A multidrug-resistance protein (MRP)-like transmembrane pump is highly expressed by resting murine T helper (Th) 2, but not Th1 cells, and is induced to equal expression levels in Th1 and Th2 cells after antigenic stimulation in vivo.

A transmembrane pump for organic anions was identified in resting murine T helper (Th) 2, but not Th1 lymphocyte cell clones, as revealed by extrusion of a fluorescent dye. Dye extrusion inhibition studies suggested that the pump may be the multidrug-resistance protein (MRP). The different expression of the pump in resting Th1 and Th2 cell clones correlated with their respective levels of MRP mRNA. The pump was inducible in Th1 cells by antigenic stimulation in vitro leading to equal expression in activated Th1 and Th2 cell clones. This suggested that dye extrusion might allow the detection of Th2 (resting or activated) or of activated Th1 cells ex vivo based on a functional parameter. To test this, mice were infected with Leishmania major parasites to activate L. major-specific T cells of either Th1 (C57BL/6 mice) or Th2 (BALB/c mice) phenotype: 2-3% of CD4+ lymph node T cells of both strains of mice extruded the dye, defining a cell subset that did not coincide with subsets defined by other activation markers. Fluorescence-activated cell-sorting revealed that the lymphokine response (Th1 or Th2, respectively) to L. major antigens was restricted to this dye-extruding subset.

Mycoplasma fermentans-derived lipid inhibits class II major histocompatibility complex expression without mediation by interleukin-6, interleukin-10, tumor necrosis factor, transforming growth factor-beta, type I interferon, prostaglandins or nitric oxide.

Mycoplasma cause several diseases in man and animals. Some strains can chronically infect humans, leading to fever or inflammatory syndromes such as arthritis, particularly in immunosuppressed patients. A set of pathogenicity factors shared by many mollicutes may be membrane components that activate macrophages to secrete cytokines and other inflammatory mediators. Mycoplasma-derived high molecular weight material (MDHM) is a macrophage-activating amphiphilic lipid which was purified from Mycoplasma fermentans. We studied the influence of MDHM on the expression of major histocompatibility complex (MHC) class II molecules by mouse resident peritoneal macrophages with an ELISA. Highly purified MDHM at 4 ng/ml and 0.8 microgram/ml crude heat-killed M. fermentans (concentrations chosen to give maximal responses) suppressed interferon (IFN)-gamma-dependent class II MHC induction when added simultaneously with IFN-gamma. MDHM was not toxic and did not result in loss of adherent cells. Kinetic data showed that MDHM first up-regulated, then down-regulated the expression of preformed class II MHC molecules, while expression of Mac-1 and F4/80 antigens remained constant. MDHM-dependent suppression of class II MHC molecule expression resulted in impaired antigen presentation to the helper T cell line D10.G4.1. We further attempted to identify hypothetical products of MDHM-stimulated macrophages as secondary mediators of class II MHC suppression such as were described for lipopolysaccharide (LPS)-stimulated macrophages. Type I IFN, prostaglandins and nitric oxide, all reported to cause down-regulation of class II MHC, could be excluded in this context. Of the cytokines tumor necrosis factor, interleukin (IL)-6, IL-10 and transforming growth factor-beta, only IL-10 inhibited class II MHC expression, although less effectively than MDHM. The involvement of IL-10 was ruled out, as no evidence for its MDHM-dependent formation could be found. Our data suggest that MDHM interferes with class II MHC expression by up-regulating its turnover, and at the same time, inhibits the formation of new class II MHC molecules.

Freshly isolated mouse 4F7+ splenic dendritic cells process and present exogenous antigens to T cells.

The antibody 4F7 was reported to recognize an epitope expressed on dendritic cells (DC) from various tissues. To study the ability of splenic 4F7+ dendritic cells to process antigen for presentation to CD4+ T cells, DC were enriched using a separation procedure avoiding overnight culture which could lead to an altered phenotype. These DC were used as antigen-presenting cells (APC) in stimulation cultures of major histocompatibility complex class II-restricted T cells. It was found that they induce antigen-dependent lymphokine production by T cells and therefore could present exogenous antigens. These processing takes place intracellularly, because fixation abrogates presentation to T cells. Moreover, antigen presentation needs intracellular processing within endo- or lysosomes as chloroquine-treatment prevents T cell activation. Titration of APC numbers revealed that contaminating APC most likely did not account for antigen-specific T cell activation by DC. No evidence was found for release of antigenic peptides or for partial antigen processing possibly done by cell surface located enzymes on DC. In conclusion, these results indicate that freshly enriched DC are able to process antigens similarly to other APC.

