Prolonged Serum Half-Life of Antineoplastic Drugs by … · antineoplastic drugs. Krieger et ai. (13), Shaw et al. (14), and Received 3/12/90; accepted 8/30/90. The costs of publication

(CANCER RESEARCH 50, 7476-7482, December I, 1990]

Prolonged Serum Half-Life of Antineoplastic Drugs by Incorporation into the Low

Density Lipoprotein

P. Chris de Smidt and Theo J. C. van BerkelDivision of Biopharmaceutics, Center for Bio-Pharmaceutical Sciences. Sylvius Laboratory, University of Â¡Beiden,P. O. Box 9503, 2300 RA leiden, The Netherlands

ABSTRACT

In vitro and in vivo data have indicated that tumor cells activelyinternalize the low density lipoprotein (LDL) from the circulation. Inorder to achieve a selective delivery of drugs to tumor cells via the LDLpathway, we have incorporated oleyl derivatives of methotrexate andfloxuridine (FdUrd) into LDL particles. Three different incorporationprocedures were studied: Method A, the dry film method; Method B, thetransfer protein method; and Method C, the delipidation-reconstitutionmethod. In all cases, 'H-labeled drug was incorporated into '"I-labeled

LDL to yield double-labeled particles so that the behavior of both drugand carrier could be followed simultaneously. The last method led to thehighest drug loading and it was possible to incorporate 50-70 moleculesof dioleoyl-FdUrd per LDL particle as compared with about 18 moleculesof drug when utilizing the transfer protein procedure.

In vitro studies on the interaction of dioleoyl-FdUrd-LDL particles,obtained by the delipidation-reconstitution method, with the hepatocel-lular carcinoma cell line Hep G2, indicated that these reconstitutedparticles were equally effective in competing for LDL binding as nativeLDL. Moreover, drug delivery to Hep G2 cells occurred at the same rateas cellular association of the apolipoprotein B. In vivo studies on the fateof dioleoyl-FdUrd-LDL complexes in rats indicated that the serum decaywas increased as compared with native LDL. The half-life of 6-9 min is,however, considerably prolonged as compared to the free drug (I* < 1min).

It is suggested that the 6-fold increased serum half-life of the drug-LDL complex accompanied by the possibly more specific tumor deliverymay lead to an increased therapeutic effect.

INTRODUCTION

LDL1 forms a homogenous class of particles present in

human blood in large quantities. One particle contains about1500 molecules of cholesteryl esters in its core which is surrounded by 500 molecules of free cholesterol and 800 moleculesof phospholipids. Regarding the substantial lipid fraction andits spherical shape, the low density lipoprotein shows someresemblance to liposomes. However, LDL is not rapidly clearedby the reticuloendothelial system and can therefore be considered as the natural equivalent of the so-called "stealth" lipo

somes (1,2).The lipid core of LDL is covered by a sole M, 514,000

protein, apoprotein B, which is responsible for recognition ofthe protein by a well-established specific receptor (3). Bindingof LDL by the LDL receptor is followed by internalization andconsequently cholesterol is provided to the cell in order to meetits requirements for building or maintaining stable membranes.The high demand for cholesterol of tumor cells may explainthe high expression of LDL receptors (4-10).

About 10 years ago various authors (5, 11, 12) suggested theapplication of the low density lipoprotein as a carrier forantineoplastic drugs. Krieger et ai. (13), Shaw et al. (14), and

Received 3/12/90; accepted 8/30/90.The costs of publication of this article were defrayed in part by the payment

of page charges. This article must therefore be hereby marked advertisement inaccordance with 18 U.S.C. Section 1734 solely to indicate this fact.

1The abbreviations used arc: LDL, low density lipoprotein; FdUrd, floxuri-dine; MTX, melhotrexate; PBS. phosphate-buffered saline; I-LDL, iodinatedLDL; apo B. apoprotcin B.

Rudling et al. (15) incorporated cytostatic agents into LDL inorder to elicit cytotoxic effects. However, although in vitropromising data were obtained which showed that LDL couldbe rendered cytotoxic, no evidence was provided to show itseffectivity in vivo.

In order to incorporate antineoplastic drugs into the cholesteryl ester core of LDL, these compounds must be renderedlipophilic which may be achieved by coupling either lipophilicside chains or retinoyl (16, 17). In the present study we havecoupled two oleyl chains to methotrexate or FdUrd and evaluated three incorporation procedures: Method A, the dry filmmethod (14, 18), the passive partitioning of drug from a solidsurface to the lipoprotein; Method B, a method involvingenzymatic transfer by plasma transfer protein; and Method C,the solvent extraction method as described by Krieger et al. (19,20).

Throughout our study we have labeled the LDL carrier with125Iand used 'H-labeled (pro)drugs for incorporation in the

lipoprotein. This enabled us to follow the fate of both the drugand carrier simultaneously.

MATERIALS AND METHODS

Materials

'"I, sodium salt (98.5% pure), was obtained from Amersham, UnitedKingdom. |3',5',7-'H]Methotrexate (sodium salt, 98% pure. No. TRK-224) was obtained from Amersham. 5-[6-'H]FdUrd (99.9% pure. No.

