Effects of Liposome Dose and the Presence of Lymphosarcoma Cells on Blood Clearance and Tissue Distribution of Large Unilamellar Liposomes in Mice1

[CANCER RESEARCH 43, 2927-2934, June 1983] 0008-5472/83/0043-0000$02.00 Effects of Liposome Dose and the Presence of Lymphosarcoma Cells on Blood Clearance and Tissue Distribution of Large Unilamellar Liposomes in Mice1

Marma Ellens, Henriette W. M. Morseli, Bert H. J. Dontje, Dharamdajal Kalicharan, Caesar E. Hulstaert, and Gerrit L. Scherphof2

Laboratory of Physiological Chemistry, University of Groningen, Bloemsingel 10, 9712 KZ [H. E., H. W. M. M., B. H. J. D., G. L S.] and Centre for Medical Electron Microscopy, University of Groningen, Oostersingel 69", 97J3 EZ [D. K., C. E. H.], Groningen, The Netherlands

ABSTRACT Depression of reticuloendothelial activity has been suggested to serve as a means to reduce liver and spleen uptake of Large unilamellar liposomes (50 to 500 ^mol of lipid per kg) liposomes and to prolong the circulation time of the vesicles (1, were injected i.v. or i.p. into normal and lymphosarcoma-bearing 12, 13). The rate of disappearance of liposomes from the blood mice. The percentage of the dose remaining in the blood and decreases with increasing dose (33), indicating that the clearance that accumulated in liver, spleen, and various other organs was mechanism is a saturable process. Concomitantly, one would measured 4 hr after injection. The results indicate that liposomes expect the proportion of liposomes taken up by the liver to cause a dose-dependent saturation of the hepatic and splenic decrease with increasing dose, but conflicting results have been clearance capacities. One day after injection of 106 lymphosar- reported on this subject (12, 33). Tumor cells may influence coma cells, the capacity of the tumor-bearing mice to eliminate reticuloendothelial activity (3), but thus far little attention has liposomes from the blood (in a 4-hr period) was inhibited 30 to been paid to the possible effects of the presence of tumor cells 50%. This could be ascribed to a decreased activity of the on the elimination of liposomes from the blood. reticuloendothelial system caused by the tumor cells, as was We confirmed that large liposome doses result in prolonged indicated by the simultaneous inhibition of carbon clearance. Six circulation time in normal as well as in tumor-bearing mice, after days after injection of the lymphosarcoma cells, the elimination both i.v. and i.p. administration. During this study, we observed, of liposomes from the blood in tumor-bearing mice was restored however, a profound influence of the presence of lymphosar to the value in normal mice. The possible involvement of tumor coma cells in the animals on blood clearance and organ distri cells in the uptake of liposomes by the liver was investigated bution of the liposomes. In addition, we obtained direct evidence morphologically after i.v. injection of peroxidase-containing lipo of the lack of in vivo liposome uptake by lymphosarcoma cells somes into lymphosarcoma-bearing mice. Liposome-entrapped situated in the liver, despite their direct accessibility from the peroxidase activity was never observed in the tumor cells. The bloodstream. results presented here indicate that the lymphosarcoma cells do not directly participate in the hepatic accumulation of liposomes, MATERIALS AND METHODS although their mere presence may have significant indirect effects on the elimination of liposomes from the blood and on their tissue Liposome Preparation. Sphingomyelm (from bovine brain) and cho distribution. lesterol were obtained from Sigma Chemical Co., St. Louis, Mo. Phos- phatidylserine, purified from bovine brain as described by Papahadjopou- los and Miller (24), was a gift from Dr. J. C. Wilschut of our laboratory. INTRODUCTION 125l-Labeled PVP3 (specific activity, 38 /iCi/mg; M, 30,000 to 40,000) was purchased from the Radiochemical Centre, Amersham, England. Unipore The potential clinical usefulness of liposomes as a system for polycarbonate filters were from Bio-Rad Laboratories, Richmond, Calif. the controlled delivery of therapeutic agents has been discussed Sepharose CL-2B and Ficoll (M, 70,000) were obtained from Pharmacia in a number of recent reviews (15, 17, 23). After i.v. or i.p. Fine Chemicals, Piscataway, N. J. Horseradish peroxidase Grade II 3,3'- injection, a substantial proportion of the liposomes is found to diaminobenzidine tetrahydrochloride Grade II were from Sigma. The accumulate in the liver and spleen. Liposomes can be taken up diammonium salt of 2,2'-azinodi(3-ethylbenzthiazolinsulfonic acid) was by tissue macrophages (mainly located in liver and spleen) as from Boehringer/Mannheim, Mannheim, Germany. has been demonstrated for Kupffer cells in the liver (27, 37), but REV were prepared essentially as described by Szoka and Papa- relatively little is known about any factors which may affect such hadjopoulos (35). Sphingomyelin, cholesterol, and phosphatidylserine uptake. In cancer chemotherapy, only a few successful applica (molar ratio, 4:5:1) were dispersed in diethyl ether to give a concentration of 20 /Â¿molof total lipid per ml (the diethyl ether was distilled in the tions of liposomes as a drug carrier system have been reported presence of sodium bisulfite immediately before use). The aqueous phase (11, 14, 19, 21, 25), and the presently available evidence sup (150 mM NaCI:5 mw Tris-HCI, pH 7.4) contained 126l-labeled PVP (2.6 ports the hypothesis that the therapeutic effects from a liposome- mg/ml with varying specific activity). Per ml of the organic phase, 0.3 ml encapsulated cytotoxic drug are due to a delayed elimination of the aqueous phase was added, and the mixture was sonicated under from the blood and a sustained release of the drug from the nitrogen in a bath-type sonicator (Branson). After formation of a homo liposomes, rather than to a selective uptake of the encapsulated geneous suspension of inverted micelles, the organic phase was re drug by the tumor cells (16, 29). moved on a rotary evaporator, and the vesicles thus formed were extruded through a polycarbonate membrane with 0.4-Mm pores (36). 1This research was supported by the Netherlands Cancer Foundation Koningin Wilhelmina Fonds, Project GUKC llx. 1To whom requests for reprints should be addressed. 3 The abbreviations used are: PVP. polyvinyl pyrrolidone; REV, reverse-phase evaporation vesicles; PBS, phosphate-buffered saline.

