Highly efficient immunoliposomes prepared with a method which is compatible with various lipid...

Vol. 165, No. 3, 1989 BIOCHEMICAL AND BIOPHYSICAL RESEARCH COMMUNICATIONS

December 29, 1989 Pages 1272-1278

HIGHLY EFFICIENT IMMUNOLIPOSOMES PREPARED WITH A METHOD WHICH IS COMPATIBLE WITH VARIOUS LIPID COMPOSITIONS*

Eric Holmberg', Kazuo Maruyamal, David C. Litzingeri, Stephen Wrighti, Mark Davis', George W. Kabalka', Stephen J. Kennel3 and Leaf Huang'

Departments of 'Biochemistry and 'Chemistry, University of Tennessee, Knoxville, TN 37996

3Biology Division, Oak Ridge National Laboratory, Oak Ridge, TN 37830

Received November 3, 1989

Summary: Monoclonal antibody was conjugated to N-glutaryl-phosphatidylethanolamine in the presence of octylglucoside by using N-hydroxysulfosuccinimide as a carboxyl-activation reagent. The conjugated antibody was then incorporated into liposomes by a simple dialysis method. The method is mild and is compatible with various lipid compositions of the liposomes. We have prepared immunoliposomes containing a lung endothelium-specific monoclonal antibody and showed excellent target binding ("75% injected dose) of the immunoliposomes in mouse. Immunoliposomes can be prepared to contain other acidic lipids such as phosphatidylserine and various amounts of cholesterol. The presence of 20% or more cholesterol in liposomes resulted in high level of target binding. We have used in these experiments a new radioactive lipid-phase marker, 11 lIn- DTPA-SA, which was very stable in vivo. The halflife of clearance in mouse -- exceeded 3 weeks. a 1989 Academic Press, Inc.

We have recently described an immunoliposome targeting system in mouse

(1). Monoclonal antibodies against a pulmonary endothelium antigen, gpl12,

have been conjugated to liposomes. Since gp112 is present in large amounts on

the lumenal surface of the capillary wall (21, intravenously injected

irmnunoliposomes gain direct access and bind efficiently to their target (1,3).

In the early study (l), antibody was conjugated to palmitic acid using N-

* The submitted manuscript has been co-authored by a contractor of the U.S.

Government under contract No. DE-AC05-840R12400. Accordingly, the U.S. Government retains a nonexclusive, royalty-free license to publish or reproduce the published form of this contribution or allow others to do SO,

for U.S. Government purposes.

Abbreviations: chol, cholesterol; GD- or In-DTPA-SA, Gadolinium or iridium diethylenetriaminepentaacetic acid stearylamide complex; HEPES, N-2-Hydroxy- ethylpiperazine-N'-2-ethanesulfonic acid; MES, 2-[N-Morpholinolethanesulfonic acid hemisodium salt; NGPE, N-glutaryl phosphatidylethanolamine; PBS, phosphate buffered saline; PC, Egg phosphatidylcholine; PS, Phosphatidyl- serine.

C1006-291X/89 $1.50 Copyright 0 1989 by Academic Press, Inc. All rights of reproduction in any form reserved. 1272


hydroxysuccinimide ester of palmitic acid (4). However, the conditions of the

reaction (37OC incubation at pH 8.0) often inactivate the monoclonal

antibodies. We have thus changed to a published method for antibody coupling

using preformed liposomes containing NGPE (5). Antibody is coupled to the

carboxylic groups of NGPE exposed on the liposome surface. Although the

method is mild (0-4°C incubation) and the immunoliposomes prepared showed

relatively high levels of target binding in mouse (3), it is not compatible

with those liposomes containing either acidic lipids or amino-lipids, or both.

Therefore, we have decided to modify the protocols such that immunoliposomes

containing various lipid compositions can be prepared under a mild condition.

MATERIALS AND METHODS

Materials: Rat monoclonal antibody 34A was purified from the ascites fluid as described (6.7). NGPE was synthesized accordins to (8). DTPA-SA and Gd-DTPA-SA have-been described (9j. . '

The synthesis of All phospholipids were

obtained from Avanti Polar Lipids. "IIn and lz51 were from New England Nuclear. Other chemicals were reagent grade. lz51-antibody (3~10~ wdl-19 protein) was prepared using the IODO-GEN (Pierce) method and purified by using BioGel P-4 (Bio-Rad, Richmond, CA) spin column. 'llIn-DTPA-SA was prepared by mixing 1 ul of 1111nC13-HC1 with 600 ul of 1 mM DTPA-SA which was dissolved in hot ethanol ("65°C) with brief sonication. MES buffer contained 10 mM MES and 150 mM NaCl, pH 5.5.

