Incorporation of new amphiphilic perfluoroalkylated bipyridine platinum and palladium complexes into liposomes: stability and structure-incorporation relationships

Biochimica et Biophysicr~ 4cm, !127 (1992)41-48 © 1992 Elsevier Science Publishers B.V. All rights reserved 0005-2760/92/$05.00

41

BBALIP 53955

Incorporation of new amphiphilic perfluoroalkylated bipyridine platinum and palladium complexes into liposomes: stability and structure-incorporation relationships Nathalie Garelli and Pierre Vierling Laboratoire tit, CTffmk, Mok;culairt; U, itt; de Recherche Associt;e au CNRS, Unh'ersitt; de Nice-Sophia Antipolis, Nice (France)

(Received 31 October 1991)

Key words: Perfluoroalkylated bipyridine; Perfluoroalkylatcd complex; Platinum; Palladium; Liposome

New perfluoroalkylatcd side-chain (formed by a perfluoroalkyl tail grafted onto a hydrocarbon spacer) bipyridine complexes of platinum, palladium and their hydrocarbon analogs, when dispersed in aqueous solutions in the presence of egg yolk phospholipids (EYP), arc !ncorporated into EYP vesicles. Most complexes retain their chemical and structural integrity when entrapped into liposomes, thermally sterilized under the FDA norms and stored for 3 months at 25°C. All complex/liposome preparations consist mainly of small unilamellar vesicles (size < 100 nm) together with a population of larger uni- or multilamellar vesicles (10U to 230 nm). These preparations are remarkably stable with respect to particle size and size distribution evolution and complex leakage: no precipitate of drug was detected even after 7 months of storage at 25°C. The impacl of the perfluoroalkyl tail and of the other structdral features of the complexes on their incorporation efficiency into EYP liposomes is assessed. A high fluorophilic character (long perfluoroalkyl tail), when tempered by an equally lipophilic one (long hydrocarbon spacer) is not detrimental to the incorporation efficiency of a perfluoroalkylated drug into hydrocarbon vesicles. This incorporation efficicacy is considerably improved by the introduction of a double bond between the perfluoroalkyl tail and the hydrocarbon spacer, forming the bipyridine side-chains.

Introduction

Over past years, cisplatin (CDDP) and its second generation derivatives [1,2] have been developed a:, antitumor drugs. The clinical success of CDDP in antitumor chemotherapy - but also its disadvantages, such as its extreme nephrotoxicity, fast elimination via the kidneys, rapid binding to plasma proteins, cross-resistance after several courses of therapy in patients with sensitive tumors, and limited efficacy against several human tumors - have, indeed, induced an all-out search for platinum [3] and non-platinum [4] analogs with lower toxicity, improved therapeutic index or a completely different activity spectrum from that of CDDP. An alternative approach to modifying the therapeutic index of CDDP analogs may be the use of drug carriers [5,6], among which liposomes are particularly attractive because they are essentially nontoxic

Correspondence: P. Vierling, Laboratoire de Chimie Mol~culaire, Unit~ de Recherche Associ~e au CNRS, Universit6 de Nice-Sophia Antipolis, B.P. 71, 06108 Nice Cedex 2, France.

biodegradable lipid vesicles that can alter the drug distributio,~ and bioavailability. Liposomes as drug carriers [7] have been exploited to improve the therapeutic index of several antimicrobiais an0 anticancer agents [6]. The liposome encapsulation of CDDP also offers therapeutic advantages in cases where the utility of the drug is limited by its toxicity [8-14]. However, formulation remains one of the major problems in the clinical development of liposome-incorporated drugs. Owing to its low water solubility and lipophilicity, CDDP is encapsulated into iiposomes with a very low efficiency (below 10%) and poor stability with respect to drug leakage [8-14]. To solve this problem, drug analogs have been developed with structural features that retain the desired antitumor effect and are more compatible with the drug carrier [15-19]. We have previously describe~ ,he synthesis and characterization [20] of new amphiphilic perfluoroalkylated bipyridine complexes of platinum or palladium (Fig. l). The amphiphilic character of these complexes arises from (i) the MCI 2 polar head coordinated to a disubstituted 2,2'-bipyridine moiety connected, in 4 and 4' through an ester (CO 2) or a methylene (CH2) junction, to (ii) a hydrophobic fragment of various

Fn -Em -X

Fn--Cm--X ~ N

i[

~1

t~N.