JUNE 1983 2927

Nonencapsulated 125l-labeled PVP was separated from encapsulated the carbon concentration was measured by determining the light absor- material by gel filtration on Sepharose CL-2B. Void volume fractions bance at 620 nm. The phagocytotic index was calculated according to containing the liposomes were pooled and, if necessary, concentrated in the method of Donald and Tennent (8). an Amicon concentration cell (XM-300 filter). Vesicle lipid concentrations Microscopic Studies. Horseradish peroxidase-containing REV were were determined by lipid phosphorus assay according to the method of injected i.v. into 3-month-old (C57BL x Gr) F, mice. The mice received Bartlett (4), as modified by BÃ¶ttcher ef al. (5). Liposomes prepared this 2.8 MfTiolof liposomal lipid containing 0.13 mg of peroxidase. Two types way had an encapsulated volume of 4 to 5 liters per mol of phospholipid. of controls were included: (a) mice receiving neither liposomes nor For encapsulation of peroxidase, a solution of 30 mg of horseradish peroxidase; and (o) mice receiving liposomes mixed with free peroxidase. peroxidase per ml of 150 ITIMNaCI:5 rriM Tris-HCI pH 7.4, was used as After 2 hr, the mice were anesthetized with ether, and the liver was the aqueous phase. Peroxidase-containing liposomes were separated perfused in situ (flow rate, 2 ml/min) at room temperature via the portal from nonentrapped enzyme by gel chromatography on Sepharose CL- vein, first with cacodylate buffer (0.1 M with 2% PVP and 0.4% NaNO2, 2B. Peroxidase activity associated with the liposomes was measured by pH 7.4) for 1 min to remove blood, subsequently with 2% glutaraldehyde a modification of the method of Steinman and Cohn (34), as described in 0.1 M cacodylate buffer for 7 to 8 min to fix the tissue, and finally by Roerdink era/. (28). The diammonium salt of 2,2'-azinodi(3-ethylbenz- again with the cacodylate buffer to remove the fixative. Vibratome thiazolin sulfonic acid) was used as a chromogen. Latency of the en sections (30 to 40 Mm; Oxford vibratome sectioning system) were trapped enzyme was defined as: thoroughly washed with PBS (137 mw NaCI, 2.6 FTIMKCI, 6.4 HIM Na2HPO4, and 1.4 mKH2PO4, pH 7.4) and incubated, for the ultrastruc Total enzyme activity - Free enzyme activity tural demonstration of peroxidase reaction with 0.075% 3,3'-diamino- x 100% Total enzyme activity benzidine tetrahydrochloride and 0.001% H2O2 in PBS, for 30 min in the dark at 37Â° (10). The sections were washed in PBS and cacodylate Total and free enzyme activities refer to the activities in the presence buffer (0.1 M with 6.8% sucrose, pH 7.4) and postfixed in 1% OsO4 in and absence of 1% Triton X-100, respectively. Latency for the vesicles cacodylate buffer, dehydrated in an ethanol series, and embedded in used in this study was 94%. Epon 812. Semithin (1 ÃŸm)and ultrathin sections were cut (LKB ultra- Animals. Three-month-old female C57BL x Gr F, mice were obtained tome). Semithin sections were stained with toluidine blue. From these locally. A B-lymphosarcoma, which originated spontaneously in a C57BL sections, areas were selected for ultrathin sectioning. Ultrathin sections mouse (7), was transplanted weekly by i.p. injection of 106 tumor cells. were stained with uranyl acetate and lead citrate and examined in a Seven days after inoculation, the spleen was significantly enlarged (from Philips EM 300 electron microscope. approximately 100 to approximately 800 mg) and consisted almost entirely of tumor cells. The lymphosarcomatous spleen was excised and squeezed through nylon gauze in 10 parts of ice-cold 0.9% NaCI solution. RESULTS The single-cell suspension thus obtained was used for tumor transplan 125l-labeled PVP was chosen as a marker of the aqueous tation. At the third day after i.p. injection, tumor cells were present in liver compartment because it is metabolically inert and in a free form and spleen, as was observed by light microscopy. In the liver, the tumor is taken up by the liver and spleen only at very low rates (9). cells were located predominantly in the sinusoids, i.e., in direct contact Once inside cells, it is also released very slowly (18); thus, 125I- with the blood compartment. In the spleen, they were mainly found in label found associated with liver and spleen can be taken to the marginal zone of the follicles of the white pulp. During the tumor reflect the uptake of vesicles without taking into consideration development, the tumor cells in the spleen concentrically invaded the their subsequent metabolic fate. With normal as well as lympho- white pulp, obliterating the red pulp. At 7 days after inoculation, a diffuse sarcoma-bearing mice, we confirmed our earlier observations of tumor cell proliferative pattern was found in the spleen, and massive infiltrations were found in the liver, lungs, bone marrow, and kidneys. rats (9), showing that in both blood and the peritoneal cavity, 125l-labeled PVP remained liposome associated for at least 6 hr Macroscopically. no tumor growth or ascites production was observed in the peritoneal cavity at any day. Tumor-bearing mice died 8 or 9 days following an i.p. injection of PVP-containing liposomes. Only very after i.p. inoculation of 106 lymphosarcoma cells. small amounts of radioactivity were free or cell associated (re Tissue Distribution Experiments. Normal or tumor-bearing mice sults not shown). These findings are compatible with our earlier (either 1 or 6 days after inoculation of 106 lymphosarcoma cells) were contention that circulating macrophages are not extensively in given i.p. or i.v. (lateral tail vein) injections of various doses of liposomes. volved in the transport of liposomes from peritoneal cavity to Four hr after injection, the animals were anesthetized with ether, and the blood. peritoneal cavities of mice that received an i.p. liposome injection were Vesicles i.v. Injected. Vesicles containing 125l-labeledPVP rinsed with 2 ml of 150 ITIM NaCI:5 mw Tris-HCI, pH 7.4, in order to were injected i.v. into normal and tumor-bearing mice in doses recover the liposomes. A blood sample (0.2 ml) was taken from the vena cava, and the liver was briefly perfused with 150 HIM NaCI:5 mw Tris- of 50 to 500 Mmol of liposomal lipid per kg of body weight. The HCI, pH 7.4, to remove the blood. Various organs were excised, rinsed mice were sacrificed 4 hr after injection, and the distribution of in buffer, blotted on paper, and weighed. Radioactivity was measured in radioactivity between blood and various organs was determined whole organs. For calculation of the proportion of the dose present in (Table 1). In all cases, 70 to 80% of the injected dose was the blood, the total blood volume of the mice was taken as 2 ml. In some recovered in blood, liver, and spleen. With increasing liposome experiments, mice were treated with lanthanum by i.v. injection of 0.4 dose, larger proportions remained in the circulation after 4 hr fjmol of La(NO3)3 (Merck, Darmstadt, Germany) in 0.2 ml of 150 HIM and, concomitantly, smaller fractions were found associated with NaCI:5 mw Tris-HCI, pH 7.4, 24 hr prior to the injection of the liposomes. liver and spleen. Control animals received only the NaChTris. Reticuloendothelial System Function Assay. Reticuloendothelial ac One day after injection of tumor cells, the mice were clearly tivity was measured by determining the rate of elimination of i.v. injected inhibited in their capacity to eliminate liposomes from the blood. carbon from the blood. Mice received 0.2 ml of a 0.55% suspension of The increase in the level of circulating liposomes in the tumor- India Ink 4415 (Higgins Ink Co., Newark, N. J.) in 150 HIM NaCI:5 mw bearing mice corresponded to a decrease in hepatic uptake. It is Tris-HCI, pH 7.4). Three and 8 min after injection of carbon, groups of interesting to note that, in the tumor-bearing mice, the splenic mice were anesthetized with ether, and blood was sampled from the component of blood clearance became of relatively more impor vena cava. Blood (0.1 ml) was diluted with 2.9 ml of distilled water, and tance than in normal mice. In tumor-bearing mice, the amount of

