Download Polymeric nanoreactors for use in enzyme replacement therapy

Faculty of Science and Bio-engineering Sciences
Department of Structural Biology Brussels
Polymeric nanoreactors for use
in enzyme replacement therapy
Thesis submitted in fulfilment of the requirement for the degree
of Doctor (Ph.D.) in Bioscience Engineering
ir. Caroline De Vocht
Academic year: 2009-2010
Promotor: Prof. Dr. ir. Jan Steyaert
Co-promotors: Prof. Dr. ir. Wim Versées
Dr. Patrick Van Gelder
Print: Silhouet, Maldegem
© 2010 Caroline De Vocht
2010 Uitgeverij VUBPRESS Brussels University Press
VUBPRESS is an imprint of ASP nv (Academic and Scientific Publishers
nv)
Ravensteingalerij 28
B-1000 Brussels
Tel. + 32 (0)2 289 26 50
Fax + 32 (0)2 289 26 59
E-mail: [email protected]
www.vubpress.be
ISBN 978 90 5487 756 1
NUR 915 / 923
Legal Deposit D/2010/11.161/080
All rights reserved. No parts of this book may be reproduced or
transmitted in any form or by any means, electronic, mechanical,
photocopying, recording, or otherwise, without the prior written
permission of the author.
Acknowledgments
I would like to start by thanking my promotor Jan Steyaert for giving me the
opportunity to do this research and for entrusting this interesting project with
high biomedical relevance to me. I would also like to express my gratitude to
Wim Versées for the terrific scientific guidance, for his critical view and helpful
suggestions. I greatly appreciated your support and learned a lot from you.
Furthermore, I would like to thank Patrick Van Gelder for his invaluable
contribution to this work and for introducing me to the wonderful world of
polymeric drug delivery. A special thanks also goes to An Ranquin upon whom
I could always depend on for experimental guidance, for revising this
manuscript and for a lot of other things as well.
Furthermore, I would like to express my gratitude to all SBB-members for
their help and for providing me with a nice and stimulating work-atmosphere!
Some people I would like to thank in particular: Adinda, An VDM, Karo, Elke,
Inge and Kim; thanks for your support and for always being there for me. I
also want to thank all the JAST-desk-colleagues for the great ambiance.
Lieven, Klaas and Mike also deserve a special thanks for coming to the rescue
several times to help me out with the most diverse problems. Many thanks to
Ronnie for helping me with the AFM experiments. I am as well very thankful to
Bruno and Maria for their helpful assistance and to Nadine for taking care of all
the paperwork. As it seemed very difficult for me to have a peaceful and longlasting relationship with my computer(s), a special thanks also goes out to the
IT-guys Wim W and Wim C, who helped me out countless times and saved me
from many heart attacks.
This project would not have been possible without the expertise of numerous
people from other departments sharing their knowledge with me. From the lab
of Cellular and Molecular Immunology, I would like to thank Jo Van
Ginderachter for his excellent support regarding the immunogenicity studies. I
also want to thank Benoit Stijlemans for his help with the LAL-test and Ella
Omasta for her expertise concerning working with mice. From the In vivo
Cellular and Molecular Imaging lab, a special thanks goes out to Tony Lahoutte
and Vicky Caveliers who were so kind to introduce me in the fascinating world
of the in vivo imaging and who guided me through the biodistribution studies.
I would also like to thank Cindy Peleman for her terrific assistance with the
animal imaging work. Furthermore, I would like to express my gratitude to
Tamara Vanhaecke, Vera Rogiers and Bart Degreef from the Department of
Toxicology for their help with the toxicology studies. Last but not least, I
gratefully thank Michio Hirano from the Department of Neurology of Columbia
University Medical Center who gave me the opportunity to join his lab in New
York for several months. It was a great experience and I learned a lot from
this period abroad.
Acknowledgements also go to IWT Vlaanderen, FWO Vlaanderen and the Vrije
Universiteit Brussel for their financial support.
Graag wil ik ook enkele vrienden bedanken die altijd en overal voor mij
klaarstonden. Bedankt Anneke voor je steun en luisterend oor. Bedankt ook
aan Nick, Evelien, Kia, Nele, Joris, Sara en Tim; de VUB heeft ons
samengebracht en ik hoop dat onze vriendschap nog lang blijft voortbestaan.
Ook Tom wil ik speciaal bedanken; tijdens de beginjaren van mijn doctoraat
was je er steeds om te luisteren en advies te geven en ook nu weet ik dat ik
steeds op je kan rekenen. Een welgemeende dankjewel ook voor de
vriendinnen-klik uit Antwerpen; Yannique, Katelijne, Marie, Els, Mary, Yoka,
Anne-Fré en Céline. Onze vele leuke activiteitjes gaven mij de energie om zo
efficiënt mogelijk met mijn tijd om te springen. Bedankt Gaëlle voor de
oppeppende skype-babbels, ondanks de kilometers afstand was je er steeds
voor me. Bedankt Thomas voor de hulp bij het ontwerpen van de mooie cover.
Bedankt Wouter om af en toe voor de nodige verstrooiing te zorgen. Bedankt
Carole om steeds die zotte spring-in-’t-veld te zijn waardoor je me wist op te
vrolijken wanneer nodig. En bedankt Michiel om me te besmetten met je
aanstekelijke drive om ook buiten de lijnen te kleuren.
Tot slot wil ik graag mijn familie bedanken voor hun onvoorwaardelijke steun;
bedankt Nathalie, Stéphanie, mama en papa om er altijd voor me te zijn en
me te stimuleren om in alles wat ik doe steeds het beste van mezelf te geven!
Table of contents
Chapter 1 Introduction
1.1 Enzyme deficiencies
1.1.1 General
1.1.2 MNGIE, mitochondrial neurogastrointestinal encephalomyopathy
1.2 Enzyme replacement therapy
1.3 Delivery of therapeutic proteins and enzymes
1.3.1 Conventional drug delivery: native and pegylated proteins
1.3.2 Nanotechnology in drug delivery
1.4 Nanoreactors for enzyme delivery
1.4.1 The nanoreactor toolbox
1.4.2 Production methods
1.4.3 Controlling the nanoreactor activity
1.4.4 Biomedical applications
1.5 References
1
3
3
4
15
16
16
18
26
26
33
34
36
40
Chapter 2 Scope of the thesis
53
Chapter 3 Production and biophysical characterization of the
nanoreactors
3.1 Introduction
3.2 Materials and methods
3.2.1 Cloning, expression and purification of thymidine phosphorylase
3.2.2 Cloning, expression and purification of porins Tsx and OmpF
3.2.3 Production of nanoreactors
3.2.4 Dynamic Light Scattering
3.2.5 Spectrophotometric assay for TP activity
3.2.6 Atomic Force Microscopy
3.2.7 Detection of endotoxins: LAL assay
3.2.8 Removal of endotoxins
3.3 Results
3.3.1 Providing the nanoreactor building blocks: enzymes and porins
3.3.2 Production, size and dispersity of the nanoreactors
3.3.3 Enzymatic activity of the nanoreactors
3.3.4 Encapsulation efficiency
3.4 Conclusion and discussion
3.5 References
59
61
62
62
63
64
65
65
65
66
67
67
67
70
72
73
75
75
Chapter 4 Stability of the nanoreactors
4.1 Introduction
4.2 Materials and methods
4.2.1 Production of nanoreactors
77
79
79
79
4.2.2 Production of liposomes
4.2.3 Dynamic Light Scattering
4.2.4 Spectrophotometric assay for TP activity
4.3 Results
4.4 Conclusion and discussion
4.5 References
80
80
80
81
84
84
Chapter 5 Toxicity of the nanoreactors
5.1 Introduction
5.2 Materials and methods
5.2.1 Production of nanoreactors
5.2.2 Hepatocyte cell culture
5.2.3 LDH test
5.2.4 MTT test
5.2.5 Statistical analysis
5.3 Results
5.4 Conclusion and discussion
5.5 References
87
89
90
90
90
90
91
91
91
94
94
Chapter 6 Immunogenicity of the nanoreactors
6.1 Introduction
6.2 Materials and methods
6.2.1 Production of nanoreactors
6.2.2 Macrophage cell culture
6.2.3 In vivo inflammatory response study
6.2.4 Cytokine analysis and NO detection
6.2.5 Immunization of animals
6.2.6 Antibody detection
6.3 Results
6.3.1 Ex vivo macrophage response study
6.3.2 In vivo macrophage response study
6.3.3 Antibody response study
6.4 Conclusion and discussion
6.5 References
97
99
100
100
100
101
101
102
102
103
103
107
108
112
114
Chapter 7 Biodistribution of the nanoreactors
7.1 Introduction
7.2 Materials and methods
7.2.1 Nanoreactor synthesis and characterization
7.2.2 Radiolabeling of particles with technetium-99m
7.2.3 Radiolabeling of enzyme with iodine
7.2.4 Animal guidelines
7.2.5 Dynamic planar gamma camera imaging procedure
7.2.6 Image acquisition and processing
7.2.7 Ex vivo analysis
117
119
120
120
120
121
122
122
123
123
7.2.8 Statistical analysis
7.3 Results
7.3.1 Kinetic analysis of short-term biodistribution
7.3.2 Long-term biodistribution analysis
7.4 Conclusion and discussion
7.5 References
124
124
124
127
130
132
Chapter 8 Efficiency of the nanoreactors
8.1 Introduction
8.2 Materials and methods
8.2.1 Production of nanoreactors
8.2.2 In vitro assay and sample processing
8.2.3 HPLC analysis
8.3 Results
8.4 Conclusion and discussion
8.5 References
135
137
138
138
138
139
140
143
143
Chapter 9 Detailed evaluation of the nanoreactor build-up
9.1 Introduction
9.2 Materials and methods
9.2.1 Production of nanoreactors
9.2.2 Characterization of TPE.coli containing nanoreactors
9.2.3 Activity measurements of TvNH containing nanoreactors
9.2.4 Protease treatment
9.2.5 Surface pressure experiments
9.2.6 Fluorescent labeling of thymidine phosphorylase
9.3 Results
9.3.1 Enzyme activity versus porin/enzyme concentration
9.3.2 Protease sensitivity of nanoreactors
9.3.3 Interaction between the enzyme and the polymeric particle
9.3.4 Nanoreactors with other enzymes
9.3.5 Nanoreactors with other polymers
9.4 Conclusion and discussion
9.5 References
145
147
147
147
148
148
148
149
149
150
150
152
154
157
158
159
161
Chapter 10 General conclusion
163
Chapter 11 Summary
171
Chapter 12 Samenvatting
181
Chapter 13 Publications and scientific manifestations
191
Chapter 1
Introduction
Introduction
1 Introduction
1.1 Enzyme deficiencies
1.1.1 General
A lot of known genetic deficiencies in human are caused by a mutation of a
crucial enzyme, leading to the accumulation of one or more of its substrates.
As a result toxic concentrations of the substrates appear in the bloodstream
and tissues and disturb the normal metabolic degradation pathways. Enzyme
deficiencies are usually due to inherited recessive autosomal metabolic defects
that give rise to a whole range of rare and fatal diseases, each with their own
clinical characteristics.
For example, adenosine deaminase (ADA) deficiency is one of the best studied
enzyme deficiency disorders. Due to the lack of the enzyme ADA there is an
accumulation of the substrates adenosine and deoxyadenosine [1]. The latter
causes an increase in S-adenosylhomocysteine which is toxic to immature
lymphocytes. As a result, the lymphocytes fail to mature and the immune
system is severely compromised or completely absent. As such, ADA
deficiency accounts for about 15% of all cases of Severe Combined
Immunodeficiency (SCID).
Other examples of enzyme deficiencies are Fabry disease [2], Pompe disease
[3], Gaucher disease [4, 5], Hurler disease (mucopolysaccharidosis type I) [6],
Lesch-Nyhan disease [7] and MNGIE (mitochondrial neurogastrointestinal
encephalomyopathy) [8]; all summarized in Table 1-1.
3
Chapter 1
Table 1-1: Examples of enzyme deficiencies with their mutated enzyme and
accumulating substrates.
Deficiency
Enzyme
Substrates
ADA deficiency
adenosine deaminase
adenosine,
deoxyadenosine
Fabry disease
-galactosidase A
trihexosylceramide
Pompe disease
-glucosidase
glycogen
Gaucher disease
glucocerebrosidase
glucocerebroside
Hurler disease
-L-iduronidase
glycosaminoglycan
Lesch-Nyhan disease
hypoxanthine-guanine
phosphoribosyltransferase
hypoxanthine, guanine
MNGIE
thymidine phosphorylase
thymidine, deoxyuridine
1.1.2 MNGIE, mitochondrial neurogastrointestinal encephalomyopathy
In this study, thymidine phosphorylase deficiency was chosen as enzyme
deficiency disease model system.
Thymidine phosphorylase
Thymidine phosphorylase (TP) is a cytosolic homodimeric enzyme that
catalyzes the reversible phosphorolysis of the nucleosides thymidine and
deoxyuridine into the nucleobases thymine and uracil respectively, and 2deoxyribose-1-phosphate (Figure 1-1) [9].
Figure 1-1: Reaction of the enzyme thymidine phosphorylase (TP).
4
Introduction
TP is widely expressed in human tissues, including the gastrointestinal system,
brain, peripheral nerve, spleen, bladder, lung and at lower levels in muscle. It
is not expressed however in kidney, gall bladder, aorta and fat [10-14].
Human TP has been studied extensively and exhibits at least three functions:
catalysis (as thymidine phosphorylase), angiogenesis and cell trophism [9, 15,
16]. TP is also called platelet-derived endothelial cell growth factor (PD-ECGF)
and endothelial cell growth factor 1 (ECGF1) [17], because of its angiogenic
properties, as well as gliostatin, to denote its inhibitory effects on glial cell
proliferation [18-20]. The angiogenic and cell trophism of TP may be related to
the generation of ribose as a byproduct of thymidine catabolism. Cancer
researchers have studied TP extensively because its expression and activity
are increased in some tumors, possibly reflecting their profuse vascularization
[9].
The three-dimensional structure of thymidine phosphorylase from Escherichia
coli [21], Bacillus stearothermophilus [22] and human [23] were solved
previously. The enzyme exists as an S-shaped dimer containing two identical
subunits with a dimeric molecular mass ranging from 90 kDa in E.coli to 110
kDa in mammals. Each subunit is composed of a small -helical domain of six
helices and a large /β domain. The /β domain includes a six-stranded mixed
β-sheet and a four-stranded antiparallel β-sheet. The small and large domains
are connected by three peptide loops. The active site lies in a cavity between
the small and the large domain. The pyrimidine binding site is located at the
small α-helical domain and the dimer interface, while the phosphate binding
site is located in the large α/β domain (Figure 1-2). The three peptide loops
act as a hinge allowing the two domains to move between the open (inactive)
and closed (active) conformations. By bringing together the active site
residues, this domain movement is critical for the enzymatic activity.
5
Chapter 1
Figure 1-2: Structure of E.coli thymidine phosphorylase. The two different monomers of
the S-shaped dimer are presented in green and cyan. The -domains and /-domains
are defined with curly brackets. The ligand SO4 (present in the crystallization solution)
binds in the phosphate pocket that is located in the large /-domain, while the thymine
binds in the small α-helical domain and the dimer interface. The PDB entry 1AZY was
used to construct the dimer and the ligands SO4 and thymine were modeled after
superposition of the 1TPT onto the 1AZY structure. Oxygen atoms are colored in red,
nitrogen in blue, sulfur in orange and carbon in pink.
Mendieta et al. described a two-step SN1 mechanism for the TPE.coli
phosphorolysis reaction [24]. In the first rate-limiting step the O2 of thymidine
interacts with a proton from the imidazole ring of the protonated histidine 85
residue, which makes the pyrimidine base a better leaving group. The
weakening of the glycosidic bond generates an oxocarbocation transition state.
The second step is a fast reaction between this oxocarbocation and a
phosphate dianion, which is generated by deprotonation of the incoming
phosphate by the same histidine 85 residu. As such, the phosphate dianion
acts as a nucleophile that gives rise to 2-deoxyribose-1-phosphate and
thymine. It is clear that in this mechanism the highly conserved histidine 85
plays a crucial role.
6
Introduction
Thymidine phosphorylase deficiency causes MNGIE
In 1999, mutations in the TYMP gene, encoding the enzyme thymidine
phosphorylase,
were
identified
as
the
cause
of
mitochondrial
neurogastrointestinal encephalomyopathy or MNGIE [8]. The disease locus was
mapped to a telomeric region of the long arm of chromosome 22 [25]. The
mutations drastically reduce the activity of this cytosolic enzyme, which results
in systemic increases of its substrates thymidine and deoxyuridine. This leads
to nucleoside pool imbalance which specifically affects the mitochondrial DNA
(mtDNA), causing mtDNA instability (depletion, multiple deletions and point
mutations). This in turn results in defects of the respiratory chain and
mitochondrial dysfunction. The selective effect on mtDNA can be explained by
three factors. First, mitochondrial dNTP pools are physically separated and are
regulated independently [26-29]. Second, mitochondrial nucleotide pools are
probably more vulnerable to the toxic effects of excessive thymidine than
nuclear nucleotide pools (which can rely also on the de novo synthesis) (Figure
1-3) [28-30]. Third, human mitochondria probably lack an effective DNA
mismatch repair system [31].
Figure 1-3: Schematic representation of thymidine metabolism in MNGIE. Due to a
decreased or absent thymidine phosphorylase (TP) activity, increased thymidine is
salvaged predominantly in mitochondria, especially in quiescent cells where the
mitochondrial thymidine kinase (TK2) is constitutively active. This results in increased
dTTP levels, nucleotide pool imbalances and mutations in mitochondrial DNA (mtDNA).
Nuclear DNA (nDNA) relies more on the de novo salvage pathway and is not affected by
increased thymidine levels.
7
Chapter 1
MNGIE, a mitochondrial disease
Mitochondria are the powerhouses of eukaryotic cells. These organelles
generate
energy
in
the
form
of
adenosine
triphosphate
(ATP)
from
carbohydrates, fats and proteins via oxidative phosphorylation. Mitochondria
are unique organelles because they possess their own genetic material,
mitochondrial DNA. This feature is due to the origin of mitochondria as
prokaryotic endosymbionts of eukaryotic cells.
MtDNA is a 16.6 kb, double-stranded, circular molecule. This small genome
contains
tightly
compacted
genes
for
22
transfer
RNAs
(tRNAs),
13
polypeptides and 2 ribosomal RNAs (rRNAs). All 13 polypeptides are subunits
of the oxidative phosphorylation system: 7 belong to Complex I (NADH-CoQ
oxidoreductase), 1 to Complex III (CoQ-cytochrome c oxidoreductase), 3 to
Complex IV (cytochrome c oxidase or COX) and 2 to Complex V (ATP
synthase). These subunits are synthesized within the mitochondrion where
they are assembled together with a large number of subunits encoded by the
nuclear DNA (nDNA). The nDNA-encoded subunits are synthesized in the
cytoplasm and afterwards transported into the mitochondrion.
Figure 1-4 : Schematic representation of the mitochondrial respiratory chain. Protons
are pumped from the matrix to the intermembrane space through complexes I, III, and
IV, and are pumped back to the matrix through complex V to produce ATP. Coenzyme Q
(CoQ) and cytochrome c (Cyt c) are electron (e−) transfer carriers.
8
Introduction
Because mtDNA derives from the oöcyte, the mode of transmission of mtDNA
and of mtDNA point mutations differs from Mendelian inheritance and is called
maternal inheritance. Thus, a disease expressed in both sexes, but no
evidence of paternal transmission is strongly suggestive of an mtDNA
mutation. Moreover, because there are thousands of mitochondria in each cell,
with an average of 5 mtDNAs per organelle, mutations in mtDNA result in two
populations of mtDNAs, mutated and wild-type, a condition known as
heteroplasmy. The phenotypic expression of a mtDNA mutation is regulated by
the threshold effect; the mutant phenotype is expressed in the heteroplasmic
cells only when the relative proportion of the mutant mtDNAs reaches a
certain critical value. That is why a respiratory chain effect becomes only
manifest in some tissues, in which the threshold of mutant mtDNAs is
surpassed, and not in others [32]. About 200 mtDNA point mutations and
innumerable partial deletions have been associated with human diseases, most
of which affect the central and peripheral nervous system. Examples of typical
mitochondrial diseases are MELAS (mitochondrial encephalomyopathy, lactic
acidosis and stroke like episodes), LS (Leigh syndrome) and LHON (Leber
hereditary optic neuropathy) [32].
Although caused by a mutation in nDNA (TYMP gene), MNGIE is a
mitochondrial disease (but with a classical Mendelian inheritance). It is caused
by defects in the intergenomic communication between the nuclear and
mitochondrial genome [33]. Thus, even though MNGIE is not caused by
primary mutations in the mitochondrial genome, it finally results in alterations
in mtDNA (point mutations, depletions and deletions [34-37]) due to
nucleotide pool imbalances.
It is still uncertain how mtDNA point mutations are produced in MNGIE, but
based on the sequence specificity of the alterations (T-to-C, T-to-A and TT-toAA mutations), two molecular events have been hypothesized as schematically
explained in Figure 1-5: a ‘next nucleotide effect’ leading to a high level of
direct misincorporation of the incorrect nucleotide [38] and ‘dislocation
mutagenesis’ [39]. It was shown that direct misincorporation of a dGMP
opposite a template thymine residue is a common event [40, 41]. After the
9
Chapter 1
misincorporation, high concentrations of dTTP in the mitochondria of MNGIE
cells may accelerate polymerase activity across the 5’-AA template region [42]
(Figure 1-5A). In addition, elevated dTMP and dTTP derived from the increased
thymidine level will compromise the mtDNA polymerase  exonuclease removal
of the mismatched nucleotide [39, 43]. The T.G misincorporation will result in
a T-to-C mutation following a subsequent round of mtDNA replication.
Secondly, a dislocation mutagenesis model could account for the T-to-A and
TT-to-AA mutations. According to this proposed mechanism, mutations are
produced by a transient single-nucleotide misalignment in the template DNA.
For example, a T residue within a homopolymeric region loops out and leads to
the incorporation of dTMP opposite the adjacent A residue. If subsequent
nucleotides are incorporated before extrahelical T realigns, a single T deletion
will result. Alternatively, if realignment occurs before the next nucleotide is
incorporated, the 3’-end T of the daughter strand will realign inappropriately
with a template T, leading to a T-to-A transversion [42] (Figure 1-5B). Also
depletions and deletions in mtDNA of MNGIE patients were observed. NADHdehydrogenase 5 has been identified as a hot-spot for mtDNA deletions [44].
The multiple deletions were identified in skeletal muscle, brain and kidney.
Figure 1-5: Proposed mechanism for site-specific point mutations of mtDNA in MNGIE.
A. Model of misincorporation and next-nucleotide effect. B. Model of dislocation
mechanism with the next-nucleotide effect [42].
10
Introduction
Physiology and diagnosis
The main clinical features of MNGIE are progressive external ophtalmoplegia,
severe
gastrointestinal
dysmotility,
cachexia,
peripheral
neuropathy,
leukoencephalopathy and mitochondrial abnormalities [45]. The disease is
chronic and progressive. Although the age-at-onset varies from 5 months to
43 years (most commonly the symptoms start in the second or third decade of
life), the clinical manifestations are remarkably homogeneous. Gastrointestinal
dysfunction is the most prominent feature and leads to progressive loss of
weight (average weight loss of 15 kg) up to severe cachexia. Patients usually
die of complications of their gastrointestinal problems and their critical
nutritional status, with an average age of death of 37 years [46]. In almost all
MNGIE patients clear evidence of mitochondrial dysfunction is found, like lactic
acidosis, defects of mitochondrial respiratory chain enzymes in muscle,
cytochrome c oxidase deficient muscle fibers, depletion in mtDNA and multiple
mtDNA deletions [47].
In patients who are suspected of having MNGIE, the determination of TP
activity in buffy coats (this is the fraction of the blood containing white blood
cells and platelets) (with a spectrophotometric assay) and the determination of
thymidine and deoxy-uridine levels in plasma (with a reversed phase HPLC
method) are diagnostic. Decreased buffy coat TP activity of lower than or
equal to 8% of the activity found in controls and increased concentrations of
thymidine (more than 3 µmol/l) and deoxyuridine (more than 5 µmol/l) in
plasma is sufficient to diagnose MNGIE (Table 1-2) [48, 49]. When these
clinical criteria are reached, mtDNA is sequenced to identify the genetic defect
that causes the TP deficiency.
Apart from these typical MNGIE patients, some late-onset MNGIE patients
(affected at 55 to 61 years old) were reported in which a partial loss of TP
activity
causes
an
incomplete
and
mild
MNGIE
phenotype
[50,
51].
Asymptomatic TYMP mutation carriers have moderate reductions of TP activity
but undetectable thymidine and deoxyuridine levels.
11
Chapter 1
Table 1-2: Thymidine and deoxyuridine concentrations in plasma (expressed in µM)
and TP activity in buffy coats (expressed as nmol thymine produced/hour/mg TP) of
MNGIE patients, late-onset MNGIE patients, TYMP mutation carriers and healthy
controls [49, 50].
Number
of persons
tested
Thymidine
Deoxyuridine
TP activity
(µM)
mean±SD
(Range)
(µM)
mean±SD
(Range)
(nmol/h/mg)
mean±SD
(Range)
MNGIE
25
8.6±3.4
(3.9-17.7)
14.2±4.4
(5.5-24.4)
10±15
(0-46)
Late-onset
MNGIE
3
1.03±0.45
(0.4-1.4)
2.9±1.9
(1.0-4.7)
84±19.5
(58-105)
Heterozygotes
14
<0.05
<0.05
222±89
(90-360)
Controls
20
<0.05
<0.05
634±217
(253-1000)
MNGIE is a rather uncommon disease with worldwide around 130 known cases
up until now (personal communication Dr. Hirano). Due to misdiagnosis (most
likely as anorexia nervosa) there are probably more cases around. All MNGIE
patients are of different ethnic groups and males and females are equally
affected.
Thymidine phosphorylase deficient mice
To further characterize the disease and to better evaluate new therapies, a
MNGIE mouse model was generated independently by two different groups
[20, 52]. Because the murine uridine phosphorylase (UP) effectively degrades
thymidine, deoxyuridine and uridine (in contrast to the human UP which
catabolizes the ribonucleoside uridine, but not the deoxyribonucleosides
thymidine and deoxyuridine) [53], a double knockout TP-/-UP-/- strain was
generated. The TP knockout mouse was produced by inserting a DNA cassette
into the pyrimidine phosphorylase consensus sequence in exon 4 of the mouse
TP gene through homologous recombination in murine embryonic stem cells.
The DNA cassette contains a phosphoglycerate kinase promotor and a
neomycin resistance gene flanked by LoxP sites. Subsequently, these cells
were microinjected into blastocysts to obtain mutant chimeric mice that
generated heterozygous TP-knockout mice [52]. By crossing these TP
12
Introduction
knockout mice with CRE-expressing mice the large neomycin cassette inserted
into the TP gene was successfully flipped out. Afterwards these mice were
backcrossed with C57BL/6 mice for 10 generations to obtain TP knockout
animals with a homogenous nuclear background. In a similar way UP
knockouts were produced and both strains were mated to obtain TP-/-UP-/double knockouts.
TP-/-UP-/- mice were characterized extensively by Lopez et al. [52]. TP activity
was undetectable in all studied tissues except liver, which had only a
decreased activity. As a consequence thymidine and deoxyuridine were
significantly increased by 4-65 fold in all tissues tested. In addition, brain of
TP-/-UP-/- mice showed partial mtDNA depletion at 14-18 months, resulting in
reduced activities of respiratory chain complex enzymes. Thus, TP-/-UP-/- mice
recapitulate several features of MNGIE patients including TP deficiency,
elevated thymidine and deoxyuridine levels in plasma and tissues, mtDNA
depletion and respiratory chain defects. Nevertheless, the last two features
could only be observed in the brain of the double knockout mice. The brain
specific phenotype may be due to the shorter life span of the mice compared
with humans and the relatively modest increases of nucleosides in mutant
mice. First, the average age-at-onset of symptoms in MNGIE is 18.7 years;
therefore the 2-year lifespan of mice may not be sufficient to allow significant
accumulation
of
mtDNA
alterations
and
consequent
respiratory
chain
dysfunctions. Secondly, in contrast to humans who have undetectable
amounts of thymidine and deoxyuridine in tissues and plasma, wild type mice
have detectable amounts of these nucleosides; therefore TP-/-UP-/- mice show
only 4-65 fold increases in nucleosides, whereas MNGIE patients show >100
fold increases [54]. The group of Dr. Hirano is now trying to enhance the
mitochondrial phenotype of the TP-/-UP-/- mice by stressing them with
exogenous thymidine and deoxyuridine (in drinking water).
Therapeutic approaches
Since systemic accumulations of thymidine and deoxyuridine are the cause of
MNGIE, strategies that are able to reduce the toxic concentrations of these
13
Chapter 1
circulating nucleosides may be therapeutic [55]. Attempts to reduce thymidine
concentrations in the blood of two patients with MNGIE through hemodialysis
indicated that although repeated dialysis could reduce the basal level of
circulating thymidine, a more continuous elimination of nucleosides is needed
for a permanent reduction [48, 56]. Because in healthy individuals platelets
have high TP activity [57], repeated platelet infusions were used to try to
restore TP enzyme activity in MNGIE patients [58]. In vitro and in vivo studies
showed that the nucleoside concentrations could be reduced to undetectable
levels. Unfortunately the decrease in toxic nucleoside levels is only transient
and a more effective restoration of the TP activity is required.
In 2006, Dr. Hirano started to use allogeneic stem cell transplantations to
provide long-term restoration of TP activity. From the 2 MNGIE patients
initially
treated
significantly
with
improved
allogeneic
her
stem
quality
of
cell
transplantations,
life
and
1
demonstrated
patient
virtually
undetectable thymidine and deoxyuridine levels in plasma, while the other
patient died several weeks after the transplantation [59]. Up until now,
transplants were attempted in total in 12 patients. Two had problems with the
conditioning regimen which include chemotherapy to partially ablate bone
marrow and make space for donor cells. From the 10 patients that underwent
transplants, 5 have died and 5 showed biochemical and clinical improvements
(personal communication Dr. Hirano). In the cases where the transplantation
failed, the patients were already severely debilitated by MNGIE and/or the
quality and source of the donor stem cells were not optimal. Rejection of the
donor stem cells is a well-known and intrinsic risk of stem cell transplantations
in general.
Since erythrocytes encapsulated with adenosine deaminase have been used
successfully in a patient with adenosine deaminase deficiency [60], a carrier
erythrocyte entrapped thymidine phosphorylase therapy (CEETP) was used to
treat one MNGIE patient [61]. Although the plasma and urine concentrations
decreased significantly, the patient’s clinical condition remained poor and she
died of pneumonia 21 days after CEETP. This approach is however promising
since the encapsulation of the TP prevents the formation of antibodies and
14
Introduction
maintains the enzyme activity during the erythrocyte life span. In general,
enzyme replacement strategies have shown their potential in other enzyme
deficiencies as well.
1.2 Enzyme replacement therapy
In enzyme replacement therapy (ERT) the deficient enzyme is administered
intravenously in its native or stabilized form. Enzyme replacement therapy is
in general a well-tolerated and safe strategy to ‘treat’ or better ‘stabilize’
enzyme deficiency diseases [62]. Because the substrates, in most cases
nucleosides,
equilibrate
rapidly
across
cell
membranes
(by
nucleoside
transporters) [63], high substrate concentrations in tissues and organs lead to
high plasma levels. Therefore, restoring the enzyme activity in plasma is
sufficient to reverse the principal abnormalities and intravenous injections of
the enzyme are administered. In most cases immunosuppressive drugs (like
cyclosporine A and azathioprine) are co-administered to prevent a strong
antibody response and induce immune tolerance to the therapeutic enzyme
[64].
Since ERT has been successfully introduced for patients with Gaucher disease
by Brady and coworkers [65], this kind of treatment has been taken into
consideration for other enzyme deficiency disorders (mostly lysosomal storage
diseases) as well. Clinical trials have demonstrated the clinical benefit of this
therapeutic approach in Fabry disease [66], mucopolysaccharidosis type I
(Hurler disease), II and VI [67] and Pompe disease [68].
In most enzyme deficiency disorders the substrate concentrations accumulate
during the first years of life until a threshold is exceeded and the disease pass
from a preclinical to a clinical state with physical symptoms. Thus, timing of
the ERT is very important; the earlier the ERT starts, the better the response
to therapy [69]. Unfortunately, good biomarkers that allow fast diagnosis and
adequate judgment of the stage of the disease are currently lacking (or not
used frequently because of the rarity of the disorders) and in most cases ERT
is only started in the clinical state, after appearance of the symptoms [70].
15
Chapter 1
The biggest limitation of ERT is probably the financial burden of the treatment.
Large pharmacological quantities of recombinant enzyme need to be produced
and an individualized approach to treatment is followed rather than a set
dosage regimen [71]. As a result, average annual costs per patient are very
high with estimated averages of 180 000 USD for Gaucher disease, 215 000
USD for Fabry disease and 260 000 USD for Hurler disease [72].
Importantly, ERT is not a cure that completely eliminates all signs and
symptoms, but a treatment that can improve the quality of life significantly.
Patients need repeated intravenous injections, probably two to four weekly
and potentially lifelong, to achieve a continuous elimination of the toxic
substrates. Prolonging the circulation time of the therapeutic enzyme would
lead to decreased frequency of administration and decreased costs of the
therapy and as such would mean a tremendous improvement for patients
suffering from such a genetic disorder.
1.3 Delivery of therapeutic proteins and enzymes
1.3.1 Conventional drug delivery: native and pegylated proteins
A lot of proteins, in particular hormones, blood factors, vaccines, interferons,
monoclonal antibodies and enzymes, can be very useful as therapeutics in
various biomedical applications. Since the approval of recombinant human
insulin in 1982, many biotechnology products were shown to be effective and
safe for the treatment of numerous diseases [73]. The first generation
pharmaceuticals were recombinant proteins with an amino acid sequence
identical to that of the native human polypeptide. These formulations however
often do not possess optimal drug characteristics and suffer from low stability
and solubility, very short circulation half-lives, sensitivity to proteolytic
degradation and immunogenicity and toxicity issues [74]. Therefore, an
increasing number of second generation protein therapeutics are engineered to
reach better efficiency.
16
Introduction
To
improve
their
stability,
proteolytic
resistance,
immunogenicity
and
circulation half-life, therapeutic proteins can be modified by covalently linking
various poly(ethylene glycol) (PEG) molecules to their surface, a technique
called pegylation [75]. PEG has a long history as a non-toxic, nonimmunogenic, hydrophilic, uncharged and non-biodegradable polymer and has
been approved by the FDA (Food and Drug Administration) as ‘generally
recognized as safe’. In the beginning, linear PEG reagents of 5 to 10 kDa were
used. Later, branched, forked and multi-arm PEG chains and PEG dendrimers
reaching molecular masses up to 60 kDa became quite common [76, 77].
Since PEG is highly flexible and can coordinate numerous water molecules,
pegylated proteins have a greatly expanded hydrodynamic volume with an
apparent molecular weight 5 to 10 times higher than that of a globular protein
with the same molecular mass. This results in reduced kidney filtration. In
addition, PEG chains exhibit stealth properties by shielding the protein from
its environment, hereby protecting it against proteolytic degradation and
recognition by the immune system (epitope shielding). Finally, pegylation can
mask or modify protein charges which prevents interaction with charged blood
components. All these features result in a prolonged circulation half-life (from
2 to 400 times longer) of the pegylated protein in comparison with the native
protein
[78].
The
FDA
has
approved
several
pegylated
proteins
as
therapeutics, and probably more are under investigation (Table 1-3).
However, pegylation also has disadvantages. The attachment of PEG molecules
to a protein can modify the interaction capabilities or active sites of the protein
that are responsible for its biological function [79]. In addition, the nonimmunogenicity of PEG became questionable since specific anti-PEG antibodies
were found in some clinical trials. Some reports indicate the loss of longcirculating properties when the PEG-protein conjugate is administered multiple
times due to a strong anti-PEG immune response [80, 81].
Although widely used, pegylation is a rather complicated and expensive
approach. In particular when the therapeutic protein is an enzyme, there is a
serious risk of a decrease or a complete loss of enzyme activity. Because the
rather expensive production of recombinant protein is followed by an
17
Chapter 1
inefficient coupling to the PEG chains and multiple purification steps, costs of
this technology are quite high [79]. Furthermore, since pegylation includes a
chemical modification, the whole developing process from bench to market
needs to pass numerous trials, requiring a lot of time and money. A more
convenient strategy to deliver therapeutic enzymes to the body is to
encapsulate them in appropriate carrier systems.
Table 1-3: Approved PEG-protein conjugates.
PEG conjugate
Company
Disease
Year to market
PEG-adenosine
deaminase (Adagen®)
[82]
Enzon
Pharmaceuticals
Severe Combined
Immunodeficiency
disease (SCID)
1990
PEG-asparaginase
(Oncaspar®) [83]
Enzon
Pharmaceuticals
Acute
lymphoblastic
leukemia
1994
PEG-interferon 2b
(PEG-Intron®) [84]
Schering-Plough
Hepatitis C
2000
PEG-interferon 2a
(Pegasys®) [85]
Hoffman-La
Roche
Hepatitis C
2002
PEG-granulocyte
colony stimulating
factor (Filgrastim,
Neulasta®) [86]
Amgen
Treating
neutropenia
during
chemotherapy
2002
PEG-growth hormone
receptor antagonist
(Pegvisomant,
Somavert®) [87]
Pfizer
Acromegaly
2002
1.3.2 Nanotechnology in drug delivery
Nanotechnology (derived from the Greek word ‘nano’ meaning dwarf) is an
emerging branch of multidisciplinary science that deals with designing tools
and devices in the nanometer-size-range. Over the last decades new
nanotechnology-based
drug
delivery
systems
were
developed
for
the
controlled release and/or the targeted delivery of proteins in the body [88,
18
Introduction
89]. The carrier acts as a depot for high concentrations of therapeutics and
provides a solubilising and protective environment. This can be used either for
local delivery, where the carrier is retained at the site of delivery, or for
systemic delivery, where it circulates through the vasculature, increasing the
circulation time of the therapeutic agent [90].
Liposomes
Liposomes are artificial phospholipid vesicles made up of an aqueous interior
surrounded by a lipid bilayer (Figure 1-6A). They are composed of one or more
bilayers of natural and/or synthetic lipids like PE (phosphatidylethanolamine),
DOPC (dioleoylphosphatidylcholine) or EPG (egg phosphatidylglycerol) which
are relatively biocompatible and biodegradable. The vesicles are formed
spontaneously when the lipids are hydrated in aqueous solutions. The major
driving force behind this self-association is the amphiphilic character of the
lipids which favors a micelle or vesicle conformation where the hydrophobic
tails are removed from the aqueous surrounding and the hydrophilic parts are
exposed into the water. Liposomes have been investigated as carriers of
various pharmacologically active agents. Drugs with varying lipophilicities can
be encapsulated in liposomes, either in the phospholipid membrane (strongly
lipophilic drugs), in the entrapped aqueous volume (strongly hydrophilic
drugs) or at the bilayer interface (drugs with intermediate lipophilicity) [91].
The sustained release of the drugs occurs gradually due to leakage of the
substance or degradation of the liposome. In addition, many stimuli-sensitive
liposomes are being developed that release their substance in one single burst
as a result of destabilization of the liposome membrane caused by a change in
the environment. Such a triggered release can be accomplished by pHsensitive [92], thermo-sensitive [93] or photo-sensitive [94] liposomes.