NET 774) was from New England Nuclear, Boston, MA.Human serum albumin and agarose were purchased from Sigma

Chemical Co., St. Louis, MO. Bio-Gel A-50 m was from Bio-RadLaboratories, Richmond, < V Fetal calf serum was obtained fromBoehringer Mannheim, Mannheim, Germany, and Dulbecco's modifiedEagle's medium was from Flow Laboratories, Irvine, Scotland. Celile545 was purchased from Fluka, Buchs, Switzerland. ['H]Dioleylmeth-

otrexate was synthesized by the neutral esterification of methotrexatewith an alkyl halide and cesium carbonate according to the method ofRosowsky and Yu (21 ).

The procedure as described by Nishizawa and Casida (22) utilizingoleoyl chloride, was followed to convert [3H]FdUrd to [3H]dioleoyl-FdUrd. Full details of the syntheses of ('H)dioleyl-MTX and [3H]-dioleoyl-FdUrd will be published elsewhere.

Lipoprotein

LDL was isolated from human plasma at density 1.024 <d< 1.055g/ml by two repetitive centrifugations according to the method ofRedgrave et al. (23) as described previously (24). The LDL preparationcontained solely apolipoprotein B (99.97%) and no degradation products were noticeable when checked by electrophoresis in sodium dodecylsulfate gels. Radioiodination of LDL was done according to the ['"!]-iodine monochloride method described by Bilheimer et al. (25). Lipo-protein-deficient serum was isolated as the bottom fraction (density >1.24 g/ml) from human plasma after centrifugaron as described previously (24).

Radioactivity Analysis

Samples were assayed for radioactivity utilizing a Packard 1500 InCarb liquid scintillation analyzer equipped with software validated for'H-'"I double-labeled samples.

7476

Research. on December 23, 2020. © 1990 American Association for Cancercancerres.aacrjournals.org Downloaded from

http://cancerres.aacrjournals.org/

LDL-ASSOCIATED ANTINEOPLASTIC DRUGS

Preparation of Drug-LDL Particles

Three different methods were used for incorporation of lipophilicprodrug into LDL. For all methods, '"I-labeled LDL was used toobtain the double-labeled drug-lipoprotein complex.

Method A (Dry Film/Celite Method). [3H]Dioleoylfloxuridine (ISO

/il. 168 Mg,dissolved in toluene) was added to a glass tube containing20 mg Celite. The drug was coated on the solid surface by evaporationunder N2. After complete evaporation of all solvent, 250 Â¿tlof a solutionof '"I-LDL (550 Mg, in PBS-1 mM EOT A, pH 7.4) were added. The

mixture was incubated under N2 while shaking continuously (overnight). The complex was then purified over a 10.0-ml Sephadex G-25column to discard unbound drug and Celite material.

Method B (Transfer Protein Procedure). This method is basically amodification from the method previously described by Blomhoff et al.(26) and is based upon the presence of a cholesterol ester transferprotein in serum. First 50 /<! (50 Mg) of [3H]dioleoylfloxuridine in

toluene were evaporated to dryness in a glass tube, the compound wasthen dissolved in 50 >i\of acetone, and 500 Â¿ilof a human serum fraction(density > 1.24 g/ml) lipoprotein-deficient serum were added. Afterevaporation of the acetone with N2, 300 M'of PBS-1 mM EDTA wereadded. This lipoprotein-deficient serum-dioleoylfloxuridine preparation was then incubated with 500 Mg '"I-LDL at 37Â°Cfor 4 h. The

mixture was then purified by density ultracentrifugation and gel diromatography.

Method C (Delipidation-Reconstitution Method). The lipophilic pro-drugs were incorporated in LDL according to a modification of theheptane-starch method described previously by Krieger et al. (11, 20).'"I-LDL (470 ng) was freeze-dried for 5 h with 8 mg potato starch in

a siliconized tube. The neutral lipids were extracted with heptane andto the residue were added subsequently 200 M'cholesteryl oleate (= 3mg) in heptane and 100 ÃŸl[3H]dioleoylfloxuridine (112 Mg,10.3 x 10'dpm 3H) in toluene. The organic solvents were then evaporated underN2 at room temperature. The dry residue was then incubated at 4Â°Cwith 1.0 ml of a 10 mM tricine (A'-tris(hydroxymethyl)methylglycine]

buffer, pH 8.4 (calbiochem), with 1 mM EDTA. After 14 h, the starchand unincorporated material were removed by centrifugation over aSepharose column that was saturated with PBS-1 mM EDTA. Additional centrifugations were performed to discard unincorporated drugand to yield a drug-LDL complex that contained 30-70 molecules ofdioleoylfloxuridine per LDL particle.

For the incorporation of [3H]dioleylmethotrexate in LDL, the same

procedure as described above could be followed.

In Vivo Studies

Male Wistar rats (250-330 g) under Nembutal anesthesia were giveninjections of 0.5-1 ml of sample in the vena cava. At regular intervals,blood samples were taken from the vena cava at least 1 cm below thesite of injection and liver lobules were excised. Blood samples werecentrifuged and 200 M!of serum were mixed with 10 ml of HIONICscintillation cocktail before assay for 3H and '"I radioactivity.

