Search

The association of dioleoylphosphatidylcholine vesicles with P388 murine tumor cells was monitored using fluorescent lipid probes incorporated into the vesicle membrane. As shown by fluorescence microscopy and by fluorescence photobleaching recovery measurements, most of the vesicles were stably adsorbed on the cell surface with little incorporation of vesicle lipid into the cell membrane. After 10 minutes of incubation only about 15% as much solute (3H-sucrose) remained cell-associated, as would have been expected on the basis of adsorbed lipid. These data indicate that the major component of interaction between these vesicles and cells is stable adsorption to the cell surface with considerable leakage of vesicle contents (cf., Szoka et al., Biochim. Biophys. Acta. (1979, 1980]). However, data in the literature indicate effects of vesicle-encapsulated solutes on physiological processes in the cytoplasm of non-endocytosing cells. Moreover, our data [Weinstein et al., Science (Wash. D.C.) 1977.] show fluorescent staining of the cytoplasm of non-endocytosing cells incubated with vesicle-encapsulated carbox-yfluorescein. We confirmed this observation using carbox-yfluorescein purified according to Ralston et al., Biochim. Biophys. Acta. [1981]. Our findings indicate that small amounts of solute are transferred from vesicles into cells. It has been suggested that this transfer could occur by leakage at the cell surface and entry into the cell before diffusion out of the unstirred layer. We here analyze such a model and show the following: such an "unstirred layer" mechanism could not increase solute incorporation beyond that for an equivalent bathing concentration of free solute unless the diffusion coefficient for the vesicles were larger than that for the free solute. Such is certainly not the case. We conclude that the small amount of vesicle-mediated entry occurs either by a low level of fusion or by some other unknown mechanism.

A new technique for the quantification of cellular receptor-mediated endocytosis has been developed based on the analysis by flow cytometry of ligand-bearing liposomes containing the fluorochrome carboxyfluorescein. Carboxyfluorescein encapsulated at high concentrations in protein A-bearing liposomes is self-quenched. Binding and internalization of such liposomes by cells via antibodies directed towards membrane surface determinants results in the release of the liposome-encapsulated carboxyfluorescein into the cytoplasm causing an increase in cell-associated fluorescence. This increase can be quantified on a flow cytofluorometer.

5' and 2' stabilized (2'-5')(A)n analogues were synthesized by chemical modifications of enzymatically polymerized (2'-5')(A)n oligomers. They exhibit an increased antiviral activity after micro-injection in HeLa cell cytoplasm in agreement with their augmented metabolic stability. Their specific in vitro delivery to mouse leukemia cells after encapsulation in targetted liposomes leads to a transient inhibition of protein synthesis and an antiviral activity.

Mouse L929 cells were incubated with antibody-targeted liposomes containing oligodeoxyribonucleotides (oligomers). When the oligomer was a 15-mer complementary to the 5'-end region of the mRNA encoding the N protein of vesicular stomatitis virus, the cells became less permissive for multiplication of that virus; greater than 95% reduction of viral multiplication was achieved. Protection was not seen for "empty" liposomes, liposomes containing a random oligomer sequence, or liposomes containing a sequence complementary to the 5' end of c-myc protooncogene mRNA targeted by the same antibody, nor was it seen when the liposomes containing the N-protein antisense oligomer were targeted by an antibody that does not bind to L929 cells. Antibody-bearing liposomes containing antisense oligomers thus have a double specificity: a particular cell selected by the targeting antibody on the liposome and a particular mRNA in the cell selected by sequence complementarity with the liposome-encapsulated oligomer. Nonencapsulated oligomers are sensitive to nucleases and usually must be administered to cells at high concentrations. Oligomers encapsulated in liposomes resist DNase and are active in amounts 1-2 orders of magnitude lower than for those reported for unencapsulated oligomer sequences.

Specific receptor-mediated delivery of the contents of small, sonicated liposomes was studied with three murine tumor cell types: an IgG Fc receptor-negative nonphagocytic line (EL4); an Fc receptor-positive phagocytic line (P388D1); and an Fc receptor-positive nonphagocytic line (P388). The liposomes (formed from phosphatidylcholines, cholesterol, and dinitrophenyl-substituted phosphatidylethanolamine) contained carboxyfluorescein as a fluorescent marker and methotrexate as a pharmacologic agent. Binding and internalization of the liposomes were observed by fluorescence microscopy and measured by flow microfluorometry. The hapten-derivatized lipid was used as a binding point on the liposome for the antibody-combining site of the immunoglobulin. In the presence of IgG anti-dinitrophenyl, but not F(ab')2 or IgA anti-dinitrophenyl, liposomes bound to the Fc receptor-bearing cells. The liposomes underwent endocytosis by the P388D1 cells and, to a lesser extent, by the P388 cells. As measured by depression of [3H]deoxyuridine incorporation, methotrexate in IgG-opsonized liposomes had a much greater pharmacologic effect on the P388D1 cells than did the same amount in unopsonized liposomes or in free solution. This observation indicates that an appropriately chosen drug, incorporated in liposomes, can exert its effect on a cytoplasmic target after endocytosis. P388 cells showed a moderate effect of the drug in liposomes. Neither P388 nor P388D1 cells bound or ingested unopsonized liposomes, and the Fc receptor-negative EL4 line neither bound nor ingested opsonized liposomes. The data demonstrate specific interaction of opsonized liposomes with the cells' IgG Fc receptor.

