DISS. ETH NO. 23258
Improving enzymatic oral therapies via sitespecific polymer conjugation
A thesis submitted to attain the degree of
DOCTOR OF SCIENCES of ETH ZURICH
(Dr. sc. ETH Zurich)
presented by
JESSICA DAGMAR SCHULZ
M. Sc. in Chemistry, ETH Zurich
born on 16.08.1987
citizen of Germany
accepted on the recommendation of
Prof. Dr. Jean-Christophe Leroux
Prof. Dr. Laura Nyström
2016
Abstract
C
ELIAC disease is an autoimmune disorder which is triggered by gluten, a main
proteinic constituent of various grains. The disease-specific inflammation of the
small intestine is developed upon gluten ingestion by individuals who carry a
genetic predisposition (~1% worldwide). Thereby, immunogenic gluten peptides are taken
up by the small intestine where they induce a cascade of inflammatory processes that leads
to e.g., villus atrophy and intraepithelial lymphocytosis, which in turn increases the risk of
developing certain cancers. Classical symptoms consist of abdominal pain, bloating and
diarrhea, but patients also often suffer from fatigue, depression and anemia. Unfortunately,
pharmaceutical treatments are not available and currently only a gluten-free diet can relieve
patients from symptoms and helps avoiding the intestinal damage. Various pharmacological
approaches have reached clinical trials among which one promising strategy to tackle celiac
disease stands out: the oral application of enzymes that can cleave the gluten peptides in
the gastro-intestinal tract. However, the harsh conditions of the oral route can deactivate
the enzymes. In order to develop an efficient oral therapy the biomolecules need to be
stabilized in vivo. To this end, in this work site-specific polymer conjugation was tested on
prolyl endopeptidases (PEPs) known to effectively break-down gluten peptides.
Conjugated polymers can stabilize proteins in the gastro-intestinal tract by protecting them
from enzymatic degradation and physical denaturation. In addition, when performed in a
site-specific manner polymer attachment can enable the enzymes to retain high activity
upon modification.
Two PEPs derived from Myxococcus xanthus (MX) and Spingomonas capsulate (SC) were
engineered to contain cysteine residues that provided the free thiol groups on their surface
to which the polymers were conjugated. A library of single, double and triple cysteinemutated PEPs was successfully produced with the aim of covering most of the enzymes’
surfaces upon polymer attachment. While mutagenesis caused activity loss in the SC
mutants, all MX mutants preserved full MX wild-type (MX WT) activity and led to the
successful generation of two threefold mutated variants (MX3.1 and MX3.2).
Covalent site-specific polymer conjugation was subsequently investigated. First, two
different sizes of the linear polymer poly(ethylene glycol) (PEG, 5 and 40 kDa) were
I
conjugated to mutant MXs. The threefold attachment of PEG to MX3.1 and MX3.2 caused
different impairments on activity (30% vs. 65% respectively) indicating that the attachment
sites are of high importance for activity preservation. Interestingly, no difference in activity
was observed when comparing the MX conjugated to three PEG5 or PEG40 chains.
Additionally, the polycationic dendrimer poly(amido amine) (PAMAM, 6.9 kDa) was
attached to the cysteine-mutated MXs. The triple attachment of PAMAM to MX3.1 led to
full MX WT activity indicating that the polymer chains do not interfere with the enzyme’s
dynamics at these positions, probably due to the dendrimer’s compact and precise structure.
The most active bioconjugates of MX3.1 with highest in vitro stability under simulated
intestinal conditions were then tested in vivo. A gluten-specific sequence carrying a quenched
dye that turns fluorescent upon peptide cleavage was orally administered to rats.
Subsequently, MX or MX–polymer conjugate were applied and the cleavage of the substrate
was detected through the rats’ abdomens over time. The MX WT and mutant MX were
directly deactivated in the gastro-intestinal tract as no signal was observed. In the case of
the conjugate with the three short PEG5 chains some fluorescence was detected but it was
statistically not distinguishable from that of the native enzyme. In contrast, the longer
PEG40 chain as well as PAMAM triply conjugated to MX3.1 led to significant signals upon
oral administration. Importantly, the stabilization effect of the polymers was also
maintained when the conjugates were administered to the digestive tract of rats 1 or 2 h
before the substrate was applied.
The findings presented in this thesis have helped gaining a better understanding of
the protection of enzymes by polymers in the gastro-intestinal tract. By attaching the
polymers site-specifically, structure-activity relationships could be established. Two of the
triply mutated MX3.1 conjugates produced in this project successfully retained their activity
in the digestive tract of rats and therefore position themselves as promising candidates for
the supportive treatment of celiac disease. Finally, the knowledge stemming from this
research is a valuable contribution to the general field of oral therapeutic proteins tract.
II
Zusammenfassung
Z
ÖLIAKIE ist eine Autoimmunerkrankung, die durch Gluten, Hauptbestandteil
vieler Getreidesorten, ausgelöst werden kann. Die krankheitspezifische
Entzündung des Dünndarms kann nach dem Konsum von Glutenprodukten in
genetisch prädispositionierten Personen hervorgerufen werden (1% der weltweiten
Bevölkerung). Immunogene Peptidfragmente, die durch den teilweisen Verdau von Gluten
entstehen, können im Dünndarm von Zöliakieerkrankten aufgenommen werden. Dort
induzieren diese eine Reihe von Entzündungsmenchanismen, die z.B. zu Zottenatrophie
und intraepithelialer Lymphozytose führen. Klassische Symptome bestehen aus
Bauchschmerzen, Blähungen und Diarrhö, wobei Patienten oftmals auch unter anderen
Auswirkungen wie z.B. chronischer Müdigkeit, Depressionen und Anämie leiden. Zudem
kann die chronische Entzündung zu einem höheren Darmkrebsrisiko führen. Bis heute ist
keine medizinische Behandlung möglich und nur eine strikte gluten-freie Ernährung beugt
Symptomen und der Entzündung im Dünndarm vor. Einige pharmakologische Ansätze,
die der Behandlung von Zöliakie dienen könnten, wurden bereits in klinischen Studien
getestet. Eine Möglichkeit der gluten-induzierte Entzündung vorzubeugen, ist die orale
Verabreichung
von
Enzymen,
welche
die
immunogenen
Glutenpeptide
im
Magendarmtrakt verdauen und somit entgiften können. Jedoch werden auch Enzyme im
Verdauungstrakt angegriffen, was oftmals zu deren Deaktivierung führt. Um nun eine
effektive orale Therapie für Zöliakie entwickeln zu können, werden Methoden benötigt,
welche Enzyme im Magendarmtrakt stabilisieren. Aufgrund dessen wurde die Strategie der
ortsspezifischen Polymerkupplung auf Prolylendopetidasen (PEP) angewendet, da diese die
immunogenen Glutenpeptide effektiv abbauen können. Die Polymere sollen hierbei die
Enzyme
von
deaktivierenden
Einflüssen
abschirmen
(z.B.
Gallensäure,
pH,
Verdauungsenzyme), wobei die Aktivität der Enzyme durch die ortsspezifische Kupplung
der Polymere erhalten bleiben soll.
Hierfür wurden gezielt Aminosäuren zweier PEPs, isoliert von Myxococcus xanthus
(MX) and Spingomonas capsulate (SC), zu Cysteinen mutiert um freie Thiolgruppen für die
Polymerkonjugation an die Oberfläche der Enzyme zu platzieren. Eine Reihe von
mutierten Enzymen wurde hergestellt und evaluiert, die ein, zwei oder drei Cysteine
enthalten, mit dem Ziel letztendlich die gesamte Oberfläche der PEPs abzudecken, sobald
III
die Polymere gekuppelt sind. Während die Mutagenese erheblichen Aktivitätsverlust der
SC Mutanten zur Folge hatte, wiesen alle MX Mutanten volle MX Wildtyp (MX WT)
Aktivität auf, was zur erfolgreichen Produktion zweier dreifach cysteinmutierten Varianten
führte (MX3.1 und MX3.2).
Aufgrund dessen wurde die kovalente ortsspezifische Polymerkonjugation
ausschließlich mit den MX Mutanten durchgeführt. Zunächst wurde das lineare
poly(ethylene glycol) (PEG, 5 und 40 kDa) in zwei verschiedenen Größen an die MX
Mutanten konjugiert. Die dreifache Kupplung von PEG an MX3.1 und MX3.2 zeigte
unterschiedliche Auswirkungen auf deren Aktivität (30% bzw. 65% Aktivitätsverlust),
womit die große Bedeutung der Kupplungsorte verdeutlich wurde. Interessanterweise gab
es keinen Aktivitätsunterschied zwischen der dreifachen Kupplung von PEG5 oder PEG40
an MX. Anschlieβend wurde das polykationische Dendrimer poly(amido amine) (PAMAM,
6.9 kDa) über die Cysteine konjugiert. Dabei behielt MX3.1 auch nach dreifacher Kupplung
von PAMAM eine der MX WT entsprechenden Aktivität. Hieraus schließen wir, dass das
Dendrimer aufgrund seiner kompakten und präzisen Struktur nicht mit der Dynamik von
MX interferiert, wenn es an diesen spezifischen Orten gekuppelt ist. Die besten Konjugate,
hinsichtlich in vitro Aktivität und Stabilität in simulierten Darmbedingungen, wurden
anschließend in einem in vivo Modell getestet. Hierbei wurde eine gluten-spezifische
Peptidsequenz, die zudem einen gequenchten Farbstoff trägt, Ratten oral verabreicht.
Anschließend wurde MX oder MX–polymer oral appliziert und sobald das Substrat
gespalten wurde, konnte das entstehende fluoreszente Signal zeitabhängig durch die
Bauchdecke der Tiere gemessen werden. MX WT und MX3.1 wurden umgehend im
Magendarmtrakt deaktiviert, da kein Signal gemessen werden konnte. Obwohl das MXKonjugat, welches drei kurze PEG5 Polymere trägt, ein Signal verursachte, war dies nicht
signifikant im Vergleich zu MX WT. Dreifach konjugierte PEG40- und PAMAMKonjugate zeigten ein intensives Signal. Der Grad der Stabilisierung wurde zudem
untersucht, indem das PEG40 und PAMAM Konjugat 1 oder 2 h vor dem Substrat Ratten
oral appliziert wurden. Die Experimente zeigten, dass beide MX–Polymere auch noch nach
einiger Inkubationszeit im Magendarmtrakt hohe katalytische Aktivität aufwiesen.
Die Erkenntnisse, die aus den präsentierten Resultaten gezogen werden können,
trugen zu einem besseren Verständnis der Stabilisierung von Enzymen durch gekuppelte
IV
Polymere im Magendarmtrakt bei. Durch die ortsspezifische Kupplung konnten
Zusammenhänge zwischen Polymerstruktur und der Aktivität der korrespondierenden
Enzymkonjugate aufgestellt und deren Effekt in vivo bestimmt werden. Zwei der dreifach
konjugierten MX3.1, welche in diesem Projekt erfolgreich hergestellt wurden, zeigten hohe
Aktivität im Verdauungstrakt von Ratten und verkörpern demnach vielversprechende
Kandidaten für eine Enzymtherapie gegen Zöliakie. Zudem hat dieses Projekt generell
erhebliche Erkenntnisse zu dem Bereich der oralen Enzymtherapien beigetragen.
V
X
Table of Contents
Abstract
I
Zusammenfassung
III
1 Background and Objectives
1
2 Improving oral drug bioavailability with polycations?
21
3 Engineering of two prolyl endopeptidases
47
4 Syntheses, in vitro and in vivo characterization of MX bioconjugates
63
5 General Conclusion & Outlook
85
6 Appendix
95
7 References
109
List of Abbreviations
131
Scientific Contributions
135
Acknowledgments
139
Chapter 1
Background and Objectives
Chapter 1: Background and Objectives
F
OOD is an essential source of nutrients and energy but also pleasure, quality of
life and more importantly health. Indeed, research has shown that diet has an
impact on maintaining health or developing (chronic) diseases (e.g., heart disease,
cancer, diabetes, Alzheimer’s disease or nutrient deficiencies).1–4 Thereby, food habits and
changes in the nourishment can affect the raising prevalence of specific diseases.
Globalization has influenced the diet as the population is nowadays generally able to
consume a variety of international food to which it did not have access to a hundred years
ago. Moreover, freshly prepared home-made meals are increasingly replaced by a take-away
culture and highly processed products with a variety of food additives, high levels of sugar,
salt and fat. However, not only convenience food or foreign nutrients that are relatively
new in the nourishment are known to affect health; also common staple (e.g., milk or nuts)
can trigger diseases such as food intolerance or allergy. These food-triggered diseases cause
a variety of symptoms affecting the gastro-intestinal (GI) tract, respiratory tract, skin and
blood circulation and are an increasing health issue in especially the Western society.5
Thereby, the cause might ultimately be linked to the change of diet and its impact on the
digestive tract.6–8 Indeed, the gut microbiota alters with nourishment and its over- or
underdevelopment can cause overreactions upon food intake.9 Additionally, genetic
predisposition and lifestyle have an impact on the development of food-triggered diseases,
contributing to the increasing number of patients suffering from food allergy or intolerance,
which is also partially explained by the elevated awareness on the topic and improved
diagnostic tools. To this end, there are a variety of complex reasons why the incidence of
food-triggered diseases has increased, highlighting the unmet medical need of
pharmaceutical treatments as a challenge to be tackled.
Food allergy is defined as an abnormal immune response to food which can cause
mild to severe effects. Generally, the immune system fights infections to maintain health
but its misregulation can lead to a protective response when specific food content is
identified as harmful. Egg, wheat, milk, peanut, tree nut or fish are major examples
containing allergens that are known to trigger an immune response in individuals
(Table 1.1). Thereby, a variety of symptoms occur e.g., shortness of breath, chest pain, itchy
mouth, diarrhea, swelling of lips, tongue, and throat. Interestingly, children may develop
tolerance over time, whereas food allergies usually remain a life-long condition in adults.10
Various studies were performed testing the impact of breast feeding, its duration and
1
Chapter 1: Background and Objectives
introduction of food on the development of food allergies showing positive effects on their
prevention. The World Health Organization has reviewed the scientific data and published
a recommendation to exclusively breastfeed infants for six months.11 Moreover, several
investigations including clinical studies have been performed in order to induce
desensitization, which was performed subcutaneously, sublingually, transdermally and
orally.12 Thereby, some success was achieved for egg, milk and peanut allergies. In these
cases the amount of food tolerated was effectively increased but major side effects such as
anaphylaxis were observed. Thus, due to its questionable safety, food immunotherapy is
not yet an established treatment. Therefore, omitting the specific triggers in the diet is to
date the only available treatment that effectively avoids symptoms.
Table 1.1. The prevalence of food-related diseases (*in Europe or **worldwide).
Food related disease
Prevalence (%)
Celiac Disease
0.4-1.2*13
Non-celiac gluten sensitivity
0.5-13**14
Wheat allergy
3.6*5
Milk allergy
6.0*5
Egg allergy
2.5*5
Peanut allergy
1.3*5
Tree nuts allergy
2.2*5
Fish and shellfish allergy
1.3*5
Lactose intolerance
Histamine intolerance
2 (Scandinavia)15
~60 (Italy)15
1**16
Food intolerance (or sensitivity) results from an enzyme deficiency in the GI tract.
The insufficient break-down of food ingredients (e.g., lactose, histamine, sucrose or
fructose, Table 1.1) can lead to GI responses, malabsorption, or metabolic problems.15,16
Fortunately, food supplements are available for some cases. These usually consist of the
deficient enzyme such as lactase and diamine oxidase to treat lactose and histamine
intolerance, respectively.17 Generally, effective strategies to avoid food intolerance
2
Chapter 1: Background and Objectives
symptoms can only be developed once the trigger has been identified. When the symptoms
are severe or no supplements are available on the market, the only effective treatment is
omitting specific ingredients in the diet. It was investigated that genetic and environmental
factors as well as drugs that inhibit digestive processes are involved but the mechanism by
which some people lack or only partially express specific enzymes in the GI tract is complex
and barely understood. Therefore, preventing the development of food intolerance or
sensitivity has not been possible yet.
Moreover, the intake of the protein mixture gluten, a major constituent of grains
(i.e., wheat, barley, rye), can lead to symptoms similar to those obtained from food
intolerance. However, because the overreaction to gluten is rather complex and cannot be
categorized as food allergy or intolerance, it is defined as celiac disease or non-celiac gluten
sensitivity (NCGS). Even though research on the pathogenesis, genetic predisposition,
diagnostics and treatments for gluten-triggered abnormalities has developed enormously
over the last two decades, various aspects are still to be understood.
Celiac disease is a chronic autoimmune disorder mainly characterized by the small
intestinal damage triggered by gluten, with a prevalence of ~1% in the western population
(Table 1.1).13 Gluten consists of a polymeric (glutenin) and monomeric (gliadin) fraction,
the latter being rich in proline and glutamine residues. The human GI system is unable to
breakdown such peptides that are usually excreted in the stools without any autoimmune
reaction. These peptides are, however, taken up by the lamina propria in genetically
predisposed individuals where they induce an inflammatory process. When the peptides
pass the epithelium the tissue transglutaminase 2 (tTG2) causes deamidation (transforming
glutamine residues to glutamate), leading to negatively charged fragments.18 The peptide
fragments, especially when deamidated, have a high binding affinity to the cell surface
receptor human leukocyte antigen DQ2 or DQ8 haplotype (HLA-DQ2/8).19 Ninety
percent of celiac patients express HLA with DQ2 and the remaining with DQ8.20 Antigen
presenting cells present (deamidated) gliadin fragments to CD4+ T lymphocytes. These
secrete pro-inflammatory cytokines (e.g., IL-18, TNF-α, IL-21 and IFN-γ), which in turn
activate B lymphocytes. Subsequently, antibodies against gliadin and tTG2 are secreted and
inflammatory CD8+ T lymphocytes and natural killer cells ultimately induce epithelial cells’
apoptosis. Mononuclear cells in the lamina propria and fibroblasts secrete metalloproteases
3
Chapter 1: Background and Objectives
which injure the intestinal wall.21 The consequences of the autoimmune response include
intestinal damage, crypt hyperplasia, villus atrophy and intraepithelial lymphocytosis.
Therefore, “classical” symptoms of celiac disease constitute of diarrhea, abdominal pain
and bloating. In addition, there are several “non-classical” symptoms which do not affect
the GI tract including fatigue, infertility, and depression.20 Diagnosis of celiac disease is
usually firstly indicated by a self-diagnosed food sensitivity. Thereafter, serological testing
for immunoglobulins IgG and IgA is performed. Additional sequence-specific PCR can
identify DQ2/8 alleles and duodenal biopsy can confirm the disease and identify the degree
of inflammation.22
Even though NCGS was first described in 1976, research studies have focused on
the gluten sensitivity intensively since 2010.14 Therefore, the prevalence of NCGS is mainly
unknown, but estimated to be 0.63-6% of the population (Table 1.1).23 NCGS is defined
as gluten sensitivity with negative celiac serology. Typical symptoms include abdominal
pain, bloating, diarrhea, headache and chronic fatigue. The symptoms are usually induced
upon gluten challenge and disappear upon gluten withdrawal. Unfortunately, there is no
direct diagnosis of NCGS, but only the exclusion of celiac disease. Thereby, the diagnosis
consists first of the self-reported food sensitivity, followed by negative celiac disease
serology testing and the absence of villous atrophy on duodenal histology (when on a
gluten-containing diet).22 Indeed, NCGS patients have normal intestinal permeability and
minor histological changes of the GI tract. However, it is not clear yet if genetic background
is as important as in celiac disease. Fifty percent of NCGS patients are tested positive for
HLA-DQ2/8 alleles, which implies that genetic testing can only exclude celiac disease in
half of the cases.24 More studies need to be performed in order to draw a defined line
between the definition of celiac disease and NCGS, and also to specify the difference with
wheat allergy since a variety of studies on NCGS have been performed with wheat rather
than with gluten only. Broader knowledge of the pathogenesis could help developing
diagnosis tools for NCGS as well as potential treatments.
As described above, the pathogenesis of celiac disease is relatively well-investigated,
but the causes for the development of the autoimmune reaction to gluten remain unclear
(Table 1.1, Figure 1.1). The genetic predisposition is an important factor, but not all
individuals expressing HLA-DQ2/8 (~40%) are sensitive to gluten (~1%), implying that
4
Chapter 1: Background and Objectives
other factors are involved. Petersen et al.25,26 found that in the process of presenting gliadin
fragments, specific residues in T-cells’ receptors are essential when complexing
immunogenic gliadin peptides. Thus, the T-cell repertoire could also be another important
genetic factor for the development of celiac disease. Although this finding needs to be
further investigated, it might open new avenues to prevent the inflammatory process in the
intestine. Moreover, various clinical and statistical studies have been performed to explore
the correlation between environmental and geographical factors, the amount of gluten
intake and the duration of breast feeding with the development of celiac disease. Indeed, it
is broadly discussed if the amount of gluten intake promotes its development. Lionetti et
al.27 have studied the correlation of the worldwide distribution of genetically predisposed
populations with the amount of gluten intake and the prevalence of celiac disease
(Figure 1.1). Interestingly, populations that consume higher amounts of wheat also have a
higher frequency of HLA-DQ2 and DQ8 haplotypes. This trend is also observed within a
nation. For example, in India the consumption of wheat is higher in the North, where the
population also expresses the DQ2 allele to a higher extent (32% compared to 9–13% in
the South).27 However, this did not correlate with the disease prevalence. In Europe the
HLA DQ2 haplotype frequency as well as wheat consumption is comparable throughout
the continent, but the prevalence of celiac disease differs among countries. Therefore, other
factors need to be considered in order to fully understand the reasons for developing the
disease. Earlier studies suggested that breastfeeding provided a protective effect against the
development of celiac disease when gluten was introduced to high-risk infants in a strict
time period (between four and six months of age).28,29 However, an international,
randomized, placebo-controlled study involving 994 infants on high risk found that the
development of celiac disease is not correlated with breastfeeding or its duration, nor with
the time of gluten introduction. These findings were further supported by other studies.30–
32
Thus, the full picture of factors leading to the development of celiac disease remains
elusive, making the prevention of celiac disease a significant challenge.33
5
Chapter 1: Background and Objectives
Figure 1.1. World map of 1) level of wheat consumption in g/person/day, 2) frequency of HLA-DQ2
in %, 3) frequency of HLA-DQ8 in %, and 4) celiac disease prevalence in %. Reproduced from Lionetti et
al.27 with permission.
6
Chapter 1: Background and Objectives
The gluten-free diet (GFD) is still the only available treatment for celiac patients that
prevents symptoms and helps curing the intestinal inflammation. Gluten consumption is
generally rather high in the western population (20 g/day) and staying on a safe GFD can
be complicated as the level of gluten content declared as safe is low (~10 mg/day) and may
differ among countries.34 Moreover, the limitation to gluten-free food decreases the quality
of life as recreational activities such as eating in restaurants or going on vacation etc. can be
rather difficult.35 The market for gluten-free products is increasing though thanks to the
raised awareness and number of diagnosed patients. However, these products are usually
expensive compared to regular groceries and may not be affordable for underprivileged
populations. Indeed, complying with a GFD can be a major complication in developing
countries as the access to gluten-free products is difficult and withdrawal of gluten
containing products is often impossible in regions where wheat constitutes a major part of
the diet. Thus, the awareness of celiac disease in developing/poor countries needs to be
supported with access to affordable gluten-free products.
Notwithstanding, for reasons that are yet to be understood not all celiac patients
respond successfully to a GFD as symptoms and especially the intestinal inflammation
might persist. One possible explanation for the poor response to a GFD could be given by
the small intestinal bacteria which are partially overgrown in celiac patients. Based on this
principle, the antibiotic rifaximin inhibiting bacterial RNA synthesis was tested in poorly
responding celiac patients. The clinical study indicated no difference between the treated
and placebo group in terms of symptoms improvement (i.e., abdominal pain, reflux and
diarrhea).36 Moreover, it is known that there is an association between exocrine pancreatic
insufficiency and celiac disease, a possible explanation for persistent symptoms on a GFD.
The benefits of administering Creon®, a pancreatic enzyme replacement therapy, is
currently being evaluated in celiac patients on its the efficacy in diminishing the GI
symptoms.37 With however general little success, studies seeking to identify the causes for
the low response to a GFD are urgently needed in order to find suitable treatments for
these severe cases.
The strategies explored by investigations seeking suitable pharmaceutical treatments
for celiac disease are very diverse.38,39 One approach focuses on the uptake of gliadin
peptides in the small intestine which is upregulated in celiac patients. In general,
7
Chapter 1: Background and Objectives
macromolecules such as the large gliadin peptide fragments are unable to pass the epithelial
layer due to tight junctions that restrict the passage to the lamina propia. However, by
interacting with the chemokine receptor CXCR3 gliadin peptides induce the expression and
activation of zonulin, a protein that modulates the permeability of tight junctions, leading
to enhanced intestinal accessibility (Figure 1.2).40 Therefore, antagonists of zonulin have
been proposed as a possible therapy for celiac disease. Larazotide acetate (AT-1001)
inhibited the opening of the tight junctions in the intestinal epithelium ex vivo and thus,
helped to prevent gliadin-induced permeability.41,42 A phase I clinical trial successfully
demonstrated that AT-1001 was well-tolerated and gave indications that it might reduce
intestinal permeability.43 However, the follow-up trials failed on longer-term studies (2- and
7-week phase of treatment), showing no effect of AT-1001 on gluten-induced permeability
of the intestinal epithelium.44,45 On the contrary, a 12-week phase IIb clinical study found
that a daily dose of 0.5 mg of the zonulin antagonist improved symptoms such as abdominal
pain, headache and tiredness (Table 1.2).46 The discrepancies of the studies could be related
to the number of patients included in each trial, which was rather low in the first studies
(86 or 177) compared to 342 in the second. Even though some success was obtained, the
overall efficacy of AT-1001 remains unknown demanding further clinical investigations.
Besides the paracellular transport of gliadin peptides, the transcellular route can also be
involved in the fragment absorption. Indeed, it was found that secretory immunoglobulin
A (sIgA) enhances the passage of gliadin peptides across the intestinal epithelium.47,48 The
peptides can complex to sIgA which binds to the transferrin receptor CD71 of epithelial
cells, leading to transcellular transport (Figure 1.2). Interestingly, it was found that CD71
is over-expressed on the apical side of the epithelium in untreated celiac patients.47
Therefore, the blockage of this transport pathway could constitute another option for
treating celiac disease. However, it is not clear to which extent transcellular or paracellular
transport of gliadin fragments is involved in celiac disease, demanding further
investigations.
8
Chapter 1: Background and Objectives
Figure 1.2. Suggested treatment options to prevent gluten-induced effects at the intestinal
epithelium. 1) A characteristic feature of untreated celiac disease is compromised epithelial barrier function
caused by abnormal expression of epithelial junction proteins, which enables paracellular passage of gluten
peptides into the subepithelial compartment. This process, promoted by zonulin, has been proposed as a
suitable therapeutic approach for celiac disease. 2) In addition to the paracellular route, intact gliadin
peptides can pass through the epithelium transcellularly by a mechanism involving gliadin–sIgA complexes
binding to transferrin receptor CD71. Blocking transcellular entry of gliadin peptides into the intestinal
mucosa could be the basis of a novel therapeutic strategy in celiac disease. 3) Toxic gliadin peptides induce
epithelial and other cells to secrete IL-15, which results in an increase in the number of intraepithelial
lymphocytes. These intraepithelial lymphocytes become activated via epithelial MICA–NKG2D interaction,
inducing cytotoxic effects on epithelial cells and ultimately increasing epithelial permeability. Blocking IL15-mediated effects has thus been suggested as a therapeutic option for celiac disease. Reproduced from
Kaukinen et al.38 with permission.
In celiac patients the toxic gliadin fragments induce the expression of the cytokine
interleukin IL-15 by epithelial and other cells. IL-15 increases the number of intraepithelial
lymphocytes and upregulates the expression of MICA in the epithelium, a cell surface
protein serving as a ligand for the NKG2D receptor. NKG2D is a transmembrane protein
being expressed by e.g., natural killer (NK) cells. The NKG2D-MICA activates T- and NK
cells responses to epithelial cells and thus, mediates enterocyte apoptosis, which in turn
leads to villous atrophy in the small intestine (Figure 1.2).49 Therefore, blocking IL-15
offers a potential therapeutic strategy to maintain the epithelial barrier functions.
Transgenic mice that overexpress IL-15 in the intestinal epithelium showed extensive
9
Chapter 1: Background and Objectives
villous atrophy that was successfully inhibited with an IL-15-blocking antibody directed
against CD122.50 Moreover, anti-IL-15 monoclonal antibodies inhibited the overexpression
of MICA, neutralized enterocyte apoptosis and downregulated the adaptive immune
response in mucosal biopsies of the small intestine from celiac patients.38,51 A phase I
clinical trial study is currently recruiting celiac patients to test the safety of a humanized
monoclonal antibody, Mik-Beta-1, that blocks IL-15 actions (Table 1.2). This approach
could be particularly important for severe cases of celiac disease because it might help
diminishing the intestinal damage. However, it would not serve as a preventive therapy for
the autoimmune response in the small intestine because other inflammatory cascades are
induced in the lamina propia.
Table 1.2. Overview of therapeutic approaches for celiac disease and *food supplements that have
reached the market. The clinical trials of the green highlighted drug formulations are still ongoing, whereas
the studies of the red highilghted potential therapeutics have been completed.
Phase I
Blockage of IL-15
Mik-Beta-1
Vaccination
Nexvax2
Phase II
Phase III On the market
Anti-gliadin antibody AGY
Gluten binder
BL-7010
Zolunin antagonist
AT-1001
Glutenase
ALV003
STAN1
AN PEP
Tolerase®G*
Parasite therapy
Hookworm
CCR9 antagonist
CCX282-B
10
Chapter 1: Background and Objectives
As mentioned above, the gluten peptides get partially deamidated by the enzyme
tTG2 once they have passed the intestinal epithelium layer. The negative charge increases
their binding affinity towards the HLA-DQ2/8 molecules which triggers the inflammatory
cascade. The inhibition of tTG2 is therefore a potential treatment for celiac disease as it
could decrease the inflammatory process in the intestinal epithelium (Figure 1.3). A variety
of reversible and irreversible tTG2 inhibitors have been developed, mostly unsuccessful
though, due to the existence of nine TG homologs that make the inhibition far too
unspecific.39
HLA-DQ2/8 molecules have been identified to be the most important genetic
factor in celiac patients. Blocking the presenting of gliadin peptides by HLA-DQ2/8 is
therefore another potential strategy to prevent the inflammatory process upon gluten intake
(Figure 1.3). To this end, peptides which compete against gliadin fragments for their
binding to HLA-DQ2/8 have been investigated for their potential anti-inflammatory
effects. Despite being a promising concept, all attempts have so far failed due to structural
similarities between gliadin peptides and the inhibitor molecules that still caused T-cell
activation.52,53
Given that pro-inflammatory cytokines are secreted upon activation of T-cells, the
immunoregulatory cytokine IL-10 has been successfully tested for its capacity to suppress
T-cell activation ex vivo in organ cultures from celiac patients (Figure 1.3).54 IL-10 was not
effective in a clinical pilot study which was, however, performed with celiac patients who
poorly respond to a GFD being complex and barely understood cases.55 T-cell activation
induces an extensive cytokine response whereas the usage of antibodies against IFN-γ, TNF
or tTG2 could thus inhibit the inflammatory process. Although this approach showed
success in other autoimmune disorders,56 these potential drug candidates have, however,
not been substantially evaluated to treat celiac patients. Moreover, other HLA-associated
diseases, e.g., rheumatoid arthritis and multiple sclerosis, showed clinical success in depleting
B-cells with anti-CD20 antibodies.57,58 Even though this approach could also be a promising
strategy for celiac disease, there has been little success so far.59 Generally, approaches
focusing on the inhibition of a specific inflammatory process have not reached clinical
relevance for celiac disease yet. This might be also explained by the various inflammatory
pathways leading to insufficient suppression of the intestinal inflammation by focusing only
11
Chapter 1: Background and Objectives
on one route. Thus, the combination of two or more treatments could lead to more
profound success.
Figure 1.3. Therapeutic options based on prevention of immunological cascades in the lamina
propria . 1) After gaining access to the lamina propria, relevant gliadin peptides are deamidated by tTG2.
tTG2 inhibitors have therefore emerged as drug candidates for celiac disease. 2) Gliadin peptides bind to
HLA-DQ2 and HLA-DQ8 molecules on antigen presenting cells. HLA-blocking compounds could thus
inhibit the presentation of gliadin peptides to T-cells, preventing downstream T-cell activation. 3) T-cell
response in the mucosa could be suppressed by exogenous IL-10, anti-CD3-antibodies or by blocking the
homing of T-cells into the small intestinal mucosa. 4) Activated T-cells secreting inflammatory cytokines
such as IFN-γ and TNF have been suggested as therapeutic targets. 5) Selective B-cell depletion by antiCD20 therapy and blockage of the binding of celiac disease autoantibodies to their target have emerged as
future therapeutic strategies. 6) A further means to prevent or treat celiac disease in the future is the
induction of tolerance by vaccinations or modifying the inflammatory immune response by hookworm
infection. 7) Other therapies include steroid therapy, which are used in refractory celiac disease. Reproduced
from Kaukinen et al.38 with permission.
12
Chapter 1: Background and Objectives
Another feature observed in celiac patients is the enhanced expression of the human
chemokine receptor CCR9 in the duodenum that supports the homing of lymphocyte to
the GI mucosa (Figure 1.3). Blocking the intestinal homing with the antagonist of CCR9
(CCX282-B) was therefore evaluated in a clinical trial with celiac patients.60 In a phase II
trial study CCX282-B was tested in celiac patients by evaluating its effect on the ratio
between villous height and crypt depth upon gluten challenge (Table 1.2). Although the
study has been completed in 2008, the results have not been published yet. Interestingly,
this approach was also tested for Crohn’s disease. Clinical studies found that CCX282-B
was well-tolerated, but the follow-up clinical trial showed that the treatment was not
effective in active Crohn’s disease patients.61,62
The hookworm infection has decreased its prevalence in human beings over time
which is explained by today’s hygienic life-style. This infection is however thought to be an
important player in the immune system due to its function as an “immune trainer”. Indeed,
due to the “hygiene hypothesis” it is believed that the increasing prevalence of inflammatory
diseases is caused by the diminishment of infections like the hookworm infection and is
thus thought to be another therapeutic approach for celiac disease (Figure 1.3). The
infection with parasitic helminth was tested for its capacity to suppress the inflammatory
immune response in celiac patients. In a first phase IIa clinical trial, immunity and glutensensitivity were tested in celiac patients upon infection with the human hookworm
necator.63 Thereby, no therapeutic effect was achieved as mucosal damage, systemic
inflammatory immune responses and symptoms were comparable to those of the control
group. In another clinical study, celiac patients who were inoculated with Necator Americanus
showed a constant height-to-crypt depth ratio upon 1 g of gluten challenge.64 However, the
number of volunteers in the study (12) as well as the amount of gluten administered were
rather low, demanding further investigations although no upcoming clinical trials have been
announced.