2928 CANCER RESEARCH VOL. 43

Table 1 Vesiclesinjected i.v. in normal and tumor-bearingmice, dose-dependentblood elimination,and organ distribution The vesicles containing 125l-labeledPVPwere injected i.v. at doses varying from 50 to 500 ^mol/kg into normal mice or into lymphosarcomatousmice 23 hr or 6 days after inoculation of tumor cells. In one set of experiments, mice received 0.4 /imol of La(N03)3i.v. 24 hr prior to injection of liposomes. organ(s)Normal Mean % of injected dose per whole

miceDose125nmol/kg mice1 -day tumor mice125iimol/kgtumor mice500 iimol/kg ^mol/kg 25 Â¿Â¿mol/kg250 /Â¿moi/kg Â¿imol/kg ^mol/kg ! Mmol/kg (7f500 (3)50 (3)1 (7) (4)500 (11)La3*-treated (4)6-Day (3)500 (3) Â±0.2655.6 BloodLiverSpleenLungsHeartKidneys0.8Â±10.134.7 Â±2.732.4 Â±8.121 Â±1.914.1 .3Â±7.410.3 Â±0.042.8 Â±7.719. Â±5.513.6 Â±6.840.8 .8Â±4.630.0 Â±3.019.7 Â±2.212.2 Â±3.525.4 7 Â±1.90.5 Â±1.60.7 Â±5.84.2 Â±4.22.7 Â±2.62.3 Â±1.61.6 Â±1.41.0 Â±0.10.0 Â±0.30.1 Â±1.50.1 Â±0.70.2 Â±0.30.3 Â±0.40.3 Â±0.10.0 Â±0.00.1 Â±0.10.8 Â±0.10.5 Â±0.11.0 Â±0.11.3 Â±0.21 Â±0.00.2 Â±0.123.9 Â± 0.23.6 Â±0.329.3 Â±0.246.5 Â±0.361 .7 Â±0.656.617.010.91.20.211.82.74.30.30.11.3 0.31.2 Â±0.120.628.722.11.60.12.92.71.20.10.00.6Â±0.1 a Numbers in parentheses, number of mice given injections. " Mean Â±S.D. liposomes deposited in the lung also had increased. Â¿000 liver spleen For the major target organs, liposome uptake, expressed in absolute amounts rather than as percentage of the injected dose, is plotted as a function of the injected dose in Chart 1. Uptake in both liver and spleen increased with the injected dose, with a Ã•3000 tendency to reach a saturation level, particularly in the 1-day tumor mice. Clearly, the maximal uptake capacities in these animals were 3- to 4-fold times lower than those in the controls. Six days after inoculation of the tumor cells, the fraction of the 0 2000- administered dose eliminated in 4 hr in tumor-bearing mice was very similar to that in normal mice (Table 1; Chart 1). The total liver uptake in the tumor-bearing mice was slightly decreased 1000 compared to that of normal mice, but the accumulation in the spleen was increased. However, as a result of a substantial increase in the weight of the liver (from approximately 1 to 1.4 g) and, particularly, the spleen (from approximately 100 to 700 250 500 250 500 mg), the accumulation per g of tissue was decreased significantly liposome dose lpmo(/kg) (not shown). This would argue against an extensive involvement Chart 1. Total uptake of i.v. injected liposomes in liver and spleen of normal of tumor cells per se in the accumulation of liposomes in these mice and tumor-bearing mice. REV, containing 125l-labeledPVPin the aqueous tissues, particularly because the increased weight correlates with compartment, were injected i.v. at 50 to 500 nmol/kg. Four hr after injection, the tissues were removed, and radioactivity was measured in the whole tissues. The an increase in the number of tumor cells, as was found morpho results are expressed as nmol of lipid (calculated from the amount of 125l-labeled logically. PVP encapsulated per nmol of lipid)taken up per whole tissue. A, normal mice; x, 1-day tumor mice; â€¢,tumormice, 6 days after inoculation of the lymphosarcoma The increased circulation time of the liposomes in mice with a cells. Bars, S.D. 1-day tumor and the concomitantly decreased hepatic uptake are indicative of a decreased activity of the reticuloendothelial Table 2 system in such animals. This was confirmed by a direct meas Effect of lymphosarcoma cells on RESactivity Lymphosarcomacells (106)were injected i.p. Reticuloendothelialsystem activity urement of the phagocytic index (by carbon clearance) in normal was measured as described in "Materials and Methods." and tumor-bearing mice (Table 2). The phagocytic index of 1- Phagocytic index8 day tumor mice was decreased compared to that of the control Control mice (5)" 0.045 Â±0.004C mice. Six days after inoculation with tumor cells, the animals had 1-Day lymphosarcoma mice (3) 0.023 Â±0.004 recovered from the depression of reticuloendothelial activity. The 6-Day lymphosarcoma mice (4) 0.099 + 0.010 carbon clearance was even substantially increased in those animals. La(NO3)3, known to depress the phagocytic activity of Phagocytic index = Kupffer cells in mice (20), caused a change in the tissue distri te â€”ti bution in normal mice very similar to that in the 1-day tumor mice Numbers in parentheses, number of animals. (Table 1). c Mean Â±S.D. Vesicles l.p. Injected. When the liposomes were injected i.p.. the results were essentially similar to those obtained with i.v.- The different doses of vesicles were injected in the same total Â¡njectedvesicles (Table 3). However, only in 1-day tumor mice volume (0.45 ml) and apparently were transported to the blood was the percentage of the dose remaining in the blood 4 hr after at a high rate since, even with the high dose, less than 10% was injection clearly dose dependent. Also, uptake in the liver and recovered from the peritoneal cavity after 4 hr. The recovery of spleen of these 1-day tumor mice was inhibited compared to the vesicles tended to be lower after i.p. than after i.v. injection. that of normal mice. Six days after inoculation with tumor cells, It is conceivable that a fraction of the vesicles in the peritoneal blood levels and tissue distribution were almost restored to cavity is not easily removed by washing and/or is taken up by control values. lymph nodes.