Liver lobules were weighed and homogenized. Homogenate (500 n\)was incubated with 750 Â¿ilof Soluene, mixed with 10 ml of HIONICscintillation cocktail, and analyzed for 'II and '"1 radioactivity.

Density Gradient Ultracentrifugations

Samples of double-labeled drug-LDL complexes obtained by different incorporation procedures were analyzed by ultracentrifugation. Anappropriate amount of sample (200-600 Â«J,in PBS-1 mM EDTA) wasleft at room temperature in a poly allumer centrifuge tube. Thereafter,1106 mg of solid KBr and PBS-1 mM EDTA were used for the completedissolution of KBr to a final volume of 4.0 ml. Consecutive layers of3.0, 2.5, and 2.5 ml of KBr solution (1.063, 1.019, and 1.0063 g/ml,respectively) were then added.

The tubes were centrifuged in a Beckman ultracentrifuge at 40,000rpm (6x15 swing-out rotor) for 22 h at 4Â°C.By taking 500-^1 samples,

starting at the bottom of the tube, the gradient was subdivided accordingto density.

Agarose Electropboresis

Aliquots of 3H-125Idrug-LDL complexes were subjected to electro-

phoresis in agarose gels at pH 8.8 (Tris-hippuric acid buffer). Afterelectrophoresis, the gel was cut into segments which were then treatedwith 750 /iI of soluene. After 20 h, 10.0 ml scintillation cocktail wereadded and the samples were assayed for 'II and '"I radioactivity.

Gel Chromatography

A small volume (100-200 //I) 3H-'"I drug LDL complex was loadedon a 0.9- x 44-cm Bio-Gel A-50 m column. The sample was eluted withPBS-1 mM EDTA, pH 7.4, at a flow rate of 4 ml/h. Aliquots (500 n\)of the different fractions were analyzed for their 3H and '"I radioactiv

ity.

Culturingof Hep G2 Cells

The Hep G2 cell line, derived from a human hepatocyte tumor, wasobtained from Dr. B. B. Knowles (Wistar Institute of Anatomy andPhysiology, Philadelphia, PA). The cells were cultured at 37Â°Cin 25-cm ' flasks (Costar) containing 0.2 ml of culture medium/cm2 supple

mented with 10% heat-inactivated fetal calf serum, penicillin, andstreptomycin under CO2:air (1:19). The medium was renewed twice aweek.

Cellular Uptake of Drug-LDL Particles

Three-4 days before the experiment, Hep G2 cells were trypsinizedand transferred to 2-cm2 Costar dishes. Twenty h before the assay, themedium was replaced with Dulbecco's modified Eagle's medium con

taining 1% human serum albumin (preincubation medium) to enhancecellular LDL receptor expression. Just prior to the experiment, thecells were washed and incubated with preincubation medium threetimes (15, 15, and 30 min, respectively). The Hep G2 cells were thenincubated with LDL in Dulbecco's modified Eagle's medium-1% human serum albumin at 37Â°Cin a humidified incubator. After 4 h thewells were cooled to 4Â°Cand washed with PBS-2.5 mM Ca2*-0.2%

bovine serum albumin (5x) followed by two final washes with PBS-2.5M Ã‡a2*.For determination of cellular association of drug and carrier,

the cells were dissolved in 500 ii\ 0.1 N NaOH and sonicated. Of thiscell homogenate 350 //I were assayed for radioactivity and proteinaccording to the method of Lowry et al. (27) with bovine serum albuminas standard.

RESULTS

Physicochemical Properties. The efficiency of incorporationof 3H-labeled prodrug into the low density lipoprotein was

evaluated by density ultracentrifugation, gel chromatography,and agarose gel electrophoresis.

Construction of a double-labeled particle enabled the directcomparison between drug and carrier, and furthermore a properestimation can be made of the incorporation efficiency andnumber of drug molecules per LDL particle.

Dry Film Incorporation Method. Fig. 1 shows density ultracentrifugation of particles obtained by the dry film method. Theessential step of this method is the partitioning of drug from asolid surface to the lipoprotein. The contact area can be greatlyenlarged by using glass beads (28) or Celite (29), the latterbeing used for this series of experiments.

The shouldered shape of the [3H]methotrexate curve, distinctfrom the I-LDL peak, suggests that only a fraction of dioleyl-methotrexate is LDL associated. The position of the smallshoulder, however, matches that of the carrier. When the di-oleoyl-FdUrd-LDL preparation is subjected to ultracentrifugation, a uniform peak is found (Fig. IB). However, this analysismay be undiscriminating because dioleoyl-FdUrd, in the absence of LDL, also flotates at a similar density. When subjected

7477



LDL-ASSOCIATED ANTINEOP1.ASTIC DRUGS

3000

20

9 1113

FRACTION

15 17 19 21 23 25MV

~-Â«4Ã¶ocT5Â£joZa.