Besides attempting to prevent specific actions in the celiac pathogenesis, an
alternative strategy includes the induction of gluten tolerance through vaccination. The
vaccine Nexvax2 was developed based on three immunodominant gluten peptides specific
for HLA-DQ2 patients and is delivered intradermally as booster shots (Figure 1.3). Phase I
clinical trial proved that the therapeutic is well-tolerated in positive HLA-DQ2 celiac
13
Chapter 1: Background and Objectives
patients when weekly injected over two weeks (Table 1.2).65 A phase II clinical study to
evaluate its safety, tolerability and pharmacokinetics in celiac patients with HLA-DQ2
predisposition is currently being carried out.
One possible alternative to a GFD is the consumption of non-toxic wheat or cereals
that are poor in gliadin components (Figure 1.4). Triticum monococcum (TM), also known as
einkorn, was tested for its toxicity in comparison to wheat gliadin. While initial ex vivo
studies carried out on intestinal biopsy samples from celiac patients showed promising
results, a clinical study conducted later on indicated that the short-term challenge with a
single-dose of 2.5 g TM was not enough to evaluate its toxicity on intestinal
permeability.66,67 A follow-up phase II study assessed the longer term safety of 6 g TM over
60 days on celiac patients following a GFD for one year and found that the dose caused
villous atrophy in four out of five volunteers, which indicated that the long-term
administration of TM would not be safe for celiac patients.68
Another approach consists of gliadin sequestration in the stomach by the synthetic
polymeric binder poly(hydroxyethyl methacrylate-co-styrene sulfonate) (P[HEMA-co-SS],
BL-7010) (Figure 1.4).69–71 Once gliadin is bound to the polymer, it is masked from
digestive enzymes. Thus, the formation of the immunogenic peptide fragments is decreased
and the gliadin-polymer complex is excreted in the feces. Indeed, P[HEMA-co-SS] reduced
the secretion of inflammatory cytokines ex vivo in mucosal biopsy specimens from patients
with celiac disease.71 Moreover, studies with celiac disease mouse models showed reduced
paracellular intestinal permeability and an attenuated systemic immune response to gluten.70
Phase I/II clinical trial evaluated the safety and systemic exposure of single and repeated
administration to celiac patients (Table 1.2). Generally BL-7010 was found to be safe and
well-tolerated upon multiple administered doses, and the dose of 1 g was selected for
further investigations. Now, the formulation will be optimized and tested in well-controlled
celiac patients upon gluten challenge. Another oral approach focuses on the administration
of anti-gliadin chicken egg yolk IgY antibodies to inhibit the peptide absorption by binding
and sequestering gliadin. Caco-2 cell culture studies showed promising results as IgY
suppressed pro-inflammatory responses.72 AGY, the IgY antibody product, is currently
being tested in celiac patients in a phase I clinical trial.
14
Chapter 1: Background and Objectives
Generally, the GI immune system majorly influences the oral tolerance. Thereby,
the microbes are responsible for nutrient absorption, barrier function and immune
development and are thus key players in maintaining health or developing diseases.73 Thus,
the degree of developed intestinal bacteria majorly influences the GI health. Interestingly,
when analyzing the microbes of celiac patients, it was found that they present altered
microbiota composition, e.g., the level of Bifidobacterium is lower than in the healthy
population.74 In vitro and in vivo bifidobacteria reduced toxic effects of gliadin peptides.75,76
Thus, Bifidobacterium infantis was evaluated in active celiac patients in a clinical study and
although no positive effects were observed on intestinal permeability or inflammatory
abnormalities, patients suffered less from diarrhea.77 Even though probiotics could offer a
new strategy to fight celiac disease, the influence of the gut microbiota has not been
completely characterized in this autoimmune disease yet (Figure 1.4). A complete celiacspecific intestinal microbiota needs to be identified and its function in the development of
the disease established. Only then reasonable probiotic-based alternatives will arise to treat
or even prevent the sensitivity to gluten.
Figure 1.4. Investigational approaches active in the lumen of the small intestine that could be used
for celiac disease treatment. Gluten is highly resistant to degradation by GI enzymes and thus fairly long
peptides enter the intestinal lumen. In patients with celiac disease, these peptides trigger deleterious
downstream effects. 1) Cereal products from non-toxic cultivars produced by selection, breeding and genetic
engineering would lack the harmful gluten peptides that induce the disease. 2) Sourdough fermentation
leads to gluten degradation during food processing. The ingestion of foods produced by this technique
should be safe for patients with celiac disease. 3) The degradation of harmful gluten peptides into non-toxic
fragments can be accomplished after ingestion by probiotics such as bifidobacteria or by oral enzymes. 4)
The binding of the polymer P(HEMA-co-SS) to gluten in the GI lumen can prevent the production of
harmful gluten peptides and their downstream effects. Reproduced from Kaukinen et al.38 with permission.
15
Chapter 1: Background and Objectives
One of the most promising approaches to treat celiac disease is the use of orally
applied enzymes. Food supplements are available on the market for various food
intolerances such as the intolerance to lactose. Unfortunately, this strategy is more difficult
to implement in the case of celiac disease as the problem does not lie on the lack of a natural
GI enzyme that can be orally applied. Herein, foreign enzymes need to be identified that
can completely detoxify gluten by breaking down the proline and glutamine-rich gliadin
peptides and that are safe and stable in the GI tract (Figure 1.4). Prolyl endopeptidase
enzymes (PEPs) cleave peptides in a post-proline-specific fashion and have been isolated
from a variety of bacteria and fungi, e.g., Mycococcus xanthus, Spingomonas capsulate, Aspergillus
niger, Flavobacterium meningosepticum.78–82 Indeed, it was shown that PEPs are able to detoxify
gliadin peptides and prevent the induction of a T-cell response in celiac patients.80,81
Moreover, the cysteine endoprotease B (EP-B2) isolated from germinating barley seeds is
able to break down the peptides at glutamine residues.83 One advantage of EP-B2 is its high
stability at acidic pH, which is important for the enzyme to efficiently break down the
peptide fragments in the stomach before they reach the small intestine, as was successfully
demonstrated in a rat model.84 The combination of both strategies (proline- and glutaminespecific enzymes) resulted in a mixture of the PEP derived from Sphingomonas capsulate (SC)
and the EP-B2 which reached clinical trials as ALV003 (Table 1.2). In a phase I study it
was proven that all tested doses of ALV003 were well-tolerated and did not cause any
allergic reactions or other symptoms.85,86 Moreover, it was shown that the enzyme mixture
were stable in fed stomachs as it cleaved gluten up to 88% in vivo.86 In a phase IIa trial celiac
patients were treated daily with 900 mg ALV003 over six weeks which led to attenuate
gluten-induced small intestinal mucosal injury upon a daily gluten intake of 2 g.87 To
evaluate the effects of different ALV003 doses on the mucosa and symptoms of celiac
patients a phase IIb clinical study is currently conducted. Moreover, the detoxification of
gluten was evaluated in vitro with a mixture of commercially available proteases
(aspergillopepsin from Aspergillus niger and dipeptidyl peptidase IV from Aspergillus oryzae).88
Phases I and II of clinical trials are currently ongoing for this mixture denominated STAN1.
Additionally, a prolyl endoprotease derived from Aspergillus niger (AN) has shown high
stability and efficacy at slight acidic pH (pH ~4–5) in vitro and in a dynamic in vitro GI model
that mimics in vivo digestion.78,89 AN was therefore tested in phase I and II clinical trials
with healthy participants to determine safety and efficacy with, however, 12 volunteers
16
Chapter 1: Background and Objectives
only.90 Another study in which AN was evaluated in NCGS participants has just been
completed but its results have not been published yet. Despite the data supporting its
efficacy as a pharmaceutical treatment for NGCS being unpublished, the AN PEP-based
enzyme supplement reached the market in the United States in the summer of 2015 under
the name Tolerase®G. Thereby, the product needs to be carefully considered since the
safety of gluten-intake by celiac patients upon Tolerase®G administration has not been
demonstrated.
Even though oral enzyme therapy constitutes a promising approach to treat celiac
disease and maybe also NCGS, enzymes themselves are prone to get digested in the GI
tract. Naturally occurring PEPs are generally rather unstable at the low acidic pH
encountered in the stomach and are not resistant to the digestive enzymes found in the GI
tract (especially pepsin). Hence, protective strategies are needed in order to deliver
functional enzymes with high activity via the oral route. Recently, Wolf et al.91 engineered a
naturally occurring acid-active peptidase by computational design in combination with
mutagenesis.92 The gliadin endopeptidase Kuma030 was designed to target specifically the
dipeptide PQ which is frequently found in the immunogenic gliadin fragments but much
less in the sequence of other proteins. This feature is important to minimize potential sidereactions in vivo. The high efficiency of the engineered enzyme was shown by the potent
degradation of immunodominant epitopes from wheat, barley and rye (> 99% within
minutes). Due to the promising results the authors announced to perform in vivo studies to
further analyze its potential as a therapeutic agent for celiac patients. Fuhrmann et al.93
developed an approach in which polymer conjugation to a PEP derived from Myxococcus
xanthus (MX) stabilized the enzyme in vivo. Polymers with different character were randomly
conjugated to MX via solvent-exposed lysine residues. The negatively charged poly(acrylic
acid) (PAA) conjugate did not show stabilization in the GI tract of rats, whereas methoxy
poly(ethylene glycol) (mPEG) covalently bound to MX improved the cleavage of a
fluorescent immunogenic peptide in vivo. The most promising candidate was, however, the
bioconjugate of MX and PG1 (poly-(3,5-bis(3-aminopropoxy)benzyl)-methacrylate), a
positively charged dendronized polymer. The comparison of the native enzyme and MX–
PG1 showed that the latter remained in the stomach of rats for longer time, indicating an
interaction of the conjugate with the negatively charged mucosa. However, this conjugate
was approximately four times less active than the native protein in vitro. Despite polymer
17
Chapter 1: Background and Objectives
conjugation being a promising strategy with various products on the market (mainly applied
to the blood), high loss in activity causes the application of higher amounts of the nonhuman material increasing the probability of side effects. Moreover, a direct comparison of
the bioconjugates was not possible in this study due to a number of reasons. Firstly,
randomly modified proteins are hard to characterize and reproduce caused by high
heterogeneity of the conjugates. Secondly, PG1, PEG and PAA differ in size (59, 4.6, 14
kDa, respectively) and finally, due to random conjugation different numbers of polymer
chains were attached to MX (1, 14, 15 polymer chains respectively).
Nevertheless, polymer conjugation offers a promising approach to overcome the
deactivation of PEPs in the GI tract while the drawbacks of random polymer attachment
need to be tackled. Thus, the general aim of this project was to synthesize PEP–polymer
conjugates using neutral and positively charged polymers attached in a site-specific manner.
The objective included the efficient stabilization of the enzymes in vitro and in vivo while
retaining high activity upon conjugation of a low number of well-placed polymer chains. In
the first part of this work (Chapter 2), a review on the application of cationic polymers in
oral drug delivery systems is presented. Herein, the effect of polycations on GI stability,
efficiency and bioavailability of drugs (small molecules, peptides and proteins) is discussed
focusing on polymer-drug solutions, nanoparticles and drug-polymer conjugates. In
Chapter 3 the engineering of PEPs for which the crystal structures have been identified is
presented (Figure 1.5). Thus, PEPs derived from MX and SC were modified by sitespecifically introducing cysteine residues on the surface of the proteins. A library of singly,
doubly and threefold mutated MXs and SCs were synthesized, expressed and characterized.
The mutants were assessed for their in vitro activity to ensure that the modifications did not
cause major deactivation. Chapter 4 presents the site-specific polymer attachment to MX
mutants and the evaluation of the best candidates in vivo. Polymers of different
characteristics were conjugated to the cysteine mutated MXs to yield well-defined
bioconjugates (Figure 1.5). Herein, a structural comparison of the conjugates was
performed in terms of polymer length, structure (linear vs. dendrimer) and charge (neutral
vs. positively charged). The parameters were evaluated in relation to their impact on
enzymatic activity and stability in vitro. Moreover, to compare different strategies of polymer
conjugation, random and site-specific polymer attachment were combined with two
different kinds of polymers. Thereafter, the most promising MX–polymer conjugates were
18
Chapter 1: Background and Objectives
evaluated for their ability to enhance enzymatic stability in the GI tract of rats. Herein, an
immunogenic peptide that turns fluorescent upon cleavage by the enzyme/conjugates was
monitored in vivo in real-time (Figure 1.5). Finally the work is concluded in Chapter 5 with
a general outlook on this topic.
Figure 1.5. Objectives of the doctoral thesis: 1) PEPs with identified 3D structures were chosen for
mutagenesis experiments. The crystal structure of MX is presented with highlighted residues (red) which
were modified to cysteines. Activity of mutants with 1-3 cysteine residues on the surface of the PEPs was
evaluated with a model substrate releasing nitro-aniline upon cleavage. 2) Polymers of different character
were site-specifically conjugated to the PEPs via the engineered cysteine residues. Additionally, site-specific
and random conjugation was combined with two different polymers. The bioconjugates were tested for
their activity and stability in vitro. 3) The most promising candidates were then evaluated for their stability in
vivo. After their oral administration to rats, the signal of a probe that turns fluorescent upon cleavage was
monitored in real-time.
19
Chapter 2
Improving oral drug bioavailability with
polycations?
This chapter is published in:
Schulz, J. D., Gauthier, M. J., and Leroux J.-C. Improving oral drug bioavailability with
polycations? Eur. J. Pharm. Biopharm. 2015; 97:427-3.
Chapter 2: Improving oral drug bioavailability with polycations?
2.1. Introduction
D
RUG administration via the oral route is widely used, and is preferred due to
its convenience and low associated costs. However, the oral delivery of
compounds which are sensitive to degradation, poorly water soluble, or poorly
membrane-permeable remains challenging. For example, biomacromolecular drugs such as
proteins and polynucleotides can easily undergo denaturation or degradation in the GI tract
due to pH fluctuations, and the presence of surfactants (e.g., bile salts) and enzymes.94,95
Furthermore, their high molecular weight and polarity lead to inefficient permeation
through mucus and the epithelium. Among the different strategies that have been
investigated to address this challenge, formulating drugs with polycations has been shown
to increase their solubility, protect labile compounds from pH changes and digestive
enzymes, and serve as permeation enhancers. In addition, polycations are particularly
interesting for oral drug delivery because of their mucoadhesive properties, which increase
the retention time of drugs in the GI tract, and their ability to promote the absorption
process by a variety of mechanisms.96,97 Even though polycations can destabilize cell
membranes and exhibit toxicity, compared to the parenteral route, the systemic exposure
to polycations delivered orally is assumed to be low, and therefore better tolerance is
expected.98,99 This contribution will begin by providing an overview of the intrinsic
mechanisms by which the most widely investigated polycations (e.g., chitosan, poly(amido
amine) (PAMAM), poly(L-lysine)) in pharmaceutical sciences can increase oral
bioavailability or induce toxicity when co-administered alongside drugs. Thereafter, aspects
of the formulation of drugs with polycations into nano/microparticles and bioconjugates
are reviewed, including a presentation of the performance of such systems in vivo. This
manuscript will not cover buccal delivery100 as well as the use of polycations in the
preparation of tablets,101–103 or polycations complexed with polyanions in layer-by-layer
assembling systems.104,105
21
Chapter 2: Improving oral drug bioavailability with polycations?
2.2. Free polycations
Because of its simplicity, co-administration of polycations with drugs in solution
(Figure 2.1A) is the most convenient and widely investigated manner to affect the oral
bioavailability of therapeutic agents. This section introduces several polycations that have
been orally co-administered with drugs, and discusses their toxicity and the mechanism by
which they increase bioavailability.
2.2.1. Chitosan
Chitosan is an attractive polymer for biomedical applications due to its biocompatibility,
biodegradability, natural origin, and low cost. It is a linear polysaccharide with a molecular
weight up to ca. 2,000 kDa, obtained by partial N-deacetylation of chitin (Figure 2.2A)
from crustacean shells or mushrooms.106 Chitosan is composed of 2-amino-2-deoxy-Dglucopyranose and 2-acetamido-2-deoxy-D-glucopyranose units randomly linked together
via β-(1,4) glycosidic bonds (Figure 2.2B). Its charge density, which influences properties
such as solubility, permeability, and toxicity, is dependent on the degree of deacetylation
(DA) because only the 2-amino-2-deoxy-D-glucopyranose units are ionizable.107 Chitosan’s
aqueous solubility is limited by its pKa of ~5.5–6.5 to acidic solutions, where the polymer
possesses a net positive charge. Although chitosan is generally considered as safe, toxicity
has been reported to be strongly dependent on the DA. In vitro, highly deacetylated chitosan
(DA 99%) exhibited toxicity towards monolayers of cultured intestinal epithelial cells
(Caco-2), induced morphological changes in the form of a decrease in the number of
microvilli, and an alteration in the organization of the terminal web.108 Conversely, chitosan
with a lower degree of DA (≤ 65%) produced less morphological changes and generated
signs of cytotoxicity only at high concentrations. In vivo, rats orally administered a chitosan
oligosaccharide (1 kDa) at 2 g/kg/day did not show any significant behavioral changes or
altered biochemical marker levels compared to control groups.109 Because of its relative
good safety profile, chitosan is the most widely investigated polycation co-administered
alongside drugs in oral delivery systems.110,111 Its ability to modulate the permeation through
Caco-2 cell monolayers was found to depend on pH, molecular weight, and DA of chitosan.
At pH 6.2, the transport of mannitol was significantly increased, whereas no effect was
observed at pH 7.4 due polymer insolubility.112,113 Chitosan, with a DA of 65% and high
22
Chapter 2: Improving oral drug bioavailability with polycations?
Figure 2.1. Drug–polymer formulations and their advantages. A: Drug formulations prepared with
polycations by either simple mixing to obtain a polycation-drug solution, encapsulating the drug in
polycationic micro- or nanoparticles, or by synthesizing a covalent drug–conjugate. B: The different
formulations should enhance drug stability in the GI tract and shield the drug from the acidic pH, GI
enzymes, and bile salts. C: Adequate drug release is required at the targeted site. This is achieved by either
releasing the drug from the polymeric particle or by cleavage of the covalent bonds attaching it to the
polymer. D: Transport across the intestinal membrane can occur via the paracellular and/or transcelluar
pathway through epithelial or M cells.
molecular weight (170 kDa), had a large impact (eight-fold increase) on the mannitol
passage through Caco-2 cell monolayers. Inversely, lower molecular weight chitosan with a
low DA was not able to affect its transport.108 However, Caco-2 cells lack the mucus layer
which is an important barrier for drug absorption in vivo. This became evident as permeation
of atenolol, a poorly absorbed hydrophilic drug, through Caco-2 cell monolayers increased
significantly during chitosan exposure at concentrations five times lower than those used
in the in situ rat perfusion model.114 Indeed, when administered as a solution to rats by single
pass intestinal perfusions, chitosan only slightly improved the bioavailability of atenolol.
Therefore, the influence of mucus on absorption enhancement by chitosan was tested on
a mucin-producing cell line (HT29-H). While the permeation of mannitol was enhanced, a
smaller effect was observed than for Caco-2 cell monolayers. Using fluorescently-labelled
chitosan, it was shown that the mucus prevented the interaction between chitosan and
23
Chapter 2: Improving oral drug bioavailability with polycations?
HT29-H cells. The role of the mucus was further revealed by reducing its thickness and
showing that the labeled polymer could then achieve closer contact with the cells.114
The mechanism by which chitosan enhances mucosal absorption is multi-faceted
and not fully understood. Interaction with the cell membrane is an important factor that
can facilitate the absorption of drugs in the GI tract. Indeed, chitosan’s positive charges
can interact electrostatically with the negatively charged sialic acid residues on mucin (pKa
2.6).96,97 The spatial distribution of zonula occludens-1 (ZO-1) (tight junction associated
protein), occludin (transmembrane protein of the tight junctions), and cytoskeletal F-actin
was altered in Caco-2 cell monolayers in the presence of chitosan.112,115,116 Additionally,
chitosan reversibly and significantly decreases the transepithelial electric resistance (TEER)
of Caco-2 cell monolayers, which is a measure of the tightness of epithelial cell layers.117
These findings are indicators that chitosan widens intercellular junctions, thus affecting
paracellular transport (Figure 2.1D). Moreover, the importance of the positive charges was
underlined by the fact that the increased transport of mannitol could be inhibited by
heparin, a polyanion that complexes chitosan.115
In addition to poor permeability, premature digestion in the GI tract can be an
important factor responsible for the low bioavailability of certain drugs, such as peptides.
Therefore, polymers that can simultaneously enhance drug permeation and inactivate (or
reduce the activity of) luminal digestive enzymes are particularly interesting (Figure 2.1B).
Electrostatic interactions between polycations and enzymes present in the GI tract, or the
change of local pH due to the buffering capacity of the numerous amino groups, could
mitigate the activity of digestive enzymes. Although chitosan in solution was found to not
intrinsically inhibit trypsin or carboxypeptidase B, it was superior in enhancing the
bioavailability of the peptide buserelin in vivo compared to poly(acrylates), which are known
to reduce the activity of luminal enzymes.118,119 Thus, for this rather stable drug, the increase
in paracellular transport seems to play a more important role on the oral bioavailability than
the inhibition effect on GI enzymes.
Conclusively, various factors, especially chitosan’s architectural properties, e.g., DA
and molecular weight, significantly influence its ability to enhance drug transport after oral
administration when the polycation is co-administered.
24
Chapter 2: Improving oral drug bioavailability with polycations?
Figure 2.2. Molecular structure of un-ionized polycations: A: chitin, B: chitosan, C: TMC, D: PAMAMNH2 G2, E: poly(L-lysine), F: poly(L-ornithine), G: poly(L-histidine), H: poly(ethylenimine). DA = degree
of deacetylation.
2.2.2. N -trimethyl chitosan (TMC)
As previously discussed, one of the challenges associated with the use of chitosan as an
absorption enhancer is its low solubility at neutral to basic pH. This can be overcome by
methylating its amino groups to produce the quaternary derivative N-trimethyl chitosan
(TMC) (Figure 2.2C), which is water soluble over a broad pH range (pH 1–9).120 Trimethylated amino groups remain charged even at higher pH values and thus, the degree of
quaternization (DQ) of TMC strongly impacts its characteristics. Lactate dehydrogenase
assay of cells exposed to TMC demonstrated destabilization of the cell membrane when
the DQ increased. This was confirmed when TMC was administered orally to mice showing
that only a low DQ (22%) TMC was non-toxic at a dose of 540 mg/kg.121 Besides toxicity,
the DQ is an important parameter for modulating the permeability of intestinal cells. At
neutral pH, a rise in the DQ led to a more pronounced effect on TEER and permeability
of mannitol through Caco-2 cell monolayers.122 In comparison, a TMC with a DQ of 40%
was 2–3 fold less efficient than one with a DQ of 60% in promoting the
25
Chapter 2: Improving oral drug bioavailability with polycations?
permeation/absorption of the peptide buserelin, both in vitro and in vivo.123 Moreover,
exposure of fluorescently-labeled TMC to Caco-2 cell monolayers showed that the polymer
remained localized in the intercellular space suggesting a paracellular pathway
(Figure 2.3).113 Importantly, no labeled TMC was observed within the cells, implying the
absence of nonspecific cell internalization. Moreover, similarly to chitosan, TMC did not
inhibit α-chymotrypsin.117
It was reported that TMC was less effective in reducing TEER and increasing the
permeability of various peptide drugs at acidic pH compared to chitosan.117 The positively
charged amino groups of TMC are sterically hindered by methyl groups that may lead to a
less efficient interaction with the mucin compared to the primary amino groups of
chitosan.113 Nevertheless, TMC’s greater solubility at neutral and basic pH offers distinct
advantages vs. chitosan leading to a more efficient absorption enhancer. However, as
toxicity and permeation enhancement by co-administration of TMC directly correlate with
DQ and thus, with the number of positive charges, it is essential to carefully choose a TMC
candidate to meet efficacy and safety.
Figure 2.3. Optical vertical cross-section (XZ image) through a Caco-2 cell monolayer after 60 min of
incubation with fluorescein isothiocyanate-labeled dextran (1 mg/mL) and TMC (0.5%). Top is apical and
bottom is basolateral. Arrow indicates paracellular transport. Reproduced from Kotzé et al.113 with
permission.
26
Chapter 2: Improving oral drug bioavailability with polycations?
2.2.3. Poly(amido amine) (PAMAM)
PAMAMs (Figure 2.2D) were first introduced in 1985, and are a class of water-soluble
dendritic macromolecules constructed by the successive grafting of amino acrylate units to
a multi-functional initiator core molecule. Each grafting step increases the “generation”
number (G) of the dendrimers. Unlike typical synthetic polymers, dendrimers are
unimolecular and have a correspondingly well-defined shape and size.124 Due to their
specific synthetic route, all PAMAMs possess internal tertiary amino groups (pKa 3–6) and
carry variable terminal groups. Particularly interesting are PAMAMs with terminal amino
groups (pKa of 8–9) which are positively charged at physiological pH and possess
mucoadhesive properties.125,126
Cytotoxicity, permeation, and transport mechanisms have been investigated for
PAMAMs and surface-modified PAMAMs using monolayers of Caco-2 cells. The
cytotoxicity of dendrimers seems to be directly related to the generation number. Although
G0–G2 PAMAM did not show a significant effect on microvilli structure at 1 mM
(Figure 2.4A), higher generation PAMAMs induced considerable cell-membrane damage
and loss of microvilli above 0.01 mM (Figure 2.4B).127 This is likely caused by the increase
in membrane interactions due to a higher number of positively charged amino groups per
dendrimer. Indeed, the conjugation of six lauroyl moieties to PAMAM reduced its charge
and decreased its cytotoxicity. In vivo, PAMAM generation was also found to impact toxicity.
G4 PAMAM was tolerated in mice after oral delivery up to 300 mg/kg, whereas G7 showed
toxicity > 30 mg/kg as revealed by changes in behavior, body weight, and biochemical
parameters.128,129
The paracellular route is one possible transport pathway for dendrimers to cross the
epithelium (Figure 2.1D) given that immunofluorescence studies showed a disruption of
the staining pattern of tight junction proteins.130,131 However, dendrimer permeation was
temperature-sensitive and was significantly reduced by an endocytosis inhibitor. This
indicates that the transport mechanism of dendrimers through Caco-2 cell monolayers is
not only via the paracellular pathway but may also involve endocytosis-mediated
transepithelial transport by a transcellular route.130 Moreover, TEER was decreased
transiently returning to > 95% of the initial values 24 h after exposure to G2 or G3
PAMAM.132 The transport of mannitol was favored, whereas co-administration of the G3
27
Chapter 2: Improving oral drug bioavailability with polycations?
dendrimer enhanced the permeability to a greater extent than G2, which may result from
the higher number of positive surface charges (64 amino groups, compared to 32 on the
G2). Moreover, Lin et al.133 have examined the co-administration of PAMAMs with various
hydrophilic molecules in an in situ intestinal rat model. It was found that the intestinal
absorption of 5(6)-carboxyfluorescein increased by up to 11-fold in the presence of a 0.5%
G2 PAMAM solution.133 The effect depended on the size of the molecule being transported
leading to a significant increase with hydrophilic molecules up to 4 kDa. Transport of larger
compounds (e.g., insulin, 5.8 kDa) could not be enhanced with PAMAMs. Thus, small
PAMAMs in solution can act as a safe absorption enhancer for small to average size
hydrophilic compounds.
A
B
C
Figure 2.4. Transmission electron microscopic images of Caco-2 cell monolayers after treatment with
PAMAM dendrimers (1 mM) for 2 h: A: G2NH2; B: G4NH2; C: control cells. The images display a
generation-dependent effect of PAMAM dendrimers on Caco-2 microvilli (magnification = 12,500×). Scale
bars = 1 µm. Reproduced from Kitchens et al.127 with permission.
28
Chapter 2: Improving oral drug bioavailability with polycations?
2.2.4. Poly(L-lysine) and other polycations
Polycations in solution other than chitosan/TMC and PAMAMs have been significantly
less studied in oral drug delivery systems. Among them, poly(L-lysine) (Figure 2.2E), a
natural homopolymer with a molecular weight generally ranging from 3 to 220 kDa, is
produced by bacterial fermentation or organic synthesis.134 Its amino groups are positively
charged at a broad pH range (pKa 9–10) creating a high charge density that is the reported
cause of its cytotoxicity.135 The reduction of charge density via cyanate-modification
successfully decreased its cytotoxicity up to 25-fold in a non-intestinal fibroblast cell line
(L929).135 Furthermore, poly(L-lysine) was shown to affect intestinal epithelial monolayers
in a dose- and time-dependent manner. More specifically, high molecular weight poly(Llysine) (300 kDa) had a higher impact on TEER and insulin passage through Caco-2 cell
monolayers than a lower molecular weight variant (30 Dka). Stereochemistry did not appear
to play a role as no difference was observed by replacing L-lysine by D-lysine.136 Exposure
to poly(L-lysine) changed the pattern of cytoskeletal F-actin, and affected the distribution
of ZO-1 leading to the hypothesis that it influences the paracellular route (Figure 2.1D).
Similarly,
other
polycations
such
as
poly(L-ornithine),
poly(L-histidine),
poly(ethyleneimine) (PEI) (Figures 2.2F–H), and protamine also induced a significant
increase in transport of hydrophilic compounds through intestinal epithelial
monolayers.137,138 Poly(L-histidine) was less effective than the other polycations, likely due
to lower charge density resulting from the lower pKa of its imidazole units (pKa 6.5).137,139
Additionally, cytotoxicity studies identified PEI as being seven-fold more toxic than
chitosan, whereas protamine exhibited only minimal cytotoxicity.138,140 Another polycation
that has been recently investigated for the oral administration of drugs is the cell-penetrating
peptide penetratin. When co-administered with insulin to diabetic mice, it significantly
reduced glycemia, but the effect remained low compared to that achieved by the parenteral
route.141 Stability experiments performed with rat intestinal fluids showed a difference
between the two isomers, with D-penetratin being the more stable form. Overall, more
detailed toxicity and permeability studies of the polycations need to be performed in GI
models in vitro and in vivo in order to fully understand the impact of these co-administered
polycations in an oral drug formulation.
29
Chapter 2: Improving oral drug bioavailability with polycations?
2.3. Polycationic micro- and nanoparticles
A successful intestinal absorption enhancer needs not only to enhance transport through
the intestinal epithelium, but also guarantee that the therapeutic agent remains intact until
it reaches this location. Although solutions of polycations co-administered with model
drugs have been shown to slightly improve the paracellular and/or transcellular uptake of
hydrophilic compounds (Figure 2.1D), little or no stabilizing effects have been
observed.114,118,123 Among the different systems that have been explored to stabilize drugs
from digestion, nano- and microparticles based on polycationic systems are particularly
promising. These can be conveniently prepared by electrostatic complexation with
(poly)anionic drugs (Figure 2.2A). A variety of strategies to prepare particle formulations
have been developed throughout the years including ionotropic gelation,142–144
condensation,145 emulsion/solvent evaporation,146 coacervation/precipitation,147,148 or
chemical crosslinking.146 Incorporation of drugs into nano/microparticles can sterically
shield them from enzymes and protect from pH changes by providing a more stable internal
chemical microenvironment (Figure 2.1B). Furthermore, such systems also provide the
means for enhancing intestinal drug transport via the mechanisms discussed in the previous
section, and via microfold cells (M cells) found in the follicle-associated epithelium of the
Peyer's patches.149 This section describes the characteristics of particle formulations, the
mechanism of permeation, the role of M cell in the uptake process and therapeutic
approaches.
2.3.1. Characteristics and mechanism of uptake
The characteristics of drug-loaded particles are influenced by several physicochemical
parameters. The molecular weight of the polycation can affect particle diameter and the
rate of drug release (through particle disassembly) (Figure 2.1C). For instance, a decrease
in size of chitosan yielded in smaller particles when proteins or plasmid DNA were
complexed.148,150 Correspondingly, high molecular weight chitosan complexed more
efficiently polynucleotides and negatively charged peptides/proteins due to larger scale
cooperative interactions. However, its lower solubility (vs. shorter chains) led to the
formation of larger particles. Additionally, higher polymer concentration led to an increase
in size, but also to enhanced loading efficiency.148 With regard to TMC, the DQ significantly
30
Chapter 2: Improving oral drug bioavailability with polycations?
influenced particle size, zeta potential, drug loading, and release rate.151 Higher DQ resulted
in smaller particles due to increased charged density leading to greater complexation of
insulin and a slower drug release rate.151
In contrast to polynucleotides, not all proteins exhibit a net negative charge over a
broad pH range and thus, nanoparticle preparation with polycations is not necessarily
possible. Helper anions that can complex the polycation are therefore employed to promote
the formation of stable particles and can significantly influence their features. For example,
polyanion hydroxypropyl methylcellulose phthalate or the smaller tripolyphosphate have
been used to complex chitosan, and incorporate insulin.152 Hydroxypropyl methylcellulose
phthalate, the macromolecular counter ion, formed larger particles, but led to significantly
higher insulin entrapment and protection against degradation in simulated gastric fluids.
Furthermore, higher loading efficiency is generally achieved when the size of the entrapped
protein decreases.148 Protein size is, however, not the sole factor, as the charge of the
incorporated drug dictates the level of interaction with the other components of the
particle. Thus, the isoelectric point of the loaded drug has a determinant role as it regulates
the drug’s overall charge which depends on the pH of the surroundings.
Various strategies have been developed to increase the mucoadhesive properties of
polymers. In addition to electrostatic interactions between polycations and mucin, van der
Waals forces, hydrogen bonding, and hydrophobic interactions can strengthen
mucoadhesion. Microparticles prepared with oleoyl-modified chitosan were slightly more
mucoadhesive than chitosan particles on rat intestinal tissue, probably due to the
contribution of hydrophobic interactions between the alkyl chains and the hydrophobic
components of the mucus glycoprotein.153 Moreover, chitosan nanoparticles modified with
up to 15% 3,4-dihydroxy-L-phenylalanine also exhibited improved mucoadhesion to rabbit
intestinal mucosa. Although the key component of 3,4-dihydroxy-L-phenylalanine,
catechol, has been identified to improve mucoadhesion of polymers, the mechanism
remains unknown, though is most likely of non-covalent nature.144 Thiolation has been an
additional modification made to polycations to improve their mucoadhesive properties.
Thiolation promotes adhesion by formation of disulfide bridges with cysteine-rich mucin
glycoproteins.151,154 Cationic thiolated nanoparticles (100–200 nm) prepared with TMC
bearing cysteines were shown to improve mucoadhesion by 2–4-fold ex vivo compared to
31
Chapter 2: Improving oral drug bioavailability with polycations?
unmodified TMC.151 Furthermore, thiolation enhanced particle stability via formation of
intra-particle disulfide bonds, and contributed to increased absorption by inhibiting the
intestinal P-glycoprotein (P-gp) and protein tyrosine phosphatase, the latter being involved
in the opening and closing of the tight junctions.151,155,156 However, strong interaction with
the mucus could inhibit the binding to the cell membrane leading to a decreased cell uptake.
