Review
Delivery
Smart polymeric carriers for
enhanced intracellular delivery of
therapeutic macromolecules
1. Introduction
Mohamed EH El-Sayed, Allan S Hoffman & Patrick S Stayton†
2. Polycarboxylate polymers
†University
and copolymers
3. Enhancement of non-viral
gene delivery using PPAA
4. Application of PPAA in
immunotoxin therapy
5. Development of pH-sensitive,
membrane-destabilising and
glutathione-reactive polymeric
drug carriers
6. ‘Encrypted’ polymeric
drug carriers
7. Expert opinion and conclusions
of Washington, Department of Bioengineering, Box 351721, Seattle, WA 98195, USA
Limited cytoplasmic delivery of enzyme-susceptible drugs remains a significant challenge facing the development of protein and nucleic acid therapies
that act in intracellular compartments. Researchers have examined several
approaches, including fusogenic proteins and protein transduction domains,
to enhance the intracellular delivery of the therapeutic cargo. This review
summarises efforts to develop ‘smart’ pH-sensitive and membrane-destabilising polymers that can shuttle therapeutic peptide, protein and nucleic acid
molecules past the endosomal membrane into the cytoplasm of targeted
cells. Several classes of ‘smart’ non-degradable polymeric carriers have been
developed that have proved effective both in vitro and in vivo in enhancing
the cytoplasmic delivery of a variety of therapeutic molecules.
P
r
o
r
f
o
Keywords: antisense delivery, encrypted polymers, gene delivery, intracellular delivery,
PDSA polymers, polymer therapeutics, poly(propylacrylic acid) polymers, smart polymers,
vaccine delivery
o
h
Expert Opin. Biol. Ther. (2005) 5(1):xxx–xxx
1. Introduction
A
Ashley Publications
www.ashley-pub.com
t
u
Recent advances in the biotechnology and pharmaceutical fields have led to the
development of several classes of therapeutic biomolecules including plasmid DNA
(pDNA), antisense oligodeoxynucleotides (ASODNs), silencing RNA, peptides,
proteins, and immunotoxins. The therapeutic activity of many of these macromolecules depends on their ability to reach intracellular targets commonly present in the
cytoplasm. However, cellular uptake of macromolecules usually occurs by passive or
receptor-mediated endocytosis, with subsequent accumulation in the endosomal–
lysosomal trafficking pathway where degradation ultimately occurs (Figure 1). The
limited cytoplasmic delivery of enzyme-susceptible drugs remains a major challenge
facing the development of effective intracellular therapies.
Several viruses and pathogenic organisms, such as the influenza virus and Corynebacterium diphtheriae, have developed sophisticated intracellular delivery systems that
include fusogenic proteins such as haemagglutinin and diphtheria toxin. These
fusogenic proteins display membrane-destabilising activity in response to endosomal
pH gradients and have been shown to enhance DNA and protein transport from the
endosomal compartment to the cytoplasm of targeted cells [1-5]. The pH of the endosomal compartment drops during maturation to values of ≤ 5.5 through the proton
pumping activity of membrane-bound ATP-dependent proton pumps. This drop in
pH triggers a conformational change in fusogenic proteins, leading to exposure of membrane-active domains that control membrane fusion, endosomal–lysosomal development and aid transport across the endosomal barrier [1-4]. The pH-sensing mechanism
of haemagglutinin and diphtheria toxin is connected to the protonation of key carboxylate residues followed by a shift in conformational equilibria towards the membraneactive state [1-4]. Synthetic peptides with similar membrane-destabilising mechanisms
10.1517/14712598.5.1.xxx © 2005 Ashley Publications Ltd ISSN 1471-2598
1
Smart polymeric carriers for enhanced intracellular delivery of therapeutic macromolecules
A) Poly(alkylacrylic acid) polymers
Lysosome
pH ~ 5.5
Endosome
pH ~ 5.5 – 6.5
*
n
*
*
n
O
HO
*
*
n
O
HO
*
O
HO
RME
DNA plasmids
antisense oligos
(ASODNs)
RNA therapeutics
Proteins, peptides
PEAA
PPAA
PBAA
B) Poly(acrylic acid-co-alkylacrylates) copolymers
Figure 1. The endosomal barrier to intracellular delivery of
therapeutic macromolecules.
ASODN: Antisense oligodeoxynucleotides; RME: Receptor-mediated endocytosis.
have been developed for use as endosomal-releasing agents in
gene and protein delivery systems [6-10].
Another approach to gain direct access to the cytoplasm of
targeted cells utilises protein transduction domains. Protein
transduction domains represent a class of peptides that includes
the Drosophila homeotic transcription factor (ANTP) [11,12],
herpes simplex virus type-1 transcription factor (VP22) [11],
and Tat peptide (Y47GRKKRRQRRR57) of the HIV-1 transactivating factor [12], which have been shown to translocate across
the cell membrane and into the cytoplasm of targeted cells in
an energy-independent [13] and receptor-less manner [14]. The
Tat peptide has been reported to improve the cytoplasmic
delivery of macromolecules, such as
β-galactosidase
protein [15], nanoparticles of 45 nm diameter [16], and pharmacologically active 2’-O-methyl phosphorothioate ASODNs [17].
The nonspecific activity of these peptides allows for a wide
range of applications; however, it can also lead to nonspecific
delivery of therapeutic molecules to undesired cell types.