3yuuu8000700060005000

4000300020001000A

Tfri

\\iÂ¿IQ1\%1

1Â°\\Â«o \i/

/ 0P*\â€¢

/ / Â°\ â€¢_iT~-o MnÂ«â€” . ..zuoou150001000050000

^f^*1*** ' '.'.M-'J-' ' 1 ^-^ 'Q0

10 20 3040FRACTION

lai

â€¢ 4000

1 3 5 7 9 11 13 15 17 19 21 23 25

FRACTION <NRÃŒ

2800

~ 2400

â€¢

2000

_Â« 1600

1200

SÃ» 400

1400

1200 ~O

1000

800

o600 Y

400 2^

200 ft

100 200

4 5 6 7 8 9 10

DISTANCE ICMI

Fig. 2. Physicochemical analysis of |'H|dioleoyl-FdUrd ({'HJFUdROl2)-'"l-

LDL complexes prepared by Ihe transfer protein method. /I, chromatography of|3H|dioleoyl-FdUrd-'"I-LDL on Bio-Gel A-50 m; B, agarose gel elcctrophoresisofdioleoyl-FdUrd-LDL.

_Â«4500O

Â£ 3000I

0123456789 10

DISTANCE ICMI

Fig. I. Physicochemical analysis of drug-LDL particles prepared by the dryfilm/Celite method. A, density gradient ultracentrifugation of |3H]dioleyl-MTX(MTXOI2)-'"\-LDL prepared by the dry film/Celite method; B, density gradientullracentrifugation of |3H]dioleoyl-FdUrd (/'HJFVdROI^-'^i-LDL prepared bythe dry film/Celite method; C. agarose gel electrophoresis of [3H]dioleoyl-FdUrd-'"l-LDL prepared by the dry film/Celite method. NR, number.

to agarose electrophoresis, a distinct migration of LDL discriminates the required particle from free dioleoyl-FdUrd. it can beseen (Fig. 1C) that a substantial part of dioleoyl-FdUrd is foundin a fraction different from I-LDL. Low incorporation efficiencies did not allow the performance of measurable gel chromatography.

Enzymatic Transfer. The second incorporation procedureevaluated was the transfer protein method (26). Fig. 2A showsthat unbound drug Ã©lÃ»tesdirectly after the void volume, whilea substantial part of drug was collected in the same fractions asLDL. These fractions were pooled and subjected to densityultracentrifugation (Fig. 3/1). The ['Hjdioleoyl-FdUrd peak appeared at the same density as I25l-apo B while [5H]dioleylmeth-

otrexate was collected at significantly higher densities (Fig.3B).

7478

1 3 5 7 9 11 13 15 17 19 21 23

FRACTION INRI

1 3 5 7 9 11 13 15 17 19 21 23 25

FRACTION (NRI

Fig. 3. Density ultracentrifugation of drug-LDL complexes prepared by thetransfer protein method. . I. density gradient ultracentrifugation of the dioleyl-FdUrd (Fl/V/AO^l-LDL-containing fraction that was previously collected afterBio-Gel A-50 m chromatography (Fig. Z4); B, density gradient ultracentrifugationof dioleyl-MTX (MTXOI2)-LDL prepared by the transfer protein method. NR,number.




Agarose gel electrophoresis of the dioleoyl-FdUrd-LDL par

ticle obtained after density ultracentrifugation confirms theuniformity of the drug-LDL complex (Fig. 2B). In variouspreparations the incorporation of dioleoyl-FdUrd varied from0.2 to 0.6% of the added prodrug. Similar experiments wereperformed with ['HJcholesteryl oleate indicating an incorpora

tion up to 70% of the added labeled substrate.Delipidation-Reconstitution. The delipidation-reconstitution

method (11, 20) involves freeze drying and heptane extraction.Both on gel chromatography and on agarose gel electrophoresis(Fig. 4, A and B), the [3H]dioleoyl-FdUrd peak coincided withthe I25I-LDL peak. Also [3H]dioleylmethotrexate showed a gelChromatographie behavior similar to that of I-LDL (Fig. 4C).

The incorporation efficiencies of the prodrugs varied from 15to 30%. Protein recovery was 50 Â±15% (SD; n = 3).

7000600050004000300020001000A*OPu11I*

1I

\r

\â€¢\k*J

W4UUU~O3000

â€”j-2000

Â§iÂ¿~-1000

2Ã¨0

10 20 30 40

FRACTION (gÃ¬

20000

4000 O

<â€¢> 5000

0123456789 10

In Vitro Incubations. The in vitro interaction of the drug-LDL complex was investigated with Hep G2 cells, a cell linewith a well-established LDL receptor (30, 31). In order to testif the drug incorporation method affects the recognition ofLDL by the LDL receptor, we compared the effect of increasingconcentrations of unlabeled native and reconstituted drug-LDLcomplexes on the binding of I-LDL (Fig. 5A). Both the controland drug-LDL complex (delipidation-reconstitution method)are equally effective in competing with '"I-LDL binding. Sim

ilarly, the dry film incorporation method does not affect theability of LDL for LDL receptor recognition (Fig. 5B).