Thus, Perera et al.157 reported a strategy where a neutral to slightly negative zeta potential
of nanoparticles was used to reduce interactions with the mucus. Once the particles have
passed the mucus layer, the zeta potential turned positive by reacting with brush border
membrane enzymes (i.e., alkaline phosphatase) leading to a higher uptake by intestinal
cells.157 This represents an interesting approach, but follow up studies need to clarify its
impact in vivo.
In the GI tract, M cells appear to be the cells being most involved in particle
transport as they possess high transcytotic activity, a thinner mucus layer, and irregular
microvilli (Figure 2.1D).158 Indeed, particles smaller than ~5–10 µm are preferably taken
up by M cells and transported into Peyer’s patches where they can degrade and release the
encapsulated drug in the lymphoid tissue.159 Immunohistochemistry analysis on ex vivo
tissues of mice orally gavaged with ovalbumin-loaded chitosan microparticles showed their
ability to transport their content into the Peyer’s patches in vivo, confirming the involvement
of M cells in chitosan particle GI absorption.160,147 Although M cells are more prone to
internalize cationic particles than epithelial cells, they only represent ~10% of all cells in the
intestine (human and mouse). However, the relative contribution of these cells towards
total absorption is not available.149 Uptake of cationic nanoparticles by epithelial cells has
been reported in a number of in vitro studies, but it is unclear how relevant this phenomenon
is in vivo (aside from an immunological perspective), especially in primates.151,161 Using
thiolated fluorescently-labeled TMC nanoparticles, it was found that particle absorption
occurred both in the Peyer’s patches (M cells) and in the non-follicular mucosa
(enterocytes) in rats, although at a slower rate.151 Roy et al.161 demonstrated in gene delivery
experiments that chitosan nanoparticles (150–300 nm) transfected mouse intestinal and
gastric cells after oral administration. However, the relative contribution towards the
internalization of the nanoparticles by the different intestinal cell types was not directly
investigated. Indeed, in another study performed in rabbits and where chitosan
nanoparticles combined with an endosomolytic peptide were instilled directly in the small
32
Chapter 2: Improving oral drug bioavailability with polycations?
intestine or colon, little reporter gene expression was seen.150 Interestingly, some expression
in the mesenteric lymph node was observed supporting particle uptake in the Peyer’s
patches. Further work is therefore needed to more precisely assess the contribution of both
cellular uptake pathways in the trancytosis of nanoparticles.
2.3.2. Nucleic acids
The successful oral delivery of nucleic acids requires their safe transport through the
stomach/gut, efficient cell uptake, DNA/RNA release, and gene expression within the cell.
When the target cell population lies outside the GI tract, then the delivery process becomes
very challenging because the particles must permeate intact through the intestinal
epithelium in order to reach distal tissues. Unfortunately, there are relatively few studies
describing the use of polycations as oral delivery agents for nucleic acid drugs.162 This might
be caused by low levels of transgene expression in such systems and the complexity of the
challenges that need be overcome.
2.3.2.1. Therapeutic genes and gene vaccines
Nucleic acids are increasingly used especially in gene therapy and gene vaccine delivery to
treat a variety of diseases. Polycations are particular interesting for the oral delivery of genes
as alternatives to traditional viral vectors because they are less immunogenic and can readily
complex and condense DNA. Thus, Chen et al.163 have prepared chitosan–DNA
nanoparticles (70–150 nm) containing plasmid DNA encoding for murine erythropoietin,
a glycoprotein stimulating red blood cell production. The oral delivery of naked
erythropoietin to mice did not cause a change in hematocrit, whereas orally-administered
nanoparticles produced an 18% increase, indicating enhanced erythropoietin secretion.163
Nevertheless, the intramuscular injection of a viral vector carrying erythropoietin produced
long-term and high-level erythropoietin expression, which has yet to be matched by oral
delivery (Figure 2.5).164 Furthermore, plasmid DNA–chitosan nanoparticles (300 nm)
encoded for factor VIII (FVIII) were orally administered to mice suffering from hemophilia
A, a chronic disorder of the blood coagulation cascade caused by defective FVIII.165
Plasmid DNA was detected by real-time polymerase chain reaction in various tissues
including the stomach, ileum, Peyer’s patches, liver, and spleen, with no difference between
the naked and encapsulated DNA. The authors provided two hypotheses to rationalize
33
Chapter 2: Improving oral drug bioavailability with polycations?
these findings. To administer the formulations, they used a gelatin base, which can act as a
protective agent and thus, stabilize naked DNA. Furthermore, the assay used to determine
the amount of DNA could not distinguish between degraded and intact DNA. However,
proof of protection through encapsulation and improved transfection efficacy was
demonstrated by higher FVIII protein production and enhanced phenotypic bleeding
correction within one month. Unfortunately, the FVIII levels achieved remained very low
(1–4%).165
Figure 2.5. Erythropoietin expression after oral delivery and intramuscular injection of DNA.
Hematocrit was measured in mice after the oral delivery of erythropoietin encapsulated in chitosan
nanoparticles (green bars, n = 9) or after intramuscularly injection of erythropoietin (blue bars, n = 2). Both
formulations contained doxycycline (200 µg/mL), an antibiotic that serves as a transcription activator. Oral
delivery led to a short term increased hematocrit count that decreased again after day 4, whereas a long-term
effect was observed after injection. Mean ± SD .163,164
Plasmid DNA encoding for a peanut allergen was loaded into chitosan nanoparticles
and then delivered orally to mice in order to treat peanut allergy.161 The level of tissue
expression of bacterial β-galactosidase was higher with nanoparticles compared to naked
DNA in both the stomach and small intestine, suggesting protection and increased transfer
to the epithelium. Moreover, the encapsulated DNA induced a protective immune response
in a peanut allergy mouse model in the form of increased antibody levels (IgA, IgG2a) and
less anaphylaxis in response to peanut challenge. Similar results were obtained by Chew et
al.166 who orally administered plasmid DNA encoding for native dust mite allergen Der p
34
Chapter 2: Improving oral drug bioavailability with polycations?
1 in chitosan nanoparticles to mice. An immune response was observed, while the
intramuscular injection of the DNA alone was ineffective.166 In both studies the oral
administration of naked DNA did not induce an immune response. These proof-ofconcept studies provided the impetus for more advanced work including multiple
sensitization and kinetic studies.
GRA1 DNA vaccination serves as a protective action against the infection with
Toxoplasma gondii, which can cause toxoplasmosis, a parasitic disease. Chitosan nanoparticles
loaded with GRA1 plasmid DNA and recombinant GRA1 protein were administered by
gavage to mice.167 Despite the fact that the nanoparticles produced an immune response, it
was not specifically protective against T. gondii. As the reported particles exhibited very high
GI stability, the gene might have been inefficiently released. Thus, it would be interesting
to gain a better understanding whether mucosal administration of encapsulated GRA1 can
cause a specific local immune response by altering the structure of the system to more
effectively release the DNA.167 Moreover, Negash et al.168 prepared cationic microparticles
with the biodegradable and biocompatible polymer poly-(DL-lactide-co-glycolide)169 in
combination with PEI170, and subsequently absorbed plasmid DNA. The DNA vaccine
encoded the virus protein 2 targeting the infectious bursal disease virus (IBDV) that causes
highly contagious viral infection in young chicken. It could be shown that the desired DNA
vaccine was immunogenic and partially protective against viral challenge.
2.3.2.2. Therapeutic siRNA
A very attractive therapeutic strategy to overcome inflammatory conditions such as
Crohn’s disease in the GI tract is the use of small interfering ribonucleic acid (siRNA) to
knockdown pro-inflammatory cytokines.171 However, RNA faces the same challenge as
DNA when orally administered. Additionally, complexes prepared with siRNA are
generally more prone to dissociate in physiological media than those obtained with plasmid
DNA due to the lower molecular weight of siRNA.172,173 Laroui et al.174 succeeded in
delivering a siRNA in vivo against tumor necrosis factor α (TNFα), a major proinflammatory cytokine that is produced by macrophages and dendritic cells in inflammatory
bowel diseases. The nucleic acid was complexed with PEI or chitosan, loaded into poly(D,L
lactide) (PLA) nanoparticles (~400 nm), and the latter sterically stabilized with poly(vinyl
alcohol). Particles prepared with PEI afforded greater protection than chitosan against
35
Chapter 2: Improving oral drug bioavailability with polycations?
degradation by RNases and allowed higher loading in the nanoparticles. These particles had
an overall slightly negative zeta potential. However, during the GI transit, the degradation
of the PLA matrix can occur,175,176 and eventually lead to the release of positively-charged
PEI/siRNA complexes. After oral administration to mice with lipopolysaccharide-induced
inflammation, TNFα concentration was reduced in the colon but not in the liver,
demonstrating the restricted delivery of the siRNA to the colonic tissue.174 Based on the
same system, CD98 siRNA was complexed with PEI, loaded into PLA nanoparticles (~450
nm), and delivered to mice with dextran sodium sulfate-induced colitis.175 The siRNA was
selected to knockdown colonic CD98, which is overexpressed in inflammatory bowel
diseases.177 It was shown that siRNA CD98-loaded nanoparticles (~500 nm) decreased
colitis and down-regulated CD98 in intestinal macrophages, and more surprisingly in the
intestinal epithelial cells.
Recently, He et al.173 encapsulated siRNA in thiolated TMC nanoparticles through
ionic gelation using sodium tripolyphosphate to mediate TNFα knockdown in
macrophages. Longer polymeric chains resulted in higher stability towards RNA
degradation tested in mouse serum and simulated GI tract fluids. In a mouse inflammatory
model, an optimal balance between stability and ability to release the therapeutic could be
achieved with 200 kDa TMC. In a subsequent study, thiolated TMC was complexed with
antitumor small hairpin RNA-expression plasmid DNA and siRNA that induce apoptosis
and inhibit angiogenesis, respectively.178 The presence of the plasmid DNA permitted the
formation of more compact particles (130–160 nm). The nanoparticles were decorated with
galactose residues to enhance uptake by hepatoma cells after systemic absorption. In vitro
stability tests suggested that the particles would protect the nucleic acids in the GI tract and
the bloodstream. After oral gavage to mice bearing subcutaneous hepatic tumors, the
nucleic acids encapsulated within both galactose-decorated and non-decorated
nanoparticles were deposited in the tumoral tissue and inhibited tumor growth.178 While
promising, these data should be viewed with caution as the tumoral levels of DNA reached
after oral intake were in the same order of magnitude as those achieved after parenteral
administration of sterically stabilized nanovectors, despite the numerous hurdles
encountered in the GI tract.179 It would be interesting to examine the generality of this
observation in different tumor models.
36
Chapter 2: Improving oral drug bioavailability with polycations?
2.3.3. Proteins, peptides and low molecular weight drugs
Apart from polynucleotides, cationic particles have been investigated as a means to deliver
proteins, peptide, and small drugs. For example, tetanus toxoid vaccine was developed in
1890 and gave high hope in fighting tetanus, a disease of the nervous system which leads
to painful muscle contractions. In the Third World, mass vaccination via oral administration
could help prevent new infections. To this end, Harde et al.180 have encapsulated the tetanus
toxoid into chitosan nanoparticles by ionic gelation with tripolyphosphate followed by
covalent cross-linking using glutaraldehyde. Moreover, glucomannosylation was expected
to facilitate particle uptake by antigen presenting cells due to high density of mannose and
glucose receptors on their surface. In vivo evaluation after oral delivery to mice showed that
naked tetanus toxoid failed due to poor protection in the GI tract and low level of uptake
by antigen presenting cells. Inversely, the nanoparticle vaccine led to a protective immune
response.180 In vivo challenge should be examined in additional animal species for a broader
view of the applicability of the concept. Moreover, oral delivery of chitosan microparticles
loaded with diphtheria toxoid increased local (IgA) and systemic (IgE) immune response in
mice.181 Even though the systemic immune response was increased by 125-times when
encapsulated diphtheria toxoid was administered compared to the negative control
(diphtheria toxoid only), the dose–response relationship stated that a four-fold higher dose
was needed to obtain an immune response comparable to that achieved after subcutaneous
injection.
The oral administration of insulin is an intensively studied topic to address the
inconvenience of frequent injections in the management of diabetes. The oral delivery of
chitosan particles loaded with insulin to diabetic rats led to a significant decrease (9.8-fold)
in blood glucose levels compared to insulin dissolved in a chitosan solution.142,143,152
Additionally, the hypoglycemic effect of insulin encapsulated in TMC microparticles
following oral intake was improved when the latter was functionalized with cysteine.151
Moreover, PAMAM G2–chitosan conjugates were prepared to encapsulate insulin in
polycationic nanoparticles.182 After oral administration to diabetic mice, the bioavailability
of insulin encapsulated in the polymer conjugate was found to be ~9% whereas delivered
insulin chitosan nanoparticles led to a bioavailability of ~5%. However, despite these
promising results which are reminiscent of other in vivo data obtained on rodents with non37
Chapter 2: Improving oral drug bioavailability with polycations?
cationic delivery systems,183 the oral bioavailability of insulin remained low, showed high
variability throughout the different in vivo studies, and may not be predictive of efficacy in
humans.184
Dendrimers, with their highly branched architecture, offer a good platform for
encapsulation and has been therefore intensively used to enhance the solubility and
bioavailability of drugs. In general, they possess hydrophilic surfaces and hydrophobic cores
leading to non-covalent interaction with drugs (e.g., hydrophobic and/or electrostatic
interactions).185 Drug complexation with polycationic dendrimers was however seldom
investigated, and only a few low molecular weight drugs such as doxorubicin, 7-ethyl-10hydroxy-camptothecin, and ketoprofen have been formulated with PAMAM (G3, G4, and
G5, respectively).186–188 In vitro studies conducted with Caco-2 cells showed an enhanced
cell uptake,188 but poor stability at acidic pH was obtained, especially regarding encapsulated
ketoprofen and 7-ethyl-10-hydroxy-camptothecin, being necessary for the oral delivery of
drugs.186,187
It has been previously described that partial shielding of positive charges can lower
the cytotoxicity of polycations. Thus, chitosan chemically modified with PEG (5 kDa) has
been used to formulate salmon calcitonin (protein used to treat hypocalcemia) in
particles.189 PEGylation decreased chitosan’s cytotoxicity by 10–20-fold and increased the
stability of the complexed peptide in simulated GI fluids. After oral gavage to rats, it was
shown that the hypocalcemic effect was inversely related to the degree of PEGylation
(Figure 2.6). This effect might be caused by the partial shielding of the positive charges of
chitosan that can reduce the electrostatic interactions with the mucosa. While the
PEGylated nanocapsules with 0.5% PEG–chitosan performed better than the control
calcitonin solution, their efficacy did not differ from that of the non-PEGylated
nanocapsules.189,190
Recently, Ma et al.191 modified PAMAM G5 with PEG (8%, 5 kDa) to improve
solubility, transepithelial transport, and the therapeutic effect of probucol, a hydrophobic
anti-hyperlipidemic drug. Liposomes which were modified with the PEGylated PAMAM
and orally administered to LDLR-/- mice suffering from hypercholesterolemia were found
to improve the effect of probucol on lowering the cholesterol and triglyceride plasma
levels.191 While these results are potentially interesting, this study did not provide sufficient
38
Chapter 2: Improving oral drug bioavailability with polycations?
information of the preparation of the control formulations to adequately assess the benefits
of this relatively complex system.
Figure 2.6. Hypocalcemic effect after oral administration in rats of calcium calcitonin in aqueous
solution, encapsulated in chitosan–PEG 0.5% and 1%. Dose of salmon calcitonin is 500 IU, mean ± SD (n
= 6). Reproduced from Prego et al.189 with permission.
39
Chapter 2: Improving oral drug bioavailability with polycations?
2.4. Polycation–drug conjugates
Drugs can be covalently bound, rather than electrostatically complexed, to polycations to
create a produg (Figure 2.1A). This can be achieved by direct conjugation to the polymer
backbone, or via a linker molecule. The desired features of a polymer–drug conjugate often
are the same as for the systems described above, namely, protect the drug from degradation,
increase the aqueous solubility of hydrophobic drugs, increase bioavailability, possess
minimal non-specific toxicity, and potentially provide mechanisms for drug targeting.
Additionally, the conjugate can also prolong the residence time of the drug in the GI tract
due to its mucoadhesive character and thus, enhance the overall uptake, as for the
particulate systems in the previous section. Furthermore, release of the drug can be
controlled by incorporating chemically or enzymatically degradable linkages between the
drug and polymer (Figure 2.1C).192,193
Dendrimer–drug conjugates composed of PAMAM G3 and 5-aminosalicylic acid, a
topical anti-inflammatory drug, have been prepared with azo linkers. Azo bonds are cleaved
by azoreductases in the colon, which provokes drug release (or reconversion) specifically
at this location. In vitro analysis indicated that the prodrug possessed the required pH and
gastric stability leading to a stable prodrug formulation under these conditions. Drug
targeting was successful, given that prodrug reconversion under colonic conditions
gradually increased over time (23–38%), whereas only small amounts were found in the
small intestinal homogenate. Additionally, differences in drug release could be obtained by
altering the chemical structure of the linker.192 For instance, naproxen, a poorly water
soluble drug, was conjugated to PAMAM G0 via different linkers (amide and ester bonds)
leading to different pH, plasma, and liver stabilities.193,194 An amide linkage provided very
high chemical and enzymatic stability, leading unsufficient drug release. Inversely, a lactic
acid ester linkage was very stable under GI conditions, and showed a slow release of
naproxen from the conjugate in human plasma and rat liver homogenate.193,194 It was found
that PAMAM increased the permeation of naproxen across Caco-2 cell monolayers,
especially when the conjugate further bore a lauroyl chain.
Known substrates of P-gp (efflux transporter that reduces the oral absorption of
several drugs) such as propranolol and terfenadine have been covalently conjugated to
PAMAMs G1 and G3.195,196 The polycationic dendrimer enhanced paracellular and
40
Chapter 2: Improving oral drug bioavailability with polycations?
transcellular transport, and the dendrimer was shown to be neither a P-gp inhibitor nor a
P-gp substrate.197 The conjugation increased the solubility of the compounds and helped
bypass the efflux system in vitro.195,198 In addition, the conjugation of drugs to PAMAM
dendrimers might also reduce the latter’s toxicity as it decreases its overall positive charge.
This was seen for lauroyl or PEG chain conjugation to PAMAM leading to a partial
shielding of its charged character, which is the reported main cause of its toxic effect.197
After drug release, the dendrimer is no longer shielded, but may potentially exhibit lower
toxicity because it becomes diluted.
One major drawback of several anticancer drugs, such as paclitaxel and docetaxel, is
their poor water solubility. Lee et al.199 have conjugated paclitaxel to a low molecular weight
chitosan (6 kDa), which significantly increased its aqueous solubility. Moreover, only a low
amount of paclitaxel was released under simulated gastric conditions, while most of it was
released in rat plasma. The conjugate was tested in a murine melanoma-bearing B16F10
mouse model. It showed a 5-fold higher tumor uptake compared to free paclitaxel injected
intravenously. The oral bioavailability of paclitaxel ranged between 20 to 85%, and was
found to be independent of P-gp inhibition. An experiment performed with an
125I-
radiolabeled polymer indicated substantial absorption of the conjugate. Although it cannot
be excluded that part of the measured radioactivity corresponded to released 125I (thyroid
levels were not measured in this work), additional data generated with a fluorescentlylabeled conjugate indicated some translocation in the ileum. A similar study was conducted
with docetaxel, where the chitosan–drug conjugate exhibited an even higher oral
absorption.200 In both cases, the antitumoral efficacy of the orally-administered conjugates
was similar to the intravenously injected drugs, despite important differences in the
pharmacokinetics and biodistribution profiles. It may be interesting to verify the generality
of this phenomenon by repeating these experiments in other animal models. In another
study, a low molecular weight chitosan (9 kDa) was conjugated to exendin-4 (4 kDa) (a
glucagon-like peptide-1 mimetic for the treatment of type 2 diabetes) with a disulfide
bond.201 The prodrug formed nanoparticles (~100 nm) due to electrostatic complexation
between the positively charged chitosan and the negatively charged peptide. This led to
increased stability towards digestion by trypsin. In vitro testing on INS-1 pancreatic β-cells
to evaluate the amount of glucose-induced insulin secretion indicated that the conjugate
retained its original biological activity. In vivo studies, conducted with a diabetic mouse
41
Chapter 2: Improving oral drug bioavailability with polycations?
model, suggested that conjugation to chitosan resulted in a relative bioavailability of 6.4%
vs. the subcutaneous route, and enhanced hypoglycemic efficacy after oral administration
compared to the free peptide.201 Unfortunately, it was not possible to evaluate the role of
the chemical linkage between the peptide and chitosan in the uptake process as control
particles consisting of only the physical complex between the two macromolecules were
not analyzed. In another study, similar data were also reported with an insulin–chitosan
conjugate.202
Polymer conjugation can also be used to increase the stability of proteins that act in
the GI tract.203 For example, Fuhrmann et al.93 recently showed that the binding of a
positively charged dendronized polymer PG1 to a prolyl endopeptidase derived from MX
prolonged the retention of the enzyme in the stomach and preserved its catalytic activity
for several hours (Figure 2.7). Such an enzyme is currently being explored as a means to
degrade the immunogenic peptidic sequences of gluten and could therefore be used as a
supportive therapy for celiac disease.204
42
Chapter 2: Improving oral drug bioavailability with polycations?
Figure 2.7. In vivo activity of MX PEP and MX–PG1 conjugate. A fluorescence-quenched model
peptide was orally administered to rats. Cleavage of the peptide by co-administered enzymes (MX) or
enzyme–polycation conjugate (MX–PG1) restores fluorescence, should they remain active in the GI tract.
Images show the evolution of the fluorescent signal throughout the GI tract. MX is rapidly deactivated and
no fluorescence is observed. MX–PG1 remains active in the stomach. A precleaved (and thus fluorescent)
peptide serves as positive control in the first column. Reproduced from Fuhrmann et al.93 with permission.
43
Chapter 2: Improving oral drug bioavailability with polycations?
2.5. Concluding remarks
There is now sufficient evidence that polycations have the ability to increase the
transepithelial passage of drugs and biomacromolecules. The uptake pathways seem to be
polymer- and system-dependent (soluble macromolecules vs. nano/microparticles vs.
prodrug). Increased bioavailability is mainly achieved by promoting paracellular passage,
though some systems may also permit transcytotic transport. Despite the numerous in vivo
studies that have reported an increase in oral bioavailability, it still remains unclear to what
extent this effect is sufficient to provide the necessary plasmatic drug concentrations for
pharmaceutical development. It is our opinion that for class III drugs (high solubility, low
permeability) only highly potent and relatively stable compounds having a high therapeutic
index or vaccine antigens could benefit from oral delivery using polycations. Regarding
nucleic acids, it is reasonable to assume that transfection would be mainly restricted to the
cells of the GI tract, which could prove useful in several diseases of the colon. For class II
drugs (low solubility, high permeability), polycations may help to improve bioavailability by
enhancing solubility, but to maximize bioavailability, the system may need to be designed
to release the low molecular weight active compound in the GI lumen before absorption
takes place. One aspect that should be investigated in more detail is the impact of prolonged
exposure to polycations, especially when the latter are non-digestible/degradable in the GI
tract. Indeed, sustained exposure to polycations may expose the organisms to antigenic
substances from the GI lumen, which would otherwise not cross the mucosa. Moreover,
some polycations may also possess bacteriostatic or bactericidal activity and consequently
may alter the gut flora or induce physiological changes. These effects are more delicate to
identify but will provide essential information for the successful development of
polycation-based drug delivery systems.
44
Chapter 3
Engineering of two prolyl endopeptidases
Part of this chapter is published in:
Schulz, J. D., Patt, M., Basler S., Kries H., Hilvert, D., Gauthier, M. A., Leroux, J.-C. Adv.
Mater. 2015; DOI: 10.1002/adma.201504797.
Chapter 3: Engineering of two prolyl endopeptidases
3.1. Introduction
P
ROTEIN engineering is a broad scientific field that has developed methods to
study or modify proteins, in particular to understand protein folding, enzymatic
catalytic processes and/or impair proteins’ properties.205 Among the existing
approaches, rational design and directed evolution are well-established strategies. Thereby,
site-directed or random mutagenesis is performed to engineer proteins, respectively.206 By
modifying amino acids it is possible to identify essential residues for the enzymatic function
or folding. Thereby, the enzyme’s mechanisms can be established or its activity enhanced
by improved substrate cleavage in the binding pocket.207 Additionally, based on known
scaffolds, computational design can develop artificial biomolecules like non-natural
occurring biocatalysts.92,208,209 Usually this approach is performed in combination with sitedirected mutagenesis to specifically enhance the enzymes’ characteristics due to the often
low initial activity of the starting design.207 Besides focusing on engineering proteins’
properties by mutagenesis, point mutations can also introduce reactive groups on the
surface of proteins which can be coupled to other molecules. Indeed, the site-specific
attachment of polymers to engineered residues of proteins constitutes a very attractive
strategy to improve their properties, e.g., prevent aggregation or enhance in vivo stability.210–
212
In order to introduce promising conjugation sites, well-chosen residues of the
protein are modified to either natural or unnatural amino acids containing side chains that
are prone to react chemically with polymers. Indeed, a variety of techniques have been
published to introduce unnatural amino acids to a protein sequence.213 Thereby, one of the
most popular techniques to couple polymers via unnatural residues is click chemistry,
whereas the side chains often contain alkyne and azide functional groups. In addition, other
groups such as aryl halide, redox-active, and photo-reactive groups have been introduced
as functionalities.213 Engineering proteins with unnatural amino acids is a promising
scientific approach including drug development but has not found yet its way to the
market.214,215 On the other hand, modifying residues to natural amino acids is a very
common and well-established method. In this case the amino acid that is placed on the
protein’s surface should be rather rare and/or naturally located in non-accessible locus of
the protein in order to avoid competitive reaction with the polymer. Additionally, the amino
47
Chapter 3: Engineering of two prolyl endopeptidases
acid should contain a reactive group for polymer attachment. Cysteine’s thiols are the
strongest nucleophilic group upon the side chains of amino acids and the proportion of
cysteines in proteins is generally rather low. Cysteines are often essential for protein
structures as they can build di-sulfide bridges and are therefore mostly not found on the
surface of proteins. Hence, modifying proteins’ residues to cysteine is the most commonly
used approach to create conjugation sites for the site-specific attachment of polymers.
This strategy offers an interesting approach to stabilize enzymes for therapeutic
applications. Bacterial PEPs are enzymes that hydrolyze peptides at proline residues. They
have the ability to degrade immunogenic peptides from gluten which are triggers for celiac
disease, and have thus been evaluated as an oral adjuvant therapy.216 However, due to the
harsh conditions in the GI tract and the lack of information that currently exists regarding
the stabilization of orally administered therapeutic enzymes, the potential of PEPs as
therapeutic agents for celiac disease has not yet fully unfolded. The site-specific polymer
attachment to engineered residues of enzymes that are orally administered has not been
tested so far, but offers a promising approach.
To this end, detailed knowledge of the protein’s 3D structure is required in order to
identify potential modification sites on which to carry out protein engineering. Fortunately,
the crystal structures of the two PEPs, MX and SC, have been solved by Shan et al.82 Both
enzymes share high similarities in their structure and are characterized by a catalytic domain,
whose activity is gated by a second, β-barrel propeller domain (Figure 3.1). Aside from the
two linear strands covalently connecting the domains, the specific enzyme structure is kept
solely by domain–domain interactions. It is proposed that peptidic substrates induce
conformational changes that lead to the opening of the domain interface. Following
substrate entry into the binding pocket, the complex is temporarily stabilized by noncovalent interactions between the catalytic domain and the peptide.82 Figure 3.1 shows MX
crystallized with a bound inhibitor and SC in an unoccupied form.
The engineering of the two PEPs is described in the following chapter. Thereby,
residues of the surfaces of the PEPs were modified in order to strategically place free and
available thiol groups that would serve as sites for site-specific polymer conjugation. The
strategy of PEPs’ engineering and the activity of the expressed mutants is presented and
discussed.
48
Chapter 3: Engineering of two prolyl endopeptidases
Figure 3.1. 3D structures of the PEPs illustrated with PyMOL. a) MX with the bound inhibitor Z-Alaprolinal shown as sticks in red and b) SC in an unoccupied conformation.82
49
Chapter 3: Engineering of two prolyl endopeptidases
3.2. Materials and Methods
All chemicals were purchased at the highest possible analytical grade, and used as received.
Site-directed mutagenesis: The gene encoding MX (Zedira, Darmstadt, Germany) and
SC (Prozomix, Northumberland, UK) were amplified using the polymerase chain reaction
(PCR) with the outer primers T7short and pET-T7up. Primers were purchased from
Microsynth AG (Balgach, Switzerland) (Table 3.1). Site-directed mutagenesis was
performed by overlap extension to mutate the residues to cysteine. The PCR products of
MX were digested with Nde I and Hind III and the products of SC with Nde I and Xho I
(20,000 U/mL, Biolabs® Inc., Ipswich, MA) and cloned into digested plasmid pET_28b
(Novagen, Schaffhausen, Switzerland). Constructs were confirmed by DNA sequencing
(Figures A1–A4), performed by Synergene (Schlieren, Switzerland).
Expression and purification of MXs and SCs: Wild-type enzymes and mutants were
expressed as recombinant proteins with C-terminal His6 tags in Escherichia coli
BL21(DE3).79,217 More specifically, inoculum was grown at 37 °C until the OD600 reached
0.6 and then protein expression induced with isopropyl β-D-1-thiogalactopyranoside
(0.1 mM, PanReac AppliChem, Darmstadt, Germany). The culture was incubated overnight
at 22 °C for MX and 18 °C for SC, and cell lysis and purification steps were performed at
4 °C. Cell culture suspension was centrifuged at 10,000 × g for 20 min and the cell pellet of
MX re-suspended in wash buffer (imidazole (15 mM), sodium phosphate (50 mM), sodium
chloride (200 mM), tris(2-carboxyethyl)phosphine (TCEP, 1 mM), pH 8). Bacteria were
again pelleted at 10,000 × g for 20 min and re-suspended in wash buffer (3 mL) per gram
of pellet. The cell pellet of SC was not washed, but directly re-suspended in 3 mL per gram
of pellet in imidazole (5 mM), benzamidine (1.5 mM; Acros, Geel, Belgium), pepstatin A (2
mg/L; Fluka, Buchs, Switzerland), leupeptin (2 mg/L; Sigma, Buchs Switzerland), Tris-HCl
(50 mM), sodium chloride (0.2 M), TCEP (1 mM), pH 8.5. Cell lysis was performed by
sonication (Sonicator UP200H, Dr. Hielscher GmbH, Teltow, Germany) and the solution
clarified by centrifugation (30,000 × g for 45 min). The supernatant was loaded onto a NiNTA resin column (QIAGEN GmbH, Hilden, Germany) and incubated for a minimum
of 2 h.
50
Chapter 3: Engineering of two prolyl endopeptidases
Table 3.1. DNA sequences of primers used for site-directed mutagenesis. The codon encoding
cysteine is highlighted in red. The blue background shows primers used for MX and the red highlighted part
primers used for SC mutagenesis experiments.
Primer
Primer sequence (5’→ 3’)
T7short
TAATACGACTCACTATAGG
pET-T7up
CGGTGATGTCGGCGATATAG
A25C_for
GCAGCCATATGTCCTACCCGTGTACCCGGGCGGAGCAGGTTGTCG
A25C_rev
CGGGTAGGACATATGGCTGC
S123C_for
CCTCTACTGGCGCCAGGGGGAGTGTGGGCAGGAGAAGGTGCTGTTGG
S123C_rev
CTCCCCCTGGCGCCAGTAGAGG
A328C_for
GGATCCGGCGAAGCCGGCCCGCTGTTCGTGGAAGGAGATTGTCC
A328C_rev
GCGGGCCGGCTTCGCCGGATCC
S338C_for
GGAGATTGTCCCCGAGGACTGTTCCGCGTCCCTGCTGTCC
S338C_rev
GTCCTCGGGGACAATCTCC
T417C_for
CGTCAAATCTACAAGACGTCCGTCAGCTGTGGGAAGTCTGAGCTCTGGGCCAAGG
T417C_rev
GCTGACGGACGTCTTGTAGATTTGACG
K461C_for
CGTGGTGCACCGCAAGGACCTGTGTCGTGACGGCAACGCGCCCACG
K461C_rev
CAGGTCCTTGCGGTGCACCACG
A96C_for
CTCAAGCAGTTGCCCGAGCGT TGTGCGCTGGAAAAGCGGATGAAGG
A96C_rev
ACGCTCGGGCAACTGCTTGAG
S351C_for
CGTGCGGGTCGATCTGTCGGGCTGTACGCCGCGCTTCGATACC
S351C_rev
GCCCGACAGATCGACCCGCACG
T441C_for
GGTGCTGGCGCTCGATCCCGCCTGTGCGAAAACCACCCCGTGG
T441C_rev
GGCGGGATCGAGCGCCAGCACC
A560C_for
GCCGCGGGCGAATGGCTGATCTGTAATGGCGTGACGCCGCGCCACG
A560C_rev
GATCAGCCATTCGCCCGCGGC
S643C_for
GCTATTCGCCCTATCACAACGTGCGTTGTGGCGTGGACTATCCGGCCATCC
S643C_rev
ACGCACGTTGTGATAGGGCGAATAGC
51
Chapter 3: Engineering of two prolyl endopeptidases
After three washing steps with wash buffer (5 mM imidazole, 50 mM Tris-HCl,
0.2 M sodium chloride, 1 mM TCEP, pH 8.5 for SC), the protein was eluted with 12 mL
imidazole (150 mM), sodium phosphate (50 mM), sodium chloride (0.2 M), TCEP (1 mM),
pH 8.0 or imidazole (0.2 M), Tris-HCl (50 mM), sodium chloride (0.2 M), TCEP (1 mM),
pH 8.5 for MX and SC, respectively. Buffer exchange was performed with sodium
phosphate (50 mM), sucrose (3%), TCEP (1 mM), pH 7.5 for MX and sodium phosphate
(50 mM), sodium chloride (0.3 M), TCEP (1 mM) pH 8.0 for SC by ultrafiltration (Amicon
30 kDa MWCO, Schaffhausen, Switzerland). The quality of protein expression was
determined by sodium dodecyl sulfate polyacrylamide gel electrophoresis (SDS-PAGE,
10%) (Figures 3.2, 3.3). Protein concentration was measured by spectrophotometry
(NanoPhotometer® Pearl, IMPLEN, Munich, Germany) using its predicted extinction
coefficients at 280 nm.
In vitro activity: Enzymes were generally stored at 4 °C and used for activity assessment
from the third day onwards after protein expression. MX or MX–polymer conjugates
(0.5 µg/mL) were incubated at 37 °C with (benzyloxycarbonyl)-glycine-proline-pnitroaniline (2.5 µM, Bachem, Rorrance, CA) in sodium phosphate buffer (50 mM, pH 7.0).
