Antibiotic Transport in Bacteria

Abstract

Survival or destruction of intramacrophage pathogen Leishmania depends in part on modulation of their host cell phagosome, capabilities of the infected macrophages to present parasite antigen to the host's immune system. Macrophages house these parasites as amastigotes in the acidic phagolysosomal compartment. Leishmania phagolysosome is the potential site for processing and presentation of its antigen as well as being the target site for chemotherapy in leishmaniasis. It is thought that the parasites are killed from macrophage activation by lymphokines secreted from either helper T1 cells or CD8⁺ T cells. Characterization of both the host and parasite molecules in the compartment in the context of biogenesis of Leishmania-phagolysosome and processing of the parasite antigen by this compartment are discussed. Trafficking of different drugs and new agents through this compartment and their role in chemotherapy and necessity of developing new drug carrier are also stressed.     

Leishmania-infected macrophages sequester endogenously synthesized parasite antigens from presentation to CD4+ T cells

CD4⁺ T cell lines raised against the protective leishmanial antigens GP46 and P8 were used to study the presentation of endogenously synthesized Leishmania antigens by infected cells. Using two different sources of macrophages, the 14.07 macrophage cell line (H-2^k) which constitutively expresses major histocompatibility complex (MHC) class II molecules, and elicited peritoneal exudate cells, we found that cells infected with Leishmania amastigotes presented little, if any endogenously synthesized parasite antigens to CD4⁺ T cells. In contrast, promastigote-infected macrophages did present endogenous parasite molecules to CD4⁺ T cells, although only for a limited time, with maximal presentation occurring within 24 h of infection and decreasing to minimal antigen presentation at 72 h post-infection. These observations suggest that once within the macrophage, Leishmania amastigote antigens are sequestered from the MHC class II pathway of antigen presentation. This allows live parasites to persist in infected hosts by evading the activation of CD4⁺ T cells, a major and critical anti-leishmanial component of the host immune system. Studies with drugs that modify fusion patterns of phagosomes suggest that the mechanism of this antigen sequestration includes targeted fusion of the parasitophorous vacuole with certain endocytic compartments.     

Major histocompatibility complex class I presentation of ovalbumin peptide 257–264 from exogenous sources: protein context influences the degree of TAP-independent presentation

Peritoneal macrophages from C57BL/6 mice process antigens from bacteria or coated on polystyrene beads for presentation by major histocompatibility complex (MHC) class I molecules. To investigate this antigen processing pathway, peritoneal macrophages from homozygous TAP1^−/− mice, which lack the transporter associated with antigen processing (TAP) and are defective in presenting endogenous antigens on MHC class I, were used. TAP1^−/− or C57BL/6 macrophages were co-incubated with either bacteria or polystyrene beads containing the 257–264 epitope from ovalbumin [OVA(257–264)], which binds the mouse class I molecule K^b. The source of the OVA(257–264) epitope was either the Crl-OVA(257–264) (Crl-OVA) fusion protein, the maltose binding protein (MBP)-Crl-OVA fusion protein, native OVA or bacterial recombinant OVA (rOVA); Crl-OVA, MBP-Crl-OVA and rOVA were each expressed in bacteria, and Crl-OVA and MBP-Crl-OVA purified from bacterial lysates and native egg OVA were coated onto polystyrene beads. The data reveal that peritoneal macrophages from C57BL/6 and TAP1^−/− mice can process bacteria expressing Crl-OVA, MBP-Crl-OVA and rOVA as well as beads coated with native OVA, purified Crl-OVA, and purified MBP-Crl-OVA and present OVA(257–264) for recognition by OVA(257–264)/K^b-specific T hybridoma cells, albeit with different relative processing efficiencies. The processing efficiency of TAP1^−/− macrophages co-incubated with bacteria or beads containing Crl-OVA or MBP-Crl-OVA was reduced approximately three to five times compared to C57BL/6 macrophages, but OVA(257–264) was presented 100 times less efficiently when the source of OVA(257–264) was full-length OVA. Chloroquine inhibition studies showed a differential requirement for acidic compartments in C57BL/6 versus TAP1^−/− macrophages, which also depended upon the source of the OVA (257–264) epitope (Crl-OVA versus full-length OVA). These data suggest that TAP1^−/− and C57BL/6 macrophages may process Crl-OVA and full-length OVA in different cellular compartments and that the protein context of the OVA(257–264) epitope influences the extent of TAP-independent processing for MHC class I presentation.     

