Novel biodegradable cationic core shell nanoparticles for codelivery of drug and DNA chapter 5 conclusions and recommendations

178


Chapter 5
Conclusions and Recommendations

5.1 Conclusions
Polycations are promising non-viral vectors for gene delivery, which provide many
advantages over viral vectors. However, low efficiency of gene transfection provided by
polycations limits their application. Codelivery of various drugs with plasmid DNA has
been reported to improve gene transfection and achieve synergistic effect of drug and
gene therapies. In this study, biodegradable cationic amphiphilic copolymers have been
developed. These copolymers can form core-shell structured micelles with a hydrophobic
core for encapsulation of drug molecules and a cationic shell for DNA binding.
Theoretically, these cationic micelles can be used to carry drug and gene simultaneously.
For the first time, it has been proved by this study that cationic core-shell structured
micelles can be used for codelivery of drug and gene. Successful in vitro and in vivo gene
transfection has been achieved using these cationic micelles. These cationic micelles may
make a promising carrier for drug and/or gene delivery.

5.1.1 Biodegradable cationic amphiphilic polymers
P(MDS-co-CES) and P(MDA-co-CEA) have been designed to have hydrophobic
pendant groups as the core-forming block and a cationic main chain as the shell-forming
block. The main chain PMDS or PMDA possesses a polyester structure carrying tertiary

179

amine and quaternary ammonium. It can be hydrolyzed and is used as the shell-forming
segment. Cholesterol is the pendant group and used as the core-forming segment. The
positive charges of the micelles come from the quaternary ammonium and protonization

of tertiary amines in acidic solution. It is believed that if the positive charges are well
distributed on the surface of the micelles, they must have well-controlled particle size
after complexation with DNA and strong DNA binding ability because there is no steric
hindrance for DNA binding.
PMDS and PMDA were synthesized by condensation reaction, in which triethylamine
was added to absorb hydrogen chloride produced. Since the monomer methyl
diethanolamine carried tertiary amine, which could also absorb hydrogen chloride.
Therefore, an excessive amount of triethylamine needed to be added to compete with the
monomer and the main chain of the produced polymer. The protonization of the tertiary
amine by hydrogen chloride might not affect this reaction but it would definitely affect
grafting of cholesterol, which was performed via quaternization reaction. Removal of
unreacted monomers after the synthesis of precursor polymer is also critical to the further
reaction. The precursor polymer was purified with either washing with or extraction with
sodium chloride- saturated water. The former method could produce PMDS or PMDA
with high molecule weight but was not efficient. The latter one provided very pure
PMDS and PMDA. For the synthesis of PEG-grafted P(MDS-co-CES) or PEG-grafted
(PMDA-co-CEA), monomethyl poly(ethyl glycol) was conjugated onto the main chain as
terminating agent of the condensation reaction.
The cholesterol group (Be-chol) was grafted onto PMDS, PMDA, PEG-PMDS and
PEG-PMDA by quaternization. The grafting degree of cholesterol could be varied by

180

changing the molar ratio of Be-chol to PMDS or PMDA in the quaternization reaction
(See Table 4.1). In fact, the hydrophilicity of the main chain, which served as the shell of
the micelles, was mainly attributed to the existence of positive charges of quaternary
ammonium. This is very different from commonly used cationic materials. This property
allowed the micelles to have DNA binding ability even in DI water. On the other hand,
there were tertiary amines existing in the main chain, which could be protonized in an
acidic environment. Therefore, to improve their DNA binding ability, the micelles were

complexed with DNA in an acidic solution such as sodium acetate buffer with pH 4.6.
The hydrophilicity of the polymers could be adjusted not only by the grafting degree of
cholesterol but also by environmental pH. Moreover, primary and secondary amines can
also be incorporated into the main chain for improvement of hydrophilicity and
endosome escaping ability. These adjustments could actually affect the stability, size,
zeta potential and ultimately gene transfection of the micelles.

