Liposome Formulations of Combretastatin A4 and Its 4 Arylcoumarin ...

ISSN 19907508, Biochemistry (Moscow) Supplement Series B: Biomedical Chemistry, 2011, Vol. 5, No. 3, pp. 276–283. © Pleiades Publishing, Ltd., 2011. Original Russian Text © E.V. Moiseeva, N.R. Kuznetsova, E.V. Svirshchevskaya, N.V. Bovin, N.S. Sitnikov, A.S. Shavyrin, I.P. Beletskaya, S. Combes, A.Yu. Fedorov, E.L. Vodovozova, 2011, published in Biomeditsinskaya Khimiya.

Liposome Formulations of Combretastatin A4 and Its 4Arylcoumarin Analogue Prodrugs: the Antitumor Effect in the Mouse Model of Breast Cancer E. V. Moiseevaa, N. R. Kuznetsovaa, E. V. Svirshchevskayaa, N. V. Bovina, N. S. Sitnikovb, A. S. Shavyrinc, I. P. Beletskayad, S. Combese, A. Yu. Fedorovb, and E. L. Vodovozovaa* a

Shemyakin–Ovchinnikov Institute of Bioorganic Chemistry, Russian Academy of Sciences, ul. MiklukhoMaklaya 16/10, Moscow V437, GSP, 117997 Russia; phone +74953306610; fax +74953306601; email: [email protected] bN.I. Lobachevsky Nizhny Novgorod State University, Nizhny Novgorod, Russia c G.A. Razuvaev Institute of Organometallic Chemistry, Russian Academy of Sciences, Nizhny Novgorod, Russia dDepartment of Chemistry, M.V. Lomonosov Moscow State University, Moscow, Russia eUMRCNRS 6264, Saint Jerome Department of Sciences, AixMarseille University 1 and 2, Marseille, Cedex 20, France Received August 31, 2010

Abstract—The antimitotic agent combretastatin A4 (CA4) has been recently proposed as an antivascular agent for anticancer therapy. In order to reduce systemic toxicity by means of administration in liposome for mulations, new lipophilic prodrugs, oleic derivatives of CA4 and its 4arylcoumarin analogue (CA4Ole and ArCOle, respectively), have been synthesized in this study. Liposomes with mean diameter of 100 nm pre pared on the basis of egg phosphatidylcholine and baker’s yeast phosphatidylinositol quantitatively included up to 15 mol% of CA4Ole, or 7 mol% of ArCOle. To achieve targeting to neovascular endothelium prodrug bearing liposomes decorated with the tetrasaccharide selectin ligand Sialyl Lewis X (SiaLeX) have been also prepared. The antitumor activity was studied in vivo using the model of slowgrowing mouse breast cancer. Under the dose used (22 mg/kg) and the administration protocol (four injections, one per a week, starting from the appearance of palpable tumors) cytostatic CA4 did not reveal any anticancer effect; moreover, it even stimulated tumor growth. The liposome formulations of CA4Ole did not demonstrate such stimulation. However, to achieve a pronounced antitumor effect, the number of injections of liposomes should be appar ently increased. The cytotoxic activity of a novel antimitotic agent ArC was one order of magnitude lower in the human breast carcinoma cell culture in vitro. Nevertheless, in vivo in the mouse model of breast cancer the antitumor effect of this compound corresponded to the double equivalent dose of CA4. The results dem onstrate perspectives of SiaLeXliposomes loaded with ArCOle: the preparation partially inhibited tumor growth already after the second injection. Thus, subsequent optimization of doses and regimens of adminis tration both for ArC and liposomal ArCOle formulations are needed. Keywords: combretastatin A4, 4arylcoumarins, lipophilic prodrugs, liposomes, Sialyl Lewis X, breast cancer. DOI: 10.1134/S1990750811030073

*

INTRODUCTION

Combretastatin A4 (CA4, Fig. 1), one of the most effective plantderived antimitotic agents, a blocker of the colchicine site of tubulin, has been recently proposed as an antivascular agent for antican cer therapy [1]. The water soluble form, CA4 phos phate (CA4P), is now under Phase III clinical trials. Chemotherapy with antivascular drugs significantly decreases blood circulation in tumors, thus causing hypoxia and metabolic impairments in them. How ever, this is accompanied by nonspecific injuries of blood vessels of normal tissues and organs including heart, brain, spleen, skin, and kidneys of patients * To whom correspondence should be addressed.

receiving such chemotherapy. In the case of the brain and heart even rather shortterm changes in blood cir culation may be a cause of serious complications. In addition, animal experiments have shown that these drugs are frequently effective only at doses exceeding tolerant ones [2]. Drug incorporation into bio and hemocompatible nanosized carriers improves biodistribution and decreases systemic toxicity due to the decrease of blood concentrations of free preparations and accu mulation of the particles in tumors and inflammatory nidi owing to enhanced permeability of neovascular endothelium also known as the EPR (enhanced per meability and retention) effect [3, 4]. Delivery systems with mean diameters of 50–150 nm are the most suit