The authors’ group has exploited key mechanistic aspects of
biological systems to develop synthetic ‘smart’ polymers that
can be utilised as intracellular delivery systems for macromolecular drugs. ‘Smart’ polymers are characterised by their ability to ‘sense’ the changes in environmental pH. They
reversibly switch from a hydrophilic stealth-like conformation
at physiological pH to a hydrophobic and membrane-destabilising state in response to acidic stimuli. The design of smart
polymeric carriers incorporates pH-sensing functionalities,
hydrophobic membrane-destabilising groups, versatile conjugation and/or complexation elements to allow drug incorporation, and an optional cell-targeting component. ‘Smart’
polymers retain the established advantages of standard polymer–drug conjugates, including high plasma levels, low renal
clearance and long circulation half-life [18,19]. ‘Smart’ polymers also improve the stability of their therapeutic cargo by
shielding against degrading enzymes. This paper reviews the
initial development of several classes of smart pH-sensitive
membrane-destabilising polymers and their ability to enhance
the cytoplasmic delivery of therapeutic molecules.
A
2
t
u
o
h
O
*
y
x
f
o
O
O
*
O
HO
P
r
Poly(AA-co-EA)
o
r
*
x
O
O
y
*
O
*
HO
Poly(AA-co-PA)
y
x
O
*
O
HO
Poly(AA-co-BA)
Figure 2. The chemical structure of: A) poly(alkylacrylic acid)
polymers, and B) poly(acrylic acid-co-alkylacrylates)
copolymers.
2. Polycarboxylate
polymers and copolymers
These polymer compositions contain the combination of carboxylate and hydrophobic groups found in naturally occurring
proteins, such as haemagglutinin, and in designed fusogenic
peptides, such as GALA. Tirrell and co-workers previously
described the pH-dependent disruption of lipid vesicles using
poly(ethylacrylic acid) (PEAA) [20]. PEAA destabilised model
lipid bilayers at acidic pH values (5.5 – 6.5) similar to that
found in the endosomal compartment. The authors initially
investigated the ability of PEAA to destabilise cell membranes
as part of a new poly(alkylacrylic acid) polymer family, including poly(propylacrylic acid) (PPAA) and poly(butylacrylic
acid) (PBAA), where the α-alkyl group progressively increased
by one methylene group from ethyl to propyl to butyl group,
respectively (Figure 2) [21,22]. This series of polymer compositions was designed to examine the effect of increasing the
length of the hydrophobic α-alkyl group on the pH-dependent, membrane-destabilising activity of the polymer. In order
to examine the effect of spatial arrangement between the carboxylic groups and the hydrophobic alkyl chains on membrane disruption, the authors developed another series of
random poly(acrylic acid-co-alkylacrylate) copolymers using
ethyl acrylate (EA), propyl acrylate (PA) and butyl acrylate
(BA) monomers (Figure 2). The membrane-destabilising activities of both series were evaluated as a function of polymer
Expert Opin. Biol. Ther. (2005) 5(1)
El-Sayed, Hoffman & Stayton
100
PPAAc
PEAAc
% Haemolysis
80
60
40
20
0
4
5
6
pH
7
8
P
r
Figure 3. The pH-dependent haemolysis of RBCs by PEAA
(100 µg/ml) (open circle) and PPAA (10 µg/ml) (filled
square). Results are the average of triplicate ± standard
deviations of the mean.
PEAA: Poly(ethylacrylic acid); PPAA: poly(propylacrylic acid);
RBC: Red blood cell.
o
h
concentration and environment pH using the standard red
blood cell (RBC) haemolysis assay.
PEAA was not active at physiological pH values (pH 7.4);
however, it became strongly haemolytic at pH 6.3 [21]. PEAA
produced 50% haemolysis of RBCs at a concentration of
35 µg/ml, indicating that it is equally effective to the membrane-destabilising peptide melittin [21]. PEAA reached its
maximum haemolytic activity at highly acidic values of
pH 5.0 and lower, whereas PPAA exhibited its maximum
haemolytic activity at pH 6.3 and lower (Figure 3) [21]. By
comparing the haemolytic activity of PEAA and PPAA with
similar molecular weights at a pH value similar to that of early
endosomes (pH 6.1), PPAA was ∼ 15 times more effective
than PEAA [21]. This result suggested the potential of PPAA
for destabilising the endosomal membrane at relatively early
development stages (i.e., early endosomes). The observed shift
in the pH-dependent, membrane-destabilising activity of PPAA
towards higher pH values (pH 6.1 – 6.3) compared with PEAA
is a result of the increased hydrophobic character of the pendant
alkyl chain by one methylene group [21]. The addition of
another methylene group to the pendant alkyl chain in PBAA
caused a significant shift in the pH profile [22]. PBAA was not
soluble at acidic pH values of 5.5; however, it became soluble at
physiological pH 7.4, causing 100% haemolysis of RBCs at
equivalent concentrations [22]. These results indicate that the
length of the hydrophobic alkyl group is a critical parameter in
determining the pH-dependent, membrane-destabilising
activity of poly(alkylacrylic acid) polymers.
A
t
u
The poly(acrylic acid-co-alkylacrylate) copolymers displayed a pH-dependent haemolytic activity that was dependent on the type of the incorporated alkylacrylate monomer
and the proportion of alkylacrylate to AA monomers [22]. A
1:1 random copolymer of AA and EA showed a haemolytic
activity similar to that of PEAA at pH 5.5 [21]. Copolymerisation of AA and PA produced a pH-sensitive membranedestabilising composition, where the observed haemolytic
activity increased with the increase in PA/AA ratio [22]. Similarly, the haemolytic activity of poly(AA-co-BA) copolymer
increased with increasing BA/AA ratio [22]. These results
indicate that the haemolytic activity of poly(acrylic acid-coalkylacrylate) copolymers increases with the increase in
length of the alkyl chain incorporated in the polymer backbone and the increase in alkylacrylate content. The
copolymers typically showed lower haemolytic activity
compared with the corresponding alkylacrylic acid homopolymers [21,22]. This is probably due to the regular spacing
of the hydrophobic alkyl groups along the homopolymer
backbone, which allows for better partition and penetration
into the lipid bilayers compared to the copolymer compositions where the alkyl chains are randomly dispersed between
the carboxylic groups [22]. Both homo- and copolymer compositions exhibited a concentration-dependent haemolytic
activity where RBC haemolysis increased with the increase in
polymer concentration [21,22].