Epitope cleavage by Leishmania endopeptidase(s) limits the efficiency of the exogenous pathway of major histocompatibility complex class I-associated antigen presentation

The activation of CD8⁺ T cell responses is commonplace during infection with a number of nonviral pathogens. Consequently, there has been much interest in the pathways of presentation of such exogenous antigens for major histocompatibility complex class I-restricted recognition. We had previously shown that Leishmania promastigotes transfected with the ovalbumin (OVA) gene could efficiently target OVA to the parasitophorous vacuole (PV), with subsequent recognition by class II-restricted T cells. We now report the results of studies aimed at evaluating the PV as a route of entry into the exogenous class I pathway. Bone marrow-derived macrophages can present soluble OVA (albeit at high concentrations) to the OVA_257–264-specific T cell hybridoma 13.13. In contrast, infection with OVA-transfected Leishmania promastigotes failed to result in the stimulation of this hybridoma. This appeared unrelated to variables such as antigen concentration, parasite survival, and macrophage activation status. These results prompted an analysis of the effects of promastigotes on class I peptide binding using RMA-S cells and OVA_257–264. Our data indicate that the major surface protease of Leishmania, gp63, inhibits this interaction by virtue of its endopeptidase activity against the OVA_257–264 peptide. The data suggest that this activity, if maintained within the PV, would result in loss of the OVA_257–264 epitope. Although we can therefore draw no conclusions from these studies regarding the efficiency of the PV as a site of entry of antigen into the exogenous class I pathway, we have identified a further means by which parasites may manipulate the immune repertoire of their host.     

Antigen presentation by Leishmania mexicana-infected macrophages: Activation of helper T cells specific for amastigote cysteine proteinases requires intracellular killing of the parasites

Leishmania mexicana amastigotes proliferate in the phagolysosomes of mammalian macrophages. The parasites abundantly synthesize lysosomal cysteine proteinases, which are encoded by the lmcpb gene family. One of these genes was overexpressed in Escherichia coli, and the purified recombinant protein was used as an antigen to induce and establish a T helper 1 (T_h1) cell line. The T cells recognize epitopes shared by the native cysteine proteinases and the recombinant protein. Infected bone marrow-derived macrophages induced to express major histocompatibility complex class II molecules by interferon (IFN)-γ do not affect parasite viability. These macrophages fail to stimulate the proliferation of the T cell line. In contrast, strong T cell stimulation is observed after the parasites are killed by treatment with L-leucine methylester, or after activation of macrophages by IFN-γ and tumor necrosis factor-α. It is concluded that infected macrophages efficiently present this lysosomal Leishmania antigen once the parasites are inactivated and degraded. This observation may be of considerable relevance for the outcome of Leishmania infections provided that it can be extended to other parasite antigens.     

Antigen presentation by Leishmania mexicana-infected macrophages: activation of helper T cells by a model parasite antigen secreted into the parasitophorous vacuole or expressed on the amastigote surface