Conjugation of antibody with NGPE: NGPE dissolved in CHC13 was dried with N2 gas, and solubilized with octylglucoside in MES buffer to a final concentration of 824 ug/ml (NGPE to octylglucoside molar ratio = 0.07). Forty ul 1-ethyl-3-(3-dimethylaminopropyl) carbodiimide (0.25M in MES buffer) and 40 ~1 N-hydroxysulfosuccinimide (0.1 M in MES buffer) were added to 200 ul of above NGPE solution and then incubated for 10 min at room temperature. The mixture was neutralized by adding HEPES buffer (100 mM, pH 7.5) and 1N NaOH to pH 7.5. Monoclonal antibody 34A and a trace amount of lz51-labeled antibody were then added, and incubated for 8 hr at 4°C with gentle stirring. Lipid mixtures and trace amounts of 'llIn-DTPA-SA were mixed and evaporated free of organic solvent with N2 gas. The dried lipid film was vacuum desiccated and solubilized with octylglucoside (100 mM in PBS (pH 7.4), 1ipid:octylglucoside = 1:5, m/m). The resultant solution was mixed vigorously with the antibody conjugated with NGPE, and then the detergent was removed by dialysis in PBS for 12-18 hr at 4°C. The immunoliposomes were extruded 4 times through a stack of two Nuclepore membranes (pore size 0.4 um). The size of liposomes was measured using a Coulter N4SD sub-micron particle size analyzer (Hialeah, FL). The immunoliposomes were separated from the unbound antibody on a Bio- Gel A 1.5M (Bio-Rad) column. Peak liposome fractions were collected and diluted to 1 mg lipid/ml PBS (pH 7.4).

Preparation of liposome by sonication: The liposome was prepared according to (9). Briefly, an eguimolar quantity (33.3 mol %) of Gd-DTPA-SA (in methanol), , . PC, and chol in chloroform, and a trace amount of 'llIn-DTPA-SA in ethanol were mixed, then dried, vacuum desiccated, and resuspended in PBS (pH 7.4). The suspension was bath-sonicated for 1 hr to produce liposomes of

1273


approximately 200 nm in average diameter, which was verified by particle size analyzer. The liposome concentration was 33 mg lipids/ml.

In vivo biodistribution of liposomes: Two hundred ul of immunoliposomes or sonicated liposomes were injected into Balb/c mice (6-8 weeks old) via the tail vein. At the desired time, mice were anesthetized lightly with diethyl ether and bled by eye puncture. Blood was collected and weighed. The mice were sacrificed by cervical dislocation and dissected. Organs were collected and counted for "'In counts in a y-counter. The results are presented as percent of the total injected dose for each organ. The total radioactivity in the blood was determined by assuming that the total volume of blood was 1.3% of the body weight (10).

RESULTS AND DISCUSSION

Lipid composition is an important factor determining the biodistribution,

clearance rate and pharmacokinetics of the liposomes and their encapsulated

drugs (11). It is essential that immunoliposomes with various lipid

compositions can be prepared by a simple method. By first conjugating a

hydrophobic anchor to the antibody molecule and then incorporating the

modified antibody to liposomes allows a flexible use of any lipid compositions

for the construction of liposomes. This is because the liposomal lipids are

not directly exposed to the conjugation reagents. We have previously used

this strategy to construct immunoliposomes of various lipid compositions (12);

however, the reaction conditions (37OC incubation) often cause inactivation of

the antibody. The conjugation reactions described by Bogdanov et al. are much

milder; the antibody is incubated at 0-4OC at neutral pH (5). Data described

in this report indicate that the reaction can be carried out in a micellar

solution of NGPE in octylglucoside and the conjugated antibody can be

incorporated into liposomes by a simple dialysis protocol.

Fig. 1 shows the biodistribution of immunoliposomes prepared by this new

method. The immunoliposome preparation contained a very high number of

antibody molecules (antibody to lipid molar ratio = 1:460). These

immunoliposomes accumulated efficiently in the target organ, i.e. lung, at 76%

of the injected dose. In comparison antibody-free liposomes prepared with

identical method did not accumulate in lung at all. We conclude that the new

method is suitable for the generation of highly efficient, target-specific

irmnunoliposomes.