Enzyme activity was calculated from the slope of the absorbance change at 410 nm due to
the release of nitroaniline, which was monitored for 5 min. The activity (slope) observed
with the commercially available MX WT was set to 100%.
Statistical analysis: The data were analyzed by parametric one-way ANOVA followed by
a Dunnett`s post hoc or Tukey`s post hoc test. Differences were considered statistical
significant at p < 0.05.
52
Chapter 3: Engineering of two prolyl endopeptidases
Figure 3.2. SDS-PAGE (10%) analysis for protein expression of MX PEP and affinity chromatography
stained with Coomassie blue for protein identification. Row identification: 1 molecular weight (MW) marker
in kDa; 2 supernatant of induced cells after sonication; 3 flow through after protein binding to Ni-NTA
resin; 4–6 wash steps 1–3; 7–13 elution fractions of the protein of interest a) MX3.1, b) MX3.2.
Figure 3.3. SDS-PAGE (10%) analysis for protein expression of SC PEP and affinity chromatography
stained with Coomassie blue for protein identification. Row identification: 1 molecular weight (MW) marker
in kDa; 2 supernatant of induced cells after sonication; 3 flow through after protein binding to Ni-NTA
resin; 4–6 wash steps 1–3; 7–14 elution fractions of the protein of interest a) SC3.1, b) SC3.2. Two protein
bands with similar size were expressed as main products. The sequence of SC contains two start codons
which are separated by 24 nucleotides. This leads to the expression of two SCs proteins with a difference
of eight amino acids.
53
Chapter 3: Engineering of two prolyl endopeptidases
3.3. Results and Discussion
The structural conformation of MX and SC and the proposed catalytic mechanism of the
PEPs suggest that extensive motions need to be involved in substrate binding and cleavage.
The two PEPs offer thus a challenging target for site-specific polymer conjugation as not
only blockage of the active site by the polymer can reduce the enzymatic activity, but also
interferences with their dynamics and shielding of amino acids on the surface being essential
for initial interaction with the substrate. To explore methods of protecting PEPs with
polymers without compromising their activity, locations for site-specific conjugation were
introduced in three successive rounds of mutagenesis. Enzyme mutants with one to three
engineered cysteine residues were produced in order to position free thiol groups on the
surface of the enzymes. The thiols were
then coupled to various polymers (vide
infra) to produce a series of bioconjugates
for analysis (see Chapter 4). However, a
strategy to retain the enzymatic activity
upon polymer conjugation and to not
change the folding of the PEPs upon
mutagenesis was essential. Firstly, mainly
residues that resemble cysteine in size (i.e.,
alanine or serine residues) were selected
for modification. The amino acids were
also not sterically hindered by neighboring
residues to ensure that the placed thiol
groups
were
available
for
polymer
attachment in order to facilitate polymer
coupling.
Figure 3.4
highlights
the
residues of MX chosen for mutagenesis Figure 3.4. Image sections of the 3D structure of
with residues being chosen for mutagenesis
experiments as red and blue sticks, MX
(MX3.1 residues in red, MX3.2 residues in blue) are
showing that no neighboring residues are presented with PyMOL. Amino acids are not sterically
hindered by neighboring residues increasing the
sterically shielding the amino acids. chance that these engineered cysteines are available
for polymer attachment. a) S123, b) S338, c) K461,
Moreover, the engineered cysteines should d) A25, e) A328, f) T417.
be located in a positively charged micro-
54
Chapter 3: Engineering of two prolyl endopeptidases
environment in order to stabilize the cysteine-polymer bond (Figure 3.5). Indeed, Shen et
al. showed that positively charged neighboring residues increase the stability of a thiol–
maleimide linkage used for coupling.218
Figure 3.5. Electrostatic charge distribution of the surface of MX. Charges are calculated with PyMOL.
Color code of the surface shows residues +1.4 kT/e blue and –1.4 kT/e red. The black framed box presents
the residues chosen for cysteine modification: a) S338, b) K461, c) S123, d) A25, e) A328, f) T417. A
negative environment of the conjugation site was specified to destabilize bioconjugates.218
Regarding MX, three areas of the enzyme were selected as sites for modification, so
that the grafted polymers would be roughly equidistant from one another (~4–7 nm) and
at opposite ends of the protein in order to cover most of its solvent-accessible surface
(Figure 3.6). Because the MX residues involved in substrate binding and the functionalities
important for conformational rearrangements (i.e., opening and closing) were not known a
priori, two sites were chosen for mutations in each region. The first area (“A” in
Figure 3.6a) is located on the catalytic domain (residues A25 and K461) opposite the
catalytic site, whereas the other two areas (“B”: S123, T417 and “C”: S338, A328) are on
the β-barrel propeller domain. The first generation of mutagenesis yielded six single
cysteine-mutated MXs (MX1.1–1.6; Figures 3.6a, b) that were expressed and assessed for
55
Chapter 3: Engineering of two prolyl endopeptidases
their activity. All six variants maintained the activity of MX WT (Table 3.2). Because no
loss in enzyme activity was obtained upon mutagenesis, two cysteine mutations were
combined in order to produce three double-mutants of MX (MX2.1–2.3, Figure 3.7a).
MX2.1 carries two cysteine residues on the β-barrel propeller domain and MX2.2 and
MX2.3 carry one cysteine residue each on the catalytic domain and the β-barrel propeller
domain. The doubly mutated enzymes were found to be as active as the MX WT control
(Table 3.2). To further probe that mutagenesis on the surface of MX does not affect the
activity, two triply-mutated forms of MX were produced by introducing cysteine residues
simultaneously in regions “A”, “B”, and “C” (Figures 3.6a, 3.7b). These maintained the
structure (Figure A5) and activity (Table 3.2) of MX WT. Hence, mutagenesis did not
impair MX activity leading to the successful development of two mutants each carrying
three cysteines on the surface that cover three different sites of the enzyme.
Figure 3.6. Engineering of MX. a) 3D structure of the closed (inhibitor-complexed) form of MX with
highlighted residues chosen for cysteine modification (MX1.1–MX1.6 in blue). The catalytic triad (Ser, Asp,
His) is depicted in yellow, and the two linear strands covalently connecting the catalytic and β-barrel
propeller domain of MX are shown in light black and indicated by an arrow. The three different regions of
MX (A, B and C) where surface-exposed cysteines were introduced are circled. The residues targeted for
mutation in each region are shown as spheres. The labeling of the residues consists of the original amino
acid and the location in the amino acid sequence. (PDB 2BKL).82 b) Nomenclature used for the conjugates
produced in the three rounds of mutagenesis. c) Distance of sites for polymer attachment. Residues chosen
for mutagenesis and thus, for site-specific polymer attachment are highlighted in red and blue (MX3.1 and
MX3.2, respectively). The distances were calculated with PyMol and are presented in Å (= 0.1 nm).
56
Chapter 3: Engineering of two prolyl endopeptidases
Table 3.3. Activity of MX mutants. The location
of the cysteine modification is presented with the
labeling of the constructs and their activity. Mean
± SD (n = 3).
Location
of mutation
Constructs
Activity [%]
± SD
None
MX WT
101 ± 2.50
S123C
MX1.1
79 ± 7.8
S338C
MX1.2
98 ± 1.8
K461C
MX1.3
91 ± 25
A328C
MX1.4
88 ± 2.0
A25C
MX1.5
82 ± 7.3
T417C
MX1.6
77 ± 10
S123C + S338C
MX2.1
96 ± 12
A328C + K461C
MX2.2
81 ± 16
A25C + T417C
MX2.3
93 ± 7.9
MX3.1
90 ± 9.6
MX3.2
89 ± 11
S123C + S338C
+ K461C
A25C + A328C
+ T417C
Figure 3.7. Second and third round of
mutagenesis. The 3D structure of MX is presented
with residues chosen for cysteine modification in
colored spheres (prepared in PyMOL), whereas the
same color code belongs to the one construct. The
catalytic triad (Ser, Asp, His) is shown in yellow
spheres. MX consists of two domains that are
covalently connected by two linear strands shown in
black.82 a) Three doubly cysteine-mutated MXs,
b) two triple cysteine-mutated MXs.
57
a
b
Chapter 3: Engineering of two prolyl endopeptidases
Figure 3.8. Engineering of SC. a) 3D structure of the unoccupied form of SC with highlighted residues
chosen for cysteine modification (SC1.1–1.5 in blue). The active site (Ser, Asp) is depicted in yellow, and
the two linear strands covalently connecting the catalytic and β-barrel propeller domain of SC are shown in
light black and are indicated by an arrow. (PDB 1YR2).82,217 b) Nomenclature used for the conjugates
produced in the three rounds of mutagenesis. c) Distance of sites for polymer attachment. Residues chosen
for mutagenesis and thus, for site-specific polymer attachment are highlighted in red and blue (SC3.1 and
SC3.2, respectively). The distances were calculated with PyMol and are presented in Å (= 0.1 nm).
Regarding SC, again three rounds of mutagenesis were performed to introduce
cysteine residues on the surface of the enzyme. In the first round, five singly mutated SCs
were designed (SC1.1–SC1.5; Figures 3.8a, b) while applying the same rules as for MX.
Residues being modified were similar in size to cysteine and located in a positively charged
microenvironment. The distance between the modifications on the same domain was
approximately 4 nm, whereas the distance between engineered cysteines across the two
domains was longer (~7 nm, Figure 3.8c). However, the interdomain distance is expected
to decrease when SC adopts a liganded conformation.82 Three cysteine modifications were
located on the catalytic domain and two on the β-barrel propeller domain (Figure 3.8a).
SC1.1–1.5 were expressed and assessed for their activity. SC1.1 and SC1.2 exhibited activity
similar to SC WT, whereas SC1.3–SC1.5 showed significantly lower activity (Table 3.3).
The cause for the partial activity loss is unclear but possibly related to impaired protein
folding, dynamics or catalytic processes (i.e., substrate binding).
58
Chapter 3: Engineering of two prolyl endopeptidases
Table 3.2. Activity of SC mutants. The location
of the cysteine modification is presented with the
labeling of the different SC variants and their
activity. *p < 0.05 vs. SC WT. Mean ± SD (n = 3).
Location
of mutation
Constructs
Activity [%]
± SD
None
SC WT
105 ± 4.30
A96C
SC1.1
81 ± 18
S351C
SC1.2
114 ± 4.65
A560C
SC1.3
54 ± 19*
T441C
SC1.4
64 ± 2.3*
S643C
SC1.5
52 ± 5.1*
A96C + S351C
SC2.1
58 ± 19*
S351C + A560C
SC2.2
50 ± 13*
S643C + T441C
SC2.3
71 ± 28
SC3.1
61 ± 13*
SC3.2
38 ± 15*
A96C + S351C
+ A560C
S351C + A560C
+ T441C
Figure
3.9. Second and third round of
mutagenesis. The 3D structure of SC is presented
with residues chosen for cysteine modification as
colored spheres (prepared in PyMOL), whereas the
same color code belongs to the one construct. The
active site (Ser, Asp) is shown as yellow spheres. SC
consists of two domains that are covalently connected
by two linear strands shown in black.82 a) Three
doubly cysteine-mutated SCs, b) two threefold
cysteine-mutated SCs.
59
a
b
Chapter 3: Engineering of two prolyl endopeptidases
To further probe the impact of cysteine mutagenesis on the surface of SC, sets of
two mutations were combined to produce three doubly mutated SCs (SC2.1–2.3) carrying
one cysteine modification on each domain (Figure 3.9a). Regarding SC2.1 and SC2.2, the
high activity of SC1.2 could not be preserved but the constructs showed activity losses in
the range of 42–50% compared to MX WT (Table 3.3). SC2.3 showed similar loss in
activity (29%) with no statistical difference to MX WT due to the large standard deviation.
Then, two triply mutated SCs were designed by combining three modifications so
that two cysteines are placed on the catalytic domain and one cysteine mutation on the βbarrel propeller domain (Figure 3.9b). SC3.1 (A96C, S351C, A560C) was based on SC2.1
and exhibited ~60% WT activity being similar to the single mutant with the lowest activity
of the three parental mutations (SC1.3, Table 3.3). SC3.2 (S351C, A560C, T441C) was
constructed from SC2.2 and exhibited the largest loss in activity upon three rounds of
mutagenesis (~40% WT activity).
In contrast to MX, most single cysteine mutations of SC have caused significant
activity losses. When combined, their detrimental effect was accumulating. Apparently,
even residues on the surface of the PEP can impact the catalytic process to a certain degree.
The catalytic triad is cleaving the peptidic substrates where the interface domain is taking a
crucial part when complexing the peptide. However, first the substrate needs to interact
with the enzyme to induce its opening. Not only the induction but also the process of
opening and closing might be partially driven by residues on the surface of the PEPs;
processes which might be disturbed by modifying the specific residues of SC to cysteines.219
Moreover, the cysteine mutations might also have affected the protein folding causing
decreased activity. However, a detailed analysis of structural changes or impairment of the
catalytic mechanism has not been part of this project. Based on the presented results, the
highly active MX mutants were chosen for site-specific polymer conjugation.
60
Chapter 3: Engineering of two prolyl endopeptidases
3.4. Conclusion
The structural similarities of MX and SC were previously proven by solving their crystal
structures.82 However, site-directed mutagenesis of MX and SC showed that the
modification of surface residues had different effects on their activity. The single cysteine
mutants of SC showed partially decreased enzyme activity. When combining two or three
cysteine modifications on SC, the sum of mutations caused a higher drop in activity. This
implicates that the sites of SC chosen for modification must be somehow involved in
catalytic processes and offer, therefore, no promising targets for polymer attachment.
Regarding MX, all three rounds of mutagenesis did not influence enzyme activity. This led
to the successful development of two highly active triply cysteine mutated MXs that serve
as promising candidates for site-specific polymer conjugation.
61
Chapter 4
Syntheses, in vitro and in vivo
characterization of MX bioconjugates
Part of this chapter is published in:
Schulz, J. D., Patt, M., Basler S., Kries H., Hilvert, D., Gauthier, M. A., Leroux, J.-C. Adv.
Mater. 2015; DOI: 10.1002/adma.201504797.
Chapter 4: MX bioconjugates, in vitro and in vivo
4.1. Introduction
T
HE catalytic properties of enzymes make them very attractive therapeutic agents
for the treatment of diverse diseases.220 Enzymes efficiently and selectively
catalyze reactions that can be harnessed to modify endogenous biomolecules and
metabolic pathways. Furthermore, enzymatic drugs are often covalently modified with
polymers such as PEG to improve their physicochemical and biological properties. Indeed,
PEGylation is used to prevent aggregation,212 increase stability against proteolytic
cleavage,203 enhance hydrodynamic volume, thereby slowing down renal clearance and
increasing biological half-life,221 and, finally, reduce immunogenicity by masking antigenic
epitopes.222 A variety of polymer-modified protein formulations have successfully reached
the market,223 mostly prepared by randomly attaching polymers to solvent exposed lysine
residues. However, when randomly attaching polymers, rather conjugation sites nor the
number of conjugated polymer chains can be controlled, leading to a heterogeneous
mixture of bioconjugates and thus, to poor reproducibility. Lysine residues are often
multiple exposed on the surface of proteins and generally also at sites that are not attractive
for polymer attachment, e.g., near the active site. Thus, random polymer attachment is
typically associated with a loss of enzyme activity, so that large amounts of the often nonhuman protein must be administered to achieve the desired in vivo activity. This is potentially
problematic since the amount of administered drug generally correlates with side-effects.
Moreover, polymer conjugation has been most successful for enzymes acting on small
substrates. Enzymes that target (oligo)peptides (such as PEPs) are potentially more
challenging because they often require conformational flexibility, a property that may be
hindered by polymer conjugation. Further, enzyme therapeutics for more aggressive
biological environments than the blood, such as the GI tract, have received substantially
less attention, in part because of severe activity loss under strongly denaturing/proteolytic
conditions.203
In the case of celiac disease random polymer conjugation was performed on PEPs
with the objective of improving enzyme stability in the GI tract. In a patent application,224
the random PEGylation of glutenases was reported. Although conjugates displaying
> 100% of the catalytic activity of the native proteins were reported, in our hands93 random
PEGylation systematically led to substantial loss in activity consistent with reports from
63
Chapter 4: MX bioconjugates, in vitro and in vivo
most other PEGylated enzymes.225 Furthermore, the patent application in question
provided only relatively limited in vitro analysis of the conjugates, and no in vivo
experiments.224 More recently, our group has shown that random modification of MX with
a polycationic dendronized mucoadhesive polymer stabilized the enzyme in the GI tract,
leading to prolonged activity in this harsh environment.93 However, this conjugate was
~four-fold less active than the MX WT, which can be tentatively attributed to impaired
protein dynamics or hindered substrate entry. As a consequence, higher doses of this
conjugate would be required for therapeutic applications, possibly enhancing side-effects
associated with the polycationic mucoadhesive polymer. Indeed, it is known that
polycations promote the opening of tight junctions in the intestinal epithelium,226 which
may allow other immunogenic or toxic substances to reach the blood.
To this end, substantial efforts are currently focused on developing strategies to
protect proteins more effectively without compromising their activity.227 These strategies
include site-specific polymer conjugation to residues remote from the active site and the
conjugation of polymers with defined architectures and bearing functional groups to better
protect the enzymes.222,228–230 Site-specific polymer conjugation offers a promising tool to
overcome the limitations of random polymer attachment due to the better control of
polymer coupling. Thereby, homogenous products can be produced with a defined number
of polymer chains at specific locations on the protein causing good reproducibility and high
activity which are important factors for the pharmaceutical industry. Moreover, attaching
polymers with well-defined structures helps to decrease the polydispersity of the
bioconjugates and functional groups can positively influence the stability in vivo.93
Hence, this chapter demonstrates that protection of MX in the challenging
environment of the GI tract can be achieved without compromising catalytic activity
relative to the native protein. Thereby, a combination of controlling the properties of
polymers and their ‘multi-site-specific’ attachment to the enzyme was performed.
64
Chapter 4: MX bioconjugates, in vitro and in vivo
4.2. Materials and Methods
All chemicals were purchased at the highest possible analytical grade, and used as received.
The
fluorescence-quenched
peptide
substrate
(HiLyte
Fluor™647)–
LPYPQPK(QXL™670) was custom-synthesized by Eurogentec (Seraing, Belgium) at 95%
purity, and was supplied with an analytical chromatogram and mass spectrum.
Site-specific polymer conjugation of MX: Mutant MX (0.65 mg/mL) was incubated
overnight (4 °C) with 100 molar eq. (relative to protein) of either 5 kDa (Sigma-Aldrich) or
40 kDa (Jenkem Technology, Plano, TX) mPEG–maleimide in sodium phosphate buffer
(50 mM, pH 8) (Figure 4.1a). Conjugates were purified by repetitive ultrafiltration (Amicon
30 kDa or 100 kDa MWCO) with sodium phosphate buffer (50 mM, pH 7.5). A generation
3 PAMAM (3.6 mM) with an ethylenediamine core and amino terminal groups (Figure
4.1b, Dendritech® Inc, Midland, MI) was modified with 2.5 eq. (relative to the dendrimer)
of N-γ-maleimidobutyryl-oxysulfosuccinimide ester (Thermo Scientific, Waltham, MA) in
3-(N-morpholino)propanesulfonic acid (MOPS, 100 µL, 100 mM), pH 7.2 for 1 h at r.t.
The resulting thiol-reactive dendrimer was purified by ultrafiltration (Amicon 5 kDa
MWCO) with MOPS (50 mM, pH 7.2) and characterized by 1H-NMR spectroscopy (D2O,
400 MHz, Bruker AV-400) (Figure A6). Subsequently, mutant MX (0.65 mg/mL) was
reacted with the singly modified dendrimer (100 eq. relative to protein) in MOPS (50 µL,
50 mM, pH 7.2) overnight at 4 °C. The conjugate was purified by repetitive ultrafiltration
(Amicon 30 kDa MWCO) with MOPS buffer (50 mM, pH 7.5). Protein concentration was
determined with the bicinchoninic acid (BCA) protein assay kit (Thermo Scientific)
according to the supplier’s protocol or by absorbance measurement at 280 nm, and purity
was confirmed by HPLC (Figures A7, A8). Complete conjugation was determined by SDSPAGE (10%) and by titrating residual thiol groups (Measure-IT™ Thiol Assay Kit, Life
Technologies).
65
Chapter 4: MX bioconjugates, in vitro and in vivo
a
b
H2NH2N NH2
H 2N
N N
H2N
O
H 3C
O
O
N
n
O
H 2N
N
N
H 2N N
H 2N
N N
H 2N
H 2N N
H 2N
H 2N
H2N NH2 NH2
NH2
N N
N
N
N
N
N
N
N
N
N
N
N
N N
H 2N
H2NH2N NH2
NH2
N NH2
N
N NH2
NH2
N N NH
2
NH2
NH2
NH2
N N
NH2
H2N NH2NH2
Figure 4.1. Molecular structure of a) mPEG–maleimide and b) PAMAM G3 with terminal amino groups.
In vitro activity: MX or MX–polymer conjugates (0.5 μg/mL) were incubated at 37 °C
with (benzyloxycarbonyl)-glycine-proline-p-nitroaniline (2.5 μM, Bachem) in sodium
phosphate buffer (50 mM, pH 7.0). Enzyme activity was calculated from the slope of the
absorbance change at 410 nm due to the release of nitroaniline, which was monitored for
5 min. The activity (slope) observed with the commercially available MX WT was set to
100%.
In vitro stability: To assess stability, MX or MX–polymer conjugates (2 μg/mL) were
incubated in simulated intestinal fluid USP pH 6.8 (10 mg/mL pancreatin from porcine
pancreas, 4 × USP, Sigma-Aldrich) at 37 °C. At specific time points an aliquot was
withdrawn to measure activity. To calculate the enzymatic half-life (t50), an exponential
curve fitting of the graphs was performed with SigmaPlot (Figure A9).
HPLC analysis: The PAMAM–GMBS conjugate was analyzed by HPLC (Merck Hitachi
LaChrom, Tokyo, Japan) using a XBridgeTM BEH 300Prep C18 5 μm (10 × 150 mm) with
1 mL*min–1 flow rate, isocratic with water (+0.1% TFA) for 5 min followed by a linear
gradient of ACN (+0.1% TFA) over 20 min (40 ̶ 60% v/v). The polymer–MXs were
analyzed by HPLC with 1 mL*min–1 flow rate, isocratic with water (+0.1% TFA) for 5 min
followed by a linear gradient of ACN (+0.1% TFA) over 20 min (10 ̶ 90% v/v).
Random PEGylation: mPEG (5 kDa, methoxy PEG propionaldehyde, JenKem
Technology) was conjugated to MX3.1 (0.325 mg/mL) in 200 µL sodium phosphate
(20 mM), sodium cyanoborohydride (20 mM), pH 6, containing 100, 500, 1000, 2000 or
66
Chapter 4: MX bioconjugates, in vitro and in vivo
5000 eq. (relative to protein) of the reactive mPEG. The reaction was maintained at 4 °C
overnight, after which time the conjugate was purified by repetitive ultrafiltration (Amicon
30 kDa MWCO) using MOPS buffer (50 mM, pH 7.5) as eluent. Protein concentration was
assayed by spectrophotometry at 280 nm. The degree of conjugation was determined by
SDS-PAGE (10%) and NMR analysis. The additional site-specific PAMAM conjugation
was performed as mentioned above. Thiol quantification assay (Measure-IT™ Thiol Assay
Kit, Life Technologies) was performed to prove complete site-specific PAMAM
conjugation.
1H
NMR spectroscopy: A known amount of MX–polymer conjugates was dissolved in
D2O and a defined amount of DMSO was added. 1H NMR spectroscopy of this solution
was performed (D2O, 500 MHz, Bruker DRX-500). The ratio of the integrals of the PEG5
backbone peak (3.6 ppm) and the DMSO yielded the concentration of the PEG5
component in the solution. With the known protein concentration (based on absorbance
at 280 nm), the grafting density of PEG5 per MX can be calculated.
In vivo activity: Animal experiments were approved by the Cantonal Veterinary Office
Zurich (226/2012). Female Sprague-Dawley rats (7–21 weeks old, 250–350 g) on glutenfree food were shaved in their abdominal region and fasted for 4 h with access to water
unless stated otherwise. For each animal, a fluorescence image was recorded prior to any
procedure. The rats were orally gavaged with the fluorescence-quenched peptide alone
(negative control) or with the pre-cleaved peptide (positive control). MX or MX–polymer
conjugates (1 µg in 200 µL 10 mM sodium phosphate buffer, 5 µg/mL, pH 7.5) were mixed
with gliadin (2 mg/mL, 10 µL) and orally gavaged to rats. The fluorescence-quenched
peptide substrate [(HiLyte Fluor647)–LPYPQPK(QXL670)] (2.5 µM in 200 µL sodium
cholate 0.5 wt%) was administered 5 min later. To determine stability in the GI tract, the
protocol above was modified to introduce a 1 h or 2 h delay between the administration of
the MX–polymer conjugate and the fluorescence-quenched peptide. For imaging, rats were
anaesthetized (2–2.5% isoflurane, 0.5 mL/min oxygen) and placed on a heated platform
(37 °C) in an in vivo imaging system (IVIS Spectrum, Caliper, Mainz, Germany).
Fluorescence images were recorded at specific time points at λex/em = 745/800 nm, binning
8, f-stop 2, exposure time 2 s. Between time points, rats were returned to their cages. Images
were analyzed with Living Image software (Caliper Life Science, Waltham, MA). Images
67
Chapter 4: MX bioconjugates, in vitro and in vivo
were smoothed by 7 × 7 pixels to reduce background noise. At each time point, the average
fluorescence intensity was normalized to the pre-gavage value. Auto-fluorescence not
related to the experiment and outside the abdominal area was removed from all displayed
images.
In vivo pH measurements: Rats (n = 4) were fasted for 4 h with access to water and
sacrificed. The stomach and intestine were removed and the organs opened. The pH was
measured three times on the opened tissues with pH indicator paper (pH gradation pH 3.8–
5.8 and 6.4–8.0, Macherey-Nagel, Düren, Germany).
Analysis of Recorded Images: Images were analyzed with Living Image software (Caliper
Life Science). A region of interest (ROI) was defined for the GI tract and stayed constant
in size and location throughout the whole experiments (Figure 4.2). At each time point,
the average fluorescent intensity was normalized to the pre-gavage status.
Figure 4.2. A ROI (blue) was defined, constant in size and position, to measure the fluorescent signal of
the cleaved peptide in the GI tract of rats.
Statistical analysis: The data from in vivo experiments were analyzed using nonparametric
Kruskal-Wallis one-way ANOVA followed by Dunn post hoc test. In vitro data were
analyzed by parametric one-way ANOVA followed by a Dunnett`s post hoc or Tukey`s
post hoc test. Differences were considered statistically significant at p < 0.05.
68
Chapter 4: MX bioconjugates, in vitro and in vivo
4.3. Results and Discussion
The single, double and triple cysteine-mutated MXs (Table 3.2) were PEGylated with 5
and 40 kDa mPEG–maleimide (PEG5 and PEG40) to yield well-defined conjugates with
1–3 polymers attached (Figure 4.3a). The functionalized PEG was reacted via a Michaeladdition to the free thiol group(s) on the surface of MX and reduced conditions prevented
disulfide bond formation. The first indication of a successful polymer coupling was
obtained by SDS-PAGE gel analysis showing that no unconjugated MX was still available
(Figure 4.4). When the mono-PEGylation of MX1.1–1.6 was performed, the SDS-gel
proved that all single cysteines were reacted. In the case of doubly and triply mutated PEPs,
this method could however not tell if possibly a heterogeneous mixture of singly, doubly
(and triply) MX–PEGs was observed. Once MX is PEGylated, it migrates differently
through a SDS-gel indicating that the size of the band does not correspond to the actual
size of the conjugate (Figure 4.4). Therefore, other measures to demonstrate the successful
coupling of the PEG chains to all free cysteines were required. A dye that turns fluorescent
once coupled to a thiol group was used to detect unreacted cysteines. In all PEGylation
reactions no free thiol groups were detected (Figure 4.5) proving that both PEG5 and
PEG40 were successfully conjugated to the all cysteines of mutant MXs.
Figure 4.3. Chemical reaction schemes of site-specific and random polymer conjugation. a) Sitespecific PEGylation of mPEG-maleimide (5 and 40 kDa) to thiol groups of engineered cysteines on MX.
b) Chemical reaction of sulfo-GMBS to PAMAM G3. Subsequently the polymer-linker was mixed with
MX–SH to form PAMAM–MX conjugates. c) Random PEGylation of mPEG–aldehyde (5 kDa) to lysine
residues of MX. Depending on the initial feed ratio, different grafting densities of MX–ran–PEG5 linked
via -NH2 groups were observed (Table 4.1).
69
Chapter 4: MX bioconjugates, in vitro and in vivo
Figure 4.4. SDS-PAGE (10%) analysis of conjugation reactions. a) The gel was stained with coomassie
blue for protein identification. b) Barium iodide staining (PEG staining) shows a brown color on the
PEGylated protein bands (rows 3 and 4). The migration behavior depends on the nature of the bioconjugate.
PEG has a higher hydrodynamic volume than proteins and therefore migrates slower in SDS gel
electrophoresis compared to a protein of the same molecular weight. Thus, the 5 kDa PEGylated enzyme
appears larger than it actually is and the 40 kDa PEGylated variant does almost not migrate at all. The
dendrimeric PAMAM has a very compact structure leading to a very similar migration to a protein of the
same size. Generally, the polydispersity of polymers in addition to the different migration behavior through
SDS gels lead to smeared bands of the polymer conjugates.
7000
Normalized fluorescence
Normalized fluorescence
8
6
4
MX3.1
MX3.1–PEG5
MX3.1–PEG40
MX3.1–PAMAM
6000
5000
4000
3000
2000
1000
0
-1000
500
2
520
540
560
580
600
620
640
660
Wavelength / nm
0
g
Ne
ve
a ti
n
co
tr o
l
MX
3 .1
MX
–P
3 .1
EG
5
MX
3 .1
E
–P
G4
MX
0
3 .1
AM
AM
P
–
MX variant
Figure 4.5. Proof of completed site-specific conjugation. A thiol-binding dye was used to detect free
thiol groups. The negative control consists of the dye only, whereas the positive control is represented by
MX3.1 carrying three free cysteines. Data were divided by the negative control and plotted in a histogram
at 523 nm. The inset presents the fluorescent signal with wavelength showing again that no free thiols were
detected in none of the reactions. Mean + SD (n = 3). Only the positive control MX3.1 was performed
with n = 1.
70
Chapter 4: MX bioconjugates, in vitro and in vivo
Once the PEGylated MXs were purified, in vitro activity measurements were
performed. In Figure 4.6a the activity of the six mono-PEGylated MX conjugates are
illustrated and typically the PEG5 conjugates retained ~100% MX WT activity. This
indicates that the short polymer did not interfere with the catalytic process of MX. One
exception, MX1.1–PEG5, possessed a slightly higher activity than the native enzyme.
Increased activity upon PEGylation has been reported previously,225,231,232 and might result
from subtle conformational changes in the protein induced by the PEG chain. Others have
suggested that polymers may favor accumulation of substrate near the active site.231
Site-specific mono-PEGylation of PEG5 to MX had no negative effects on the
enzymatic activity, whereas the longer mPEG (PEG40) showed discrimination between the
attachment sites. Most conjugates with the longer PEG40 possessed ~50% of the activity
of MX WT, except MX1.1–PEG40 that was as active as the native protein, and MX1.2–
PEG40 that maintained ~70% MX WT activity (Figure 4.6a), indicating that the
conjugation site is important when full activity of the conjugates should be preserved. It
should be noted that PEG40 has a hydrodynamic diameter of ~13 nm,233 which is ~2 to 3
times larger than MX itself (based on PDB 2BKL). Thus, in addition to PEG40 sterically
hindering the approach of the substrate, the polymer may also mask substrate binding sites
on the protein associated with the catalytic mechanism.
To address this point, three double-PEGylated MXs were created and analyzed
(Figures 3.7a, 4.6b). MX2.1–PEG5 possesses two PEG5 chains on the β-barrel propeller
domain, and MX2.2–PEG5 and MX2.3–PEG5 have one PEG5 chain on the catalytic
domain and another on the β-barrel propeller domain. As seen in Figure 4.6b, the PEG5–
conjugate MX2.1 retained ~80% MX WT activity, whereas MX2.2–PEG5 and MX2.3–
PEG5 are ~55% active. The partial loss of activity suggests that the collective interactions
between PEG and the protein are influenced by the total amount of PEG added (i.e., either
the molecular weight of a single chain, or the combined molecular weight of several
chains).234 Because both PEG5 chains are on the same domain of the protein in MX2.1–
PEG5, it is conceivable that the approach of the substrate to its binding site at the interdomain interface as well as the conformational reorganization of the protein are less
adversely affected than when the polymers are present on both domains (MX2.2–PEG5
and MX2.3–PEG5). It is notable that increasing the PEG length from 5 to 40 kDa on the
71
Chapter 4: MX bioconjugates, in vitro and in vivo
doubly mutated MXs did not significantly reduce activity. This suggests that the steric
hindrance associated with PEG eventually saturates when the amount (as overall molecular
weight) reaches a threshold value (vide infra). Indeed, Chiu et al. reported similar results with
PEGylated trypsin where conjugates carrying multiple PEG chains of different sizes (2, 5
and 10 kDa) showed identical activity.232
Figure 4.6. In vitro activity of MX WT, mutated and polymer conjugated MXs. a) Singly mutated and
PEGylated MXs, b) doubly mutated and PEGylated MXs, c) triply mutated, PEGylated and PAMAM
conjugated MXs. d) Random PEG5 conjugates of MX3.1 and the latter with additional site-specifically
attached PAMAMs. Activity was assessed by post-proline cleavage of Z-Gly-Pro-pNA at 37 °C. *p < 0.05
vs. MX WT. Mean + SD (n = 3). MX WT activity bar is identical in a), b) and c).
72
Chapter 4: MX bioconjugates, in vitro and in vivo
To probe this phenomenon further, two triply PEGylated forms of MX were created
by attaching polymers simultaneously in regions “A”, “B” and “C” so that both variants
carry two cysteine mutations on the β-barrel propeller domain and one on the catalytic
domain (Figure 4.7). The first variant, MX3.1 (i.e., S123C, S338C, K461C), was based on
MX2.1 and did not show any loss in activity upon the attachment of three PEG5 chains
compared to the corresponding di-PEGylated enzyme (MX2.1–PEG5, Figure 4.6c, b).
The second, MX3.2 (i.e., A25C, A328C, T417C) was constructed from MX2.3, and its
threefold PEGylated form (MX3.2–PEG5) showed less activity than its di-PEGylated
counterpart MX2.3–PEG5 (~55% vs. ~35%, respectively; Figures 4.6b, c). Furthermore,
as for the doubly-PEGylated conjugates, activity was independent of PEG molecular
weight. These results support the idea that the detrimental effect of PEGylation on activity
of MX eventually plateaus when the overall molecular weight of PEG per conjugate
increases above a threshold value in the 15–40 kDa range. Moreover, the analysis of the
library of conjugates suggests that the difference between the activity of PEGylated MX3.1
and MX3.2 (~70% vs. ~35%, Figure 4.6c) reflects the importance of the conjugation sites
rather than PEG length on the overall catalytic process of the protein (substrate binding
and conformational reorganization).