....cl

iCl

/M~cI Fn--

Fn --Em--X / v

Cm-X/~N M = Pt, Pd

X =-O-C(O)-, -CH2 -

FnCmMXand FnEmMX [ Fn = CnF2n+lcomplex series

~ C m = - (CHgm -

: n =4,6,8 m = 2, 5, I !

Em:-HC=CH-(CHz)m.2- m=5,ll

HCraMX and iiEmMX analogs

t(I. Cm Fn:= • H. (CH2)m" Em:

: m :¢ 13, 14 -HC:CH-(CH2)m.2 - : m : II

Fig, I, Structure of the pcrfluoroalkylatcd hipyridine PI and Pd complexes (FnCmMX and FnEmMX) and of their hydrocarbon analol~ (| !CmMX and l t E m M X ) ,

lengths consisting of a hydrocarbon-saturated (Cm) or unsaturated (Era) spacer, terminated by a highly perfluoroalkyl tail (Fn). Owing to their amphiphilic character, these complexes were susceptible to being incorporated not only into liposomes but also into injectable oxygen-delivering fluorocarbon emulsions. Such complex/emulsion formulations could thus act, at the same time, as an oxygen carrier and a drug delivery system, extending the potential scope of fluorocarbon emulsions to numerous other biomedical applications. The presence of perfluoroalkylated chains is effectively known to increase hydrophobicity and fluorophilicity and is, therefore, intended to facilitate the incorporation of the new perfluoroaikylated complexes, respactivcly, into liposomcs and into fluorocarbon emulsions. Furthermore, the incorporation of a drug into fluorocarbon emulsions is expected to combine the numerous advantages of drug encapsulation with the capacity of the fluorocarbon to deliver oxygen in radio[21,22] and theme-resistant [23] tumors, increasing the efficacy of radio- and chemotherapies. Such a synergic effect has been shown in the treatment of tumors with alkylating agent,,;, antimetabolitcs and antibiotics used in conjonction with fluorocarbon emulsions [24]. in addition, the use of fluorocarbon emulsions in therapy as drug delivery systems is particularly attractive, in view of their high intravascular persistence and tendency to concentrate around tumors [25]. Our main goal is to develop drug-loaded fluorocarbon emulsions, indeed, several 2,2'-bipyridyl metal complexes are found to be endowed with antimicrobial [26], antifungai [27] and antineoplasmic activity [28]. Our complexes, which are based on perfluoroalkylated bipyridine ligands, are thus susceptible of acting as potential antitumor agents and showing a synergic effect when used in combination with fluorocarbon emUl-

sions. Presently the most efficacious emulsions with respect to high O2-transport capability and shelf stability utilize egg yolk phospholipids (EYP) as emulsifier [29,30]. It was, therefore, of particular interest to investigate the incorporation of the perfluoroalkylated drugs into EYP vesicles and to control the characteristics and properties of the resulting perfluoroalkylated drug/liposome systems prior to studying the more complex drug/fluorocarbon/EYP systems themselves. This paper reports the formulation and incorporation efficiency of the new amphiphilic perfluoroalkylated bipyridine complexes of platinum or palladium (FnCmMX and FnEmMX) into EYP-based liposomes together with the stability of the resulting preparations (during sterilization and upon ageing). Besides these perfluoroalkylatcd complexes, we have investigated their hydrocarbon analogs (Fn = H in Fig. 1) in order to assess the impact of the Fn tails on these characteristics. Furthermore, the structural modular design of the complexes was destined to allow the assessment of structure/incorporation relationships: variable relative weights of the Fn perfluoroalkyi tail and Em or Cm hydrocarbon spacer ( F n / E m or F n / C m ratio), hence various fluorophilic/lipophilic balances, besides changes in chemical properties (nature of the central metal, M = Pt or Pd, anti of the connector, X = CH 2 or CO.,) were selected in order to evaluate the part played by these variables on the incorporation efficiency. The in vitro antitumoral activity of the resulting preparations will be presented elsewhere. Materials and Methods

Chemicals The egg yolk phospholipids (EYP) Lipoid E 80, from Lipoid KG (Ludwigshafen, Germany) contain phosphatidyi cholines (80 to 83% w/w), phosphatidyl ethanolamines (8 to 11%), lysophosphatidyl cholines ( < 3%), sphingomyeline ( < 3%), nonpolar lipids (< 5%) and cholesterol ( < 1%). The FnCmMX, FnEmMX, HCmMX and HEmMX platinum and palladium complexes used throughout this study have been prepared according to Ref. 2(I. Their elemental and spectroscopic (IH, 13C, I')F NMR and IR) analyses are in accordance with their chemical structure. Phosphate buffer saline (PBS) was purchased from Gibco (Paisley, UK).