The delipidation-reconstitution method allowed labeled drugincorporation yields that were sufficient to perform in vitrodrug delivery studies. Fig. 6 shows the effect of increasing LDLconcentrations on the cell association of ['Hjdioleoyl-FdUrd

and iodinated LDL in the absence and presence of an excessunlabeled LDL (200 ^g/ml). Uptake of both the drug and LDLclearly has a saturable component and the cell association ofboth particles is decreased by an equal level in the presence ofexcess LDL.

In Vivo Studies. Previous studies have shown that nativeLDL has a half-life in the rat of 5 to 6 h (32). The free drugsdioleoyl-FdUrd and dioleyl-MTX in an albumin-stabilizedphosphate-buffered saline solution are cleared within minutesfrom plasma (half-lives < 1 min) and taken up primarily by theliver (Fig. 7).

500

30 60 90 120

ILDLI tuO/mL/

40 80 120 160

[LDL1 tutl/mUFig. 5. Ability of dioleoyl-FdUrd-LDL to compete with the association of I25I-

LDL with HepG2 cells. Left, dioleoyl-FdUrd-LDL prepared by the delipidation-reconstitution method (O) native LDL (â€¢);righi, dioleoyl-FdUrd-LDL preparedby the dry film method (O) native LDL (â€¢).

_!Â» 2000

FRACTION (gl

Fig. 4. Physicochemical analysis of drug-LDL complexes prepared by thedelipidation-reconstitution procedure. A, Bio-Gel A-50 m chromatography of|JH]dioleoyl-FdUrd (/V//-/Y/iÂ«O/2)-1"I-LDL: B. agarose gel electrophoresis of|3H]dioleoyl-FdUrd-'"I-LDL; C, chromatography of |3H]dioleyl-MTX (f'HJ-MTXOI2)-'"\-WL on Bio-Gel A-50 m.

ILDLI luO/mLJ

Fig. 6. Relation of the concentration of dioleoyl-FdUrd-LDL to the extent ofcell association with HepG2 cells. O, â€¢,'"I activity; A, A, ['Hjdioleoyl-FdUrd.

â€¢,A, cellular association of drug or LDL. O, A, cellular association of drug/LDLwith incubation of excess (200 jig/ml) native LDL.

7479




20 30 0 10 20 30 40

TIME (MINUTES/ TIME (MINUTES}

Fig. 7. Serum decay and liver association of dioleoyl-FdUrd and LDL. Left,serum decay of dioleoyl-FdUrd (A). Also shown (â€¢)is the serum decay of nativeLDL (separate rats). Data are average values of two rats; right, liver uptake ofdioleoyl-FdUrd (A) and reference native LDL III. Bars, SD.

When double-labeled ['HJFdUrd-1 "I-LDL, prepared by the

dry film method, was injected in the rat, a separation of theserum decay of drug and carrier was observed (Fig. 8/1). Boththe decay and liver uptake of drug exceed uptake of labeledLDL. The decay and liver uptake of '"I-LDL appear to be

slightly affected by exposure of LDL to the incorporationmethod.

Fig. SB shows serum decay and liver uptake of ['HJdioleoyl-FdUrd-'"I-LDL particles obtained by the transfer protein

method. Here, it is also noticed that the decay of drug is morerapid and the liver uptake percentual higher than for the I25I-LDL. The serum decay as well as the liver uptake of I25I label

is essentially the same as with native LDL (Fig. 7).The delipidation-reconstitution method is generating drug-

LDL-particles with an in vivo behavior which differs from theaforementioned method (Fig. 9). Drug and carrier follow similar decay rates and reach similar liver uptake values.

As compared to the half-life of the albumin-bound drug, thehalf-life of the dioleylmethotrexate is, however, significantlyincreased. A similar prolongation of half-life was also observedfor dioleoyl-FdUrd. However, the decay of I-LDL appears tobe strongly stimulated, mostly as a consequence of an increasedliver uptake.

DISCUSSION

Three methods for the incorporation of drugs into LDL wereevaluated on basis of quantitative association, //; vitro modification of receptor recognition, and cell uptake and finally, mostimportantly, on basis of the in vivo fate of both the LDL andthe drugs. ['HjMTX and [3H]FdUrd were derivatized in order

to increase the affinity of the drugs for the hydrophobic core ofthe LDL molecule. For an adequate evaluation of the LDLparticle and the drug, double-labeled drug-LDL complexes were

used.The dry film/Celite method is based upon the passive parti

tioning of drug from a solid surface into the LDL. The transferprotein method, described previously for the incorporation ofradiolabeled cholesteryl oleate into lipoproteins (26), utilizestransfer enzyme activity present in plasma. Experimental conditions for drug incorporation with these two methods arerelatively mild for the LDL particle, because either spontaneousor enzyme-facilitated drug incorporation is applied. In accordance with this assumption, in vitro receptor recognition of dryfilm ['Hjdioleoyl-FdUrd-'"I-LDL was identical to native LDL

and serum decay of the apo B moiety of this particle was also

too

TIME (MINUTES! TIME (MINUTES!