JUNE 1983 2929

Microscopie Visualization of Liposome-entrapped Horse the sinusoids was conspicuous, but they did not contain any radish Peroxidase Â¡nthe Liver. Peroxidase-containingREV peroxidase reaction product. Apparently, the tumor cells did not were injected into normal or lymphosarcoma-bearing mice. The take up liposome-encapsulated or free peroxidase whereas, in distribution of the peroxidase reaction product, as seen in the Kupffer cells, located in close proximity to the tumor cells, light microscope 2 hr after injection of the liposomes into a peroxidase reaction product was abundantly present. In Figs. 3 normal mouse, is shown Â¡nFig. 1. The reaction product was and 4, electron micrographs of the livers of a normal mouse and prominent in the Kupffer cells but could also be detected in a 6-day tumor mouse that received peroxidase-containing lipo endothelial cells and hepatocytes. When a similar dose of free somes are shown. Peroxidase activity was located in membrane- peroxidase was injected, mixed with empty liposomes, peroxi bound vacuoles in the cytoplasm of Kupffer cells. In agreement dase reaction product was detected only in endothelial cells, and with our light microscopic observations, no peroxidase activity no reaction product was found in Kupffer cells or hepatocytes was detected in the tumor cells. Light and electron microscopic (not shown). The endogenous peroxidase activity of Kupffer cells observations on the localization of the peroxidase reaction prod could easily be distinguished from the exogenous enzyme by uct in the livers of 1-day tumor mice (results not shown) were virtue of the essentially different localization (nuclear envelope the same as in normal and 6-day tumor mice. and endoplasmic reticulum) of the endogenous enzyme. Six days after injection of the tumor cells, the distribution of DISCUSSION peroxidase reaction product in the liver was identical to that observed Â¡nnormalmice (Fig. 2). The presence of tumor cells in Our results confirm those of others (2, 12, 13, 32) in that i.v.-

Table 3 Vesicles injected i.p. and organ distribution Vesicles containing 125l-labeled PVP were injected i.p. at 125 and 500 ^mol/kg into normal mice, 1-day tumor mice, and 6-day tumor mice. Four hr after injection, the peritoneal cavity was rinsed with 150 mu NaCI:5 mM Tris-HCI (pH 7.4), a blood sample was taken from the vena cava, and liver and spleen were removed and weighed. Radioactivity was determined Â¡nthe whole organs. Tissue radioactivity (% of injected dose per whole organ) Dose mol/kg)Normal (tÃ cavity3.0 Â±0.2a mice 125 Â± 2.2 Â±1.7 Â±0.6 500 11.3 Â±3.0 39.9 Â± 4.4 13.7 Â±2.1 6.0 Â±2.0 1-Day tumor mice 125 19.7 Â±8.2 16.7 Â± 2.9 22.3 Â±3.2 8.0 Â±2.8 500 51.7 Â±3.6 10.6Â±4.836.4 14.3 Â±2.6 7.2 Â±2.4 6-Day tumor mice 125 1.1 Â±0.2 Â± 1.9 15.9 Â±1.7 5.6 Â±2.2 2.0" 500Blood0.9 6.4 Â±5.7Liver46.633.4 Â±12.5Spleen14.418.6 Â±3.4Peritoneal6.0 Â± Mean Â±S.D.of 5 mice.

V -*â€¢., Â¿*, J ec ^â€¢"iL*â€¢ . / "â€¢A.-. SÃ¤Ã„^.> â€¢ . Â¿-:'* â€¢â€¢':â€¢ * + k.Â«.'â€¢â€¢Â» -u\, *-Y tc J \r' r '.- *&*'^...â€¢.Â» â€¢A r-|;Â». , \ _Â» k* â€¢_-!>â€¢'â€¢ _ '".' Af\eo â€¢.

M'. -v'. ?..\ :- â€¢ .'â€¢'

Fig. 1. Micrograph of normal mouse liver 2 hr after i.v. injection of liposomes with entrapped peroxidase (2.8 ^mo\ of liposomal lipid containing 0.13 mg of peroxidase). Peroxidase reaction product is conspicuously present in Kupffer cells (kc) but can also be detected in endothelial cells (ec) and in hepatocytes (bold arrows). Bar, 10 ^m. Fig. 2. Micrograph of the liver of a 6-day tumor mouse 2 hr after i.v. injection of liposomes with entrapped peroxidase (2.8 //mol of liposomal lipid containing 0 13 mg of peroxidase). Peroxidase reaction product is present in Kupffer cells (kc) but can also be detected in endothelial cells (ec) and in hepatocvtes Tumor cells (to> are devoid of peroxidase activity. Bar, 10 ^m. ,

