellular uptake. J Biol Chem 278, 585-90 (2003).

142. Pokorny, A., Birkbeck, T. H. & Almeida, P. F. Mechanism and ki-netics of delta-lysin interaction with phospholipid vesicles. Bioche-mistry 41, 11044-56 (2002).

143. Violini, S., Sharma, V., Prior, J. L., Dyszlewski, M. & Piwnica-Worms, D. Evidence for a plasma membrane-mediated permeability barrier to Tat basic domain in well-differentiated epithelial cells: lack of correlation with heparan sulfate. Biochemistry 41, 12652-61 (2002).

57

144. Gump, J. M. & Dowdy, S. F. TAT transduction: the molecular me-chanism and therapeutic prospects. Trends Mol Med 13, 443-8 (2007).

145. Kaplan, I. M., Wadia, J. S. & Dowdy, S. F. Cationic TAT peptide transduction domain enters cells by macropinocytosis. J Control Re-lease 102, 247-53 (2005).

146. Al-Taei, S. et al. Intracellular traffic and fate of protein transduction domains HIV-1 TAT peptide and octaarginine. Implications for their utilization as drug delivery vectors. Bioconjug Chem 17, 90-100 (2006).

147. Richard, J. P. et al. Cellular uptake of unconjugated TAT peptide involves clathrin-dependent endocytosis and heparan sulfate recep-tors. J Biol Chem 280, 15300-6 (2005).

148. Fittipaldi, A. et al. Cell membrane lipid rafts mediate caveolar endo-cytosis of HIV-1 Tat fusion proteins. J Biol Chem 278, 34141-9 (2003).

149. Ferrari, A. et al. Caveolae-mediated internalization of extracellular HIV-1 tat fusion proteins visualized in real time. Mol Ther 8, 284-94 (2003).

150. Fittipaldi, A. & Giacca, M. Transcellular protein transduction using the Tat protein of HIV-1. Adv Drug Deliv Rev 57, 597-608 (2005).

151. Duchardt, F., Fotin-Mleczek, M., Schwarz, H., Fischer, R. & Brock, R. A comprehensive model for the cellular uptake of cationic cell-penetrating peptides. Traffic 8, 848-66 (2007).

152. Padari, K. et al. Cell transduction pathways of transportans. Biocon-jug Chem 16, 1399-410 (2005).

153. Fretz, M. M. et al. Temperature-, concentration- and cholesterol-dependent translocation of L- and D-octa-arginine across the plasma and nuclear membrane of CD34+ leukaemia cells. Biochem J 403, 335-42 (2007).

154. Tünnemann, G. et al. Cargo-dependent mode of uptake and bioavai-lability of TAT-containing proteins and peptides in living cells. Fa-seb J 20, 1775-84 (2006).

155. Herce, H. D. & Garcia, A. E. Molecular dynamics simulations sug-gest a mechanism for translocation of the HIV-1 TAT peptide across lipid membranes. Proc Natl Acad Sci U S A 104, 20805-10 (2007).

156. Deshayes, S. et al. Insight into the mechanism of internalization of the cell-penetrating carrier peptide Pep-1 through conformational analysis. Biochemistry 43, 1449-57 (2004).

157. Gerbal-Chaloin, S. et al. First step of the cell-penetrating peptide mechanism involves Rac1 GTPase-dependent actin-network remo-delling. Biol Cell 99, 223-38 (2007).

158. Andersson, A., Danielsson, J., Gräslund, A. & Mäler, L. Kinetic models for peptide-induced leakage from vesicles and cells. Eur Bi-ophys J 36, 621-35 (2007).

159. Kabouridis, P. S. Biological applications of protein transduction technology. Trends Biotechnol 21, 498-503 (2003).

58

160. Snyder, E. L. & Dowdy, S. F. Cell penetrating peptides in drug deli-very. Pharm Res 21, 389-93 (2004).

161. Trehin, R. & Merkle, H. P. Chances and pitfalls of cell penetrating peptides for cellular drug delivery. Eur J Pharm Biopharm 58, 209-23 (2004).

162. Johansson, H. J. et al. Characterization of a novel cytotoxic cell-penetrating peptide derived from p14ARF protein. Mol Ther 16, 115-23 (2008).

163. Yin, H. et al. Cell-penetrating peptide-conjugated antisense oligo-nucleotides restore systemic muscle and cardiac dystrophin expres-sion and function. Hum Mol Genet (2008).

