Soluble proteins delivered to dendritic cells via pH-sensitive liposomes induce primary cytotoxic T lymphocyte responses in vitro.

Effective immunity to many infectious agents, particularly viruses, requires a CD8+ cytotoxic T lymphocyte (CTL) response. Understanding how to achieve CTL induction with soluble proteins is important for vaccine development since such antigens are usually not processed appropriately to induce CTL. In the present report, we have demonstrated that a potent primary CTL response against a soluble protein can be achieved by delivering antigen in pH-sensitive liposomes to dendritic cells (DC) either in vivo or in vitro. Since the pH-sensitive liposome delivery system is efficient and easy to use, the approach promises to be valuable both in the study of basic mechanisms in antigen processing, and as a practical means of immunization.

Prolonged survival of thymoma-bearing mice after vaccination with a soluble protein antigen entrapped in liposomes: a model study.

EG7-OVA cells are mouse thymoma EL4 cells stably transfected with the complementary DNA of chicken ovalbumin (OVA) and thus express OVA epitopes as a unique antigen. Cytotoxic T lymphocytes specific to OVA can be elicited by immunization of mice with OVA osmotically loaded into syngeneic splenocytes or entrapped in liposomes. Cytotoxic T lymphocytes thus induced can specifically cytolyse the EG7-OVA cells in vitro in an antigen-specific and major histocompatibility complex-restricted manner. In the present study, we have examined in this model system whether immunization with liposomal OVA can protect mice against tumors induced by EG7-OVA cells. Vaccination with OVA either entrapped in liposomes or osmotically loaded in the syngeneic splenocytes prolonged the survival of mice which had been challenged with EG7-OVA cells, but not those mice challenged with the parent EL4 cells. The antitumor effect was attributed to the induced OVA-specific cytotoxic T lymphocyte activity, since other forms of acquired immunity such as interaction of tumor cells with specific antibody could not be detected. Our results demonstrate that immunization with antigen incorporated in liposomes could be a useful means of inducing a protective antitumor response.

An improved method of loading pH-sensitive liposomes with soluble proteins for class I restricted antigen presentation.

We have recently shown that ovalbumin (OVA) entrapped in pH-sensitive liposomes could sensitize mouse thymoma cells for lysis by MHC class I-restricted cytotoxic T lymphocytes (CTL) (Reddy et al. (1991) J. Immunol. Methods, 141, 157-163). The present studies were designed to optimize the antigen delivery system. A simple freeze-thaw method was developed to load OVA into pH-sensitive liposomes, and the protocol was optimized in terms of the choice of buffer, pH and ionic strength of the medium, lipid composition, lipid and OVA concentrations and the number of freeze-thaw cycles. Under optimized conditions, approximately 25% of OVA could be entrapped in pH-sensitive liposomes at 172 micrograms protein/mg lipid. This compares to only about 5% entrapment (70 micrograms protein/mg lipid) using the previous method. OVA loaded to pH-sensitive liposomes using the improved method led to a sensitive measure of CTL activity. The approach promises to be suitable to measure CTL against less available soluble antigens such as viral proteins.

Class I restricted CTL recognition of a soluble protein delivered by liposomes containing lipophilic polylysines.

CD8+ cytotoxic lymphocytes recognize peptides derived from endogenous antigens complexed with class I major histocompatibility complex while CD4+ helper cells recognize peptides from exogenous antigens bound to class II MHC molecules. A soluble protein can be introduced into the class I pathway of antigen processing and presentation using an appropriate vehicle to deliver the antigen into the cytosol. Cationic liposomes containing lipophilic polylysine readily form complexes with an anionic, soluble protein ovalbumin. Mouse thymoma EL4 cells incubated with such complexes can be sensitized for killing by OVA-specific CTL effector cells. This method of target sensitization by a soluble antigen is more sensitive than the osmotic loading method previously reported.

pH sensitive liposomes provide an efficient means of sensitizing target cells to class I restricted CTL recognition of a soluble protein.