5.1.2 Cationic micelles
The cationic amphiphilic copolymers P(MDS-co-CES) and P(MDA-co-CEA) could
self-assemble into core-shell structured micelles in DI water and sodium acetate buffer
with pH of 5.6 and 4.6 at very low concentrations. For example, the CMC of P(MDS-co-
CES) (Batch No. 120902b), obtained from the I
338
/I
333
ratios, was 1.9, 1.9 and 1.5 mg/L
in DI water and 0.02M sodium acetate buffer with pH of 5.6 and 4.6 respectively. The
CMC of P(MDA-co-CEA) (Batch No. 111002b) in DI water, obtained from the I
338
/I
333

ratios, was 2.4 mg/L. After PEG conjugation, the CMC increased slightly. For example,
the CMC of PEG5000-P(MDS-co-CES), obtained from the I
338
/I
333
ratios was 13.2, 18.8,

181

13.2 mg/L in DI water and 0.02M sodium acetate buffer with pH of 5.6 and 4.6
respectively. The molecular weight of PEG used could be optimized to yield micelles
with stable core-shell structure at low concentrations. These results suggest that the
micelles could form under neutral and acidic conditions, and they had a very stable core-
shell structure because of the rigid chemical structure of the core-forming segment
cholesterol. The TEM scans of the micelles also evidenced the formation of the
nanoparticles.
The stability of the micelles in DI water, acidic buffer, PBS buffer and PBS buffer
containing serum or BSA of different concentrations was characterized by particle size.
The buffer concentration (ionic strength) and pH were the two main factors to influence
the size of the micelles. At lower pH and ionic strength, the particle size was smaller. The
micelles remained stable in DI water for more than one month. However the size of the
micelles in PBS buffer tended to increase slightly as a function of time because of the
neutralization of the positive charges of the micelles by anions existing in the buffer. The
presence of proteins in the blood affected the stability of the micelles. An increased
protein concentration yielded greater size of the micelles because of the adsorption of
proteins on the surface. In PBS buffer containing 10% serum, particle size increased
immediately up to 425 nm from about 255 nm. In the presence of 1% and 3% BSA, the
size of the micelles changed in a narrow range from 135 to 161 nm. This may be due to
the presence of small BSA particles having a size of about 10 to 20 nm. However, the
conjugation of PEG could improve the stability of the micelles, which were designed for
systemic delivery of drug and/or gene.


182

5.1.3 Encapsulation of drug molecules
Two model drug compounds, indomethacin and pyrene, were encapsulated into
P(MDS-co-CES) and P(MDA-co-CEA) micelles by dialysis in DI water, 0.02M sodium

acetate buffer with pH of 5.6 and 4.6. P(MDA-co-CEA) micelles provided greater
loading capacity probably due to the presence of a larger amount of small micelles. After
drug loading, the particle size increased while the zeta potential decreased. An increased
drug loading level led to bigger micelles. A reduced pH value resulted in greater
encapsulation efficiency of both pyrene and indomethacin. Compared to indomethacin,
pyrene had a more rigid structure and might need more time to interact with the core-
forming segment cholesterol and assemble into the micelles, resulting in lower
encapsulation efficiency. However, at lower pH, the hydrophilicity and hydrophobicity of
the polymer might be well balanced, providing more time for pyrene to assemble into the
core of the micelles during the dialysis process. This improved the loading efficiency of
pyrene. Moreover, the weight ratio of polymer to drug also affected the loading
efficiency of drugs. An increased polymer/drug weight ratio yielded greater
encapsulation efficiency and slight improvement of drug loading level. An increased drug
loading level led to bigger micelles. More importantly, after drug loading, size
distribution became much more uniform and narrow possibility because of the interaction
between the drug and the core-forming segment cholesterol. The drug-loaded micelles
had suitable size for in vivo application and high zeta potential for DNA binding.