276

THE ANTITUMOR EFFECT OF PRODRUG LIPOSOME FORMULATIONS

277

o

A

MeO

C

MeO

OMe

OH

MeO

OMe

O

MeO

OMe

OMe

CA4 O

O

CA4Ole O

O

A O

C O

OH OMe

ArC

ArCOle

OH OH OH H3C CH3 HO OH O − COOOH O OH NH HO O O O O HN O O OH O O OH OH CH 3

SialylLewis X, SiaLeX

Phospholipids Prodrugs SiaLeXPEGDG

OMe

OOC

O N H

O

H N

nO

O

O O

OOC

n = 8–14

Neu5Acα23Galβ14 GlcNAcβ1 Fucα13 SiaLeXPEGDG

Fig. 1. Structures of the lipophilic prodrugs of combretastatin A4 and 4arylcoumarin, the lipophilic SiaLeX conjugate and sche matic presentation of a targeted drug liposome.

able for such passive transport. Liposomal carriers are already used in chemotherapy for systemic adminis tration of drugs; grafting of liposome surfaces with polyethylene glycol (PEG) residues protects them against early elimination from circulation by reticu loendothelial cells (Stealth® liposomes). For exam ple, encapsulation of the antitumor antibiotics doxo rubicin into liposomes with mean diameter of 100 nm significantly reduced unwanted side effects of the drug substance in clinical practice [6].

lipids) incorporated in the lipid bilayer of liposomes of mean diameter ~100 nm [8]. Ester bonds could be eas ily hydrolyzed by intracellular esterases as they are less specific than amidases and abundant in all tissues. At the same time, in liposomes prodrugs were resistant to premature hydrolysis by human plasma esterases [9]. The antitumor effect of the liposomal formulation of melphalan was demonstrated in the experimental lym pholeukemia [10] and the mouse model of breast can cer [11].

However, a technologically reasonable mode of encapsulation of watersoluble drug substances in the inner volume of liposomes, the method of remote loading against gradients of potassium acetate or ammonium sulfate concentrations is realizable only for limited number of drugs, which structurally repre sent weak amphiphilic acids of bases, such as anthra cycline antibiotics [7]. In this connection lipophilic prodrugs, which can be incorporated into the lipid bilayer of liposomes, attract much interest. Such approach not only simplifies the technology of liposo mal drug preparation but also improves their pharma cokinetics due to decreased losses of drug substance in the case of liposomal membrane damage both in cir culation and during interaction with a target cell. Ear lier we synthesized lipophilic prodrugs of the antitu mor drugs melphalan (sarcolysin) and methotrexate, which are widely used in clinical practice; they were prepared as drug diglyceride ester conjugates and the drugs were effectively (up to 10 mol% versus matrix

We have demonstrated that decoration of lipophilic prodrug loaded liposomes with the tetrasaccharide selectin ligand Sialyl Lewis X (SiaLeX, Fig. 1) as an oligo(ethylene glycol) octadecyl conjugate signifi cantly increased the antitumor effect [11]. Selectins are carbohydratebinding adhesive proteins that are increasingly expressed on activated endothelial cells, leukocytes, and platelets. They participate in primary interactions between circulating leukocytes and endothelial cells and are involved into numerous (patho)physiological processes including develop ment of inflammatory response, metastases, etc. [12]. Recently SiaLeX liposomes have been used for doxoru bicin delivery after experimental surgery in rats for stenosis prevention [13]. Pattillo et al. [14] proposed to include CA4P into liposomes carrying a fullsized antiEselectin mono clonal antibody attached to the distal end of PEG. These authors employed the phenomenon of increased expression of adhesion molecules on tumor

BIOCHEMISTRY (MOSCOW) SUPPLEMENT SERIES B: BIOMEDICAL CHEMISTRY

Vol. 5

No. 3

2011

278

MOISEEVA et al.