164. EL Andaloussi, S., Johansson, H. J., Lundberg, P. & Langel, Ü. In-duction of splice correction by cell-penetrating peptide nucleic acids. J Gene Med 8, 1262-73 (2006).

165. Jones, L. R. et al. Releasable luciferin-transporter conjugates: tools for the real-time analysis of cellular uptake and release. J Am Chem Soc 128, 6526-7 (2006).

166. Wender, P. A. et al. Real-time analysis of uptake and bioactivatable cleavage of luciferin-transporter conjugates in transgenic reporter mice. Proc Natl Acad Sci U S A 104, 10340-5 (2007).

167. Wadia, J. S., Stan, R. V. & Dowdy, S. F. Transducible TAT-HA fusogenic peptide enhances escape of TAT-fusion proteins after li-pid raft macropinocytosis. Nat Med 10, 310-5 (2004).

168. Lundberg, P., EL Andaloussi, S., Sütlü, T., Johansson, H. & Langel, Ü. Delivery of short interfering RNA using endosomolytic cell-penetrating peptides. Faseb J 21, 2664-71 (2007).

169. Morris, M. C. et al. A non-covalent peptide-based carrier for in vivo delivery of DNA mimics. Nucleic Acids Res 35, e49 (2007).

170. Torchilin, V. P., Rammohan, R., Weissig, V. & Levchenko, T. S. TAT peptide on the surface of liposomes affords their efficient intracellular delivery even at low temperature and in the presence of metabolic inhibitors. Proc Natl Acad Sci U S A 98, 8786-91 (2001).

171. Gupta, B., Levchenko, T. S. & Torchilin, V. P. TAT peptide-modified liposomes provide enhanced gene delivery to intracranial human brain tumor xenografts in nude mice. Oncol Res 16, 351-9 (2007).

172. Torchilin, V. P. Tat peptide-mediated intracellular delivery of phar-maceutical nanocarriers. Adv Drug Deliv Rev 60, 548-58 (2008).

173. Meade, B. R. & Dowdy, S. F. Enhancing the cellular uptake of siR-NA duplexes following noncovalent packaging with protein trans-duction domain peptides. Adv Drug Deliv Rev 60, 530-6 (2008).

174. Deshayes, S., Morris, M., Heitz, F. & Divita, G. Delivery of proteins and nucleic acids using a non-covalent peptide-based strategy. Adv Drug Deliv Rev 60, 537-47 (2008).

175. Vives, E., Schmidt, J. & Pelegrin, A. Cell-penetrating and cell-targeting peptides in drug delivery. Biochim Biophys Acta (2008).

59

176. Saric, T., Graef, C. I. & Goldberg, A. L. Pathway for degradation of peptides generated by proteasomes: a key role for thimet oligopepti-dase and other metallopeptidases. J Biol Chem 279, 46723-32 (2004).

177. Geier, E. et al. A giant protease with potential to substitute for some functions of the proteasome. Science 283, 978-81 (1999).

178. Beninga, J., Rock, K. L. & Goldberg, A. L. Interferon-gamma can stimulate post-proteasomal trimming of the N terminus of an anti-genic peptide by inducing leucine aminopeptidase. J Biol Chem 273, 18734-42 (1998).

179. Mo, X. Y., Cascio, P., Lemerise, K., Goldberg, A. L. & Rock, K. Distinct proteolytic processes generate the C and N termini of MHC class I-binding peptides. J Immunol 163, 5851-9 (1999).

180. Reits, E. et al. A major role for TPPII in trimming proteasomal de-gradation products for MHC class I antigen presentation. Immunity 20, 495-506 (2004).

181. Neijssen, J. et al. Cross-presentation by intercellular peptide transfer through gap junctions. Nature 434, 83-8 (2005).

182. Tomkinson, B. & Lindas, A. C. Tripeptidyl-peptidase II: a multi-purpose peptidase. Int J Biochem Cell Biol 37, 1933-7 (2005).

183. Conlin, C. A. & Miller, C. G. Dipeptidyl carboxypeptidase and oli-gopeptidase A from Escherichia coli and Salmonella typhimurium. Methods Enzymol 248, 567-79 (1995).