Leishmania are protozoan parasites which invade mammalian macrophages and multiply as amastigotes in phagolysosomes (parasitophorous vacuoles). Using L. mexicana and bone marrow-derived macrophages (BMM), the question is addressed whether infected BMM induced to express major histocompatibility complex class II molecules can present defined antigens to specific T helper type 1 cells. As a model antigen, a membrane-bound acid phosphatase (MAP), a minor protein associated with intracellular vesicles in amastigotes, was either overexpressed at the surface of the parasites or overexpressed in a soluble form leading to antigen secretion into the parasitophorous vacuole. Presentation of MAP epitopes by these three types of amastigotes was then compared for macrophages containing live parasites or amastigotes inactivated by drug treatment. It is shown that surface-exposed and secreted MAP can be efficiently presented to T cells by macrophages harboring live amastigotes. Therefore, the parasitophorous vacuole communicates by vesicular membrane traffic with the plasmalemma of the host cell. The intracellular MAP of wild-type cells or the abundant lysosomal cysteine proteinases are not or only inefficiently presented, respectively. After killing of the parasites, abundant proteins such as overexpressed MAP and the cysteine proteinases efficiently stimulate T cells, while wild-type MAP levels are not effective. We conclude that intracellular proteins of intact amastigotes are not available for presentation, while after parasite inactivation, presentation depends on antigen abundance and possibly stability. The cell biological and possible immunological consequences of these results are discussed.     

MHC I presentation of Toxoplasma gondii immunodominant antigen does not require Sec22b and is regulated by antigen orientation at the vacuole membrane

The intracellular Toxoplasma gondii parasite replicates within a parasitophorous vacuole (PV). T. gondii secretes proteins that remain soluble in the PV space, are inserted into PV membranes or are exported beyond the PV boundary. In addition to supporting T. gondii growth, these proteins can be processed and presented by MHC I for CD8⁺ T‐cell recognition. Yet it is unclear whether membrane binding influences the processing pathways employed and if topology of membrane antigens impacts their MHC I presentation. Here we report that the MHC I pathways of soluble and membrane‐bound antigens differ in their requirement for host ER recruitment. In contrast to the soluble SAG1‐OVA model antigen, we find that presentation of the membrane‐bound GRA6 is independent from the SNARE Sec22b, a key molecule for transfer of host endoplasmic reticulum components onto the PV. Using parasites modified to secrete a transmembrane antigen with opposite orientations, we further show that MHC I presentation is highly favored when the C‐terminal epitope is exposed to the host cell cytosol, which corresponds to GRA6 natural orientation. Our data suggest that the biochemical properties of antigens released by intracellular pathogens critically guide their processing pathway and are valuable parameters to consider for vaccination strategies.     

Dendritic cell/macrophage precursors capture exogenous antigen for MHC class I presentation by dendritic cells

Presentation of MHC class I antigens by professional antigen-presenting cells (APC) is an important pathway in priming cytotoxic T lymphocyte responses in vivo. This study sought to identify the nature of the professional APC responsible for indirect class I presentation by examining a special feature of professional APC, namely their ability to process exogenous forms of antigen for class I presentation. Incubation of highly purified bone marrow-derived precursor cells with chicken ovalbumin (OVA) led to the efficient presentation of the major class I-restricted OVA determinant by mature dendritic cells (DC), but not by macrophages (Mϕ) derived from the precursor population. DC as well as macrophages were, however, able to mediate class II presentation of OVA, suggesting that macrophages were deficient in class I processing but not in capturing exogenous OVA. The majority of mature DC, i.e. over 80 %, generated from the precursor cells pulsed with OVA, presented the class I OVA epitope. Upon maturation, class I presentation of OVA by DC was greatly reduced, suggesting that class I processing of exogenous antigen is modulated during DC maturation in a manner similar to class II antigen processing. This study shows that bone marrow-derived DC/Mϕ progenitors capture exogenous antigen for class I presentation, and that cells of the DC lineage can be functionally distinguished from cells of the macrophage lineage based on their ability to process exogenous antigen for class I presentation.     