Figure 4.7. 3D structure of MX with
highlighted residues chosen for polymer
attachment (MX3.1 residues in red and
MX3.2 in blue). The catalytic triad (Ser,
Asp, His) is depicted in yellow, and the
two linear strands covalently connecting
the catalytic and β-barrel propeller
domain of MX are shown in light black
and indicated by an arrow. The three
different regions of MX (A, B and C)
where surface-exposed cysteines were
introduced are circled.82
73
Chapter 4: MX bioconjugates, in vitro and in vivo
Figure 4.8. SDS-PAGE (10%)
visualizing unconjugated MX3.1 and
MX3.1 reacted with different molar
equivalents
of
PEG5.
Row
identification: 1 molecular weight
marker, 2 MX3.1, 3 MX3.1–ran –
PEG5 1, 4 MX3.1–ran –PEG5 3, 5
MX3.1–ran –PEG5 5, 6 MX3.1–
ran –PEG5 8, 7 MX3.1–ran –PEG5
14. Higher molecular weight for the
MX conjugates is produced when the
molar amount of PEG5 is increasing
(Table 4.1). A smear is obtained for
the PEG5 conjugated samples on the
gel, hinting towards heterogeneous
mixtures of the bioconjugates.
This conclusion was tested by randomly conjugating mPEG propionaldehyde
(5 kDa) to solvent-exposed lysine residues of MX3.1 to different extents (MX3.1–ran –
PEG5; Figures 4.3c, 4.8). The multi-PEGylated conjugates displayed decreasing activity
with increasing grafting density of PEG5 from 1 to 14 chains per enzyme
(Figure 4.6d, 4.9). These results support the notion that conjugates prepared by appending
polymers site-specifically to appropriate attachment sites are generally superior to randomly
modified proteins.
Prior in silico analyses of PEG–protein conjugates performed by other groups have
suggested that PEG can interact with hydrophobic areas on proteins and by doing so has a
large “footprint” on the solvent-exposed surface.235 To examine this point, MX3.1 and
MX3.2 were site-specifically modified with a G3 PAMAM (6.5 kDa) dendrimer with
terminal amino groups, which adopts a compact spherical structure (hydrodynamic
diameter of 3.1 nm; Figure 4.1b).125,236 Thereby, a two-step reaction was performed,
whereas a bi-functional linker was first reacted with an amino group of PAMAM and then
coupled to MXs’ thiol(s) (Figure 4.3b). The site-specific coupling of three dendrimers to
MX3.1 and MX3.2 succeeded as shown in Figures 4.4 and 4.5. PAMAM is not expected
to significantly interact with the protein except in the immediate vicinity of the conjugation
site due to its charge and rigid structure giving it little flexibility. Interestingly, MX3.1–
PAMAM maintained 100% of the activity of MX WT, while MX3.2–PAMAM possessed
74
Chapter 4: MX bioconjugates, in vitro and in vivo
the same low activity as its PEGylated analogs (Figure 4.6c), again underscoring the
importance of the conjugation sites. It was also possible to modify MX3.1–ran –PEG5
conjugates with three PAMAM units without further loss of activity (Figures 4.10, 4.11).
These results indicate that attachment of large PAMAM units on the surface of MX is
possible in combination with random PEGylation, suggesting that the surface of the protein
remains fairly exposed despite the presence of multiple PEG5 chains.
a
PEG
backbone
DMSO
MX3.1–PEG5
b
MX3.1–ran–PEG5 3
c
MX3.1–ran–PEG5 8
d
MX3.1–ran–PEG5 14
Figure 4.9. 1H NMR spectra of a) MX3.1–PEG5 site-specifically attached, and b)–d) three random
PEG5 conjugates: b) MX3.1–ran –PEG5 3, c) MX3.1–ran –PEG5 8, d) MX3.1–ran –PEG5 14 (D2O,
500 MHz, Bruker DRX-500). A known mass of the MX3.1–PEG5 conjugate was dissolved in D2O
containing 1 µl DMSO that served as an internal standard. The quantification of the PEG backbone signal
resulted in the following number of PEG chains attached per MX: a) 3, b) 3, c) 8, d) 14. The grafting density
of all conjugates is presented in Table 4.1.
75
Chapter 4: MX bioconjugates, in vitro and in vivo
Figure 4.10. 3D structure of MX highlighting all lysine residues (blue), engineered cysteine residues
(red) and the active site (yellow) as spheres. a) MX3.1 carries 44 lysine residues. The crystal structure shows
that most of the lysines are located on the surface of the enzyme and thus a high number should be available
for random PEGylation. However, some are closely located to the active site leading to one explanation of
loss of activity upon random PEGylation. b) The three cysteine residues of MX3.1 are not closely located
to any lysine residue which gives an explanation why the thiols are still available for PAMAM conjugation
upon random PEGylation.
Normalized fluorescence
10
8
6
4
2
M
X3
.1
M –ra MX
3.
X3
n–
.1
PE 1
–
r
M
G
5
X 3 an
–P
1
.1
E
–
ra
M
G
5
X3
n–
.1
PE 3
M
–
M
X3
X3 ran G5
.1
.1
–P
5
–
–r
r
M
an E G
X 3 an
–P
–P 5 8
.1
EG
M –ra EG
5
X3
n–
5
1
.1
PE
–P 14
–
AM
r
M
G
5
X 3 an
AM
–P
3–
.1
P
E
–
M
X3 ran G5 AM
AM
.1
–P
5–
–r
an E G P A
–P 5 8 MA
M
E G –P
AM
5
14
–P AM
AM
AM
0
MX variant
Figure 4.11. Proof of completed site-specific conjugation after random PEG5 attachment to MX3.1.
A thiol-binding dye was used to detect free thiol groups. The positive control is represented by MX3.1
carrying three free cysteines. Data were divided by the negative control (dye only, Figure 4.5) and plotted
in a histogram at 523 nm. Free thiols were detected in the positive control and in the random conjugated
MXs with no statistical difference upon them. Mean + SD (n = 3). Only when PAMAM was site-specifically
added to the engineered cysteines no free thiols were detected. Herein, n = 1 was performed because
previous experiments showed that the standard deviation was very low and therefore the measurement is
precise when the thiols are reacted (Figure 4.5).
76
Chapter 4: MX bioconjugates, in vitro and in vivo
Based on these findings, MX3.1 and MX3.1–polymer conjugates were selected for
analysis in vivo, as they displayed the highest activity. A model peptide substrate mimicking
an immunogenic sequence of gluten peptide was designed. The intact substrate bears a
fluorophore/quencher pair allowing in vivo activity to be monitored because digestion of
the peptide releases the fluorophore, which recovers its fluorescence (Figure 4.12a).216 Rats
were orally gavaged with MX or MX–polymer conjugates on an equal mass basis (very close
to equal activity basis), together with gliadin (sacrificial substrate), followed by the
administration of the fluorescence-quenched substrate. This protocol is consistent with the
envisaged therapy, in which celiac patients would ingest the enzyme–conjugate prior to
ingesting gluten-containing products. Enzyme activity was monitored using a whole animal
imaging system (Figure 4.12b). For better reproducibility, a single large region of interest
was defined that encompassed the entire abdomen of the rats (Figure 4.2). Moreover, in
order to avoid a high background signal with the negative MX WT control, the dose of
enzyme administered was set at 1 µg/rat. The in vivo activity of MX WT and MX3.1 was
found to be equivalent to that of the negative control (peptide alone) under these conditions
(Figures 4.12b, c). Even though MX3.1–PEG5 showed higher in vivo activity, no statistical
differences to the negative control were observed (Figure 4.12b). These results reinforce
the idea that the three PEG5 chains poorly cover the surface of the protein, leaving it
accessible to proteases and susceptible to pH-induced deactivation. Increasing PEG length
to 40 kDa significantly improved stability, as shown by an increase of the fluorescence
signal in the GI tract (Figures 4.12c, d). Interestingly, PAMAM had an equivalent effect to
PEG40, despite the fact that PAMAM is smaller than PEG5 (hydrodynamic diameter of
3.6 vs. ~5 nm, respectively, Figures 4.12c, d). The cationic charge of the polymer may play
a role in stabilizing MX against proteases, possibly by complexing endogenous negatively
charged macromolecules, such as mucin. Alternatively, PAMAM could complex negativelycharged proteases such as pepsin (isoelectric point 1.0) or even repel positively-charged
proteases such as chymotrypsin or trypsin (isoelectric points of 8.5 and 10.5, respectively;
see Figure 4.13 for pH values along the GI tract).237
77
Chapter 4: MX bioconjugates, in vitro and in vivo
Figure 4.12. Enzymatic activity of MX3.1–polymer conjugates in the GI tract of rats. a) Scheme of
the cleavage of the fluorescence-quenched peptide. The model peptide (HiLyte Fluor647)–
LPYPQPK(QXL670) was orally gavaged to rats after the administration of MX WT, MX3.1 or MX3.1–
polymer conjugates. After the cleavage of the peptide, fluorescence emission was detected by live imaging
of the GI tract. b) Schematic illustration of the in vivo experimental set-up. 5 min, 1 or 2 h prior to the
substrate, MX and MX–polymers were orally applied. At specific time points (1–4 h after peptide
application) in vivo imagining was performed. c) Normalized fluorescent signal of the cleaved peptide vs.
time. MX or MX–polymer conjugates were orally administered to rats 5 min prior to the peptide application.
Fluorescent signal was measured at specific time points (1–4 h). *p < 0.05 vs. MX WT. Mean + SD (n = 8–
10). Data are also partially presented in Figure 4.14. d) Images show peptide cleavage over time in the GI
tract when the peptide was administered 5 min after the enzyme/conjugate. The positive control is
presented by the pre-cleaved peptide. All conjugates were active over time and the PEG40 and PAMAM
conjugates showed highest fluorescent signals.
78
Chapter 4: MX bioconjugates, in vitro and in vivo
9
8
pH
7
6
5
4
3
ac
om
St
h
Du
e
od
nu
m
ju
Je
nu
m
u
Ille
m
up
pe
r
u
Ille
m
low
er
Tissue
Figure 4.13. In vivo pH of GI tract content. Mean + SD (n = 4). No standard deviation was obtained
when measuring the stomach pH with pH paper.
Although it is difficult to pin down the precise origins of this effect, conjugation of
another high-molecular weight polycation to MX has previously been shown to promote
stability and retention in the stomach of rats, suggesting that the underlying mechanism is
general.93 It is also worth noting that the fluorescence intensity observed for MX3.1–
PAMAM and MX3.1–PEG40 was statistically indistinguishable from the positive control,
confirming that these constructs retain high activity. Indeed, in vitro analysis of the stability
of MX and its polymer conjugates in simulated intestinal fluid (pH 6.8, containing
pancreatin) reflects the trends observed in vivo (significant stabilization with MX3.1–PEG40
and MX3.1–PAMAM, Table 4.1). This model fluid has been shown to be a good predictor
of in vivo stability in previous work with this type of enzyme.93 Thereby, most of the
conjugates of MX3.2 and the random PEG5 conjugates of MX3.1 showed low stabilization
success even though the latter are carrying a high number of polymers (≤ 14) which could
have possibly led to more efficient enzyme protection. This indicates that high activity as
only obtained for the site-specific MX3.1 conjugates and which is mainly caused by the
specific conjugation sites is an essential factor for the effective stabilization under
physiological conditions.
79
Chapter 4: MX bioconjugates, in vitro and in vivo
Table 4.1. Half-life of enzymatic activity in simulated intestinal fluid. MX or MX–polymer was
exposed to simulated intestinal conditions (pancreatin, pH 6.8, 37 °C) and activity was assessed at different
time points. The table presents the time needed to inactivate the enzymes by 50% (T50). *p < 0.05 vs. MX
WT. Mean (n = 3). The location of the cysteine modifications (color code as in Figure 4.7) is presented
with the labeling of the different MX variants. Regarding the random conjugation, the initial ratios of PEG
are presented with the grafting densities of PEG chains to MX3.1.
Cysteine mutation
MX variant
t50 [min] ± SD
None
MX WT
6.8 ± 1.0
S123C + S338C + K461C
MX3.1
5.5 ± 1.4
“
MX3.1–PEG5
9.8 ± 1.2
“
MX3.1–PEG40
13.9 ± 1.8*
“
MX3.1–PAMAM
11.7 ± 1.9*
A25C + A328C + T417C
MX3.2
2.9 ± 0.3*
“
MX3.2–PEG5
5.7 ± 0.6
“
MX3.2–PEG40
10.6 ± 0.2*
“
MX3.2–PAMAM
PEG5 (eq.)
1.9 ± 0.4
# PEGs attached
100
1
2.1 ± 0.1
500
3
3.4 ± 0.8
1000
5
4.3 ± 1.4
2000
8
5.5 ± 1.1
5000
14
13.6 ± 7.9
80
Chapter 4: MX bioconjugates, in vitro and in vivo
To further investigate the stability of the conjugates of MX3.1 in the GI tract, they
were orally applied to rats and allowed to “incubate” in this environment for 1 or 2 h before
the fluorescence-quenched substrate was administered (Figure 4.12b). As illustrated in
Figure 4.14, the conjugates maintained full activity, even after prolonged exposure to the
harsh environment of the GI tract which is underlining the ability of the polymers to protect
MX in the digestive tract.
Figure 4.14. In vivo activity of MX–polymer conjugates. a)–b) Normalized fluorescent signal of the
cleaved peptide vs. time. MX, (a) MX3.1–PEG40 and (b) MX3.1–PAMAM were gavaged to rats 5 min, 1 h
or 2 h prior to the peptide. Fluorescent signal was measured at specific time points (1–4 h on x-axis) after
peptide administration. *p < 0.05 vs. peptide only; †p < 0.05 vs. positive control (pre-cleaved peptide). Mean
+ SD (n = 8–10). The pre-cleaved peptide, peptide only and MX WT curves are identical in c) and d) and
the data is also presented in Figure 4.12. c) In vivo images present the fluorescent signal when MX3.1–
PEG40 or MX3.1–PAMAM was orally administered 2 h prior to the peptide.
81
Chapter 4: MX bioconjugates, in vitro and in vivo
It should also be mentioned that, while the objective of this work was not to
investigate the innocuousness of the conjugates, no signs of animal distress were observed,
despite repeated dosing of the conjugates, or PAMAM alone (Figure 4.15). Indeed,
PAMAM G3 was found to be safe in other studies when orally administered.129,238
a
b
Figure 4.15. Analysis of body weight changes of rats. a) Evaluating the body weight of rats which were
fed with MX and MX-polymer. Arrow indicates the time point where the fastening was changed from 4 h
to overnight which caused a temporary drop of body weight. The blue highlighted region indicates the time
frame of the in vivo activity experiments (Figure 4.12). The green frame shows the time period when the in
vivo kinetic studies were performed (Figure 4.14). Mean ± SD (n = 4). b) After three months, three rats
were orally administered with PAMAM for two weeks. The body weight is illustrated over these specific
two weeks. No loss in body weight was obtained.
82
Chapter 4: MX bioconjugates, in vitro and in vivo
4.4. Conclusion
In summary, when analyzing the in vitro results of the bioconjugates created in this work,
three major findings can be concluded. The site of polymer attachment importantly
influences enzymatic activity, whereas the impact of polymer length of PEG is negligible
upon a threshold value, but polymeric architecture matters. Thereby, these findings became
most evident with the triple conjugated MXs as MX3.1 showed higher activity upon
PEGlyation than MX3.2, underlining the importance of conjugation sites and in both cases
no difference between attached PEG5 and PEG40 was observed. Additionally, the
relevance of polymeric architecture was evaluated by attaching the dendrimer PAMAM to
MX3.1. The bioconjugate with the compact polymeric structure retained full WT activity,
whereas the linear PEGs attached to MX3.1 caused a drop in activity.
When orally applied to rats, the results also demonstrated that judiciously chosen
attachment sites and polymer types were crucial to the success of the modified enzymes in
vivo. MX3.1–PEG40 showed higher activity in the GI tract of rats compared to the
corresponding PEG5 conjugated, indicating that the shorter polymer did not efficiently
cover the enzyme’s surface. PAMAM with a size comparable to PEG5 stabilized MX
similar to PEG40 though. This reveals that a compact polymeric architecture as well as
functional groups can enhance the stability in vivo. Moreover, the two bioconjugates of
MX3.1 coupled to PEG40 and PAMAM showed high activity even when they were
incubated in the GI tract 1 or 2 h before the application of the substrate. Thus, the strategy
of triple site-specific conjugation of PEG40 or PAMAM efficiently stabilized MX3.1
without compromising its catalytic activity. As potential adjuvant therapy for celiac disease
and other gluten-related disorders such as non-celiac-gluten-sensitivity,239 less of our
conjugate may therefore be needed to achieve a desired activity level in the body. Lower
doses are favorable for reducing the risk of potential side-effects from the synthetic
polymer.
83
Chapter 5
General Conclusion & Outlook
Chapter 5: Conclusion &Outlook
T
HE work presented in this thesis describes a strategy to successfully stabilize
enzymes in the GI tract without compromising their catalytic activity. Celiac
disease served as an ideal platform to test the developed approach in view of the
lack of available pharmaceutical treatments and of the benefits that the oral administration
of enzymes would bring to patients suffering from this pathology and possibly also from
NCGS.
With a prevalence of ~1%, celiac disease is an autoimmune disorder of the small
intestine triggered by gluten, a mixture of proteins found in a variety of grains.13 A GFD is
so far the only option available for celiac patients. Notwithstanding, staying on a safe GFD
is rather difficult because even small amounts of (hidden) gluten can often lead to intestinal
damage. An enzymatic therapy would offer an approach to help digest hidden or
accidentally consumed gluten. It was reported that PEPs can break down the immunogenic
peptides which prevents the inflammation in the intestine.79 However, PEPs themselves
are prone to get degraded or denatured under the harsh conditions in the GI tract, especially
in the acidic environment of the stomach. A common strategy to overcome this problem
is to encapsulate the enzymes in stomach-resistant material in order to deliver them safely
to the intestine. This approach is unfortunately not suitable for celiac disease treatment
because in order to prevent the inflammation in the small intestine the gluten peptides need
to be degraded in the upper part of the GI tract, for which the enzyme needs to retain its
activity in the stomach. In light of this, celiac disease constitutes a challenging platform for
which to develop an oral adjuvant therapy.
The attachment of polymers to proteins is a very common approach to improve the
physicochemical and biological properties of biomacromolecular drugs used in the
treatment of a variety of pathological conditions (Figure 5.1). For example, the
subcutaneous injection of a phenylalanine-degrading enzyme could potentially treat
phenylketonuria, a disease caused by impaired activity of phenylalanine hydroxylase leading
to neurocognitive dysfunction.240 In order to reduce its immunogenicity the enzyme was
PEGylated, reaching clinical trial Phase II as rAvPAL–PEG.241 Moreover, the PEGylation
of interferon α-2a in the formulation PEGASYS® increased the protein’s half-life upon
subcutaneous administration, enabling it to reach the market as an effective treatment for
chronic hepatitis C in adults.242 Most biomacromolecular drugs so far described consist of
85
Chapter 5: Conclusion &Outlook
injectable formulations of therapeutic proteins on which random polymer conjugation has
been performed. However, introducing random modifications in therapeutic proteins can
wane their biological activity and also compromise the formulation’s homogeneity.
Although other conjugation and protection strategies exist, they have received substantially
less attention.210 In this regard the site-specific attachment of polymeric molecules to
engineered residues of proteins offers a very promising alternative strategy. Despite being
considerably more laborious, this strategy preserves the proteins’ catalytic activity upon
modification and was therefore selected to stabilize the gluten-degrading PEPs enzymes in
the GI tract.243,244
Figure 5.1. Positive effects of polymer conjugation on therapeutic proteins.
86
Chapter 5: Conclusion &Outlook
In Chapter 3 two PEPs, MX and SC, were engineered to strategically place cysteine
residues on the surface of the enzymes.245 In both cases a library of singly, doubly and triply
mutated variants was designed and expressed. Thereby, this protein engineering had
different impacts on the two PEP variants. In the case of SC, all three rounds of
mutagenesis resulted in activity loss. It was therefore not possible to create a catalytically
active SC mutant with three available cysteines on different sites of the surface (which
would have allowed covering most of the enzyme’s surface by polymer conjugation). The
causes for the activity loss upon mutagenesis remain unclear, but the data indicated that the
activity drop was not related to the modification of a specific amino acid or to one particular
site of SC. However, the unique catalytic processes of PEPs involve immense motions that
could be partially driven by surface residues. On the contrary, the mutagenesis of MX was
very successful and two threefold cysteine-mutants that retained the activity of MX WT
were produced. These two constructs (MX3.1 and MX3.2) offered three conjugation sites
each that enabled full coverage of the enzymes’ surfaces upon polymer attachment,
maximizing the degree of shielding from the harsh conditions in the GI tract (Figure 5.2a).
To this end, only MX mutants were coupled to polymers and analyzed in terms of
enzymatic activity and stability under physiological conditions. In Chapter 4, the single
cysteine-mutated MXs were first conjugated to linear PEG of two different molecular
weights (5 and 40 kDa). Whereas the mono-conjugation of PEG5 did not affect the activity
of the mutant MXs, the single attachment of PEG40 partially decreased their activity. To
better understand the effect of PEGylation conjugation on MX’s functionality, the PEGs
were coupled to the doubly mutated variants. The constructs were less active than MX WT,
but interestingly no difference was observed between the PEG5 or PEG40 conjugates. This
indicated that the amount of conjugated PEG (determined by polymer length or number
of polymer chains attached) plateaued above a certain threshold. Similarly, the conjugation
of the PEGs to the three-fold cysteine-mutated MXs followed the same trend, showing no
difference in activity between the PEG5 and PEG40 bioconjugates (Figure 5.2a).
87
Chapter 5: Conclusion &Outlook
Figure 5.2. Summary of main findings of this doctoral thesis. a) MX was engineered and two variants
(MX3.1 (red) and MX3.2 (blue)) with three cysteine residues on the surface were produced and analyzed
for their catalytic activity. b) The two variants were conjugated to linear PEG (5 and 40 kDa) and PAMAM
(G3) with terminal amino groups, and tested for their activity in comparison to that of MX WT. c) The
polymer conjugates of MX3.1 were analyzed in vivo. The bioconjugates were orally administered to rats
together with a quenched substrate. The fluorescent signal generated upon substrate cleavage was measured
during 4 h; the images show the fluorescence 2 h after treatment administration.
However, given that the catalytic activity of the triply PEGylated MX3.1 was substantially
superior to that of MX3.2 conjugates, the importance of the conjugation sites was
demonstrated (Figure 5.2a, b). Thereafter, the polycationic dendrimer PAMAM (6.9 kDa)
was attached to MX3.1 and MX3.2. Dendrimers offer advantages over “traditional”
polymers in terms of their precise structure and size, biocompatibility, aqueous solubility
and versatility given by adaptable terminal functional groups. PAMAM can thus generally
serve as a stability, solubility and/or cellular uptake enhancer.226 Although PAMAM–drug
conjugates have been tested for a variety of applications,246 the use of this dendrimer on
enzyme stabilization, especially when orally applied, has been barely investigated. Herein,
the triple PAMAM conjugation to MX3.1 led to full WT activity whereas MX3.2 showed
88
Chapter 5: Conclusion &Outlook
considerable activity loss, similarly to the PEGylated counterparts (Figure 5.2a, b). The
fact that full WT activity was retained with MX3.1–PAMAM (in contrast to some activity
loss with the PEGylated mutant) indicated that the dendrimer had no impact on the
dynamics and catalytic processes of MX when attached at these specific conjugation sites,
which could be attributed to the dendrimer’s rigidity.
To further understand the impact of polymer conjugation, the best mutant MX3.1
was randomly conjugated to PEG5 via surface-exposed lysine residues. Thereby, it was
found that higher grafting densities of the polymer further aggravated activity loss, which
emphasized the importance of controlling the conjugation site for enzymatic activity
preservation. Thereafter, the randomly PEG5-conjugated MX3.1s were successfully sitespecifically conjugated to three additional PAMAM dendrimers, indicating that the PEG5
chains only poorly covered the surface of the enzyme. No additional activity loss was
observed upon PAMAM addition, highlighting once more the great value of the three
conjugation sites of MX3.1 and the importance of polymer structure.
The best candidates, namely the site-specifically polymer-conjugated MX3.1, were
then tested in vivo. The enzyme and conjugates were orally applied to rats by gavage,
followed by a model substrate consisting of a cleavable gluten-specific sequence and a dyequencher system. Upon enzymatic cleavage the substrate’s dye is de-quenched, emitting a
fluorescent signal that can be detected through the rats’ abdomens. As expected, for control
MX WT and MX3.1, no signal was detected due to the fast deactivation of the unprotected
enzymes in the GI tract. MX3.1–PEG5 gave a small signal but was not statistically
significant compared to MX WT (Figure 5.2c). In contrast, the administration of MX3.1–
PEG40 increased the fluorescence signal significantly, showing that the longer PEG chains
better shielded MX from the harsh conditions in the GI tract. In the case of PAMAM, its
conjugate stabilized the enzyme to an extent comparable to that of PEG40 despite having
a hydrodynamic volume smaller than that of PEG5.
To further confirm that the attached PEG40 and PAMAM durably stabilized the
enzymes, the corresponding MX3.1 conjugates were orally applied and incubated in the
digestive tract for 1 or 2 h before administering the substrate. In both cases full in vivo
activity was observed, demonstrating that PEG40 as well as PAMAM protected MX
efficiently in the GI tract over longer periods of time.
89
Chapter 5: Conclusion &Outlook
As described in Chapter 1, some enzymatic formulations for the oral treatment of
celiac disease have successfully reached clinical trials.86,90 Such studies will determine
whether any of these formulations are safe and efficient enough to reach the market.
Because insufficient in vivo stability leading to poor efficacy is a common pitfall at which
many promising formulations fail, strategies like polymer conjugation to enhance in vivo
activity need to be developed. To this end, our conjugates offer a promising approach and
may be considered as therapeutic candidates for celiac disease. Tremendous effort has so
far been put into the characterization of the MX conjugates and a crucial next step to
demonstrate their clinical relevance will be to test their ability to prevent intestinal
inflammation upon gluten ingestion. Celiac disease rodent models have been established247
and could be used to test the efficacy of the bioconjugates and further evaluate the potential
of the protected vs. unprotected delivery of MX. Moreover, costs and scale-up challenges
need to be determined, which might be best estimated after comparison with similar
methods such as the random polymer conjugation of proteins. Finding a suitable
bioconjugate for each individual disease with the presented approach is more costly and
time-consuming as it implies engineering the enzyme, producing the respective conjugates
and then screening the generated library. The production of the enzyme itself will be equally
required for the modified and the native variant. The conjugation reaction can be very
similar in the two strategies; although in some cases a two-step reaction might be necessary
if the polymer is coupled to thiol groups for instance (like the PAMAM conjugation
presented in this work), which would ultimately increase production costs. However, once
a potent candidate is identified, the advantage of site-specific conjugation over randomly
attached polymers is conclusive. When polymers are randomly attached to proteins a
heterogeneous mixture is generally produced due to the accessibility of numerous residues
of the same kind. This not only leads to poor reproducibility and difficult characterization
of the conjugate but also complicates the purification of the bioconjugates into a single
defined product. In contrast, site-specific polymer attachment enables to produce a much
more homogenous product and only the excess of polymer in the reaction mixture needs
to be removed in order to produce a pure product. Moreover, another advantage of the
site-specific strategy is that it usually renders bioconjugates with higher activity. Thereby,
less material would need to be administered to patients to achieve the same effect, leading
90
Chapter 5: Conclusion &Outlook
to reduced production costs and an improved formulation’s safety profile which would be
vastly preferable in view of a potential translation to the clinic.
The strategy presented herein may be also applicable to the protection of other orally
applied therapeutic enzymes. Lactose intolerance is a very prominent example, currently
treated with an orally administered enzyme. In spite of the enzymatic drug’s widespread
use, its deactivation in the GI tract leads to reduced efficiency.248 To address this issue
random PEGylation was performed on the enzyme, which despite leading to enhanced
stability at acidic pH in vitro, it did not reach clinical relevance.249 Based on what was shown
in this work, the site-specific polymer conjugation of the enzyme could offer a more
efficient approach to tackle the deactivation in vivo. Moreover, attaching polymers sitespecifically might also help to develop therapeutics for other diseases treatable with orally
applied enzymes, such as other food intolerances, exocrine pancreatic insufficiency or
phenylketonuria.203 Given that protein-based medicines have become increasingly
important in pharmaceutical research and that oral administration is largely preferred
among patients, the site-specific polymer conjugation to proteins could be of interest to
many medical areas. Polymer conjugation has been previously performed on various
biomolecules not only to increase the stability in the GI tract but also to enhance drug
bioavailability (Figure 5.1).226 However, when the therapeutic target of the orally applied
protein is not in the GI tract itself but is located systemically, the challenges become even
more complex. The multiple barriers faced by this approach include the administration of
a sufficiently stable biomolecule to the intestine, its effective absorption through the
mucosa, the first-pass effect, and in some cases also the release of the drug at the desired
location in an active form. Increasing the bioavailability of orally administered biologics is
particularly difficult as the absorption mechanisms need to be altered and this can in turn
be associated with side effects such as damage of the mucosa, chronic disruption of
junctions and intestinal inflammation.95,250 The great effort to tackle the hurdles of orally
applied drugs with a systemic target is shown on the intense research performed on the
oral delivery of insulin, which despite episodical preclinical and clinical encouraging data is
still not available as an oral therapeutic treatment.251 The formulation of PEGylated insulin
reached even clinical trial Phase III (IN-105) but failed due to insufficient efficacy.251 Thus,
oral drug formulations may not be suitable or available for some diseases, but in the cases
where it is, polymer conjugation can contribute effectively to the improvement of
91
Chapter 5: Conclusion &Outlook
biopharmaceuticals.226 Additionally, other application routes could benefit from this
strategy because therapeutic proteins need to be stabilized in most biological environments.
Thereby, polymer conjugation does not only serve to protect proteins from degradation,
but offers also other positive effects such as preventing the drug’s aggregation, decreasing
its immunogenicity, enhancing its solubility or increasing its hydrodynamic volume to
prolong the plasma half-life due to reduced kidney clearance (Figure 5.1).252 Due to the
simplicity of random conjugation, this strategy has prevailed and has led to the approval of
some therapeutic bioconjugates (PEG-conjugated mainly).223 Thus, old formulations could
be optimized and new potential therapeutic proteins developed by attaching polymers sitespecifically. The strategy developed in this work could help to overcome the drawbacks
presented by established methods, giving rise to a new generation of therapeutic
bioconjugates.
92
Chapter 6
Appendix
Chapter 6: Appendix
a
MX WT:
ATGGGCAGCA
ATGTCCTACC
GCGGACCCGT
GCGCAGAACG
GCGCGCTTCA
CGCTTCTTCT
GGGGAGAGCG
GTGTCCCTGG
CCCAACGCCG
AAGGTGGACG
GGCTTCTATT
TACACCACCA
GAGCGCACCG
CTGTTCGTCT
GAGAAGGACT
AAGGACCGCT
GATCCGGCGA
TCCCTGCTGT
ACGTCCGAGG
GGCGTGGGCG
TTCACGTCCT
GAGCTCTGGG
TTCTACGCGT
AAGCGTGACG
GAGGCCAACT
GCCAACCTGC
AAGAAGCAGA
TACACGCAGC
GCGGCCATGA
GACATGGTGC
GCGGAGAAGC
CCGGACGTGC
CCCATGCACG
GCCCTGCTGC
ATTGAGTCCA
CAGGGTGGGG
CACCAC
GCCATCATCA
CGGCGACCCG
ATCGCTGGCT
CGCACGCCCG
AGGAGCTGTT
ACGTCCGCAC
GGCAGGAGAA
GGACGTGGGC
CGGATGAGGC
TCATCGAGGG
ACGAGTGGCT
TCCGCTACCA
GCGACCCGAC
ACATCCTCCG
TCCGCCTGCT
TCTACGTCCT
AGCCGGCCCG
CCGTCAGCAT
TGCGCGTGGC
CGGCGTCCAA
TCACCACGCC
CCAAGGTGGA
CCAAGGACGG
GCAACGCGCC
TCCGCTCGAG
GCGGCGGCGG
ACGTCTTCGA
CCAAGCGGCT
CGCAGCGGCC
GCTACCACCT
CCGAGGACTT
GCTACCCCGC
CCCGCAAGTT
GCATCGAGGC
GCGTGGACCT
TGGCGGCGCA
TCATCATCAC
GGCGGAGCAG
CGAGGACGAG
CGAAGCGCTG
CTACACCGAC
CCACAAGGAC
GGTGCTGTTG
CGTGTCCTGG
GGTGCTGCAC
CGGCAAGTAC
GCCCACGGAC
CACGCTGGGC
GACGTTCCTC
CGGCTGGAGC
GGTGAAGGGC
CACCGACGAG
CGCGTCGTGG
CGTCGGCGGG
CACGCTGAAG
CCTGATGGGG
CCGTCAAATC
CGTGCCCATG
GACGAAGGTG
CACGCTGCTC
CATCCTGCCC
CGAGTACGGC
CGACTTCCAC
GGCCATCTAC
GGAGCTGTAC
CTTCGGCAGC
CAAGACGCTG
GCTGCTGATG
CGTGGCGGCG
CAACGCGGGC
GTATTCGTTC
GGGCCGCAAG
AGCAGCGGCC
GTTGTCGACA
AAGGCCCCCG
GCGAAGTTCC
TCCGTCTCCA
AAGGAGAAGG
GATCCGAACG
GACGGCAAGA
GTCATCGACG
GCCACGCCCA
CCGTCCATCA
ACGGAGCCGT
CAGAGCGACC
GAGAACGACG
GTGGGCGCCA
GGCGCCCCGC
AAGGAGATTG
CACCTGTCGC
GGCAAGCCGG
CTGGAGGACC
TACAAGACGT
AACCCGGAGC
CCCATGTTCG
TACGGCTACG
TGGCTGGACG
AAGGCCTGGC
GCGGCGGCCG
GGCGGCAGCA
GGCGCGGTGG
GGCCGGACCT
CACGCCTACT
ATGGCGGCGG
GTGCAGAACT
CACGGTGGCG
CTGTTCCAAG
CTTGCGGCCG
TGGTGCCGCG
CGTTGCACGG
AGGTCCAGAC
CCGGCCGTGA
CCCCGTCGCG
CCATCCTCTA
GCTGGAGCAA
AGGTGGCCTT
TGGACTCTGG
AGTGGACGCC
AGGTGGACGA
CGAAGGACAC
TGAGCCGCGA
TCTACTGGAA
AGTACGAGGT
GCCAGCGCGT
TCCCCGAGGA
TGGAGTACCT
TGCGCACGGT
TGGATGACGC
CCGTCAGCAC
AGTACCAGGT
TGGTGCACCG
GCGGCTTCAA
CGGGCGGCGT
ACGACGCCGG
AGTACTTGGT
ACGGCGGCCT
TGTGCGCGGT
GGATTCCGGA
CGCCCTACCA
ACCACGACGA
CGCCCGGAAA
CGGATCAGGT
TCCTGGATGT
CACTCGAGCA
Figure A1a. Deoxyribonucleic acid (DNA) sequence of MX WT.