endothelium after irradiation and demonstrated sig nificant inhibition of experimental breast cancer in mice during systemic administration of immunolipo somes after irradiation [14]. Earlier the same group prepared PEGylated liposomes containing the water insoluble (initial) form CA4 in the lipid bilayer and targeted by a cyclic RGD peptide to αvβ3integrins [15]. However, loading of such liposomes with CA4 was less than 3 mol% and therefore they did not attract interest for subsequent systemic administration to ani mals. Indeed, it is reasonable to incorporate a small hydrophobic molecule of CA4 into the apolar part of the lipid bilayer, however, for reliable retention in lipo somes by hydrophobic interactions it requires modifi cation with an additional alkyl membrane anchor. The aim of this study was to develop liposomal forms including those “equipped” with the SiaLeX ligand for the acyl prodrugs combretastatin A4 and its 4aryl coumarin analogue (ArC; Fig. 1) and investigate their antitumor activity using the experimental model of breast cancer in mice. The presence of two noncopla nar synaryl fragments A and C linked by a rigid C⎯C bond in 4arylcoumarins determines their structural analogies with CA4 [16, 17]. Indeed, 4arylcoumarin analogues of CA4 also block cell cycle G2/M phase, exhibit comparable levels of cytotoxic and antimitotic activity with CA4 in cultures of human tumor cell lines, and also similar values of binding constants to tubulin and therefore both types of these molecules may be referred to the ligands of the colchicine site of tubulin [16, 18]. An important advantage of ArC and other 4[2(hydroxymethyl)aryl]coumarins over CA 4 is their rigid structure [1618], while easy Z/E isomerization of CA4 results in significant decrease of its antitumor activity during use and storage [19]. MATERIALS AND METHODS Syntheses of Prodrugs Combretastatin A4 was prepared using the stan dard method [20]. 4Arylcoumarin ArC was synthe sized by analogy with other 4[2(hydroxyme thyl)aryl]coumarins [16]. 1HNMR and 13CNMR spectra were registered using a Bruker ARX 400 (USA) spectrometer in CDCl3 at 300.13 and 75.54 MHz, respectively; chemical shifts are shown in a δ scale (ppm) using residual solvent signals (1HNMR δ 7.24; 13 CNMR δ 77.0). C and Hanalysis was performed using a PerkinElmer Series II CHN/O Analysis 2400 instrument (USA). Column chromatography was car ried out using Silicagel 60 (70–230 mesh, Alfa Aesar, Germany). Other reagents were obtained from Sigma Aldrich (USA) or Lancaster (UK). Solvents were puri fied by standard methods. After purification reaction products were dried at 20 Pa. 5(3,4,5Trimethoxystyryl)2methoxyphenyl ole ate (CA4Ole). Oxalyl chloride (80 μl, 0.87 mmol) was

added in an inert atmosphere to an oleic acid solution (0.21 g, 0.76 mmol) in anhydrous tetrahydrofuran (TNF, 2 ml). The mixture was stirred at 65°C for 1 h and then evaporated and dried. The dry residue was dissolved in 2 ml THF in the inert atmosphere and was added dropwise to CA4 sodium phenolate, which was obtained by adding NaH (0.015 g, 0.64 mmol) as 60% suspension in mineral oil to CA4 solution (0.18 g, 0.58 mmol) in anhydrous THF (1 ml). The reaction mixture was stirred at 65°C for 2 h and then evapo rated. The residue was dissolved in ethyl acetate and extracted three times with 5% NaOH. An organic layer was dried over anhydrous Na2SO4 and the resultant product was purified by chromatography on a silicagel column using ethyl acetate—petroleum ether (40/65, 1 : 4) as an eluent. The yield of CA4Ole (colorless oil) was 0.23 g (67%). Found (%): C, 74.29; H, 9.04. C36H52O6. Calculated (%):C, 74.45; H, 9.02. 1H NMR spectrum: (δ, ppm; J, Hz): 0.88 (m, 3H, CH3); 1.31 (m, 20H, CH2); 1.72 (m, 2H, C(O)CH2CH2); 2.03 (m, 4H, CH2CH=CH); 2.52 (t, 2H, C(O)CH2CH2, J 8.0); 3.70 (s, 6H, OCH3); 3.79 (s, 3H, OCH3), 3.83 (s, 3H, OCH3); 5.35 (m, 2H, CH=CH); 6.45 (s, 2H, ArCH=CHAr’); 6,50 (s, 2H, H2’, H6’); 6.84 (d, 1H, H5”, J 8.0); 6.99 (d, 1H, H2”, J 2.0); 7.11 (d.d., 1H, H6”, J 2.0, J 8.0). 13CNMR spectrum (δ, ppm): 14.1; 22.7; 25.0; 27.2; 29.0; 29.2; 29.3; 29.5; 29.7; 31.9; 33.9; 55.8; 55.9; 60.9; 105.9; 112.0; 123.2; 127.6; 128.6; 129.5; 129.7; 130.0; 132.4; 137.2; 139.5; 150.3; 153.0; 171.7. 4(4'Methoxy3'oleoyloxyphenyl)chromen2 one (ArCOle) was synthesized similarly to CA4Ole. Using 0.03 g of ArC 0.04 g of ArCOle (76%) was obtained as yellow oil. Found (%): C, 76.42, H, 8.28. C34H44O5. Calculated (%): C 76.66, H, 8.33. 1H NMR spectrum: (δ, ppm; J, Hz): 0.85 (m, 3H, CH3); 1.30 (m, 20H, CH2); 1.78 (m, 2H, C(O)CH2CH2), 2.02 (m, 4H, CH2CH); 2.61 (t, 2H, C(O)CH2, J 7.4); 3.91 (s, 3H, OCH3); 5.34 (m, 2H, CH); 6.36 (s, 1H, H3); 7.12 (m, 2H, H5’, H6’); 7.34 (m, 3H, H2’, H6, H8); 7.55 (m, 2H, H5, H7). 13CNMR spectrum (δ, ppm): 14.1; 22.6; 25.0; 27.1; 27.1; 29.0; 29.1; 29.2; 29.3; 29.5; 29.7; 29.8; 31.6; 31.9; 34.0; 56.0; 112.6; 115.0; 117.3; 118.8; 123.3; 124.2; 126.9; 127.0; 127.6; 129.7; 130.0; 131.9; 140.0; 152.5; 154.2; 154.3; 160.7; 171.6. Preparation of liposomal drug dispersions was car ried out using egg yolk phosphatidylcholine (PC) and S. cerevisiae phosphatidylinositol (PI) produced by Reakhim (Russia). The glycoconjugate SiaLeXPEG DG was synthesized in three steps by the method of activated esters, using PEG biscarboxymethyl ether of mean molecular mass of 600 Da (Aldrich, USA), SiaLeX 3aminopropyl glycoside and rac1,2dio leoyl3(3aminopropionyl)glycerol. The following buffers containing 1 mM EDTA were prepared: phos phate buffered saline, pH 7.06 (PBS): KH2PO4, 0.2 g/l; NaH2PO4 ⋅ 2 H2O, 0.15 g/l; Na2HPO4, 1.0 g/l; KCl, 0.2 g/l; NaCl, 8.0 g/l; HEPES buffered saline,