1274


LLNG BLOOD LIVER KlCNEY SPLEBI HEART

Fig. 1. detergent lipids) or

Biodistribution of immunoliposomes and bare liposomes prepared by dialysis method. Immunoliposomes (0) (100 ng antibody, 200 !4J

bare liposomes (m) (200 )lg lipids) labeled with illIn-DTPA-SA were in;ected i.v. Both liposome preparations contained PC:chol (65:35 molar ratio) and were 250 nm in size. Percent injected dose was measured 15 min. after injection. Bar is S.D. (n=3).

The immunoliposomes used in the experiment shown in Fig. 1 were composed

of PC and chol; both do not contain carboxylic or amino groups which could

potentially interfere with the antibody coupling reaction. Fig. 2 shows the

result of an experiment in which immunoliposomes composed of Pc:chol:Ps

(10:5:1 molar ratio) were prepared with the new method. Biodistribution of

lllIn-labeled immunoliposomes was measured at different' intervals after

injection. Substantial amount ("40% of injected dose at 15 min) of lung

accumulation was found. Lung accumulation of immunoliposomes shown in Fig. 2

is lower than that shown in Fig. 1. This is because the antibody content of

the two immunoliposome preparations were very different. We have previously

60 I

0 5 IO 15

TIME (min)

Fig. 2. Biodistribution of irmnunoliposomes containing PS. Immunoliposomes composed of PC:chol:PS (10:5:1 molar ratio) (13 Pg antibody, 200 Pg lipids, 270 nm in size) were injected i.v. Percent injected dose in lung (a), blood ( 6 ), liver (m) and spleen (A) were measured at indicated time intervals. Bar is S.D. (n=3).

1275


60 m ,60

0 20 35 50 MOL% CHOLESTEROL

Fig. 3. Effects of chol on antibody incorporation and lung accumulation of immunoliposomes. Immunoliposomes composed of PC and 0, 20, 35 or 50 mol% chol and the indicated lipid-to-antibody weight ratio (a). Immunoliposomes (150 pg lipid, 250-350 nm in size) labeled with IllIn-DTPA-SA were injected i.v. Percent injected dose in lung (M) was measured 15 min. after injection. Bar is S.D. (n=3).

shown that the target binding of inununoliposomes increases with increasing

number of antibody molecules per liposome (3). Immunoliposomes used in the

experiment shown in Fig. 2 contained PS which has a carboxylic group as well

as an amino group on the molecule. The data shown in Fig. 2 indicate that the

new method of immunoliposome preparation is compatible with a lipid

composition containing these functional groups.

To illustrate the fact that the lipid composition can be easily changed

with the new method, we have varied the chol content of the immunoliposomes

(Fig. 3). Change in chol content did not affect the efficiency of antibody

incorporation into the liposomes up to 35% chol; the lipid to antibody molar

ratio stayed about the same. However, inclusion of chol (20 and 35%) in the

immunoliposomes had enhanced the level of target binding as compared to the

chol-free immunoliposomes. This is probably due to the instability of the

chol-free liposomes in serum or plasma (13). Apolipoproteins, particularly

those found in HDL, readily insert into the liposome membranes and transfer

the liposomal lipids to the HDL particles (14). Inclusion of chol in the

liposomes effectively prevents this deleterious effect (15). Fig. 3 also

shows that too much chol (50%) is not favorable for the target binding. This

is probably due to a lower level of antibody incorporation into these chol-

1276


Table 1. Biodistribution of liposomes containing trace 'l'In-DTPA-SA

% Injected Dose

Organ Day 1 Day 4 Day 8 Day 12

Lung 5.5kO.2 l.OkO.2 1.3f1.4 1.6kO.7

Blood 0.5+0.1 0.1+0.0 0.1to.o 0.2io.o

Liver 55.8f3.7 51.3k4.8 48.1k2.0 43.Oi2.5

Kidney 0.4fO.l 0.4kO.l 0.4+0.1 0.5+0.0

Spleen 7.2fl.O 6.8kO.2 5.9fl.O 5.5t1.2

Heart O.l?O.O o.o+o.o o.o*o.o O.OfO.0

Total Recovered 69.5k7.4 59.6k4.0 55.923.6 50.8f4.4

Liposomes (6.6 mg) composed of Ga-DTPA-SA:PC:chol (l:l:l, m/m) with a trace of IllIn-DTPA-SA were injected i.v. on day 0. IllIn counts of various organs were corrected for radioactive decay. Mean f S.D. of data from 3 mice are shown.