Class II major histocompatibility complex molecules are integral membrane glycoproteins whose distribution is limited to certain tissues. To identify the molecular basis for such specificity, the chromatin configuration of the class II A alpha gene was examined in intact nuclei from various cell types. We show that there are three DNase I-hypersensitive sites in the A alpha gene. One of these sites, located near the promoter region, is specific to cells that normally express class II molecules at some stage of differentiation. Furthermore, this tissue-specific site appears to be under trans-regulatory control.

Antibody-bearing fluorescent liposomes containing methotrexate became bound to cells expressing determinants recognized by the antibody. The number of bound liposomes could be evaluated by fluorometry, and the internalization of liposomes was evaluated by the methotrexate-mediated inhibition of radio-labeled deoxyuridine incorporation. The effect of methotrexate transferred from the liposomes into the cells was a function not of the number of liposomes bound but of the nature of the cells and of the target molecules. Liposomes bearing antibodies with specificity for the H-2K or Mr 94,000 and 180,000 molecules were much more effective at drug delivery into T than B cells, even though these determinants were expressed by both cell types. B cells were more sensitive to the effect of methotrexate in anti-H-2 I-A and I-E liposomes than in anti-H-2K liposomes. Inhibition of the methotrexate effect by NH4Cl suggested that methotrexate entered the cell by endocytosis of the liposomes. The results are consistent with differential internalization of H-2K, I-A, I-E, and Mr 94,000 and 180,000 cell surface molecules by mitogen-stimulated T and B cells.

Previous studies revealed that antisense oligodeoxynucleotides to specific regions of the human immunodeficiency virus-1 (HIV-1) are potent inhibitors of replication of HIV-1 in vitro (Zamecnik, P. C., Goodchild, J., Taguchi, Y., and Sarin, P. S. (1986) Proc. Natl. Acad. Sci. U. S. A. 83, 4143-4146). We now report that antisense RNA, synthesized in vitro using T7 and SP6 RNA polymerase, displayed an anti-HIV-1 effect in the HTLV-IIIB/H9 system in vitro. Treatment of HIV-1-infected H9 cells with viral env region antisense RNA encapsulated in liposomes targeted by antibodies specific for the T cell receptor molecule CD3 almost completely inhibited HIV-1 production. The viral env segment covered a part of exon II of HIV-1 tat gene. No anti-HIV activity could be detected with similarly targeted liposome-encapsulated sense env RNA or with pol RNA synthesized in either the sense or antisense orientations, or with env region antisense RNA free in solution, or encapsulated in liposomes in the absence of the targeting antibody. A semiquantitative evaluation revealed that 4000-7000 RNA molecules became cell-bound in targeted liposomes; the half-life of the intracellularly present hybridizable antisense env RNA was approximately 12 h. Western blots showed that antisense env RNA suppressed tat gene expression by approximately 90% and gp160 production by 100%. These data were confirmed by immunoprecipitation studies. Northern blots (using an env probe) demonstrated the existence of all major HIV RNA species (9.3-, 4.3-, and 2.0-kb mRNA) in HIV-infected cells treated with antisense env RNA although at a reduced level. We conclude that the antisense env RNA inhibited viral protein production at the translational level.

In a study designed to evaluate the potential for in vivo manipulation of the circulation and tissue distribution of injected liposomes, mice were passively injected with antidinitrophenyl (anti-DNP) monoclonal antibodies of the IgG2a or IgG2b subclasses or were immunized with the nitrophenyl hapten bound to a protein carrier. They were then injected i.v. with 125I- and carboxyfluorescein-labeled, DNP-bearing liposomes. Circulation time of the DNP-bearing liposomes was markedly reduced in actively and passively immune mice, with increased deposition of liposomes in the liver. The increased clearance of liposomes could be abrogated by injection of a monoclonal antibody directed against the murine IgG Fc receptor (2.4G2). The results suggest that clearance of ligand-bearing reagent in the face of an immune response may be modified by specific immunologic manipulation in vivo.