o
r
3. Enhancement
f
o
of non-viral gene delivery
using PPAA
The excellent pH-dependent haemolytic activity of PPAA
motivated an investigation of its ability to enhance gene
expression and serum stability when incorporated in cationic
lipoplexes formulations [23]. Murine mouse fibroblasts
(NIH3T3) were treated with ternary mixtures of the cationic
lipid dioleyltrimethylammonium propane (DOTAP), the
pCMVβ plasmid DNA, and PPAA with theoretical cationic
to anionic charge ratio of 1.0 – 1.9 using binary DOTAP/
DNA mixtures as a control. In serum-free media, ternary
DOTAP/DNA/PPAA particles exhibited higher transfection
efficiencies compared with the binary DOTAP/DNA particles
at all +/- charge ratios, reaching maximum β-galactosidase
gene expression at a +/- charge ratio of 1.3 [23]. Binary
DOTAP/DNA particles produced maximum β-galactosidase
gene expression at +/- charge ratio of 1.9 [23]. In the presence
of serum, binary DOTAP/DNA complexes exhibited lower
transfection efficiencies compared with PPAA-containing ternary lipoplexes (Figure 4). The effects of serum proteins on the
stability, cellular uptake, and transfection efficiency of
DOTAP/DNA lipoplexes with or without PPAA were
recently studied at concentrations matching their average levels in whole blood [24]. DNA condensation, cell uptake, and
transfection results collectively showed that incorporation of
PPAA greatly improved cellular uptake and complex stability
in the presence of serum proteins [24].
Expert Opin. Biol. Ther. (2005) 5(1)
3
Smart polymeric carriers for enhanced intracellular delivery of therapeutic macromolecules
β-Gal activity (mu/mg protein)
2500
TSP2-null wounds. These experiments with knockout and
wild type mice collectively demonstrate that inclusion of
PPAA in cationic lipoplexes formulations greatly enhanced
transfection and resulted in the localised modulation of the
wound healing response.
DOTAP/DNA 1.9
DOTAP/DNA/PPAA 1.3
2000
1500
4. Application
500
0
0
10
20
30
40
50
% Serum
Figure 4. The effects of varying serum concentration on
gene transfer efficiency. The ability of DOTAP/DNA and DOTAP/
DNA/PPAA formulations were assessed in media containing FBS
levels up to 50% of the media. Each formulation was tested in
triplicate at each serum level.
DOTAP: Dioleyltrimethylammonium propane; FBS: Fetal bovine serum;
PPAA: poly(propylacrylic acid).
The favourable transfection results encouraged further
in vivo evaluation. A mouse model of wound healing was utilised for the in vivo evaluation of cationic lipoplexes with or
without PPAA [25]. This model is based on previous studies
that demonstrated that excisional wound healing is accelerated in thrombospondin-2 (TSP2)-null knockout mice [26]. In
the absence of TSP2, the excisional wounds exhibit irregular
deposition of extracellular matrix and enhanced vascularisation that is associated with significantly accelerated wound
healing. These results with the knockout mouse model suggested that delivery of a plasmid encoding an ASODN to
inhibit TSP2 expression could enhance healing in the wildtype mouse [27]. In addition, delivery of plasmid encoding for
TSP2 should reverse the TSP2-null phenotype as a built-in
control of the delivery and the mechanism.
The deposition of TSP2 in wild-type mice wounds was
found to be absent during the early inflammatory phase and
peaking on day 10, coinciding with the period of maximal
vascular regression [28]. The ternary DOTAP/DNA/PPAA formulations were injected into the wound on days 4, 8, and 12,
followed by wound evaluation on day 14 [25]. In the TSP2null mice, the DOTAP/DNA/PPAA formulations resulted in
a significantly higher TSP2 expression in comparison with
DOTAP/DNA controls [25]. In the wild-type mice, the delivery of PPAA-containing lipoplexes incorporating an antisenseencoding plasmid resulted in significantly enhanced disorganisation of the wound extracellular matrix, resembling that
seen with the TSP2-null wound healing process [25]. Immunohistochemical staining of wound sections for the endothelial
marker platelet–endothelial cell adhesion molecule-1 demonstrated that PPAA addition to the lipoplexes also resulted in
significantly greater vascularisation similar to that seen in
A
4
of PPAA in immunotoxin therapy
1000
t
u
o
h
Immunotoxins are a class of therapeutics that exploit monoclonal antibody specificity for the delivery of toxins to cellular
and tissue targets. Antibody-directed toxins are internalised
by endocytosis and enter the endosomal–lysosomal trafficking pathway where they are degraded to a much greater
extent [29] than the native toxin that is more efficiently translocated into the cytoplasm [30]. In order to modulate the
trafficking of antibody-targeted therapies and enhance their
cytoplasmic delivery, we examined the ability of PPAA to
serve as an endosomal releasing agent through its
pH-dependent
membrane-destabilising
affect.
The
pH-dependent, haemolytic activity of PPAA was retained
after being conjugated to a model protein [31].
