Microcalorimetric Studies on the Physical Stability of PEG-Liposome

(1)

Microcalorimetric studies on the physical stability of

poly-ethylene glycol-grafted liposome

Der-Zen Liu

a,

, Yi-Ling Hsieh

a

, Sheng-Yen Chang

a

, Wen-Yih Chen

b

a

Graduate Institute of Biomedical Materials, Taipei Medical University, Taipei, Taiwan, ROC

b

Department of Chemical Engineering National Central University, Chung-Li, Taiwan, ROC Received 1 November 2001; accepted 24 June 2002

Abstract

Liposomes were prepared from mixtures of egg-PC, cholesterols and distearoyl-phosphatidyl-ethanolamine

covalently attached poly-ethylene glycol (PEG) with a molecular of weight 2000 (DSPE /PEG2000). In this work

examines how PEG2000-grafted lipids affect the surface properties of the egg phosphatidylcholine (egg-PC) liposomal

bilayer membrane through zeta potential and interaction potential measurements using microcalorimetry. Experimental

results demonstrate that the absolute value of the zeta potential of PEG2000-grafted PC liposomes decreased from /19

to /8 mV when increasing DSPE /PEG2000 from 0 to 7 mol fraction, and the repulsive interaction potential of

PEG2000-grafted PC liposomes decreased compared with those liposomes without PEG-grafting. However, the

phenomenon of fusion between the liposomes incorporated with PEG-grafted PC was reduced. In brief, this result of fusion is contrary to the expectation of the interaction potential measurement; therefore, we believe that the steric

hindrance of the grafted PEG2000molecules on the liposomal surface contribute to a major imposition on the approach

between liposomal surface and the formation of inverted micelles which are suggested as the necessary steps of liposome fusion.

Keywords: Poly-ethylene glycol-liposome; Physical stability; Zeta potential; Interaction potential

1. Introduction

Liposomes (lipid vesicles) have a unique closed structure and special physical and chemical prop-erties. As early as 1973, researchers recognized that liposomes could be an effective drug delivery vesicle [1]. The way the structure of liposomes

minics that of cell membranes enables liposomes to serve a model for physical and biochemical studies of biomembranes [2 /5]. However, in vitro and in

vivo instability of liposomes has limited not only their widespread application but also some of their potential advantages. Recently, extensive research has sought to improve the stability of liposomes resulting in novel liposome systems with better in vitro and in vivo physical stabilities.

Liposome stability plays important roles at various stages of medical practice. However,

Corresponding author. Tel.: /886-2-2739-0581

E-mail address: [email protected](D.-Z. Liu).

(2)

liposomes are somewhat biologically unstable as a parenteral drug delivery system owing to their rapid uptake and clearance from circulation by cells of the mononuclear phagocytic system (MPS) located mainly in the liver and spleen [6,7]. A major breakthrough occurred in 1990, as liposo-mal bilayer membranes containing lipids with covalently attached poly-ethylene glycol (PEG-lipid) by which the membrane surface steric inhibits protein and cellular interactions with liposomes drastically prolonging the blood circu-lation time when injected in animal models [8]. These vesicles are the so-called Stealth liposome[9]

or sterically stabilized liposomes (SSL) [10]. The steric repulsion of the stealth liposome appears to not only stabilize liposome suspensions against aggregation but also inhibit the adsorption of various opsonins onto the liposomal surface and degrade biological interactions.

Considering the interactive forces between lipo-somes, incorporating lipids with bulky polar head-groups increases the repulsive forces between lipid bilayers, as measured by X-ray, neutron diffrac-tion, micro-pipette manipulation [11], and small-angle X-ray scattering (SAXS) [12]. Needham et al. [13] demonstrated that including 10 mol% of monosialogangliosides (GM1) in PC bilayers in-creased the interbilayer separation more then two fold as compared to PC bilayers alone, while PC bilayers containing 5 mol% of DSPE /PEG₁₉₀₀

increased the separation even more. Furthermore, Kuhl et al. [14] focused on short-chain ethylene oxide headgroups grafted onto liposomal surfaces; according to their results, incorporating a slight amount of dielaidoylphosphatidylethanolamine -polyethylene oxide (DEPE /PEO₄₅) lipid greatly

