CISPLATINU COMPRISING PHARMACEUTICAL.
Summary of the invention The subject invention lies in the field of pharmaceuticals. In particular the invention is concerned with a novel pharmaceutical composition for treatment of neoplasms . The novel pharmaceutical is a complexed form of a component belonging to the family of cisplatinum compounds. Particularly suitable embodiments comprise a complex of 1,2- diacyl-sn-glycero-3-phosphatidylserine with cisdiammineplatinum(II) . The novel form should ensure reduced toxicity. In a preferred embodiment the pharmaceutical composition has a slow release form such as a micelle or liposome structure. The pharmaceutical composition according to the invention ensures a reduction of the severe side effects associated with cisplatinum treatment.
Background of the invention
Since the discovery of it's cytostatic activity in 1964 and I969 the compound cis-[PtCl2(NH3)2] which is clinically known as cisplatin has become a very succesful antitumour drug[I.l]. Cisplatin is regularly used for treatment of testicular and ovarian cancer. The survival rate for testicular cancer is over 95° thanks to this drug. It also finds application in treatment of other cancer forms like head and neck tumours, cervical cancer and bladder tumours [1.2]. Recently lung tumors have also been treated. Cisplatin is one of the top three selling antitumor agents world wide.
It is also known that cisplatin forms adducts with many biological molecules but strong evidence suggests its principal target is DNA [II.2]. It was determined in early mechanistic studies reviewed in [1.2 and 1.3] that cisplatin specifically binds to guanine N7 atoms of DNA. There is good evidence that after administration either by injection/infusion or orally that the drug circulates in the blood. Primarily this will occur as the chloride (for cisplatin) due to the rather high concentration of chloride ions in blood. Relatively slow hydrolysis occurs(tl/2 of a few hours at 37°C) upon passing through cell membranes of both normal and tumor cells either inside the cell or during passsage into the cell [III.11] followed by binding to nucleic acids and probably many other targets. It has been found with in vitro competition tests between DNA and proteins that the degree of binding to DNA (guanine
N7) and S donor atoms is determined by the degree of hydrolysis of cis- Pt. Although the major target seems to be nucleic acid many low molecular weight competitors for cisplatin-DNA reactions in the cell are known and intracellular reactions with peptides and proteins may take place. Examples are C1-, (HPO,,)2-, 0H-, H20, His, Met, Cys, gluthathione, metallothionein, ATP, amino acids and polyphosphates . The sites at proteins are believed to be the most likely origin of the several toxic side effects of cis-Pt and its several derivatives [III. .10 and 14]. Significant amounts of administered cis-Pt are lost due to binding to proteins .
The disadvantage of antitumor agents that target nucleic acid is that not only tumor cells but also other healthy cells that also replicate at a high rate are automatically targeted. Examples are the inner lining of the stomache thus causing nausea, the white and red blood cells thus causing anaemia and fatigue, liver and kidney cells thus causing liver damage and nephrotoxicity . It is often these side effects that restrict the physician with regard to administration of more of the often extremely reactive cytostatic compound. Toxic side effects of cisplatin are severe and have stimulated a lot of research into derivatives of cisplatin.
The toxic side effects have also resulted in the development of special protocols regarding drug dosage [I]. By slow administration of the drug e.g. via infusion lasting several hours rather than administration of a single bolus a high toxic peak concentration in blood can be avoided. In addition the use of rescue agents like sodium dithiocarbamate has been developed in order to counter toxicity by rescuing the cis-Pt bonds to proteins as described above. An example is thiourea [III.I5].
Derivatives belonging to the cisplatinum family of compounds that are capable of similar or improved biological activity are known. Up to now only 10-15 have reached clinical trials. The cis platinum family of compounds comprises compounds of the structural class cis- [PtX2(amine)2] , wherein X is the leaving anionic group and amine is any primary or secondary amine. It has been possible to formulate structure- activity relationships for platinum compounds. The cis-geometry of two amines and the presence of at least one N-H group on the amine as well as the leaving groups with a weaker trans effect than the amine have been found to be necessary. The amines can be symmetric, asymmetric, chelating or nonchelating. Water solubility should be good and toxic side effects
should be minimised. Possible reactions in the "blood with ligands containing S donor atoms should be suppressed [I].
New platinum complexes containing tissue specific carrier molecules as ligands for achieving higher concentrations have been described(I.7] • New platinum complexes attached to other chemotherapeutic agents like intercalators [1.8] or phosρhonocarboxylates[I.9] as coligands to obtain some sort of synergistic effect have been disclosed. Alternative compounds have more than one platinum atom connected by a bridge [I.10], contain radiosensitizers as ligands [I.11] and have protecting groups released by antibody linked enzymes only at the surface of the specific tumour cells [1.12].
The second generation platinum drug carboplatin, [Pt(C6H60z,) (NH3)2] has been developed and is in routine use. This compound has less toxic side effects than cisplatin. It has a lower reactivity which allows a higher dose to be delivered. The dosage of carboplatin can be up to 2000mg/dose[1.2] . The cisplatin dosage is usually around 100 mg/day for up to five consecutive days . The second generation drugs are generally referred to as CBDCA derivatives after the parent compound.
Beside the classical platinum compounds completely different structural classes of platinum compounds have also been developed. These include Pt(IV) and dinuclear compounds. In short a great amount of research in lots of different directions has been carried out around cisplatinum derivatives in order to improve the pharmaceutical. The references [III. 10,11,12 and 13] provide examples of the cisplatinum compounds known to date and are incorporated herein by reference.