Vol. 5

No. 3

2011


pH 7.2 (HBS): 25 mM HEPESNa, 140 mM NaCl; HEPES—N2(hydroxyethyl)piperazineN' 2ethanesulfonic acid (Flow Laboratories). Mixtures PC/PI/CA4Ole(+/–SiaLeXPEG DG), 7.5 : 1 : 1.5 (±0.2) (mol) or PC/PI/ArC Ole((±SiaLeXPEGDG), 8 : 1 : 0.7 (±0.2) (mol) were evaporated in roundbottom tubes in a rotor evapora tor at 40°C or below. The mixtures contained 54 mg PC, 8 mg PI, 8 mg (14 μmol) CA4Ole and 4 mg SiaLeXPEGDG (~1.87 μmol) or 81 mg PC. 11.4 mg PI, 5 mg (9.2 μmol) ArCOle and 5.6 mg SiaLeX PEGDG (~2.6 μmol). The lipid films were dried at 5 Pa for 30 min and then hydrated at room tempera ture for 2 h in 2 ml of PBS (during liposome prepara tion of experiments with cell cultures or animals) or HBS (for determination of CA4Ole and ArCOle incorporation into liposomes). The suspensions were shaken and subjected to 5 freezethaw cycles (liquid nitrogen – +40°C) and extruded 20 times through the polycarbonate membrane filters with the pore size of 100 nm (Nucleopore, USA) using an Avanti Polar Lip ids Miniextruder (USA). According to dynamic laser light scattering measurements performed on a Brookhaven Particle Analyzer 90+ (USA) the mean value of the liposome diameter varied for different preparations from 92 ± 30 to 105 ± 33. Prodrug con centrations in dispersions were determined after lipo some degradation (caused by 20fold dilution in etha nol). Registration of UV spectra and determination of optical density at absorption maxima (CA4Ole: λ = 287 nm, ε ~ 13440; ArCOle: λ = 307 nm, ε ~ 10200) using a SF256UVI doublebeam spectrophotometer (Lomophotonika, StPetersburg, Russia). Drug losses on filters were controlled by determining their content in solutions prepared by soaking filters in ethanol fol lowed by registration of UV spectra. The losses did not exceed 3–5%. Liposomal composition was deter mined after gelchromatography of Sepharose CL4B by analyzing fractions for phospholipid phosphorus (using a colorimetric method) and prodrugs (spectro photometrically) as described earlier for other prepa rations [8, 9]. CA4Ole and ArCOle were included almost completely into liposomes. Liposome disper sions were stored at +4°C for not more than 2 days.

279

1%. Control cells were incubated with a PBS aliquot and 1 % DMSO in the medium. Incubations were performed in the cultivation medium during 48 h in 24well plates. The number of living cells was deter mined by the Trypan blue exclusion test; percent of living cells was calculated as (number of living cells in the experimental sample/number of living cells in control) × 100. Each experiment was performed in duplicates. Cytotoxic activity (IC50) was calculated by the program Origin 6.0 (MicroCal Software Inc., USA). Tests for antitumor activity in vivo The breast cancer cell line Wnt1 of C57BL/6J obtained from the National Cancer Institute (NCI, Bethesda, USA) was maintained in vivo. Female C57BL/6 Lac Sto mice were obtained from the Stol bovaya nursery. Tumor cells (106 per mice) were inoc ulated into left hind paws. After appearance of palpa ble tumors female mice were subdivided into groups indistinguishable by these parameters on the average. CA4 and its derivatives were tested on 35 mice, while ArC and its derivatives were tested on 23 mice. Drug preparations were injected into the tail vein four times (with a oneweek interval between each injection): 0.2 ml of 7 mM CA4 (22 mg/kg) in PBS— 5% Tween 80 (n = 6, group 1); dispersions of the lipo somes PC/PI/CA4Ole(±SiaLeXPEGDG), 7.5 : 1 : 1.5 (± 0.2) (mol) at a dose equivalent to CA4 (n = 6⎯8, groups 2 and 3, respectively). Control groups (in each n = 6–7) received injections of PBS without the detergent or PBS—5% Tween 80 (groups C1 and C2). In experiments with ArC this compound was injected at a dose of 37 mg/kg (13.8 mM in PBS—5% Tween 80) and liposomal preparations contained 1/3 equiva lents of the ArCOle dose (4.6 mM in 0.2 ml PBS; the liposome composition 8 : 1 : 0.7 : 0.2). Table summa rizes results of these tests. The antitumor effect was evaluated by dynamics of tumor growth (evaluated by mean tumor diameter) and improvement of survival of tumor bearing mice. RESULTS AND DISCUSSION