Some liposomal formulations have reached the market or are now entering
clinical trials [95, 96]. Ambisome® (Gilead Sciences) in which the encapsulated
drug is the antifungal amphotericin B [97], Myocet® (Elan Pharmaceuticals,
Inc) encapsulating the anticancer agent doxorubicin [98] and Daunoxome®
(Gilead Sciences) in which the drug incorporated is daunorubicin [99], are
19
Chapter 1
some examples of approved liposomal drugs. Note that plasma half-lives are
around 7-10 hours for Ambisome®, 2-3 hours for Myocet® and 5 hours for
Daunoxome® in comparison with less than 0.5 hours for the free drugs.
In addition, several studies reported the encapsulation of enzymes inside
liposomes [100]. The enzyme must be liberated from the particle before it can
catalyze a particular reaction occurring outside the liposome. Enzymecontaining
liposomes
are
under
development
for
various
biomedical
applications, including enzyme replacement therapy. Table 1-4 gives an
overview of enzyme containing lipid particles constructed and investigated for
their use in enzyme replacement therapy.
Table 1-4: Selected examples of enzyme-containing liposomes for potential use in
enzyme replacement therapy [101].
Investigation
Reference
Amyloglucosidase
Enzyme
First proposal to use enzyme-containing
liposomes in treatment for Pompe
disease
[102]
Dextranase
Development of a model system for a
lysosomal storage disease in rat
[103]
-Galactosidase
In vitro evaluation of the use of enzymecontaining liposomes for the treatment
of cultured gangliosidosis fibroblasts
[104]
Glucocerebrosidase
Treatment of a patient with Gaucher
disease
[105]
-Glucuronidase
In vivo investigation in mice of the effect
of charged enzyme-containing liposomes
[106]
Liposomes
protect
the
encapsulated
molecules
from
rapid
degradation
resulting in prolonged circulation persistence. Due to their enhanced size the
clearance route shifts from the kidneys (size-selective cut-off for glomerular
filtration is approximately 60 kDa) to the reticuloendothelial system (RES),
also called mononuclear phagocyte system (MPS) [107, 108]. RES clearance is
based on the phagocytotic uptake most often by macrophages especially in the
20
Introduction
liver, spleen and lymph nodes. However, when the target site is beyond the
RES system, liposome uptake by macrophages and their consequent removal
from circulation is one of the main disadvantages for the use of liposomes as
drug delivery systems. RES does not recognize the liposomes themselves but
rather the opsonins which are serum proteins bound to the surface of
liposomes [109, 110]. Complement components, which play a role in host
defense against invading pathogens, comprise another important system able
to recognize liposomes [111]. This system acts through initiating membrane
lysis and enhancing uptake by phagocytotic cells. Another limitation of
liposomes is their instability in plasma due to their interaction with high (HDL)
and low density (LDL) lipoproteins, since this interaction results in the
premature release of the encapsulated drug into the plasma [112].
Pegylated liposomes
An effective strategy to increase the circulation time of liposomes is to
incorporate PEG polymers in the lipid bilayer (Figure 1-6B) [113]. These PEG
chains form a protective coat on the liposome surface and decrease
recognition by opsonins and uptake by the RES system [114]. Allen and
coworkers [107] were the first to show the long-circulating effect of pegylated
liposomes and the names ‘stealth liposomes’ (‘stealth liposomes’ is a
registered trademark of Liposome Technology, Inc.) or ‘sterically stabilised
liposomes’ have been given to this new class of liposomes.
By reducing RES uptake, long-circulating liposomes can passively accumulate
inside tissues and organs. This phenomenon, called passive targeting, is
especially useful in solid tumors [115]. Due to the leaky vasculature, longcirculating liposomes are preferentially accumulated in the tumor area [116].
As such, sterically stabilized liposomes are explored to encapsulate anti-cancer
drugs
[117].
Some
examples
of
stealth
liposomal-based
anti-cancer
®
pharmaceuticals are Doxil , a pegylated liposomal formulation of doxorubicin
(half-life of 1.5 to 45 hours) [118], Lipoplatin®, a pegylated liposomal
formulation of cisplatin (half-life of 60-117 hours) [119] and SPI-077®,
another pegylated liposomal formulation of cisplatin (half-life of 60-117 hours)
21
Chapter 1
[120]. Stealth liposomes can also be used for active targeting to specific
tissues by coupling targeting moieties like ligands, antibodies or growth factors
to the terminal of the PEG chain or the phospholipid head group [121-124].
Although PEG remains the standard polymer for the steric stabilization of
liposomes, other polymers also possess stealth-like properties such as
polyvinyl pyrrolidone [125], polyvinyl alcohol [126], polyacril amide [127] and
polymethyl and polyethyl oxazoline [128, 129]. These polymers appear to be
attractive alternatives for designing long-circulation liposomes.
Unfortunately, stealth liposomes still have some shortcomings and are not
ideal as drug delivery vehicles. Fast drug leakage due to the rather unstable
character of lipidic particles in plasma [130, 131] and accelerated blood
clearance remain major drawbacks of these liposomal delivery systems. The
latter was reported by Ishida et al. [132, 133] and Laverman et al. [134]
which
show
that
intravenous
injection
of
PEG-grafted
liposomes
may
significantly alter the pharmacokinetic behavior of a second dose. Recent
evidence also shows that sterically stabilized liposomes are still prone to
opsonization. The dense PEG brush does delay opsonization, but selective
protein deposition eventually occurs after 10 to 20 hours [135, 136]. The
prolonged circulation longevity of stealth liposomes is now believed to be
rather due to diminished aggregation of liposomes.
Figure 1-6: Schematic representation of liposomes (A) and stealth liposomes (B).
22
Introduction
Polymeric particles
Owing to their low molecular weight (MW < 1 kDa), aggregation of lipids
results in molecularly thin membranes that possess a high dynamical and
physical softness. As a consequence, many liposome properties like stability,
degradation and drug retention are difficult to control. In order to obtain more
robust membranes with controllable properties, extensive efforts were made
within the last decades to design polymeric vesicles.
The polymeric nanoparticles described in the literature so far are constructed
from block copolymers. Such a block copolymer consist of distinct polymer
chains covalently linked in a series of two or more segments [137].
Amphiphilic block copolymers are composed of at least one hydrophilic block
and one hydrophobic block, causing self-assembly in aqueous solutions to
nanometer-sized structures. The driving forces for the self-organization are
the difference in solubility of the blocks and the constraint imposed by the
chemical linkage between the blocks. Depending on their concentration,
molecular
weight,
hydrophobic-to-hydrophilic
balance
and
block-length,
amphiphilic block copolymers can aggregate into micelles (Figure 1-7A),
vesicles, cylinders or rod-like structures [138]. Compared to the selfassembled structures formed by lower molecular weight amphiphilic molecules
such as lipids or surfactants, block copolymers self-assemble into significantly
thicker, less permeable and mechanically more stable particles [139]. This
higher stability is due to the larger size of the hydrophobic part and the slower
dynamics of the underlying copolymer molecules caused by a higher
entanglement [140, 141]. It is this increased stability, along with their selfassembled nanometer-sized structures that make block copolymers so
attractive for biomedical applications such as drug delivery.
Numerous natural and synthetic polymers have already been used for the
production of drug delivery nanoparticles. Examples of polymers for the
hydrophobic block include inert PEE (polyethylethylene), PS (polystyrene),
PDMS (polydimethylsiloxane), the degradable PLA (polylactic acid) and PCL
(polycaprolactone). Hydrophilic blocks have been synthesized from PEG
23
Chapter 1
(polyethylene glycol) or PEO (polyethylene oxide), the negatively charged PAA
(polyacrylic acid) and the crosslinkable PMOXA (polymethyloxazoline) [142].
Up until now, PLGA (polylactic-co-glycolic acid) is the most frequently used
copolymer, synthesized from the two monomers lactic acid and glycolic acid.
This polymer is FDA approved for the use in therapeutic devices, due to its
biodegradability and biocompatibility [143].
Polymeric particles can be composed of diblock copolymers (Figure 1-7B)
[144], triblock copolymers (Figure 1-7C), branched polymeric dendrimers or
mixed polymer-lipid composites [145]. Analogous to liposomes, stimuliresponsive polymeric particles have been developed based on the triggered
release of their content by destabilization of the polymeric membrane upon
specific stimuli like pH, temperature, redox potential or ultrasound [146].
Injection of polymeric particles into the blood circulation of rats and mice
shows prolonged circulation times. Clearance half-lives range from 15 to 30
hours and significantly exceed the circulation times of stealth liposomes in
similar dose injections [147].
Figure
(A), in
triblock
colored
1-7: Schematic representation of proteins incorporated in polymeric micelles
particles composed of diblock copolymers (B) and in particles composed of
copolymers (C). Hydrophilic blocks are colored in blue, hydrophobic blocks are
in grey.
Various proteins have been entrapped in nanoparticles (typically between 50250 nm), either as model system, or as possible therapeutic application (Table
1-5 gives a snapshot of some examples). Enzymes, such as -galactosidase
24
Introduction
entrapped
in
nanoparticles
of
polybutylcyanoacrilate
(PBCA)
[148],
or
superoxide dismutase loaded in polylactide-co-glycolide (PLGA) nanoparticles
[149], were delivered to the brain. In addition, some examples are reported of
nanoparticles constructed for the potential use in enzyme replacement
therapy. Genta et al. first encapsulated the enzyme prolidase inside particles
consisting of PLGA [150], later inside nanoparticles consisting of chitosan
[151] and tested them in the enzyme therapy for prolidase deficiency.
Garnacho and coworkers designed a new delivery carrier, based on acid
sphingomyelinase
encapsulation
in
polystyrene
and
PLGA
nanocarriers
targeted to an endothelial surface protein to use in enzyme replacement
therapy for type B Niemann-Pick disease [152].
Nanotechnology, especially nanomedicine, is a rapidly growing field and will
lead to more medical benefits in the coming years. Nevertheless, the
development of biocompatible and biodegradable drug carriers possessing
small particle size, high loading capacity, extended circulation time and ability
to accumulate in required pathological sites in the body still has many
unresolved issues and remains a challenging field of research.
Table 1-5: Examples of polymeric nanoparticles used as carrier for proteins.
Polymer
Payload
Proposed application
Reference
PEG-PLA
catalase
model system
[153]
chitosan
FITC-BSA
vaccine
[154]
polyacrylamide
ovalbumine
vaccine
[155]
PEG-PLA and PEGPCL
paclitaxel and
doxorubicin
cancer therapy
[156]
PLGA
superoxide
dismutase
oxidative stress
[157]
PLA
cyanine dye
imaging
[158]
PEO-PPS-PEO
glucose oxidase
stimuli responsive particles
releasing their content upon
exposure to oxidative
environment
[159]
25
Chapter 1
1.4 Nanoreactors for enzyme delivery
Nanoreactors are defined as submicrometer hollow spheres that encapsulate
enzymes in their aqueous interior and that incorporate channel forming
proteins inside their wall, hereby permeabilizing the nanoreactor. The enzyme
remains inside the particles and the substrates and products can diffuse in and
out. As such, the enzymatic reaction is restricted to the inner volume of the
nanocontainer, which is in contrast to the ‘classical’ enzyme delivery particles
where the enzymes need to be released before they can catalyse the reaction.
As a consequence, nanoreactors avoid the problems of release profile
optimization or the release of therapeutics in inappropriate compartments
compared with the conventional carriers.
1.4.1 The nanoreactor toolbox
Polymers
Although few examples are known of liposomal reactors [160-162], the
majority
of
nanoreactors
reported
thus
far
are
constructed
of
block
copolymers. The polymer that is most commonly used is the PMOXA-PDMSPMOXA
or
poly(2-methyloxazoline)-poly(dimethylsiloxane)-poly(2-
methyloxazoline) triblock copolymer (Figure 1-8) [163]. The main advantage
of this triblock copolymer is the versatility of the self-assembled particles; 1)
the shell has a hydrophilic, biocompatible low-protein-binding surface and an
extremely high mechanical and thermal stability, 2) the shell can act as a
biomembrane for protein reconstitution, 3) the nanovesicle can encapsulate
hydrophilic substances, and 4) the shell material has enormous possibilities for
molecular functionalization (like crosslinking and targeting). Also, the nontoxic and biodegradable PEG-oligo(DTO suberate)-PEG or
glycol)-block-oligo(desamino-tyrosyltyrosine
octyl
ester
poly(ethylene
suberate)-block-
poly(ethylene glycol) [164] and the PEO-PDMS-PMOXA or poly(ethylene
oxide)-block-poly(dimethylsiloxane)-block-poly(2-methyloxazoline)
as
an
asymmetric block copolymer [138] have been used to construct nanoreactors.
26
Introduction
Figure 1-8: Structure of poly(2-methyloxazoline)-poly(dimethylsiloxane)-poly(2methyloxazoline) or PMOXA-PDMS-PMOXA triblock copolymer.
Porines
Nature provides a great variety of specific and non specific channel proteins
which allow translocation of substrates across biological membranes. It is clear
that membrane-like structures formed by amphiphilic block copolymers closely
resemble typical biological membranes. Actually, it has been shown that
membrane proteins can be successfully reconstituted in such artificial polymer
membranes [165]. The block copolymer membranes are considerably thicker
(e.g. 10 nm) than conventional lipid bilayers (3-5nm) due to the larger size of
the underlying copolymer molecules. This raises the question whether the
proteins can preserve their activity within a block copolymer membrane.
However, the high flexibility of the hydrophobic block and the conformational
freedom of the polymer molecules allows a block copolymer membrane to
adapt to the specific geometric and dynamic requirements of membrane
proteins without considerable loss of free energy and thus minimizing the
hydrophobic mismatch [166] (Figure 1-9). This technology can be used on the
one hand to study protein-membrane interactions (e.g. antimicrobial proteins
like alamethicin) [167] and on the other hand to permeabilize polymeric
membranes for substrates and products and make functionalized nanoreactors
for drug delivery applications.
27
Chapter 1
Figure 1-9: A molecular dynamics simulation of stable insertion of a mimetic protein
pore into either thin (A) or thick (B) block copolymer membranes (taken from [142]).
Up to now most channels used in nanoreactor design are derived from the
Gram-negative outer membrane [168]. These proteins form β-barrels and can
be produced and isolated in large quantities (tens of mg/L culture) [169] and
are extremely rigid (stable up to 70°C in the presence of 2% sodium dodecyl
sulphate, in organic solvents and 4 M guanidinium hydrochloride).
The integration of protein channels into block copolymer membranes was
initially realized by Meier and coworkers. They demonstrated that the general
diffusion porin OmpF (Outer membrane protein F) can be integrated in
PMOXA-PDMS-PMOXA
copolymer
membranes
[170].
Surprisingly,
the
functionality of the OmpF protein was fully preserved despite the artificial
surrounding.
This
channel
functions
as
a
molecular
sieve,
allowing
concentration-driven diffusion of molecules with a molecular weight below 600
Da. There is no selectivity for the transported molecules, although OmpF
shows slight cation selectivity [160, 171]. PhoE is also a general diffusion porin
but it favours transport of more anionic molecules [168]. Up until now, PhoE
however has not been used to permeabilize nanoreactors.
Secondly, Graff et al. reconstituted the bacterial channel forming protein LamB
(or Maltoporin) in ABA-copolymer vesicles [172]. The outer membrane protein
LamB is a specific transporter for maltodextrins but also serves as a receptor
for λ phage and triggers the injection of DNA. Since DNA translocation was
28
Introduction
observed, this study proved again that porins can preserve their activity within
a completely synthetic block copolymer membrane (Figure 1-10).
Figure 1-10: Schematic representation of a DNA-loaded PMOXA–PDMS–PMOXA vesicle.
λ-phage binds a LamB protein, and the DNA is transferred across the block copolymer
membrane.
Tsx is an outer membrane transporter of Gram-negative bacteria constructed
of a monomeric 12-stranded -barrel. In contrast to the general porines, Tsx
allows specific transport of nucleosides and nucleotides. Since it has at least
three binding sites for nucleosides in the interior of the channel, rapid
transport even at low concentrations is possible [173]. The channel is shaped
like a keyhole; upon binding the base moiety of the nucleoside is located in the
narrow part of the keyhole, while the bulky sugar occupies the wider opening
(Figure 1-11). Pairs of aromatic residues and flanking ionizable residues are
involved in nucleoside binding. Nucleoside transport occurs by diffusion from
one binding site to the next [174]. The direction of transport is driven by the
concentration gradient across the membrane. Apparently the Tsx-dependent
uptake differs slightly for the various nucleosides. Among the five nucleosides
investigated, the binding increased in the following order: cytidine, guanosine,
uridine, adenosine and thymidine. Despite its specificity, the Tsx pore also
allows concentration-driven aspecific diffusion of small molecules unrelated to
its specific substrates [175].
29
Chapter 1
Figure 1-11: Nucleoside binding sites in the Tsx channel. 1. Surface representation
showing the keyhole shape of the Tsx pore (left panel). The right panel shows a close-up
of the channel with a nucleoside bound at Nuc1. 2. Stereo view of the nucleoside binding
sites in thymidine-soaked crystals. The nucleosides that could be built in the density at
Nuc1 and Nuc2 are shown in green. The aromatic residues that line the channel and that
are involved in nucleoside binding are indicated in cyan. The front (F) and the back (B)
of the barrel are indicated (taken from [174]).
All previously discussed porins are rather small in diameter (7-11 Å) and
molecules with a molecular weight greater than 600 Da have difficulties
passing through these small channels. To overcome this, Nallani and
coworkers introduced a two deletion mutant of the protein FhuA to
permeabilize nanoreactors which they dubbed synthosomes. This monomeric
protein is an outer membrane receptor for the uptake of iron scavenging
siderophores and consists of 22 β-strands wrapped around a plug-domain.
30
Introduction
Removal of this internal domain results in a channel with a large pore diameter
(39-46 Å) that ensures rapid compound flux [176].
Kumar and coworkers described the incorporation of the bacterial waterchannel protein Aquaporin Z into a PMOXA-PDMS-PMOXA triblock copolymer
membrane with a large hydrophobic-to-hydrophilic block ratio. The water
permeability varied with the porin/polymer ratio and showed a clear channelmediated transport. Furthermore, the selectivity of the membrane for water
over small solutes such as salt, glucose, urea and glycerol was demonstrated
[177].
Symmetry and orientation of the membrane plays a crucial role for the
insertion of integral membrane proteins into artificial membrane systems.
Stoenescu and coworkers demonstrated this using an Aquaporin that was
fused to a poly-his-tag. Using antibodies that are specifically directed against
the his-tag, it was possible to investigate the orientation of the proteins in the
artificial membrane. The asymmetric ‘ABC’ triblock copolymer membranes
always favor one orientation, while there is no preferred orientation in
symmetric ‘ABA’-polymer membranes [138].
Enzymes
In contrast to the long list of enzymes entrapped in liposomes (eg: alkaline
phosphatase, asparaginase, chymotrypsin, elastase, lysozyme, peroxidase,
proteinase K, … [101]), up until now only few examples are reported where
enzymes are entrapped inside the aqueous space of polymeric nanoreactors.
As proof of principle for the nanoreactor-technology Meier and coworkers
encapsulated the enzyme β-lactamase in PMOXA-PDMS-PMOXA particles. The
enzyme is able to hydrolyze β-lactam antibiotics like ampicillin which in turn
can reduce iodine to iodide. This reduction was used to monitor the activity
and functionality of the enzyme inside the nanoreactor. In this nanoreactormodelsystem the enzyme is encapsulated in the aqueous interior of the
nanoreactor and the substrate ampicillin is transported across the membrane
via the porin OmpF [178]. As these nanoreactors clearly showed enzyme
31
Chapter 1
activity, it was evident that both the channel protein and the encapsulated
enzyme remain functional after incorporation in a nanoreactor system [179,
180].
Also
acetylcholinesterase,
which
hydrolyses
the
neurotransmitter
acetylcholine, could succesfully be entrapped in PMOXA-PDMS-PMOXA particles
whithout loss of function. This enzyme was chosen to provide an easy and
accurate biosensor to detect insecticide residuals [171].
The encapsulation of a prodrug activating enzyme led the nanoreactor strategy
to its first biomedical proof of principle. Prodrug activating enzymes are
promising tools in cancer therapy as they can convert a non-toxic prodrug into
a cytotoxic agent. In Antibody Directed Enzyme Prodrug Therapy (ADEPT), the
enzyme is targeted to the tumor in a fusion with a monoclonal antibody in
order to avoid systemic toxicity. Although this strategy has advantages
compared
to
non-targeted
chemotherapy,
the
major
obstacle
is
the
immunogenicity of the exogenous enzyme. The encapsulation of the prodrug
activating enzyme nucleoside hydrolase of Trypanosoma vivax (TvNH) inside
DOPC/EPG liposomes permeabilized with OmpF was a first attempt to improve
this prodrug therapy [161]. Unfortunately such lipidic particles are unstable in
blood serum and need to be grafted with poly(ethylene glycol) (PEG) or other
polymers to increase the average circulation time. For this reason, a more
promising kind of nanoreactor, composed of PMOXA-PDMS-PMOXA triblock
copolymer was constructed. These nanoreactors encapsulate the prodrug
activating enzyme TvNH and are further functionalized by incorporating
bacterial porins in the reactor wall, thus allowing transport of solutes across
the membrane. The reactors can efficiently hydrolyze different substrates
including the prodrug 2-fluoroadenosine, resulting in the release of the
cytotoxic molecule, 2-fluoroadenine [181].
Recently, more enzyme-containing nanoreactors are being developed for
various biomedical applications (see chapter 1.4.4).
32
Introduction
1.4.2 Production methods
Up until now there are 3 different methods described to prepare functional
nanoreactors: the solvent evaporation method, the lamellar film rehydration
method and the direct dispersion method. The reported preparation methods
are only used on laboratory scale and produce a few milliliters of nanoreactor
suspension.
Solvent evaporation method
In the solvent evaporation method, a solution of purified porin is mixed with
copolymer dissolved in an appropriate solvent (like ethanol), to the desired
molar ratio of protein to polymer. For encapsulation of the enzyme in the
interior of the vesicles, the homogeneous polymer-porin solution is added
dropwise to an aqueous solution containing the enzyme and stirred for several
hours at room temperature. During this incubation period the nanoreactors are
formed by self-assembly and the solvent is evaporated. Then, the resulted
dispersion is repeatedly extruded through filters of a desired pore size. This
leads to monodisperse particles [163].
Lamellar film rehydration method
Block copolymers and porins are mixed in solvent as described in the previous
method. This solution is then dried at the bottom of a glass tube under
vacuum to remove all the remaining solvent. Subsequently, the lamellar film is
rehydrated by adding an aqueous solution containing the enzyme. Finally,
extrusion produces monolamellar and monodisperse vesicles [182].
Direct dispersion method
In the direct dispersion method the polymers are suspended in an aqueous
solution containing the enzyme and gently stirred at room temperature
overnight
to
form
self-assembled
nanocompartments.
Triton
X-100
is
subsequently added and the suspension is sonicated twice for 10 seconds to
achieve destabilization of the polymeric membrane. Porins are added shortly
after the sonication step and will incorporate inside the polymeric wall of the
33
Chapter 1
nanoreactors. Biobeads are added to remove residual detergent [145].
Repeated extrusion through a polycarbonate filter of desired pore size can be
used to obtain uniformly sized vesicles.
In the preparation methods described above, the encapsulation efficiency is
usually around 10% to 30%. Therefore the non-entrapped enzyme molecules
have to be separated from the enzyme-containing vesicles in a last step of the
preparation procedure. This is usually carried out either by size-exclusion
chromatography (e.g. using sepharose 4B), by affinity chromatography or by
dialysis. If the encapsulation efficiency or entrapment efficiency (EE) is
unacceptably low, freeze-thaw cycles can be used to enhance entrapment.
Also, using charged polymers that have affinity for the encapsulated
compound [183] or using higher concentrations of enzyme might improve
encapsulation efficiency [101]. However, in such cases removing the adhering
molecules from the outside of the reactor might be more difficult.
1.4.3 Controlling the nanoreactor activity
To further increase the stability of nanoreactors, the polymer membrane can
be stabilized by crosslinking the polymers. To this end, Nardin et al. used
methacrylated PMOXA-PDMS-PMOXA polymers [170]. These polymers carry
reactive polymerizable methacrylate endgroups. After formation of the
nanoreactors, the polymers were crosslinked via UV irradiation. Since every
polymer molecule carries two polymerizable groups, the polymerization leads
to the formation of crosslinked polymer structures. As a result, the
polymerized
nanocontainers
possess
solid
state
properties
like
shape
persistence and elasticity. This polymerization did not affect the shape of the
reactors nor did it modify the activity of the nanoreactors.
By choosing the appropriate channel forming protein, the translocation of
substrates and products across the membrane can be controlled. Aspecific
porins allow transport of solutes below a size threshold while specific porins
transport specific solutes as discussed above. Additionally, porins like OmpF
and PhoE are voltage-gating porins, which means that they can open and close
34
Introduction
depending on the transmembrane voltage potential, also called the Donnan
potential. When a Donnan potential of above 100 mV is applied, the porins
close [184]. This mechanism has evolved in Gram-negative bacteria to protect
the cell against drastic changes in the environment and the voltage-gating
characteristic can be used to switch on and off the activity of nanoreactors by
external stimuli. The polymer-protein hybrid membrane can be regarded as a
semipermeable membrane separating the internal volume from the external
solution. Nardin et al. successfully controlled the activity of nanoreactors
composed of PMOXA-PDMS-PMOXA polymers through voltage-gating of the
OmpF porin [170]. To apply a Donnan potential they used sodium-polystyrenesulfonate (Na-PSS), a polyectrolyte ion. When Na-PSS is added to the
external solution, a Donnan potential is created since Na-PSS cannot permeate
the membrane. When the Donnan potential exceeded 100 mV, they saw a
complete deactivation of the nanoreactors due to the closure of OmpF. This
closure was reversed by adding NaCl, hereby decreasing the Donnan potential
below 100 mV.
Another way to control the activity of the nanoreactors is by using an enzyme
that is switched on and off by external factors such as pH and temperature.
For instance, it is well known that the extracellular environment of solid
tumors is acidic (with a pH ranging from 6.5 to 7.2) compared to the pH of the
blood and normal tissue (7.5) [185, 186]. By choosing an enzyme that is
inactive at neutral pH but active at mild acidic conditions, the activity of the
nanoreactor is restricted to the extracellular matrix of solid tumors. Broz et al.
demonstrated this method by encapsulating the plant-derived enzyme acid
phosphatase in OmpF permeabilized nanoreactors [187]. This enzyme is pHdependent
and
catalyses
fast
dephosphorylation
of
various
phosphate
substrates at mildly acidic pH ranging from 4-6.5 while it is inactive at highly
acidic, neutral or basic pH. By varying the pH of the surrounding solution, the
nanoreactor was able to change its state of activity as demonstrated by the
conversion of a soluble non-fluorescent substrate into a water-insoluble
fluorescent dye inside the polymeric particle [187]. They successfully created
35
Chapter 1
an environment-controlled and triggerable sensor-effector nanoreactor that is
only active at mildly acidic pH.
1.4.4 Biomedical applications
The delivery of enzymes can be used for a wide range of purposes. Although
most applications reported up until now are applications of polymeric
nanocarriers without channel proteins, we expect that nanoreactors will soon
be exploited as improved drug carriers for the same diagnostic and therapeutic
uses. In this chapter a snapshot of reported applications of nanoreactor-based
strategies are summarized.
Various polymeric drug delivery systems are currently applied or under
development for their use in cancer therapy. Entrapping anti-cancer drugs
inside nanoparticles, allows scientists to minimize drug degradation, to prevent
undesirable side effects exercised on normal cells by cytotoxic drugs and to
increase drug bioavailability [188]. The use of nanoreactors in cancer therapy
makes it possible to encapsulate a prodrug-activating enzyme rather than the
drug itself. In this strategy the free non-toxic prodrug is administered
systemically and is converted to a toxic drug by the active nanoreactors which
can be targeted to the tumor, hence decreasing toxic side effects to a
minimum. The encapsulation of a prodrug-activating enzyme inside polymeric
nanoreactors is a revolutionary new anti-cancer therapy introduced by Ranquin
et al. [181]. This work opens the door to the encapsulation of various other
prodrug-activating enzymes.
Because of their cell-targeting potential, nanoreactors can also be used as
diagnostic tools and targeted drug delivery systems for a wide range of
diseases. For example, Broz and coworkers constructed nanoreactors that may
be of particular value for the detection and treatment of vulnerable plaque
macrophages in atherosclerotic diseases [189]. Current diagnostic techniques
for the detection of plaque macrophages are often limited by insufficient
sensitivity and selectivity. Nanoreactors with macrophage-specific delivery
[190] are promising candidates for improved diagnostic strategies. In addition,
36
Introduction
such nanoreactors were evaluated for targeted drug delivery. Receptor-specific
targeting using statin-loaded nanometer-sized triblock copolymer vesicles with
macrophage targeting moieties were constructed to rupture macrophage-rich
atherosclerotic plaques [191]. The formation of such plaques in the coronary
arteries is the main cause of heart attacks.
Polymeric vesicles can also be used as new drug carrier systems for brain
delivery. Targeting drugs to the brain by crossing the blood-brain barrier
(BBB) is quite a challenge. Adequate brain delivery systems must have longcirculation
properties
and
appropriate
surface
characteristics
to
allow
interactions with BBB endothelial cells. Several studies indicated that the
surface modification of particles by coating them with polysorbate-80 or
polyethylene glycol is effective in brain drug delivery [192, 193]. By choosing
the right surface polymer, enzyme-loaded nanoreactors could have great
potential in the treatment of metabolic brain diseases, such as Lesch-Nyhan
syndrome. This is a neurogenetic disorder caused by the deficiency of the
purine
salvage
enzyme
hypoxanthine-guanine
phosphoribosyltransferase
(HPRT).
Nanoreactors can also be applied as biosensors. Although using lipids instead
of polymers, Vamvakaki et al. used the nanoreactor-strategy to develop a
novel nano-biosensor for pesticide analysis. The enzyme acetylcholinesterase
(AChE) is encapsulated in the internal environment of liposomes and substrate
entrance is facilitated by the incorporation of the porin OmpF (Figure 1-12).
Within the liposomes, the pH sensitive fluorescent indicator pyranine was also
immobilized for the optical transduction of the enzymatic activity. Increasing
amounts of pesticides lead to a decrease of the enzymatic activity for the
hydrolysis of the acetylcholine and thus to a decrease in the fluorescent signal
of the pH indicator. This biosensor system has been applied successfully for
the detection of two widely used organophosphorus pesticides in drinking
water [194].
37
Chapter 1
Figure 1-12: Schematic diagram of the acetylcholinesterase-based liposome biosensor
for pesticide detection.
Microparticle systems are also developed to produce crystals under
controlled conditions. Michel and coworkers used liposomes containing alkaline
phophatase as the encapsulated enzyme. This enzyme converts p-nitrophenyl
phosphate into phosphate and subsequently into calcium phosphate crystals,
since Ca2+ ions are already present inside the vesicles [195]. In this case the
substrate diffuses through the lipidic membrane without the help of channel
proteins. Reconstitution of the phosphate-induced PhoE porin might be an
improvement of this reactor set-up [196]. Sauer et al. reported a quiet similar
system of ion carrier-assisted precipitation of calcium phosphate in giant
PMOXA-PDMS-PMOXA
(Lasalocid
A,
vesicles.
Alamethicin
In
and
this
study
three
different
ionofores
N,N-dicyclohexyl-N’,N’-dioctadecyl-3-
oxapentane-1,5-diamide) were used for selective or unselective calcium
transport over the polymeric membrane into the intravesicular space. Calcium
phosphate crystals start to grow at the inner surface of the membrane [197].
This strategy allows the inorganic particles to grow only in restricted regions
and opens the route to control crystal size in biomimetic systems.
In addition, antioxidant nanoreactors based on encapsulated superoxide
dismutase (SOD) were constructed [198, 199]. The polymeric membrane of
the nanoreactors is capable of protecting the SOD and allowing penetration of
the inner space by dioxygen, where the conversion to hydrogen peroxide and
oxygen is catalyzed. In this example, the level of superoxide permeability
38
Introduction
obviates the necessity of inserting channel proteins. Oxidative stress has been
shown to play a significant role in many disease states including arthritis,
Parkinson’s disease, cancer and AIDS. Therefore, such nanoreactors open a
new direction to the development of new antioxidant drugs.
The nanoreactor system can also be applied as a novel method for the
construction of biologically active surfaces for analytics and sensors with
space-controlled
reactions.
Recently,
biotinylated
nanoreactors
were
immobilized on a glass substrate. The surface of the glass substrate was first
structured with bovine serum albumin labeled with biotin. Exposure of
biotinylated protein patterns to streptavidin followed by incubation with
biotinylated particles lead to their immobilization. The model reaction chosen
for enzymatic conversion inside the nanoreactors was the dephosphorylation of
the fluorogenic substrate ELF97 by acid phosphatase (Figure 1-13) [200].
Figure 1-13: Schematic representation of the immobilization of a nanoreactor system
on a glass surface. Acid phosphatase was encapsulated in the OmpF bearing biotinylated
vesicles and immobilized on the surface of the glass structure with streptavidin (taken
from [200]).
39
Chapter 1
From the above it is clear that nanoreactors possess great potential in various
biomedical applications. Nevertheless, this technology is still in the infantile
stage. At the moment nanoreactors are only tested in cell cultures and to our
knowledge there are no published results about the behavior of this type of
polymer vesicles in a real in vivo environment. Crucial points toward future
use of nanoreactors in vivo are biodistribution, biodegradability, efficacy and
short-term and long-term toxicity.
1.5 References
[1] R. Hirschhorn, Adenosine deaminase deficiency: molecular basis and
recent developments, Clin Immunol Immunopathol, 76 (1995) S219-227.
[2] R.O. Brady, R. Schiffmann, Clinical features of and recent advances in
therapy for Fabry disease, Jama, 284 (2000) 2771-2775.
[3] N.A. van der Beek, M.L. Hagemans, A.T. van der Ploeg, A.J. Reuser, P.A.
van Doorn, Pompe disease (glycogen storage disease type II): clinical features
and enzyme replacement therapy, Acta Neurol Belg, 106 (2006) 82-86.
[4] E. Beutler, Gaucher disease: new molecular approaches to diagnosis and
treatment, Science, 256 (1992) 794-799.
[5] P. Guggenbuhl, B. Grosbois, G. Chales, Gaucher disease, Joint Bone Spine,
75 (2008) 116-124.
[6] H.S. Scott, S. Bunge, A. Gal, L.A. Clarke, C.P. Morris, J.J. Hopwood,
Molecular genetics of mucopolysaccharidosis type I: diagnostic, clinical, and
biological implications, Hum Mutat, 6 (1995) 288-302.
[7] R.A. Harkness, G.M. McCreanor, R.W. Watts, Lesch-Nyhan syndrome and
its pathogenesis: purine concentrations in plasma and urine with metabolite
profiles in CSF, J Inherit Metab Dis, 11 (1988) 239-252.
[8] I. Nishino, A. Spinazzola, M. Hirano, Thymidine phosphorylase gene
mutations in MNGIE, a human mitochondrial disorder, Science, 283 (1999)
689-692.
[9] N.S. Brown, R. Bicknell, Thymidine phosphorylase, 2-deoxy-D-ribose and
angiogenesis, Biochem J, 334 ( Pt 1) (1998) 1-8.
[10] A. Yoshimura, Y. Kuwazuru, T. Furukawa, H. Yoshida, K. Yamada, S.
Akiyama, Purification and tissue distribution of human thymidine
phosphorylase; high expression in lymphocytes, reticulocytes and tumors,
Biochim Biophys Acta, 1034 (1990) 107-113.
[11] J. Waltenberger, K. Usuki, B. Fellstrom, K. Funa, C.H. Heldin, Plateletderived endothelial cell growth factor. Pharmacokinetics, organ distribution
and degradation after intravenous administration in rats, FEBS Lett, 313
(1992) 129-132.
[12] P.A. Eccleston, K. Funa, C.H. Heldin, Neurons of the peripheral nervous
system express thymidine phosphorylase, Neurosci Lett, 192 (1995) 137-141.
[13] S.B. Fox, A. Moghaddam, M. Westwood, H. Turley, R. Bicknell, K.C.
Gatter, A.L. Harris, Platelet-derived endothelial cell growth factor/thymidine
40
Introduction
phosphorylase expression in normal tissues: an immunohistochemical study, J
Pathol, 176 (1995) 183-190.
[14] K. Matsukawa, A. Moriyama, Y. Kawai, K. Asai, T. Kato, Tissue
distribution of human gliostatin/platelet-derived endothelial cell growth factor
(PD-ECGF) and its drug-induced expression, Biochim Biophys Acta, 1314
(1996) 71-82.
[15] L. Griffiths, I.J. Stratford, Platelet-derived endothelial cell growth factor
thymidine phosphorylase in tumour growth and response to therapy, Br J
Cancer, 76 (1997) 689-693.
[16] F. Focher, D. Ubiali, M. Pregnolato, C. Zhi, J. Gambino, G.E. Wright, S.
Spadari, Novel nonsubstrate inhibitors of human thymidine phosphorylase, a
potential target for tumor-dependent angiogenesis, J Med Chem, 43 (2000)
2601-2607.
[17] N.W. Barton, R.O. Brady, J.M. Dambrosia, A.M. Di Bisceglie, S.H. Doppelt,
S.C. Hill, H.J. Mankin, G.J. Murray, R.I. Parker, C.E. Argoff, et al., Replacement
therapy
for
inherited
enzyme
deficiency--macrophage-targeted
glucocerebrosidase for Gaucher's disease, N Engl J Med, 324 (1991) 14641470.
[18] F. Ishikawa, K. Miyazono, U. Hellman, H. Drexler, C. Wernstedt, K.
Hagiwara, K. Usuki, F. Takaku, W. Risau, C.H. Heldin, Identification of
angiogenic activity and the cloning and expression of platelet-derived
endothelial cell growth factor, Nature, 338 (1989) 557-562.
[19] H. Koide, T. Asai, K. Hatanaka, S. Akai, T. Ishii, E. Kenjo, T. Ishida, H.
Kiwada, H. Tsukada, N. Oku, T cell-independent B cell response is responsible
for ABC phenomenon induced by repeated injection of PEGylated liposomes,
Int J Pharm.
[20] S. Akiyama, T. Furukawa, T. Sumizawa, Y. Takebayashi, Y. Nakajima, S.
Shimaoka, M. Haraguchi, The role of thymidine phosphorylase, an angiogenic
enzyme, in tumor progression, Cancer Sci, 95 (2004) 851-857.
[21] A. Walter, G. Kuehl, K. Barnes, G. VanderWaerdt, The vesicle-to-micelle
transition of phosphatidylcholine vesicles induced by nonionic detergents:
effects of sodium chloride, sucrose and urea, Biochim Biophys Acta, 1508
(2000) 20-33.
[22] M.J. Pugmire, S.E. Ealick, The crystal structure of pyrimidine nucleoside
phosphorylase in a closed conformation, Structure, 6 (1998) 1467-1479.
[23] R.A. Norman, S.T. Barry, M. Bate, J. Breed, J.G. Colls, R.J. Ernill, R.W.
Luke, C.A. Minshull, M.S. McAlister, E.J. McCall, H.H. McMiken, D.S. Paterson,
D. Timms, J.A. Tucker, R.A. Pauptit, Crystal structure of human thymidine
phosphorylase in complex with a small molecule inhibitor, Structure, 12 (2004)
75-84.
[24] J. Mendieta, S. Martin-Santamaria, E.M. Priego, J. Balzarini, M.J.