In order to reduce development of resistance to the cytostatic further derivatives, the socalled third generation derivatives have evolved. These include platinum(IV) derivatives that can be administered orally. An example of a third generation drug [1.5] is [Pt(Cl2(RC02)2(amine)2] , wherein R=alkyl. Treatment to try and prevent resistance occurring comprises administering a number of chemotherapeutic agents rather than one and applying a varied scheme of administration. In particular upon treatment of ovarian cancer a major limitation of cisplatin is acquired resistance. The dose escalation required to overcome even a small increase in cellular resistance can cause severe toxicity.
In summary as regrettably for the past 25 years with perhaps the exception of taxol no new strongly improved antitumor agents have been developed a lot of time and effort has gone into improving existing
therapies and administration methods. A lot of research into derivatives of cisplatin has been aimed at either reducing toxicity and/or circumventing resistance.
The subject invention is directed at formulating novel and improved cisplatinum derivatives. The improved pharmaceutical composition comprises a stable cis-platinum derivative in an inert form in the bloodstream. The novel composition thus exhibits reduced toxicity in the bloodstream. Additionally, loss of active compound due to binding of phosphatidylserine in vivo which can occur with the prior art compounds can now also be prevented. Furthermore upon entry into the cell the complexed form can be released thereby providing a cisplatinum derivative capable of becoming the most active form of the cytostatic. The Cl- concentration and pH and temperature in the bloodstream provide an environment ensuring stable binding of the cis-platinum compound such that it does not occur in the hydrolysed active form in the bloodstream to the degree the corresponding non complexed compound does. The Cl- concentration and/or presence of other components in the cell known to compete for cisplatinum binding provide the circumstances required for the cis-platinum derivative to be released in it's active form. The complex preferably releases cis-platinol which is considered to be the pharmacologically most active form for treating neoplasms directly in the cell. This form binds to nucleic acid and exerts it's activity. The pharmaceutical composition thus is present in an inactive form in the bloodstream thereby avoiding the toxicity effects due to the binding to phosphatidylserine. In addition the toxicity effects due to the presence of cisplatinol in the bloodstream common to the prior art compounds can also be avoided. In particular the complexed form can slowly release the active form in the bloodstream thereby avoiding plasma peaks of the extremely toxic cisplatinum compound in a form other than the complexed form according to the invention. Further advantage is provided by quick release of the pharmaceutically active cisplatinol becoming possible in a cell which is ultimately the target of the cytostatic compound. An essential, aspect of the invention is the reversibility of the complex formation. The cisplatinum derivative must be releasable from the anionic organic binding partner.
The pharmaceutical composition according to the invention is a pharmaceutical composition for use in treatment of a subject having neoplasms with a view to reducing size and/or number of neoplasms, said pharmaceutical composition comprising a complex of a cytostatic compound
of the cis platinum family and an anionic organic compound as complexed form of the active pharmaceutical component, said complex having the following general formula 1
wherein
Q = -CO- or -P0(0~)-;
Y = -CHR'-, -CO-, -P0(0")- or a direct bond;
Zl, Z2 = independently 0 or NR' ;
R1,R2=H or a hydrocarbon group optionally substituted, optionally linked to each other or to R3 and/or R4 wherein R3, 4=H or hydrocarbon group optionally substituted, optionally linked to each other or to Rl and/or R2, wherein any of the nitrogen atoms can be protonated or quaternised and wherein
R'=H, or a hydrocarbon group optionally substituted, and wherein
R"= a hydrocarbon group substituted with a group capable of liposome formation.
Liposome technology has been available for a number of years and has been the subject of enormous amounts of research. The cosmetic industry and also the pharmaceutical world have produced a number of useful products with this technology. Vaccines and other medicaments have been introduced into liposomes. Upon formation of a liposome with a correct composition the liposome is degraded slowly and the medicament is released after a delay. To date the only antitumor agent commercially available in a liposome is doxorubicin. The dosage can be ten times higher than without the liposome. Experiments are ongoing with
daunorubicin a related compound of doxorubicin. A vinca alkaloid has also been described as being present in a liposome. Generally speaking there are two options when forming liposomes capable of carrying medicaments. First the construction of a multilamellar bilayer vesicle with water comprised between the layers. The medicament is comprised within this water fraction. It is not linked to the membrane structure of the liposome but is captured in the water layer. Alternatively it has been possible to incorporate compounds with hydrophobic groups within the membrane structure of the liposome or attached to the membrane layer. European patent application EP-A-0.664.116 provides a method for producing liposomes wherein lecithin is used as a cell component. Lecithin consists of phosphatidylcholine and does not comprise phosphatidylserine. This publication suggests using such liposomes for delivering a large number of compounds. One of the compounds mentioned is cisplatin. However no illustration of this is provided. In fact such an embodiment as suggested would not be feasible. The cisplatin is a small hydrophilic molecule which is not expected to remain trapped in such a liposome structure as taught in the document for a sufficient length of time to provide a suitable slow release medicament. It is possible in alternative embodiments of the invention to provide unilamellar vesicles or multilamellar vesicles. Both embodiments are useful versions of a pharmaceutical composition according to the invention. It is also possible to produce micelles comprising the complex according to the invention thus providing a simple form of cytostatic exhibiting reduced toxicity. The substituent R" thus may have many embodiments which are obvious to a person skilled in the art of liposome formation or micelle formation. The requirements for the membrane components to produce a liposome suitable for application to a mammal are well known to a person skilled in the art as they are also for micelle formation. Such general information is readily available in general handbooks and text books on liposomes .