The ability of PPAA to enhance the cytoplasmic delivery of
the model antibody-targeted protein complex was studied
with a biotinylated anti-CD3 antibody/streptavidin
complex [32]. The monoclonal anti-CD3 antibody 64.1
(MoAb) was previously shown to traffic to the lysosome with
minimal translocation to the cytoplasm [29,33,34]. This allowed
for a rigorous evaluation of PPAA effect on the cytoplasmic
delivery of MoAb-targeted systems [32]. The uptake of antibody–streptavidin systems was shown to be mediated by the
antibody and to be independent of the PPAA in flow cytometry experiments [32]. Changes in the trafficking of antibody–
protein complexes with or without PPAA–biotin were evaluated by visualising the intracellular distribution of different
complexes in Jurkat cells using confocal microscopy techniques [32]. Incorporation of PPAA–biotin in the antibody–
protein complex composition resulted in a diffuse intracellular staining of the cells indicating the cytoplasmic delivery of
the protein complex [32]. In contrast, protein complexes with
no PPAA–biotin showed a punctate fluorescence, indicating
entrapment of the protein complexes in the endosomal/lysosomal vesicles of the cells [32]. The quantity of the protein
complex present in the cytoplasmic fraction was compared
with that present in the total cell homogenate using quantitative western blotting techniques [32]. Results showed that
only the protein complexes with PPAA–biotin were detected
(∼ 73%) in the cyoplasmic fraction of cell homogenate [32].
The physical mixture of PPAA polymer with the MoAb–
biotin/streptavidin complex yielded some cytoplasmic release
(∼ 29%), indicating that PPAA can function in trans to
enhance cytoplasmic release [32]. These results collectively
demonstrate that PPAA is a potent endosomal releasing agent
and can be successfully utilised to enhance the cytoplasmic
delivery of therapeutic antibody-targeted conjugates that are
internalised through receptor-mediated endocytosis.
P
r
Expert Opin. Biol. Ther. (2005) 5(1)
o
r
f
o
El-Sayed, Hoffman & Stayton
Targeting
ligand
14C-labelled
COO
COOH
Endocytosis
S
S
SH
SH
pH
Ionically complexed
ODN
+
SH
Cytosol
Figure 5. The design and mechanism of endosomal release
of PDSA polymer–ODN ionic complexes. ODN molecules
(squares) are ionically complexed to the cationic peptides grafted
onto the PDSA polymers. As the pH drops in the endosome, the
COO- group of the pH-responsive monomer gets protonated to
COOH and the polymer backbone becomes hydrophobic, disrupts
the endosomal membrane, and releases the ODN into the
cytoplasm by dissociation of the ionic complexes.
ODN: Oligodeoxynucleotide; PDSA: Pyridyl disulfide acrylate.
5. Development
of pH-sensitive,
membrane-destabilising and
glutathione-reactive polymeric drug carriers
P
r
Given the activity of PPAA in enhancing the cytoplasmic delivery of pDNA [23,25] and model proteins [32], the next step was
to develop ‘functionalised’ pH-sensitive, membrane-destabilising polymers to carry different drug molecules by conjugation
or complexation. A new functionalised monomer named pyridyl disulfide acrylate (PDSA) was designed to allow direct conjugation of therapeutic molecules, complexing agents, and/or
targeting moieties via simple disulfide linkages (Figure 5) [35].
The first PDSA-containing polymer was prepared by copolymerisation with pH-sensitive (methylacrylic acid [MAA]) and
hydrophobic (BA) monomers. Several compositions and
molecular weight variants of this terpolymer were prepared by
varying the monomer feed ratios, solvent/monomer ratio, and
initiator’s molar concentration. The activity of the new carrier
was evaluated by examining the uptake and trafficking of
peptide/DNA-polymer conjugates.
A cationic peptide [(Lys)6-(Gly)3-(Cys)] was conjugated to the
new polymer backbone by reacting the PDSA units on the polymer backbone with the cysteine residues of the peptide to form
disulfide linkages. This cationic conjugate was used to complex
FITC-labelled ASODN (20 bases, MW: 6350 Da) through
electrostatic interactions to form polymer–ASODN ionic complexes. The uptake and intracellular trafficking of polymer–
ASODN ionic complexes was compared with that of free
ASODN using confocal microscopy analysis with THP-1 macrophage-like cells. The free ASODN exhibited a punctate fluorescence pattern indicating localisation in vesicular
compartments such as the endosomes or the lysosomes, whereas
the polymer–ASODN ionic complexes showed higher cellular
uptake and a diffuse staining indicating cytoplasmic delivery of
the incorporated ASODN [35]. These results were further
supported by the observed increase in THP-1 cell uptake of the
A
t
u
o
h
polymer with increasing polymer concentration [35].
The effect of the polymer’s molecular weight on its pH-dependent haemolytic activity was evaluated, showing the haemolytic
activity to increase with increasing polymer concentration and/or
molecular weight [35]. In addition, the new PDSA polymer was
found to be non-toxic neither to the NIH3T3 fibroblasts nor to
the THP-1 macrophage-like cells at concentrations up to
240 µg/ml and an incubation time of 16 h [35].
The influence of monomer composition on the pH-sensitive membrane-destabilising activity of PDSA polymers was
investigated by studying i) the influence of increasing the
length of the hydrophobic α-alkyl group substituted onto the
pH-sensitive monomer; and ii) the effect of incorporating a
hydrophobic monomer, such as BA, on the pH sensitivity and
membrane destabilising activity of new polymer
compositions [36]. A series of new polymer compositions was
synthesised by random free radical polymerisation of PDSA
monomers with pH-sensitive monomers containing an
α-substituted alkyl group that progressively increased in
length by one methylene group, including MAA, ethylacrylic
acid (EAA) and propylacrylic acid (PAA). The second series of
PDSA polymers was prepared by incorporating the hydrophobic BA monomers in the polymer backbone. The monomer
feed ratio, solvent/monomer ratio and initiator concentration
were controlled to allow the incorporation of ∼ 5 mole% of
PDSA units in the polymer backbone and to achieve an average molecular weight of 30 – 35 KDa. The haemolytic activity of these new polymer compositions was characterised as a
function of pH and polymer concentration to identify the key
structural features essential for pH sensitivity and membrane
destabilising activity of PDSA polymers.