95
CGGCAGCCAT
CGTCCAGGTA
GTGGATGACG
GGCCCTGGCC
CCGCAACGGG
CTGGCGCCAG
GGACGGCACC
CGCCCAGAAG
CGAGTGGTCC
CGACAGCAAG
GCGCCCCGGC
CGTGGTGCAC
CGGCAAGTAC
GCGGCCGGGT
GCACGCCTGG
CTTCGAGGTG
CTCGTCCGCG
CAAGGACGCG
GCAGCTGCCG
CTACTACGTC
CGGGAAGTCT
CGAGCAGGTC
CAAGGACCTG
CGTGAACATG
GTACGCGGTG
CCGCCTGGAC
CCAGCAGAAG
GCTGGTGGGC
GCCCCTGCTG
GTACGGCACG
CCACGTGCGG
CCGGGTGGAC
CCCGGCGACG
GGCCAAGGCC
CCAGGGGGCA
CCACCACCAC
Chapter 6: Appendix
b
MX3.1: S123C, S338C, K461C
ATGGGCAGCA
ATGTCCTACC
GCGGACCCGT
GCGCAGAACG
GCGCGCTTCA
CGCTTCTTCT
GGGGAGTGTG
GTGTCCCTGG
CCCAACGCCG
AAGGTGGACG
GGCTTCTATT
TACACCACCA
GAGCGCACCG
CTGTTCGTCT
GAGAAGGACT
AAGGACCGCT
GATCCGGCGA
TCCCTGCTGT
ACGTCCGAGG
GGCGTGGGCG
TTCACGTCCT
GAGCTCTGGG
TTCTACGCGT
TGTCGTGACG
GAGGCCAACT
GCCAACCTGC
AAGAAGCAGA
TACACGCAGC
GCGGCCATGA
GACATGGTGC
GCGGAGAAGC
CCGGACGTGC
CCCATGCACG
GCCCTGCTGC
ATTGAGTCCA
CAGGGTGGGG
CACCAC
GCCATCATCA
CGGCGACCCG
ATCGCTGGCT
CGCACGCCCG
AGGAGCTGTT
ACGTCCGCAC
GGCAGGAGAA
GGACGTGGGC
CGGATGAGGC
TCATCGAGGG
ACGAGTGGCT
TCCGCTACCA
GCGACCCGAC
ACATCCTCCG
TCCGCCTGCT
TCTACGTCCT
AGCCGGCCCG
CCGTCAGCAT
TGCGCGTGGC
CGGCGTCCAA
TCACCACGCC
CCAAGGTGGA
CCAAGGACGG
GCAACGCGCC
TCCGCTCGAG
GCGGCGGCGG
ACGTCTTCGA
CCAAGCGGCT
CGCAGCGGCC
GCTACCACCT
CCGAGGACTT
GCTACCCCGC
CCCGCAAGTT
GCATCGAGGC
GCGTGGACCT
TGGCGGCGCA
TCATCATCAC
GGCGGAGCAG
CGAGGACGAG
CGAAGCGCTG
CTACACCGAC
CCACAAGGAC
GGTGCTGTTG
CGTGTCCTGG
GGTGCTGCAC
CGGCAAGTAC
GCCCACGGAC
CACGCTGGGC
GACGTTCCTC
CGGCTGGAGC
GGTGAAGGGC
CACCGACGAG
CGCGTCGTGG
CGTCGGCGGG
CACGCTGAAG
CCTGATGGGG
CCGTCAAATC
CGTGCCCATG
GACGAAGGTG
CACGCTGCTC
CATCCTGCCC
CGAGTACGGC
CGACTTCCAC
GGCCATCTAC
GGAGCTGTAC
CTTCGGCAGC
CAAGACGCTG
GCTGCTGATG
CGTGGCGGCG
CAACGCGGGC
GTATTCGTTC
GGGCCGCAAG
AGCAGCGGCC
GTTGTCGACA
AAGGCCCCCG
GCGAAGTTCC
TCCGTCTCCA
AAGGAGAAGG
GATCCGAACG
GACGGCAAGA
GTCATCGACG
GCCACGCCCA
CCGTCCATCA
ACGGAGCCGT
CAGAGCGACC
GAGAACGACG
GTGGGCGCCA
GGCGCCCCGC
AAGGAGATTG
CACCTGTCGC
GGCAAGCCGG
CTGGAGGACC
TACAAGACGT
AACCCGGAGC
CCCATGTTCG
TACGGCTACG
TGGCTGGACG
AAGGCCTGGC
GCGGCGGCCG
GGCGGCAGCA
GGCGCGGTGG
GGCCGGACCT
CACGCCTACT
ATGGCGGCGG
GTGCAGAACT
CACGGTGGCG
CTGTTCCAAG
CTTGCGGCCG
TGGTGCCGCG
CGTTGCACGG
AGGTCCAGAC
CCGGCCGTGA
CCCCGTCGCG
CCATCCTCTA
GCTGGAGCAA
AGGTGGCCTT
TGGACTCTGG
AGTGGACGCC
AGGTGGACGA
CGAAGGACAC
TGAGCCGCGA
TCTACTGGAA
AGTACGAGGT
GCCAGCGCGT
TCCCCGAGGA
TGGAGTACCT
TGCGCACGGT
TGGATGACGC
CCGTCAGCAC
AGTACCAGGT
TGGTGCACCG
GCGGCTTCAA
CGGGCGGCGT
ACGACGCCGG
AGTACTTGGT
ACGGCGGCCT
TGTGCGCGGT
GGATTCCGGA
CGCCCTACCA
ACCACGACGA
CGCCCGGAAA
CGGATCAGGT
TCCTGGATGT
CACTCGAGCA
CGGCAGCCAT
CGTCCAGGTA
GTGGATGACG
GGCCCTGGCC
CCGCAACGGG
CTGGCGCCAG
GGACGGCACC
CGCCCAGAAG
CGAGTGGTCC
CGACAGCAAG
GCGCCCCGGC
CGTGGTGCAC
CGGCAAGTAC
GCGGCCGGGT
GCACGCCTGG
CTTCGAGGTG
CTGTTCCGCG
CAAGGACGCG
GCAGCTGCCG
CTACTACGTC
CGGGAAGTCT
CGAGCAGGTC
CAAGGACCTG
CGTGAACATG
GTACGCGGTG
CCGCCTGGAC
CCAGCAGAAG
GCTGGTGGGC
GCCCCTGCTG
GTACGGCACG
CCACGTGCGG
CCGGGTGGAC
CCCGGCGACG
GGCCAAGGCC
CCAGGGGGCA
CCACCACCAC
Figure A1b. DNA sequence of MX3.1 containing three cysteine mutations. Mutations are highlighted in
color. The labeling of the mutations includes the original amino acid, the location in the amino acid sequence
and the mutated residue.
96
Chapter 6: Appendix
c
MX3.2: A25C, A328C, T417C
ATGGGCAGCA
ATGTCCTACC
GCGGACCCGT
GCGCAGAACG
GCGCGCTTCA
CGCTTCTTCT
GGGGAGAGCG
GTGTCCCTGG
CCCAACGCCG
AAGGTGGACG
GGCTTCTATT
TACACCACCA
GAGCGCACCG
CTGTTCGTCT
GAGAAGGACT
AAGGACCGCT
GATCCGTGTA
TCCCTGCTGT
ACGTCCGAGG
GGCGTGGGCG
TTCACGTCCT
GAGCTCTGGG
TTCTACGCGT
AAGCGTGACG
GAGGCCAACT
GCCAACCTGC
AAGAAGCAGA
TACACGCAGC
GCGGCCATGA
GACATGGTGC
GCGGAGAAGC
CCGGACGTGC
CCCATGCACG
GCCCTGCTGC
ATTGAGTCCA
CAGGGTGGGG
CACCAC
GCCATCATCA
CGTGTACCCG
ATCGCTGGCT
CGCACGCCCG
AGGAGCTGTT
ACGTCCGCAC
GGCAGGAGAA
GGACGTGGGC
CGGATGAGGC
TCATCGAGGG
ACGAGTGGCT
TCCGCTACCA
GCGACCCGAC
ACATCCTCCG
TCCGCCTGCT
TCTACGTCCT
AGCCGGCCCG
CCGTCAGCAT
TGCGCGTGGC
CGGCGTCCAA
TCACCACGCC
CCAAGGTGGA
CCAAGGACGG
GCAACGCGCC
TCCGCTCGAG
GCGGCGGCGG
ACGTCTTCGA
CCAAGCGGCT
CGCAGCGGCC
GCTACCACCT
CCGAGGACTT
GCTACCCCGC
CCCGCAAGTT
GCATCGAGGC
GCGTGGACCT
TGGCGGCGCA
TCATCATCAC
GGCGGAGCAG
CGAGGACGAG
CGAAGCGCTG
CTACACCGAC
CCACAAGGAC
GGTGCTGTTG
CGTGTCCTGG
GGTGCTGCAC
CGGCAAGTAC
GCCCACGGAC
CACGCTGGGC
GACGTTCCTC
CGGCTGGAGC
GGTGAAGGGC
CACCGACGAG
CGCGTCGTGG
CGTCGGCGGG
CACGCTGAAG
CCTGATGGGG
CCGTCAAATC
CGTGCCCATG
GACGAAGGTG
CACGCTGCTC
CATCCTGCCC
CGAGTACGGC
CGACTTCCAC
GGCCATCTAC
GGAGCTGTAC
CTTCGGCAGC
CAAGACGCTG
GCTGCTGATG
CGTGGCGGCG
CAACGCGGGC
GTATTCGTTC
GGGCCGCAAG
AGCAGCGGCC
GTTGTCGACA
AAGGCCCCCG
GCGAAGTTCC
TCCGTCTCCA
AAGGAGAAGG
GATCCGAACG
GACGGCAAGA
GTCATCGACG
GCCACGCCCA
CCGTCCATCA
ACGGAGCCGT
CAGAGCGACC
GAGAACGACG
GTGGGCGCCA
GGCGCCCCGC
AAGGAGATTG
CACCTGTCGC
GGCAAGCCGG
CTGGAGGACC
TACAAGACGT
AACCCGGAGC
CCCATGTTCG
TACGGCTACG
TGGCTGGACG
AAGGCCTGGC
GCGGCGGCCG
GGCGGCAGCA
GGCGCGGTGG
GGCCGGACCT
CACGCCTACT
ATGGCGGCGG
GTGCAGAACT
CACGGTGGCG
CTGTTCCAAG
CTTGCGGCCG
TGGTGCCGCG
CGTTGCACGG
AGGTCCAGAC
CCGGCCGTGA
CCCCGTCGCG
CCATCCTCTA
GCTGGAGCAA
AGGTGGCCTT
TGGACTCTGG
AGTGGACGCC
AGGTGGACGA
CGAAGGACAC
TGAGCCGCGA
TCTACTGGAA
AGTACGAGGT
GCCAGCGCGT
TCCCCGAGGA
TGGAGTACCT
TGCGCACGGT
TGGATGACGC
CCGTCAGCTG
AGTACCAGGT
TGGTGCACCG
GCGGCTTCAA
CGGGCGGCGT
ACGACGCCGG
AGTACTTGGT
ACGGCGGCCT
TGTGCGCGGT
GGATTCCGGA
CGCCCTACCA
ACCACGACGA
CGCCCGGAAA
CGGATCAGGT
TCCTGGATGT
CACTCGAGCA
CGGCAGCCAT
CGTCCAGGTA
GTGGATGACG
GGCCCTGGCC
CCGCAACGGG
CTGGCGCCAG
GGACGGCACC
CGCCCAGAAG
CGAGTGGTCC
CGACAGCAAG
GCGCCCCGGC
CGTGGTGCAC
CGGCAAGTAC
GCGGCCGGGT
GCACGCCTGG
CTTCGAGGTG
CTCGTCCGCG
CAAGGACGCG
GCAGCTGCCG
CTACTACGTC
TGGGAAGTCT
CGAGCAGGTC
CAAGGACCTG
CGTGAACATG
GTACGCGGTG
CCGCCTGGAC
CCAGCAGAAG
GCTGGTGGGC
GCCCCTGCTG
GTACGGCACG
CCACGTGCGG
CCGGGTGGAC
CCCGGCGACG
GGCCAAGGCC
CCAGGGGGCA
CCACCACCAC
Figure A1c. DNA sequence of MX3.2 containing three cysteine mutations. Mutations are highlighted in
color.
97
Chapter 6: Appendix
a
MX WT:
MGSSHHHHHH
AQNAHAREAL
GESGQEKVLL
KVDVIEGGKY
ERTGDPTTFL
KDRFYVLTDE
TSEVRVATLK
ELWAKVDVPM
EANFRSSILP
YTQPKRLAIY
AEKPEDFKTL
ALLRIEANAG
HH
SSGLVPRGSH
AKFPGREALA
DPNGWSKDGT
ATPKWTPDSK
QSDLSRDGKY
GAPRQRVFEV
GKPVRTVQLP
NPEQYQVEQV
WLDAGGVYAV
GGSNGGLLVG
HAYSPYHHVR
HGGADQVAKA
MSYPATRAEQ
ARFKELFYTD
VSLGTWAVSW
GFYYEWLPTD
LFVYILRGWS
DPAKPARASW
GVGAASNLMG
FYASKDGTKV
ANLRGGGEYG
AAMTQRPELY
PDVRYPALLM
IESSVDLYSF
VVDTLHGVQV
SVSTPSRRNG
DGKKVAFAQK
PSIKVDERPG
ENDVYWKRPG
KEIVPEDSSA
LEDLDDAYYV
PMFVVHRKDL
KAWHDAGRLD
GAVVCAVPLL
MAADHDDRVD
LFQVLDVQGA
ADPYRWLEDE
RFFYVRTHKD
PNAADEAVLH
YTTIRYHTLG
EKDFRLLVKG
SLLSVSIVGG
FTSFTTPRQI
KRDGNAPTLL
KKQNVFDDFH
DMVRYHLFGS
PMHARKFVAA
QGGVAAQGRK
KAPEVQTWMT
KEKAILYWRQ
VIDVDSGEWS
TEPSKDTVVH
VGAKYEVHAW
HLSLEYLKDA
YKTSVSTGKS
YGYGGFNVNM
AAAEYLVQQK
GRTWIPEYGT
VQNSPGNPAT
LAAALEHHHH
VVDTLHGVQV
SVSTPSRRNG
DGKKVAFAQK
PSIKVDERPG
ENDVYWKRPG
KEIVPEDCSA
LEDLDDAYYV
PMFVVHRKDL
KAWHDAGRLD
GAVVCAVPLL
MAADHDDRVD
LFQVLDVQGA
ADPYRWLEDE
RFFYVRTHKD
PNAADEAVLH
YTTIRYHTLG
EKDFRLLVKG
SLLSVSIVGG
FTSFTTPRQI
CRDGNAPTLL
KKQNVFDDFH
DMVRYHLFGS
PMHARKFVAA
QGGVAAQGRK
KAPEVQTWMT
KEKAILYWRQ
VIDVDSGEWS
TEPSKDTVVH
VGAKYEVHAW
HLSLEYLKDA
YKTSVSTGKS
YGYGGFNVNM
AAAEYLVQQK
GRTWIPEYGT
VQNSPGNPAT
LAAALEHHHH
VVDTLHGVQV
SVSTPSRRNG
DGKKVAFAQK
PSIKVDERPG
ENDVYWKRPG
KEIVPEDSSA
LEDLDDAYYV
PMFVVHRKDL
KAWHDAGRLD
GAVVCAVPLL
MAADHDDRVD
LFQVLDVQGA
ADPYRWLEDE
RFFYVRTHKD
PNAADEAVLH
YTTIRYHTLG
EKDFRLLVKG
SLLSVSIVGG
FTSFTTPRQI
KRDGNAPTLL
KKQNVFDDFH
DMVRYHLFGS
PMHARKFVAA
QGGVAAQGRK
KAPEVQTWMT
KEKAILYWRQ
VIDVDSGEWS
TEPSKDTVVH
VGAKYEVHAW
HLSLEYLKDA
YKTSVSCGKS
YGYGGFNVNM
AAAEYLVQQK
GRTWIPEYGT
VQNSPGNPAT
LAAALEHHHH
b
MX3.1: S123C, S338C, K461C
MGSSHHHHHH
AQNAHAREAL
GECGQEKVLL
KVDVIEGGKY
ERTGDPTTFL
KDRFYVLTDE
TSEVRVATLK
ELWAKVDVPM
EANFRSSILP
YTQPKRLAIY
AEKPEDFKTL
ALLRIEANAG
HH
SSGLVPRGSH
AKFPGREALA
DPNGWSKDGT
ATPKWTPDSK
QSDLSRDGKY
GAPRQRVFEV
GKPVRTVQLP
NPEQYQVEQV
WLDAGGVYAV
GGSNGGLLVG
HAYSPYHHVR
HGGADQVAKA
MSYPATRAEQ
ARFKELFYTD
VSLGTWAVSW
GFYYEWLPTD
LFVYILRGWS
DPAKPARASW
GVGAASNLMG
FYASKDGTKV
ANLRGGGEYG
AAMTQRPELY
PDVRYPALLM
IESSVDLYSF
c
MX3.2: A25C, A328C, T417C
MGSSHHHHHH
AQNAHAREAL
GESGQEKVLL
KVDVIEGGKY
ERTGDPTTFL
KDRFYVLTDE
TSEVRVATLK
ELWAKVDVPM
EANFRSSILP
YTQPKRLAIY
AEKPEDFKTL
ALLRIEANAG
HH
SSGLVPRGSH
AKFPGREALA
DPNGWSKDGT
ATPKWTPDSK
QSDLSRDGKY
GAPRQRVFEV
GKPVRTVQLP
NPEQYQVEQV
WLDAGGVYAV
GGSNGGLLVG
HAYSPYHHVR
HGGADQVAKA
MSYPCTRAEQ
ARFKELFYTD
VSLGTWAVSW
GFYYEWLPTD
LFVYILRGWS
DPCKPARASW
GVGAASNLMG
FYASKDGTKV
ANLRGGGEYG
AAMTQRPELY
PDVRYPALLM
IESSVDLYSF
Figure A2. Amino acid sequence of a) MX WT and the two variants, b) MX3.1 and c) MX3.2 containing
three cysteine mutations. Mutations are highlighted in color.
98
Chapter 6: Appendix
a
SC WT:
ATGAAGAACC
TTTGCCCAGA
CCCGCCAGCC
TGGCGCTGGC
CAAAGCGCCT
CGGATGAAGG
GTCTTCTACA
GATGCGCCGG
GGCGCCACTG
GTGCAGGATG
CCGCTGGCCG
GCGCTGCTCT
TATAACCAGA
TTCGCGACGC
TGGGTGGTGA
GTCACCAACG
GATTTCGTCG
AAGAAGATCG
GAAAGCAAGG
ATCCACGATG
GTCAGCCTGC
GCCTATCTCT
ACGGCGAAAA
GTGGAGCAGG
CGCAAGGATG
GCGCTTACCC
GCCCTGGCAA
CGCGACAAGA
AATGGCGTGA
ATCGGCGCGG
GTGATGGACA
GGCTATCCCG
GTGCGTTCGG
GTCGTGCCGG
AAGCCGCACC
AAGCAGATCG
CCGCGCCCCC
GCTTGTGGCT
CGCCGCCCAC
CGCAAGTGCC
TGGAAGCCGA
ATACCGCAGC
CGCTGATCGA
GCTGGAACAG
TGGGCACCAA
CGCTCGATGC
GCGGTTCGGA
ACGAACTCAA
ATTCCCGCTT
CGGTGTGGCT
CCGAGCTGCC
TCACCAGCAG
GCAAGATTGG
ATGGCGTGGG
TGCGGGTCGA
ACAACCTCGA
CCAAGAGCCA
CCGGCATCGG
CGTTCAGCAG
CCACCCCGTG
TGTTCTATCC
CCAAGGGCCC
CATGGTTTTC
ACCTACGCGG
AGCAGAACGT
CGCCGCGCCA
TGACCAACCA
TGCTGCGCTT
AAAAGGAAGC
GCGTGGACTA
GGCACAGCTT
TGATCCGCAT
AGGAAACCGC
TCGAGCACCA
GGCCATGGCC
CCTTGCCAAG
GCTGGTGGAG
NGTGCGCACC
CTATCTCAAG
TTACGAACGG
CNGCTTGATG
GGGACGGGTA
CTGGGCGGCA
TTGGCGGACG
NTGGGTGAAG
TGCCGAGCCA
CCACCGCCTG
CAAGCGCGGC
CGAGGGCACC
GCCGGTTACC
CGATCAGCTG
TCTGTCGGGC
ATCCGTCGGC
GGTCCTGGCG
CAGCGCCTCG
CTTTACCCAG
GGAGCCGGTG
CAGCAANGAC
GCTGCCCACG
CGCCGGTTTC
CGGCGGCGAA
CTTCGACGAT
CGGCCTGGCG
GCGGCCCGAT
CGACCAGTTC
CGATTGGCGC
TCCGGCCATC
CAAATATACT
CGAGACGCGC
CGATGTGCAG
CCACCACCAC
GCCCCGCTTG
GACCAGGCCA
GACCATTTCG
GATGCCAAGG
CAGTTGCCCG
TTTGGCCTGC
AACCAGTCGC
CTGCTCGATC
TCNGATGACG
GTGAAGTTTG
TTTTCCGGCC
AAGGAAGGCC
GGCACGCCGC
CACGGCGCCA
GATCCGGTCA
GCACTGATCC
TGGTTCGTGT
TGTACGCCGC
ATTGCCGGCA
TTCGATCTTG
GGCCTGAGCG
CCTGCCACGG
CACCTGACGT
GGCACCAAGG
CTGCTCTATG
ATGACCTGGA
TATGGCGATG
TTCATTGCCG
ATCGAGGGTG
CTGTTTGCGG
ACCGCCGGGC
GTGCTGCGCC
CTGNTGACCA
GCGGCGCTGC
GCCGGCCATG
GCGTTTCTGG
CAC
CGCTTGCCAC
TGCCGTCACT
GTGAAAAGGT
TGGCGGCCTG
AGCGTTGTGC
CGCAGCGCCG
AGCTGCTGGT
CCAATACCTG
GGCGCCTGCT
TCGGCGTGGC
TGGCCTGGCT
AGGCCTTCCA
AAAGCGCCGA
GCGTTTCGAG
ACACCGTGCA
CCGATCTCAA
CGGGCGATGG
GCTTCGATAC
ACCGCCTGTT
ATGGCAAGCC
GCCGGCCCGG
TGCTGGCGCT
TCGATCCGGC
TGCCGATGTT
GCTATGGCGG
TCGACAGCGG
CGTGGCACGA
CGGGCGAATG
GCTCCAACGG
CGGCCAGCCC
GCTATTGGGT
GCTATTCGCC
CGGCCGACAC
AGACCGCCGC
GTTCGGGCAA
CCCATTTCAC
CCCGGTTGCG
GCCGCCCTAT
GTCCGATCCC
GGTCCAGGCG
GCTGGAAAAG
TGGCGCTTCG
GCGCCCTGCC
GGCCAAGGAT
GGCCTATTCG
CGACGGCAAG
GGGCAACGAT
GGCGCTGAAC
TCAGCCGGTG
CGATGGGCGC
CGTTGCGCGC
GGCGCAATGG
CGCCCCGCTC
CGTGGTGCCC
CGCCAGCTAT
GGCCGGTGCG
CGATCGCCAT
CGATCCCGCC
CGATTTCCGG
CATCGTGCGC
CTTCAATGTG
CGGCGCCTTT
TGCCGGCCGG
GCTGATCTGT
TGGCCTGCTG
TGCCGTGGGC
GGATGACTAT
CTATCACAAC
CGACGATCGC
GATCGGGCCC
GCCGATCGAC
CGGGCTGACG
Figure A3a. DNA sequence of SC WT. The two start codons are highlighted in blue.
99
Chapter 6: Appendix
b
SC3.1: A96C, S351C, A560C
ATGAAGAACC
TTTGCCCAGA
CCCGCCAGCC
TGGCGCTGGC
CAAAGCGCCT
CGGATGAAGG
GTCTTCTACA
GATGCGCCGG
GGCGCCACTG
GTGCAGGATG
CCGCTGGCCG
GCGCTGCTCT
TATAACCAGA
TTCGCGACGC
TGGGTGGTGA
GTCACCAACG
GATTTCGTCG
AAGAAGATCG
GAAAGCAAGG
ATCCACGATG
GTCAGCCTGC
GCCTATCTCT
ACGGCGAAAA
GTGGAGCAGG
CGCAAGGATG
GCGCTTACCC
GCCCTGGCAA
CGCGACAAGA
AATGGCGTGA
ATCGGCGCGG
GTGATGGACA
GGCTATCCCG
GTGCGTTCGG
GTCGTGCCGG
AAGCCGCACC
AAGCAGATCG
CCGCGCCCCC
GCTTGTGGCT
CGCCGCCCAC
CGCAAGTGCC
TGGAAGCCGA
ATACCGCAGC
CGCTGATCGA
GCTGGAACAG
TGGGCACCAA
CGCTCGATGC
GCGGTTCGGA
ACGAACTCAA
ATTCCCGCTT
CGGTGTGGCT
CCGAGCTGCC
TCACCAGCAG
GCAAGATTGG
ATGGCGTGGG
TGCGGGTCGA
ACAACCTCGA
CCAAGAGCCA
CCGGCATCGG
CGTTCAGCAG
CCACCCCGTG
TGTTCTATCC
CCAAGGGCCC
CATGGTTTTC
ACCTACGCGG
AGCAGAACGT
CGCCGCGCCA
TGACCAACCA
TGCTGCGCTT
AAAAGGAAGC
GCGTGGACTA
GGCACAGCTT
TGATCCGCAT
AGGAAACCGC
TCGAGCACCA
GGCCATGGCC
CCTTGCCAAG
GCTGGTGGAG
CGTGCGCACC
CTATCTCAAG
TTACGAACGG
CGGCTTGATG
GGGACGGGTA
CTGGGCGGCA
TTGGCGGACG
GTGGGTGAAG
TGCCGAGCCA
CCACCGCCTG
CAAGCGCGGC
CGAGGGCACC
GCCGGTTACC
CGATCAGCTG
TCTGTCGGGC
ATCCGTCGGC
GGTCCTGGCG
CAGCGCCTCG
CTTTACCCAG
GGAGCCGGTG
CAGCAAGGAC
GCTGCCCACG
CGCCGGTTTC
CGGCGGCGAA
CTTCGACGAT
CGGCCTGGCG
GCGGCCCGAT
CGACCAGTTC
CGATTGGCGC
TCCGGCCATC
CAAATATACT
CGAGACGCGC
CGATGTGCAG
CCACCACCAC
GCCCCGCTTG
GACCAGGCCA
GACCATTTCG
GATGCCAAGG
CAGTTGCCCG
TTTGGCCTGC
AACCAGTCGC
CTGCTCGATC
TCGGATGACG
GTGAAGTTTG
TTTTCCGGCC
AAGGAAGGCC
GGCACGCCGC
CACGGCGCCA
GATCCGGTCA
GCACTGATCC
TGGTTCGTGT
TGTACGCCGC
ATTGCCGGCA
TTCGATCTTG
GGCCTGAGCG
CCTGCCACGG
CACCTGACGT
GGCACCAAGG
CTGCTCTATG
ATGACCTGGA
TATGGCGATG
TTCATTGCCG
ATCGAGGGTG
CTGTTTGCGG
ACCGCCGGGC
GTGCTGCGCC
CTGGTGACCA
GCGGCGCTGC
GCCGGCCATG
GCGTTTCTGG
CAC
CGCTTGCCAC
TGCCGTCACT
GTGAAAAGGT
TGGCGGCCTG
AGCGTTGTGC
CGCAGCGCCG
AGCTGCTGGT
CCAATACCTG
GGCGCCTGCT
TCGGCGTGGC
TGGCCTGGCT
AGGCCTTCCA
AAAGCGCCGA
GCGTTTCGAG
ACACCGTGCA
CCGATCTCAA
CGGGCGATGG
GCTTCGATAC
ACCGCCTGTT
ATGGCAAGCC
GCCGGCCCGG
TGCTGGCGCT
TCGATCCGGC
TGCCGATGTT
GCTATGGCGG
TCGACAGCGG
CGTGGCACGA
CGGGCGAATG
GCTCCAACGG
CGGCCAGCCC
GCTATTGGGT
GCTATTCGCC
CGGCCGACAC
AGACCGCCGC
GTTCGGGCAA
CCCATTTCAC
CCCGGTTGCG
GCCGCCCTAT
GTCCGATCCC
GGTCCAGGCG
GCTGGAAAAG
TGGCGCTTCG
GCGCCCTGCC
GGCCAAGGAT
GGCCTATTCG
CGACGGCAAG
GGGCAACGAT
GGCGCTGAAC
TCAGCCGGTG
CGATGGGCGC
CGTTGCGCGC
GGCGCAATGG
CGCCCCGCTC
CGTGGTGCCC
CGCCAGCTAT
GGCCGGTGCG
CGATCGCCAT
CGATCCCGCC
CGATTTCCGG
CATCGTGCGC
CTTCAATGTG
CGGCGCCTTT
TGCCGGCCGG
GCTGATCTGT
TGGCCTGCTG
TGCCGTGGGC
GGATGACTAT
CTATCACAAC
CGACGATCGC
GATCGGGCCC
GCCGATCGAC
CGGGCTGACG
Figure A3b. DNA sequence of SC3.1 containing three cysteine mutations. Mutations are highlighted in
color.
100
Chapter 6: Appendix
c
SC3.2: A351C, T441C, A560C
ATGAAGAACC
TTTGCCCAGA
CCCGCCAGCC
TGGCGCTGGC
CAAAGCGCCT
CGGATGAAGG
GTCTTCTACA
GATGCGCCGG
GGCGCCACTG
GTGCAGGATG
CCGCTGGCCG
GCGCTGCTCT
TATAACCAGA
TTCGCGACGC
TGGGTGGTGA
GTCACCAACG
GATTTCGTCG
AAGAAGATCG
GAAAGCAAGG
ATCCACGATG
GTCAGCCTGC
GCCTATCTCT
TGTGCGAAAA
GTGGAGCAGG
CGCAAGGATG
GCGCTTACCC
GCCCTGGCAA
CGCGACAAGA
AATGGCGTGA
ATCGGCGCGG
GTGATGGACA
GGCTATCCCG
GTGCGTTCGG
GTCGTGCCGG
AAGCCGCACC
AAGCAGATCG
CCGCGCCCCC
GCTTGTGGCT
CGCCGCCCAC
CGCAAGTGCC
TGGAAGCCGA
ATACCGCAGC
CGCTGATCGA
GCTGGAACAG
TGGGCACCAA
CGCTCGATGC
GCGGTTCGGA
ACGAACTCAA
ATTCCCGCTT
CGGTGTGGCT
CCGAGCTGCC
TCACCAGCAG
GCAAGATTGG
ATGGCGTGGG
TGCGGGTCGA
ACAACCTCGA
CCAAGAGCCA
CCGGCATCGG
CGTTCAGCAG
CCACCCCGTG
TGTTCTATCC
CCAAGGGCCC
CATGGTTTTC
ACCTACGCGG
AGCAGAACGT
CGCCGCGCCA
TGACCAACCA
TGCTGCGCTT
AAAAGGAAGC
GCGTGGACTA
GGCACAGCTT
TGATCCGCAT
AGGAAACCGC
TCGAGCACCA
GGCCATGGCC
CCTTGCCAAG
GCTGGTGGAG
CGTGCGCACC
CTATCTCAAG
TTACGAACGG
CGGCTTGATG
GGGACGGGTA
CTGGGCGGCA
TTGGCGGACG
GTGGGTGAAG
TGCCGAGCCA
CCACCGCCTG
CAAGCGCGGC
CGAGGGCACC
GCCGGTTACC
CGATCAGCTG
TCTGTCGGGC
ATCCGTCGGC
GGTCCTGGCG
CAGCGCCTCG
CTTTACCCAG
GGAGCCGGTG
CAGCAAGGAC
GCTGCCCACG
CGCCGGTTTC
CGGCGGCGAA
CTTCGACGAT
CGGCCTGGCG
GCGGCCCGAT
CGACCAGTTC
CGATTGGCGC
TCCGGCCATC
CAAATATACT
CGAGACGCGC
CGATGTGCAG
CCACCACCAC
GCCCCGCTTG
GACCAGGCCA
GACCATTTCG
GATGCCAAGG
CAGTTGCCCG
TTTGGCCTGC
AACCAGTCGC
CTGCTCGATC
TCGGATGACG
GTGAAGTTTG
TTTTCCGGCC
AAGGAAGGCC
GGCACGCCGC
CACGGCGCCA
GATCCGGTCA
GCACTGATCC
TGGTTCGTGT
TGTACGCCGC
ATTGCCGGCA
TTCGATCTTG
GGCCTGAGCG
CCTGCCACGG
CACCTGACGT
GGCACCAAGG
CTGCTCTATG
ATGACCTGGA
TATGGCGATG
TTCATTGCCG
ATCGAGGGTG
CTGTTTGCGG
ACCGCCGGGC
GTGCTGCGCC
CTGGTGACCA
GCGGCGCTGC
GCCGGCCATG
GCGTTTCTGG
CAC
CGCTTGCCAC
TGCCGTCACT
GTGAAAAGGT
TGGCGGCCTG
AGCGTGCCGC
CGCAGCGCCG
AGCTGCTGGT
CCAATACCTG
GGCGCCTGCT
TCGGCGTGGC
TGGCCTGGCT
AGGCCTTCCA
AAAGCGCCGA
GCGTTTCGAG
ACACCGTGCA
CCGATCTCAA
CGGGCGATGG
GCTTCGATAC
ACCGCCTGTT
ATGGCAAGCC
GCCGGCCCGG
TGCTGGCGCT
TCGATCCGGC
TGCCGATGTT
GCTATGGCGG
TCGACAGCGG
CGTGGCACGA
CGGGCGAATG
GCTCCAACGG
CGGCCAGCCC
GCTATTGGGT
GCTATTCGCC
CGGCCGACAC
AGACCGCCGC
GTTCGGGCAA
CCCATTTCAC
CCCGGTTGCG
GCCGCCCTAT
GTCCGATCCC
GGTCCAGGCG
GCTGGAAAAG
TGGCGCTTCG
GCGCCCTGCC
GGCCAAGGAT
GGCCTATTCG
CGACGGCAAG
GGGCAACGAT
GGCGCTGAAC
TCAGCCGGTG
CGATGGGCGC
CGTTGCGCGC
GGCGCAATGG
CGCCCCGCTC
CGTGGTGCCC
CGCCAGCTAT
GGCCGGTGCG
CGATCGCCAT
CGATCCCGCC
CGATTTCCGG
CATCGTGCGC
CTTCAATGTG
CGGCGCCTTT
TGCCGGCCGG
GCTGATCTGT
TGGCCTGCTG
TGCCGTGGGC
GGATGACTAT
CTATCACAAC
CGACGATCGC
GATCGGGCCC
GCCGATCGAC
CGGGCTGACG
Figure A3c. DNA sequence of SC3.2 containing three cysteine mutations. Mutations are highlighted in
color.