The pharmaceutical composition according to the invention suitably comprises as substituents R1.R2.R3 and R4 substituents common to what are generally known in the art as belonging to the cisplatinum family of compounds. These have been described in the introductory part of the description and are all incorporated herein by reference. In particular the first and second generation derivatives are suitable. Preferable compounds comprise cisplatinum substituents in combinations already in use clinically. The most preferred embodiment comprises the
compound most widely used as cytostatic i.e. wherein the cytostatic compound is a cis diammineplatinum(II) derivative i.e. NR1R2=NR3R4=NH3.
A suitable embodiment of a pharmaceutical composition according to invention comprises as anionic organic compound the structure of formula 1, wherein R" = -A-0-S02-0-R" ' or preferably R" = -A-0-P02-0-R" ' wherein A=CnH2n, n=l-4, optionally substituted with hydroxy or phenyl; R"'=H, or a hydrocarbon group such as alkyl, alkenyl, optionally substituted by amino, hydroxy, acylamino and/or acyloxy. The R" ' can comprise one or more hydrocarbon group chains . The hydrocarbon group chains can be substituted or non substituted. It can comprise one or more alkyl, alkenyl, amino, hydroxy, acylamino and/or acyloxy chains or a combination thereof. It is e.g. possible to use a mono- or diacyloxyalkyl compound as R" ' . A suitable pharmaceutical composition according to the invention is one as defined above wherein R'" comprises an acylgroup of 10-26 carbon atoms in length. For micelle formation a length of 10-15 will be suitable. For liposome formation longer chains are generally required in order to provide the desired hydrophobicity . Chain lengths of 12-24, even 14-24 are suitable. Saturated or non saturated fatty acids can be used in both instances. Cis or trans fatty acids can be applied. As stated above this is common to liposome or micelle technology. With regard to chain lengths this counts also in general terms to R" and R" ' , not only for fatty acids as embodiment of R" ' . Such chains can be alkyl, alkenyl, optionally substituted by amino, hydroxy and/or acyloxy or a mixture thereof. Such hydrophobic chains may comprise one or more fatty acid chains. These can vary in length or be of the same length. In a suitable embodiment naturally occurring fatty acids can be used. In general fatty acids with a chain length of 10-24 are suitable. A list of suitable chains is for example available in a general text book on fat constitution e.g. "The chemical constitution of natural fats" by T.P. Hilditch 2nd ed. 1949 page 6 and f . The list of fatty acids listed there is hereby incorporated by reference. Other non natural derivatives are also suitable and will be apparent to a person skilled in the art. Oleoyl and myristoyl chains are particularly suited for example.
A pharmaceutical composition according to the invention wherein R"'= mono or diacylglyceryl forms a suitable embodiment of the invention. A glycerol derivative is an extremely suitable example of R" ' . Alternatively a more complex alcohol derivative can be applied e.g. sphingosine. Other alternatives will be apparent to a person skilled in the art. Shingosine is particularly suited as it forms lipids in nature.
A pharmaceutical composition according to the invention preferably comprises a phosphate group in the anionic organic compound. A particularly suitable embodiment comprises a phosphatide as R" . Alternatively the anionic organic compound can comprise a sulphate group. Such is particularly suitable in the embodiment disclosed above with A and R" ' . A pharmaceutical composition according to any of the preceding embodiments , wherein the anionic organic compound comprises a phospholipid forms a preferred embodiment of the invention. In particular a pharmaceutical composition according to any of the preceding embodiments, wherein the anionic organic compound is a phosphoglyceride is suitable. A good example of a compound according to the invention ccomprises as anionic organic compound a 1 , 2-diacyl-sn-glycero-3_ phosphate .
Examples of the anionic organic complexant are amino acid residues (-0C0-CHR"-NR' -) , malonic acid derivatives (-0C0-CHR"-C00-) , α- phosphonocarboxylic acid derivatives (-0C0-CHR"-P0(0~) -0-) and the corresponding amides .
A pharmaceutical composition wherein the anionic organic compound comprises a phosphorylated tyrosine, serine, homoserine or threonine structure is a preferred embodiment of the invention. The presence of the structures corresponding to other amino acids either natural or derived i.e. comprising an amino and carboxyl group also falls within the scope of the invention. A pharmaceutical composition according to the invention wherein the anionic organic compound is derived from serine (A = CH2) , forms a preferred embodiment. In particular in the embodiment disclosed above with A and R" ' , wherein the amino acids are comprised in the structure to the left of the phosphate group form suitable embodiments as does a pharmaceutical composition according to any of the preceding embodiments mutatis mutandis wherein A=CH2.
A pharmaceutical composition according to the invention preferably releases the active cytostatic compound upon contact with gluthathione . In general it is preferable that the complex is more stable in the bloodstream than in the cell. A pharmaceutical composition according to the invention comprises the complex in a pharmaceutically acceptable dosage form. Such dosage forms are immediately apparent to a person skilled in the art. The dosage form will depend on the patient, the severity of the malignancy, the dosage regime selected by the physician and can have
numerous embodiments as is well known in the art. A number are provided here. A pharmaceutically acceptable dosage form can be selected from the group comprising tablets, lozenges etc, ingestible liquids, injectable solutions or freezedried products to be solubilised that can be administered via infusion i.a. The amount of active compound in a single bolus dosage form according to the invention can surpass that common to the prior art compositions comprising the corresponding cisplatinum derivative, due to the reduced toxicity of the subject compositions vis a vis the prior art compositions. A pharmaceutical composition must be sterile. It is preferably isotonic upon administration and should be free of undesirable compounds e.g. toxic compounds other than a cytostatic compound and impurities.