In the first series of PDSA polymers, only the poly(PAA-coPDSA) copolymer exhibited a pH-dependent haemolytic
activity, whereas the poly(MAA-co-PDSA) and poly(EAA-coPDSA) copolymers were non-haemolytic at all the examined
concentrations and pH values (Figure 6). These results showed
that long alkyl chains (3 carbon atoms) are essential to allow
significant destabilisation of the RBC membrane. In the
second series of PDSA polymers, both the poly(EAA-co-BAco-PDSA) and poly(PAA-co-BA-co-PDSA) terpolymers exhibited pH- and concentration-dependent haemolytic activities
(Figure 6). At a polymer concentration of 5 µg/ml, the
poly(EAA-co-BA-co-PDSA) terpolymer exhibited high haemolytic activity (80% haemolysis of RBCs) at acidic pH values of
5.8 and 6.6. The poly(EAA-co-BA-co-PDSA) terpolymer also
showed a gradual increase in the haemolytic activity with the
increasing polymer concentration at pH 7.4 due to the significant increase in the hydrophobic character of the polymer
backbone upon the incorporation of BA monomers [36]. The
poly(PAA-co-BA-co-PDSA) terpolymer exhibited a similar
haemolysis profile except that the high hydrophobic character
of this polymer at low pH led to its precipitation out of solution
at pH 5.8 [36]. It is interesting to note that both the poly(EAAco-BA-co-PDSA) and poly(PAA-co-BA-co-PDSA) terpolymers
exhibited this high haemolytic/membrane destabilising activity
Expert Opin. Biol. Ther. (2005) 5(1)
o
r
f
o
5
Smart polymeric carriers for enhanced intracellular delivery of therapeutic macromolecules
A
100
90
% Haemolysis of RBCs
80
pH 5.8
pH 6.6
pH 7.4
68
70
60
50
40
30
17
20
10
-2
-1
-1
1
0
15
1
0
Poly(MAA-co-PDSA) Poly(EAA-co-PDSA) Poly(PAA-co-PDSA)
B
100
90
% Haemolysis of RBCs
80
79
70
60
50
40
30
20
10
t
u
10
1
0
A
o
r
90
pH 5.8
pH 6.6
pH 7.4
o
h
P
r
82
37
f
o
64
36
1
Poly(MAA-coBA-co-PDSA)
Poly(EAA-coBA-co-PDSA)
Poly(PAA-coBA-co-PDSA)
Figure 6. The influence of: (A) increasing the length of the hydrophobic α-alkyl group substituted on the pH-sensitive
monomer, and (B) incorporating the hydrophobic BA monomer in the polymer backbone on the polymer’s haemolytic
activity at a polymer concentration of 5 µg/ml.
BA: Butyl acrylate.
at low molecular weights of 9 and 12 kDa, respectively. This is
a significant finding as it suggests active chains can be designed
at molecular weights that allow for renal excretion after delivery of the therapeutic cargo. These results showed that the
incorporation of hydrophobic BA monomers had a significant
effect on the pH-responsiveness and membrane-destabilising
activity of PDSA polymers.
The haemolytic activity of the poly(PAA-co-BA-co-PDSA)
polymer conjugated to a cationic [(Cys)-(Gly)3-(Lys)6] was
determined along with the ionic mixtures of this conjugate
with model ASODN (18 bases, MW: 5699 Da) [36].
The poly(PAA-co-BA-co-PDSA)-[(Cys)-(Gly)3-(Lys)6] conjugate and the ionic complexes with ASODN showed pH- and
6
concentration-dependent haemolytic activities similar to that
of the parent polymer backbone [36]. These results indicate that
the poly(PAA-co-BA-co-PDSA) polymer could retain its
pH-dependent membrane-destabilising activity after drug loading. At present, the authors are evaluating the activity of these
new PDSA polymer compositions both in vitro and in vivo.
6. ‘Encrypted’
polymeric drug carriers
A related family of ‘smart’ polymeric carriers known as
‘encrypted’ polymers has been recently developed. These
polymers were termed ‘encrypted’ by analogy to encrypted
protein domains present in nature, where active domains of
Expert Opin. Biol. Ther. (2005) 5(1)
El-Sayed, Hoffman & Stayton
Membrane-disruptive
backbone ('masked')
Acid-degradable
acetal linkers
C
O
R
S
S
C
O
R
O
S
S
Endocytosis
O
R
R
Disrupted
endosome
pH
S
S
S
S
Conjugated
peptide DRUG
PEG grafts
'mask' the backbone
'Unmasked' backbone
disrupts endosomal
membrane
+
+
-
Ionically
complexed
ODN DRUG
f
o
Free DRUG delivered
into cytoplasm
o
r
Figure 7. Schematic diagram of the encrypted polymer design. The polymers are designed to be PEGylated and serum stable at
pH 7.4, but become disruptive to the endosomal membrane at the acidic pHs within the endosome. The polymers have the following
components: a membrane-disruptive backbone, acid-degradable linkers, PEG grafts, conjugated or ionically complexed drug molecules,
hexalysine peptide and targeting ligands. At pH 7.4 the polymers are PEGylated or ‘masked’; however, after endocytosis the aciddegradable linker hydrolyses and the polymer backbone becomes de-PEGylated or ‘unmasked’, thus causing endosomal membrane
disruption. The PEGs may be conjugated to the backbone via both acid-degradable linkages and disulfide bonds. The latter are reduced
in the cytoplasm to release the free drug.