2930 CANCER RESEARCH VOL. 43

Downloaded from cancerres.aacrjournals.org on September 25, 2021. © 1983 American Association for Cancer Research. Liposomes in Lymphosarcoma-bearing Mice injected liposomes are eliminated from the blood compartment noted that, for the large liposome dose, splenic uptake of lipo in a dose-dependent way. Higher doses are cleared more slowly, somes in 6-day tumor mice also prevailed over control values indicating that the clearance mechanism is saturable, a well- (Chart 1). The high phagocytic index for carbon in these animals known phenomenon for the clearance of large particles (6, 32). may, similarly, be caused by splenic macrophages whereas, in After i.p. injection, the elimination was less clearly dose depend control animals, mostly the Kupffer cells will be involved in carbon ent. The delaying effect of the transport from the peritoneal clearance. cavity to the blood, although small as compared to our previous The depression of Kupffer cell activity in the 1-day tumor mice observations of rats (9), may be responsible for this phenome causes a shift of liposome accumulation towards the spleen, non. Only in the 1-day tumor rats, in which reticuloendothelial especially at the lower lipid doses. Apparently, there is an over activity was depressed, was dose-dependent blood elimination flow of liposomes to the spleen, which in turn tends to become of i.p.-administered liposomes manifest. saturated at the higher lipid doses. There is a small but significant The slower clearance of high doses was paralleled by de increase of liposome accumulation in the lungs of 1-day tumor creased hepatic and splenic uptake, as might be expected. mice. It remains to be seen whether these liposomes are only Souhami ef al. (33), on the other hand, did not observe decreased physically trapped or are associated with macrophages outside liver uptake concomitant with decreased blood elimination, which the vascular bed. led these authors to suggest that the observed saturation of Morphological examination of the livers of mice that received blood clearance capacity was related to an extrahepatic site of liposome-entrapped peroxidase revealed that the peroxidase uptake. On the other hand, the size of the vesicles used by these reaction product was concentrated in the Kupffer cells but was authors (average diameter, 58 nm) may have allowed a larger also present in endothelial cells and hepatocytes. Comparison proportion of the liposomes to gain access to the hepatocytes with the adequate controls showed that the presence of peroxi through the open endothelial fenestrations (38). Rahman ef a/. dase in the Kupffer cells and presumably also in the hepatocytes (26) recently presented indirect evidence that small unilamellar must be attributed to uptake of liposome-entrapped enzyme vesicles preferentially associate with liver parenchyma! cells. since at this dose free peroxidase is seen only in endothelial Experiments from our own laboratory provided direct evidence cells. The extrusion procedure used for the preparation of the of high uptake of both lipid and entrapped solute of small liposomes yields a preparation in which a fraction of the lipo unilamellar vesicles in hepatocytes and low uptake in nonparen- somes will have a diameter smaller than 100 nm (35, 36). It may chymal cells.4 In the present study, we used liposomes with a be this fraction that is allowed access to the parenchymal cells mean diameter larger than 100 nm (36), so that most of the through the fenestrations in the endothelial sieve plates. Reaction liposomes are likely to be prevented access from and thus product in the endothelial cells could be due to the uptake of association with the parenchymal cells of the liver. It is conceiv free peroxidase that had leaked from the liposomes, as was also able that differences in the distribution between the cell types in suggested by Roerdink ef al. (28). the liver are responsible for the discrepancy between the results Our study suggests that if liposomes are to be used as a of Souhami ef al. and the data presented in this paper. carrier system in the administration of cytostatic drugs, their Clearance and organ distribution of i.v. injected liposomes advantage is to be found in their depot function, either intracel- were very similar in 1-day tumor mice and in lanthanum-treated lularly in cells such as Kupffer cells or extracellularly during mice. Both conditions probably involve a depression of Kupffer prolonged presence in the circulation, rather than in their uptake cell activity, as measured by carbon clearance in the tumor by tumor cells per se. In our model system, absolutely no uptake animals. A change in the reticuloendothelial function following was observed of liposomal contents by tumor cells localized in tumor cell challenge has been reported by various investigators the liver sinusoids in direct contact with the blood compartment. (22, 31) and may be due to tumor components in the serum or Modest uptake by hepatocytes and, more abundantly, by Kupffer to altered plasma opsonin levels. It is not likely that the (tempo cells in close proximity to the tumor cells was at the same time rary) depression in blood elimination of liposomes and of carbon clearly manifest. When considering liposomes as a circulating particles following the injection of tumor cells is caused by direct drug depot, one should recognize that attempts to prolong interaction of the tumor cells with the Kupffer cells. In our liposomal circulation time by means of blocking reticuloendo morphological studies, no evidence was found of the presence thelial system activity may invoke adverse side effects. Although of tumor cells in the liver before the third day after administration. we and others have observed that even such drastic blockade In 6-day tumor mice, elimination from the blood and hepatic as caused by lanthanides is transient and does not seem to uptake of liposomes were comparable to those in control ani cause any permanent damage (20, 27), too little is known about mals. Apparently, the tumor mice had recovered from the re the functioning tissue macrophages to assume that such treat ticuloendothelial depression. Surprisingly, in the 6-day tumor ment is essentially harmless. mice, carbon clearance was increased to an even higher value Finally, our results call for caution in tumor model studies with than that found in control animals. At present, we do not know liposomes, because the administration of a moderate dose of what causes this phenomenon. Saba and Antikatzides (31) re tumor cells may profoundly influence in vivo clearance kinetics ported that Walker 256 tumor cell-induced changes of reticu and disposition of liposomes. loendothelial carbon clearance in rats correlated well with ACKNOWLEDGMENTS changes in plasma opsonin levels. Thus, increased levels of opsonic protein in the plasma may be involved in the increased We thank Dr. A. W. T. Konings of the Laboratory of Radiopathology of this carbon clearance, as observed in our experiments. It may be university for making available to us the lymphosarcoma cell line used in this study. We also wish to thank the Netherlands Cancer Foundation. Koningin Wilhelmina Fonds, for generous financial support. The skillful secretarial work of Karin van Wijk and Rinske Kuperus is gratefully ' G. L. Scherphof, unpublished data. acknowledged.