Two documents describe solutions of phosphatidylserine and cisplatin as aquated and/or chloride species. These documents do not deal however with production of a pharmaceutical composition according to the invention. In fact they do not disclose pharmaceutical compositions as such merely solutions with these two components. The solutions do not form compositions even inherently suitable as pharmaceutical compositions due to he presence of undesirable additives such as EGTA and absence of isotonic and sterile conditions.
The article by Taylor et al (IV) is directed at formation of complexes between these components in a gel like environment. It is stated that the interaction between DMPS and DDP seems to be very slow and difficult whereas cellular uptake of DDP seems to be quite fast. Also both DDP-OH and DDP-C1 appear to be bound. They indicate the phosphate group as being a binding partner. It is also suggested the aquated species binds to a greater extent than the Cl species due to interaction with the phosphate group. This suggests that DDP-OH would be stably bound to phosphatidyl serine in the cell. The cited article would thus seem to preclude application of such a complex for treatment requiring release of cisplatinumcompound in the cell. According to the cited article a person skilled in the art would assume DDP-OH level in the cell to be lower than DDP-C1 due to more stable binding. It also appears to tech that presence of a CP-PS complex of whatever type is not a likely event. Nothing is taught about any such complex being a suitable adminstration form of a cytostatic or suggests any advantages in this direction that could be used by a person skilled in the art faced with the problem of improved dosage forms of cisplatinum compounds.
The second article is written by the inventors (V) and
describes the phenomenon that divalent ions electrostatically interact with various negatively charged membrane components i.a. phosphatidyl serine. They also disclose this phenomenon for phosphatidic acid, cardiolipin, phosphatidylinositol and phosphatidylglycerol . Nothing is taught or suggested about specific complex formation with cisplatin derivatives of any type certainly not of the type according to the invention with all the inherent advantages thereof resolving a long felt need.
A method of treating a subject having neoplasms with a compound of the cisplatinum family of cytostatic compounds said method comprising application of a pharmaceutical composition in any embodiment described above as the invention in a pharmaceutically acceptable dosage and form to said subject in order to reduce the number and/or size of neoplasms also falls within the scope of the invention. The neoplasms that can be treated are those which are already known in the prior art susceptible to treatment with cis platinum derivatives. Other forms may also be treated due to the higher dosages now possible in a more effective manner. A method according to the invention can reduce the number and/or size of neoplasms with reduced side effects related to treatment with said cytostatic compound were it to be applied as a direct release pharmaceutical composition per se. In particular a method according to the invention can reduce the number and/or size of neoplasms with reduced nephrotoxicity related to treatment with said cytostatic compound were it to be applied as a direct release pharmaceutical composition per se. The invention also comprises use of a complex defined in the general formula of claim 1 in any of the embodiments of the subsequent claims as active component for preparing a pharmaceutical composition suitable for treating a subject having neoplasms with a view to reducing size and/or number of neoplasms, said pharmaceutical composition exhibiting reduced side effects in comparison to the cytostatic compound provided by said complex were said cytostatic compound to be applied as a direct release composition per se.
The following examples elucidate the invention but are not considered to restrict the scope of the invention. The content of the listed references is deemed to be incorporated by reference.
EXAMPLE
Cisplatin complexes with phosphatidylserine in membranes*
INTRODUCTION
cis-Diamminedichloroplatinum(II) (cisplatin)1 is a commonly used anti-cancer drug. It is a reactive compound which can occur in various species in aqueous solution depending on the pH and chloride ion concentration (1-4). It is well established that cisplatin interacts with DNA and thereby causes inhibition of DNA synthesis which might be a primary therapeutic action (5~7)- In addition the molecule interacts with proteins (8, 9). Whether cisplatin has an affinity for cellular lipids is not known. Knowledge about cisplatin-lipid interactions might contribute to a better understanding of the antitumor activity of cisplatin but might also shed light on the severe side-effects of the drug of which nephrotoxicity is the most serious (4, 10, 11). A relationship between nephrotoxicity and interaction of positively charged aminoglucosides with negatively charged phospholipids (12, 13) has been suggested.
We recently reported the binding of cisplatin to model membranes of varying phospholipid composition (14). It was observed that the positively charged aquated species of cisplatin binds to different anionic lipids which can result in changes in lipid packing. In the course of these investigations it was noticed that cisplatin specifically shows a strong interaction with phosphatidylserine (PS) (14). Also in another recent study a binding of cisplatin to PS was observed (15). In the present study the formation of a stable complex with PS is described. It is shown that both in model and in biological membranes cisplatin coordinatess specifically with PS involving a loss of two chloride ions (or in case of aquated cisPt two water molecules) and coordination of platinum to the amine and carboxylate group of the serine moiety. Because PS is involved in many cellular functions such as transmembrane signaling (16, 17) this implies that cisplatin-PS complex formation quite likely interferes with these functions.
EXPERIMENTAL PROCEDURES
Materials - Cisplatin was purchased from Sigma (St. Louis, MO,
USA). Phospholipids were purchased from Avanti Polar Lipids, Inc. (Birmingham, AL, USA) and contained two oleoyl fatty acyl chains, except cardiolipin, derived from bovine heart, sphingomyelin, derived from egg, and phosphatidylinositol, derived from soybean.