PEG: Polyethylene glycol.
o
h
P
r
several extracellular and matricellular proteins are initially
masked but become exposed and activated by the action of
proteolytic enzymes at controlled time points. Similarly, the
encrypted polymers contain a ‘masked’ membrane-destabilising
backbone that becomes activated in response to acidic
endosomal pH gradients [37,38]. The authors’ first encrypted
polymer (E1) utilised a hydrophobic backbone synthesised by
free radical polymerisation of a styrene acetal monomer with
butyl methacrylate and dimethylaminoethyl methacrylate
monomers [37]. The hydrophobic backbone was ‘masked’ by
the direct conjugation of hydrophilic polyethylene glycol
(PEG) chains through acid-sensitive acetal linkages, which
was designed to improve the solubility and serum stability of
the polymer backbone [37]. Endocytosis of encrypted polymers would result in their accumulation in the endosomes
where the acid-sensitive acetal linkages between the PEG
chains and the hydrophobic backbone become hydrolysed.
The ‘unmasking’ of the hydrophobic backbone in turn causes
destabilisation of the endosomal membrane and release of the
therapeutic cargo into the cytoplasm (Figure 7).
The hydrolysis kinetics of the PEG grafts from the E1 polymer backbone was two orders of magnitude faster at pH 5.4 than
at pH 7.4, with an average half-life of 15 min [37]. A 5 µg/ml
polymer solution caused 100% haemolysis of RBCs at pH 5.0
while exhibiting no haemolysis at pH 7.4 [37]. The authors
investigated the ability of the E1 polymer backbone ‘masked’ by
heterobifunctional PEG grafts modified with (Lys)3-(mannose)3
A
t
u
groups to enhance the cytoplasmic delivery of ASODN for
inducible nitric oxide synthase (iNOS). The ionic complexes of
polymer (E1)-(Lys)3-(mannose)3 and ASODN for iNOS produced an 80% reduction in nitric oxide (NO) production,
compared with a 25% reduction by the free ASODN (Figure 8).
The observed increase in oligodeoxynucleotide (ODN) activity
was sequence specific as scrambled ODN showed no effect on
the production of NO. These results showed that the mannosetargeted E1 polymer could enhance the cytoplasmic delivery of
therapeutic ASODN in cultured cells.
The authors also evaluated the ability of the E1 polymer
backbone to enhance the endosomal escape of model polypeptides, [FITC-(His)6-(Gly)4-Cys], in macrophage RAW cells.
Confocal microscopy analysis demonstrated a punctate fluorescence pattern, indicating that the E1 polymer was not sufficiently hydrophobic to allow the release of the polypeptide into
the cytoplasm [37]. A new polymer backbone (E3) was designed
and synthesised to incorporate an additional hydrophobic monomer, BA. The E3 polymer showed enhanced haemolytic activity that was higher than that of the E1 polymer [37]. The FITC(His)6-(Gly)4-Cys peptide and methoxy-PEG-SH polymer were
conjugated to the E3 polymer backbone via disulfide linkages
and the intracellular distribution was examined by confocal
microscopy. Fluorescence images showed higher cellular uptake
and a more diffuse staining in the case of the E3–peptide conjugate compared with the punctate and localised staining
observed with the fluorescent peptide alone [37]. Results showed
Expert Opin. Biol. Ther. (2005) 5(1)
7
Smart non-degradable polymeric carriers for enhanced intracellular delivery of therapeutic macromolecules
iNOS (% of control)
100
80
60
40
20
0
Control
(no ODN,
no polymer)
ASODN
ASODN
+ polymer
Scrambled ODN
+ polymer
f
o
Figure 8. ASODN delivery with the mannose-targeted E1 polymer. Horizontal lines: 104 µg/ml of mannose-targeted E1 polymer
mixed with 10 µg/ml of ‘scrambled’ ODN. Black filling: 104 µg/ml of the mannose-targeted E1 polymer mixed with 10 µg/ml of ASODN
targeted against iNOS. Grey filling: 10 µg/ml of ASODN targeted against iNOS. White filling: A control in which cells were incubated with
10 mg/ml of LPS and 10 units/ml of IFN-γ. This sample was used as the 100% reference. After incubation with the appropriate ODN
formulation, each of the samples were stimulated with 10 µg/ml of LPS and 10 units/ml of IFN-γ. The cell supernatants were collected
after 8 h of stimulation and assayed for NO content.
AS: Antisense; iNOS: Inducible NO synthase; NO: Nitric oxide; ODN: Oligodeoxynucleotide.
that the E3 polymer enhanced the endosomal escape and
cytoplasmic delivery of the conjugated polypeptide.
These results motivated the testing of the encrypted polymers in a more challenging model system. Targeting via the
asialoglycoprotein (ASGP) receptor of the hepatocytes is
known to rapidly traffic the conjugates taken up by endocytosis directly to the lysosomal compartment where they get
degraded [39]. Lactose was conjugated to the PEG grafts
attached to the E1 polymer backbone to serve as a targeting
moiety for the ASGP receptor. The trafficking and intracellular distribution of FITC–PEG chains attached to the E1 backbone, together with the rhodamine-labelled ODN, were
characterised in mouse hepatocytes [38]. The fluorescence
images showed a more diffuse staining of FITC–PEG in the
case of the polymer–PEG–FITC conjugate compared with
punctate and localised staining observed with the FITC–PEG
alone [38]. Similarly, the encrypted polymer showed enhanced
uptake and there was a strongly enhanced cytoplasmic and
nuclear distribution of the rhodamine-labelled ODN. The free
ODN showed limited uptake that was restricted to vesicular
compartments, such as the endosomes/lysosomes [38]. These
results collectively indicate that encrypted polymers are effective carriers for enhancing the endosomal escape and the
cytoplasmic delivery of therapeutic molecules.