Determination of cytotoxic activity in vitro Human mammary gland carcinoma cells HBL100 were cultivated in the atmosphere of 4% CO2 at 37°C in RPMI1640 medium (ICN Biomedicals Inc., USA) supplemented with 0.2% NaHCO3, 2 mM Lglutamine, 50 μg/ml gentamycin G, 100 μg/ml streptomycin and 10% calf fetal serum (heatinacti vated) (Gibco BRL, UK), pH 7.4, and passaged twice a week. Cells were incubated with various liposomes containing 0.005–1.0 μM CA4Ole or 0.8–80 μM ArCOle or with initial CA4 (0.001–0.2 μM) or ArC (0.01–1.0 μM) added as dimethyl sulfoxide (DMSO) solutions. Final DMSO concentration did not exceed

Liposomes were obtained using mixtures of natural phospholipids with prodrugs and the glycolconjugate by means of the standard method of extrusion through the filters with pore size of 100 nm [8, 9, 21]. This results in formation of unilamellar (i.e. with a single lipid bilayer) liposomes of requested size, which carry a modified drug and the carbohydrate ligand. Figure 1 shows a schematic liposome construct. Significant exposure of SiaLeX carbohydrate groups on a reason able distance from the liposomal membrane by means of flexible hydrophilic PEGs (polymerization degree 8⎯14) determines effective contact with surface recep tors on target cells.


Vol. 5

No. 3

2011

280

MOISEEVA et al.

Average life span and percent of survived animals in the groups of mice with inoculated breast cancer after treatment with combretastatin A4 and 4arylcoumarinbased preparations Dose µmol/kg

Groups Control PBS (C1) Control PBS/Tween 80 (C2) CA4/Tween 80 (group 1) CA4Ole in liposomes (group 2) CA4Ole SiaLeXliposomes (group 3) Control PBS (C) ArC/Tween 80 (group 1) ArCOle in liposomes (group 2) ArCOle SiaLeXliposomes (group 3)

— — 70 70 70 — 138 46 46

Dose mg/kg

n*

ALS

Survival, %

— — 22 40 40 — 37 25 25

7 6 6 7 8 5 6 6 6

75.6 66.5 72.8 82.3 81.1 78.8 86.5 79.5 83.5

14 17 33 25 25 60 83 67 67

Note: *Number of mice in the group.

The Stealth® liposomes are formed using phos pholipids containing saturated acyl residues (usually, hydrogenated soy lecithin) and cholesterol (up to 30%): a rigid membrane is required to reduce leakage of drug from formulation in blood [5, 6]. Lipophilic prodrugs are components of the liposomal lipid bilayer and after its damage they are retained within lipo somes as we have shown using lipid conjugates of watersoluble drugs [8, 9]. A liquid lipid bilayer can accommodate more prodrug molecules. Moreover, it exhibits easier fusion with tumor cell membranes [22]. Thus, in our case phospholipids containing about half saturated acyl (palmitoyl and stearoyl) residues and

CA4 CA4Ole ArC ArCOle

Cell survival

100 80 60 40 20 0 1E3

0.01

0.1

1 10 100 Concentration, μM

Fig. 2. Cytotoxicity of combretastatin A4 and 4arylcou marin and liposomal preparations of the lipophilic pro drugs in human mammary gland carcinoma cells HBL 100. The number of living cells was determined by the Try pan blue exclusion test after incubation for 48 h. Percent of living cells was calculated as (number of living cells in the experimental sample/number of living cells in control) × 100. Data represent mean ± SEM for each experiment per formed in duplicate.