enhances liposome stability. Kuhl et al. contended that short polymer chains can provide a physical barrier or force buffer around the bilayer to prevent contact between liposomes. In addition, both Kenworthy et al. [15]and Lasic [16] proved that the attachment of PEG molecules to liposo-mal surfaces can strongly degrade the attractive forces and increase the repulsive steric forces imposed on each other between liposomes. How-ever, no direct evidence or measurements have been presented on the effect of PEG-lipid on the interactive forces between liposomes. The effects

of PEG-lipids on the interactive forces include the electrostatic force, van der Waal force, and even the short-range hydration force.

Therefore, as to the investigation aims of this study, we examined the interaction potential by measuring the dilution heat of PC liposomes incorporated with various amounts of DSPE /

PEG2000 by isothermal titration calorimetry (ITC). The parameter of interaction potential for PC liposomes was then calculated from the empirical dilution heat data combined with a simple statistical thermodynamic model [17]. To sum up, aggregation and fusion behaviors are explained by the interaction potential between PC liposomes incorporated with varying amounts of DSPE /PEG₂₀₀₀.

2. Materials and methods 2.1. Materials

Distearoyl-phosphatidylethanolamine (DSPE)

covalently attached poly(ethylene-glycol)

(DSPE /PEG2000) was supplied by Shearwater Polymers, while egg-phosphatidylcholine (PC) ( /99%), cholesterol, and a-tocopherol were

pur-chased from Sigma (USA). Chloroform, methanol, Tris, and sodium chloride were obtained from Merk (Germany). All chemicals were used as received without further purification.

2.2. Preparation of stealth liposome suspension solution

Large multilamellar PEG2000-grafted liposomes were prepared with the method reported by Bangham et al. [18]. Briefly, the lipid mixture, 0.59 mg ml1 of a-tocopherol, egg PC (3.75 mg ml1), cholesterol (9.6 mg ml1), and various amounts of DSPE /PEG2000 (from 0 to 7 mol%)

were dissolved in a chloroform/methanol mixture (2:1 volume ratio). The solvent was evaporated in a round-bottomed flask at 50 8C, and nitrogen was used to blow over the dried lipid film for approximately 30 min in order to remove traces of organic solvents. The lipid bilayer film was then hydrated in 10 ml of buffer solution composed of

(3)

20 mM Tris, 100 mM NaCl and 1 mM EDTA (with the pH being adjusted to 7.4 with HCl) at 323 K.

Frozen then thawed (FT) the MLVs liposomes in order to reduce lamellarity[19]. These liposomes were subjected to freezing (in boiling nitrogen) and thawing (in a water bath at 323 K) five times, and were then sonicated for 30 min using Bandelin HD-200 sonication probe at a power setting of KE 76/D under nitrogen to prepare small unilamellar vesicles (SUVs). The temperature was maintained at 323 K by circulating the water in round-bottomed flask. Then the final optically clear suspension was filtered through a 0.22-mm Milli-pore filter for three cycles.

In our experimental system, liposomes were added to the same amounts of a-tocopherol and cholesterol. a-Tocopherol is believed to act as an antioxidant in lipid membranes to avoid oxidative damage caused by interactions between free radi-cals and PC [20]. Therefore, pressure on the chemical stability was diminished.

2.3. Calculation of the number of liposomes The number of PC lipids within a liposome can be estimated by calculating the ratio between the head-group surface areas of egg-PC (0.70 nm2)

[21], distearoyl-phosphatidylethanolamin (0.43 nm2) [15], and cholesterol (0.28 nm2) [21]. This calculation allows us to estimate the number of liposomes originating from 3.75 mg ml1 of egg PC[22].