Destructive proteolysis by cysteine proteases in antigen presentation of ovalbumin

Most native antigens require digestion by acidic proteases in order to be recognized in the context of major histocompatibility complex class II by T helper cells (Th). We have studied the roles of three different acidic proteases, cathepsin D, cathepsin B and cathepsin L, in the processing of ovalbumin (OVA) for presentation in the context of I-A^d. We report that digestion of OVA in vitro with the aspartyl protease cathepsin D generates the epitope OVA_322–336, which is recognized by I-A^d-restricted OVA-specific Th in the presence of paraformaldehyde-fixed antigen-presenting cells (APC). In contrast, digestion of OVA with the cysteine proteases cathepsin B and L not only failed to generate an epitope, but also destroyed OVA_322–336. In the presence of fixed APC expressing I-A^d, OVA_322–336 was protected from destructive proteolysis by cathepsin L. These results illustrate the dependence of epitope selection on the intracellular proteolytic environment in APC, and suggest that mechanisms must exist for protection of epitopes from destructive proteolysis in the processing compartments.     

The role of endosomes and lysosomes in MHC class II functioning

B cells, dendritic cells and macrophages present peptides derived from exogenous antigens to CD4⁺ T cells. After uptake, antigens are processed into peptides, bound to major histocompatibility complex (MHC) class II molecules and transported to the plasma membrane. Here, Hans Geuze reviews the current controversy surrounding the identity of the compartments that play a role in antigen processing and peptide loading.     

Antigen presentation by macrophages from bacille Calmette-Guérin (BCG)-resistant and -susceptible mice

We have compared the antigen-presenting capacity of macrophages from congenic BALB/c.Bcg^r and BALB/c.Bcg^s mice that differentially express MHC class II glycoproteins. Several different criteria were used to evaluate the presentation of a protein antigen, ovalbumin (OVA), including limiting the concentration of antigen or the numbers of macrophages, and using both native OVA and OVA peptide 323–339. No differences in the capacity of macrophages from Bcg^r and Bcg^s mice to present antigen to a OVA-specific T cell hybridoma were found. Splenic macrophages from BCG-infected congenic mice also induced an equivalent amount of IL-2 production by the T cell hybridoma. The relationship of these findings to other differences that have been attributed to Bcg are discussed.     

The nature of antigen in the eye has a profound effect on the cytokine milieu and resultant immune response

The eye is endowed with a number of mechanisms that protect it from immune-mediated injury. One such mechanism, termed anterior chamber-associated immune deviation (ACAID), evokes the antigen-specific, systemic down-regulation of Th1 responses to antigen inoculated into the anterior chamber of the eye. ACAID has been correlated with the selective production of IL-10 by the antigen-presenting cells (APC) and the development of a cross-regulatory Th2-like response. A small subset of antigens do not induce ACAID, but instead provoke IL-12 and normal Th1 immunity. Remarkably, all soluble antigens tested are capable of inducing ACAID; only cell-associated antigens do not induce ACAID. We hypothesized that the nature of antigen plays a decisive role in the resultant immune response. This hypothesis was tested with two well-characterized antigens, ovalbumin (OVA) and SV40 large T antigen (SV40 Lg T Ag). The soluble forms of OVA and SV40 Lg T Ag induced ACAID in both in vivo and in vitro models of the eye. In contrast, the particulate forms of these antigens, i. e. OVA passively absorbed onto inert latex beads (OVA-latex) and SV40 Lg T Ag expressed in two different cell lines, 99E1 and SV-T2, did not induce ACAID in either in vivo or in vitro models of the eye. In addition, the cytokine profiles of ocular APC pulsed with OVA or OVA-latex showed that soluble OVA induced the production of IL-10, whereas OVA-latex induced the production of IL-12. These data suggest that the nature of the antigen in the eye, whether soluble or particulate, is a crucial determinant in the resultant immune response. Moreover, they suggest a mechanism in which soluble antigens preferentially induce the release of ACAID-inducing IL-10 whereas particulate antigens preferentially induce the release of Th1-inducing IL-12 by responding APC.     