Preparation of the liposomal-complexes Typically 6 ml liposomal-complex formulations are prepared as follows. Weighed amounts of dry EYP and the chosen complex were dissolved in chloroform; the organic solvent was then slowly evaporated under reduced pressure at 40-45°C using a rotary evaporator. The resulting film containing the lipids and the complex was flushed with argon and further dried under

43 high vacuum for 1 h. The dried film was then rehydrated under argon at room temperature with 6 ml of a PBS solution (pH; 7.4) for 30 min under agitation (Vortex). The resulting suspension was placed in a water bath at 10 or 40°C, then submitted to ultrasonic disruption using a Bioblock sonicator (600 W, 13 mm titanium probe, dial 7, 50% pulsed) for several periods of 3 min each, with a break of 1 min between each sonication period, to form liposomes containing the complex. All the resulting preparations were filtered under argon through a 0.22 /zm filter to remove the titanium particles from the sonication probe and the unincorporated complex (the analysis of the solid residue remaining on the filter showed also the presence of traces of phospholipids, less than 5% of the initial amount of EYP introduced in the preparation). The filtrate was then sterilized under the standard FDA norms (1210C, 15 rain, 1() "~ N m - ' ) . Formation of liposomes in the presence of the complexes was assessed by negative staining electron microscopy using the drop method for a 1/50 complex/EYP molar ratio formulation. Typically, a drop of the dispersions obtained as described above, was applied for 1 rain to the formvar-coated grid and drawn off with a filter paper. Then, a drop of 2% phosphotungstic acid was immediately applied for 1 min and drawn off. The grid was allowed to dry at 60°C. Examination of the preparations using a Philips microscope model CM 12 at 80 kV under magnification of 30000 to 60000, indicated that the dispersions consist effectively of iiposomes (results not shown). The amount of metal complex in the crude samples, and in the supernatants after the liposome complex suspensions were centrifuged at 105000 x g for 17 h at 4°C, was determined. Quantitative analysis of Pt was performed by atomic absoption spectrophotometry (AAS) using a Perkin-Elmer model 3030 Atomic Absorption spectrophotometer with background correction by the Zeeman effect for trace analysis. Pt samples were diluted 1/200 to 1/400 and the supernatants diluted (or not) 1/2 using a 0.2% HNO 3 solution containing 0.01% Triton XI00. The injected volume was 20 #!. A standard curve ((I.162, 0.325 and 0.65/zg Pt/ml) was automatically plotted by an auto-sampler using a sl~iked blank plasma specimen that had previously been diluted to 1/2 (0.65 /zg/ml). Analysis ineluded the following steps: drying at I I0°C for 40 s, ashing at 1400"C for 20 s and atomisation at 2650°C with a 3-s step flow, then tube cleaning at 2650°C for 4 s. Quantitative analysis of Pd was performed by AAS with a acetylene/air flame Atomic Absorption Spectrophotometer (PYE LINICAM SP9 Model) after 1/25 dilution with a 0.1% HNO 3 solution. The injected volume was 1 ml. A standard curve (0.2, 0.6, 1, 2, 5 and 10 mg Pd/ml) was previously measured using a diluted standard solution of PdCI 2 (10 mg/ml of Pd).

The entrapment yield (%) is calculated by the following formula: (amount of metal complex in the sample after preparation-amount of metal complex in the supernatant (which is negligible in most cases)/amount of metal complex added at the preparation) × 100. The efficiency of incorporation is defined as the entrapped complex ( × 10-3)/EYP molar ratio (an average molecular weight of 730 for EYP was used according to its composition). The chemical stability and the purity of the iiposome entrapped complexes was controlled after preparation and after heat sterilization by analytical TLC (precoated silicagel 60 F2s4, Merck, ultraviolet detection), tH NMR (Bruker AC200, CDCI.0 and IR (KBr pellets). In each case, the sample was ultracentrifuged and the supernatant separated and analyzed by AAS. Chloroform (10 rrll) was added to the residue to solubilize the complex and the lecithin. The organic layer was reduced and dried, the lecithin was then eliminated by washing with ethanol to leave the insoluble complex to be analysed (analysis by AAS of the ethanolic fraction showed the absence of any trace of metal). All the preparations were regularly checked for the presence of free drug and analyzed after a 3-month ageing period at room temperature, in the same way as just described, to assess the chemical stability of the liposome-incorporated complexes during storage. Particle size and size distribution measurements were performed by light scattering spectroscopy (LSS) on a Coulter Model N4SD Sub-Micron Particle Anaiyser after filtration, after heat sterilisation and upon ageing for up to 7 months at room temperature in order to check the morphological stability of the liposome-complex systems and the absence of complex precipitate. Results