JUNE 1983 2931

REFERENCES 20. Lazar, G. The reticuloendothelial-blocking effect of rare earth metals in rats. J. Reticuloendothel. Soc., 73: 231-237, 1973. 21. Mayhew, E., Papahadjopoulos, D., Rustum, Y. M., and Dave, C. Inhibition of 1. Abra, R. M., Bosworth, M. E., and Hunt, C. A. Liposome disposition in vivo. tumor cell growth in vitro and in vivo by 1-fi-D-arabinofuranosylcytosine en Effects of predosing with liposomes. Res. Commun. Chem. Pathol. Pharmacol.. trapped within phospholipid vesicles. Cancer Res., 36: 4406-4411,1976. 29:349-360,1980. 22. Otu, A. A., RÃ¼ssel,R. J., Wilkinson, P. C., and White, R. G. Alterations of 2. Abra, R. M., and Hunt, C. A. Liposome disposition in vivo. III. Dose and vesicle mononuclear phagocyte function induced by Lewis lung carcinoma in C57BL size effects. Biochim. Biophys. Acta, 666. 493-503, 1981. mice. Br. J. Cancer, 36: 330-340, 1977. 3. Antikatzides, G. T., and Saba, T. M. Kupffer cell clearance function following 23. Pagano, R. E., and Weinstein, J. N. Interactions of liposomes with mammalian intravenous tumor cell challenge. J. Reticuloendothel. Soc.. 22: 1-12, 1977. cells. Annu. Rev. Biophys. Bioeng., 7: 435-468,1978. 4 Bartlett, G. R. Phosphorus assay in column chromatography. J. Biol. Chem., 24. Papahadjopoulos. D., and Miller, N. Phospholipid model membranes. I. Struc 234:466-468.1959. tural characteristics of hydrated liquid crystals. Biochim. Biophys. Acta, 735: 5. BÃ¶ttcher. C. J. F., Van Gent, C. M., and Pries, C. A rapid and sensitive sub- 624-638, 1967. micro phosphorus determination. Anal. Chim. Acta, 24: 203-204, 1961. 25. Rahman, Y-E., Cemy, E. A., Tollaksen, G. L.. Wright, B. J., Nance, S. L., and 6. Bradfield, J. W. B. A new look at reticuloendothelial blockade. Br. J. Exp. Thomson, J. F. Liposome-encapsulated Actinomycin-D: potential in cancer Pathol., 67:617-623, 1980. chemotherapy. Proc. Soc. Exp. Biol. Med., 746: 1173-1176, 1974. 7. De Vries, M. J., and Vos, 0. Treatment of mouse lymphosarcoma by total- 26. Rahman, Y-E., Lau, E. H., and Wright, B. J. Application of liposomes to metal body X-irradiation and by injection of bone-marrow and lymph node cells. J. chelation therapy. In: B. H. Tom and H. R. Six (eds.), Liposomes and Immu- Nati. Cancer Inst., 21: 1117-1129,1958. nobiology, pp. 285-299. Elsevier/North-Holland BiomÃ©dicalPress, 1980. 8. Donald, K. J., and Tennent, R. J. The relative roles of platelets and macro 27. Roerdink, F., Dijkstra, J., Hartman, G., Bolscher, B., and Scherphof, G. The phages in clearing particles from the blood: the value of carbon clearance as involvement of parenchymal, Kupffer, and endothelial liver cells in the hepatic a measure of reticuloendothelial phagocytosis. J. Pathol., 117: 235-245,1975. uptake of intravenously injected liposomes. Biochim. Biophys. Acta, 677: 79- 9. Ellens, H., Morseli, H. W. M., and Scherphof, G. L. In vivo fate of large 89, 1981. unilamellar sphingomyelin-cholesterol liposomes after Â¡ntraperitonealand intra 28 Roerdink, F. H.. Wisse. E., Morselt, H. W. M., Van der Meulen, J., and venous injection into rats. Biochim. Biophys. Acta. 674: 10-18, 1981. Scherphof, G. L. Cellular distribution of intravenously injected protein-contain 10. Graham, R. C., and Kamovsky, M. J. The early stages of absorption of injected ing liposomes in the rat liver. In: E. Wisse and D. L. Knook (eds.), Kupffer cells horse-radish peroxidase in the proximal tubules of mouse kidney: ultrastruc and other liver sinusoidal cells, pp. 263-272. Elsevier/North-Holland BiomÃ©dical tural cytochemistry by a new technique. J. Histochem. Cytochem., 14: 291- Press, 1977. 