Cisplatin incubations with phospholipids dispersions - Lipid dispersions (2 mM phospholipid) were prepared by adding buffer containing no or 5 niM cisplatin to a dry lipid film followed by 10 times freeze- thawing and agitation with a Vortex mixer. In all cases cisplatin solutions were always freshly prepared by dissolving the compound in the appropriate buffer which was facilitated by warming to 40-50 °C. Buffers employed were 10 mM Pipes, 50 mM Na2S0ή, 1 mM egta, pH 7.4; 10 mM Mes, 50 mM Na2S0^, 1 mM egta, pH 6.0; 10 mM Pipes, 100 mM NaCl, 1 mM egta, pH 7.4; and 10 mM Mes, 100 mM NaCl, 1 mM egta, pH 6.0. The samples were incubated in the dark at room temperature for the indicated times after which the lipids were extracted (18) and analyzed by TLC on silicagel eluting with chloroform/methanol/acetic acid (7/2/1, v/v) , chloroform /methanol /ammonium/water (34/14/1/1, v/v), or chloroform/methanol/water (35/13/2 v/v). Lipids were identified by spraying with either 10$ H2S0/, followed by charring, phosphorus or ninhydrin spray reagents. For quantitative analysis lipid spots were scrapped off after which the material was destructed by perchloric acid and the amount of phosphorus was quantified (19).
Cisplatin incubations with red blood ce lls ghost - Human red blood cells ghost isolated according to (20). 5 mM cisplatin (final concentration) was added to the membranes (2 mM phospholipid-Pi) and incubation was performed at 37 °C, in 10 mM Mes, 0 mM Na2Sθ , 1 mM egta, pH 6.0. Since these ghosts are open structures, cisplatin is able to reach both the inner and outer leaflet of the membrane. Samples were taken at different time points and phospholipids were extracted according to (18). TLC experiments were performed in 1 dimension with chloroform/methanol/acetic acid (7/2/1, v/v) as eluens and sprayed with ninhydrin or in 2 dimensions with in the first dimension chloroform/methanol/ammonium/water (34/14/1/1, v/v) and in the second dimension chloroform/methanol/acetic acid (7/2/1, v/v) after which lipids
were identified by spraying with phosphorus spray reagent.
Mass spectvometry - Cisplatin was incubated with 1 ,2-dioleoyl- src-glycero-phosphoserine (DOPS) as described above after which the lipids were extracted (18). The chloroform phase was used to analyze the product. Positive FAB mass spectra and CID MS-MS spectra were obtained using a Jeol JMS-SX102/102A four sector instrument. For the FAB mass spectra NBA was used as a matrix. The MS-MS spectra were acquired with sufficient resolution of MSI using nitrogen as collision gas. The gun was operated at 6 kV ad at mA current . The sample was dissolved in CHC13/CH30H and added to the matrix. The mass spectrum shows a cluster of protonated molecular ions at m/z 1014, 1015. and 1016. The intensity ratio and the mass value are in agreement with the elemental formula of the proposed Pt complexes. From the ions mentioned above, representing the most abundant Pt isotopes, MS-MS spectra were recorded.
NMR spectroscopy- Cisplatin was incubated with DOPS after which the lipids were extracted as described above. 1H, 13C, 31P, and 195Pt
NMR spectra were recorded in saturated solutions of CDC13 (:H, 13C, and 31P) and in CDCl3/Me0H 9:1 (13C, 31P and 195Pt) ; spectra were referenced to
TMS, TMP (trimethyl phosphate) and K2PtCl,, (-1600 ppm) , respectively.
RESULTS AND DISCUSSION
Cisplatin spec ifical ly complexes wi th phosphatidylserine in mode l membranes - When DOPS is dispersed in a chloride-ion free buffer of pH 6.0 (10 mM MES, 50 mM Na2SO4, 1 mM egta) containing 5 mM cisplatin and incubated at room temperature for 3 days a new species (X) can be observed by TLC (Fig. 1A) . X has a lower Rr value then DOPS in an acidic eluens, it is phosphorus positive and in contrast to DOPS it does not stain purple (but brownish) when stained with the amino-group-specific ninhydrin reagent (data not shown) . Dispersions of DOPE, D0PC and sphingomyelin, representatives of the other major mammalian plasma membrane phospholipids , do not show the appearance of new reaction products upon incubation with cisplatin under these conditions (Fig. 1A) . This is also not observed for a D0PE/D0PC (1:1, molar) mixed dispersion (data not shown) which is organized in a lamellar phase at room temperature (pure DOPE forms a Hu phase [21]). Furthermore, new reaction products are also absent when the negatively charged phospholipids
cardiolipin, phosphatidylglycerol and phosphatidylinositol are incubated with cisplatin (data not shown) .
Formation of X at room temperature is a slow process with an estimated half-time of 70 hours (Fig. IB) and is paralleled by a reduction in free DOPS suggesting that X is the product of the DOPS - cisplatin interaction.
The most reactive species of cisplatin is not the native drug but the positively charged, aquated species (4) . When cisplatin becomes in contact with water, chloride ions are stepwise replaced by water molecules resulting in mono- [ (NH3)2PtCl(H20) +] and diaquated [ (NH3)2Pt(H20)2 2+] species. This process is inhibited by Cl". The mono- and diaquated species are weak acids with pKa's of 6.4l and 5-37. respectively (1, 2). To test which species is responsible for the reaction with DOPS the lipid was dispersed and incubated at 37 °C in buffers with no or 100 mM NaCl and at pH values of 6.0 and 7.4. Because the properties of anionic lipids like PS depend on the nature and amount of cations in the buffer the chloride free buffer was supplemented with 50 mM Na2S0/,. After certain time intervals the samples were analyzed by TLC and the percentage of conversion of PS was quantified (Fig. 2) . Comparing the incubation of [Cl-] = 0, pH 6.0 in Fig. IB (room temperature incubation) with that in Fig. 2B immediately reveals that at the clinically relevant temperature of 37 °C the conversion of PS is greatly increased, with a half-time of the reaction of 5 hours. In the absence of Cl" the reaction rate is slightly lower at the higher pH value. Furthermore it is clear from Fig. 2 that also in the Cl" containing buffers the reaction occurs to a significant extent, but at an 20- to 30-fold reduced rate for the two pH values tested.