Camarasa, M.J. Perez-Perez, F. Gago, Role of histidine-85 in the catalytic
mechanism of thymidine phosphorylase as assessed by targeted molecular
dynamics simulations and quantum mechanical calculations, Biochemistry, 43
(2004) 405-414.
[25] M. Hirano, J. Garcia-de-Yebenes, A.C. Jones, I. Nishino, S. DiMauro, J.R.
Carlo, A.N. Bender, A.F. Hahn, L.M. Salberg, D.E. Weeks, T.G. Nygaard,
Mitochondrial neurogastrointestinal encephalomyopathy syndrome maps to
chromosome 22q13.32-qter, Am J Hum Genet, 63 (1998) 526-533.
41
Chapter 1
[26] A.J. Berk, D.A. Clayton, A genetically distinct thymidine kinase in
mammalian mitochondria. Exclusive labeling of mitochondrial deoxyribonucleic
acid, J Biol Chem, 248 (1973) 2722-2729.
[27] R.K. Bestwick, C.K. Mathews, Unusual compartmentation of precursors
for nuclear and mitochondrial DNA in mouse L cells, J Biol Chem, 257 (1982)
9305-9308.
[28] R.K. Bestwick, G.L. Moffett, C.K. Mathews, Selective expansion of
mitochondrial nucleoside triphosphate pools in antimetabolite-treated HeLa
cells, J Biol Chem, 257 (1982) 9300-9304.
[29] P. Ferraro, G. Pontarin, L. Crocco, S. Fabris, P. Reichard, V. Bianchi,
Mitochondrial deoxynucleotide pools in quiescent fibroblasts: a possible model
for mitochondrial neurogastrointestinal encephalomyopathy (MNGIE), J Biol
Chem, 280 (2005) 24472-24480.
[30] D. Bogenhagen, D.A. Clayton, Thymidylate nucleotide supply for
mitochondrial DNA synthesis in mouse L-cells. Effect of 5-fluorodeoxyuridine
and methotrexate in thymidine kinase plus and thymidine kinase minus cells, J
Biol Chem, 251 (1976) 2938-2944.
[31] D.L. Croteau, R.H. Stierum, V.A. Bohr, Mitochondrial DNA repair
pathways, Mutat Res, 434 (1999) 137-148.
[32] S. DiMauro, E.A. Schon, Mitochondrial disorders in the nervous system,
Annu Rev Neurosci, 31 (2008) 91-123.
[33] I. Nishino, A. Spinazzola, M. Hirano, MNGIE: from nuclear DNA to
mitochondrial DNA, Neuromuscul Disord, 11 (2001) 7-10.
[34] Y.C. Kocaefe, S. Erdem, M. Ozguc, E. Tan, Four novel thymidine
phosphorylase gene mutations in mitochondrial neurogastrointestinal
encephalomyopathy syndrome (MNGIE) patients, Eur J Hum Genet, 11 (2003)
102-104.
[35] G. Pontarin, P. Ferraro, M.L. Valentino, M. Hirano, P. Reichard, V. Bianchi,
Mitochondrial DNA depletion and thymidine phosphate pool dynamics in a
cellular model of mitochondrial neurogastrointestinal encephalomyopathy, J
Biol Chem, 281 (2006) 22720-22728.
[36] R. Laforce, Jr., P.N. Valdmanis, N. Dupre, G.A. Rouleau, A.F. Turgeon, M.
Savard, A novel TYMP mutation in a French Canadian patient with
mitochondrial
neurogastrointestinal
encephalomyopathy,
Clin
Neurol
Neurosurg, 111 (2009) 691-694.
[37] J. Gamez, M.C. Lara, F. Mearin, C. Oliveras-Ley, N. Raguer, M. Olive, A.T.
Leist, A. Perello, M. Perona, C. Cervera, A.L. Andreu, R. Marti, M. Hirano, A
novel thymidine phosphorylase mutation in a Spanish MNGIE patient, J Neurol
Sci, 228 (2005) 35-39.
[38] G. Phear, J. Nalbantoglu, M. Meuth, Next-nucleotide effects in mutations
driven by DNA precursor pool imbalances at the aprt locus of Chinese hamster
ovary cells, Proc Natl Acad Sci U S A, 84 (1987) 4450-4454.
[39] E.C. Butcher, E.L. Berg, E.J. Kunkel, Systems biology in drug discovery,
Nat Biotechnol, 22 (2004) 1253-1259.
[40] P.J. Wilderman, A.I. Vasil, Z. Johnson, M.L. Vasil, Genetic and
biochemical analyses of a eukaryotic-like phospholipase D of Pseudomonas
aeruginosa suggest horizontal acquisition and a role for persistence in a
chronic pulmonary infection model, Mol Microbiol, 39 (2001) 291-303.
42
Introduction
[41] S. Ferraris, S. Clark, E. Garelli, G. Davidzon, S.A. Moore, R.H. Kardon,
R.J. Bienstock, M.J. Longley, M. Mancuso, P. Gutierrez Rios, M. Hirano, W.C.
Copeland, S. DiMauro, Progressive external ophthalmoplegia and vision and
hearing loss in a patient with mutations in POLG2 and OPA1, Arch Neurol, 65
(2008) 125-131.
[42] Y. Nishigaki, R. Marti, W.C. Copeland, M. Hirano, Site-specific somatic
mitochondrial DNA point mutations in patients with thymidine phosphorylase
deficiency, J Clin Invest, 111 (2003) 1913-1921.
[43] S.E. Lim, W.C. Copeland, Differential incorporation and removal of
antiviral deoxynucleotides by human DNA polymerase gamma, J Biol Chem,
276 (2001) 23616-23623.
[44] Y. Nishigaki, R. Marti, M. Hirano, ND5 is a hot-spot for multiple atypical
mitochondrial
DNA
deletions
in
mitochondrial
neurogastrointestinal
encephalomyopathy, Hum Mol Genet, 13 (2004) 91-101.
[45] M. Hirano, G. Silvestri, D.M. Blake, A. Lombes, C. Minetti, E. Bonilla, A.P.
Hays, R.E. Lovelace, I. Butler, T.E. Bertorini, et al., Mitochondrial
neurogastrointestinal encephalomyopathy (MNGIE): clinical, biochemical, and
genetic features of an autosomal recessive mitochondrial disorder, Neurology,
44 (1994) 721-727.
[46] M.C. Lara, M.L. Valentino, J. Torres-Torronteras, M. Hirano, R. Marti,
Mitochondrial neurogastrointestinal encephalomyopathy (MNGIE): biochemical
features and therapeutic approaches, Biosci Rep, 27 (2007) 151-163.
[47] M. Hirano, Y. Nishigaki, R. Marti, Mitochondrial neurogastrointestinal
encephalomyopathy (MNGIE): a disease of two genomes, Neurologist, 10
(2004) 8-17.
[48] A. Spinazzola, R. Marti, I. Nishino, A.L. Andreu, A. Naini, S. Tadesse, I.
Pela, E. Zammarchi, M.A. Donati, J.A. Oliver, M. Hirano, Altered thymidine
metabolism due to defects of thymidine phosphorylase, J Biol Chem, 277
(2002) 4128-4133.
[49] R. Marti, A. Spinazzola, S. Tadesse, I. Nishino, Y. Nishigaki, M. Hirano,
Definitive diagnosis of mitochondrial neurogastrointestinal encephalomyopathy
by biochemical assays, Clin Chem, 50 (2004) 120-124.
[50] R. Marti, J.J. Verschuuren, A. Buchman, I. Hirano, S. Tadesse, A.B. van
Kuilenburg, A.H. van Gennip, B.J. Poorthuis, M. Hirano, Late-onset MNGIE due
to partial loss of thymidine phosphorylase activity, Ann Neurol, 58 (2005) 649652.
[51] D.L. Dorset, A. Engel, M. Haner, A. Massalski, J.P. Rosenbusch, Twodimensional crystal packing of matrix porin. A channel forming protein in
Escherichia coli outer membranes, J Mol Biol, 165 (1983) 701-710.
[52] L.C. Lopez, H.O. Akman, A. Garcia-Cazorla, B. Dorado, R. Marti, I.
Nishino, S. Tadesse, G. Pizzorno, D. Shungu, E. Bonilla, K. Tanji, M. Hirano,
Unbalanced deoxynucleotide pools cause mitochondrial DNA instability in
thymidine phosphorylase-deficient mice, Hum Mol Genet, 18 (2009) 714-722.
[53] D. Cao, J.J. Leffert, J. McCabe, B. Kim, G. Pizzorno, Abnormalities in
uridine homeostatic regulation and pyrimidine nucleotide metabolism as a
consequence of the deletion of the uridine phosphorylase gene, J Biol Chem,
280 (2005) 21169-21175.
[54] M.L. Valentino, R. Marti, S. Tadesse, L.C. Lopez, J.L. Manes, J. Lyzak, A.
Hahn, V. Carelli, M. Hirano, Thymidine and deoxyuridine accumulate in tissues
43
Chapter 1
of patients with mitochondrial neurogastrointestinal encephalomyopathy
(MNGIE), FEBS Lett, 581 (2007) 3410-3414.
[55] P.F. Chinnery, J. Vissing, Treating MNGIE: is reducing blood nucleosides
the first cure for a mitochondrial disorder?, Neurology, 67 (2006) 1330-1332.
[56] G. la Marca, S. Malvagia, B. Casetta, E. Pasquini, I. Pela, M. Hirano, M.A.
Donati, E. Zammarchi, Pre- and post-dialysis quantitative dosage of thymidine
in urine and plasma of a MNGIE patient by using HPLC-ESI-MS/MS, J Mass
Spectrom, 41 (2006) 586-592.
[57] C. Desgranges, G. Razaka, M. Rabaud, P. Picard, F. Dupuch, H. Bricaud,
The human blood platelet: a cellular model to study the degradation of
thymidine and its inhibition, Biochem Pharmacol, 31 (1982) 2755-2759.
[58] M.C. Lara, B. Weiss, I. Illa, P. Madoz, L. Massuet, A.L. Andreu, M.L.
Valentino, Y. Anikster, M. Hirano, R. Marti, Infusion of platelets transiently
reduces nucleoside overload in MNGIE, Neurology, 67 (2006) 1461-1463.
[59] M. Hirano, R. Marti, C. Casali, S. Tadesse, T. Uldrick, B. Fine, D.M.
Escolar, M.L. Valentino, I. Nishino, C. Hesdorffer, J. Schwartz, R.G. Hawks,
D.L. Martone, M.S. Cairo, S. DiMauro, M. Stanzani, J.H. Garvin, Jr., D.G.
Savage,
Allogeneic
stem
cell
transplantation
corrects
biochemical
derangements in MNGIE, Neurology, 67 (2006) 1458-1460.
[60] B.E. Bax, M.D. Bain, L.D. Fairbanks, A.D. Webster, P.W. Ind, M.S.
Hershfield, R.A. Chalmers, A 9-yr evaluation of carrier erythrocyte
encapsulated adenosine deaminase (ADA) therapy in a patient with adult-type
ADA deficiency, Eur J Haematol, 79 (2007) 338-348.
[61] N.F. Moran, M.D. Bain, M.M. Muqit, B.E. Bax, Carrier erythrocyte
entrapped thymidine phosphorylase therapy for MNGIE, Neurology, 71 (2008)
686-688.
[62] R.O. Brady, R. Schiffmann, Enzyme-replacement therapy for metabolic
storage disorders, Lancet Neurol, 3 (2004) 752-756.
[63] J.B. Rose, I.R. Coe, Physiology of nucleoside transporters: back to the
future, Physiology (Bethesda), 23 (2008) 41-48.
[64] E. Kakkis, T. Lester, R. Yang, C. Tanaka, V. Anand, J. Lemontt, M.
Peinovich, M. Passage, Successful induction of immune tolerance to enzyme
replacement therapy in canine mucopolysaccharidosis I, Proc Natl Acad Sci U S
A, 101 (2004) 829-834.
[65] N.W. Barton, F.S. Furbish, G.J. Murray, M. Garfield, R.O. Brady,
Therapeutic response to intravenous infusions of glucocerebrosidase in a
patient with Gaucher disease, Proc Natl Acad Sci U S A, 87 (1990) 1913-1916.
[66] W.R. Wilcox, M. Banikazemi, N. Guffon, S. Waldek, P. Lee, G.E. Linthorst,
R.J. Desnick, D.P. Germain, Long-term safety and efficacy of enzyme
replacement therapy for Fabry disease, Am J Hum Genet, 75 (2004) 65-74.
[67] E.D. Kakkis, J. Muenzer, G.E. Tiller, L. Waber, J. Belmont, M. Passage, B.
Izykowski, J. Phillips, R. Doroshow, I. Walot, R. Hoft, E.F. Neufeld, Enzymereplacement therapy in mucopolysaccharidosis I, N Engl J Med, 344 (2001)
182-188.
[68] A. Amalfitano, A.R. Bengur, R.P. Morse, J.M. Majure, L.E. Case, D.L.
Veerling, J. Mackey, P. Kishnani, W. Smith, A. McVie-Wylie, J.A. Sullivan, G.E.
Hoganson, J.A. Phillips, 3rd, G.B. Schaefer, J. Charrow, R.E. Ware, E.H.
Bossen, Y.T. Chen, Recombinant human acid alpha-glucosidase enzyme
44
Introduction
therapy for infantile glycogen storage disease type II: results of a phase I/II
clinical trial, Genet Med, 3 (2001) 132-138.
[69] M.A. Hamdan, M.H. Almalik, H.M. Mirghani, Early administration of
enzyme replacement therapy for Pompe disease: Short-term follow-up results,
J Inherit Metab Dis, (2008).
[70] M. Beck, New therapeutic options for lysosomal storage disorders:
enzyme replacement, small molecules and gene therapy, Hum Genet, 121
(2007) 1-22.
[71] T.A. Burrow, R.J. Hopkin, N.D. Leslie, B.T. Tinkle, G.A. Grabowski,
Enzyme reconstitution/replacement therapy for lysosomal storage diseases,
Curr Opin Pediatr, 19 (2007) 628-635.
[72] J.E. Wraith, Limitations of enzyme replacement therapy: current and
future, J Inherit Metab Dis, 29 (2006) 442-447.
[73] G. Walsh, Biopharmaceutical benchmarks--2003, Nat Biotechnol, 21
(2003) 865-870.
[74] H. Schellekens, Immunogenicity of therapeutic proteins: clinical
implications and future prospects, Clin Ther, 24 (2002) 1720-1740; discussion
1719.
[75] F.M. Veronese, G. Pasut, PEGylation, successful approach to drug
delivery, Drug Discov Today, 10 (2005) 1451-1458.
[76] J.M. Harris, R.B. Chess, Effect of pegylation on pharmaceuticals, Nat Rev
Drug Discov, 2 (2003) 214-221.
[77] M.J. Roberts, M.D. Bentley, J.M. Harris, Chemistry for peptide and protein
PEGylation, Adv Drug Deliv Rev, 54 (2002) 459-476.
[78] P. Caliceti, F.M. Veronese, Pharmacokinetic and biodistribution properties
of poly(ethylene glycol)-protein conjugates, Adv Drug Deliv Rev, 55 (2003)
1261-1277.
[79] V. Gaberc-Porekar, I. Zore, B. Podobnik, V. Menart, Obstacles and pitfalls
in the PEGylation of therapeutic proteins, Curr Opin Drug Discov Devel, 11
(2008) 242-250.
[80] X. Wang, T. Ishida, H. Kiwada, Anti-PEG IgM elicited by injection of
liposomes is involved in the enhanced blood clearance of a subsequent dose of
PEGylated liposomes, J Control Release, 119 (2007) 236-244.
[81] T. Ishida, X. Wang, T. Shimizu, K. Nawata, H. Kiwada, PEGylated
liposomes elicit an anti-PEG IgM response in a T cell-independent manner, J
Control Release, 122 (2007) 349-355.
[82] Y. Levy, M.S. Hershfield, C. Fernandez-Mejia, S.H. Polmar, D. Scudiery,
M. Berger, R.U. Sorensen, Adenosine deaminase deficiency with late onset of
recurrent infections: response to treatment with polyethylene glycol-modified
adenosine deaminase, J Pediatr, 113 (1988) 312-317.
[83] M.L. Graham, Pegaspargase: a review of clinical studies, Adv Drug Deliv
Rev, 55 (2003) 1293-1302.
[84] Y.S. Wang, S. Youngster, M. Grace, J. Bausch, R. Bordens, D.F. Wyss,
Structural and biological characterization of pegylated recombinant interferon
alpha-2b and its therapeutic implications, Adv Drug Deliv Rev, 54 (2002) 547570.
[85] P. Bailon, A. Palleroni, C.A. Schaffer, C.L. Spence, W.J. Fung, J.E. Porter,
G.K. Ehrlich, W. Pan, Z.X. Xu, M.W. Modi, A. Farid, W. Berthold, M. Graves,
Rational design of a potent, long-lasting form of interferon: a 40 kDa branched
45
Chapter 1
polyethylene glycol-conjugated interferon alpha-2a for the treatment of
hepatitis C, Bioconjug Chem, 12 (2001) 195-202.
[86] O.B. Kinstler, D.N. Brems, S.L. Lauren, A.G. Paige, J.B. Hamburger, M.J.
Treuheit, Characterization and stability of N-terminally PEGylated rhG-CSF,
Pharm Res, 13 (1996) 996-1002.
[87] P.J. Trainer, W.M. Drake, L. Katznelson, P.U. Freda, V. Herman-Bonert,
A.J. van der Lely, E.V. Dimaraki, P.M. Stewart, K.E. Friend, M.L. Vance, G.M.
Besser, J.A. Scarlett, M.O. Thorner, C. Parkinson, A. Klibanski, J.S. Powell, A.L.
Barkan, M.C. Sheppard, M. Malsonado, D.R. Rose, D.R. Clemmons, G.
Johannsson, B.A. Bengtsson, S. Stavrou, D.L. Kleinberg, D.M. Cook, L.S.
Phillips, M. Bidlingmaier, C.J. Strasburger, S. Hackett, K. Zib, W.F. Bennett,
R.J. Davis, Treatment of acromegaly with the growth hormone-receptor
antagonist pegvisomant, N Engl J Med, 342 (2000) 1171-1177.
[88] S.D. Caruthers, S.A. Wickline, G.M. Lanza, Nanotechnological applications
in medicine, Curr Opin Biotechnol, 18 (2007) 26-30.
[89] S. Sandhiya, S.A. Dkhar, A. Surendiran, Emerging trends of
nanomedicine: an overview, Fundam Clin Pharmacol, 23 (2009) 263-269.
[90] Y. Shi, L.C. Li, Current advances in sustained-release systems for
parenteral drug delivery, Expert Opin Drug Deliv, 2 (2005) 1039-1058.
[91] G. Gregoriadis, Drug entrapment in liposomes, FEBS Lett, 36 (1973) 292296.
[92] C.J. Chu, J. Dijkstra, M.Z. Lai, K. Hong, F.C. Szoka, Efficiency of
cytoplasmic delivery by pH-sensitive liposomes to cells in culture, Pharm Res,
7 (1990) 824-834.
[93] K. Kono, Thermosensitive polymer-modified liposomes, Adv Drug Deliv
Rev, 53 (2001) 307-319.
[94] R.H. Bisby, C. Mead, C.G. Morgan, Active uptake of drugs into
photosensitive liposomes and rapid release on UV photolysis, Photochem
Photobiol, 72 (2000) 57-61.
[95] P. Goyal, K. Goyal, S.G. Vijaya Kumar, A. Singh, O.P. Katare, D.N.
Mishra, Liposomal drug delivery systems--clinical applications, Acta Pharm, 55
(2005) 1-25.
[96] M.L. Immordino, F. Dosio, L. Cattel, Stealth liposomes: review of the
basic science, rationale, and clinical applications, existing and potential, Int J
Nanomedicine, 1 (2006) 297-315.
[97] P.R. Veerareddy, V. Vobalaboina, Lipid-based formulations of
amphotericin B, Drugs Today (Barc), 40 (2004) 133-145.
[98] D.S. Alberts, F.M. Muggia, J. Carmichael, E.P. Winer, M. Jahanzeb, A.P.
Venook, K.M. Skubitz, E. Rivera, J.A. Sparano, N.J. DiBella, S.J. Stewart, J.J.
Kavanagh, A.A. Gabizon, Efficacy and safety of liposomal anthracyclines in
phase I/II clinical trials, Semin Oncol, 31 (2004) 53-90.
[99] T.M. Allen, F.J. Martin, Advantages of liposomal delivery systems for
anthracyclines, Semin Oncol, 31 (2004) 5-15.
[100] G. Gregoriadis, Enzyme entrapment in liposomes, Methods Enzymol, 44
(1976) 218-227.
[101] P. Walde, S. Ichikawa, Enzymes inside lipid vesicles: preparation,
reactivity and applications, Biomol Eng, 18 (2001) 143-177.
[102] G. Gregoriadis, P.D. Leathwood, B.E. Ryman, Enzyme entrapment in
liposomes, FEBS Lett, 14 (1971) 95-99.
46
Introduction
[103] C.M. Colley, B.E. Ryman, The use of a liposomally entrapped enzyme in
the treatment of an artificial storage condition, Biochim Biophys Acta, 451
(1976) 417-425.
[104] G.C. Reynolds, H.J. Baker, R.H. Reynolds, Enzyme replacement using
liposome carriers in feline Gm1 gangliosidosis fibroblasts, Nature, 275 (1978)
754-755.
[105] P.E. Belchetz, J.C. Crawley, I.P. Braidman, G. Gregoriadis, Treatment of
Gaucher's
disease
with
liposome-entrapped
glucocerebroside:
betaglucosidase, Lancet, 2 (1977) 116-117.
[106] L.D. Steger, R.J. Desnick, Enzyme therapy. VI: Comparative in vivo
fates and effects on lysosomal integrity of enzyme entrapped in negatively and
positively charged liposomes, Biochim Biophys Acta, 464 (1977) 530-546.
[107] T.M. Allen, C. Hansen, J. Rutledge, Liposomes with prolonged circulation
times: factors affecting uptake by reticuloendothelial and other tissues,
Biochim Biophys Acta, 981 (1989) 27-35.
[108] T. Ishida, H. Harashima, H. Kiwada, Liposome clearance, Biosci Rep, 22
(2002) 197-224.
[109] Y.J. Kao, R.L. Juliano, Interactions of liposomes with the
reticuloendothelial system. Effects of reticuloendothelial blockade on the
clearance of large unilamellar vesicles, Biochim Biophys Acta, 677 (1981) 453461.
[110] J.H. Senior, Fate and behavior of liposomes in vivo: a review of
controlling factors, Crit Rev Ther Drug Carrier Syst, 3 (1987) 123-193.
[111] J. Szebeni, The interaction of liposomes with the complement system,
Crit Rev Ther Drug Carrier Syst, 15 (1998) 57-88.
[112] T.M. Allen, A study of phospholipid interactions between high-density
lipoproteins and small unilamellar vesicles, Biochim Biophys Acta, 640 (1981)
385-397.
[113] A.L. Klibanov, K. Maruyama, V.P. Torchilin, L. Huang, Amphipathic
polyethyleneglycols effectively prolong the circulation time of liposomes, FEBS
Lett., 268 (1990) 235-237.
[114] G. Blume, G. Cevc, Molecular mechanism of the lipid vesicle longevity in
vivo, Biochim Biophys Acta, 1146 (1993) 157-168.
[115] D. Papahadjopoulos, T.M. Allen, A. Gabizon, E. Mayhew, K. Matthay,
S.K. Huang, K.D. Lee, M.C. Woodle, D.D. Lasic, C. Redemann, et al., Sterically
stabilized liposomes: improvements in pharmacokinetics and antitumor
therapeutic efficacy, Proc Natl Acad Sci U S A, 88 (1991) 11460-11464.
[116] A. Gabizon, D. Papahadjopoulos, Liposome formulations with prolonged
circulation time in blood and enhanced uptake by tumors, Proc Natl Acad Sci U
S A, 85 (1988) 6949-6953.
[117] L. Cattel, M. Ceruti, F. Dosio, From conventional to stealth liposomes: a
new frontier in cancer chemotherapy, Tumori, 89 (2003) 237-249.
[118] A.A. Gabizon, Y. Barenholz, M. Bialer, Prolongation of the circulation
time of doxorubicin encapsulated in liposomes containing a polyethylene
glycol-derivatized phospholipid: pharmacokinetic studies in rodents and dogs,
Pharm Res, 10 (1993) 703-708.
[119] T. Boulikas, G.P. Stathopoulos, N. Volakakis, M. Vougiouka, Systemic
Lipoplatin infusion results in preferential tumor uptake in human studies,
Anticancer Res, 25 (2005) 3031-3039.
47
Chapter 1
[120] E.S. Kim, C. Lu, F.R. Khuri, M. Tonda, B.S. Glisson, D. Liu, M. Jung,
W.K. Hong, R.S. Herbst, A phase II study of STEALTH cisplatin (SPI-77) in
patients with advanced non-small cell lung cancer, Lung Cancer, 34 (2001)
427-432.
[121] T.M. Allen, I. Ahmad, D.E. Lopes de Menezes, E.H. Moase,
Immunoliposome-mediated targeting of anti-cancer drugs in vivo, Biochem
Soc Trans, 23 (1995) 1073-1079.
[122] O.P. Medina, Y. Zhu, K. Kairemo, Targeted liposomal drug delivery in
cancer, Curr Pharm Des, 10 (2004) 2981-2989.
[123] P. Sapra, T.M. Allen, Ligand-targeted liposomal anticancer drugs, Prog
Lipid Res, 42 (2003) 439-462.
[124] K. Maruyama, O. Ishida, T. Takizawa, K. Moribe, Possibility of active
targeting to tumor tissues with liposomes, Adv Drug Deliv Rev, 40 (1999) 89102.
[125] V.P. Torchilin, T.S. Levchenko, K.R. Whiteman, A.A. Yaroslavov, A.M.
Tsatsakis, A.K. Rizos, E.V. Michailova, M.I. Shtilman, Amphiphilic poly-Nvinylpyrrolidones: synthesis, properties and liposome surface modification,
Biomaterials, 22 (2001) 3035-3044.
[126] X. Mu, Z. Zhong, Preparation and properties of poly(vinyl alcohol)stabilized liposomes, Int J Pharm, 318 (2006) 55-61.
[127] V.P. Torchilin, M.I. Shtilman, V.S. Trubetskoy, K. Whiteman, A.M.
Milstein, Amphiphilic vinyl polymers effectively prolong liposome circulation
time in vivo, Biochim Biophys Acta, 1195 (1994) 181-184.
[128] M.C. Woodle, C.M. Engbers, S. Zalipsky, New amphipatic polymer-lipid
conjugates forming long-circulating reticuloendothelial system-evading
liposomes, Bioconjug Chem, 5 (1994) 493-496.
[129] M.C. Woodle, Controlling liposome blood clearance by surface-grafted
polymers, Adv Drug Deliv Rev, 32 (1998) 139-152.
[130] C. Kirby, G. Gregoriadis, Plasma-induced release of solutes from small
unilamellar liposomes is associated with pore formation in the bilayers,
Biochem J, 199 (1981) 251-254.
[131] A. Sharma, U.S. Sharma, Liposomes in drug delivery: progress and
limitations, International Journal of Pharmaceutics, 154 (1997) 123-140.
[132] T. Ishida, R. Maeda, M. Ichihara, K. Irimura, H. Kiwada, Accelerated
clearance of PEGylated liposomes in rats after repeated injections, J Control
Release, 88 (2003) 35-42.
[133] T. Ishida, M. Harada, X.Y. Wang, M. Ichihara, K. Irimura, H. Kiwada,
Accelerated blood clearance of PEGylated liposomes following preceding
liposome injection: effects of lipid dose and PEG surface-density and chain
length of the first-dose liposomes, J Control Release, 105 (2005) 305-317.
[134] P. Laverman, M.G. Carstens, O.C. Boerman, E.T. Dams, W.J. Oyen, N.
van Rooijen, F.H. Corstens, G. Storm, Factors affecting the accelerated blood
clearance of polyethylene glycol-liposomes upon repeated injection, J
Pharmacol Exp Ther, 298 (2001) 607-612.
[135] N. Dos Santos, C. Allen, A.M. Doppen, M. Anantha, K.A. Cox, R.C.
Gallagher, G. Karlsson, K. Edwards, G. Kenner, L. Samuels, M.S. Webb, M.B.
Bally, Influence of poly(ethylene glycol) grafting density and polymer length
on liposomes: relating plasma circulation lifetimes to protein binding, Biochim
Biophys Acta, 1768 (2007) 1367-1377.
48
Introduction
[136] S.M. Moghimi, J. Szebeni, Stealth liposomes and long circulating
nanoparticles: critical issues in pharmacokinetics, opsonization and proteinbinding properties, Prog Lipid Res, 42 (2003) 463-478.
[137] D.E. Discher, A. Eisenberg, Polymer vesicles, Science., 297 (2002) 967973.
[138] R. Stoenescu, A. Graff, W. Meier, Asymmetric ABC-Triblock Copolymer
Membranes Induce a Directed Insertion of Membrane Proteins, Macromol
Biosci, 4 (2004) 930-935.
[139] A. Taubert, A. Napoli, W. Meier, Self-assembly of reactive amphiphilic
block copolymers as mimetics for biological membranes, Curr Opin Chem Biol,
8 (2004) 598-603.
[140] W. Meier, Polymer nanocapsules, Chem. Soc. Rev., 29 (2000) 295-303.
[141] G. Battaglia, A.J. Ryan, Bilayers and interdigitation in block copolymer
vesicles, J Am Chem Soc., 127 (2005) 8757-8764.
[142] D.E. Discher, Emerging applications of polymersomes in delivery: From
molecular dynamics to shrinkage of tumors, Progress in Polymer Science, 32
(2007) 838-857.
[143] J. Panyam, V. Labhasetwar, Biodegradable nanoparticles for drug and
gene delivery to cells and tissue, Adv Drug Deliv Rev, 55 (2003) 329-347.
[144] L.G. Ding J, Water-Soluble Hollow Nanospheres as Potential Drug
Carriers, J. Phys. Chem. B, 102 (1998) 6107-6113.
[145] T. Ruysschaert, A.F. Sonnen, T. Haefele, W. Meier, M. Winterhalter, D.
Fournier, Hybrid nanocapsules: interactions of ABA block copolymers with
liposomes, J Am Chem Soc, 127 (2005) 6242-6247.
[146] O. Onaca, R. Enea, D.W. Hughes, W. Meier, Stimuli-responsive
polymersomes as nanocarriers for drug and gene delivery, Macromol Biosci, 9
(2009) 129-139.
[147] P.J. Photos, L. Bacakova, B. Discher, F.S. Bates, D.E. Discher, Polymer
vesicles in vivo: correlations with PEG molecular weight, J Control Release, 90
(2003) 323-334.
[148] L. Hasadsri, J. Kreuter, H. Hattori, T. Iwasaki, J.M. George, Functional
protein delivery into neurons using polymeric nanoparticles, J Biol Chem, 284
(2009) 6972-6981.
[149] M.K. Reddy, L. Wu, W. Kou, A. Ghorpade, V. Labhasetwar, Superoxide
dismutase-loaded PLGA nanoparticles protect cultured human neurons under
oxidative stress, Appl Biochem Biotechnol, 151 (2008) 565-577.
[150] I. Genta, P. Perugini, F. Pavanetto, K. Maculotti, T. Modena, B. Casado,
A. Lupi, P. Iadarola, B. Conti, Enzyme loaded biodegradable microspheres in
vitro ex vivo evaluation, J Control Release., 77 (2001) 287-295.
[151] C. Colonna, B. Conti, P. Perugini, F. Pavanetto, T. Modena, R. Dorati, P.
Iadarola, I. Genta, Ex vivo evaluation of prolidase loaded chitosan
nanoparticles for the enzyme replacement therapy, Eur J Pharm Biopharm, 70
(2008) 58-65.
[152] C. Garnacho, R. Dhami, E. Simone, T. Dziubla, J. Leferovich, E.H.
Schuchman, V. Muzykantov, S. Muro, Delivery of acid sphingomyelinase in
normal and niemann-pick disease mice using intercellular adhesion molecule1-targeted polymer nanocarriers, J Pharmacol Exp Ther, 325 (2008) 400-408.
49
Chapter 1
[153] T.D. Dziubla, A. Karim, V.R. Muzykantov, Polymer nanocarriers
protecting active enzyme cargo against proteolysis, J Control Release, 102
(2005) 427-439.
[154] F. Chen, Z.R. Zhang, F. Yuan, X. Qin, M. Wang, Y. Huang, In vitro and in
vivo study of N-trimethyl chitosan nanoparticles for oral protein delivery, Int J
Pharm, 349 (2008) 226-233.
[155] S.M. Standley, I. Mende, S.L. Goh, Y.J. Kwon, T.T. Beaudette, E.G.
Engleman, J.M. Frechet, Incorporation of CpG oligonucleotide ligand into
protein-loaded particle vaccines promotes antigen-specific CD8 T-cell
immunity, Bioconjug Chem, 18 (2007) 77-83.
[156] F. Ahmed, R.I. Pakunlu, A. Brannan, F. Bates, T. Minko, D.E. Discher,
Biodegradable polymersomes loaded with both paclitaxel and doxorubicin
permeate and shrink tumors, inducing apoptosis in proportion to accumulated
drug, J Control Release., 116 (2006) 150-158. Epub 2006 Jul 2020.
[157] M.K. Reddy, V. Labhasetwar, Nanoparticle-mediated delivery of
superoxide dismutase to the brain: an effective strategy to reduce ischemiareperfusion injury, Faseb J, 23 (2009) 1384-1395.
[158] R. Tong, V.J. Coyle, L. Tang, A.M. Barger, T.M. Fan, J. Cheng,
Polylactide nanoparticles containing stably incorporated cyanine dyes for in
vitro and in vivo imaging applications, Microsc Res Tech.
[159] A. Napoli, M.J. Boerakker, N. Tirelli, R.J. Nolte, N.A. Sommerdijk, J.A.
Hubbell, Glucose-oxidase based self-destructing polymeric vesicles, Langmuir,
20 (2004) 3487-3491.
[160] M. Nasseau, Y. Boublik, W. Meier, M. Winterhalter, D. Fournier,
Substrate-permeable encapsulation of enzymes maintains effective activity,
stabilizes against denaturation, and protects against proteolytic degradation,
Biotechnol Bioeng, 75 (2001) 615-618.
[161] G. Huysmans, A. Ranquin, L. Wyns, J. Steyaert, P. Van Gelder,
Encapsulation of therapeutic nucleoside hydrolase in functionalised
nanocapsules, J Control Release, 102 (2005) 171-179.
[162] M. Yoshimoto, S. Wang, K. Fukunaga, D. Fournier, P. Walde, R. Kuboi,
K. Nakao, Novel immobilized liposomal glucose oxidase system using the
channel protein OmpF and catalase, Biotechnol Bioeng, 90 (2005) 231-238.
[163] C. Nardin, Hirt, T., Leukel, J., Meier, W., Polymerized ABA Triblock
Copolymer Vesicles, Langmuir, 16 (2000) 1035-1041.
[164] C. Nardin, D. Bolikal, J. Kohn, Nontoxic block copolymer nanospheres:
design and characterization, Langmuir., 20 (2004) 11721-11725.
[165] W. Meier, C. Nardin, M. Winterhalter, Reconstitution of Channel Proteins
in (Polymerized) ABA Triblock Copolymer Membranes Angew Chem Int Ed
Engl., 39 (2000) 4599-4602.
[166] V. Pata, N. Dan, The effect of chain length on protein solubilization in
polymer-based vesicles (polymersomes), Biophys J, 85 (2003) 2111-2118.
[167] D.D. Vijayan K, Lal J, Janmey P, Goulian M, Interactions of MembraneActive Peptides with Thick, Neutral, Nonzwitterionic Bilayers, J. Phys. Chem. B,
109 (2005) 14356-14364.
[168] R. Koebnik, K.P. Locher, P. Van Gelder, Structure and function of
bacterial outer membrane proteins: barrels in a nutshell Mol Microbiol, 37
(2000) 239-253.
50
Introduction
[169] A. Prilipov, P.S. Phale, P. Van Gelder, J.P. Rosenbusch, R. Koebnik,
Coupling site-directed mutagenesis with high-level expression: large scale
production of mutant porins from E. coli, FEMS Microbiol Lett, 163 (1998) 6572.
[170] C. Nardin, J. Widmer, M. Winterhalter, W. Meier, Amphiphilic block
copolymer nanocontainers as bioreactors, Eur. Phys. J. E, 4 (2001) 403-410.
[171] M. Winterhalter, Hilty, C., Bezrukov, S.M., Nardin, C., Meier, W.,
Fournier, D., Controlling membrane permeability with bacterial porins:
applications to encapsulated enzymes, Talanta, 55 (2001) 965-971.
[172] A. Graff, M. Sauer, P. Van Gelder, W. Meier, Virus-assisted loading of
polymer nanocontainer, Proc Natl Acad Sci U S A, 99 (2002) 5064-5068. Epub
2002 Mar 5026.
[173] C. Maier, E. Bremer, A. Schmid, R. Benz, Pore-forming activity of the
Tsx protein from the outer membrane of Escherichia coli. Demonstration of a
nucleoside-specific binding site, J Biol Chem, 263 (1988) 2493-2499.
[174] J. Ye, B. Van Den Berg, Crystal structure of the bacterial nucleoside
transporter Tsx, EMBO J, 23 (2004) 3187-3195. Epub 2004 Jul 3122.
[175] R. Benz, A. Schmid, C. Maier, E. Bremer, Characterization of the
nucleoside-binding site inside the Tsx channel of Escherichia coli outer
membrane, Eur J Biochem, 176 (1988) 699-705.
[176] M. Nallani, S. Benito, O. Onaca, A. Graff, M. Lindemann, M.
Winterhalter, W. Meier, U. Schwaneberg, A nanocompartment system
(Synthosome) designed for biotechnological applications, J Biotechnol, 123
(2006) 50-59. Epub 2005 Dec 2020.
[177] M. Kumar, M. Grzelakowski, J. Zilles, M. Clark, W. Meier, Highly
permeable polymeric membranes based on the incorporation of the functional
water channel protein Aquaporin Z, Proc Natl Acad Sci U S A, 104 (2007)
20719-20724.
[178] C. Nardin, S. Thoeni, J. Widmer, M. Winterhalter, W. Meier,
Nanoreactors based on (polymerized) ABA-triblock copolymer vesicles, Chem
Comm, (2000) 1433-1434.
[179] C. Nardin, W. Meier, Polymerizable amphiphilic block copolymers: From
nanostructured hydrogels to nanoreactors and ultrathin films, Chimia, 55
(2001) 142-146.
[180] C. Nardin, W. Meier, Hybrid materials from amphiphilic block copolymers
and membrane proteins, J Biotechnol, 90 (2002) 17-26.
[181] A. Ranquin, W. Versees, W. Meier, J. Steyaert, P. Van Gelder,
Therapeutic Nanoreactors: Combining Chemistry and Biology in a Novel
Triblock Copolymer Drug Delivery System, Nano Lett, 5 (2005) 2220-2224.
[182] T. Ruysschaert, M. Germain, J.F. Gomes, D. Fournier, G.B. Sukhorukov,
W. Meier, M. Winterhalter, Liposome-based nanocapsules, IEEE Trans
Nanobioscience, 3 (2004) 49-55.
[183] J.P. Colletier, B. Chaize, M. Winterhalter, D. Fournier, Protein
encapsulation in liposomes: efficiency depends on interactions between protein
and phospholipid bilayer, BMC Biotechnol, 2 (2002) 9.
[184] H. Samartzidou, A.H. Delcour, E.coli PhoE porin has an opposite voltagedependence to the homologous OmpF, Embo J, 17 (1998) 93-100.
