Characterization of Pegylated Liposomal Mitomycin C Lipid-Based

Subscriber access provided by UNIV OF CALIFORNIA SAN DIEGO LIBRARIES

Article

Characterization of pegylated liposomal mitomycin C lipidbased prodrug (Promitil®) by high sensitivity differential scanning calorimetry and cryogenic transmission electron microscopy Xiaohui Wei, Yogita Patil, Patricia Ohana, Yasmine Amitay, Hilary Shmeeda, Alberto Gabizon, and Yechezkel Barenholz Mol. Pharmaceutics, Just Accepted Manuscript • DOI: 10.1021/acs.molpharmaceut.6b00865 • Publication Date (Web): 03 Jan 2017 Downloaded from http://pubs.acs.org on January 4, 2017

Just Accepted “Just Accepted” manuscripts have been peer-reviewed and accepted for publication. They are posted online prior to technical editing, formatting for publication and author proofing. The American Chemical Society provides “Just Accepted” as a free service to the research community to expedite the dissemination of scientific material as soon as possible after acceptance. “Just Accepted” manuscripts appear in full in PDF format accompanied by an HTML abstract. “Just Accepted” manuscripts have been fully peer reviewed, but should not be considered the official version of record. They are accessible to all readers and citable by the Digital Object Identifier (DOI®). “Just Accepted” is an optional service offered to authors. Therefore, the “Just Accepted” Web site may not include all articles that will be published in the journal. After a manuscript is technically edited and formatted, it will be removed from the “Just Accepted” Web site and published as an ASAP article. Note that technical editing may introduce minor changes to the manuscript text and/or graphics which could affect content, and all legal disclaimers and ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errors or consequences arising from the use of information contained in these “Just Accepted” manuscripts.

Molecular Pharmaceutics is published by the American Chemical Society. 1155 Sixteenth Street N.W., Washington, DC 20036 Published by American Chemical Society. Copyright © American Chemical Society. However, no copyright claim is made to original U.S. Government works, or works produced by employees of any Commonwealth realm Crown government in the course of their duties.

Page 1 of 28

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Molecular Pharmaceutics 1

Characterization of pegylated liposomal mitomycin C lipid-based prodrug (Promitil®) by high sensitivity differential scanning calorimetry and cryogenic transmission electron microscopy.

Xiaohui Weia,b, Yogita Patilc,d, Patricia Ohanae, Yasmine Amitaye, Hilary Shmeedac, Alberto Gabizonc,d,e,*, Yechezkel Barenholza,* a

Laboratory of Membrane and Liposome Research, The Hebrew University-Hadassah Medical School,

IMRIC, Jerusalem, Israel School of Pharmacy, Shanghai Jiao Tong University, Shanghai, China c Oncology Institute, Shaare Zedek Medical Center, Jerusalem, Israel d Hebrew University-School of Medicine, Jerusalem, Israel e Lipomedix Pharmaceuticals, Jerusalem, Israel

b

*Co-Corresponding authors: - Yechezkel Barenholz, PhD; Laboratory of Membrane and Liposome Research, The Hebrew University-Hadassah Medical School, IMRIC, POB 12272, Jerusalem 9112102, Israel. Fax: (972)-2-6757499; Number: (972)-2-6757615 Email: [email protected]; [email protected]

- Alberto Gabizon, MD, PhD; Oncology Institute, Shaare Zedek Medical Center, Jerusalem, Israel Fax: (972) -2-6555080; Number: (972) 2-6555361 Email: [email protected]; [email protected]

ACS Paragon Plus Environment

Molecular Pharmaceutics

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 2 of 28

2

Abstract The effect of a lipidated prodrug of mitomycin C (MLP) on the membrane of a pegylated liposome formulation (PL-MLP), also known as Promitil®, was characterized through high-sensitivity differential scanning calorimetry (DSC), and cryo-TEM. The thermodynamic analysis demonstrated that MLP led to the formation of heterogeneous domains in the membrane plane of PL-MLP. MLP concentrated in prodrug-rich domains, arranged in high-ordered crystal-like structures, as suggested by the sharp and high enthalpy endotherm in the 1st heating scanning. After thiolytic cleavage of mitomycin C from MLP by dithiotreitol (DTT) treatment, the crystal-like prodrug domain disappears and a homogeneous membrane with stronger lipid interactions and higher phase transition temperature compared with the blank (MLP-free) liposomes is observed by DSC. In parallel, the rod-like discoid liposomes and the “kissing liposomes” seen by cryo-TEM in the PL-MLP formulation disappear, and liposome mean size and polydispersity increase after DTT treatment. Both MLP and the residual post-cleavage lipophilic moiety of the prodrug increased the rigidity of the liposome membrane as indicated by DSC. These results confirm that MLP is inserted in the PL-MLP liposome membrane via its lipophilic anchor, and its mitomycin C moiety located mainly at the region of the phospholipid glycerol backbone and polar head-group. We hypothesize that π-π stacking between the planar aromatic rings of the mitomycin C moieties leads to the formation of prodrug-rich domains with highly ordered structure on the PL-MLP liposome membrane. This thermodynamically stable conformation may explain the high stability of the PL-MLP formulation. These results also provide us with an interesting example of the application of high sensitivity DSC in understanding the composition-structure-behavior dynamics of liposomal nanocarriers having a lipid-based drug as pharmaceutical ingredient.

