Novel biodegradable cationic core shell nanoparticles for codelivery of drug and DNA 4

4.1.3 Synthesis and characterization of PMDS-co-CES and PMDA-co-CEA The synthesis of PMDS-co-CES and PMDA-co-CEA was performed by grafting Be-chol onto PMDS and PMDA through quaterniz

N-(2-bromoethyl) carbarmoyl cholesterol (Be-chol) has a bromoethyl group that was

used to quaternize the main chain at the amino group and produce positive charges at the same sites Be-chol was also designed as the random dispersed hydrophobic pendant chains It was synthesized by connecting bromoethylamine onto the cholesterol molecule through the amidation reaction with cholesteryl chloroformate as showed in Scheme 1 Be-chol was obtained in yield of ~ 78% after twice consecutive-purification by re-crystallization with ethanol and acetone, respectively TLC analysis showed one point at

Rf of 0.68 in the mixture of toluene, hexane and methanol (8:8:1), indicating that Be-chol was pure Figure 4.1 and Figure 4.2 display the 1H-NMR and FTIR spectra of purified Be-chol, respectively As showed in Figure 4.1, the 1H peak at δ 5.10 (Signal HN) was

due to the amide groups (CONH) (See Fig 4.1) δ 3.60 (Signal H4) and 3.61 (Signal H5) were attributed to the 2-bromoethyl groups δ 4.52 (H1) and 5.40 (H2) were associated with the cholesterol units The ratio of the H1, H2, HN, H4 and H5 peak areas was determined to be 1:1:1:2:2, confirming the successful synthesis of Be-chol The FTIR spectrum of Be-chol also evidenced its successful synthesis The IR peak at 3325 cm-1

was due to -NH- stretching (see Figure 4.2) Peaks from -C=O stretching and -NH-

bending overlapped at 1685 cm-1 The peak at 1536 cm-1 was attributed to -C-N-

stretching In summary, pure Be-chol was successfully synthesized The purity of Be-chol

is important for its further grafting onto the main chain For instance, the impurity, cholesteryl formic acid, may act as a catalyst to promote the hydrolysis of the main chain (PMDS or PMDA)

Figure 4.1 1H-NMR spectrum of N-(2-bromoethyl) carbarmoyl cholesterol (Be-chol)

NH

Br

O

H H

H O

c d

Figure 4.2 FTIR spectrum of N-(2-bromoethyl) carbarmoyl cholesterol (Be-chol)

4.1.2 Synthesis and characterization of PMDS and PMDA

Poly(N-methyldiethylamine sebacate) (PMDS) and Poly(N-methyldiethylamine

adipate) (PMDA) are the main chains of the designed polymers The successful synthesis

of PMDS was verified by 1H-NMR and FTIR spectra as shown in Figures 4.3 and Figure 4.4, respectively NMR peaks at δ 2.71–2.73 (signal a), δ 1.62 (signal b) and δ 1.32

(signals c and d) were attributed to the protons of four different -CH2- groups from the sebacate units (see Figures 4.3) Peaks at δ 4.17–4.19 (signal e) and δ 2.30-2.37 (signals f

and g) were due to protons of two different -CH2- groups and the -CH3 group linked to the nitrogen atom IR spectrum also confirmed the polyester formation (see Figure 4.4)

The -C=O stretching shifted to a lower wavenumber (1736 cm-1) compared to carbonyl halide (1805 cm-1) due to the inductive effect of halide The peak at 1172 cm-1 was

attributed to C-O

The 1H-NMR and FTIR spectra of PMDA confirmed its successful synthesis (Figure

4.5 and Figure 4.6) NMR peak at δ 1.63 (signal b) was attributed to the -CH2- groups of

adipate, which were connected to another two -CH 2 - groups Peaks at δ 2.67-2.76 (signal

a) were attributed to the -CH2- groups of adipate (Figure 4.5), which were connected to

carboxyl group and another -CH 2 - group Peaks at δ 2.27-2.37 (signals d and e) came

from the protons of the -CH 2 - and -CH 3 groups connected with the nitrogen Peaks at δ

4.1-4.2 (signal c) were attributed to the -CH 2 - groups connected to the oxygen The FTIR

spectrum of PMDA also evidenced its successful synthesis (Figure 4.6) Stretching

vibration of C-O at 1174 cm-1 and stretching vibration of C=O at 1732 cm-1 were observed, indicating the existence of ester group in the polymer

From the 1H-NMR spectra of PMDS and PMDA, no impurity peaks were observed, especially the peaks of triethylamine that may influence the subsequent reaction In addition, it indicates that the purification method applied was suitable and effective

n

O N

O C

O

C O a

b c d

e f

g

CH3

Figure 4.4 FTIR spectrum of PMDS

Figure 4.5 1H-NMR spectrum of PMDA

n

O N

O C

O

C O a

Figure 4.6 FTIR spectrum of PMDA

4.1.3 Synthesis and characterization of P(MDS-co-CES) and P(MDA-co-CEA)

The synthesis of P(MDS-co-CES) and P(MDA-co-CEA) was performed by grafting

Be-chol onto PMDS and PMDA through quaternization reaction This reaction needs to

be performed at a relatively high temperature when alkyl bromide is used as the reagent for quaternization The purposes to introduce the cholesteryl group onto PMDS and PMDA are to use the cholesteryl group as the core-forming block and to produce positive

charges on the main chain The successful synthesis of P(MDS-co-CES) and

P(MDA-co-CEA) was evidenced by 1H-NMR and FTIR spectra Figure 4.7 and Figure 4.8 show the