[185] I.F. Tannock, D. Rotin, Acid pH in tumors and its potential for
therapeutic exploitation, Cancer Res, 49 (1989) 4373-4384.
51
Chapter 1
[186] M. Stubbs, P.M. McSheehy, J.R. Griffiths, Causes and consequences of
acidic pH in tumors: a magnetic resonance study, 13-30.
[187] P. Broz, S. Driamov, J. Ziegler, N. Ben-Haim, S. Marsch, W. Meier, P.
Hunziker, Toward Intelligent Nanosize Bioreactors: A pH-Switchable, ChannelEquipped, Functional Polymer Nanocontainer, Nano Lett, 6 (2006) 2349-2353.
[188] V.P. Torchilin, Micellar nanocarriers: pharmaceutical perspectives, Pharm
Res., 24 (2007) 1-16. Epub 2006 Nov 2016.
[189] P. Broz, S. Marsch, P. Hunziker, Targeting of vulnerable plaque
macrophages with polymer-based nanostructures, Trends Cardiovasc Med., 17
(2007) 190-196.
[190] P. Broz, S.M. Benito, C. Saw, P. Burger, H. Heider, M. Pfisterer, S.
Marsch, W. Meier, P. Hunziker, Cell targeting by a generic receptor-targeted
polymer nanocontainer platform, J Control Release, 102 (2005) 475-488.
[191] P. Broz, N. Ben-Haim, M. Grzelakowski, S. Marsch, W. Meier, P.
Hunziker, Inhibition of Macrophage Phagocytotic Activity by a Receptortargeted Polymer Vesicle-based Drug Delivery Formulation of Pravastatin, J
Cardiovasc Pharmacol, 51 (2008) 246-252.
[192] P. Calvo, B. Gouritin, H. Chacun, D. Desmaele, J. D'Angelo, J.P. Noel, D.
Georgin, E. Fattal, J.P. Andreux, P. Couvreur, Long-circulating PEGylated
polycyanoacrylate nanoparticles as new drug carrier for brain delivery, Pharm
Res., 18 (2001) 1157-1166.
[193] A. Ambruosi, H. Yamamoto, J. Kreuter, Body distribution of polysorbate80 and doxorubicin-loaded [14C]poly(butyl cyanoacrylate) nanoparticles after
i.v. administration in rats, J Drug Target., 13 (2005) 535-542.
[194] V. Vamvakaki, N.A. Chaniotakis, Pesticide detection with a liposomebased nano-biosensor, Biosens Bioelectron, 22 (2007) 2848-2853.
[195] M. Michel, M. Winterhalter, L. Darbois, J. Hemmerle, J.C. Voegel, P.
Schaaf, V. Ball, Giant liposome microreactors for controlled production of
calcium phosphate crystals, Langmuir, 20 (2004) 6127-6133.
[196] P. Van Gelder, H. De Cock, J. Tommassen, Detergent-induced folding of
the outer-membrane protein PhoE, a pore protein induced by phosphate
limitation, Eur J Biochem, 226 (1994) 783-787.
[197] M. Sauer, T. Haefele, A. Graff, C. Nardin, W. Meier, Ion-carrier
controlled precipitation of calcium phosphate in giant ABA triblock copolymer
vesicles, Chem Commun (Camb), (2001) 2452-2453.
[198] F. Axthelm, O. Casse, W.H. Koppenol, T. Nauser, W. Meier, C.G. Palivan,
Antioxidant nanoreactor based on superoxide dismutase encapsulated in
superoxide-permeable vesicles, J Phys Chem B, 112 (2008) 8211-8217.
[199] O. Onaca, D.W. Hughes, V. Balasubramanian, M. Grzelakowski, W.
Meier, C.G. Palivan, SOD Antioxidant Nanoreactors: Influence of Block
Copolymer Composition on the Nanoreactor Efficiency, Macromol Biosci.
[200] M. Grzelakowski, O. Onaca, P. Rigler, M. Kumar, W. Meier, Immobilized
protein-polymer nanoreactors, Small, 5 (2009) 2545-2548.
52
Chapter 2
Scope of the thesis
Scope of the thesis
2 Scope of the thesis
Due to their low bioavailability, many naturally occurring enzymes cannot be
used in their native form as therapeutics in enzyme replacement. A large
variety of carrier systems have been developed to protect and deliver
enzymes, based on lipids, polymers or conjugates. Nevertheless, to solve the
specific pathological situation in enzyme deficiencies, a long-lasting restoration
of enzyme activity is required and the delivery of the enzyme needs further
improvement. We suggest this can be accomplished by developing enzyme
loaded nanoreactors based on biocompatible, biodegradable and non-toxic
polymer systems. Due to their adapted size, stealth surface properties and
multi-functionality, they can cope with the complexity of the in vivo biological
conditions.
The scope of this work was to develop new enzyme-containing polymeric
nanoreactors
and
to
investigate
the
therapeutic
potential
of
these
nanoreactors as drug delivery vehicles in enzyme replacement therapy. As
proof of principle, we focused on the enzyme deficiency MNGIE (mitochondrial
neurogastrointestinal encephalomyopathy), which is caused by a deficiency of
the enzyme thymidine phosphorylase. This model system was chosen because
1) the lab has experience in working with enzymes of the purine and
pyrimidine salvage pathway, 2) a mouse model of this disease is available, 3)
there is proof of a clear correlation between the amount of toxic substrates in
plasma/urine and the pathology of the disease and 4) there is still no therapy
available for this devastating disease.
The nanoreactors investigated in this study are constructed of the amphiphilic
triblock
copolymer
poly(methyloxazoline)-poly(dimethylsiloxane)-
poly(methyloxazoline) (PMOXA-PDMS-PMOXA). The main advantage of this
triblock copolymer is the versatility of the self-assembled particles; 1) the shell
has a hydrophilic, biocompatible low-protein-binding surface and an extremely
high mechanical and thermal stability, 2) the shell can act as a biomembrane
for protein reconstitution, 3) the nanovesicle can encapsulate hydrophilic
55
Chapter 2
substances, and 4) the shell material has enormous possibilities for molecular
functionalization.
The
amphiphilic
®
formulated by Ciba-Vision
PMOXA-PDMS-PMOXA
was
initially
as a material for contact lenses, because of its
biocompatibility and low-protein binding capacity. In addition, the potential of
the PMOXA polymer as outer shell in drug delivery particles is reported (and
patented) by Zalipsky and Woodle since this polymer seems to possess stealth
properties comparable to the PEG polymer.
In this study the porin Tsx was selected to permeabilize the polymeric
membrane,
allowing substrates and products to go in and out of the
nanoparticle. This nucleoside specific porin is known to be very efficient in
transporting nucleosides and nucleobases even at low concentrations. As such,
the rather low but toxic concentrations of the substrates thymidine and
deoxyuridine in the plasma of MNGIE patients can be transported across the
polymeric membrane into the nanocapsule where they can be degraded into
their corresponding products.
All
together,
the
thymidine
phosphorylase
containing
nanoreactors,
schematically represented in Figure 2-1, serve both to protect the enzyme
from degradation and to allow the enzyme to act in situ. As such, these
nanoparticles could have advantages as enzyme delivery vehicles for the use
in enzyme replacement therapy of MNGIE. The aim of this thesis was to
construct such enzyme containing nanoreactors and to evaluate them as
potential drug delivery particles.
Most of the aforementioned studies start from the assumption that the
PMOXA-PDMS-PMOXA nanoparticles are more stable than liposomes and it is
often implied that those particles are non-immunogenic and inert towards
macrophages. However, until now systematic analysis to unequivocally support
these claims is lacking. Therefore, this study not only focuses on the
construction of new thymidine phosphorylase containing nanoreactors, but
also
on
the
investigation
of
their
stability,
toxicity,
immunogenicity,
biodistribution and efficacy, not only in vitro but also in vivo. Up until now,
there is a clear lack of in vivo studies of different polymeric particles. The
56
Scope of the thesis
available published data are scattered bits and pieces mostly focusing on in
vitro tests on cell cultures. Hence, this work is of high relevance for the further
exploitation of the PMOXA-PDMS-PMOXA particles as drug delivery devices.
Figure 2-1: Schematic representation of the envisioned thymidine phosphorylase (TP)
containing PMOXA-PDMS-PMOXA nanoreactor with integrated Tsx porins within the
polymeric membrane.
57
Chapter 3
Production and biophysical
characterization of the nanoreactors
Production and biophysical characterization of the nanoreactors
3 Production and biophysical
characterization of the nanoreactors
3.1 Introduction
The construction and characterization of new polymeric nanometer-sized
bioreactors,
so
called
nanoreactors,
as
a
new
approach
for
enzyme
replacement therapy for MNGIE is investigated in this study. The therapeutic
enzyme, thymidine phosphorylase, is encapsulated in polymeric particles
constructed of the amphiphilic triblock copolymer poly(2-methyloxazoline)block-poly(dimethylsiloxane)-block-poly(2-methyloxazoline)
(PMOXA-PDMS-
PMOXA). PMOXA was chosen as hydrophilic outer block for the nanoreactors
because this polymer possesses protein-repellent and stealth properties,
shielding the encapsulated enzyme from aggression by external agents [1, 2].
The nanoparticles are permeabilized for substrates and products by integrating
bacterial channel proteins in their polymeric wall, a strategy introduced by
Wolfgang Meier and co-workers [3]. They showed that channel forming
proteins can be incorporated without loss of function inside a polymeric wall,
despite a membrane thickness three times wider than a biological lipid
membrane [4]. This results in reactors where the enzymatic reaction is
restricted to the inner volume of the polymeric nanocontainer. This is a
relatively new concept that could give advantages over the conventional drug
delivery systems for the delivery of enzymes.
Recently, other PMOXA-PDMS-PMOXA nanoreactors were constructed for their
use in various biomedical applications like prodrug cancer therapy [5],
prevention of heart attack [6] and development of bio-sensors [7]. However,
the use of such nanoreactors as enzyme delivery devices in enzyme
replacement therapy has never been investigated before. For enzyme
replacement applications the circulation time of active nanoreactors in the
bloodstream is extremely important. The longer the particles remain active in
61
Chapter 3
the bloodstream, the less frequently administrations need to be performed.
Even a rather small improvement in the circulation half-life would mean a
tremendous progress for patients suffering from a fatal enzyme deficiency
disorder (by improving their quality of life as well as by reducing the high costs
of the therapy). In this context it is important to note that the nanoreactors
investigated in this study are developed as proof of principle for the concept of
using enzyme containing PMOXA-PDMS-PMOXA nanoreactors for enzyme
replacement purposes. Since the choice of working with the enzyme thymidine
phosphorylase was predominantly made on practical considerations, the
results
must
be
placed
in
a
broader
perspective,
knowing
that
the
nanoreactors still have many tunable properties.
In this first chapter the construction and biophysical characterization of new
thymidine
phosphorylase
containing
nanoreactors
is
described.
Size,
dispersity, morphology, activity, encapsulation efficiency and pyrogenicity are
important characteristics for the use of nanoparticles as drug delivery systems
in general and are thus the first parameters to determine.
3.2 Materials and methods
3.2.1 Cloning, expression and purification of thymidine phosphorylase
The gene encoding the Escherichia coli C thymidine phosphorylase (TPE.coli)
was cloned via the HindIII-BamHI restriction sites in the pQE30 vector
(Qiagen). The protein was expressed in E.coli WK6 cells containing the pQE30TPE.coli construct. Overnight cultures were used to inoculate baffled flasks with
Terrific Broth and 50 mg/ml ampicillin and were incubated at 37°C while
shaking until the cells reached an optical density at 600 nm between 0.6-0.8.
The medium was then cooled to 28°C and expression of the recombinant
protein was induced by adding isopropyl β-D-1-thiogalactopyranoside (IPTG)
(0.5 mM final concentration). The next morning cells were harvested and
resuspended in 20 mM Tris pH 7.5 with 1 M NaCl, 10 mM imidazole and the
protease inhibitors leupeptine (1 µg/ml) and 4-(2-aminoethyl) benzenesulfonyl
fluoride hydrochloride (AEBSF) (0.1 mg/ml). The cells were lysed via a
62
Production and biophysical characterization of the nanoreactors
passage through the cell crack (Atlas Copco) and the disrupted cell suspension
was cleared via centrifugation (30 minutes at 15000 rpm). The presence of an
N-terminal His6-tag allows a two-step purification scheme, consisting of a NiNTA affinity chromatography (Qiagen) followed by a gelfiltration on a
Superdex 200 16/90 column (Amersham Biosciences) (in 20 mM Hepes, 150
mM NaCl and 1 mM CaCl2 pH 7.5).
For the recombinant expression of human TP (TPhuman) two plasmids, calTP and
GroESL, were kindly provided by Dr. Krosky (SAIC Frederick, Inc). The calTP
plasmid (derived from the pCAL-n-EK vector from Stratagene) contains an
open reading frame encoding the TPhuman with an N-terminal calmodulinbinding peptide-tag. The GroESL plasmid encodes for the chaperons GroES
and GroEL which enhance correct folding of the human enzyme. Both plasmids
were expressed in E.coli BL21 cells which were induced with IPTG after 2 hours
growth at 37°C and allowed to grow further for 4 hours at 20°C. After cell
harvesting, the pellet was resuspended in lysisbuffer containing 2.9 mM CaCl2,
2.7 mM KCl, 1.5 mM KH2PO4, 0.5 mM MgCl2, 200 mM NaCl, 8.1 mM Na2HPO4,
1 mM DTT and 0.1% NP40 and lysed via 3 passages through a French Press
(Beun de Ronde Serlabo) at 11000 psi. Due to the calmodulin-binding peptidetag
the
protein
could
be
purified
via
calmodulin-sepharose
affinity
chromatography (Stratagene). Finally, a gelfiltration step on a Superdex 200
16/90 column (Amersham Biosciences) was performed (in 50 mM Tris-HCl Ph
8.0, 150 mM NaCl and 1 mM DTT).
3.2.2 Cloning, expression and purification of porins Tsx and OmpF
OmpF was cloned into the pGOmpF vector and expressed into a BL21(DE3)
ΔlamB ΔompC ΔompA OmpF::Tn5 strain as described by Prilipov et al. [8].
After cell harvesting by centrifugation, the cell pellets were resuspended in a
20 mM Tris-HCl pH 8 lysisbuffer containing 2% SDS and shaken for 1 hour at
60°C to obtain cell lysis. Lysates were pelleted by ultracentrifugation and
resuspended in 20 mM NaH2PO4 pH 7.3 + 0.125% oPOE for pre-extraction at
37°C for 45 minutes. After a second cycle of centrifugation, pellets were
resuspended in the same phosphate buffer with 3% oPOE for extraction at
63
Chapter 3
37°C for 45 minutes. After centrifugation and concentration, a gel filtration on
a Superdex 200 16/90 column (Amersham Biosciences) was performed (in 20
mM NaH2PO4, 150 mM NaCl and 1% oPOE pH 7.3) to remove contaminating
proteins.
Tsx was cloned into a pGTsx vector, which was derived from the pGOmpF
vector by introducing the sequences for an N-terminal His6-tag and a thrombin
cleaving site into the PstI site. BL21(DE3) ΔlamB ompR ΔompA strain was
used for overproduction of Tsx as described by Prilipov et al. [8]. The Tsx
porin was purified by detergent extraction out of the bacterial membrane by
using octylpolyoxyethylene (oPOE). After cell harvesting by centrifugation, the
cell pellets were resuspended in 20 mM Tris-HCl pH 8.0, 150 mM NaCl, 5%
w/v sucrose and 2 mM EDTA. The cells were lysed via a passage through the
cell crack (Atlas Copco). Lysates were pelleted by centrifugation and
resuspended in 50 mM Tris-HCl pH 8.0 + 0.125% oPOE for pre-extraction at
37°C for 45 minutes. After a second cycle of centrifugation, pellets were
resuspended in 50 mM Tris-HCl pH 8.0 with 250 mM NaCl and 3% oPOE for
extraction at 37°C for 45 minutes. Since Tsx was expressed with a His6-tag, it
was further purified on a Ni-NTA affinity column (Qiagen). Finally, a gel
filtration on a Superdex 200 16/90 column (Amersham Biosciences) was
performed in a buffer containing 50 mM Tris-HCl pH 7.3, 250 mM NaCl and 1%
oPOE to remove contaminating proteins.
3.2.3 Production of nanoreactors
PMOXA (poly 2-methyloxazoline)20 - PDMS (poly dimethylsiloxane)54 - PMOXA20
triblock copolymers with a molecular weight of 7484 g/mol were purchased
from Polymer Source Inc. (Montreal, Canada). The thymidine phosphorylase
encapsulating nanoreactors (TP-NRs) were constructed via the solvent
evaporation method [9]. First, 20 mg of the PMOXA-PDMS-PMOXA polymer
was dissolved in 300 µl ethanol and porin was added at the desired molecular
ratio of protein to polymer. Secondly, this solution was dripped in PBS
containing
100
µM
of
thymidine
phosphorylase
(resulting
in
a
final
concentration of 10 mg polymer/ml) and stirred for 1 hour at room
64
Production and biophysical characterization of the nanoreactors
temperature in order to evaporate the ethanol. Afterwards, continuous stirring
for several days at 4°C results in self-assembled nanoreactors. If necessary,
successive extrusion through a polycarbonate membrane (Avestin) with pore
diameter of 400 nm and 200 nm consecutively was performed. Finally, the
non-encapsulated enzymes were removed via Ni-NTA affinity chromatography
for TPE.coli or via calmodulin affinity chromatography for TPhuman.
3.2.4 Dynamic Light Scattering
The size and polydispersity of the nanoreactors was determined via Dynamic
Light Scattering (DLS) using a laser-spectroscatter (RiNA GmbH, Berlin,
Germany) at 532 nm with a scattering angle of 90°. Ten data sets were
recorded and the autocorrelation curve, size distribution and counts of each
data set were analyzed via the CONTIN software.
3.2.5 Spectrophotometric assay for TP activity
The concentrations of thymidine phosphorylase and thymidine stock solutions
were determined using the following extinction coefficients: 280
mM cm
for TPE.coli, 280
-1
-1
for thymidine. All kinetic experiments were performed in 200 mM
mM cm
= 28.99 mM cm
-1
= 24.41
-1
nm
-1
nm
-1
for TPhuman and 267
nm
= 9.65
potassium phosphate pH 6.8 at 37°C. Product formation was determined
spectrophotometrically
using
the
difference
in
absorption
substrate thymidine and the product thymine (∆290
nm
between
-1
the
-1
= -1 mM cm ). To
measure the enzymatic activity of the nanoreactors, 900 µM thymidine was
mixed with 20 µl nanoreactors (in a cuvet with a total volume of 500 µl) and
the decrease in absorption was measured at 290 nm using a Cary 100Bio UV
visible spectrophotometer (Varian).
3.2.6 Atomic Force Microscopy
Freshly cleaved mica was coated with APTES (3-aminopropyltriethoxy silane)
by the evaporation method [10]. In short, 30 µl of APTES was dropped into a
plastic cap and 10 µl of N,N-diisopropylethylamine in another cap in a
dessicator filled with argon. Freshly cleaved mica was mounted at the top of
65
Chapter 3
the dessicator and left for the reaction to proceed for 1-2 hours. Next, the cap
with APTES was removed and purged with argon for 2 minutes. The mica
sheets were cured for 1-2 days. The nanoreactors
prepared via the solvent
evaporation method were deposited onto the APTES-mica and left to adsorb
for 30 minutes. They were gently washed with pure water (resistivity was over
18 MΩ.cm). Images were obtained with a multimode AFM equipped with a
Nanoscope IIIa controller (Veeco, Santa Barbara, CA) using a glass fluid cell.
All measurements were made with standard Veeco NP-S silicon nitride
cantilevers with lengths of 200 µm and nominal spring constants of 0.06 N/m.
AFM measurements were performed in pure water using tapping mode at a
scan rate around 1.5-2 Hz.
3.2.7 Detection of endotoxins: LAL assay
The level of bacterial endotoxins, especially lipopolysaccharide (LPS), in the
protein or nanoreactor suspensions were determined with the Limulus
ameboecyte
lysate
assay
(LAL-assay)
according
to
the
manufacturer’s
instructions (Cambrex, Inc.). The principle of the test is based on the fact that
bacteria cause intravascular coagulation in the horseshoe crab, Limulus
polyphemus. The agent responsible for the clotting phenomena resides in the
crab’s ameboecytes and a lysate of these ameboecytes (Limulus Ameboecyte
Lysate or LAL) can therefore be used for the detection of endotoxins. The
Kinetic-QCL® LAL assay is a quantitative, kinetic assay for the detection of
Gram negative bacterial endotoxins. The samples and standards are first
incubated with the LAL reagent and placed in a 96 well plate. During this
incubation, the bacterial endotoxins present in the samples will catalyze the
activation of a pro-enzyme present in the LAL reagent to an activated enzyme.
Secondly, a colorless substrate is added which will be consumed by the
activated enzyme thereby releasing the chromophore p-nitroaniline. The plate
is placed in an incubating plate reader and monitored over time (via
WinKQCL® Software) for the appearance of the yellow color.
66
Production and biophysical characterization of the nanoreactors
3.2.8 Removal of endotoxins
A first technique used to remove endotoxins from the porin samples is a MonoP chromatofocusing. Chromatofocusing is a chromatography method where
proteins with the same pI are focused in high concentration and resolved from
other proteins that have different pI in a pH gradient matter. In this case, the
endotoxins present in the sample are washed away from the Mono-P column
during a slow pH gradient, hereafter the porin itself is eluted from the column
at a pH = pI. A start buffer of 0.025 M bis-Tris/HCl pH 6.2 and an eluent
Polybuffer 74/HCl pH 3.4 (1/10 diluted) was applied so that the porin is eluted
after 2/3 of the pH gradient. A Mono-P column (Tricorn) was used at a flow
rate of 1 ml/min.
Secondly, polymyxin B affinity chromatography (Sigma) was used to remove
endotoxins from the samples. Since polymyxin B has affinity for the lipid A
part of LPS it is able to separate the porin from the LPS fraction. After
equilibrating the resin with endotoxin-free 0.1 M ammonium bicarbonate
buffer pH 8.0, the sample was loaded and eluted at a flow rate of 0.2 ml/min.
The last method applied to remove endotoxins is the use of Prosep Remtox
beads (Millipore). A small amount of the beads were mixed with the sample
and incubated for 1.5 hour at 4°C while mixing slowly. Afterwards, the beads
were separated from the samples via centrifugation.
3.3 Results
3.3.1 Providing the nanoreactor building blocks: enzymes and porins
Thymidine phosphorylase (TP) is a cytosolic dimeric enzyme that catalyzes the
reversible phosphorolysis of the nucleosides thymidine and deoxyuridine into
the nucleobases thymine and uracil. Two different enzymes were initially
purified: the Escherichia coli TP (TPE.coli) and the human TP (TPhuman). The
TPE.coli enzyme was expressed in high levels in E.coli WK6 cells and purified via
Ni-NTA affinity chromatography followed by a gelfiltration. A quasi pure protein
67
Chapter 3
could be obtained with a subunit molecular weight of around 52 kDa as
determined with sodium dodecyl sulfate polyacrylamide gel electrophoresis
(SDS-PAGE) analysis and a yield of 13 mg/l culture medium (Figure 3-1). Pure
TPhuman was obtained with a subunit molecular weight of around 54 kDa and a
yield of 3 mg/l culture medium via a calmodulin-affinity-chromatography
followed by a gelfiltration (Figure 3-1).
The catalytic parameters kcat, KM and kcat/KM of the purified enzymes were
determined via initial rate experiments. The initial rate of conversion of
different concentrations of thymidine was measured spectrophotometrically at
290 nm in 200 mM potassium phosphate pH 6.8 at 37°C. The results were
fitted with the Michaelis-Menten equation (Origin software) to determine the
catalytic parameters of the enzymes, which are summarized in Table 3-1.
Table 3-1: Catalytic parameters of purified thymidine phosphorylases.
kcat (s-1)
KM (µM)
kcat/KM (s-1 µM-1)
TPE.coli
413.6 ± 71.1
643.2 ± 268.7
0.64 ± 0.29
TPhuman
16.9 ± 1.1
82.5 ± 19.7
0.2 ± 0.05
Initially two different porins, OmpF and Tsx, were chosen for characterization
as potential transporters in our nanoreactors. Both were recombinantly
expressed and purified. OmpF is a trimeric protein of which each monomer
forms a 16-stranded transmembrane -barrel that functions as a molecular
sieve, allowing concentration driven diffusion of solutes smaller than 600 Da
[11]. Tsx on the other hand is a monomer that forms a 12-stranded -barrel
and allows specific transport of nucleosides and nucleotides [12]. Since Tsx
has specific binding sites for nucleosides in the interior of the channel, rapid
transport of nucleosides at low concentrations is possible compared to slow
diffusion through the non-specific porin OmpF. Detergent extraction with oPOE
of the porins from the cell envelope fraction yielded highly pure material. The
Tsx porin is observed on SDS-PAGE as a band corresponding well to the
molecular weight of the protein (31 kDa) (Figure 3-1) and was produced with
a yield of 8 mg/l culture medium. OmpF is observed on SDS-PAGE as a band
68
Production and biophysical characterization of the nanoreactors
which corresponds well with the expected subunit molecular weight of 37 kDa
(Figure 3-1). For this porin, a yield of 5 mg/l culture medium was obtained.
Figure 3-1: Sodium dodecyl sulfate polyacrylamide gel electrophoresis (SDS-PAGE) of
the purified proteins. The protein molecular weight markers used are the SM1811 from
Fermentas (left gel), the Mark 12 from Invitrogen (middle gel) and the SM0431 from
Fermentas (right gel).
Since the porins are extracted from the outer bacterial membrane it is
expected that the samples contain residual lipopolysaccharides. However, full
stripping of lipopolysaccharides is needed since we aim to use the porins as
building blocks (incorporated inside the polymeric membrane) for nanoreactors
that we want to administer in vivo. LPS induce strong immune responses after
injection in mammals and such unwanted side-effects need to be avoided in
our study [13]. Therefore, the amount of endotoxin in the purified Tsx and
OmpF samples was determined via the LAL assay. This is the most sensitive
and FDA-recommended test to detect endotoxins in therapeutic samples [14].
With the LAL test the amount of endotoxin in the sample is calculated in
endotoxin units (EU). The maximum amount of EU allowed to be administered
in vivo without inducing immune responses is 10 EU per mouse.
After purification, the Tsx porin has an endotoxin load lower than 0.05 EU/ml.
As such, for this porin no further purification is needed. The endotoxin load
measured in purified OmpF samples is however very high (above the detection
69
Chapter 3
range of the assay, > 50 EU/ml).
endotoxins
from
the
porin
Three different methods to remove the
were
compared:
polymyxin
B
affinity
chromatography, Mono-P chromatofocusing and Prosep Remtox beads. After
polymyxin B affinity chromatography, the LAL test still displays too high
amounts of endotoxin units (> 50 EU/ml). In addition, not only the LPS but
the entire OmpF-LPS aggregates are sticking to the column and the major
amount of porin was only released from the polymyxin beads after
regenerating the column. This indicates that the affinity between polymyxin B
and OmpF is too high to use this technique for the removal of LPS from the
porin sample. Secondly, incubating the OmpF sample with Prosep Remtox
beads was used to remove the LPS. Despite its easy and fast procedure the
endotoxin load after incubation with Remtox beads is still higher than 50
EU/ml. In a last attempt, a Mono-P chromatofocusing was performed to
remove LPS from the OmpF samples. After loading the samples at pH 6.2 a pH
gradient was introduced and the OmpF porin was eluted at a pH around 4.6
which corresponds well with the pI of the porin. The pooled fractions were
analyzed with the LAL test and showed an endotoxin amount of 1.774 EU/ml
which is low enough for in vivo use.
In conclusion, the nucleoside specific porin Tsx has a higher yield and doesn’t
need extra purification steps to remove endotoxins. Moreover, earlier work
describes the efficient transport of thymidine through these porins [15]. In
addition, the successful incorporation of the porin Tsx into a PMOXA-PDMSPMOXA polymeric membrane, with evidence of protein functionality, was
proved previously [5]. Therefore, Tsx was the preferred porin to use for
nanoreactor construction.
3.3.2 Production, size and dispersity of the nanoreactors
The amphiphilic copolymer PMOXA20-PDMS54-PMOXA20 was used to construct
thymidine phosphorylase containing nanoreactors (TP-NRs).
TP-NRs were
prepared via the solvent evaporation method as described above, which
included extensive stirring for several days. If the stirring was performed in a
70
Production and biophysical characterization of the nanoreactors
continuous and homogeneous matter, the nanoparticle sample spontaneously
has a high degree of monodispersity (as measured with DLS) and no extrusion
was needed. Otherwise,
successive extrusion through a filter with pore
diameter of 400 nm and 200 nm was performed. Finally, the non-encapsulated
enzymes were removed via affinity chromatography. The samples were
analyzed by dynamic light scattering to determine the size and the
polydispersity of the nanoreactors. Nanoreactors with a mean size around 200
nm diameter were obtained with low batch-to-batch variability (Figure 3-2).
Figure 3-2: Size distribution curve (log-scale) of free thymidine phosphorylase in blue
and thymidine phosphorylase containing nanoreactors in red. The mean radius of the
nanoreactors is around 100 nm.
The TP-NRs were further analyzed via Atomic Force Microscopy (AFM). As
shown in Figure 3-3 the nanoreactors appear as relatively monodisperse and
round particles. Nevertheless, the adsorption of the nanoreactors onto APTES
resulted in a larger base area. This behaviour has also been observed for
vesicle-bilayer complexes [16]. The images indicate that the polymeric
particles are very flexible: the upper part of the particle is easily deformed by
the tip interaction during scanning. Increasing the tip force on the nanoreactor
also resulted in reduced height of the particles. The adsorption and tip-sample
interaction during scanning resulted in significantly lower dimensions of the
nanoreactors obtained with AFM than ensemble measurements with DLS [17].
71
Chapter 3
Analysis of the cross section of the nanoreactors shows a diameter of
approximately 100 nm and a height of 10-20 nm (Figure 3-3C), while DLS
measurements of the same particles showed a mean radius of 104 nm.
Figure 3-3: AFM image of enzyme filled nanoreactors: (A) Tapping mode height image
(scale bar 300 nm). (B) Zoomed-in 3D-image of the nanoreactor indicated by the arrow
on image (A). (C) Cross-sectional analysis of 3 nanoreactors; blue colored section
corresponds to the nanoreactor marked with 2 blue "+" - signs and the white crosssection line in image (A), red curve corresponds to the 2 marked nanoreactors in the
middle of image (A). The distance between the "+" - signs on figure (A) is indicated by
the vertical lines on figure (C).
3.3.3 Enzymatic activity of the nanoreactors
The
in
vitro
enzymatic
activity
of
the
TP-NRs
was
confirmed
spectrophotometrically by measuring the decrease in absorption at 290 nm
due to thymidine consumption. The thymidine concentration used is 900 µM
which is above the KM of the enzyme. As such, the enzyme can work in theory
close
to
its
maximal
turnover
efficiency.
However,
this
concentration
corresponds to the substrate concentration outside the particles. Since the
activity of the nanoreactors is probably limited by the transport of substrate
72
Production and biophysical characterization of the nanoreactors
and product over the membrane, the substrate concentration inside the
particle that is available for the enzyme is probably lower.
TP-NRs
induce,
in
comparison
to
control
samples
(PBS
and
empty
nanoparticles), a decrease in absorption due to the conversion of thymidine to
thymine. These results show that the nanoreactors possess thymidine
phosphorylase activity. However, in comparison to the same amounts of free
enzyme, the enzyme activity of the nanoreactors is rather low.
Figure 3-4: Activity measurements showing the change in absorption as a function of
time measured at 290 nm. For each measurement 900 µM thymidine was mixed with 20
µl nanoparticles or control samples (in a cuvet with a total volume of 500 µl) and the
reaction was followed for 10 minutes at 37°C.
3.3.4 Encapsulation efficiency
The efficiency of encapsulation (i.e. the amount of TPE.coli found in the
nanoparticles after separation versus the starting amount of protein) was
determined by comparative SDS-PAGE analysis. Known quantities of free
TPE.coli (0.5 µg, 1 µg, 2 µg, 4 µg, 8 µg, 12 µg, 24 µg) were run on an SDSPAGE together with an aliquot (20 µl) of two different nanoreactor samples.
The nanoparticles were first disrupted completely by boiling them for 15
minutes in the presence of SDS-containing loading buffer. The intensity of
each coomassie blue stained TPE.coli band was measured with Quantity One
73
Chapter 3
Software (ChemiDocXRS, Biorad). The amount of TPE.coli incorporated in the
nanoreactor sample was determined via a standard curve that plots the
protein amount versus band intensity of known amounts of protein. The
overall enzyme concentration in the sample is 10 µM (0.5 mg/ml) (Figure 3-5).
As such, an enzyme encapsulation efficiency of around 10% is found. Take
care, 10 µM is the overall concentration in the sample, the local enzyme
concentration inside the particles is higher (ideally 183 molecules per
particle*).
Figure 3-5: SDS-PAGE analysis of two TP-NR samples. The nanoparticles with enzyme
and porin (NP+TP+Tsx) show encapsulation of TPE.coli (±52 kDa) and incorporation of
porin Tsx (±31 kDa). The nanoreactor-sample without porins (NP+TP-Tsx) shows only
an analogues enzyme band. Known concentrations of TPE.coli (lane 1 = 0.5 µg; lane 2 = 1
µg; lane 3 = 2 µg; lane 4 = 4 µg; lane 5 = 8 µg; lane 6 = 12 µg; lane 7 = 24 µg) serve
as standard to calculate the encapsulation efficiency of the enzyme inside the
nanoreactors. The protein molecular weight marker used is SMO431 from Fermentas.
* The volume of the nanoreactors with an internal radius of 90 nm is 3.05 x 10-18 l. With an initial
enzyme concentration of 100 µM and the assumption that there is no attraction or repulsion of the
enzyme to the particles, but a simple encapsulation of the specific volume, there will be ideally 183
molecules TP inside each nanoreactor.
74
Production and biophysical characterization of the nanoreactors
3.4 Conclusion and discussion
New thymidine phosphorylase containing nanoreactors composed of PMOXAPDMS-PMOXA triblock copolymers were constructed. The nanoreactors are
permeabilized with the bacterial membrane protein Tsx. We were able to
produce monodisperse nanoreactors with a mean size of 200 nm. This is the
ideal size if we intend long-circulating particles [18]. Furthermore, we
demonstrated that the enzyme thymidine phosphorylase can be encapsulated
inside the nanoreactors with an encapsulation efficiency around 10%,
comparable to enzyme encapsulation in liposomes, with a high batch-to-batch
reproducibility. In addition, these nanoreactors are able to convert thymidine
into thymine and thus show enzymatic activity. Although rather low, the
activity
is
significantly
above
the
background
signal.
Moreover,
the
pyrogenicity of the porins by means of endotoxin contamination was
addressed and purification steps were introduced if necessary to ensure the
absence of LPS in the porin and concomitant nanoreactor samples. All
together, the physicochemical properties of the nanoreactors are promising
and further steps in the preclinical characterization of the particles can be
carried out.
3.5 References
[1] M.C. Woodle, C.M. Engbers, S. Zalipsky, New amphipatic polymer-lipid
conjugates forming long-circulating reticuloendothelial system-evading
liposomes, Bioconjug Chem, 5 (1994) 493-496.
[2] S. Zalipsky, C.B. Hansen, J.M. Oaks, T.M. Allen, Evaluation of blood
clearance rates and biodistribution of poly(2-oxazoline)-grafted liposomes, J
Pharm Sci, 85 (1996) 133-137.
[3] C. Nardin, J. Widmer, M. Winterhalter, W. Meier, Amphiphilic block
copolymer nanocontainers as bioreactors, Eur. Phys. J. E, 4 (2001) 403-410.
[4] M. Winterhalter, Hilty, C., Bezrukov, S.M., Nardin, C., Meier, W., Fournier,
D., Controlling membrane permeability with bacterial porins: applications to
encapsulated enzymes, Talanta, 55 (2001) 965-971.
[5] A. Ranquin, W. Versees, W. Meier, J. Steyaert, P. Van Gelder, Therapeutic
Nanoreactors: Combining Chemistry and Biology in a Novel Triblock Copolymer
Drug Delivery System, Nano Lett, 5 (2005) 2220-2224.
[6] P. Broz, N. Ben-Haim, M. Grzelakowski, S. Marsch, W. Meier, P. Hunziker,
Inhibition of Macrophage Phagocytotic Activity by a Receptor-targeted Polymer
75
Chapter 3
Vesicle-based Drug Delivery Formulation of Pravastatin, J Cardiovasc
Pharmacol, 51 (2008) 246-252.
[7] M. Grzelakowski, O. Onaca, P. Rigler, M. Kumar, W. Meier, Immobilized
protein-polymer nanoreactors, Small, 5 (2009) 2545-2548.
[8] A. Prilipov, P.S. Phale, P. Van Gelder, J.P. Rosenbusch, R. Koebnik,
Coupling site-directed mutagenesis with high-level expression: large scale
production of mutant porins from E. coli, FEMS Microbiol Lett, 163 (1998) 6572.
[9] C. Nardin, Hirt, T., Leukel, J., Meier, W., Polymerized ABA Triblock
Copolymer Vesicles, Langmuir, 16 (2000) 1035-1041.
[10] M. Bezanilla, M. Srinivas, D. Laney, Y. Lyubchenko, H. Hansma,
Adsorption of DNA to mica, silylated mica, and minerals : characterization by
atomic force microscopy, Langmuir 11 (1995) 655-659
[11] R. Koebnik, K.P. Locher, P. Van Gelder, Structure and function of bacterial
outer membrane proteins: barrels in a nutshell Mol Microbiol, 37 (2000) 239253.
[12] J. Ye, B. Van Den Berg, Crystal structure of the bacterial nucleoside
transporter Tsx, EMBO J, 23 (2004) 3187-3195. Epub 2004 Jul 3122.
[13] A.J. Ulmer, H. Flad, T. Rietschel, T. Mattern, Induction of proliferation and
cytokine production in human T lymphocytes by lipopolysaccharide (LPS),
Toxicology, 152 (2000) 37-45.
[14] J.B. Hall, M.A. Dobrovolskaia, A.K. Patri, S.E. McNeil, Characterization of
nanoparticles for therapeutics, Nanomedicine (Lond), 2 (2007) 789-803.
[15] R. Benz, A. Schmid, C. Maier, E. Bremer, Characterization of the
nucleoside-binding site inside the Tsx channel of Escherichia coli outer
membrane, Eur J Biochem, 176 (1988) 699-705.
[16] S. Kumar, J.H. Hoh, Direct visualization of vesicle-bilayer complexes by
atomic force microscopy, Langmuir, 16 (2000) 9936-9940.
[17] Y. Ebenstein, Nahum, E., Banin, U., Tapping Mode Atomic Force
Microscopy for Nanoparticle Sizing: Tip-Sample Interaction Effects, Nano Lett,
2 (2002) 945-950.
[18] S.M. Moghimi, A.C. Hunter, J.C. Murray, Long-circulating and targetspecific nanoparticles: theory to practice, Pharmacol Rev, 53 (2001) 283-318.
76
Chapter 4
Stability of the nanoreactors
Stability of the nanoreactors
4 Stability of the nanoreactors
4.1 Introduction
Polymeric nanoreactors functionalized by encapsulating the enzyme thymidine
phosphorylase and permeabilized by incorporating the membrane protein Tsx
in the reactor wall were constructed. The nanoreactors have unique physical
properties which make them promising particles for enzyme delivery. In
theory, the stability of such polymeric particles should be better than the
stability of liposomes [1, 2]. This is due to the higher molecular weight and the
lower flexibility of the underlying block copolymers. However, the interaction
of PMOXA-PDMS-PMOXA nanoreactors with plasma proteins and other blood
components is still unknown. Such interactions may influence uptake and
clearance and hence potentially affect distribution and delivery of the
therapeutic
enzyme
[3].