also unsaturated oleoyl and smaller amount of lino leoyl residues represent a basis of the liposomal bilayer (up to 90 mol%). Phosphatidylinositol (PI) protects against opsonization in blood circulation: PI inositol residues representing about 10% of the bilayer form a highly hydrated sterically stabilized shell on the lipo somal surface like Stealth® liposomal PEG residues (polymerization degree of 45–48) [23]. In addition, it seems unlikely that PI can induce side effects includ ing allergic reactions and mucositis, which accom pany application of the Stealth® liposomes [24]. CA4Ole loading capacity of liposomes with mean diameter of 100 nm was 15 mol% and the prodrug was incorporated in the lipid bilayer almost completely (i.e. with efficiency close to 100%). The same effi ciency of bilayer loading with ArCOle was achieved under extrusion conditions at 22–40°C for the lipo some capacity of 7 mol%. The antiproliferative activity of CA4, its analogue and liposomal preparation of these prodrugs were investigated in human mammary gland carcinoma cells in vitro. After incubation with all tested drug for mulations for 48 h the following changes in cell mor phology were observed: the adhesion type cells with characteristic spreading over plate wells acquired a rounded shape suggesting blockade of the mitotic stage. Figure 2 shows data on cell survival. The calcu lated IC50 values (concentration required for 50% inhibition of cell growth) were 0.0075 ± 0.002, 0.096 ± 0.003, 0.023 ± 0.004, and 0.79 ± 0.17 μM for CA4, ArC, and corresponding liposomal CA4Ole and ArC Ole, respectively. In experiments with cell cultures usual multiple (up to tens of times) decrease in toxicity of drugs entrapped into nanosized carriers is explained by altered mechanism of endocytosis and also by an additional stage of intracellular unloading of the carriers. The stage of prodrug hydrolysis can also slow down manifestations of cytotoxic activity. It should be also noted that the cytotoxicity of CA4


Vol. 5

No. 3

2011

22 20 18 16 14 12 10 8 6 4 2

281

(a)

C1 C2 CA4 L/CA4Ole LSiaLe X/CA4Ole 0

20 30 40 50 10 Days after the beginning of treatment (b)

100 80 Survival

analogue (ArC) was about 12 times lower than that of CA4 itself (Fig. 2). The effect of drugs on tumor growth in vivo was investigated in mice with inoculated breast cancer cells, with the intravenous treatment of mice being started after appearance of palpable tumors. Mice were subdivided into groups equivalent by tumor sizes. In each group tumor sizes varied from 0.6 to 8 mm. For intravenous administration of the hydrophobic substances CA4 and ArC were solubilized in an aque ous phase by the Pharmacopia detergent Tween 80 added up to 5% (v/v) and therefore one of control groups of mice (C2) received injections of PBS–5% Tween 80. The first experiment was performed with CA4 and its derivatives. Figure 3 shows dynamics of tumor growth (Fig. 3a) and dynamics of mice survival (Fig. 3b) in the investigated groups. No significant changes in animal weights from different groups were observed (data not shown). There was a marked ten dency in body weight loss in group C2 (treated with the PBSTween); this possibly was a cause of death of the first animals in this group (Fig. 3b). It appears that sys temic administration of the detergent in the concen tration used was toxic for animals. Both control groups (C1 and C2) were character ized by the same tumor growth. Starting from the sec ond injection intact CA4 insignificantly stimulated tumor growth. The liposomal preparations of CA4 Ole tended to attenuate this effect and no differences between usual liposomes (group 2) and SiaLeXlipo somes (groups 3) were found. Improved survival of treated animals (Fig. 3b) ver sus control group C1 was found only in mice treated with CA4Ole liposomes (group 2). This resulted in a small increase of the average life span (ALS; table): 82.3 versus 75.6 days (+8.8%). However, differences in ALS did not reach the level of statistical significance. Nevertheless, there was some increase in ALS in the group of CA4treated mice compared with C2 (+9.5%) and also in the group of mice treated with liposomes with CA4Ole compared with CA4 (+13%). The targeted form of the liposome prepara tion (group 3) did not improve survival of animals and the ALS value was the same as in the liposome prepa rations without SiaLeX (ALS = 81.1 days). Thus, in the used dose (22 mg/kg) and the protocol of drug administration CA4 did not exhibit any anti tumor effect, moreover, it even slightly stimulated tumor growth. The liposomal forms of CA4Ole did not stimulate tumor growth. Lack of the antitumor activity in CA4 may be associated with an insuffi ciently high dose of the drug. For example, Patillo et al. demonstrated [14] that single administration of CA4P to mice with transplanted mammary gland tumor MCa4 at the clinically accepted dose (81 mg/kg) did not inhibit tumor growth. A marked inhibition (by 30% within first 10 days) was achieved

Mean tumor diameter, mm


60 40 20 0 40

50

60 70 80 90 100 Days after the beginning of treatment

Fig. 3. Dynamics of tumor growth (a) and survival (b) of C57Bl/6 mice with the transplanted mammary gland tumor Wnt1. After appearance of palpable tumors (from 0.6 to 8 mm in diameter) females of each group received four intravenous injections (with oneweek interval between each injection) 0.2 ml of PBS (C1; n = 7), PBS 5% Tween 80 (C2; n = 6), solution of CA4 (22 mg/kg) in PBS5% Tween 80 (CA4; n = 6) or equivalent dose of CA4Ole in liposomes (L/CA4Ole; n = 7) or SiaLeX liposomes (LSiaLeX/CA4Ole; n = 8). Vertical arrows mark days of injections.

only during drug administration to irradiated mice and the tumor growth curve was basically the same as in irradiated (control) mice, which did not receive any medication. A similar level of tumor growth inhibition was found in mice, which received 4 injections of the same dose of CA4P (with oneday interval between each injection). Potent inhibition of breast cancer (by 80% during 15–20 days) was achieved after CA4P administration (15 mg/kg) in stealth immunolipo somes carrying monoclonal antibody (mAb) to Eselectin, but only after tumor irradiation, which sig nificantly increased selectin expression [14]. Without irradiation of animals the effect of immunoliposomes


Vol. 5

No. 3

2011

282

MOISEEVA et al.