101
Chapter 6: Appendix
a
SC WT:
MKNRLWLAMA
WRWLEADVRT
VFYSWNSGLM
VQDGGSDWRT
YNQTVWLHRL
VTNGKIGPVT
ESKDNLESVG
AYLSFSSFTQ
RKDAKGPLPT
RDKKQNVFDD
VMDMLRFDQF
VVPGHSFKYT
PRPLEHHHHH
APLALATPVA
DAKVAAWVQA
NQSQLLVRPA
VKFVGVADGK
GTPQSADQPV
ALIPDLKAQW
IAGNRLFASY
PATVLALDPA
LLYGYGGFNV
FIAAGEWLIA
TAGRYWVDDY
AALQTAAIGP
H
FAQTPPTLAK
QSAYTAAYLK
DAPVGTKGRV
PLADELKWVK
FATPELPKRG
DFVDGVGDQL
IHDAKSQVLA
TAKTTPWEPV
ALTPWFSAGF
NGVTPRHGLA
GYPEKEADWR
KPHLIRIETR
DQAMPSLPPY
QLPERAALEK
LLDPNTWAKD
FSGLAWLGND
HGASVSSDGR
WFVSGDGAPL
FDLDGKPAGA
HLTFDPADFR
MTWIDSGGAF
IEGGSNGGLL
VLRRYSPYHN
AGHGSGKPID
PASPQVPLVE
RMKALIDYER
GATALDAWAA
ALLYSRFAEP
WVVITSSEGT
KKIVRVDLSG
VSLPGIGSAS
VEQVFYPSKD
ALANLRGGGE
IGAVTNQRPD
VRSGVDYPAI
KQIEETADVQ
DHFGEKVSDP
FGLPQRRGAS
SDDGRLLAYS
KEGQAFQALN
DPVNTVHVAR
STPRFDTVVP
GLSGRPGDRH
GTKVPMFIVR
YGDAWHDAGR
LFAAASPAVG
LVTTADTDDR
AFLAHFTGLT
DQAMPSLPPY
QLPERCALEK
LLDPNTWAKD
FSGLAWLGND
HGASVSSDGR
WFVSGDGAPL
FDLDGKPAGA
HLTFDPADFR
MTWIDSGGAF
IEGGSNGGLL
VLRRYSPYHN
AGHGSGKPID
PASPQVPLVE
RMKALIDYER
GATALDAWAA
ALLYSRFAEP
WVVITSSEGT
KKIVRVDLSG
VSLPGIGSAS
VEQVFYPSKD
ALANLRGGGE
IGAVTNQRPD
VRSGVDYPAI
KQIEETADVQ
DHFGEKVSDP
FGLPQRRGAS
SDDGRLLAYS
KEGQAFQALN
DPVNTVHVAR
CTPRFDTVVP
GLSGRPGDRH
GTKVPMFIVR
YGDAWHDAGR
LFAAASPAVG
LVTTADTDDR
AFLAHFTGLT
DQAMPSLPPY
QLPERAALEK
LLDPNTWAKD
FSGLAWLGND
HGASVSSDGR
WFVSGDGAPL
FDLDGKPAGA
HLTFDPADFR
MTWIDSGGAF
IEGGSNGGLL
VLRRYSPYHN
AGHGSGKPID
PASPQVPLVE
RMKALIDYER
GATALDAWAA
ALLYSRFAEP
WVVITSSEGT
KKIVRVDLSG
VSLPGIGSAS
VEQVFYPSKD
ALANLRGGGE
IGAVTNQRPD
VRSGVDYPAI
KQIEETADVQ
DHFGEKVSDP
FGLPQRRGAS
SDDGRLLAYS
KEGQAFQALN
DPVNTVHVAR
CTPRFDTVVP
GLSGRPGDRH
GTKVPMFIVR
YGDAWHDAGR
LFAAASPAVG
LVTTADTDDR
AFLAHFTGLT
b
SC3.1: A96C, S351C, A560C
MKNRLWLAMA
WRWLEADVRT
VFYSWNSGLM
VQDGGSDWRT
YNQTVWLHRL
VTNGKIGPVT
ESKDNLESVG
AYLSFSSFTQ
RKDAKGPLPT
RDKKQNVFDD
VMDMLRFDQF
VVPGHSFKYT
PRPLEHHHHH
APLALATPVA
DAKVAAWVQA
NQSQLLVRPA
VKFVGVADGK
GTPQSADQPV
ALIPDLKAQW
IAGNRLFASY
PATVLALDPA
LLYGYGGFNV
FIAAGEWLIC
TAGRYWVDDY
AALQTAAIGP
H
FAQTPPTLAK
QSAYTAAYLK
DAPVGTKGRV
PLADELKWVK
FATPELPKRG
DFVDGVGDQL
IHDAKSQVLA
TAKTTPWEPV
ALTPWFSAGF
NGVTPRHGLA
GYPEKEADWR
KPHLIRIETR
c
SC3.2: A351C, T441C, A560C
MKNRLWLAMA
WRWLEADVRT
VFYSWNSGLM
VQDGGSDWRT
YNQTVWLHRL
VTNGKIGPVT
ESKDNLESVG
AYLSFSSFTQ
RKDAKGPLPT
RDKKQNVFDD
VMDMLRFDQF
VVPGHSFKYT
PRPLEHHHHH
APLALATPVA
DAKVAAWVQA
NQSQLLVRPA
VKFVGVADGK
GTPQSADQPV
ALIPDLKAQW
IAGNRLFASY
PATVLALDPA
LLYGYGGFNV
FIAAGEWLIC
TAGRYWVDDY
AALQTAAIGP
H
FAQTPPTLAK
QSAYTAAYLK
DAPVGTKGRV
PLADELKWVK
FATPELPKRG
DFVDGVGDQL
IHDAKSQVLA
CAKTTPWEPV
ALTPWFSAGF
NGVTPRHGLA
GYPEKEADWR
KPHLIRIETR
Figure A4. Amino acid sequence of a) SC WT and the two variants, b) SC3.1 and c) SC3.2 containing
three cysteine mutations. Mutations are highlighted in color.
102
Chapter 6: Appendix
8
MX WT
MX3.1
MX3.2
6
CD (mdeg)
4
2
0
-2
-4
-6
-8
200
220
240
Wavelength (nm)
Figure A5. Circular dichroism (CD) spectra of MX WT, MX3.1 and MX3.2. Mean (n = 3).
103
Chapter 6: Appendix
a
b
c
Figure A6. 1H NMR spectra of a) sulfo-GMBS, b) GMBS-modified PAMAM G3, and c) PAMAM G3
(D2O, 400 MHz, Bruker Av-400). One intact maleimide per dendrimer was quantified.
104
Chapter 6: Appendix
a
b
Figure A7. Analytical HPLC analysis of a) MX WT, PEG5 and MX3.1–PEG5 and b) MX WT, PEG40
and MX3.1–PEG40.
105
Chapter 6: Appendix
a
b
Figure A8. Analytical HPLC analysis of a) GMBS, PAMAM and PAMAM–GMBS and b) MX WT,
PAMAM and MX3.1–PAMAM.
106
Chapter 6: Appendix
Figure A9. Enzymatic stability in simulated intestinal fluid: a) MX3.1 and MX3.1–polymer, b) MX3.2
and MX3.2–polymer, c) MX3.1–ran –PEG5–PAMAMs. While incubated with pancreases at pH 6.8,
activity was assed at specific time point. The graphs show the fitting of the deactivation over time. Mean (n
= 3).
107
Chapter 7
References
Chapter 7: References
1.
Creegan, R., Hunt, W., McManus, A. & Rainey-Smith, S. R. Diet, nutrients and
metabolism: cogs in the wheel driving Alzheimer’s disease pathology? Br. J. Nutr. 113,
1499–517 (2015).
2.
Wicki, A. & Hagmann, J. Diet and cancer. Swiss Med. Wkly. 141, w13250 (2011).
3.
Newmark, H. L. et al. A Western-style diet induces benign and malignant neoplasms in the
colon of normal C57Bl/6 mice. Carcinogenesis 22, 1871–75 (2001).
4.
Anand, P. et al. Cancer is a preventable disease that requires major lifestyle changes. Pharm.
Res. 25, 2097–116 (2008).
5.
Nwaru, B. I. et al. Prevalence of common food allergies in Europe: A systematic review
and meta-analysis. Eur. J. Allergy Clin. Immunol. 69, 992–1007 (2014).
6.
De Filippo, C. et al. Impact of diet in shaping gut microbiota revealed by a comparative
study in children from Europe and rural Africa. Proc. Natl. Acad. Sci. 107, 14691–96 (2010).
7.
Turnbaugh, P. J. et al. The effect of Diet on the Human Gut Microbiom: A Metagenomic
Analysis in Humanized Gnotobiotic Mice. Sci. Transl. Med. 1, 1–19 (2009).
8.
Maslowski, K. M. & Mackay, C. R. Diet, gut microbiota and immune responses. Nat.
Immunol. 12, 5–9 (2011).
9.
Jandhyala, S. M. et al. Role of the normal gut microbiota. World J. Gastroenterol. 21, 8787–
803 (2015).
10.
Turnbull, J. L., Adams, H. N. & Gorard, D. A. Review article: the diagnosis and
management of food allergy and food intolerances. Aliment. Pharmacol. Ther. 41, 3–25
(2015).
11.
World Health Organization. Global strategy for infant and young child feeding: The optimal duration
of exclusive breastfeeding. Fifty-Fourth World Health Assembly (2001).
12.
Umetsu, D. T., Rachid, R. & Schneider, L. C. Oral immunotherapy and anti-IgE antibody
treatment for food allergy. World Allergy Organ. J. 8, 20 (2015).
13.
Gujral, N., Freeman, H. J. & Thomson, A. B. R. Celiac disease: Prevalence, diagnosis,
pathogenesis and treatment. World J. Gastroenterol. 18, 6036–59 (2012).
14.
Molina-Infante, J., Santolaria, S., Sanders, D. S. & Fernández-Bañares, F. Systematic
review: noncoeliac gluten sensitivity. Aliment. Pharmacol. Ther. 41, 807–20 (2015).
109
Chapter 7: References
15.
Vesa, T. H., Marteau, P. & Korpela, R. Lactose Intolerance. J. Am. Coll. Nutr. 19, 165S–
75S (2000).
16.
Maintz, L. & Novak, N. Histamine and histamine intolerance. Am. J. Clin. Nutr. 85, 1185–
96 (2007).
17.
Ortolani, C. & Pastorello, E. A. Food allergies and food intolerances. Best Pract. Res. Clin.
Gastroenterol. 20, 467–83 (2006).
18.
Dørum, S., Qiao, S., Sollid, L. M. & Fleckenstein, B. A Quantitative Analysis of
Transglutaminase 2-Mediated Deamidation of Gluten Peptides: Implications for the T-cell
Response in Celiac Disease. J. Proteome Res. 8, 1748–55 (2009).
19.
Fleckenstein, B. et al. Molecular characterization of covalent complexes between tissue
transglutaminase and gliadin peptides. J. Biol. Chem. 279, 17607–16 (2004).
20.
Tack, G. J., Verbeek, W. H. M., Schreurs, M. W. J. & Mulder, C. J. J. The spectrum of
celiac disease: epidemiology, clinical aspects and treatment. Nat. Rev. Gastroenterol. Hepatol.
7, 204–13 (2010).
21.
Koning, F., Schuppan, D., Cerf-Bensussan, N. & Sollid, L. M. Pathomechanisms in celiac
disease. Best Pract. Res. Clin. Gastroenterol. 19, 373–87 (2005).
22.
Elli, L. et al. Diagnosis of gluten related disorders: Celiac disease, wheat allergy and nonceliac gluten sensitivity. World J. Gastroenterol. 21, 7110–19 (2015).
23.
Czaja-Bulsa, G. Non coeliac gluten sensitivity – A new disease with gluten intolerance.
Clin. Nutr. 34, 189–94 (2015).
24.
Kabbani, T. A. et al. Celiac disease or non-celiac gluten sensitivity? An approach to clinical
differential diagnosis. Am. J. Gastroenterol. 109, 741–6 (2014).
25.
Petersen, J. et al. Determinants of Gliadin-Specific T Cell Selection in Celiac Disease. J.
Immunol. 194, 6112–22 (2015).
26.
Petersen, J. et al. T-cell receptor recognition of HLA-DQ2-gliadin complexes associated
with celiac disease. Nat. Struct. Mol. Biol. 21, 480–8 (2014).
27.
Lionetti, E., Gatti, S., Pulvirenti, A. & Catassi, C. Celiac disease from a global perspective.
Best Pract. Res. Clin. Gastroenterol. 29, 365–79 (2015).
110
Chapter 7: References
28.
Norris, J. M. et al. Risk of celiac disease autoimmunity and timing of gluten introduction in
the diet of infants at increased risk of disease. J. Am. Med. Assoc. 293, 2343–51 (2005).
29.
Mearin, M. L. The prevention of coeliac disease. Best Pract. Res. Clin. Gastroenterol. 29, 493–
501 (2015).
30.
Vriezinga, S. L. et al. Randomized Feeding Intervention in Infants at High Risk for Celiac
Disease. N. Engl. J. Med. 371, 1304–15 (2014).
31.
Lionetti, E. et al. Introduction of Gluten, HLA Status, and the Risk of Celiac Disease in
Children. N. Engl. J. Med. 371, 1295–303 (2014).
32.
Størdal, K., White, R. A. & Eggesbø, M. Early feeding and risk of celiac disease in a
prospective birth cohort. Pediatrics 132, 1202–9 (2013).
33.
Greenhill, C. Coeliac disease: Does age at introduction to gluten affect risk of coeliac
disease? Nat. Rev. Gastroenterol. Hepatol. (2014). doi:10.1038/nrgastro.2014.176
34.
Pinier, M., Fuhrmann, G., Verdu, E. F., Verdu, E. & Leroux, J.-C. Prevention measures
and exploratory pharmacological treatments of celiac disease. Am. J. Gastroenterol. 105,
2551–61 (2010).
35.
Lee, A. & Newman, J. M. Celiac diet: Its impact on quality of life. J. Am. Diet. Assoc. 103,
1533–5 (2003).
36.
Chang, M. S. et al. Double-Blind Randomized Controlled Trial of Rifaximin for Persistent
Symptoms in Patients with Celiac Disease. Dig. Dis. Sci. 56, 2939–46 (2011).
37.
Evans, K. E., Leeds, J. S., Morley, S. & Sanders, D. S. Pancreatic Insufficiency in Adult
Celiac Disease: Do Patients Require Long-Term Enzyme Supplementation? Dig. Dis. Sci.
55, 2999–3004 (2010).
38.
Kaukinen, K., Lindfors, K. & Mäki, M. Advances in the treatment of coeliac disease: an
immunopathogenic perspective. Nat. Rev. Gastroenterol. Hepatol. 11, 36–44 (2014).
39.
Plugis, N. M. & Khosla, C. Therapeutic approaches for celiac disease. Best Pract. Res. Clin.
Gastroenterol. 29, 503–21 (2015).
40.
Lammers, K. M. et al. Gliadin Induces an Increase in Intestinal Permeability and Zonulin
Release by Binding to the Chemokine Receptor CXCR3. Gastroenterology 135, 194–204
(2008).
111
Chapter 7: References
41.
Drago, S. et al. Gliadin, zonulin and gut permeability: Effects on celiac and non-celiac
intestinal mucosa and intestinal cell lines. Scand. J. Gastroenterol. 41, 408–19 (2006).
42.
Clemente, M. G. et al. Early effects of gliadin on enterocyte intracellular signalling involved
in intestinal barrier function. Gut 52, 218–23 (2003).
43.
Paterson, B. M., Lammers, K. M., Arrieta, M. C., Fasano, A. & Meddings, J. B. The safety,
tolerance, pharmacokinetic and pharmacodynamic effects of single doses of AT-1001 in
coeliac disease subjects: a proof of concept study. Aliment. Pharmacol. Ther. 26, 757–66
(2007).
44.
Leffler, D. A. et al. A Randomized, Double-Blind Study of Larazotide Acetate to Prevent
the Activation of Celiac Disease During Gluten Challenge. Am. J. Gastroenterol. 107, 1554–
62 (2012).
45.
Kelly, C. P. et al. Larazotide acetate in patients with coeliac disease undergoing a gluten
challenge: A randomised placebo-controlled study. Aliment. Pharmacol. Ther. 37, 252–62
(2013).
46.
Leffler, D. A. et al. Larazotide acetate for persistent symptoms of celiac disease despite a
gluten-free diet: a randomized controlled trial. Gastroenterology 148, 1311–9 (2015).
47.
Matysiak-Budnik, T. et al. Secretory IgA mediates retrotranscytosis of intact gliadin
peptides via the transferrin receptor in celiac disease. J. Exp. Med. 205, 143–54 (2008).
48.
Rauhavirta, T. et al. Epithelial transport and deamidation of gliadin peptides: a role for
coeliac disease patient immunoglobulin A. Clin. Exp. Immunol. 164, 127–36 (2011).
49.
Maiuri, L. et al. Association between innate response to gliadin and activation of
pathogenic T cells in coeliac disease. Lancet 362, 30–7 (2003).
50.
Yokoyama, S. et al. Antibody-mediated blockade of IL-15 reverses the autoimmune
intestinal damage in transgenic mice that overexpress IL-15 in enterocytes. Proc. Natl.
Acad. Sci. 106, 15849–54 (2009).
51.
Sarra, M. et al. IL-15 positively regulates IL-21 production in celiac disease mucosa. Mucosal
Immunol. 6, 244–55 (2013).
52.
Xia, J. et al. Cyclic and dimeric gluten peptide analogues inhibiting DQ2-mediated antigen
presentation in celiac disease. Bioorganic Med. Chem. 15, 6565–73 (2007).
53.
Kapoerchan, V. V. et al. Design of azidoproline containing gluten peptides to suppress
CD4+ T-cell responses associated with Celiac disease. Bioorganic Med. Chem. 16, 2053–62
112
Chapter 7: References
(2008).
54.
Salvati, V. M. et al. Recombinant human interleukin 10 suppresses gliadin dependent T cell
activation in ex vivo cultured coeliac intestinal mucosa. Gut 54, 46–53 (2005).
55.
Mulder, C. J., Wahab, P. J., Meijer, J. W. R. & Metselaar, E. A pilot study of recombinant
human interleukin-10 in adults with refractory coeliac disease. Eur. J. Gastroenterol. Hepatol.
13, 1183–8 (2001).
56.
Sollid, L. M. & Khosla, C. Novel Therapies for Coeliac Disease. J. Intern. Med. 296, 604–13
(2011).
57.
Edwards, J. C. W. et al. Efficacy of B-cell-targeted therapy with rituximab in patients with
rheumatoid arthritis. N. Engl. J. Med. 350, 2572–81 (2004).
58.
Hauser, S. L. et al. B-cell depletion with rituximab in relapsing-remitting multiple sclerosis.
N. Engl. J. Med. 358, 676–88 (2008).
59.
Sollid, L. M., Molberg, Ø., McAdam, S. & Lundin, K. E. A. Autoantibodies in coeliac
disease: tissue transglutaminase - guilt by association? Gut 41, 851–52 (1997).
60.
Walters, M. J. et al. Characterization of CCX282-B, an orally bioavailable antagonist of the
CCR9 chemokine receptor, for treatment of inflammatory bowel disease. J. Pharmacol. Exp.
Ther. 335, 61–9 (2010).
61.
Feagan, B. G. et al. Randomised clinical trial: vercirnon, an oral CCR9 antagonist, vs.
placebo as induction therapy in active Crohn’s disease. Aliment. Pharmacol. Ther. 42, 1170–
81 (2015).
62.
Keshav, S. et al. A Randomized Controlled Trial of the Efficacy and Safety of CCX282-B,
an Orally-Administered Blocker of Chemokine Receptor CCR9, for Patients with Crohn’s
Disease. PLoS One 8, 1–12 (2013).
63.
Daveson, A. J. et al. Effect of hookworm infection on wheat challenge in celiac disease - a
randomised double-blinded placebo controlled trial. PLoS One 6, e17366 (2011).
64.
Croese, J. et al. Experimental hookworm infection and gluten microchallenge promote
tolerance in celiac disease. J. Allergy Clin. Immunol. 135, 508–16 (2015).
65.
Brown, G. J. et al. A Phase I Study to Determine Safety, Tolerability and Bioactivity of
Nexvax2® in HLA DQ2+ Volunteers With Celiac Disease Following a Long-Term, Strict
Gluten-Free Diet. AGA Abstr. S–437–S–438 (2011). doi:10.1016/S0016-5085(11)61794-0
113
Chapter 7: References
66.
Pizzuti, D. et al. Lack of intestinal mucosal toxicity of Triticum monococcum in celiac
disease patients. Scand. J. Gastroenterol. 41, 1305–11 (2006).
67.
Zanini, B. et al. Search for atoxic cereals: a single blind, cross-over study on the safety of a
single dose of Triticum monococcum, in patients with celiac disease. BMC Gastroenterol. 13,
92 (2013).
68.
Zanini, B., Villanacci, V., De Leo, L. & Lanzini, A. Triticum monococcum in patients with
celiac disease: a phase II open study on safety of prolonged daily administration. Eur. J.
Nutr. 54, 1027–9 (2015).
69.
Pinier, M. et al. Polymeric binders suppress gliadin-induced toxicity in the intestinal
epithelium. Gastroenterology 136, 288–98 (2009).
70.
McCarville, J. L. et al. BL-7010 Demonstrates Specific Binding to Gliadin and Reduces
Gluten-Associated Pathology in a Chronic Mouse Model of Gliadin Sensitivity. PLoS One
9, e109972 (2014).
71.
Pinier, M. et al. The copolymer P(HEMA-co-SS) binds gluten and reduces immune
response in gluten-sensitized mice and human tissues. Gastroenterology 142, 316–25 (2012).
72.
Gujral, N., Suh, J. W. & Sunwoo, H. H. Effect of anti-gliadin IgY antibody on epithelial
intestinal integrity and inflammatory response induced by gliadin. BMC Immunol. 16, 41
(2015).
73.
Verdu, E. F., Galipeau, H. J. & Jabri, B. Novel players in coeliac disease pathogenesis: role
of the gut microbiota. Nat. Rev. Gastroenterol. Hepatol. 12, 497–506 (2015).
74.
Cenit, M. C., Olivares, M., Codoñer-Franch, P. & Sanz, Y. Intestinal Microbiota and
Celiac Disease: Cause, Consequence or Co-Evolution? Nutrients 7, 6900–23 (2015).
75.
Laparra, J. M. & Sanz, Y. Bifidobacteria inhibit the inflammatory response induced by
gliadins in intestinal epithelial cells via modifications of toxic peptide generation during
digestion. J. Cell. Biochem. 109, 801–7 (2010).
76.
Lindfors, K. et al. Live probiotic Bifidobacterium lactis bacteria inhibit the toxic effects
induced by wheat gliadin in epithelial cell culture. Clin. Exp. Immunol. 152, 552–8 (2008).
77.
Smecuol, E. et al. Exploratory, randomized, double-blind, placebo-controlled study on the
effects of Bifidobacterium infantis natren life start strain super strain in active celiac
disease. J. Clin. Gastroenterol. 47, 139–47 (2013).
78.
Stepniak, D. et al. Highly efficient gluten degradation with a newly identified prolyl
114
Chapter 7: References
endoprotease: implications for celiac disease. Am. J. Physiol. Gastrointest. Liver Physiol. 291,
G621–9 (2006).
79.
Gass, J., Ehren, J., Strohmeier, G., Isaacs, I. & Khosla, C. Fermentation, purification,
formulation, and pharmacological evaluation of a prolyl endopeptidase from Myxococcus
xanthus: implications for Celiac Sprue therapy. Biotechnol. Bioeng. 92, 674–84 (2005).
80.
Gass, J. & Khosla, C. Prolyl endopeptidases. Cell. Mol. life Sci. 64, 345–55 (2007).
81.
Shan, L., Marti, T., Sollid, L. M., Gray, G. M. & Khosla, C. Comparative biochemical
analysis of three bacterial prolyl endopeptidases: implications for coeliac sprue. Biochem. J.
383, 311–8 (2004).
82.
Shan, L., Mathews, I. I. & Khosla, C. Structural and mechanistic analysis of two prolyl
endopeptidases: role of interdomain dynamics in catalysis and specificity. Proc. Natl. Acad.
Sci. 102, 3599–604 (2005).
83.
Bethune, M. T., Strop, P., Tang, Y., Sollid, L. M. & Khosla, C. Heterologous Expression,
Purification, Refolding, and Structural-Functional Characterization of EP-B2, a SelfActivating Barley Cysteine Endoprotease. Chem. Biol. 13, 637–47 (2006).
84.
Gass, J., Vora, H., Bethune, M. T., Gray, G. M. & Khosla, C. Effect of barley
endoprotease EP-B2 on gluten digestion in the intact rat. J. Pharmacol. Exp. Ther. 318,
1178–86 (2006).
85.
Tye-Din, J. A. et al. The effects of ALV003 pre-digestion of gluten on immune response
and symptoms in celiac disease in vivo. Clin. Immunol. 134, 289–95 (2010).
86.
Siegel, M. et al. Safety, Tolerability, and Activity of ALV003: Results from Two Phase 1
Single, Escalating-Dose Clinical Trials. Dig. Dis. Sci. 57, 440–50 (2012).
87.
Lähdeaho, M.-L. et al. Glutenase ALV003 attenuates gluten-induced mucosal injury in
patients with celiac disease. Gastroenterology 146, 1649–58 (2014).
88.
Ehren, J. et al. A Food-Grade Enzyme Preparation with Modest Gluten Detoxification
Properties. PLoS One 4, e6313 (2009).
89.
Mitea, C. et al. Efficient degradation of gluten by a prolyl endoprotease in a gastrointestinal
model: implications for coeliac disease. Gut 57, 25–32 (2008).
90.
Salden, B. N. et al. Randomised clinical study: Aspergillus niger -derived enzyme digests
gluten in the stomach of healthy volunteers. Aliment. Pharmacol. Ther. 42, 273–85 (2015).
115
Chapter 7: References
91.
Wolf, C. et al. Engineering of Kuma030: A Gliadin Peptidase That Rapidly Degrades
Immunogenic Gliadin Peptides in Gastric Conditions. J. Am. Chem. Soc. 137, 13106–13
(2015).
92.
Gordon, S. R. et al. Computational Design of an α-Gliadin Peptidase. J. Am. Chem. Soc.
134, 20513–20 (2012).
93.
Fuhrmann, G. et al. Sustained gastrointestinal activity of dendronized polymer-enzyme
conjugates. Nat. Chem. 5, 582–9 (2013).
94.
Choonara, B. F. et al. A review of advanced oral drug delivery technologies facilitating the
protection and absorption of protein and peptide molecules. Biotechnol. Adv. 32, 1269–82
(2014).
95.
Park, K., Kwon, I. C. & Park, K. Oral protein delivery: Current status and future prospect.
React. Funct. Polym. 71, 280–7 (2011).
96.
Deacon, M. P. et al. Atomic force microscopy of gastric mucin and chitosan mucoadhesive
systems. Biochem. J. 348, 557–63 (2000).
97.
Lehr, C.-M., Bouwstra, J. A., Schacht, E. H. & Junginger, H. E. In vitro evaluation of
mucoadhesive properties of chitosan and some other natural polymers. Int. J. Pharm. 78,
43–8 (1992).
98.
Lv, H., Zhang, S., Wang, B., Cui, S. & Yan, J. Toxicity of cationic lipids and cationic
polymers in gene delivery. J. Control. Release 114, 100–9 (2006).
99.
Kircheis, R. et al. Polycation-Based DNA Complexes for Tumor-Targeted Gene Delivery
in vivo. J. Gene Med. 1, 111–20 (1999).
100.
Xu, J., Strandman, S., Zhu, J. X. X., Barralet, J. & Cerruti, M. Genipin-crosslinked
catechol-chitosan mucoadhesive hydrogels for buccal drug delivery. Biomaterials 37, 395–
404 (2014).
101.
Luo, Y. & Wang, Q. Recent development of chitosan-based polyelectrolyte complexes
with natural polysaccharides for drug delivery. Int. J. Biol. Macromol. 64, 353–67 (2014).
102.
Khunawattanakul, W., Puttipipatkhachorn, S., Rades, T. & Pongjanyakul, T. Novel
chitosan-magnesium aluminum silicate nanocomposite film coatings for modified-release
tablets. Int. J. Pharm. 407, 132–41 (2011).
103.
Sogias, I. A., Williams, A. C. & Khutoryanskiy, V. V. Chitosan-based mucoadhesive tablets
for oral delivery of ibuprofen. Int. J. Pharm. 436, 602–10 (2012).
116
Chapter 7: References
104.
Imperiale, J. C., Nejamkin, P., Del Sole, M. J., E Lanusse, C. & Sosnik, A. Novel protease
inhibitor-loaded Nanoparticle-in-Microparticle Delivery System leads to a dramatic
improvement of the oral pharmacokinetics in dogs. Biomaterials 37C, 383–94 (2014).
105.
Song, L., Zhi, Z. & Pickup, J. C. Nanolayer encapsulation of insulin- chitosan complexes
improves efficiency of oral insulin delivery. Int. J. Nanomedicine 9, 2127–36 (2014).
106.
Rinaudo, M. Chitin and chitosan: Properties and applications. Prog. Polym. Sci. 31, 603–32
(2006).
107.
Varum, K. M., Ottoy, M. H. & Smidsrod, O. Water-solubility of partially N-acetylated
chitosans as a function of pH: effect of chemical composition and depolymerisation.
Carbohydr. Polym. 25, 65–70 (1994).
108.
Schipper, N. G. M., Vårum, K. M. & Artursson, P. Chitosans as Absorption Enhancers
for Poorly Absorbable Drugs. 1: Influence of Molecular Weight and Degree of Acetylation
on Drug Transport Across Human Intestinal Epithelial (Caco-2) Cells. Pharm. Res. 13,
1686–92 (1996).
109.
Kim, S.-K., Park, P.-J., Yang, H.-P. & Han, S.-S. Subacute Toxicity of Chitosan
Oligosaccharide in Sprague-Dawley Rats. Arzneimittelforschung 51, 769– 74 (2001).
110.
Chen, M.-C. et al. Recent advances in chitosan-based nanoparticles for oral delivery of
macromolecules. Adv. Drug Deliv. Rev. 65, 865–79 (2013).
111.
Thanou, M., Verhoef, J. C. & Junginger, H. E. Oral drug absorption enhancement by
chitosan and its derivatives. Adv. Drug Deliv. Rev. 52, 117–26 (2001).
112.
Artursson, P., Lindmark, T., Davis, S. S. & Illum, L. Effect of Chitosan on the
Permeability of Monolayers of Intestinal Epithelial Cells (Caco-2). Pharm. Res. 11, 1358–61
(1994).
113.
Kotzé, A. F. et al. Comparison of the effect of different chitosan salts and N-trimethyl
chitosan chloride on the permeability of intestinal epithelial cells (Caco-2). J. Control. Release
51, 35–46 (1998).
114.
Schipper, N. G. M. et al. Chitosans as absorption enhancers of poorly absorbable drugs. 3:
Influence of mucus on absorption enhancement. Eur. J. Pharm. Sci. 8, 335–43 (1999).
115.
Schipper, N. G. M. et al. Chitosans as Absorption Enhancers for Poorly Absorbable Drugs
2: Mechanism of Absorption Enhancement. Pharm. Res. 14, 923–9 (1997).
117
Chapter 7: References
116.
Dodane, V., Khan, A. M. & Merwin, J. R. Effect of chitosan on epithelial permeability and
structure. Int. J. Pharm. 182, 21–32 (1999).
117.
Kotzé, A. F. et al. Chitosans for enhanced delivery of therapeutic peptides across intestinal
epithelia: in vitro evaluation in Caco-2 cell monolayers. Int. J. Pharm. 159, 243–53 (1997).
118.
Lueβen, H. L. et al. Mucoadhesive polymers in peroral peptide drug delivery. I. Influence
of mucoadhesive excipients on the proteolytic activity of intestinal enzymes. Eur. J. Pharm.
Sci. 4, 117–28 (1996).
119.
Lueβen, H. L. et al. Mucoadhesive polymers in peroral peptide drug delivery. IV.
Polycarbophil and chitosan are potent enhancers of peptide transport across intestinal
mucosae in vitro. J. Control. Release 45, 15–23 (1997).
120.
Kotzé, A. F. et al. N-Trimethyl Chitosan Chloride as a Potential Absorption Enhancer
Across Mucosal Surfaces: In Vitro Evaluation in Intestinal Epithelial Cells (Caco-2).
Pharm. Res. 14, 1197–202 (1997).
121.
Biswas, S., Chattopadhyay, M., Sen, K. K., Saha, M. K. & Maji, H. S. Structure-Toxicity
Relationship of Chemically Modified Chitosan as an Oral Protein Drug Delivery Carrier. J.
Pharm. Sci. Pharmacol. 1, 131–40 (2014).
122.
Thanou, M. M. et al. Effect of degree of quaternization of N-trimethyl chitosan chloride
for enhanced transport of hydrophilic compounds across intestinal Caco-2 cell
monolayers. J. Control. Release 64, 15–25 (2000).
123.
Thanou, M., Florea, B. I., Langemeÿer, M. W., Verhoef, J. C. & Junginger, H. E. Ntrimethylated chitosan chloride (TMC) improves the intestinal permeation of the peptide
drug buserelin in vitro (Caco-2 cells) and in vivo (rats). Pharm. Res. 17, 27–31 (2000).
124.
Tomalia, D. A. et al. A New Class of Polymers: Starburst-Dendritic Macromolecules.
Polym. J. 17, 117–32 (1984).
125.
Tomalia, D. A., Naylor, A. M. & Goddard, W. A. Starburst Dendrimers: Molecular-Level
Control of Size, Shape, Surface Chemistry, Topology, and Flexibility from Atoms to
Macroscopic Matter. Angew. Chemie 29, 138–75 (1990).
126.
Wiwattanapatapee, R., Carreno-Gomez, B., Malik, N. & Duncan, R. PAMAM dendrimers
as a potential oral drug delivery system: uptake by everted rat intestinal sacs in-vitro. J.
Pharmacol. Pharmacother. 50, 99 (1998).
127.