184. Buchler, M., Tisljar, U. & Wolf, D. H. Proteinase yscD (oligopepti-dase yscD). Structure, function and relationship of the yeast enzyme with mammalian thimet oligopeptidase (metalloendopeptidase, EP 24.15). Eur J Biochem 219, 627-39 (1994).

185. Roelse, J., Gromme, M., Momburg, F., Hammerling, G. & Neefjes, J. Trimming of TAP-translocated peptides in the endoplasmic reticu-lum and in the cytosol during recycling. J Exp Med 180, 1591-7 (1994).

186. Reits, E. A., Vos, J. C., Gromme, M. & Neefjes, J. The major sub-strates for TAP in vivo are derived from newly synthesized proteins. Nature 404, 774-8 (2000).

187. Reits, E. et al. Peptide diffusion, protection, and degradation in nuc-lear and cytoplasmic compartments before antigen presentation by MHC class I. Immunity 18, 97-108 (2003).

188. Van Wart, H. E. & Birkedal-Hansen, H. The cysteine switch: a prin-ciple of regulation of metalloproteinase activity with potential appli-cability to the entire matrix metalloproteinase gene family. Proc Natl Acad Sci U S A 87, 5578-82 (1990).

189. Jongeneel, C. V., Bouvier, J. & Bairoch, A. A unique signature iden-tifies a family of zinc-dependent metallopeptidases. FEBS Lett 242, 211-4 (1989).

190. Vallee, B. L. & Auld, D. S. Short and long spacer sequences and other structural features of zinc binding sites in zinc enzymes. FEBS Lett 257, 138-40 (1989).

60

191. Visse, R. & Nagase, H. Matrix metalloproteinases and tissue inhibi-tors of metalloproteinases: structure, function, and biochemistry. Circ Res 92, 827-39 (2003).

192. Edwards, D. R., Handsley, M. M. & Pennington, C. J. The ADAM metalloproteinases. Mol Aspects Med (2008).

193. Vassar, R. Beta-secretase (BACE) as a drug target for Alzheimer's disease. Adv Drug Deliv Rev 54, 1589-602 (2002).

194. Hällbrink, M., Oehlke, J., Papsdorf, G. & Bienert, M. Uptake of cell-penetrating peptides is dependent on peptide-to-cell ratio rather than on peptide concentration. Biochim Biophys Acta 1667, 222-8 (2004).

195. Elmquist, A. & Langel, Ü. In vitro uptake and stability study of pVEC and its all-D analog. Biol Chem 384, 387-93 (2003).

196. Lindgren, M. E., Hällbrink, M. M., Elmquist, A. M. & Langel, Ü. Passage of cell-penetrating peptides across a human epithelial cell layer in vitro. Biochem J 377, 69-76 (2004).

197. Trehin, R. et al. Metabolic cleavage of cell-penetrating peptides in contact with epithelial models: human calcitonin (hCT)-derived pep-tides, Tat(47-57) and penetratin(43-58). Biochem J 382, 945-56 (2004).

198. Rennert, R., Wespe, C., Beck-Sickinger, A. G. & Neundorf, I. De-veloping novel hCT derived cell-penetrating peptides with improved metabolic stability. Biochim Biophys Acta 1758, 347-54 (2006).

199. Youngblood, D. S., Hatlevig, S. A., Hassinger, J. N., Iversen, P. L. & Moulton, H. M. Stability of cell-penetrating peptide-morpholino oligomer conjugates in human serum and in cells. Bioconjug Chem 18, 50-60 (2007).

200. Amantana, A. et al. Pharmacokinetics, biodistribution, stability and toxicity of a cell-penetrating peptide-morpholino oligomer conju-gate. Bioconjug Chem 18, 1325-31 (2007).

201. Lehrer, R. I. & Ganz, T. Antimicrobial peptides in mammalian and insect host defence. Curr Opin Immunol 11, 23-7 (1999).

202. Hancock, R. E. & Diamond, G. The role of cationic antimicrobial peptides in innate host defences. Trends Microbiol 8, 402-10 (2000).

203. Zasloff, M. Antimicrobial peptides of multicellular organisms. Na-ture 415, 389-95 (2002).

204. Steiner, H., Hultmark, D., Engstrom, A., Bennich, H. & Boman, H. G. Sequence and specificity of two antibacterial proteins involved in insect immunity. Nature 292, 246-8 (1981).