Structure of the cisplatin-phosphatidylserine complex - The lower RF value in an acidic eluens (Fig. 1A) and higher RF in an alkalic or neutral eluens (data not shown) indicate that the complex is less negatively charged than DOPS and probably neutral . In order to examine the structure of the cisPt - DOPS complex mass spectrometry was performed. The mass spectrum shows a (M+H)+ cluster around m/z 1015. (Fig- 3) • Based on the intensity ratio and the m/z values of the ions in this cluster the molecule must contain platinum. A calculated isotope distribution of a molecule containing 42 C, 83 H 3 N 100τ 1 P, and 1 Pt is inserted in Fig. 3-
The (M+H)+ ions at m/z 1014, 1015, and 1016 were subjected to
MS-MS. In the MS-MS spectra two dominant fragments ion clusters are observed m/z 412, 413. and 4l4, and m/z 314, 31 , and 316 (Fig. 4). In the product ion spectrum of m/z 1014 only m/z 4l2 and m/z 31 are observed (Fig. 4A) . In the MS-MS spectrum of m/z 1015 the peak intensity ratios I(m/z 4l2)/I(m/z 413) and I(m/z 3l4)/I(m/z 315) correspond to the ratio between the 13C and 195Pt contribution (35:65) in the selected precursor ion (Fig. 4B) . Three major components contribute to the m/z 1016 ion, namely: (13C)2 19"Pt, 13C195Pt, and 196Pt in an intensity ratio of approximately 10:35:55- In the product ion spectrum of m/z 1016 we observed again two fragment ion clusters each containing 3 peaks with relative intensities corresponding to the component distribution in the precursor ion. These observations allow to conclude that the fragment ions mentioned above all contain a platinum atom.
The existence of these fragments can only be explained if the cisPt - PS complex has a structure as shown in Fig. 5A, in which the platinum is coordinated to the carboxylate and amine group of the serine moiety. This molecule has a nominal molecular weight of 1015 dalton in its protonated form, and upon fragmentation of this complex by MS-MS platinum containing fragment ions of m/z 4l2 and m/z 31 dalton are obtained. The former fragment indicates that platinum is not associated with the fatty acyl chain part or the glycerol backbone but with the headgroup (Fig. 5B) . The latter fragment can be explained by a loss of H3P0/t from the serine moiety (Fig. 50) . The structure shown in Fig. A is at a physiologically relevant pH zwitterionic. From 31P spectroscopy a shift observed at O.67 agrees with phosphate groups not coordinated to platinum. The signal observed in the 195Pt spectrum at -2159 PPm is in excellent agreement with a chromophore [PtN30]; although theoretically also unreacted cisplatin could be in this range. As a reaction has occurred because the 13C signal at 184.6 ppm agrees with platinated carboxylate group, it can be concluded that all NMR data agree with the structure proposed in Fig. A.
Interaction of cisplatin with phosphatidylserine in erythrocyte membranes - Subsequently, it was examined whether the formation of cisplatin-phosphatidylserine coordination complex also occurs in biological membranes, where mixtures of phospholipids and proteins are present. Human erythrocyte ghosts were used to study the interaction of cisplatin. A two-dimensional TLC was performed in order to get a good separation of the phospholipid mixture. Fig. 6A demonstrates that in the
presence of cisplatin the cisplatin-PS product appears next to the PS spot with a higher RF value in the first dimension (alkalic eluens) and a lower RF value in the secon dimension (acid eluens). From Fig. 6B it is clear that after incubation with cisplatin the amount of PS is significantly reduced in a time-dependent manner, whereas the amount of phosphatidylethanolamine remains constant. Furthermore, in control experiments without cisplatin present during incubation the amount of PS remains constant in time. Therefore, the interaction of cisplatin with PS not only takes place in model membranes , but also in biological membranes .
PS is exclusively located in the inner leaflet of the plasma membrane and contributes 20 % to the phospholipids present in this leaflet (22) . At this side of the membrane the pH and Cl" concentration are lower than that of the extracellular compartment, i.e. the highest concentration of aquated, reactive cisplatin is formed at the side were PS is located. PS plays a crucial role in cellular processes such as signal transduction (16) , cell proliferation and apoptosis (23), and the blood clotting cascade (24) the biogenesis of mitochondria (275. and is necessary for the activity of enzymes such as Na*/K+-ATPase (26) and protein kinase C (16) . For the latter enzyme the carboxyl acid moiety of PS is involved (16, 17). which is now obscured by the interaction with cisplatin. Indeed, effects of cisplatin on the activity of this enzyme have been reported (28, 29) . It is therefore highly interesting to study the effect of the cisplatin -PS interaction on PS-dependent processes, and its consequences for the mechanism of cisplatin toxicity and side- effects, such as nephrotoxicity.
REFERENCES
Document I : Reedijk J. Chem. Commun, 1996, pp. 801-806
"Improved understanding in platinum chemistry" Document II: Zamble, Deborah B. , and Lippard, Stephen J., TIBS 20,Oct.