Exogenous antigens are normally endocytosed and enter the class II pathway of processing and presentation. It had been shown earlier that soluble antigen could be introduced into the class I pathway of processing and presentation by osmotic loading. In this report, we have demonstrated that OVA containing liposomes that destabilize on exposure to low pH, referred to as pH sensitive liposomes, could sensitize target cells to lysis by class I MHC-restricted OVA-specific CTL. However, OVA-containing pH insensitive liposomes, native OVA, or OVA subjected to the same protocol as was used to make the liposomes, failed to sensitize targets to OVA-specific CTL lysis. The pH sensitive liposomal approach was less toxic and more efficient (about 20-fold) in delivering than the osmotic loading approach. The pH liposome approach may prove valuable to study CTL recognition characteristics of less available proteins such as viral proteins.

Inhibition of herpes simplex virus replication in the mouse cornea by drug containing immunoliposomes.

Monoclonal antibody to HSV glycoprotein D was derivatized with palmitic acid and incorporated into liposomes. These immunoliposomes bound specifically to intact mouse corneas infected with HSV-1 in vitro. Furthermore, in yield reduction assays, anti-gD immunoliposomes loaded with acyclovir proved far more effective at inhibiting viral replication in the cornea than free drug or drug delivered in untargeted liposomes. Site-specific sustained release immunoliposomes of this type are potentially an improved vehicle for drug delivery in the treatment of ocular HSV.

Targeting of drug loaded immunoliposomes to herpes simplex virus infected corneal cells: an effective means of inhibiting virus replication in vitro.

The goal of our studies was to develop liposomes containing antiviral drugs and targeted with antiviral antibody (immunoliposomes) that would be effective at inhibiting replication of herpes simplex virus (HSV) in vitro. To achieve this, a monoclonal antibody to glycoprotein D of HSV was derivatized with palmitic acid and was incorporated into the lamellae of dehydration-rehydration vesicles. The gD containing immunoliposomes were shown to bind specifically to HSV-infected rabbit corneal cells in vitro, whereas control immunoliposomes prepared with a monoclonal antibody of the same class as the anti-gD failed to preferentially bind to virus-infected cells. The gD immunoliposome binding was inhibitable by pretreatment with rabbit anti-HSV serum but not by aggregated normal serum. Thus liposome binding was judged to represent an antigen-antibody reaction not binding to Fc receptors expressed by cells infected with HSV. Immunoliposomes loaded with iododeoxyuridine (IUDR) leaked drug rapidly at 37 degrees C, whereas acyclovir (ACV)-loaded liposomes still contained 48% of drug after 24 hr at 37 degrees C. The ACV-liposomes retained 44% of drug after 14 days at 4 degrees C. The ability of immunoliposomes to inhibit virus replication was compared with that of untargeted and empty liposomes by means of virus yield assays in vitro, Immunoliposomes loaded with either IUDR or ACV inhibited virus replication, although ACV-containing immunoliposomes were the most efficacious. The implications of our in vitro results for the development of immunoliposomes suitable for the treatment of ocular herpes infection are briefly discussed.

Liposomes as antigen delivery systems in viral immunity.

Since their first description, liposomes have been put to a wide variety of uses. Encapsulation or incorporation of antigens into liposomes markedly enhances the immunogenicity of the antigen. The type of immune response elicited by the liposome is found to depend on their chemical and structural properties. Immunization with viral glycoproteins encapsulated in liposomes has resulted in enhancement of the humoral response seen as a rise in the serum antibody levels which is several fold higher than that elicited by free antigen alone. Furthermore, liposome encapsulation of peptide antigens which are poor immunogens by themselves not only increases the immunogenicity of the peptide but also play an important role in delivery. The adjuvant effect afforded by liposomes can be further enhanced by the concomitant encapsulation of adjuvants like lipid A or muramyl tripeptide-phosphatidylethanolamine. Formation of liposomes with special characteristics such as pH sensitivity has resulted in the use of liposomes to deliver soluble antigen to the cytosol where it can undergo class I pathway of processing and presentation. Therefore liposomes could provide valuable tools to further understand the pathways of antigen processing and the requirements for induction of cell mediated immunity.