Keywords: mitomycin C, lipophilic drug, prodrug, liposome, nanoparticle, DSC, thermodynamics, phase transition


Page 3 of 28

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


Graphics Abstract:



1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 4 of 28

4

1. Introduction Pegylated Liposomal–Mitomycin C Lipid-Based Prodrug formulation (PL-MLP), also known as Promitil®, is a liposome-based anticancer nanomedicine consisting of the same lipid components as the clinically approved formulation of DOXIL® (pegylated liposomal doxorubicin) except for a reduction of the cholesterol fraction and addition of a prodrug which is a lipidated mitomycin C (referred as MLP). Chemically it is described as 2,3-(distearoyloxy)propane-1-dithio-4'-benzyloxycarbonylmitomycin C. The lipid moiety, which is composed of two stearoyl chains, allows MLP stable insertion into the liposome lipid bilayer. PL-MLP was previously characterized for its pharmacokinetics, pharmacodynamics, and mechanism of action 1-4, and is currently undergoing clinical evaluation 5. PL-MLP has demonstrated improved antitumor activity and reduced toxicity compared to equivalent doses of free mitomycin C (MMC) in a mouse multi-drug resistant carcinoma model 1 , in a human MDR ovarian cancer tumor model 2 and in mouse models of human gastroentero-pancreatic tumors 3. Various physicochemical and morphological aspects of PL-MLP have been previously reported4,5. The PL-MLP formulation results in high entrapment efficiency of MLP, nearly 100% relative to lipid recovery, and high physico-chemical stability for at least 30 months at 5°C 4. Cryo-TEM analysis shows that most of the nanoparticles are spherical vesicles, predominantly unilamellar liposomes of ~100-nm diameter, with a small fraction of rod-like or discoidal structures, of slightly smaller diameter. Recently we demonstrated the power of high-sensitivity differential scanning calorimetry (DSC) as a tool for the physicochemical characterization of liposomal drugs. DSC enables analysis of the thermotropic behavior of liposomal nano-drugs6-9. DSC can precisely determine if the liposome


Page 5 of 28

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


membrane and/or the liposome associated drug undergo a phase transition and the features of such phase transitions9. This includes: Tm, the temperature in the phase transition range at which the analyzed sample undergo the highest heat capacity change; ∆H, the enthalpy of the phase transition, a measure of the strength of intra-vesicle interaction between the liposome components related to packing of the lipid and/or drug molecules; and ∆T1/2, the width of the transition at half peak height which defines the phase transition cooperativity 10. DSC has been applied previously for the characterization of liposomes composed of binary or tertiary lipid mixtures, including characterization of the effects of cholesterol and DSPE-PEG11-13. The thermodynamics properties of liposomes are primarily due to liposome lipid composition and are influenced by the liposome size and lamellarity 11, 14, the mole % of cholesterol present on the liposome membrane13, 15 and the guest molecules 16, 17. In addition, high-sensitivity DSC provides important information about the state of the drug in the liposomal formulation, for instance, in the case of Pegylated Liposomal Doxorubicin, doxorubicin-sulfate forms intra-liposomal rod nano-crystals having typical thermotropic behavior9 but does not affect the membrane lipid phase transition; while, in liposomes remote-loaded with the local anesthetic bupivacaine, DSC shows that the drug interacts with the liposome membrane8. These thermodynamic parameters are important for a better understanding of the structure as well as for the optimization of liposomal drugs. The thermotropic behavior is also highly relevant to the physical stability of the liposomal product in vitro and in vivo, and therefore to the pharmacokinetics, biodistribution, drug release and bioavailability18-20. Moreover, cycled DSC scan (heating-cooling-reheating) is a useful tool to study if a phase transition is a reversible process and if the system is thermodynamically stable. Irreversible phase transitions would happen in unstable or metastable systems. For instance, celecoxib beta-casein micelles showed an irreversible phase transition



1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 6 of 28

6

due to the release of celecoxib during the first heating scan7. In contrast, the intraliposome doxorubicin-sulfate nanocrystals in pegylated liposomal doxorubicin (DOXIL®) show a reversible phase transition (melting/recrystallization) in cycled DSC scans from 15 to 90°C, indicating the high stability of this formulation9. In view of all the above, regulatory agencies recommend to perform thermotropic characterization by DSC as a part of the characterization of liposomal and other products21. In this study, DSC analysis of PL-MLP clearly demonstrates that the lipidated prodrug of MMC has major effects on the carrier liposome membrane structure and organization. The ability to remove the MMC moiety by thiolysis under mild conditions allows differentiating between the contributions of MMC and that of its lipophilic anchor to the thermotropic behavior of PL-MLP. This study is also an example of the validity of DSC characterization of liposomes with a lipidated drug in their membrane.