Trypanosoma cruzi infection does not impair major histocompatibility complex class I presentation of antigen to cytotoxic T lymphocytes

Trypanosoma cruzi (T. cruzi), the etiological agent of Chagas' disease, lives free within the cytoplasm of infected host cells. This intracellular niche suggests that parasite antigens may be processed and presented on major histocompatibility complex (MHC) class I molecules for recognition by CD8⁺ T cells. However, the parasite persists indefinitely in the mammalian host, indicating its success at evading immune clearance. It has been shown that T. cruzi interferes with processing and presentation of antigenic peptides in the MHC class II pathway. This investigation sought to determine whether interference in MHC class I processing and presentation occurs with T. cruzi infection. Surface expression of MHC class I molecules was found to be unaffected or up-regulated by T. cruzi infection in vitro. A model system employing a β-galactosidase (β-gal)-specific murine cytotoxic T lymphocyte (CTL) line (0805B) showed: (i) in vitro infection of mouse peritoneal macrophages or J774 cells with T. cruzi did not inhibit MHC class I presentation of exogenous peptide (a nine-amino acid epitope of β-gal) to the CTL line, (ii) in vitro infection of a β-gal-expressing 3T3 cell line (LZEJ) with T. cruzi did not inhibit MHC class I presentation of the endogenous protein to the CTL line and (iii) mouse renal adenocarcinoma cells infected with T. cruzi and subsequently infected with adenovirus expressing β-gal were able to present antigen to the β-gal-specific CTL line. These findings indicate that the failure of the immune response to clear T. cruzi does not result from global interference by the parasite with MHC class I processing and presentation. Parasites engineered to express β-gal were unable to sensitize infected antigen-presenting cells in vitro to lysis by the CTL 0805B line. This was probably due to the intracellular localization of the β-gal within the parasite and its inaccessibility to the host cell cytoplasm.     

Macrophage-T cell interaction in experimental visceral leishmaniasis: failure to express costimulatory molecules on Leishmania-infected macrophages and its implication in the suppression of cell-mediated immunity

The most important immunopathological consequence of infection with Leishmania seen in murine and human hosts is the suppression of T cell-mediated immune responses to both mitogens and leishmanial antigens. It has been suggested that this suppression is mediated by macrophages, either by defective antigen processing and presentation or by the elaboration of suppressive mediators like prostaglandins. Optimum activation of T helper cells requires not only T cell receptor occupancy by the antigen-Ia complex, but also costimulatory signals provided by the antigen-presenting cells. We investigated the status of several costimulatory molecules on infected macrophages from both genetically susceptible BALB/c and resistant C57BL/6 mice. Our results demonstrate that upon parasitization, the macrophages become unable to deliver costimulatory signals to T helper cells, and that this effects is mediated by prostaglandins, as the inhibition of its synthesis by indomethacin recovered the defect. Upon infection with L. donovani, B7-1 expression was decreased, while ICAM-1 was marginally increased in BALB/c macrophages and there was no significant change in the expression of B7-1 and ICAM-1 in Leishmania-infected C57BL/6 macrophages. Expression of VCAM-1 did not change during infection. This selective alteration in the expression of costimulatory molecules on L. donovani-infected BALB/c macrophages was caused by the living parasite, as shown by the fact that killing of the parasites by stibogluconate led to no alteration in the levels of costimulatory molecules. We found that the change in B7-1 expression on the surface of infected macrophages resulted in the inhibition of delayed-type hypersensitivity-mediating functions of T helper cells from BALB/c mice. The results described in this study not only throw light on the possible mechanism of leishmanial pathogenesis, but also open up the possibility of immunotherapy of leishmaniasis by selective manipulation of costimulatory molecules.     