Induction of cytotoxic T lymphocytes in vivo with protein antigen entrapped in membranous vehicles.

Intravenous administration of APC such as splenocytes loaded with a soluble protein Ag has been shown to prime for an Ag-specific CTL response. It is thought that the APC directly presents loaded Ag in a MHC-restricted manner. However, it is demonstrated in this study that allogeneic splenocytes, MHC-free RBC, and even synthetic lipid vesicles (liposomes) after loading with OVA can elicit an OVA-specific and MHC-restricted CTL response. Biodistribution studies of these Ag-associated vehicles showed that the liver, spleen, and lung were the major organs responsible to scavenge these carriers, suggesting that the monocyte-macrophage system was involved in the Ag presentation for CTL. Depletion of macrophages by a specific macrophage killer, Cl2MDP, containing liposomes, abolished the CTL induction by immunization with OVA Ag carried by these vehicles except the induction by syngeneic splenocytes. Thus, the syngeneic splenocytes present Ag directly to the T cells, but other membranous vehicles carry the Ag to the host APC including macrophages, which then present it to the T cells. These results indicate that formulation of an Ag in membranous/colloidal vehicles may be a way to prime for a CTL response.

Lymphocyte activation by cell separation procedures.

Cell separation techniques normally used to obtain subpopulations of lymphocytes were shown, under certain conditions, to render the cells cytotoxic towards a number of target-cells including autologous lymphocytes. To cause cytotoxicity, it was necessary to pass cells through glass wool and/or nylon wool columns equilibrated with media containing fresh plasma or serum. Cells lost activation upon overnight in vitro culture or treatment with trypsin. In addition to direct cell cytotoxicity, cells released heat labile cytotoxic factors and antibody during the separation procedures. The implications of the results for the interpretation of cell separation studies designed to attribute immunological effects to one or another cell type were discussed.

Influence of peptide acylation, liposome incorporation, and synthetic immunomodulators on the immunogenicity of a 1-23 peptide of glycoprotein D of herpes simplex virus: implications for subunit vaccines.

A peptide corresponding to residues 1 to 23 of glycoprotein D of herpes simplex virus type 1 was chemically synthesized and coupled to a fatty acid carrier by standard Merrifield synthesis procedures. The resulting peptide-palmitic acid conjugate (acylpeptide) exhibited enhanced immunogenicity in mice as compared with that exhibited by the free form of the peptide. Incorporation of the acylpeptide into liposomes further increased the immunogenicity of the peptide, while inclusion of the immunomodulators muramyl tripeptide phosphatidylethanolamine and monophosphoryl lipid A into the same liposome stimulated the strongest response. The humoral immune responses induced by the acylpeptide-liposome construct were greater than those induced by peptide in Freund complete adjuvant, and cellular responses were equal. The acylpeptide-immunomodulator-liposome formulation also induced significant levels of protective immunity, although the immunity was less than that induced by herpes simplex virus infection. Acylated peptides, especially in liposomes, were taken up more effectively by draining lymph nodes, which possibly accounts in part for the enhanced immunogenicity of the peptides. Since the acylpeptide-immunoliposome formulation used was nontoxic, it could represent a useful way to enhance immunogenicity of subunit peptides used for vaccine purpose in humans and animals.

Destabilization of target-sensitive immunoliposomes by antigen binding--a rapid assay for virus.