TIME (MINUTES; TIME (MINUTESÂ»

Fig. 8. Serum decay and liver association of dioleoyl-FdUrd-LDL particles.A, serum decay (Â¡eft)and liver association (right) of dioleoyl-FdUrd-LDL particlesprepared by the dry film procedure. â€¢'"I-LDL; A, dioleoyl-FdUrd (n = 2). B,serum decay (left) and liver association (right) of dioleoyl-FdUrd-LDL particlesprepared by the transfer protein method. â€¢.'"I-LDL; A, dioleoyl-FdUrd (n = 2).

Ã•

ft*

TIME (MINUTES) TIME (MINUTES!

100Â»

a

S

iSu*

TIME (MINUTES! TIME (MINUTES)

Fig. 9. Serum decay and liver association of drug-LDL complexes preparedby the delipidation-reconstitution method.. I, serum decay (left) and liver uptake(right) of dioleyl-MTX-LDL particles. â€¢'"I-LDL; A, dioleyl-MTX. B, serumdecay (left) and liver uptake (right) of dioleoyl-FdUrd-LDL particles. â€¢ '"I-LDL; A, dioleoyl-FdUrd.

7480




unchanged. The serum half-life of the drug, however, was notidentical to its LDL carrier, which probably was due to thepresence of non-firmly LDL-associated drug in the preparation(Fig. 1C). In the case of dioleyl-MTX and dioleoyl-FdUrd thedry film procedure gave low incorporation values into LDLand, moreover, "incorporated" drug may not be associated withinner core of the LDL particle. Such a "loose" incorporation

was found earlier with benzo(a)pyrene-LDL complexes obtained by a similar procedure (33).

The enzymatically catalyzed incorporation method (26) isunexplored for the incorporation of drugs in LDL. The methodis very successful (up to 80% incorporation efficiency) in thecase of incorporation of radiolabeled cholesteryl oleate. Densityultracentrifugation pointed out that dioleoyl-FdUrd can anddioleyl-MTX cannot be incorporated into LDL with this procedure, indicating a high degree of structure specificity.

Although the serum halflife of apo B was not significantlyaltered by the transfer protein procedure (Fig. SB), surprisinglyliver uptake of dioleoyl-FdUrd was higher than that of LDLitself which may be due to "loose" incorporation of drugs into

LDL.Of the three methods tested, the delipidation-reconstitution

procedure had the highest incorporation efficiency. Incorporation of 10-30% of the amount of added drugs were found.Using this procedure, LDL-drug complexes of 50-70 moleculesof dioleoyl-FdUrd per LDL particle were produced but occasionally a drug loading up to 400 molecules could be achieved.Agarose gel electrophoresis and gel chromatography (Fig. 4, Aand B) gave evidence that physicochemically well-defined particles can be obtained with this procedure. Migration of theparticle in agarose gels was not significantly different thanmigration of native LDL. In vitro examination of the interactionof such reconstituted LDL with LDL receptors on Hep G2 cellsindicated that the competition with the cell association of 125I-

LDL was identical for reconstituted and native LDL. Thisindicates that the receptor recognition properties of apo Bremained intact by the drug incorporation method. With [JH]-dioleoyl-FdUrd-'"I-LDL particles obtained by the delipidation-

reconstitution procedure, a saturable component in the cellassociation was found and apparently equimolar uptake of thedrug and particle association occurred. Furthermore, the presence of 200 i/g/ml native LDL showed a substantial effect onthe association of both the drug and the LDL particle. Thus itappears possible to produce LDL particles that show a receptor-dependent controlled delivery of dioleoyl-FdUrd to cells.

However, in vivo, the delipidation-reconstitution method revealed particles which were cleared more rapidly from thecirculation than the native particle. The liver appeared to beresponsible for this increased decay. Although this behavior ofLDL is undesired, it may be stated that the incorporation ofdioleoyl-FdUrd into LDL led to a 6-fold retarded half-life inthe blood. Although a high initial plasma concentration is notnecessarily an advantage for the effect on the tumor, the anticipated higher LDL-mediated tumor cell uptake may lead to aconsiderably improved therapeutic effect.

The type of LDL modification responsible for the increasedliver uptake is unclear at the moment. Modification of LDL,for instance by acetylation (26, 34) or oxidation (35, 36), leadsto an increased uptake by the liver. This uptake is mediated byhighly active receptors on liver endothelial and Kupffer cells(35, 37). The mechanism of this additional liver uptake is nowunder investigation and methods to prevent such an uptakemust be further explored.

It is clear from the present results that promising physico-chemical and //; vitro data are not necessarily representative forsimilar results at the physiological level. In vivo serum decayand liver uptake studies indicated that two of the methodsdescribed in this paper did not significantly alter in vivo behaviorof the carriers. However, further work on the efficiency of thesemethods is needed before in vivo application will be possible.

Further investigations are needed to analyze the mechanismof the induced liver uptake before adequate prevention of suchmodifications can be achieved. However, simultaneously clinical studies are initiated in order to test if the 6-fold increasedserum half-life of the cytostatic prodrugs, accompanied by thepossible specific tumor delivery, will lead to an increased therapeutic effect.