Nanoparticle-protein-binding
occurs
almost
instantaneously once the particle enters biological fluid and the physical
properties of the drug delivery particle may significantly change. Ideally, the
integrity of the nanoreactors is not disturbed in the complex biological
environment. As such, understanding the nanoreactor hematocompatibility is
an important step during the preclinical characterization of the nanoreactors as
enzyme delivery vehicles. Therefore, in this chapter the nanoreactors are
characterized in terms of their in vitro stability in serum, thereby mimicking
the in vivo situation in the bloodstream.
4.2 Materials and methods
4.2.1 Production of nanoreactors
The TPE.coli (thymidine phosphorylase from E.coli) encapsulating nanoreactors
(TP-NRs) were constructed via the solvent evaporation method as described in
detail in section 3.2.3 [4].
79
Chapter 4
4.2.2 Production of liposomes
The production of liposomes was initially described by Nikaido and coworkers
[5, 6]. Briefly, DOPC/ePG lipids dissolved in chloroform (Avanti) were mixed in
a ratio of 4:1 and the solution was dried under a gentle stream of nitrogen
until a smooth film was formed. Afterwards, 200 µl high quality diethyl ether
was added and this solution was incubated for 10 minutes at room
temperature in order to dissolve the film completely. Next, the lipid solution
was dried while shaking in a 50°C water bath. Drying was completed by
incubating the tubes for 30 minutes in a dessicator. Hydration of this lipid film
was performed by adding 1 ml of PBS containing the TPE.coli enzyme (100 µM)
and the Tsx porin (30 µg) and the vesicles were allowed to swell for at least 30
minutes. Next, the solution was gently sonicated in a water bath sonicator
(about 5 seconds) and a homogenous and turbid solution was obtained.
Subsequently, eight freeze-thaw cycles and extrusion through a 200 nm
polycarbonate filter provided uniform and unilammellar proteoliposomes. The
non-encapsulated
enzyme
was
further
removed
via
Ni-NTA
affinity
chromatography.
4.2.3 Dynamic Light Scattering
The size and polydispersity of the nanoreactors was determined via Dynamic
Light Scattering (DLS) using a laser-spectroscatter (RiNA GmbH, Berlin,
Germany) at 532 nm with a scattering angle of 90°. Ten data sets were
recorded and the auto correlation curve, size distribution and counts of each
data set were analyzed via the CONTIN software.
4.2.4 Spectrophotometric assay for TP activity
All kinetic experiments were performed in 200 mM potassium phosphate pH
6.8 at 37°C. Product formation was determined spectrophotometrically using
the difference in absorption between the substrate thymidine and the product
thymine (∆290 nm = -1 mM-1cm-1). To measure the enzymatic activity of TPE.coli
inside nanoreactors and liposomes, 900 µM thymidine was mixed with 40 µl of
80
Stability of the nanoreactors
the nanoreactor/serum solution or 40 µl of the liposome/serum solution. For
the activity measurement of free TPE.coli incubated in serum, 20 µl of a 1000x
dilution was added to 900 µM thymidine. The decrease in absorption was
measured at 290 nm during 10 minutes using a Cary 100Bio UV visible
spectrophotometer (Varian). Every measurement was performed twice and the
mean decrease in absorption was calculated.
4.3 Results
The stability of the thymidine phosphorylase containing nanoreactors (TP-NRs)
in physiological conditions is a very important requirement if we aim the
intravenous administration of the particles for enzyme replacement purposes.
For stability studies, the permeabilized TP-NRs were incubated for several days
in 50% naïve mouse serum at 37°C. As a reference, the same experiments
were done with PBS instead of blood serum. At different time points both the
size distribution (by DLS) and the enzyme activity (290 nm) was measured.
Also, liposomes with encapsulated TPE.coli and permeabilized with the porin Tsx
(TP-LIPs) were constructed, analogous to previously reported liposomal
reactors [7, 8]. TP-LIPs and free TPE.coli were used as control samples.
DLS measurements show that the morphology and size of the TP-NRs in
mouse
serum
remains
stable
over
time
(Figure
4-1).
Although
the
polydispersity of the nanoreactor samples increases slightly, we can conclude
that the nanoreactors are still intact after 4 days of incubation at 37°C. As a
negative control, detergent (TritonX-100) was added to the same sample after
96 hours of incubation. This results in a clear shift of the DLS pattern to lower
sizes, demonstrating the destruction of the particles. The same results were
obtained when the reactors are incubated in PBS (data not shown), although in
this case a more narrow dispersity window is found due to the absence of
serum proteins.
81
Chapter 4
Figure 4-1: DLS distribution patterns (intensity in function of radius for ten consecutive
measurements) of TP-NRs incubated in 50% naive mouse serum after (A) 0 h and (B)
96 h at 37°C. (C) As negative control, detergent TritonX-100 was added to the sample
after 96 h of incubation, resulting in a destruction of the particles.
To measure the enzyme activity of the TP-NRs, TP-LIPs and free TP incubated
in serum, an aliquot was taken after the corresponding times and used for the
spectrophotometric measurement as described above. The enzyme activity of
the TP-LIPs increases significantly during their incubation period at 37°C
(Figure 4-2). Such an increase in apparent activity of the liposomeencapsulated enzyme is expected if enzyme is leaking out of the liposomes, as
previously reported [9, 10]. Indeed, for the encapsulated enzymes the netto
rate of substrate conversion is limited by the rate of diffusion of the substrate
through the porins. Hence, release of the enzyme to the medium leads to the
relief of diffusion barriers with concomitant higher relative rates of substrate
consumption. In this context it is noteworthy that the activity values plotted in
figure 4-2 represent activities relative to the value at the zero-time-point for
each formulation and that hence at all times the absolute activity of the
encapsulated enzyme preparations remains below the activity of the free
enzyme. In contrast to the TP-LIPs, no increase in activity is observed for the
TP-NRs. Thus, the enzyme is contained inside the polymeric particles without
leakage. This is a clear advantage for their in vivo use because the release of a
foreign enzyme will elicit an immune response. Note that the enzyme activity
of the TP-NRs decreases slightly over time. After three days half of the initial
activity remained. This decrease in activity cannot be due to the degradation
82
Stability of the nanoreactors
of the enzyme, because the activity of the free enzyme in serum remains
constant. One possible hypothesis for this decrease in activity could be a
partial obstruction of the porins by serum proteins. This can be due to the
artificial and static in vitro setting and has to be clarified further in the future.
In the control experiments where the free TP, TP-NRs and TP-LIPs are
incubated in PBS the same results were obtained only no leakage of enzyme
out of the liposomes was detected as no increase in relative activity was
observed (data not shown). These results confirm once again that upon
interaction with serum proteins the liposomes become leaky, while the
nanoreactors remain their integrity and keep the enzyme encapsulated and
protected.
Figure 4-2: Residual activity of TP-NRs (●), TP-LIPs (■) and free TP (▲) during their
incubation in 50% mouse serum. For each enzyme formulation the relative activity
scaled towards its zero-time-point activity is plotted as function of time. The data
represent at least three independent experiments.
83
Chapter 4
4.4 Conclusion and discussion
Here, we present for the first time stability measurements of nanometer-sized
PMOXA-PDMS-PMOXA reactors. We can conclude that, although a slight
decrease in enzymatic activity was observed over time, the TP-NRs are stable
and not leaky at 37°C in serum for several days. The study clearly shows that
the nanoreactors are more suited for in vivo use than analogous particles
constructed of lipids.
Independently from us and at the same time, Litvinchuk and co-workers also
investigated the stability of PMOXA-PDMS-PMOXA nanoparticles (without
porins) [11]. They as well find that the nanoparticles are considerably more
stable than conventional liposomes. In plasma 50% of the encapsulated
calcein was released after 48 hours. This release was however attributed to
the presence of sodium azide that was added to the blood plasma as
sterilization agent.
Due to the more robust nature of polymeric nanoreactors compared to
liposomes, their application as enzyme delivery vehicles is promising. In the
next chapters, important in vivo properties of the nanoreactors (like toxicity,
immunogenicity and biodistribution) will be addressed in order to further
evaluate the nanoreactors as enzyme replacement particles.
4.5 References
[1] D.E. Discher, Emerging applications of polymersomes in delivery: From
molecular dynamics to shrinkage of tumors, Progress in Polymer Science, 32
(2007) 838-857.
[2] J.C. Lee, H. Bermudez, B.M. Discher, M.A. Sheehan, Y.Y. Won, F.S. Bates,
D.E. Discher, Preparation, stability, and in vitro performance of vesicles made
with diblock copolymers, Biotechnol Bioeng, 73 (2001) 135-145.
[3] P. Aggarwal, J.B. Hall, C.B. McLeland, M.A. Dobrovolskaia, S.E. McNeil,
Nanoparticle interaction with plasma proteins as it relates to particle
biodistribution, biocompatibility and therapeutic efficacy, Adv Drug Deliv Rev,
61 (2009) 428-437.
[4] C. Nardin, Hirt, T., Leukel, J., Meier, W., Polymerized ABA Triblock
Copolymer Vesicles, Langmuir, 16 (2000) 1035-1041.
84
Stability of the nanoreactors
[5] M. Luckey, H. Nikaido, Diffusion of solutes through channels produced by
phage lambda receptor protein of Escherichia coli: inhibition by higher
oligosaccharides of maltose series, Biochem Biophys Res Commun, 93 (1980)
166-171.
[6] H. Nikaido, Proteins forming large channels from bacterial and
mitochondrial outer membranes: porins and phage lambda receptor protein,
Methods Enzymol, 97 (1983) 85-100.
[7] B. Chaize, J.P. Colletier, M. Winterhalter, D. Fournier, Encapsulation of
enzymes in liposomes: high encapsulation efficiency and control of substrate
permeability, Artif Cells Blood Substit Immobil Biotechnol, 32 (2004) 67-75.
[8] G. Huysmans, A. Ranquin, L. Wyns, J. Steyaert, P. Van Gelder,
Encapsulation of therapeutic nucleoside hydrolase in functionalised
nanocapsules, J Control Release, 102 (2005) 171-179.
[9] S.J. Comiskey, T.D. Heath, Serum-induced leakage of negatively charged
liposomes at nanomolar lipid concentrations, Biochemistry, 29 (1990) 36263631.
[10] T. Ishida, H. Harashima, H. Kiwada, Liposome clearance, Biosci Rep, 22
(2002) 197-224.
[11] S. Litvinchuk, Z. Lu, P. Rigler, T.D. Hirt, W. Meier, Calcein release from
polymeric vesicles in blood plasma and PVA hydrogel, Pharm Res, 26 (2009)
1711-1717.
85
Chapter 5
Toxicity of the nanoreactors
Toxicity of the nanoreactors
5 Toxicity of the nanoreactors
5.1 Introduction
An important prerequisite of the new enzyme-containing nanoreactors is that
they may not provoke toxic effects on healthy cells and tissues. Although
polymers made of PMOXA-PDMS-PMOXA blocks are approved by the FDA for
use in contact-lens material, the toxicity of the polymers for in vivo use is still
not determined. To date, there are only a few studies investigating the toxic
effects of nanomaterials and no clear guidelines are presently available to
quantify these effects [1]. However, it is very important to investigate the
toxicity of the nanoreactors as early as possible in their development to avoid
clinical trial failure due to the toxic effects of these new drug compounds.
Since hepatocytes are a rich source of enzymes involved in xenobiotic
biotransformations, the liver is the predominant target for xenobiotic-induced
toxicity. For economical and ethical reasons, the use of primary hepatocytes in
culture has gained interest as an alternative to in vivo toxicity studies [2]. The
importance of short-term hepatocyte cultures in the characterization of acute
hepatotoxicity during early drug development is now widely recognized [3]. In
contrast, long-term models which can address sub-chronic and chronic
hepatotoxicity are not yet available.
In this chapter, the potential toxic effect of the thymidine phosphorylase
containing nanoreactors is evaluated in an in vitro model derived from rat
hepatocytes.
89
Chapter 5
5.2 Materials and methods
5.2.1 Production of nanoreactors
The TPE.coli (thymidine phosphorylase from E.coli) encapsulating nanoreactors
(TP-NRs) were constructed via the solvent evaporation method as described in
detail in section 3.2.3 [4].
5.2.2 Hepatocyte cell culture
Male Sprague Dawley rats (around 250 g) were kept under controlled
environmental conditions (12-hour light–dark cycle) and fed a standard diet
(Animalabo A04 and water ad libitum). Experiments were carried out in
accordance to the regulations of the local ethical committee for animal
experiments of the Vrije Universiteit Brussel (VUB). Hepatocytes were isolated
(viability > 80% as assessed by trypan blue dye exclusion) and plated at a
density of 0.57 × 105 cells/cm2 at 37°C, in an atmosphere of 5% CO2 and 95%
air and 100% relative humidity [5]. They were cultured in Williams medium E
(GIBCO 22551) supplemented with different antibiotics (50 µg/ml kanamycin
monosulphate, 10 µg/ml sodiumampicillin, 7.3 IU/ml sodiumbenzylpenicillin,
50 µg/ml streptomycin sulphate) and additives (2 mM L-glutamine, 5 µg/ml
insulin, 7 ng/ml glucagon, 25 µg/ml hydrocortisone-sodium succinate). Cells
were allowed to attach to the plastic substratum for 4 hours. Then, serumcontaining medium was removed and fresh, serum-free culture medium
supplemented with different concentrations of nanoreactors (50 µg/ml, 100
µg/ml, 500 µg/ml and 1000 µg/ml) was added to the cultures. The medium
was further renewed daily.
5.2.3 LDH test
The leakage of the enzyme lactate dehydrogenase (LDH) was measured using
the Merckotest (VWR International, Leuven, Belgium). Briefly, a mixture of
pyruvate and NADH is added to the supernatant and/or cell debris. The
decrease in NADH concentration is measured spectrophotometrically at 340
90
Toxicity of the nanoreactors
nm and is an amount for the LDH activity. The LDH leakage was calculated in
percentage by the ratio: (100 x LDH activity in supernatant)/(LDH activity in
(supernatant + cells)).
5.2.4 MTT test
The cytotoxicity in the MTT-assay was determined by measuring the reduction
of
water
soluble
yellow
MTT
(3-(4,5-dimethyl-2-thiazolyl)-2,5-
diphenyltetrazolium bromide, Sigma) to water insoluble dark blue MTT
formazan crystals in living cells, at 570 nm. The results are calculated as
relative values (%) to the negative control, whereas untreated control is set to
be 100% viable.
5.2.5 Statistical analysis
Results are expressed as the mean ± standard deviation of at least three
independent experiments. Statistical analysis was performed using a one-way
ANOVA test, followed by a Bonferroni correction to study differences amongst
individual mean values. P-values < 0.05 are considered significant.
5.3 Results
Since the liver is the predominant organ in which biotransformation of foreign
compounds takes place, it is the ultimate organ to test xenobiotic–induced
toxicity [6, 7]. Therefore, possible toxic effects of the TP-NRs on hepatocytes
were investigated. As such, hepatocytes were isolated from male outbred
Sprague-Dawley rats by a two-step collagenase perfusion method [8] and
cultured in Williams medium E as described previously [5]. After 4 hours of cell
adhesion, serum-containing medium was removed and fresh, serum-free
culture medium supplemented with different concentrations of TP-NRs (50
µg/ml, 100 µg/ml, 500 µg/ml and 1000 µg/ml) was added to the hepatocyte
cultures.
Cytotoxicity was tested by cellular morphology, membrane leakage of lactate
dehydrogenase (LDH assay) and mitochondrial function (MTT assay) as a
91
Chapter 5
function of culture time. The lactate dehydrogenase (LDH) release assay and
the
3-(4,5-dimethyl-2-thiazolyl)-2,5-diphenyltetrazolium
bromide
(MTT)
reduction assay have been widely used for evaluating cell viability in culture.
The LDH assay measures the release of the intracellular enzyme LDH upon
damage of the plasma membrane. Therefore, the amount of LDH release
corresponds to the amount of death cells. On the other hand, the MTT assay
measures the conversion of MTT into purple-colored MTT formazan by the
redox activity of living cells and a decrease in cellular MTT reduction
corresponds to cell damage. Medium alone and medium with addition of PBS
were used as negative controls. SuperFasLigand™ (human recombinant; Alexis
Biochemicals)
in
combination
with
cycloheximide
(Sigma),
two
known
apoptosis-inducers, acted as the positive control [9, 10].
The results show no significant difference in LDH leakage between the various
conditions up to 48 hours of incubation (Figure 5-1). Only after prolonged
exposure and at higher doses of TP-NRs (96 hours for the 500 µg/ml and 72
hours for the 1000 µg/ml concentrations), elevated levels of LDH leakage were
measured.
This cytotoxic effect was confirmed by visual inspection of the morphology of
the hepatocytes: after prolonged incubation with TP-NRs at high doses, the
hepatocyte-morphology became apoptotic in comparison to healthy untreated
cells (data not shown). However, TP-NRs at lower doses (50 µg/ml and 100
µg/ml) did not affect the viability of the hepatocytes throughout culture time.
Indeed, the hepatocytes treated with TP-nanoreactors at these concentrations
still looked healthy after 96 h of incubation (Figure 5-2).
The cytotoxic effect of nanoreactors was also tested via the MTT-test.
Unfortunately this assay displayed false positive results. An increase in
viability was observed after incubation with TP-nanoreactors, and changes in
concentration or incubation time had no significant influence on the viability
(data not shown). Such inconsistencies were also reported by others [11, 12]
and could be due to an interference of the MTT-formazan with the
nanoparticles.
92
Toxicity of the nanoreactors
Figure 5-1: LDH leakage of hepatocytes incubated with different concentrations of TPNRs as a function of time. Medium with or without TP-NRs was renewed daily. Medium
and PBS serve as negative controls, a combination of SuperFasLigand with
cycloheximide is used as the positive control. The data represent at least three
independent experiments. Statistical analysis were performed using a one-way ANOVA
test, followed by a Bonferroni post hoc test. (*p < 0.05 and ***p < 0.001 compared to
medium and PBS control values).
Figure 5-2: Morphology of cultured hepatocytes incubated for 96 h with (A) medium as
negative control, (B) SuperFasLigand-cycloheximide as positive control, (C) TP-NRs at
50 µg/ml and (D) TP-NRs at 100 µg/ml. All images were acquired with constant
microscope settings (magnification = 100x).
93
Chapter 5
5.4 Conclusion and discussion
Overall, we can conclude that the thymidine phosphorylase encapsulating
nanoreactors of concentrations up to 100 µg polymer/ml have no acute
cytotoxic effect on hepatocytes. This is evidenced by the maintenance of the
cellular morphology and the absence of LDH leakage of hepatocytes during
their
incubation
with
nanoreactors.
It
is
moreover
questionable
if
concentrations of 100 μg/ml and higher reflect biological relevant conditions.
Indeed,
the
nanoreactors
will
circulate
in
the
bloodstream
and
local
concentrations at liver tissue will probably never reach these levels.
In conclusion, our results demonstrate that enzyme containing PMOXA-PDMSPMOXA nanoreactors up to 100 µg/ml do not affect the viability of hepatocytes
under the experimental conditions of this study. These results are an
important step in the evaluation of the nanoreactors as drug delivery devices
and opens the door to further examination of their in vivo behavior. In
addition, since little information is available on nanoparticle toxicity in general,
these data are also a valuable contribution to the toxicity risk assessment
knowledge of nanomaterials.
5.5 References
[1] S.M. Hussain, K.L. Hess, J.M. Gearhart, K.T. Geiss, J.J. Schlager, In vitro
toxicity of nanoparticles in BRL 3A rat liver cells, Toxicol In Vitro, 19 (2005)
975-983.
[2] G. Eisenbrand, B. Pool-Zobel, V. Baker, M. Balls, B.J. Blaauboer, A. Boobis,
A. Carere, S. Kevekordes, J.C. Lhuguenot, R. Pieters, J. Kleiner, Methods of in
vitro toxicology, Food Chem Toxicol, 40 (2002) 193-236.
[3] T. Vanhaecke, V. Rogiers, Hepatocyte cultures in drug metabolism and
toxicological research and testing, Methods Mol Biol, 320 (2006) 209-227.
[4] C. Nardin, Hirt, T., Leukel, J., Meier, W., Polymerized ABA Triblock
Copolymer Vesicles, Langmuir, 16 (2000) 1035-1041.
[5] T. Henkens, P. Papeleu, G. Elaut, M. Vinken, V. Rogiers, T. Vanhaecke,
Trichostatin A, a critical factor in maintaining the functional differentiation of
primary cultured rat hepatocytes, Toxicol Appl Pharmacol, 218 (2007) 64-71.
[6] V. Rogiers, A. Vercruysse, Rat hepatocyte cultures and co-cultures in
biotransformation studies of xenobiotics, Toxicology, 82 (1993) 193-208.
94
Toxicity of the nanoreactors
[7] E.F. Brandon, C.D. Raap, I. Meijerman, J.H. Beijnen, J.H. Schellens, An
update on in vitro test methods in human hepatic drug biotransformation
research: pros and cons, Toxicol Appl Pharmacol, 189 (2003) 233-246.
[8] P. Papeleu, T. Vanhaecke, T. Henkens, G. Elaut, M. Vinken, S. Snykers, V.
Rogiers, Isolation of rat hepatocytes, Methods Mol Biol, 320 (2006) 229-237.
[9] S. Nagata, P. Golstein, The Fas death factor, Science, 267 (1995) 14491456.
[10] K. Ito, N. Kiyosawa, K. Kumagai, S. Manabe, N. Matsunuma, T. Yamoto,
Molecular mechanism investigation of cycloheximide-induced hepatocyte
apoptosis in rat livers by morphological and microarray analysis, Toxicology,
219 (2006) 175-186.
[11] J.M. Worle-Knirsch, K. Pulskamp, H.F. Krug, Oops they did it again!
Carbon nanotubes hoax scientists in viability assays, Nano Lett, 6 (2006)
1261-1268.
[12] T. Laaksonen, H. Santos, H. Vihola, J. Salonen, J. Riikonen, T. Heikkila, L.
Peltonen, N. Kumar, D.Y. Murzin, V.P. Lehto, J. Hirvonen, Failure of MTT as a
toxicity testing agent for mesoporous silicon microparticles, Chem Res Toxicol,
20 (2007) 1913-1918.
95
Chapter 6
Immunogenicity of the nanoreactors
Immunogenicity of the nanoreactors
6 Immunogenicity of the nanoreactors
6.1 Introduction
The enzyme containing nanoreactors constructed in this study have unique
physicochemical properties which make them promising candidates for drug
delivery. However, the application of the nanoreactors as intravenous enzyme
delivery vehicles depends on avoiding rapid elimination from the systemic
circulation by cells of the immune system. Nanoparticle recognition and uptake
by immune cells may occur through different pathways [1]. The common initial
step is the adsorption of opsonins to the surface of the nanoparticles. Opsonins
are blood proteins that bind to the particle, facilitate the recognition by special
immune cells and/or signal immune cells to ingest or destroy the particle.
Nanoparticles whose surfaces are not shielded to prevent adsorption of
opsonins are removed from the bloodstream within seconds [2]. The protein
binding to the surface of nanoparticles depends on their surface characteristics
like size, charge, hydrophobicity and composition [3]. In this context, the
polymethyloxazoline (PMOXA) polymer was chosen as outer block for the
nanoreactors since it has good stealth properties, able to evade binding of
opsonins
[4,
nanoreactors
5].
However,
constructed
of
the
immunogenicity
PMOXA-PDMS-PMOXA
of
multi-component
triblock
copolymers,
enzymes and porins has never been investigated before. In theory, the
proteins are protected from the environment since the porins are embedded
inside the thick polymeric membrane while the enzymes are encapsulated
inside the inner space of the particles.
Consequently, it is very important to test the interaction of the thymidine
phosphorylase
containing
nanoreactors
with
the
immune
system.
As
nanotechnology-based products are relatively new drug candidates, there are
currently
no
guidelines
or
harmonized
standards
for
assessing
their
immunogenicity. The main immunogenic properties addressed here are the
inflammatory and antigenic potency of the nanoparticles. These two key
99
Chapter 6
questions will not reveal the complete immunogenic picture, but will provide
essential insight in the behavior of the nanoreactors in a complex in vivo
environment. Important preclinical parameters to examine in this context are;
endotoxin load, uptake by macrophages, cytokine production, nitric oxide
production and nanoparticle-specific antibody responses [6]. Therefore, this
chapter describes the investigation of the above mentioned parameters in
order to evaluate the nanoreactors further as potential enzyme delivery
particles.
6.2 Materials and methods
6.2.1 Production of nanoreactors
The TPE.coli (thymidine phosphorylase from E.coli) encapsulating nanoreactors
were constructed via the solvent evaporation method as described in detail in
section 3.2.3 [7]. In addition to the nanoreactors containing enzyme and porin
(NP+TP+Tsx), other nanoreactor-compositions were prepared and used in
parallel as control samples; nanoparticles with enzyme without porin (NP+TPTsx) and empty nanoparticles (NP-TP-Tsx).
6.2.2 Macrophage cell culture
For the in vitro macrophage studies, 5 C57BL/6 mice were each administered
with 3 ml thioglycolate solution intraperitoneally (ip). After 4 days, mice were
euthanized and peritoneal macrophages were collected by flushing the
peritoneal cavity with 10 ml sucrose solution. The cell suspension was
centrifuged at 1400 rpm for 8 min at 4°C and resuspended in complete RPMI
1640 medium (supplemented with 10% heat-inactivated Fetal Calf Serum, 100
mg/ml penicillin, 100 mg/ml streptomycin, 0.03% L-glutamine, 1 mM nonessential amino-acids, 1 mM Na-pyruvate and 0.02 mM mercapto-ethanol).
These thioglycolate-elicited peritoneal macrophages were plated out at 106
cells/ml and incubated at 37°C. After 4 hours of adhesion the non-adherent
cells were washed away and half of the adherent cells (> 98% macrophages)
were pre-stimulated with 100 U/ml interferon-γ
100
(IFNγ) to prime the
Immunogenicity of the nanoreactors
macrophages. The following day different concentrations of TP-NRs (50 µg
polymer/ml, 100 µg polymer/ml and 500 µg polymer/ml) were added to the
macrophage cultures and duplicate samples were prepared. After 24 h, 48 h
and 72 h of incubation at 37°C, the cell culture supernatants were harvested
and stored at -20°C until they were analyzed for their nitric oxide and cytokine
load.
6.2.3 In vivo inflammatory response study
To study the inflammatory potency of the nanoparticles in vivo, naïve C57BL/6
mice (5 mice per experimental group) received a intraperitoneal injection of
100 µl TP-NRs (50 µg polymer/mouse). Ip injection of PBS and a sub-lethal
dose of LPS (50 µg/mouse) were used as negative and positive control,
respectively. After 6 hours mice were euthanized and peritoneal macrophages
and blood samples were collected. Peritoneal macrophages were cultured at
37°C and after 48 hours the supernatant was analyzed for the detection of the
inflammatory cytokines IL-6, IL-1β and TNF- and for NO.
Similarly, the
cytokines were evaluated in the serum.
6.2.4 Cytokine analysis and NO detection
Concentrations of TNF- (DuoSet ELISA Development System, R&D Systems),
IL-6 and IL-1 (Pharmingen, BD Biosciences) in serum and cell supernatants
were determined by sandwich ELISA as recommended by the suppliers.
The amount of nitric oxide (NO) in the cell supernatants was measured by
detecting the more stable NO-derivative nitrite (NO2-). Nitrite quantification in
cell supernatants was assayed by a standard Griess reaction [8]. Therefore,
100 μl of cell-free supernatants were added to an equal volume of Griess
reagent
(0.5%
sulfanilamide
and
0.05%
N-1
naphthylethylenediamine
hydrochloride in 2.5% phosphoric acid). After a 10-minute incubation at room
temperature absorbance at 540 nm was recorded. A standard curve was
generated with known concentrations of NaNO2 in culture medium. All
reagents for the Griess reaction were obtained from Sigma.
101
Chapter 6
6.2.5 Immunization of animals
Naive C57BL/6 mice were split into 6 groups each with 4 animals per
experimental group. Mice in group 1 received an intravenous (iv) injection of
200 µl sterile PBS, mice in group 2 received an iv injection of 200 µl sterile
PBS containing NP+TP+Tsx (final concentration of 666 µg polymer/mouse, 60
µg TP/mouse and 5 µg Tsx/mouse), mice in group 3 received an iv injection of
200 µl NP+TP-Tsx (final concentration of 666 µg polymer/mouse and 60 µg
TP/mouse), mice in group 4 received an iv injection of 200 µl NP-TP-Tsx (final
concentration of 666 µg polymer/mouse), mice in group 5 received an iv
injection of 200 µl sterile PBS containing free TPE.coli (final concentration of 60
µg TP/mouse) and mice in group 6 received an intraperitoneal (ip) injection of
200 µl sterile PBS containing free TPE.coli bound to the adjuvant Alum (final
concentration of 60 µg TP/mouse). Imject® Alum (Pierce) is a solution of
aluminium hydroxide and magnesium hydroxide and was mixed with TPE.coli
according to the manufacturer instructions. After the initial injection, 2 boost
injections were performed with every time ten days in between each injection.
Several hours before each injection, approximately 100 µl blood was collected
via orbital blood extraction of each animal and blood was stored at 4°C. The
samples were then centrifuged and serum was stored at -20°C until it was
further analyzed.
6.2.6 Antibody detection
Immunized mice sera were analyzed for specific anti-TP or anti-Tsx antibodies
by ELISA. Briefly, microtiter plates (MaxiSorp®, Nunc) were each coated with 1
µg/ml TPE.coli or Tsx in 0.1 M NaHCO3 pH 8.2 at 4°C overnight. The wells were
washed and residual protein binding sites were blocked by adding 200 µl/well
of 10% fetal calf serum in PBS for 2 hours at room temperature. Afterwards
serial dilutions of sera in blocking buffer were added and incubated overnight
at 4°C. Bound specific antibodies were detected by using a goat anti-mouse
antibody conjugated to Horse Radish Peroxidase (HRP) (Sigma).
102
Immunogenicity of the nanoreactors
After adding the substrate tetramethylbenzidine (diluted in a substrate-buffer
containing H2O2) the reaction was stopped after three minutes with 50 µl of
0.1 N H2SO4 and the optical density was measured at 405 nm. In between
each step the wells were washed extensively with PBS containing 0.05%
Tween. Uncoated plates were treated in exactly the same way as controls to
detect background signal.
6.3 Results
6.3.1 Ex vivo macrophage response study
An important prerequisite of enzyme delivery devices for the use in enzyme
replacement therapy is that they have long circulation times within the
bloodstream
and
do
not
alert
the
immune
system
or
induce
overt
inflammation. An important factor in the clearance of foreign compounds from
the circulation and the initiation of inflammatory responses is their recognition
by the mononuclear phagocyte system, most notably macrophages [9].
Macrophages are cells that stimulate other immune cells and remove foreign
materials by ingesting them through a process known as phagocytosis. The
rapid sequestration of intravenously injected nanoreactors from the blood by
macrophages would be problematic for their application as long circulating
devices in enzyme replacement therapy. Therefore, it is very important that
TP-NRs are not recognized by macrophages and are immunologically inert. To
investigate this, the impact of TP-NRs on the inflammatory status of
macrophages was tested.
For ex vivo studies the effect of nanoreactors on thioglycolate-elicited
peritoneal macrophages isolated from C57BL/6 mice was analyzed [10]. Cells
were plated in complete RPMI medium and incubated at 37°C. After 4 hours of
adhesion, non-adherent cells were washed away, leaving almost pure (> 98%)
macrophages in the plate. Half of these cells were pre-stimulated with 100
U/ml of the known macrophage activating cytokine interferon- (IFN) to prime
the macrophages [11]. The following day, different concentrations of TP-NRs
(50 µg polymer/ml, 100 µg polymer/ml and 500 µg polymer/ml) were added
103
Chapter 6
to the macrophage cultures and duplicate samples were prepared to assess
the induction of inflammatory mediators. Plain polymeric nanoparticles without
porin and without enzyme (NP-TP-Tsx) and nanoreactors without porin but
with enzyme (NP+TP-Tsx) were used as controls and added at the same
concentrations as the complete TP-NRs (NP+TP+Tsx). The free bacterial
enzyme TPE.coli (5 µM) was added to investigate the shielding effectiveness of
encapsulation.
PBS
was
the
negative
control,
while
the
known
pro-
inflammatory bacterial compound lipopolysaccharide (LPS) (100 ng/ml) served
as positive control.
In this context it is very important to note that all the nanoparticle
preparations used in this experiment need to be clean of LPS, in order to
exclude LPS dependent immune responses [12]. Given that Tsx is a Gramnegative bacterial outer membrane protein, the risk of LPS contamination, a
potent endotoxin, is considerable. Therefore, nanoreactor samples used in this
study were routinely checked by the LAL-assay to determine the LPS content
(see also Chapter 3). Only the samples with an endotoxin unit lower than 50
EU/ml were used.
After 24 h, 48 h and 72 h the cell culture supernatants were harvested and
analyzed via ELISA for the presence of the inflammatory cytokines tumor
necrosis factor- (TNF-) and interleukin-6 (IL-6) [13]. These cytokines are
produced by activated macrophages and are the principal mediators of the
acute
phase
immune
response.
Another
indicator
for
inflammatory
macrophage activation is nitric oxide (NO) production [14].
The results show that none of the nanoparticles tested (NP-TP-Tsx, NP+TPTsx, NP+TP+Tsx) induce NO, TNF- or IL-6 production by naïve macrophages
(Figure 6-1). Interestingly, the free enzyme significantly provoked the
secretion
of
all
three
inflammatory
mediators.
This
means
that
the
encapsulation of the enzyme in polymer particles indeed protects the protein
from the immune system. Additionally, these results unequivocally indicate
that the polymers possess good stealth properties towards peritoneal
104
Immunogenicity of the nanoreactors
macrophages and hence are not recognized by naïve macrophages as foreign
molecules.
However, on IFN- primed macrophages, complete TP-NRs (NP+TP+Tsx) dosedependently elevate the amounts of NO, TNF- and IL-6 produced (Figure 61). Since no significant responses were found in the other conditions (NP-TPTsx, NP+TP-Tsx) we can conclude that the induction of inflammatory
mediators is porin-dependent. Importantly however, TP-NR-induced TNF-
levels and IL-6 levels are much lower as compared to the positive control and
the free enzyme. This indicates that even on pre-activated macrophages the
inflammatory effect of the nanoreactors is very limited; again highlighting that
encapsulation efficiently shields the inflammatory potency of free TP.
105
Chapter 6
Figure 6-1: Macrophage response to different compositions of nanoparticles after 24 h,
48 h and 72 h of incubation. PBS was the negative control, while free TPE.coli (5 µM) and
LPS (100 ng/ml) served as positive controls. (A) NO production (measured as the more
stable NO-derivative NO2-) of naive macrophages, (B) NO production (measured as the
more stable NO-derivative NO2-) of IFN- primed macrophages, (C) TNF- production of
naive macrophages, (D) TNF- production of IFN- primed macrophages, (E) IL-6
production of naive macrophages, (F) IL-6 production of IFN- primed macrophages.
106
Immunogenicity of the nanoreactors
6.3.2 In vivo macrophage response study
The inflammatory potency of the nanoparticles was also investigated in vivo.
Therefore,
naive C57BL/6 mice (5 mice per experimental group) received a
intraperitoneal injection of 100 µl TP-NRs (final concentration of 500 µg
polymer/ml). Intraperitoneal injection of PBS and a sub-lethal dose of LPS (50
µg/mouse) were used as negative and positive control, respectively. After 6 h
mice were euthanized and peritoneal macrophages and blood samples were
collected. Peritoneal macrophages were cultured at 37°C and after 48 h the
supernatant was analyzed for the detection of the inflammatory cytokines IL6, IL-1β and TNF- and for NO. Similarly, the cytokines were evaluated in the
serum. Only low amounts of IL-6, IL-1β and TNF- could be found in
macrophage-supernatant and in serum of mice treated with TP-NRs, while high
concentrations were detected in mice injected with LPS (Table 6-1).
Table 6-1: Concentrations of NO2- and inflammatory cytokines produced by peritoneal
macrophages (PECs) and serum after ip administration of PBS, TP-NRs (500 µg/ml
NP+TP+Tsx) and LPS (50 µg/mouse) in mice. (*n.e. = not evaluated)
NO2- produced by
PECs (µM)
NO2- produced in
serum (µM)
TNF- produced by
PECs (pg/ml)
TNF- produced in
serum (pg/ml)
IL-6 produced by
PECs (pg/ml)
IL-6 produced in
serum (pg/ml)
IL-1 produced by
PECs (pg/ml)
IL-1 produced in
serum (pg/ml)
PBS
TP-NRs
LPS
0.0
1.5
>100
n.e.*
n.e.*
n.e.*
147.5
81.9
560.2
0.0
0.0
324.0
7298.3
7668.8
28666.0
98.6
45.5
38811.0
7.6
0.0
1164.1
0.0
0.0
183.5
107
Chapter 6
6.3.3 Antibody response study
Other important components of the immune system that are involved in the
removal of pathogens from the circulation are antibodies. The formation of
antibodies against antigenic epitopes of the nanoreactors (and subsequent
antigen-binding) can contribute to altered nanoparticle biodistribution in the
form of rapid clearance. The antibody-coated particle can either directly be
recognized by phagocytes via binding to Fc-receptors or they can activate the
complement system leading to complement receptor-mediated phagocytosis.
The complement system consists of different proteins (opsonins) that bind to
the antigen-antibody complex. Subsequently, this system triggers a cascade of
biochemical cleaving reactions resulting in the removal of the antigen from the
circulation.
Since the nanoreactors contain bacterial proteins, there is a realistic chance
that upon contact with the blood, production of antibodies against these
foreign proteins is induced. Nevertheless, as long as the nanoreactors remain
stable in the bloodstream, the proteins are in theory shielded by the polymer
and
not
(completely)
exposed
to
the
environment.
The
longer
the
nanoreactors can escape from recognition by the immune system, the longer
they can provide their enzymatic activity in situ and the more promising their
use in enzyme replacement therapy will be. As such, the investigation of
antibody responses against the nanoreactors is an important issue in
evaluating their non-immunogenicity and shielding capacity.
For this study, again different formulations of nanoparticles were prepared:
nanoparticles with enzyme and porin (NP+TP+Tsx), nanoparticles with enzyme
without porin (NP+TP-Tsx) and empty nanoreactors (NP-TP-Tsx). The size (via
DLS), enzyme activity (via spectrophotometer) and encapsulation efficiency
(via comparative SDS-PAGE analysis) of the nanoreactors was determined as
previously described. The nanoparticles containing enzyme are around 200 nm
in diameter, are enzymatically active and show an encapsulation efficiency of
10% resulting in a final enzyme concentration of 0.6 µg/µl. The empty
nanoparticles are considerably smaller, with a diameter around 100 nm, and
108
Immunogenicity of the nanoreactors
obviously lack enzymatic activity. In addition, all samples were confirmed to
be endotoxin-free (via the LAL-test) before administration in vivo. PBS was
used as negative control and free TP and TP adsorbed to the adjuvant Alum
were used as positive controls. This last control was used because adding an
adjuvant to the antigen stimulates an improved immune response compared
to the antigen alone [15]. 6 groups of C57BL/6 mice (4 mice per experimental
group) were immunized with the different nanoreactor formulations and
control samples, as schematically represented in Figure 6-2. At different time
points blood samples were collected and the sera of the immunized animals
were analyzed for the presence of specific antibodies against the enzyme TP or
the porin Tsx using ELISA.