2. Experimental Section 2.1. Materials Hydrogenated soybean phosphatidylcholine (HSPC) and N-(carbamoyl - methoxypolyethylene glycol 2000)-1,2-distearoyl-snglycerol-3-phosphoethanolamine (mPEG2000-DSPE) sodium salt were purchased from Lipoid GmbH, Ludwigshafen, Germany; mitomycin C was from Dexcel Pharma, Or Akiva, Israel. MLP, the lipophilic prodrug of MMC, was synthesized by Laurus, Hyderabad, India, by a process based on Zalipsky & Gabizon 22. Cholesterol (Chol) and dithiotreitol (DTT) were purchased from Sigma Chemicals, St. Louis, MO; Methanol and isopropanol, HPLC grade, were purchased from Biolab Ltd, Israel.


Page 7 of 28

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


2.2. Methods 2.2.1. Liposome preparation PL-MLP and a preparation of blank liposomes having the same lipid composition as PL-MLP but without MLP were manufactured by Northern Lipids Inc. (Burnaby, BC, Canada) and provided by Lipomedix Pharm. Ltd. (Jerusalem, Israel). Briefly, a mixture of HSPC, Chol, mPEG2000-DSPE, and the prodrug conjugate MLP at a mole ratio 55:30:5:10, respectively, was dissolved in ethanol/tertiary butanol (50:50, v/v) and mixed with 5% dextrose /sodium phosphate buffer 20mM, pH 7.0, at a 20:80 v/v ratio. The liposome suspension was extruded under high pressure through stacked polycarbonate membranes of 0.08 or 0.10 µm pore size at 65°C. Organic solvents were removed by dialysis or diafiltration against the buffer. The liposome formulation was adjusted to a final MLP concentration of 5 mg/ml and sterile-filtered through 0.22 µm membranes. In the case of blank liposomes, the lipid composition was HSPC:Chol:mPEG2000-DSPE at molar ratio of 61:33:6, respectively. All other steps were as for PL-MLP. 2.2.2. Liposome characterization The liposome size and its distribution as poly dispersity index (PDI) was measured by Dynamic Light Scattering (DLS) (Malvern Instruments, Worcestershire, UK) at 25◦C. The sample was illuminated by He-Ne red laser at 633 nm, and the scattered light intensity was measured by the detector positioned at 173°. The Stokes–Einstein equation was used to obtain the average diameter of the liposomes from the intensity weighted diffusion coefficient. Three measurements were made per sample.



1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 8 of 28

8

Liposome samples were imaged using Cryo-transmission electron microscopy (Cryo-TEM). Sample preparation and examination by cryo-TEM was carried out at the Hebrew University Center for Nanoscience and Nanotechnology, Edmond J. Safra Campus, (Jerusalem, Israel) on a FEI Tecnai 12 G2 TEM, operated at 120 kV. Images were recorded on a 4K x 4K FEI Eagle CCD camera. The concentration of phospholipid for all liposomal dispersions used in this study was determined by the modified Bartlett method 23. Prodrug MLP was determined by reverse phase High Performance Liquid Chromatography (HPLC). Liposome samples were extracted by 1/10 dilution in isopropanol (IPA), then filtered and run on a LaChrom Merck Hitachi HPLC system with a Phenomenex Hypersil BDS C18 column, in a mobile phase composed of methanol: isopropanol 70/30, at a flow rate of 1ml/min, with UV detection at 360nm. MLP retention time under these conditions was 5.5 min. Peak areas were quantified and compared to standard curves of MLP. 2.2.3. DTT induced release of mitomycin C from the lipidated produg of Promitil Active free MMC can be released from the MLP prodrug of PL-MLP by reductive thiolysis 4. A standardized assay of the release was developed for quality control analysis and used in this study to explore the thermotropic impact of drug release. Ten mg MLP (2ml PL-MLP at 5mg/ml) were incubated with 10 mM DTT (final concentration in the reaction tube) at 37 ˚C for 1 hour with shaking. The reaction was stopped by extraction with IPA (1:10). Disappearance of the MLP peak was followed by HPLC. Liposomes were dialyzed against phosphate-buffered 5% dextrose to remove free MMC and then drug-free liposomes were analyzed by DSC. Due to the high permeability of the liposomes to DTT the lipidated prodrug in both leaflets of the liposome membrane is cleaved. The MLP cleavage by DTT