1

H-NMR and FTIR spectra of P(MDS-co-CES), respectively The 1H-NMR spectrum of P(MDS-co-CES) illustrates peaks at δ 2.7–2.8 (signal a), 1.5–1.7 (signal b), 1.2–1.4 (signals c and d), 4.0–4.2 (signal e) and 2.2–2.4 (signals f and g) due to protons from the

groups (signal h) The peak at δ 0.7 was from the methyl group directly linked to the cyclic hydrocarbon (signal i) The information provided by the 1H-NMR spectrum of

P(MDS-co-CES) proved that the cholesteryl group was successfully grafted onto the PMDS main chain Figure 4.8 shows the FTIR spectrum of P(MDS-co-CES), which also

evidenced the successful quaternization The peak at 1252 cm-1 was attributed to C-N

stretching of amine The shift and increased intensity of this peak compared with that of PMDS (1240 cm-1) illustrated the formation of a quaternary ammonium salt

The 1H-NMR and FTIR spectra of P(MDA-co-CEA) also gave similar results as P(MDS-co-CES) As shown in Figure 4.9, the wide peak at δ 2.66 (signal a) and the peak

at δ 1.67 (signal b) were from the protons of the methylene groups (-CH2-) in the adipate segments The multiple peaks at δ 4.0-4.2 (signal c) came from the methylene group (-

CH2-) in the N-methyldiethanolamine segments Signal d of another methylene group linked to the nitrogen atom of N-methyldiethanolamine was overlapped with signal e of

the methyl group directly linked to the nitrogen at δ 2.3-2.4 The peaks at δ 0.7-1.2 were from the cholesterol group In particular, the peaks at δ 5.37 (signal f) and δ 0.69 (signal

g) came from the protons linked to the carbon with double bond (=CH-) and the methyl

group linked to the cyclic hydrocarbon, respectively Moreover, Figure 4.10 illustrates the peak at 1251 cm-1 from the C-N stretching of amine The shift and increased intensity

of this peak compared with that of PMDA (1240 cm-1) illustrated the formation of a quaternary ammonium salt

The degree of grafting (R g ), defined as the ratio of the amines quaternized by

N-(2-bromoethyl) carbarmoyl cholesterol to the total number of amines on the PMDS main chain, can be estimated as follows,

R g = (∆A p N Hm /∆A m N Hp) × 100%,

Where ∆A p is the area of the selected peak from the pendant chain, ∆A m is the area of the

selected peak from the main chain, N Hp is the number of hydrogen atoms in the selected

group from the pendant chain, and N Hm is the number of hydrogen atoms in the selected group from the main chain Only suitable protons from the pendant chain and the main chain of the polymers were selected in the calculation The proton signal selected should not overlap with signals from other protons Furthermore, those protons affected by the

quaternized amines should not be used For P(MDS-co-CES), the proton of the methylene

group linked to the carbonyl group of the main chain (signal a), as well as the proton of

the methylidyne group (-CH=) linked to the double bond (signal h) and the proton of the

methyl group linked to the hexane and pentane cycles of the pendant chain (signal i) were

considered suitable for use in the estimation of R g For the proton of methylene linked to the carbonyl group overlapped with other signals The proton of methylene linked to the

ester (O=C-O-CH 2 -, signal c) on the main chain was chosen since this proton is far from

the nitrogen atom of the quaternary ammonium The inductive effect of the quaternary ammonium on the proton was neglected For the pendant chain, the same protons were

used as P(MDS-co-CES) (i.e signal f and signal g) Based on the peak areas of signal a and signal i, R g for P(MDS-co-CES) (Batch No 120902b) was estimated to be 27.0% (i.e R g=∆AHi×4×100%/3×∆AHa=2.046×4×100%/3×10.1=27.0%) Based on the peak

areas of signal c and signal g, R g for P(MDA-co-CEA) (Batch No 110102b) was estimated to be 56.0% (i.e R g=∆AHg×4×100%/3×∆AHc=2.59×4×100%/6.17×3=56.0%)

By changing the molar ratio of the pendant chain to the PMDS or PMDA main chain, R g

of the cholesterol moiety and the positive charge of P(MDS-co-CES) could be

modulated

The polymers with different cholesteryl grafting degree were synthesized by changing the amount of Be-chol precursor The purity of PMDS and PMDA may influence the grafting degree of Be-chol onto PMDS and PMDA For example, the residue of triethylamine added to absorb HCl could undergo quaternization reaction with Be-chol Therefore, even a small amount of triethylamine and its salt form can affect the grafting reaction significantly