Figure 6-2: Schematic representation of the followed immunization scheme and the
protocol of the antibody response study.
The results clearly show elevated anti-TP antibody levels in mice immunized
with nanoparticles containing TP (Figure 6-3). More surprisingly, the amount of
specific TP antibodies found in mice immunized with the nanoreactors was
even higher than the antibody titer detected in mice treated with the free TP.
Figure 6-4 demonstrates visibly that the responses to the TP containing
nanoparticles are stronger than the responses to the free TP. They are even in
109
Chapter 6
the same range than the responses to the free enzyme in complex with the
adjuvant Alum. These results are rather unexpected and demonstrate that the
bacterial
enzyme
is
not
protected
against
antibody
response
by
the
nanoparticle at all. Although the enzyme is strongly associated with the
particle, it is clearly not shielded from the environment in the nanoreactor
samples used for this study. Hence, the particle appears to act as an adjuvant
in presenting the enzyme to the immune system. The adjuvant effect of these
nanoreactors is in strong contrast with the envisioned shielding effect and
questions their composition and thereby their therapeutic potential.
Although the antibody response to the enzyme is the most important
characteristic to evaluate at this point, the antibody response to the porin was
also measured in the sera of the immunized mice. The level of anti-Tsx
antibodies detected in mice immunized with nanoparticles containing Tsx
(NP+TP+Tsx) are slightly higher than the response measured in the mice
immunized with the formulations lacking Tsx (Figure 6-4 and Figure 6-5).
Nevertheless, this increase is rather low and does not significantly differ from
the anti-Tsx response in mice that did not receive the antigen at all. Note that
the amount of Tsx present in the administered dose of nanoreactors is
considerably lower than the amount of TP (5 µg Tsx to 60 µg TP per mouse).
However, from these data we can conclude that in response to a relevant dose
of nanoreactors, the specific antibody response to the porin is rather low.
110
Immunogenicity of the nanoreactors
Figure 6-3: ELISA results of TP specific antibody titers measured in sera of mice
immunized with different formulations of nanoreactors. PBS acts as negative control, TP
and TP + Alum act as positive controls. Nanoreactors containing the enzyme TP elicit
higher antibody titers than free enzyme. The antibody response to the nanoreactors is in
the same range than the response to the enzyme in complex with Alum adjuvant. Error
bars represent mean + standard deviation (n = 4).
Figure 6-4: ELISA results of serum samples after the second boost injection (30 days
after the first injection). The upper horizontal row of plates are the uncoated plates, the
middle row are the plates coated with Tsx and the lowest row are the plates coated with
TP. The sera of the mice immunized with six different formulations (6 vertical rows)
were added in different dilutions (in duplicate) and the presence of specific anti-TP or
anti-Tsx antibodies was detected.
111
Chapter 6
Figure 6-5: ELISA results of Tsx specific antibody titers measured in sera of mice
immunized with different formulations of nanoreactors. Nanoreactors containing the
porin Tsx elicit slight increased antibody titers. Error bars represent mean + standard
deviation (n = 4).
6.4 Conclusion and discussion
The data from the ex vivo and in vivo macrophage activation studies show that
the effects of TP-NRs on macrophage activation and concomitant inflammation
are minor. Only a slight pro-inflammatory effect was seen on strongly IFN-
primed macrophages. Importantly, no inflammatory effect is detected in vivo.
Since there is only a limited diffusion of particles from the peritoneum to the
bloodstream, the level of cytokines present in the serum after intraperitoneal
injection of the nanoreactors is rather an indirect indication of the potent
inflammatory response. However, strong inflammatory responses in the
peritoneum are usually reflected in the blood as well. Since the ip injected
nanoparticles do not induce high cytokine levels in the local peritoneal
macrophages nor in the serum, we can safely conclude that the nanoreactors
are
immunologically
inert
towards
macrophages.
Clearly,
the
stealthy
polymers are able to protect the enzyme from recognition by these
112
Immunogenicity of the nanoreactors
phagocytotic
cells.
The
slight
inflammatory
effect
on
IFN-
primed
macrophages in vitro is probably due to surface exposed parts of the bacterial
porins that are recognized by macrophages. Previously, it was reported that
the porins are mainly burried in the polymeric membrane that is two- to threefold thicker than conventional lipid bilayers [16, 17]. The immunogenic
response to the porins can probably be minimized by choosing thicker PMOXA
side-blocks that shield the incorporated porins better from the environment.
These results, demonstrating that the TP-NRs induce no acute innate immune
response after intraperitoneal injection, are a first important step in the
immunologic characterization of the nanoparticles. These observations are in
good agreement with the findings of Broz and coworkers [18] who also
reported
the
absence
of
non-specific
binding
of
PMOXA-PDMS-PMOXA
nanoreactors to macrophages. Only when a specific ligand was attached to the
nanoparticles, able to bind to the scavenger receptor A1 of the macrophages,
receptor-specific binding and uptake could be demonstrated. This low
nonspecific binding supports the stealth properties of the carrier and makes
them promising candidates for use in drug delivery.
In a next phase, the nanoreactors were administered intravenously and
antibody responses were analyzed in the sera after several boost injections.
Surprisingly, the antibody response to the nanoreactor-encapsulated-enzyme
is higher than to the free enzyme. These results were rather unforeseen as
they show the opposite outcome than expected. The nanoreactors even seem
to act as an adjuvant increasing the level of specific anti-TP antibody
production. Because it is generally accepted that physical association of
antigen and adjuvant is required for adjuvant activity, the observed effect is
probably not due to the release of free enzyme by fast degradation of the
nanoreactors. More likely, the adjuvant effect of the nanoreactors is caused by
nanoreactor-associated enzymes that are not nicely encapsulated inside the
particles but rather exposed to the environment. As such, the nanoreactors
present the enzyme as repeats on their surface acting as an antigenpresenting platform, thereby provoking higher antibody responses. A number
of other studies also described adjuvant properties of nanoparticles [19-23].
113
Chapter 6
For
example,
polymethylmethacrylate
nanoparticles
induce
long-lasting
antibody titers in HIV whole virus vaccine in mice. The antibody response was
even 100 times higher than when a traditional adjuvant was used [24].
Obviously, for the development of vaccines the adjuvant effect is beneficial.
However, since strong antibody responses will cause fast elimination from the
circulation, such effect is not desirable for enzyme replacement therapy
purposes. Furthermore, the antibody response will become stronger after
several boost injections, hereby devastating the therapeutic effect of life-long
needed enzyme replacement therapy.
Taken together, the nanoreactors do not provoke acute inflammatory
responses on macrophages, but they do induce a specific antibody response
against the enzyme. These observations limit the potential use of these
particular nanoreactors as enzyme delivery particles for enzyme replacement
purposes
significantly.
Comprehensive
studies
to
reveal
the
detailed
composition of the nanoreactors are necessary to further understand the
critical
parameters
that
determine
their
antigenic
and
even
adjuvant
properties.
6.5 References
[1] B.S. Zolnik, A. Gonzalez-Fernandez, N. Sadrieh, M.A. Dobrovolskaia,
Nanoparticles and the immune system, Endocrinology, 151 458-465.
[2] D.E. Owens, 3rd, N.A. Peppas, Opsonization, biodistribution, and
pharmacokinetics of polymeric nanoparticles, Int J Pharm, 307 (2006) 93-102.
[3] M.A. Dobrovolskaia, S.E. McNeil, Immunological properties of engineered
nanomaterials, Nat Nanotechnol, 2 (2007) 469-478.
[4] M.C. Woodle, C.M. Engbers, S. Zalipsky, New amphipatic polymer-lipid
conjugates forming long-circulating reticuloendothelial system-evading
liposomes, Bioconjug Chem, 5 (1994) 493-496.
[5] S. Zalipsky, C.B. Hansen, J.M. Oaks, T.M. Allen, Evaluation of blood
clearance rates and biodistribution of poly(2-oxazoline)-grafted liposomes, J
Pharm Sci, 85 (1996) 133-137.
[6] M.A. Dobrovolskaia, P. Aggarwal, J.B. Hall, S.E. McNeil, Preclinical Studies
To Understand Nanoparticle Interaction with the Immune System and Its
Potential Effects on Nanoparticle Biodistribution, Mol Pharm, (2008).
[7] C. Nardin, Hirt, T., Leukel, J., Meier, W., Polymerized ABA Triblock
Copolymer Vesicles, Langmuir, 16 (2000) 1035-1041.
114
Immunogenicity of the nanoreactors
[8] L.C. Green, D.A. Wagner, J. Glogowski, P.L. Skipper, J.S. Wishnok, S.R.
Tannenbaum, Analysis of nitrate, nitrite, and [15N]nitrate in biological fluids,
Anal Biochem, 126 (1982) 131-138.
[9] A.U. Daniels, F.H. Barnes, S.J. Charlebois, R.A. Smith, Macrophage
cytokine response to particles and lipopolysaccharide in vitro, J Biomed Mater
Res, 49 (2000) 469-478.
[10] I. Catelas, A. Petit, R. Marchand, D.J. Zukor, L. Yahia, O.L. Huk,
Cytotoxicity and macrophage cytokine release induced by ceramic and
polyethylene particles in vitro, J Bone Joint Surg Br, 81 (1999) 516-521.
[11] H.W. Murray, G.L. Spitalny, C.F. Nathan, Activation of mouse peritoneal
macrophages in vitro and in vivo by interferon-gamma, J Immunol, 134
(1985) 1619-1622.
[12] V.M. Hitchins, K. Merritt, Decontaminating particles exposed to bacterial
endotoxin (LPS), J Biomed Mater Res, 46 (1999) 434-437.
[13] A. Prabhu, C.E. Shelburne, D.F. Gibbons, Cellular proliferation and
cytokine responses of murine macrophage cell line J774A.1 to
polymethylmethacrylate and cobalt-chrome alloy particles, J Biomed Mater
Res, 42 (1998) 655-663.
[14] P. Tripathi, P. Tripathi, L. Kashyap, V. Singh, The role of nitric oxide in
inflammatory reactions, FEMS Immunol Med Microbiol, 51 (2007) 443-452.
[15] H.F. Stills, Jr., Adjuvants and antibody production: dispelling the myths
associated with Freund's complete and other adjuvants, Ilar J, 46 (2005) 280293.
[16] C. Nardin, W. Meier, Hybrid materials from amphiphilic block copolymers
and membrane proteins, J Biotechnol, 90 (2002) 17-26.
[17] V. Pata, N. Dan, The effect of chain length on protein solubilization in
polymer-based vesicles (polymersomes), Biophys J, 85 (2003) 2111-2118.
[18] P. Broz, S.M. Benito, C. Saw, P. Burger, H. Heider, M. Pfisterer, S.
Marsch, W. Meier, P. Hunziker, Cell targeting by a generic receptor-targeted
polymer nanocontainer platform, J Control Release, 102 (2005) 475-488.
[19] B. Slutter, L. Plapied, V. Fievez, M.A. Sande, A. des Rieux, Y.J. Schneider,
E. Van Riet, W. Jiskoot, V. Preat, Mechanistic study of the adjuvant effect of
biodegradable nanoparticles in mucosal vaccination, J Control Release, 138
(2009) 113-121.
[20] S. Ribeiro, S.G. Rijpkema, Z. Durrani, A.T. Florence, PLGA-dendron
nanoparticles enhance immunogenicity but not lethal antibody production of a
DNA vaccine against anthrax in mice, Int J Pharm, 331 (2007) 228-232.
[21] T. Uto, X. Wang, K. Sato, M. Haraguchi, T. Akagi, M. Akashi, M. Baba,
Targeting of antigen to dendritic cells with poly(gamma-glutamic acid)
nanoparticles induces antigen-specific humoral and cellular immunity, J
Immunol, 178 (2007) 2979-2986.
[22] J. Kreuter, Nanoparticles as adjuvants for vaccines, Pharm Biotechnol, 6
(1995) 463-472.
[23] A.K. Jain, A.K. Goyal, P.N. Gupta, K. Khatri, N. Mishra, A. Mehta, S.
Mangal, S.P. Vyas, Synthesis, characterization and evaluation of novel triblock
copolymer based nanoparticles for vaccine delivery against hepatitis B, J
Control Release, 136 (2009) 161-169.
115
Chapter 6
[24] F. Stieneker, J. Kreuter, J. Lower, High antibody titres in mice with
polymethylmethacrylate nanoparticles as adjuvant for HIV vaccines, Aids, 5
(1991) 431-435.
116
Chapter 7
Biodistribution of the nanoreactors
Biodistribution of the nanoreactors
7 Biodistribution of the nanoreactors
7.1 Introduction
The delivery of therapeutic proteins to the body is a challenge due to their
limited circulation time, which prevents long-term therapeutic efficacy. During
the past decades many efforts were made to develop appropriate carrier
systems to enhance the size and thereby the circulation half-life of therapeutic
drugs. Nanoparticles are attractive drug delivery candidates because they are
large enough to avoid fast clearance by the kidneys (size-selective cut-off for
glomerular filtration is approximately 60 kDa) [1] and small enough to prevent
removal from the circulation by simple filtration of the first capillary barrier
encountered. Nevertheless, the clearance behavior and tissue distribution of
intravenously injected nanoparticles is also greatly influenced by their surface
characteristics. Many conventional nanoparticles are rapidly removed from the
bloodstream due to clearance by macrophages of the mononuclear phagocyte
system (MPS), also called the reticuloendothelial system (RES). In particular
Kupffer cells, which are specific macrophages from liver and spleen, are known
to be involved in recognizing and eliminating nanoparticles [2]. This fast
clearance strongly prevents their application in controlled drug delivery.
However, surface engineering can lead to particles that can evade, at least to
some extent, MPS uptake and thereby provoke prolonged residence in the
blood. Coating the nanoparticle surface with a suitable hydrophilic polymer
such as poly(ethylene glycol) (PEG) has been shown to confer long circulation
properties [3]. The presence of the hydrophilic coating is thought to sterically
stabilize the particles against opsonization and phagocytosis [4]. Except from
PEG, also other hydrophilic polymers are found to be effective steric
stabilizers. Amongst them are the oxazoline based polymers which are claimed
by Zalipsky and Woodle to possess stealth properties to the same extend as
PEG [5]. It has been shown that the poly(methyl-oxazoline) or PMOXA
polymer, conjugated to a lipid, is able to convey long circulation and low
119
Chapter 7
hepatosplenic uptake [6]. Despite these initial encouraging results, it is not
verified further whether the usefulness of the PMOXA polymer also extends to
other applications.
In this chapter, the in vivo behavior of the new thymidine phosphorylase
containing
PMOXA-PDMS-PMOXA
nanoreactors
is
investigated.
The
biodistribution of such complex nanoparticle formulations (including polymer,
enzyme and porine) is not yet known and it is very important to unravel their
in vivo fate. This is not only of high relevance for their potential use in enzyme
replacement therapy, but also for the further exploitation of the PMOXA-PDMSPMOXA particles as drug delivery devices in general.
7.2 Materials and methods
7.2.1 Nanoreactor synthesis and characterization
The TPE.coli (thymidine phosphorylase from E.coli) encapsulating nanoreactors
(TP-NRs) were constructed via the solvent evaporation method as described in
detail in section 3.2.3 [7]. Radiolabeling was achieved, either by labeling the
nanoparticles directly after their production process, or by incorporating
radiolabeled enzyme, as will be explained in the next sections.
7.2.2 Radiolabeling of particles with technetium-99m
The labeling of the nanoreactors with technetium (99mTc) was performed after
their production, on the hydroxyl end-groups of the polymer, according to
Wunderlich et al. [8]. Briefly, 200 µl of the nanoreactor suspension (6.6 mg
polymer/ml) was mixed with 20 µl of freshly prepared SnCl2 dissolved in 0.05
M HCl (0.1 mg/ml). After shaking for 15 minutes (at room temperature), 200
µl of
99m
TcO4 (± 2 mCi or 74 MBq) was added and the vial was shaken again
for 15 minutes.
99m
Tc pertechnetate was obtained from a
99
Mo/99mTc
radionuclide generator (Drytec, GE Healthcare). Stannous chloride is needed
for the reduction of pertechnetate to create reactive technetium. After the
radiolabeling of the nanoreactors, the radiolabeling yield and stability was
120
Biodistribution of the nanoreactors
determined by instant thin layer chromatography on silica gel strips (ITLC-SG)
(Pall Life Sciences) in physiologic H2O (0.9% NaCl) as solvent. This quality
control was performed directly after and several hours after the radiolabeling.
A drop of the labeled sample was deposited on one end of the strip and the
strip was placed vertically in a beaker containing the solvent. The solvent was
allowed to migrate to the end of the strip. Subsequently, the strips were
removed and cut in half, and the radioactivity of each section was measured in
a gamma counter. Free unreacted
99m
Tc will migrate to the solvent front while
99m
Tc-labeled nanoreactor stays at the application point. The labeling-efficiency
was calculated in percentage as: (counts at application point / (counts at
application point + counts at solvent front)) * 100. An aliquot of the particles
was analyzed (after the radioactive decay) via DLS (size) and SDS-PAGE
(encapsulation efficiency) in order to confirm their integrity and quality after
labeling.
7.2.3 Radiolabeling of enzyme with iodine
The enzyme thymidine phosphorylase was labeled with Iodine-123 (I123) and
Iodine-125 (I125) according to the standard iodination protocol for proteins
using iodogen as oxidizing agent [9, 10]. Iodogen vials were prepared by
dissolving 1 mg iodogen (1,3,4,6-tetrachloro-3,6-di-phenylglycouril) in 1 ml
dichloromethane. 100 µl of this solution was added per glass vial and the
solvent was allowed to evaporate until a dry film at the bottom of the vial was
obtained. The iodogen vials were stored at -20°C until further use. For the
labeling, 1 ml PBS and ±1 mCi I123 (GE Healthcare, UK) or I125 (MP Biomedicals
LLC, California) was added to an iodogen coated vial. For labeling with I123
some cold NaI was added (5 µl of a 0.2 mg/ml stock solution) in order to lower
the specific activity of the radionuclide. As such, the specific activity is in the
same order of magnitude as the I125 specific activity. In a next step, 200 µl
enzyme solution (1 mg TP dissolved in PBS) was added and the vials were
gently agitated at 5-min time intervals over a period of at least 1 hour.
Afterwards, the reaction was stopped by adding 10 µl Na2SO3 (0.15 mg/ml)
and the sample was transferred to a clean glass vial. Quality control was
121
Chapter 7
performed via solid phase extraction by loading an aliquot of the labeled
sample onto a C18 seppak column (Waters). After elution with PBS, the
labeling-efficiency was calculated by comparing the amount of radioactivity on
the column (containing the labeled protein) with the flow through (containing
the free iodine). The labeled enzyme solution was purified further via several
passages through an Ag-filter (Millipore), which selectively captures the free
iodine.
7.2.4 Animal guidelines
Wistar rats and C57BL/6 mice (both from Harlan, Horst, The Netherlands)
were
used
nanoparticles
to
study
and
the
labeled
biodistribution
of
free
Minimum
enzyme.
labeled
enzyme-containing
three
animals
per
experimental group were used. During the entire study, the animals were
housed in stainless steel cages with sawdust bedding. They were kept at an
average room temperature of 24°C, a relative humidity of 50% and a 12-hour
day/night cycle. The study protocol was approved by the Ethics Committee for
animal studies of the Vrije Universiteit Brussel. Guidelines of the National
Institute of Health Principles of Laboratory Animal Care were followed.
7.2.5 Dynamic planar gamma camera imaging procedure
For in vivo dynamic planar gamma camera imaging studies, the animals were
anesthetized
by
intraperitoneal
injection
of
pentobarbital
(50
mg/kg
bodyweight) (CEVA, Brussels, Belgium). The animals were placed on the
collimator surface of a gamma camera and a tail vein intravenous (iv) injection
of
99m
Tc labeled nanoparticles (TP-NRs-99mTc, ±10 mCi/animal) (half-life = 6
hours; energy = 141 keV) or I123 labeled enzyme (TP-I123, ±1 mCi/animal)
(half-life = 13 hours; energy = 159 keV) was performed (n = 3 per tracer).
Images were acquired for 65 minutes. Directly after the imaging period, the
animals were sacrificed by a lethal dose of sodium pentobarbital.
122
Biodistribution of the nanoreactors
7.2.6 Image acquisition and processing
Images were acquired in prone position using a gamma camera (e.cam180,
Siemens Medical Systems) mounted with a low energy high resolution
collimator. A total of 480 frames were acquired in two steps: 300 frames of 1 s
followed by 180 frames of 20 s. The energy window was set at 20% around
141 keV for
99m
Tc and 20% around 159 keV for I123.
All images were processed using AMIDE software [11]. Regions of interest
(ROIs) were drawn around the liver, spleen, lung, heart (left ventricle) and
bladder. Also, one ROI encompassing the total body was specified. All images
were analyzed using the same ROIs. Average counts per minute (cpm) were
measured in all frames and corrected for injected activity based on the
average cpm in the total body ROI for each rat. Time activity curves were
generated for the different organs.
7.2.7 Ex vivo analysis
For long-term ex vivo biodistribution studies, an iv injection of I125 labeled
nanoreactors (NP-TP-I125) or I125 labeled free enzyme (TP-I125) (±100 µCi)
(half-life = 59.6 days; energy = 20 to 40 keV) was performed. The animals
were anesthetized via inhalation of isoflurane during the iv injection of the
labeled compounds. Bloodsamples were collected using microcapillaries at
different time-points after injection to obtain a time-activity blood curve. The
animals were sacrificed 1 hour or 3 days after the injection by intravenous
administration of a lethal dose of pentobarbital (CEVA, Brussels, Belgium).
Major organs/tissues (liver, spleen, lung, heart, kidney and small intestines)
and blood were collected and weighted. The radioactivity content of the blood
samples and the dissected organs and tissues was measured in a gamma
counter. Tissue and organ uptake was calculated as percentage of injected
activity (IA) per tissue or organ corrected for decay.
123
Chapter 7
7.2.8 Statistical analysis
The amount of NP-TP-I125 and TP-I125 in the blood was expressed as average ±
standard deviation (in % injected activity/total blood volume or %IA/TBV) and
compared with a Mann-Whitney test using Graphpad Prism. The test was
considered statistically significant when the P-value was less than 0.05.
7.3 Results
7.3.1 Kinetic analysis of short-term biodistribution
A short-term biodistribution analysis of the nanoreactors and the free enzyme
was performed using in vivo dynamic planar gamma camera imaging with the
following constructs: TP-NRs-99mTc and TP-I123. The labeling-efficiency of the
TP-NRs-99mTc construct was 98%. The stability of the labeled particles was
tested in serum and their size (diameter around 200 nm) was confirmed via
DLS. For the TP-I123 construct, a labeling-efficiency of more than 90% was
obtained after the iodination reaction. Three rats per experimental group
received an iv injection of TP-NRs-99mTc or TP-I123 followed by dynamic planar
imaging during 65 minutes.
During the imaging period a clear difference in biodistribution between the
enzyme containing nanoreactors and the free enzyme was observed. The free
enzyme was almost completely cleared by the hepatobiliary system into the
intestines and to a lesser extent by the kidneys. A slight signal was also found
in the thyroid gland, indicating a fraction of free I123 released from the
enzyme. In contrast, the TP containing nanoreactors were trapped in a large
scale in the liver and the spleen (Figure 7-1).
These results were confirmed by analyzing the time activity curves for the
different organs (Figure 7-2). These curves were obtained by processing the
images using the AMIDE software.
124
Biodistribution of the nanoreactors
Figure 7-1: Gamma-camera images of a rat 5, 10, 20, 40 and 60 minutes after iv
injection of TP-I123 and TP-NRs-99mTc. Each image was scaled to its own maximum pixel
value.
125
Chapter 7
Figure 7-2: Kinetic plots of the % injected activity (% IA) over time in different organs
after iv injection of TP-I123 and TP-NRs-99mTc.
The accumulation of the nanoreactors in liver and spleen suggests interaction
with the mononuclear phagocyte system (MPS). However, the radioactive
signal in these organs is rather high: 60% for liver and 10% for spleen of the
total injected activity. It is possible that the obtained biodistribution pattern of
the nanoparticles is influenced by the presence of colloidal particles of tin
126
Biodistribution of the nanoreactors
oxides which can be formed by the reduction of technetium by stannous
chloride during the radiolabeling process (and which cannot be resolved and
quantified via ITLC). Such interference of colloidal tin was also reported by
others [12] and is correlated to accumulations in liver and spleen due to the
absorbance of the nanoreactors to the surface of the colloidal aggregates. As
such, the organ uptake values can be slightly over-estimated. To circumvent
these problems efforts were made to label the nanoreactors by encapsulating
the labeled enzyme TP-I123. Unfortunately, this was practically not achievable
due to the too short half-life of this radioligand in comparison to the time
needed for the production process of the nanoreactors.
7.3.2 Long-term biodistribution analysis
To perform long-term biodistribution studies, the enzyme was linked to the
radioisotope I125, which has a half-life of 59.6 days. A labeling-efficiency
around 80% was obtained after labeling the enzyme with I125. Purification on
an Ag-filter resulted in a labeling-efficiency high enough (> 90%) for in vivo
use. Subsequently, the labeled enzyme was used to construct nanoreactors.
The difference in biodistribution between the free TP (TP-I125) and the TP
containing nanoreactors (NP-TP-I125) was investigated 1 hour and 3 days after
a single bolus iv injection in rats. Since the energy of the gamma radiation
emitted from I125 is too low for gamma-camera detection, the biodistribution
was analyzed ex vivo via dissection of the major organs and tissues followed
by gamma counting. Also blood samples were collected at different time-points
and analyzed for their radioactive load.
The results indicate that the biodistribution of free and encapsulated TP differs
significantly in Wistar rats (Figure 7-3). Already 1 hour after iv injection, a
considerable higher value for the nanoreactors is observed in the liver and the
spleen compared to the free enzyme. After 3 days, the difference in
hepatosplenic uptake between the two formulations is even more pronounced.
Importantly, no significant difference in blood-load was observed between TPI125 and NP-TP-I125.
127
Chapter 7
Figure 7-3: Biodistribution of TP-I125 (blue) and NP-TP-I125 (red) 1 hour (upper graph)
and 3 days (lower graph) after iv injection in Wistar rats. Three animals per
experimental group were analyzed and means and standard deviations are displayed in
% injected activity (IA) per organ or per tissue.
128
Biodistribution of the nanoreactors
The blood-pool activity was assessed at different time-points after the injection
and the blood-clearance curves (Figure 7-4) indicate that the nanoreactors are
in the end as rapidly cleared from the circulation as the free enzyme. During
the first hour after administration, the blood values are consistently higher for
the nanoreactors compared to the free enzyme (p < 0.05). However, from 1
hour onwards, the two blood curves are very similar, underlining the fast
clearance of the nanoreactors from the blood circulation.
Figure 7-4: The blood-clearance curves of TP-I125 (dashed blue line) and NP-TP-I125 (full
red line). Rats were given a single bolus iv injection. The % injected activity (IA) per
total body volume (TBV) was plotted over time. The TBV was calculated as 7% of the
total body weight. Three animals per experimental group were analyzed and means and
standard deviations are plotted. The result is statistically significant, because the P value
based on a Mann-Whitney test, is 0.023 (< 0.05).
In order to investigate whether similar results could be observed in another
species, the long-term biodistribution study was repeated in C57BL/6 mice.
Three days after iv injection of TP-I125 and NP-TP-I125, the main organs and
tissues were analyzed for their radioactive load using a gamma counter. As
shown in Figure 7-5, exactly the same results could be obtained as in the rats.
Also in mice, the liver and spleen values of NP-TP-I125 are significantly higher
than those for TP-I125, while no difference in blood-pool activity was observed.
129
Chapter 7
Figure 7-5: Biodistribution of TP-I125 (blue) and NP-TP-I125 (red) 3 days after iv injection
in C57BL/6 mice. Three animals per experimental group were analyzed and means and
standard deviations are displayed in % injected activity (IA) per organ or per tissue.
7.4 Conclusion and discussion
In this chapter, we describe the in vivo biodistribution of the new thymidine
phosphorylase
containing
nanoreactors
(TP-NRs).
By
encapsulating
the
enzyme in polymeric nanoreactors, we aim to increase the blood circulation
half-life of the therapeutic enzyme, which is an important prerequisite for their
use in enzyme replacement therapy. First, we investigated the short-term
biodistribution using in vivo gamma camera imaging. Taken together, the
imaging studies with the I123 labeled enzyme revealed an expected fast
clearance of the free enzyme via the hepatobiliary system. The
99m
Tc labeled
nanoreactors however were also cleared rapidly from the blood circulation and
were predominantly trapped in the liver and the spleen. This suggests uptake
of the nanoreactors by the mononuclear phagocyte system (MPS). The
absolute uptake values in liver and spleen might however be somewhat
overestimated due to the formation of colloidal oxidative
130
99m
Tc species formed
Biodistribution of the nanoreactors
during the labeling procedure. Nevertheless, the technetium labeling of the
nanoreactors is in general a rather mild process used to label other
nanoparticles
as
well
[13].
123
encapsulation of the I
Attempts
to
image
the
nanoreactors
via
labeled enzyme failed due to practical reasons.
From the ex vivo biodistribution studies with I125 labeled nanoreactors and I125
labeled enzyme in rats, again, a clear difference in biodistribution between the
free enzyme and the nanoreactor-associated enzyme was observed. The
nanoreactors accumulated mainly in liver and spleen, while the values in these
organs were clearly lower for the free enzyme. Here, the absolute organ
uptake values are considerably lower in comparison with the
99m
Tc experiment.
This is most likely caused by inevitable release of the radionuclide by in vivo
de-iodination and possibly also a partial release of the enzyme from the
particle. Also here, no difference in blood circulation half-life could be
demonstrated between the free enzyme and the nanoreactor-associated
enzyme. Moreover, these results could be confirmed in C57BL/6 mice.
All together, these data indicate that the nanoreactors do not possess the ideal
size and/or surface characteristics to evade hepatosplenic uptake and to
prolong the circulation time of the incorporated enzyme significantly. As such,
the potential advantageous effect of the nanoreactors for use in enzyme
replacement therapy is strongly compromised. It has been emphasized that
the clearance behavior and tissue distribution of intravenously injected drug
carriers are greatly influenced by their surface characteristics and their size.
The
surface
characteristics
determine
whether
the
nanoparticles
are
recognized and taken up by phagocytotic cells of the MPS. Since many
nanoparticles do not possess ideal shielding properties, different approaches
have been developed to circumvent rapid hepatosplenic uptake. One approach
is to administer large doses of placebo carriers in an effort to impair the
phagocytic
capacity
of
macrophages,
thereby
allowing
subsequently
administered particles to remain in systemic circulation for prolonged periods
[14]. Another strategy is transient apoptotic destruction of liver and spleen
macrophages by prior administration of gadolinium chloride [15]. These
strategies, although successful in experimental models, have however little
131
Chapter 7
justification in practice as they suppress the essential defense system of the
body. The size and deformability of nanoparticles plays another critical role in
their clearance by the liver and spleen. Ideally, the size of an engineered longcirculating particle should not exceed 200 nm. If larger, the particle must be
deformable enough to bypass splenic filtration [16]. In addition, there is also a
clear-cut relationship between particle size and the extent to which they reach
the hepatic parenchyma. As a rough approximation, the size of the particles
should be in the range of 120 to 200 nm in diameter to substantially avoid
particle trapping in the liver.
To conclude, our nanoreactors of 200 nanometer size covered with the
stealthy polymer PMOXA had promising characteristics to act as longcirculating devices in theory. Nevertheless, in practice we have observed that
they are not able to evade hepatosplenic uptake and to prolong circulation
times in the blood significantly. Whether this behavior is due to their surface
characteristics (complex core of porin/PMOXA-PDMS-PMOXA), their size (just
too big), their rigidity or to a combination of several parameters requires
further investigation.
7.5 References
[1] A. Edwards, B.S. Daniels, W.M. Deen, Ultrastructural model for size
selectivity in glomerular filtration, Am J Physiol, 276 (1999) F892-902.
[2] E. Sadauskas, H. Wallin, M. Stoltenberg, U. Vogel, P. Doering, A. Larsen,
G. Danscher, Kupffer cells are central in the removal of nanoparticles from the
organism, Part Fibre Toxicol, 4 (2007) 10.
[3] A.L. Klibanov, K. Maruyama, V.P. Torchilin, L. Huang, Amphipathic
polyethyleneglycols effectively prolong the circulation time of liposomes, FEBS
Lett., 268 (1990) 235-237.
[4] D.D. Lasic, F.J. Martin, A. Gabizon, S.K. Huang, D. Papahadjopoulos,
Sterically stabilized liposomes: a hypothesis on the molecular origin of the
extended circulation times, Biochim Biophys Acta, 1070 (1991) 187-192.
[5] M.C. Woodle, C.M. Engbers, S. Zalipsky, New amphipatic polymer-lipid
conjugates forming long-circulating reticuloendothelial system-evading
liposomes, Bioconjug Chem, 5 (1994) 493-496.
[6] S. Zalipsky, C.B. Hansen, J.M. Oaks, T.M. Allen, Evaluation of blood
clearance rates and biodistribution of poly(2-oxazoline)-grafted liposomes, J
Pharm Sci, 85 (1996) 133-137.
[7] C. Nardin, Hirt, T., Leukel, J., Meier, W., Polymerized ABA Triblock
Copolymer Vesicles, Langmuir, 16 (2000) 1035-1041.
132
Biodistribution of the nanoreactors
[8] G. Wunderlich, T. Gruning, B.R. Paulke, A. Lieske, J. Kotzerke, 99mTc
labelled model drug carriers - labeling, stability and organ distribution in rats,
Nucl Med Biol, 31 (2004) 87-92.
[9] T.B. Schulz, R. Jorde, H.L. Waldum, P.G. Burhol, Preparation of 125Ilabeled gastric inhibitory polypeptide, using water-insoluble 1,2,4,6tetrachloro-3 alpha,6 alpha-diphenylglycoluril (Iodo-gen) as the oxidizing
agent, Scand J Gastroenterol, 17 (1982) 379-382.
[10] G.B. Saha, J. Whitten, R.T. Go, Conditions of radioiodination with iodogen
as oxidizing agent, Int J Rad Appl Instrum B, 16 (1989) 431-433.
[11] A.M. Loening, S.S. Gambhir, AMIDE: a free software tool for
multimodality medical image analysis, Mol Imaging, 2 (2003) 131-137.
[12] T. Banerjee, A.K. Singh, R.K. Sharma, A.N. Maitra, Labeling efficiency and
biodistribution of Technetium-99m labeled nanoparticles: interference by
colloidal tin oxide particles, Int J Pharm, 289 (2005) 189-195.
[13] M. Stevanovic, T. Maksin, J. Petkovic, M. Filipic, D. Uskokovic, An
innovative, quick and convenient labeling method for the investigation of
pharmacological behavior and the metabolism of poly(DL-lactide-co-glycolide)
nanospheres, Nanotechnology, 20 (2009) 335102.
[14] S.M. Moghimi, S.S. Davis, Innovations in avoiding particle clearance from
blood by Kupffer cells: cause for reflection, Crit Rev Ther Drug Carrier Syst, 11
(1994) 31-59.
[15] M.J. Hardonk, F.W. Dijkhuis, C.E. Hulstaert, J. Koudstaal, Heterogeneity
of rat liver and spleen macrophages in gadolinium chloride-induced elimination
and repopulation, J Leukoc Biol, 52 (1992) 296-302.
[16] S.M. Moghimi, C.J. Porter, I.S. Muir, L. Illum, S.S. Davis, Non-phagocytic
uptake of intravenously injected microspheres in rat spleen: influence of
particle size and hydrophilic coating, Biochem Biophys Res Commun, 177
(1991) 861-866.
133
Chapter 8
Efficiency of the nanoreactors
Efficiency of the nanoreactors
8 Efficiency of the nanoreactors
8.1 Introduction
MNGIE patients are characterized by thymidine phosphorylase (TP) gene
mutations in association with drastically reduced TP activity and strikingly
elevated levels of the substrates thymidine and deoxyuridine in plasma.
Circulating levels of these substrates range from 4 to 24 µmol/l in MNGIE
patients, while their levels are undetectable in healthy persons [1, 2]. To
characterize the disease and to evaluate new therapies, a MNGIE mouse model
was generated by our collaborator Dr. Hirano and his team from the
Department of Neurology at Columbia University Medical Center. Beside the
enzyme TP, the murine uridine phosphorylase (UP) is also able to convert
thymidine. Therefore, a double knock-out TP-/-UP-/- strain was constructed in
order to establish a MNGIE mouse model [3]. These mice develop several, but
not all, of the typical MNGIE features, such as TP deficiency, elevated
thymidine and deoxyuridine levels, mtDNA depletion and respiratory chain
defects in the brain. The relatively mild phenotype in the animal model in
comparison to the human disease is probably due to a combination of two
factors. First, the levels of nucleosides in tissues of the knock-out animals are
only modestly increased (4- to 65-fold elevated) compared to the more than
100-fold increases of both deoxyuridine and thymidine in tissues of MNGIE
patients. Second, the short lifespan of the animals may not be sufficient to
develop severe alterations of mitochondrial DNA. In order to intensify the
biochemical abnormalities and augment mitochondrial DNA alterations, the TP/-
UP-/- mice are treated with exogenous thymidine, deoxyuridine, or both
(supplemented in drink water). Nevertheless, since the concentrations of the
toxic substrates in the knock-out mice are significantly higher than the
untraceable low levels in naïve mice, the TP-/-UP-/- mice are a powerful model
system to evaluate the potential effect of our TP containing nanoreactors as
enzyme replacement devices. The effect of TP administrations in TP-/-UP-/- mice
137
Chapter 8
has not yet been investigated before, neither in a native, pegylated nor
controlled release form. The only reported attempt of
using enzyme
replacement to treat MNGIE was performed by Moran and coworkers who
reported the use of an erythrocyte entrapped thymidine phosphorylase to treat
one MNGIE patient [4]. Although the substrate concentrations in the plasma
decreased significantly, the patient died several weeks after the therapy due to
pneumonia. Except from this preliminary and single case study, no further
investigations in this area were performed.
In this chapter, the efficiency of the nanoreactors as potential new enzyme
delivery devices for the enzyme replacement therapy of MNGIE is evaluated.
The aim of this study was to investigate whether the thymidine phosphorylase
containing nanoreactors are able to reduce the toxic substrate concentrations
of thymidine and deoxyuridine in vitro and in vivo using a TP-/-UP-/- mouse
model.
8.2 Materials and methods
8.2.1 Production of nanoreactors
The TPE.coli (thymidine phosphorylase from E.coli) encapsulating nanoreactors
(TP-NRs) were constructed via the solvent evaporation method as described in
detail in section 3.2.3 [5].
8.2.2 In vitro assay and sample processing
Plasma samples of a MNGIE patient (100 µl) were incubated with TP-NRs (25
µl containing 12 µg TP) for 30 minutes at 37°C. Free thymidine phosphorylase
(25 µl containing 12 µg TP) and PBS (25 µl) were used as positive and
negative control respectively. Subsequently, the samples were treated with
perchloric acid (PCA) to a final concentration of 0.35 M to precipitate the
proteins. After a short incubation on ice, the samples were centrifuged (10
minutes, 17900 g) and the supernatant was collected. This supernatant was
analyzed immediately or stored at -20°C until further use.