Page 9 of 28

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


is a continuous reaction and is best described by mono-exponential decay with 1st order kinetics4, suggesting that there is no difference in the rate of cleavage between inner and outer leaflet of the bilayer. 2.2.4. DSC measurements The “high-sensitivity” MicroCal™ VP-DSC system (Malvern, Worcestershire, UK) was used. The following liposome dispersions were analyzed by DSC: blank liposomes, and PL-MLP liposomes with or without DTT treatment. Scanning was carried out at a slow scanning rate to achieve a condition close to the thermodynamic equilibrium 11. Based on our previous experience with Pegylated Liposomal Doxorubicin 9, 1.0°C/min was selected as the routine scanning rate in the study. The samples were vacuum degassed by the MicroCal™ ThermoVac system (Malvern, Worcestershire, UK). The samples and the references were scanned in a cycle of heating-cooling-reheating from 15°C to 90°C at the rate of 1.0°C/min, using a filtering period of 10 sec and feedback mode of “Mid”. The reference for all the liposomal samples was 5% dextrose in 20mM phosphate buffered saline (PBS, pH 7.4), identical to the dispersion medium of PL-MLP liposomes. The calculations of thermodynamic parameters were carried out with the workstation of Microcal™ LLC DSC software (Malvern, Worcestershire, UK). First, the original thermograms were corrected by baseline subtraction and normalized to phosphatidylcholine (PC) or MLP concentrations. The “lipid-related” endotherms/exotherms were normalized using total PC concentration, and the ones related to MLP in the MLP-loaded liposomes were normalized with the MLP concentration. For the MLP- loaded liposomes after DTT treatment, the normalization was carried out only with the liposomal



1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 10 of 28

10

PC concentration. The normalized thermograms were analyzed by one-, two- or three-peak nonlinear modeling at the smallest chi-square or reduced chi-square values, and the thermodynamic parameters were calculated: the endotherms/exotherms in the model with the closest Tm to the Tm of the lipid-rich or the prodrug-rich domains were selected, and the enthalpy (∆H) and ∆T1/2 were calculated correspondingly. Tm was determined by the workstation from the normalized thermograms and labeled in the thermograms (the modeled value of Tm is slightly different from that automatically determined). Quantitative comparison of the thermodynamic parameters between different structural domains on the liposome membrane were based on the modeled values of Tm, ∆H and ∆T1/2.

3. Results and Discussion 3.1. Liposome characterization The lipidated prodrug, MLP, which is a component of the liposome membrane, is the source of MMC that is released in vivo by thiolysis and the actual active pharmaceutical ingredient (API) of the formulation. MLP constitutes 10 mole % of the liposome lipid bilayer. The molecular structure of MLP is shown in Figure 1. Liposome mean size, phospholipid and MLP concentrations for PL-MLP and blank (MLP-free) liposomes are presented in Table 1. The currently used clinical formulation of PL-MLP is prepared by dissolving the lipids in a 50:50 mixture of tertiary butanol and ethanol. When this solvent mixture is used, cryo-TEM analysis shows that the final product includes a significant fraction (~15%) of rods (Figure 2). These rods have been identified previously by 3D cryo-TEM as side views of thin disks, or incompletely swollen vesicles 4. The presence of t-butanol facilitates the dissolution of the lipids and MLP and results in liposomes of the


Page 11 of 28

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


desired size for therapeutic applications.

Figure 1: (a) Molecular structure of mitomycin C prodrug (MLP); (b) Thiolytic reduction of the prodrug MLP with DTT. Table 1: Main characteristics of Liposomes tested Before DTT treatment

After DTT treatment

Components/Properties Blank

PL-MLP

PL-MLP

HSPC (µmoles/ml)

21.9

17.9

17.7

MLP (µmoles/ml)

-

4.03

-

MLP/total PL (mol/mol)

-

0.23

-

Diameter (nm)

72

93.1

149

PDI

0.091

0.13

0.2



1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 12 of 28

12

Figure 2: Cryo-transmission electron microscopy of: (a) Blank liposomes (arrows point to rods/disks), (b) PL-MLP liposomes (arrows point at some of the rod-shaped particles), (c) PL-MLP liposomes (arrows point at some of the ”kissing vesicles”), (d) PL-MLP treated with DTT. Note some large liposomes with diameters approaching 200 nm, and wide size variability.


Page 13 of 28

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


3.2. The effect of the lipidated prodrug (MLP), on the thermotropic behavior of pegylated nano-liposomes The molecular structure of MLP with its two stearoyl acyl chains supports a stable insertion in the liposome bilayer. One of the most sensitive ways to characterize the effect of a broad spectrum of guest molecules on the liposome lipid bilayer is through the thermotropic behavior of the liposomes11. We used DSC to investigate whether insertion of MLP into the membrane modifies the phase transition temperature and alters the uniform lipid arrangement of the bilayer. Indeed, we found important effects of MLP on the thermotropic behavior of the liposomes. As shown in Fig. 3 the thermograms demonstrate that the endotherms are strongly affected by the lipidated prodrug.