rich liposomes. In all of these experiments we used a new radioactive, lipid

phase marker, i.e. 'llIn-DTPA-SA. "iIn is a y-emitting isotope which can be

easily counted in a y-counter. However, the metal-DTPA complex is H20 soluble

and is rapidly excreted in urine (16). We have attached two stearylamine

groups to the chelater and the modified, lipid-soluble, metal-chelater complex

readily incorporates into the liposome membranes. Furthermore, this marker is

very stable in the animal as shown by the data in Table 1. Antibody-free

liposomes were injected via tail vein into mice and the biodistribution of

"'In counts was assayed over a 12-day period. As can be seen in Table 1, the

counts were mainly found in liver and spleen and decreased very slowly over

this period. At 12 days after injection, there was still 11% of the day-l

dose in the liver, and 76% in the spleen. In fact, the decrease of counts in

liver paralleled the decrease of the total counts recovered from various

organs including lung, blood, liver, kidney, spleen and heart, indicating that

the marker is very slowly eliminated from the mice. The calculated halflife

of the marker elimination is approximately 22 days. To our knowledge, this is

1277


the most stable liposome marker ever reported in the literature, This superb

characteristic of the marker is important for the pharmacokinetic studies of

immunoliposomes which require the use of a non- or slowly metabolizable

marker.

In conclusion, we have developed a method which is suitable for preparing

irrmunoliposomes composed of various lipids and different amounts of antibody.

The in vivo biodistribution of the immunoliopsomes can be easily measured by --

using a stable, y-emitting, lipid-phase marker, i.e. IllIn-DTPA-SA. These

technical advances should allow us to rigorously study the pharmacokinetics of

immunoliposomes and their encapsulated drugs as a function of the lipid

composition.

Acknowledgments : This work is supported by NIH grants CA24553 and AI25834 to L.H., GM39081 to G.W.K, and NIH fellowship CA08331 to S.W. S.J.K. is supported by the Office of Health and Environmental Research, U.S. Department of Energy, under contract DE-AC05-840R21400 with the Martin Marietta Energy Systems, Inc.

1.

2.

3.

4.

5.

6.

7.

8.

9.

10.

11.

12. 13. 14. 15.

16.

REFERENCES

Hughes, B. J., Kennel, S., Lee, R. and Huang, L. (1989) Cancer Res. in press. Rorvik, M. C., Allison, D. P., Hotchkiss, J. A., Witchi, H. P. and Kennel, S. J. (1988) J. Histochem. Cytochem. 26, 741-749. Maruyama, K., Holmberg, E., Kennel, S. J., Klibanov, A. L., Torchilin, V. P. and Huang, L. (1989) submitted. Huang, A., Huang, L. and Kennel, S. J. (1980) J. Biol. Chem. 255, 8015-

8018. Bogdanov, A. A., Jr., Klibanov, A. L. and Torchilin, V. P. (1988) FEBS Lett. 231, 381-384. Kennel, S. J., Foote, L. J. and Lankford, P. K. (1981) Cancer Res. 41, 3465-3470. Kennel, S. J., Hotchkiss, J. A., Rorvik, M. C., Allison, D. P. and Foote, L. J. (1987) Exp. and Mol. Patho. 110, 47-52. Weissig, V., Lasch, J., Klibanov, A. L. and Torchilin, V. P. (1986) FEBS Lett. 202, 86-90. Kabalka, G. W., Buonocore, E., Hubner, K., Moss, T., Norley, N. and Huang, L. (1987) Radiology 163, 255-258. Wu, M. S., Robbins, J. C., Bugianesi, R. L., Ponpipom, M. M. and Shen, T. Y. (1981) Biochim. Biophys. Acta 674, 19-26. Senior, J. H. (1987) in "Critical Reviews TM in Therapeutic Drug Carrier Systems" Bruck, S. D. ed. pp. 123-193, CRC Press, Boca Raton. Wright, S. and Huang, L. (1989) Adv. Drug Delivery Rev. 3, 343-389. Kirby, C. and Gregoriadis, G. (1980) Life Sci. 27, 2223-2230. Bonte, F. and Juliano, R.L. (1986) Chem. Phys. Lipids 40, 359-372. Scherphof, G.L., Daemen, T. and Wilschut, J. (1984) in "Liposome Technology" Vol. 3 (Gregoriadis, G., ed.), Chap. 14, pp. 205-224, CRC Press, Boca Raton. Carr, D. H., Brown, J., Bydder, G. M. (1984) Lancet 1, 484-486.

1278

Highly efficient immunoliposomes prepared with a method which is compatible with various lipid...

Documents

Transcript of Highly efficient immunoliposomes prepared with a method which is compatible with various lipid...