The I-Ab-restricted alloresponse of D10.G4.1 T cells is based on the recognition of an endogenous peptide.

The alloreactivity of T cells is thought to be based on the cross-reactive recognition of allogeneic major histocompatibility complex (MHC) molecules which have bound peptides derived from self antigens or, in the case of cultured T cells, from serum components. While studying the processing requirements of conalbumin (CA) that is recognized by D10.G4.1 T cells in combination with I-Ak molecules we also analysed the cross-reactive stimulation of clone D10.G4.1 T cells by allogeneic, I-Ab expressing stimulator cells which is shown here to be CD4 dependent. In order to distinguish between an endogenous or exogenous origin of a peptide that is presumably co-recognized with I-Ab different types of stimulatory/antigen-presenting cells (APC) were treated with drugs that are known to influence the processing and/or presentation of antigens. It was found that the alloreactive response of D10.G4.1 cells was abolished if the APC were treated with brefeldin A or inhibitors of protein biosynthesis. Under the same conditions neither the CA-specific response of D10.G4.1 cells nor the activation of control T cells by ovalbumin (OVA) or insulin was affected. On the other hand, the use of lysosomotropic agents or inhibitors of glycoprotein trimming had no influence on the ability of the APC to induce the alloresponse of D10.G4.1 cells, whilst the presentation of CA and other protein antigens by the APC was prevented. In addition, treatment of APC with pronase to remove surface MHC molecules or acidic buffer to remove peptides from the binding groove of MHC class II molecules at the surface of APC strongly diminished their ability to induce an alloresponse. However, this capacity was restored by incubating the APC for 2 hr in serum-free medium. These data indicate that the alloreactive response of D10.G4.1 cells is based on the recognition of newly synthesized endogenous peptide(s) in combination with I-Ab.

In vitro processing of insulin for recognition by murine T cells results in the generation of A chains with free CysSH.

Studies on the processing of insulin as an Ag for the presentation to MHC class II-restricted T cells revealed that the amino acid residues 1-14 of the insulin A chain are recognized by insulin-specific T cells. An A1-14 peptide containing three cys-residues that were protected by S-sulfonate groups still needed processing by APC for efficient presentation similar to native insulin. We suspected that reductive deblocking or opening of disulfide bonds that generates CysSH-residues may be an essential processing step for these Ag. Due to the instability of SH-groups it was not possible to test A chain peptides with free SH-groups in the usual way for processing-independent presentation by fixed APC. However, under acidic conditions (pH 5) during APC pulsing with the Ag we could demonstrate that the freshly reduced A1-14 fragment as well as reduced insulin are able to bind to Ia Ag and to stimulate appropriate T cells without further processing. Various substitutions of cys-residues by Ser within this peptide revealed that only CysA7 is critical for Ia binding and/or T cell recognition. In intact insulin, this residue links the A chain containing the T cell epitope to the B chain. Therefore, we propose that insulin processing is not dependent on proteolysis or on the generation of a conformational determinant but on the separation of A and B chains resulting in A chains whose cys-residues are converted into CysSH.

The endo/lysosomal protease cathepsin B is able to process conalbumin fragments for presentation to T cells.

The protein antigens conalbumin (CA) and ovalbumin (OVA) are known to require uptake into antigen-presenting cells (APC) for their presentation to major histocompatibility complex (MHC) class II-restricted T cells. In both cases proteolytic cleavage is thought to be a necessary step for the generation of the respective antigenic peptides. A specific inhibitor of the endosomal protease cathepsin B, Cbz-Phe-Ala-CHN2, blocks the presentation of both CA and OVA, whereas this inhibitor has no effect on the presentation of a processing-independent OVA peptide. Furthermore, the presentation of insulin, an antigen that needs processing but no proteolytic cleavage, is enhanced when cathepsin B is inhibited during antigen pulsing. When the APC were treated with an inhibitor of acid proteases, the CA response was not affected, while the presentation of OVA was diminished under these conditions. To estimate the relevance of these findings for the generation of the antigenic CA peptide, extracellular digestions of CA by cathepsin B were carried out. The fragment(s) present in these digests was recognized by T cells without further processing. Furthermore, the time-course of intra- and extracellular CA processing with respect to the capacity to stimulate T cells was similar. Taken together these data suggest that degradation by cathepsin B may be sufficient in vivo to generate the antigenic CA fragment. On the other hand, the blocking of cathepsin B does not appear to have an adverse effect on the general mechanisms of antigen presentation.