138
Efficiency of the nanoreactors
8.2.3 HPLC analysis
Separation of nucleosides was carried out through High Pressure Liquid
Chromatography (HPLC) with gradient elution and UV detection, according to a
previously described method with modifications [6-8]. A reversed-phase C18
column (Alltima C18, 100 Å pore size, 5 µm particle size, 250 mm x 4.6 mm,
catalogue number 88056) protected by a guard column was used to separate
the sample components. Injected samples (50 µl) were eluted over 60 minutes
through a methanol-gradient buffered mobile phase. The gradient was
obtained by employing a combination of 3 eluents; eluent A: 20 mM potassium
phosphate pH 5.6, eluent B: methanol and eluent C: water. The elution
gradient was programmed as described in Table 8-1. The column was kept at
30°C during the entire experiment and the absorption of the eluate was
measured at 267 nm.
Table 8-1: Gradient conditions of three eluents applied to separate nucleosides
in plasma samples via HPLC. Eluent A = 20 mM potassium phosphate pH 5.6;
eluent B = methanol; eluent C = water.
Time (min)
Flow (ml/min)
% eluent A
% eluent B
% eluent C
0
1.5
100
0
0
5
1.5
100
0
0
25
1.5
82.6
17.4
0
26
1.5
0
0
100
30
1.5
0
0
100
31
1.5
0
100
0
35
2.0
0
100
0
45
2.0
0
100
0
46
1.5
0
100
0
47
1.5
0
0
100
50
1.5
0
0
100
51
1.5
100
0
0
60
1.5
100
0
0
139
Chapter 8
8.3 Results
To investigate the in vitro efficiency of the TP containing nanoreactors, plasma
of a MNGIE patient was incubated with the nanoreactors and the nucleoside
concentrations were subsequently determined by HPLC. PBS and free TP were
used as negative and positive control respectively. Quantification of thymidine
and deoxyuridine levels was based on external standards which were treated
in exact the same way as the plasma samples. The area under the peaks was
correlated to the amount of thymidine or deoxyuridine via a corresponding
standard curve. The identification of the peaks was based upon retention time
and always confirmed by treatment of a second aliquot of each sample with a
large excess of thymidine phosphorylase to eliminate the thymidine and
deoxyuridine peaks.
The results show a clear thymidine and deoxyuridine peak in untreated (only
PBS was added) MNGIE plasma samples (Figure 8-1). From the standard
curves (data not shown) the concentrations of thymidine and deoxyuridine
were calculated as 8.0 µM and 13.1 µM respectively. These results correspond
well to previous reported data (8.6 µM thymidine and 14.2 µM deoxyuridine as
mean values from 25 MNGIE patients) [2]. After treatment with free TP, the
thymidine peak disappears (< 0.05 µM) and the deoxyuridine peak decreases
to a concentration of 0.6 µM (Figure 8-2). As expected the free enzyme is able
to decrease the substrate concentrations efficiently in this in vitro setup.
Moreover, we observe an increase of the thymine peak, which confirms
substrate-product conversion.
However, after treatment with the TP-NRs, only a slight decrease in substrate
concentrations could be observed. Concentrations of 6.8 µM thymidine and
11.1 µM deoxyuridine were detected (Figure 8-3). Thus, the nanoreactors are
not able to decrease the toxic substrate amounts significantly. Nevertheless,
the nanoreactor formulation contains a significant amount of enzyme as
verified by SDS-PAGE analysis (data not shown) and the same amount of
enzyme was added to the plasma sample as was used in the free enzyme
condition.
140
Efficiency of the nanoreactors
In conclusion, the nanoreactor associated enzyme is not active enough to
convert the substrates efficiently. Therefore, the in vivo experiments in the TP/-
UP-/- mice were cancelled.
Figure 8-1: HPLC diagram of plasma of a MNGIE patient pre-treated with PBS. The
relevant peaks (thymidine, deoxyuridine and thymine) are identified by their typical
retention times.
141
Chapter 8
Figure 8-2: HPLC diagram of plasma of a MNGIE patient pre-treated with free TP. The
relevant peaks are identified by their typical retention times.
Figure 8-3: HPLC diagram of plasma of a MNGIE patient pre-treated with TP-NRs. The
relevant peaks (thymidine, deoxyuridine and thymine) are identified by their typical
retention times.
142
Efficiency of the nanoreactors
8.4 Conclusion and discussion
Unfortunately, the TP containing nanoreactors, as presented in their current
composition, are not efficient in reducing the toxic substrate concentrations in
MNGIE plasma in vitro. Therefore, the in vivo experiments in the TP-/-UP-/mice were cancelled at this point. Although the enzyme is clearly associated to
the nanoparticle, it is not able to efficiently convert the substrates into its
products. Probably, the enzyme is stuck into or onto the particle in a nonactive conformation. As such, these nanoreactors fail in their role of being
active enzyme carriers and cannot be used in enzyme replacement therapy. In
order to produce better enzyme containing nanoreactors, the construction and
composition of the nanoreactors need to be further scrutinized critically.
8.5 References
[1] M.C. Lara, M.L. Valentino, J. Torres-Torronteras, M. Hirano, R. Marti,
Mitochondrial neurogastrointestinal encephalomyopathy (MNGIE): biochemical
features and therapeutic approaches, Biosci Rep, 27 (2007) 151-163.
[2] R. Marti, A. Spinazzola, S. Tadesse, I. Nishino, Y. Nishigaki, M. Hirano,
Definitive diagnosis of mitochondrial neurogastrointestinal encephalomyopathy
by biochemical assays, Clin Chem, 50 (2004) 120-124.
[3] L.C. Lopez, H.O. Akman, A. Garcia-Cazorla, B. Dorado, R. Marti, I. Nishino,
S. Tadesse, G. Pizzorno, D. Shungu, E. Bonilla, K. Tanji, M. Hirano,
Unbalanced deoxynucleotide pools cause mitochondrial DNA instability in
thymidine phosphorylase deficient mice, Hum Mol Genet, (2008).
[4] N.F. Moran, M.D. Bain, M.M. Muqit, B.E. Bax, Carrier erythrocyte entrapped
thymidine phosphorylase therapy for MNGIE, Neurology, 71 (2008) 686-688.
[5] C. Nardin, Hirt, T., Leukel, J., Meier, W., Polymerized ABA Triblock
Copolymer Vesicles, Langmuir, 16 (2000) 1035-1041.
[6] R.A. Hartwick, S.P. Assenza, P.R. Brown, Identification and quantitation of
nucleosides, bases and other UV-absorbing compounds in serum, using
reversed-phase high-performance liquid chromatography. I. Chromatographic
methodology, J Chromatogr, 186 (1979) 647-658.
[7] R.A. Hartwick, A.M. Krstulovic, P.R. Brown, Identification and quantitation
of nucleosides, bases and other UV-absorbing compounds in serum, using
reversed-phase high-performance liquid chromatography. II. Evaluation of
human sera, J Chromatogr, 186 (1979) 659-676.
[8] R. Marti, Y. Nishigaki, M. Hirano, Elevated plasma deoxyuridine in patients
with thymidine phosphorylase deficiency, Biochem Biophys Res Commun, 303
(2003) 14-18.
143
Chapter 9
Detailed evaluation of the nanoreactor
build-up
Detailed evaluation of the nanoreactor build-up
9 Detailed evaluation of the nanoreactor
build-up
9.1 Introduction
Although the first characterization of the thymidine phosphorylase containing
nanoreactors looked promising, the latest data indicate that the activity and in
vivo behavior of the nanoreactors is not like expected/hoped (see chapters 6,
7 and 8). Indeed, we observed a specific antibody response against the
enzyme, hepatosplenic uptake and low enzymatic activity. These observations
suggest that the nanoreactor composition is not like it used to be and indicate
that the enzyme is not encapsulated properly in an active form inside the
particles. Such a configuration is however an absolute prerequisite in order to
shield the protein from proteolytic degradation and to allow the enzyme to
catalyze the substrates that are transported to the inner cavity of the
nanoparticles. The hampered enzyme activity and specific immune responses
are clearly unwanted properties that limit the use of these nanoreactors in
enzyme replacement therapy. Especially for this specific purpose, nanoreactors
with better immunogenic and catalytic properties are needed. This chapter
summarizes the efforts that were made to unravel the exact nanoreactor
composition and to try to resolve the problems encountered.
9.2 Materials and methods
9.2.1 Production of nanoreactors
The TPE.coli (thymidine phosphorylase from E.coli) containing nanoreactors
were constructed via the solvent evaporation method as described in detail in
section 3.2.3 [1]. Nanoreactors with varying porin concentrations (100 µg, 150
µg and 300 µg Tsx) were constructed and analyzed. As control sample,
nanoparticles with enzyme without porin were used. In addition, nanoreactors
147
Chapter 9
with other enzymes; the human thymidine phosphorylase (TPhuman), the
commercial Escherichia coli thymidine phosphorylase (TPE.coli,Sigma from Sigma)
and the Trypanosoma vivax nucleoside hydrolase (TvNH), were prepared in
exactly the same way and used as reference samples.
9.2.2 Characterization of TPE.coli containing nanoreactors
The TPE.coli containing nanoreactors were analyzed via DLS (to determine their
size and dispersity), via SDS-PAGE analysis (to determine the amount of
enzyme
and
porin
incorporated
in
the
nanoreactors)
and
via
spectrophotometry (to determine the enzymatic activity) as described in detail
in chapter 3.
9.2.3 Activity measurements of TvNH containing nanoreactors
The enzymatic activity (i.e. hydrolysis of nucleosides) of the TvNH containing
nanoreactors was determined with inosine as substrate making use of a
reducing sugar colorimetric assay as described by Parkin and Ranquin [2, 3].
Briefly, 150 µl nanoreactors were added to 1 mM inosine and the enzymatic
reaction was stopped at different time-points by adding a CuSO4 solution. The
Cu2+ is reduced to Cu+ by the reaction product ribose. This reduced Cu+ can
than react with neocuproin to form a complex. Color development (yellow) of
this complex was achieved by heating the solution 8 min at 95°C and the OD
was measured at 450 nm. A standard curve with known ribose concentrations
was used to determine the concentration of ribose formed in the nanoreactortreated samples under the assay conditions.
9.2.4 Protease treatment
In order to investigate their proteolytic resistance, the nanoreactors were
treated with proteinase K according to Axthelm et al. [4]. Proteinase K is a
protease with a broad specificity and therefore very efficient in digesting
proteins. 1 mg/ml proteinase K (stock solution of 20 mg/ml, Roche) was added
to the nanoreactors and the mixture was incubated for 4 h at 37°C. Before and
148
Detailed evaluation of the nanoreactor build-up
after the proteinase treatment, the protein content, size and activity of the
samples
were
analyzed
via
SDS-PAGE,
DLS
and
spectrophotometry
respectively.
7.2.5 Surface pressure experiments
For surface pressure experiments a MicroTrough S (Kibron Inc.) was used. The
microwell-trough was filled with the aqueous sub-phase (3 ml), either PBS or
PBS containing 1 M NaCl. A PMOXA-PDMS-PMOXA monolayer was formed by
adding 2 µl of polymer dissolved in ethanol (10 mg/ml) at the air-liquid
surface. After the formation of a stable monolayer with a surface pressure
around 20 mN/m, proteins were added to a final concentration of 0.08 µM with
a microsyringe at the bottom of the trough. Continuous recording of the
surface pressure was performed. Prior to experiments the trough was
thoroughly cleaned and the surface pressure detection wire probe was cleaned
by flaming. The experiments were performed at room temperature. For
protection of impurities the trough was housed in a Plexiglas cabinet.
9.2.6 Fluorescent labeling of thymidine phosphorylase
To fluorescently label the enzyme TPE.coli, we used the amine reactive
Alexa®555 dye (Invitrogen). The spectra of Alexa®555 (excitation/emission
maxima ~555/565 nm) are an almost exact match to those of the Cy3 dye.
The amine reactive succinimidyl ester of Alexa®555 forms a stable amide bond
with amine groups on the protein. Before coupling the Alexa®555 dye, the
enzyme was dialyzed to 0.1 M sodium bicarbonate buffer pH 8.3 and
concentrated to a final concentration of 20 mg/ml. Alexa®555 was dissolved in
DMSO to a final concentration of 10 mg/ml. Then, 1 ml of the TP sample was
mixed with 100 µl Alexa®555 and incubated for 1 h at room temperature
under continuous stirring. The reaction was stopped by adding 1.5 M
hydroxylamine pH 8.5. The non-bonded Alexa®555 was subsequently removed
via gelfiltration on a superdex 75 high resolution column (Amersham
Biosciences) with PBS as running buffer.
149
Chapter 9
9.3 Results
9.3.1 Enzyme activity versus porin/enzyme concentration
In order to identify the cause of the low enzymatic activity of the
nanoreactors, kinetic experiments were performed using nanoreactors with
two different porin concentrations (150 µg and 300 µg Tsx). Nanoparticles
encapsulating TPE.coli without porin Tsx were used as controls. The conversion
of 900 µM thymidine was followed spectrophotometrically at 290 nm for 10
minutes. Theoretically, when more porin is added, the overall transport of the
substrate across the polymeric membrane occurs faster and thus, the
enzymatic activity of the nanoreactors should be higher. This was previously
described by Ranquin et al. for TvNH containing nanoreactors [3].
For the TPE.coli containing nanoparticles however, an opposite trend is
observed. Namely, the enzymatic activity of the nanoreactors increases with
decreasing porin concentrations. Even more surprisingly, control samples
without reconstituted channel proteins are also enzymatically active and their
enzymatic activity is higher than the activity of the particles containing the
membrane protein Tsx (Figure 9-1). Clearly, these results show that transport
of thymidine across the membrane is not necessary for enzymatic activity. The
presence of Tsx even seems to have a negative effect on the enzymatic
activity. Furthermore, SDS-PAGE analysis shows that the encapsulation
efficiency of TPE.coli decreases with increasing Tsx concentrations. The protein
band representing the enzyme becomes slightly smaller when more Tsx is
added (Figure 9-1).
150
Detailed evaluation of the nanoreactor build-up
Figure 9-1: Left: Thymidine conversion catalyzed by TPE.coli containing nanoreactors
with different porin concentrations measured as the decrease in optical density (OD) at
290 nm in function of time. Right: SDS-PAGE analysis of the nanoreactors showing the
amount of TPE.coli (52 kDa) and Tsx (31 kDa) present in the samples. The protein
molecular weight marker used is Mark 12 (Invitrogen).
Subsequently, the initial rate of conversion of different concentrations of
thymidine by nanoreactors containing enzyme TPE.coli and porin Tsx (100 µg)
was measured spectrophotometrically at 290 nm (in 200 mM potassium
phosphate pH 6.8 at 37°C). The results were fitted to the Michaelis-Menten
equation (Origin software) to determine the catalytic parameters of the TPcontaining nanoreactors (Figure 9-2). As such, an apparent kcat,app of 0.073 ±
0.003 s-1 and a KM,app of 760 ± 69 µM were obtained. In comparison to the
catalytic parameters of the free enzyme (kcat = 413.6 ± 71.1 s-1 and KM =
643.2 ± 268.7 µM ; see chapter 3), the enzymatic activity of the nanoreactors
is extremely low. These results are also not in agreement with the results
obtained by Ranquin et al. [3], where the KM,app increased tenfold, indicating
that transport over the membrane was the rate limiting step. In case of the TP
encapsulating nanoreactors, the KM,app is in the same range as the KM of the
free enzyme. This again suggests that transport of thymidine is not necessary
for enzymatic conversion. However, roughly estimated, the catalytic turnover
151
Chapter 9
of the enzymes is reduced by a factor 104. This finding probably reflects that
most enzyme molecules are present in a catalytically inactive conformation.
Figure 9-2: Initial product formation rate of nanoreactors (composed of enzyme TPE.coli
and 100 µg Tsx) as a function of thymidine concentration. The data were fitted to a
hyperbolic curve using Origin software to determine the apparent catalytic parameters
kcat,app and KM,app.
From these experiments we conclude that since the porins are not required for
activity, the enzyme is not situated inside the particles. The enzyme is
probably located or associated into or onto the polymeric membrane. Such
association or partial insertion of enzymes in the membrane also explains the
observed low activity.
9.3.2 Protease sensitivity of nanoreactors
If the TP enzymes are associated with the polymeric membrane and located at
least partially at the outside of the particle, they are prone to degradation by
proteases.
152
Detailed evaluation of the nanoreactor build-up
To verify whether the enzyme is solvent exposed and can be attacked by
proteases, proteinase K was added to the nanoreactors (NP+TP+Tsx) and the
mixture was incubated for 4 hours at 37°C. Nanoparticles with enzyme without
porine (NP+TP-Tsx) and free enzyme (TP) were used as control samples. The
SDS-PAGE analysis of the samples before and after proteinase K treatment
demonstrates that the nanoreactor associated enzyme (either with or without
porin) is not protected against proteolytic degradation (Figure 9-3). No protein
band could be detected in the proteinase K treated samples. The nanoparticle
associated enzyme is degraded to the same extend as the free enzyme. Also,
no enzymatic activity could be measured after proteinase K treatment in none
of the samples (data not shown). Analysis of the size distribution of the
nanoparticles by DLS reveals that the particles are not completely degraded
after proteinase K treatment (Figure 9-4). However, a small population at
lower sizes could be observed. This indicates that the nanoreactors undergo
slight compositional changes and are not completely unaffected by the
protease treatment.
Figure 9-3: SDS-PAGE analysis of TP containing nanoreactors and free TP before and
after proteinase K (prot K) treatment. Lane 1: NP+TP+Tsx; lane 2: NP+TP+Tsx without
prot K after 4 h incubation at 37°C; lane 3: NP+TP+Tsx with prot K after 4 h incubation
at 37°C. Lane 4: NP+TP-Tsx; lane 5: NP+TP-Tsx without prot K after 4 h incubation at
37°C; lane 6: NP+TP-Tsx with prot K after 4 h incubation at 37°C. Lane 7: free TP
without prot K after 4 h incubation at 37°C; lane 8: free TP with prot K after 4 h
incubation at 37°C. The protein molecular weight marker used is SM0431 (Fermentas).
153
Chapter 9
Figure 9-4: DLS distribution patterns of nanoreactors (A) before and (B) after
proteinase K treatment. The intensity was plotted for ten consecutive measurements as
a function of the radius.
9.3.3 Interaction between the enzyme and the polymeric particle
The motivation of this study was to test whether the enzyme TPE.coli has affinity
for the polymer and tends to stick to the polymeric membrane. This could
explain why straightforward encapsulation of the enzyme inside the aqueous
space of the nanoparticles is prevented. The enzyme TPhuman is used as
comparative sample in this study. The interaction of monolayers of PMOXAPDMS-PMOXA (at air-water interface) with the enzymes TPE.coli and TPhuman was
investigated by surface pressure studies. Changes of the surface pressure
upon addition of the protein, indicate interactions between protein and
polymer. The porin Tsx acts as a positive control for these experiments,
because it has already been proven that this porin interacts with such a
polymeric membrane [3]. If diluted to a concentration below the critical
micellar concentration of the detergent used for porin extraction and storage,
the porin is forced to incorporate inside the hydrophobic block of the
membrane in order to shield its hydrophobic surfaces and avoid precipitation.
154
Detailed evaluation of the nanoreactor build-up
At physiological salt concentrations (PBS), our results show beside a clear
interaction with Tsx, also an interaction with TPE.coli. No changes in surface
pressure were observed after adding TPhuman (Figure 9-5A). However, the
observed changes are rather surprising, since interactions are normally
associated with an increase in surface pressure rather than a decrease. The
behavior of PMOXA-PDMS-PMOXA
triblock copolymers at the air-water
interface has previously been reported by Haefele and coworkers [5]. Different
phases of monolayer arrangement were identified ranging from an expanded
conformation at large molecular areas to mushroom and brush conformations
by stretching the chains due to space limitations. Probably, upon interaction,
the polymers reorganize from a mushroom conformation into a more
expanded conformation resulting in a decrease in surface pressure. The same
experiments were therefore performed in high salt conditions (1 M NaCl),
where the polymers are probably directly organized in the expanded
conformation and where the expected increase in surface pressure was
observed each time the porin Tsx was added (Figure 9-5B). In this condition,
the interaction of Tsx is characterized by an initial strong insertion into the
polymer layer followed by a slow decrease of the surface pressure. This might
hint to a possible rearrangement of the polymer phase.
In response to the
enzymes TPE.coli and TPhuman also an increase in surface pressure was detected,
although the kinetics were different as compared to the Tsx interaction (Figure
9-5B). This suggests no real incorporation is occurring, but rather an
association of the enzymes with the PMOXA blocks in the solution.
The surface pressure experiments are however difficult to interpret and it is
questionable
whether
these
results
are
really
indicative
for
relevant
interactions between the triblock copolymer and the enzymes. Moreover, the
monolayer approach deviates from fully hydrated membrane systems like
nanoreactors, where the polymers are probably organized in different
conformations and behave in a different way than at air-water surfaces.
155
Chapter 9
Figure 9-5: Surface pressure measurements of monolayers composed of PMOXA-PDMSPMOXA triblock copolymers in PBS (A) and in high salt solution (B). Arrows indicate the
addition of 0.08 µM protein to the solution. The surface pressure changes are recorded
over time.
The association of the enzyme TP with the polymeric particle was further
investigated using fluorescently labeled enzyme. By making use of a TPAlexa®555, we want to distinguish between association occurring by binding of
the enzyme to the outside of the particles (which should be able to take place
after nanoparticle synthesis) and association by (partial) incorporation of the
enzyme into the polymeric wall during the production process. Premade empty
nanovesicles were mixed with free TP-Alexa®555 in order to verify whether the
enzyme interacts with the surface of the nanoparticles. After the elimination of
unbound/unassociated enzyme by Ni affinity chromatography, the sample did
not show any fluorescence signal. In contrast, when using the fluorescently
labeled enzyme to produce nanoreactors, an estimated amount of 10% of the
fluorescent signal was retained in the nanoreactor sample after purification on
a Ni affinity column. This is in close agreement with the previously determined
‘encapsulation’ efficiencies using SDS-PAGE analysis (see section 3.3.4).
Nevertheless, from the above we now know that no real encapsulation occurs,
but rather an association of the enzyme to the particles. In conclusion, since
156
Detailed evaluation of the nanoreactor build-up
no strong non-specific interaction and binding of the enzyme to the surface of
the particles is detected, the association of the enzyme takes place during the
production process of the nanoreactors. The exact localization of the enzymes
in the particle and the precise nanoparticle/enzyme conformation is however
not clear at the moment. Unfortunately, the resolution of the fluorescence
microscope is not high enough to analyze the exact localization of the enzyme
onto the 200 nm-sized particles in more detail.
9.3.4 Nanoreactors with other enzymes
Next, we wanted to investigate whether the problems we encounter with
TPE.coli incorporation into the polymer wall of the nanoreactors are enzyme
specific. Indeed, previous studies reported the successful encapsulation of
other enzymes inside similar PMOXA-PDMS-PMOXA nanoreactors. For example
-lactamase and nucleoside hydrolase from Trypanosoma vivax (TvNH) were
used and their activity was preserved after incorporation in the nanoreactors
[3, 6]. To this end, we used our polymers and production procedure to
incorporate different enzymes. As test cases TvNH and other sources of TP;
human TP (TPhuman) and TPE.coli from a commercial source (Sigma, hereafter
called TPE.coli,Sigma) were used. Strikingly, we were not able to make active
nanoreactors with none of these enzymes. Also for these enzyme-reactor
systems, the enzymes lose the majority of their activity during the production
process.
For
example,
figure
9-6
shows
the
results
of
the
TvNH
containing
nanoreactors. In its active conformation, this enzyme is able to cleave
nucleosides into the nucleobase and the ribose. The activity of the enzyme can
be followed by a reducing sugar assay that measures the ribose production.
The results demonstrate that also here the nanoreactors without porins are the
most active ones (Figure 9-6 left). Moreover, the amount of enzyme
associated to the particle decreases when more porin was added (Figure 9-6
right). In conclusion, the problems encountered are clearly not enzyme
specific. Since the same protocol and conditions were used as described in the
paper by Ranquin et al. [3], the TvNH containing nanoreactors constructed
157
Chapter 9
here exhibit a totally different behavior. This seems to indicate that the
problems are rather polymer dependent.
Figure 9-6: Left: Inosine conversion catalyzed by TvNH containing nanoreactors with
different porin concentrations measured as the increase in product (ribose) formation in
function of time. Right: SDS-PAGE analysis of the nanoreactors showing the amount of
TvNH (37 kDa) and Tsx (31 kDa) present in the samples. The protein molecular weight
marker used is SM0431 (Fermentas).
9.3.5 Nanoreactors with other polymers
The previous experiments suggest that failure to encapsulate enzymes in the
interior of the polymeric nanoparticles is not enzyme dependent but rather
polymer dependent. From the beginning of this study we worked with the
same batch of commercially available PMOXA20-PDMS54-PMOXA20 triblock
copolymers (from Polymer Source Inc.), in order to reduce batch-to-batch
variations. Since it is very difficult to precisely control the synthesis and length
of both hydrophilic and hydrophobic blocks of block copolymers, there is a high
risk for polymer batch variability. However, it is possible that, due to longterm storage, the polymer degraded during the proceedings of this study.
Therefore, we ordered and examined fresh polymer batches from different
sources in order to investigate whether different, preferentially better,
nanoreactor properties can be obtained. To this end, 3 batches of commercial
158
Detailed evaluation of the nanoreactor build-up
PMOXA20-PDMS54-PMOXA20 polymers from Polymer Source Inc. and 1 batch of
PMOXA20-PDMS42-PMOXA20 polymers, kindly provided by Prof. Meier from the
Department of Chemistry from Basel University, were used. Although the
commercial and non-commercial polymers show a slight difference in
hydrophobic block length, the hydrophobic-to-hydrophilic ratio is still very
similar and we expect both triblock copolymers to form nanoreactors.
After visual inspection of the various polymer batches, already clear
differences in the morphology, texture and color of the polymer powder could
be observed. The construction of TPE.coli containing nanoparticles with the
different PMOXA-PDMS-PMOXA polymer batches, resulted in nanoparticles of
slightly different sizes and turbidity (data not shown). Unfortunately, we were
not able to obtain more active nanoreactors with any of the polymer batches
available.
9.4 Conclusion and discussion
Since the TP containing nanoreactors showed very low enzymatic activity and
specific immune responses against the enzyme, we performed various
experiments in order to pinpoint the exact problem. First, we were able to
show, much to our surprise, that incorporation of the nucleoside specific porin
Tsx in the nanoreactor wall is not necessary for enzymatic conversion of
thymidine by the TP enzyme. Curiously, the highest enzymatic activity was
found for nanoreactors that have no Tsx incorporated. So, instead of
improving enzymatic activity by incorporating a nucleoside specific transporter
in the polymeric membrane, the enzymatic activity is decreased as a function
of Tsx incorporation. This is very much in contrast to other findings [3, 7]
where enzymatic activity is only obtained after permeabilization of the reactor
wall and enzymatic activity increases proportional to porin concentration.
Further, we demonstrated that although our nanoreactors contain a relatively
high amount of enzyme, they show very poor activity. The apparent catalytic
turnover of the enzyme containing nanoreactors was drastically decreased in
comparison to the free enzyme. Moreover, they are sensitive to proteolytic
159
Chapter 9
degradation. From these experiments it is clear that the enzyme is not nicely
encapsulated inside the particles like it is supposed to be, but that the enzyme
is rather associated to the outside of the particle in a non-active conformation.
Since no strong attraction and binding of the enzyme to the particle surface
and polymeric membrane was observed, the association of the enzyme to the
particles is occurring during the nanoreactor production process. Thereby, the
enzyme is incorporated into and/or onto the polymeric wall, losing its
functional integrity. These data also support the adjuvant effect of the
nanoreactors as described in chapter 6.
Most likely, the problems are rather polymer-dependent than enzymedependent, because attempts to encapsulate other enzymes encountered the
same inconsistencies. Even when exactly the same conditions were used to
construct previously reported TvNH containing nanoreactors, encapsulation
remained unsuccessful. In addition, analyzing the structure of the enzyme TP
gave us an idea of the amount of hydrophobic residues on the surface of the
enzyme.
Such hydrophobic patches can cause aggregation (cf. membrane
proteins) and prevent encapsulation of the enzyme inside the aqueous cavity
of the nanoreactors. Nevertheless, we could not detect significant hydrophobic
patches on the surface of the TPE.coli. As such, we suggest that the TP enzyme
is at least theoretically perfectly suited for encapsulation. Unfortunately, thus
far, we were not able to resolve the problems using other polymer batches of
different sources. Further in depth investigation is needed to identify the
problems encountered and to make better enzyme containing nanoreactors
with high reproducibility. It is possible that the polymers underwent transitions
during the enrolment of this study due to long-term storage. Air humidity can
be the cause of polymer degradation. As a result, the polymers form rather
nanospheres that have the same size as vesicles, but are not hollow. In such
case, encapsulation of the enzyme is not possible and it is realistic that the
enzyme is incorporated in the polymeric matrix forming protein/polymer
aggregates instead of vesicles. The presence of impurities in the polymer
batch can also cause problems in producing good reproducible nanoreactors.
160
Detailed evaluation of the nanoreactor build-up
All together, we suggest that polymer degradation, poor polymer purity and
batch-to-batch variability of the polymer stocks are the main causes of our
problems. Nanoreactors based on triblock copolymers are a completely new
and promising field. Therefore, if pharmacological applications on an industrial
scale are envisioned it is of uttermost importance that the synthesis of the
polymers can be precisely controlled in order to obtain identical nanoreactors
of different polymer batches. Further investigation is necessary to better
control the polymer chemistry in order to construct nanoreactors exhibiting
exactly the same properties and behaving in precisely the same way every
time a new batch of reactors is made from a new batch of polymers.
9.5 References
[1] C. Nardin, Hirt, T., Leukel, J., Meier, W., Polymerized ABA Triblock
Copolymer Vesicles, Langmuir, 16 (2000) 1035-1041.
[2] D.W. Parkin, Purine-specific nucleoside N-ribohydrolase from Trypanosoma
brucei brucei. Purification, specificity, and kinetic mechanism, J Biol Chem, 271
(1996) 21713-21719.
[3] A. Ranquin, W. Versees, W. Meier, J. Steyaert, P. Van Gelder, Therapeutic
Nanoreactors: Combining Chemistry and Biology in a Novel Triblock Copolymer
Drug Delivery System, Nano Lett, 5 (2005) 2220-2224.
[4] F. Axthelm, O. Casse, W.H. Koppenol, T. Nauser, W. Meier, C.G. Palivan,
Antioxidant nanoreactor based on superoxide dismutase encapsulated in
superoxide-permeable vesicles, J Phys Chem B, 112 (2008) 8211-8217.
[5] T. Haefele, K. Kita-Tokarczyk, W. Meier, Phase behavior of mixed Langmuir
monolayers from amphiphilic block copolymers and an antimicrobial peptide,
Langmuir, 22 (2006) 1164-1172.
[6] C. Nardin, S. Thoeni, J. Widmer, M. Winterhalter, W. Meier, Nanoreactors
based on (polymerized) ABA-triblock copolymer vesicles, Chem Comm, (2000)
1433-1434.
[7] M. Grzelakowski, O. Onaca, P. Rigler, M. Kumar, W. Meier, Immobilized
protein-polymer nanoreactors, Small, 5 (2009) 2545-2548.
161
Chapter 10
General conclusion
General conclusion
10 General conclusion
In this study we propose a new therapeutic tool for enzyme delivery in enzyme
replacement therapy, based on nanoreactors composed of PMOXA-PDMSPMOXA triblock copolymers. As a proof of principle we chose to develop
nanoreactors for the potential use in enzyme replacement therapy of MNGIE.
This is a severe disorder caused by the deficiency of the crucial enzyme
thymidine
phosphorylase
(TP).
At
the
moment,
allogeneic
stem
cell
transplantation is the only available therapy for MNGIE. Unfortunately, the
treatment is associated with significant morbidity and mortality. Furthermore,
for 30-to-40% of the patients, no matching related or unrelated donors are
available. Therefore, a safer and more universal treatment for TP-deficiency
would be beneficial to MNGIE patients and enzyme replacement therapy could
provide a promising alternative strategy.
The aim of this study was to develop and evaluate self-assembling PMOXAPDMS-PMOXA nanoreactors that are filled with active thymidine phosphorylase
and are permeabilized for small molecules by incorporating a bacterial porin
Tsx in the reactor wall. The nucleoside specific Tsx porin should allow the
transport of the substrates and products across the particle membrane,
resulting in nanometer-sized bioreactors. Such nanoreactors are significantly
different from the classical drug delivery particles, because the enzyme doesn’t
need to be released from the particle to fulfill its biological function. They
represent a new and innovative concept in the area of drug delivery research.
Especially for the delivery of enzymes, there is an urgent need for better
delivery systems, because their catalytic activity is easily impaired by the
smallest
conformational
change
or
environmental
attack.
Although
nanoreactors have been suggested already for the therapeutic delivery of
enzymes
[1-5],
detailed
studies
concerning
their
stability,
toxicity,
immunogenicity, biodistribution and in vivo efficacy are lacking. These
properties are however of crucial importance for the further exploitation and
evaluation of such nanoreactors as therapeutic enzyme delivery devices.
165
Chapter 10
For this study, we prepared stable TP-containing nanoreactors by the solvent
evaporation method, based on previously described successful nanoreactor
production protocols as reported by the group of Wolfgang Meier [6-9]. The
production process resulted in monodisperse particles with a diameter of
approximately 200 nm, as determined by DLS. Suspension in blood serum had
no immediate effect on vesicle stability and no leakage of the enzyme from the
particles was detected. On the contrary, enzyme leakage was observed when
TP was encapsulated in liposomes, showing that our polymeric nanoparticles
exhibit higher mechanical stability. In addition, no stimulation of phagocytes
was detected when nanoreactors were incubated with naïve macrophages in
vitro. On IFN- primed macrophages, a minor porin dependent activation was
observed upon incubation with the nanoreactors. If needed, this response
could possibly be circumvented by substituting the bacterial-derived Tsx
nucleoside pore with a human homologue, such as the eENT or VDAC porin for
example [10]. Subsequent in vivo experiments demonstrated however that the
bacterial porins are sufficiently shielded by the polymer, since no acute
inflammatory effects on macrophages could be detected after intraperitoneal
injection of the nanoreactors. Moreover, proliferating liver cells are unaffected
when
cultured
for
an
extended
time
with
relevant
concentrations
of
nanoreactors. These effects are comparable to those obtained with PEGmodified particles (i.e. sterically stabilized liposomes) [11]. Taken together,
the
nanoreactors
exhibit
high
stability
and
low
acute
toxicity
and
immunogenicity, making them promising towards in vivo applications at a first
view.
Unfortunately, further investigation revealed that the nanoreactors induce a
strong enzyme-specific antibody response after repeated iv injections in mice.
This implies that the enzyme is not shielded from the environment as stated
before. Moreover, hepatosplenic uptake of the nanoreactors prevents longterm blood circulation in living mice and rats. In addition, the enzymatic
activity of the TP-containing nanoreactors seemed insufficient to significantly
reduce the thymidine levels in the plasma of MNGIE patients. Taken together,
these characteristics are major drawbacks and limit the use of this type of
166
General conclusion
nanoreactors in enzyme replacement therapy. The impaired enzymatic activity
of the nanoreactors was unexpected and not in agreement with previous
reports on PMOXA-PDMS-PMOXA nanoreactors [2, 4, 5], suggesting an
incorrect nanoreactor composition. Moreover, the specific anti-TP immune
response indicates that the enzyme is exposed to the environment. This
prompted us to re-evaluate the nanoreactor build-up. The subsequent
experiments
unequivocally
indicated
that
the
enzyme
indeed
is
not
encapsulated properly inside the particles, but tends to be associated to the
surface of the particles in a predominantly inactive conformation. The exact
cause of the disturbed encapsulation is not elucidated completely yet and
needs further investigation. At our opinion, the polymers are probably not able
to form hollow particles and instead, homogenic dense polymer-matrices are
formed that incorporate both the enzyme and the porin. This incorporation
renders the enzyme inactive, probably due to conformational changes.
All together, we can conclude that although the polymer-based nanoreactors
present a promising new tool in biomedical applications, the TP-containing
nanoreactors as presented in this study do not possess ideal properties for use
in enzyme replacement therapy. Nevertheless, this investigation revealed
important information on the stability, toxicity and immunogenicity of such
nanoreactors and was of high relevance for the further exploitation of PMOXAPDMS-PMOXA triblock copolymer based particles as drug delivery devices.
Although we were not successful in generating highly active nanoreactors, we
could demonstrate overall low immunogenicity and toxicity of the PMOXAPDMS-PMOXA particles. Moreover, we were the first to provide in vivo data on
PMOXA-PDMS-PMOXA nanoparticles, elucidating the potential of this type of
vesicles for other applications. Since our nanoreactors seem to be very
effective in inducing a specific immune response against the associated
enzyme and are particularly stable under physiological conditions, the
nanoparticles could for example be tested for their use in vaccination
strategies.
Future studies need to focus more on the reproducibility of nanoreactor
preparation and the improvement of polymer quality. Beside the problems we
167
Chapter 10
encountered here with the use of PMOXA-PDMS-PMOXA triblock copolymers,
other
studies
also
report
clear
batch-to-batch
differences
(personal
communication with Prof. Meier and Dr. Ranquin). Given that the phase
behavior of triblock copolymers in aqueous solutions depends on the
composition, length, structure and molecular weight distribution of the
individual blocks as well as on the molecular architecture of the whole
polymer, we believe it is very important to better control these properties to
assure the reproducibility of the experiments performed with them. However,
due to the nature of the synthesis of the polymer, it is very difficult to
precisely control the length and molecular weight distribution. Fine tuning the
production method of the polymers and the nanoreactors will be a key step in
the exploitation of triblock based nanoreactors as drug delivery devices. In
addition, other polymers than PMOXA-PDMS-PMOXA may prove to be more
effective for the production of enzyme-containing nanoreactors. Such polymers
should ideally be biodegradable, which is not the case for the PDMS middle
block of our triblock copolymer.
Overall, the development of enzyme-containing nanoparticles for use in
enzyme replacement therapy is a challenge, because life-time administrations
of appropriate enzyme formulations are required to restore the enzyme
activity. Therefore, the ideal enzyme replacement delivery device needs to
possess long blood circulation times, high activity and low immunogenicity
properties. Recent studies underline the difficulty of constructing nanocarriers
that provide long circulation together with low immunogenicity [12-14]. The
long circulation property can be obtained by grafting stealthy polymers like
PEG to the outer core of the particle in order to sterically stabilize the particle
and prevent aggregation [15]. However, repeated administrations of such
sterically stabilized particles result in the formation of PEG specific (or drug
specific) antibodies [16, 17]. Immunogenic aspects are less important if only
a single bolus injection is required like it is the case for particular cancer
therapy purposes. This is probably the reason why nanoparticle-mediated drug
delivery has booked its first successes and is entering clinical trials for such
applications [18-20].
168
General conclusion
Nevertheless, the development of controlled drug delivery systems remains a
challenging field of research. In addition to the highly innovative and
engineered drug delivery strategies making use of complex combinations of
drugs, carriers and ligands, other drug delivery systems focus more on the
way nature provides solutions for efficient delivery of compounds to the body
(like drugs loaded on erythrocytes [21] and drugs coupled to human serum
albumin
[22]).
However,
each
strategy
has
its
own
advantages
and
disadvantages and every application needs its own optimization. At this stage
there are no simple predictors that will guarantee the best delivery strategy
for a given protein. The design and optimization of carriers represent an
important step in the development of new protein delivery approaches. Both
the
physical
(such
as
compartmentalization)
and
size,
the
shape,
chemical
mechanical
(such
as
properties
pegylation,
and
polymer
composition and ligand coupling) properties of the delivery systems need to be
further optimized to better control their behavior in a biological context.