Entrapment yield and efficiency of #icorporation All the amphiphilic perfluoroalkylated Pt and Pd derivatives presented here are soluble in halogenated organic solvents. They are neither soluble nor dispersible in water, even after ultrasonication, as shown by AAS measurements of their sonicated and then 0.22-~m filtered suspensions in water. However, when the dispersion was achieved from a mixed film formed by the complex and EYP, AAS, I H-NMR and IR analyses indicated the presence of the complex in the liposomal preparation. Examination of the EYP dispersions containing the complexes by negative staining electron microscopy showed them to, indeed, consist of liposomes. We investigated the evolution of the entrapment yield and of the efficiency of incorporation of the F8EI1PtCH 2 complex into EYP liposomes as a function of the complex/EYP molar ratio (from 1/20,

44 1/50, 1/100, 1/500 to 1 / l l 0 0 ) f o r a sonication time of 6 min and temperature of 40°C. In these preparations, the initial amount of complex was fixed in order to obtain liposomal formulations of 6. l0 -5 M complex concentration, in the case of 100% entrapment, and the amount of EYP was varied accort~ingly. Table I shows that the entrapment yield increases when lowering the complex/EYP molar ratio and reaches a value of 97% for a molar ratio of 1/1100. The incorporation efficiency is highest for an initial 1/20 to 1/50 complex/EYP molar ratio. The 1/20 complex/EYP molar ratio was then selected for the optimization of the sonication conditions to improve the incorporation of the complexes. These experiments were designed in order to obtain complex-liposomal preparations containing 2 . 1 0 -2 M of EYP and 10 -'~ M of complex in the case of 100% incorporation. Fig. 2 shows that the efficiency of incorporation of complexes FSEI 1PtCH 2 or F4E 11PtCH 2 into EYP liposomes is improved on raising the temper-

TABLE I

Entrapment yield, efficiency of incorporation and drag leakage of FSEIIPtCH, incorporated into EYP iiposomes, as a function of the FSEI IPtCH, / EYP molar ratio The samples were prepared and analyzed as described in Materials and Methods. Molar ratio" FSEI! PtCH 2 /EYP

Entrapment yield (%)

Incorporation efficiency (mmol of complex/mol of EYP)

Pt in the supernatant at the preparation (%) b

1/20 I/50 I / 100 I/200 l/5{X) ! / 1100

9 ~; 37 45 72

4.59 5,02 3,67 2.25 1,43 0.88

3.33 0.75 0.13

07

0.04

(}.3 -

" I F S E I I P t C H , , ] - 6 ' 10 -s M, ~' (Amount of metal in the supernatant after ultracentrifugation/ amount of metal entrapped into EYP liposomes)x I00.

TABLE !1

Entrapm¢,nt yidd, incorporation efficienty and complex leakage after preparation and after 3 mmzths of ageing at 25°C fi~r comph,xes hworporatt,d into EYP ;q osomt's The samples were prepared using the optimized conditions (sonication at 40°C and for 18 min) starting from a 1/20 (i.e., 10~ "~ M/2.10--" M) complex/EYP molar ratio and analyzed as described in Materials and Methods, Complexes

Fn/Em or F n / C m ratio R ( R = n/m)

Entrapment yield (¢/~,)"

Incorporation efficiency (mmol of complex/real of EYP)"

Metal in the supernatant (%)t, After After 3 months preparation ageing at 25°C

0,36 0,73 1,2 I.h

49 43 33 I!