Figure 4.7 1H-NMR spectrum of P(MDS-co-CES)

p

O C

Br + C

e f

g

h i

Figure 4.8 FTIR spectrum of P(MDS-co-CES)

p

O C

CH 3

Br + C

O

C

O a

b

c d

e

f g

Figure 4.10 FTIR spectrum of P(MDA-co-CEA)

4.1.4 PEGylation of PMDS, PMDA and P(MDS-co-CES), P(MDA-co-CEA)

PEGylation means conjugation of PEG onto the polymer The main purpose of

introducing PEG onto the cationic polymer is to increase the stability of micelles/DNA complexes by preventing protein absorption In this study, PEG with molecular weight (Mn) of 5000, 2000, 1000 and 550 (labeled as PEG5000, PEG2000, PEG1100 and

PEG550 respectively) was conjugated onto PMDS and P(MDS-co-CES) as a terminal

group of the polymers

Figure 4.11 to Figure 4.15 show the 1H-NMR spectra of PEG550, PEG1100, PEG2000, and PEG5000-conjugated PMDS and PEG5000-conjugated PMDA, respectively The peak at δ 3.65 was from PEG An increased molecular weight of PEG led to increased intensity of the peak, indicating the increase in the relative length of PEG

of the polymer Figure 4.16 to Figure 4.20 display the 1H-NMR spectra of PEG550,

PEG1100, PEG2000, and conjugated P(MDS-co-CES) and conjugated P(MDA-co-CEA), respectively Similarly, the intensity of the peak at δ 3.65

PEG5000-increased as increasing the molecular weight of PEG, indicating the increase in the relative length of PEG of the polymer Peaks at δ 0.7-1.2 were from the cholesteryl groups The single peak at δ 0.69 was attributed to the methyl group linked to the cyclic hydrocarbon These results show that the grafting of cholesteryl group and the conjugation of PEG onto PMDS and PMDA were successful However, the grafting degree of cholesteryl group onto PEGylated PMDS and PMDA was much lower than that onto PMDS and PMDA (Table 4.1, and Figure 4.16-4.20) That means that PEG influenced the quaternization reaction of Be-chol with the tertiary amine on the PMDS and PMDA main chains It is possibly because the conjugation of PEG onto PMDS and PMDA changed the polarity of the polymer and the compatibility of the polymer with toluene, which restricted fully stretching of the polymer chain in toluene and thus provide steric hindrance to the quaternization reaction

Figure 4.11 1H-NMR spectrum of PEG550-PMDS

Figure 4.12 1H-NMR spectrum of PEG1100-PMDS

Figure 4.13 1H-NMR spectrum of PEG2000-PMDS

Figure 4.14 1H-NMR spectrum of PEG5000-PMDS

Figure 4.15 1H-NMR spectrum of PEG5000-PMDA

Figure 4.16 1H-NMR spectrum of PEG550-P(MDS-co-CES)

Figure 4.17 1H-NMR spectrum of PEG1100-P(MDS-co-CES)

Figure 4.18 1H-NMR spectrum of PEG2000-P(MDS-co-CES)

Figure 4.19 1H-NMR spectrum of PEG5000-P(MDS-co-CES)

Figure 4.20 1H-NMR spectrum of PEG5000-P(MDA-co-CEA)

4.1.5 Molecular weight, grafting degree as well as PEG contents of the polymers

The molecular weight, grafting degree and PEG contents of the polymers are listed in

from the 1H-NMR spectra The weight average molecular weight (Mw) of PMDS could

reach as high as 18.5 kDa while the Mw of P(MDS-co-CES) could be up to 7.9 kDa The molecular weight of P(MDS-co-CES) was usually lower than the PMDS, from which the P(MDS-co-CES) was synthesized This indicates that the grafting reaction may result in

the degradation of the main chain, decreasing the molecular weight The cholesteryl

grafting degree of P(MDS-co-CES) ranged from 9.4% to 56.2%, depending on the purity

of PMDS and the amount of Be-chol added As discussed in the previous sections, triethylamine in PMDS could significantly affect the grafting degree The 1H-NMR spectra of PMDS evidenced that extracting PMDS/toluene solution using NaCl-saturated aqueous solution for at least 4 times effectively removed the triethylamine and thus increased the grafting degree when other conditions remained the same The amount of Be-chol used in the quaternization reaction also influenced the grafting degree For example, polymers with the batch numbers of 191003a and 191003b were synthesized by using Be-chol to PMDS unit ratios of 0.5 and 0.2 respectively The grafting degree obtained were 13.6% and 9.4% respectively (Table 4.1) However, the grafting degree seldom exceeded 60% even though the molar ratio of Be-chol to PMDS unit increased to 1.5 (Batch No.010902) This is possibly because the structure of the cholesteryl group

provided steric hindrance for the reaction For PEGylated P(MDS-co-CES) polymers, the

grafting degrees ranged from 4.25% to 16.35%, where the molar ratio of Be-chol to

PMDS unit was at 1.0 except for PEG2000-P(MDS-co-CES) (Batch No 051103b), for which the ratio of 0.5 was used Obviously, the grafting degrees of PEGylated P(MDS-

co-CES) were much lower than that of P(MDS-co-CES) As discussed in Section 4.1.4,