Leishmania infantum-specific T cell lines derived from asymptomatic dogs that lyse infected macrophages in a major histocompatibility complex-restricted manner

Protective immunity to leishmaniasis has been demonstrated in murine models to be mediated by T cells and the cytokines they produce. We have previously shown that resistance to experimental Leishmania infantum infection in the dog, a natural host and reservoir of the parasite, is associated with the proliferation of peripheral blood mononuclear cells (PBMC) to parasite antigen and to the production of interleukin-2 and tumour necrosis factor. In this study we show that PBMC from asymptomatic experimentally infected dogs produce interferon-γ upon parasite antigen-specific stimulation, whereas lymphocytes from symptomatic dogs do not. In addition, we report for the first time the lysis of L. infantum-infected macrophages by PBMC from asymptomatic dogs and by parasite-specific T cell lines derived from these animals. These T cell lines were generated by restimulation in vitro with parasite soluble antigen and irradiated autologous PBMC as antigen-presenting cells. We show that lysis of infected macrophages by T cell lines is major histocompatibility complex restricted. Characterization of parasite-specific cytotoxic T cell lines revealed that the responding cells are CD8⁺. However, for some animals, CD4⁺ T cells that lyse infected macrophages were also found. In contrast to asymptomatic dogs, lymphocytes from symptomatic dogs failed to proliferate and produce interferon-γ after Leishmania antigen stimulation in vitro and were not capable of lysing infected macrophages. These results suggest that both the production of interferon-γ and the destruction of the parasitized host cells by Leishmania-specific T cells play an important role in resistance to visceral leishmaniasis.     

The delivery of an antigen from the endocytic compartment into the cytosol for cross-presentation is restricted to early immature dendritic cells

Dendritic cells (DCs) are the only antigen-presenting cell population having a cross-presentation capacity. For cross-presentation, however, the intracellular antigen-processing pathway and its regulatory mechanism have not been defined. Here we report the differences in cross-presentation ability among murine bone marrow-derived immature DC, early immature day8-DC and late immature day10-DC, and fully mature day10 + lipopolysaccharide DC. Day8-DCs and day10-DCs show an immature phenotypic profile but are different in morphology. Day8-DCs can internalize an abundant volume of exogenous soluble ovalbumin (OVA) and result in cross-presentation. In contrast, day10-DCs are not able to cross-present, although they maintain efficient macropinocytosis. Exogenously internalized OVA antigens are stored in the endocytic compartments. The endocytic compartments are temporarily maintained at mildly acidic pH in day8-DCs and are rapidly acidified in day10-DCs after uptake of antigens. We show that OVA antigens accumulated in the endocytic compartments move into the cytosol in day8-DCs but do not in day10-DCs. NH(4)Cl-treatment, which neutralizes the acidic endocytic compartments and/or delays endosomal maturation, restores day10-DCs for transport the stored OVA antigens from the endocytic compartments into the cytosol. Diphenyleneiodonium chloride-treatment, which acidifies the endocytic compartments, decreases an amount of transported OVA antigen into the cytosol in day8-DCs. These data indicate that only the early immature stage of DC interferes with endosomal maturation, even after uptake of exogenous antigens, and then transports the antigens into the cytosol.     

Primary CD8+ T-cell response to soluble ovalbumin is improved by chloroquine treatment in vivo

The efficiency of cross-presentation of exogenous antigens by dendritic cells (DCs) would seem to be related to the level of antigen escape from massive degradation mediated by lysosomal proteases in an acidic environment. Here, we demonstrate that a short course of treatment with chloroquine in mice during primary immunization with soluble antigens improved the cross-priming of naïve CD8⁺ T lymphocytes in vivo. More specifically, priming of chloroquine-treated mice with soluble ovalbumin (OVA), OVA associated with alum, or OVA pulsed on DCs was more effective in inducing OVA-specific CD8⁺ T lymphocytes than was priming of untreated mice. We conclude that chloroquine treatment improves the cross-presentation capacity of DCs and thus the size of effector and memory CD8⁺ T cells during vaccination.     