245 21,5 16.5 5.5

I) I) 0 -

0.5 0.8 3.5 -

0 0

47 15

23 7,5

0.2 -

-

0.73 1.2

60 40

30 20

0 0.2

0.3 3.0

0,73 1,2 3

I0 15 14

5 8 7 0.5 11.3

!.3 0.3

F~EmPrCH :

F4E I I PtCH: FSE ! ! PtCH, F6ESPtCH, FSE~PICH,

Hydrocarb~m PtCtl : cmalog,~ HE I I PIC|!: HCI4PICH,

1.0

F,£mPt('O: FSE I I PtCO, F6ESPtCO:

FnCmi~CO: FSC! iPtCO, F6CSPtCO~ F6C2PtCO,

I~leocarbon PtCO : analogs HEI IPtCO, HCI3PtCY),

0 0

69 77

34 38.5

F4~ I I PdCII ~ FSEI IPdCH ~

0,3(~ 0,73

83 56

42 28

< 0.4 < 0.6

0,73 1,2

18 18

9 9

< 1.84 c < 1.84 ~

FnCmPdCO: FSCI IPdCO2 FbC5PdCO 2

~' Data are presented as mean values (S.D. = 15%) of several experiments (from four to eight). b (Amount of metal in the supernatant after ultracentrifugation/amount of metal entrapped into EYP liposomes)× 100. c The samples are not heat sterilized after preparation.

45 ~" 30

were then determined using the optimized conditions (sonication at 40°C and for 18 min) starting from a 1/20 complex/EYP molar ratio (to obtain complexiiposomal preparations containing 2 . 1 0 -2 M of EYP and 10 - 3 M of complex in the case of 100% incorporation) and the results are collected in Table I1. For each complex, several experiments have been performed and our results indicate that they are reproducible within 15%. These preparations were further used fol the study of their stability (see below). Owing to their common structural features (Table II), five series of perfluoroalkylated complexes (FnEmMCH 2 and FnCmMCO 2, with M = Pt, Pd, and FnEmPtCO,) and two series of hydrocarbon complexes (HEm- or HCmPtCH, and HCmPtCO 2) can be defined in order to establish structure/incorporation efficiency relationships. Our results are discussed in the next section considering the impact on the incorporation efficiency of the different structural features length and/or relative weight of the Fn tail and hydro-

18 min

,8

18 min

25

40°C

o.~

~ u 10 .~. o ,

F4EI IPtCH2

FSEI lPtCH2

Fig. 2. Efficiency of incorporation of F4E ! i PtCH, and F8EI 1PtCH, complexes into EYP liposomes for various durations and temperatures of sonication of the complex/EYP film (molar ratio of 1/20).

ature of sonication from 10 to 40°C and on increasing the sonication time to 18 min (a sonication duration exceeding 18 min has no further influence). The entrapment yield and the efficiency of in~,orporation of the vario~s complexes into EYP liposomes

TABLE Ill

Average partick, sizes and size distril)utions of the complex-lil)o~omal samples after: (i) filtration; (ii) heat sterilization and; (iii) after 7 months of

ageing at 25"C The data were determined by LSS on samples prepared using the optimized conditions (sonication at 4WC and for 18 min) starting from a 1/21) (i.e., I11-a M / 2 . 1 0 --~ M) complex/EYP molar ratio. Complcx-liposomal

Average Izarticle size (nm) and size distribution (%)

samples

After filtration

n m ( S D ") d

After filtration, and heat stel ilization

After filtration, heat sterilization. 7 months ageing at 25°C

(e~>)d

n m ( S D ") d

(%) d

,/ 60 (30) 270 (85) / 45 (15) 145 (20) /' 50(30) 230 (60) / 45 (15) 165 (40) 50 (30) 70 (30)

7q 21 88 12 80 20 69 31 I00 100

125 (50)

100

130 (6(I)

I00

125 (85)

100

I I0 (30)

I00

105 (40)

100

I00 (20)

100

85 (40)

I00

I O0 (30)

100

F6C2PtCO: F4EI IPdCH 2

85 (45) 45 (15)

l(10 86

100 (30) 214 (90)

100 I00

F8EI I PdCH 2

f 40 (10) i00 (25) 45 (15) 23O (30) f 50 (20) 280 (70) (35(15) 95 (lO)

52 48 77 23 80

214 (50)

100

9O (20) 2000 (500) 100 (35) I I0 (35) 95 (20)" 23(1 (70) 100 (30) 80 (20)" 2200 ((')50) 210 (70)

8,', i5 I00 100 80 20 I00 St) 20 I00

I00 (30)

10{1

F6E5PtCH2 | t E l IPtCH 2 F6E5PtCO 2

HEIIPtCO 2 FSC! I PtCO 2 F6C5 PtCO :,

F8C 11PdCO 2 FbC5PdCO 2 EYP

" t, c d

2(1 18 82

h

t,

t,

h

(75(30) 230 (50)

68 32

n m ( S D ") d

The preparations were stored for 7 months at 25°C without heat sterilization. Heat sterilization caused degradation of complexes FSCI I PdCO 2 and FOC5PdCO v Standard deviation in nm. In some cases and, for a given sample, two populations of liposomes were found; they are both indicated.