302, 1966. 29. Rustum, Y. M.. Mayhew, E., Szoka, F., and Campbell, J. Inability of liposome 11. Gregoriadis, G., and Neerunjun, D. E. Treatment of tumor-bearing mice with encapsulated 1-J-D-arabinofuranosylcytosine nucleotides to overcome drug liposome-entrapped Actinomycin D prolongs their survival. Res. Commun. resistance in Li2io cells. Eur. J. Cancer, 77. 809-817, 1981. Chem. Pathol. Pharmacol., 10: 351-361, 1975. 30. Rusznyak, I., Foldi. M., and Szabo, G. Absorption of corpuscular particles from 12. Haynes, D. H., and Kang, C. H. Saturation of uptake of liposomes by the serous cavities. In: I. Rusnyak, M. Foldi, and G. Szabo (eds.), Lymphatics and reticuloendothelial system: possible use for increased tumor specificity. Ann. Lymph Circulation, pp. 478-487, 1967. N. Y. Acad. Sci., 308: 440, 1979. 31. Saba, T. M., and Antikatzides, T. G. Humoral mediated macrophage response 13. Kao, Y. J., and Juliano, R. L. Interactions of liposomes with the reticuloendo during tumour growth. Br. J. Cancer, 32: 417-482. 1975. thelial system. Effects of reticuloendothelial blockade on the clearance of large 32. Souhami, R. L. The effect of colloidal carbon on the organ distribution of sheep unilamellar vesicles. Biochim. Biophys. Acta, 677: 453-461. 1981. red cells and the immune response. Immunology, 22: 685-694, 1972. 14. Kataoka, T., and Kobayashi, T. Enhancement of chemotherapeutic effect by 33. Souhami, R. L., Patel, H. M., and Ryman. B. E. The effect of reticuloendothelial entrapping 1-fi-D-arabinofuranosylcytosine in lipid vesicles and its mode of blockade on the blood clearance and tissue distribution of liposomes. Biochim. action. Ann. N. Y. Acad. Sci., 308: 387, 1978. Biophys. Acta, 674: 354-371, 1981. 15. Kaye. S. B. Liposomesâ€”problems and promise as selective drug carriers. 34. Steinman, R. M., and Cohn, Z. A. The interaction of soluble horseradish Cancer Treat. Rev., 8. 27-50, 1981. peroxidase with mouse peritoneal macrophages in vitro. J. Cell Biol., 55: 186- 16. Kaye. S. B., Boden, J. A., and Ryman, B. E. The effect of liposome (phospho- 204, 1972. lipid vesicle) entrapment of Actinomycin D and Methotrexate on the in vivo 35. Szoka, F., and Papahadjopoulos, D. Procedure for preparation of liposomes treatment of sensitive and resistant solid murine tumors. Eur. J. Cancer, 77: with large internal aqueous space and high capture by reverse-phase evapo 279-289,1981. ration. Proc. Nati. Acad. Sei. U. S. A., 75: 4194-4198. 1978. 17. Kimelberg, H. K., and Mayhew, E. G. Properties and biological effects of 36. Szoka, F. C., Olson, F.. Heath, T., Vail, W., Mayhew, E., and Papahadjopoulos, liposomes and their uses in pharmacology and toxicology. Crit. Rev. Toxico!., D. Preparation of unilamellar liposomes of intermediate size (0.1-0.2 >im) by a 6:25-79, 1978. combination of reverse phase evaporation and extrusion through polycarbon 18. Kooistra, T. Endocytosis of proteins by liver cells. Ph.D. Dissertation, University ate membranes. Biochim. Biophys. Acta, 607: 559-571, 1980. of Groningen, The Netherlands, 1979. 37. Tyrrell, D. A., Heath, T. D., Colley. C. M., and Ryman, B. E. New aspects of 19. Kosloski, M. J., Rosen, F., Milholland, R. J., and Papahadjopoulos, D. Effect liposomes. Biochim. Biophys. Acta, 457: 259-302, 1976. of lipid vesicle (liposome) encapsulation of methotrexate on its chemothera 38. Wisse, E. An electron microscopic study of the fenestrated endothelial lining peutic efficacy in solid rodent tumors. Cancer Res., 38: 2848-2853, 1978. of rat liver sinusoids. J. Ultrastruct. Res., 37: 125-150, 1970.