2.4. Determination of particle size distribution of liposomes by PCS

Liposome size distribution was determined by photon correlation spectroscopy (PCS) that em-ployed a zetasizer 3000 (Malvern Instruments, UK). The samples (liposome suspension solution) were diluted in a buffer solution until their concentrations were low enough to avoid hydro-dynamic and electrostatic interactions between vesicles but high enough that reliable autocorrela-tion funcautocorrela-tions from PCS measurements could be achieved. All PCS measurements were taken at 298 K and the viscosity of the sample solution was 1

cp. The liposome suspension solution was placed in a sample holder, and laser light (helium-neon laser of wavelength 633 nm and 5 mW) was focused on the sample. Scattered light was de-tected at 908 to the incident beam using a photomutiplier tube. Five replicate measurements were made of each liposome suspension, and the mean average particle diameter and polydispersity index were obtained using Malvern software supported comulant analysis.

2.5. g-potential measurements of liposome

The electrophoretic mobility and zeta potential of liposome suspension solutions were determined by laser Doppler electrophoretic mobility mea-surements with the zetasizer 3000 (Malvern In-struments).

2.6. Dilution heat by ITC and interaction potential of PEG-grafted liposomes

The deviation of a solution from ideal behavior due to two-body interactions in a suitable con-centration of solution can described by vant Hoff’s law [23]. Therefore, it is desirable to obtain the value of a second virial coefficient and b2/B0as the interaction potential (o ) index between liposomes. This study extend own development of measuring second virial coefficient of the colloidal system by isothermal titration calorimetry to the liposome system[17,24].

PEG-grafted liposome suspension solution dilu-tion heat was determined by applying isothermal titration calorimetry (ITC). Briefly, 2 ml of lipo-some suspension solution was placed in an am-poule. A series of 150 ml buffer solutions was then titrated into the dispersion liposomes suspension solution using a Hamilton microliter syringe at 30-min intervals after the ampoule and the heat sink reached thermo-equilibrium. The liposome suspen-sion solution was titrated five times per

experi-ment. All experiments were performed at

temperatures of 298 and 310 K, respectively. Dilution heats of PEG-grafted liposome suspen-sion solutions with the number density of the liposome suspension solution can be generally expressed as a function of a second order

(4)

poly-nomial inEq. (1): d q NkBT dr $ d E NkBT dr b2b3rb4r 2 (1) For an open liquid system, pressure changes and titration volume are neglected in the system total volume. The observed dilution heat of each titra-tion is comparable to the internal energy change. Furthermore, the internal energy change can be expressed by the virial coefficient and the number density inEq. (2): E NkBT 3 2T X i1 1 i dB_i1 dT r i (2) ComparingEq. (1) withEq. (2) reveals that the coefficient of the dilution heat polynomial can be affiliated with the virial coefficients as inEq. (3): b₂TdB2

dT (3)

According to statistical mechanics, the relation between the second virial coefficient and the interaction potential energy function U (r ) is

B₂(T ) 2p

g

0 [exp U(r) k_BT 1]r2_dr ₍₄₎

and if a square-well energy potential function is picked and plugged into Eq. (4), thenEq. (3)can be rewritten as:

b2T dB₂

dTB0(l 3

1)eo=kBT_(o=k

BT ) (5)

b2=B0 (l 3

1)eo=kBT_(o=k

BT) (6)

where B0/(16/3)pRHS3 , l /(2RHS/s)/2RHS and s, o denotes the width and depth (or energy barrier) of the square-well energy potential func-tion.

In our experimental system, the value of b2/B0 can be described qualitatively from the interaction potential (o ) between PEG2000-grafted PC lipo-somes.

3. Results and discussion

3.1. Zeta potential measurements

This study measured the zeta potential (z-potential) of PEG-grafted PC liposomes incorpo-rated with varying amounts of DSPE /PEG₂₀₀₀.

The data (shown in Table 1) correspond to the results of Carrion et al.[25]and McLaughlin et al.

[26] indicating that there is a weak electrostatic repulsive force between PC liposomes at pH 7.4. This experiment also demonstrated that the in-corporation of DSPE /PEG₂₀₀₀into the PC bilayer

decreases the negative zeta potential. The absolute value of the zeta potential of PEG2000-grafted liposomes decreased from /19 to /8 mV as the

concentration of DSPE /PEG increased from 0 to

7 mol%. This observation can be explained by the exposed PEG chains shielding the negatively charged liposomal surface, thus decreasing the absolute zeta potential of the liposome[27]. More-over, Woodle et al.[28]believed that the phenom-enon was attributable to an increase in the hydrodynamic radius from the steric effects of PEG. Furthermore, Kostarelos et al.[29]reported that adsorption or incorporation of the tri-block copolymer (PEO /PPO /PEO) shifted the shear

plane away from the liposomal surface, and, subsequently, reduced the absolute values of the zeta potential of the liposome.