(e/~),~

6 carbon spacer (Cm or Era:~, hence fluorophilic/ lipophilic balance, nature of the hydrocarbon spacer, of the junction and of the metal. These results indeed show that the efficiency of incorporation is dependent upon the fluorophilic/lipophilic balance (Fn/Em or Fn/Cm ratio) and, most importantly, upon the saturated ( C m ) o r unsaturated (Em) nature of the hydrocarbon spacer,

amellar or multilamellar vesicles (100 to 280 nm). After heat sterilization, only a single population of uni- or multilamellar vesicles with an average size of 85 to 215 nm was measured. In most cases, no further evolution in size nor in size distribution even after 7 months of storage under argon and at 25°C was detected (in two cases only, a small population of large particles (2 ~tm) was measured).

Stability of the preparations Chetnical and structural stability of the entrapped complexes. Except the PdCO2 derivatives which are

Discussion

destroyed by the heat sterilization proc0durc, all the complexes remain chemically and structurally unchanged after their liposomal formulation, sterilization and upon ageing. This was assessed by analysis of the supernatant and of the residue that deposited after uitracentrifugation. As show~~ by AAS, the supernatants contain, when present, only traces of metal even after 3 months of storage at 25°C (Table Ii), which could be, due either to incomplete sedimentation of the vesicles or to degradation of the complex (the amount of metal found was even too low to allow the identification of the species present in the supernatant). TLC, nH NMR and IR analysis of the ultracentrifugation residues after removal of EYP (see Materials and Methods) showed them to consist of the pure non-degraded complexes. Stability of the complea'-entrapped liposomes. At any stage of the liposome complex system, no precipitate of free drug was seen. The particle size and size distribution and their evolution have been measured by LSS for several complex/liposome preparations (Table ill). When measured after sonication and filtration, they consist mainly of small unilamellar lilY,seines (particles < 100 nm) together with a population of larger unil-

50~

0

II ~0'~

FnEnd~H2

• I"1 F~.mPt¢202

\

li°[_ 10

0 0.0

0,5

1.0

1.5

2.0

2.5

3.0

R - Fn/C~mor Fn/Em Fig, 3. Evolution of the efficiency of ":ncorporation into EYP liposeines against the FNCm or F n / E m ratio R ( = n / m ) in analogous series of FnEmMX and FnCmMX comple~cs (Fn = H - or CnF,, ~l - . M = Pt or Pd, X = - C H 2 - or - O C ( O ) - .

The presence of perfluoroalkyl tails grafted onto hydrocarbon spacers confers hydrophobicity to the complexes investigated here. Thus, all complexes are neither soluble nor dispersible in water. However, in the presence of EYP, they are incorporated, in a reproducible way, into the EYP-based liposomes, whose formation has been confirmed by negative staining electron microscopy. Where the stability of the liposome-entrapped complexes is concerned, it should be pointed out that, except for the PdCO 2 complexes, which were destroyed by the heat sterilization procedure, all the complexes in this study retained their chemical and structural integrity when: (i) entrapped into the liposomes; (ii) sterilized under the FDA norms and; (iii) stored for 3 months at 25°C. This is attested by the spectroscopic analysis of the extract obtained as described in Materials and Methods. It is also noteworthy that all the complex/liposome systems are particularly stable with respect to drug leakage or degradation and to particle size and size distribution. Thus, even after heat sterilization and 7 months of storage at 25°C, no precipitate of free complex could be seen in the samples. Further, after 3 months of storage at 25°C, only traces of metal were detected in the supernatants obtained after ultracentrifugation. The morphological parameters of the liposomal complex preparations ale only slightly affected by the heat sterilization procedure, which induces a homogenisation of the distribution of vesicles accompanied by a small average size increase. No further evolution in size nor in size distribution was observed by LSS measurements after 7 months of storage at 25°C. it is established that hydrophobic or amphiphilic compounds are incorporated into the liposome lipidic bilayers. The entrapment yield and the efficiency of incorporation of such derivatives into liposomes is known to be particularly influenced by the hydrophobic and lipophilic nature of the compounds to be incorporated. The most original feature of the new Pt and Pd complexes in this study lies in the presence of perfluoroalkyl tails which confer fluorophilicity on these derivatives. Therefore, the Fn tails will modulate the