PEG-conjugated PMDS may influence the conformation of the chain in the solvent and

thus hinder the grafting reaction This is possibly the main factor that renders the low

grafting degree of PEG-P(MDS-co-CES) In in vitro gene expression experiments these

polymers, with or without grafted PEG or cholesteryl groups, will be tested to uncover the influence of polymer structure on gene transfection level

4.1.6 Thermal properties of the polymers

The thermal properties of polymeric materials are a key factor that determines their storage conditions and sterilization methods to be employed The thermal properties of

PMDA, P(MDA-co-CEA), PMDS and P(MDS-co-CES) polymers were studied using TGA and DSC Figure 4.21 to Figure 4.24 show the TGA spectra of PMDA and P(MDA-

co-CEA), PMDS and P(MDS-co-CES), respectively The maximal degradation

temperature of P(MDS-co-CES) and P(MDA-co-CEA) appeared at 339.7ºC and 339.8ºC,

respectively while those of PMDS and PMDA occurred at temperatures higher than

400ºC The similar thermal degradation patterns between P(MDS-co-CES) and

P(MDA-co-CEA) as well as between PMDS and PMDA may be due to their similar chemical

structure The lower degradation temperature of P(MDS-co-CES) and P(MDA-co-CEA)

compared with PMDS and PMDA suggests that the quaternization of the tertiary amine

on the main chain may make the polymer more degradable The dissociation of cholesteryl groups from the main chain may also cause the degradation to occur at lower temperatures From the cumulative degradation curves of these polymers, there was very small weight loss starting from 100ºC, which was water loss The small water loss shows that the moisture existing in the polymers was not high This may be due to the

hydrophobic nature of the polymers This indicates that the storage and transportation conditions should not be stringent

From Figure 4.25 and Figure 4.26, it can be seen that the glass transition temperature

of P(MDS-co-CES) and P(MDA-co-CEA) was 73.5ºC and 24.3ºC, respectively The melting point of P(MDS-co-CES) was also observed at 138.2ºC

Figure 4.21 TGA spectrum of PMDA (Batch No 110902)

Figure 4.22 TGA spectrum of P(MDA-co-CEA) (Batch No 111002a)

Figure 4.23 TGA spectrum of PMDS (Batch No 120902a)

Figure 4.24 TGA spectrum of P(MDS-co-CES) (Batch No 120902b)

Figure 4.25 DSC spectrum of P(MDS-co-CES) (Batch No 120902b)

Figure 4.26 DSC spectrum of P(MDA-co-CEA) (Batch No 111002a)

4.1.7 In vitro degradation of P(MDS-co-CES) and P(MDA-co-CEA)

The main chains of P(MDS-co-CES) and P(MDA-co-CEA) are polyester It is known

that polyester is degradable in aqueous solution, especially in an acidic environment Degradable materials are easily accepted by scientists in the biomedical field since the degradable materials can be cleared out of the body through the renal system during the degrading process and prevent the accumulation of the foreign materials in the body The

in vitro degradation tests of P(MDS-co-CES) and P(MDA-co-CEA) were performed in

PBS (pH 7.4) Figure 4.27 shows the weight loss of the polymers as function of

incubation time On the third day, the weight of P(MDS-co-CES) slightly increased

because of water uptake and ions from the PBS buffer After that, it underwent rapid

weight loss At Week 8, its weight loss was about 54% P(MDA-co-CEA) experienced

much slower degradation After eight weeks incubation, only 25% weight loss was found The samples harvested after the weight loss test can neither dissolve in water nor organic solvents including DMF, which makes the measurement of molecular weight of the

degraded polymers difficult using GPC This also makes the characterization of the residue by 1H-NMR and FTIR very difficult The weight loss of P(MDS-co-CES) after

the third day is mainly due to the hydrolysis of the main PMDS chain Both sebacic acid and methyl diethanolamine produced after hydrolysis were soluble in PBS Thus, the main residue of the polymers after eight weeks of degradation should be cholesteryl

group, which was not soluble in PBS Similarly, P(MDA-co-CEA) had a great amount of residue (75%) after eight weeks of degradation Because the grafting degree of P(MDA-

co-CEA) used was higher than that of P(MDS-co-CES) (56.0% versus 27.0%) and the

unit molecular weight of PMDA is also lower than that of PMDS, the weight loss of

P(MDA-co-CEA) was slower than that of P(MDS-co-CES) However, the figure 4.27 also shows that P(MDA-co-CEA) can be hydrolyzed more easily, since , unlike P(MDS-

co-CES), the weight of PMDA-co-CEA didn’t increase at the third day

In summary, the weight loss of polymer in the period of 8 weeks can surely evidence the degradation of the polymer Although the difficulty of detecting degraded product due to the ionization of the polymer make the study of degradation mechanism impossible, it is reasonable to say that the polymer should be degraded by hydrolysis since it possess ester structure

0 20 40 60 80 100

The critical micelle concentration (CMC) is an important factor to evaluate the micelles formation by amphiphilic polymers through self-assembly [Jones M-C., 1999]

In addition, it is also a critical factor to understand the stability of micelles in the post blood administration