Undoubtedly,
further
scientific
research
will
lead
to
improved
protein
therapeutics that will provide more efficacious drugs and favorable clinical
outcomes.
References
[1] C. Nardin, S. Thoeni, J. Widmer, M. Winterhalter, W. Meier, Nanoreactors
based on (polymerized) ABA-triblock copolymer vesicles, Chem Comm, (2000)
1433-1434.
[2] A. Ranquin, W. Versees, W. Meier, J. Steyaert, P. Van Gelder, Therapeutic
Nanoreactors: Combining Chemistry and Biology in a Novel Triblock Copolymer
Drug Delivery System, Nano Lett, 5 (2005) 2220-2224.
[3] P. Broz, S. Driamov, J. Ziegler, N. Ben-Haim, S. Marsch, W. Meier, P.
Hunziker, Toward Intelligent Nanosize Bioreactors: A pH-Switchable, ChannelEquipped, Functional Polymer Nanocontainer, Nano Lett, 6 (2006) 2349-2353.
[4] F. Axthelm, O. Casse, W.H. Koppenol, T. Nauser, W. Meier, C.G. Palivan,
Antioxidant nanoreactor based on superoxide dismutase encapsulated in
superoxide-permeable vesicles, J Phys Chem B, 112 (2008) 8211-8217.
[5] M. Grzelakowski, O. Onaca, P. Rigler, M. Kumar, W. Meier, Immobilized
protein-polymer nanoreactors, Small, 5 (2009) 2545-2548.
[6] C. Nardin, J. Widmer, M. Winterhalter, W. Meier, Amphiphilic block
copolymer nanocontainers as bioreactors, Eur. Phys. J. E, 4 (2001) 403-410.
[7] A. Graff, M. Sauer, P. Van Gelder, W. Meier, Virus-assisted loading of
polymer nanocontainer, Proc Natl Acad Sci U S A, 99 (2002) 5064-5068. Epub
2002 Mar 5026.
169
Chapter 10
[8] P. Broz, S.M. Benito, C. Saw, P. Burger, H. Heider, M. Pfisterer, S. Marsch,
W. Meier, P. Hunziker, Cell targeting by a generic receptor-targeted polymer
nanocontainer platform, J Control Release, 102 (2005) 475-488.
[9] M. Nallani, S. Benito, O. Onaca, A. Graff, M. Lindemann, M. Winterhalter,
W. Meier, U. Schwaneberg, A nanocompartment system (Synthosome)
designed for biotechnological applications, J Biotechnol, 123 (2006) 50-59.
Epub 2005 Dec 2020.
[10] A. Deniaud, C. Rossi, A. Berquand, J. Homand, S. Campagna, W. Knoll, C.
Brenner, J. Chopineau, Voltage-dependent anion channel transports calcium
ions through biomimetic membranes, Langmuir, 23 (2007) 3898-3905.
[11] X. Shan, C. Liu, Y. Yuan, F. Xu, X. Tao, Y. Sheng, H. Zhou, In vitro
macrophage uptake and in vivo biodistribution of long-circulation nanoparticles
with poly(ethylene-glycol)-modified PLA (BAB type) triblock copolymer,
Colloids Surf B Biointerfaces, 72 (2009) 303-311.
[12] H. Koide, T. Asai, K. Hatanaka, K. Shimizu, M. Yokoyama, T. Ishida, H.
Kiwada, N. Oku, [Elucidation of accelerated blood clearance phenomenon
caused by repeat injection of PEGylated nanocarriers], Yakugaku Zasshi, 129
(2009) 1445-1451.
[13] S.M. Moghimi, J. Szebeni, Stealth liposomes and long circulating
nanoparticles: critical issues in pharmacokinetics, opsonization and proteinbinding properties, Prog Lipid Res, 42 (2003) 463-478.
[14] S.M. Moghimi, A.C. Hunter, Capture of stealth nanoparticles by the body's
defences, Crit Rev Ther Drug Carrier Syst, 18 (2001) 527-550.
[15] A.L. Klibanov, K. Maruyama, V.P. Torchilin, L. Huang, Amphipathic
polyethyleneglycols effectively prolong the circulation time of liposomes, FEBS
Lett., 268 (1990) 235-237.
[16] T. Ishida, X. Wang, T. Shimizu, K. Nawata, H. Kiwada, PEGylated
liposomes elicit an anti-PEG IgM response in a T cell-independent manner, J
Control Release, 122 (2007) 349-355.
[17] H. Koide, T. Asai, K. Hatanaka, S. Akai, T. Ishii, E. Kenjo, T. Ishida, H.
Kiwada, H. Tsukada, N. Oku, T cell-independent B cell response is responsible
for ABC phenomenon induced by repeated injection of PEGylated liposomes,
Int J Pharm.
[18] A. Surendiran, S. Sandhiya, S.C. Pradhan, C. Adithan, Novel applications
of nanotechnology in medicine, Indian J Med Res, 130 (2009) 689-701.
[19] S.D. Caruthers, S.A. Wickline, G.M. Lanza, Nanotechnological applications
in medicine, Curr Opin Biotechnol, 18 (2007) 26-30.
[20] V. Wagner, A. Dullaart, A.K. Bock, A. Zweck, The emerging nanomedicine
landscape, Nat Biotechnol., 24 (2006) 1211-1217.
[21] B.E. Bax, M.D. Bain, L.D. Fairbanks, A.D. Webster, P.W. Ind, M.S.
Hershfield, R.A. Chalmers, A 9-yr evaluation of carrier erythrocyte
encapsulated adenosine deaminase (ADA) therapy in a patient with adult-type
ADA deficiency, Eur J Haematol, 79 (2007) 338-348.
[22] W.J. Gradishar, Albumin-bound paclitaxel: a next-generation taxane,
Expert Opin Pharmacother, 7 (2006) 1041-1053.
170
Chapter 11
Summary
Summary
11 Summary
Many known genetic deficiencies in human are caused by a mutation in a gene
encoding a crucial enzyme, leading to the accumulation of its substrates. As a
result, toxic amounts of the substrates appear in the bloodstream, disturbing
the
normal
metabolic
neurogastrointestinal
degradation
pathways.
encephalomyopathy
is
MNGIE
such
a
or
mitochondrial
unique
autosomal-
recessive disorder associated with mitochondrial DNA alterations [1]. In 1998,
Nishino et al. mapped the disease locus and identified a deficiency in the
enzyme thymidine phosphorylase (TP) as the cause of the disorder [2]. Lossof-function mutations in the TP gene lead to systemic accumulations of the
substrates thymidine and deoxyuridine [3]. Current therapies for MNGIE are
still in the developmental stage and include platelet infusion [4] and allogeneic
stem cell transplantation [5]. Unfortunately, these treatments are far from
ideal and suffer from the lack of sustained TP activity and serious safety risks
[6]. A more reliable and generally well-tolerated strategy to treat enzyme
deficiencies is enzyme replacement therapy [7, 8].
In enzyme replacement therapy the deficient enzyme is administered
intravenously in its native or stabilized form [9]. Direct administration of
native enzymes is usually limited due to rapid protein elimination from the
circulation
[10].
To
improve
their
stability,
proteolytic
resistance,
immunogenicity and circulation half-life, therapeutic proteins can be modified
by covalently linking various poly(ethylene glycol) (PEG) molecules to their
surface, a technique called pegylation [11]. However, the pegylation process is
complex, expensive, and often compromises enzyme activity [12, 13]. A more
convenient strategy to deliver therapeutic enzymes to the body is to
encapsulate them in appropriate carrier systems [14]. In a first attempt,
proteins were trapped in liposomes [15]. Unfortunately, these lipidic carrier
systems suffer from rapid leakage of the protein, poor stability and fast
clearance by the mononuclear phagocyte system [16, 17]. By grafting PEG on
the liposome surface, a new generation of long-circulating or stealth liposomes
173
Chapter 11
was introduced [18]. However, fast drug leakage and the accelerated blood
clearance [19] remain major drawbacks of these lipidic delivery systems. In
order to obtain more robust membranes with controllable properties, extensive
efforts were made within the last decade to design polymeric vesicles [20-23].
The aim of this study was to develop and evaluate polymeric nanometer-sized
reactors for use in enzyme replacement therapy, with MNGIE as proof of
concept. Therefore, the therapeutic enzyme, thymidine phosphorylase, is
encapsulated in polymeric particles constructed of the amphiphilic triblock
copolymer
PMOXA-PDMS-PMOXA
(poly(2-methyloxazoline)-block-
poly(dimethylsiloxane)-block-poly(2-methyloxazoline))
(Figure
11-1).
The
nanoparticles are permeabilized for substrates and products by integrating
bacterial channel proteins (or porins) in their polymeric wall, a strategy
introduced by Meier and co-workers [24]. The nucleoside specific porin Tsx
was selected as channel-forming protein for the nanoreactors, since it contains
specific nucleoside binding sites. In this way, the Tsx porin allows the efficient
transport of the substrates and products through the capsule wall. This results
in reactors where the enzymatic reaction is restricted to the inner volume of
the polymeric nanocontainer.
Figure 11-1: Schematic representation of the envisioned thymidine phosphorylase
encapsulating nanoreactor.
174
Summary
We constructed thymidine phosphorylase encapsulating nanoreactors (TP-NRs)
according to the method of Ranquin et al. [25]. Monodisperse nanoreactors
with a mean diameter around 200 nm were obtained. SDS-PAGE analysis
revealed an enzyme encapsulation efficiency of ±10% and showed a clear
incorporation of the porin Tsx. The in vitro enzymatic activity of the TP-NRs
was confirmed spectrophotometrically by measuring a decrease in absorption
at 290 nm due to thymidine consumption. The stability of the permeabilized
TP-NRs was investigated by incubating them for several days in mouse serum
at 37°C. TP containing liposomes were used as control samples. Although a
slight decrease in enzymatic activity was observed over time, the TP-NRs are
stable and not leaky at 37°C in serum for several days. The high mechanical
stability of the nanoreactors indicate that they are potentially more suited for
in vivo use than analogous particles constructed of lipids. Next, possible toxic
effects of the TP-NRs on hepatocytes were investigated. Hepatocytes were
isolated from rats and cultured with different concentrations of TP-NRs.
Cytotoxicity was tested by cellular morphology and membrane leakage of
lactate dehydrogenase (LDH assay) as a function of time. The results showed
no significant influence on the viability of the hepatocytes up to 48 h of
incubation. Only after prolonged exposure and at higher doses of TP-NRs,
elevated levels of LDH leakage were measured. This low cytotoxic effect was
confirmed by visual inspection of the morphology of the hepatocytes.
Further, the impact of TP-NRs on the inflammatory status of macrophages was
tested ex vivo and in vivo. For ex vivo studies the effect of nanoreactors on
isolated peritoneal macrophages was analyzed. The results showed that the
nanoparticles do not induce inflammatory cytokine production by naïve
macrophages. Interestingly, the free enzyme significantly provoked the
secretion
of
all
inflammatory
mediators
tested.
This
means
that
the
encapsulation of the enzyme in polymer particles indeed protects the protein
from the immune system. On IFN-γ primed macrophages, a slight porindependent effect was observed. Subsequently, the inflammatory potency of
the nanoparticles was tested in vivo. Only low amounts of cytokines could be
found in macrophage-supernatant and in serum of mice treated with an
175
Chapter 11
intraperitoneal injection of TP-NRs, while high concentrations were detected in
control mice injected with the highly inflammatory compound LPS. Hence, in a
physiologically relevant in vivo setting, the TP-NRs do not provoke acute
inflammatory responses. In a next phase, the nanoreactors were administered
intravenously and enzyme-specific antibody responses were analyzed in the
sera after several boost injections. Surprisingly, the antibody response to the
nanoreactor-encapsulated enzyme was higher than to the free enzyme. The
nanoreactors even seem to act as an adjuvant, increasing the level of specific
anti-TP antibody production. This effect is not desirable for their use in enzyme
replacement therapy, because such antibody responses will cause fast
elimination of the TP-NRs from the blood circulation. Moreover, the antibody
response will only become stronger after several boost injections, hereby
devastating the therapeutic effect of life-long needed enzyme replacement
therapy.
Furthermore, the biodistribution of the nanoreactors was investigated by
comparing the in vivo fate of radiolabeled enzyme-containing nanoreactors
with radiolabeled free enzyme. The results indicated that the biodistribution of
free and encapsulated TP differs significantly. The nanoreactors accumulated
mainly in liver and spleen, while the concentrations in these organs were
clearly lower for the free enzyme. No significant difference in blood circulation
half-life could be demonstrated between the free enzyme and the nanoreactorassociated enzyme. These data indicate that the nanoreactors do not possess
the ideal size and/or surface characteristics to evade hepatosplenic uptake and
to prolong the blood circulation time of the incorporated enzyme significantly.
To investigate whether the nanoreactors are able to decrease the toxic
substrate concentrations in MNGIE patients, the TP-NRs were incubated with
plasma of a MNGIE patient at 37°C. The thymidine and deoxyuridine
concentrations were determined via HPLC. The results showed that the
nanoreactors are not efficient enough to convert the high amount of substrates
to normal concentrations. At that point, a detailed re-evaluation of the
nanoreactor build-up was performed. The results revealed that the enzyme is
not encapsulated properly inside the inner cavity of the nanoreactors, but is
176
Summary
rather associated to the outside of the particle in a non-active conformation.
This explains the enzyme specific antibody response and low enzymatic
activity of the particles. Unfortunately, thus far, we were not able to resolve
the encapsulation problems. Further in depth investigation is needed to
identify the problems encountered and to make better enzyme containing
nanoreactors with high reproducibility.
In
conclusion,
we
constructed
thymidine
phosphorylase
containing
nanoparticles and evaluated their potential as new enzyme delivery vehicles.
All together, we can conclude that although the results potentiate a broad new
class of technologically useful polymer-based nanoreactors, the TP-NRs as
presented in this study do not possess ideal properties for use in enzyme
replacement therapy. Nevertheless, this investigation revealed important
information on the stability, toxicity and immunogenicity of such nanoreactors
and was of high relevance for the further exploitation of these particles as
drug delivery devices.
References
[1] I. Nishino, A. Spinazzola, M. Hirano, MNGIE: from nuclear DNA to
mitochondrial DNA, Neuromuscul Disord, 11 (2001) 7-10.
[2] I. Nishino, A. Spinazzola, M. Hirano, Thymidine phosphorylase gene
mutations in MNGIE, a human mitochondrial disorder, Science, 283 (1999)
689-692.
[3] M.L. Valentino, R. Marti, S. Tadesse, L.C. Lopez, J.L. Manes, J. Lyzak, A.
Hahn, V. Carelli, M. Hirano, Thymidine and deoxyuridine accumulate in tissues
of patients with mitochondrial neurogastrointestinal encephalomyopathy
(MNGIE), FEBS Lett, 581 (2007) 3410-3414.
[4] M.C. Lara, B. Weiss, I. Illa, P. Madoz, L. Massuet, A.L. Andreu, M.L.
Valentino, Y. Anikster, M. Hirano, R. Marti, Infusion of platelets transiently
reduces nucleoside overload in MNGIE, Neurology, 67 (2006) 1461-1463.
[5] M. Hirano, R. Marti, C. Casali, S. Tadesse, T. Uldrick, B. Fine, D.M. Escolar,
M.L. Valentino, I. Nishino, C. Hesdorffer, J. Schwartz, R.G. Hawks, D.L.
Martone, M.S. Cairo, S. DiMauro, M. Stanzani, J.H. Garvin, Jr., D.G. Savage,
Allogeneic stem cell transplantation corrects biochemical derangements in
MNGIE, Neurology, 67 (2006) 1458-1460.
[6] P.F. Chinnery, J. Vissing, Treating MNGIE: is reducing blood nucleosides
the first cure for a mitochondrial disorder?, Neurology, 67 (2006) 1330-1332.
[7] T.A. Burrow, R.J. Hopkin, N.D. Leslie, B.T. Tinkle, G.A. Grabowski, Enzyme
reconstitution/replacement therapy for lysosomal storage diseases, Curr Opin
Pediatr, 19 (2007) 628-635.
177
Chapter 11
[8] W.R. Wilcox, M. Banikazemi, N. Guffon, S. Waldek, P. Lee, G.E. Linthorst,
R.J. Desnick, D.P. Germain, Long-term safety and efficacy of enzyme
replacement therapy for Fabry disease, Am J Hum Genet, 75 (2004) 65-74.
[9] N.W. Barton, R.O. Brady, J.M. Dambrosia, A.M. Di Bisceglie, S.H. Doppelt,
S.C. Hill, H.J. Mankin, G.J. Murray, R.I. Parker, C.E. Argoff, et al., Replacement
therapy
for
inherited
enzyme
deficiency--macrophage-targeted
glucocerebrosidase for Gaucher's disease, N Engl J Med, 324 (1991) 14641470.
[10] D.T. Achord, F.E. Brot, C.E. Bell, W.S. Sly, Human beta-glucuronidase: in
vivo clearance and in vitro uptake by a glycoprotein recognition system on
reticuloendothelial cells, Cell, 15 (1978) 269-278.
[11] J.M. Harris, R.B. Chess, Effect of pegylation on pharmaceuticals, Nat Rev
Drug Discov, 2 (2003) 214-221.
[12] M.S. Hershfield, PEG-ADA replacement therapy for adenosine deaminase
deficiency: an update after 8.5 years, Clin Immunol Immunopathol, 76 (1995)
S228-232.
[13] V. Gaberc-Porekar, I. Zore, B. Podobnik, V. Menart, Obstacles and pitfalls
in the PEGylation of therapeutic proteins, Curr Opin Drug Discov Devel, 11
(2008) 242-250.
[14] T.M. Allen, P.R. Cullis, Drug delivery systems: entering the mainstream,
Science, 303 (2004) 1818-1822.
[15] P. Walde, S. Ichikawa, Enzymes inside lipid vesicles: preparation,
reactivity and applications, Biomol Eng, 18 (2001) 143-177.
[16] M. Yamauchi, K. Tsutsumi, M. Abe, Y. Uosaki, M. Nakakura, N. Aoki,
Release of drugs from liposomes varies with particle size, Biol Pharm Bull, 30
(2007) 963-966.
[17] A. Gabizon, R. Chisin, S. Amselem, S. Druckmann, R. Cohen, D. Goren, I.
Fromer, T. Peretz, A. Sulkes, Y. Barenholz, Pharmacokinetic and imaging
studies in patients receiving a formulation of liposome-associated adriamycin,
Br J Cancer, 64 (1991) 1125-1132.
[18] A.L. Klibanov, K. Maruyama, V.P. Torchilin, L. Huang, Amphipathic
polyethyleneglycols effectively prolong the circulation time of liposomes, FEBS
Lett., 268 (1990) 235-237.
[19] P. Laverman, M.G. Carstens, O.C. Boerman, E.T. Dams, W.J. Oyen, N.
van Rooijen, F.H. Corstens, G. Storm, Factors affecting the accelerated blood
clearance of polyethylene glycol-liposomes upon repeated injection, J
Pharmacol Exp Ther, 298 (2001) 607-612.
[20] D.E. Discher, Emerging applications of polymersomes in delivery: From
molecular dynamics to shrinkage of tumors, Progress in Polymer Science, 32
(2007) 838-857.
[21] D.E. Discher, A. Eisenberg, Polymer vesicles, Science., 297 (2002) 967973.
[22] R. Gref, Y. Minamitake, M.T. Peracchia, V. Trubetskoy, V. Torchilin, R.
Langer, Biodegradable long-circulating polymeric nanospheres, Science, 263
(1994) 1600-1603.
[23] J. Panyam, V. Labhasetwar, Biodegradable nanoparticles for drug and
gene delivery to cells and tissue, Adv Drug Deliv Rev, 55 (2003) 329-347.
178
Summary
[24] C. Nardin, S. Thoeni, J. Widmer, M. Winterhalter, W. Meier, Nanoreactors
based on (polymerized) ABA-triblock copolymer vesicles, Chem Comm, (2000)
1433-1434.
[25] A. Ranquin, W. Versees, W. Meier, J. Steyaert, P. Van Gelder, Therapeutic
Nanoreactors: Combining Chemistry and Biology in a Novel Triblock Copolymer
Drug Delivery System, Nano Lett, 5 (2005) 2220-2224.
179
Chapter 12
Samenvatting
Samenvatting
12 Samenvatting
Heel wat beschreven genetische afwijkingen in de mens hebben tot gevolg dat
een levensnoodzakelijk enzym ontbreekt of slecht functioneert. Hierdoor
accumuleren hoge, vaak toxische concentraties van de substraten, wat de
normale stofwisseling verstoort en aanleiding geeft tot een brede waaier aan
pathologieën. Mitochondriaal neurogastrointestinaal encephalomyopatie of
MNGIE is zo een unieke autosomaal-recessieve aandoening die gepaard gaat
met wijzigingen in het mitochondriale DNA [1]. In 1998 ontdekte Nishino et al.
dat een deficiëntie in het enzym thymidine phosphorylase (TP) de oorzaak is
van deze ziekte [2]. Mutaties in het TP-gen leiden tot een geheel of
gedeeltelijk
functieverlies
van
het
enzym
en
veroorzaken
systemische
accumulaties van de substraten thymidine en deoxyuridine [3]. De beschreven
therapieën voor MNGIE, zoals transfusie van bloedplaatjes [4] en allogene
stamcel transplantaties [5], staan nog in de kinderschoenen. Daarenboven zijn
deze behandelingen verre van ideaal en hebben ze heel wat tekortkomingen
zoals een onvoldoende lange werking van de enzymactiviteit en ernstige
veiligheidsrisico’s [6]. Een veiligere en algemeen aanvaarde strategie voor de
behandeling van enzymdeficiënties is de enzymvervangingstherapie [7, 8].
In de enzymvervangingstherapie wordt het deficiënt enzym intraveneus
toegediend in zijn natieve of gestabiliseerde vorm [9].
In de praktijk is de
intraveneuze toediening van het deficiënt enzym in zijn natieve vorm echter
weinig efficiënt, aangezien eiwitten een korte verblijfstijd in het lichaam
hebben, waardoor het doseringsinterval te kort wordt voor praktisch gebruik
[10].
Om
het
doseringsinterval
te
vergroten
en
de
stabiliteit
en
de
proteaseresistentie te verhogen, worden therapeutische eiwitten vaak covalent
gekoppeld aan poly-ethyleenglycol (PEG) moleculen, een proces dat pegylatie
wordt genoemd [11]. Desalniettemin is pegylatie een complexe en dure
strategie waarbij het enzym vaak een deel van zijn activiteit verliest [12, 13].
Het afleveren van therapeutische enzymen met behulp van passende
dragersystemen is een betere en meer algemeen toepasbare strategie [14].
183
Chapter 12
Zo worden eiwitten vaak ingekapseld in liposomen [15]. Spijtig genoeg
vertonen ook deze lipide partikels nadelen, zoals een snelle vrijgave van het
enzym, lage stabiliteit en snelle klaring door het mononucleair fagocytsysteem
[16, 17]. Door liposomen te construeren die PEG moleculen op hun oppervlak
dragen, ontstond een nieuwe generatie van lang-circulerende (of ‘stealth’)
liposomen [18]. Toch blijven de snelle vrijlating van het enzym en de snelle
klaring
belangrijke
tekortkomingen
van
deze
afleveringssystemen
[19].
Vandaar werden in het afgelopen decennium aanzienlijke inspanningen
geleverd om polymere partikels te vervaardigen met beter controleerbare en
afstembare eigenschappen [20-23].
De bedoeling van deze studie was om polymere nanometer-schaal reactors te
construeren
en
deze
te
evalueren
voor
hun
gebruik
in
enzym-
vervangingstherapie, met MNGIE als modelsysteem. Daarom werd het
therapeutisch enzym, thymidine phosphorylase, ingekapseld in polymere
partikels bestaande uit het triblok copolymeer PMOXA-PDMS-PMOXA (poly(2methyloxazoline)-block-poly(dimethylsiloxane)-block-poly(2-methyloxazoline))
(Figuur 12-1). De wand van de partikels wordt gepermeabiliseerd voor de
substraten en producten door een porievormend membraaneiwit (ook porine
genoemd) in de polymeerwand te integreren. Deze strategie werd voor het
eerst geïntroduceerd door de groep van Meier [24]. Het nucleoside-specifiek
porine Tsx werd geselecteerd als kanaal-vormend eiwit omdat het over
specifieke bindingsplaatsen voor nucleosiden beschikt. Op deze manier zorgt
het Tsx voor een efficiënt transport van de substraten en de producten
doorheen de polymere wand. Dit resulteert in reactoren waar de enzymatische
reactie beperkt is tot de binnenste caviteit van de polymere nanocontainer.
De
thymidine
phosphorylase-bevattende
nanoreactors
(TP-NRs)
werden
vervaardigd volgens Ranquin et al. [25]. Monodisperse nanoreactors met een
gemiddelde diameter van 200 nm werden bekomen. SDS-PAGE analyse
demonstreerde een TP-inkapselefficiëntie van ±10% en toonde aan dat ook
het porine Tsx duidelijk geïncorporeerd zat in de nanoreactor wand. De in vitro
enzymatische activiteit werd bevestigd via spectrofotometrie. De stabiliteit van
de TP-NRs
184
werd
onderzocht
door
ze
enkele
dagen
te
incuberen
in
Samenvatting
muizenserum
bij
37°C.
TP-bevattende
liposomen
werden
gebruikt
als
controlestaal. Ondanks een lichte daling in enzymatische activiteit, bleken de
TP-NRs erg stabiel bij 37°C en vertoonden ze geen snelle vrijlating van het
enzym. Hiermee toont deze studie aan dat de TP-NRs beter geschikt zijn voor
in vivo gebruik dan analoge partikels vervaardigt uit lipiden. Vervolgens werd
onderzocht
of
de
TP-NRs
toxisch
zijn
voor
levercellen.
Verschillende
concentraties TP-NRs werden aan rat hepatocyten toegevoegd en hun effect
op de leefbaarheid van de levercellen werd onderzocht via een lactaat
dehydrogenase test (LDH test) en via een microscopisch onderzoek van de
cellulaire morfologie. Er werd geen significant effect op de leefbaarheid van de
cellen vastgesteld gedurende een incubatieperiode van 48 uur. Enkel na
langere tijd van blootstelling en bij hogere concentraties aan TP-NRs werd een
verhoogde LDH-lekkage vastgesteld. Deze lage toxiciteit werd bevestigd door
een visuele inspectie van de levercellen.
Figuur 12-1: Schematische voorstelling van de beoogde thymidine phosphorylase
bevattende nanoreactoren.
185
Chapter 12
Vervolgens werd de impact van de TP-NRs op de inflammatoire status van
macrofagen onderzocht zowel ex vivo als in vivo. Voor de ex vivo studies werd
het effect van de nanoreactors op geïsoleerde peritoneale macrofagen
geanalyseerd. De resultaten toonden aan dat er geen inflammatoire cytokines
werden geproduceerd door naïeve macrofagen als respons op de TP-NRs.
Nochtans
induceerde
het
vrij
enzym
wel
duidelijk
de
secretie
van
inflammatoire cytokines. Dit betekent dat de inkapseling van het enzym in de
partikels
inderdaad
bescherming
biedt
tegen
inflammatoire
immune
responsen. Op IFN- gestimuleerde macrofagen werd een licht porineafhankelijk effect gemeten. Verder werd het inflammatoir karakter van de TPNRs
ook
in
vivo
getest.
Enkel
lage
cytokine
hoeveelheden
werden
gedetecteerd in het macrofaag supernatant en het serum van muizen die een
intraperitoneale injectie kregen van TP-NRs. In de controle muizen, waar de
sterk inflammatoire component LPS werd toegediend, kon wel een hoge
cytokine respons teruggevonden worden. In een volgende fase werden de TPNRs herhaaldelijk intraveneus toegediend en werd de specifieke antilichaam
respons
op
het
enzym
geanalyseerd.
De
antilichaamrespons
op
de
nanoreactors was tot onze verbazing groter dan de respons op het vrij enzym.
De
nanoreactors
tegenstelling
gedragen
tot
wat
we
zich
dus
als
verwachtten
adjuvans
een
en
hoge
veroorzaken
specifieke
in
anti-TP
antilichaamrespons. Dit effect is geenszins voordelig voor hun gebruik in
enzymvervangingstherapie aangezien dit aanleiding geeft tot een snelle
eliminatie
van
de
TP-NRs
uit
de
bloedcirculatie.
Bovendien
zal
de
antilichaamrespons steeds groter worden na meerdere injecties, wat het
therapeutisch
effect
van
de
levenslang
noodzakelijke
enzymvervangingstherapie teniet zal doen.
Verder werd de biodistributie van de TP-NRs onderzocht en vergeleken met die
van het vrij enzym aan de hand van radiogelabelde TP-NRs en radiogelabeld
vrij TP. De resultaten toonden aan dat de in vivo distributie van de partikels en
het vrij enzym significant verschillen. De nanoreactoren zijn vooral terug te
vinden in de lever en de milt, terwijl de waarden in deze organen duidelijk
lager zijn voor het vrij enzym. Belangrijk is ook dat er geen duidelijk verschil
186
Samenvatting
in bloedcirculatietijden tussen de TP-NRs en het vrij TP vastgesteld kon
worden.
Deze
data
wijzen
erop
dat
de
nanoreactors
geen
ideale
eigenschappen bezitten om te ontsnappen aan opname door het mononucleair
fagocyt systeem in de lever en milt. Hierdoor zijn de TP-NRs ook niet in staat
de halfwaardetijd van het enzym in het bloed aanzienlijk te verhogen.
Tot slot werd onderzocht of de TP-NRs in staat zijn de toxische concentraties
van de substraten in het bloed van MNGIE patiënten te verlagen. Daarvoor
werden de TP-NRs geïncubeerd met plasma van een MNGIE patiënt, waarna de
thymidine- en deoxyuridineconcentraties bepaald werden via HPLC. Spijtig
genoeg werd vastgesteld dat de activiteit van de TP-NRs niet voldoende is om
de substraatconcentraties tot een normaal niveau te verlagen. Deze eerder
onverwachte resultaten dwongen ons de nanoreactorcompositie in vraag te
stellen en een grondige evaluatie van de nanoreactorsamenstelling door te
voeren. De uitgevoerde experimenten brachten aan het licht dat het enzym
niet mooi ingekapseld zit, maar eerder geassocieerd is aan de buitenkant van
het partikel, waardoor het zich in een niet-actieve conformatie bevindt. Dit
verklaard meteen ook de sterke antilichaamrespons tegen het enzym. Tot
dusver werd er nog geen oplossing gevonden voor dit probleem. Verder
onderzoek is nodig om de oorzaak van de inkapselingsproblemen
te
identificeren en betere nanoreactors te vervaardigen.
In conclusie werden er in deze studie enzym bevattende nanopartikels
gemaakt en werd hun therapeutisch potentieel geëvalueerd in het kader van
de
enzymvervangingstherapie
voor
MNGIE.
Hoewel
nieuwe
polymeer-
gebaseerde nanoreactors geconstrueerd werden, moet besloten worden dat de
TP-NRs, zoals hier beschreven, niet de ideale eigenschappen vertonen voor
gebruik
in
belangrijke
enzymvervangingstherapie.
informatie
vergaard
over
Desalniettemin
de
werd
stabiliteit,
heel
wat
toxiciteit
en
immunogeniciteit van deze nanoreactoren en is dit werk van aanzienlijk belang
voor de verdere exploitatie van zulke partikels als enzymafleveringssystemen.
187
Chapter 12
Referenties
[1] I. Nishino, A. Spinazzola, M. Hirano, MNGIE: from nuclear DNA to
mitochondrial DNA, Neuromuscul Disord, 11 (2001) 7-10.
[2] I. Nishino, A. Spinazzola, M. Hirano, Thymidine phosphorylase gene
mutations in MNGIE, a human mitochondrial disorder, Science, 283 (1999)
689-692.
[3] M.L. Valentino, R. Marti, S. Tadesse, L.C. Lopez, J.L. Manes, J. Lyzak, A.
Hahn, V. Carelli, M. Hirano, Thymidine and deoxyuridine accumulate in tissues
of patients with mitochondrial neurogastrointestinal encephalomyopathy
(MNGIE), FEBS Lett, 581 (2007) 3410-3414.
[4] M.C. Lara, B. Weiss, I. Illa, P. Madoz, L. Massuet, A.L. Andreu, M.L.
Valentino, Y. Anikster, M. Hirano, R. Marti, Infusion of platelets transiently
reduces nucleoside overload in MNGIE, Neurology, 67 (2006) 1461-1463.
[5] M. Hirano, R. Marti, C. Casali, S. Tadesse, T. Uldrick, B. Fine, D.M. Escolar,
M.L. Valentino, I. Nishino, C. Hesdorffer, J. Schwartz, R.G. Hawks, D.L.
Martone, M.S. Cairo, S. DiMauro, M. Stanzani, J.H. Garvin, Jr., D.G. Savage,
Allogeneic stem cell transplantation corrects biochemical derangements in
MNGIE, Neurology, 67 (2006) 1458-1460.
[6] P.F. Chinnery, J. Vissing, Treating MNGIE: is reducing blood nucleosides
the first cure for a mitochondrial disorder?, Neurology, 67 (2006) 1330-1332.
[7] T.A. Burrow, R.J. Hopkin, N.D. Leslie, B.T. Tinkle, G.A. Grabowski, Enzyme
reconstitution/replacement therapy for lysosomal storage diseases, Curr Opin
Pediatr, 19 (2007) 628-635.
[8] W.R. Wilcox, M. Banikazemi, N. Guffon, S. Waldek, P. Lee, G.E. Linthorst,
R.J. Desnick, D.P. Germain, Long-term safety and efficacy of enzyme
replacement therapy for Fabry disease, Am J Hum Genet, 75 (2004) 65-74.
[9] N.W. Barton, R.O. Brady, J.M. Dambrosia, A.M. Di Bisceglie, S.H. Doppelt,
S.C. Hill, H.J. Mankin, G.J. Murray, R.I. Parker, C.E. Argoff, et al., Replacement
therapy
for
inherited
enzyme
deficiency--macrophage-targeted
glucocerebrosidase for Gaucher's disease, N Engl J Med, 324 (1991) 14641470.
[10] D.T. Achord, F.E. Brot, C.E. Bell, W.S. Sly, Human beta-glucuronidase: in
vivo clearance and in vitro uptake by a glycoprotein recognition system on
reticuloendothelial cells, Cell, 15 (1978) 269-278.
[11] J.M. Harris, R.B. Chess, Effect of pegylation on pharmaceuticals, Nat Rev
Drug Discov, 2 (2003) 214-221.
[12] M.S. Hershfield, PEG-ADA replacement therapy for adenosine deaminase
deficiency: an update after 8.5 years, Clin Immunol Immunopathol, 76 (1995)
S228-232.
[13] V. Gaberc-Porekar, I. Zore, B. Podobnik, V. Menart, Obstacles and pitfalls
in the PEGylation of therapeutic proteins, Curr Opin Drug Discov Devel, 11
(2008) 242-250.
[14] T.M. Allen, P.R. Cullis, Drug delivery systems: entering the mainstream,
Science, 303 (2004) 1818-1822.
[15] P. Walde, S. Ichikawa, Enzymes inside lipid vesicles: preparation,
reactivity and applications, Biomol Eng, 18 (2001) 143-177.
188
Samenvatting
[16] M. Yamauchi, K. Tsutsumi, M. Abe, Y. Uosaki, M. Nakakura, N. Aoki,
Release of drugs from liposomes varies with particle size, Biol Pharm Bull, 30
(2007) 963-966.
[17] A. Gabizon, R. Chisin, S. Amselem, S. Druckmann, R. Cohen, D. Goren, I.
Fromer, T. Peretz, A. Sulkes, Y. Barenholz, Pharmacokinetic and imaging
studies in patients receiving a formulation of liposome-associated adriamycin,
Br J Cancer, 64 (1991) 1125-1132.
[18] A.L. Klibanov, K. Maruyama, V.P. Torchilin, L. Huang, Amphipathic
polyethyleneglycols effectively prolong the circulation time of liposomes, FEBS
Lett., 268 (1990) 235-237.
[19] P. Laverman, M.G. Carstens, O.C. Boerman, E.T. Dams, W.J. Oyen, N.
van Rooijen, F.H. Corstens, G. Storm, Factors affecting the accelerated blood
clearance of polyethylene glycol-liposomes upon repeated injection, J
Pharmacol Exp Ther, 298 (2001) 607-612.
[20] D.E. Discher, Emerging applications of polymersomes in delivery: From
molecular dynamics to shrinkage of tumors, Progress in Polymer Science, 32
(2007) 838-857.
[21] D.E. Discher, A. Eisenberg, Polymer vesicles, Science., 297 (2002) 967973.
[22] R. Gref, Y. Minamitake, M.T. Peracchia, V. Trubetskoy, V. Torchilin, R.
Langer, Biodegradable long-circulating polymeric nanospheres, Science, 263
(1994) 1600-1603.
[23] J. Panyam, V. Labhasetwar, Biodegradable nanoparticles for drug and
gene delivery to cells and tissue, Adv Drug Deliv Rev, 55 (2003) 329-347.
[24] C. Nardin, S. Thoeni, J. Widmer, M. Winterhalter, W. Meier, Nanoreactors
based on (polymerized) ABA-triblock copolymer vesicles, Chem Comm, (2000)
1433-1434.
[25] A. Ranquin, W. Versees, W. Meier, J. Steyaert, P. Van Gelder, Therapeutic
Nanoreactors: Combining Chemistry and Biology in a Novel Triblock Copolymer
Drug Delivery System, Nano Lett, 5 (2005) 2220-2224.
189
Chapter 13
Publications and scientific
manifestations
Publications and scientific manifestations
13 Publications and scientific manifestations
13.1 Publications
De Vocht C., Ranquin A., Willaert R., Van Ginderachter J., Vanhaecke T.,
Rogiers V., Versées W., Van Gelder P. and Steyaert J., Assessment of stability,
toxicity and immunogenicity of new polymeric nanoreactors for use in enzyme
replacement therapy of MNGIE. J. Contr. Release, 137 (3), 246-254, 2009.
Ranquin A., De Vocht C. and Van Gelder P., Polymer-based nanoreactors for
medical applications. In: Polymer-based nanostructures: Medical Applications.
Broz P., Editor. Royal Society of Chemistry. 315-332, 2010.
Ranquin A., De Vocht C., Wilkinson H., Steyaert J. and Van Gelder P., Prodrug
activation by enzyme encapsulating triblock copolymer nanoreactors.
In
submission.
13.2 Scientific manifestations
11th European Symposium on Controlled Drug Delivery, April 2010, Egmond
aan Zee, The Netherlands, publication in abstract book.
13th International Symposium on Purine and Pyrimidine Metabolism in Man,
June 2009, Stockholm, Sweden, oral presentation.
SBE’s Fourth International Conference on Bioengineering and Nanotechnology,
July 2008, Dublin, Ireland, poster presentation.
Knowledge for Growth, June 2008, Gent, Belgium, oral presentation.
Mini-symposium on polymeric drug delivery systems, April 2008, Brussels,
Belgium, oral presentation.
193
Chapter 13
VIB
Seminar
2008,
March
2008,
Blankenberge,
Belgium,
poster
presentation.
Recent Advances in Drug Delivery Systems, February 2007, Salt Lake City,
Utah, USA, poster presentation.
Symposium
of
the
Belgian
Society
of
Biochemistry,
December
2006,
Gembloux, Belgium, poster presentation.
NSTI
Nanotech,
Mai
2006,
Boston,
Massachusetts,
USA,
poster
Belgium,
poster
presentation.
VIB
Seminar
presentation.
194
2006,
March
2006,
Blankenberge,