205. Zasloff, M. Magainins, a class of antimicrobial peptides from Xeno-pus skin: isolation, characterization of two active forms, and partial cDNA sequence of a precursor. Proc Natl Acad Sci U S A 84, 5449-53 (1987).

206. Oren, Z. & Shai, Y. Mode of action of linear amphipathic alpha-helical antimicrobial peptides. Biopolymers 47, 451-63 (1998).

207. Park, C. B., Kim, M. S. & Kim, S. C. A novel antimicrobial peptide from Bufo bufo gargarizans. Biochem Biophys Res Commun 218, 408-13 (1996).

61

208. Ehrenstein, G. & Lecar, H. Electrically gated ionic channels in lipid bilayers. Q Rev Biophys 10, 1-34 (1977).

209. Shai, Y. Mode of action of membrane active antimicrobial peptides. Biopolymers 66, 236-48 (2002).

210. Tossi, A., Sandri, L. & Giangaspero, A. Amphipathic, alpha-helical antimicrobial peptides. Biopolymers 55, 4-30 (2000).

211. Hancock, R. E. & Rozek, A. Role of membranes in the activities of antimicrobial cationic peptides. FEMS Microbiol Lett 206, 143-9 (2002).

212. Zhu, W. L. et al. Design and mechanism of action of a novel bacte-ria-selective antimicrobial peptide from the cell-penetrating peptide Pep-1. Biochem Biophys Res Commun 349, 769-74 (2006).

213. Haukland, H. H., Ulvatne, H., Sandvik, K. & Vorland, L. H. The antimicrobial peptides lactoferricin B and magainin 2 cross over the bacterial cytoplasmic membrane and reside in the cytoplasm. FEBS Lett 508, 389-93 (2001).

214. Takeshima, K., Chikushi, A., Lee, K. K., Yonehara, S. & Matsuzaki, K. Translocation of analogues of the antimicrobial peptides magai-nin and buforin across human cell membranes. J Biol Chem 278, 1310-5 (2003).

215. Sandgren, S. et al. The human antimicrobial peptide LL-37 transfers extracellular DNA plasmid to the nuclear compartment of mamma-lian cells via lipid rafts and proteoglycan-dependent endocytosis. J Biol Chem 279, 17951-6 (2004).

216. Matzinger, P. The danger model: a renewed sense of self. Science 296, 301-5 (2002).

217. Kudela, P. et al. Effective gene transfer to melanoma cells using bacterial ghosts. Cancer Lett (2007).

218. Paukner, S. et al. DNA-loaded bacterial ghosts efficiently mediate reporter gene transfer and expression in macrophages. Mol Ther 11, 215-23 (2005).

219. Sanger, F. et al. Nucleotide sequence of bacteriophage phi X174 DNA. Nature 265, 687-95 (1977).

220. Young, K. D. & Young, R. Lytic action of cloned phi X174 gene E. J Virol 44, 993-1002 (1982).

221. Buckley, K. J. & Hayashi, M. Lytic activity localized to membrane-spanning region of phi X174 E protein. Mol Gen Genet 204, 120-5 (1986).

222. Witte, A., Wanner, G., Sulzner, M. & Lubitz, W. Dynamics of PhiX174 protein E-mediated lysis of Escherichia coli. Arch Micro-biol 157, 381-8 (1992).

223. Bläsi, U., Henrich, B. & Lubitz, W. Lysis of Escherichia coli by cloned phi X174 gene E depends on its expression. J Gen Microbiol 131, 1107-14 (1985).

224. Burke, B., Sumner, S., Maitland, N. & Lewis, C. E. Macrophages in gene therapy: cellular delivery vehicles and in vivo targets. J Leukoc Biol 72, 417-28 (2002).

62

225. Paukner, S., Kohl, G., Jalava, K. & Lubitz, W. Sealed bacterial ghosts--novel targeting vehicles for advanced drug delivery of wa-ter-soluble substances. J Drug Target 11, 151-61 (2003).

226. Felnerova, D. et al. T cell-specific immune response induced by bacterial ghosts. Med Sci Monit 10, BR362-70 (2004).

227. Haslberger, A. G. et al. Activation, stimulation and uptake of bac-terial ghosts in antigen presenting cells. J Biotechnol 83, 57-66 (2000).

228. Groger, M. et al. Dermal microvascular endothelial cells express the 180-kDa macrophage mannose receptor in situ and in vitro. J Immu-nol 165, 5428-34 (2000).