Pyrene is a commonly used fluorescent probe to measure CMC of amphiphilic polymers With the formation of micelles, hydrophobic pyrene goes into the hydrophobic core of micelles In this procedure the microenvironment of pyrene changes from

hydrophilic to hydrophobic, the ratio of I 1 /I 3 from the emission spectra of pyrene

decreases but the ratio of I 338 /I 333 from its excitation spectra increases abruptly near the

CMC Since the I 1 /I 3 ratio is affected by the excitation wavelength and may result in an erroneous CMC value [Jones M-C., 1999], both emission and excitation spectra were

used to ascertain the formation of micelles Since P(MDS-co-CES) is a polyelectrolyte, the effect of pH and ionic strength on the CMC of P(MDS-co-CES) was also studied Figure 4.28 and Figure 4.29 show the CMC of P(MDS-co-CES) in deionized water and

the sodium acetate buffer with different pH and ionic strength, measured by the changes

of I 1 /I 3 and I 338 /I 333 as a function of polymer concentration The CMC measured from the emission spectra was 6.1 mg/L in deionized water and 7.4 mg/L in the sodium acetate buffer (0.1M, pH 4.6) (see Figure 4.28) However, from the excitation spectra, it was 1.9

mg/L and 2.4 mg/L respectively The CMC of P(MDS-co-CES) in the sodium acetate

buffer (0.02M, pH 4.6) was 5.2 mg/L and 1.5 mg/L, obtained from the emission and

excitation spectra respectively The CMC of P(MDS-co-CES) in the sodium acetate

buffer (0.02M, pH 5.6) was 6.0 mg/L and 1.9 mg/L, measured from the emission and excitation spectra respectively The real CMC of the polymer should be between the two

values obtained from the excitation spectra and emission spectra [Astafieva I et al.,

1993] Compared with other amphiphilic polymers reported previously [Jones M-C.,

1999; Chung J E., 2000], the CMC of P(MDS-co-CES) was much lower For instance,

the CMCs of PEO-PBLA and PNIPA-PS were between 5 mg/L and 20 mg/L Some amphiphilic copolymers like ploxamers exhibited even much higher CMC, reaching up to

100 mg/L to 100,000 mg/L [Jones M-C., 1999] Amphiphilic copolymers with higher CMC may not be suitable as a drug or gene carrier because the micelles formed may dissociate after injected into the blood (the dilution effect) In addition, if the core of the micelles is not stable enough (in the case of high CMC), the micelles may tend to dissociate after adsorption of proteins in the blood since the interactions between the shell and the proteins may damage the balance between the core and the shell The dissociation

of the micelles may also render a fast clearance Therefore, having a low CMC is a great

advantage Furthermore, the CMC of P(MDS-co-CES) in the sodium acetate buffer

[0.1M, pH 4.6] was 2.4 mg/L, obtained from the I338/I333 ratios, which was very close to

that in deionized water (1.9 mg/L) From Figure 4.29, the CMC of P(MDS-co-CES) obtained from the I 338 /I 333 ratios in 0.02M sodium acetate buffer with pH of 4.6 and 5.6

was similar (1.5 mg/L versus 1.9 mg/L) In addition, the CMC of P(MDS-co-CES) in the

sodium acetate buffer (0.1M) was similar to that in the buffer (0.02M) with a lower ionic strength (see Figure 4.28 and Figure 4.29) These findings suggest that pH and ionic strength did not affect the CMC significantly This property is important for DNA

binding of the polymer at low pH and for in vivo applications where the blood contains a

large amount of salts On the other hand, the successful determination of CMC evidences

Similarly, the CMC of P(MDA-co-CEA) in deionized water was determined to be 7.4 mg/L and 2.4 mg/L, obtained from the I 1 /I 3 ratios and I 338 /I 333 ratios respectively (figure

4.30) Compared to P(MDS-co-CES), P(MDA-co-CEA) had a slightly higher CMC This

is possibly due to the relatively lower molecular weight of P(MDA-co-CEA) compared with P(MDS-co-CES) After the grafting reaction, the molecules degraded more significantly (see Table 4.1) than that of P(MDS-co-CES) It was much more difficult to increase the molecular weight of P(MDA-co-CEA) In general, small molecular

amphiphiles have higher water solubility, resulting in greater CMC

Moreover, the CMC values of PEG5000-P(MDS-co-CES) and

PEG5000-P(MDA-co-CEA) were also studied The results are shown in Figure 4.31 and Figure 4.32 From

Figure 4.31, the CMC (I 1 /I 3 ) of PEG5000-P(MDS-co-CES) in deionized water, the

acetate buffer (0.02M) with pH of 5.6 and 4.6 was 42.7, 74.7 and 74.3 mg/L respectively

The CMC (I 338 /I 333) in deionized water, the acetate buffer (0.02M) with pH of 5.6 and 4.6

was 13.2, 18.8 and 13.2 mg/L respectively Similar with that of P(MDS-co-CES), pH did not pose great influence on the CMC of PEG5000-P(MDS-co-CES) However, compared with that of P(MDS-co-CES), the CMC of PEG5000-P(MDS-co-CES) increased almost

one magnitude This is because of the presence of highly water-soluble PEG It was reported that the presence of a large portion of hydrophilic segment might prevent the

amphiphilic copolymer from self-assembling into micelles [Kwon G S et al., 1996] Figure 4.32 shows the CMC (I 1 /I 3 ) of PEG5000-P(MDA-co-CEA) The CMC (I 1 /I 3) of