Mean tumor diameter, mm

30

(a)

25 20 15 C ArC L/ArCOle

10 5

LSiaLeX/ArCOle

0

10

20 30 40 50 60 70 80 Days after the beginning of treatment (b)

100

Survival

90 80 70 60 50 20

40 50 60 70 80 30 Days after the beginning of treatment

Fig. 4. Dynamics of tumor growth (a) and survival of mice (b) treated with 4arylcoumarin preparations. The model of breast cancer and the mode of drug administration were the same as shown in the legend to Fig. 3. Groups: control PBS (K; n = 5); ArC solution (37 mg/kg) in PBS5% Tween 80 (ArC; n = 6), ArCOle, 1/3 of the equivalent dose in liposomes (L/ArCOle; n = 6) or in SiaLeXlipo somes (LSiaLeX/ArCOle; n = 6).

on tumor growth insignificantly differed from the effect of CA4P after irradiation (see above); however, the dose of drug administered in liposomes was 5.4 times lower [14]. It should be noted that the authors of that study [14] employed the model of rap idly growing mammary tumor (within 10 days after the beginning of therapy tumors in the control group increased their volume from 1 cm3 to 4 cm3 [14]. How ever, according to clinical practice, rapidly growing aggressive tumors are much more sensitive to chemo therapy [25]. We have used the model of slowgrowing mammary tumor, which better reflects corresponding disease in humans.

In the second experiment we tested various prepa rations based on the 4arylcoumarin analogue of CA4. The ArC preparation was administered at the dose of 37 mg/kg; this corresponds to the double (molar) dose of CA4 (used at the dose 22 mg/kg). The liposomal formulations of ArCOle contained 1/3 (mol) equivalents of the ArC dose, because at the lipo some capacity of 7% an increase in ArCOle dosage would require additional administration of significant quantities of both matrix phospholipids and SiaLeX PEGDG. Finally the molar dose of liposomal ArOle was 1.5 times lower than the dose of liposomal CA4 Ole (table). Figure 4 shows dynamics of tumor growth and sur vival of various groups of animals. ArC itself and its liposome analogue (ArCOle) had no influence on tumor growth, while the targeted liposomal form with SiaLeX caused inhibition of tumor growth observed already after second injection (Fig. 4a). No significant differences in survival of groups of mice were found (Fig. 4b). Although differences in ALS and percent of live female mice by day 96 did not reach the level of statistical significance there was a tendency for improved survival in ArCtreated mice (Fig. 4b and table). Interestingly, in contrast to CA4, ArC did not cause tumor hemorrhages . We suggest that the preparations used in this study can demonstrate essential antitumor properties during prolonged administration prior appearance of palpa ble tumor, i.e. tight after inoculation of tumor cells. CONCLUSIONS A novel liposomal construct carrying the antimi totic agent combretastatin A4 in the form of the lipo philic prodrug has been proposed in this study. Employment of CA4 as the antivascular agent for anticancer therapy may be improved by insertion of its lipophilic prodrug into liposomes containing the car bohydrate ligand SiaLeX, specific for vascular endot helium of tumors. A pronounced antitumor effect may be achieved after a prolonged course of injections of the liposomal formulation. Antitumor efficiency of the liposomal formulation with SiaLeX may be increased by preliminary tumor irradiation as it has been demonstrated in [14]. 4Arylcoumarin (ArC), a novel antimitotic agent (an analogue of CA4), was one order of magnitude less toxic in the culture of human mammary gland car cinoma cells. Nevertheless, in the mouse model of slowgrowing breast cancer its antitumor effect corre sponded to the double equivalent dose of CA4. Under these conditions, CA4 did not exhibit any antitumor activity. For preparation of 4arylcoumarinbased liposo mal formulation the lipophilic prodrug ArCOle has been synthesized by analogy with CA4. Experiments on the mouse model of breast cancer demonstrate per