REFERENCES

9.

10.

11.

12.

13.

14.

15.

16.

17.

18.

19.

20.

21.

Allen, T. M and Munin. A. Large unilamellar liposomes with low uptakeinto the rctieuloendothelial system. FEBS Lett., 223: 42-46, 1987.Gabizon, A., and Papahadjopoulos, D. Liposome formulations with prolonged circulation time in blood and enhanced uptake by tumors. Proc. Nail.Acad. Sci. USA, Â«5:6949-6953, 1988.Brown, M. S., and Goldslein, J. L. A receptor-mediated pathway for cholesterol homeostasis. Science (Washington DC), 232: 34-47, 1986.Ho, Y. K., Smith, R. G., Brown, M. S., and Goldstein, J. L. Low densitylipoprotein (LDL) receptor activity in human acute myclogenous leukemiacells. Blood, 52: 1099-1114, 1978.Gal, D., Ohashi, M., MacDonald, P. C., Bucksbaum, H. J., and Simpson, E.R. Low density lipoprotein as a potential vehicle for chemotherapeutic agentsand radionucleotides in the management of gynecologic neoplasms. Am. J.Obstel. Gynccol., 133: 877-885, 1981.Gal, D., MacDonald, P. C, Porter, J. C, and Simpson, E. R. Cholesterolmetabolism in cancer cells in monolayer culture. 111.Low density lipoproteinmetabolism. Int. J. Cancer, 28: 315-319, 1981.Vitols, S., Gahrton, G., Ost, A., and Peterson, C. Elevated low densitylipoprotein receptor activity in leukemic cells with monocytic differentiation.Blood, 63: 1186-1193, 1984.Vitols, S., Bjorkholm, M., Gahrton, G., and Peterson, C. Hypocholesterole-mia in malignancy due to elevated low density lipoprotein receptor activityin tumor cells: evidence from patients with leukemia. Lancet, 2: 1150-1154,1985.Norata, G., Canti, G., Ricci, L., Nicolin, A., Trezzi, E., and Catapano, A. L.In fit-Â«assimilation of low density lipoproteins by a fibrosarcoma tumor linein mice. Cancer Lett., 25: 203-208, 1984.Hynds. S. A., Welsh, J., Stewart, J. M., Jack, A., Soukop, M., McArdle, C.S., Calman, K. C., Packard, C. J., and Shepherd, J. Low density lipoproteinmetabolism in mice with soft tissue tumors. Biochim. Biophys. Acta, 795:589-595, 1984.Krieger. M., Brown. M. S., Faust, J. R., and Goldstein, J. L. Replacementof endogenous cholesteryl esters of low density lipoprotein with exogenouscholesteryl linoleate. J. Biol. Chem., 255:4093-4101, 1978.Counsell, R. E., and Pohland, R. C. Lipoproteins as potential site-specificdelivery systems for diagnostic and therapeutic agents. J. Med. Chem., 25:1115-1120, 1982.Krieger, M., Smith, L. C., Anderson, R. G. W., Goldstein, J. L., Kao, Y. J.,Pownall, H. J., Gotto, A. M., Jr., and Brown, M. S. Reconstituted lowdensity lipoprotein: a vehicle for the delivery of hydrophobic fluorescentprobes to cells. J. Supramol. Struct., 10: 467-478, 1979.Shaw. J. M.. Shaw. K. V., Yanovich, S., Iwanik. M., Futch, W. S., Rosowsky,A., and Schook, L. B. Delivery of lipophilic drugs using lipoproteins. Ann.NY Acad. Sci., 507: 252-271. 1987.Rudling, M. J., Collins, V. P., and Peterson, C. O. Delivery of aclacinomycinA to human glioma cells in vitro by the low density lipoprotein pathway.Cancer Res., 43: 4600-4605, 1983.Vitols, S. G.. Masquelier, M., and Peterson, C. D. Selective uptake of a toxiclipophilic anthracyclinc derivative by the low-density lipoprotein receptorpathway in cultured fibroblasts. J. Biol. Chem., 28: 451-454. 1985.Firestone, R. A., Pisano, J. M., Falck. J. R., McPhaul, M. M., and Krieger,M. Selective delivery of cytotoxic compounds to cells by the LDL pathway.J. Mcd. Chem., 27: 1037-1043, 1984.Iwanik, M. J., Shaw, K. V., Ledwith, B. J., Yanovich, S., and Shaw, J. M.Preparation and interaction of a low-density lipoprotein: daunomycin complex with P388 leukemic cells. Cancer Res., 44: 1206-1215, 1984.Krieger, M., McPhaul, M. J., Goldstein, J. L., and Brown, M. S. Replacementof neutral lipids of low density lipoprotein with esters of long chain unsatu-rated fatty acids. J. Biol. Chem., 256: 3845-3853, 1979.Krieger, M. Reconstitution of the hydrophobic core of low-density lipoproteins. Methods F.nzymol., 128: 608-613, 1986.Rosowsky, A., and Yu, C.-S. New methods of neutral cstcrification ofmethotrexate and related compounds. Dev. Biochem., 4: 273-277, 1978.