PEG5000-P(MDA-co-CEA) in deionized water, the acetate buffer with pH of 5.6 and 4.6 was 162.5, 89.7 and 150.9 mg/L respectively Compared to PEG5000-P(MDS-co-CES), PEG5000-P(MDS-co-CES) had higher CMC In addition, the I 338 /I 333 curves did not

show smooth and inerratic change, indicating that the micelles obtained were not stable This is possibly because of the greater hydrophilictity of the shell-forming segment Figures 4.33 and 4.34 provide direct observation of the spectra change of pyrene with

the change of P(MDS-co-CES) and P(MDA-co-CEA) concentration in deionized water From these figures, the I 3 peak of the emission spectra can be observed to increase greatly with the increase of concentration while in the excitation spectra the slope of the curves from 333nm to 339nm also increased obviously with increase of polymer concentration Successful measurements of the CMC prove that the polymers synthesized in this study can form core-shell structured micelles by self-assembly in aqueous solution, which provides the possibility of encapsulating drug molecules Other parameters including particle size, stability and zeta potential of the micelles need to be studied to evaluate the

possibility of using the micelles for in vivo codelivery of drug and gene

0.4 0.5 0.6 0.7 0.8 0.9 1.0 1.1 1.2 1.3 1.4 1.5 1.6 1.7 1.8 1.9 2.0

Figure 4.28 CMC determination of P(MDS-co-CES) by a fluorescence spectrometer

using pyrene as the probe (left: deionized water; right: 0.1M sodium acetate buffer with

pH 4.6) (Polymer Batch No 120902b) [In deionized water, CMC (I 1 /I 3) = 6.1 mg/L and

CMC (I 338 /I 333 ) = 1.9 mg/L; at pH 4.6, CMC (I 1 /I 3 ) = 7.4 mg/L and CMC (I 338 /I 333) = 2.4 mg/L]

Figure 4.29 CMC determination of P(MDS-co-CES) by a fluorescence spectrometer

using pyrene as the probe (left: 0.02M sodium acetate buffer with pH 5.6; right: 0.02M sodium acetate buffer with pH 4.6) (Polymer Batch No 120902b) [In pH 5.6 and 0.02M

sodium acetate buffer, CMC(I 1 /I 3 )=6.0mg/l , CMC (I 338 /I 333)=1.9mg/L; In pH 4.6 and

0.02M sodium acetate buffer, CMC(I 1 /I 3 )=5.2mg/l and CMC(I 338 /I 333)=1.5mg/L]

0.4 0.6 0.8 1.0 1.2 1.4 1.6 1.8 2.0

Figure 4.30 CMC determination of P(MDA-co-CEA) in deionized water by a

fluorescence spectrometer using pyrene as the probe (Polymer Batch No 111002a) [In

deionized water, CMC (I 1 /I 3 )=8.9mg/l, CMC(I 338/I333)=4.3mg/l]

0.8 0.9 1.0 1.1 1.2 1.3

CMC=74.7ppm

LgC (mg/L)

0.0 0.5 1.0 1.5 2.0 2.5 3.0 0.6

0.7 0.8 0.9 1.0 1.1 1.2

0.6 0.8 1.0 1.2 1.4 1.6 1.8

0.6 0.8 1.0 1.2 1.4 1.6 1.8

CMC(I 1 /I 3 )=74.7mg/L, CMC(I 338 /I 333 )=18.8mg/L; In pH 4.6, CMC(I 1 /I 3)=74.3mg/L,

0.5 0.6 0.7 0.8 0.9 1.0 1.1 1.2 1.3

0.7 0.8 0.9 1.0 1.1 1.2 1.3

Figure 4.32 CMC determination of PEG5000-P(MDA-co-CEA) in 0.02M sodium acetate

buffer by a fluorescence spectrometer using pyrene as the probe (left: deionized water; middle: pH 5.6; right: pH 4.6) (Polymer Batch No 310703a) [In deionized water,

CMC(I 1 /I3)=162.5mg/L; In pH 5.6, CMC(I1/I3)=89.7mg/L; In pH 4.6,

CMC(I1/I3)=150.9mg/L]

0 100 200 300 400 500 600 700

Figure 4.33 Fluorescence spectra of pyrene in aqueous solutions of P(MDS-co-CES) at

different concentrations (left, λem=339 nm for the emission spectra; right, λex=390 nm

for the excitation spectra)

0 100 200 300 400 500 600

Figure 4.34 Fluorescence spectra of pyrene in aqueous solutions of P(MDA-co-CEA) at

different concentrations (left, λem=339 nm for the emission spectra; right, λex=390 nm

for the excitation spectra)

4.2.2 Size and stability of the polymeric micelles

Size and zeta potential of the micelles are the important properties, which determine

whether the micelles can be used for drug and gene delivery As introduced in Chapter 2,

nanoscale size of polymer micelles makes them an ideal carrier for drug or gene delivery