5.1.4 DNA binding ability

183

The DNA binding ability of P(MDS-co-CES), P(MDA-co-CEA) and PEG5000-
P(MDS-co-CES) micelles were studied at different pH and buffer concentration by
competition binding assays and agarose gel electrophoresis. The DNA binding ability of
P(MDS-co-CES) and P(MDA-co-CEA) micelles was affected by pH and ionic strength.
With decreasing pH from 7.0 (DI water) to 4.6, the DNA binding ability of the micelles
increased because more tertiary amines of the polymer were protonized and greater zeta
potential was obtained. Increasing ionic strength in the buffer weakened the DNA
binding ability of the micelles probably because ions could compete either with DNA for

the micelles or with the micelles for DNA. P(MDS-co-CES) and P(MDA-co-CEA)
micelles fabricated in 0.02M sodium acetate buffer with pH of 4.6 exhibited strong DNA
binding ability.
The PEGylation of P(MDS-co-CES) reduced the DNA binding ability of the micelles.
In particular, PEG5000-P(MDS-co-CES) micelles showed much lower DNA binding
ability than P(MDS-co-CES) micelles because of the shielding effect of PEG. Using PEG
of low molecular weight is expected to improve their DNA binding ability.
Indomethacin- and pyrene-loaded P(MDS-co-CES) micelles also exhibited strong
DNA binding ability. Compared to P(MDS-co-CES) micelles, the DNA binding ability of
the drug-loaded P(MDS-co-CES) micelles was slightly lower because of the reduced zeta
potential and the increased particle size after drug loading. The complete retardation of
DNA was achieved at the N/P ratio of about 2 and 3 for the blank P(MDS-co-CES)
micelles and the indomethacin- or pyrene-loaded P(MDS-co-CES) micelles respectively.
The size of the micelles/DNA complexes was slightly bigger than the blank micelles.
The complexes were stable in aqueous solution even at the zero zeta potential point when

184

they were prepared by adding the micelles into DNA solution. However, when DNA was
added into the micelle solution, the micelles tendered to aggregate at the zero zeta
potential point.

5.1.5 Structural integrity of drug-loaded micelles during the DNA binding process
The structural integrity of drug-loaded micelles during the DNA binding process and in
the growth medium with 10% FBS were evaluated by using pyrene as a probe. When
pyrene partitioned into the core of P(MDS-co-CES) micelles from an aqueous solution,
the I
338
/I
333

ratio increased, indicating that the core of the micelles was more hydrophobic
than the aqueous solution. With DNA binding at the N/P ratio of 0.2 to 6, the I
338
/I
333

ratio further increased, suggesting that the microenvironment of pyrene was more
hydrophobic after DNA binding. DNA molecules bound on the surface of nanoparticles
enhanced the stability of core-shell structure. On the other hand, the size of the micelles
ranged from 152 to 299 nm at the N/P ratio of 0.4 to 10, indicating that the micelles did
not collapse during the process of DNA binding. Furthermore, the growth medium has
also been evidenced to further enhance the structural integrity of the micelles. This
indicates that unlike other cationic polymer/DNA complexes the complexes can maintain
the structural integrity with the presence of serum. This is significant for the gene
transfection,

5.1.6 In vitro and in vivo gene expression
In vitro gene transfection was performed in various cell lines and primary human
dermal fibroblasts using P(MDS-co-CES), PMDS and pegylated P(MDS-co-CES)

185

micelles as the vector. P(MDS-co-CES) micelles provided similar level of luciferase
expression in HEK293 and HepG2 cells as PEI and higher transfection level than PEI in
4T1. However, in HeLa cells and human dermal fibroblasts, the luciferase expression
efficiency obtained from P(MDS-co-CES) micelles/DNA complexes was lower than that
from PEI/DNA complexes. At the same N/P ratio, PEG2000-P(MDS-co-CES) micelles
showed lower luciferase expression efficiency than P(MDS-co-CES) micelles because of
their lower DNA binding ability. However, increasing the N/P ratio could significantly
improve the gene transfection efficiency of PEG2000-P(MDS-co-CES) micelle. In