3.2. Interactions between phosphatidylcholine liposomes

Although the interactive forces between lipo-somes have been thoroughly discussed [30,31],

Table 1

Values of zeta potential of PEG-grafted PC liposomes with various amount of incorporated DSPE /PEG2000(values are the

mean9/standard deviation, n /3)

Mole fraction of DSPE /PEG (mol%)

[DSPE /PEG/PC/VitE/Chol]

Zeta potential (mV) (298 K) 0 /18.89/1.6 3 /13.29/1.3 5 /10.39/1.2 7 /8.79/0.7

(5)

they have not yet been able to directly measure them. For that, this study proposed a thermo-dynamic approach to examine how PEG2000 -grafted lipids affect the interaction potential, not the interaction forces, of intervesicle interactions between PEG2000-grafted liposomes. The interac-tion potential can be used to explain the stability of PEG-grafted PC liposomes thermodynamically. The isothermal titration calorimetry was utilized to measure the dilution heat of PC liposomes without and with various levels of added DSPE /

PEG2000and to derive values of b2and b2/B0from the dilution heat measurements (as discussed in

Section 2). Differences in depth (o ) and location of the interaction potential energy minimum between any two liposomes caused by adding different amounts of DSPE /PEG2000 are discussed herein.

The ITC experiments of this study show that the system of PC liposomes with or without added DSPE /PEG2000 has a positive b2 value as indi-cated in Table 2. This finding reveals that the overall net interactive forces and potential between any the two PC liposomes is repulsive, thus preventing aggregation or fusion and enhancing the liposomal stability. Previously, Needham et al.

[32]confirmed that incorporating DSPE /PEG₁₉₀₀

increases both the magnitude and the range of the repulsive pressure between adjacent SOPC/PC bilayers. They attributed the repulsive pressure to the steric repulsive pressure since it did not result from increasing the electrostatic, undulation, or hydration repulsive forces. However, in our ex-perimental system, the value of b2/B0 (Table 2) decreased when adding DSPE /PEG₂₀₀₀ to PC

liposomes; this observation shows that the

repul-sive interaction potential (o ) is negative (as an energy barrier), and the repulsive interaction potential of PEG2000-grafted liposome is decreased (lower values ofjo j and s ). These results indicate that the net interactive repulsive potential or energy barrier decreases when DSPE /PEG₂₀₀₀ is

incorporated into PC liposomes. Possible explana-tions are that the electrostatic repulsive forces between PEG2000-grafted PC liposomes decrease and hydrophilic interactions between PEG-head-groups increase

3.3. Changes of liposomal particle size

Experimental results show that the initial lipo-somal size decreased when the amount of DSPE /

PEG2000 of the liposome composition was in-creased (Fig. 1). This result is considered due to a steric repulsion among PEG chains exposed from the outer and inner leaflet of the liposomal bilayer membrane. PEG chains exposed from the outer leaflet of the bilayer menbrane will increase the liposome particle curvature, whereas the PEG chains exposed to the inner leaflet do the opposite. In fact, more DSPE /PEG₂₀₀₀ exists in the outer

leaflet than in the inner leaflet of liposomal bilayer membranes [27]. Therefore, adding DSPE /

PEG2000reduces the initial liposomal size.

Fig. 1 displays size changes of PC liposomes with or without added DSPE /PEG2000at 298 K as a function of incubation time. The experimental results imply that PC liposomes can hold the electrostatic repulsive forces to temporarily pause in the secondary minimum state [33] (or in reversible aggregation) from the interaction

poten-Table 2

Values of second-order polynomial b2 and the B0 of virial coefficients of PEG-grafted PC liposomes with various amount of

incorporated DSPE /PEG2000

Reaction temperature Mole fraction of DSPE /PEG (mol%)

[DSPE /PEG/PC/VitE/Chol]

b2(nm 3 ) B0virial coefficient (nm 3 ) b2/Bo 298 K 0 7.5 /1012 1.7 /106 4.4 /106 3 1.1 /1012 9.9 /105 1.1 /106 5 5.2 /1011 7.2 /105 7.2 /105 7 3.8 /1011 6.6 /105 5.8 /105

(6)

tial perspective. Finally, PC liposome will reach the primarily minimum state (or irreversible ag-gregation) due to the effective collisions among liposomes, and the effective collisions may be promoted by the oxidation of lipids.