[Jones M-C, 1999; Merdan T., 2002] Nanoparticles can avoid uptake by the MPS and

RES systems and realize prolonged circulation in the blood In addition, nanoparticles

exhibit enhanced permeability and retention at tumor sites and achieve passive targeting

to leaky tissues such as tumor or inflamed tissues Due to the complication of the in vivo

environment, a variety of factors that influence the size and stability of the polymeric micelles, including ionic strength, pH, temperature, serum and the protocol used to prepare the polymeric micelles, need to be investigated

4.2.2.1 Influence of pH, ionic strength and concentration on the size of the polymeric micelles

Figure 4.35 shows the effects of pH and ionic strength on the size of P(MDS-co-CES)

micelles When the concentration of the salts was 0.01M or 0.02M, the sizes of the micelles decreased with decreasing pH value For instance, at the concentration of 0.02M, the particle size decreased from 187 nm to 86 nm when the pH value decreased from 6.4 (deionized water, measured by pH meter) to 4.6 However, when the concentration of the salts was 0.2 M, the particle size increased at pH 5.6 but dropped when pH continued to decrease to 4.6 At pH 4.6, the concentration of the salts did not affect the particle size significantly However, at pH of 5.6, an increased salt concentration yielded an increase

in particle size Both the concentration of salts and pH played key roles in the formation

of micelles In general, at lower pH, the tertiary amines of P(MDS-co-CES) could be

more easily protonized, yielding more positive charges on the surface of the polymeric micelles The positive charges provided repulsive force between polymeric micelles and consequently prevented the aggregation of the micelles In addition, the repulsive force between the polymer chains allowed the polymer molecules to assemble in a lower

pH and at the salt concentrations of 0.01M and 0.02M In addition, according to DLCO theory, the positive charges on the surface of micelles can increase the repulsive interaction between the micelles At the same time, on the surface of the micelles, a double layer can form because of the electrostatic interaction The thickness of the double layer is depended on the ionic strength The anion in the buffer could interact with the positive charges on the surface of the micelles Therefore, the higher was the concentration of the salts, the larger the particle size The shielding of the positive charges by anions may also lead to aggregation of micelle particles When pH of the buffer was not low enough (e.g 5.6), the concentration of salts became a dominant factor

to determine the particle size and an increased concentration of the salts led to bigger micelles However, at low pH (e.g 4.6), the surface of the micelles was easily protonized and the produced positive charges were enough to override the effect of salts (see Figure 4.35) In summary, pH and ionic strength were two factors affecting particle size in a reverse way There was an optimized pH and salt concentration to obtain small size of the micelles To study the influence of pH value, and buffer concentration on the dynamic

changes of size distribution, the overlapped curves of the size distribution of

P(MDS-co-CES) micelles measured for 5 times were also shown in figure 4.36 Figure 4.36 provides the direct observation of the size change of the micelles under the effect of pH value and buffer concentration In DI water, there was a single and narrow peak, indicating a good size distribution and stability Under the other conditions such as the presence of salts with lower pH, multiple peaks appeared For instance, in the 0.02M sodium acetate buffer with pH of 5.6, peak at about 50 nm and a big peak at about 200 nm were observed However, when pH decreased to 4.6, the intensity of the peak at around 50 nm increased

and the big peak shifted to a smaller size, resulting in a smaller average diameter A similar phenomenon was observed in the case of 0.01M sodium acetate buffer However,

in 0.2M sodium acetate buffer with pH of 5.6, a big peak appeared between 500 nm and

1000 nm In 0.2M sodium acetate buffer with pH of 4.6, much smaller micelles were produced and a single peak appeared around 100 nm due to more complete protonization

of the surface of the micelles Obviously, re-assembly of the micelles occurred due to the effect of pH value and ions In addition, the size distribution kept changing in 5 times measurement, indicating that the micelles with different size is reversible The existence

of multiple peaks is possibly due to the broad polydispersity of molecular weight The polymer chain with higher molecular weight and lower grafting degree may need lower

pH value to be protonized, which cause the existence of multiple peaks and the shift of

the peaks to smaller size when pH value decreased For P(MDA-co-CEA), possibly

because of the denser tertiary amine and quaternary ammonium groups on the main chain, the influence of positive charge or pH value was even greater compared with

P(MDS-co-CES) Like figure 4.36, figure 4.37 shows the effect of ions and pH value on size distribution of the nanoparticles prepared by P(MDA-co-CEA) Most of the particles

had a diameter of about 200 nm in DI water However, in 0.01M sodium acetate buffer of

pH 4.6 or 5.6, the particles were of a size around 10 nm In 0.2M sodium acetate buffer, a big peak appeared at around 40 to 50 nm A smaller peak was also observed around 10

nm It seemed that micelles were not well formed with this polymer Therefore, in the

following study, we will focus on P(MDS-co-CES)