229. Molteni, M., Marabella, D., Orlandi, C. & Rossetti, C. Melanoma cell lines are responsive in vitro to lipopolysaccharide and express TLR-4. Cancer Lett 235, 75-83 (2006).

230. Kim, J. J. et al. Macrophage colony-stimulating factor can modulate immune responses and attract dendritic cells in vivo. Hum Gene Ther 11, 305-21 (2000).

231. de Vries, I. J. et al. Phenotypical and functional characterization of clinical grade dendritic cells. J Immunother 25, 429-38 (2002).

232. Parmiani, G., Rivoltini, L., Andreola, G. & Carrabba, M. Cytokines in cancer therapy. Immunol Lett 74, 41-4 (2000).

233. Paukner, S., Kohl, G. & Lubitz, W. Bacterial ghosts as novel ad-vanced drug delivery systems: antiproliferative activity of loaded doxorubicin in human Caco-2 cells. J Control Release 94, 63-74 (2004).

234. Ebensen, T. et al. Bacterial ghosts are an efficient delivery system for DNA vaccines. J Immunol 172, 6858-65 (2004).

235. Ferber, A., Yaen, C., Sarmiento, E. & Martinez, J. An octapeptide in the juxtamembrane domain of VE-cadherin is important for p120ctn binding and cell proliferation. Exp Cell Res 274, 35-44 (2002).

236. Binah, O., Liu, C. C., Young, J. D. & Berke, G. Channel formation and [Ca2+]i accumulation induced by perforin N-terminus peptides: comparison with purified perforin and whole lytic granules. Biochem Biophys Res Commun 240, 647-50 (1997).

237. Merrifield, R. B. Solid-Phase Peptide Synthesis. 3. An Improved Synthesis of Bradykinin. Biochemistry 3, 1385-90 (1964).

238. Palm-Apergi, C., Lorents, A., Padari, K., Pooga, M. & Hällbrink, M. The membrane repair response masks membrane disturbances caused by cell-penetrating peptide uptake. Faseb J (2008).

239. Rajarao, G. K., Nekhotiaeva, N. & Good, L. Peptide-mediated deli-very of green fluorescent protein into yeasts and bacteria. FEMS Mi-crobiol Lett 215, 267-72 (2002).

240. Yu, K. et al. Solution structure of a cathelicidin-derived antimicrobi-al peptide, CRAMP as determined by NMR spectroscopy. J Pept Res 60, 1-9 (2002).

63

241. Cheng, L., Li, T. Y. & Zhang, Y. Rapid preparation of total nucleic acids from E. coli for multi-purpose applications. J Biochem Mol Biol 37, 351-5 (2004).

242. Saar, K. et al. Cell-penetrating peptides: a comparative membrane toxicity study. Anal Biochem 345, 55-65 (2005).

243. Elmquist, A., Hansen, M. & Langel, Ü. Structure-activity relation-ship study of the cell-penetrating peptide pVEC. Biochim Biophys Acta 1758, 721-9 (2006).

244. Palm, C., Netzereab, S. & Hallbrink, M. Quantitatively determined uptake of cell-penetrating peptides in non-mammalian cells with an evaluation of degradation and antimicrobial effects. Peptides 27, 1710-6 (2006).

245. Palm, C., Jayamanne, M., Kjellander, M. & Hallbrink, M. Peptide degradation is a critical determinant for cell-penetrating peptide up-take. Biochim Biophys Acta 1768, 1769-76 (2007).

246. Kosuge, M., Takeuchi, T., Nakase, I., Jones, A. T. & Futaki, S. Cel-lular internalization and distribution of arginine-rich peptides as a function of extracellular peptide concentration, serum, and plasma membrane associated proteoglycans. Bioconjug Chem 19, 656-64 (2008).

247. Henrich, B., Lubitz, W. & Plapp, R. Lysis of Escherichia coli by induction of cloned phi X174 genes. Mol Gen Genet 185, 493-7 (1982).

248. Monick, M. M. et al. Respiratory syncytial virus up-regulates TLR4 and sensitizes airway epithelial cells to endotoxin. J Biol Chem 278, 53035-44 (2003).

249. Farhood, H., Serbina, N. & Huang, L. The role of dioleoyl phospha-tidylethanolamine in cationic liposome mediated gene transfer. Bio-chim Biophys Acta 1235, 289-95 (1995).