Figure 3: Overlapped thermograms of blank (black) and PL-MLP liposomes (red) in the 1st heating scan normalized to HSPC concentration.



1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 14 of 28

14

This difference between the blank liposomes and the PL-MLP liposomes is also expressed in thermodynamic parameters that characterize the phase transitions observed (Table 2), and demonstrate a significant effect of MLP on the endotherm related to membrane lipid organization. A sharp MLP-related endotherm, with Tm at 56.25°C and a narrow phase transition (∆T1/2 of 0.85°C), partially overlapping with the expected blank liposome lipid-associated endotherm at 52-53oC (∆T1/2 of 8.00°C), were observed for PL-MLP. This is in contrast to a single, shallow, and smeared blank liposome-associated endotherm at 52-53oC (∆T1/2 of 13.05°C). Such type of membrane lipid phase transition is expected from a cholesterol-rich membrane9,13,19. The complex thermogram of PL-MLP suggests the formation of prodrug-rich domains in the PL-MLP membrane plane, which are tightly packed in highly-ordered crystal-like structure and displaying a higher phase transition than HSPC. The DSC analysis also suggests a lateral phase separation between the prodrug-rich domains and the membrane-lipid-rich domains. The membrane lipid-rich domains of PL-MLP have a similar Tm (52.58°C) as that of the blank liposome (Tm of 52.98°C), but with tighter packing. This is supported by the differences in ∆T1/2 (8.00°C and 13.05°C) and ∆H (4.18 kcal/mole and 2.57 kcal/mole) between PL-MLP and blank liposomes respectively (Table 2). The increased enthalpy and cooperativity of the membrane-lipid-rich domains indicates a strengthened lipid-lipid interaction as a result of the prodrug presence in the membrane. Thus, the prodrug is partly mixed with the membrane lipids and partly in phase-separated prodrug-rich domains. The interaction between MMC moieties is not limited only to the membrane plane of the same liposomes but also occurs between neighboring vesicles as suggested by the presence of the “kissing liposomes” observed in cryoTEM of PL-MLP but not in blank liposomes (Fig. 2a and 2c).


Page 15 of 28

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


Table 2: Thermodynamic parameters of blank and PL-MLP liposomes in the 1st heating DSC scan Membrane lipid-related phase transition

MLP-related phase transition

Liposome Tm (°C)

∆T1/2 (°C)

∆H (kcal/mole)

Tm (°C)

∆T1/2 (°C)

∆H (kcal/mole)

Blank

52.98

13.05

2.57

-

-

-

PL-MLP

52.58

8.00

4.18

56.25

0.85

4.18

3.3. Effect of MLP cleavage and MMC release on the thermotropic behavior of PL-MLP liposomes In order to differentiate between the contribution of the lipophilic anchor of MLP and that of the relatively water-soluble MMC moiety on the PL-MLP membrane arrangement, we cleaved the liposomal prodrug with a thiolytic reducing agent. For this we used DTT, a potent dithiol which releases all MMC from the prodrug in both liposome bilayer leaflets, while keeping the lipophilic anchor of the prodrug inserted in the liposome bilayer. DTT causes release of MMC in nearly stoichiometric order, that can be kinetically described using nonlinear curve fitting by a simple one-phase decay equation 4. Most of the generated MMC can be dialyzed out from the liposome suspension indicating that it is free and not retained by the liposomes (data not shown). We then compared the cryoTEM appearance and thermodynamic properties of liposomes with and without MMC. Cryo-TEM examination of the PL-MLP formulation after DTT treatment shows important morphologic changes (Figure 2d and Table 1): vesicle size and PDI increase significantly, the rods or disks disappear, and no “kissing” liposomes



1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 16 of 28

16

are observed. Altogether this suggests that major changes in membrane organization and inter-liposome surface interactions are taking place as a consequence of MLP cleavage. DSC analysis reveals clearly different thermotropic behaviors when intact PL-MLP and cleaved PL-MLP (in which MMC was removed by thiolysis) liposomes are compared (Table 3, Figure 4a). In the 1st heating scan (Figure 4a), the biphasic endotherm pattern of PL-MLP disappeared. Instead, a much more organized membrane with a higher phase transition temperature around 58.5°C and smaller ∆T1/2 (2.67°C) than that of blank liposomes was observed. This result confirms that the sharp endotherm at 56.25°C observed in the 1st heating in PL-MLP liposomes without DTT treatment (Figures 3 and 4a, red line) is attributed to the prodrug present on the liposome membrane, and most probably to the π-π stacking between the planar aromatic rings of the mitomycin-C moiety in the bilayer-water interface. Similarly, π-π stacking of the benzyl spacer group may also contribute to MLP intermolecular bonding. The strong intermolecular stacking leads to the formation of a prodrug-rich domain on the liposome membrane with highly-ordered crystal-like structure. π-π stacking between prodrug domains of neighboring vesicles may also explain the occasional cryo-TEM detection of pairs of “kissing vesicles” (Figure 2c) with a highly dense liner band at the liposome contact plane. The thiolated diacyl-glycerol lipid anchor of the prodrug remaining in the membrane after prodrug cleavage appears to strengthen the lipid-lipid interactions due to its small head group resulting in an increase of the phase transition temperature (Fig. 4a). Therefore, the membrane is more homogeneous and rigid after the cleavage of MLP as compared to blank liposomes and PL-MLP liposomes.