Vol. 5

No. 3

2011


spective of ArCOle administration in liposomes con taining SiaLeX. However, subsequent optimization of doses and regimens of administration both for ArC and liposomal ArCOle formulations are clearly needed. ACKNOWLEDGMENTS This work was supported by the Russian Founda tion for Basic Research (grant nos. 090300647a and 100401021a) and also by the Grant Council of the President of Russian Federation for State support of young scientists and leading scientific schools of Rus sian Federation (grant no. MD5606.2010.3) and the Federal Purpose Program (contract 16.740.11.0476). REFERENCES 1. Thorpe, P.E., Clin. Cancer Res., 2004, vol. 10, pp. 415– 427. 2. Grosios, K., Holwell, S.E., McGown, A.T., Pettit, G.R., and Bibby, M.C., Br. J. Cancer, 1999, vol. 81, pp. 1318– 1327. 3. Fenske, D.B. and Cullis, P.R., Expert Opin. Drug Deliv., 2008, vol. 5, pp. 25–44. 4. Maeda, H., Sawa, T., and Konno, T., J. Control. Release, 2001, vol. 74, pp. 47–61. 5. Lasic, D.D., and Papahadjopoulos, D., Science, 1995, vol. 267, pp. 1275–1276. 6. Gabizon, A., Schmeeda, H., and Barenholz, Y., Clin. Pharmacokinet., 2003, vol. 42, pp. 419–436. 7. Zucker, D., Marcus, D., Barenholz, Y., and Goldblum, A., J. Control. Release, 2009, vol. 139, pp. 73–80. 8. Vodovozova, E.L., Kuznetsova, N.R., Kadykov, V.A., Khutsyan, S.S., Gaenko, G. P., and Molotkovsky, J.G., Nanotechnologies in Russia , 2008, vol. 3, no. 3–4, pp. 228–239. 9. Kuznetsova, N., Kandyba, A., Vostrov, I., Kadykov, V., Gaenko, G., Molotkovsky, J., and Vodovozova, E., J. Drug Deliv. Sci. Techn., 2009, vol. 19, pp. 51–59. 10. Kozlov, A.M., Korchagina, E.Yu., Vodovozova, E.L., Bovin, N.V., Molotkovsky, J.G., and Syrkin, A.B., Byul. Eksper. Biol. Med., 1997, vol. 123, pp. 439441.

283

11. Vodovozova, E.L., Moiseeva, E.V., Grechko, G.K., Gayenko, G.P., Nifant’ev, N.E., Bovin, N.V., and Molotkovsky, J.G., Eur. J. Cancer, 2000, vol. 36, pp. 942–949. 12. Ehrhardt, C., Kneuer, C., and Bakowsky, U., Adv. Drug. Deliv. Rev., 2004, vol. 56, pp. 527–549. 13. Tsuruta, W., Tsurushima, H., Yamamoto, T., Suzuki, K., Yamazaki, N., and Matsumura, A., Biomaterials, 2009, vol. 3, pp. 118–125. 14. Pattillo, C.B., Venegas, B., Donelson, F.J., Valle, L.D., Knight, L.C., Chong, P.L.G., and Kiani, M.F., Pharm. Res., 2009, vol. 26, pp. 1093–1100. 15. Nallamothu, R., Wood, G., Pattillo, C.B., Scott, R.C., Kiani, M.F., Moore, B.M., and Thoma, L.A., AAPS PharmSciTech., 2006, vol. 7, no. 2, pp. E1–E10. 16. Bailly, C., Bal, C., Barbier, P., Combes, S., Finet, J.P., Hildebrand, M.P., Peyrot, V., and Wattez, N., J. Med. Chem., 2003, vol. 46, pp. 5437–5444. 17. Ganina, O.G., Daras, E., BourgarelRey, V., Peyrot, V., Andresyuk, A.N., Finet, J.P., Fedorov, A.Yu., Beletskaya, I.P., and Combes, S., Bioorg. Med. Chem., 2008, vol. 16, pp. 8806–8812. 18. Rappl, C., Barbier, P., BourgarelRey, V., Gregoire, C., Gilli, R., Carre, M., Combes, S., Finet, J.P., and Pey rot, V., Biochemistry, 2006, vol. 45, pp. 9210–9218. 19. Tron, G.C., Pirali, T., Sorba, G., Pagliai, F., Busacca, S., and Genazzani, A., J. Med. Chem., 2006, vol. 49, pp. 3033– 3044. 20. LaraOchoa, F. and EspinosaPérez, G., Tetrahedron Lett., 2007, vol. 48, pp. 7007–7010. 21. Mayer, L.D., Hope, M.J., and Cullis, P.R., Biochim. Biophys. Acta, 1986, vol. 858, pp. 161–168. 22. Funaki, N.O., Tanaka, J., Kohmoto, M., Sugiyama, T., Ohshio, G., Nonaka, A., Yotsumoto, F., Takeda, Y., and Imamura, M., Oncol. Rep., 2001, vol. 8, pp. 527– 532. 23. Gabizon, A. and Papahadjopoulos, D., Proc. Natl. Acad. Sci. USA, 1988, vol. 85, pp. 69496953. 24. Romberg, B., Metselaar, J.M., Baranyi, L., Snel, C.J., Bunger, R., Hennink, W.E., Szebeni, J., and Storm, G., Int. J. Pharm., 2007, vol. 331, pp. 186–189. 25. Bonetti, A., Zaninelli, M., Rodella, S., Molino, A., Sperotto, L., Piubello, Q., Bonetti, F., Nortilli, R., Turazza, M., and Cetto, G.L., Breast Cancer Res. Treat., 1996, vol. 38, pp. 289–297.


Vol. 5

No. 3

2011