There are two continuous steps for liposomes to be fused, as suggested: (i) liposomes have to be aggregated, and (ii) intermediate structures such as inverted micelles or inverse hexagonal structures have to occur after close juxtaposition and local perturbation in the liposomal bilayer packing of the aggregated liposomes[34]. In our experiment, results show that incorporating DSPE /PEG2000 into the PC bilayer decreases the repulsive inter-action potential between PEG-grafted liposomes according to the data of zeta potential and the b2/ B0 value obtained. But, fusion between PEG-grafted PC liposomes was inhibited compared to that for PC liposomes without PEG-grafting at the temperature of 298 K (Fig. 1). Simply speaking, incorporating DSPE /PEG₂₀₀₀. into the PC bilayer

decreases the liposomes interaction energy barrier

compared with those liposomes without added DSPE /PEG2000 from the perspective of interac-tion potential. An explanainterac-tion for this apparent inconsistency is that the steric hindrance of the grafted PEG2000 molecular on the liposomal sur-face imposes resistance to the approach between the liposomal surface and the formation of in-verted micelle which was suggested as a necessary step for liposome fusion.

4. Conclusions

This research used the interaction potential to show the physical stability of DSPE /PEG2000 liposome. In this study, we demonstrate that the incorporation of a small amount of DSPE /

PEG2000 into the PC liposomal bilayer decreases the absolute value of the zeta potential and the repulsive interaction potential. These results in-dicate that the decrease in the repulsive potential can be attributed to a decrement in the electrostatic

Fig. 1. Sizes of PC liposomes with various amount of incorporated DSPE /PEG2000at 298 K with time. The number density of the

(7)

potential and an increment in hydrophilic attractive interactions occurring between PEG2000-grafted PC liposomes. However, the fusion taking place between PEG2000-grafted PC liposomes is reduced as a small amount of DSPE /PEG₂₀₀₀ is

incorpo-rated into the PC liposomal bilayer. Therefore, it is our strong belief that the phenomenon does not result from repulsive forces between PEG2000 -grafted PC liposomes. Instead, the major contribu-tion is due to PEG2000grafted onto the liposomal surface that acts as a steric barrier inhibiting the close approach of liposomal bilayers between liposomes. Furthermore, our work has proven the feasibility of isothermal titration calorimetry for studies of liposomal interactions and provides more information for discussions of liposomal interaction mechanisms.

Acknowledgements

The authors would like to thank the Bureau of Animal and Plant Health Inspection and Quar-antine Council of Agriculture no. 91AS-3.1.3-BQ-B1.

Appendix A: Nomenclature

q dilution heat of DSPE /PEG₂₀₀₀egg-PC

liposome suspension solution (mJ) E internal energy (mJ)

N number of liposomes

kB Boltzmann constant (J K1) T absolute temperature (K)

r number density of liposome suspension solution (number ml1)

b2 second virial coefficient of the fitting poly-nomial of the dilution heat of a DSPE /

PEG2000 egg-PC liposome suspension solu-tion

B0 virial coefficient

RHS the radius of liposome (nm)

s the width of the square-well potential func-tion

o the depth of the square-well potential func-tion (or energy barrier)

References

[1] G. Gregoriadis, FEBS. Lett. 36 (1973) 292.

[2] M. Costuleanu, E. Brailoiu, C.M. Filipeanu, O. Baltatu, S. Slatineanu, L. Saila, L.M. Nechifor, D.D. Branisteanu, Eur. J. Pharmacol. 281 (1995) 89.