Page 17 of 28

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


Figure 4: Overlapped thermograms of PL-MLP with and without DTT treatment and the blank liposome in cycled DSC scanning

However, MLP-cleaved PL-MLP liposomes were not able to maintain the uniform arrangement of the membrane lipids during heat-cool-reheat cycled DSC scanning (Figure 4a-c). Split exotherms indicating obvious existence of different phases on the liposome membrane plane were observed in the 1st cooling and 2nd heating cycles (Figure 4b-c), suggesting that the organization may be metastable. The very sharp exotherm at 56.52°C with ∆T1/2 of 0.83°C of the 1st cooling cycle can be attributed to a lipophilic anchor-rich domain with very strong lipid-lipid interactions (Figure 4b). In the 2nd heating cycle, MLP-cleaved PL-MLP liposomes separated into two membrane domains (Figure 4c), as reflected by the two partially overlapping but distinct endotherms with Tm of 52.69°C and 59.22°C, respectively. Both domains are of ordered structure with ∆T1/2 of 1.34°C and 1.67°C, respectively (Table 3). These two domains may be related to the interaction of the liposome lipids with the prodrug cleavage by-products (thiolated lipophilic anchor and thiobenzyl spacer group).



1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48

Page 18 of 28

18

Table 3: Thermodynamic parameters of the blank and PL-MLP liposomes with (+) and without (-) DTT treatment* DTT treatment Blank

Endotherm/Exotherm 1 Tm (°C)

∆T1/2

/


∆T1/2

(°C)

∆H (kcal/mole)

52.98

13.05

2.57

+

55.50

7.83

−

47.91

+


∆T1/2

(°C)

∆H (kcal/mole)

(°C)

∆H (kcal/mole)

48.79

12.75

-3.41

52.19

14.90

3.80

2.32

58.65

2.67

4.61

-

-

-

12.34

2.06

53.43

5.17

2.46

56.25

1.00

4.41

35.92

2.50

-1.05

53.01

10.38

-4.90

56.53

0.83

-0.75

−

46.27

8.03

-3.14

53.45

9.52

-1.30

49.28

2.50

-7.68

+

52.52

1.34

0.75

54.20

10.70

4.97

59.05

2.34

1.52

−

48.20

11.53

3.97

53.55

4.00

1.29

55.71

1.67

2.98

PL-MLP 1st heat

1st cool

2nd heat

*For DTT (+) liposomes, all the normalization is based on the HSPC concentration; For DTT (-) liposomes, endotherm/exotherm 1 is normalized to HSPC concentration; endotherm/exotherm 2 (main membrane lipid-related peak) is normalized to prodrug concentration in the liposome in TWO-peak modeling; or normalized to HSPC concentration in THREE-peak modeling; endotherm/exotherm 3 (main MLP-related or MLP cleavage product-related peak) is normalized to the prodrug concentration in the liposomes. All the concentrations are in mM units. Minus (-) before the enthalpy values stands for exotherms.


Page 19 of 28

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


In comparison to MLP-cleaved PL-MLP liposomes, the membrane of MLP-intact PL-MLP liposomes maintained relatively reversible phase transitions for both lipid-rich and prodrug-rich domains in the cycled DSC scanning, but with peaks with lower degree of sharpness, suggesting increased mixing of the prodrug with the liposomal lipids and a stronger effect of cholesterol with smearing of the phase transition (Figure 4 a-c). An endotherm at 55.38°C representing the melting of the prodrug-rich domain was observed in the 2nd heating scan (Figure 4c), just below the Tm in the 1st heating (56.25°C), with some broadening of the peak (∆T1/2 of 1.3°C compared with 1.00°C in the 1st heating scan). These results demonstrate again that the MMC moiety in the prodrug plays a role in the membrane structure arrangement of PL-MLP. Thus, DSC characterization of PL-MLP liposomes, before and after cleavage and release of MMC, demonstrates that the presence of the lipidated prodrug greatly impacts the membrane organization of PL-MLP. The amphipathic MMC moiety is an important driving force for the formation of the prodrug-rich domain on the liposome membrane plane with high-ordered crystal-like structure most probably through the strong π-π stacking between MMC molecules. This explains the instability of the bilayer when MMC is released by cleavage following DTT-induced thiolysis, with disappearance of the “kissing” vesicles and disks, and an increase in liposome size (Figure 2c) probably as a result of fusion of liposomes and water swelling. The in vivo significance of this de-stabilization of the liposome bilayer upon MLP cleavage is probably nil, since this effect will be triggered after MMC has left the liposome, and thiolysis of MLP does not occur in plasma but only after liposome deposition in tissues, where molecules with free thiols can attack MLP upon contact with cell membranes or upon access to the intra-cellular compartment 4. The lipophilic anchor contributes to the strengthened interactions between the prodrug and the