1995 "Cisplatin and DNA repair in cancer chemotherapy" Document III: Reedijk J. Inorganica Chimica Acta 198-200 (1992 873-881 "The relevance of hydrogen bonding in the machanism of action of platinum antitumor compounds"
I. Reedijk J. Chem. Commun, 1996, pp. 801-806
"Improved understanding in platinum chemistry"
1. Reedijk J. Inorg. Chim. Acta 1992, 198-200, 873
2. Barnard, C.F.J., Plat.Met. Rev., 1989, 33. 162; Semtin, Oncol., 1992, 19. 720.
3. Bloemink M.J. and Reedijk J., in Metal ions in biological systems, ed. Sigel H. and Sigel A.,
Dekker M. , New York, 1996, vol. 32, pp. 641-685. 5. Giandomenico CM. et al . , Inorg. Chem. 1995, 34, 1015 7. Bloemink M.J. et al . , Inorg. Chem. 1994, 33, H27
8. Bowler B.E. et al., J.A. Chem. Soc, 1989, HI, 1299
10. Farrell N, Comm. Inorg. Chem. 1995. 16, 373
11. Farrell N and Skov K.A. , J. Chem. Soc. Chem. Commun. 1987, 1043
12. Hanassian S and Wang S, Can. J. Chem. 1993, 71 , 897
II.
2. Rosenberg B. (1985) Cancer 55, 2303-2316
III. 2. Rosenberg B. , L. Van Camp, Grimley E.B. and Thomson A.J, J. Biol . Chem. 242 (I967) 13 7- 5. Lempers E.L.M. and Reedijk J., Adv. Inorg. Chem.
37 (199D 175-217. 10. Van der Vijgh, Clin. Pharmacokin, 21 (1191) 242. 11. Miller S.E. and House D.A. , Inorg. Chim. Acta, 166 (I989), 189; 173 (1990) 53
12. Kralingen C.G et al . , Inorg. Chem. 19 U980) 1481
13. Hollis S.L. et al . , J.Med. Chem. 32 (I989) 128; Farrell N. et al., J. Med. Chem., 33 (I990) 2179-2184; Kδpf-Maier P and Kopf
H. Naturwissenschaften, 73 (1986) 239, Pasini A et al. Angew. Chem. 26 (I987) 615; Keppler B.K. New J. Chem., 14 (1990) 389- 403 14. Corden, Inorg. Chim. Acta, 137 (1987) 12 - 15. Bodenner, D.L. , Dedon, P.C., Keng P.C. and Borch R.F., Cancer, Res. 46 (1986) 2745; Borch R.F. and Markman, Pharmacol. Ther. 41 (1989) 371. From the Example
1. Berners-Price, S. J., Frenkiel, T. A., Frey, U., Ranford, J. D., and Sadler, P. J. J. (1992) J. Soc . Chem. Commun. , 789-791 2. Sadler, P. J. J. (1992) J. Soc . Chem. Commun. , 7δ9~791
Miller, S. E. , and House, D. A. (1991) Inorg. Chim. Acta I87, 125- 132 3- Reedijk, J. (1992) Inorg. Chim. Acta 198-200, 873-882
4. Reedijk, J. (1996) J. Chem. Soc;Chem. Comm . 801-806
5. Garner, M. , Green, M., Boogaard, N. , and Reedijk, J. (1993) J- Chem. Soc , Chem. Commun. , 290-291
6. Bloemink, M. J., and Reedijk, J. (1996) Metal ions in bio logical systems 32 (Sigel, H. , and Sigel, A. eds) pp. 641-685. M. Dekker,
New York
7. Yang, D., Van Boom, S. S. G. E. , Reedijk, J., Van Boom, J. H. , Farell, N., and Wang, A. H.-J. (1995) Nat . Struct. Bio l. 2 , 577-586
8. Howe-Grant, M. E. , and Lippard, S. J. (1980) Metal ions in bio logical systems 11 (Sigel, H. ed) pp. 63-125, M. Dekker, New York
9. Courjault-Gautier, F. , Legrimellec, C, Gioconidi, M. C, and Toutain, H. J. (1995) Kidney Int . 47, 1048-1056
10. Tay, L. K., Bregman, C. L. , Masters, B. A., and Williams, P. D. (1988) Cancer Res. 48, 2 38-2 43 11. Daugaard, G., and Abilgaard, U. (I989) Cancer Chemother. Pharmacol . 25 , 1-9
12. Mingeot-leClerq, M. P., Tulkers , P. M. , and Brasseur, R. Biochem. Pharmaco l . 44 , 1967-1975
13. Todd, J. H. and Hottendorf, G. H. (1995) J. Pharmaco l . Exp. Therap. 274 , 285-263
14. Speelmans, G. Sips, W. H. H. M. , Grisel, R. J. H. , Staffhorst, R. W. H. M., Fichtinger-Schepman, A. M. J., Reedijk, J. and de Kruijff, B. (1996) Biochim. Biophys . Acta . in press
15. Taylor, K.D., Goel, R. , Shirazi, F.H. , Molepo, M. , Popovic, P.,
Stewart, D.J. , and Wong, P.T.T. (1995) Br. J. Cancer, 72, 1400-1405
16. Rando, R. R. (1988) FASEB J. 2, 2348-2355
17. De Kruijff, B. (1994) FEBS Lett. 346, 78-82
18. Bligh, E. G., and Dyer, W. L. (1959) Can. J. Biochem. Physiol. 37, 911-917
19. Rouser, G., Fleischer, S. , and Yamamoto, A. (1970) Lipids 5, 4-496
20. Auland, M. E. , and Roufogalis, B. D. , Devaux, P. F. , and Zachowski, A. (1994) Proc. Natl. Acad. Sci. USA 91, 10938-10942
21. Cullis, P. R., and de Kruijff, B. (1978) Biochim. Biophys. Acta 513, 31-42
22. Op den Kamp, J. A. F. (1979) Ann. Rev. Biochem. 48, 7~71
23. Sambrano, G. R. (1995) Proc. Natl. Acad. Sci .USA 92, 1396-1400