47 iipophilicity of the complexes and consequently their interactions with EYP and, most important, those which govern their incorporation into the lipidic area of the liposomes. The structural modular design of the complexes - variable perfluoroalkyl tail (Fn) and hydrocarbon spacer (Em or Cm) lengths, hence variable relative weight of the Fn and Em (or Cm) segments ( F n / E m or F n / C m ratio), hence variable fluorophilic/lipophilic balances - was destined to permit the assessment of the impact of these variables on the incorporation efficiency of the perfluoroalkylated complexes into phospholipidic membranes such as those forming lipotomes, but also those stabilizing injectable fluorocarbon emulsions. Fig. 3 illustrates the variation of the incorporation efficiency against the F n / E m or F n / C m ratio (R = n / m ) for a series of analogous complexes i.e., FnEmPtCH,, FnEmPdCH,, FnEmPtCO, and FnCmMCO., (M = Pt or Pd). The main feature shown by this figure is that the effect on the incorporation efficiency of introducing a perfluoroalkyt tail will essentially bc dependent upon the resulting R value, hence upon its relative weight in the whole molecule. Thus, in the FnEm complex series, their incorporation efficiency decreases slowly when the R value is increased from 0 (for Fn = H) to 0.73. It decreases more rapidly for higher values of R and then reaches a plateau for R >i 1.6. These results indicate that a high fluorophilic character (long Fn tail), when tempered by an equally lipophilic one (long hydrocarbon spacer), is not detrimental to the incorporation efficiency of a perfluoroalkylated drug into hydrocarbon vesicles. However, this is not the case for the FnCm complexes: the replacement of H by a perfluoroalkyl tail results here in a rapid decrease in their incorporation efficiency even for a low R value (R = 0.73) and the plateau is already reached for this low R value. Furthermore, for identical R ratios (R = 0.73 or 1.2), and as shown also in Fig. 4, the efficiency of incorporation

35

I

30

~'8 25 .~ ~ 20 °~ N 15 Q ~

F8CII

~

Saturated Unsaturated

,'8[

u

.o o 10 r,J ,,,~

0

R = 0.73

R = 1.2

Fig. 4. Comparison of the incorporation efficiency into EYP liposomes of the FnCmPtCO2 and FnEmPtCO, complexes differing in the nature of their hydrocarbon spacer - saturated (Cm) or unsaturated (Em) - and for a given R ratio.

of the FnCm complexes is much lower than that of their FnEm analogs. These results most likely indicate a pronounced effect of the presence of a double bond on the incorporation into liposomes of these two series oi complexes, whicl', differ or)ly in the presence of a double bond in the hydrophobic spacers for the FnEm series. It is also noticeable that the nature of the connector and of the metal influences the incorporation efficiency. Thus, in the FnEm complex series, and more particularly for lower R values, it appears that the incorporation is favored for a CO2 connector and for palladium rather than for a CH2 junction and for platinum, respectively. In conclusion, we have succeeded in entrapping perfluoroalkylated Pt and Pd complexes in remarkably stable liposomal preparations. As reported elsewhere, their in vitro antitumoral activity is most encouraging.

Acknowledgements We would like to thank ATTA, FEGEFLUC and the CNRS for financial support.