Antigen requirements for efficient priming of CD8+ T cells by Leishmania major-infected dendritic cells

CD4(+) and CD8(+) T-cell responses have been shown to be critical for the development and maintenance of acquired resistance to infections with the protozoan parasite Leishmania major. Monitoring the development of immunodominant or clonally restricted T-cell subsets in response to infection has been difficult, however, due to the paucity of known epitopes. We have analyzed the potential of L. major transgenic parasites, expressing the model antigen ovalbumin (OVA), to be presented by antigen-presenting cells to OVA-specific OT-II CD4(+) or OT-I CD8(+) T cells. Truncated OVA was expressed in L. major as part of a secreted or nonsecreted chimeric protein with L. donovani 3' nucleotidase (NT-OVA). Dendritic cells (DC) but not macrophages infected with L. major that secreted NT-OVA could prime OT-I T cells to proliferate and release gamma interferon. A diminished T-cell response was observed when DC were infected with parasites expressing nonsecreted NT-OVA or with heat-killed parasites. Inoculation of mice with transgenic parasites elicited the proliferation of adoptively transferred OT-I T cells and their recruitment to the site of infection in the skin. Together, these results demonstrate the possibility of targeting heterologous antigens to specific cellular compartments in L. major and suggest that proteins secreted or released by L. major in infected DC are a major source of peptides for the generation of parasite-specific CD8(+) T cells. The ability of L. major transgenic parasites to activate OT-I CD8(+) T cells in vivo will permit the analysis of parasite-driven T-cell expansion, differentiation, and recruitment at the clonal level.     

Leishmania mexicana promastigotes induce cytotoxic T lymphocytes in vivo that do not recognize infected macrophages

The question is addressed whether antigens of Leishmania, a parasite residing in the endosomal compartment of macrophages, can be presented in the context of major histocompatibility complex class I molecules. We used E. coli β-galactosidase as a model antigen which can be expressed in high levels in L. mexicana promastigotes (L. mexicana-gal). Infection of BALB/c mice with L. mexicanagal induces β-galactosidase-specific cytotoxic T cells (CTL), which can be isolated using a β-galactosidase-expressing mastocytoma line as an antigen-presenting cell. These CTL recognize epitopes of β-galactosidase in the context of H-2K^d; however, they do not recognize L. mexicanagal-infected macrophages even after killing of the intracellular amastigotes by drug treatment or macrophage activation by lymphokines, although class I-peptide interaction and the presentation of endogenously produced antigens is normal. It is concluded that parasite antigens can induce a CTL response in vivo but that these CTL cannot recognize infected macrophages because the relevant epitopes cannot gain access to class I molecules. The effect of priming in vivo may be explained by the well-known but ill-understood phenomenon of cross-priming.     

Injection of detergent-denatured ovalbumin primes murine class I-restricted cytotoxic T cells in vivo

Complex adjuvant formulations have been used to introduce soluble protein antigens into the “endogenous” processing pathway and hence to elicit specific, major histocompatibility complex class I-restricted cytotoxic T lymphocyte (CTL) responses. We tested if simple modifications of a model protein antigen, i.e. ovalbumin (OVA), can render it immunogenic for murine class I-restricted CTL when injected into mice in soluble form. Injection of 1—100 μg native OVA into C57BL/6 (H-2^b) mice did not stimulate a class I-restricted CTL response. In contrast, immunization of mice with 0.5 to 10 μg sodium dodecyl sulfate (SDS)-or deoxycholate (DOC)-denatured OVA efficiently primed CD8⁺ CTL specific for the well-characterized K^b-restricted OVA_257—264 epitope. Gel-purified SDS-denatured OVA devoid of protein fragments and excess detergent efficiently stimulated a specific CTL response in vivo. OVA preparations denatured by heat or urea treatment were not immunogenic for murine CTL. Injection of non-treated or detergent-treated, antigenic OVA_257—264 peptide into mice did not elicit a CTL response. Thus, denaturation of OVA by simple detergents such as SDS or DOC dramatically enhances its immunogenicity for class I-restricted CTL but not all modes of denaturation are equally effective.