24. Krishnaswamy, S., Field, K. A., Edgington, T. S. , Morrissey, J. H. , and Mann, K. G. (1992) J. Biol. Chem. 67, 26110-26120 25. Voelker, D. R. (1984) Proc. Natl. Acad. Sci. USA 81, 2669-2673
26. De Pont, J. J., van Prooijen-van Eeden, A., and Bonting, S. L. (1978) Biochim. Biophys. Acta 508, 467-477
27. Igarashi, K. , Kaneda, M. , Yamaji, A., Saido, T. C, Kikkawa, U., Ono, Y., Inoue, K. , and Umeda, M. (1995) J. Biol. Chem. 270, 2907 - 29078
28. Kupper, M. , Koster, M. , Schmidtspaniol , I., Wagnergillen, I., and Issinger, 0. G. (1994) Cell. Mol. Biol. Res. 40, δ7~592
29. Rubin, E. , Kharbanda, S., Gunji, H. , Weichselbaum, R., and Kufe, R. (1991) Cancer Res. 51, 878-882
1 The following abbreviations were used:
CID Collision induced dissociation cisplatin cis-diamminedichloroplatinum( II )
DOPC 1 , 2-dioleoyl-s?-glycero-3-phosphocholine
DOPS 1 , 2-dioleoyl-sn-glycero-3-phosphoserine
DOPE 1 ,2-dioleoyl-sn-glycero-3-phosphoethanolamine egta ethyleneglycol bis ( β-aminoethyl ethe ) -N , N,N ', N' tetraacetic acid
FAB Fast Atom Bombardment
Mes 2-morpholinoethanesulphonic acid
MS mass spectrometry
MS-MS tandem mass spectrometry
NBA methyl-Nitrobenzylalcohol
Pipes piperazine-1 ,4-diethanesulfonic acid
PC phosphatidylcholine
PE phosphatidylethanolamine
PS phosphatidylserine
TLC thin layer chromatography
LEGENDS TO THE FIGURES
Fig. 1: Formation of a product of DOPS upon incubation with cisplatin. (A) Thin layer chromatography of DOPS, DOPE, DOPC and sphingomyelin. LUVET composed of these lipids were incubated in buffer composed of 10 mM Mes, 50 mM Na2S0/,, 1 mM egta, pH 6.0 for 7 h at room temperature in the presence or absence of 5 M cisplatin. Lipids were extracted and subjected to TLC with as eluens chloroform/methanol/acetic acid (7/2/1 v/v) and visualized with phosphorous spray. (B) Conversion of DOPS in time. DOPS was incubated with 5 mM cisplatin at room temperature in the buffer described above. Samples were taken and extracted and subjected to TLC. The increase in the DOPS - cisplatin reaction product was based on the phosphor content of the product spot on the TLC plate. 100 % is the sum of the amount of phosphor in the product spot and in the DOPS spot .
Fig. 2: Conversion of DOPS at 37 °C in buffers of various composition. DOPS was incubated in the presence of 5 mM cisplatin. Samples were taken and analyzed as described in the legend of fig. 1. (A) Incubation in buffer composed of 10 mM Pipes, 50 mM Na2S0, , 1 mM egta, pH 7.4. (B)
Incubation in buffer composed of 50 mM Mes, 50 mM Na2S0;,, 1 mM egta, pH
6.0. (C) Incubation in buffer composed of 10 mM Pipes, 100 mM NaCl, 1 mM egta, pH 7.^, and (D) Incubation in buffer composed of 10 mM Mes, 100 mM NaCl, 1 mM egta, pH 6.0.
Fig. 3: Mass spectrometry of the product formed upon incubation of DOPS with cisplatin. The inset shows a calculation of the isotope distribution as can be expected for a molecule with a bruto formula of 42 C, 83 H, 3 N, 10 0, 1 P, and 1 Pt.
Fig. 4: MS-MS spectrometry of fragment ions obtained from the ion of m/z 1014 (A), m/z 1015 (B) , and m/z 1016 (C) .
Fig. 5*. Proposed structure of the product formed upon incubation of DOPS with cisplatin with a molecular mass of 1015 dalton (A) . Proposed structure of the large fragment ion obtained from MS-MS (B) . Proposed structure of the small fragment obtained by MS-MS (C) .
Fig. 6: Formation of a cisplatin-PS complex in biological membranes. Ghosts of red blood cells were incubated with or without cisplatin as described in the legend of fig. 1. (A) Lipids were analyzed with 2- dimensional TLC with in the first dimension chloroform/methanol/ammonium/water (34/14/1/1, v/v) and in the second dimension chloroform/methanol/acetic acid (7/2/1, v/v) after which lipids were identified by spraying with phosphorus spray reagent. The upper panel shows lipids after incubation without cisplatin and the lower panel shows the incubation in the presence of cisplatin. (B) Lipids were analyzed after incubation with cisplatin for various periods of time, and extraction with 1-dimensional TLC with chloroform/methanol/acetic acid (7/2/1, v/v) as eluens after which the plate was sprayed with ninhydrin.