Ret~rences I Sherman, S.E. and Lippard, S.J. (198,7) Chem. Rev., 87, 11531181. 2 Barnard, C.F.J., Cleare, M.J. and Hydes, P.C. (1986) Chem. Britain, I(101-1004. 3 Dabrowiak, J.C. and Bradner, W.T. (19871 in Progress in Medicinal Chemistry (Ellis, G.P. md West, G.B., eds.), Vol. 24, pp. ! 29-158, Elsevier, Amsterdam. 4 Keppler, B.K., Berger, M.R., Klenner, T.It. and lteim, M.E. (19901 Adv. Drug Res., 19, 243-310. 5 Weinstein, J.N. and Lesennan, L.IS. (1984) Pharnlac. Thor., 24, 2117-233. 6 Mayhew, E. and Papahadjopouh)s, D. (19831 in Liposomcs (Ostr(), M.J., ud.), pp. 289-341, M. Dekker, New York. 7 Gregoriadis, G. (1988) in Liposomes as Drug Carriers, John Wiley & Sons, New York. 8 Yatvin, M.B., Muhlensiep~r), H., Porshen, W., Weinstein, J.N. and Feinendegen, L.E. (1981)Cancer Res., 41, 1602-1(')07. 9 Freise, J., Mueller, W.H., Magerstedt, P. and Schmoll, H.J. (1982) Arch. Int. Pharmacodyn. 258, 180-192. I(I Sur, B., Ray, R.R., Sur, P. and Roy, D.K. (19831 Oncology, 40, 372-376. 11 Steerenberg, P.A., Storm, G., de Groot, G., Claessen, A., Bergers, J.J., Franken, M.A.M., van Hoesel, Q.G.C.M., Wubs, K.L. and de Jong, W.H. (1988) Cancer Chemother. Pharmacol., 21, 299-307. 12 Kaledin, V.I., Matienko, N.A., Nikolin, V.P., Grutenko, Y.V. and Budker, V.G. (1981)J. Nail. Cancer Inst., 00, 881-1187. 13 Reszka, R., Fichtner, !., Nissen, E., Arndt, D. and Ladhoff, A.-M. (1987) J. Microencapsulation, 4, 2(11-212. 14 Steerenberg, P.A., Storm, G., de Grool, G., Bergers, J.J., Claessen, A. and de Jong, W.H. (19871 Int. J. Pharm., 40. 51-t12. 15 Khokhar, A.R., AI-Baker, S. and Perez-Soler, R. (19881 Anticancer Drug Design, 3, 177-184. 16 Perez-Soler, R., Lopez-Berenslein, G. and Khokhar, A.R. (1~189) in Liposomes in the Therapy of Infectious Diseases and Cancer

48 (Lopez-Berenstein. G. and Fidler, l.J, eds.), pp. 59-69, Liss A.R., New York. 17 Perez-Soler, R., Yang, L.Y., Drewinko, B., Lauterzstain, ,I. and Khokhar, A.R. (1988)Cancer Res., 48, 4509-4512. 18 Perez-Soler, R., Khokhar, A.R. and l_opez-Berenstein, G. (1988) in Liposomes as Drug Carriers (Gregoriadis, G., ed.), pp. 401-417, Wiley, New York. 19 Khokhar, R , AI,Baker, S., Brown, T. and Perez-Soler, R. (1991) J: Med: Chem., 34, 325-329. 20 Garelli, N. and Vierling, P, (1992) Inorg, Chim. Acta, in press. 21 Rockwell, S. (1985) Int. J. Radiation Oncology Biol. Phys., 11, 97-103. 22 Thomas, C,, Riess, J.G, and Guichard, M, (1991) Int, J. Radiat, Biol. 59, 433-445. 23 Teicher, B,A,, Herman, T,S,, Tanaka, J., Eder, J,P., Holden, S,A,, Bubl~y, O., Coleman, C.N, and Fr¢i !!! E, (1991)Cancer Res, 51, I086= 1091.

24 Teicher, B.A., Mclntosh-Lowe, N.L. and Rose, C.M. (1988) in Biomat. Art. Cells, Art. Org. (Chang, T.M.S. and Geyer, R., eds.), Voi. 16, pp. 533-546, M. Dekker, New York. 25 Mattrey, R.F. (1989) Am. J. Radiol. 152, 247-252. 26 Craciunescu, D., Scarcia, V., Furlani, A., Gbirvu, C., Ravalico, L. and Doadrio, A. (1987) Acta Pharmaceutica Fennica 96, 157-166. 27 Cristalli, G., Franchetti, P., Nasini, E., Vittori, S., Grifimtini, M,, Barzi, A., Lepri, E. and Ripa, S. (1988) Eur. J. Med. Chem. 23, 301-305. 28 Gill, D.S. (1984) in Platinum Coordination Complexes in Cancer Chemotherapy (Hacker, M,P, Douple, D.P. and Krakoff, H.I., eds,), pp. 267-27,';, Martinus Nijshoff, Boston. 29 Riess, J,G. (1991) Artif, Org, 15, 408-413, 30 Riess, J,G,, Dafors, J.L., Hanna, G.K,, Klein, D.H., Krafft, M.P., Pclura, TJ. and Schutt, E,G, Biomat, Art, Cells Immob, Biotech., ~ll press,