A
t
u
Bibliography
1.
2.
8
SKEHEL JJ, WILEY DC: Receptor binding
and membrane fusion in virus entry: the
influenza hemagglutinin. Annu. Rev.
Biochem. (2000) 69:531-569.
WILEY DC, SKEHEL JJ: The structure
and function of the hemagglutinin
o
h
P
r
7. Expert
o
r
opinion and conclusions
Advances in the area of gene, antisense, and vaccine therapy
require the development of specialised delivery systems that
can successfully shuttle a wide range of therapeutic molecules
into the cytoplasm of targeted cells. The authors’ group has
focused on developing ‘smart’ polymeric carriers that mimic
the features of the fusogenic peptides/proteins present in
viruses and other pathogenic organisms. These polymers are
designed to be hydrophilic at physiological pH, but become
hydrophobic and membrane-destabilising once they get trafficked into the endosome. The authors have developed several classes of smart, pH-sensitive, membrane-destabilising
polymers that proved effective both in vitro and in vivo. The
modular design provides a range of chemistries for the conjugation of drug molecules and/or targeting moieties. However,
one important goal that still requires development is the formulation of smart and biodegradable polymeric carriers.
These polymers would retain their pH-dependent membrane-destabilising activity until they deliver their therapeutic cargo, but would then be hydrolysed and degraded into
smaller fragments directly afterwards so that they can be
eliminated from the body without causing prolonged
accumulation or toxicity.
membrane glycoprotein of influenza virus.
Annu. Rev. Biochem. (1987) 56:365-394.
3.
4.
HUGHSON FM: Structural
characterization of viral fusion proteins.
Curr. Biol. (1995) 5:265-274.
REN J, SHARPE JC, COLLIER RJ,
LONDON E: Membrane translocation of
Expert Opin. Biol. Ther. (2005) 5(1)
charged residues at the tips of hydrophobic
helices in the T domain of diphtheria toxin.
Biochemistry (1999) 38:976-984.
5.
GRIMM D, KAY MA: From virus
evolution to vector revolution: use of
naturally occurring serotypes of adenoassociated virus (AAV) as novel vectors for
El-Sayed, Hoffman & Stayton
In vivo protein transduction: delivery of a
biologically active protein into the mouse.
Science (1999) 285(5433):1569-1572.
human gene therapy. Curr. Gene Ther.
(2003) 3:281-304.
6.
7.
WAGNER E, PLANK C,
ZATLOUKAL K, COTTON M,
BIRNSTIEL ML: Influenza virus
hemagglutinin HA-2 N-terminal fusogenic
peptides augment gene transfer by
transferring-polylysine-DNA complexes:
toward a synthetic virus-like gene transfer
vehicle. Proc. Natl. Acad. Sci. USA (1992)
89:7934-7938.
PLANK C, OBERHAUSER B,
MECHTLER K, KOCH C, WAGNER E:
The influence of endosome-disruptive
peptides on gene transfer using synthetic
virus-like gene transfer systems. J. Biol.
Chem. (1994) 269:12918-12924.
8.
TOLSTIKOV VV, COLE R, FANG H,
PINCUS SH: Influence of endosomedestabilizing peptides on efficacy of antiHIV immunotoxins. Bioconjug. Chem.
(1997) 8:38-43.
9.
SUBBARAO NK, PARENTE RA,
SZOKA FC JR, NADASDI L,
PONGRACZ K: pH-dependent bilayer
destabilization by an amphipathic peptide.
Biochemistry (1987) 26:2964-2972.
10.
PARENTE RA, NIR S, SZOKA FC JR:
pH-dependent fusion of
phosphatidylcholine small vesicles. J. Biol.
Chem. (1988) 263:4724-4730.
11.
LUNDBERG M, JOHANSSON M:
Positively charged DNA-binding proteins
cause apparent cell membrane translocation.
Biochem. Biophys. Res. Commun. (2002)
291(2):367-371.
12.
CONSOLE S, MARTY C, GARCIAECHEVERRIA C, SCHWENDENER R,
BALLMER-HOFER K: Antennapedia and
HIV transactivator of transcription (TAT)
‘protein transduction domains’ promote
endocytosis of high molecular weight cargo
upon binding to cell surface
glycosaminoglycans. J. Biol. Chem. (2003)
278(37):35109-35114.
13.
A
t
u
VIVES E, BRODIN P, LEBLEU B: A
truncated HIV-1 Tat protein basic domain
rapidly translocates through the plasma
membrane and accumulates in the cell
nucleus. J. Biol. Chem. (1997)
272(25):16010-16017.
14.
SCHWARZE SR, HRUSKA K,
DOWDY SF: Protein transduction:
unrestricted delivery into all cells? Trends
Cell Biol. (2000) 10(7):290-295.
15.
SCHWARZE SR, HO A,
VOCERO-AKBANI A, DOWDY SF:
16.
LEWIN M, CARLESSO N, TUNG CH
et al.: Tat peptide-derivatized magnetic
nanoparticles allow in vivo tracking and
recovery of progenitor cells. Nat. Biotechnol.
(2000) 18(4):410-414.
17.
ASTRIAB-FISHER A, SERGUEEV D,
FISHER M, SHAW BR, JULIANO RL:
Conjugates of antisense oligonucleotides
with the Tat and Antennapedia cellpenetrating peptides: Effects on cellular
uptake, binding to target sequences, and
biologic actions. Pharm. Res. (2002)
19(6):744-754.
18.
19.
TAKAKURA Y, HASHIDA M:
Macromolecular carrier systems for
targeting drug delivery: pharmacokinetic
considerations on biodistribution.