1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

20

membrane lipids, and results eventually in the formation of a homogeneous, albeit metastable, high phase transition membrane structure after the cleavage of MLP as revealed by DSC during the 1st heating cycle. These DSC studies shed light on the molecular model underlying the high stability of PL-MLP in aqueous buffers and plasma, and its long in vivo circulation half-life3-5.

Conclusion Using DSC as a liposome characterization tool, we demonstrate a clear pattern of effects of the lipidated prodrug of mitomycin C on the membrane organization of PL-MLP. From the thermodynamic analysis, it follows that the liposome membrane of PL-MLP is heterogeneous with lateral phase separation of MLP prodrug-rich domains. The π-π stacking between the planar aromatic rings of the mitomycin C moieties leads to the formation of prodrug-rich domains with highly ordered structure on the liposome membrane plane. Both mitomycin C and lipophilic anchor moieties of the prodrug strengthened the arrangement of the lipid-rich domain via interactions of the prodrug and/or its lipophilic anchor with the membrane lipids. This stable thermodynamic conformation may explain the high stability of the PL-MLP formulation. This study also provides us with an interesting example of the application of high sensitivity DSC in understanding the composition-structure-behavior dynamics of liposomal formulations of lipophilic API’s.

Acknowledgement This study was supported by the following: (1) the Barenholz Fund; (2) Postdoctoral fellowships


Page 20 of 28

Page 21 of 28

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


jointly sponsored by the Planning and Budgeting Committee of the Council for Higher Education in Israel and the Hebrew University of Jerusalem to Xiaohui Wei and Yogita Patil; (3) National Natural Science Foundation of China grant to Xiaohui Wei (No.81202475); (4) Lipomedix Pharmaceuticals Inc.

References 1.

Gabizon, A.; Tzemach, D.; Horowitz, A. T.; Shmeeda, H.; Yeh, J.; Zalipsky, S.

Reduced

Toxicity and Superior Therapeutic Activity of a Mitomycin C Lipid-Based Prodrug Incorporated in Pegylated Liposomes. Clinical Cancer Research 2006, 12, (6), 1913-1920. 2.

Zalipsky, S.; Saad, M.; Kiwan, R.; Ber, E.; Yu, N.; Minko, T.

Antitumor activity of new

liposomal prodrug of mitomycin C in multidrug resistant solid tumor: Insights of the mechanism of action. Journal of Drug Targeting 2007, 15, (7-8), 518-530. 3.

Gabizon, A.; Amitay, Y.; Tzemach, D.; Gorin, J.; Shmeeda, H.; Zalipsky, S.

Therapeutic

efficacy of a lipid-based prodrug of mitomycin C in pegylated liposomes: Studies with human gastro-entero-pancreatic ectopic tumor models. Journal of controlled release 2012, 160, (2), 245. 4.

Amitay, Y.; Shmeeda, H.; Patil, Y.; Gorin, J.; Tzemach, D.; Mak, L.; Ohana, P.; Gabizon, A.

Pharmacologic Studies of a Prodrug of Mitomycin C in Pegylated Liposomes (Promitil(®)): High Stability in Plasma and Rapid Thiolytic Prodrug Activation in Tissues. Pharmaceutical research 2016, 33, (3), 686. 5.

Golan, T.; Grenader, T.; Ohana, P.; Amitay, Y.; Shmeeda, H.; La-Beck, N. M.; Tahover, E.;

Berger, R.; Gabizon, A. A.

Pegylated liposomal mitomycin C prodrug enhances tolerance of

mitomycin C: a phase 1 study in advanced solid tumor patients. Cancer Medicine 2015, 4, (10), 1472-1483. 6.

Turjeman, K.; Bavli, Y.; Kizelsztein, P.; Schilt, Y.; Allon, N.; Katzir, T. B.; Sasson, E.; Raviv, U.;

Ovadia, H.; Barenholz, Y. Nano-drugs based on nano sterically stabilized liposomes for the



1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 22 of 28

22

treatment of inflammatory neurodegenerative diseases. PloS one 2015, 10, (7), e0130442. 7.

Perlstein, H.; Turovsky, T.; Gimeson, P.; Cohen, R.; Rubinstein, A.; Danino, D.; Barenholz, Y.

Thermotropic behaviour of celecoxib-loaded beta-casein micelles: relevance to the improved bioavailability. European Journal of Nanomedicine 2015, 7, (4), 303–312.. 8.