Interactions of antibody stabilized phosphatidylethanolamine (PE) immunoliposomes with Herpes Simplex virus (HSV) and virus infected cells were studied by detecting the immune-dependent lysis of liposomes. Employing PE immunoliposomes bearing anti-HSV glycoprotein D (gD) IgG, immune-specificity of these liposomes were documented by the sole ability of HSV and the HSV-infected L cells to induce immunoliposome lysis. In addition, inhibition of PE immunoliposome lysis by free anti-gD IgG, but not anti-HSV glycoprotein B IgG, indicated the target antigen specificity of these immunoliposomes. Based on these observations, alkaline phosphate encapsulated PE liposomes were used to directly detect HSV in fluid phase. This immunoliposome assay which does not require washing was shown to be very rapid and sensitive: 35pfu of HSV-1 in 5ul could be detected within 1.5hr.

Target-sensitive immunoliposomes: preparation and characterization.

A novel target-sensitive immunoliposome was prepared and characterized. In this design, target-specific binding of antibody-coated liposomes was sufficient to induce bilayer destabilization, resulting in a site-specific release of liposome contents. Unilamellar liposomes were prepared by using a small quantity of palmitoyl-immunoglobulin G (pIgG) to stabilize the bilayer phase of the unsaturated dioleoylphosphatidylethanolamine (PE) which by itself does not form stable liposomes. A mouse monoclonal IgG antibody to the glycoprotein D of Herpes simplex virus (HSV) and PE were used in this study. A minimal coupling stoichiometry of 2.2 palmitic acids per IgG was essential for the stabilization activity of pIgG. In addition, the minimal pIgG to PE molar ratio for stable liposomes was 2.5 X 10(-4). PE immunoliposomes bound with HSV-infected mouse L929 cells with an apparent Kd of 1.00 X 10(-8) M which was approximately the same as that of the native antibody. When 50 mM calcein was encapsulated in the PE immunoliposomes as an aqueous marker, binding of the liposomes to HSV-infected cells resulted in a cell concentration dependent lysis of the liposomes as detected by the release of the encapsulated calcein. Neither uninfected nor Sendai virus infected cells caused a significant amount of calcein release. Therefore, the release of calcein from PE immunoliposomes was target specific. Dioleoylphosphatidylcholine immunoliposomes were not lysed upon contact with infected cells under the same conditions, indicating that PE was essential for the target-specific liposome destabilization.(ABSTRACT TRUNCATED AT 250 WORDS)

In vivo cytotoxic T lymphocyte induction with soluble proteins administered in liposomes.

The in vivo induction of a CTL response usually requires that Ag be endogenously synthesized so that appropriate processing can occur. In most of the few examples where successful CTL induction was reported with proteins and peptides, unacceptable adjuvants or means of Ag formulation were used. In the present report, liposomes were used to incorporate the soluble proteins OVA and beta-galactosidase. This simple and convenient to use approach, which requires minimal amounts of Ag, results in priming for a CD8+ CTL response and the establishment of immunologic memory. The liposome approach may not only prove a convenient means of inducing CTL responses in vivo but may also be useful to study the mechanisms of Ag processing.

Priming for virus-specific CD8+ but not CD4+ cytotoxic T lymphocytes with synthetic lipopeptide is influenced by acylation units and liposome encapsulation.

Synthetic peptides of the herpes simplex virus glycoprotein B synthesized either as a free form or derivatized with one (PAM1) or three palmitic acids (PAM3Cys) were used to assess the in vivo priming efficacy of high affinity virus-specific CTL induction. The peptide and its derivatives were delivered in vivo with or without liposome encapsulation. Neither the free peptide nor the PAM1 derivative primed for high affinity virus specific CD8+ CTL induction, whether delivered via liposomes or not. On the other hand, the PAM3Cys derivative was able to prime for low levels of high affinity virus specific CD8+ CTL induction in the absence of liposome encapsulation. However, the efficiency of virus-specific CD8+ CTL induction with PAM3Cys derivative was enhanced following encapsulation in the liposomes. In contrast, all forms of the peptides induced both CD4+ T cell proliferative response as well as high affinity virus-specific CD4+ CTL. In addition, the efficiency of the PAM3Cys derivative to prime for CD4+ or CD8+ CTL was found to be influenced by the liposome encapsulation. When delivered via liposomes, the PAM3Cys derivative effectively primed for CD8+ CTL. However, liposomal delivery was not necessary for efficient priming for CD4+ CTL induction. Thus, both the acylation units as well as liposomal delivery appear to influence the in vivo priming of CD4+ and CD8+ T cell responses with synthetic peptides.