Pharm. Res. (1996) 13:820-831.
o
h
20.
THOMAS JL, TIRRELL DA:
Polyelectrolyte-sensitized phospholipid
vesicles. Acc. Chem. Res. (1992) 25:336-342.
21.
MURTHY N, ROBICHAUD JR,
TIRRELL DT, STAYTON PS,
HOFFMAN AS: The design and synthesis
of polymers for eukaryotic membrane
disruption. J. Control. Release (1999)
61:137-143.
22.
23.
24.
25.
MURTHY N, CHANG I, STAYTON PS,
HOFFMAN AS: pH-sensitive hemolysis by
random copolymers of alkyl acrylates and
acrylic acid. Macromolec. Symp. (2001)
172:49-55.
CHEUNG CY, MURTHY N,
STAYTON PS, HOFFMAN AS:
A pH-sensitive polymer that enhances
cationic lipid-mediated gene transfer.
Bioconjug. Chem. (2001) 12:906-910.
CHEUNG CY, STAYTON PS,
HOFFMAN AS: Poly(propylacrylic acid)mediated reduction in serum component
inactivation of cationic lipoplexes.
Submitted (2004).
KYRIAKIDES TR, CHEUNG CY,
MURTHY N et al.: pH-sensitive polymers
that enhance intracellular drug delivery
in vivo. J. Control. Release (2002)
78:295-303.
Expert Opin. Biol. Ther. (2005) 5(1)
KYRIAKIDES TR, ZHU YN, SMITH LT
et al.: Mice that lack thrombospondin 2
display connective tissue abnormalities that
are associated with disordered collagen
fibrillogenesis, an increased vascular density,
and a bleeding diathesis. J. Cell Biol. (1998)
140:419-430.
27.
KYRIAKIDES TR, HARTZEL T,
HUYEN G, BORNSTEIN P: Modulation
of angiogenesis and matrix remodeling by
localized, matrix-mediated, antisense gene
delivery. Mol. Ther. (2001) 3:842-849.
28.
KYRIAKIDES TR, TAM JW,
BORNSTEIN P: Accelerated wound
healing in mice with a disruption of the
thrombosponding 2 gene. J. Invest.
Dermatol. (1999) 113:782-787.
29.
PRESS OW, HANSEN JA, FARR A,
MARTIN PJ: Endocytosis and degradation
of murine anti-human CD3 monoclonal
antibodies by normal and malignant
T-lymphocytes. Cancer Res. (1988)
48:2249-2257.
30.
WAWRZYNCZAK EJ: Systemic
immunotoxin therapy of cancer: advances
and prospects. Br. J. Cancer (1991)
64:624-630.
31.
LACKEY CA, MURTHY N, PRESS OW
et al.: Hemolytic activity of pH-responsive
polymer-streptavidin bioconjugates.
Bioconjug. Chem. (1999) 10:401-405.
32.
LACKEY CA, PRESS OW,
HOFFMAN AS, STAYTON PS:
A biomimetic pH-responsive polymer
directs endosomal release and intracellular
delivery of an endocytosed antibody
complex. Bioconjug. Chem. (2002)
13:996-1001.
33.
GEISSLER F, ANDERSON SK, PRESS O:
Intracellular catabolism of radiolabeled antiCD3 antibodies by leukemic T cells.
Cell. Immunol. (1991) 137:96-110.
34.
SCHAFFAR L, DALLANEGRA A,
BREITTMAYER JP, CARREL S,
FEHLMANN M: Monoclonal antibody
internalization and degradation during
modulation of the CD3/T-cell receptor
complex. Cell. Immunol. (1988) 116:52-59.
35.
BULMUS V, WOODWARD M, LIN L
et al.: A new pH-responsive and
glutathione-reactive, endosomal membranedisruptive polymeric carrier for intracellular
delivery of biomolecular drugs. J. Control.
Release (2003) 93:105-120.
36.
EL-SAYED MEH, HOFFMAN AS,
STAYTON PS: Rational design of
composition and activity correlations for
o
r
NOGUCHI Y, WU J, DUNCAN R et al.:
Early phase tumor accumulation of
macromolecules: a great difference between
the tumor versus normal tissue in their
clearance rate. Jpn. J. Cancer Res. (1998)
89:307-314.
P
r
26.
f
o
9
Smart polymeric carriers for enhanced intracellular delivery of therapeutic macromolecules
pH-sensitive and glutathione-reactive
polymer therapeutics. J. Control. Release
(2004) (In Press).
37.
MURTHY N, CAMPBELL J, FAUSTO N,
HOFFMAN AS, STAYTON PS:
Bioinspired polymeric carriers that enhance
intracellular delivery of biomolecular
therapeutics. Bioconjug. Chem. (2003)
14:412-419.
A
10
38.
39.
t
u
MURTHY N, CAMPBELL J, FAUSTO N,
HOFFMAN AS, STAYTON PS: Design
and synthesis of pH-responsive polymeric
carriers that target uptake and enhance the
intracellular delivery of oligonucleotides.
J. Control. Release (2003) 89(3):365-374.
TYCKO B, KEITH CH, MAXFIELD FR:
Rapid acidification of endocytic vesicles
containing asialoglycoprotein in cells of a
human hepatoma cell line. J. Cell Biol.
(1983) 94:3191-3197.
o
h
P
r
Expert Opin. Biol. Ther. (2005) 5(1)
Affiliation
Mohamed EH El-Sayed, Allan S Hoffman &
Patrick S Stayton PhD†
†Author for correspondence
University of Washington, Department of
Bioengineering, Box 351721, Seattle,
WA 98195, USA
Tel: +1 206 685 8148; Fax: +1 206 685 8526;
E-mail: [email protected]
o
r
f
o