Besides ionic strength and pH, polymer concentration also had an effect on the size of

concentration in PBS buffer (pH 7.4) When the polymer concentration increased from 1

to 3 mg/mL, the size of the resulting micelles decreased from 170 to 129 nm High polymer concentration accelerated micelle formation This phenomenon is similar to that

of crystal formation As is known, denser concentration leads to the formation of smaller crystal Since the formation of micelles and crystal share a similar thermo dynamical procedure That is the decrease of enthalpy [Gao Z and Eisenberg A., 1993] This may be the cause for the smaller micelle size in higher polymer concentration

In summary, pH, ionic strength and polymer concentration can influence the size of the micelles In the following sections, their effects on DNA binding ability will also be investigated

40 60 80 100 120 140 160 180 200 220

DI water, 0.02M pH5.6 0.02M pH4.6

0.2M, pH=5.6 0.2M pH=4.6

0.01M, pH=5.6 0.01M, pH=4.6

Figure 4.36 SDP peak intensity change of P(MDS-co-CES) polymeric micelles (Batch

No 120902b) at different buffer concentration and pH, measured using a COULTER N4 Plus Submicron Particle Sizer

DI water 0.02M pH5.6 0.02M pH4.6

0.2M, pH=5.6 Size undetectable 0.2M, pH=4.6 Size undetectable

Figure 4.37 SDP peak intensity change of P(MDA-co-CEA) (Batch No 111002b) at

different buffer concentration and pH, measured using a COULTER N4 Plus Submicron Particle Sizer

80 90 100 110 120 130 140 150 160 170 180

in the figure were the actual concentration of the micelles in PBS) [Polymer:

P(MDS-co-CES), Batch No 120902b]

4.2.2.2 Stability of the polymeric micelles in deionized water, PBS buffer, serum and bovine serum albumin (BSA)

Stability of the micelles in the physiological environment is our primary concern since

it determines the feasibility of using these micelles for in vivo drug and gene delivery

Figure 4.39 shows the size of the polymeric micelles in deionized water as a function of

storage time The micelles made from both P(MDS-co-CES) and P(MDA-co-CEA) were

stable in deionized water during the testing period of time (44 days) and the size did not vary significantly This means the micelles neither aggregated nor dissociated in deionized water in a long period of time because of their positively charged surface

Figure 4.40 shows the size of the micelles in pH 7.4 PBS as a function of storage time The size of the micelles increased from 120 nm to 250 nm in 28 days As discussed previously, the ions of the PBS buffer can be absorbed onto the positively charged surface of the micelles through electrostatic interactions and form a double layer which can provide serious shielding to the positive charges This was evidenced by zeta

potential measurements The P(MDS-co-CES) micelles prepared in PBS buffer had much

lower zeta potential value than those prepared in deionized water For example, at 25ºC, the zeta potential of the micelles in PBS was 38.0±3.2mV while in deionized water the zeta potential was 69.4±3.9mV The different degree of protonization of the polymer in

DI water (pH=6.4) and PBS (pH=7.4) may also contribute to the different zeta potential

of the micelles in DI water and PBS buffer respectively The shielding of the positive charges of the micelle surface led to easy aggregation of the micelles Although the micelles aggregated slowly in PBS, it should be noted that they were quite stable in at least several days, which is enough for the micelles to find the target tissues in the case of drug and gene delivery

It needs to mention that the stability of the polymeric micelles in deionized water and PBS was measured at room temperature (25ºC) Giving that the human body temperature

is 37ºC and in some inflammatory site the temperature is a little higher, the stability of the micelles in PBS needs to be studied around the body temperature Figure 4.41 shows the micelle size in PBS as a function of temperature from 20ºC to 40ºC It can be seen that the micelle size did not change significantly with the temperature increased from 20ºC to 40ºC, suggesting that the micelles remained stable in the temperature range This

property is especially important for in vivo applications For example, the temperature

change from room temperature to 37ºC in the process of in vivo experiments would not

change the particle size

Although in many cases PBS is used to simulate the physiological condition in human body, the ingredients in serum especially the proteins may affect the stability of the micelles For example, in the case of gene delivery, the proteins in the blood may cause aggregation and/or dissociation of the cations/DNA complexes [Merdan T., 2002]

Therefore, when performing in vitro gene transfection in many experiments, the

complexes need to be incubated in serum-free medium [Suh W et al, 2001; Mannisto M

et al, 2002; Mao H-Q et al, 2001] to avoid the disruption and aggregation of the complexes and achieve high gene transfection level It is necessary to study the stability

of the micelles in serum- or protein-containing PBS Figure 4.42 shows the micelle size change in PBS containing 10% serum as a function of time The original size of the micelles in PBS was 255nm After the particles were incubated with PBS containing 10% FBS, the average micelle size immediately increased to about 423 nm at 37ºC In 4 hours, the size slowly went up to around 566 nm The temperature did not affect the stability of the micelles in serum-containing PBS buffer The proteins absorbed onto the surface of the micelles and neutralized the positive charges This might lead to the aggregation of the micelles However, the micelles did not dissociate in the presence of serum This is quite different from many other polymer/DNA complexes, in which the absorption of the proteins in serum easily caused the dissociation of the complexes [Merdan T., 2002] The micelles synthesized in this study may be more stable than other polycationic gene carriers