[3] R.M. Fielding, Clin. Pharmacokinet. 21 (1991) 155. [4] D.D. Lasic, H. Strey, M.C. Stuart, R. Podgornic, P.M.

Frederic, J. Am. Chem. Soc. 119 (1997) 832.

[5] C. Karau, M. Petszulat, P.C. Schmidt, Int. J. Pharm. 128 (1996) 89.

[6] P.L. Beaumier, K.J. Hwang, Biochim. Biophys. Acta 731 (1983) 23.

[7] C.R. Alving, N.M. Wassef, J. Liposome Res. 2 (1992) 383.

[8] T.M. Allen, C. Hansen, F. Martin, C. Redemann, A. Yau-Young, Biochim. Biophys. Acta 1066 (1991) 29. [9] M.C. Woodle, D.D. Lasic, Biochim. Biophys. Acta 1113

(1992) 171.

[10] M.C. Woodle, Adv. Drug Deliv. Rev. 32 (1998) 139. [11] D.D. Lasic (Ed.) Liposome: from Physics to Application,

Elsevier, Amsterdam, 1993.

[12] G. Blume, J. Liposome Res. 2 (1992) 355.

[13] D. Needham, K. Hristova, T.J. McIntosh, M. Wu, N. Dewhirst, D.D. Lasic, J. Liposome Res. 2 (1992) 411. [14] T.L. Kuhl, D.E. Leckband, D.D. Lasic, J.N. Israelachvili,

Biophys. J. 66 (1994) 1479.

[15] A.K. Kenworthy, K. Hristova, D. Needham, T.J. McIn-tosh, Biophys. J. 68 (1995) 1921.

[16] D.D. Lasic, F. Martin, Stealth Liposome, CRC Press, Boca Raton, FL, 1995.

[17] W.Y. Chen, C.S. Kuo, D.Z. Liu, Langmuir 16 (2000) 300.

[18] A.D. Bangham, M.M. Standish, J.C. Watkins, J. Mol. Biol. 13 (1965) 238.

[19] W. Zhang, E.C.A. Van Winden, J.A. Bouwstra, D.J.A. Crommelin, Cryobiology 35 (1997) 277.

[20] T. Koga, J. Terao, Biosci. Biotechnol. Biochem. 60 (1996) 043.

[21] F. Szoka, D. Papahadjopoulos, Annu. Rev. Biophys. Bioeng. 9 (1980) 467.

[22] R. Xu, Langmuir 14 (1998) 2593. [23] W. Bruns, Macromolecules 29 (1996) 2641.

[24] D.Z. Liu, W.Y. Chen, L.M. Tasi, S.P. Yang, Colloids Surf. A 172 (2000) 57.

[25] F.J. Carrion, A..D. Maza, J.L. Parra, J. Colloid Interface Sci. 164 (1994) 78.

[26] A. McLaughlin, W.K. Eng, G. Vaio, T. Wilson, S. McLaughlin, J. Membr. Biol. 76 (1983) 183.

[27] A. Yoshida, K. Hashizaki, H. Yamauchi, H. Sakai, S. Yokoyama, M. Abe, Langmuir 15 (1999) 2333.

[28] M.C. Woodle, L.R. Collins, E. Sponsler, N. Kossovsky, D. Papahadjopoulos, F.J. Martin, Biophys. J. 61 (1992) 902.

[29] K. Kostarelos, Th.F. Tadros, P.E. Luckham, Langmuir 15 (1999) 369 /376.

(8)

[30] J.N. Israelachvili, Intermolecular and Surface Forces, Academic Press, New York, 1992.

[31] T.L. Kuhl, D.E. Leckb, D.D. Lasic, J.N. Israelachvili, Biophys. J. 66 (1994) 1479.

[32] D. Needham, T.J. McIntosh, D.D. Lasic, Biochim. Bio-phys. Acta 1108 (1992) 40.

[33] S. Nir, J. Bentz, J. Wilschut, N. Duzgunes, Prog. Surf. Sci. 13 (1983) 1.

[34] P.M. Frederik, A.J. Verkleij, M.C.A. Stuart, Biochim. Biophys. Acta. 979 (1989) 275.

Microcalorimetric Studies on the Physical Stability of PEG-Liposome