Kitchens, K. M., Foraker, A. B., Kolhatkar, R. B., Swaan, P. W. & Ghandehari, H.
Endocytosis and Interaction of Poly (Amidoamine) Dendrimers with Caco-2 Cells. Pharm.
Res. 24, 2138–45 (2007).
118
Chapter 7: References
128.
Greish, K. et al. Size and surface charge significantly influence the toxicity of silica and
dendritic nanoparticles. Nanotoxicology 6, 713–23 (2012).
129.
Thiagarajan, G., Greish, K. & Ghandehari, H. Charge affects the oral toxicity of
poly(amidoamine) dendrimers. Eur. J. Pharm. Biopharm. 84, 330–4 (2013).
130.
El-Sayed, M., Rhodes, C. A., Ginski, M. & Ghandehari, H. Transport mechanism(s) of
poly (amidoamine) dendrimers across Caco-2 cell monolayers. Int. J. Pharm. 265, 151–7
(2003).
131.
Kitchens, K. M., Kolhatkar, R. B., Swaan, P. W., Eddington, N. D. & Ghandehari, H.
Transport of Poly(Amidoamine) Dendrimers across Caco-2 Cell Monolayers: Influence of
Size, Charge and Fluorescent Labeling. Pharm. Res. 23, 2818–26 (2006).
132.
Jevprasesphant, R., Penny, J., Attwood, D., McKeown, N. B. & D`Emanuele, A.
Engineering of Dendrimer Surfaces to Enhance Transepithelial Transport and Reduce
Cytotoxicity. Pharm. Res. 20, 1543–50 (2003).
133.
Lin, Y. et al. Polyamidoamine dendrimers as novel potential absorption enhancers for
improving the small intestinal absorption of poorly absorbable drugs in rats. J. Control.
Release 149, 21–8 (2011).
134.
Shukla, S. C., Singh, A., Pandey, A. K. & Mishra, A. Review on production and medical
applications of ɛ -polylysine. Biochem. Eng. J. 65, 70–81 (2012).
135.
Ekrami, H. M. & Shen, W.-C. Carbamylation Decreases the Cytotoxicity but not the
Drug-Carrier Properties of Polylysines. J. Drug Target. 2, 469–75 (1995).
136.
Ranaldi, G., Marigliano, I., Vespignani, I., Perozzi, G. & Sambuy, Y. The effect of chitosan
and other polycations on tight junction permeability in the human intestinal Caco-2 cell
line. J. Nutr. Biochem. 13, 157–67 (2002).
137.
McEwan, G. T. A., Jepson, M. A., Hirst, B. H. & Simmons, N. L. Polycation-induced
enhancement of epithelial paracellular permeability is independent of tight junctional
characteristics. Biochim. Biophys. Acta 1148, 51–60 (1993).
138.
Peterson, M. W. & Gruenhaupt, D. Protamine increases the permeability of cultured
epithelial monolayers. J. Appl. Physiol. 68, 220–7 (1990).
139.
Kim, G. M., Bae, Y. H. & Jo, W. H. pH-induced micelle formation of poly(histidine-cophenylalanine)-block-poly(ethylene glycol) in aqueous media. Macromol. Biosci. 5, 1118–24
(2005).
119
Chapter 7: References
140.
Köping-Höggård, M. et al. Chitosan as a nonviral gene delivery system. Structure-property
relationships and characteristics compared with polyethylenimine in vitro and after lung
administration in vivo. Gene Ther. 8, 1108–21 (2001).
141.
Nielsen, E. J. B. et al. In vivo proof of concept of oral insulin delivery based on a coadministration strategy with the cell-penetrating peptide penetratin. J. Control. Release 189,
19–24 (2014).
142.
Ma, Z., Lim, T. M. & Lim, L.-Y. Pharmacological activity of peroral chitosan-insulin
nanoparticles in diabetic rats. Int. J. Pharm. 293, 271–80 (2005).
143.
Pan, Y. et al. Bioadhesive polysaccharide in protein delivery system: chitosan nanoparticles
improve the intestinal absorption of insulin in vivo. Int. J. Pharm. 249, 139–47 (2002).
144.
Soliman, G. M., Zhang, Y. L., Merle, G., Cerruti, M. & Barralet, J. Hydrocaffeic acidchitosan nanoparticles with enhanced stability, mucoadhesion and permeation properties.
Eur. J. Pharm. Biopharm. 88, 1026–37 (2014).
145.
Vijayanathan, V., Thomas, T. & Thomas, T. J. DNA Nanoparticles and Development of
DNA Delivery Vehicles for Gene Therapy. Biochemistry 41, 14085–94 (2002).
146.
Albrecht, K. & Bernkop-Schnürch, A. Thiomers: forms, functions and applications to
nanomedicine. Nanomedicine 2, 41–50 (2007).
147.
van der Lubben, I. M., Verhoef, J. C., van Aelst, A. C., Borchard, G. & Junginger, H. E.
Chitosan microparticles for oral vaccination: preparation, characterization and preliminary
in vivo uptake studies in murine Peyer’s patches. Biomaterials 22, 687–94 (2001).
148.
Koppolu, B. P. et al. Controlling chitosan-based encapsulation for protein and vaccine
delivery. Biomaterials 35, 4382–9 (2014).
149.
Clark, M. A., Jepson, M. A. & Hirst, B. H. Exploiting M cells for drug and vaccine
delivery. Adv. Drug Deliv. Rev. 50, 81–106 (2001).
150.
Maclaughlin, F. C. et al. Chitosan and depolymerized chitosan oligomers as condensing
carriers for in vivo plasmid delivery. J. Control. Release 56, 259–72 (1998).
151.
Yin, L. et al. Drug permeability and mucoadhesion properties of thiolated trimethyl
chitosan nanoparticles in oral insulin delivery. Biomaterials 30, 5691–700 (2009).
152.
Makhlof, A., Tozuka, Y. & Takeuchi, H. Design and evaluation of novel pH-sensitive
120
Chapter 7: References
chitosan nanoparticles for oral insulin delivery. Eur. J. Pharm. Sci. 42, 445–51 (2011).
153.
Sonia, T. A., Rekha, M. R. & Sharma, C. P. Bioadhesive Hydrophobic Chitosan
Microparticles for Oral Delivery of Insulin: In Vitro Characterization and In Vivo Uptake
Studies. J. Appl. Polym. Sci. 119, 2902–10 (2010).
154.
Bonengel, S. & Bernkop-Schnürch, A. Thiomers - From bench to market. J. Control. Release
195, 120–9 (2014).
155.
Bernkop-Schnürch, A., Hornof, M. & Guggi, D. Thiolated chitosans. Eur. J. Pharm.
Biopharm. 57, 9–17 (2004).
156.
Föger, F., Schmitz, T. & Bernkop-Schnürch, A. In vivo evaluation of an oral delivery
system for P-gp substrates based on thiolated chitosan. Biomaterials 27, 4250–5 (2006).
157.
Perera, G., Zipser, M., Bonengel, S., Salvenmoser, W. & Bernkop-Schnürch, A.
Development of phosphorylated nanoparticles as zeta potential inverting systems. Eur. J.
Pharm. Biopharm. 97, 250–6 (2015).
158.
Neutra, M. R., Mantis, N. J. & Kraehenbuhl, J.-P. Collaboration of epithelial cells with
organized mucosal lymphoid tissues. Nat. Immunol. 2, 1004–9 (2001).
159.
Eldridge, J. H. et al. Controlled Vaccine Release In The Gut-Associated Lymphoid Tissues.
I. Orally Administered Biodegradable Microspheres Target The Peyer’s Patches. J. Control.
Release 11, 205–14 (1990).
160.
van der Lubben, I. M., Konings, F. A. J., Borchard, G., Verhoef, J. C. & Junginger, H. E.
In Vivo Uptake of Chitosan Microparticles by Murine Peyer’s Patches: Visualization
Studies using Confocal Laser Scanning Microscopy and Immunohistochemistry. J. Drug
Target. 9, 39–47 (2001).
161.
Roy, K., Mao, H.-Q., Huang, S.-K. & Leong, K. W. Oral gene delivery with chitosanDNA nanoparticles generates immunologic protection in a murine model of peanut
allergy. Nat. Med. 5, 387–91 (1999).
162.
Agnihotri, S. A., Mallikarjuna, N. N. & Aminabhavi, T. M. Recent advances on chitosanbased micro- and nanoparticles in drug delivery. J. Control. Release 100, 5–28 (2004).
163.
Chen, J. et al. Transfection of mEpo gene to intestinal epithelium in vivo mediated by oral
delivery of chitosan-DNA nanoparticles. World J. Gastroenterol. 10, 112–6 (2004).
164.
Bohl, B. D., Salvetti, A., Moullier, P. & Heard, J. M. Control of Erythropoietin Delivery by
Doxycycline in Mice After Intramuscular Injection of Adeno-Associated Vector. Blood 92,
121
Chapter 7: References
1512–7 (2015).
165.
Bowman, K., Sarkar, R., Raut, S. & Leong, K. W. Gene transfer to hemophilia A mice via
oral delivery of FVIII-chitosan nanoparticles. J. Control. Release 132, 252–9 (2008).
166.
Chew, J. L., Wolfowicz, C. B., Mao, H.-Q., Leong, K. W. & Chua, K. Y. Chitosan
nanoparticles containing plasmid DNA encoding house dust mite allergen, Der p 1 for
oral vaccination in mice. Vaccine 21, 2720–9 (2003).
167.
Bivas-Benita, M. et al. Generation of Toxoplasma gondii GRA1 protein and DNA vaccine
loaded chitosan particles: preparation, characterization, and preliminary in vivo studies. Int.
J. Pharm. 266, 17–27 (2003).
168.
Negash, T., Liman, M. & Rautenschlein, S. Mucosal application of cationic poly(D,Llactide-co-glycolide) microparticles as carriers of DNA vaccine and adjuvants to protect
chickens against infectious bursal disease. Vaccine 31, 3656–62 (2013).
169.
Danhier, F. et al. PLGA-based nanoparticles: an overview of biomedical applications. J.
Control. Release 161, 505–22 (2012).
170.
Boussif, O. et al. A versatile vector for gene and oligonucleotide transfer into cells in
culture and in vivo: Polyethylenimine. Proc. Natl. Acad. Sci. 92, 7297–301 (1995).
171.
Forbes, D. C. & Peppas, N. A. Oral delivery of small RNA and DNA. J. Control. Release
162, 438–45 (2012).
172.
Scholz, C. & Wagner, E. Therapeutic plasmid DNA versus siRNA delivery: common and
different tasks for synthetic carriers. J. Control. Release 161, 554–65 (2012).
173.
He, C., Yin, L., Tang, C. & Yin, C. Trimethyl chitosan-cysteine nanoparticles for systemic
delivery of TNF-α siRNA via oral and intraperitoneal routes. Pharm. Res. 30, 2596–606
(2013).
174.
Laroui, H. et al. Functional TNFα gene silencing mediated by polyethyleneimine/TNFα
siRNA nanocomplexes in inflamed colon. Biomaterials 32, 1218–28 (2011).
175.
Laroui, H. et al. Targeting intestinal inflammation with CD98 siRNA/PEI-loaded
nanoparticles. Mol. Ther. 22, 69–80 (2014).
176.
Landry, F. B., Bazile, D. V., Spenlehauer, G., Veillard, M. & Kreuter, J. Degradation of
poly (D,L-lactic acid) nanoparticles coated with albumin in model digestive fluids (USP
XXII). Biomaterials 17, 715–23 (1996).
122
Chapter 7: References
177.
Yan, Y., Dalmasso, G., Sitaraman, S. & Merlin, D. Characterization of the human
intestinal CD98 promoter and its regulation by interferon-gamma. Am. J. Physiol.
Gastrointest. Liver Physiol. 292, G535–45 (2007).
178.
Han, L., Tang, C. & Yin, C. Oral delivery of shRNA and siRNA via multifunctional
polymeric nanoparticles for synergistic cancer therapy. Biomaterials 35, 4589–600 (2014).
179.
Duncan, R. Polymer conjugates as anticancer nanomedicines. Nat. Rev. Cancer 6, 688–701
(2006).
180.
Harde, H., Agrawal, A. K. & Jain, S. Tetanus toxoids loaded glucomannosylated chitosan
based nanohoming vaccine adjuvant with improved oral stability and immunostimulatory
response. Pharm. Res. 32, 122–34 (2015).
181.
van der Lubben, I. M. et al. Chitosan microparticles for mucosal vaccination against
diphtheria: oral and nasal efficacy studies in mice. Vaccine 21, 1400–8 (2003).
182.
Mukhopadhyay, P., Sarkar, K., Bhattacharya, S., Mishra, R. & Kundu, P. P. Efficient oral
insulin delivery by dendronized chitosan: in vitro and in vivo studies. RSC Adv. 4, 43890–
902 (2014).
183.
Damgé, C., Michel, C., Aprahamian, M. & Couvreur, P. New Approach for Oral
Administration of Insulin With Polyalkylcyanoacrylate Nanocapsules as Drug Carrier.
Diabetes 37, 246–51 (1988).
184.
Heinemann, L. & Jacques, Y. Oral insulin and buccal insulin: a critical reappraisal. J.
Diabetes Sci. Technol. 3, 568–84 (2009).
185.
Jain, N. K. & Gupta, U. Application of dendrimer-drug complexation in the enhancement
of drug solubility and bioavailability. Expert Opin. Drug Metab. Toxicol. 4, 1035–52 (2008).
186.
Ke, W., Zhao, Y., Huang, R., Jiang, C. & Pei, Y. Enhanced Oral Bioavailability of
Doxorubicin in a Dendrimer Drug Delivery System. J. Pharm. Sci. 97, 2208–16 (2008).
187.
Na, M. et al. Dendrimers as potential drug carriers. Part II. Prolonged delivery of
ketoprofen by in vitro and in vivo studies. Eur. J. Med. Chem. 41, 670–4 (2006).
188.
Kolharkar, R. B., Swaan, P. & Ghandehari, H. Potential Oral Delivery of 7-Ethyl-10Hydroxy-Camptothecin (SN-38) using Poly (amidoamine) Dendrimers. Pharm. Res. 25,
1723–9 (2009).
189.
Prego, C. et al. Chitosan-PEG nanocapsules as new carriers for oral peptide delivery:
Effect of chitosan pegylation degree. J. Control. Release 111, 299–308 (2006).
123
Chapter 7: References
190.
Prego, C., García, M., Torres, D. & Alonso, M. J. Transmucosal macromolecular drug
delivery. J. Control. Release 101, 151–62 (2005).
191.
Ma, Q. et al. Oral Absorption Enhancement of Probucol by PEGylated G5 PAMAM
Dendrimer Modi fi ed Nanoliposomes. Mol. Pharm. 12, 665–74 (2015).
192.
Wiwattanapatapee, R., Lomlim, L. & Saramunee, K. Dendrimers conjugates for colonic
delivery of 5-aminosalicylic acid. J. Control. Release 88, 1–9 (2003).
193.
Najlah, M., Freeman, S., Attwood, D. & D’Emanuele, A. Synthesis, characterization and
stability of dendrimer prodrugs. Int. J. Pharm. 308, 175–82 (2006).
194.
Najlah, M., Freeman, S., Attwood, D. & D’Emanuele, A. In vitro evaluation of dendrimer
prodrugs for oral drug delivery. Int. J. Pharm. 336, 183–90 (2007).
195.
Najlah, M., Freeman, S., Attwood, D. & D’Emanuele, A. Synthesis and Assessment of
First-Generation Polyamidoamine Dendrimer Prodrugs to Enhance the Cellular
Permeability of P-gp Substrates. Bioconjug. Chem. 18, 937–46 (2007).
196.
D’Emanuele, A., Jevprasesphant, R., Penny, J. & Attwood, D. The use of a dendrimerpropranolol prodrug to bypass efflux transporters and enhance oral bioavailability. J.
Control. Release 95, 447–53 (2004).
197.
Jevprasesphant, R. et al. The influence of surface modification on the cytotoxicity of
PAMAM dendrimers. Int. J. Pharm. 252, 263–6 (2003).
198.
Svenson, S. Dendrimers as versatile platform in drug delivery applications. Eur. J. Pharm.
Biopharm. 71, 445–62 (2009).
199.
Lee, E. et al. Conjugated chitosan as a novel platform for oral delivery of paclitaxel. J. Med.
Chem. 51, 6442–9 (2008).
200.
Lee, E., Kim, H., Lee, I.-H. & Jon, S. In vivo antitumor effects of chitosan-conjugated
docetaxel after oral administration. J. Control. Release 140, 79–85 (2009).
201.
Ahn, S. et al. Oral delivery of an anti-diabetic peptide drug via conjugation and
complexation with low molecular weight chitosan. J. Control. Release 170, 226–32 (2013).
202.
Lee, E., Lee, J. & Jon, S. A novel approach to oral delivery of insulin by conjugating with
low molecular weight chitosan. Bioconjug. Chem. 21, 1720–3 (2010).
124
Chapter 7: References
203.
Fuhrmann, G. & Leroux, J.-C. Improving the stability and activity of oral therapeutic
enzymes-recent advances and perspectives. Pharm. Res. 31, 1099–105 (2014).
204.
Matoori, S., Fuhrmann, G. & Leroux, J.-C. Celiac disease: a challenging disease for
pharmaceutical scientists. Pharm. Res. 30, 619–26 (2013).
205.
Kourist, R., Brundiek, H. & Bornscheuer, U. T. Protein engineering and discovery of
lipases. Eur. J. Lipid Sci. Technol. 112, 64–74 (2010).
206.
Leisola, M. & Turunen, O. Protein engineering: opportunities and challenges. Appl.
Microbiol. Biotechnol. 75, 1225–32 (2007).
207.
Preiswerk, N. et al. Impact of scaffold rigidity on the design and evolution of an artificial
Diels-Alderase. Proc. Natl. Acad. Sci. 111, 8013–8 (2014).
208.
Jiang, L. et al. De novo computational design of retro-aldol enzymes. Science 319, 1387–91
(2008).
209.
Siegel, J. B. et al. Computational Design of an Enzyme Catalyst for a Stereoselective
Bimolecular Diels-Alder Reaction. Science. 329, 309–13 (2010).
210.
Rosendahl, M. S. et al. A long-acting, highly potent interferon alpha-2 conjugate created
using site-specific PEGylation. Bioconjug. Chem. 16, 200–7 (2005).
211.
Doherty, D. H. et al. Site-specific PEGylation of engineered cysteine analogues of
recombinant human granulocyte-macrophage colony-stimulating factor. Bioconjug. Chem.
16, 1291–8 (2005).
212.
Uchio, T., Baudyš, M., Liu, F., Song, S. C. & Kim, S. W. Site-specific insulin conjugates
with enhanced stability and extended action profile. Adv. Drug Deliv. Rev. 35, 289–306
(1999).
213.
de Graaf, A. J., Kooijman, M., Hennink, W. E. & Mastrobattista, E. Nonnatural amino
acids for site-specific protein conjugation. Bioconjug. Chem. 20, 1281–95 (2009).
214.
Zimmerman, E. S. et al. Production of site-specific antibody-drug conjugates using
optimized non-natural amino acids in a cell-free expression system. Bioconjug. Chem. 25,
351–61 (2014).
215.
Stevenazzi, A., Marchini, M., Sandrone, G., Vergani, B. & Lattanzio, M. Amino acidic
scaffolds bearing unnatural side chains: An old idea generates new and versatile tools for
the life sciences. Bioorg. Med. Chem. Lett. 24, 5349–56 (2014).
125
Chapter 7: References
216.
Fuhrmann, G. & Leroux, J.-C. In vivo fluorescence imaging of exogenous enzyme activity
in the gastrointestinal tract. Proc. Natl. Acad. Sci. 108, 9032–7 (2011).
217.
Ehren, J., Govindarajan, S., Morón, B., Minshull, J. & Khosla, C. Protein engineering of
improved prolyl endopeptidases for celiac sprue therapy. Protein Eng. Des. Sel. 21, 699–707
(2008).
218.
Shen, B.-Q. et al. Conjugation site modulates the in vivo stability and therapeutic activity of
antibody-drug conjugates. Nat. Biotechnol. 30, 184–9 (2012).
219.
Gandhi, M., Jangi, M. & Goode, B. L. Functional Surfaces on the Actin-binding Protein
Coronin Revealed by Systematic Mutagenesis. J. Biol. Chem. 285, 34899–908 (2010).
220.
Leader, B., Baca, Q. J. & Golan, D. E. Protein therapeutics: a summary and
pharmacological classification. Nat. Rev. Drug Discov. 7, 21–39 (2008).
221.
Kontermann, R. E. Strategies for extended serum half-life of protein therapeutics. Curr.
Opin. Biotechnol. 22, 868–76 (2011).
222.
Liu, M., Johansen, P., Zabel, F., Leroux, J.-C. & Gauthier, M. A. Semi-permeable coatings
fabricated from comb-polymers efficiently protect proteins in vivo. Nat. Commun. 5, 5526
(2014).
223.
Alconcel, S. N. S., Baas, A. S. & Maynard, H. D. FDA-approved poly(ethylene glycol)–
protein conjugate drugs. Polym. Chem. 2, 1442–8 (2011).
224.
Robic, S. Pegylated glutenase polypeptides. Pat. Appl. WO2007047303 A2 (2009).
225.
Gauthier, M. A. & Klok, H.-A. Polymer–protein conjugates: an enzymatic activity
perspective. Polym. Chem. 1, 1352–73 (2010).
226.
Schulz, J. D., Gauthier, M. A. & Leroux, J. Improving oral drug bioavailability with
polycations? Eur. J. Pharm. Biopharm. 97, 427–37 (2015).
227.
MacDonald, J. I., Munch, H. K., Moore, T. & Francis, M. B. One-step site-specific
modification of native proteins with 2-pyridinecarboxyaldehydes. Nat. Chem. Biol. 11, 326–
31 (2015).
228.
Shaunak, S. et al. Site-specific PEGylation of native disulfide bonds in therapeutic proteins.
Nat. Chem. Biol. 2, 312–3 (2006).
229.
Liu, M. et al. Molecular sieving on the surface of a protein provides protection without loss
126
Chapter 7: References
of activity. Adv. Funct. Mater. 23, 2007–15 (2013).
230.
Duncan, R. The dawning era of polymer therapeutics. Nat. Rev. Drug Discov. 2, 347–60
(2003).
231.
Ding, Z., Chen, G. & Hoffman, A. S. Unusual properties of thermally sensitive oligomer –
enzyme conjugates of poly(N-isopropylacrylamide)– trypsin. J. Biomed. Mater. Res. 39, 498–
505 (1998).
232.
Chiu, K. et al. Effects of polymer molecular weight on the size, activity, and stability of
PEG-functionalized trypsin. Biomacromolecules 11, 3688–92 (2010).
233.
Armstrong, J. K., Wenby, R. B., Meiselman, H. J. & Fisher, T. C. The hydrodynamic radii
of macromolecules and their effect on red blood cell aggregation. Biophys. J. 87, 4259–70
(2004).
234.
Yang, C., Lu, D. & Liu, Z. How PEGylation enhances the stability and potency of insulin:
A molecular dynamics simulation. Biochemistry 50, 2585–93 (2011).
235.
Arakawa, T. & Timasheff, S. N. Mechanism of poly(ethylene glycol) interaction with
proteins. Biochemistry 24, 6756–62 (1985).
236.
Hartmann, L. & Börner, H. G. Precision polymers: monodisperse, monomer-sequencedefined segments to target future demands of polymers in medicine. Adv. Mater. 21, 3425–
31 (2009).
237.
Ciolkowski, M., Rozanek, M., Bryszewska, M. & Klajnert, B. The influence of PAMAM
dendrimers surface groups on their interaction with porcine pepsin. Biochim. Biophys. Acta
1834, 1982–7 (2013).
238.
Sadekar, S. & Ghandehari, H. Transepithelial transport and toxicity of PAMAM
dendrimers: implications for oral drug delivery. Adv. Drug Deliv. Rev. 64, 571–88 (2012).
239.
Aziz, I., Hadjivassiliou, M. & Sanders, D. S. The spectrum of noncoeliac gluten sensitivity.
Nat. Rev. Gastroenterol. Hepatol. 12, 516–26 (2015).
240.
Levy, H. L. Phenylketonuria: old disease, new approach to treatment. Proc. Natl. Acad. Sci.
96, 1811–3 (1999).
241.
Longo, N. et al. Single-dose, subcutaneous recombinant phenylalanine ammonia lyase
conjugated with polyethylene glycol in adult patients with phenylketonuria: an open-label,
multicentre, phase 1 dose-escalation trial. Lancet 384, 37–44 (2014).
127
Chapter 7: References
242.
Rajender Reddy, K., Modi, M. W. & Pedder, S. Use of peginterferon alfa-2a (40 KD)
(Pegasys®) for the treatment of hepatitis C. Adv. Drug Deliv. Rev. 54, 571–86 (2002).
243.
Maleki, A. et al. Design, modeling, and expression of erythropoietin cysteine analogs in
Pichia pastoris: improvement of mean residence times and in vivo activities through
cysteine-specific PEGylation. Eur. J. Pharm. Biopharm. 80, 499–507 (2012).
244.
Pettit, D. K. et al. Structure-function studies of interleukin 15 using site-specific
mutagenesis, polyethylene glycol conjugation, and homology modeling. J. Biol. Chem. 272,
2312–8 (1997).
245.
Schulz, J. D. et al. Site-Specifi c Polymer Conjugation Stabilizes Therapeutic Enzymes in
the Gastrointestinal Tract. Adv. Mater. (2015). doi:10.1002/adma.201504797
246.
Menjoge, A. R., Kannan, R. M. & Tomalia, D. A. Dendrimer-based drug and imaging
conjugates: design considerations for nanomedical applications. Drug Discov. Today 15, 171–
85 (2010).
247.
Korneychuk, N., Meresse, B. & Cerf-Bensussan, N. Lessons from rodent models in celiac
disease. Mucosal Immunol. 8, 18–28 (2015).
248.
Suarez, F. L., Savaiano, D. A. & Levitt, M. D. Review article: the treatment of lactose
intolerance. Aliment. Pharmacol. Ther. 9, 589–97 (1995).
249.
Turner, K. M., Pasut, G., Veronese, F. M., Boyce, A. & Walsh, G. Stabilization of a
supplemental digestive enzyme by post-translational engineering using chemicallyactivated polyethylene glycol. Biotechnol. Lett. 33, 617–21 (2011).
250.
Aungst, B. J. Absorption Enhancers: Applications and Advances. AAPS J. 14, 10–8
(2012).
251.
Zijlstra, E., Heinemann, L. & Plum-Mörschel, L. Oral Insulin Reloaded: A Structured
Approach. J. Diabetes Sci. Technol. 8, 458–65 (2014).
252.
Haag, R. & Kratz, F. Polymer therapeutics: concepts and applications. Angew. Chemie 45,
1198–215 (2006).
128
List of Abbreviations
List of Abbreviations
3D
three-dimensional
AN
(PEP from) Aspergillus niger
CCR9
C-C chemokine receptor type 9
CD
cluster of differentiation
GFD
gluten-free diet
GI tract
gastro-intestinal tract
DA
degree of deacetylation
DQ
degree of quaternization
FVIII
factor VIII
G
generation
HLA
human leukocyte antigen
IFN
interferon
Ig
immunoglobin
IL
interleukin
MICA
MHC class I polypeptide-related sequence A
mPEG
methoxy poly(ethylene glycole)
M cells
microfold cells
MX
(PEP from) Myxococcus xanthus
NCGS
non-celiac gluten sensitivity
NK
natural killer cells
PAA
poly(acrylic acid)
PAMAM
poly(amido amine)
PEI
poly(ethylenimine)
PEP
prolyl endopeptidase
P-gp
P-glycoprotein
PG1
poly-(3,5-bis(3-aminopropoxy)benzyl)-methacrylate
131
List of Abbreviations
PLA
poly(D,L lactide)
SC
(PEP from) Sphingomonas capsulate
sIg
secretory immunoglobulin
siRNA
small interfering RNA
TEER
transepithelial electric resistance
TM
Triticum monococcum
TMC
N-trimethyl chitosan
TNF
tumor necrosis factors
tTG2
tissue transglutaminase 2
WT
wild-type
132
Scientific Contributions
Scientific Contributions
Publications
Schulz, J. D., Patt, M., Basler, S., Kries, H., Hilvert, D., Gauthier, M. A., Leroux, J.-C. Site-Specific
Polymer Conjugation Stabilizes Therapeutic Enzymes in the Gastro-Intestinal Tract. Adv. Mater.
2015; DOI: 10.1002/adma.201504797.
Schulz, J. D., Gauthier, M. A., Leroux, J.-C. Improving oral drug bioavailability with polycations?
Eur. J. Pharm. Biopharm. 2015; 97:427-37.
Preiswerk, N., Beck, T., Schulz, J. D., Milovník, P., Mayer, C., Siegel, J. B., Bakeb, D., Hilvert, D.
Impact of scaffold rigidity on the design and evolution of an artificial Diels-Alderase. Proc. Natl.
Acad. Sci. 2014; 111:8013-18.
Galipeau, H. J., Wiepjes, M., Motta, J.-P., Schulz, J. D. et al. Novel role of the serine protease
inhibitor elafin in gluten-related disorders. Am. J. Gastroenterol. 2014; 109:748-56.
Fuhrmann, K., Schulz, J. D., Gauthier, M. A., Leroux, J.-C. PEG Nanocages as Non-sheddable
Stabilizers for Drug Nanocrystals. ACS Nano 2012; 6:1667-76.
Matoori S., Fuhrmann G., Schulz J. D., Leroux J.-C. Gluten Binden und Spalten – neue adjuvante
Therapieansätze für die Zöliakie. Swiss Medical Forum 2012; 12:716-7.
135
Scientific Contributions
Oral Presentations
Schulz, J. D. 160th ETH day, Zurich, Switzerland. November 21, 2015.
Schulz, J. D., Patt, M., Kries, H., Hilvert, D., Leroux, J.-C. Improving Enzymatic Oral Therapies
via Polymer Conjugation. Zurich-Basel-Geneva Joint Seminar Series in Drug Formulation &
Delivery; University of Basel, Switzerland. January 22–23, 2015.
Schulz, J. D., Patt, M., Kries, H., Hilvert, D., Leroux, J.-C. Improving Enzymatic Oral Therapies
via Polymer Conjugation. Germany local chapter meeting CRS; Muttenz, Switzerland. January 12–
13, 2015.
Schulz, J. D., Patt, M., Kries, H., Hilvert, D., Leroux, J.-C. Celiac Disease: Improving Enzymatic
Oral Therapies via Polymer Conjugation. Doktorandentag (doctoral students’ day); Institute of
Pharmaceutical Sciences, ETH Zurich, Switzerland. September 10, 2014.
Poster Presentations
Schulz, J. D., Patt, M., Kries, H., Hilvert, D., Leroux, J.-C. Improvement of Prolylendopeptidase
Stability by Site-Specific Polymer Conjugation; 16th International Celiac Disease Symposium
(ICDS); Prague, Czech Republic. June 21–24, 2014.
Schulz, J. D., Patt, M., Kries, H., Hilvert, D., Leroux, J.-C. Improving Enzymatic Oral Therapies
via Polymer Conjugation; 15th ICDS; Chicago, USA. September 22–25, 2013.
136
Acknowledgments
Acknowledgments
The work in this doctoral thesis was supported by many individuals who contributed
differently to the success of this project and to whom I want to express my gratitude.
I am grateful to Prof. Jean-Christophe Leroux for giving me the chance to work in
his unique research group and on this interesting, challenging and interdisciplinary project.
I am thankful for his scientific input and availability for discussions.
The research project presented in this work was financially supported by the Swiss
National Science Foundation (310030_135732).
I also thank Katharina Holzinger and Dennis Mollenhauser, who helped performing
the in vivo experiments presented in Chapter 4 with great patience and supporting advice. I
am also grateful to Dr. Franziska Bootz and Dr. Sinem Karaman for their technical support
with the ex vivo studies. I thank Dr. Max Pillong, Dr. Steven Proulx, Dr. Giovanni Pellegrini
and Dr. Bernhard Pfeiffer for their technical support, Dr. Davide Brambilla for his support
in the graphical design of figures and Dr. Gregor Fuhrmann for helpful discussions.
I thank Prof. Marc Gauthier from the Institut National de la Recherche Scientifique
in Canada for his collaboration, availability and expert advice. Thanks go also to Prof.
Donald Hilvert and Dr. Hajo Kries from the Laboratory of Organic Chemistry at ETH
Zurich for their collaboration and helpful support of the work presented in Chapter 3.
I am grateful to my Master’s students Melanie Patt, Sophie Basler, Valerija
Kostadinova and Rydvikha Govender for their motivation and contribution to my research.
Thanks go also to Prof. Donald Hilvert and Prof. Michael Detmar and their respective
group members for the use of their instruments and to Monica Langfritz for IT support.
Many thanks go to Prof. Laura Nyström for agreeing to be my co-examiner and to
Dr. Jong Ah Kim, Dr. Giovanna Giacalone and Matthias Westphal for proofreading my
thesis.
I also thank all current and former members of PSA and WiNS for the fruitful
discussions, hand-in-hand work and friendships.
I want to express great gratitude to all members of the Drug & Deliver group. The
strong support for group members is extraordinary and served as an important motivation
139
Acknowledgments
throughout the years. I am grateful for the time inside and outside the lab, the unique
experiences that would not have been possible without the multi-cultural friendships that
have developed over time. I want to thank former group members who supported and
helped me integrate well at the start of this project, namely Dr. Arnaud Felber, Dr. Mattias
Ivarsson, Dr. Vincent Forster, Dr. Lorine Brülisauer, Dr. Kathrin Fuhrmann and Dr. Soo
Hyeon Lee. I especially thank Athanasia Dasargyri and Diana Andina for the everyday
support, discussions and the good atmosphere in our office. Special thanks go also to Anna
Polomska, Elena Moroz, Maurizio Roveri, Xiangang Huan, Peter Tiefenböck, Sandhya
Ananta, Prof. Paula Luciani, Prof. Bastien Castagner, Dr. Giovanna Giacalone, Dr. Ander
Estella, Dr. Jong Ah Kim, Anna Pratsinis, Virginie Rusca, Dr. Valentina Agostoni and
Dr. Davide Brambilla for the good time, help and friendships.
Finally, I want to thank my friends outside of the research group who have always
been an important and stable part in my life. I am grateful to my family, especially to my
parents and my brother who have supported me throughout my life. A big thank you goes
to my boyfriend, Stephan, for his motivation, positive spirit, and also for helpful scientific
discussions.
140

Untitled

YAKUGAKU ZASSHI

Cover Page The handle http://hdl.handle.net/26936 holds

詳細 - バイオベースマテリアル学専攻

Plan Vivo Terms of Reference for Project Validation

Details for your registration

Diana Mager/Abeer Alzaben, University of Alberta

Lorenzo Massimo Polgar, Netherlands

導電性高分子ハイブリッド アルミ電解コンデンサ 導電性高分子

FTIR Spectroscopy XRD, SEM EDX and AFM Studies On Natural