Presentation of insulin and insulin A chain peptides to mouse T cells: involvement of cysteine residues.

The requirements for insulin presentation and recognition by A alpha b A beta b- and A alpha b A beta k-restricted mouse T cells were studied using a variety of derivatives of the insulin A chain. It was found that A chain peptides with irreversibly blocked Cys residues are non-stimulatory for the T cells. This suggests that at least one of the Cys residues is essential for recognition. On the other hand, all A chain peptides containing Cys residues modified in a way reversible by reaction with thiols are stimulatory yet differ in antigenic potency. All these A chain derivatives including a 14 amino acid fragment require uptake by antigen presenting cells (APC) for efficient presentation. Differences in stimulatory potency between the A chain peptides derived from the same insulin appear to be mainly due to the efficiency of uptake and/or processing by APC. Based on these findings we propose that processing in the case of insulin and its A chain derivatives involves the reductive deblocking of Cys residues or the rearrangement of disulfide bonds apart from a possible proteolytic cleavage. The same may apply to other proteins if Cys residues in the presented peptides are important for the interaction with Ia or the T cell receptor.

Processing without proteolytic cleavage is required for recognition of insulin by T cells.

Beef insulin as well as a chymotryptic A-chain fragment [BI-A1-14(SSO3-)3] need uptake by antigen-presenting cells (APC) for efficient presentation in combination with major histocompatibility complex class II molecules to insulin-specific T cells. This could be shown by the inability of aldehyde-fixed APC to present these antigens to T cells. Furthermore, presentation of the insulin fragment as well as presentation of ovalbumin (OVA) was inhibited by treatment of APC with chloroquine, cerulenin or tunicamycin. This was not the case for a processing-independent OVA peptide. Treatment of APC during antigen pulsing with various protease inhibitors, active on all classes of proteases, did not block presentation of insulin although some of these reagents did interfere with the presentation of OVA. Several inhibitors especially of serine or thiol proteases rather enhanced the presentation of insulin. This indicates that intracellular proteolytic cleavage of insulin does not seem to be required for generation of the antigenic determinant but, if it occurs, rather destroys the antigenic peptide. Insulin and its A-chain fragment may, therefore, represent a model for a processing-dependent antigen not requiring proteolytic cleavage but other modifications.

Processing requirements for the recognition of insulin fragments by murine T cells.

In this study we investigated aspects of antigen processing using insulin and insulin A chain-derived fragments as model antigens in Ab alpha Ak beta-restricted T-cell stimulation. Similarly to other proteins, the immunodominant region of insulin recognized by these T cells is limited in size. It is located on the insulin A chain and encompasses a portion of the molecule that is represented faithfully by peptide A1-14(SSO3-)3. Efficient presentation of intact insulin and its entire A chain is dependent on uptake and processing by APC. Whereas peptides stemming from various globular proteins are known to be presented to T cells by APC without requiring processing, this is not the case with A-chain fragment A1-14 (SSO3-)3. This observation suggested that, in addition to proteolytic degradation, other mechanisms might play a role in the processing of these antigens. Three cys-residues are located in close proximity to those amino acid residues of the insulin A chain that are inferred to participate in the specific interaction with MHC class II molecules and the TcR. In A-chain derivatives that are stimulatory for the T cells or in intact insulin these cys residues are engaged in disulfide bonds or are S-sulfonated. Both linkages can be reversibly modified by reaction with thiols. Functional data indicate that from intact insulin and from structurally distinct A-chain derivatives a closely similar or identical peptide is formed and bound to class II molecules for recognition by the T cells. The question arises as to whether, in this processed peptide, the cys residues are present in reduced form, engaged in disulfide bonds, or are modified in some other way. Taken together, these findings suggest that modification of cys residues or isomerization of disulfide bonds may play a role in insulin processing. It can be expected that other proteins carrying cys residues in their immunodominant peptides may show similar processing requirements. The inhibition of N-glycosylation of proteins by tunicamycin in APC blocked the processing and presentation of insulin and OvA whereas, under the same conditions, the presentation of a processing-independent peptide was not affected. Furthermore, an autoreactive T-cell clone was capable of recognizing tunicamycin-treated APC. Since the expression of class II molecules was found to be unaltered as demonstrated by cytofluorometric analysis the deficient N-glycosylation appears to have little influence on class II antigen-mediated T-cell recognition but interferes with uptake of antigen and/or its processing by APC.