contrast, PMDS provided much lower luciferase transfection efficiency compared to
P(MDS-co-CES) and PEG2000-P(MDS-co-CES. This proved that the core-shell structure
with positive charge on the surface is crucial for the high gene transfection level of the
polymer. Furthermore, the results of GFP gene transfection performed in different cell
lines also indicate that the high gene transfections of P(MDS-co-CES) were mainly
achieved through its high cells uptake efficiency. This shows that besides the buffering
ability of the cationic polymer in acidic environment, the stability of the particles is also
another concern to improve the gene transfection efficiency.
In addition, the enhancement of gene expression by cyclosporin A in KB-31-MA cells
and by paclitaxel in 4T1 cells has proved that the micelles is a kind of excellent
codelivery system for hydrophobic drug and nucleic acid including DNA, RNA.
For proof of the principle, in vivo gene transfection was performed in the cochlea of
guinea pigs and subcutaneous breast cancer model established in balb/C mice using
P(MDS-co-CES) micelles. After 24 hours, the micelles/DNA complexes crossed the
round window membrane and luciferase as well as GFP expression was observed in the

186

cochlea of guinea pigs. The P(MDS-co-CES) micelles may be used as a carrier for gene
delivery to the inner ear for correcting hearing loss. At the same time, the luciferase gene
expression in the breast cancer performed by local injection shows that PMDS-co-CES
complexes had higher gene expression level than PEI complexes. The experiment
performed by tail vein injection shows that P(MDS-co-CES) complexes can successfully
avoid entrapment by the lung and possess higher passive targeting ability than PEI
complexes. This further evidenced that P(MDS-co-CES) complexes is a very stable DNA
carrier.
Furthermore, the in vivo codelivery of paclitaxel and luciferase gene to the tumor model
via local injection also shows that the gene expression level has been improved
significantly. This indicates that the system is an excellent codelivery system not only in
vitro but also in vivo to realize synergistic effect.

In conclusion, a biodegradable cationic micelle system has been designed, which can
carry both drugs and DNA simultaneously. These micelles exhibited a much lower
cytotoxicity compared to conventionally used PEI, and yielded a comparable luciferase
expression level and greater percentage of GFP-positive cells when compared to PEI.
Gene expression was also successfully achieved in the cochlea of guinea pigs with the
micelles/DNA complexes. With this unique cationic micelle system, a variety of
compounds could be codelivered with plasmid DNA to enhance gene transfection and/or
achieve a synergy between drug and gene therapies. Such a system could be used to carry
an anti-cancer drug (such as paclitaxel or doxorubicin) in its hydrophobic core, while
binding a nucleic acid agent on its cationic shell. The nucleic acid component might be a
vector encoding an antisense molecule directed against the P-glycoprotein mRNA in the

187

target cell. Such a system could inhibit P-glycoprotein expression by the target, and
hence, incapacitate its ability to establish multi-drug resistance, a common trait among
cancer cells. This, coupled with the cytotoxic effects of the anti-cancer drug, should
enhance the therapeutic effect of the system.

5.2 Recommendations for future work
1. To test PEG-grafted P(MDS-co-CES) micelles on a multi-drug resistant tumor
model through systemic delivery (i.e. intravenous injection). An anticancer drug,
paclitaxel and si-RNA encoded plasmids can be chosen to gain the benefits that the
codelivery system can provide. The results should be compared with those obtained
from free paclitaxel in the absence of si-RNA.
2. To conjugate biological ligands onto the shell of the micelles for active targeting.
One of the examples is to use folate. It is well known that folate receptor is over
expressed on the surface of most of the cancer cell types. With the conjugated folate,
active targeting of the above system, the micelles/paclitaxel/si-RNA, to tumor
tissues may be realized.

3. Moreover, a chemical compound such as chloroquine that can help break down
endosome membrane can be codelivered with DNA to improve gene expression in
some cell lines such as human dermal fibroblasts, in which the current cationic
micelles provided low gene expression efficiency.
4. Primary and secondary amines can be chemically incorporated into the main chain
of the cationic polymer to improve its buffering effect, which may eventually induce

188

the breakage of endosome membrane to increase gene expression efficiency of the
cationic micelles.