In vivo induction of anti-herpes simplex virus immune response by type 1 antigens and lipid A incorporated into liposomes.

To establish the requirements for a potential subunit vaccine against herpes simplex virus (HSV), we analyzed the effects on immunogenicity of incorporating detergent-extracted glycoprotein-enriched HSV type 1 (HSV-1) antigens into liposomes alone or with the adjuvant lipid A. Incorporating HSV-1 antigens into liposomes enhanced their immunogenicity for antibody production as detected by radioimmunoassay. Antibody levels to free and liposome-bound antigens were enhanced by administering lipid A as an adjuvant. The maximum immunogenic effect was obtained by incorporating lipid A into liposomes containing the HSV-1 proteins. Such liposomes induced secondary antibody responses higher than those engendered by virus infection. Whereas infectious virus induced cell-mediated immunity detectable by the delayed-type hypersensitivity reactions and cytotoxic T lymphocyte production, none of the liposome preparations induced cell-mediated immunity.

Cell-mediated immunity to herpes simplex virus: induction of cytotoxic T lymphocyte responses by viral antigens incorporated into liposomes.

The immunogenicity of inactivated herpes simplex virus type 1 (HSV-1) antigens incorporated into liposomes was measured by their ability to induce secondary anti-HSV-1 specific cytotoxic T lymphocyte (CTL) responses in splenocyte cultures from virus-primed mice. Such virus-specific CTL could be induced provided the liposomes contained virus along with plasma membrane antigen of the same H-2 type as that of the virus-primed responser cells. Responses did not occur in cultures stimulated with liposomes containing only viral antigens or with a mixture of liposomes composed respectively of lipid and virus with those composed of lipid and plasma membrane proteins. Moreover F1 responder cells stimulated with liposomes composed of virus and plasma membrane protein of one of the parental haplotypes produced CTL restricted in their cytotoxicity to infected targets of the same haplotype as was used in the liposome. These results show that liposomes can be used to induce anti-HSV-1 CTL with inactivated viral antigens but recognition of both viral and H-2 antigen is required for this process to occur in vitro. The implications of our findings to the preparation of subunit vaccines against HSV-1 are briefly discussed.

Role of macrophages and dendritic cells in primary cytotoxic T lymphocyte responses.

The successful induction of class I restricted cytotoxic T lymphocytes (CTL) responses with soluble non-replicating antigens relies upon vehicles which deliver antigen in vivo appropriately to antigen presenting cells (APC), which for CTL may be dendritic cells (DC). In this study, we have followed the distribution of liposomes and their incorporated antigen and compared the efficacy of splenic macrophages (Mo) and DC at inducing primary CTL responses in vitro. Our results show that whereas liposomes locate predominantly in the splenic red pulp and marginal zone locations, some of their soluble antigen contents redistribute to the central white pulp, comprising mainly DC and T cells. Such antigen redistribution was most apparent following administration of pH-sensitive liposomes. In comparisons of the APC activity of Mo and DC taken at various times post-injection, DC were always superior to Mo. However, if Mo were depleted prior to antigen exposure, DC were ineffective APC for CTL induction. Furthermore, addition of supernatant from OVA-liposome treated Mo to naive DC-T cell cultures in vitro resulted in OVA-specific T cell responses. These studies indicate a role for Mo in enhancing the antigen presenting function of DC.