Cohen, R.; Steiner, A.; Kanaan, H.; Barenholz, Y.

Chemical and physical characterization of

remotely loaded bupivacaine liposomes: comparison between large multivesicular vesicles and small unilamellar vesicles. Journal of Materials Chemistry B 2013, 1, (36), 4619-4627. 9.

Wei, X.; Cohen, R.; Barenholz, Y.

Insights into composition/structure/function relationships of

Doxil® gained from "high-sensitivity" differential scanning calorimetry. European Journal of Pharmaceutics and Biopharmaceutics 2016, 104, 260-270. 10. Lichtenberg, D.; Menashe, M.; Donaldson, S.; Biltonen, R. L. Thermodynamic characterization of the pretransition of unilamellar dipalmitoyl-phosphatidylcholine vesicles. Lipids 1984, 19, (6), 395-400. 11. Biltonen, R. L.; Lichtenberg, D.

The use of differential scanning calorimetry as a tool to

characterize liposome preparations. Chemistry and physics of lipids 1993, 64, (1), 129-142. 12. Barenholz, Y.; Cevc, G., Structure and properties of membranes. In Physical Chemistry of Biological Surfaces, Baszkin, A., Norde, W, Ed. Marcel Dekker: New York, 2000; pp 171-241. 13. Kitayama, H.; Takechi, Y.; Tamai, N.; Matsuki, H.; Yomota, C.; Saito, H.

Thermotropic phase

behavior of hydrogenated soybean phosphatidylcholine–cholesterol binary liposome membrane. Chemical and Pharmaceutical Bulletin 2014, 62, (1), 58-63. 14. Lichtenberg, D.; Freire, E.; Schmidt, C.; Barenholz, Y.; Felgner, P.; Thompson, T. Effect of surface curvature on stability, thermodynamic behavior, and osmotic activity of dipalmitoylphosphatidylcholine single lamellar vesicles. Biochemistry 1981, 20, (12), 3462-3467. 15. Hjort Ipsen, J.; Karlström, G.; Mourtisen, O.; Wennerström, H.; Zuckermann, M. Phase equilibria in the phosphatidylcholine-cholesterol system. Biochimica et Biophysica Acta


Page 23 of 28

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


(BBA)-Biomembranes 1987, 905, (1), 162-172. 16. Maswadeh, H.; Demetzos, C.; Daliani, I.; Kyrikou, I.; Mavromoustakos, T.; Tsortos, A.; Nounesis, G.

A molecular basis explanation of the dynamic and thermal effects of vinblastine

sulfate upon dipalmitoylphosphatidylcholine bilayer membranes. Biochim Biophys Acta 2002, 23, 1-2. 17. Koukoulitsa, C.; Kyrikou, I.; Demetzos, C.; Mavromoustakos, T.

The role of the anticancer

drug vinorelbine in lipid bilayers using differential scanning calorimetry and molecular modeling. Chemistry and physics of lipids 2006, 144, (1), 85-95. 18. Clas, S. D.; Dalton, C. R.; Hancock, B. C.

Differential scanning calorimetry: applications in

drug development. Pharm Sci Technolo Today 1999, 2, (8), 311-320. 19. Demetzos, C. Differential Scanning Calorimetry (DSC): a tool to study the thermal behavior of lipid bilayers and liposomal stability. J Liposome Res 2008, 18, (3), 159-73. 20. Gardikis, K.; Hatziantoniou, S.; Viras, K.; Wagner, M.; Demetzos, C., Interaction of Dendrimers with Model Lipid Membranes Assessed by DSC and Raman Spectroscopy. In Nanocarrier Technologies: Frontiers of Nanotherapy, Mozafari, M. R., Ed. Springer Netherlands: Dordrecht, 2006; pp 207-220. 21. Tinkle, S.; McNeil, S. E.; Mühlebach, S.; Bawa, R.; Borchard, G.; Barenholz, Y.; Tamarkin, L.; Desai, N. Nanomedicines: addressing the scientific and regulatory gap. Annals of the New York Academy of Sciences 2014, 1313, (1), 35-56. 22. Zalipsky, S.; Gabizon, A. Conjugate having a cleavable linkage for use in a liposome. United State Patent No: US 6,365,179 B1, 2002. 23. Gabizon, A.; Horowitz, A. T.; Goren, D.; Tzemach, D.; Shmeeda, H.; Zalipsky, S.

In vivo fate

of folate-targeted polyethylene-glycol liposomes in tumor-bearing mice. Clin Cancer Res 2003, 9, (17), 6551-9.



1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Graphic Abstract 338x190mm (96 x 96 DPI)


Page 24 of 28

Page 25 of 28

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


Figure 1 338x190mm (96 x 96 DPI)



1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60



Page 26 of 28

Page 27 of 28

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60





1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60



Page 28 of 28

Characterization of Pegylated Liposomal Mitomycin C Lipid-Based

Recommend Documents