7481




22. Nishizawa, Y., and Casida, J. E. 3',5' Diesters of 5-fluoro-2'-deoxyuridine:synthesis and biological activity. Biochem. Pharmacol., 14:1605-1619, 1965.

23. Redgrave, T. G., Roberts, D. C. K., and West. C. E. Separation of plasmalipoproteins by density gradient ultracentrifugation. Anal. Biochem., 65:42-49, 1975.

24. Van Berkel, T. J. C., Kruijt, J. K., Spanjer, H. H., Nagelkerke, J. F., Harkes,L., and Kempen. H. J. M. The effect of a water-soluble tris-galactoside-terminated cholesterol derivative on the fate of low density lipoproteins andliposomes. J. Biol. Chem., 260: 2694-2699, 1985.

25. Bilheimer, D. W., Eisenberg, S., and Levy, R. 1. The metabolism of very lowdensity lipoproteins. I: Preliminary in vitro and in vivoobservations. Biochim.Biophys. Acta, 280: 212-221, 1972.

26. Blomhoff, R., Drevon, C. A., Eskild, W., Helgerud, P., Norum, K., and Berg,T. Clearance of acetyl low density lipoprotein by rat liver endothelial cells.J. Biol. Chem., 259: 8898-8903, 1984.

27. Lowry, O. H., Rosebrough, N. J., Fair, A. L., and Randall, R. J. Proteinmeasurement with the Folin phenol reagent. J. Biol. Chem., 93: 265-275,1951.

28. Shu, H. P., and Nichols, A. V. Benzo(a)pyrene uptake by human plasmalipoproteins in vitro. Cancer Res., 39: 1224-1230, 1979.

29. Seki, J., Okita, A., Watanabe, M., Nakagawa, T., Honda, K., Tatewaki, N.,and Sugiyama, M. Plasma lipoproteins as drug carriers: pharmacologicalactivity and disposition of the complex of Â£f-sitosteryl-/}-i>-glucopyranoside

with plasma lipoproteins. J. Pharm. Sci., 74: 1259-1264, 1985.30. Havekes, L., Van Hinsbergh, V. W. M., Kempen, H. J. M., and Emeis, J. J.

The metabolism in vitro of human low density lipoprotein by the humanhcpatoma cell line Hep G2. Biochem. J., 214:951-958, 1983.

31. Wu, G. Y., Wu, C. H., Rifici, V. A., and Stocken, R. J. Activity and regulationof low density lipoprotein receptors in a human hcpatoblastoma cell line.Hepalology. 4: 1190-1194, 1984.

32. Harkes, L., and Van Berkel, T. J. C. Cellular localization of the receptor-dependent and receptor-independent uptake of human IDI in the liver ofnormal and 17<r-ethinylestradiol-treated rats. FEBS Lett., Â¡54:75-79, 1983.

33. Rcmsen, J. F.. and Shircman, R. B. Effect of LDL on the incorporation ofbenzojaIpyrene by cultured cells. Cancer Res., 41: 3179-3185, 1981.

34. Van Berkel, T.J.C., Nagelkerke, J. F., Harkes, L., and Kruijt, J. K. Processingof acctylated human low density lipoprotein by parenchyma! and non-paren-chymal cells. Biochem. J., 20Â«:493-503, 1982.

35. Parthasarathy, S., Printz. D. J., Boyd, D., Joy, L., and Steinberg, D. Macrophage oxidation of low density lipoprotein generates a modified formrecognized by the scavenger receptor. Arteriosclerosis, 6: 505-510, 1986.

36. Steinberg, D. Metabolism of lipoproteins and their role in the pathogenesisof atherosclerosis. Artherosclerosis Rev., 18: 1-23, 1988.

37. Nagelkerke, J. F., Barto, K. P., and Van Berkel, T.J. C. In vivo and in vitrouptake and degradation of acetylatcd low density lipoprotein by rat liverendothelial, Kupffer and parenchyma! cells. J. Biol. Chem., 25Â«:12221-12227, 1983.

7482



1990;50:7476-7482. Cancer Res P. Chris de Smidt and Theo J. C. van Berkel Incorporation into the Low Density LipoproteinProlonged Serum Half-Life of Antineoplastic Drugs by

Updated version

http://cancerres.aacrjournals.org/content/50/23/7476

Access the most recent version of this article at:

E-mail alerts related to this article or journal.Sign up to receive free email-alerts

Subscriptions

Reprints and

[email protected] at

To order reprints of this article or to subscribe to the journal, contact the AACR Publications

Permissions

Rightslink site. Click on "Request Permissions" which will take you to the Copyright Clearance Center's (CCC)

.http://cancerres.aacrjournals.org/content/50/23/7476To request permission to re-use all or part of this article, use this link



http://cancerres.aacrjournals.org/cgi/alerts

mailto:[email protected]



Documents

Prolonged Serum Half-Life of Antineoplastic Drugs by … · antineoplastic drugs. Krieger et ai. (13), Shaw et al. (14), and Received 3/12/90; accepted 8/30/90. The costs of publication