2932 CANCER RESEARCH VOL. 43

Fig. 3. Electron micrograph of the liver of a normal mouse given an injection of liposome-encapsulated peroxidase. Details as in Fig. 1. Peroxidase reaction product is present in vacuoles in the cytoplasm of the Kupffer cell (kc). h. hepatocyte; s, sinusoid. Bar, 2.5 ^m.

JUNE 1983 2933

Fig. 4. Electron micrograph of the liver of a 6-day tumor mouse. Details as in Fig. 2. Peroxidase reaction product is present in vacuoles in the cytoplasm of the Kupffer cell (kc). Tumor cells (to) are devoid of peroxidase activity. Bar, 2.5 urn.

2934 CANCER RESEARCH VOL. 43

Downloaded from cancerres.aacrjournals.org on September 25, 2021. © 1983 American Association for Cancer Research. Effects of Liposome Dose and the Presence of Lymphosarcoma Cells on Blood Clearance and Tissue Distribution of Large Unilamellar Liposomes in Mice

Harma Ellens, Henriëtte W. M. Morselt, Bert H. J. Dontje, et al.

Cancer Res 1983;43:2927-2934.

Updated version Access the most recent version of this article at: http://cancerres.aacrjournals.org/content/43/6/2927

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

Reprints and To order reprints of this article or to subscribe to the journal, contact the AACR Publications Subscriptions Department at [email protected].

Permissions To request permission to re-use all or part of this article, use this link http://cancerres.aacrjournals.org/content/43/6/2927. Click on "Request Permissions" which will take you to the Copyright Clearance Center's (CCC) Rightslink site.